From fdc7bd909a5f38793468e9cf9b6a9063d96c6234 Mon Sep 17 00:00:00 2001 From: Sam Edwards Date: Sat, 29 Mar 2025 09:50:17 -0700 Subject: [PATCH 001/555] arm64: dts: rockchip: Allow Turing RK1 cooling fan to spin down The RK3588 thermal sensor driver only receives interrupts when a higher-temperature threshold is crossed; it cannot notify when the sensor cools back off. As a result, the driver must poll for temperature changes to detect when the conditions for a thermal trip are no longer met. However, it only does so if the DT enables polling. Before this patch, the RK1 DT did not enable polling, causing the fan to continue running at the speed corresponding to the highest temperature reached. Follow suit with similar RK3588 boards by setting a polling-delay of 1000ms, enabling the driver to detect when the sensor cools back off, allowing the fan speed to decrease as appropriate. Fixes: 7c8ec5e6b9d6 ("arm64: dts: rockchip: Enable automatic fan control on Turing RK1") Cc: stable@kernel.org # v6.13+ Signed-off-by: Sam Edwards Reviewed-by: Dragan Simic Link: https://lore.kernel.org/r/20250329165017.3885-1-CFSworks@gmail.com Signed-off-by: Heiko Stuebner --- arch/arm64/boot/dts/rockchip/rk3588-turing-rk1.dtsi | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/arm64/boot/dts/rockchip/rk3588-turing-rk1.dtsi b/arch/arm64/boot/dts/rockchip/rk3588-turing-rk1.dtsi index 711ac4f2c7cb..60ad272982ad 100644 --- a/arch/arm64/boot/dts/rockchip/rk3588-turing-rk1.dtsi +++ b/arch/arm64/boot/dts/rockchip/rk3588-turing-rk1.dtsi @@ -214,6 +214,8 @@ }; &package_thermal { + polling-delay = <1000>; + trips { package_active1: trip-active1 { temperature = <45000>; From e0bd7ecf6b2dc71215af699dffbf14bf0bc3d978 Mon Sep 17 00:00:00 2001 From: Dragan Simic Date: Mon, 24 Mar 2025 12:00:43 +0100 Subject: [PATCH 002/555] arm64: dts: rockchip: Remove overdrive-mode OPPs from RK3588J SoC dtsi The differences in the vendor-approved CPU and GPU OPPs for the standard Rockchip RK3588 variant [1] and the industrial Rockchip RK3588J variant [2] come from the latter, presumably, supporting an extended temperature range that's usually associated with industrial applications, despite the two SoC variant datasheets specifying the same upper limit for the allowed ambient temperature for both variants. However, the lower temperature limit is specified much lower for the RK3588J variant. [1][2] To be on the safe side and to ensure maximum longevity of the RK3588J SoCs, only the CPU and GPU OPPs that are declared by the vendor to be always safe for this SoC variant may be provided. As explained by the vendor [3] and according to the RK3588J datasheet, [2] higher-frequency/higher-voltage CPU and GPU OPPs can be used as well, but at the risk of reducing the SoC lifetime expectancy. Presumably, using the higher OPPs may be safe only when not enjoying the assumed extended temperature range that the RK3588J, as an SoC variant targeted specifically at higher-temperature, industrial applications, is made (or binned) for. Anyone able to keep their RK3588J-based board outside the above-presumed extended temperature range at all times, and willing to take the associated risk of possibly reducing the SoC lifetime expectancy, is free to apply a DT overlay that adds the higher CPU and GPU OPPs. With all this and the downstream RK3588(J) DT definitions [4][5] in mind, let's delete the RK3588J CPU and GPU OPPs that are not considered belonging to the normal operation mode for this SoC variant. To quote the RK3588J datasheet [2], "normal mode means the chipset works under safety voltage and frequency; for the industrial environment, highly recommend to keep in normal mode, the lifetime is reasonably guaranteed", while "overdrive mode brings higher frequency, and the voltage will increase accordingly; under the overdrive mode for a long time, the chipset may shorten the lifetime, especially in high-temperature condition". To sum the RK3588J datasheet [2] and the vendor-provided DTs up, [4][5] the maximum allowed CPU core, GPU and NPU frequencies are as follows: IP core | Normal mode | Overdrive mode ------------+-------------+---------------- Cortex-A55 | 1,296 MHz | 1,704 MHz Cortex-A76 | 1,608 MHz | 2,016 MHz GPU | 700 MHz | 850 MHz NPU | 800 MHz | 950 MHz Unfortunately, when it comes to the actual voltages for the RK3588J CPU and GPU OPPs, there's a discrepancy between the RK3588J datasheet [2] and the downstream kernel code. [4][5] The RK3588J datasheet states that "the max. working voltage of CPU/GPU/NPU is 0.75 V under the normal mode", while the downstream kernel code actually allows voltage ranges that go up to 0.95 V, which is still within the voltage range allowed by the datasheet. However, the RK3588J datasheet also tells us to "strictly refer to the software configuration of SDK and the hardware reference design", so let's embrace the voltage ranges provided by the downstream kernel code, which also prevents the undesirable theoretical outcome of ending up with no usable OPPs on a particular board, as a result of the board's voltage regulator(s) being unable to deliver the exact voltages, for whatever reason. The above-described voltage ranges for the RK3588J CPU OPPs remain taken from the downstream kernel code [4][5] by picking the highest, worst-bin values, which ensure that all RK3588J bins will work reliably. Yes, with some power inevitably wasted as unnecessarily generated heat, but the reliability is paramount, together with the longevity. This deficiency may be revisited separately at some point in the future. The provided RK3588J CPU OPPs follow the slightly debatable "provide only the highest-frequency OPP from the same-voltage group" approach that's been established earlier, [6] as a result of the "same-voltage, lower-frequency" OPPs being considered inefficient from the IPA governor's standpoint, which may also be revisited separately at some point in the future. [1] https://wiki.friendlyelec.com/wiki/images/e/ee/Rockchip_RK3588_Datasheet_V1.6-20231016.pdf [2] https://wmsc.lcsc.com/wmsc/upload/file/pdf/v2/lcsc/2403201054_Rockchip-RK3588J_C22364189.pdf [3] https://lore.kernel.org/linux-rockchip/e55125ed-64fb-455e-b1e4-cebe2cf006e4@cherry.de/T/#u [4] https://raw.githubusercontent.com/rockchip-linux/kernel/604cec4004abe5a96c734f2fab7b74809d2d742f/arch/arm64/boot/dts/rockchip/rk3588s.dtsi [5] https://raw.githubusercontent.com/rockchip-linux/kernel/604cec4004abe5a96c734f2fab7b74809d2d742f/arch/arm64/boot/dts/rockchip/rk3588j.dtsi [6] https://lore.kernel.org/all/20240229-rk-dts-additions-v3-5-6afe8473a631@gmail.com/ Fixes: 667885a68658 ("arm64: dts: rockchip: Add OPP data for CPU cores on RK3588j") Fixes: a7b2070505a2 ("arm64: dts: rockchip: Split GPU OPPs of RK3588 and RK3588j") Cc: stable@vger.kernel.org Cc: Heiko Stuebner Cc: Alexey Charkov Helped-by: Quentin Schulz Reviewed-by: Quentin Schulz Signed-off-by: Dragan Simic Link: https://lore.kernel.org/r/eeec0d30d79b019d111b3f0aa2456e69896b2caa.1742813866.git.dsimic@manjaro.org Signed-off-by: Heiko Stuebner --- arch/arm64/boot/dts/rockchip/rk3588j.dtsi | 53 ++++++++--------------- 1 file changed, 17 insertions(+), 36 deletions(-) diff --git a/arch/arm64/boot/dts/rockchip/rk3588j.dtsi b/arch/arm64/boot/dts/rockchip/rk3588j.dtsi index bce72bac4503..3045cb3bd68c 100644 --- a/arch/arm64/boot/dts/rockchip/rk3588j.dtsi +++ b/arch/arm64/boot/dts/rockchip/rk3588j.dtsi @@ -11,20 +11,15 @@ compatible = "operating-points-v2"; opp-shared; - opp-1416000000 { - opp-hz = /bits/ 64 <1416000000>; + opp-1200000000 { + opp-hz = /bits/ 64 <1200000000>; opp-microvolt = <750000 750000 950000>; clock-latency-ns = <40000>; opp-suspend; }; - opp-1608000000 { - opp-hz = /bits/ 64 <1608000000>; - opp-microvolt = <887500 887500 950000>; - clock-latency-ns = <40000>; - }; - opp-1704000000 { - opp-hz = /bits/ 64 <1704000000>; - opp-microvolt = <937500 937500 950000>; + opp-1296000000 { + opp-hz = /bits/ 64 <1296000000>; + opp-microvolt = <775000 775000 950000>; clock-latency-ns = <40000>; }; }; @@ -33,9 +28,14 @@ compatible = "operating-points-v2"; opp-shared; + opp-1200000000{ + opp-hz = /bits/ 64 <1200000000>; + opp-microvolt = <750000 750000 950000>; + clock-latency-ns = <40000>; + }; opp-1416000000 { opp-hz = /bits/ 64 <1416000000>; - opp-microvolt = <750000 750000 950000>; + opp-microvolt = <762500 762500 950000>; clock-latency-ns = <40000>; }; opp-1608000000 { @@ -43,25 +43,20 @@ opp-microvolt = <787500 787500 950000>; clock-latency-ns = <40000>; }; - opp-1800000000 { - opp-hz = /bits/ 64 <1800000000>; - opp-microvolt = <875000 875000 950000>; - clock-latency-ns = <40000>; - }; - opp-2016000000 { - opp-hz = /bits/ 64 <2016000000>; - opp-microvolt = <950000 950000 950000>; - clock-latency-ns = <40000>; - }; }; cluster2_opp_table: opp-table-cluster2 { compatible = "operating-points-v2"; opp-shared; + opp-1200000000{ + opp-hz = /bits/ 64 <1200000000>; + opp-microvolt = <750000 750000 950000>; + clock-latency-ns = <40000>; + }; opp-1416000000 { opp-hz = /bits/ 64 <1416000000>; - opp-microvolt = <750000 750000 950000>; + opp-microvolt = <762500 762500 950000>; clock-latency-ns = <40000>; }; opp-1608000000 { @@ -69,16 +64,6 @@ opp-microvolt = <787500 787500 950000>; clock-latency-ns = <40000>; }; - opp-1800000000 { - opp-hz = /bits/ 64 <1800000000>; - opp-microvolt = <875000 875000 950000>; - clock-latency-ns = <40000>; - }; - opp-2016000000 { - opp-hz = /bits/ 64 <2016000000>; - opp-microvolt = <950000 950000 950000>; - clock-latency-ns = <40000>; - }; }; gpu_opp_table: opp-table { @@ -104,10 +89,6 @@ opp-hz = /bits/ 64 <700000000>; opp-microvolt = <750000 750000 850000>; }; - opp-850000000 { - opp-hz = /bits/ 64 <800000000>; - opp-microvolt = <787500 787500 850000>; - }; }; }; From 7fc027e894fae7c8661b52b1fde223004b2a8e0c Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= Date: Wed, 19 Mar 2025 12:31:38 +0100 Subject: [PATCH 003/555] arm64: dts: rockchip: Add pinmuxing for eMMC on QNAP TS433 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Until now the emmc worked when booting because the bootrom set up the pin config correctly to load the initial bootloader from it. So when the kernel started it "just" reused this setup but never made sure it was actually correct. This then breaks when the system is started via some other means, like downloading the initial bootloader via the bootrom usb download. With actual emmc pin-config added, barebox is able to access the eMMC even when booted via USB. Fixes: 9da1c0327d58 ("arm64: dts: rockchip: Add basic support for QNAP TS-433") Signed-off-by: Uwe Kleine-König [refined commit message to explain that we're currently just running on bootom-goodwill] Link: https://lore.kernel.org/r/20250319113138.125192-2-uwe@kleine-koenig.org Signed-off-by: Heiko Stuebner --- arch/arm64/boot/dts/rockchip/rk3568-qnap-ts433.dts | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/arm64/boot/dts/rockchip/rk3568-qnap-ts433.dts b/arch/arm64/boot/dts/rockchip/rk3568-qnap-ts433.dts index 7bd32d230ad2..b80d628c426b 100644 --- a/arch/arm64/boot/dts/rockchip/rk3568-qnap-ts433.dts +++ b/arch/arm64/boot/dts/rockchip/rk3568-qnap-ts433.dts @@ -619,6 +619,8 @@ bus-width = <8>; max-frequency = <200000000>; non-removable; + pinctrl-names = "default"; + pinctrl-0 = <&emmc_bus8 &emmc_clk &emmc_cmd &emmc_datastrobe>; status = "okay"; }; From 6d03811d7a99e08d5928f58120acb45b8ba22b08 Mon Sep 17 00:00:00 2001 From: Gustavo Silva Date: Tue, 4 Mar 2025 15:01:02 -0300 Subject: [PATCH 004/555] iio: imu: bmi270: fix initial sampling frequency configuration In the bmi270_configure_imu() function, the accelerometer and gyroscope configuration registers are incorrectly written with the mask BMI270_PWR_CONF_ADV_PWR_SAVE_MSK, which is unrelated to these registers. As a result, the accelerometer's sampling frequency is set to 200 Hz instead of the intended 100 Hz. Remove the mask to ensure the correct bits are set in the configuration registers. Fixes: 3ea51548d6b2 ("iio: imu: Add i2c driver for bmi270 imu") Signed-off-by: Gustavo Silva Reviewed-by: Alex Lanzano Link: https://patch.msgid.link/20250304-bmi270-odr-fix-v1-1-384dbcd699fb@gmail.com Signed-off-by: Jonathan Cameron --- drivers/iio/imu/bmi270/bmi270_core.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/drivers/iio/imu/bmi270/bmi270_core.c b/drivers/iio/imu/bmi270/bmi270_core.c index a86be5af5ccb..2e4469f30d53 100644 --- a/drivers/iio/imu/bmi270/bmi270_core.c +++ b/drivers/iio/imu/bmi270/bmi270_core.c @@ -918,8 +918,7 @@ static int bmi270_configure_imu(struct bmi270_data *data) FIELD_PREP(BMI270_ACC_CONF_ODR_MSK, BMI270_ACC_CONF_ODR_100HZ) | FIELD_PREP(BMI270_ACC_CONF_BWP_MSK, - BMI270_ACC_CONF_BWP_NORMAL_MODE) | - BMI270_PWR_CONF_ADV_PWR_SAVE_MSK); + BMI270_ACC_CONF_BWP_NORMAL_MODE)); if (ret) return dev_err_probe(dev, ret, "Failed to configure accelerometer"); @@ -927,8 +926,7 @@ static int bmi270_configure_imu(struct bmi270_data *data) FIELD_PREP(BMI270_GYR_CONF_ODR_MSK, BMI270_GYR_CONF_ODR_200HZ) | FIELD_PREP(BMI270_GYR_CONF_BWP_MSK, - BMI270_GYR_CONF_BWP_NORMAL_MODE) | - BMI270_PWR_CONF_ADV_PWR_SAVE_MSK); + BMI270_GYR_CONF_BWP_NORMAL_MODE)); if (ret) return dev_err_probe(dev, ret, "Failed to configure gyroscope"); From 38f67d0264929762e54ae5948703a21f841fe706 Mon Sep 17 00:00:00 2001 From: Lothar Rubusch Date: Sun, 9 Mar 2025 19:35:15 +0000 Subject: [PATCH 005/555] iio: accel: adxl367: fix setting odr for activity time update Fix setting the odr value to update activity time based on frequency derrived by recent odr, and not by obsolete odr value. The [small] bug: When _adxl367_set_odr() is called with a new odr value, it first writes the new odr value to the hardware register ADXL367_REG_FILTER_CTL. Second, it calls _adxl367_set_act_time_ms(), which calls adxl367_time_ms_to_samples(). Here st->odr still holds the old odr value. This st->odr member is used to derrive a frequency value, which is applied to update ADXL367_REG_TIME_ACT. Hence, the idea is to update activity time, based on possibilities and power consumption by the current ODR rate. Finally, when the function calls return, again in _adxl367_set_odr() the new ODR is assigned to st->odr. The fix: When setting a new ODR value is set to ADXL367_REG_FILTER_CTL, also ADXL367_REG_TIME_ACT should probably be updated with a frequency based on the recent ODR value and not the old one. Changing the location of the assignment to st->odr fixes this. Fixes: cbab791c5e2a5 ("iio: accel: add ADXL367 driver") Signed-off-by: Lothar Rubusch Reviewed-by: Marcelo Schmitt Link: https://patch.msgid.link/20250309193515.2974-1-l.rubusch@gmail.com Signed-off-by: Jonathan Cameron --- drivers/iio/accel/adxl367.c | 10 +++------- 1 file changed, 3 insertions(+), 7 deletions(-) diff --git a/drivers/iio/accel/adxl367.c b/drivers/iio/accel/adxl367.c index add4053e7a02..0c04b2bb7efb 100644 --- a/drivers/iio/accel/adxl367.c +++ b/drivers/iio/accel/adxl367.c @@ -601,18 +601,14 @@ static int _adxl367_set_odr(struct adxl367_state *st, enum adxl367_odr odr) if (ret) return ret; + st->odr = odr; + /* Activity timers depend on ODR */ ret = _adxl367_set_act_time_ms(st, st->act_time_ms); if (ret) return ret; - ret = _adxl367_set_inact_time_ms(st, st->inact_time_ms); - if (ret) - return ret; - - st->odr = odr; - - return 0; + return _adxl367_set_inact_time_ms(st, st->inact_time_ms); } static int adxl367_set_odr(struct iio_dev *indio_dev, enum adxl367_odr odr) From 159ca7f18129834b6f4c7eae67de48e96c752fc9 Mon Sep 17 00:00:00 2001 From: Silvano Seva Date: Tue, 11 Mar 2025 09:49:47 +0100 Subject: [PATCH 006/555] iio: imu: st_lsm6dsx: fix possible lockup in st_lsm6dsx_read_fifo Prevent st_lsm6dsx_read_fifo from falling in an infinite loop in case pattern_len is equal to zero and the device FIFO is not empty. Fixes: 290a6ce11d93 ("iio: imu: add support to lsm6dsx driver") Signed-off-by: Silvano Seva Acked-by: Lorenzo Bianconi Link: https://patch.msgid.link/20250311085030.3593-2-s.seva@4sigma.it Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_buffer.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_buffer.c b/drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_buffer.c index 0a7cd8c1aa33..480a9b31065c 100644 --- a/drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_buffer.c +++ b/drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_buffer.c @@ -392,6 +392,9 @@ int st_lsm6dsx_read_fifo(struct st_lsm6dsx_hw *hw) if (fifo_status & cpu_to_le16(ST_LSM6DSX_FIFO_EMPTY_MASK)) return 0; + if (!pattern_len) + pattern_len = ST_LSM6DSX_SAMPLE_SIZE; + fifo_len = (le16_to_cpu(fifo_status) & fifo_diff_mask) * ST_LSM6DSX_CHAN_SIZE; fifo_len = (fifo_len / pattern_len) * pattern_len; From 8114ef86e2058e2554111b793596f17bee23fa15 Mon Sep 17 00:00:00 2001 From: Silvano Seva Date: Tue, 11 Mar 2025 09:49:49 +0100 Subject: [PATCH 007/555] iio: imu: st_lsm6dsx: fix possible lockup in st_lsm6dsx_read_tagged_fifo Prevent st_lsm6dsx_read_tagged_fifo from falling in an infinite loop in case pattern_len is equal to zero and the device FIFO is not empty. Fixes: 801a6e0af0c6 ("iio: imu: st_lsm6dsx: add support to LSM6DSO") Signed-off-by: Silvano Seva Acked-by: Lorenzo Bianconi Link: https://patch.msgid.link/20250311085030.3593-4-s.seva@4sigma.it Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_buffer.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_buffer.c b/drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_buffer.c index 480a9b31065c..8a9d2593576a 100644 --- a/drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_buffer.c +++ b/drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_buffer.c @@ -626,6 +626,9 @@ int st_lsm6dsx_read_tagged_fifo(struct st_lsm6dsx_hw *hw) if (!fifo_len) return 0; + if (!pattern_len) + pattern_len = ST_LSM6DSX_TAGGED_SAMPLE_SIZE; + for (read_len = 0; read_len < fifo_len; read_len += pattern_len) { err = st_lsm6dsx_read_block(hw, ST_LSM6DSX_REG_FIFO_OUT_TAG_ADDR, From 839f81de397019f55161c5982d670ac19d836173 Mon Sep 17 00:00:00 2001 From: Simon Xue Date: Wed, 12 Mar 2025 14:20:16 +0800 Subject: [PATCH 008/555] iio: adc: rockchip: Fix clock initialization sequence clock_set_rate should be executed after devm_clk_get_enabled. Fixes: 97ad10bb2901 ("iio: adc: rockchip_saradc: Make use of devm_clk_get_enabled") Signed-off-by: Simon Xue Reviewed-by: Heiko Stuebner Link: https://patch.msgid.link/20250312062016.137821-1-xxm@rock-chips.com Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/adc/rockchip_saradc.c | 17 ++++++++--------- 1 file changed, 8 insertions(+), 9 deletions(-) diff --git a/drivers/iio/adc/rockchip_saradc.c b/drivers/iio/adc/rockchip_saradc.c index 9a099df79518..5e28bd28b81a 100644 --- a/drivers/iio/adc/rockchip_saradc.c +++ b/drivers/iio/adc/rockchip_saradc.c @@ -520,15 +520,6 @@ static int rockchip_saradc_probe(struct platform_device *pdev) if (info->reset) rockchip_saradc_reset_controller(info->reset); - /* - * Use a default value for the converter clock. - * This may become user-configurable in the future. - */ - ret = clk_set_rate(info->clk, info->data->clk_rate); - if (ret < 0) - return dev_err_probe(&pdev->dev, ret, - "failed to set adc clk rate\n"); - ret = regulator_enable(info->vref); if (ret < 0) return dev_err_probe(&pdev->dev, ret, @@ -555,6 +546,14 @@ static int rockchip_saradc_probe(struct platform_device *pdev) if (IS_ERR(info->clk)) return dev_err_probe(&pdev->dev, PTR_ERR(info->clk), "failed to get adc clock\n"); + /* + * Use a default value for the converter clock. + * This may become user-configurable in the future. + */ + ret = clk_set_rate(info->clk, info->data->clk_rate); + if (ret < 0) + return dev_err_probe(&pdev->dev, ret, + "failed to set adc clk rate\n"); platform_set_drvdata(pdev, indio_dev); From 82c51ac74071b80b3199d9e200ae1a5399f4deb0 Mon Sep 17 00:00:00 2001 From: David Lechner Date: Thu, 20 Mar 2025 11:21:52 -0500 Subject: [PATCH 009/555] iio: adc: ad7380: disable offload before using SPI bus MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Move disabling of the SPI offload before attempting to use the SPI bus to write a register in ad7380_offload_buffer_predisable(). This caused a crash in the spi_engine_irq() interrupt handler due to being in an invalid state. Fixes: bbeaec81a03e ("iio: ad7380: add support for SPI offload") Signed-off-by: David Lechner Reviewed-by: Nuno Sá Reviewed-by: Angelo Dureghello Link: https://patch.msgid.link/20250320-iio-adc-ad7380-fix-spi-offload-buffer-predisable-v1-1-6912ac8c0ae0@baylibre.com Signed-off-by: Jonathan Cameron --- drivers/iio/adc/ad7380.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/drivers/iio/adc/ad7380.c b/drivers/iio/adc/ad7380.c index 4fcb49fdf566..a2b41980c942 100644 --- a/drivers/iio/adc/ad7380.c +++ b/drivers/iio/adc/ad7380.c @@ -1211,6 +1211,9 @@ static int ad7380_offload_buffer_predisable(struct iio_dev *indio_dev) struct ad7380_state *st = iio_priv(indio_dev); int ret; + spi_offload_trigger_disable(st->offload, st->offload_trigger); + spi_unoptimize_message(&st->offload_msg); + if (st->seq) { ret = regmap_update_bits(st->regmap, AD7380_REG_ADDR_CONFIG1, @@ -1222,10 +1225,6 @@ static int ad7380_offload_buffer_predisable(struct iio_dev *indio_dev) st->seq = false; } - spi_offload_trigger_disable(st->offload, st->offload_trigger); - - spi_unoptimize_message(&st->offload_msg); - return 0; } From f063a28002e3350088b4577c5640882bf4ea17ea Mon Sep 17 00:00:00 2001 From: Luca Ceresoli Date: Fri, 21 Mar 2025 19:10:00 +0100 Subject: [PATCH 010/555] iio: light: opt3001: fix deadlock due to concurrent flag access The threaded IRQ function in this driver is reading the flag twice: once to lock a mutex and once to unlock it. Even though the code setting the flag is designed to prevent it, there are subtle cases where the flag could be true at the mutex_lock stage and false at the mutex_unlock stage. This results in the mutex not being unlocked, resulting in a deadlock. Fix it by making the opt3001_irq() code generally more robust, reading the flag into a variable and using the variable value at both stages. Fixes: 94a9b7b1809f ("iio: light: add support for TI's opt3001 light sensor") Cc: stable@vger.kernel.org Signed-off-by: Luca Ceresoli Link: https://patch.msgid.link/20250321-opt3001-irq-fix-v1-1-6c520d851562@bootlin.com Signed-off-by: Jonathan Cameron --- drivers/iio/light/opt3001.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/iio/light/opt3001.c b/drivers/iio/light/opt3001.c index 65b295877b41..393a3d2fbe1d 100644 --- a/drivers/iio/light/opt3001.c +++ b/drivers/iio/light/opt3001.c @@ -788,8 +788,9 @@ static irqreturn_t opt3001_irq(int irq, void *_iio) int ret; bool wake_result_ready_queue = false; enum iio_chan_type chan_type = opt->chip_info->chan_type; + bool ok_to_ignore_lock = opt->ok_to_ignore_lock; - if (!opt->ok_to_ignore_lock) + if (!ok_to_ignore_lock) mutex_lock(&opt->lock); ret = i2c_smbus_read_word_swapped(opt->client, OPT3001_CONFIGURATION); @@ -826,7 +827,7 @@ static irqreturn_t opt3001_irq(int irq, void *_iio) } out: - if (!opt->ok_to_ignore_lock) + if (!ok_to_ignore_lock) mutex_unlock(&opt->lock); if (wake_result_ready_queue) From 5257d80e22bf27009d6742e4c174f42cfe54e425 Mon Sep 17 00:00:00 2001 From: David Lechner Date: Tue, 18 Mar 2025 17:52:09 -0500 Subject: [PATCH 011/555] iio: adc: ad7606: check for NULL before calling sw_mode_config() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Check that the sw_mode_config function pointer is not NULL before calling it. Not all buses define this callback, which resulted in a NULL pointer dereference. Fixes: e571c1902116 ("iio: adc: ad7606: move scale_setup as function pointer on chip-info") Reviewed-by: Nuno Sá Signed-off-by: David Lechner Link: https://patch.msgid.link/20250318-iio-adc-ad7606-improvements-v2-1-4b605427774c@baylibre.com Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/adc/ad7606.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/drivers/iio/adc/ad7606.c b/drivers/iio/adc/ad7606.c index 1a314fddd7eb..703556eb7257 100644 --- a/drivers/iio/adc/ad7606.c +++ b/drivers/iio/adc/ad7606.c @@ -1236,9 +1236,11 @@ static int ad7616_sw_mode_setup(struct iio_dev *indio_dev) st->write_scale = ad7616_write_scale_sw; st->write_os = &ad7616_write_os_sw; - ret = st->bops->sw_mode_config(indio_dev); - if (ret) - return ret; + if (st->bops->sw_mode_config) { + ret = st->bops->sw_mode_config(indio_dev); + if (ret) + return ret; + } /* Activate Burst mode and SEQEN MODE */ return ad7606_write_mask(st, AD7616_CONFIGURATION_REGISTER, @@ -1268,6 +1270,9 @@ static int ad7606b_sw_mode_setup(struct iio_dev *indio_dev) st->write_scale = ad7606_write_scale_sw; st->write_os = &ad7606_write_os_sw; + if (!st->bops->sw_mode_config) + return 0; + return st->bops->sw_mode_config(indio_dev); } From 83ded7cfaccccd2f4041769c313b58b4c9e265ad Mon Sep 17 00:00:00 2001 From: Zhang Lixu Date: Mon, 31 Mar 2025 13:50:20 +0800 Subject: [PATCH 012/555] iio: hid-sensor-prox: Restore lost scale assignments The variables `scale_pre_decml`, `scale_post_decml`, and `scale_precision` were assigned in commit d68c592e02f6 ("iio: hid-sensor-prox: Fix scale not correct issue"), but due to a merge conflict in commit 9c15db92a8e5 ("Merge tag 'iio-for-5.13a' of https://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio into staging-next"), these assignments were lost. Add back lost assignments and replace `st->prox_attr` with `st->prox_attr[0]` because commit 596ef5cf654b ("iio: hid-sensor-prox: Add support for more channels") changed `prox_attr` to an array. Cc: stable@vger.kernel.org # 5.13+ Fixes: 9c15db92a8e5 ("Merge tag 'iio-for-5.13a' of https://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio into staging-next") Signed-off-by: Zhang Lixu Acked-by: Srinivas Pandruvada Link: https://patch.msgid.link/20250331055022.1149736-2-lixu.zhang@intel.com Signed-off-by: Jonathan Cameron --- drivers/iio/light/hid-sensor-prox.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/iio/light/hid-sensor-prox.c b/drivers/iio/light/hid-sensor-prox.c index 76b76d12b388..1dc6fb7cf614 100644 --- a/drivers/iio/light/hid-sensor-prox.c +++ b/drivers/iio/light/hid-sensor-prox.c @@ -257,6 +257,11 @@ static int prox_parse_report(struct platform_device *pdev, st->num_channels = index; + st->scale_precision = hid_sensor_format_scale(hsdev->usage, + &st->prox_attr[0], + &st->scale_pre_decml, + &st->scale_post_decml); + return 0; } From 8b518cdb03f5f6e06d635cbfd9583d1fdbb39bfd Mon Sep 17 00:00:00 2001 From: Zhang Lixu Date: Mon, 31 Mar 2025 13:50:21 +0800 Subject: [PATCH 013/555] iio: hid-sensor-prox: support multi-channel SCALE calculation With the introduction of multi-channel support in commit 596ef5cf654b ("iio: hid-sensor-prox: Add support for more channels"), each channel requires an independent SCALE calculation, but the existing code only calculates SCALE for a single channel. Addresses the problem by modifying the driver to perform independent SCALE calculations for each channel. Cc: stable@vger.kernel.org Fixes: 596ef5cf654b ("iio: hid-sensor-prox: Add support for more channels") Signed-off-by: Zhang Lixu Acked-by: Srinivas Pandruvada Link: https://patch.msgid.link/20250331055022.1149736-3-lixu.zhang@intel.com Signed-off-by: Jonathan Cameron --- .../hid-sensors/hid-sensor-attributes.c | 4 ++++ drivers/iio/light/hid-sensor-prox.c | 24 ++++++++++--------- 2 files changed, 17 insertions(+), 11 deletions(-) diff --git a/drivers/iio/common/hid-sensors/hid-sensor-attributes.c b/drivers/iio/common/hid-sensors/hid-sensor-attributes.c index ad1882f608c0..2055a03cbeb1 100644 --- a/drivers/iio/common/hid-sensors/hid-sensor-attributes.c +++ b/drivers/iio/common/hid-sensors/hid-sensor-attributes.c @@ -66,6 +66,10 @@ static struct { {HID_USAGE_SENSOR_HUMIDITY, 0, 1000, 0}, {HID_USAGE_SENSOR_HINGE, 0, 0, 17453293}, {HID_USAGE_SENSOR_HINGE, HID_USAGE_SENSOR_UNITS_DEGREES, 0, 17453293}, + + {HID_USAGE_SENSOR_HUMAN_PRESENCE, 0, 1, 0}, + {HID_USAGE_SENSOR_HUMAN_PROXIMITY, 0, 1, 0}, + {HID_USAGE_SENSOR_HUMAN_ATTENTION, 0, 1, 0}, }; static void simple_div(int dividend, int divisor, int *whole, diff --git a/drivers/iio/light/hid-sensor-prox.c b/drivers/iio/light/hid-sensor-prox.c index 1dc6fb7cf614..941508e58286 100644 --- a/drivers/iio/light/hid-sensor-prox.c +++ b/drivers/iio/light/hid-sensor-prox.c @@ -34,9 +34,9 @@ struct prox_state { struct iio_chan_spec channels[MAX_CHANNELS]; u32 channel2usage[MAX_CHANNELS]; u32 human_presence[MAX_CHANNELS]; - int scale_pre_decml; - int scale_post_decml; - int scale_precision; + int scale_pre_decml[MAX_CHANNELS]; + int scale_post_decml[MAX_CHANNELS]; + int scale_precision[MAX_CHANNELS]; unsigned long scan_mask[2]; /* One entry plus one terminator. */ int num_channels; }; @@ -116,9 +116,12 @@ static int prox_read_raw(struct iio_dev *indio_dev, ret_type = IIO_VAL_INT; break; case IIO_CHAN_INFO_SCALE: - *val = prox_state->scale_pre_decml; - *val2 = prox_state->scale_post_decml; - ret_type = prox_state->scale_precision; + if (chan->scan_index >= prox_state->num_channels) + return -EINVAL; + + *val = prox_state->scale_pre_decml[chan->scan_index]; + *val2 = prox_state->scale_post_decml[chan->scan_index]; + ret_type = prox_state->scale_precision[chan->scan_index]; break; case IIO_CHAN_INFO_OFFSET: *val = hid_sensor_convert_exponent( @@ -249,6 +252,10 @@ static int prox_parse_report(struct platform_device *pdev, st->prox_attr[index].size); dev_dbg(&pdev->dev, "prox %x:%x\n", st->prox_attr[index].index, st->prox_attr[index].report_id); + st->scale_precision[index] = + hid_sensor_format_scale(usage_id, &st->prox_attr[index], + &st->scale_pre_decml[index], + &st->scale_post_decml[index]); index++; } @@ -257,11 +264,6 @@ static int prox_parse_report(struct platform_device *pdev, st->num_channels = index; - st->scale_precision = hid_sensor_format_scale(hsdev->usage, - &st->prox_attr[0], - &st->scale_pre_decml, - &st->scale_post_decml); - return 0; } From 79dabbd505210e41c88060806c92c052496dd61c Mon Sep 17 00:00:00 2001 From: Zhang Lixu Date: Mon, 31 Mar 2025 13:50:22 +0800 Subject: [PATCH 014/555] iio: hid-sensor-prox: Fix incorrect OFFSET calculation The OFFSET calculation in the prox_read_raw() was incorrectly using the unit exponent, which is intended for SCALE calculations. Remove the incorrect OFFSET calculation and set it to a fixed value of 0. Cc: stable@vger.kernel.org Fixes: 39a3a0138f61 ("iio: hid-sensors: Added Proximity Sensor Driver") Signed-off-by: Zhang Lixu Acked-by: Srinivas Pandruvada Link: https://patch.msgid.link/20250331055022.1149736-4-lixu.zhang@intel.com Signed-off-by: Jonathan Cameron --- drivers/iio/light/hid-sensor-prox.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/iio/light/hid-sensor-prox.c b/drivers/iio/light/hid-sensor-prox.c index 941508e58286..4c65b32d34ce 100644 --- a/drivers/iio/light/hid-sensor-prox.c +++ b/drivers/iio/light/hid-sensor-prox.c @@ -124,8 +124,7 @@ static int prox_read_raw(struct iio_dev *indio_dev, ret_type = prox_state->scale_precision[chan->scan_index]; break; case IIO_CHAN_INFO_OFFSET: - *val = hid_sensor_convert_exponent( - prox_state->prox_attr[chan->scan_index].unit_expo); + *val = 0; ret_type = IIO_VAL_INT; break; case IIO_CHAN_INFO_SAMP_FREQ: From 2d7b60f33da324abe7824037b4829ff7df70e435 Mon Sep 17 00:00:00 2001 From: David Lechner Date: Wed, 2 Apr 2025 18:55:58 -0500 Subject: [PATCH 015/555] iio: adc: ad7380: fix event threshold shift Add required bit shift to the event threshold read function to get correct scaling. When alert support was added, the write function correctly included the required shift needed to convert the threshold register value to the same scale as the raw ADC value. However, the shift got missed in the read function. Fixes: 27d1a4dbe1e1 ("iio: adc: ad7380: add alert support") Signed-off-by: David Lechner Reviewed-by: Julien Stephan Link: https://patch.msgid.link/20250402-iio-adc-ad7380-fix-event-threshold-shift-v1-1-ad4975c296b2@baylibre.com Signed-off-by: Jonathan Cameron --- drivers/iio/adc/ad7380.c | 25 +++++++++++++++++++------ 1 file changed, 19 insertions(+), 6 deletions(-) diff --git a/drivers/iio/adc/ad7380.c b/drivers/iio/adc/ad7380.c index a2b41980c942..aef85093eb16 100644 --- a/drivers/iio/adc/ad7380.c +++ b/drivers/iio/adc/ad7380.c @@ -1610,11 +1610,25 @@ static int ad7380_write_event_config(struct iio_dev *indio_dev, return ret; } -static int ad7380_get_alert_th(struct ad7380_state *st, +static int ad7380_get_alert_th(struct iio_dev *indio_dev, + const struct iio_chan_spec *chan, enum iio_event_direction dir, int *val) { - int ret, tmp; + struct ad7380_state *st = iio_priv(indio_dev); + const struct iio_scan_type *scan_type; + int ret, tmp, shift; + + scan_type = iio_get_current_scan_type(indio_dev, chan); + if (IS_ERR(scan_type)) + return PTR_ERR(scan_type); + + /* + * The register value is 12-bits and is compared to the most significant + * bits of raw value, therefore a shift is required to convert this to + * the same scale as the raw value. + */ + shift = scan_type->realbits - 12; switch (dir) { case IIO_EV_DIR_RISING: @@ -1624,7 +1638,7 @@ static int ad7380_get_alert_th(struct ad7380_state *st, if (ret) return ret; - *val = FIELD_GET(AD7380_ALERT_HIGH_TH, tmp); + *val = FIELD_GET(AD7380_ALERT_HIGH_TH, tmp) << shift; return IIO_VAL_INT; case IIO_EV_DIR_FALLING: ret = regmap_read(st->regmap, @@ -1633,7 +1647,7 @@ static int ad7380_get_alert_th(struct ad7380_state *st, if (ret) return ret; - *val = FIELD_GET(AD7380_ALERT_LOW_TH, tmp); + *val = FIELD_GET(AD7380_ALERT_LOW_TH, tmp) << shift; return IIO_VAL_INT; default: return -EINVAL; @@ -1647,7 +1661,6 @@ static int ad7380_read_event_value(struct iio_dev *indio_dev, enum iio_event_info info, int *val, int *val2) { - struct ad7380_state *st = iio_priv(indio_dev); int ret; switch (info) { @@ -1655,7 +1668,7 @@ static int ad7380_read_event_value(struct iio_dev *indio_dev, if (!iio_device_claim_direct(indio_dev)) return -EBUSY; - ret = ad7380_get_alert_th(st, dir, val); + ret = ad7380_get_alert_th(indio_dev, chan, dir, val); iio_device_release_direct(indio_dev); return ret; From fd15594ba7d559d9da741504c322b9f57c4981e5 Mon Sep 17 00:00:00 2001 From: Charles Keepax Date: Wed, 9 Apr 2025 13:22:39 +0100 Subject: [PATCH 016/555] soundwire: bus: Fix race on the creation of the IRQ domain The SoundWire IRQ domain needs to be created before any slaves are added to the bus, such that the domain is always available when needed. Move the call to sdw_irq_create() before the calls to sdw_acpi_find_slaves() and sdw_of_find_slaves(). Fixes: 12a95123bfe1 ("soundwire: bus: Allow SoundWire peripherals to register IRQ handlers") Signed-off-by: Charles Keepax Link: https://lore.kernel.org/r/20250409122239.1396489-1-ckeepax@opensource.cirrus.com Signed-off-by: Vinod Koul --- drivers/soundwire/bus.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/soundwire/bus.c b/drivers/soundwire/bus.c index 6f8a20014e76..39aecd34c641 100644 --- a/drivers/soundwire/bus.c +++ b/drivers/soundwire/bus.c @@ -122,6 +122,10 @@ int sdw_bus_master_add(struct sdw_bus *bus, struct device *parent, set_bit(SDW_GROUP13_DEV_NUM, bus->assigned); set_bit(SDW_MASTER_DEV_NUM, bus->assigned); + ret = sdw_irq_create(bus, fwnode); + if (ret) + return ret; + /* * SDW is an enumerable bus, but devices can be powered off. So, * they won't be able to report as present. @@ -138,6 +142,7 @@ int sdw_bus_master_add(struct sdw_bus *bus, struct device *parent, if (ret < 0) { dev_err(bus->dev, "Finding slaves failed:%d\n", ret); + sdw_irq_delete(bus); return ret; } @@ -156,10 +161,6 @@ int sdw_bus_master_add(struct sdw_bus *bus, struct device *parent, bus->params.curr_bank = SDW_BANK0; bus->params.next_bank = SDW_BANK1; - ret = sdw_irq_create(bus, fwnode); - if (ret) - return ret; - return 0; } EXPORT_SYMBOL(sdw_bus_master_add); From c8c59bcac9300c22abf1afd7e236978a413c3644 Mon Sep 17 00:00:00 2001 From: "Rob Herring (Arm)" Date: Wed, 9 Apr 2025 15:50:46 -0500 Subject: [PATCH 017/555] arm64: dts: rockchip: Use "regulator-fixed" for btreg on px30-engicam for vcc3v3-btreg The vcc3v3-btreg regulator only has 1 state and no state gpios defined, so "regulator-gpio" is not the correct binding to use. "regulator-fixed" is the correct binding to use. It supports an enable GPIO which is needed in this case. Signed-off-by: "Rob Herring (Arm)" Link: https://lore.kernel.org/r/20250409205047.1522943-1-robh@kernel.org Signed-off-by: Heiko Stuebner --- arch/arm64/boot/dts/rockchip/px30-engicam-common.dtsi | 3 +-- arch/arm64/boot/dts/rockchip/px30-engicam-ctouch2.dtsi | 2 +- .../boot/dts/rockchip/px30-engicam-px30-core-edimm2.2.dts | 2 +- 3 files changed, 3 insertions(+), 4 deletions(-) diff --git a/arch/arm64/boot/dts/rockchip/px30-engicam-common.dtsi b/arch/arm64/boot/dts/rockchip/px30-engicam-common.dtsi index 1edfd643b25a..a334ef0629d1 100644 --- a/arch/arm64/boot/dts/rockchip/px30-engicam-common.dtsi +++ b/arch/arm64/boot/dts/rockchip/px30-engicam-common.dtsi @@ -31,7 +31,7 @@ }; vcc3v3_btreg: vcc3v3-btreg { - compatible = "regulator-gpio"; + compatible = "regulator-fixed"; enable-active-high; pinctrl-names = "default"; pinctrl-0 = <&bt_enable_h>; @@ -39,7 +39,6 @@ regulator-min-microvolt = <3300000>; regulator-max-microvolt = <3300000>; regulator-always-on; - states = <3300000 0x0>; }; vcc3v3_rf_aux_mod: regulator-vcc3v3-rf-aux-mod { diff --git a/arch/arm64/boot/dts/rockchip/px30-engicam-ctouch2.dtsi b/arch/arm64/boot/dts/rockchip/px30-engicam-ctouch2.dtsi index 80db778c9684..b60e68faa83a 100644 --- a/arch/arm64/boot/dts/rockchip/px30-engicam-ctouch2.dtsi +++ b/arch/arm64/boot/dts/rockchip/px30-engicam-ctouch2.dtsi @@ -26,5 +26,5 @@ }; &vcc3v3_btreg { - enable-gpios = <&gpio1 RK_PC3 GPIO_ACTIVE_HIGH>; + gpios = <&gpio1 RK_PC3 GPIO_ACTIVE_HIGH>; }; diff --git a/arch/arm64/boot/dts/rockchip/px30-engicam-px30-core-edimm2.2.dts b/arch/arm64/boot/dts/rockchip/px30-engicam-px30-core-edimm2.2.dts index 165d09ccb942..5886b802c520 100644 --- a/arch/arm64/boot/dts/rockchip/px30-engicam-px30-core-edimm2.2.dts +++ b/arch/arm64/boot/dts/rockchip/px30-engicam-px30-core-edimm2.2.dts @@ -39,5 +39,5 @@ }; &vcc3v3_btreg { - enable-gpios = <&gpio1 RK_PC2 GPIO_ACTIVE_HIGH>; + gpios = <&gpio1 RK_PC2 GPIO_ACTIVE_HIGH>; }; From 6833cbdc733c9e1088fe9936b2dad95cc7d4c580 Mon Sep 17 00:00:00 2001 From: "Rob Herring (Arm)" Date: Wed, 9 Apr 2025 15:50:39 -0500 Subject: [PATCH 018/555] arm64: dts: rockchip: Fix mmc-pwrseq clock name on rock-pi-4 The defined name for "mmc-pwrseq-simple" clock is "ext_clock". Signed-off-by: "Rob Herring (Arm)" Link: https://lore.kernel.org/r/20250409205040.1522754-1-robh@kernel.org Signed-off-by: Heiko Stuebner --- arch/arm64/boot/dts/rockchip/rk3399-rock-pi-4.dtsi | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm64/boot/dts/rockchip/rk3399-rock-pi-4.dtsi b/arch/arm64/boot/dts/rockchip/rk3399-rock-pi-4.dtsi index 541dca12bf1a..046dbe329017 100644 --- a/arch/arm64/boot/dts/rockchip/rk3399-rock-pi-4.dtsi +++ b/arch/arm64/boot/dts/rockchip/rk3399-rock-pi-4.dtsi @@ -43,7 +43,7 @@ sdio_pwrseq: sdio-pwrseq { compatible = "mmc-pwrseq-simple"; clocks = <&rk808 1>; - clock-names = "lpo"; + clock-names = "ext_clock"; pinctrl-names = "default"; pinctrl-0 = <&wifi_enable_h>; reset-gpios = <&gpio0 RK_PB2 GPIO_ACTIVE_LOW>; From b47158fb42959c417ff2662075c0d46fb783d5d1 Mon Sep 17 00:00:00 2001 From: Wayne Chang Date: Tue, 8 Apr 2025 11:09:05 +0800 Subject: [PATCH 019/555] phy: tegra: xusb: Use a bitmask for UTMI pad power state tracking The current implementation uses bias_pad_enable as a reference count to manage the shared bias pad for all UTMI PHYs. However, during system suspension with connected USB devices, multiple power-down requests for the UTMI pad result in a mismatch in the reference count, which in turn produces warnings such as: [ 237.762967] WARNING: CPU: 10 PID: 1618 at tegra186_utmi_pad_power_down+0x160/0x170 [ 237.763103] Call trace: [ 237.763104] tegra186_utmi_pad_power_down+0x160/0x170 [ 237.763107] tegra186_utmi_phy_power_off+0x10/0x30 [ 237.763110] phy_power_off+0x48/0x100 [ 237.763113] tegra_xusb_enter_elpg+0x204/0x500 [ 237.763119] tegra_xusb_suspend+0x48/0x140 [ 237.763122] platform_pm_suspend+0x2c/0xb0 [ 237.763125] dpm_run_callback.isra.0+0x20/0xa0 [ 237.763127] __device_suspend+0x118/0x330 [ 237.763129] dpm_suspend+0x10c/0x1f0 [ 237.763130] dpm_suspend_start+0x88/0xb0 [ 237.763132] suspend_devices_and_enter+0x120/0x500 [ 237.763135] pm_suspend+0x1ec/0x270 The root cause was traced back to the dynamic power-down changes introduced in commit a30951d31b25 ("xhci: tegra: USB2 pad power controls"), where the UTMI pad was being powered down without verifying its current state. This unbalanced behavior led to discrepancies in the reference count. To rectify this issue, this patch replaces the single reference counter with a bitmask, renamed to utmi_pad_enabled. Each bit in the mask corresponds to one of the four USB2 PHYs, allowing us to track each pad's enablement status individually. With this change: - The bias pad is powered on only when the mask is clear. - Each UTMI pad is powered on or down based on its corresponding bit in the mask, preventing redundant operations. - The overall power state of the shared bias pad is maintained correctly during suspend/resume cycles. The mutex used to prevent race conditions during UTMI pad enable/disable operations has been moved from the tegra186_utmi_bias_pad_power_on/off functions to the parent functions tegra186_utmi_pad_power_on/down. This change ensures that there are no race conditions when updating the bitmask. Cc: stable@vger.kernel.org Fixes: a30951d31b25 ("xhci: tegra: USB2 pad power controls") Signed-off-by: Wayne Chang Reviewed-by: Jon Hunter Tested-by: Jon Hunter Link: https://lore.kernel.org/r/20250408030905.990474-1-waynec@nvidia.com Signed-off-by: Vinod Koul --- drivers/phy/tegra/xusb-tegra186.c | 44 +++++++++++++++++++------------ 1 file changed, 27 insertions(+), 17 deletions(-) diff --git a/drivers/phy/tegra/xusb-tegra186.c b/drivers/phy/tegra/xusb-tegra186.c index fae6242aa730..cc7b8a6a999f 100644 --- a/drivers/phy/tegra/xusb-tegra186.c +++ b/drivers/phy/tegra/xusb-tegra186.c @@ -237,6 +237,8 @@ #define DATA0_VAL_PD BIT(1) #define USE_XUSB_AO BIT(4) +#define TEGRA_UTMI_PAD_MAX 4 + #define TEGRA186_LANE(_name, _offset, _shift, _mask, _type) \ { \ .name = _name, \ @@ -269,7 +271,7 @@ struct tegra186_xusb_padctl { /* UTMI bias and tracking */ struct clk *usb2_trk_clk; - unsigned int bias_pad_enable; + DECLARE_BITMAP(utmi_pad_enabled, TEGRA_UTMI_PAD_MAX); /* padctl context */ struct tegra186_xusb_padctl_context context; @@ -603,12 +605,8 @@ static void tegra186_utmi_bias_pad_power_on(struct tegra_xusb_padctl *padctl) u32 value; int err; - mutex_lock(&padctl->lock); - - if (priv->bias_pad_enable++ > 0) { - mutex_unlock(&padctl->lock); + if (!bitmap_empty(priv->utmi_pad_enabled, TEGRA_UTMI_PAD_MAX)) return; - } err = clk_prepare_enable(priv->usb2_trk_clk); if (err < 0) @@ -667,17 +665,8 @@ static void tegra186_utmi_bias_pad_power_off(struct tegra_xusb_padctl *padctl) struct tegra186_xusb_padctl *priv = to_tegra186_xusb_padctl(padctl); u32 value; - mutex_lock(&padctl->lock); - - if (WARN_ON(priv->bias_pad_enable == 0)) { - mutex_unlock(&padctl->lock); + if (!bitmap_empty(priv->utmi_pad_enabled, TEGRA_UTMI_PAD_MAX)) return; - } - - if (--priv->bias_pad_enable > 0) { - mutex_unlock(&padctl->lock); - return; - } value = padctl_readl(padctl, XUSB_PADCTL_USB2_BIAS_PAD_CTL1); value |= USB2_PD_TRK; @@ -690,13 +679,13 @@ static void tegra186_utmi_bias_pad_power_off(struct tegra_xusb_padctl *padctl) clk_disable_unprepare(priv->usb2_trk_clk); } - mutex_unlock(&padctl->lock); } static void tegra186_utmi_pad_power_on(struct phy *phy) { struct tegra_xusb_lane *lane = phy_get_drvdata(phy); struct tegra_xusb_padctl *padctl = lane->pad->padctl; + struct tegra186_xusb_padctl *priv = to_tegra186_xusb_padctl(padctl); struct tegra_xusb_usb2_port *port; struct device *dev = padctl->dev; unsigned int index = lane->index; @@ -705,9 +694,16 @@ static void tegra186_utmi_pad_power_on(struct phy *phy) if (!phy) return; + mutex_lock(&padctl->lock); + if (test_bit(index, priv->utmi_pad_enabled)) { + mutex_unlock(&padctl->lock); + return; + } + port = tegra_xusb_find_usb2_port(padctl, index); if (!port) { dev_err(dev, "no port found for USB2 lane %u\n", index); + mutex_unlock(&padctl->lock); return; } @@ -724,18 +720,28 @@ static void tegra186_utmi_pad_power_on(struct phy *phy) value = padctl_readl(padctl, XUSB_PADCTL_USB2_OTG_PADX_CTL1(index)); value &= ~USB2_OTG_PD_DR; padctl_writel(padctl, value, XUSB_PADCTL_USB2_OTG_PADX_CTL1(index)); + + set_bit(index, priv->utmi_pad_enabled); + mutex_unlock(&padctl->lock); } static void tegra186_utmi_pad_power_down(struct phy *phy) { struct tegra_xusb_lane *lane = phy_get_drvdata(phy); struct tegra_xusb_padctl *padctl = lane->pad->padctl; + struct tegra186_xusb_padctl *priv = to_tegra186_xusb_padctl(padctl); unsigned int index = lane->index; u32 value; if (!phy) return; + mutex_lock(&padctl->lock); + if (!test_bit(index, priv->utmi_pad_enabled)) { + mutex_unlock(&padctl->lock); + return; + } + dev_dbg(padctl->dev, "power down UTMI pad %u\n", index); value = padctl_readl(padctl, XUSB_PADCTL_USB2_OTG_PADX_CTL0(index)); @@ -748,7 +754,11 @@ static void tegra186_utmi_pad_power_down(struct phy *phy) udelay(2); + clear_bit(index, priv->utmi_pad_enabled); + tegra186_utmi_bias_pad_power_off(padctl); + + mutex_unlock(&padctl->lock); } static int tegra186_xusb_padctl_vbus_override(struct tegra_xusb_padctl *padctl, From d784552e76a23c4ffad0e383670cd1d86064a6be Mon Sep 17 00:00:00 2001 From: Nitin Rawat Date: Mon, 7 Apr 2025 17:40:08 +0530 Subject: [PATCH 020/555] phy: qcom-qmp-ufs: check for mode type for phy setting Generally all target supports Rate B but for very few like SM8550, two sets of UFS PHY settings are provided, one set is to support HS-G5 Rate A and another set is to support HS-G4 and lower gears with Rate B. Commit b02cc9a17679("phy: qcom-qmp-ufs: Add PHY Configuration support for sm8750") apply Rate B setting for SM8550 gear 5 without checking for mode value (Rate A or Rate B) from Controller driver which caused issue as SM8550 support rate A for Gear 5. Fix this by adding mode check before applying Rat B phy setting. Fixes: b02cc9a17679 ("phy: qcom-qmp-ufs: Add PHY Configuration support for sm8750") Reported-by: Neil Armstrong Closes: https://lore.kernel.org/all/430ed11c-0490-45be-897b-27cad9682371@quicinc.com/ Tested-by: Neil Armstrong # on SM8550-QRD Signed-off-by: Nitin Rawat Link: https://lore.kernel.org/r/20250407121008.22230-1-quic_nitirawa@quicinc.com Signed-off-by: Vinod Koul --- drivers/phy/qualcomm/phy-qcom-qmp-ufs.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/phy/qualcomm/phy-qcom-qmp-ufs.c b/drivers/phy/qualcomm/phy-qcom-qmp-ufs.c index 45b3b792696e..b33e2e2b5014 100644 --- a/drivers/phy/qualcomm/phy-qcom-qmp-ufs.c +++ b/drivers/phy/qualcomm/phy-qcom-qmp-ufs.c @@ -1754,7 +1754,8 @@ static void qmp_ufs_init_registers(struct qmp_ufs *qmp, const struct qmp_phy_cfg qmp_ufs_init_all(qmp, &cfg->tbls_hs_overlay[i]); } - qmp_ufs_init_all(qmp, &cfg->tbls_hs_b); + if (qmp->mode == PHY_MODE_UFS_HS_B) + qmp_ufs_init_all(qmp, &cfg->tbls_hs_b); } static int qmp_ufs_com_init(struct qmp_ufs *qmp) From e153fdea9db04dd0e2e536e2eb125b16bbbc2af7 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Thu, 20 Mar 2025 16:15:42 +0100 Subject: [PATCH 021/555] phy: can-transceiver: Re-instate "mux-states" property presence check On the Renesas Gray Hawk Single development board: can-transceiver-phy can-phy0: /can-phy0: failed to get mux-state (0) "mux-states" is an optional property for CAN transceivers. However, mux_get() always prints an error message in case of an error, including when the property is not present, confusing the user. Fix this by re-instating the property presence check (this time using the proper API) in a wrapper around devm_mux_state_get(). When the multiplexer subsystem gains support for optional muxes, the wrapper can just be removed. In addition, propagate all real errors upstream, instead of ignoring them. Fixes: d02dfd4ceb2e9f34 ("phy: can-transceiver: Drop unnecessary "mux-states" property presence check") Signed-off-by: Geert Uytterhoeven Tested-by: Biju Das Reviewed-by: Biju Das Reviewed-by: Vincent Mailhol Link: https://lore.kernel.org/r/3d7e0d723908284e8cf06ad1f7950c03173178f3.1742483710.git.geert+renesas@glider.be Signed-off-by: Vinod Koul --- drivers/phy/phy-can-transceiver.c | 22 +++++++++++++++------- 1 file changed, 15 insertions(+), 7 deletions(-) diff --git a/drivers/phy/phy-can-transceiver.c b/drivers/phy/phy-can-transceiver.c index 2bec70615449..f59caff4b3d4 100644 --- a/drivers/phy/phy-can-transceiver.c +++ b/drivers/phy/phy-can-transceiver.c @@ -93,6 +93,16 @@ static const struct of_device_id can_transceiver_phy_ids[] = { }; MODULE_DEVICE_TABLE(of, can_transceiver_phy_ids); +/* Temporary wrapper until the multiplexer subsystem supports optional muxes */ +static inline struct mux_state * +devm_mux_state_get_optional(struct device *dev, const char *mux_name) +{ + if (!of_property_present(dev->of_node, "mux-states")) + return NULL; + + return devm_mux_state_get(dev, mux_name); +} + static int can_transceiver_phy_probe(struct platform_device *pdev) { struct phy_provider *phy_provider; @@ -114,13 +124,11 @@ static int can_transceiver_phy_probe(struct platform_device *pdev) match = of_match_node(can_transceiver_phy_ids, pdev->dev.of_node); drvdata = match->data; - mux_state = devm_mux_state_get(dev, NULL); - if (IS_ERR(mux_state)) { - if (PTR_ERR(mux_state) == -EPROBE_DEFER) - return PTR_ERR(mux_state); - } else { - can_transceiver_phy->mux_state = mux_state; - } + mux_state = devm_mux_state_get_optional(dev, NULL); + if (IS_ERR(mux_state)) + return PTR_ERR(mux_state); + + can_transceiver_phy->mux_state = mux_state; phy = devm_phy_create(dev, dev->of_node, &can_transceiver_phy_ops); From 9cf118aafd6682793c40dde31b5f24d271da3996 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Fri, 21 Mar 2025 17:36:14 +0300 Subject: [PATCH 022/555] phy: rockchip-samsung-dcphy: Add missing assignment The "ret = " was accidentally dropped so the error handling doesn't work. Fixes: b2a1a2ae7818 ("phy: rockchip: Add Samsung MIPI D-/C-PHY driver") Signed-off-by: Dan Carpenter Reviewed-by: Heiko Stuebner Link: https://lore.kernel.org/r/e64265a4-9543-4728-a49f-ea910fccef7c@stanley.mountain Signed-off-by: Vinod Koul --- drivers/phy/rockchip/phy-rockchip-samsung-dcphy.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/phy/rockchip/phy-rockchip-samsung-dcphy.c b/drivers/phy/rockchip/phy-rockchip-samsung-dcphy.c index 08c78c1bafc9..28a052e17366 100644 --- a/drivers/phy/rockchip/phy-rockchip-samsung-dcphy.c +++ b/drivers/phy/rockchip/phy-rockchip-samsung-dcphy.c @@ -1653,7 +1653,7 @@ static __maybe_unused int samsung_mipi_dcphy_runtime_resume(struct device *dev) return ret; } - clk_prepare_enable(samsung->ref_clk); + ret = clk_prepare_enable(samsung->ref_clk); if (ret) { dev_err(samsung->dev, "Failed to enable reference clock, %d\n", ret); clk_disable_unprepare(samsung->pclk); From 0cd34d98dfd4f2b596415b8f12faf7b946613458 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Sun, 6 Apr 2025 22:01:42 +0200 Subject: [PATCH 023/555] iio: accel: fxls8962af: Fix wakeup source leaks on device unbind Device can be unbound, so driver must also release memory for the wakeup source. Signed-off-by: Krzysztof Kozlowski Link: https://patch.msgid.link/20250406-b4-device-wakeup-leak-iio-v1-1-2d7d322a4a93@linaro.org Signed-off-by: Jonathan Cameron --- drivers/iio/accel/fxls8962af-core.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/iio/accel/fxls8962af-core.c b/drivers/iio/accel/fxls8962af-core.c index 48e4282964a0..bf1d3923a181 100644 --- a/drivers/iio/accel/fxls8962af-core.c +++ b/drivers/iio/accel/fxls8962af-core.c @@ -1226,8 +1226,11 @@ int fxls8962af_core_probe(struct device *dev, struct regmap *regmap, int irq) if (ret) return ret; - if (device_property_read_bool(dev, "wakeup-source")) - device_init_wakeup(dev, true); + if (device_property_read_bool(dev, "wakeup-source")) { + ret = devm_device_init_wakeup(dev); + if (ret) + return dev_err_probe(dev, ret, "Failed to init wakeup\n"); + } return devm_iio_device_register(dev, indio_dev); } From ad3764b45c1524872b621d5667a56f6a574501bd Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Sun, 6 Apr 2025 22:01:43 +0200 Subject: [PATCH 024/555] iio: adc: qcom-spmi-iadc: Fix wakeup source leaks on device unbind Device can be unbound, so driver must also release memory for the wakeup source. Signed-off-by: Krzysztof Kozlowski Link: https://patch.msgid.link/20250406-b4-device-wakeup-leak-iio-v1-2-2d7d322a4a93@linaro.org Signed-off-by: Jonathan Cameron --- drivers/iio/adc/qcom-spmi-iadc.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/iio/adc/qcom-spmi-iadc.c b/drivers/iio/adc/qcom-spmi-iadc.c index 7fb8b2499a1d..b64a8a407168 100644 --- a/drivers/iio/adc/qcom-spmi-iadc.c +++ b/drivers/iio/adc/qcom-spmi-iadc.c @@ -543,7 +543,9 @@ static int iadc_probe(struct platform_device *pdev) else return ret; } else { - device_init_wakeup(iadc->dev, 1); + ret = devm_device_init_wakeup(iadc->dev); + if (ret) + return dev_err_probe(iadc->dev, ret, "Failed to init wakeup\n"); } ret = iadc_update_offset(iadc); From 4551383e78d59b34eea3f4ed28ad22df99e25d59 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Sun, 6 Apr 2025 22:01:44 +0200 Subject: [PATCH 025/555] iio: imu: st_lsm6dsx: Fix wakeup source leaks on device unbind Device can be unbound, so driver must also release memory for the wakeup source. Signed-off-by: Krzysztof Kozlowski Acked-by: Lorenzo Bianconi Link: https://patch.msgid.link/20250406-b4-device-wakeup-leak-iio-v1-3-2d7d322a4a93@linaro.org Signed-off-by: Jonathan Cameron --- drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c b/drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c index 4fdcc2acc94e..96c6106b95ee 100644 --- a/drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c +++ b/drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c @@ -2719,8 +2719,11 @@ int st_lsm6dsx_probe(struct device *dev, int irq, int hw_id, } if (device_property_read_bool(dev, "wakeup-source") || - (pdata && pdata->wakeup_source)) - device_init_wakeup(dev, true); + (pdata && pdata->wakeup_source)) { + err = devm_device_init_wakeup(dev); + if (err) + return dev_err_probe(dev, err, "Failed to init wakeup\n"); + } return 0; } From 8772cc49e0b8ab782e475ce5ef659eedab601a09 Mon Sep 17 00:00:00 2001 From: Matthias Schiffer Date: Wed, 16 Apr 2025 20:37:56 +0200 Subject: [PATCH 026/555] batman-adv: fix duplicate MAC address check batadv_check_known_mac_addr() is both too lenient and too strict: - It is called from batadv_hardif_add_interface(), which means that it checked interfaces that are not used for batman-adv at all. Move it to batadv_hardif_enable_interface(). Also, restrict it to hardifs of the same mesh interface; different mesh interfaces should not interact at all. The batadv_check_known_mac_addr() argument is changed from `struct net_device` to `struct batadv_hard_iface` to achieve this. - The check only cares about hardifs in BATADV_IF_ACTIVE and BATADV_IF_TO_BE_ACTIVATED states, but interfaces in BATADV_IF_INACTIVE state should be checked as well, or the following steps will not result in a warning then they should: - Add two interfaces in down state with different MAC addresses to a mesh as hardifs - Change the MAC addresses so they conflict - Set interfaces to up state Now there will be two active hardifs with the same MAC address, but no warning. Fix by only ignoring hardifs in BATADV_IF_NOT_IN_USE state. The RCU lock can be dropped, as we're holding RTNL anyways when the function is called. Fixes: c6c8fea29769 ("net: Add batman-adv meshing protocol") Signed-off-by: Matthias Schiffer Signed-off-by: Simon Wunderlich --- net/batman-adv/hard-interface.c | 31 ++++++++++++++++++------------- 1 file changed, 18 insertions(+), 13 deletions(-) diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c index f145f9662653..d099434d3dfa 100644 --- a/net/batman-adv/hard-interface.c +++ b/net/batman-adv/hard-interface.c @@ -506,28 +506,32 @@ batadv_hardif_is_iface_up(const struct batadv_hard_iface *hard_iface) return false; } -static void batadv_check_known_mac_addr(const struct net_device *net_dev) +static void batadv_check_known_mac_addr(const struct batadv_hard_iface *hard_iface) { - const struct batadv_hard_iface *hard_iface; + const struct net_device *mesh_iface = hard_iface->mesh_iface; + const struct batadv_hard_iface *tmp_hard_iface; - rcu_read_lock(); - list_for_each_entry_rcu(hard_iface, &batadv_hardif_list, list) { - if (hard_iface->if_status != BATADV_IF_ACTIVE && - hard_iface->if_status != BATADV_IF_TO_BE_ACTIVATED) + if (!mesh_iface) + return; + + list_for_each_entry(tmp_hard_iface, &batadv_hardif_list, list) { + if (tmp_hard_iface == hard_iface) continue; - if (hard_iface->net_dev == net_dev) + if (tmp_hard_iface->mesh_iface != mesh_iface) continue; - if (!batadv_compare_eth(hard_iface->net_dev->dev_addr, - net_dev->dev_addr)) + if (tmp_hard_iface->if_status == BATADV_IF_NOT_IN_USE) + continue; + + if (!batadv_compare_eth(tmp_hard_iface->net_dev->dev_addr, + hard_iface->net_dev->dev_addr)) continue; pr_warn("The newly added mac address (%pM) already exists on: %s\n", - net_dev->dev_addr, hard_iface->net_dev->name); + hard_iface->net_dev->dev_addr, tmp_hard_iface->net_dev->name); pr_warn("It is strongly recommended to keep mac addresses unique to avoid problems!\n"); } - rcu_read_unlock(); } /** @@ -764,6 +768,8 @@ int batadv_hardif_enable_interface(struct batadv_hard_iface *hard_iface, hard_iface->net_dev->name, hardif_mtu, required_mtu); + batadv_check_known_mac_addr(hard_iface); + if (batadv_hardif_is_iface_up(hard_iface)) batadv_hardif_activate_interface(hard_iface); else @@ -902,7 +908,6 @@ batadv_hardif_add_interface(struct net_device *net_dev) batadv_v_hardif_init(hard_iface); - batadv_check_known_mac_addr(hard_iface->net_dev); kref_get(&hard_iface->refcount); list_add_tail_rcu(&hard_iface->list, &batadv_hardif_list); batadv_hardif_generation++; @@ -989,7 +994,7 @@ static int batadv_hard_if_event(struct notifier_block *this, if (hard_iface->if_status == BATADV_IF_NOT_IN_USE) goto hardif_put; - batadv_check_known_mac_addr(hard_iface->net_dev); + batadv_check_known_mac_addr(hard_iface); bat_priv = netdev_priv(hard_iface->mesh_iface); bat_priv->algo_ops->iface.update_mac(hard_iface); From 97994333de2b8062d2df4e6ce0dc65c2dc0f40dc Mon Sep 17 00:00:00 2001 From: Purva Yeshi Date: Thu, 10 Apr 2025 16:32:16 +0530 Subject: [PATCH 027/555] dmaengine: idxd: cdev: Fix uninitialized use of sva in idxd_cdev_open Fix Smatch-detected issue: drivers/dma/idxd/cdev.c:321 idxd_cdev_open() error: uninitialized symbol 'sva'. 'sva' pointer may be used uninitialized in error handling paths. Specifically, if PASID support is enabled and iommu_sva_bind_device() returns an error, the code jumps to the cleanup label and attempts to call iommu_sva_unbind_device(sva) without ensuring that sva was successfully assigned. This triggers a Smatch warning about an uninitialized symbol. Initialize sva to NULL at declaration and add a check using IS_ERR_OR_NULL() before unbinding the device. This ensures the function does not use an invalid or uninitialized pointer during cleanup. Signed-off-by: Purva Yeshi Reviewed-by: Dave Jiang Acked-by: Vinicius Costa Gomes Link: https://lore.kernel.org/r/20250410110216.21592-1-purvayeshi550@gmail.com Signed-off-by: Vinod Koul --- drivers/dma/idxd/cdev.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/dma/idxd/cdev.c b/drivers/dma/idxd/cdev.c index ff94ee892339..7bd031a60894 100644 --- a/drivers/dma/idxd/cdev.c +++ b/drivers/dma/idxd/cdev.c @@ -222,7 +222,7 @@ static int idxd_cdev_open(struct inode *inode, struct file *filp) struct idxd_wq *wq; struct device *dev, *fdev; int rc = 0; - struct iommu_sva *sva; + struct iommu_sva *sva = NULL; unsigned int pasid; struct idxd_cdev *idxd_cdev; @@ -317,7 +317,7 @@ failed_set_pasid: if (device_user_pasid_enabled(idxd)) idxd_xa_pasid_remove(ctx); failed_get_pasid: - if (device_user_pasid_enabled(idxd)) + if (device_user_pasid_enabled(idxd) && !IS_ERR_OR_NULL(sva)) iommu_sva_unbind_device(sva); failed: mutex_unlock(&wq->wq_lock); From df180e65305f8c1e020d54bfc2132349fd693de1 Mon Sep 17 00:00:00 2001 From: Nathan Lynch Date: Thu, 3 Apr 2025 11:24:19 -0500 Subject: [PATCH 028/555] dmaengine: Revert "dmaengine: dmatest: Fix dmatest waiting less when interrupted" Several issues with this change: * The analysis is flawed and it's unclear what problem is being fixed. There is no difference between wait_event_freezable_timeout() and wait_event_timeout() with respect to device interrupts. And of course "the interrupt notifying the finish of an operation happens during wait_event_freezable_timeout()" -- that's how it's supposed to work. * The link at the "Closes:" tag appears to be an unrelated use-after-free in idxd. * It introduces a regression: dmatest threads are meant to be freezable and this change breaks that. See discussion here: https://lore.kernel.org/dmaengine/878qpa13fe.fsf@AUSNATLYNCH.amd.com/ Fixes: e87ca16e9911 ("dmaengine: dmatest: Fix dmatest waiting less when interrupted") Signed-off-by: Nathan Lynch Link: https://lore.kernel.org/r/20250403-dmaengine-dmatest-revert-waiting-less-v1-1-8227c5a3d7c8@amd.com Signed-off-by: Vinod Koul --- drivers/dma/dmatest.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/dma/dmatest.c b/drivers/dma/dmatest.c index d891dfca358e..91b2fbc0b864 100644 --- a/drivers/dma/dmatest.c +++ b/drivers/dma/dmatest.c @@ -841,9 +841,9 @@ static int dmatest_func(void *data) } else { dma_async_issue_pending(chan); - wait_event_timeout(thread->done_wait, - done->done, - msecs_to_jiffies(params->timeout)); + wait_event_freezable_timeout(thread->done_wait, + done->done, + msecs_to_jiffies(params->timeout)); status = dma_async_is_tx_complete(chan, cookie, NULL, NULL); From 8ca9590c39b69b55a8de63d2b21b0d44f523b43a Mon Sep 17 00:00:00 2001 From: Yemike Abhilash Chandra Date: Thu, 17 Apr 2025 13:25:21 +0530 Subject: [PATCH 029/555] dmaengine: ti: k3-udma: Use cap_mask directly from dma_device structure instead of a local copy Currently, a local dma_cap_mask_t variable is used to store device cap_mask within udma_of_xlate(). However, the DMA_PRIVATE flag in the device cap_mask can get cleared when the last channel is released. This can happen right after storing the cap_mask locally in udma_of_xlate(), and subsequent dma_request_channel() can fail due to mismatch in the cap_mask. Fix this by removing the local dma_cap_mask_t variable and directly using the one from the dma_device structure. Fixes: 25dcb5dd7b7c ("dmaengine: ti: New driver for K3 UDMA") Cc: stable@vger.kernel.org Signed-off-by: Vaishnav Achath Acked-by: Peter Ujfalusi Reviewed-by: Udit Kumar Signed-off-by: Yemike Abhilash Chandra Link: https://lore.kernel.org/r/20250417075521.623651-1-y-abhilashchandra@ti.com Signed-off-by: Vinod Koul --- drivers/dma/ti/k3-udma.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/dma/ti/k3-udma.c b/drivers/dma/ti/k3-udma.c index b223a7aacb0c..08ed8cd7f1dd 100644 --- a/drivers/dma/ti/k3-udma.c +++ b/drivers/dma/ti/k3-udma.c @@ -4246,7 +4246,6 @@ static struct dma_chan *udma_of_xlate(struct of_phandle_args *dma_spec, struct of_dma *ofdma) { struct udma_dev *ud = ofdma->of_dma_data; - dma_cap_mask_t mask = ud->ddev.cap_mask; struct udma_filter_param filter_param; struct dma_chan *chan; @@ -4278,7 +4277,7 @@ static struct dma_chan *udma_of_xlate(struct of_phandle_args *dma_spec, } } - chan = __dma_request_channel(&mask, udma_dma_filter_fn, &filter_param, + chan = __dma_request_channel(&ud->ddev.cap_mask, udma_dma_filter_fn, &filter_param, ofdma->of_node); if (!chan) { dev_err(ud->dev, "get channel fail in %s.\n", __func__); From fca280992af8c2fbd511bc43f65abb4a17363f2f Mon Sep 17 00:00:00 2001 From: Ronald Wahl Date: Mon, 14 Apr 2025 19:31:13 +0200 Subject: [PATCH 030/555] dmaengine: ti: k3-udma: Add missing locking Recent kernels complain about a missing lock in k3-udma.c when the lock validator is enabled: [ 4.128073] WARNING: CPU: 0 PID: 746 at drivers/dma/ti/../virt-dma.h:169 udma_start.isra.0+0x34/0x238 [ 4.137352] CPU: 0 UID: 0 PID: 746 Comm: kworker/0:3 Not tainted 6.12.9-arm64 #28 [ 4.144867] Hardware name: pp-v12 (DT) [ 4.148648] Workqueue: events udma_check_tx_completion [ 4.153841] pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--) [ 4.160834] pc : udma_start.isra.0+0x34/0x238 [ 4.165227] lr : udma_start.isra.0+0x30/0x238 [ 4.169618] sp : ffffffc083cabcf0 [ 4.172963] x29: ffffffc083cabcf0 x28: 0000000000000000 x27: ffffff800001b005 [ 4.180167] x26: ffffffc0812f0000 x25: 0000000000000000 x24: 0000000000000000 [ 4.187370] x23: 0000000000000001 x22: 00000000e21eabe9 x21: ffffff8000fa0670 [ 4.194571] x20: ffffff8001b6bf00 x19: ffffff8000fa0430 x18: ffffffc083b95030 [ 4.201773] x17: 0000000000000000 x16: 00000000f0000000 x15: 0000000000000048 [ 4.208976] x14: 0000000000000048 x13: 0000000000000000 x12: 0000000000000001 [ 4.216179] x11: ffffffc08151a240 x10: 0000000000003ea1 x9 : ffffffc08046ab68 [ 4.223381] x8 : ffffffc083cabac0 x7 : ffffffc081df3718 x6 : 0000000000029fc8 [ 4.230583] x5 : ffffffc0817ee6d8 x4 : 0000000000000bc0 x3 : 0000000000000000 [ 4.237784] x2 : 0000000000000000 x1 : 00000000001fffff x0 : 0000000000000000 [ 4.244986] Call trace: [ 4.247463] udma_start.isra.0+0x34/0x238 [ 4.251509] udma_check_tx_completion+0xd0/0xdc [ 4.256076] process_one_work+0x244/0x3fc [ 4.260129] process_scheduled_works+0x6c/0x74 [ 4.264610] worker_thread+0x150/0x1dc [ 4.268398] kthread+0xd8/0xe8 [ 4.271492] ret_from_fork+0x10/0x20 [ 4.275107] irq event stamp: 220 [ 4.278363] hardirqs last enabled at (219): [] _raw_spin_unlock_irq+0x38/0x50 [ 4.287183] hardirqs last disabled at (220): [] el1_dbg+0x24/0x50 [ 4.294879] softirqs last enabled at (182): [] handle_softirqs+0x1c0/0x3cc [ 4.303437] softirqs last disabled at (177): [] __do_softirq+0x1c/0x28 [ 4.311559] ---[ end trace 0000000000000000 ]--- This commit adds the missing locking. Fixes: 25dcb5dd7b7c ("dmaengine: ti: New driver for K3 UDMA") Cc: Peter Ujfalusi Cc: Vignesh Raghavendra Cc: Vinod Koul Cc: dmaengine@vger.kernel.org Cc: stable@vger.kernel.org Signed-off-by: Ronald Wahl Acked-by: Peter Ujfalusi Link: https://lore.kernel.org/r/20250414173113.80677-1-rwahl@gmx.de Signed-off-by: Vinod Koul --- drivers/dma/ti/k3-udma.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/dma/ti/k3-udma.c b/drivers/dma/ti/k3-udma.c index 08ed8cd7f1dd..b6255c0601bb 100644 --- a/drivers/dma/ti/k3-udma.c +++ b/drivers/dma/ti/k3-udma.c @@ -1091,8 +1091,11 @@ static void udma_check_tx_completion(struct work_struct *work) u32 residue_diff; ktime_t time_diff; unsigned long delay; + unsigned long flags; while (1) { + spin_lock_irqsave(&uc->vc.lock, flags); + if (uc->desc) { /* Get previous residue and time stamp */ residue_diff = uc->tx_drain.residue; @@ -1127,6 +1130,8 @@ static void udma_check_tx_completion(struct work_struct *work) break; } + spin_unlock_irqrestore(&uc->vc.lock, flags); + usleep_range(ktime_to_us(delay), ktime_to_us(delay) + 10); continue; @@ -1143,6 +1148,8 @@ static void udma_check_tx_completion(struct work_struct *work) break; } + + spin_unlock_irqrestore(&uc->vc.lock, flags); } static irqreturn_t udma_ring_irq_handler(int irq, void *data) From 1d2d8524eaffc4d9a116213520d2c650e07c9cc6 Mon Sep 17 00:00:00 2001 From: David Lechner Date: Thu, 17 Apr 2025 11:52:39 -0500 Subject: [PATCH 031/555] iio: imu: inv_mpu6050: align buffer for timestamp Align the buffer used with iio_push_to_buffers_with_timestamp() to ensure the s64 timestamp is aligned to 8 bytes. Fixes: 0829edc43e0a ("iio: imu: inv_mpu6050: read the full fifo when processing data") Signed-off-by: David Lechner Link: https://patch.msgid.link/20250417-iio-more-timestamp-alignment-v1-7-eafac1e22318@baylibre.com Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/imu/inv_mpu6050/inv_mpu_ring.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iio/imu/inv_mpu6050/inv_mpu_ring.c b/drivers/iio/imu/inv_mpu6050/inv_mpu_ring.c index 3d3b27f28c9d..273196e647a2 100644 --- a/drivers/iio/imu/inv_mpu6050/inv_mpu_ring.c +++ b/drivers/iio/imu/inv_mpu6050/inv_mpu_ring.c @@ -50,7 +50,7 @@ irqreturn_t inv_mpu6050_read_fifo(int irq, void *p) u16 fifo_count; u32 fifo_period; s64 timestamp; - u8 data[INV_MPU6050_OUTPUT_DATA_SIZE]; + u8 data[INV_MPU6050_OUTPUT_DATA_SIZE] __aligned(8); size_t i, nb; mutex_lock(&st->lock); From bb49d940344bcb8e2b19e69d7ac86f567887ea9a Mon Sep 17 00:00:00 2001 From: David Lechner Date: Thu, 17 Apr 2025 11:52:37 -0500 Subject: [PATCH 032/555] iio: chemical: sps30: use aligned_s64 for timestamp MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Follow the pattern of other drivers and use aligned_s64 for the timestamp. This will ensure that the timestamp is correctly aligned on all architectures. Fixes: a5bf6fdd19c3 ("iio:chemical:sps30: Fix timestamp alignment") Signed-off-by: David Lechner Reviewed-by: Nuno Sá Link: https://patch.msgid.link/20250417-iio-more-timestamp-alignment-v1-5-eafac1e22318@baylibre.com Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/chemical/sps30.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iio/chemical/sps30.c b/drivers/iio/chemical/sps30.c index 6f4f2ba2c09d..a7888146188d 100644 --- a/drivers/iio/chemical/sps30.c +++ b/drivers/iio/chemical/sps30.c @@ -108,7 +108,7 @@ static irqreturn_t sps30_trigger_handler(int irq, void *p) int ret; struct { s32 data[4]; /* PM1, PM2P5, PM4, PM10 */ - s64 ts; + aligned_s64 ts; } scan; mutex_lock(&state->lock); From 6ffa698674053e82e811520642db2650d00d2c01 Mon Sep 17 00:00:00 2001 From: David Lechner Date: Thu, 17 Apr 2025 11:52:36 -0500 Subject: [PATCH 033/555] iio: chemical: pms7003: use aligned_s64 for timestamp MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Follow the pattern of other drivers and use aligned_s64 for the timestamp. This will ensure that the timestamp is correctly aligned on all architectures. Also move the unaligned.h header while touching this since it was the only one not in alphabetical order. Fixes: 13e945631c2f ("iio:chemical:pms7003: Fix timestamp alignment and prevent data leak.") Signed-off-by: David Lechner Reviewed-by: Nuno Sá Link: https://patch.msgid.link/20250417-iio-more-timestamp-alignment-v1-4-eafac1e22318@baylibre.com Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/chemical/pms7003.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/iio/chemical/pms7003.c b/drivers/iio/chemical/pms7003.c index d0bd94912e0a..e05ce1f12065 100644 --- a/drivers/iio/chemical/pms7003.c +++ b/drivers/iio/chemical/pms7003.c @@ -5,7 +5,6 @@ * Copyright (c) Tomasz Duszynski */ -#include #include #include #include @@ -19,6 +18,8 @@ #include #include #include +#include +#include #define PMS7003_DRIVER_NAME "pms7003" @@ -76,7 +77,7 @@ struct pms7003_state { /* Used to construct scan to push to the IIO buffer */ struct { u16 data[3]; /* PM1, PM2P5, PM10 */ - s64 ts; + aligned_s64 ts; } scan; }; From f79aeb6c631b57395f37acbfbe59727e355a714c Mon Sep 17 00:00:00 2001 From: Jonathan Cameron Date: Sun, 13 Apr 2025 11:34:36 +0100 Subject: [PATCH 034/555] iio: temp: maxim-thermocouple: Fix potential lack of DMA safe buffer. The trick of using __aligned(IIO_DMA_MINALIGN) ensures that there is no overlap between buffers used for DMA and those used for driver state storage that are before the marking. It doesn't ensure anything above state variables found after the marking. Hence move this particular bit of state earlier in the structure. Fixes: 10897f34309b ("iio: temp: maxim_thermocouple: Fix alignment for DMA safety") Reviewed-by: David Lechner Link: https://patch.msgid.link/20250413103443.2420727-14-jic23@kernel.org Signed-off-by: Jonathan Cameron --- drivers/iio/temperature/maxim_thermocouple.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iio/temperature/maxim_thermocouple.c b/drivers/iio/temperature/maxim_thermocouple.c index c28a7a6dea5f..555a61e2f3fd 100644 --- a/drivers/iio/temperature/maxim_thermocouple.c +++ b/drivers/iio/temperature/maxim_thermocouple.c @@ -121,9 +121,9 @@ static const struct maxim_thermocouple_chip maxim_thermocouple_chips[] = { struct maxim_thermocouple_data { struct spi_device *spi; const struct maxim_thermocouple_chip *chip; + char tc_type; u8 buffer[16] __aligned(IIO_DMA_MINALIGN); - char tc_type; }; static int maxim_thermocouple_read(struct maxim_thermocouple_data *data, From 1bb942287e05dc4c304a003ea85e6dd9a5e7db39 Mon Sep 17 00:00:00 2001 From: Jonathan Cameron Date: Sun, 13 Apr 2025 11:34:27 +0100 Subject: [PATCH 035/555] iio: accel: adxl355: Make timestamp 64-bit aligned using aligned_s64 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The IIO ABI requires 64-bit aligned timestamps. In this case insufficient padding would have been added on architectures where an s64 is only 32-bit aligned. Use aligned_s64 to enforce the correct alignment. Fixes: 327a0eaf19d5 ("iio: accel: adxl355: Add triggered buffer support") Reported-by: David Lechner Reviewed-by: Nuno Sá Reviewed-by: David Lechner Link: https://patch.msgid.link/20250413103443.2420727-5-jic23@kernel.org Signed-off-by: Jonathan Cameron --- drivers/iio/accel/adxl355_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iio/accel/adxl355_core.c b/drivers/iio/accel/adxl355_core.c index e8cd21fa77a6..cbac622ef821 100644 --- a/drivers/iio/accel/adxl355_core.c +++ b/drivers/iio/accel/adxl355_core.c @@ -231,7 +231,7 @@ struct adxl355_data { u8 transf_buf[3]; struct { u8 buf[14]; - s64 ts; + aligned_s64 ts; } buffer; } __aligned(IIO_DMA_MINALIGN); }; From 5097eaae98e53f9ab9d35801c70da819b92ca907 Mon Sep 17 00:00:00 2001 From: Jonathan Cameron Date: Sun, 13 Apr 2025 11:34:26 +0100 Subject: [PATCH 036/555] iio: adc: dln2: Use aligned_s64 for timestamp MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Here the lack of marking allows the overall structure to not be sufficiently aligned resulting in misplacement of the timestamp in iio_push_to_buffers_with_timestamp(). Use aligned_s64 to force the alignment on all architectures. Fixes: 7c0299e879dd ("iio: adc: Add support for DLN2 ADC") Reported-by: David Lechner Reviewed-by: Andy Shevchenko Reviewed-by: Nuno Sá Reviewed-by: David Lechner Link: https://patch.msgid.link/20250413103443.2420727-4-jic23@kernel.org Signed-off-by: Jonathan Cameron --- drivers/iio/adc/dln2-adc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iio/adc/dln2-adc.c b/drivers/iio/adc/dln2-adc.c index a1e48a756a7b..359e26e3f5bc 100644 --- a/drivers/iio/adc/dln2-adc.c +++ b/drivers/iio/adc/dln2-adc.c @@ -466,7 +466,7 @@ static irqreturn_t dln2_adc_trigger_h(int irq, void *p) struct iio_dev *indio_dev = pf->indio_dev; struct { __le16 values[DLN2_ADC_MAX_CHANNELS]; - int64_t timestamp_space; + aligned_s64 timestamp_space; } data; struct dln2_adc_get_all_vals dev_data; struct dln2_adc *dln2 = iio_priv(indio_dev); From ffbc26bc91c1f1eb3dcf5d8776e74cbae21ee13a Mon Sep 17 00:00:00 2001 From: Jonathan Cameron Date: Sun, 13 Apr 2025 11:34:25 +0100 Subject: [PATCH 037/555] iio: adc: ad7768-1: Fix insufficient alignment of timestamp. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit On architectures where an s64 is not 64-bit aligned, this may result insufficient alignment of the timestamp and the structure being too small. Use aligned_s64 to force the alignment. Fixes: a1caeebab07e ("iio: adc: ad7768-1: Fix too small buffer passed to iio_push_to_buffers_with_timestamp()") # aligned_s64 newer Reported-by: David Lechner Reviewed-by: Nuno Sá Reviewed-by: David Lechner Link: https://patch.msgid.link/20250413103443.2420727-3-jic23@kernel.org Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/adc/ad7768-1.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iio/adc/ad7768-1.c b/drivers/iio/adc/ad7768-1.c index 5a863005aca6..5e0be36af0c5 100644 --- a/drivers/iio/adc/ad7768-1.c +++ b/drivers/iio/adc/ad7768-1.c @@ -168,7 +168,7 @@ struct ad7768_state { union { struct { __be32 chan; - s64 timestamp; + aligned_s64 timestamp; } scan; __be32 d32; u8 d8[2]; From 52d349884738c346961e153f195f4c7fe186fcf4 Mon Sep 17 00:00:00 2001 From: Jonathan Cameron Date: Sun, 13 Apr 2025 11:34:24 +0100 Subject: [PATCH 038/555] iio: adc: ad7266: Fix potential timestamp alignment issue. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit On architectures where an s64 is only 32-bit aligned insufficient padding would be left between the earlier elements and the timestamp. Use aligned_s64 to enforce the correct placement and ensure the storage is large enough. Fixes: 54e018da3141 ("iio:ad7266: Mark transfer buffer as __be16") # aligned_s64 is much newer. Reported-by: David Lechner Reviewed-by: Nuno Sá Reviewed-by: David Lechner Link: https://patch.msgid.link/20250413103443.2420727-2-jic23@kernel.org Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/adc/ad7266.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iio/adc/ad7266.c b/drivers/iio/adc/ad7266.c index 18559757f908..7fef2727f89e 100644 --- a/drivers/iio/adc/ad7266.c +++ b/drivers/iio/adc/ad7266.c @@ -45,7 +45,7 @@ struct ad7266_state { */ struct { __be16 sample[2]; - s64 timestamp; + aligned_s64 timestamp; } data __aligned(IIO_DMA_MINALIGN); }; From 2e922956277187655ed9bedf7b5c28906e51708f Mon Sep 17 00:00:00 2001 From: Gabriel Shahrouzi Date: Mon, 14 Apr 2025 11:40:49 -0400 Subject: [PATCH 039/555] staging: iio: adc: ad7816: Correct conditional logic for store mode MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The mode setting logic in ad7816_store_mode was reversed due to incorrect handling of the strcmp return value. strcmp returns 0 on match, so the `if (strcmp(buf, "full"))` block executed when the input was not "full". This resulted in "full" setting the mode to AD7816_PD (power-down) and other inputs setting it to AD7816_FULL. Fix this by checking it against 0 to correctly check for "full" and "power-down", mapping them to AD7816_FULL and AD7816_PD respectively. Fixes: 7924425db04a ("staging: iio: adc: new driver for AD7816 devices") Cc: stable@vger.kernel.org Signed-off-by: Gabriel Shahrouzi Acked-by: Nuno Sá Link: https://lore.kernel.org/stable/20250414152920.467505-1-gshahrouzi%40gmail.com Link: https://patch.msgid.link/20250414154050.469482-1-gshahrouzi@gmail.com Signed-off-by: Jonathan Cameron --- drivers/staging/iio/adc/ad7816.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/staging/iio/adc/ad7816.c b/drivers/staging/iio/adc/ad7816.c index 6c14d7bcdd67..081b17f49863 100644 --- a/drivers/staging/iio/adc/ad7816.c +++ b/drivers/staging/iio/adc/ad7816.c @@ -136,7 +136,7 @@ static ssize_t ad7816_store_mode(struct device *dev, struct iio_dev *indio_dev = dev_to_iio_dev(dev); struct ad7816_chip_info *chip = iio_priv(indio_dev); - if (strcmp(buf, "full")) { + if (strcmp(buf, "full") == 0) { gpiod_set_value(chip->rdwr_pin, 1); chip->mode = AD7816_FULL; } else { From 936a25ef11f5d6c3e3e6736bb8b28e28dfb77918 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Wed, 16 Apr 2025 10:45:17 -0700 Subject: [PATCH 040/555] input/joystick: magellan: Mark __nonstring look-up table GCC 15's new -Wunterminated-string-initialization notices that the 16 character lookup table "nibbles" (which is not used as a C-String) needs to be marked as "nonstring": drivers/input/joystick/magellan.c: In function 'magellan_crunch_nibbles': drivers/input/joystick/magellan.c:51:44: warning: initializer-string for array of 'unsigned char' truncates NUL terminator but destination lacks 'nonstring' attribute (17 chars into 16 available) [-Wunterminated-string-initialization] 51 | static unsigned char nibbles[16] = "0AB3D56GH9:K Link: https://lore.kernel.org/r/20250416174513.work.662-kees@kernel.org Signed-off-by: Dmitry Torokhov --- drivers/input/joystick/magellan.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/input/joystick/magellan.c b/drivers/input/joystick/magellan.c index 2eaa25c9c68c..7622638e5bb8 100644 --- a/drivers/input/joystick/magellan.c +++ b/drivers/input/joystick/magellan.c @@ -48,7 +48,7 @@ struct magellan { static int magellan_crunch_nibbles(unsigned char *data, int count) { - static unsigned char nibbles[16] = "0AB3D56GH9:K Date: Fri, 18 Apr 2025 18:37:18 -0700 Subject: [PATCH 041/555] Input: sparcspkr - avoid unannotated fall-through Fix follow warnings with clang-21i (and reformat for clarity): drivers/input/misc/sparcspkr.c:78:3: warning: unannotated fall-through between switch labels [-Wimplicit-fallthrough] 78 | case SND_TONE: break; | ^ drivers/input/misc/sparcspkr.c:78:3: note: insert 'break;' to avoid fall-through 78 | case SND_TONE: break; | ^ | break; drivers/input/misc/sparcspkr.c:113:3: warning: unannotated fall-through between switch labels [-Wimplicit-fallthrough] 113 | case SND_TONE: break; | ^ drivers/input/misc/sparcspkr.c:113:3: note: insert 'break;' to avoid fall-through 113 | case SND_TONE: break; | ^ | break; 2 warnings generated. Signed-off-by: WangYuli Link: https://lore.kernel.org/r/6730E40353C76908+20250415052439.155051-1-wangyuli@uniontech.com Signed-off-by: Dmitry Torokhov --- drivers/input/misc/sparcspkr.c | 22 ++++++++++++++++------ 1 file changed, 16 insertions(+), 6 deletions(-) diff --git a/drivers/input/misc/sparcspkr.c b/drivers/input/misc/sparcspkr.c index 8d7303fc13bc..1cfadd73829f 100644 --- a/drivers/input/misc/sparcspkr.c +++ b/drivers/input/misc/sparcspkr.c @@ -74,9 +74,14 @@ static int bbc_spkr_event(struct input_dev *dev, unsigned int type, unsigned int return -1; switch (code) { - case SND_BELL: if (value) value = 1000; - case SND_TONE: break; - default: return -1; + case SND_BELL: + if (value) + value = 1000; + break; + case SND_TONE: + break; + default: + return -1; } if (value > 20 && value < 32767) @@ -109,9 +114,14 @@ static int grover_spkr_event(struct input_dev *dev, unsigned int type, unsigned return -1; switch (code) { - case SND_BELL: if (value) value = 1000; - case SND_TONE: break; - default: return -1; + case SND_BELL: + if (value) + value = 1000; + break; + case SND_TONE: + break; + default: + return -1; } if (value > 20 && value < 32767) From c6cb8bf79466ae66bd0d07338c7c505ce758e9d7 Mon Sep 17 00:00:00 2001 From: Hugo Villeneuve Date: Thu, 10 Apr 2025 14:46:32 -0400 Subject: [PATCH 042/555] Input: cyttsp5 - ensure minimum reset pulse width The current reset pulse width is measured to be 5us on a Renesas RZ/G2L SOM. The manufacturer's minimum reset pulse width is specified as 10us. Extend reset pulse width to make sure it is long enough on all platforms. Also reword confusing comments about reset pin assertion. Fixes: 5b0c03e24a06 ("Input: Add driver for Cypress Generation 5 touchscreen") Cc: stable@vger.kernel.org Acked-by: Alistair Francis Signed-off-by: Hugo Villeneuve Link: https://lore.kernel.org/r/20250410184633.1164837-1-hugo@hugovil.com Signed-off-by: Dmitry Torokhov --- drivers/input/touchscreen/cyttsp5.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/input/touchscreen/cyttsp5.c b/drivers/input/touchscreen/cyttsp5.c index eafe5a9b8964..14c43f0a6c21 100644 --- a/drivers/input/touchscreen/cyttsp5.c +++ b/drivers/input/touchscreen/cyttsp5.c @@ -870,13 +870,16 @@ static int cyttsp5_probe(struct device *dev, struct regmap *regmap, int irq, ts->input->phys = ts->phys; input_set_drvdata(ts->input, ts); - /* Reset the gpio to be in a reset state */ + /* Assert gpio to be in a reset state */ ts->reset_gpio = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_HIGH); if (IS_ERR(ts->reset_gpio)) { error = PTR_ERR(ts->reset_gpio); dev_err(dev, "Failed to request reset gpio, error %d\n", error); return error; } + + fsleep(10); /* Ensure long-enough reset pulse (minimum 10us). */ + gpiod_set_value_cansleep(ts->reset_gpio, 0); /* Need a delay to have device up */ From 712dc3e7b979ae9ae1afeb1e87ec92ebb72b9529 Mon Sep 17 00:00:00 2001 From: Asahi Lina Date: Wed, 9 Apr 2025 18:11:21 +0900 Subject: [PATCH 043/555] mailmap: Update email for Asahi Lina Add an alias so I can more easily filter kernel-related emails. Signed-off-by: Asahi Lina Link: https://lore.kernel.org/r/20250409-mailmap-lina-email-v1-1-265d05848ae3@asahilina.net Signed-off-by: Sven Peter --- .mailmap | 1 + 1 file changed, 1 insertion(+) diff --git a/.mailmap b/.mailmap index 4f7cd8e23177..8f0cc0b08ed0 100644 --- a/.mailmap +++ b/.mailmap @@ -102,6 +102,7 @@ Ard Biesheuvel Arnaud Patard Arnd Bergmann Arun Kumar Neelakantam +Asahi Lina Ashok Raj Nagarajan Ashwin Chaugule Asutosh Das From dc5befecbe2683ac49fc8dc76aade35e62f4cf30 Mon Sep 17 00:00:00 2001 From: Janne Grunau Date: Wed, 16 Apr 2025 20:06:18 +0200 Subject: [PATCH 044/555] arm64: dts: apple: touchbar: Mark ps_dispdfr_be as always-on The driver depends on boot loader initialized state which resets when the ps_dispdfr_be power-domain is powered off. This happens on suspend or when the driver is missing during boot. Mark the domain as always on until the driver can handle this. Fixes: 7275e795e520 ("arm64: dts: apple: Add touchbar screen nodes") Signed-off-by: Janne Grunau Reviewed-by: Alyssa Rosenzweig Link: https://lore.kernel.org/r/20250416-arm64_dts_apple_touchbar-v1-1-e1c0b53b9125@jannau.net Signed-off-by: Sven Peter --- arch/arm64/boot/dts/apple/t8103-j293.dts | 10 ++++++++++ arch/arm64/boot/dts/apple/t8112-j493.dts | 10 ++++++++++ 2 files changed, 20 insertions(+) diff --git a/arch/arm64/boot/dts/apple/t8103-j293.dts b/arch/arm64/boot/dts/apple/t8103-j293.dts index 2dfe7b895b2b..e2d9439397f7 100644 --- a/arch/arm64/boot/dts/apple/t8103-j293.dts +++ b/arch/arm64/boot/dts/apple/t8103-j293.dts @@ -77,6 +77,16 @@ }; }; +/* + * The driver depends on boot loader initialized state which resets when this + * power-domain is powered off. This happens on suspend or when the driver is + * missing during boot. Mark the domain as always on until the driver can + * handle this. + */ +&ps_dispdfr_be { + apple,always-on; +}; + &display_dfr { status = "okay"; }; diff --git a/arch/arm64/boot/dts/apple/t8112-j493.dts b/arch/arm64/boot/dts/apple/t8112-j493.dts index 3d73f9ee2f46..be86d34c6696 100644 --- a/arch/arm64/boot/dts/apple/t8112-j493.dts +++ b/arch/arm64/boot/dts/apple/t8112-j493.dts @@ -40,6 +40,16 @@ }; }; +/* + * The driver depends on boot loader initialized state which resets when this + * power-domain is powered off. This happens on suspend or when the driver is + * missing during boot. Mark the domain as always on until the driver can + * handle this. + */ +&ps_dispdfr_be { + apple,always-on; +}; + &display_dfr { status = "okay"; }; From f81b33582f9339d2dc17c69b92040d3650bb4bae Mon Sep 17 00:00:00 2001 From: Zhu Yanjun Date: Sat, 12 Apr 2025 09:57:14 +0200 Subject: [PATCH 045/555] RDMA/rxe: Fix slab-use-after-free Read in rxe_queue_cleanup bug Call Trace: __dump_stack lib/dump_stack.c:94 [inline] dump_stack_lvl+0x7d/0xa0 lib/dump_stack.c:120 print_address_description mm/kasan/report.c:378 [inline] print_report+0xcf/0x610 mm/kasan/report.c:489 kasan_report+0xb5/0xe0 mm/kasan/report.c:602 rxe_queue_cleanup+0xd0/0xe0 drivers/infiniband/sw/rxe/rxe_queue.c:195 rxe_cq_cleanup+0x3f/0x50 drivers/infiniband/sw/rxe/rxe_cq.c:132 __rxe_cleanup+0x168/0x300 drivers/infiniband/sw/rxe/rxe_pool.c:232 rxe_create_cq+0x22e/0x3a0 drivers/infiniband/sw/rxe/rxe_verbs.c:1109 create_cq+0x658/0xb90 drivers/infiniband/core/uverbs_cmd.c:1052 ib_uverbs_create_cq+0xc7/0x120 drivers/infiniband/core/uverbs_cmd.c:1095 ib_uverbs_write+0x969/0xc90 drivers/infiniband/core/uverbs_main.c:679 vfs_write fs/read_write.c:677 [inline] vfs_write+0x26a/0xcc0 fs/read_write.c:659 ksys_write+0x1b8/0x200 fs/read_write.c:731 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xaa/0x1b0 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f In the function rxe_create_cq, when rxe_cq_from_init fails, the function rxe_cleanup will be called to handle the allocated resources. In fact, some memory resources have already been freed in the function rxe_cq_from_init. Thus, this problem will occur. The solution is to let rxe_cleanup do all the work. Fixes: 8700e3e7c485 ("Soft RoCE driver") Link: https://paste.ubuntu.com/p/tJgC42wDf6/ Tested-by: liuyi Signed-off-by: Zhu Yanjun Link: https://patch.msgid.link/20250412075714.3257358-1-yanjun.zhu@linux.dev Reviewed-by: Daisuke Matsuda Signed-off-by: Leon Romanovsky --- drivers/infiniband/sw/rxe/rxe_cq.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/drivers/infiniband/sw/rxe/rxe_cq.c b/drivers/infiniband/sw/rxe/rxe_cq.c index fec87c9030ab..fffd144d509e 100644 --- a/drivers/infiniband/sw/rxe/rxe_cq.c +++ b/drivers/infiniband/sw/rxe/rxe_cq.c @@ -56,11 +56,8 @@ int rxe_cq_from_init(struct rxe_dev *rxe, struct rxe_cq *cq, int cqe, err = do_mmap_info(rxe, uresp ? &uresp->mi : NULL, udata, cq->queue->buf, cq->queue->buf_size, &cq->queue->ip); - if (err) { - vfree(cq->queue->buf); - kfree(cq->queue); + if (err) return err; - } cq->is_user = uresp; From 80f2ab46c2ee16f046b55306dc4db4be53125016 Mon Sep 17 00:00:00 2001 From: Michal Swiatkowski Date: Mon, 14 Apr 2025 18:42:30 -0500 Subject: [PATCH 046/555] irdma: free iwdev->rf after removing MSI-X Currently iwdev->rf is allocated in irdma_probe(), but free in irdma_ib_dealloc_device(). It can be misleading. Move the free to irdma_remove() to be more obvious. Freeing in irdma_ib_dealloc_device() leads to KASAN use-after-free issue. Which can also lead to NULL pointer dereference. Fix this. irdma_deinit_interrupts() can't be moved before freeing iwdef->rf, because in this case deinit interrupts will be done before freeing irqs. The simplest solution is to move kfree(iwdev->rf) to irdma_remove(). Reproducer: sudo rmmod irdma Minified splat(s): BUG: KASAN: use-after-free in irdma_remove+0x257/0x2d0 [irdma] Call Trace: ? __pfx__raw_spin_lock_irqsave+0x10/0x10 ? kfree+0x253/0x450 ? irdma_remove+0x257/0x2d0 [irdma] kasan_report+0xed/0x120 ? irdma_remove+0x257/0x2d0 [irdma] irdma_remove+0x257/0x2d0 [irdma] auxiliary_bus_remove+0x56/0x80 device_release_driver_internal+0x371/0x530 ? kernfs_put.part.0+0x147/0x310 driver_detach+0xbf/0x180 bus_remove_driver+0x11b/0x2a0 auxiliary_driver_unregister+0x1a/0x50 irdma_exit_module+0x40/0x4c [irdma] Oops: general protection fault, probably for non-canonical address 0xdffffc0000000000: 0000 [#1] PREEMPT SMP KASAN NOPTI KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007] RIP: 0010:ice_free_rdma_qvector+0x2a/0xa0 [ice] Call Trace: ? ice_free_rdma_qvector+0x2a/0xa0 [ice] irdma_remove+0x179/0x2d0 [irdma] auxiliary_bus_remove+0x56/0x80 device_release_driver_internal+0x371/0x530 ? kobject_put+0x61/0x4b0 driver_detach+0xbf/0x180 bus_remove_driver+0x11b/0x2a0 auxiliary_driver_unregister+0x1a/0x50 irdma_exit_module+0x40/0x4c [irdma] Reported-by: Marcin Szycik Closes: https://lore.kernel.org/netdev/8e533834-4564-472f-b29b-4f1cb7730053@linux.intel.com/ Fixes: 3e0d3cb3fbe0 ("ice, irdma: move interrupts code to irdma") Reviewed-by: Marcin Szycik Signed-off-by: Michal Swiatkowski Signed-off-by: Tatyana Nikolova Link: https://patch.msgid.link/20250414234231.523-1-tatyana.e.nikolova@intel.com Signed-off-by: Leon Romanovsky --- drivers/infiniband/hw/irdma/main.c | 2 ++ drivers/infiniband/hw/irdma/verbs.c | 1 - 2 files changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/irdma/main.c b/drivers/infiniband/hw/irdma/main.c index 1ee8969595d3..d10fd16dcec3 100644 --- a/drivers/infiniband/hw/irdma/main.c +++ b/drivers/infiniband/hw/irdma/main.c @@ -255,6 +255,8 @@ static void irdma_remove(struct auxiliary_device *aux_dev) ice_rdma_update_vsi_filter(pf, iwdev->vsi_num, false); irdma_deinit_interrupts(iwdev->rf, pf); + kfree(iwdev->rf); + pr_debug("INIT: Gen2 PF[%d] device remove success\n", PCI_FUNC(pf->pdev->devfn)); } diff --git a/drivers/infiniband/hw/irdma/verbs.c b/drivers/infiniband/hw/irdma/verbs.c index eeb932e58730..1e8c92826de2 100644 --- a/drivers/infiniband/hw/irdma/verbs.c +++ b/drivers/infiniband/hw/irdma/verbs.c @@ -4871,5 +4871,4 @@ void irdma_ib_dealloc_device(struct ib_device *ibdev) irdma_rt_deinit_hw(iwdev); irdma_ctrl_deinit_hw(iwdev->rf); - kfree(iwdev->rf); } From 4bcc063939a560f05b05b34be68d20045a646e6e Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Mon, 14 Apr 2025 18:42:31 -0500 Subject: [PATCH 047/555] ice, irdma: fix an off by one in error handling code If we don't allocate the MIN number of IRQs then we need to free what we have and return -ENOMEM. The problem is this loop is off by one so it frees an entry that wasn't allocated and it doesn't free the first entry where i == 0. Fixes: 3e0d3cb3fbe0 ("ice, irdma: move interrupts code to irdma") Signed-off-by: Dan Carpenter Signed-off-by: Tatyana Nikolova Link: https://patch.msgid.link/20250414234231.523-2-tatyana.e.nikolova@intel.com Reviewed-by: Michal Swiatkowski Signed-off-by: Leon Romanovsky --- drivers/infiniband/hw/irdma/main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/irdma/main.c b/drivers/infiniband/hw/irdma/main.c index d10fd16dcec3..7599e31b5743 100644 --- a/drivers/infiniband/hw/irdma/main.c +++ b/drivers/infiniband/hw/irdma/main.c @@ -221,7 +221,7 @@ static int irdma_init_interrupts(struct irdma_pci_f *rf, struct ice_pf *pf) break; if (i < IRDMA_MIN_MSIX) { - for (; i > 0; i--) + while (--i >= 0) ice_free_rdma_qvector(pf, &rf->msix_entries[i]); kfree(rf->msix_entries); From e4570f4bb231f01e32d44fd38841665f340d6914 Mon Sep 17 00:00:00 2001 From: David Lechner Date: Fri, 18 Apr 2025 11:17:13 -0500 Subject: [PATCH 048/555] iio: imu: adis16550: align buffers for timestamp Align the buffers used with iio_push_to_buffers_with_timestamp() to ensure the s64 timestamp is aligned to 8 bytes. Fixes: bac4368fab62 ("iio: imu: adis16550: add adis16550 support") Signed-off-by: David Lechner Link: https://patch.msgid.link/20250418-iio-more-timestamp-alignment-v2-1-d6a5d2b1c9fe@baylibre.com Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/imu/adis16550.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iio/imu/adis16550.c b/drivers/iio/imu/adis16550.c index b14ea8937c7f..28f0dbd0226c 100644 --- a/drivers/iio/imu/adis16550.c +++ b/drivers/iio/imu/adis16550.c @@ -836,7 +836,7 @@ static irqreturn_t adis16550_trigger_handler(int irq, void *p) u16 dummy; bool valid; struct iio_poll_func *pf = p; - __be32 data[ADIS16550_MAX_SCAN_DATA]; + __be32 data[ADIS16550_MAX_SCAN_DATA] __aligned(8); struct iio_dev *indio_dev = pf->indio_dev; struct adis16550 *st = iio_priv(indio_dev); struct adis *adis = iio_device_get_drvdata(indio_dev); From ffcd19e9f4cca0c8f9e23e88f968711acefbb37b Mon Sep 17 00:00:00 2001 From: David Lechner Date: Fri, 18 Apr 2025 11:17:14 -0500 Subject: [PATCH 049/555] iio: pressure: mprls0025pa: use aligned_s64 for timestamp Follow the pattern of other drivers and use aligned_s64 for the timestamp. This will ensure the struct itself it also 8-byte aligned. While touching this, convert struct mpr_chan to an anonymous struct to consolidate the code a bit to make it easier for future readers. Fixes: 713337d9143e ("iio: pressure: Honeywell mprls0025pa pressure sensor") Signed-off-by: David Lechner Link: https://patch.msgid.link/20250418-iio-more-timestamp-alignment-v2-2-d6a5d2b1c9fe@baylibre.com Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/pressure/mprls0025pa.h | 17 ++++++----------- 1 file changed, 6 insertions(+), 11 deletions(-) diff --git a/drivers/iio/pressure/mprls0025pa.h b/drivers/iio/pressure/mprls0025pa.h index 9d5c30afa9d6..d62a018eaff3 100644 --- a/drivers/iio/pressure/mprls0025pa.h +++ b/drivers/iio/pressure/mprls0025pa.h @@ -34,16 +34,6 @@ struct iio_dev; struct mpr_data; struct mpr_ops; -/** - * struct mpr_chan - * @pres: pressure value - * @ts: timestamp - */ -struct mpr_chan { - s32 pres; - s64 ts; -}; - enum mpr_func_id { MPR_FUNCTION_A, MPR_FUNCTION_B, @@ -69,6 +59,8 @@ enum mpr_func_id { * reading in a loop until data is ready * @completion: handshake from irq to read * @chan: channel values for buffered mode + * @chan.pres: pressure value + * @chan.ts: timestamp * @buffer: raw conversion data */ struct mpr_data { @@ -87,7 +79,10 @@ struct mpr_data { struct gpio_desc *gpiod_reset; int irq; struct completion completion; - struct mpr_chan chan; + struct { + s32 pres; + aligned_s64 ts; + } chan; u8 buffer[MPR_MEASUREMENT_RD_SIZE] __aligned(IIO_DMA_MINALIGN); }; From f083f8a21cc785ebe3a33f756a3fa3660611f8db Mon Sep 17 00:00:00 2001 From: Angelo Dureghello Date: Fri, 18 Apr 2025 20:37:53 +0200 Subject: [PATCH 050/555] iio: adc: ad7606: fix serial register access Fix register read/write routine as per datasheet. When reading multiple consecutive registers, only the first one is read properly. This is due to missing chip select deassert and assert again between first and second 16bit transfer, as shown in the datasheet AD7606C-16, rev 0, figure 110. Fixes: f2a22e1e172f ("iio: adc: ad7606: Add support for software mode for ad7616") Reviewed-by: David Lechner Signed-off-by: Angelo Dureghello Link: https://patch.msgid.link/20250418-wip-bl-ad7606-fix-reg-access-v3-1-d5eeb440c738@baylibre.com Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/adc/ad7606_spi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iio/adc/ad7606_spi.c b/drivers/iio/adc/ad7606_spi.c index 885bf0b68e77..179115e90988 100644 --- a/drivers/iio/adc/ad7606_spi.c +++ b/drivers/iio/adc/ad7606_spi.c @@ -131,7 +131,7 @@ static int ad7606_spi_reg_read(struct ad7606_state *st, unsigned int addr) { .tx_buf = &st->d16[0], .len = 2, - .cs_change = 0, + .cs_change = 1, }, { .rx_buf = &st->d16[1], .len = 2, From 609bc31eca06c7408e6860d8b46311ebe45c1fef Mon Sep 17 00:00:00 2001 From: Gabriel Shahrouzi Date: Mon, 21 Apr 2025 09:15:39 -0400 Subject: [PATCH 051/555] iio: adis16201: Correct inclinometer channel resolution The inclinometer channels were previously defined with 14 realbits. However, the ADIS16201 datasheet states the resolution for these output channels is 12 bits (Page 14, text description; Page 15, table 7). Correct the realbits value to 12 to accurately reflect the hardware. Fixes: f7fe1d1dd5a5 ("staging: iio: new adis16201 driver") Cc: stable@vger.kernel.org Signed-off-by: Gabriel Shahrouzi Reviewed-by: Marcelo Schmitt Link: https://patch.msgid.link/20250421131539.912966-1-gshahrouzi@gmail.com Signed-off-by: Jonathan Cameron --- drivers/iio/accel/adis16201.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/iio/accel/adis16201.c b/drivers/iio/accel/adis16201.c index 8601b9a8b8e7..5127e58eebc7 100644 --- a/drivers/iio/accel/adis16201.c +++ b/drivers/iio/accel/adis16201.c @@ -211,9 +211,9 @@ static const struct iio_chan_spec adis16201_channels[] = { BIT(IIO_CHAN_INFO_CALIBBIAS), 0, 14), ADIS_AUX_ADC_CHAN(ADIS16201_AUX_ADC_REG, ADIS16201_SCAN_AUX_ADC, 0, 12), ADIS_INCLI_CHAN(X, ADIS16201_XINCL_OUT_REG, ADIS16201_SCAN_INCLI_X, - BIT(IIO_CHAN_INFO_CALIBBIAS), 0, 14), + BIT(IIO_CHAN_INFO_CALIBBIAS), 0, 12), ADIS_INCLI_CHAN(Y, ADIS16201_YINCL_OUT_REG, ADIS16201_SCAN_INCLI_Y, - BIT(IIO_CHAN_INFO_CALIBBIAS), 0, 14), + BIT(IIO_CHAN_INFO_CALIBBIAS), 0, 12), IIO_CHAN_SOFT_TIMESTAMP(7) }; From 3409f843c04df9434beb637602382eeee90e0bee Mon Sep 17 00:00:00 2001 From: Martin Blumenstingl Date: Sun, 20 Apr 2025 18:47:58 +0200 Subject: [PATCH 052/555] ARM: dts: amlogic: meson8: fix reference to unknown/untested PWM clock Device-tree expects absent clocks to be specified as <0> (instead of using <>). This fixes using the FCLK4/FCLK3 clocks as they are now seen at their correct index (while before they were recognized, but at the correct index - resulting in the hardware using a different clock than what the kernel sees). Fixes: 802cff460aab ("ARM: dts: amlogic: meson8: switch to the new PWM controller binding") Signed-off-by: Martin Blumenstingl Reviewed-by: Neil Armstrong Link: https://lore.kernel.org/r/20250420164801.330505-2-martin.blumenstingl@googlemail.com Signed-off-by: Neil Armstrong --- arch/arm/boot/dts/amlogic/meson8.dtsi | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/arm/boot/dts/amlogic/meson8.dtsi b/arch/arm/boot/dts/amlogic/meson8.dtsi index 847f7b1f1e96..f785e0de0847 100644 --- a/arch/arm/boot/dts/amlogic/meson8.dtsi +++ b/arch/arm/boot/dts/amlogic/meson8.dtsi @@ -451,7 +451,7 @@ pwm_ef: pwm@86c0 { compatible = "amlogic,meson8-pwm-v2"; clocks = <&xtal>, - <>, /* unknown/untested, the datasheet calls it "Video PLL" */ + <0>, /* unknown/untested, the datasheet calls it "Video PLL" */ <&clkc CLKID_FCLK_DIV4>, <&clkc CLKID_FCLK_DIV3>; reg = <0x86c0 0x10>; @@ -705,7 +705,7 @@ &pwm_ab { compatible = "amlogic,meson8-pwm-v2"; clocks = <&xtal>, - <>, /* unknown/untested, the datasheet calls it "Video PLL" */ + <0>, /* unknown/untested, the datasheet calls it "Video PLL" */ <&clkc CLKID_FCLK_DIV4>, <&clkc CLKID_FCLK_DIV3>; }; @@ -713,7 +713,7 @@ &pwm_cd { compatible = "amlogic,meson8-pwm-v2"; clocks = <&xtal>, - <>, /* unknown/untested, the datasheet calls it "Video PLL" */ + <0>, /* unknown/untested, the datasheet calls it "Video PLL" */ <&clkc CLKID_FCLK_DIV4>, <&clkc CLKID_FCLK_DIV3>; }; From a994b58f9d1163c4f559bd169721f0fc15866919 Mon Sep 17 00:00:00 2001 From: Martin Blumenstingl Date: Sun, 20 Apr 2025 18:47:59 +0200 Subject: [PATCH 053/555] ARM: dts: amlogic: meson8b: fix reference to unknown/untested PWM clock Device-tree expects absent clocks to be specified as <0> (instead of using <>). This fixes using the FCLK4/FCLK3 clocks as they are now seen at their correct index (while before they were recognized, but at the correct index - resulting in the hardware using a different clock than what the kernel sees). Fixes: dbf921861985 ("ARM: dts: amlogic: meson8b: switch to the new PWM controller binding") Signed-off-by: Martin Blumenstingl Reviewed-by: Neil Armstrong Link: https://lore.kernel.org/r/20250420164801.330505-3-martin.blumenstingl@googlemail.com Signed-off-by: Neil Armstrong --- arch/arm/boot/dts/amlogic/meson8b.dtsi | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/arm/boot/dts/amlogic/meson8b.dtsi b/arch/arm/boot/dts/amlogic/meson8b.dtsi index 0876611ce26a..fdb0abe23a0c 100644 --- a/arch/arm/boot/dts/amlogic/meson8b.dtsi +++ b/arch/arm/boot/dts/amlogic/meson8b.dtsi @@ -406,7 +406,7 @@ compatible = "amlogic,meson8b-pwm-v2", "amlogic,meson8-pwm-v2"; reg = <0x86c0 0x10>; clocks = <&xtal>, - <>, /* unknown/untested, the datasheet calls it "Video PLL" */ + <0>, /* unknown/untested, the datasheet calls it "Video PLL" */ <&clkc CLKID_FCLK_DIV4>, <&clkc CLKID_FCLK_DIV3>; #pwm-cells = <3>; @@ -680,7 +680,7 @@ &pwm_ab { compatible = "amlogic,meson8b-pwm-v2", "amlogic,meson8-pwm-v2"; clocks = <&xtal>, - <>, /* unknown/untested, the datasheet calls it "Video PLL" */ + <0>, /* unknown/untested, the datasheet calls it "Video PLL" */ <&clkc CLKID_FCLK_DIV4>, <&clkc CLKID_FCLK_DIV3>; }; @@ -688,7 +688,7 @@ &pwm_cd { compatible = "amlogic,meson8b-pwm-v2", "amlogic,meson8-pwm-v2"; clocks = <&xtal>, - <>, /* unknown/untested, the datasheet calls it "Video PLL" */ + <0>, /* unknown/untested, the datasheet calls it "Video PLL" */ <&clkc CLKID_FCLK_DIV4>, <&clkc CLKID_FCLK_DIV3>; }; From 511d388a4007ba580feeb2fd2e9ba35a614c093f Mon Sep 17 00:00:00 2001 From: Martin Blumenstingl Date: Sun, 20 Apr 2025 18:48:00 +0200 Subject: [PATCH 054/555] arm64: dts: amlogic: gx: fix reference to unknown/untested PWM clock Device-tree expects absent clocks to be specified as <0> (instead of using <>). This fixes using the FCLK4/FCLK3 clocks as they are now seen at their correct index (while before they were recognized, but at the correct index - resulting in the hardware using a different clock than what the kernel sees). Fixes: a526eeef9a81 ("arm64: dts: amlogic: gx: switch to the new PWM controller binding") Signed-off-by: Martin Blumenstingl Reviewed-by: Neil Armstrong Link: https://lore.kernel.org/r/20250420164801.330505-4-martin.blumenstingl@googlemail.com Signed-off-by: Neil Armstrong --- arch/arm64/boot/dts/amlogic/meson-gxbb.dtsi | 6 +++--- arch/arm64/boot/dts/amlogic/meson-gxl.dtsi | 6 +++--- 2 files changed, 6 insertions(+), 6 deletions(-) diff --git a/arch/arm64/boot/dts/amlogic/meson-gxbb.dtsi b/arch/arm64/boot/dts/amlogic/meson-gxbb.dtsi index 8ebce7114a60..6c134592c7bb 100644 --- a/arch/arm64/boot/dts/amlogic/meson-gxbb.dtsi +++ b/arch/arm64/boot/dts/amlogic/meson-gxbb.dtsi @@ -741,7 +741,7 @@ &pwm_ab { clocks = <&xtal>, - <>, /* unknown/untested, the datasheet calls it "vid_pll" */ + <0>, /* unknown/untested, the datasheet calls it "vid_pll" */ <&clkc CLKID_FCLK_DIV4>, <&clkc CLKID_FCLK_DIV3>; }; @@ -752,14 +752,14 @@ &pwm_cd { clocks = <&xtal>, - <>, /* unknown/untested, the datasheet calls it "vid_pll" */ + <0>, /* unknown/untested, the datasheet calls it "vid_pll" */ <&clkc CLKID_FCLK_DIV4>, <&clkc CLKID_FCLK_DIV3>; }; &pwm_ef { clocks = <&xtal>, - <>, /* unknown/untested, the datasheet calls it "vid_pll" */ + <0>, /* unknown/untested, the datasheet calls it "vid_pll" */ <&clkc CLKID_FCLK_DIV4>, <&clkc CLKID_FCLK_DIV3>; }; diff --git a/arch/arm64/boot/dts/amlogic/meson-gxl.dtsi b/arch/arm64/boot/dts/amlogic/meson-gxl.dtsi index 2dc2fdaecf9f..19b8a39de6a0 100644 --- a/arch/arm64/boot/dts/amlogic/meson-gxl.dtsi +++ b/arch/arm64/boot/dts/amlogic/meson-gxl.dtsi @@ -811,7 +811,7 @@ &pwm_ab { clocks = <&xtal>, - <>, /* unknown/untested, the datasheet calls it "vid_pll" */ + <0>, /* unknown/untested, the datasheet calls it "vid_pll" */ <&clkc CLKID_FCLK_DIV4>, <&clkc CLKID_FCLK_DIV3>; }; @@ -822,14 +822,14 @@ &pwm_cd { clocks = <&xtal>, - <>, /* unknown/untested, the datasheet calls it "vid_pll" */ + <0>, /* unknown/untested, the datasheet calls it "vid_pll" */ <&clkc CLKID_FCLK_DIV4>, <&clkc CLKID_FCLK_DIV3>; }; &pwm_ef { clocks = <&xtal>, - <>, /* unknown/untested, the datasheet calls it "vid_pll" */ + <0>, /* unknown/untested, the datasheet calls it "vid_pll" */ <&clkc CLKID_FCLK_DIV4>, <&clkc CLKID_FCLK_DIV3>; }; From a08b28c1ed454502abeb90ffa4a55445dae1d22a Mon Sep 17 00:00:00 2001 From: Martin Blumenstingl Date: Sun, 20 Apr 2025 18:48:01 +0200 Subject: [PATCH 055/555] arm64: dts: amlogic: g12: fix reference to unknown/untested PWM clock Device-tree expects absent clocks to be specified as <0> (instead of using <>). This fixes using the FCLK4/FCLK3 clocks as they are now seen at their correct index (while before they were recognized, but at the correct index - resulting in the hardware using a different clock than what the kernel sees). Fixes: e6884f2e4129 ("arm64: dts: amlogic: g12: switch to the new PWM controller binding") Signed-off-by: Martin Blumenstingl Reviewed-by: Neil Armstrong Link: https://lore.kernel.org/r/20250420164801.330505-5-martin.blumenstingl@googlemail.com Signed-off-by: Neil Armstrong --- arch/arm64/boot/dts/amlogic/meson-g12-common.dtsi | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/arm64/boot/dts/amlogic/meson-g12-common.dtsi b/arch/arm64/boot/dts/amlogic/meson-g12-common.dtsi index ab2b3f15ef19..69834b49673d 100644 --- a/arch/arm64/boot/dts/amlogic/meson-g12-common.dtsi +++ b/arch/arm64/boot/dts/amlogic/meson-g12-common.dtsi @@ -2313,7 +2313,7 @@ "amlogic,meson8-pwm-v2"; reg = <0x0 0x19000 0x0 0x20>; clocks = <&xtal>, - <>, /* unknown/untested, the datasheet calls it "vid_pll" */ + <0>, /* unknown/untested, the datasheet calls it "vid_pll" */ <&clkc CLKID_FCLK_DIV4>, <&clkc CLKID_FCLK_DIV3>; #pwm-cells = <3>; @@ -2325,7 +2325,7 @@ "amlogic,meson8-pwm-v2"; reg = <0x0 0x1a000 0x0 0x20>; clocks = <&xtal>, - <>, /* unknown/untested, the datasheet calls it "vid_pll" */ + <0>, /* unknown/untested, the datasheet calls it "vid_pll" */ <&clkc CLKID_FCLK_DIV4>, <&clkc CLKID_FCLK_DIV3>; #pwm-cells = <3>; @@ -2337,7 +2337,7 @@ "amlogic,meson8-pwm-v2"; reg = <0x0 0x1b000 0x0 0x20>; clocks = <&xtal>, - <>, /* unknown/untested, the datasheet calls it "vid_pll" */ + <0>, /* unknown/untested, the datasheet calls it "vid_pll" */ <&clkc CLKID_FCLK_DIV4>, <&clkc CLKID_FCLK_DIV3>; #pwm-cells = <3>; From 18853ba782bef65fc81ef2b3370382e5b479c5eb Mon Sep 17 00:00:00 2001 From: Andrea Righi Date: Tue, 22 Apr 2025 10:26:32 +0200 Subject: [PATCH 056/555] sched_ext: Track currently locked rq Some kfuncs provided by sched_ext may need to operate on a struct rq, but they can be invoked from various contexts, specifically, different scx callbacks. While some of these callbacks are invoked with a particular rq already locked, others are not. This makes it impossible for a kfunc to reliably determine whether it's safe to access a given rq, triggering potential bugs or unsafe behaviors, see for example [1]. To address this, track the currently locked rq whenever a sched_ext callback is invoked via SCX_CALL_OP*(). This allows kfuncs that need to operate on an arbitrary rq to retrieve the currently locked one and apply the appropriate action as needed. [1] https://lore.kernel.org/lkml/20250325140021.73570-1-arighi@nvidia.com/ Suggested-by: Tejun Heo Signed-off-by: Andrea Righi Acked-by: Changwoo Min Signed-off-by: Tejun Heo --- kernel/sched/ext.c | 152 +++++++++++++++++++++++++--------------- kernel/sched/ext_idle.c | 2 +- 2 files changed, 95 insertions(+), 59 deletions(-) diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c index fdbf249d1c68..585bf6d8238b 100644 --- a/kernel/sched/ext.c +++ b/kernel/sched/ext.c @@ -1118,8 +1118,38 @@ static void scx_kf_disallow(u32 mask) current->scx.kf_mask &= ~mask; } -#define SCX_CALL_OP(mask, op, args...) \ +/* + * Track the rq currently locked. + * + * This allows kfuncs to safely operate on rq from any scx ops callback, + * knowing which rq is already locked. + */ +static DEFINE_PER_CPU(struct rq *, locked_rq); + +static inline void update_locked_rq(struct rq *rq) +{ + /* + * Check whether @rq is actually locked. This can help expose bugs + * or incorrect assumptions about the context in which a kfunc or + * callback is executed. + */ + if (rq) + lockdep_assert_rq_held(rq); + __this_cpu_write(locked_rq, rq); +} + +/* + * Return the rq currently locked from an scx callback, or NULL if no rq is + * locked. + */ +static inline struct rq *scx_locked_rq(void) +{ + return __this_cpu_read(locked_rq); +} + +#define SCX_CALL_OP(mask, op, rq, args...) \ do { \ + update_locked_rq(rq); \ if (mask) { \ scx_kf_allow(mask); \ scx_ops.op(args); \ @@ -1127,11 +1157,14 @@ do { \ } else { \ scx_ops.op(args); \ } \ + update_locked_rq(NULL); \ } while (0) -#define SCX_CALL_OP_RET(mask, op, args...) \ +#define SCX_CALL_OP_RET(mask, op, rq, args...) \ ({ \ __typeof__(scx_ops.op(args)) __ret; \ + \ + update_locked_rq(rq); \ if (mask) { \ scx_kf_allow(mask); \ __ret = scx_ops.op(args); \ @@ -1139,6 +1172,7 @@ do { \ } else { \ __ret = scx_ops.op(args); \ } \ + update_locked_rq(NULL); \ __ret; \ }) @@ -1153,31 +1187,31 @@ do { \ * scx_kf_allowed_on_arg_tasks() to test whether the invocation is allowed on * the specific task. */ -#define SCX_CALL_OP_TASK(mask, op, task, args...) \ +#define SCX_CALL_OP_TASK(mask, op, rq, task, args...) \ do { \ BUILD_BUG_ON((mask) & ~__SCX_KF_TERMINAL); \ current->scx.kf_tasks[0] = task; \ - SCX_CALL_OP(mask, op, task, ##args); \ + SCX_CALL_OP(mask, op, rq, task, ##args); \ current->scx.kf_tasks[0] = NULL; \ } while (0) -#define SCX_CALL_OP_TASK_RET(mask, op, task, args...) \ +#define SCX_CALL_OP_TASK_RET(mask, op, rq, task, args...) \ ({ \ __typeof__(scx_ops.op(task, ##args)) __ret; \ BUILD_BUG_ON((mask) & ~__SCX_KF_TERMINAL); \ current->scx.kf_tasks[0] = task; \ - __ret = SCX_CALL_OP_RET(mask, op, task, ##args); \ + __ret = SCX_CALL_OP_RET(mask, op, rq, task, ##args); \ current->scx.kf_tasks[0] = NULL; \ __ret; \ }) -#define SCX_CALL_OP_2TASKS_RET(mask, op, task0, task1, args...) \ +#define SCX_CALL_OP_2TASKS_RET(mask, op, rq, task0, task1, args...) \ ({ \ __typeof__(scx_ops.op(task0, task1, ##args)) __ret; \ BUILD_BUG_ON((mask) & ~__SCX_KF_TERMINAL); \ current->scx.kf_tasks[0] = task0; \ current->scx.kf_tasks[1] = task1; \ - __ret = SCX_CALL_OP_RET(mask, op, task0, task1, ##args); \ + __ret = SCX_CALL_OP_RET(mask, op, rq, task0, task1, ##args); \ current->scx.kf_tasks[0] = NULL; \ current->scx.kf_tasks[1] = NULL; \ __ret; \ @@ -2172,7 +2206,7 @@ static void do_enqueue_task(struct rq *rq, struct task_struct *p, u64 enq_flags, WARN_ON_ONCE(*ddsp_taskp); *ddsp_taskp = p; - SCX_CALL_OP_TASK(SCX_KF_ENQUEUE, enqueue, p, enq_flags); + SCX_CALL_OP_TASK(SCX_KF_ENQUEUE, enqueue, rq, p, enq_flags); *ddsp_taskp = NULL; if (p->scx.ddsp_dsq_id != SCX_DSQ_INVALID) @@ -2269,7 +2303,7 @@ static void enqueue_task_scx(struct rq *rq, struct task_struct *p, int enq_flags add_nr_running(rq, 1); if (SCX_HAS_OP(runnable) && !task_on_rq_migrating(p)) - SCX_CALL_OP_TASK(SCX_KF_REST, runnable, p, enq_flags); + SCX_CALL_OP_TASK(SCX_KF_REST, runnable, rq, p, enq_flags); if (enq_flags & SCX_ENQ_WAKEUP) touch_core_sched(rq, p); @@ -2283,7 +2317,7 @@ out: __scx_add_event(SCX_EV_SELECT_CPU_FALLBACK, 1); } -static void ops_dequeue(struct task_struct *p, u64 deq_flags) +static void ops_dequeue(struct rq *rq, struct task_struct *p, u64 deq_flags) { unsigned long opss; @@ -2304,7 +2338,7 @@ static void ops_dequeue(struct task_struct *p, u64 deq_flags) BUG(); case SCX_OPSS_QUEUED: if (SCX_HAS_OP(dequeue)) - SCX_CALL_OP_TASK(SCX_KF_REST, dequeue, p, deq_flags); + SCX_CALL_OP_TASK(SCX_KF_REST, dequeue, rq, p, deq_flags); if (atomic_long_try_cmpxchg(&p->scx.ops_state, &opss, SCX_OPSS_NONE)) @@ -2337,7 +2371,7 @@ static bool dequeue_task_scx(struct rq *rq, struct task_struct *p, int deq_flags return true; } - ops_dequeue(p, deq_flags); + ops_dequeue(rq, p, deq_flags); /* * A currently running task which is going off @rq first gets dequeued @@ -2353,11 +2387,11 @@ static bool dequeue_task_scx(struct rq *rq, struct task_struct *p, int deq_flags */ if (SCX_HAS_OP(stopping) && task_current(rq, p)) { update_curr_scx(rq); - SCX_CALL_OP_TASK(SCX_KF_REST, stopping, p, false); + SCX_CALL_OP_TASK(SCX_KF_REST, stopping, rq, p, false); } if (SCX_HAS_OP(quiescent) && !task_on_rq_migrating(p)) - SCX_CALL_OP_TASK(SCX_KF_REST, quiescent, p, deq_flags); + SCX_CALL_OP_TASK(SCX_KF_REST, quiescent, rq, p, deq_flags); if (deq_flags & SCX_DEQ_SLEEP) p->scx.flags |= SCX_TASK_DEQD_FOR_SLEEP; @@ -2377,7 +2411,7 @@ static void yield_task_scx(struct rq *rq) struct task_struct *p = rq->curr; if (SCX_HAS_OP(yield)) - SCX_CALL_OP_2TASKS_RET(SCX_KF_REST, yield, p, NULL); + SCX_CALL_OP_2TASKS_RET(SCX_KF_REST, yield, rq, p, NULL); else p->scx.slice = 0; } @@ -2387,7 +2421,7 @@ static bool yield_to_task_scx(struct rq *rq, struct task_struct *to) struct task_struct *from = rq->curr; if (SCX_HAS_OP(yield)) - return SCX_CALL_OP_2TASKS_RET(SCX_KF_REST, yield, from, to); + return SCX_CALL_OP_2TASKS_RET(SCX_KF_REST, yield, rq, from, to); else return false; } @@ -2945,7 +2979,7 @@ static int balance_one(struct rq *rq, struct task_struct *prev) * emitted in switch_class(). */ if (SCX_HAS_OP(cpu_acquire)) - SCX_CALL_OP(SCX_KF_REST, cpu_acquire, cpu_of(rq), NULL); + SCX_CALL_OP(SCX_KF_REST, cpu_acquire, rq, cpu_of(rq), NULL); rq->scx.cpu_released = false; } @@ -2990,7 +3024,7 @@ static int balance_one(struct rq *rq, struct task_struct *prev) do { dspc->nr_tasks = 0; - SCX_CALL_OP(SCX_KF_DISPATCH, dispatch, cpu_of(rq), + SCX_CALL_OP(SCX_KF_DISPATCH, dispatch, rq, cpu_of(rq), prev_on_scx ? prev : NULL); flush_dispatch_buf(rq); @@ -3104,7 +3138,7 @@ static void set_next_task_scx(struct rq *rq, struct task_struct *p, bool first) * Core-sched might decide to execute @p before it is * dispatched. Call ops_dequeue() to notify the BPF scheduler. */ - ops_dequeue(p, SCX_DEQ_CORE_SCHED_EXEC); + ops_dequeue(rq, p, SCX_DEQ_CORE_SCHED_EXEC); dispatch_dequeue(rq, p); } @@ -3112,7 +3146,7 @@ static void set_next_task_scx(struct rq *rq, struct task_struct *p, bool first) /* see dequeue_task_scx() on why we skip when !QUEUED */ if (SCX_HAS_OP(running) && (p->scx.flags & SCX_TASK_QUEUED)) - SCX_CALL_OP_TASK(SCX_KF_REST, running, p); + SCX_CALL_OP_TASK(SCX_KF_REST, running, rq, p); clr_task_runnable(p, true); @@ -3193,8 +3227,7 @@ static void switch_class(struct rq *rq, struct task_struct *next) .task = next, }; - SCX_CALL_OP(SCX_KF_CPU_RELEASE, - cpu_release, cpu_of(rq), &args); + SCX_CALL_OP(SCX_KF_CPU_RELEASE, cpu_release, rq, cpu_of(rq), &args); } rq->scx.cpu_released = true; } @@ -3207,7 +3240,7 @@ static void put_prev_task_scx(struct rq *rq, struct task_struct *p, /* see dequeue_task_scx() on why we skip when !QUEUED */ if (SCX_HAS_OP(stopping) && (p->scx.flags & SCX_TASK_QUEUED)) - SCX_CALL_OP_TASK(SCX_KF_REST, stopping, p, true); + SCX_CALL_OP_TASK(SCX_KF_REST, stopping, rq, p, true); if (p->scx.flags & SCX_TASK_QUEUED) { set_task_runnable(rq, p); @@ -3348,7 +3381,7 @@ bool scx_prio_less(const struct task_struct *a, const struct task_struct *b, * verifier. */ if (SCX_HAS_OP(core_sched_before) && !scx_rq_bypassing(task_rq(a))) - return SCX_CALL_OP_2TASKS_RET(SCX_KF_REST, core_sched_before, + return SCX_CALL_OP_2TASKS_RET(SCX_KF_REST, core_sched_before, NULL, (struct task_struct *)a, (struct task_struct *)b); else @@ -3385,7 +3418,7 @@ static int select_task_rq_scx(struct task_struct *p, int prev_cpu, int wake_flag *ddsp_taskp = p; cpu = SCX_CALL_OP_TASK_RET(SCX_KF_ENQUEUE | SCX_KF_SELECT_CPU, - select_cpu, p, prev_cpu, wake_flags); + select_cpu, NULL, p, prev_cpu, wake_flags); p->scx.selected_cpu = cpu; *ddsp_taskp = NULL; if (ops_cpu_valid(cpu, "from ops.select_cpu()")) @@ -3430,8 +3463,8 @@ static void set_cpus_allowed_scx(struct task_struct *p, * designation pointless. Cast it away when calling the operation. */ if (SCX_HAS_OP(set_cpumask)) - SCX_CALL_OP_TASK(SCX_KF_REST, set_cpumask, p, - (struct cpumask *)p->cpus_ptr); + SCX_CALL_OP_TASK(SCX_KF_REST, set_cpumask, NULL, + p, (struct cpumask *)p->cpus_ptr); } static void handle_hotplug(struct rq *rq, bool online) @@ -3444,9 +3477,9 @@ static void handle_hotplug(struct rq *rq, bool online) scx_idle_update_selcpu_topology(&scx_ops); if (online && SCX_HAS_OP(cpu_online)) - SCX_CALL_OP(SCX_KF_UNLOCKED, cpu_online, cpu); + SCX_CALL_OP(SCX_KF_UNLOCKED, cpu_online, rq, cpu); else if (!online && SCX_HAS_OP(cpu_offline)) - SCX_CALL_OP(SCX_KF_UNLOCKED, cpu_offline, cpu); + SCX_CALL_OP(SCX_KF_UNLOCKED, cpu_offline, rq, cpu); else scx_ops_exit(SCX_ECODE_ACT_RESTART | SCX_ECODE_RSN_HOTPLUG, "cpu %d going %s, exiting scheduler", cpu, @@ -3550,7 +3583,7 @@ static void task_tick_scx(struct rq *rq, struct task_struct *curr, int queued) curr->scx.slice = 0; touch_core_sched(rq, curr); } else if (SCX_HAS_OP(tick)) { - SCX_CALL_OP_TASK(SCX_KF_REST, tick, curr); + SCX_CALL_OP_TASK(SCX_KF_REST, tick, rq, curr); } if (!curr->scx.slice) @@ -3627,7 +3660,7 @@ static int scx_ops_init_task(struct task_struct *p, struct task_group *tg, bool .fork = fork, }; - ret = SCX_CALL_OP_RET(SCX_KF_UNLOCKED, init_task, p, &args); + ret = SCX_CALL_OP_RET(SCX_KF_UNLOCKED, init_task, NULL, p, &args); if (unlikely(ret)) { ret = ops_sanitize_err("init_task", ret); return ret; @@ -3668,9 +3701,10 @@ static int scx_ops_init_task(struct task_struct *p, struct task_group *tg, bool static void scx_ops_enable_task(struct task_struct *p) { + struct rq *rq = task_rq(p); u32 weight; - lockdep_assert_rq_held(task_rq(p)); + lockdep_assert_rq_held(rq); /* * Set the weight before calling ops.enable() so that the scheduler @@ -3684,20 +3718,22 @@ static void scx_ops_enable_task(struct task_struct *p) p->scx.weight = sched_weight_to_cgroup(weight); if (SCX_HAS_OP(enable)) - SCX_CALL_OP_TASK(SCX_KF_REST, enable, p); + SCX_CALL_OP_TASK(SCX_KF_REST, enable, rq, p); scx_set_task_state(p, SCX_TASK_ENABLED); if (SCX_HAS_OP(set_weight)) - SCX_CALL_OP_TASK(SCX_KF_REST, set_weight, p, p->scx.weight); + SCX_CALL_OP_TASK(SCX_KF_REST, set_weight, rq, p, p->scx.weight); } static void scx_ops_disable_task(struct task_struct *p) { - lockdep_assert_rq_held(task_rq(p)); + struct rq *rq = task_rq(p); + + lockdep_assert_rq_held(rq); WARN_ON_ONCE(scx_get_task_state(p) != SCX_TASK_ENABLED); if (SCX_HAS_OP(disable)) - SCX_CALL_OP_TASK(SCX_KF_REST, disable, p); + SCX_CALL_OP_TASK(SCX_KF_REST, disable, rq, p); scx_set_task_state(p, SCX_TASK_READY); } @@ -3726,7 +3762,7 @@ static void scx_ops_exit_task(struct task_struct *p) } if (SCX_HAS_OP(exit_task)) - SCX_CALL_OP_TASK(SCX_KF_REST, exit_task, p, &args); + SCX_CALL_OP_TASK(SCX_KF_REST, exit_task, task_rq(p), p, &args); scx_set_task_state(p, SCX_TASK_NONE); } @@ -3835,7 +3871,7 @@ static void reweight_task_scx(struct rq *rq, struct task_struct *p, p->scx.weight = sched_weight_to_cgroup(scale_load_down(lw->weight)); if (SCX_HAS_OP(set_weight)) - SCX_CALL_OP_TASK(SCX_KF_REST, set_weight, p, p->scx.weight); + SCX_CALL_OP_TASK(SCX_KF_REST, set_weight, rq, p, p->scx.weight); } static void prio_changed_scx(struct rq *rq, struct task_struct *p, int oldprio) @@ -3851,8 +3887,8 @@ static void switching_to_scx(struct rq *rq, struct task_struct *p) * different scheduler class. Keep the BPF scheduler up-to-date. */ if (SCX_HAS_OP(set_cpumask)) - SCX_CALL_OP_TASK(SCX_KF_REST, set_cpumask, p, - (struct cpumask *)p->cpus_ptr); + SCX_CALL_OP_TASK(SCX_KF_REST, set_cpumask, rq, + p, (struct cpumask *)p->cpus_ptr); } static void switched_from_scx(struct rq *rq, struct task_struct *p) @@ -3913,7 +3949,7 @@ int scx_tg_online(struct task_group *tg) struct scx_cgroup_init_args args = { .weight = tg->scx_weight }; - ret = SCX_CALL_OP_RET(SCX_KF_UNLOCKED, cgroup_init, + ret = SCX_CALL_OP_RET(SCX_KF_UNLOCKED, cgroup_init, NULL, tg->css.cgroup, &args); if (ret) ret = ops_sanitize_err("cgroup_init", ret); @@ -3935,7 +3971,7 @@ void scx_tg_offline(struct task_group *tg) percpu_down_read(&scx_cgroup_rwsem); if (SCX_HAS_OP(cgroup_exit) && (tg->scx_flags & SCX_TG_INITED)) - SCX_CALL_OP(SCX_KF_UNLOCKED, cgroup_exit, tg->css.cgroup); + SCX_CALL_OP(SCX_KF_UNLOCKED, cgroup_exit, NULL, tg->css.cgroup); tg->scx_flags &= ~(SCX_TG_ONLINE | SCX_TG_INITED); percpu_up_read(&scx_cgroup_rwsem); @@ -3968,7 +4004,7 @@ int scx_cgroup_can_attach(struct cgroup_taskset *tset) continue; if (SCX_HAS_OP(cgroup_prep_move)) { - ret = SCX_CALL_OP_RET(SCX_KF_UNLOCKED, cgroup_prep_move, + ret = SCX_CALL_OP_RET(SCX_KF_UNLOCKED, cgroup_prep_move, NULL, p, from, css->cgroup); if (ret) goto err; @@ -3982,8 +4018,8 @@ int scx_cgroup_can_attach(struct cgroup_taskset *tset) err: cgroup_taskset_for_each(p, css, tset) { if (SCX_HAS_OP(cgroup_cancel_move) && p->scx.cgrp_moving_from) - SCX_CALL_OP(SCX_KF_UNLOCKED, cgroup_cancel_move, p, - p->scx.cgrp_moving_from, css->cgroup); + SCX_CALL_OP(SCX_KF_UNLOCKED, cgroup_cancel_move, NULL, + p, p->scx.cgrp_moving_from, css->cgroup); p->scx.cgrp_moving_from = NULL; } @@ -4001,8 +4037,8 @@ void scx_cgroup_move_task(struct task_struct *p) * cgrp_moving_from set. */ if (SCX_HAS_OP(cgroup_move) && !WARN_ON_ONCE(!p->scx.cgrp_moving_from)) - SCX_CALL_OP_TASK(SCX_KF_UNLOCKED, cgroup_move, p, - p->scx.cgrp_moving_from, tg_cgrp(task_group(p))); + SCX_CALL_OP_TASK(SCX_KF_UNLOCKED, cgroup_move, NULL, + p, p->scx.cgrp_moving_from, tg_cgrp(task_group(p))); p->scx.cgrp_moving_from = NULL; } @@ -4021,8 +4057,8 @@ void scx_cgroup_cancel_attach(struct cgroup_taskset *tset) cgroup_taskset_for_each(p, css, tset) { if (SCX_HAS_OP(cgroup_cancel_move) && p->scx.cgrp_moving_from) - SCX_CALL_OP(SCX_KF_UNLOCKED, cgroup_cancel_move, p, - p->scx.cgrp_moving_from, css->cgroup); + SCX_CALL_OP(SCX_KF_UNLOCKED, cgroup_cancel_move, NULL, + p, p->scx.cgrp_moving_from, css->cgroup); p->scx.cgrp_moving_from = NULL; } out_unlock: @@ -4035,7 +4071,7 @@ void scx_group_set_weight(struct task_group *tg, unsigned long weight) if (scx_cgroup_enabled && tg->scx_weight != weight) { if (SCX_HAS_OP(cgroup_set_weight)) - SCX_CALL_OP(SCX_KF_UNLOCKED, cgroup_set_weight, + SCX_CALL_OP(SCX_KF_UNLOCKED, cgroup_set_weight, NULL, tg_cgrp(tg), weight); tg->scx_weight = weight; } @@ -4224,7 +4260,7 @@ static void scx_cgroup_exit(void) continue; rcu_read_unlock(); - SCX_CALL_OP(SCX_KF_UNLOCKED, cgroup_exit, css->cgroup); + SCX_CALL_OP(SCX_KF_UNLOCKED, cgroup_exit, NULL, css->cgroup); rcu_read_lock(); css_put(css); @@ -4261,7 +4297,7 @@ static int scx_cgroup_init(void) continue; rcu_read_unlock(); - ret = SCX_CALL_OP_RET(SCX_KF_UNLOCKED, cgroup_init, + ret = SCX_CALL_OP_RET(SCX_KF_UNLOCKED, cgroup_init, NULL, css->cgroup, &args); if (ret) { css_put(css); @@ -4758,7 +4794,7 @@ static void scx_ops_disable_workfn(struct kthread_work *work) } if (scx_ops.exit) - SCX_CALL_OP(SCX_KF_UNLOCKED, exit, ei); + SCX_CALL_OP(SCX_KF_UNLOCKED, exit, NULL, ei); cancel_delayed_work_sync(&scx_watchdog_work); @@ -4965,7 +5001,7 @@ static void scx_dump_task(struct seq_buf *s, struct scx_dump_ctx *dctx, if (SCX_HAS_OP(dump_task)) { ops_dump_init(s, " "); - SCX_CALL_OP(SCX_KF_REST, dump_task, dctx, p); + SCX_CALL_OP(SCX_KF_REST, dump_task, NULL, dctx, p); ops_dump_exit(); } @@ -5012,7 +5048,7 @@ static void scx_dump_state(struct scx_exit_info *ei, size_t dump_len) if (SCX_HAS_OP(dump)) { ops_dump_init(&s, ""); - SCX_CALL_OP(SCX_KF_UNLOCKED, dump, &dctx); + SCX_CALL_OP(SCX_KF_UNLOCKED, dump, NULL, &dctx); ops_dump_exit(); } @@ -5069,7 +5105,7 @@ static void scx_dump_state(struct scx_exit_info *ei, size_t dump_len) used = seq_buf_used(&ns); if (SCX_HAS_OP(dump_cpu)) { ops_dump_init(&ns, " "); - SCX_CALL_OP(SCX_KF_REST, dump_cpu, &dctx, cpu, idle); + SCX_CALL_OP(SCX_KF_REST, dump_cpu, NULL, &dctx, cpu, idle); ops_dump_exit(); } @@ -5328,7 +5364,7 @@ static int scx_ops_enable(struct sched_ext_ops *ops, struct bpf_link *link) scx_idle_enable(ops); if (scx_ops.init) { - ret = SCX_CALL_OP_RET(SCX_KF_UNLOCKED, init); + ret = SCX_CALL_OP_RET(SCX_KF_UNLOCKED, init, NULL); if (ret) { ret = ops_sanitize_err("init", ret); cpus_read_unlock(); diff --git a/kernel/sched/ext_idle.c b/kernel/sched/ext_idle.c index cb343ca889e0..e67a19a071c1 100644 --- a/kernel/sched/ext_idle.c +++ b/kernel/sched/ext_idle.c @@ -674,7 +674,7 @@ void __scx_update_idle(struct rq *rq, bool idle, bool do_notify) * managed by put_prev_task_idle()/set_next_task_idle(). */ if (SCX_HAS_OP(update_idle) && do_notify && !scx_rq_bypassing(rq)) - SCX_CALL_OP(SCX_KF_REST, update_idle, cpu_of(rq), idle); + SCX_CALL_OP(SCX_KF_REST, update_idle, rq, cpu_of(rq), idle); /* * Update the idle masks: From a11d6784d7316a6c77ca9f14fb1a698ebbb3c1fb Mon Sep 17 00:00:00 2001 From: Andrea Righi Date: Tue, 22 Apr 2025 10:26:33 +0200 Subject: [PATCH 057/555] sched_ext: Fix missing rq lock in scx_bpf_cpuperf_set() scx_bpf_cpuperf_set() can be used to set a performance target level on any CPU. However, it doesn't correctly acquire the corresponding rq lock, which may lead to unsafe behavior and trigger the following warning, due to the lockdep_assert_rq_held() check: [ 51.713737] WARNING: CPU: 3 PID: 3899 at kernel/sched/sched.h:1512 scx_bpf_cpuperf_set+0x1a0/0x1e0 ... [ 51.713836] Call trace: [ 51.713837] scx_bpf_cpuperf_set+0x1a0/0x1e0 (P) [ 51.713839] bpf_prog_62d35beb9301601f_bpfland_init+0x168/0x440 [ 51.713841] bpf__sched_ext_ops_init+0x54/0x8c [ 51.713843] scx_ops_enable.constprop.0+0x2c0/0x10f0 [ 51.713845] bpf_scx_reg+0x18/0x30 [ 51.713847] bpf_struct_ops_link_create+0x154/0x1b0 [ 51.713849] __sys_bpf+0x1934/0x22a0 Fix by properly acquiring the rq lock when possible or raising an error if we try to operate on a CPU that is not the one currently locked. Fixes: d86adb4fc0655 ("sched_ext: Add cpuperf support") Signed-off-by: Andrea Righi Acked-by: Changwoo Min Signed-off-by: Tejun Heo --- kernel/sched/ext.c | 27 +++++++++++++++++++++++---- 1 file changed, 23 insertions(+), 4 deletions(-) diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c index 585bf6d8238b..ac79067dc87e 100644 --- a/kernel/sched/ext.c +++ b/kernel/sched/ext.c @@ -7113,13 +7113,32 @@ __bpf_kfunc void scx_bpf_cpuperf_set(s32 cpu, u32 perf) } if (ops_cpu_valid(cpu, NULL)) { - struct rq *rq = cpu_rq(cpu); + struct rq *rq = cpu_rq(cpu), *locked_rq = scx_locked_rq(); + struct rq_flags rf; + + /* + * When called with an rq lock held, restrict the operation + * to the corresponding CPU to prevent ABBA deadlocks. + */ + if (locked_rq && rq != locked_rq) { + scx_ops_error("Invalid target CPU %d", cpu); + return; + } + + /* + * If no rq lock is held, allow to operate on any CPU by + * acquiring the corresponding rq lock. + */ + if (!locked_rq) { + rq_lock_irqsave(rq, &rf); + update_rq_clock(rq); + } rq->scx.cpuperf_target = perf; + cpufreq_update_util(rq, 0); - rcu_read_lock_sched_notrace(); - cpufreq_update_util(cpu_rq(cpu), 0); - rcu_read_unlock_sched_notrace(); + if (!locked_rq) + rq_unlock_irqrestore(rq, &rf); } } From 8dfa57aabff625bf445548257f7711ef294cd30e Mon Sep 17 00:00:00 2001 From: Vinicius Costa Gomes Date: Mon, 21 Apr 2025 10:03:37 -0700 Subject: [PATCH 058/555] dmaengine: idxd: Fix allowing write() from different address spaces Check if the process submitting the descriptor belongs to the same address space as the one that opened the file, reject otherwise. Fixes: 6827738dc684 ("dmaengine: idxd: add a write() method for applications to submit work") Signed-off-by: Vinicius Costa Gomes Signed-off-by: Dave Jiang Link: https://lore.kernel.org/r/20250421170337.3008875-1-dave.jiang@intel.com Signed-off-by: Vinod Koul --- drivers/dma/idxd/cdev.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/drivers/dma/idxd/cdev.c b/drivers/dma/idxd/cdev.c index 7bd031a60894..a2fb2b847752 100644 --- a/drivers/dma/idxd/cdev.c +++ b/drivers/dma/idxd/cdev.c @@ -407,6 +407,9 @@ static int idxd_cdev_mmap(struct file *filp, struct vm_area_struct *vma) if (!idxd->user_submission_safe && !capable(CAP_SYS_RAWIO)) return -EPERM; + if (current->mm != ctx->mm) + return -EPERM; + rc = check_vma(wq, vma, __func__); if (rc < 0) return rc; @@ -473,6 +476,9 @@ static ssize_t idxd_cdev_write(struct file *filp, const char __user *buf, size_t ssize_t written = 0; int i; + if (current->mm != ctx->mm) + return -EPERM; + for (i = 0; i < len/sizeof(struct dsa_hw_desc); i++) { int rc = idxd_submit_user_descriptor(ctx, udesc + i); @@ -493,6 +499,9 @@ static __poll_t idxd_cdev_poll(struct file *filp, struct idxd_device *idxd = wq->idxd; __poll_t out = 0; + if (current->mm != ctx->mm) + return -EPERM; + poll_wait(filp, &wq->err_queue, wait); spin_lock(&idxd->dev_lock); if (idxd->sw_err.valid) From 305245a2e1d633e5f821178c98c6d6132cea2bdb Mon Sep 17 00:00:00 2001 From: Basavaraj Natikar Date: Mon, 21 Apr 2025 17:12:15 +0530 Subject: [PATCH 059/555] dmaengine: ptdma: Move variable condition check to the first place and remove redundancy The variable is dereferenced without first checking if it's null, leading to the following warning: 'Variable dereferenced before check: desc.' drivers/dma/amd/ptdma/ptdma-dmaengine.c: pt_cmd_callback_work() warn: variable dereferenced before check 'desc' Therefore, move the condition check for the 'desc' variable to the first place and remove the redundant extra condition check. Reported-by: Dan Carpenter Closes: https://lore.kernel.org/all/bfa0a979-ce9f-422d-92c3-34921155d048@stanley.mountain/ Fixes: 656543989457 ("dmaengine: ptdma: Utilize the AE4DMA engine's multi-queue functionality") Signed-off-by: Basavaraj Natikar Link: https://lore.kernel.org/r/20250421114215.1687073-1-Basavaraj.Natikar@amd.com Signed-off-by: Vinod Koul --- drivers/dma/amd/ptdma/ptdma-dmaengine.c | 19 ++++++++++--------- 1 file changed, 10 insertions(+), 9 deletions(-) diff --git a/drivers/dma/amd/ptdma/ptdma-dmaengine.c b/drivers/dma/amd/ptdma/ptdma-dmaengine.c index 715ac3ae067b..81339664036f 100644 --- a/drivers/dma/amd/ptdma/ptdma-dmaengine.c +++ b/drivers/dma/amd/ptdma/ptdma-dmaengine.c @@ -342,6 +342,9 @@ static void pt_cmd_callback_work(void *data, int err) struct pt_dma_chan *chan; unsigned long flags; + if (!desc) + return; + dma_chan = desc->vd.tx.chan; chan = to_pt_chan(dma_chan); @@ -355,16 +358,14 @@ static void pt_cmd_callback_work(void *data, int err) desc->status = DMA_ERROR; spin_lock_irqsave(&chan->vc.lock, flags); - if (desc) { - if (desc->status != DMA_COMPLETE) { - if (desc->status != DMA_ERROR) - desc->status = DMA_COMPLETE; + if (desc->status != DMA_COMPLETE) { + if (desc->status != DMA_ERROR) + desc->status = DMA_COMPLETE; - dma_cookie_complete(tx_desc); - dma_descriptor_unmap(tx_desc); - } else { - tx_desc = NULL; - } + dma_cookie_complete(tx_desc); + dma_descriptor_unmap(tx_desc); + } else { + tx_desc = NULL; } spin_unlock_irqrestore(&chan->vc.lock, flags); From 55cd617566ef14ed607b0b50eff4c0dbd304661b Mon Sep 17 00:00:00 2001 From: Peter Hutterer Date: Mon, 24 Mar 2025 15:36:25 +1000 Subject: [PATCH 060/555] HID: bpf: fix BTN_STYLUS for the XP Pen ACK05 remote Usage_Dig_BarrelSwitch was applied in the UsagePage_Button which incorrectly mapped to BTN_TOOL_PENCIL Link: https://gitlab.freedesktop.org/libevdev/udev-hid-bpf/-/merge_requests/193 Fixes: 834da375 ("bpf: add a v6.11+ compatible BPF fixup for the XPPen ACK05 remote") Link: https://patchwork.kernel.org/project/linux-input/patch/20250207-bpf-import-2025-02-07-v1-7-6048fdd5a206@kernel.org/ Signed-off-by: Peter Hutterer Signed-off-by: Jiri Kosina --- drivers/hid/bpf/progs/XPPen__ACK05.bpf.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/hid/bpf/progs/XPPen__ACK05.bpf.c b/drivers/hid/bpf/progs/XPPen__ACK05.bpf.c index 1a0aeea6a081..a754710fc90b 100644 --- a/drivers/hid/bpf/progs/XPPen__ACK05.bpf.c +++ b/drivers/hid/bpf/progs/XPPen__ACK05.bpf.c @@ -157,6 +157,7 @@ static const __u8 fixed_rdesc_vendor[] = { ReportCount(5) // padding Input(Const) // Byte 4 in report - just exists so we get to be a tablet pad + UsagePage_Digitizers Usage_Dig_BarrelSwitch // BTN_STYLUS ReportCount(1) ReportSize(1) From c14e02e68b43f208417891c5e21308723f03e9e6 Mon Sep 17 00:00:00 2001 From: Aditya Garg Date: Wed, 9 Apr 2025 18:42:51 +0530 Subject: [PATCH 061/555] HID: hid-appletb-kbd: Fix wrong date and kernel version in sysfs interface docs The driver hid-appletb-kbd was upstreamed in kernel 6.15. But, due to an oversight on my part, I didn't change the kernel version and expected date while upstreaming the driver, thus it remained as 6.5, the original kernel version when the driver was developed for downstream. This commit should fix this. Signed-off-by: Aditya Garg Signed-off-by: Jiri Kosina --- Documentation/ABI/testing/sysfs-driver-hid-appletb-kbd | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/Documentation/ABI/testing/sysfs-driver-hid-appletb-kbd b/Documentation/ABI/testing/sysfs-driver-hid-appletb-kbd index 2a19584d091e..8c9718d83e9d 100644 --- a/Documentation/ABI/testing/sysfs-driver-hid-appletb-kbd +++ b/Documentation/ABI/testing/sysfs-driver-hid-appletb-kbd @@ -1,6 +1,6 @@ What: /sys/bus/hid/drivers/hid-appletb-kbd//mode -Date: September, 2023 -KernelVersion: 6.5 +Date: March, 2025 +KernelVersion: 6.15 Contact: linux-input@vger.kernel.org Description: The set of keys displayed on the Touch Bar. From 09d546303b370113323bfff456c4e8cff8756005 Mon Sep 17 00:00:00 2001 From: Qasim Ijaz Date: Thu, 27 Mar 2025 23:11:46 +0000 Subject: [PATCH 062/555] HID: thrustmaster: fix memory leak in thrustmaster_interrupts() In thrustmaster_interrupts(), the allocated send_buf is not freed if the usb_check_int_endpoints() check fails, leading to a memory leak. Fix this by ensuring send_buf is freed before returning in the error path. Fixes: 50420d7c79c3 ("HID: hid-thrustmaster: Fix warning in thrustmaster_probe by adding endpoint check") Signed-off-by: Qasim Ijaz Signed-off-by: Jiri Kosina --- drivers/hid/hid-thrustmaster.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/hid/hid-thrustmaster.c b/drivers/hid/hid-thrustmaster.c index 3b81468a1df2..0bf70664c35e 100644 --- a/drivers/hid/hid-thrustmaster.c +++ b/drivers/hid/hid-thrustmaster.c @@ -174,6 +174,7 @@ static void thrustmaster_interrupts(struct hid_device *hdev) u8 ep_addr[2] = {b_ep, 0}; if (!usb_check_int_endpoints(usbif, ep_addr)) { + kfree(send_buf); hid_err(hdev, "Unexpected non-int endpoint\n"); return; } From e1ca5f39c2e37a3a8cdae005b94c3fc385be4240 Mon Sep 17 00:00:00 2001 From: Qasim Ijaz Date: Sat, 29 Mar 2025 00:20:03 +0000 Subject: [PATCH 063/555] HID: wacom: handle kzalloc() allocation failure in wacom_wac_queue_flush() During wacom_wac_queue_flush() the code calls kzalloc() to allocate a zero initialised buffer which it uses as a storage buffer to get data from the fifo via kfifo_out(). However it does not check kzalloc() for allocation failure which returns NULL and could potentially lead to a NULL deref. Fix this by checking for kzalloc() failure and skipping the current entry if allocation failure occurs. Fixes: 5e013ad20689 ("HID: wacom: Remove static WACOM_PKGLEN_MAX limit") Signed-off-by: Qasim Ijaz Reviewed-by: Jason Gerecke Signed-off-by: Jiri Kosina --- drivers/hid/wacom_sys.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/hid/wacom_sys.c b/drivers/hid/wacom_sys.c index 97393a3083ca..43d892810c9e 100644 --- a/drivers/hid/wacom_sys.c +++ b/drivers/hid/wacom_sys.c @@ -70,10 +70,16 @@ static void wacom_wac_queue_flush(struct hid_device *hdev, { while (!kfifo_is_empty(fifo)) { int size = kfifo_peek_len(fifo); - u8 *buf = kzalloc(size, GFP_KERNEL); + u8 *buf; unsigned int count; int err; + buf = kzalloc(size, GFP_KERNEL); + if (!buf) { + kfifo_skip(fifo); + continue; + } + count = kfifo_out(fifo, buf, size); if (count != size) { // Hard to say what is the "right" action in this From fd34bf79a617f6298b13b274dc255f192a987e2a Mon Sep 17 00:00:00 2001 From: Qasim Ijaz Date: Mon, 14 Apr 2025 19:33:09 +0100 Subject: [PATCH 064/555] HID: wacom: fix memory leak on size mismatch in wacom_wac_queue_flush() In wacom_wac_queue_flush() the code allocates zero initialised buffer which it uses as a storage buffer for copying data from a fifo via kfifo_out(). The kfifo_out() function returns the number of elements it has copied. The code checks if the number of copied elements does not equal the size of the fifo record, if it does not it simply skips the entry and continues to the next iteration. However it does not release the storage buffer leading to a memory leak. Fix the memory leak by freeing the buffer on size mismatch. Fixes: 5e013ad20689 ("HID: wacom: Remove static WACOM_PKGLEN_MAX limit") Reviewed-by: Jason Gerecke Signed-off-by: Qasim Ijaz Signed-off-by: Jiri Kosina --- drivers/hid/wacom_sys.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/hid/wacom_sys.c b/drivers/hid/wacom_sys.c index 43d892810c9e..95a796b3e9f2 100644 --- a/drivers/hid/wacom_sys.c +++ b/drivers/hid/wacom_sys.c @@ -87,6 +87,7 @@ static void wacom_wac_queue_flush(struct hid_device *hdev, // to flush seems reasonable enough, however. hid_warn(hdev, "%s: removed fifo entry with unexpected size\n", __func__); + kfree(buf); continue; } err = hid_report_raw_event(hdev, HID_INPUT_REPORT, buf, size, false); From bd07f751208ba190f9b0db5e5b7f35d5bb4a8a1e Mon Sep 17 00:00:00 2001 From: Henry Martin Date: Tue, 1 Apr 2025 17:48:53 +0800 Subject: [PATCH 065/555] HID: uclogic: Add NULL check in uclogic_input_configured() devm_kasprintf() returns NULL when memory allocation fails. Currently, uclogic_input_configured() does not check for this case, which results in a NULL pointer dereference. Add NULL check after devm_kasprintf() to prevent this issue. Fixes: dd613a4e45f8 ("HID: uclogic: Correct devm device reference for hidinput input_dev name") Signed-off-by: Henry Martin Signed-off-by: Jiri Kosina --- drivers/hid/hid-uclogic-core.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/hid/hid-uclogic-core.c b/drivers/hid/hid-uclogic-core.c index d8008933c052..321c43fb06ae 100644 --- a/drivers/hid/hid-uclogic-core.c +++ b/drivers/hid/hid-uclogic-core.c @@ -142,11 +142,12 @@ static int uclogic_input_configured(struct hid_device *hdev, suffix = "System Control"; break; } - } - - if (suffix) + } else { hi->input->name = devm_kasprintf(&hdev->dev, GFP_KERNEL, "%s %s", hdev->name, suffix); + if (!hi->input->name) + return -ENOMEM; + } return 0; } From 6bf8ab7774a20e1e60030e20f42ac8cc804fa457 Mon Sep 17 00:00:00 2001 From: Qasim Ijaz Date: Mon, 14 Apr 2025 19:33:17 +0100 Subject: [PATCH 066/555] HID: wacom: fix shift OOB in kfifo allocation for zero pktlen During wacom_parse_and_register() the code calls wacom_devm_kfifo_alloc to allocate a fifo. During this operation it passes kfifo_alloc a fifo_size of 0. Kfifo attempts to round the size passed to it to the next power of 2 via roundup_pow_of_two (queue-type data structures do this to maintain efficiency of operations). However during this phase a problem arises when the roundup_pow_of_two() function utilises a shift exponent of fls_long(n-1), where n is the fifo_size. Since n is 0 in this case and n is also an unsigned long, doing n-1 causes unsigned integer wrap-around to occur making the fifo_size 4294967295. So the code effectively does fls_long(4294967295) which results in 64. Returning back to roundup_pow_of_two(), the code utilises a shift exponent of 64. When a shift exponent of 64 is used on a 64-bit type such as 1UL it results in a shift-out-of-bounds. The root cause of the issue seems to stem from insufficient validation of wacom_compute_pktlen(), since in this case the fifo_size comes from wacom_wac->features.pktlen. During wacom_parse_and_register() the wacom_compute_pktlen() function sets the pktlen as 0. To fix this, we should handle cases where wacom_compute_pktlen() results in 0. Reported-by: syzbot Closes: https://syzkaller.appspot.com/bug?extid=d5204cbbdd921f1f7cad Fixes: 5e013ad20689 ("HID: wacom: Remove static WACOM_PKGLEN_MAX limit") Tested-by: Qasim Ijaz Reviewed-by: Jason Gerecke Cc: stable@vger.kernel.org Signed-off-by: Qasim Ijaz Signed-off-by: Jiri Kosina --- drivers/hid/wacom_sys.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/hid/wacom_sys.c b/drivers/hid/wacom_sys.c index 95a796b3e9f2..2d052653ba63 100644 --- a/drivers/hid/wacom_sys.c +++ b/drivers/hid/wacom_sys.c @@ -2368,6 +2368,8 @@ static int wacom_parse_and_register(struct wacom *wacom, bool wireless) unsigned int connect_mask = HID_CONNECT_HIDRAW; features->pktlen = wacom_compute_pktlen(hdev); + if (!features->pktlen) + return -ENODEV; if (!devres_open_group(&hdev->dev, wacom, GFP_KERNEL)) return -ENOMEM; From 0cc2effbc8f522af6b9d871cd27678e6aed9d56c Mon Sep 17 00:00:00 2001 From: Mario Limonciello Date: Mon, 21 Apr 2025 16:32:09 -0500 Subject: [PATCH 067/555] HID: amd_sfh: Fix SRA sensor when it's the only sensor On systems that only have an SRA sensor connected to SFH the sensor doesn't get enabled due to a bad optimization condition of breaking the sensor walk loop. This optimization is unnecessary in the first place because if there is only one device then the loop only runs once. Drop the condition and explicitly mark sensor as enabled. Reported-by: Yijun Shen Tested-By: Yijun Shen Fixes: d1c444b47100d ("HID: amd_sfh: Add support to export device operating states") Cc: stable@vger.kernel.org Signed-off-by: Mario Limonciello Acked-by: Basavaraj Natikar Signed-off-by: Jiri Kosina --- drivers/hid/amd-sfh-hid/sfh1_1/amd_sfh_init.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/drivers/hid/amd-sfh-hid/sfh1_1/amd_sfh_init.c b/drivers/hid/amd-sfh-hid/sfh1_1/amd_sfh_init.c index 25f0ebfcbd5f..c1bdf1e0d44a 100644 --- a/drivers/hid/amd-sfh-hid/sfh1_1/amd_sfh_init.c +++ b/drivers/hid/amd-sfh-hid/sfh1_1/amd_sfh_init.c @@ -134,9 +134,6 @@ static int amd_sfh1_1_hid_client_init(struct amd_mp2_dev *privdata) for (i = 0; i < cl_data->num_hid_devices; i++) { cl_data->sensor_sts[i] = SENSOR_DISABLED; - if (cl_data->num_hid_devices == 1 && cl_data->sensor_idx[0] == SRA_IDX) - break; - if (cl_data->sensor_idx[i] == SRA_IDX) { info.sensor_idx = cl_data->sensor_idx[i]; writel(0, privdata->mmio + amd_get_p2c_val(privdata, 0)); @@ -145,8 +142,10 @@ static int amd_sfh1_1_hid_client_init(struct amd_mp2_dev *privdata) (privdata, cl_data->sensor_idx[i], ENABLE_SENSOR); cl_data->sensor_sts[i] = (status == 0) ? SENSOR_ENABLED : SENSOR_DISABLED; - if (cl_data->sensor_sts[i] == SENSOR_ENABLED) + if (cl_data->sensor_sts[i] == SENSOR_ENABLED) { + cl_data->is_any_sensor_enabled = true; privdata->dev_en.is_sra_present = true; + } continue; } From f32e8c8095490152b5bc5f467d5034387a4bbd1b Mon Sep 17 00:00:00 2001 From: Mario Limonciello Date: Mon, 21 Apr 2025 16:32:10 -0500 Subject: [PATCH 068/555] HID: amd_sfh: Avoid clearing reports for SRA sensor SRA sensor doesn't allocate any memory for reports. Skip trying to clear memory for that sensor in cleanup path. Suggested-by: Basavaraj Natikar Signed-off-by: Mario Limonciello Acked-by: Basavaraj Natikar Signed-off-by: Jiri Kosina --- drivers/hid/amd-sfh-hid/sfh1_1/amd_sfh_init.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/hid/amd-sfh-hid/sfh1_1/amd_sfh_init.c b/drivers/hid/amd-sfh-hid/sfh1_1/amd_sfh_init.c index c1bdf1e0d44a..0a9b44ce4904 100644 --- a/drivers/hid/amd-sfh-hid/sfh1_1/amd_sfh_init.c +++ b/drivers/hid/amd-sfh-hid/sfh1_1/amd_sfh_init.c @@ -83,6 +83,9 @@ static int amd_sfh_hid_client_deinit(struct amd_mp2_dev *privdata) case ALS_IDX: privdata->dev_en.is_als_present = false; break; + case SRA_IDX: + privdata->dev_en.is_sra_present = false; + break; } if (cl_data->sensor_sts[i] == SENSOR_ENABLED) { @@ -237,6 +240,8 @@ static int amd_sfh1_1_hid_client_init(struct amd_mp2_dev *privdata) cleanup: amd_sfh_hid_client_deinit(privdata); for (i = 0; i < cl_data->num_hid_devices; i++) { + if (cl_data->sensor_idx[i] == SRA_IDX) + continue; devm_kfree(dev, cl_data->feature_report[i]); devm_kfree(dev, in_data->input_report[i]); devm_kfree(dev, cl_data->report_descr[i]); From acae9d5b51cf8d4da87ed13140e3de4970669213 Mon Sep 17 00:00:00 2001 From: Jiapeng Chong Date: Wed, 16 Apr 2025 10:58:03 +0800 Subject: [PATCH 069/555] HID: hid-steam: Remove the unused variable connected MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Variable connected is not effectively used, so delete it. drivers/hid/hid-steam.c:1153:7: warning: variable ‘connected’ set but not used. Reported-by: Abaci Robot Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=20462 Signed-off-by: Jiapeng Chong Signed-off-by: Jiri Kosina --- drivers/hid/hid-steam.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/hid/hid-steam.c b/drivers/hid/hid-steam.c index dfd9d22ed559..949d307c66a8 100644 --- a/drivers/hid/hid-steam.c +++ b/drivers/hid/hid-steam.c @@ -1150,11 +1150,9 @@ static void steam_client_ll_close(struct hid_device *hdev) struct steam_device *steam = hdev->driver_data; unsigned long flags; - bool connected; spin_lock_irqsave(&steam->lock, flags); steam->client_opened--; - connected = steam->connected && !steam->client_opened; spin_unlock_irqrestore(&steam->lock, flags); schedule_work(&steam->unregister_work); From fa9fdeea1b7d6440c22efa6d59a769eae8bc89f1 Mon Sep 17 00:00:00 2001 From: Milton Barrera Date: Wed, 9 Apr 2025 00:04:28 -0600 Subject: [PATCH 070/555] HID: quirks: Add ADATA XPG alpha wireless mouse support This patch adds HID_QUIRK_ALWAYS_POLL for the ADATA XPG wireless gaming mouse (USB ID 125f:7505) and its USB dongle (USB ID 125f:7506). Without this quirk, the device does not generate input events properly. Signed-off-by: Milton Barrera Signed-off-by: Jiri Kosina --- drivers/hid/hid-ids.h | 4 ++++ drivers/hid/hid-quirks.c | 2 ++ 2 files changed, 6 insertions(+) diff --git a/drivers/hid/hid-ids.h b/drivers/hid/hid-ids.h index 288a2b864cc4..1062731315a2 100644 --- a/drivers/hid/hid-ids.h +++ b/drivers/hid/hid-ids.h @@ -41,6 +41,10 @@ #define USB_VENDOR_ID_ACTIONSTAR 0x2101 #define USB_DEVICE_ID_ACTIONSTAR_1011 0x1011 +#define USB_VENDOR_ID_ADATA_XPG 0x125f +#define USB_VENDOR_ID_ADATA_XPG_WL_GAMING_MOUSE 0x7505 +#define USB_VENDOR_ID_ADATA_XPG_WL_GAMING_MOUSE_DONGLE 0x7506 + #define USB_VENDOR_ID_ADS_TECH 0x06e1 #define USB_DEVICE_ID_ADS_TECH_RADIO_SI470X 0xa155 diff --git a/drivers/hid/hid-quirks.c b/drivers/hid/hid-quirks.c index 646171598e41..0731473cc9b1 100644 --- a/drivers/hid/hid-quirks.c +++ b/drivers/hid/hid-quirks.c @@ -27,6 +27,8 @@ static const struct hid_device_id hid_quirks[] = { { HID_USB_DEVICE(USB_VENDOR_ID_AASHIMA, USB_DEVICE_ID_AASHIMA_GAMEPAD), HID_QUIRK_BADPAD }, { HID_USB_DEVICE(USB_VENDOR_ID_AASHIMA, USB_DEVICE_ID_AASHIMA_PREDATOR), HID_QUIRK_BADPAD }, + { HID_USB_DEVICE(USB_VENDOR_ID_ADATA_XPG, USB_VENDOR_ID_ADATA_XPG_WL_GAMING_MOUSE), HID_QUIRK_ALWAYS_POLL }, + { HID_USB_DEVICE(USB_VENDOR_ID_ADATA_XPG, USB_VENDOR_ID_ADATA_XPG_WL_GAMING_MOUSE_DONGLE), HID_QUIRK_ALWAYS_POLL }, { HID_USB_DEVICE(USB_VENDOR_ID_AFATECH, USB_DEVICE_ID_AFATECH_AF9016), HID_QUIRK_FULLSPEED_INTERVAL }, { HID_USB_DEVICE(USB_VENDOR_ID_AIREN, USB_DEVICE_ID_AIREN_SLIMPLUS), HID_QUIRK_NOGET }, { HID_USB_DEVICE(USB_VENDOR_ID_AKAI_09E8, USB_DEVICE_ID_AKAI_09E8_MIDIMIX), HID_QUIRK_NO_INIT_REPORTS }, From a2620f8932fa9fdabc3d78ed6efb004ca409019f Mon Sep 17 00:00:00 2001 From: Mikhail Lobanov Date: Mon, 14 Apr 2025 20:12:06 +0300 Subject: [PATCH 071/555] KVM: SVM: Forcibly leave SMM mode on SHUTDOWN interception Previously, commit ed129ec9057f ("KVM: x86: forcibly leave nested mode on vCPU reset") addressed an issue where a triple fault occurring in nested mode could lead to use-after-free scenarios. However, the commit did not handle the analogous situation for System Management Mode (SMM). This omission results in triggering a WARN when KVM forces a vCPU INIT after SHUTDOWN interception while the vCPU is in SMM. This situation was reprodused using Syzkaller by: 1) Creating a KVM VM and vCPU 2) Sending a KVM_SMI ioctl to explicitly enter SMM 3) Executing invalid instructions causing consecutive exceptions and eventually a triple fault The issue manifests as follows: WARNING: CPU: 0 PID: 25506 at arch/x86/kvm/x86.c:12112 kvm_vcpu_reset+0x1d2/0x1530 arch/x86/kvm/x86.c:12112 Modules linked in: CPU: 0 PID: 25506 Comm: syz-executor.0 Not tainted 6.1.130-syzkaller-00157-g164fe5dde9b6 #0 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.12.0-1 04/01/2014 RIP: 0010:kvm_vcpu_reset+0x1d2/0x1530 arch/x86/kvm/x86.c:12112 Call Trace: shutdown_interception+0x66/0xb0 arch/x86/kvm/svm/svm.c:2136 svm_invoke_exit_handler+0x110/0x530 arch/x86/kvm/svm/svm.c:3395 svm_handle_exit+0x424/0x920 arch/x86/kvm/svm/svm.c:3457 vcpu_enter_guest arch/x86/kvm/x86.c:10959 [inline] vcpu_run+0x2c43/0x5a90 arch/x86/kvm/x86.c:11062 kvm_arch_vcpu_ioctl_run+0x50f/0x1cf0 arch/x86/kvm/x86.c:11283 kvm_vcpu_ioctl+0x570/0xf00 arch/x86/kvm/../../../virt/kvm/kvm_main.c:4122 vfs_ioctl fs/ioctl.c:51 [inline] __do_sys_ioctl fs/ioctl.c:870 [inline] __se_sys_ioctl fs/ioctl.c:856 [inline] __x64_sys_ioctl+0x19a/0x210 fs/ioctl.c:856 do_syscall_x64 arch/x86/entry/common.c:51 [inline] do_syscall_64+0x35/0x80 arch/x86/entry/common.c:81 entry_SYSCALL_64_after_hwframe+0x6e/0xd8 Architecturally, INIT is blocked when the CPU is in SMM, hence KVM's WARN() in kvm_vcpu_reset() to guard against KVM bugs, e.g. to detect improper emulation of INIT. SHUTDOWN on SVM is a weird edge case where KVM needs to do _something_ sane with the VMCB, since it's technically undefined, and INIT is the least awful choice given KVM's ABI. So, double down on stuffing INIT on SHUTDOWN, and force the vCPU out of SMM to avoid any weirdness (and the WARN). Found by Linux Verification Center (linuxtesting.org) with Syzkaller. Fixes: ed129ec9057f ("KVM: x86: forcibly leave nested mode on vCPU reset") Cc: stable@vger.kernel.org Suggested-by: Sean Christopherson Signed-off-by: Mikhail Lobanov Link: https://lore.kernel.org/r/20250414171207.155121-1-m.lobanov@rosa.ru [sean: massage changelog, make it clear this isn't architectural behavior] Signed-off-by: Sean Christopherson --- arch/x86/kvm/smm.c | 1 + arch/x86/kvm/svm/svm.c | 4 ++++ 2 files changed, 5 insertions(+) diff --git a/arch/x86/kvm/smm.c b/arch/x86/kvm/smm.c index 699e551ec93b..9864c057187d 100644 --- a/arch/x86/kvm/smm.c +++ b/arch/x86/kvm/smm.c @@ -131,6 +131,7 @@ void kvm_smm_changed(struct kvm_vcpu *vcpu, bool entering_smm) kvm_mmu_reset_context(vcpu); } +EXPORT_SYMBOL_GPL(kvm_smm_changed); void process_smi(struct kvm_vcpu *vcpu) { diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c index d5d0c5c3300b..c5470d842aed 100644 --- a/arch/x86/kvm/svm/svm.c +++ b/arch/x86/kvm/svm/svm.c @@ -2231,6 +2231,10 @@ static int shutdown_interception(struct kvm_vcpu *vcpu) */ if (!sev_es_guest(vcpu->kvm)) { clear_page(svm->vmcb); +#ifdef CONFIG_KVM_SMM + if (is_smm(vcpu)) + kvm_smm_changed(vcpu, false); +#endif kvm_vcpu_reset(vcpu, true); } From a476cadf8ef1fbb9780581316f0199dfc62a81f2 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Mon, 24 Mar 2025 13:51:28 +0300 Subject: [PATCH 072/555] KVM: x86: Check that the high 32bits are clear in kvm_arch_vcpu_ioctl_run() The "kvm_run->kvm_valid_regs" and "kvm_run->kvm_dirty_regs" variables are u64 type. We are only using the lowest 3 bits but we want to ensure that the users are not passing invalid bits so that we can use the remaining bits in the future. However "sync_valid_fields" and kvm_sync_valid_fields() are u32 type so the check only ensures that the lower 32 bits are clear. Fix this by changing the types to u64. Fixes: 74c1807f6c4f ("KVM: x86: block KVM_CAP_SYNC_REGS if guest state is protected") Signed-off-by: Dan Carpenter Link: https://lore.kernel.org/r/ec25aad1-113e-4c6e-8941-43d432251398@stanley.mountain Signed-off-by: Sean Christopherson --- arch/x86/kvm/x86.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index df5b99ea1f18..9896fd574bfc 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -4597,7 +4597,7 @@ static bool kvm_is_vm_type_supported(unsigned long type) return type < 32 && (kvm_caps.supported_vm_types & BIT(type)); } -static inline u32 kvm_sync_valid_fields(struct kvm *kvm) +static inline u64 kvm_sync_valid_fields(struct kvm *kvm) { return kvm && kvm->arch.has_protected_state ? 0 : KVM_SYNC_X86_VALID_FIELDS; } @@ -11493,7 +11493,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu) { struct kvm_queued_exception *ex = &vcpu->arch.exception; struct kvm_run *kvm_run = vcpu->run; - u32 sync_valid_fields; + u64 sync_valid_fields; int r; r = kvm_mmu_post_init_vm(vcpu->kvm); From 98698ca0e58734bc5c1c24e5bbc7429f981cd186 Mon Sep 17 00:00:00 2001 From: Dave Stevenson Date: Wed, 23 Apr 2025 11:47:15 +0100 Subject: [PATCH 073/555] staging: bcm2835-camera: Initialise dev in v4l2_dev Commit 42a2f6664e18 ("staging: vc04_services: Move global g_state to vchiq_state") changed mmal_init to pass dev->v4l2_dev.dev to vchiq_mmal_init, however nothing iniitialised dev->v4l2_dev, so we got a NULL pointer dereference. Set dev->v4l2_dev.dev during bcm2835_mmal_probe. The device pointer could be passed into v4l2_device_register to set it, however that also has other effects that would need additional changes. Fixes: 42a2f6664e18 ("staging: vc04_services: Move global g_state to vchiq_state") Cc: stable@vger.kernel.org Signed-off-by: Dave Stevenson Reviewed-by: Stefan Wahren Link: https://lore.kernel.org/r/20250423-staging-bcm2835-v4l2-fix-v2-1-3227f0ba4700@raspberrypi.com Signed-off-by: Greg Kroah-Hartman --- drivers/staging/vc04_services/bcm2835-camera/bcm2835-camera.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/staging/vc04_services/bcm2835-camera/bcm2835-camera.c b/drivers/staging/vc04_services/bcm2835-camera/bcm2835-camera.c index b839b50ac26a..fa7ea4ca4c36 100644 --- a/drivers/staging/vc04_services/bcm2835-camera/bcm2835-camera.c +++ b/drivers/staging/vc04_services/bcm2835-camera/bcm2835-camera.c @@ -1900,6 +1900,7 @@ static int bcm2835_mmal_probe(struct vchiq_device *device) __func__, ret); goto free_dev; } + dev->v4l2_dev.dev = &device->dev; /* setup v4l controls */ ret = bcm2835_mmal_init_controls(dev, &dev->ctrl_handler); From 2ca34b508774aaa590fc3698a54204706ecca4ba Mon Sep 17 00:00:00 2001 From: Gabriel Shahrouzi Date: Fri, 18 Apr 2025 21:29:37 -0400 Subject: [PATCH 074/555] staging: axis-fifo: Correct handling of tx_fifo_depth for size validation Remove erroneous subtraction of 4 from the total FIFO depth read from device tree. The stored depth is for checking against total capacity, not initial vacancy. This prevented writes near the FIFO's full size. The check performed just before data transfer, which uses live reads of the TDFV register to determine current vacancy, correctly handles the initial Depth - 4 hardware state and subsequent FIFO fullness. Fixes: 4a965c5f89de ("staging: add driver for Xilinx AXI-Stream FIFO v4.1 IP core") Cc: stable@vger.kernel.org Signed-off-by: Gabriel Shahrouzi Link: https://lore.kernel.org/r/20250419012937.674924-1-gshahrouzi@gmail.com Signed-off-by: Greg Kroah-Hartman --- drivers/staging/axis-fifo/axis-fifo.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/drivers/staging/axis-fifo/axis-fifo.c b/drivers/staging/axis-fifo/axis-fifo.c index 7540c20090c7..04b3dc3cfe7c 100644 --- a/drivers/staging/axis-fifo/axis-fifo.c +++ b/drivers/staging/axis-fifo/axis-fifo.c @@ -775,9 +775,6 @@ static int axis_fifo_parse_dt(struct axis_fifo *fifo) goto end; } - /* IP sets TDFV to fifo depth - 4 so we will do the same */ - fifo->tx_fifo_depth -= 4; - ret = get_dts_property(fifo, "xlnx,use-rx-data", &fifo->has_rx_fifo); if (ret) { dev_err(fifo->dt_device, "missing xlnx,use-rx-data property\n"); From c6e8d85fafa7193613db37da29c0e8d6e2515b13 Mon Sep 17 00:00:00 2001 From: Gabriel Shahrouzi Date: Fri, 18 Apr 2025 20:43:06 -0400 Subject: [PATCH 075/555] staging: axis-fifo: Remove hardware resets for user errors The axis-fifo driver performs a full hardware reset (via reset_ip_core()) in several error paths within the read and write functions. This reset flushes both TX and RX FIFOs and resets the AXI-Stream links. Allow the user to handle the error without causing hardware disruption or data loss in other FIFO paths. Fixes: 4a965c5f89de ("staging: add driver for Xilinx AXI-Stream FIFO v4.1 IP core") Cc: stable@vger.kernel.org Signed-off-by: Gabriel Shahrouzi Link: https://lore.kernel.org/r/20250419004306.669605-1-gshahrouzi@gmail.com Signed-off-by: Greg Kroah-Hartman --- drivers/staging/axis-fifo/axis-fifo.c | 11 +++-------- 1 file changed, 3 insertions(+), 8 deletions(-) diff --git a/drivers/staging/axis-fifo/axis-fifo.c b/drivers/staging/axis-fifo/axis-fifo.c index 04b3dc3cfe7c..351f983ef914 100644 --- a/drivers/staging/axis-fifo/axis-fifo.c +++ b/drivers/staging/axis-fifo/axis-fifo.c @@ -393,16 +393,14 @@ static ssize_t axis_fifo_read(struct file *f, char __user *buf, bytes_available = ioread32(fifo->base_addr + XLLF_RLR_OFFSET); if (!bytes_available) { - dev_err(fifo->dt_device, "received a packet of length 0 - fifo core will be reset\n"); - reset_ip_core(fifo); + dev_err(fifo->dt_device, "received a packet of length 0\n"); ret = -EIO; goto end_unlock; } if (bytes_available > len) { - dev_err(fifo->dt_device, "user read buffer too small (available bytes=%zu user buffer bytes=%zu) - fifo core will be reset\n", + dev_err(fifo->dt_device, "user read buffer too small (available bytes=%zu user buffer bytes=%zu)\n", bytes_available, len); - reset_ip_core(fifo); ret = -EINVAL; goto end_unlock; } @@ -411,8 +409,7 @@ static ssize_t axis_fifo_read(struct file *f, char __user *buf, /* this probably can't happen unless IP * registers were previously mishandled */ - dev_err(fifo->dt_device, "received a packet that isn't word-aligned - fifo core will be reset\n"); - reset_ip_core(fifo); + dev_err(fifo->dt_device, "received a packet that isn't word-aligned\n"); ret = -EIO; goto end_unlock; } @@ -433,7 +430,6 @@ static ssize_t axis_fifo_read(struct file *f, char __user *buf, if (copy_to_user(buf + copied * sizeof(u32), tmp_buf, copy * sizeof(u32))) { - reset_ip_core(fifo); ret = -EFAULT; goto end_unlock; } @@ -542,7 +538,6 @@ static ssize_t axis_fifo_write(struct file *f, const char __user *buf, if (copy_from_user(tmp_buf, buf + copied * sizeof(u32), copy * sizeof(u32))) { - reset_ip_core(fifo); ret = -EFAULT; goto end_unlock; } From e7dcd1304b9ac08c428c84fdb96ce6a04ff2114f Mon Sep 17 00:00:00 2001 From: Andrea Righi Date: Fri, 25 Apr 2025 19:57:02 +0200 Subject: [PATCH 076/555] sched_ext: Remove duplicate BTF_ID_FLAGS definitions Some kfuncs specific to the idle CPU selection policy are registered in both the scx_kfunc_ids_any and scx_kfunc_ids_idle blocks, even though they should only be defined in the latter. Remove the duplicates from scx_kfunc_ids_any. Fixes: 337d1b354a297 ("sched_ext: Move built-in idle CPU selection policy to a separate file") Signed-off-by: Andrea Righi Signed-off-by: Tejun Heo --- kernel/sched/ext.c | 6 ------ 1 file changed, 6 deletions(-) diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c index ac79067dc87e..dddb0af36f8d 100644 --- a/kernel/sched/ext.c +++ b/kernel/sched/ext.c @@ -7369,12 +7369,6 @@ BTF_ID_FLAGS(func, scx_bpf_nr_cpu_ids) BTF_ID_FLAGS(func, scx_bpf_get_possible_cpumask, KF_ACQUIRE) BTF_ID_FLAGS(func, scx_bpf_get_online_cpumask, KF_ACQUIRE) BTF_ID_FLAGS(func, scx_bpf_put_cpumask, KF_RELEASE) -BTF_ID_FLAGS(func, scx_bpf_get_idle_cpumask, KF_ACQUIRE) -BTF_ID_FLAGS(func, scx_bpf_get_idle_smtmask, KF_ACQUIRE) -BTF_ID_FLAGS(func, scx_bpf_put_idle_cpumask, KF_RELEASE) -BTF_ID_FLAGS(func, scx_bpf_test_and_clear_cpu_idle) -BTF_ID_FLAGS(func, scx_bpf_pick_idle_cpu, KF_RCU) -BTF_ID_FLAGS(func, scx_bpf_pick_any_cpu, KF_RCU) BTF_ID_FLAGS(func, scx_bpf_task_running, KF_RCU) BTF_ID_FLAGS(func, scx_bpf_task_cpu, KF_RCU) BTF_ID_FLAGS(func, scx_bpf_cpu_rq) From 7ec0987da2c903d58167573e58b39d7ed19fc627 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Thu, 24 Apr 2025 10:47:29 +0200 Subject: [PATCH 077/555] arm64: dts: rockchip: Align wifi node name with bindings in CB2 Since commit 3c3606793f7e ("dt-bindings: wireless: bcm4329-fmac: Use wireless-controller.yaml schema"), bindings expect 'wifi' as node name: rk3566-bigtreetech-cb2-manta.dtb: sdio-wifi@1: $nodename:0: 'sdio-wifi@1' does not match '^wifi(@.*)?$' Signed-off-by: Krzysztof Kozlowski Link: https://lore.kernel.org/r/20250424084729.105182-1-krzysztof.kozlowski@linaro.org Signed-off-by: Heiko Stuebner --- arch/arm64/boot/dts/rockchip/rk3566-bigtreetech-cb2.dtsi | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm64/boot/dts/rockchip/rk3566-bigtreetech-cb2.dtsi b/arch/arm64/boot/dts/rockchip/rk3566-bigtreetech-cb2.dtsi index a48351471764..e7ba477e75f9 100644 --- a/arch/arm64/boot/dts/rockchip/rk3566-bigtreetech-cb2.dtsi +++ b/arch/arm64/boot/dts/rockchip/rk3566-bigtreetech-cb2.dtsi @@ -775,7 +775,7 @@ rockchip,default-sample-phase = <90>; status = "okay"; - sdio-wifi@1 { + wifi@1 { compatible = "brcm,bcm4329-fmac"; reg = <1>; interrupt-parent = <&gpio2>; From 5e6a4ee9799b202fefa8c6264647971f892f0264 Mon Sep 17 00:00:00 2001 From: Tom Vincent Date: Thu, 17 Apr 2025 09:17:53 +0100 Subject: [PATCH 078/555] arm64: dts: rockchip: Assign RT5616 MCLK rate on rk3588-friendlyelec-cm3588 The Realtek RT5616 audio codec on the FriendlyElec CM3588 module fails to probe correctly due to the missing clock properties. This results in distorted analogue audio output. Assign MCLK to 12.288 MHz, which allows the codec to advertise most of the standard sample rates per other RK3588 devices. Fixes: e23819cf273c ("arm64: dts: rockchip: Add FriendlyElec CM3588 NAS board") Signed-off-by: Tom Vincent Link: https://lore.kernel.org/r/20250417081753.644950-1-linux@tlvince.com Signed-off-by: Heiko Stuebner --- arch/arm64/boot/dts/rockchip/rk3588-friendlyelec-cm3588.dtsi | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/arch/arm64/boot/dts/rockchip/rk3588-friendlyelec-cm3588.dtsi b/arch/arm64/boot/dts/rockchip/rk3588-friendlyelec-cm3588.dtsi index 1af0a30866f6..af431fdcbea7 100644 --- a/arch/arm64/boot/dts/rockchip/rk3588-friendlyelec-cm3588.dtsi +++ b/arch/arm64/boot/dts/rockchip/rk3588-friendlyelec-cm3588.dtsi @@ -222,6 +222,10 @@ compatible = "realtek,rt5616"; reg = <0x1b>; #sound-dai-cells = <0>; + assigned-clocks = <&cru I2S0_8CH_MCLKOUT>; + assigned-clock-rates = <12288000>; + clocks = <&cru I2S0_8CH_MCLKOUT>; + clock-names = "mclk"; }; }; From 56651128e2fbad80f632f388d6bf1f39c928267a Mon Sep 17 00:00:00 2001 From: Marco Crivellari Date: Thu, 3 Apr 2025 18:11:42 +0200 Subject: [PATCH 079/555] MIPS: Fix idle VS timer enqueue MIPS re-enables interrupts on its idle routine and performs a TIF_NEED_RESCHED check afterwards before putting the CPU to sleep. The IRQs firing between the check and the 'wait' instruction may set the TIF_NEED_RESCHED flag. In order to deal with this possible race, IRQs interrupting __r4k_wait() rollback their return address to the beginning of __r4k_wait() so that TIF_NEED_RESCHED is checked again before going back to sleep. However idle IRQs can also queue timers that may require a tick reprogramming through a new generic idle loop iteration but those timers would go unnoticed here because __r4k_wait() only checks TIF_NEED_RESCHED. It doesn't check for pending timers. Fix this with fast-forwarding idle IRQs return address to the end of the idle routine instead of the beginning, so that the generic idle loop handles both TIF_NEED_RESCHED and pending timers. CONFIG_CPU_MICROMIPS has been removed along with the nop instructions. There, NOPs are 2 byte in size, so change the code with 3 _ssnop which are always 4 byte and remove the ifdef. Added ehb to make sure the hazard is always cleared. Fixes: c65a5480ff29 ("[MIPS] Fix potential latency problem due to non-atomic cpu_wait.") Signed-off-by: Marco Crivellari Signed-off-by: Maciej W. Rozycki Acked-by: Frederic Weisbecker Signed-off-by: Thomas Bogendoerfer --- arch/mips/include/asm/idle.h | 3 +- arch/mips/kernel/genex.S | 62 +++++++++++++++++++++--------------- arch/mips/kernel/idle.c | 7 ---- 3 files changed, 37 insertions(+), 35 deletions(-) diff --git a/arch/mips/include/asm/idle.h b/arch/mips/include/asm/idle.h index 0992cad9c632..2bc3678455ed 100644 --- a/arch/mips/include/asm/idle.h +++ b/arch/mips/include/asm/idle.h @@ -6,8 +6,7 @@ #include extern void (*cpu_wait)(void); -extern void r4k_wait(void); -extern asmlinkage void __r4k_wait(void); +extern asmlinkage void r4k_wait(void); extern void r4k_wait_irqoff(void); static inline int using_rollback_handler(void) diff --git a/arch/mips/kernel/genex.S b/arch/mips/kernel/genex.S index a572ce36a24f..46d975d00298 100644 --- a/arch/mips/kernel/genex.S +++ b/arch/mips/kernel/genex.S @@ -104,42 +104,52 @@ handle_vcei: __FINIT - .align 5 /* 32 byte rollback region */ -LEAF(__r4k_wait) - .set push - .set noreorder - /* start of rollback region */ - LONG_L t0, TI_FLAGS($28) - nop - andi t0, _TIF_NEED_RESCHED - bnez t0, 1f - nop - nop - nop -#ifdef CONFIG_CPU_MICROMIPS - nop - nop - nop - nop -#endif + /* Align to 32 bytes for the maximum idle interrupt region size. */ + .align 5 +LEAF(r4k_wait) + /* Keep the ISA bit clear for calculations on local labels here. */ +0: .fill 0 + /* Start of idle interrupt region. */ + local_irq_enable + /* + * If an interrupt lands here, before going idle on the next + * instruction, we must *NOT* go idle since the interrupt could + * have set TIF_NEED_RESCHED or caused a timer to need resched. + * Fall through -- see rollback_handler below -- and have the + * idle loop take care of things. + */ +1: .fill 0 + /* The R2 EI/EHB sequence takes 8 bytes, otherwise pad up. */ + .if 1b - 0b > 32 + .error "overlong idle interrupt region" + .elseif 1b - 0b > 8 + .align 4 + .endif +2: .fill 0 + .equ r4k_wait_idle_size, 2b - 0b + /* End of idle interrupt region; size has to be a power of 2. */ .set MIPS_ISA_ARCH_LEVEL_RAW +r4k_wait_insn: wait - /* end of rollback region (the region size must be power of two) */ -1: +r4k_wait_exit: + .set mips0 + local_irq_disable jr ra - nop - .set pop - END(__r4k_wait) + END(r4k_wait) + .previous .macro BUILD_ROLLBACK_PROLOGUE handler FEXPORT(rollback_\handler) .set push .set noat MFC0 k0, CP0_EPC - PTR_LA k1, __r4k_wait - ori k0, 0x1f /* 32 byte rollback region */ - xori k0, 0x1f + /* Subtract/add 2 to let the ISA bit propagate through the mask. */ + PTR_LA k1, r4k_wait_insn - 2 + ori k0, r4k_wait_idle_size - 2 + .set noreorder bne k0, k1, \handler + PTR_ADDIU k0, r4k_wait_exit - r4k_wait_insn + 2 + .set reorder MTC0 k0, CP0_EPC .set pop .endm diff --git a/arch/mips/kernel/idle.c b/arch/mips/kernel/idle.c index 5abc8b7340f8..80e8a04a642e 100644 --- a/arch/mips/kernel/idle.c +++ b/arch/mips/kernel/idle.c @@ -35,13 +35,6 @@ static void __cpuidle r3081_wait(void) write_c0_conf(cfg | R30XX_CONF_HALT); } -void __cpuidle r4k_wait(void) -{ - raw_local_irq_enable(); - __r4k_wait(); - raw_local_irq_disable(); -} - /* * This variant is preferable as it allows testing need_resched and going to * sleep depending on the outcome atomically. Unfortunately the "It is From b713f27e32d87c35737ec942dd6f5ed6b7475f48 Mon Sep 17 00:00:00 2001 From: Marco Crivellari Date: Thu, 3 Apr 2025 18:11:43 +0200 Subject: [PATCH 080/555] MIPS: Move r4k_wait() to .cpuidle.text section Fix missing .cpuidle.text section assignment for r4k_wait() to correct backtracing with nmi_backtrace(). Fixes: 97c8580e85cf ("MIPS: Annotate cpu_wait implementations with __cpuidle") Signed-off-by: Marco Crivellari Signed-off-by: Maciej W. Rozycki Acked-by: Frederic Weisbecker Signed-off-by: Thomas Bogendoerfer --- arch/mips/kernel/genex.S | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/mips/kernel/genex.S b/arch/mips/kernel/genex.S index 46d975d00298..2cf312d9a3b0 100644 --- a/arch/mips/kernel/genex.S +++ b/arch/mips/kernel/genex.S @@ -104,6 +104,7 @@ handle_vcei: __FINIT + .section .cpuidle.text,"ax" /* Align to 32 bytes for the maximum idle interrupt region size. */ .align 5 LEAF(r4k_wait) From cc3e3d3a9d09456cf21694b7ea8b9d781e85fda3 Mon Sep 17 00:00:00 2001 From: Marco Crivellari Date: Sat, 5 Apr 2025 16:37:05 +0200 Subject: [PATCH 081/555] MIPS: rename rollback_handler with skipover_handler Recently the rollback region has been changed into an idle interrupt region [1]. This patch make the appropriate changes renaming functions and macro, to reflect the change. [1] https://lore.kernel.org/linux-mips/20250403161143.361461-2-marco.crivellari@suse.com/ Signed-off-by: Marco Crivellari Signed-off-by: Thomas Bogendoerfer --- arch/mips/include/asm/idle.h | 2 +- arch/mips/kernel/genex.S | 10 +++++----- arch/mips/kernel/traps.c | 10 +++++----- 3 files changed, 11 insertions(+), 11 deletions(-) diff --git a/arch/mips/include/asm/idle.h b/arch/mips/include/asm/idle.h index 2bc3678455ed..c7d75807d13f 100644 --- a/arch/mips/include/asm/idle.h +++ b/arch/mips/include/asm/idle.h @@ -9,7 +9,7 @@ extern void (*cpu_wait)(void); extern asmlinkage void r4k_wait(void); extern void r4k_wait_irqoff(void); -static inline int using_rollback_handler(void) +static inline int using_skipover_handler(void) { return cpu_wait == r4k_wait; } diff --git a/arch/mips/kernel/genex.S b/arch/mips/kernel/genex.S index 2cf312d9a3b0..08c0a01d9a29 100644 --- a/arch/mips/kernel/genex.S +++ b/arch/mips/kernel/genex.S @@ -116,7 +116,7 @@ LEAF(r4k_wait) * If an interrupt lands here, before going idle on the next * instruction, we must *NOT* go idle since the interrupt could * have set TIF_NEED_RESCHED or caused a timer to need resched. - * Fall through -- see rollback_handler below -- and have the + * Fall through -- see skipover_handler below -- and have the * idle loop take care of things. */ 1: .fill 0 @@ -139,8 +139,8 @@ r4k_wait_exit: END(r4k_wait) .previous - .macro BUILD_ROLLBACK_PROLOGUE handler - FEXPORT(rollback_\handler) + .macro BUILD_SKIPOVER_PROLOGUE handler + FEXPORT(skipover_\handler) .set push .set noat MFC0 k0, CP0_EPC @@ -156,7 +156,7 @@ r4k_wait_exit: .endm .align 5 -BUILD_ROLLBACK_PROLOGUE handle_int +BUILD_SKIPOVER_PROLOGUE handle_int NESTED(handle_int, PT_SIZE, sp) .cfi_signal_frame #ifdef CONFIG_TRACE_IRQFLAGS @@ -276,7 +276,7 @@ NESTED(except_vec_ejtag_debug, 0, sp) * This prototype is copied to ebase + n*IntCtl.VS and patched * to invoke the handler */ -BUILD_ROLLBACK_PROLOGUE except_vec_vi +BUILD_SKIPOVER_PROLOGUE except_vec_vi NESTED(except_vec_vi, 0, sp) SAVE_SOME docfi=1 SAVE_AT docfi=1 diff --git a/arch/mips/kernel/traps.c b/arch/mips/kernel/traps.c index 39e248d0ed59..8ec1e185b35c 100644 --- a/arch/mips/kernel/traps.c +++ b/arch/mips/kernel/traps.c @@ -77,7 +77,7 @@ #include "access-helper.h" extern void check_wait(void); -extern asmlinkage void rollback_handle_int(void); +extern asmlinkage void skipover_handle_int(void); extern asmlinkage void handle_int(void); extern asmlinkage void handle_adel(void); extern asmlinkage void handle_ades(void); @@ -2066,7 +2066,7 @@ void *set_vi_handler(int n, vi_handler_t addr) { extern const u8 except_vec_vi[]; extern const u8 except_vec_vi_ori[], except_vec_vi_end[]; - extern const u8 rollback_except_vec_vi[]; + extern const u8 skipover_except_vec_vi[]; unsigned long handler; unsigned long old_handler = vi_handlers[n]; int srssets = current_cpu_data.srsets; @@ -2095,7 +2095,7 @@ void *set_vi_handler(int n, vi_handler_t addr) change_c0_srsmap(0xf << n*4, 0 << n*4); } - vec_start = using_rollback_handler() ? rollback_except_vec_vi : + vec_start = using_skipover_handler() ? skipover_except_vec_vi : except_vec_vi; #if defined(CONFIG_CPU_MICROMIPS) || defined(CONFIG_CPU_BIG_ENDIAN) ori_offset = except_vec_vi_ori - vec_start + 2; @@ -2426,8 +2426,8 @@ void __init trap_init(void) if (board_be_init) board_be_init(); - set_except_vector(EXCCODE_INT, using_rollback_handler() ? - rollback_handle_int : handle_int); + set_except_vector(EXCCODE_INT, using_skipover_handler() ? + skipover_handle_int : handle_int); set_except_vector(EXCCODE_MOD, handle_tlbm); set_except_vector(EXCCODE_TLBL, handle_tlbl); set_except_vector(EXCCODE_TLBS, handle_tlbs); From 7f74c066e5d920b3a2f0f936060984e3b3709250 Mon Sep 17 00:00:00 2001 From: Thorsten Blum Date: Mon, 7 Apr 2025 18:32:21 +0200 Subject: [PATCH 082/555] MIPS: CPS: Fix potential NULL pointer dereferences in cps_prepare_cpus() Check the return values of kcalloc() and exit early to avoid potential NULL pointer dereferences. Compile-tested only. Cc: stable@vger.kernel.org Fixes: 75fa6a583882e ("MIPS: CPS: Introduce struct cluster_boot_config") Fixes: 0856c143e1cd3 ("MIPS: CPS: Boot CPUs in secondary clusters") Signed-off-by: Thorsten Blum Signed-off-by: Thomas Bogendoerfer --- arch/mips/kernel/smp-cps.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/arch/mips/kernel/smp-cps.c b/arch/mips/kernel/smp-cps.c index e85bd087467e..cc26d56f3ab6 100644 --- a/arch/mips/kernel/smp-cps.c +++ b/arch/mips/kernel/smp-cps.c @@ -332,6 +332,8 @@ static void __init cps_prepare_cpus(unsigned int max_cpus) mips_cps_cluster_bootcfg = kcalloc(nclusters, sizeof(*mips_cps_cluster_bootcfg), GFP_KERNEL); + if (!mips_cps_cluster_bootcfg) + goto err_out; if (nclusters > 1) mips_cm_update_property(); @@ -348,6 +350,8 @@ static void __init cps_prepare_cpus(unsigned int max_cpus) mips_cps_cluster_bootcfg[cl].core_power = kcalloc(BITS_TO_LONGS(ncores), sizeof(unsigned long), GFP_KERNEL); + if (!mips_cps_cluster_bootcfg[cl].core_power) + goto err_out; /* Allocate VPE boot configuration structs */ for (c = 0; c < ncores; c++) { From 2d7f844ffac683dea7b1697c016e63081199383e Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sun, 6 Apr 2025 10:38:52 +0200 Subject: [PATCH 083/555] NFSv4: Handle fatal ENETDOWN and ENETUNREACH errors Ensure that the NFSv4 error handling code recognises the RPC_TASK_NETUNREACH_FATAL flag, and handles the ENETDOWN and ENETUNREACH errors accordingly. Signed-off-by: Trond Myklebust Reviewed-by: Jeff Layton --- fs/nfs/nfs4proc.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 970f28dbf253..1f7cc260b007 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -671,6 +671,15 @@ nfs4_async_handle_exception(struct rpc_task *task, struct nfs_server *server, struct nfs_client *clp = server->nfs_client; int ret; + if ((task->tk_rpc_status == -ENETDOWN || + task->tk_rpc_status == -ENETUNREACH) && + task->tk_flags & RPC_TASK_NETUNREACH_FATAL) { + exception->delay = 0; + exception->recovering = 0; + exception->retry = 0; + return -EIO; + } + ret = nfs4_do_handle_exception(server, errorcode, exception); if (exception->delay) { int ret2 = nfs4_exception_should_retrans(server, exception); From 440caf8ee2743d44ae5a6da209854188110993fa Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sun, 6 Apr 2025 11:05:27 +0200 Subject: [PATCH 084/555] NFSv4/pnfs: Layoutreturn on close must handle fatal networking errors If we have a fatal ENETDOWN or ENETUNREACH error, then the layoutreturn on close code should also handle that as fatal, and free the layouts. Signed-off-by: Trond Myklebust Reviewed-by: Jeff Layton --- fs/nfs/pnfs.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index 5f582713bf05..10fdd065a61c 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -1661,6 +1661,18 @@ int pnfs_roc_done(struct rpc_task *task, struct nfs4_layoutreturn_args **argpp, /* Was there an RPC level error? If not, retry */ if (task->tk_rpc_status == 0) break; + /* + * Is there a fatal network level error? + * If so release the layout, but flag the error. + */ + if ((task->tk_rpc_status == -ENETDOWN || + task->tk_rpc_status == -ENETUNREACH) && + task->tk_flags & RPC_TASK_NETUNREACH_FATAL) { + *ret = 0; + (*respp)->lrs_present = 0; + retval = -EIO; + break; + } /* If the call was not sent, let caller handle it */ if (!RPC_WAS_SENT(task)) return 0; From bead8b4953f466514f28e552269110b23a19d6f0 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Mon, 7 Apr 2025 14:36:41 +0200 Subject: [PATCH 085/555] pNFS/flexfiles: Record the RPC errors in the I/O tracepoints When debugging I/O issues, we want to see not just the NFS level errors, but also the RPC level problems, so record both in the tracepoints. Signed-off-by: Trond Myklebust --- fs/nfs/flexfilelayout/flexfilelayout.c | 6 ++--- fs/nfs/nfs4trace.h | 34 +++++++++++++++++--------- 2 files changed, 25 insertions(+), 15 deletions(-) diff --git a/fs/nfs/flexfilelayout/flexfilelayout.c b/fs/nfs/flexfilelayout/flexfilelayout.c index 61ad269c825f..e6909cafab68 100644 --- a/fs/nfs/flexfilelayout/flexfilelayout.c +++ b/fs/nfs/flexfilelayout/flexfilelayout.c @@ -1329,7 +1329,7 @@ static int ff_layout_read_done_cb(struct rpc_task *task, hdr->args.offset, hdr->args.count, &hdr->res.op_status, OP_READ, task->tk_status); - trace_ff_layout_read_error(hdr); + trace_ff_layout_read_error(hdr, task->tk_status); } err = ff_layout_async_handle_error(task, hdr->args.context->state, @@ -1502,7 +1502,7 @@ static int ff_layout_write_done_cb(struct rpc_task *task, hdr->args.offset, hdr->args.count, &hdr->res.op_status, OP_WRITE, task->tk_status); - trace_ff_layout_write_error(hdr); + trace_ff_layout_write_error(hdr, task->tk_status); } err = ff_layout_async_handle_error(task, hdr->args.context->state, @@ -1551,7 +1551,7 @@ static int ff_layout_commit_done_cb(struct rpc_task *task, data->args.offset, data->args.count, &data->res.op_status, OP_COMMIT, task->tk_status); - trace_ff_layout_commit_error(data); + trace_ff_layout_commit_error(data, task->tk_status); } err = ff_layout_async_handle_error(task, NULL, data->ds_clp, diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h index bc67fe6801b1..deab4c0e21a0 100644 --- a/fs/nfs/nfs4trace.h +++ b/fs/nfs/nfs4trace.h @@ -2051,13 +2051,15 @@ TRACE_EVENT(fl_getdevinfo, DECLARE_EVENT_CLASS(nfs4_flexfiles_io_event, TP_PROTO( - const struct nfs_pgio_header *hdr + const struct nfs_pgio_header *hdr, + int error ), - TP_ARGS(hdr), + TP_ARGS(hdr, error), TP_STRUCT__entry( __field(unsigned long, error) + __field(unsigned long, nfs_error) __field(dev_t, dev) __field(u32, fhandle) __field(u64, fileid) @@ -2073,7 +2075,8 @@ DECLARE_EVENT_CLASS(nfs4_flexfiles_io_event, TP_fast_assign( const struct inode *inode = hdr->inode; - __entry->error = hdr->res.op_status; + __entry->error = -error; + __entry->nfs_error = hdr->res.op_status; __entry->fhandle = nfs_fhandle_hash(hdr->args.fh); __entry->fileid = NFS_FILEID(inode); __entry->dev = inode->i_sb->s_dev; @@ -2088,7 +2091,8 @@ DECLARE_EVENT_CLASS(nfs4_flexfiles_io_event, TP_printk( "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " - "offset=%llu count=%u stateid=%d:0x%08x dstaddr=%s", + "offset=%llu count=%u stateid=%d:0x%08x dstaddr=%s " + "nfs_error=%lu (%s)", -__entry->error, show_nfs4_status(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), @@ -2096,28 +2100,32 @@ DECLARE_EVENT_CLASS(nfs4_flexfiles_io_event, __entry->fhandle, __entry->offset, __entry->count, __entry->stateid_seq, __entry->stateid_hash, - __get_str(dstaddr) + __get_str(dstaddr), __entry->nfs_error, + show_nfs4_status(__entry->nfs_error) ) ); #define DEFINE_NFS4_FLEXFILES_IO_EVENT(name) \ DEFINE_EVENT(nfs4_flexfiles_io_event, name, \ TP_PROTO( \ - const struct nfs_pgio_header *hdr \ + const struct nfs_pgio_header *hdr, \ + int error \ ), \ - TP_ARGS(hdr)) + TP_ARGS(hdr, error)) DEFINE_NFS4_FLEXFILES_IO_EVENT(ff_layout_read_error); DEFINE_NFS4_FLEXFILES_IO_EVENT(ff_layout_write_error); TRACE_EVENT(ff_layout_commit_error, TP_PROTO( - const struct nfs_commit_data *data + const struct nfs_commit_data *data, + int error ), - TP_ARGS(data), + TP_ARGS(data, error), TP_STRUCT__entry( __field(unsigned long, error) + __field(unsigned long, nfs_error) __field(dev_t, dev) __field(u32, fhandle) __field(u64, fileid) @@ -2131,7 +2139,8 @@ TRACE_EVENT(ff_layout_commit_error, TP_fast_assign( const struct inode *inode = data->inode; - __entry->error = data->res.op_status; + __entry->error = -error; + __entry->nfs_error = data->res.op_status; __entry->fhandle = nfs_fhandle_hash(data->args.fh); __entry->fileid = NFS_FILEID(inode); __entry->dev = inode->i_sb->s_dev; @@ -2142,14 +2151,15 @@ TRACE_EVENT(ff_layout_commit_error, TP_printk( "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " - "offset=%llu count=%u dstaddr=%s", + "offset=%llu count=%u dstaddr=%s nfs_error=%lu (%s)", -__entry->error, show_nfs4_status(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, __entry->fhandle, __entry->offset, __entry->count, - __get_str(dstaddr) + __get_str(dstaddr), __entry->nfs_error, + show_nfs4_status(__entry->nfs_error) ) ); From c457dc1ec770a22636b473ce5d35614adfe97636 Mon Sep 17 00:00:00 2001 From: Li Lingfeng Date: Thu, 17 Apr 2025 15:25:08 +0800 Subject: [PATCH 086/555] nfs: handle failure of nfs_get_lock_context in unlock path When memory is insufficient, the allocation of nfs_lock_context in nfs_get_lock_context() fails and returns -ENOMEM. If we mistakenly treat an nfs4_unlockdata structure (whose l_ctx member has been set to -ENOMEM) as valid and proceed to execute rpc_run_task(), this will trigger a NULL pointer dereference in nfs4_locku_prepare. For example: BUG: kernel NULL pointer dereference, address: 000000000000000c PGD 0 P4D 0 Oops: Oops: 0000 [#1] SMP PTI CPU: 15 UID: 0 PID: 12 Comm: kworker/u64:0 Not tainted 6.15.0-rc2-dirty #60 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.3-2.fc40 Workqueue: rpciod rpc_async_schedule RIP: 0010:nfs4_locku_prepare+0x35/0xc2 Code: 89 f2 48 89 fd 48 c7 c7 68 69 ef b5 53 48 8b 8e 90 00 00 00 48 89 f3 RSP: 0018:ffffbbafc006bdb8 EFLAGS: 00010246 RAX: 000000000000004b RBX: ffff9b964fc1fa00 RCX: 0000000000000000 RDX: 0000000000000000 RSI: fffffffffffffff4 RDI: ffff9ba53fddbf40 RBP: ffff9ba539934000 R08: 0000000000000000 R09: ffffbbafc006bc38 R10: ffffffffb6b689c8 R11: 0000000000000003 R12: ffff9ba539934030 R13: 0000000000000001 R14: 0000000004248060 R15: ffffffffb56d1c30 FS: 0000000000000000(0000) GS:ffff9ba5881f0000(0000) knlGS:00000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 000000000000000c CR3: 000000093f244000 CR4: 00000000000006f0 Call Trace: __rpc_execute+0xbc/0x480 rpc_async_schedule+0x2f/0x40 process_one_work+0x232/0x5d0 worker_thread+0x1da/0x3d0 ? __pfx_worker_thread+0x10/0x10 kthread+0x10d/0x240 ? __pfx_kthread+0x10/0x10 ret_from_fork+0x34/0x50 ? __pfx_kthread+0x10/0x10 ret_from_fork_asm+0x1a/0x30 Modules linked in: CR2: 000000000000000c ---[ end trace 0000000000000000 ]--- Free the allocated nfs4_unlockdata when nfs_get_lock_context() fails and return NULL to terminate subsequent rpc_run_task, preventing NULL pointer dereference. Fixes: f30cb757f680 ("NFS: Always wait for I/O completion before unlock") Signed-off-by: Li Lingfeng Reviewed-by: Jeff Layton Link: https://lore.kernel.org/r/20250417072508.3850532-1-lilingfeng3@huawei.com Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 1f7cc260b007..b1d2122bd5a7 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -7083,10 +7083,18 @@ static struct nfs4_unlockdata *nfs4_alloc_unlockdata(struct file_lock *fl, struct nfs4_unlockdata *p; struct nfs4_state *state = lsp->ls_state; struct inode *inode = state->inode; + struct nfs_lock_context *l_ctx; p = kzalloc(sizeof(*p), GFP_KERNEL); if (p == NULL) return NULL; + l_ctx = nfs_get_lock_context(ctx); + if (!IS_ERR(l_ctx)) { + p->l_ctx = l_ctx; + } else { + kfree(p); + return NULL; + } p->arg.fh = NFS_FH(inode); p->arg.fl = &p->fl; p->arg.seqid = seqid; @@ -7094,7 +7102,6 @@ static struct nfs4_unlockdata *nfs4_alloc_unlockdata(struct file_lock *fl, p->lsp = lsp; /* Ensure we don't close file until we're done freeing locks! */ p->ctx = get_nfs_open_context(ctx); - p->l_ctx = nfs_get_lock_context(ctx); locks_init_lock(&p->fl); locks_copy_lock(&p->fl, fl); p->server = NFS_SERVER(inode); From 6b9785dc8b13d9fb75ceec8cf4ea7ec3f3b1edbc Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Thu, 10 Apr 2025 16:42:03 -0400 Subject: [PATCH 087/555] nfs: don't share pNFS DS connections between net namespaces Currently, different NFS clients can share the same DS connections, even when they are in different net namespaces. If a containerized client creates a DS connection, another container can find and use it. When the first client exits, the connection will close which can lead to stalls in other clients. Add a net namespace pointer to struct nfs4_pnfs_ds, and compare those value to the caller's netns in _data_server_lookup_locked() when searching for a nfs4_pnfs_ds to match. Reported-by: Omar Sandoval Reported-by: Sargun Dillon Closes: https://lore.kernel.org/linux-nfs/Z_ArpQC_vREh_hEA@telecaster/ Tested-by: Sargun Dillon Signed-off-by: Jeff Layton Reviewed-by: Benjamin Coddington Link: https://lore.kernel.org/r/20250410-nfs-ds-netns-v2-1-f80b7979ba80@kernel.org Signed-off-by: Trond Myklebust --- fs/nfs/filelayout/filelayoutdev.c | 6 +++--- fs/nfs/flexfilelayout/flexfilelayoutdev.c | 6 +++--- fs/nfs/pnfs.h | 4 +++- fs/nfs/pnfs_nfs.c | 9 +++++---- 4 files changed, 14 insertions(+), 11 deletions(-) diff --git a/fs/nfs/filelayout/filelayoutdev.c b/fs/nfs/filelayout/filelayoutdev.c index 4fa304fa5bc4..29d9234d5c08 100644 --- a/fs/nfs/filelayout/filelayoutdev.c +++ b/fs/nfs/filelayout/filelayoutdev.c @@ -76,6 +76,7 @@ nfs4_fl_alloc_deviceid_node(struct nfs_server *server, struct pnfs_device *pdev, struct page *scratch; struct list_head dsaddrs; struct nfs4_pnfs_ds_addr *da; + struct net *net = server->nfs_client->cl_net; /* set up xdr stream */ scratch = alloc_page(gfp_flags); @@ -159,8 +160,7 @@ nfs4_fl_alloc_deviceid_node(struct nfs_server *server, struct pnfs_device *pdev, mp_count = be32_to_cpup(p); /* multipath count */ for (j = 0; j < mp_count; j++) { - da = nfs4_decode_mp_ds_addr(server->nfs_client->cl_net, - &stream, gfp_flags); + da = nfs4_decode_mp_ds_addr(net, &stream, gfp_flags); if (da) list_add_tail(&da->da_node, &dsaddrs); } @@ -170,7 +170,7 @@ nfs4_fl_alloc_deviceid_node(struct nfs_server *server, struct pnfs_device *pdev, goto out_err_free_deviceid; } - dsaddr->ds_list[i] = nfs4_pnfs_ds_add(&dsaddrs, gfp_flags); + dsaddr->ds_list[i] = nfs4_pnfs_ds_add(net, &dsaddrs, gfp_flags); if (!dsaddr->ds_list[i]) goto out_err_drain_dsaddrs; trace_fl_getdevinfo(server, &pdev->dev_id, dsaddr->ds_list[i]->ds_remotestr); diff --git a/fs/nfs/flexfilelayout/flexfilelayoutdev.c b/fs/nfs/flexfilelayout/flexfilelayoutdev.c index e58bedfb1dcc..4a304cf17c4b 100644 --- a/fs/nfs/flexfilelayout/flexfilelayoutdev.c +++ b/fs/nfs/flexfilelayout/flexfilelayoutdev.c @@ -49,6 +49,7 @@ nfs4_ff_alloc_deviceid_node(struct nfs_server *server, struct pnfs_device *pdev, struct nfs4_pnfs_ds_addr *da; struct nfs4_ff_layout_ds *new_ds = NULL; struct nfs4_ff_ds_version *ds_versions = NULL; + struct net *net = server->nfs_client->cl_net; u32 mp_count; u32 version_count; __be32 *p; @@ -80,8 +81,7 @@ nfs4_ff_alloc_deviceid_node(struct nfs_server *server, struct pnfs_device *pdev, for (i = 0; i < mp_count; i++) { /* multipath ds */ - da = nfs4_decode_mp_ds_addr(server->nfs_client->cl_net, - &stream, gfp_flags); + da = nfs4_decode_mp_ds_addr(net, &stream, gfp_flags); if (da) list_add_tail(&da->da_node, &dsaddrs); } @@ -149,7 +149,7 @@ nfs4_ff_alloc_deviceid_node(struct nfs_server *server, struct pnfs_device *pdev, new_ds->ds_versions = ds_versions; new_ds->ds_versions_cnt = version_count; - new_ds->ds = nfs4_pnfs_ds_add(&dsaddrs, gfp_flags); + new_ds->ds = nfs4_pnfs_ds_add(net, &dsaddrs, gfp_flags); if (!new_ds->ds) goto out_err_drain_dsaddrs; diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h index 30d2613e912b..91ff877185c8 100644 --- a/fs/nfs/pnfs.h +++ b/fs/nfs/pnfs.h @@ -60,6 +60,7 @@ struct nfs4_pnfs_ds { struct list_head ds_node; /* nfs4_pnfs_dev_hlist dev_dslist */ char *ds_remotestr; /* comma sep list of addrs */ struct list_head ds_addrs; + const struct net *ds_net; struct nfs_client *ds_clp; refcount_t ds_count; unsigned long ds_state; @@ -415,7 +416,8 @@ int pnfs_generic_commit_pagelist(struct inode *inode, int pnfs_generic_scan_commit_lists(struct nfs_commit_info *cinfo, int max); void pnfs_generic_write_commit_done(struct rpc_task *task, void *data); void nfs4_pnfs_ds_put(struct nfs4_pnfs_ds *ds); -struct nfs4_pnfs_ds *nfs4_pnfs_ds_add(struct list_head *dsaddrs, +struct nfs4_pnfs_ds *nfs4_pnfs_ds_add(const struct net *net, + struct list_head *dsaddrs, gfp_t gfp_flags); void nfs4_pnfs_v3_ds_connect_unload(void); int nfs4_pnfs_ds_connect(struct nfs_server *mds_srv, struct nfs4_pnfs_ds *ds, diff --git a/fs/nfs/pnfs_nfs.c b/fs/nfs/pnfs_nfs.c index dbef837e871a..2ee20a0f0b36 100644 --- a/fs/nfs/pnfs_nfs.c +++ b/fs/nfs/pnfs_nfs.c @@ -604,12 +604,12 @@ _same_data_server_addrs_locked(const struct list_head *dsaddrs1, * Lookup DS by addresses. nfs4_ds_cache_lock is held */ static struct nfs4_pnfs_ds * -_data_server_lookup_locked(const struct list_head *dsaddrs) +_data_server_lookup_locked(const struct net *net, const struct list_head *dsaddrs) { struct nfs4_pnfs_ds *ds; list_for_each_entry(ds, &nfs4_data_server_cache, ds_node) - if (_same_data_server_addrs_locked(&ds->ds_addrs, dsaddrs)) + if (ds->ds_net == net && _same_data_server_addrs_locked(&ds->ds_addrs, dsaddrs)) return ds; return NULL; } @@ -716,7 +716,7 @@ out_err: * uncached and return cached struct nfs4_pnfs_ds. */ struct nfs4_pnfs_ds * -nfs4_pnfs_ds_add(struct list_head *dsaddrs, gfp_t gfp_flags) +nfs4_pnfs_ds_add(const struct net *net, struct list_head *dsaddrs, gfp_t gfp_flags) { struct nfs4_pnfs_ds *tmp_ds, *ds = NULL; char *remotestr; @@ -734,13 +734,14 @@ nfs4_pnfs_ds_add(struct list_head *dsaddrs, gfp_t gfp_flags) remotestr = nfs4_pnfs_remotestr(dsaddrs, gfp_flags); spin_lock(&nfs4_ds_cache_lock); - tmp_ds = _data_server_lookup_locked(dsaddrs); + tmp_ds = _data_server_lookup_locked(net, dsaddrs); if (tmp_ds == NULL) { INIT_LIST_HEAD(&ds->ds_addrs); list_splice_init(dsaddrs, &ds->ds_addrs); ds->ds_remotestr = remotestr; refcount_set(&ds->ds_count, 1); INIT_LIST_HEAD(&ds->ds_node); + ds->ds_net = net; ds->ds_clp = NULL; list_add(&ds->ds_node, &nfs4_data_server_cache); dprintk("%s add new data server %s\n", __func__, From d5fb22a7c585b12ec3e6cef150689f7386e8cfd0 Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Thu, 10 Apr 2025 16:42:04 -0400 Subject: [PATCH 088/555] nfs: move the nfs4_data_server_cache into struct nfs_net Since struct nfs4_pnfs_ds should not be shared between net namespaces, move from a global list of objects to a per-netns list and spinlock. Tested-by: Sargun Dillon Signed-off-by: Jeff Layton Reviewed-by: Benjamin Coddington Link: https://lore.kernel.org/r/20250410-nfs-ds-netns-v2-2-f80b7979ba80@kernel.org Signed-off-by: Trond Myklebust --- fs/nfs/client.c | 7 +++++++ fs/nfs/netns.h | 6 +++++- fs/nfs/pnfs_nfs.c | 31 +++++++++++++++++-------------- 3 files changed, 29 insertions(+), 15 deletions(-) diff --git a/fs/nfs/client.c b/fs/nfs/client.c index 02c916a55020..2115c1189c2d 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -1199,6 +1199,10 @@ void nfs_clients_init(struct net *net) INIT_LIST_HEAD(&nn->nfs_volume_list); #if IS_ENABLED(CONFIG_NFS_V4) idr_init(&nn->cb_ident_idr); +#endif +#if IS_ENABLED(CONFIG_NFS_V4_1) + INIT_LIST_HEAD(&nn->nfs4_data_server_cache); + spin_lock_init(&nn->nfs4_data_server_lock); #endif spin_lock_init(&nn->nfs_client_lock); nn->boot_time = ktime_get_real(); @@ -1216,6 +1220,9 @@ void nfs_clients_exit(struct net *net) nfs_cleanup_cb_ident_idr(net); WARN_ON_ONCE(!list_empty(&nn->nfs_client_list)); WARN_ON_ONCE(!list_empty(&nn->nfs_volume_list)); +#if IS_ENABLED(CONFIG_NFS_V4_1) + WARN_ON_ONCE(!list_empty(&nn->nfs4_data_server_cache)); +#endif } #ifdef CONFIG_PROC_FS diff --git a/fs/nfs/netns.h b/fs/nfs/netns.h index a68b21603ea9..6ba3ea39e928 100644 --- a/fs/nfs/netns.h +++ b/fs/nfs/netns.h @@ -31,7 +31,11 @@ struct nfs_net { unsigned short nfs_callback_tcpport; unsigned short nfs_callback_tcpport6; int cb_users[NFS4_MAX_MINOR_VERSION + 1]; -#endif +#endif /* CONFIG_NFS_V4 */ +#if IS_ENABLED(CONFIG_NFS_V4_1) + struct list_head nfs4_data_server_cache; + spinlock_t nfs4_data_server_lock; +#endif /* CONFIG_NFS_V4_1 */ struct nfs_netns_client *nfs_client; spinlock_t nfs_client_lock; ktime_t boot_time; diff --git a/fs/nfs/pnfs_nfs.c b/fs/nfs/pnfs_nfs.c index 2ee20a0f0b36..91ef486f40b9 100644 --- a/fs/nfs/pnfs_nfs.c +++ b/fs/nfs/pnfs_nfs.c @@ -16,6 +16,7 @@ #include "nfs4session.h" #include "internal.h" #include "pnfs.h" +#include "netns.h" #define NFSDBG_FACILITY NFSDBG_PNFS @@ -504,14 +505,14 @@ EXPORT_SYMBOL_GPL(pnfs_generic_commit_pagelist); /* * Data server cache * - * Data servers can be mapped to different device ids. - * nfs4_pnfs_ds reference counting + * Data servers can be mapped to different device ids, but should + * never be shared between net namespaces. + * + * nfs4_pnfs_ds reference counting: * - set to 1 on allocation * - incremented when a device id maps a data server already in the cache. * - decremented when deviceid is removed from the cache. */ -static DEFINE_SPINLOCK(nfs4_ds_cache_lock); -static LIST_HEAD(nfs4_data_server_cache); /* Debug routines */ static void @@ -604,12 +605,12 @@ _same_data_server_addrs_locked(const struct list_head *dsaddrs1, * Lookup DS by addresses. nfs4_ds_cache_lock is held */ static struct nfs4_pnfs_ds * -_data_server_lookup_locked(const struct net *net, const struct list_head *dsaddrs) +_data_server_lookup_locked(const struct nfs_net *nn, const struct list_head *dsaddrs) { struct nfs4_pnfs_ds *ds; - list_for_each_entry(ds, &nfs4_data_server_cache, ds_node) - if (ds->ds_net == net && _same_data_server_addrs_locked(&ds->ds_addrs, dsaddrs)) + list_for_each_entry(ds, &nn->nfs4_data_server_cache, ds_node) + if (_same_data_server_addrs_locked(&ds->ds_addrs, dsaddrs)) return ds; return NULL; } @@ -653,10 +654,11 @@ static void destroy_ds(struct nfs4_pnfs_ds *ds) void nfs4_pnfs_ds_put(struct nfs4_pnfs_ds *ds) { - if (refcount_dec_and_lock(&ds->ds_count, - &nfs4_ds_cache_lock)) { + struct nfs_net *nn = net_generic(ds->ds_net, nfs_net_id); + + if (refcount_dec_and_lock(&ds->ds_count, &nn->nfs4_data_server_lock)) { list_del_init(&ds->ds_node); - spin_unlock(&nfs4_ds_cache_lock); + spin_unlock(&nn->nfs4_data_server_lock); destroy_ds(ds); } } @@ -718,6 +720,7 @@ out_err: struct nfs4_pnfs_ds * nfs4_pnfs_ds_add(const struct net *net, struct list_head *dsaddrs, gfp_t gfp_flags) { + struct nfs_net *nn = net_generic(net, nfs_net_id); struct nfs4_pnfs_ds *tmp_ds, *ds = NULL; char *remotestr; @@ -733,8 +736,8 @@ nfs4_pnfs_ds_add(const struct net *net, struct list_head *dsaddrs, gfp_t gfp_fla /* this is only used for debugging, so it's ok if its NULL */ remotestr = nfs4_pnfs_remotestr(dsaddrs, gfp_flags); - spin_lock(&nfs4_ds_cache_lock); - tmp_ds = _data_server_lookup_locked(net, dsaddrs); + spin_lock(&nn->nfs4_data_server_lock); + tmp_ds = _data_server_lookup_locked(nn, dsaddrs); if (tmp_ds == NULL) { INIT_LIST_HEAD(&ds->ds_addrs); list_splice_init(dsaddrs, &ds->ds_addrs); @@ -743,7 +746,7 @@ nfs4_pnfs_ds_add(const struct net *net, struct list_head *dsaddrs, gfp_t gfp_fla INIT_LIST_HEAD(&ds->ds_node); ds->ds_net = net; ds->ds_clp = NULL; - list_add(&ds->ds_node, &nfs4_data_server_cache); + list_add(&ds->ds_node, &nn->nfs4_data_server_cache); dprintk("%s add new data server %s\n", __func__, ds->ds_remotestr); } else { @@ -755,7 +758,7 @@ nfs4_pnfs_ds_add(const struct net *net, struct list_head *dsaddrs, gfp_t gfp_fla refcount_read(&tmp_ds->ds_count)); ds = tmp_ds; } - spin_unlock(&nfs4_ds_cache_lock); + spin_unlock(&nn->nfs4_data_server_lock); out: return ds; } From d82e86c15364d42706eb8b5249640a839d61a681 Mon Sep 17 00:00:00 2001 From: Sergey Shtylyov Date: Tue, 8 Apr 2025 23:53:42 +0300 Subject: [PATCH 089/555] nfs: direct: drop useless initializer in nfs_direct_write_completion() In nfs_direct_write_completion(), the local variable req isn't used outside the *while* loop and is assigned to right at the start of that loop's body, so its initializer appears useless -- drop it; then move the declaration to the loop body (which happens to have a pointless empty line anyway)... Found by Linux Verification Center (linuxtesting.org) with the Svace static analysis tool. Signed-off-by: Sergey Shtylyov Reviewed-by: Benjamin Coddington Link: https://lore.kernel.org/r/416219f5-7983-484b-b5a7-5fb7da9561f7@omp.ru Signed-off-by: Trond Myklebust --- fs/nfs/direct.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index f32f8d7c9122..48d89716193a 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -757,7 +757,6 @@ static void nfs_direct_write_completion(struct nfs_pgio_header *hdr) { struct nfs_direct_req *dreq = hdr->dreq; struct nfs_commit_info cinfo; - struct nfs_page *req = nfs_list_entry(hdr->pages.next); struct inode *inode = dreq->inode; int flags = NFS_ODIRECT_DONE; @@ -786,6 +785,7 @@ static void nfs_direct_write_completion(struct nfs_pgio_header *hdr) spin_unlock(&inode->i_lock); while (!list_empty(&hdr->pages)) { + struct nfs_page *req; req = nfs_list_entry(hdr->pages.next); nfs_list_remove_request(req); From c367eea5041c2e5ef6836fe0ba8c5dc75a965b1b Mon Sep 17 00:00:00 2001 From: Sergey Shtylyov Date: Wed, 9 Apr 2025 23:36:33 +0300 Subject: [PATCH 090/555] nfs: nfs3acl: drop useless assignment in nfs3_get_acl() In nfs3_get_acl(), the local variable status is assigned the result of nfs_refresh_inode() inside the *switch* statement, but that value gets overwritten in the next *if* statement's true branch and is completely ignored if that branch isn't taken... Found by Linux Verification Center (linuxtesting.org) with the Svace static analysis tool. Signed-off-by: Sergey Shtylyov Reviewed-by: Benjamin Coddington Link: https://lore.kernel.org/r/c32dced7-a4fa-43c0-aafe-ef6c819c2f91@omp.ru Signed-off-by: Trond Myklebust --- fs/nfs/nfs3acl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/nfs/nfs3acl.c b/fs/nfs/nfs3acl.c index 18d8f6529f61..a126eb31f62f 100644 --- a/fs/nfs/nfs3acl.c +++ b/fs/nfs/nfs3acl.c @@ -104,7 +104,7 @@ struct posix_acl *nfs3_get_acl(struct inode *inode, int type, bool rcu) switch (status) { case 0: - status = nfs_refresh_inode(inode, res.fattr); + nfs_refresh_inode(inode, res.fattr); break; case -EPFNOSUPPORT: case -EPROTONOSUPPORT: From 8b0ba61df5a1c44e2b3cf683831a4fc5e24ea99d Mon Sep 17 00:00:00 2001 From: Stephen Smalley Date: Thu, 24 Apr 2025 11:28:20 -0400 Subject: [PATCH 091/555] fs/xattr.c: fix simple_xattr_list to always include security.* xattrs The vfs has long had a fallback to obtain the security.* xattrs from the LSM when the filesystem does not implement its own listxattr, but shmem/tmpfs and kernfs later gained their own xattr handlers to support other xattrs. Unfortunately, as a side effect, tmpfs and kernfs-based filesystems like sysfs no longer return the synthetic security.* xattr names via listxattr unless they are explicitly set by userspace or initially set upon inode creation after policy load. coreutils has recently switched from unconditionally invoking getxattr for security.* for ls -Z via libselinux to only doing so if listxattr returns the xattr name, breaking ls -Z of such inodes. Before: $ getfattr -m.* /run/initramfs $ getfattr -m.* /sys/kernel/fscaps $ setfattr -n user.foo /run/initramfs $ getfattr -m.* /run/initramfs user.foo After: $ getfattr -m.* /run/initramfs security.selinux $ getfattr -m.* /sys/kernel/fscaps security.selinux $ setfattr -n user.foo /run/initramfs $ getfattr -m.* /run/initramfs security.selinux user.foo Link: https://lore.kernel.org/selinux/CAFqZXNtF8wDyQajPCdGn=iOawX4y77ph0EcfcqcUUj+T87FKyA@mail.gmail.com/ Link: https://lore.kernel.org/selinux/20250423175728.3185-2-stephen.smalley.work@gmail.com/ Signed-off-by: Stephen Smalley Link: https://lore.kernel.org/20250424152822.2719-1-stephen.smalley.work@gmail.com Fixes: b09e0fa4b4ea66266058ee ("tmpfs: implement generic xattr support") Signed-off-by: Christian Brauner --- fs/xattr.c | 24 ++++++++++++++++++++++++ 1 file changed, 24 insertions(+) diff --git a/fs/xattr.c b/fs/xattr.c index fabb2a04501e..8ec5b0204bfd 100644 --- a/fs/xattr.c +++ b/fs/xattr.c @@ -1428,6 +1428,15 @@ static bool xattr_is_trusted(const char *name) return !strncmp(name, XATTR_TRUSTED_PREFIX, XATTR_TRUSTED_PREFIX_LEN); } +static bool xattr_is_maclabel(const char *name) +{ + const char *suffix = name + XATTR_SECURITY_PREFIX_LEN; + + return !strncmp(name, XATTR_SECURITY_PREFIX, + XATTR_SECURITY_PREFIX_LEN) && + security_ismaclabel(suffix); +} + /** * simple_xattr_list - list all xattr objects * @inode: inode from which to get the xattrs @@ -1460,6 +1469,17 @@ ssize_t simple_xattr_list(struct inode *inode, struct simple_xattrs *xattrs, if (err) return err; + err = security_inode_listsecurity(inode, buffer, remaining_size); + if (err < 0) + return err; + + if (buffer) { + if (remaining_size < err) + return -ERANGE; + buffer += err; + } + remaining_size -= err; + read_lock(&xattrs->lock); for (rbp = rb_first(&xattrs->rb_root); rbp; rbp = rb_next(rbp)) { xattr = rb_entry(rbp, struct simple_xattr, rb_node); @@ -1468,6 +1488,10 @@ ssize_t simple_xattr_list(struct inode *inode, struct simple_xattrs *xattrs, if (!trusted && xattr_is_trusted(xattr->name)) continue; + /* skip MAC labels; these are provided by LSM above */ + if (xattr_is_maclabel(xattr->name)) + continue; + err = xattr_list_one(&buffer, &remaining_size, xattr->name); if (err) break; From b71f9804f66c2592d4c3a2397b7374a4039005a5 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 18 Apr 2025 22:46:52 -0700 Subject: [PATCH 092/555] timekeeping: Prevent coarse clocks going backwards Lei Chen raised an issue with CLOCK_MONOTONIC_COARSE seeing time inconsistencies. Lei tracked down that this was being caused by the adjustment: tk->tkr_mono.xtime_nsec -= offset; which is made to compensate for the unaccumulated cycles in offset when the multiplicator is adjusted forward, so that the non-_COARSE clockids don't see inconsistencies. However, the _COARSE clockid getter functions use the adjusted xtime_nsec value directly and do not compensate the negative offset via the clocksource delta multiplied with the new multiplicator. In that case the caller can observe time going backwards in consecutive calls. By design, this negative adjustment should be fine, because the logic run from timekeeping_adjust() is done after it accumulated approximately multiplicator * interval_cycles into xtime_nsec. The accumulated value is always larger then the mult_adj * offset value, which is subtracted from xtime_nsec. Both operations are done together under the tk_core.lock, so the net change to xtime_nsec is always always be positive. However, do_adjtimex() calls into timekeeping_advance() as well, to apply the NTP frequency adjustment immediately. In this case, timekeeping_advance() does not return early when the offset is smaller then interval_cycles. In that case there is no time accumulated into xtime_nsec. But the subsequent call into timekeeping_adjust(), which modifies the multiplicator, subtracts from xtime_nsec to correct for the new multiplicator. Here because there was no accumulation, xtime_nsec becomes smaller than before, which opens a window up to the next accumulation, where the _COARSE clockid getters, which don't compensate for the offset, can observe the inconsistency. This has been tried to be fixed by forwarding the timekeeper in the case that adjtimex() adjusts the multiplier, which resets the offset to zero: 757b000f7b93 ("timekeeping: Fix possible inconsistencies in _COARSE clockids") That works correctly, but unfortunately causes a regression on the adjtimex() side. There are two issues: 1) The forwarding of the base time moves the update out of the original period and establishes a new one. 2) The clearing of the accumulated NTP error is changing the behaviour as well. User-space expects that multiplier/frequency updates are in effect, when the syscall returns, so delaying the update to the next tick is not solving the problem either. Commit 757b000f7b93 was reverted so that the established expectations of user space implementations (ntpd, chronyd) are restored, but that obviously brought the inconsistencies back. One of the initial approaches to fix this was to establish a separate storage for the coarse time getter nanoseconds part by calculating it from the offset. That was dropped on the floor because not having yet another state to maintain was simpler. But given the result of the above exercise, this solution turns out to be the right one. Bring it back in a slightly modified form. Thus introduce timekeeper::coarse_nsec and store that nanoseconds part in it, switch the time getter functions and the VDSO update to use that value. coarse_nsec is set on operations which forward or initialize the timekeeper and after time was accumulated during a tick. If there is no accumulation the timestamp is unchanged. This leaves the adjtimex() behaviour unmodified and prevents coarse time from going backwards. [ jstultz: Simplified the coarse_nsec calculation and kept behavior so coarse clockids aren't adjusted on each inter-tick adjtimex call, slightly reworked the comments and commit message ] Fixes: da15cfdae033 ("time: Introduce CLOCK_REALTIME_COARSE") Reported-by: Lei Chen Signed-off-by: Thomas Gleixner Signed-off-by: John Stultz Signed-off-by: Thomas Gleixner Signed-off-by: Ingo Molnar Link: https://lore.kernel.org/all/20250419054706.2319105-1-jstultz@google.com Closes: https://lore.kernel.org/lkml/20250310030004.3705801-1-lei.chen@smartx.com/ --- include/linux/timekeeper_internal.h | 8 +++-- kernel/time/timekeeping.c | 50 ++++++++++++++++++++++++----- kernel/time/vsyscall.c | 4 +-- 3 files changed, 49 insertions(+), 13 deletions(-) diff --git a/include/linux/timekeeper_internal.h b/include/linux/timekeeper_internal.h index e39d4d563b19..785048a3b3e6 100644 --- a/include/linux/timekeeper_internal.h +++ b/include/linux/timekeeper_internal.h @@ -51,7 +51,7 @@ struct tk_read_base { * @offs_real: Offset clock monotonic -> clock realtime * @offs_boot: Offset clock monotonic -> clock boottime * @offs_tai: Offset clock monotonic -> clock tai - * @tai_offset: The current UTC to TAI offset in seconds + * @coarse_nsec: The nanoseconds part for coarse time getters * @tkr_raw: The readout base structure for CLOCK_MONOTONIC_RAW * @raw_sec: CLOCK_MONOTONIC_RAW time in seconds * @clock_was_set_seq: The sequence number of clock was set events @@ -76,6 +76,7 @@ struct tk_read_base { * ntp shifted nano seconds. * @ntp_err_mult: Multiplication factor for scaled math conversion * @skip_second_overflow: Flag used to avoid updating NTP twice with same second + * @tai_offset: The current UTC to TAI offset in seconds * * Note: For timespec(64) based interfaces wall_to_monotonic is what * we need to add to xtime (or xtime corrected for sub jiffy times) @@ -100,7 +101,7 @@ struct tk_read_base { * which results in the following cacheline layout: * * 0: seqcount, tkr_mono - * 1: xtime_sec ... tai_offset + * 1: xtime_sec ... coarse_nsec * 2: tkr_raw, raw_sec * 3,4: Internal variables * @@ -121,7 +122,7 @@ struct timekeeper { ktime_t offs_real; ktime_t offs_boot; ktime_t offs_tai; - s32 tai_offset; + u32 coarse_nsec; /* Cacheline 2: */ struct tk_read_base tkr_raw; @@ -144,6 +145,7 @@ struct timekeeper { u32 ntp_error_shift; u32 ntp_err_mult; u32 skip_second_overflow; + s32 tai_offset; }; #ifdef CONFIG_GENERIC_TIME_VSYSCALL diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index 1e67d076f195..a009c91f7b05 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -164,10 +164,34 @@ static inline struct timespec64 tk_xtime(const struct timekeeper *tk) return ts; } +static inline struct timespec64 tk_xtime_coarse(const struct timekeeper *tk) +{ + struct timespec64 ts; + + ts.tv_sec = tk->xtime_sec; + ts.tv_nsec = tk->coarse_nsec; + return ts; +} + +/* + * Update the nanoseconds part for the coarse time keepers. They can't rely + * on xtime_nsec because xtime_nsec could be adjusted by a small negative + * amount when the multiplication factor of the clock is adjusted, which + * could cause the coarse clocks to go slightly backwards. See + * timekeeping_apply_adjustment(). Thus we keep a separate copy for the coarse + * clockids which only is updated when the clock has been set or we have + * accumulated time. + */ +static inline void tk_update_coarse_nsecs(struct timekeeper *tk) +{ + tk->coarse_nsec = tk->tkr_mono.xtime_nsec >> tk->tkr_mono.shift; +} + static void tk_set_xtime(struct timekeeper *tk, const struct timespec64 *ts) { tk->xtime_sec = ts->tv_sec; tk->tkr_mono.xtime_nsec = (u64)ts->tv_nsec << tk->tkr_mono.shift; + tk_update_coarse_nsecs(tk); } static void tk_xtime_add(struct timekeeper *tk, const struct timespec64 *ts) @@ -175,6 +199,7 @@ static void tk_xtime_add(struct timekeeper *tk, const struct timespec64 *ts) tk->xtime_sec += ts->tv_sec; tk->tkr_mono.xtime_nsec += (u64)ts->tv_nsec << tk->tkr_mono.shift; tk_normalize_xtime(tk); + tk_update_coarse_nsecs(tk); } static void tk_set_wall_to_mono(struct timekeeper *tk, struct timespec64 wtm) @@ -708,6 +733,7 @@ static void timekeeping_forward_now(struct timekeeper *tk) tk_normalize_xtime(tk); delta -= incr; } + tk_update_coarse_nsecs(tk); } /** @@ -804,8 +830,8 @@ EXPORT_SYMBOL_GPL(ktime_get_with_offset); ktime_t ktime_get_coarse_with_offset(enum tk_offsets offs) { struct timekeeper *tk = &tk_core.timekeeper; - unsigned int seq; ktime_t base, *offset = offsets[offs]; + unsigned int seq; u64 nsecs; WARN_ON(timekeeping_suspended); @@ -813,7 +839,7 @@ ktime_t ktime_get_coarse_with_offset(enum tk_offsets offs) do { seq = read_seqcount_begin(&tk_core.seq); base = ktime_add(tk->tkr_mono.base, *offset); - nsecs = tk->tkr_mono.xtime_nsec >> tk->tkr_mono.shift; + nsecs = tk->coarse_nsec; } while (read_seqcount_retry(&tk_core.seq, seq)); @@ -2161,7 +2187,7 @@ static bool timekeeping_advance(enum timekeeping_adv_mode mode) struct timekeeper *real_tk = &tk_core.timekeeper; unsigned int clock_set = 0; int shift = 0, maxshift; - u64 offset; + u64 offset, orig_offset; guard(raw_spinlock_irqsave)(&tk_core.lock); @@ -2172,7 +2198,7 @@ static bool timekeeping_advance(enum timekeeping_adv_mode mode) offset = clocksource_delta(tk_clock_read(&tk->tkr_mono), tk->tkr_mono.cycle_last, tk->tkr_mono.mask, tk->tkr_mono.clock->max_raw_delta); - + orig_offset = offset; /* Check if there's really nothing to do */ if (offset < real_tk->cycle_interval && mode == TK_ADV_TICK) return false; @@ -2205,6 +2231,14 @@ static bool timekeeping_advance(enum timekeeping_adv_mode mode) */ clock_set |= accumulate_nsecs_to_secs(tk); + /* + * To avoid inconsistencies caused adjtimex TK_ADV_FREQ calls + * making small negative adjustments to the base xtime_nsec + * value, only update the coarse clocks if we accumulated time + */ + if (orig_offset != offset) + tk_update_coarse_nsecs(tk); + timekeeping_update_from_shadow(&tk_core, clock_set); return !!clock_set; @@ -2248,7 +2282,7 @@ void ktime_get_coarse_real_ts64(struct timespec64 *ts) do { seq = read_seqcount_begin(&tk_core.seq); - *ts = tk_xtime(tk); + *ts = tk_xtime_coarse(tk); } while (read_seqcount_retry(&tk_core.seq, seq)); } EXPORT_SYMBOL(ktime_get_coarse_real_ts64); @@ -2271,7 +2305,7 @@ void ktime_get_coarse_real_ts64_mg(struct timespec64 *ts) do { seq = read_seqcount_begin(&tk_core.seq); - *ts = tk_xtime(tk); + *ts = tk_xtime_coarse(tk); offset = tk_core.timekeeper.offs_real; } while (read_seqcount_retry(&tk_core.seq, seq)); @@ -2350,12 +2384,12 @@ void ktime_get_coarse_ts64(struct timespec64 *ts) do { seq = read_seqcount_begin(&tk_core.seq); - now = tk_xtime(tk); + now = tk_xtime_coarse(tk); mono = tk->wall_to_monotonic; } while (read_seqcount_retry(&tk_core.seq, seq)); set_normalized_timespec64(ts, now.tv_sec + mono.tv_sec, - now.tv_nsec + mono.tv_nsec); + now.tv_nsec + mono.tv_nsec); } EXPORT_SYMBOL(ktime_get_coarse_ts64); diff --git a/kernel/time/vsyscall.c b/kernel/time/vsyscall.c index 01c2ab1e8971..32ef27c71b57 100644 --- a/kernel/time/vsyscall.c +++ b/kernel/time/vsyscall.c @@ -98,12 +98,12 @@ void update_vsyscall(struct timekeeper *tk) /* CLOCK_REALTIME_COARSE */ vdso_ts = &vc[CS_HRES_COARSE].basetime[CLOCK_REALTIME_COARSE]; vdso_ts->sec = tk->xtime_sec; - vdso_ts->nsec = tk->tkr_mono.xtime_nsec >> tk->tkr_mono.shift; + vdso_ts->nsec = tk->coarse_nsec; /* CLOCK_MONOTONIC_COARSE */ vdso_ts = &vc[CS_HRES_COARSE].basetime[CLOCK_MONOTONIC_COARSE]; vdso_ts->sec = tk->xtime_sec + tk->wall_to_monotonic.tv_sec; - nsec = tk->tkr_mono.xtime_nsec >> tk->tkr_mono.shift; + nsec = tk->coarse_nsec; nsec = nsec + tk->wall_to_monotonic.tv_nsec; vdso_ts->sec += __iter_div_u64_rem(nsec, NSEC_PER_SEC, &vdso_ts->nsec); From 8d16dd7b651b75ce7c79da3539553a25e60668f5 Mon Sep 17 00:00:00 2001 From: Hongbo Li Date: Thu, 24 Apr 2025 11:06:53 +0800 Subject: [PATCH 093/555] MAINTAINERS: erofs: add myself as reviewer I have a solid background in file systems and since much of my recent work has focused on EROFS, I am familiar with it. Now I have the time and am willing to help review EROFS patches. I hope my participation can be helpful to the EROFS patch review process. Signed-off-by: Hongbo Li Acked-by: Chao Yu Acked-by: Gao Xiang Link: https://lore.kernel.org/r/20250424030653.3308358-1-lihongbo22@huawei.com Signed-off-by: Gao Xiang --- MAINTAINERS | 1 + 1 file changed, 1 insertion(+) diff --git a/MAINTAINERS b/MAINTAINERS index 3cbf9ac0d83f..b53a268e4a92 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -8726,6 +8726,7 @@ M: Chao Yu R: Yue Hu R: Jeffle Xu R: Sandeep Dhavale +R: Hongbo Li L: linux-erofs@lists.ozlabs.org S: Maintained W: https://erofs.docs.kernel.org From fa7ab64f1e2fdc8f2603aab8e0dd20de89cb10d9 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Mon, 21 Apr 2025 14:43:34 -0400 Subject: [PATCH 094/555] NFS/localio: Fix a race in nfs_local_open_fh() Once the clp->cl_uuid.lock has been dropped, another CPU could come in and free the struct nfsd_file that was just added. To prevent that from happening, take the RCU read lock before dropping the spin lock. Fixes: 86e00412254a ("nfs: cache all open LOCALIO nfsd_file(s) in client") Signed-off-by: Trond Myklebust Reviewed-by: Mike Snitzer --- fs/nfs/localio.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/nfs/localio.c b/fs/nfs/localio.c index 5c21caeae075..4ec952f9f47d 100644 --- a/fs/nfs/localio.c +++ b/fs/nfs/localio.c @@ -278,6 +278,7 @@ nfs_local_open_fh(struct nfs_client *clp, const struct cred *cred, new = __nfs_local_open_fh(clp, cred, fh, nfl, mode); if (IS_ERR(new)) return NULL; + rcu_read_lock(); /* try to swap in the pointer */ spin_lock(&clp->cl_uuid.lock); nf = rcu_dereference_protected(*pnf, 1); @@ -287,7 +288,6 @@ nfs_local_open_fh(struct nfs_client *clp, const struct cred *cred, rcu_assign_pointer(*pnf, nf); } spin_unlock(&clp->cl_uuid.lock); - rcu_read_lock(); } nf = nfs_local_file_get(nf); rcu_read_unlock(); From 20a6cff3b283f0601048ace87ad1bc89627e36f2 Mon Sep 17 00:00:00 2001 From: Yan Zhao Date: Tue, 18 Mar 2025 09:33:33 +0800 Subject: [PATCH 095/555] KVM: x86/mmu: Check and free obsolete roots in kvm_mmu_reload() Check request KVM_REQ_MMU_FREE_OBSOLETE_ROOTS to free obsolete roots in kvm_mmu_reload() to prevent kvm_mmu_reload() from seeing a stale obsolete root. Since kvm_mmu_reload() can be called outside the vcpu_enter_guest() path (e.g., kvm_arch_vcpu_pre_fault_memory()), it may be invoked after a root has been marked obsolete and before vcpu_enter_guest() is invoked to process KVM_REQ_MMU_FREE_OBSOLETE_ROOTS and set root.hpa to invalid. This causes kvm_mmu_reload() to fail to load a new root, which can lead to kvm_arch_vcpu_pre_fault_memory() being stuck in the while loop in kvm_tdp_map_page() since RET_PF_RETRY is always returned due to is_page_fault_stale(). Keep the existing check of KVM_REQ_MMU_FREE_OBSOLETE_ROOTS in vcpu_enter_guest() since the cost of kvm_check_request() is negligible, especially a check that's guarded by kvm_request_pending(). Export symbol of kvm_mmu_free_obsolete_roots() as kvm_mmu_reload() is inline and may be called outside of kvm.ko. Fixes: 6e01b7601dfe ("KVM: x86: Implement kvm_arch_vcpu_pre_fault_memory()") Suggested-by: Sean Christopherson Signed-off-by: Yan Zhao Link: https://lore.kernel.org/r/20250318013333.5817-1-yan.y.zhao@intel.com Signed-off-by: Sean Christopherson --- arch/x86/kvm/mmu.h | 3 +++ arch/x86/kvm/mmu/mmu.c | 1 + 2 files changed, 4 insertions(+) diff --git a/arch/x86/kvm/mmu.h b/arch/x86/kvm/mmu.h index 050a0e229a4d..f2b36d32ef40 100644 --- a/arch/x86/kvm/mmu.h +++ b/arch/x86/kvm/mmu.h @@ -104,6 +104,9 @@ void kvm_mmu_track_write(struct kvm_vcpu *vcpu, gpa_t gpa, const u8 *new, static inline int kvm_mmu_reload(struct kvm_vcpu *vcpu) { + if (kvm_check_request(KVM_REQ_MMU_FREE_OBSOLETE_ROOTS, vcpu)) + kvm_mmu_free_obsolete_roots(vcpu); + /* * Checking root.hpa is sufficient even when KVM has mirror root. * We can have either: diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c index 63bb77ee1bb1..387464ebe8e8 100644 --- a/arch/x86/kvm/mmu/mmu.c +++ b/arch/x86/kvm/mmu/mmu.c @@ -5974,6 +5974,7 @@ void kvm_mmu_free_obsolete_roots(struct kvm_vcpu *vcpu) __kvm_mmu_free_obsolete_roots(vcpu->kvm, &vcpu->arch.root_mmu); __kvm_mmu_free_obsolete_roots(vcpu->kvm, &vcpu->arch.guest_mmu); } +EXPORT_SYMBOL_GPL(kvm_mmu_free_obsolete_roots); static u64 mmu_pte_write_fetch_gpte(struct kvm_vcpu *vcpu, gpa_t *gpa, int *bytes) From 9a046c1d21f0ae14c73b5e106e5a501dd902b6a9 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Wed, 9 Apr 2025 14:22:55 +0200 Subject: [PATCH 096/555] Input: stmpe-ts - use module alias instead of device table When compile tested with W=1 on x86_64 with driver as built-in: stmpe-ts.c:371:34: error: unused variable 'stmpe_ts_ids' [-Werror,-Wunused-const-variable] Ideally this would be referenced from the platform_driver, but since the compatible string is already matched by the mfd driver for its parent device, that would break probing. In this case, the of_device_id table just serves as a module alias for loading the driver, while the device itself is probed using the platform device name. Remove the table and instead use a module alias that reflects how the driver is actually probed. Link: https://lore.kernel.org/lkml/20240403080702.3509288-8-arnd@kernel.org/ Link: https://lore.kernel.org/lkml/181dbdb8-c050-4966-8cb4-2f39495ff3f9@app.fastmail.com/ Signed-off-by: Arnd Bergmann Reviewed-by: Krzysztof Kozlowski Link: https://lore.kernel.org/r/20250409122314.2848028-3-arnd@kernel.org Signed-off-by: Dmitry Torokhov --- drivers/input/touchscreen/stmpe-ts.c | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) diff --git a/drivers/input/touchscreen/stmpe-ts.c b/drivers/input/touchscreen/stmpe-ts.c index a94a1997f96b..af0fb38bcfdc 100644 --- a/drivers/input/touchscreen/stmpe-ts.c +++ b/drivers/input/touchscreen/stmpe-ts.c @@ -366,12 +366,7 @@ static struct platform_driver stmpe_ts_driver = { }; module_platform_driver(stmpe_ts_driver); -static const struct of_device_id stmpe_ts_ids[] = { - { .compatible = "st,stmpe-ts", }, - { }, -}; -MODULE_DEVICE_TABLE(of, stmpe_ts_ids); - +MODULE_ALIAS("platform:stmpe-ts"); MODULE_AUTHOR("Luotao Fu "); MODULE_DESCRIPTION("STMPEXXX touchscreen driver"); MODULE_LICENSE("GPL"); From b8ac485a179d8819ae291afbf13a2fd89d76d4f3 Mon Sep 17 00:00:00 2001 From: Mattijs Korpershoek Date: Mon, 28 Apr 2025 10:35:13 +0200 Subject: [PATCH 097/555] dt-bindings: mediatek,mt6779-keypad: Update Mattijs' email address Update Mattijs Korpershoek's email address to @kernel.org. Signed-off-by: Mattijs Korpershoek Acked-by: Rob Herring (Arm) Link: https://lore.kernel.org/r/20250428-keypad-email-v1-1-dde6ac76725b@kernel.org Signed-off-by: Dmitry Torokhov --- .../devicetree/bindings/input/mediatek,mt6779-keypad.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/input/mediatek,mt6779-keypad.yaml b/Documentation/devicetree/bindings/input/mediatek,mt6779-keypad.yaml index 517a4ac1bea3..e365413732e7 100644 --- a/Documentation/devicetree/bindings/input/mediatek,mt6779-keypad.yaml +++ b/Documentation/devicetree/bindings/input/mediatek,mt6779-keypad.yaml @@ -7,7 +7,7 @@ $schema: http://devicetree.org/meta-schemas/core.yaml# title: Mediatek's Keypad Controller maintainers: - - Mattijs Korpershoek + - Mattijs Korpershoek allOf: - $ref: /schemas/input/matrix-keymap.yaml# From 6a10a2f1e0502c1f23e3095291c985d9bd8c8488 Mon Sep 17 00:00:00 2001 From: Mattijs Korpershoek Date: Mon, 28 Apr 2025 10:40:17 +0200 Subject: [PATCH 098/555] MAINTAINERS: .mailmap: update Mattijs Korpershoek's email address Update Mattijs Korpershoek's email address to @kernel.org. Signed-off-by: Mattijs Korpershoek Link: https://lore.kernel.org/r/20250428-keypad-maintainers-v1-1-4e9c4afba415@kernel.org Signed-off-by: Dmitry Torokhov --- .mailmap | 1 + MAINTAINERS | 2 +- 2 files changed, 2 insertions(+), 1 deletion(-) diff --git a/.mailmap b/.mailmap index a897c16d3bae..86e8c8e55ed9 100644 --- a/.mailmap +++ b/.mailmap @@ -472,6 +472,7 @@ Matthias Fuchs Matthieu Baerts Matthieu CASTET Matti Vaittinen +Mattijs Korpershoek Matt Ranostay Matt Ranostay Matt Ranostay Matthew Ranostay diff --git a/MAINTAINERS b/MAINTAINERS index 82eb3de2d76c..765219b393db 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -14744,7 +14744,7 @@ F: Documentation/devicetree/bindings/media/mediatek-jpeg-*.yaml F: drivers/media/platform/mediatek/jpeg/ MEDIATEK KEYPAD DRIVER -M: Mattijs Korpershoek +M: Mattijs Korpershoek S: Supported F: Documentation/devicetree/bindings/input/mediatek,mt6779-keypad.yaml F: drivers/input/keyboard/mt6779-keypad.c From 7675b5efd81fe6d524e29d5a541f43201e98afa8 Mon Sep 17 00:00:00 2001 From: Mikael Gonella-Bolduc Date: Wed, 23 Apr 2025 09:52:43 -0400 Subject: [PATCH 099/555] Input: cyttsp5 - fix power control issue on wakeup The power control function ignores the "on" argument when setting the report ID, and thus is always sending HID_POWER_SLEEP. This causes a problem when trying to wakeup. Fix by sending the state variable, which contains the proper HID_POWER_ON or HID_POWER_SLEEP based on the "on" argument. Fixes: 3c98b8dbdced ("Input: cyttsp5 - implement proper sleep and wakeup procedures") Cc: stable@vger.kernel.org Signed-off-by: Mikael Gonella-Bolduc Signed-off-by: Hugo Villeneuve Reviewed-by: Alistair Francis Link: https://lore.kernel.org/r/20250423135243.1261460-1-hugo@hugovil.com Signed-off-by: Dmitry Torokhov --- drivers/input/touchscreen/cyttsp5.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/input/touchscreen/cyttsp5.c b/drivers/input/touchscreen/cyttsp5.c index 14c43f0a6c21..071b7c9bf566 100644 --- a/drivers/input/touchscreen/cyttsp5.c +++ b/drivers/input/touchscreen/cyttsp5.c @@ -580,7 +580,7 @@ static int cyttsp5_power_control(struct cyttsp5 *ts, bool on) int rc; SET_CMD_REPORT_TYPE(cmd[0], 0); - SET_CMD_REPORT_ID(cmd[0], HID_POWER_SLEEP); + SET_CMD_REPORT_ID(cmd[0], state); SET_CMD_OPCODE(cmd[1], HID_CMD_SET_POWER); rc = cyttsp5_write(ts, HID_COMMAND_REG, cmd, sizeof(cmd)); From 22cd66a5db56a07d9e621367cb4d16ff0f6baf56 Mon Sep 17 00:00:00 2001 From: Lode Willems Date: Tue, 22 Apr 2025 13:24:27 +0200 Subject: [PATCH 100/555] Input: xpad - add support for 8BitDo Ultimate 2 Wireless Controller This patch adds support for the 8BitDo Ultimate 2 Wireless Controller. Tested using the wireless dongle and plugged in. Signed-off-by: Lode Willems Link: https://lore.kernel.org/r/20250422112457.6728-1-me@lodewillems.com Cc: stable@vger.kernel.org Signed-off-by: Dmitry Torokhov --- drivers/input/joystick/xpad.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/input/joystick/xpad.c b/drivers/input/joystick/xpad.c index c33e6f33265b..c933e47173bd 100644 --- a/drivers/input/joystick/xpad.c +++ b/drivers/input/joystick/xpad.c @@ -387,6 +387,7 @@ static const struct xpad_device { { 0x2dc8, 0x3106, "8BitDo Ultimate Wireless / Pro 2 Wired Controller", 0, XTYPE_XBOX360 }, { 0x2dc8, 0x3109, "8BitDo Ultimate Wireless Bluetooth", 0, XTYPE_XBOX360 }, { 0x2dc8, 0x310a, "8BitDo Ultimate 2C Wireless Controller", 0, XTYPE_XBOX360 }, + { 0x2dc8, 0x310b, "8BitDo Ultimate 2 Wireless Controller", 0, XTYPE_XBOX360 }, { 0x2dc8, 0x6001, "8BitDo SN30 Pro", 0, XTYPE_XBOX360 }, { 0x2e24, 0x0652, "Hyperkin Duke X-Box One pad", 0, XTYPE_XBOXONE }, { 0x2e24, 0x1688, "Hyperkin X91 X-Box One pad", 0, XTYPE_XBOXONE }, From bbfe756dc3062c1e934f06e5ba39c239aa953b92 Mon Sep 17 00:00:00 2001 From: Max Kellermann Date: Tue, 29 Apr 2025 01:09:33 +0200 Subject: [PATCH 101/555] fs/erofs/fileio: call erofs_onlinefolio_split() after bio_add_folio() If bio_add_folio() fails (because it is full), erofs_fileio_scan_folio() needs to submit the I/O request via erofs_fileio_rq_submit() and allocate a new I/O request with an empty `struct bio`. Then it retries the bio_add_folio() call. However, at this point, erofs_onlinefolio_split() has already been called which increments `folio->private`; the retry will call erofs_onlinefolio_split() again, but there will never be a matching erofs_onlinefolio_end() call. This leaves the folio locked forever and all waiters will be stuck in folio_wait_bit_common(). This bug has been added by commit ce63cb62d794 ("erofs: support unencoded inodes for fileio"), but was practically unreachable because there was room for 256 folios in the `struct bio` - until commit 9f74ae8c9ac9 ("erofs: shorten bvecs[] for file-backed mounts") which reduced the array capacity to 16 folios. It was now trivial to trigger the bug by manually invoking readahead from userspace, e.g.: posix_fadvise(fd, 0, st.st_size, POSIX_FADV_WILLNEED); This should be fixed by invoking erofs_onlinefolio_split() only after bio_add_folio() has succeeded. This is safe: asynchronous completions invoking erofs_onlinefolio_end() will not unlock the folio because erofs_fileio_scan_folio() is still holding a reference to be released by erofs_onlinefolio_end() at the end. Fixes: ce63cb62d794 ("erofs: support unencoded inodes for fileio") Fixes: 9f74ae8c9ac9 ("erofs: shorten bvecs[] for file-backed mounts") Cc: stable@vger.kernel.org Signed-off-by: Max Kellermann Reviewed-by: Gao Xiang Tested-by: Hongbo Li Link: https://lore.kernel.org/r/20250428230933.3422273-1-max.kellermann@ionos.com Signed-off-by: Gao Xiang --- fs/erofs/fileio.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/erofs/fileio.c b/fs/erofs/fileio.c index 4fa0a0121288..60c7cc4c105c 100644 --- a/fs/erofs/fileio.c +++ b/fs/erofs/fileio.c @@ -150,10 +150,10 @@ io_retry: io->rq->bio.bi_iter.bi_sector = io->dev.m_pa >> 9; attached = 0; } - if (!attached++) - erofs_onlinefolio_split(folio); if (!bio_add_folio(&io->rq->bio, folio, len, cur)) goto io_retry; + if (!attached++) + erofs_onlinefolio_split(folio); io->dev.m_pa += len; } cur += len; From 690d722e02819ef978f90cd7553973eba1007e6c Mon Sep 17 00:00:00 2001 From: Mario Limonciello Date: Wed, 23 Apr 2025 08:18:44 -0500 Subject: [PATCH 102/555] drivers/platform/x86/amd: pmf: Check for invalid sideloaded Smart PC Policies MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit If a policy is passed into amd_pmf_get_pb_data() that causes the engine to fail to start there is a memory leak. Free the memory in this failure path. Fixes: 10817f28e5337 ("platform/x86/amd/pmf: Add capability to sideload of policy binary") Signed-off-by: Mario Limonciello Link: https://lore.kernel.org/r/20250423132002.3984997-2-superm1@kernel.org Reviewed-by: Ilpo Järvinen Signed-off-by: Ilpo Järvinen --- drivers/platform/x86/amd/pmf/tee-if.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/platform/x86/amd/pmf/tee-if.c b/drivers/platform/x86/amd/pmf/tee-if.c index 14b99d8b63d2..e008ac079fad 100644 --- a/drivers/platform/x86/amd/pmf/tee-if.c +++ b/drivers/platform/x86/amd/pmf/tee-if.c @@ -364,9 +364,14 @@ static ssize_t amd_pmf_get_pb_data(struct file *filp, const char __user *buf, amd_pmf_hex_dump_pb(dev); ret = amd_pmf_start_policy_engine(dev); if (ret < 0) - return ret; + goto cleanup; return length; + +cleanup: + kfree(dev->policy_buf); + dev->policy_buf = NULL; + return ret; } static const struct file_operations pb_fops = { From 8e81b9cd6e95188d12c9cc25d40b61dd5ea05ace Mon Sep 17 00:00:00 2001 From: Mario Limonciello Date: Wed, 23 Apr 2025 08:18:45 -0500 Subject: [PATCH 103/555] drivers/platform/x86/amd: pmf: Check for invalid Smart PC Policies MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit commit 376a8c2a14439 ("platform/x86/amd/pmf: Update PMF Driver for Compatibility with new PMF-TA") added support for platforms that support an updated TA, however it also exposed a number of platforms that although they have support for the updated TA don't actually populate a policy binary. Add an explicit check that the policy binary isn't empty before initializing the TA. Reported-by: Christian Heusel Closes: https://lore.kernel.org/platform-driver-x86/ae644428-5bf2-4b30-81ba-0b259ed3449b@heusel.eu/ Fixes: 376a8c2a14439 ("platform/x86/amd/pmf: Update PMF Driver for Compatibility with new PMF-TA") Signed-off-by: Mario Limonciello Tested-by: Christian Heusel Link: https://lore.kernel.org/r/20250423132002.3984997-3-superm1@kernel.org Reviewed-by: Ilpo Järvinen Signed-off-by: Ilpo Järvinen --- drivers/platform/x86/amd/pmf/tee-if.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/drivers/platform/x86/amd/pmf/tee-if.c b/drivers/platform/x86/amd/pmf/tee-if.c index e008ac079fad..d3bd12ad036a 100644 --- a/drivers/platform/x86/amd/pmf/tee-if.c +++ b/drivers/platform/x86/amd/pmf/tee-if.c @@ -334,6 +334,11 @@ static int amd_pmf_start_policy_engine(struct amd_pmf_dev *dev) return 0; } +static inline bool amd_pmf_pb_valid(struct amd_pmf_dev *dev) +{ + return memchr_inv(dev->policy_buf, 0xff, dev->policy_sz); +} + #ifdef CONFIG_AMD_PMF_DEBUG static void amd_pmf_hex_dump_pb(struct amd_pmf_dev *dev) { @@ -361,6 +366,11 @@ static ssize_t amd_pmf_get_pb_data(struct file *filp, const char __user *buf, dev->policy_buf = new_policy_buf; dev->policy_sz = length; + if (!amd_pmf_pb_valid(dev)) { + ret = -EINVAL; + goto cleanup; + } + amd_pmf_hex_dump_pb(dev); ret = amd_pmf_start_policy_engine(dev); if (ret < 0) @@ -533,6 +543,12 @@ int amd_pmf_init_smart_pc(struct amd_pmf_dev *dev) memcpy_fromio(dev->policy_buf, dev->policy_base, dev->policy_sz); + if (!amd_pmf_pb_valid(dev)) { + dev_info(dev->dev, "No Smart PC policy present\n"); + ret = -EINVAL; + goto err_free_policy; + } + amd_pmf_hex_dump_pb(dev); dev->prev_data = kzalloc(sizeof(*dev->prev_data), GFP_KERNEL); From 0581d384f344ed0a963dd27cbff3c7af80c189e7 Mon Sep 17 00:00:00 2001 From: Suma Hegde Date: Fri, 25 Apr 2025 10:23:57 +0000 Subject: [PATCH 104/555] platform/x86/amd/hsmp: Make amd_hsmp and hsmp_acpi as mutually exclusive drivers MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit amd_hsmp and hsmp_acpi are intended to be mutually exclusive drivers and amd_hsmp is for legacy platforms. To achieve this, it is essential to check for the presence of the ACPI device in plat.c. If the hsmp ACPI device entry is found, allow the hsmp_acpi driver to manage the hsmp and return an error from plat.c. Additionally, rename the driver from amd_hsmp to hsmp_acpi to prevent "Driver 'amd_hsmp' is already registered, aborting..." error in case both drivers are loaded simultaneously. Also, support both platform device based and ACPI based probing for family 0x1A models 0x00 to 0x0F, implement only ACPI based probing for family 0x1A, models 0x10 to 0x1F. Return false from legacy_hsmp_support() for this platform. This aligns with the condition check in is_f1a_m0h(). Link: https://lore.kernel.org/platform-driver-x86/aALZxvHWmphNL1wa@gourry-fedora-PF4VCD3F/ Fixes: 7d3135d16356 ("platform/x86/amd/hsmp: Create separate ACPI, plat and common drivers") Reviewed-by: Naveen Krishna Chatradhi Co-developed-by: Gregory Price Signed-off-by: Gregory Price Signed-off-by: Suma Hegde Link: https://lore.kernel.org/r/20250425102357.266790-1-suma.hegde@amd.com Reviewed-by: Ilpo Järvinen Signed-off-by: Ilpo Järvinen --- drivers/platform/x86/amd/hsmp/acpi.c | 3 +-- drivers/platform/x86/amd/hsmp/hsmp.h | 1 + drivers/platform/x86/amd/hsmp/plat.c | 6 +++++- 3 files changed, 7 insertions(+), 3 deletions(-) diff --git a/drivers/platform/x86/amd/hsmp/acpi.c b/drivers/platform/x86/amd/hsmp/acpi.c index c1eccb3c80c5..eaae044e4f82 100644 --- a/drivers/platform/x86/amd/hsmp/acpi.c +++ b/drivers/platform/x86/amd/hsmp/acpi.c @@ -27,9 +27,8 @@ #include "hsmp.h" -#define DRIVER_NAME "amd_hsmp" +#define DRIVER_NAME "hsmp_acpi" #define DRIVER_VERSION "2.3" -#define ACPI_HSMP_DEVICE_HID "AMDI0097" /* These are the strings specified in ACPI table */ #define MSG_IDOFF_STR "MsgIdOffset" diff --git a/drivers/platform/x86/amd/hsmp/hsmp.h b/drivers/platform/x86/amd/hsmp/hsmp.h index af8b21f821d6..d58d4f0c20d5 100644 --- a/drivers/platform/x86/amd/hsmp/hsmp.h +++ b/drivers/platform/x86/amd/hsmp/hsmp.h @@ -23,6 +23,7 @@ #define HSMP_CDEV_NAME "hsmp_cdev" #define HSMP_DEVNODE_NAME "hsmp" +#define ACPI_HSMP_DEVICE_HID "AMDI0097" struct hsmp_mbaddr_info { u32 base_addr; diff --git a/drivers/platform/x86/amd/hsmp/plat.c b/drivers/platform/x86/amd/hsmp/plat.c index b9782a078dbd..81931e808bbc 100644 --- a/drivers/platform/x86/amd/hsmp/plat.c +++ b/drivers/platform/x86/amd/hsmp/plat.c @@ -11,6 +11,7 @@ #include +#include #include #include #include @@ -266,7 +267,7 @@ static bool legacy_hsmp_support(void) } case 0x1A: switch (boot_cpu_data.x86_model) { - case 0x00 ... 0x1F: + case 0x00 ... 0x0F: return true; default: return false; @@ -288,6 +289,9 @@ static int __init hsmp_plt_init(void) return ret; } + if (acpi_dev_present(ACPI_HSMP_DEVICE_HID, NULL, -1)) + return -ENODEV; + hsmp_pdev = get_hsmp_pdev(); if (!hsmp_pdev) return -ENOMEM; From 11cdb506d0fbf5ac05bf55f5afcb3a215c316490 Mon Sep 17 00:00:00 2001 From: Gary Bisson Date: Tue, 29 Apr 2025 09:16:29 -0700 Subject: [PATCH 105/555] Input: mtk-pmic-keys - fix possible null pointer dereference In mtk_pmic_keys_probe, the regs parameter is only set if the button is parsed in the device tree. However, on hardware where the button is left floating, that node will most likely be removed not to enable that input. In that case the code will try to dereference a null pointer. Let's use the regs struct instead as it is defined for all supported platforms. Note that it is ok setting the key reg even if that latter is disabled as the interrupt won't be enabled anyway. Fixes: b581acb49aec ("Input: mtk-pmic-keys - transfer per-key bit in mtk_pmic_keys_regs") Signed-off-by: Gary Bisson Cc: stable@vger.kernel.org Signed-off-by: Dmitry Torokhov --- drivers/input/keyboard/mtk-pmic-keys.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/input/keyboard/mtk-pmic-keys.c b/drivers/input/keyboard/mtk-pmic-keys.c index 5ad6be914160..061d48350df6 100644 --- a/drivers/input/keyboard/mtk-pmic-keys.c +++ b/drivers/input/keyboard/mtk-pmic-keys.c @@ -147,8 +147,8 @@ static void mtk_pmic_keys_lp_reset_setup(struct mtk_pmic_keys *keys, u32 value, mask; int error; - kregs_home = keys->keys[MTK_PMIC_HOMEKEY_INDEX].regs; - kregs_pwr = keys->keys[MTK_PMIC_PWRKEY_INDEX].regs; + kregs_home = ®s->keys_regs[MTK_PMIC_HOMEKEY_INDEX]; + kregs_pwr = ®s->keys_regs[MTK_PMIC_PWRKEY_INDEX]; error = of_property_read_u32(keys->dev->of_node, "power-off-time-sec", &long_press_debounce); From e38be1c7647c8c78304ce6d931b3b654e27948b3 Mon Sep 17 00:00:00 2001 From: Andrea Righi Date: Mon, 28 Apr 2025 23:43:20 +0200 Subject: [PATCH 106/555] sched_ext: Fix rq lock state in hotplug ops The ops.cpu_online() and ops.cpu_offline() callbacks incorrectly assume that the rq involved in the operation is locked, which is not the case during hotplug, triggering the following warning: WARNING: CPU: 1 PID: 20 at kernel/sched/sched.h:1504 handle_hotplug+0x280/0x340 Fix by not tracking the target rq as locked in the context of ops.cpu_online() and ops.cpu_offline(). Fixes: 18853ba782bef ("sched_ext: Track currently locked rq") Reported-by: Tejun Heo Signed-off-by: Andrea Righi Tested-by: Changwoo Min Signed-off-by: Tejun Heo --- kernel/sched/ext.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c index dddb0af36f8d..4e37b40ce280 100644 --- a/kernel/sched/ext.c +++ b/kernel/sched/ext.c @@ -3477,9 +3477,9 @@ static void handle_hotplug(struct rq *rq, bool online) scx_idle_update_selcpu_topology(&scx_ops); if (online && SCX_HAS_OP(cpu_online)) - SCX_CALL_OP(SCX_KF_UNLOCKED, cpu_online, rq, cpu); + SCX_CALL_OP(SCX_KF_UNLOCKED, cpu_online, NULL, cpu); else if (!online && SCX_HAS_OP(cpu_offline)) - SCX_CALL_OP(SCX_KF_UNLOCKED, cpu_offline, rq, cpu); + SCX_CALL_OP(SCX_KF_UNLOCKED, cpu_offline, NULL, cpu); else scx_ops_exit(SCX_ECODE_ACT_RESTART | SCX_ECODE_RSN_HOTPLUG, "cpu %d going %s, exiting scheduler", cpu, From 4bf593be2e462623c4c34c7e3b604eb3f8f9de45 Mon Sep 17 00:00:00 2001 From: Nicolas Frattaroli Date: Tue, 29 Apr 2025 18:51:55 +0200 Subject: [PATCH 107/555] arm64: dts: rockchip: fix Sige5 RTC interrupt pin Someone made a typo when they added the RTC to the Sige5 DTS, which resulted in it using interrupts from GPIO0 B0 instead of GPIO0 A0. The pinctrl entry for it wasn't typoed though, curiously enough. The Sige5 v1.1 schematic was used to verify that GPIO0 A0 is the correct pin for the RTC wakeup interrupt, so let's change it to that. Fixes: 40f742b07ab2 ("arm64: dts: rockchip: Add rk3576-armsom-sige5 board") Signed-off-by: Nicolas Frattaroli Link: https://lore.kernel.org/r/20250429-sige5-rtc-oopsie-v1-1-8686767d0f1f@collabora.com Signed-off-by: Heiko Stuebner --- arch/arm64/boot/dts/rockchip/rk3576-armsom-sige5.dts | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm64/boot/dts/rockchip/rk3576-armsom-sige5.dts b/arch/arm64/boot/dts/rockchip/rk3576-armsom-sige5.dts index 828bde7fab68..314067ba6f3c 100644 --- a/arch/arm64/boot/dts/rockchip/rk3576-armsom-sige5.dts +++ b/arch/arm64/boot/dts/rockchip/rk3576-armsom-sige5.dts @@ -610,7 +610,7 @@ reg = <0x51>; clock-output-names = "hym8563"; interrupt-parent = <&gpio0>; - interrupts = ; + interrupts = ; pinctrl-names = "default"; pinctrl-0 = <&hym8563_int>; wakeup-source; From 9fcd53c3206618166facf03ee3d465f66a107e01 Mon Sep 17 00:00:00 2001 From: Hongbo Li Date: Tue, 29 Apr 2025 07:50:56 +0000 Subject: [PATCH 108/555] erofs: remove unused enum type Opt_err is not used in EROFS, we can remove it. Signed-off-by: Hongbo Li Reviewed-by: Gao Xiang Link: https://lore.kernel.org/r/20250429075056.689570-1-lihongbo22@huawei.com Signed-off-by: Gao Xiang --- fs/erofs/super.c | 1 - 1 file changed, 1 deletion(-) diff --git a/fs/erofs/super.c b/fs/erofs/super.c index cadec6b1b554..da6ee7c39290 100644 --- a/fs/erofs/super.c +++ b/fs/erofs/super.c @@ -357,7 +357,6 @@ static void erofs_default_options(struct erofs_sb_info *sbi) enum { Opt_user_xattr, Opt_acl, Opt_cache_strategy, Opt_dax, Opt_dax_enum, Opt_device, Opt_fsid, Opt_domain_id, Opt_directio, - Opt_err }; static const struct constant_table erofs_param_cache_strategy[] = { From 3e6244429ba38f8dee3336b8b805948276b281ab Mon Sep 17 00:00:00 2001 From: Ze Huang Date: Mon, 28 Apr 2025 17:24:36 +0800 Subject: [PATCH 109/555] riscv: dts: sophgo: fix DMA data-width configuration for CV18xx MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The "snps,data-width" property[1] defines the AXI data width of the DMA controller as: width = 8 × (2^n) bits (0 = 8 bits, 1 = 16 bits, 2 = 32 bits, ..., 6 = 512 bits) where "n" is the value of "snps,data-width". For the CV18xx DMA controller, the correct AXI data width is 32 bits, corresponding to "snps,data-width = 2". Test results on Milkv Duo S can be found here [2]. Link: https://github.com/torvalds/linux/blob/master/Documentation/devicetree/bindings/dma/snps%2Cdw-axi-dmac.yaml#L74 [1] Link: https://gist.github.com/Sutter099/4fa99bb2d89e5af975983124704b3861 [2] Fixes: 514951a81a5e ("riscv: dts: sophgo: cv18xx: add DMA controller") Co-developed-by: Yu Yuan Signed-off-by: Yu Yuan Signed-off-by: Ze Huang Link: https://lore.kernel.org/r/20250428-duo-dma-config-v1-1-eb6ad836ca42@whut.edu.cn Signed-off-by: Inochi Amaoto Signed-off-by: Chen Wang Signed-off-by: Chen Wang --- arch/riscv/boot/dts/sophgo/cv18xx.dtsi | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/riscv/boot/dts/sophgo/cv18xx.dtsi b/arch/riscv/boot/dts/sophgo/cv18xx.dtsi index c18822ec849f..58cd546392e0 100644 --- a/arch/riscv/boot/dts/sophgo/cv18xx.dtsi +++ b/arch/riscv/boot/dts/sophgo/cv18xx.dtsi @@ -341,7 +341,7 @@ 1024 1024 1024 1024>; snps,priority = <0 1 2 3 4 5 6 7>; snps,dma-masters = <2>; - snps,data-width = <4>; + snps,data-width = <2>; status = "disabled"; }; From c4eb2f88d2796ab90c5430e11c48709716181364 Mon Sep 17 00:00:00 2001 From: Jacek Lawrynowicz Date: Fri, 25 Apr 2025 11:28:22 +0200 Subject: [PATCH 110/555] accel/ivpu: Increase state dump msg timeout Increase JMS message state dump command timeout to 100 ms. On some platforms, the FW may take a bit longer than 50 ms to dump its state to the log buffer and we don't want to miss any debug info during TDR. Fixes: 5e162f872d7a ("accel/ivpu: Add FW state dump on TDR") Cc: stable@vger.kernel.org # v6.13+ Reviewed-by: Jeff Hugo Signed-off-by: Jacek Lawrynowicz Link: https://lore.kernel.org/r/20250425092822.2194465-1-jacek.lawrynowicz@linux.intel.com --- drivers/accel/ivpu/ivpu_hw.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/accel/ivpu/ivpu_hw.c b/drivers/accel/ivpu/ivpu_hw.c index ec9a3629da3a..633160470c93 100644 --- a/drivers/accel/ivpu/ivpu_hw.c +++ b/drivers/accel/ivpu/ivpu_hw.c @@ -119,7 +119,7 @@ static void timeouts_init(struct ivpu_device *vdev) else vdev->timeout.autosuspend = 100; vdev->timeout.d0i3_entry_msg = 5; - vdev->timeout.state_dump_msg = 10; + vdev->timeout.state_dump_msg = 100; } } From f2ecc700d1ef53127c14a3463442a8ac2a6a5cf5 Mon Sep 17 00:00:00 2001 From: Jacek Lawrynowicz Date: Fri, 25 Apr 2025 11:33:40 +0200 Subject: [PATCH 111/555] accel/ivpu: Fix pm related deadlocks in cmdq ioctls Fix deadlocks in ivpu_cmdq_create_ioctl() and ivpu_cmdq_destroy_ioctl() related to runtime suspend. Runtime suspend acquires file_priv->lock mutex by calling ivpu_cmdq_reset_all_contexts(). The same lock is acquired in the cmdq ioctls. If one of the cmdq ioctls is called while runtime suspend is in progress, it can lead to a deadlock. Call stacks from example deadlock below. Runtime suspend thread: [ 3443.179717] Call Trace: [ 3443.179724] __schedule+0x4b6/0x16b0 [ 3443.179732] ? __mod_timer+0x27d/0x3a0 [ 3443.179738] schedule+0x2f/0x140 [ 3443.179741] schedule_preempt_disabled+0x19/0x30 [ 3443.179743] __mutex_lock.constprop.0+0x335/0x7d0 [ 3443.179745] ? xas_find+0x1ed/0x260 [ 3443.179747] ? xa_find+0x8e/0xf0 [ 3443.179749] __mutex_lock_slowpath+0x13/0x20 [ 3443.179751] mutex_lock+0x41/0x60 [ 3443.179757] ivpu_cmdq_reset_all_contexts+0x82/0x150 [intel_vpu a9bd091a97f28f0235f161316b29f8234f437295] [ 3443.179786] ivpu_pm_runtime_suspend_cb+0x1f1/0x3f0 [intel_vpu a9bd091a97f28f0235f161316b29f8234f437295] [ 3443.179850] pci_pm_runtime_suspend+0x6e/0x1f0 [ 3443.179870] ? __pfx_pci_pm_runtime_suspend+0x10/0x10 [ 3443.179886] __rpm_callback+0x48/0x130 [ 3443.179899] rpm_callback+0x64/0x70 [ 3443.179911] rpm_suspend+0x12c/0x630 [ 3443.179922] ? __schedule+0x4be/0x16b0 [ 3443.179941] pm_runtime_work+0xca/0xf0 [ 3443.179955] process_one_work+0x188/0x3d0 [ 3443.179971] worker_thread+0x2b9/0x3c0 [ 3443.179984] kthread+0xfb/0x220 [ 3443.180001] ? __pfx_worker_thread+0x10/0x10 [ 3443.180013] ? __pfx_kthread+0x10/0x10 [ 3443.180029] ret_from_fork+0x47/0x70 [ 3443.180044] ? __pfx_kthread+0x10/0x10 [ 3443.180059] ret_from_fork_asm+0x1a/0x30 User space thread: [ 3443.180128] Call Trace: [ 3443.180138] __schedule+0x4b6/0x16b0 [ 3443.180159] schedule+0x2f/0x140 [ 3443.180163] rpm_resume+0x1a7/0x6a0 [ 3443.180165] ? __pfx_autoremove_wake_function+0x10/0x10 [ 3443.180169] __pm_runtime_resume+0x56/0x90 [ 3443.180171] ivpu_rpm_get+0x28/0xb0 [intel_vpu a9bd091a97f28f0235f161316b29f8234f437295] [ 3443.180181] ivpu_ipc_send_receive+0x6d/0x120 [intel_vpu a9bd091a97f28f0235f161316b29f8234f437295] [ 3443.180193] ? free_frozen_pages+0x395/0x670 [ 3443.180199] ? __free_pages+0xa7/0xc0 [ 3443.180202] ivpu_jsm_hws_destroy_cmdq+0x76/0xf0 [intel_vpu a9bd091a97f28f0235f161316b29f8234f437295] [ 3443.180213] ? locks_dispose_list+0x6c/0xa0 [ 3443.180219] ? kmem_cache_free+0x342/0x470 [ 3443.180222] ? vm_area_free+0x19/0x30 [ 3443.180225] ? xas_load+0x17/0xf0 [ 3443.180229] ? xa_load+0x72/0xb0 [ 3443.180230] ivpu_cmdq_unregister.isra.0+0xb1/0x100 [intel_vpu a9bd091a97f28f0235f161316b29f8234f437295] [ 3443.180241] ivpu_cmdq_destroy_ioctl+0x8d/0x130 [intel_vpu a9bd091a97f28f0235f161316b29f8234f437295] [ 3443.180251] ? __pfx_ivpu_cmdq_destroy_ioctl+0x10/0x10 [intel_vpu a9bd091a97f28f0235f161316b29f8234f437295] [ 3443.180260] drm_ioctl_kernel+0xb3/0x110 [ 3443.180265] drm_ioctl+0x2ca/0x580 [ 3443.180266] ? __pfx_ivpu_cmdq_destroy_ioctl+0x10/0x10 [intel_vpu a9bd091a97f28f0235f161316b29f8234f437295] [ 3443.180275] ? __fput+0x1ae/0x2f0 [ 3443.180279] ? kmem_cache_free+0x342/0x470 [ 3443.180282] __x64_sys_ioctl+0xa9/0xe0 [ 3443.180286] x64_sys_call+0x13b7/0x26f0 [ 3443.180289] do_syscall_64+0x62/0x180 [ 3443.180291] entry_SYSCALL_64_after_hwframe+0x71/0x79 Fixes: 465a3914b254 ("accel/ivpu: Add API for command queue create/destroy/submit") Reviewed-by: Jeff Hugo Signed-off-by: Jacek Lawrynowicz Link: https://lore.kernel.org/r/20250425093341.2202895-1-jacek.lawrynowicz@linux.intel.com --- drivers/accel/ivpu/ivpu_job.c | 33 ++++++++++++++++++++++++--------- 1 file changed, 24 insertions(+), 9 deletions(-) diff --git a/drivers/accel/ivpu/ivpu_job.c b/drivers/accel/ivpu/ivpu_job.c index 863e3cd6ace5..e17b3deda201 100644 --- a/drivers/accel/ivpu/ivpu_job.c +++ b/drivers/accel/ivpu/ivpu_job.c @@ -874,15 +874,21 @@ int ivpu_cmdq_submit_ioctl(struct drm_device *dev, void *data, struct drm_file * int ivpu_cmdq_create_ioctl(struct drm_device *dev, void *data, struct drm_file *file) { struct ivpu_file_priv *file_priv = file->driver_priv; + struct ivpu_device *vdev = file_priv->vdev; struct drm_ivpu_cmdq_create *args = data; struct ivpu_cmdq *cmdq; + int ret; - if (!ivpu_is_capable(file_priv->vdev, DRM_IVPU_CAP_MANAGE_CMDQ)) + if (!ivpu_is_capable(vdev, DRM_IVPU_CAP_MANAGE_CMDQ)) return -ENODEV; if (args->priority > DRM_IVPU_JOB_PRIORITY_REALTIME) return -EINVAL; + ret = ivpu_rpm_get(vdev); + if (ret < 0) + return ret; + mutex_lock(&file_priv->lock); cmdq = ivpu_cmdq_create(file_priv, ivpu_job_to_jsm_priority(args->priority), false); @@ -891,6 +897,8 @@ int ivpu_cmdq_create_ioctl(struct drm_device *dev, void *data, struct drm_file * mutex_unlock(&file_priv->lock); + ivpu_rpm_put(vdev); + return cmdq ? 0 : -ENOMEM; } @@ -900,28 +908,35 @@ int ivpu_cmdq_destroy_ioctl(struct drm_device *dev, void *data, struct drm_file struct ivpu_device *vdev = file_priv->vdev; struct drm_ivpu_cmdq_destroy *args = data; struct ivpu_cmdq *cmdq; - u32 cmdq_id; + u32 cmdq_id = 0; int ret; if (!ivpu_is_capable(vdev, DRM_IVPU_CAP_MANAGE_CMDQ)) return -ENODEV; + ret = ivpu_rpm_get(vdev); + if (ret < 0) + return ret; + mutex_lock(&file_priv->lock); cmdq = xa_load(&file_priv->cmdq_xa, args->cmdq_id); if (!cmdq || cmdq->is_legacy) { ret = -ENOENT; - goto err_unlock; + } else { + cmdq_id = cmdq->id; + ivpu_cmdq_destroy(file_priv, cmdq); + ret = 0; } - cmdq_id = cmdq->id; - ivpu_cmdq_destroy(file_priv, cmdq); mutex_unlock(&file_priv->lock); - ivpu_cmdq_abort_all_jobs(vdev, file_priv->ctx.id, cmdq_id); - return 0; -err_unlock: - mutex_unlock(&file_priv->lock); + /* Abort any pending jobs only if cmdq was destroyed */ + if (!ret) + ivpu_cmdq_abort_all_jobs(vdev, file_priv->ctx.id, cmdq_id); + + ivpu_rpm_put(vdev); + return ret; } From 75680b7cd461b169c7ccd2a0fba7542868b7fce2 Mon Sep 17 00:00:00 2001 From: Karol Wachowski Date: Fri, 25 Apr 2025 11:36:56 +0200 Subject: [PATCH 112/555] accel/ivpu: Correct mutex unlock order in job submission The mutex unlock for vdev->submitted_jobs_lock was incorrectly placed before unlocking file_priv->lock. Change order of unlocks to avoid potential race conditions. Fixes: 5bbccadaf33e ("accel/ivpu: Abort all jobs after command queue unregister") Signed-off-by: Karol Wachowski Reviewed-by: Jeff Hugo Signed-off-by: Jacek Lawrynowicz Link: https://lore.kernel.org/r/20250425093656.2228168-1-jacek.lawrynowicz@linux.intel.com --- drivers/accel/ivpu/ivpu_job.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/accel/ivpu/ivpu_job.c b/drivers/accel/ivpu/ivpu_job.c index e17b3deda201..b28da35c30b6 100644 --- a/drivers/accel/ivpu/ivpu_job.c +++ b/drivers/accel/ivpu/ivpu_job.c @@ -681,8 +681,8 @@ static int ivpu_job_submit(struct ivpu_job *job, u8 priority, u32 cmdq_id) err_erase_xa: xa_erase(&vdev->submitted_jobs_xa, job->job_id); err_unlock: - mutex_unlock(&vdev->submitted_jobs_lock); mutex_unlock(&file_priv->lock); + mutex_unlock(&vdev->submitted_jobs_lock); ivpu_rpm_put(vdev); return ret; } From c44572e0cc13c9afff83fd333135a0aa9b27ba26 Mon Sep 17 00:00:00 2001 From: Thorsten Blum Date: Sun, 27 Apr 2025 13:34:24 +0200 Subject: [PATCH 113/555] MIPS: Fix MAX_REG_OFFSET Fix MAX_REG_OFFSET to point to the last register in 'pt_regs' and not to the marker itself, which could allow regs_get_register() to return an invalid offset. Fixes: 40e084a506eb ("MIPS: Add uprobes support.") Suggested-by: Maciej W. Rozycki Signed-off-by: Thorsten Blum Signed-off-by: Thomas Bogendoerfer --- arch/mips/include/asm/ptrace.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/mips/include/asm/ptrace.h b/arch/mips/include/asm/ptrace.h index 85fa9962266a..ef72c46b5568 100644 --- a/arch/mips/include/asm/ptrace.h +++ b/arch/mips/include/asm/ptrace.h @@ -65,7 +65,8 @@ static inline void instruction_pointer_set(struct pt_regs *regs, /* Query offset/name of register from its name/offset */ extern int regs_query_register_offset(const char *name); -#define MAX_REG_OFFSET (offsetof(struct pt_regs, __last)) +#define MAX_REG_OFFSET \ + (offsetof(struct pt_regs, __last) - sizeof(unsigned long)) /** * regs_get_register() - get register value from its offset From 87ec7d5249bb8ebf40261420da069fa238c21789 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Radim=20Kr=C4=8Dm=C3=A1=C5=99?= Date: Thu, 3 Apr 2025 13:25:24 +0200 Subject: [PATCH 114/555] KVM: RISC-V: reset smstateen CSRs MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Not resetting smstateen is a potential security hole, because VU might be able to access state that VS does not properly context-switch. Fixes: 81f0f314fec9 ("RISCV: KVM: Add sstateen0 context save/restore") Signed-off-by: Radim Krčmář Link: https://lore.kernel.org/r/20250403112522.1566629-8-rkrcmar@ventanamicro.com Signed-off-by: Anup Patel --- arch/riscv/kvm/vcpu.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/riscv/kvm/vcpu.c b/arch/riscv/kvm/vcpu.c index 60d684c76c58..02635bac91f1 100644 --- a/arch/riscv/kvm/vcpu.c +++ b/arch/riscv/kvm/vcpu.c @@ -77,6 +77,8 @@ static void kvm_riscv_reset_vcpu(struct kvm_vcpu *vcpu) memcpy(cntx, reset_cntx, sizeof(*cntx)); spin_unlock(&vcpu->arch.reset_cntx_lock); + memset(&vcpu->arch.smstateen_csr, 0, sizeof(vcpu->arch.smstateen_csr)); + kvm_riscv_vcpu_fp_reset(vcpu); kvm_riscv_vcpu_vector_reset(vcpu); From 59820fde001500c167342257650541280c622b73 Mon Sep 17 00:00:00 2001 From: Wayne Chang Date: Fri, 18 Apr 2025 16:12:28 +0800 Subject: [PATCH 115/555] usb: gadget: tegra-xudc: ACK ST_RC after clearing CTRL_RUN We identified a bug where the ST_RC bit in the status register was not being acknowledged after clearing the CTRL_RUN bit in the control register. This could lead to unexpected behavior in the USB gadget drivers. This patch resolves the issue by adding the necessary code to explicitly acknowledge ST_RC after clearing CTRL_RUN based on the programming sequence, ensuring proper state transition. Fixes: 49db427232fe ("usb: gadget: Add UDC driver for tegra XUSB device mode controller") Cc: stable Signed-off-by: Wayne Chang Link: https://lore.kernel.org/r/20250418081228.1194779-1-waynec@nvidia.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/udc/tegra-xudc.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/usb/gadget/udc/tegra-xudc.c b/drivers/usb/gadget/udc/tegra-xudc.c index c7fdbc55fb0b..2957316fd3d0 100644 --- a/drivers/usb/gadget/udc/tegra-xudc.c +++ b/drivers/usb/gadget/udc/tegra-xudc.c @@ -1749,6 +1749,10 @@ static int __tegra_xudc_ep_disable(struct tegra_xudc_ep *ep) val = xudc_readl(xudc, CTRL); val &= ~CTRL_RUN; xudc_writel(xudc, val, CTRL); + + val = xudc_readl(xudc, ST); + if (val & ST_RC) + xudc_writel(xudc, ST_RC, ST); } dev_info(xudc->dev, "ep %u disabled\n", ep->index); From 241e2ce88e5a494be7a5d44c0697592f1632fbee Mon Sep 17 00:00:00 2001 From: Pawel Laszczak Date: Fri, 18 Apr 2025 04:55:16 +0000 Subject: [PATCH 116/555] usb: cdnsp: Fix issue with resuming from L1 In very rare cases after resuming controller from L1 to L0 it reads registers before the clock UTMI have been enabled and as the result driver reads incorrect value. Most of registers are in APB domain clock but some of them (e.g. PORTSC) are in UTMI domain clock. After entering to L1 state the UTMI clock can be disabled. When controller transition from L1 to L0 the port status change event is reported and in interrupt runtime function driver reads PORTSC. During this read operation controller synchronize UTMI and APB domain but UTMI clock is still disabled and in result it reads 0xFFFFFFFF value. To fix this issue driver increases APB timeout value. The issue is platform specific and if the default value of APB timeout is not sufficient then this time should be set Individually for each platform. Fixes: 3d82904559f4 ("usb: cdnsp: cdns3 Add main part of Cadence USBSSP DRD Driver") Cc: stable Signed-off-by: Pawel Laszczak Acked-by: Peter Chen Link: https://lore.kernel.org/r/PH7PR07MB953846C57973E4DB134CAA71DDBF2@PH7PR07MB9538.namprd07.prod.outlook.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/cdns3/cdnsp-gadget.c | 29 +++++++++++++++++++++++++++++ drivers/usb/cdns3/cdnsp-gadget.h | 3 +++ drivers/usb/cdns3/cdnsp-pci.c | 12 ++++++++++-- drivers/usb/cdns3/core.h | 3 +++ 4 files changed, 45 insertions(+), 2 deletions(-) diff --git a/drivers/usb/cdns3/cdnsp-gadget.c b/drivers/usb/cdns3/cdnsp-gadget.c index 87f310841735..7f5534db2086 100644 --- a/drivers/usb/cdns3/cdnsp-gadget.c +++ b/drivers/usb/cdns3/cdnsp-gadget.c @@ -139,6 +139,26 @@ static void cdnsp_clear_port_change_bit(struct cdnsp_device *pdev, (portsc & PORT_CHANGE_BITS), port_regs); } +static void cdnsp_set_apb_timeout_value(struct cdnsp_device *pdev) +{ + struct cdns *cdns = dev_get_drvdata(pdev->dev); + __le32 __iomem *reg; + void __iomem *base; + u32 offset = 0; + u32 val; + + if (!cdns->override_apb_timeout) + return; + + base = &pdev->cap_regs->hc_capbase; + offset = cdnsp_find_next_ext_cap(base, offset, D_XEC_PRE_REGS_CAP); + reg = base + offset + REG_CHICKEN_BITS_3_OFFSET; + + val = le32_to_cpu(readl(reg)); + val = CHICKEN_APB_TIMEOUT_SET(val, cdns->override_apb_timeout); + writel(cpu_to_le32(val), reg); +} + static void cdnsp_set_chicken_bits_2(struct cdnsp_device *pdev, u32 bit) { __le32 __iomem *reg; @@ -1798,6 +1818,15 @@ static int cdnsp_gen_setup(struct cdnsp_device *pdev) pdev->hci_version = HC_VERSION(pdev->hcc_params); pdev->hcc_params = readl(&pdev->cap_regs->hcc_params); + /* + * Override the APB timeout value to give the controller more time for + * enabling UTMI clock and synchronizing APB and UTMI clock domains. + * This fix is platform specific and is required to fixes issue with + * reading incorrect value from PORTSC register after resuming + * from L1 state. + */ + cdnsp_set_apb_timeout_value(pdev); + cdnsp_get_rev_cap(pdev); /* Make sure the Device Controller is halted. */ diff --git a/drivers/usb/cdns3/cdnsp-gadget.h b/drivers/usb/cdns3/cdnsp-gadget.h index 84887dfea763..87ac0cd113e7 100644 --- a/drivers/usb/cdns3/cdnsp-gadget.h +++ b/drivers/usb/cdns3/cdnsp-gadget.h @@ -520,6 +520,9 @@ struct cdnsp_rev_cap { #define REG_CHICKEN_BITS_2_OFFSET 0x48 #define CHICKEN_XDMA_2_TP_CACHE_DIS BIT(28) +#define REG_CHICKEN_BITS_3_OFFSET 0x4C +#define CHICKEN_APB_TIMEOUT_SET(p, val) (((p) & ~GENMASK(21, 0)) | (val)) + /* XBUF Extended Capability ID. */ #define XBUF_CAP_ID 0xCB #define XBUF_RX_TAG_MASK_0_OFFSET 0x1C diff --git a/drivers/usb/cdns3/cdnsp-pci.c b/drivers/usb/cdns3/cdnsp-pci.c index a51144504ff3..8c361b8394e9 100644 --- a/drivers/usb/cdns3/cdnsp-pci.c +++ b/drivers/usb/cdns3/cdnsp-pci.c @@ -28,6 +28,8 @@ #define PCI_DRIVER_NAME "cdns-pci-usbssp" #define PLAT_DRIVER_NAME "cdns-usbssp" +#define CHICKEN_APB_TIMEOUT_VALUE 0x1C20 + static struct pci_dev *cdnsp_get_second_fun(struct pci_dev *pdev) { /* @@ -139,6 +141,14 @@ static int cdnsp_pci_probe(struct pci_dev *pdev, cdnsp->otg_irq = pdev->irq; } + /* + * Cadence PCI based platform require some longer timeout for APB + * to fixes domain clock synchronization issue after resuming + * controller from L1 state. + */ + cdnsp->override_apb_timeout = CHICKEN_APB_TIMEOUT_VALUE; + pci_set_drvdata(pdev, cdnsp); + if (pci_is_enabled(func)) { cdnsp->dev = dev; cdnsp->gadget_init = cdnsp_gadget_init; @@ -148,8 +158,6 @@ static int cdnsp_pci_probe(struct pci_dev *pdev, goto free_cdnsp; } - pci_set_drvdata(pdev, cdnsp); - device_wakeup_enable(&pdev->dev); if (pci_dev_run_wake(pdev)) pm_runtime_put_noidle(&pdev->dev); diff --git a/drivers/usb/cdns3/core.h b/drivers/usb/cdns3/core.h index 921cccf1ca9d..801be9e61340 100644 --- a/drivers/usb/cdns3/core.h +++ b/drivers/usb/cdns3/core.h @@ -79,6 +79,8 @@ struct cdns3_platform_data { * @pdata: platform data from glue layer * @lock: spinlock structure * @xhci_plat_data: xhci private data structure pointer + * @override_apb_timeout: hold value of APB timeout. For value 0 the default + * value in CHICKEN_BITS_3 will be preserved. * @gadget_init: pointer to gadget initialization function */ struct cdns { @@ -117,6 +119,7 @@ struct cdns { struct cdns3_platform_data *pdata; spinlock_t lock; struct xhci_plat_priv *xhci_plat_data; + u32 override_apb_timeout; int (*gadget_init)(struct cdns *cdns); }; From 732f35cf8bdfece582f6e4a9c659119036577308 Mon Sep 17 00:00:00 2001 From: Jim Lin Date: Tue, 22 Apr 2025 19:40:01 +0800 Subject: [PATCH 117/555] usb: host: tegra: Prevent host controller crash when OTG port is used When a USB device is connected to the OTG port, the tegra_xhci_id_work() routine transitions the PHY to host mode and calls xhci_hub_control() with the SetPortFeature command to enable port power. In certain cases, the XHCI controller may be in a low-power state when this operation occurs. If xhci_hub_control() is invoked while the controller is suspended, the PORTSC register may return 0xFFFFFFFF, indicating a read failure. This causes xhci_hc_died() to be triggered, leading to host controller shutdown. Example backtrace: [ 105.445736] Workqueue: events tegra_xhci_id_work [ 105.445747] dump_backtrace+0x0/0x1e8 [ 105.445759] xhci_hc_died.part.48+0x40/0x270 [ 105.445769] tegra_xhci_set_port_power+0xc0/0x240 [ 105.445774] tegra_xhci_id_work+0x130/0x240 To prevent this, ensure the controller is fully resumed before interacting with hardware registers by calling pm_runtime_get_sync() prior to the host mode transition and xhci_hub_control(). Fixes: f836e7843036 ("usb: xhci-tegra: Add OTG support") Cc: stable Signed-off-by: Jim Lin Signed-off-by: Wayne Chang Link: https://lore.kernel.org/r/20250422114001.126367-1-waynec@nvidia.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/host/xhci-tegra.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/usb/host/xhci-tegra.c b/drivers/usb/host/xhci-tegra.c index b5c362c2051d..0c7af44d4dae 100644 --- a/drivers/usb/host/xhci-tegra.c +++ b/drivers/usb/host/xhci-tegra.c @@ -1364,6 +1364,7 @@ static void tegra_xhci_id_work(struct work_struct *work) tegra->otg_usb3_port = tegra_xusb_padctl_get_usb3_companion(tegra->padctl, tegra->otg_usb2_port); + pm_runtime_get_sync(tegra->dev); if (tegra->host_mode) { /* switch to host mode */ if (tegra->otg_usb3_port >= 0) { @@ -1393,6 +1394,7 @@ static void tegra_xhci_id_work(struct work_struct *work) } tegra_xhci_set_port_power(tegra, true, true); + pm_runtime_mark_last_busy(tegra->dev); } else { if (tegra->otg_usb3_port >= 0) @@ -1400,6 +1402,7 @@ static void tegra_xhci_id_work(struct work_struct *work) tegra_xhci_set_port_power(tegra, true, false); } + pm_runtime_put_autosuspend(tegra->dev); } #if IS_ENABLED(CONFIG_PM) || IS_ENABLED(CONFIG_PM_SLEEP) From 8e3820271c517ceb89ab7442656ba49fa23ee1d0 Mon Sep 17 00:00:00 2001 From: Prashanth K Date: Tue, 22 Apr 2025 16:02:29 +0530 Subject: [PATCH 118/555] usb: gadget: f_ecm: Add get_status callback When host sends GET_STATUS to ECM interface, handle the request from the function driver. Since the interface is wakeup capable, set the corresponding bit, and set RW bit if the function is already armed for wakeup by the host. Cc: stable Fixes: 481c225c4802 ("usb: gadget: Handle function suspend feature selector") Signed-off-by: Prashanth K Reviewed-by: Thinh Nguyen Link: https://lore.kernel.org/r/20250422103231.1954387-2-prashanth.k@oss.qualcomm.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/function/f_ecm.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/usb/gadget/function/f_ecm.c b/drivers/usb/gadget/function/f_ecm.c index 80841de845b0..027226325039 100644 --- a/drivers/usb/gadget/function/f_ecm.c +++ b/drivers/usb/gadget/function/f_ecm.c @@ -892,6 +892,12 @@ static void ecm_resume(struct usb_function *f) gether_resume(&ecm->port); } +static int ecm_get_status(struct usb_function *f) +{ + return (f->func_wakeup_armed ? USB_INTRF_STAT_FUNC_RW : 0) | + USB_INTRF_STAT_FUNC_RW_CAP; +} + static void ecm_free(struct usb_function *f) { struct f_ecm *ecm; @@ -960,6 +966,7 @@ static struct usb_function *ecm_alloc(struct usb_function_instance *fi) ecm->port.func.disable = ecm_disable; ecm->port.func.free_func = ecm_free; ecm->port.func.suspend = ecm_suspend; + ecm->port.func.get_status = ecm_get_status; ecm->port.func.resume = ecm_resume; return &ecm->port.func; From 5977a58dd5a4865198b0204b998adb0f634abe19 Mon Sep 17 00:00:00 2001 From: Prashanth K Date: Tue, 22 Apr 2025 16:02:30 +0530 Subject: [PATCH 119/555] usb: gadget: Use get_status callback to set remote wakeup capability Currently when the host sends GET_STATUS request for an interface, we use get_status callbacks to set/clear remote wakeup capability of that interface. And if get_status callback isn't present for that interface, then we assume its remote wakeup capability based on bmAttributes. Now consider a scenario, where we have a USB configuration with multiple interfaces (say ECM + ADB), here ECM is remote wakeup capable and as of now ADB isn't. And bmAttributes will indicate the device as wakeup capable. With the current implementation, when host sends GET_STATUS request for both interfaces, we will set FUNC_RW_CAP for both. This results in USB3 CV Chapter 9.15 (Function Remote Wakeup Test) failures as host expects remote wakeup from both interfaces. The above scenario is just an example, and the failure can be observed if we use configuration with any interface except ECM. Hence avoid configuring remote wakeup capability from composite driver based on bmAttributes, instead use get_status callbacks and let the function drivers decide this. Cc: stable Fixes: 481c225c4802 ("usb: gadget: Handle function suspend feature selector") Signed-off-by: Prashanth K Reviewed-by: Thinh Nguyen Link: https://lore.kernel.org/r/20250422103231.1954387-3-prashanth.k@oss.qualcomm.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/composite.c | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) diff --git a/drivers/usb/gadget/composite.c b/drivers/usb/gadget/composite.c index 869ad99afb48..8dbc132a505e 100644 --- a/drivers/usb/gadget/composite.c +++ b/drivers/usb/gadget/composite.c @@ -2011,15 +2011,13 @@ composite_setup(struct usb_gadget *gadget, const struct usb_ctrlrequest *ctrl) if (f->get_status) { status = f->get_status(f); + if (status < 0) break; - } else { - /* Set D0 and D1 bits based on func wakeup capability */ - if (f->config->bmAttributes & USB_CONFIG_ATT_WAKEUP) { - status |= USB_INTRF_STAT_FUNC_RW_CAP; - if (f->func_wakeup_armed) - status |= USB_INTRF_STAT_FUNC_RW; - } + + /* if D5 is not set, then device is not wakeup capable */ + if (!(f->config->bmAttributes & USB_CONFIG_ATT_WAKEUP)) + status &= ~(USB_INTRF_STAT_FUNC_RW_CAP | USB_INTRF_STAT_FUNC_RW); } put_unaligned_le16(status & 0x0000ffff, req->buf); From 2372f1caeca433c4c01c2482f73fbe057f5168ce Mon Sep 17 00:00:00 2001 From: Prashanth K Date: Tue, 22 Apr 2025 16:02:31 +0530 Subject: [PATCH 120/555] usb: dwc3: gadget: Make gadget_wakeup asynchronous Currently gadget_wakeup() waits for U0 synchronously if it was called from func_wakeup(), this is because we need to send the function wakeup command soon after the link is active. And the call is made synchronous by polling DSTS continuosly for 20000 times in __dwc3_gadget_wakeup(). But it observed that sometimes the link is not active even after polling 20K times, leading to remote wakeup failures. Adding a small delay between each poll helps, but that won't guarantee resolution in future. Hence make the gadget_wakeup completely asynchronous. Since multiple interfaces can issue a function wakeup at once, add a new variable wakeup_pending_funcs which will indicate the functions that has issued func_wakup, this is represented in a bitmap format. If the link is in U3, dwc3_gadget_func_wakeup() will set the bit corresponding to interface_id and bail out. Once link comes back to U0, linksts_change irq is triggered, where the function wakeup command is sent based on bitmap. Cc: stable Fixes: 92c08a84b53e ("usb: dwc3: Add function suspend and function wakeup support") Signed-off-by: Prashanth K Acked-by: Thinh Nguyen Link: https://lore.kernel.org/r/20250422103231.1954387-4-prashanth.k@oss.qualcomm.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/dwc3/core.h | 4 +++ drivers/usb/dwc3/gadget.c | 60 +++++++++++++++------------------------ 2 files changed, 27 insertions(+), 37 deletions(-) diff --git a/drivers/usb/dwc3/core.h b/drivers/usb/dwc3/core.h index aaa39e663f60..27eae4cf223d 100644 --- a/drivers/usb/dwc3/core.h +++ b/drivers/usb/dwc3/core.h @@ -1164,6 +1164,9 @@ struct dwc3_scratchpad_array { * @gsbuscfg0_reqinfo: store GSBUSCFG0.DATRDREQINFO, DESRDREQINFO, * DATWRREQINFO, and DESWRREQINFO value passed from * glue driver. + * @wakeup_pending_funcs: Indicates whether any interface has requested for + * function wakeup in bitmap format where bit position + * represents interface_id. */ struct dwc3 { struct work_struct drd_work; @@ -1394,6 +1397,7 @@ struct dwc3 { int num_ep_resized; struct dentry *debug_root; u32 gsbuscfg0_reqinfo; + u32 wakeup_pending_funcs; }; #define INCRX_BURST_MODE 0 diff --git a/drivers/usb/dwc3/gadget.c b/drivers/usb/dwc3/gadget.c index 8c30d86cc4e3..321361288935 100644 --- a/drivers/usb/dwc3/gadget.c +++ b/drivers/usb/dwc3/gadget.c @@ -276,8 +276,6 @@ int dwc3_send_gadget_generic_command(struct dwc3 *dwc, unsigned int cmd, return ret; } -static int __dwc3_gadget_wakeup(struct dwc3 *dwc, bool async); - /** * dwc3_send_gadget_ep_cmd - issue an endpoint command * @dep: the endpoint to which the command is going to be issued @@ -2359,10 +2357,8 @@ static int dwc3_gadget_get_frame(struct usb_gadget *g) return __dwc3_gadget_get_frame(dwc); } -static int __dwc3_gadget_wakeup(struct dwc3 *dwc, bool async) +static int __dwc3_gadget_wakeup(struct dwc3 *dwc) { - int retries; - int ret; u32 reg; @@ -2390,8 +2386,7 @@ static int __dwc3_gadget_wakeup(struct dwc3 *dwc, bool async) return -EINVAL; } - if (async) - dwc3_gadget_enable_linksts_evts(dwc, true); + dwc3_gadget_enable_linksts_evts(dwc, true); ret = dwc3_gadget_set_link_state(dwc, DWC3_LINK_STATE_RECOV); if (ret < 0) { @@ -2410,27 +2405,8 @@ static int __dwc3_gadget_wakeup(struct dwc3 *dwc, bool async) /* * Since link status change events are enabled we will receive - * an U0 event when wakeup is successful. So bail out. + * an U0 event when wakeup is successful. */ - if (async) - return 0; - - /* poll until Link State changes to ON */ - retries = 20000; - - while (retries--) { - reg = dwc3_readl(dwc->regs, DWC3_DSTS); - - /* in HS, means ON */ - if (DWC3_DSTS_USBLNKST(reg) == DWC3_LINK_STATE_U0) - break; - } - - if (DWC3_DSTS_USBLNKST(reg) != DWC3_LINK_STATE_U0) { - dev_err(dwc->dev, "failed to send remote wakeup\n"); - return -EINVAL; - } - return 0; } @@ -2451,7 +2427,7 @@ static int dwc3_gadget_wakeup(struct usb_gadget *g) spin_unlock_irqrestore(&dwc->lock, flags); return -EINVAL; } - ret = __dwc3_gadget_wakeup(dwc, true); + ret = __dwc3_gadget_wakeup(dwc); spin_unlock_irqrestore(&dwc->lock, flags); @@ -2479,14 +2455,10 @@ static int dwc3_gadget_func_wakeup(struct usb_gadget *g, int intf_id) */ link_state = dwc3_gadget_get_link_state(dwc); if (link_state == DWC3_LINK_STATE_U3) { - ret = __dwc3_gadget_wakeup(dwc, false); - if (ret) { - spin_unlock_irqrestore(&dwc->lock, flags); - return -EINVAL; - } - dwc3_resume_gadget(dwc); - dwc->suspended = false; - dwc->link_state = DWC3_LINK_STATE_U0; + dwc->wakeup_pending_funcs |= BIT(intf_id); + ret = __dwc3_gadget_wakeup(dwc); + spin_unlock_irqrestore(&dwc->lock, flags); + return ret; } ret = dwc3_send_gadget_generic_command(dwc, DWC3_DGCMD_DEV_NOTIFICATION, @@ -4353,6 +4325,8 @@ static void dwc3_gadget_linksts_change_interrupt(struct dwc3 *dwc, { enum dwc3_link_state next = evtinfo & DWC3_LINK_STATE_MASK; unsigned int pwropt; + int ret; + int intf_id; /* * WORKAROUND: DWC3 < 2.50a have an issue when configured without @@ -4428,7 +4402,7 @@ static void dwc3_gadget_linksts_change_interrupt(struct dwc3 *dwc, switch (next) { case DWC3_LINK_STATE_U0: - if (dwc->gadget->wakeup_armed) { + if (dwc->gadget->wakeup_armed || dwc->wakeup_pending_funcs) { dwc3_gadget_enable_linksts_evts(dwc, false); dwc3_resume_gadget(dwc); dwc->suspended = false; @@ -4451,6 +4425,18 @@ static void dwc3_gadget_linksts_change_interrupt(struct dwc3 *dwc, } dwc->link_state = next; + + /* Proceed with func wakeup if any interfaces that has requested */ + while (dwc->wakeup_pending_funcs && (next == DWC3_LINK_STATE_U0)) { + intf_id = ffs(dwc->wakeup_pending_funcs) - 1; + ret = dwc3_send_gadget_generic_command(dwc, DWC3_DGCMD_DEV_NOTIFICATION, + DWC3_DGCMDPAR_DN_FUNC_WAKE | + DWC3_DGCMDPAR_INTF_SEL(intf_id)); + if (ret) + dev_err(dwc->dev, "Failed to send DN wake for intf %d\n", intf_id); + + dwc->wakeup_pending_funcs &= ~BIT(intf_id); + } } static void dwc3_gadget_suspend_interrupt(struct dwc3 *dwc, From a5c7973539b010874a37a0e846e62ac6f00553ba Mon Sep 17 00:00:00 2001 From: Alexey Charkov Date: Fri, 25 Apr 2025 18:11:11 +0400 Subject: [PATCH 121/555] usb: uhci-platform: Make the clock really optional Device tree bindings state that the clock is optional for UHCI platform controllers, and some existing device trees don't provide those - such as those for VIA/WonderMedia devices. The driver however fails to probe now if no clock is provided, because devm_clk_get returns an error pointer in such case. Switch to devm_clk_get_optional instead, so that it could probe again on those platforms where no clocks are given. Cc: stable Fixes: 26c502701c52 ("usb: uhci: Add clk support to uhci-platform") Signed-off-by: Alexey Charkov Link: https://lore.kernel.org/r/20250425-uhci-clock-optional-v1-1-a1d462592f29@gmail.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/host/uhci-platform.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/usb/host/uhci-platform.c b/drivers/usb/host/uhci-platform.c index a7c934404ebc..62318291f566 100644 --- a/drivers/usb/host/uhci-platform.c +++ b/drivers/usb/host/uhci-platform.c @@ -121,7 +121,7 @@ static int uhci_hcd_platform_probe(struct platform_device *pdev) } /* Get and enable clock if any specified */ - uhci->clk = devm_clk_get(&pdev->dev, NULL); + uhci->clk = devm_clk_get_optional(&pdev->dev, NULL); if (IS_ERR(uhci->clk)) { ret = PTR_ERR(uhci->clk); goto err_rmr; From 9f657a92805cfc98e11cf5da9e8f4e02ecff2260 Mon Sep 17 00:00:00 2001 From: Lukasz Czechowski Date: Fri, 25 Apr 2025 17:18:06 +0200 Subject: [PATCH 122/555] usb: misc: onboard_usb_dev: fix support for Cypress HX3 hubs The Cypress HX3 USB3.0 hubs use different PID values depending on the product variant. The comment in compatibles table is misleading, as the currently used PIDs (0x6504 and 0x6506 for USB 3.0 and USB 2.0, respectively) are defaults for the CYUSB331x, while CYUSB330x and CYUSB332x variants use different values. Based on the datasheet [1], update the compatible usb devices table to handle different types of the hub. The change also includes vendor mode PIDs, which are used by the hub in I2C Master boot mode, if connected EEPROM contains invalid signature or is blank. This allows to correctly boot the hub even if the EEPROM will have broken content. Number of vcc supplies and timing requirements are the same for all HX variants, so the platform driver's match table does not have to be extended. [1] https://www.infineon.com/dgdl/Infineon-HX3_USB_3_0_Hub_Consumer_Industrial-DataSheet-v22_00-EN.pdf?fileId=8ac78c8c7d0d8da4017d0ecb53f644b8 Table 9. PID Values Fixes: b43cd82a1a40 ("usb: misc: onboard-hub: add support for Cypress HX3 USB 3.0 family") Cc: stable Signed-off-by: Lukasz Czechowski Link: https://lore.kernel.org/r/20250425-onboard_usb_dev-v2-1-4a76a474a010@thaumatec.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/misc/onboard_usb_dev.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/drivers/usb/misc/onboard_usb_dev.c b/drivers/usb/misc/onboard_usb_dev.c index 75ac3c6aa92d..f5372dfa241a 100644 --- a/drivers/usb/misc/onboard_usb_dev.c +++ b/drivers/usb/misc/onboard_usb_dev.c @@ -569,8 +569,14 @@ static void onboard_dev_usbdev_disconnect(struct usb_device *udev) } static const struct usb_device_id onboard_dev_id_table[] = { - { USB_DEVICE(VENDOR_ID_CYPRESS, 0x6504) }, /* CYUSB33{0,1,2}x/CYUSB230x 3.0 HUB */ - { USB_DEVICE(VENDOR_ID_CYPRESS, 0x6506) }, /* CYUSB33{0,1,2}x/CYUSB230x 2.0 HUB */ + { USB_DEVICE(VENDOR_ID_CYPRESS, 0x6500) }, /* CYUSB330x 3.0 HUB */ + { USB_DEVICE(VENDOR_ID_CYPRESS, 0x6502) }, /* CYUSB330x 2.0 HUB */ + { USB_DEVICE(VENDOR_ID_CYPRESS, 0x6503) }, /* CYUSB33{0,1}x 2.0 HUB, Vendor Mode */ + { USB_DEVICE(VENDOR_ID_CYPRESS, 0x6504) }, /* CYUSB331x 3.0 HUB */ + { USB_DEVICE(VENDOR_ID_CYPRESS, 0x6506) }, /* CYUSB331x 2.0 HUB */ + { USB_DEVICE(VENDOR_ID_CYPRESS, 0x6507) }, /* CYUSB332x 2.0 HUB, Vendor Mode */ + { USB_DEVICE(VENDOR_ID_CYPRESS, 0x6508) }, /* CYUSB332x 3.0 HUB */ + { USB_DEVICE(VENDOR_ID_CYPRESS, 0x650a) }, /* CYUSB332x 2.0 HUB */ { USB_DEVICE(VENDOR_ID_CYPRESS, 0x6570) }, /* CY7C6563x 2.0 HUB */ { USB_DEVICE(VENDOR_ID_GENESYS, 0x0608) }, /* Genesys Logic GL850G USB 2.0 HUB */ { USB_DEVICE(VENDOR_ID_GENESYS, 0x0610) }, /* Genesys Logic GL852G USB 2.0 HUB */ From 364618c89d4c57c85e5fc51a2446cd939bf57802 Mon Sep 17 00:00:00 2001 From: Andrei Kuchynski Date: Thu, 24 Apr 2025 08:44:28 +0000 Subject: [PATCH 123/555] usb: typec: ucsi: displayport: Fix deadlock This patch introduces the ucsi_con_mutex_lock / ucsi_con_mutex_unlock functions to the UCSI driver. ucsi_con_mutex_lock ensures the connector mutex is only locked if a connection is established and the partner pointer is valid. This resolves a deadlock scenario where ucsi_displayport_remove_partner holds con->mutex waiting for dp_altmode_work to complete while dp_altmode_work attempts to acquire it. Cc: stable Fixes: af8622f6a585 ("usb: typec: ucsi: Support for DisplayPort alt mode") Signed-off-by: Andrei Kuchynski Reviewed-by: Heikki Krogerus Link: https://lore.kernel.org/r/20250424084429.3220757-2-akuchynski@chromium.org Signed-off-by: Greg Kroah-Hartman --- drivers/usb/typec/ucsi/displayport.c | 19 +++++++++------- drivers/usb/typec/ucsi/ucsi.c | 34 ++++++++++++++++++++++++++++ drivers/usb/typec/ucsi/ucsi.h | 2 ++ 3 files changed, 47 insertions(+), 8 deletions(-) diff --git a/drivers/usb/typec/ucsi/displayport.c b/drivers/usb/typec/ucsi/displayport.c index 420af5139c70..acd053d4e38c 100644 --- a/drivers/usb/typec/ucsi/displayport.c +++ b/drivers/usb/typec/ucsi/displayport.c @@ -54,7 +54,8 @@ static int ucsi_displayport_enter(struct typec_altmode *alt, u32 *vdo) u8 cur = 0; int ret; - mutex_lock(&dp->con->lock); + if (!ucsi_con_mutex_lock(dp->con)) + return -ENOTCONN; if (!dp->override && dp->initialized) { const struct typec_altmode *p = typec_altmode_get_partner(alt); @@ -100,7 +101,7 @@ static int ucsi_displayport_enter(struct typec_altmode *alt, u32 *vdo) schedule_work(&dp->work); ret = 0; err_unlock: - mutex_unlock(&dp->con->lock); + ucsi_con_mutex_unlock(dp->con); return ret; } @@ -112,7 +113,8 @@ static int ucsi_displayport_exit(struct typec_altmode *alt) u64 command; int ret = 0; - mutex_lock(&dp->con->lock); + if (!ucsi_con_mutex_lock(dp->con)) + return -ENOTCONN; if (!dp->override) { const struct typec_altmode *p = typec_altmode_get_partner(alt); @@ -144,7 +146,7 @@ static int ucsi_displayport_exit(struct typec_altmode *alt) schedule_work(&dp->work); out_unlock: - mutex_unlock(&dp->con->lock); + ucsi_con_mutex_unlock(dp->con); return ret; } @@ -202,20 +204,21 @@ static int ucsi_displayport_vdm(struct typec_altmode *alt, int cmd = PD_VDO_CMD(header); int svdm_version; - mutex_lock(&dp->con->lock); + if (!ucsi_con_mutex_lock(dp->con)) + return -ENOTCONN; if (!dp->override && dp->initialized) { const struct typec_altmode *p = typec_altmode_get_partner(alt); dev_warn(&p->dev, "firmware doesn't support alternate mode overriding\n"); - mutex_unlock(&dp->con->lock); + ucsi_con_mutex_unlock(dp->con); return -EOPNOTSUPP; } svdm_version = typec_altmode_get_svdm_version(alt); if (svdm_version < 0) { - mutex_unlock(&dp->con->lock); + ucsi_con_mutex_unlock(dp->con); return svdm_version; } @@ -259,7 +262,7 @@ static int ucsi_displayport_vdm(struct typec_altmode *alt, break; } - mutex_unlock(&dp->con->lock); + ucsi_con_mutex_unlock(dp->con); return 0; } diff --git a/drivers/usb/typec/ucsi/ucsi.c b/drivers/usb/typec/ucsi/ucsi.c index e8c7e9dc4930..01ce858a1a2b 100644 --- a/drivers/usb/typec/ucsi/ucsi.c +++ b/drivers/usb/typec/ucsi/ucsi.c @@ -1922,6 +1922,40 @@ void ucsi_set_drvdata(struct ucsi *ucsi, void *data) } EXPORT_SYMBOL_GPL(ucsi_set_drvdata); +/** + * ucsi_con_mutex_lock - Acquire the connector mutex + * @con: The connector interface to lock + * + * Returns true on success, false if the connector is disconnected + */ +bool ucsi_con_mutex_lock(struct ucsi_connector *con) +{ + bool mutex_locked = false; + bool connected = true; + + while (connected && !mutex_locked) { + mutex_locked = mutex_trylock(&con->lock) != 0; + connected = UCSI_CONSTAT(con, CONNECTED); + if (connected && !mutex_locked) + msleep(20); + } + + connected = connected && con->partner; + if (!connected && mutex_locked) + mutex_unlock(&con->lock); + + return connected; +} + +/** + * ucsi_con_mutex_unlock - Release the connector mutex + * @con: The connector interface to unlock + */ +void ucsi_con_mutex_unlock(struct ucsi_connector *con) +{ + mutex_unlock(&con->lock); +} + /** * ucsi_create - Allocate UCSI instance * @dev: Device interface to the PPM (Platform Policy Manager) diff --git a/drivers/usb/typec/ucsi/ucsi.h b/drivers/usb/typec/ucsi/ucsi.h index 3a2c1762bec1..9c5278a0c5d4 100644 --- a/drivers/usb/typec/ucsi/ucsi.h +++ b/drivers/usb/typec/ucsi/ucsi.h @@ -94,6 +94,8 @@ int ucsi_register(struct ucsi *ucsi); void ucsi_unregister(struct ucsi *ucsi); void *ucsi_get_drvdata(struct ucsi *ucsi); void ucsi_set_drvdata(struct ucsi *ucsi, void *data); +bool ucsi_con_mutex_lock(struct ucsi_connector *con); +void ucsi_con_mutex_unlock(struct ucsi_connector *con); void ucsi_connector_change(struct ucsi *ucsi, u8 num); From 312d79669e71283d05c05cc49a1a31e59e3d9e0e Mon Sep 17 00:00:00 2001 From: Andrei Kuchynski Date: Thu, 24 Apr 2025 08:44:29 +0000 Subject: [PATCH 124/555] usb: typec: ucsi: displayport: Fix NULL pointer access This patch ensures that the UCSI driver waits for all pending tasks in the ucsi_displayport_work workqueue to finish executing before proceeding with the partner removal. Cc: stable Fixes: af8622f6a585 ("usb: typec: ucsi: Support for DisplayPort alt mode") Signed-off-by: Andrei Kuchynski Reviewed-by: Heikki Krogerus Reviewed-by: Benson Leung Link: https://lore.kernel.org/r/20250424084429.3220757-3-akuchynski@chromium.org Signed-off-by: Greg Kroah-Hartman --- drivers/usb/typec/ucsi/displayport.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/usb/typec/ucsi/displayport.c b/drivers/usb/typec/ucsi/displayport.c index acd053d4e38c..8aae80b457d7 100644 --- a/drivers/usb/typec/ucsi/displayport.c +++ b/drivers/usb/typec/ucsi/displayport.c @@ -299,6 +299,8 @@ void ucsi_displayport_remove_partner(struct typec_altmode *alt) if (!dp) return; + cancel_work_sync(&dp->work); + dp->data.conf = 0; dp->data.status = 0; dp->initialized = false; From 8614ecdb1570e4fffe87ebdc62b613ed66f1f6a6 Mon Sep 17 00:00:00 2001 From: Pawel Laszczak Date: Fri, 25 Apr 2025 05:55:40 +0000 Subject: [PATCH 125/555] usb: cdnsp: fix L1 resume issue for RTL_REVISION_NEW_LPM version The controllers with rtl version larger than RTL_REVISION_NEW_LPM (0x00002700) has bug which causes that controller doesn't resume from L1 state. It happens if after receiving LPM packet controller starts transitioning to L1 and in this moment the driver force resuming by write operation to PORTSC.PLS. It's corner case and happens when write operation to PORTSC occurs during device delay before transitioning to L1 after transmitting ACK time (TL1TokenRetry). Forcing transition from L1->L0 by driver for revision larger than RTL_REVISION_NEW_LPM is not needed, so driver can simply fix this issue through block call of cdnsp_force_l0_go function. Fixes: 3d82904559f4 ("usb: cdnsp: cdns3 Add main part of Cadence USBSSP DRD Driver") Cc: stable Signed-off-by: Pawel Laszczak Acked-by: Peter Chen Link: https://lore.kernel.org/r/PH7PR07MB9538B55C3A6E71F9ED29E980DD842@PH7PR07MB9538.namprd07.prod.outlook.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/cdns3/cdnsp-gadget.c | 2 ++ drivers/usb/cdns3/cdnsp-gadget.h | 3 +++ drivers/usb/cdns3/cdnsp-ring.c | 3 ++- 3 files changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/usb/cdns3/cdnsp-gadget.c b/drivers/usb/cdns3/cdnsp-gadget.c index 7f5534db2086..4824a10df07e 100644 --- a/drivers/usb/cdns3/cdnsp-gadget.c +++ b/drivers/usb/cdns3/cdnsp-gadget.c @@ -1793,6 +1793,8 @@ static void cdnsp_get_rev_cap(struct cdnsp_device *pdev) reg += cdnsp_find_next_ext_cap(reg, 0, RTL_REV_CAP); pdev->rev_cap = reg; + pdev->rtl_revision = readl(&pdev->rev_cap->rtl_revision); + dev_info(pdev->dev, "Rev: %08x/%08x, eps: %08x, buff: %08x/%08x\n", readl(&pdev->rev_cap->ctrl_revision), readl(&pdev->rev_cap->rtl_revision), diff --git a/drivers/usb/cdns3/cdnsp-gadget.h b/drivers/usb/cdns3/cdnsp-gadget.h index 87ac0cd113e7..12534be52f39 100644 --- a/drivers/usb/cdns3/cdnsp-gadget.h +++ b/drivers/usb/cdns3/cdnsp-gadget.h @@ -1360,6 +1360,7 @@ struct cdnsp_port { * @rev_cap: Controller Capabilities Registers. * @hcs_params1: Cached register copies of read-only HCSPARAMS1 * @hcc_params: Cached register copies of read-only HCCPARAMS1 + * @rtl_revision: Cached controller rtl revision. * @setup: Temporary buffer for setup packet. * @ep0_preq: Internal allocated request used during enumeration. * @ep0_stage: ep0 stage during enumeration process. @@ -1414,6 +1415,8 @@ struct cdnsp_device { __u32 hcs_params1; __u32 hcs_params3; __u32 hcc_params; + #define RTL_REVISION_NEW_LPM 0x2700 + __u32 rtl_revision; /* Lock used in interrupt thread context. */ spinlock_t lock; struct usb_ctrlrequest setup; diff --git a/drivers/usb/cdns3/cdnsp-ring.c b/drivers/usb/cdns3/cdnsp-ring.c index 46852529499d..fd06cb85c4ea 100644 --- a/drivers/usb/cdns3/cdnsp-ring.c +++ b/drivers/usb/cdns3/cdnsp-ring.c @@ -308,7 +308,8 @@ static bool cdnsp_ring_ep_doorbell(struct cdnsp_device *pdev, writel(db_value, reg_addr); - cdnsp_force_l0_go(pdev); + if (pdev->rtl_revision < RTL_REVISION_NEW_LPM) + cdnsp_force_l0_go(pdev); /* Doorbell was set. */ return true; From 054c5145540e5ad5b80adf23a5e3e2fc281fb8aa Mon Sep 17 00:00:00 2001 From: Oliver Neukum Date: Wed, 30 Apr 2025 15:48:10 +0200 Subject: [PATCH 126/555] USB: usbtmc: use interruptible sleep in usbtmc_read usbtmc_read() calls usbtmc_generic_read() which uses interruptible sleep, but usbtmc_read() itself uses uninterruptble sleep for mutual exclusion between threads. That makes no sense. Both should use interruptible sleep. Fixes: 5b775f672cc99 ("USB: add USB test and measurement class driver") Cc: stable Signed-off-by: Oliver Neukum Link: https://lore.kernel.org/r/20250430134810.226015-1-oneukum@suse.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/class/usbtmc.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/usb/class/usbtmc.c b/drivers/usb/class/usbtmc.c index 34e46ef308ab..634c3bcbb413 100644 --- a/drivers/usb/class/usbtmc.c +++ b/drivers/usb/class/usbtmc.c @@ -1380,7 +1380,10 @@ static ssize_t usbtmc_read(struct file *filp, char __user *buf, if (!buffer) return -ENOMEM; - mutex_lock(&data->io_mutex); + retval = mutex_lock_interruptible(&data->io_mutex); + if (retval < 0) + goto exit_nolock; + if (data->zombie) { retval = -ENODEV; goto exit; @@ -1503,6 +1506,7 @@ static ssize_t usbtmc_read(struct file *filp, char __user *buf, exit: mutex_unlock(&data->io_mutex); +exit_nolock: kfree(buffer); return retval; } From e918d3959b5ae0e793b8f815ce62240e10ba03a4 Mon Sep 17 00:00:00 2001 From: RD Babiera Date: Tue, 29 Apr 2025 23:47:01 +0000 Subject: [PATCH 127/555] usb: typec: tcpm: delay SNK_TRY_WAIT_DEBOUNCE to SRC_TRYWAIT transition This patch fixes Type-C Compliance Test TD 4.7.6 - Try.SNK DRP Connect SNKAS. The compliance tester moves into SNK_UNATTACHED during toggling and expects the PUT to apply Rp after tPDDebounce of detection. If the port is in SNK_TRY_WAIT_DEBOUNCE, it will move into SRC_TRYWAIT immediately and apply Rp. This violates TD 4.7.5.V.3, where the tester confirms that the PUT attaches Rp after the transitions to Unattached.SNK for tPDDebounce. Change the tcpm_set_state delay between SNK_TRY_WAIT_DEBOUNCE and SRC_TRYWAIT to tPDDebounce. Fixes: a0a3e04e6b2c ("staging: typec: tcpm: Check for Rp for tPDDebounce") Cc: stable Signed-off-by: RD Babiera Reviewed-by: Badhri Jagan Sridharan Reviewed-by: Heikki Krogerus Link: https://lore.kernel.org/r/20250429234703.3748506-2-rdbabiera@google.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/typec/tcpm/tcpm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/usb/typec/tcpm/tcpm.c b/drivers/usb/typec/tcpm/tcpm.c index a99db4e025cd..8adf6f954633 100644 --- a/drivers/usb/typec/tcpm/tcpm.c +++ b/drivers/usb/typec/tcpm/tcpm.c @@ -5965,7 +5965,7 @@ static void _tcpm_cc_change(struct tcpm_port *port, enum typec_cc_status cc1, case SNK_TRY_WAIT_DEBOUNCE: if (!tcpm_port_is_sink(port)) { port->max_wait = 0; - tcpm_set_state(port, SRC_TRYWAIT, 0); + tcpm_set_state(port, SRC_TRYWAIT, PD_T_PD_DEBOUNCE); } break; case SRC_TRY_WAIT: From 95deee37a12364f410d22c6a8383f59738a2fef3 Mon Sep 17 00:00:00 2001 From: Will McVicker Date: Thu, 24 Apr 2025 11:04:19 -0700 Subject: [PATCH 128/555] platform: Fix race condition during DMA configure at IOMMU probe time To avoid a race between the IOMMU probing thread and the device driver async probing thread during configuration of the platform DMA, update `platform_dma_configure()` to read `dev->driver` once and test if it's NULL before using it. This ensures that we don't de-reference an invalid platform driver pointer if the device driver is asynchronously bound while configuring the DMA. Fixes: bcb81ac6ae3c ("iommu: Get DT/ACPI parsing into the proper probe path") Signed-off-by: Will McVicker Reviewed-by: Robin Murphy Link: https://lore.kernel.org/r/20250424180420.3928523-1-willmcvicker@google.com Signed-off-by: Greg Kroah-Hartman --- drivers/base/platform.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/base/platform.c b/drivers/base/platform.c index 1813cfd0c4bd..cfccf3ff36e7 100644 --- a/drivers/base/platform.c +++ b/drivers/base/platform.c @@ -1440,7 +1440,7 @@ static void platform_shutdown(struct device *_dev) static int platform_dma_configure(struct device *dev) { - struct platform_driver *drv = to_platform_driver(dev->driver); + struct device_driver *drv = READ_ONCE(dev->driver); struct fwnode_handle *fwnode = dev_fwnode(dev); enum dev_dma_attr attr; int ret = 0; @@ -1451,8 +1451,8 @@ static int platform_dma_configure(struct device *dev) attr = acpi_get_dma_attr(to_acpi_device_node(fwnode)); ret = acpi_dma_configure(dev, attr); } - /* @drv may not be valid when we're called from the IOMMU layer */ - if (ret || !dev->driver || drv->driver_managed_dma) + /* @dev->driver may not be valid when we're called from the IOMMU layer */ + if (ret || !drv || to_platform_driver(drv)->driver_managed_dma) return ret; ret = iommu_device_use_default_domain(dev); From 53e3e5babc0963a92d856a5ec0ce92c59f54bc12 Mon Sep 17 00:00:00 2001 From: Namjae Jeon Date: Wed, 30 Apr 2025 11:18:28 +0900 Subject: [PATCH 129/555] ksmbd: prevent rename with empty string Client can send empty newname string to ksmbd server. It will cause a kernel oops from d_alloc. This patch return the error when attempting to rename a file or directory with an empty new name string. Cc: stable@vger.kernel.org Reported-by: Norbert Szetei Tested-by: Norbert Szetei Signed-off-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/server/smb2pdu.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/fs/smb/server/smb2pdu.c b/fs/smb/server/smb2pdu.c index 46aa08245742..f2a2be8467c6 100644 --- a/fs/smb/server/smb2pdu.c +++ b/fs/smb/server/smb2pdu.c @@ -633,6 +633,11 @@ smb2_get_name(const char *src, const int maxlen, struct nls_table *local_nls) return name; } + if (*name == '\0') { + kfree(name); + return ERR_PTR(-EINVAL); + } + if (*name == '\\') { pr_err("not allow directory name included leading slash\n"); kfree(name); From eb4447bcce915b43b691123118893fca4f372a8f Mon Sep 17 00:00:00 2001 From: Wang Zhaolong Date: Wed, 30 Apr 2025 11:16:23 +0800 Subject: [PATCH 130/555] ksmbd: fix memory leak in parse_lease_state() The previous patch that added bounds check for create lease context introduced a memory leak. When the bounds check fails, the function returns NULL without freeing the previously allocated lease_ctx_info structure. This patch fixes the issue by adding kfree(lreq) before returning NULL in both boundary check cases. Fixes: bab703ed8472 ("ksmbd: add bounds check for create lease context") Signed-off-by: Wang Zhaolong Acked-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/server/oplock.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/fs/smb/server/oplock.c b/fs/smb/server/oplock.c index 81a29857b1e3..03f606afad93 100644 --- a/fs/smb/server/oplock.c +++ b/fs/smb/server/oplock.c @@ -1496,7 +1496,7 @@ struct lease_ctx_info *parse_lease_state(void *open_req) if (le16_to_cpu(cc->DataOffset) + le32_to_cpu(cc->DataLength) < sizeof(struct create_lease_v2) - 4) - return NULL; + goto err_out; memcpy(lreq->lease_key, lc->lcontext.LeaseKey, SMB2_LEASE_KEY_SIZE); lreq->req_state = lc->lcontext.LeaseState; @@ -1512,7 +1512,7 @@ struct lease_ctx_info *parse_lease_state(void *open_req) if (le16_to_cpu(cc->DataOffset) + le32_to_cpu(cc->DataLength) < sizeof(struct create_lease)) - return NULL; + goto err_out; memcpy(lreq->lease_key, lc->lcontext.LeaseKey, SMB2_LEASE_KEY_SIZE); lreq->req_state = lc->lcontext.LeaseState; @@ -1521,6 +1521,9 @@ struct lease_ctx_info *parse_lease_state(void *open_req) lreq->version = 1; } return lreq; +err_out: + kfree(lreq); + return NULL; } /** From 11854fe263eb1b9a8efa33b0c087add7719ea9b4 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Fri, 25 Apr 2025 15:45:06 -0700 Subject: [PATCH 131/555] binfmt_elf: Move brk for static PIE even if ASLR disabled In commit bbdc6076d2e5 ("binfmt_elf: move brk out of mmap when doing direct loader exec"), the brk was moved out of the mmap region when loading static PIE binaries (ET_DYN without INTERP). The common case for these binaries was testing new ELF loaders, so the brk needed to be away from mmap to avoid colliding with stack, future mmaps (of the loader-loaded binary), etc. But this was only done when ASLR was enabled, in an attempt to minimize changes to memory layouts. After adding support to respect alignment requirements for static PIE binaries in commit 3545deff0ec7 ("binfmt_elf: Honor PT_LOAD alignment for static PIE"), it became possible to have a large gap after the final PT_LOAD segment and the top of the mmap region. This means that future mmap allocations might go after the last PT_LOAD segment (where brk might be if ASLR was disabled) instead of before them (where they traditionally ended up). On arm64, running with ASLR disabled, Ubuntu 22.04's "ldconfig" binary, a static PIE, has alignment requirements that leaves a gap large enough after the last PT_LOAD segment to fit the vdso and vvar, but still leave enough space for the brk (which immediately follows the last PT_LOAD segment) to be allocated by the binary. fffff7f20000-fffff7fde000 r-xp 00000000 fe:02 8110426 /sbin/ldconfig.real fffff7fee000-fffff7ff5000 rw-p 000be000 fe:02 8110426 /sbin/ldconfig.real fffff7ff5000-fffff7ffa000 rw-p 00000000 00:00 0 ***[brk will go here at fffff7ffa000]*** fffff7ffc000-fffff7ffe000 r--p 00000000 00:00 0 [vvar] fffff7ffe000-fffff8000000 r-xp 00000000 00:00 0 [vdso] fffffffdf000-1000000000000 rw-p 00000000 00:00 0 [stack] After commit 0b3bc3354eb9 ("arm64: vdso: Switch to generic storage implementation"), the arm64 vvar grew slightly, and suddenly the brk collided with the allocation. fffff7f20000-fffff7fde000 r-xp 00000000 fe:02 8110426 /sbin/ldconfig.real fffff7fee000-fffff7ff5000 rw-p 000be000 fe:02 8110426 /sbin/ldconfig.real fffff7ff5000-fffff7ffa000 rw-p 00000000 00:00 0 ***[oops, no room any more, vvar is at fffff7ffa000!]*** fffff7ffa000-fffff7ffe000 r--p 00000000 00:00 0 [vvar] fffff7ffe000-fffff8000000 r-xp 00000000 00:00 0 [vdso] fffffffdf000-1000000000000 rw-p 00000000 00:00 0 [stack] The solution is to unconditionally move the brk out of the mmap region for static PIE binaries. Whether ASLR is enabled or not does not change if there may be future mmap allocation collisions with a growing brk region. Update memory layout comments (with kernel-doc headings), consolidate the setting of mm->brk to later (it isn't needed early), move static PIE brk out of mmap unconditionally, and make sure brk(2) knows to base brk position off of mm->start_brk not mm->end_data no matter what the cause of moving it is (via current->brk_randomized). For the CONFIG_COMPAT_BRK case, though, leave the logic unchanged, as we can never safely move the brk. These systems, however, are not using specially aligned static PIE binaries. Reported-by: Ryan Roberts Closes: https://lore.kernel.org/lkml/f93db308-4a0e-4806-9faf-98f890f5a5e6@arm.com/ Fixes: bbdc6076d2e5 ("binfmt_elf: move brk out of mmap when doing direct loader exec") Link: https://lore.kernel.org/r/20250425224502.work.520-kees@kernel.org Reviewed-by: Ryan Roberts Tested-by: Ryan Roberts Signed-off-by: Kees Cook --- fs/binfmt_elf.c | 71 ++++++++++++++++++++++++++++++++----------------- 1 file changed, 47 insertions(+), 24 deletions(-) diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 584fa89bc877..4c1ea6b52a53 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -830,6 +830,7 @@ static int load_elf_binary(struct linux_binprm *bprm) struct elf_phdr *elf_ppnt, *elf_phdata, *interp_elf_phdata = NULL; struct elf_phdr *elf_property_phdata = NULL; unsigned long elf_brk; + bool brk_moved = false; int retval, i; unsigned long elf_entry; unsigned long e_entry; @@ -1097,15 +1098,19 @@ out_free_interp: /* Calculate any requested alignment. */ alignment = maximum_alignment(elf_phdata, elf_ex->e_phnum); - /* - * There are effectively two types of ET_DYN - * binaries: programs (i.e. PIE: ET_DYN with PT_INTERP) - * and loaders (ET_DYN without PT_INTERP, since they - * _are_ the ELF interpreter). The loaders must - * be loaded away from programs since the program - * may otherwise collide with the loader (especially - * for ET_EXEC which does not have a randomized - * position). For example to handle invocations of + /** + * DOC: PIE handling + * + * There are effectively two types of ET_DYN ELF + * binaries: programs (i.e. PIE: ET_DYN with + * PT_INTERP) and loaders (i.e. static PIE: ET_DYN + * without PT_INTERP, usually the ELF interpreter + * itself). Loaders must be loaded away from programs + * since the program may otherwise collide with the + * loader (especially for ET_EXEC which does not have + * a randomized position). + * + * For example, to handle invocations of * "./ld.so someprog" to test out a new version of * the loader, the subsequent program that the * loader loads must avoid the loader itself, so @@ -1118,6 +1123,9 @@ out_free_interp: * ELF_ET_DYN_BASE and loaders are loaded into the * independently randomized mmap region (0 load_bias * without MAP_FIXED nor MAP_FIXED_NOREPLACE). + * + * See below for "brk" handling details, which is + * also affected by program vs loader and ASLR. */ if (interpreter) { /* On ET_DYN with PT_INTERP, we do the ASLR. */ @@ -1234,8 +1242,6 @@ out_free_interp: start_data += load_bias; end_data += load_bias; - current->mm->start_brk = current->mm->brk = ELF_PAGEALIGN(elf_brk); - if (interpreter) { elf_entry = load_elf_interp(interp_elf_ex, interpreter, @@ -1291,27 +1297,44 @@ out_free_interp: mm->end_data = end_data; mm->start_stack = bprm->p; - if ((current->flags & PF_RANDOMIZE) && (snapshot_randomize_va_space > 1)) { + /** + * DOC: "brk" handling + * + * For architectures with ELF randomization, when executing a + * loader directly (i.e. static PIE: ET_DYN without PT_INTERP), + * move the brk area out of the mmap region and into the unused + * ELF_ET_DYN_BASE region. Since "brk" grows up it may collide + * early with the stack growing down or other regions being put + * into the mmap region by the kernel (e.g. vdso). + * + * In the CONFIG_COMPAT_BRK case, though, everything is turned + * off because we're not allowed to move the brk at all. + */ + if (!IS_ENABLED(CONFIG_COMPAT_BRK) && + IS_ENABLED(CONFIG_ARCH_HAS_ELF_RANDOMIZE) && + elf_ex->e_type == ET_DYN && !interpreter) { + elf_brk = ELF_ET_DYN_BASE; + /* This counts as moving the brk, so let brk(2) know. */ + brk_moved = true; + } + mm->start_brk = mm->brk = ELF_PAGEALIGN(elf_brk); + + if ((current->flags & PF_RANDOMIZE) && snapshot_randomize_va_space > 1) { /* - * For architectures with ELF randomization, when executing - * a loader directly (i.e. no interpreter listed in ELF - * headers), move the brk area out of the mmap region - * (since it grows up, and may collide early with the stack - * growing down), and into the unused ELF_ET_DYN_BASE region. + * If we didn't move the brk to ELF_ET_DYN_BASE (above), + * leave a gap between .bss and brk. */ - if (IS_ENABLED(CONFIG_ARCH_HAS_ELF_RANDOMIZE) && - elf_ex->e_type == ET_DYN && !interpreter) { - mm->brk = mm->start_brk = ELF_ET_DYN_BASE; - } else { - /* Otherwise leave a gap between .bss and brk. */ + if (!brk_moved) mm->brk = mm->start_brk = mm->brk + PAGE_SIZE; - } mm->brk = mm->start_brk = arch_randomize_brk(mm); + brk_moved = true; + } + #ifdef compat_brk_randomized + if (brk_moved) current->brk_randomized = 1; #endif - } if (current->personality & MMAP_PAGE_ZERO) { /* Why this, you ask??? Well SVr4 maps page 0 as read-only, From cac01bd178d6a2a23727f138d647ce1a0e8a73a1 Mon Sep 17 00:00:00 2001 From: Dave Penkler Date: Fri, 2 May 2025 09:09:39 +0200 Subject: [PATCH 132/555] usb: usbtmc: Fix erroneous get_stb ioctl error returns wait_event_interruptible_timeout returns a long The return was being assigned to an int causing an integer overflow when the remaining jiffies > INT_MAX resulting in random error returns. Use a long return value and convert to int ioctl return only on error. When the return value of wait_event_interruptible_timeout was <= INT_MAX the number of remaining jiffies was returned which has no meaning for the user. Return 0 on success. Reported-by: Michael Katzmann Fixes: dbf3e7f654c0 ("Implement an ioctl to support the USMTMC-USB488 READ_STATUS_BYTE operation.") Cc: stable@vger.kernel.org Signed-off-by: Dave Penkler Link: https://lore.kernel.org/r/20250502070941.31819-2-dpenkler@gmail.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/class/usbtmc.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/drivers/usb/class/usbtmc.c b/drivers/usb/class/usbtmc.c index 634c3bcbb413..e91f11c1ab5f 100644 --- a/drivers/usb/class/usbtmc.c +++ b/drivers/usb/class/usbtmc.c @@ -482,6 +482,7 @@ static int usbtmc_get_stb(struct usbtmc_file_data *file_data, __u8 *stb) u8 *buffer; u8 tag; int rv; + long wait_rv; dev_dbg(dev, "Enter ioctl_read_stb iin_ep_present: %d\n", data->iin_ep_present); @@ -511,16 +512,17 @@ static int usbtmc_get_stb(struct usbtmc_file_data *file_data, __u8 *stb) } if (data->iin_ep_present) { - rv = wait_event_interruptible_timeout( + wait_rv = wait_event_interruptible_timeout( data->waitq, atomic_read(&data->iin_data_valid) != 0, file_data->timeout); - if (rv < 0) { - dev_dbg(dev, "wait interrupted %d\n", rv); + if (wait_rv < 0) { + dev_dbg(dev, "wait interrupted %ld\n", wait_rv); + rv = wait_rv; goto exit; } - if (rv == 0) { + if (wait_rv == 0) { dev_dbg(dev, "wait timed out\n"); rv = -ETIMEDOUT; goto exit; @@ -539,6 +541,8 @@ static int usbtmc_get_stb(struct usbtmc_file_data *file_data, __u8 *stb) dev_dbg(dev, "stb:0x%02x received %d\n", (unsigned int)*stb, rv); + rv = 0; + exit: /* bump interrupt bTag */ data->iin_bTag += 1; From a9747c9b8b59ab4207effd20eb91a890acb44e16 Mon Sep 17 00:00:00 2001 From: Dave Penkler Date: Fri, 2 May 2025 09:09:40 +0200 Subject: [PATCH 133/555] usb: usbtmc: Fix erroneous wait_srq ioctl return wait_event_interruptible_timeout returns a long The return was being assigned to an int causing an integer overflow when the remaining jiffies > INT_MAX resulting in random error returns. Use a long return value, converting to the int ioctl return only on error. Fixes: 739240a9f6ac ("usb: usbtmc: Add ioctl USBTMC488_IOCTL_WAIT_SRQ") Cc: stable@vger.kernel.org Signed-off-by: Dave Penkler Link: https://lore.kernel.org/r/20250502070941.31819-3-dpenkler@gmail.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/class/usbtmc.c | 23 +++++++++++------------ 1 file changed, 11 insertions(+), 12 deletions(-) diff --git a/drivers/usb/class/usbtmc.c b/drivers/usb/class/usbtmc.c index e91f11c1ab5f..4761a5ecadd6 100644 --- a/drivers/usb/class/usbtmc.c +++ b/drivers/usb/class/usbtmc.c @@ -606,9 +606,9 @@ static int usbtmc488_ioctl_wait_srq(struct usbtmc_file_data *file_data, { struct usbtmc_device_data *data = file_data->data; struct device *dev = &data->intf->dev; - int rv; u32 timeout; unsigned long expire; + long wait_rv; if (!data->iin_ep_present) { dev_dbg(dev, "no interrupt endpoint present\n"); @@ -622,25 +622,24 @@ static int usbtmc488_ioctl_wait_srq(struct usbtmc_file_data *file_data, mutex_unlock(&data->io_mutex); - rv = wait_event_interruptible_timeout( - data->waitq, - atomic_read(&file_data->srq_asserted) != 0 || - atomic_read(&file_data->closing), - expire); + wait_rv = wait_event_interruptible_timeout( + data->waitq, + atomic_read(&file_data->srq_asserted) != 0 || + atomic_read(&file_data->closing), + expire); mutex_lock(&data->io_mutex); /* Note! disconnect or close could be called in the meantime */ if (atomic_read(&file_data->closing) || data->zombie) - rv = -ENODEV; + return -ENODEV; - if (rv < 0) { - /* dev can be invalid now! */ - pr_debug("%s - wait interrupted %d\n", __func__, rv); - return rv; + if (wait_rv < 0) { + dev_dbg(dev, "%s - wait interrupted %ld\n", __func__, wait_rv); + return wait_rv; } - if (rv == 0) { + if (wait_rv == 0) { dev_dbg(dev, "%s - wait timed out\n", __func__); return -ETIMEDOUT; } From 4e77d3ec7c7c0d9535ccf1138827cb9bb5480b9b Mon Sep 17 00:00:00 2001 From: Dave Penkler Date: Fri, 2 May 2025 09:09:41 +0200 Subject: [PATCH 134/555] usb: usbtmc: Fix erroneous generic_read ioctl return wait_event_interruptible_timeout returns a long The return value was being assigned to an int causing an integer overflow when the remaining jiffies > INT_MAX which resulted in random error returns. Use a long return value, converting to the int ioctl return only on error. Fixes: bb99794a4792 ("usb: usbtmc: Add ioctl for vendor specific read") Cc: stable@vger.kernel.org Signed-off-by: Dave Penkler Link: https://lore.kernel.org/r/20250502070941.31819-4-dpenkler@gmail.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/class/usbtmc.c | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) diff --git a/drivers/usb/class/usbtmc.c b/drivers/usb/class/usbtmc.c index 4761a5ecadd6..740d2d2b19fb 100644 --- a/drivers/usb/class/usbtmc.c +++ b/drivers/usb/class/usbtmc.c @@ -833,6 +833,7 @@ static ssize_t usbtmc_generic_read(struct usbtmc_file_data *file_data, unsigned long expire; int bufcount = 1; int again = 0; + long wait_rv; /* mutex already locked */ @@ -945,19 +946,24 @@ static ssize_t usbtmc_generic_read(struct usbtmc_file_data *file_data, if (!(flags & USBTMC_FLAG_ASYNC)) { dev_dbg(dev, "%s: before wait time %lu\n", __func__, expire); - retval = wait_event_interruptible_timeout( + wait_rv = wait_event_interruptible_timeout( file_data->wait_bulk_in, usbtmc_do_transfer(file_data), expire); - dev_dbg(dev, "%s: wait returned %d\n", - __func__, retval); + dev_dbg(dev, "%s: wait returned %ld\n", + __func__, wait_rv); - if (retval <= 0) { - if (retval == 0) - retval = -ETIMEDOUT; + if (wait_rv < 0) { + retval = wait_rv; goto error; } + + if (wait_rv == 0) { + retval = -ETIMEDOUT; + goto error; + } + } urb = usb_get_from_anchor(&file_data->in_anchor); From f31fe8165d365379d858c53bef43254c7d6d1cfd Mon Sep 17 00:00:00 2001 From: Naman Jain Date: Fri, 2 May 2025 13:18:10 +0530 Subject: [PATCH 135/555] uio_hv_generic: Fix sysfs creation path for ring buffer On regular bootup, devices get registered to VMBus first, so when uio_hv_generic driver for a particular device type is probed, the device is already initialized and added, so sysfs creation in hv_uio_probe() works fine. However, when the device is removed and brought back, the channel gets rescinded and the device again gets registered to VMBus. However this time, the uio_hv_generic driver is already registered to probe for that device and in this case sysfs creation is tried before the device's kobject gets initialized completely. Fix this by moving the core logic of sysfs creation of ring buffer, from uio_hv_generic to HyperV's VMBus driver, where the rest of the sysfs attributes for the channels are defined. While doing that, make use of attribute groups and macros, instead of creating sysfs directly, to ensure better error handling and code flow. Problematic path: vmbus_process_offer (A new offer comes for the VMBus device) vmbus_add_channel_work vmbus_device_register |-> device_register | |... | |-> hv_uio_probe | |... | |-> sysfs_create_bin_file (leads to a warning as | the primary channel's kobject, which is used to | create the sysfs file, is not yet initialized) |-> kset_create_and_add |-> vmbus_add_channel_kobj (initialization of the primary channel's kobject happens later) Above code flow is sequential and the warning is always reproducible in this path. Fixes: 9ab877a6ccf8 ("uio_hv_generic: make ring buffer attribute for primary channel") Cc: stable@kernel.org Suggested-by: Saurabh Sengar Suggested-by: Michael Kelley Reviewed-by: Michael Kelley Tested-by: Michael Kelley Reviewed-by: Dexuan Cui Signed-off-by: Naman Jain Link: https://lore.kernel.org/r/20250502074811.2022-2-namjain@linux.microsoft.com Signed-off-by: Greg Kroah-Hartman --- drivers/hv/hyperv_vmbus.h | 6 +++ drivers/hv/vmbus_drv.c | 100 ++++++++++++++++++++++++++++++++++- drivers/uio/uio_hv_generic.c | 39 ++++++-------- include/linux/hyperv.h | 6 +++ 4 files changed, 128 insertions(+), 23 deletions(-) diff --git a/drivers/hv/hyperv_vmbus.h b/drivers/hv/hyperv_vmbus.h index 29780f3a7478..0b450e53161e 100644 --- a/drivers/hv/hyperv_vmbus.h +++ b/drivers/hv/hyperv_vmbus.h @@ -477,4 +477,10 @@ static inline int hv_debug_add_dev_dir(struct hv_device *dev) #endif /* CONFIG_HYPERV_TESTING */ +/* Create and remove sysfs entry for memory mapped ring buffers for a channel */ +int hv_create_ring_sysfs(struct vmbus_channel *channel, + int (*hv_mmap_ring_buffer)(struct vmbus_channel *channel, + struct vm_area_struct *vma)); +int hv_remove_ring_sysfs(struct vmbus_channel *channel); + #endif /* _HYPERV_VMBUS_H */ diff --git a/drivers/hv/vmbus_drv.c b/drivers/hv/vmbus_drv.c index 8d3cff42bdbb..0f16a83cc2d6 100644 --- a/drivers/hv/vmbus_drv.c +++ b/drivers/hv/vmbus_drv.c @@ -1802,6 +1802,27 @@ static ssize_t subchannel_id_show(struct vmbus_channel *channel, } static VMBUS_CHAN_ATTR_RO(subchannel_id); +static int hv_mmap_ring_buffer_wrapper(struct file *filp, struct kobject *kobj, + const struct bin_attribute *attr, + struct vm_area_struct *vma) +{ + struct vmbus_channel *channel = container_of(kobj, struct vmbus_channel, kobj); + + /* + * hv_(create|remove)_ring_sysfs implementation ensures that mmap_ring_buffer + * is not NULL. + */ + return channel->mmap_ring_buffer(channel, vma); +} + +static struct bin_attribute chan_attr_ring_buffer = { + .attr = { + .name = "ring", + .mode = 0600, + }, + .size = 2 * SZ_2M, + .mmap = hv_mmap_ring_buffer_wrapper, +}; static struct attribute *vmbus_chan_attrs[] = { &chan_attr_out_mask.attr, &chan_attr_in_mask.attr, @@ -1821,6 +1842,11 @@ static struct attribute *vmbus_chan_attrs[] = { NULL }; +static struct bin_attribute *vmbus_chan_bin_attrs[] = { + &chan_attr_ring_buffer, + NULL +}; + /* * Channel-level attribute_group callback function. Returns the permission for * each attribute, and returns 0 if an attribute is not visible. @@ -1841,9 +1867,24 @@ static umode_t vmbus_chan_attr_is_visible(struct kobject *kobj, return attr->mode; } +static umode_t vmbus_chan_bin_attr_is_visible(struct kobject *kobj, + const struct bin_attribute *attr, int idx) +{ + const struct vmbus_channel *channel = + container_of(kobj, struct vmbus_channel, kobj); + + /* Hide ring attribute if channel's ring_sysfs_visible is set to false */ + if (attr == &chan_attr_ring_buffer && !channel->ring_sysfs_visible) + return 0; + + return attr->attr.mode; +} + static const struct attribute_group vmbus_chan_group = { .attrs = vmbus_chan_attrs, - .is_visible = vmbus_chan_attr_is_visible + .bin_attrs = vmbus_chan_bin_attrs, + .is_visible = vmbus_chan_attr_is_visible, + .is_bin_visible = vmbus_chan_bin_attr_is_visible, }; static const struct kobj_type vmbus_chan_ktype = { @@ -1851,6 +1892,63 @@ static const struct kobj_type vmbus_chan_ktype = { .release = vmbus_chan_release, }; +/** + * hv_create_ring_sysfs() - create "ring" sysfs entry corresponding to ring buffers for a channel. + * @channel: Pointer to vmbus_channel structure + * @hv_mmap_ring_buffer: function pointer for initializing the function to be called on mmap of + * channel's "ring" sysfs node, which is for the ring buffer of that channel. + * Function pointer is of below type: + * int (*hv_mmap_ring_buffer)(struct vmbus_channel *channel, + * struct vm_area_struct *vma)) + * This has a pointer to the channel and a pointer to vm_area_struct, + * used for mmap, as arguments. + * + * Sysfs node for ring buffer of a channel is created along with other fields, however its + * visibility is disabled by default. Sysfs creation needs to be controlled when the use-case + * is running. + * For example, HV_NIC device is used either by uio_hv_generic or hv_netvsc at any given point of + * time, and "ring" sysfs is needed only when uio_hv_generic is bound to that device. To avoid + * exposing the ring buffer by default, this function is reponsible to enable visibility of + * ring for userspace to use. + * Note: Race conditions can happen with userspace and it is not encouraged to create new + * use-cases for this. This was added to maintain backward compatibility, while solving + * one of the race conditions in uio_hv_generic while creating sysfs. + * + * Returns 0 on success or error code on failure. + */ +int hv_create_ring_sysfs(struct vmbus_channel *channel, + int (*hv_mmap_ring_buffer)(struct vmbus_channel *channel, + struct vm_area_struct *vma)) +{ + struct kobject *kobj = &channel->kobj; + + channel->mmap_ring_buffer = hv_mmap_ring_buffer; + channel->ring_sysfs_visible = true; + + return sysfs_update_group(kobj, &vmbus_chan_group); +} +EXPORT_SYMBOL_GPL(hv_create_ring_sysfs); + +/** + * hv_remove_ring_sysfs() - remove ring sysfs entry corresponding to ring buffers for a channel. + * @channel: Pointer to vmbus_channel structure + * + * Hide "ring" sysfs for a channel by changing its is_visible attribute and updating sysfs group. + * + * Returns 0 on success or error code on failure. + */ +int hv_remove_ring_sysfs(struct vmbus_channel *channel) +{ + struct kobject *kobj = &channel->kobj; + int ret; + + channel->ring_sysfs_visible = false; + ret = sysfs_update_group(kobj, &vmbus_chan_group); + channel->mmap_ring_buffer = NULL; + return ret; +} +EXPORT_SYMBOL_GPL(hv_remove_ring_sysfs); + /* * vmbus_add_channel_kobj - setup a sub-directory under device/channels */ diff --git a/drivers/uio/uio_hv_generic.c b/drivers/uio/uio_hv_generic.c index 1b19b5647495..69c1df0f4ca5 100644 --- a/drivers/uio/uio_hv_generic.c +++ b/drivers/uio/uio_hv_generic.c @@ -131,15 +131,12 @@ static void hv_uio_rescind(struct vmbus_channel *channel) vmbus_device_unregister(channel->device_obj); } -/* Sysfs API to allow mmap of the ring buffers +/* Function used for mmap of ring buffer sysfs interface. * The ring buffer is allocated as contiguous memory by vmbus_open */ -static int hv_uio_ring_mmap(struct file *filp, struct kobject *kobj, - const struct bin_attribute *attr, - struct vm_area_struct *vma) +static int +hv_uio_ring_mmap(struct vmbus_channel *channel, struct vm_area_struct *vma) { - struct vmbus_channel *channel - = container_of(kobj, struct vmbus_channel, kobj); void *ring_buffer = page_address(channel->ringbuffer_page); if (channel->state != CHANNEL_OPENED_STATE) @@ -149,15 +146,6 @@ static int hv_uio_ring_mmap(struct file *filp, struct kobject *kobj, channel->ringbuffer_pagecount << PAGE_SHIFT); } -static const struct bin_attribute ring_buffer_bin_attr = { - .attr = { - .name = "ring", - .mode = 0600, - }, - .size = 2 * SZ_2M, - .mmap = hv_uio_ring_mmap, -}; - /* Callback from VMBUS subsystem when new channel created. */ static void hv_uio_new_channel(struct vmbus_channel *new_sc) @@ -178,8 +166,7 @@ hv_uio_new_channel(struct vmbus_channel *new_sc) /* Disable interrupts on sub channel */ new_sc->inbound.ring_buffer->interrupt_mask = 1; set_channel_read_mode(new_sc, HV_CALL_ISR); - - ret = sysfs_create_bin_file(&new_sc->kobj, &ring_buffer_bin_attr); + ret = hv_create_ring_sysfs(new_sc, hv_uio_ring_mmap); if (ret) { dev_err(device, "sysfs create ring bin file failed; %d\n", ret); vmbus_close(new_sc); @@ -350,10 +337,18 @@ hv_uio_probe(struct hv_device *dev, goto fail_close; } - ret = sysfs_create_bin_file(&channel->kobj, &ring_buffer_bin_attr); - if (ret) - dev_notice(&dev->device, - "sysfs create ring bin file failed; %d\n", ret); + /* + * This internally calls sysfs_update_group, which returns a non-zero value if it executes + * before sysfs_create_group. This is expected as the 'ring' will be created later in + * vmbus_device_register() -> vmbus_add_channel_kobj(). Thus, no need to check the return + * value and print warning. + * + * Creating/exposing sysfs in driver probe is not encouraged as it can lead to race + * conditions with userspace. For backward compatibility, "ring" sysfs could not be removed + * or decoupled from uio_hv_generic probe. Userspace programs can make use of inotify + * APIs to make sure that ring is created. + */ + hv_create_ring_sysfs(channel, hv_uio_ring_mmap); hv_set_drvdata(dev, pdata); @@ -375,7 +370,7 @@ hv_uio_remove(struct hv_device *dev) if (!pdata) return; - sysfs_remove_bin_file(&dev->channel->kobj, &ring_buffer_bin_attr); + hv_remove_ring_sysfs(dev->channel); uio_unregister_device(&pdata->info); hv_uio_cleanup(dev, pdata); diff --git a/include/linux/hyperv.h b/include/linux/hyperv.h index 675959fb97ba..d6ffe01962c2 100644 --- a/include/linux/hyperv.h +++ b/include/linux/hyperv.h @@ -1002,6 +1002,12 @@ struct vmbus_channel { /* The max size of a packet on this channel */ u32 max_pkt_size; + + /* function to mmap ring buffer memory to the channel's sysfs ring attribute */ + int (*mmap_ring_buffer)(struct vmbus_channel *channel, struct vm_area_struct *vma); + + /* boolean to control visibility of sysfs for ring buffer */ + bool ring_sysfs_visible; }; #define lock_requestor(channel, flags) \ From 65995e97a1caacf0024bebda3332b8d1f0f443c4 Mon Sep 17 00:00:00 2001 From: Naman Jain Date: Fri, 2 May 2025 13:18:11 +0530 Subject: [PATCH 136/555] Drivers: hv: Make the sysfs node size for the ring buffer dynamic The ring buffer size varies across VMBus channels. The size of sysfs node for the ring buffer is currently hardcoded to 4 MB. Userspace clients either use fstat() or hardcode this size for doing mmap(). To address this, make the sysfs node size dynamic to reflect the actual ring buffer size for each channel. This will ensure that fstat() on ring sysfs node always returns the correct size of ring buffer. Reviewed-by: Michael Kelley Tested-by: Michael Kelley Reviewed-by: Dexuan Cui Signed-off-by: Naman Jain Link: https://lore.kernel.org/r/20250502074811.2022-3-namjain@linux.microsoft.com Signed-off-by: Greg Kroah-Hartman --- drivers/hv/vmbus_drv.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/drivers/hv/vmbus_drv.c b/drivers/hv/vmbus_drv.c index 0f16a83cc2d6..e3d51a316316 100644 --- a/drivers/hv/vmbus_drv.c +++ b/drivers/hv/vmbus_drv.c @@ -1820,7 +1820,6 @@ static struct bin_attribute chan_attr_ring_buffer = { .name = "ring", .mode = 0600, }, - .size = 2 * SZ_2M, .mmap = hv_mmap_ring_buffer_wrapper, }; static struct attribute *vmbus_chan_attrs[] = { @@ -1880,11 +1879,21 @@ static umode_t vmbus_chan_bin_attr_is_visible(struct kobject *kobj, return attr->attr.mode; } +static size_t vmbus_chan_bin_size(struct kobject *kobj, + const struct bin_attribute *bin_attr, int a) +{ + const struct vmbus_channel *channel = + container_of(kobj, struct vmbus_channel, kobj); + + return channel->ringbuffer_pagecount << PAGE_SHIFT; +} + static const struct attribute_group vmbus_chan_group = { .attrs = vmbus_chan_attrs, .bin_attrs = vmbus_chan_bin_attrs, .is_visible = vmbus_chan_attr_is_visible, .is_bin_visible = vmbus_chan_bin_attr_is_visible, + .bin_size = vmbus_chan_bin_size, }; static const struct kobj_type vmbus_chan_ktype = { From d9ec73301099ec5975505e1c3effbe768bab9490 Mon Sep 17 00:00:00 2001 From: Max Kellermann Date: Tue, 29 Apr 2025 20:58:27 +0200 Subject: [PATCH 137/555] fs/eventpoll: fix endless busy loop after timeout has expired After commit 0a65bc27bd64 ("eventpoll: Set epoll timeout if it's in the future"), the following program would immediately enter a busy loop in the kernel: ``` int main() { int e = epoll_create1(0); struct epoll_event event = {.events = EPOLLIN}; epoll_ctl(e, EPOLL_CTL_ADD, 0, &event); const struct timespec timeout = {.tv_nsec = 1}; epoll_pwait2(e, &event, 1, &timeout, 0); } ``` This happens because the given (non-zero) timeout of 1 nanosecond usually expires before ep_poll() is entered and then ep_schedule_timeout() returns false, but `timed_out` is never set because the code line that sets it is skipped. This quickly turns into a soft lockup, RCU stalls and deadlocks, inflicting severe headaches to the whole system. When the timeout has expired, we don't need to schedule a hrtimer, but we should set the `timed_out` variable. Therefore, I suggest moving the ep_schedule_timeout() check into the `timed_out` expression instead of skipping it. brauner: Note that there was an earlier fix by Joe Damato in response to my bug report in [1]. Fixes: 0a65bc27bd64 ("eventpoll: Set epoll timeout if it's in the future") Cc: Joe Damato Cc: stable@vger.kernel.org Signed-off-by: Max Kellermann Link: https://lore.kernel.org/20250429153419.94723-1-jdamato@fastly.com [1] Link: https://lore.kernel.org/20250429185827.3564438-1-max.kellermann@ionos.com Reviewed-by: Jan Kara Signed-off-by: Christian Brauner --- fs/eventpoll.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/fs/eventpoll.c b/fs/eventpoll.c index 4bc264b854c4..d4dbffdedd08 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -2111,9 +2111,10 @@ static int ep_poll(struct eventpoll *ep, struct epoll_event __user *events, write_unlock_irq(&ep->lock); - if (!eavail && ep_schedule_timeout(to)) - timed_out = !schedule_hrtimeout_range(to, slack, - HRTIMER_MODE_ABS); + if (!eavail) + timed_out = !ep_schedule_timeout(to) || + !schedule_hrtimeout_range(to, slack, + HRTIMER_MODE_ABS); __set_current_state(TASK_RUNNING); /* From b662b162c3d06f120749eea0351ec9317d9dd905 Mon Sep 17 00:00:00 2001 From: Feng Jiang Date: Wed, 9 Apr 2025 09:46:33 +0800 Subject: [PATCH 138/555] drm: Fix potential overflow issue in event_string array MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When calling scnprintf() to append recovery method to event_string, the second argument should be `sizeof(event_string) - len`, otherwise there is a potential overflow problem. Fixes: b7cf9f4ac1b8 ("drm: Introduce device wedged event") Signed-off-by: Feng Jiang Reviewed-by: André Almeida Reviewed-by: Raag Jadav Link: https://lore.kernel.org/r/20250409014633.31303-1-jiangfeng@kylinos.cn Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/drm_drv.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c index 17fc5dc708f4..60e5ac179c15 100644 --- a/drivers/gpu/drm/drm_drv.c +++ b/drivers/gpu/drm/drm_drv.c @@ -549,7 +549,7 @@ int drm_dev_wedged_event(struct drm_device *dev, unsigned long method) if (drm_WARN_ONCE(dev, !recovery, "invalid recovery method %u\n", opt)) break; - len += scnprintf(event_string + len, sizeof(event_string), "%s,", recovery); + len += scnprintf(event_string + len, sizeof(event_string) - len, "%s,", recovery); } if (recovery) From 0c314cda93258cd1f0055a278a6576b5d4aeabf5 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Thomas=20Wei=C3=9Fschuh?= Date: Wed, 30 Apr 2025 11:20:13 +0200 Subject: [PATCH 139/555] arm64: vdso: Work around invalid absolute relocations from GCC MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit All vDSO code needs to be completely position independent. Symbol references are marked as hidden so the compiler emits PC-relative relocations. However GCC emits absolute relocations for symbol-relative references with an offset >= 64KiB. After recent refactorings in the vDSO code this is the case in __arch_get_vdso_u_timens_data() with a page size of 64KiB. Work around the issue by preventing the optimizer from seeing the offsets. Fixes: 83a2a6b8cfc5 ("vdso/gettimeofday: Prepare do_hres_timens() for introduction of struct vdso_clock") Reported-by: Jan Stancek Signed-off-by: Thomas Weißschuh Signed-off-by: Thomas Gleixner Acked-by: Catalin Marinas Link: https://lore.kernel.org/all/20250430-vdso-absolute-reloc-v2-1-5efcc3bc4b26@linutronix.de Link: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=120002 Closes: https://lore.kernel.org/lkml/aApGPAoctq_eoE2g@t14ultra/ --- arch/arm64/include/asm/vdso/gettimeofday.h | 13 +++++++++++++ 1 file changed, 13 insertions(+) diff --git a/arch/arm64/include/asm/vdso/gettimeofday.h b/arch/arm64/include/asm/vdso/gettimeofday.h index 92a2b59a9f3d..3322c7047d84 100644 --- a/arch/arm64/include/asm/vdso/gettimeofday.h +++ b/arch/arm64/include/asm/vdso/gettimeofday.h @@ -99,6 +99,19 @@ static __always_inline u64 __arch_get_hw_counter(s32 clock_mode, return res; } +#if IS_ENABLED(CONFIG_CC_IS_GCC) && IS_ENABLED(CONFIG_PAGE_SIZE_64KB) +static __always_inline const struct vdso_time_data *__arch_get_vdso_u_time_data(void) +{ + const struct vdso_time_data *ret = &vdso_u_time_data; + + /* Work around invalid absolute relocations */ + OPTIMIZER_HIDE_VAR(ret); + + return ret; +} +#define __arch_get_vdso_u_time_data __arch_get_vdso_u_time_data +#endif /* IS_ENABLED(CONFIG_CC_IS_GCC) && IS_ENABLED(CONFIG_PAGE_SIZE_64KB) */ + #endif /* !__ASSEMBLY__ */ #endif /* __ASM_VDSO_GETTIMEOFDAY_H */ From 35e4079bf1a2570abffce6ababa631afcf8ea0e5 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ma=C3=ADra=20Canal?= Date: Wed, 30 Apr 2025 17:51:52 -0300 Subject: [PATCH 140/555] drm/v3d: Add job to pending list if the reset was skipped MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When a CL/CSD job times out, we check if the GPU has made any progress since the last timeout. If so, instead of resetting the hardware, we skip the reset and let the timer get rearmed. This gives long-running jobs a chance to complete. However, when `timedout_job()` is called, the job in question is removed from the pending list, which means it won't be automatically freed through `free_job()`. Consequently, when we skip the reset and keep the job running, the job won't be freed when it finally completes. This situation leads to a memory leak, as exposed in [1] and [2]. Similarly to commit 704d3d60fec4 ("drm/etnaviv: don't block scheduler when GPU is still active"), this patch ensures the job is put back on the pending list when extending the timeout. Cc: stable@vger.kernel.org # 6.0 Reported-by: Daivik Bhatia Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/12227 [1] Closes: https://github.com/raspberrypi/linux/issues/6817 [2] Reviewed-by: Iago Toral Quiroga Acked-by: Tvrtko Ursulin Link: https://lore.kernel.org/r/20250430210643.57924-1-mcanal@igalia.com Signed-off-by: Maíra Canal --- drivers/gpu/drm/v3d/v3d_sched.c | 28 +++++++++++++++++++++------- 1 file changed, 21 insertions(+), 7 deletions(-) diff --git a/drivers/gpu/drm/v3d/v3d_sched.c b/drivers/gpu/drm/v3d/v3d_sched.c index 4a7701a33cf8..eb35482f6fb5 100644 --- a/drivers/gpu/drm/v3d/v3d_sched.c +++ b/drivers/gpu/drm/v3d/v3d_sched.c @@ -744,11 +744,16 @@ v3d_gpu_reset_for_timeout(struct v3d_dev *v3d, struct drm_sched_job *sched_job) return DRM_GPU_SCHED_STAT_NOMINAL; } -/* If the current address or return address have changed, then the GPU - * has probably made progress and we should delay the reset. This - * could fail if the GPU got in an infinite loop in the CL, but that - * is pretty unlikely outside of an i-g-t testcase. - */ +static void +v3d_sched_skip_reset(struct drm_sched_job *sched_job) +{ + struct drm_gpu_scheduler *sched = sched_job->sched; + + spin_lock(&sched->job_list_lock); + list_add(&sched_job->list, &sched->pending_list); + spin_unlock(&sched->job_list_lock); +} + static enum drm_gpu_sched_stat v3d_cl_job_timedout(struct drm_sched_job *sched_job, enum v3d_queue q, u32 *timedout_ctca, u32 *timedout_ctra) @@ -758,9 +763,16 @@ v3d_cl_job_timedout(struct drm_sched_job *sched_job, enum v3d_queue q, u32 ctca = V3D_CORE_READ(0, V3D_CLE_CTNCA(q)); u32 ctra = V3D_CORE_READ(0, V3D_CLE_CTNRA(q)); + /* If the current address or return address have changed, then the GPU + * has probably made progress and we should delay the reset. This + * could fail if the GPU got in an infinite loop in the CL, but that + * is pretty unlikely outside of an i-g-t testcase. + */ if (*timedout_ctca != ctca || *timedout_ctra != ctra) { *timedout_ctca = ctca; *timedout_ctra = ctra; + + v3d_sched_skip_reset(sched_job); return DRM_GPU_SCHED_STAT_NOMINAL; } @@ -800,11 +812,13 @@ v3d_csd_job_timedout(struct drm_sched_job *sched_job) struct v3d_dev *v3d = job->base.v3d; u32 batches = V3D_CORE_READ(0, V3D_CSD_CURRENT_CFG4(v3d->ver)); - /* If we've made progress, skip reset and let the timer get - * rearmed. + /* If we've made progress, skip reset, add the job to the pending + * list, and let the timer get rearmed. */ if (job->timedout_batches != batches) { job->timedout_batches = batches; + + v3d_sched_skip_reset(sched_job); return DRM_GPU_SCHED_STAT_NOMINAL; } From 5fea0c6c0ebe0a645c247f29fc683852cf51af70 Mon Sep 17 00:00:00 2001 From: Tom Lendacky Date: Mon, 28 Apr 2025 13:55:31 -0500 Subject: [PATCH 141/555] KVM: SVM: Update dump_ghcb() to use the GHCB snapshot fields Commit 4e15a0ddc3ff ("KVM: SEV: snapshot the GHCB before accessing it") updated the SEV code to take a snapshot of the GHCB before using it. But the dump_ghcb() function wasn't updated to use the snapshot locations. This results in incorrect output from dump_ghcb() for the "is_valid" and "valid_bitmap" fields. Update dump_ghcb() to use the proper locations. Fixes: 4e15a0ddc3ff ("KVM: SEV: snapshot the GHCB before accessing it") Signed-off-by: Tom Lendacky Reviewed-by: Liam Merwick Link: https://lore.kernel.org/r/8f03878443681496008b1b37b7c4bf77a342b459.1745866531.git.thomas.lendacky@amd.com [sean: add comment and snapshot qualifier] Signed-off-by: Sean Christopherson --- arch/x86/kvm/svm/sev.c | 32 +++++++++++++++++++------------- 1 file changed, 19 insertions(+), 13 deletions(-) diff --git a/arch/x86/kvm/svm/sev.c b/arch/x86/kvm/svm/sev.c index 0bc708ee2788..a7a7dc507336 100644 --- a/arch/x86/kvm/svm/sev.c +++ b/arch/x86/kvm/svm/sev.c @@ -3173,9 +3173,14 @@ skip_vmsa_free: kvfree(svm->sev_es.ghcb_sa); } +static u64 kvm_ghcb_get_sw_exit_code(struct vmcb_control_area *control) +{ + return (((u64)control->exit_code_hi) << 32) | control->exit_code; +} + static void dump_ghcb(struct vcpu_svm *svm) { - struct ghcb *ghcb = svm->sev_es.ghcb; + struct vmcb_control_area *control = &svm->vmcb->control; unsigned int nbits; /* Re-use the dump_invalid_vmcb module parameter */ @@ -3184,18 +3189,24 @@ static void dump_ghcb(struct vcpu_svm *svm) return; } - nbits = sizeof(ghcb->save.valid_bitmap) * 8; + nbits = sizeof(svm->sev_es.valid_bitmap) * 8; - pr_err("GHCB (GPA=%016llx):\n", svm->vmcb->control.ghcb_gpa); + /* + * Print KVM's snapshot of the GHCB values that were (unsuccessfully) + * used to handle the exit. If the guest has since modified the GHCB + * itself, dumping the raw GHCB won't help debug why KVM was unable to + * handle the VMGEXIT that KVM observed. + */ + pr_err("GHCB (GPA=%016llx) snapshot:\n", svm->vmcb->control.ghcb_gpa); pr_err("%-20s%016llx is_valid: %u\n", "sw_exit_code", - ghcb->save.sw_exit_code, ghcb_sw_exit_code_is_valid(ghcb)); + kvm_ghcb_get_sw_exit_code(control), kvm_ghcb_sw_exit_code_is_valid(svm)); pr_err("%-20s%016llx is_valid: %u\n", "sw_exit_info_1", - ghcb->save.sw_exit_info_1, ghcb_sw_exit_info_1_is_valid(ghcb)); + control->exit_info_1, kvm_ghcb_sw_exit_info_1_is_valid(svm)); pr_err("%-20s%016llx is_valid: %u\n", "sw_exit_info_2", - ghcb->save.sw_exit_info_2, ghcb_sw_exit_info_2_is_valid(ghcb)); + control->exit_info_2, kvm_ghcb_sw_exit_info_2_is_valid(svm)); pr_err("%-20s%016llx is_valid: %u\n", "sw_scratch", - ghcb->save.sw_scratch, ghcb_sw_scratch_is_valid(ghcb)); - pr_err("%-20s%*pb\n", "valid_bitmap", nbits, ghcb->save.valid_bitmap); + svm->sev_es.sw_scratch, kvm_ghcb_sw_scratch_is_valid(svm)); + pr_err("%-20s%*pb\n", "valid_bitmap", nbits, svm->sev_es.valid_bitmap); } static void sev_es_sync_to_ghcb(struct vcpu_svm *svm) @@ -3266,11 +3277,6 @@ static void sev_es_sync_from_ghcb(struct vcpu_svm *svm) memset(ghcb->save.valid_bitmap, 0, sizeof(ghcb->save.valid_bitmap)); } -static u64 kvm_ghcb_get_sw_exit_code(struct vmcb_control_area *control) -{ - return (((u64)control->exit_code_hi) << 32) | control->exit_code; -} - static int sev_es_validate_vmgexit(struct vcpu_svm *svm) { struct vmcb_control_area *control = &svm->vmcb->control; From 9129633d568edd36aa22bf703b12835153cec985 Mon Sep 17 00:00:00 2001 From: Sean Christopherson Date: Wed, 30 Apr 2025 15:09:54 -0700 Subject: [PATCH 142/555] KVM: x86/mmu: Prevent installing hugepages when mem attributes are changing When changing memory attributes on a subset of a potential hugepage, add the hugepage to the invalidation range tracking to prevent installing a hugepage until the attributes are fully updated. Like the actual hugepage tracking updates in kvm_arch_post_set_memory_attributes(), process only the head and tail pages, as any potential hugepages that are entirely covered by the range will already be tracked. Note, only hugepage chunks whose current attributes are NOT mixed need to be added to the invalidation set, as mixed attributes already prevent installing a hugepage, and it's perfectly safe to install a smaller mapping for a gfn whose attributes aren't changing. Fixes: 8dd2eee9d526 ("KVM: x86/mmu: Handle page fault for private memory") Cc: stable@vger.kernel.org Reported-by: Michael Roth Tested-by: Michael Roth Link: https://lore.kernel.org/r/20250430220954.522672-1-seanjc@google.com Signed-off-by: Sean Christopherson --- arch/x86/kvm/mmu/mmu.c | 89 ++++++++++++++++++++++++++++++------------ 1 file changed, 63 insertions(+), 26 deletions(-) diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c index 387464ebe8e8..8d1b632e33d2 100644 --- a/arch/x86/kvm/mmu/mmu.c +++ b/arch/x86/kvm/mmu/mmu.c @@ -7670,32 +7670,6 @@ void kvm_mmu_pre_destroy_vm(struct kvm *kvm) } #ifdef CONFIG_KVM_GENERIC_MEMORY_ATTRIBUTES -bool kvm_arch_pre_set_memory_attributes(struct kvm *kvm, - struct kvm_gfn_range *range) -{ - /* - * Zap SPTEs even if the slot can't be mapped PRIVATE. KVM x86 only - * supports KVM_MEMORY_ATTRIBUTE_PRIVATE, and so it *seems* like KVM - * can simply ignore such slots. But if userspace is making memory - * PRIVATE, then KVM must prevent the guest from accessing the memory - * as shared. And if userspace is making memory SHARED and this point - * is reached, then at least one page within the range was previously - * PRIVATE, i.e. the slot's possible hugepage ranges are changing. - * Zapping SPTEs in this case ensures KVM will reassess whether or not - * a hugepage can be used for affected ranges. - */ - if (WARN_ON_ONCE(!kvm_arch_has_private_mem(kvm))) - return false; - - /* Unmap the old attribute page. */ - if (range->arg.attributes & KVM_MEMORY_ATTRIBUTE_PRIVATE) - range->attr_filter = KVM_FILTER_SHARED; - else - range->attr_filter = KVM_FILTER_PRIVATE; - - return kvm_unmap_gfn_range(kvm, range); -} - static bool hugepage_test_mixed(struct kvm_memory_slot *slot, gfn_t gfn, int level) { @@ -7714,6 +7688,69 @@ static void hugepage_set_mixed(struct kvm_memory_slot *slot, gfn_t gfn, lpage_info_slot(gfn, slot, level)->disallow_lpage |= KVM_LPAGE_MIXED_FLAG; } +bool kvm_arch_pre_set_memory_attributes(struct kvm *kvm, + struct kvm_gfn_range *range) +{ + struct kvm_memory_slot *slot = range->slot; + int level; + + /* + * Zap SPTEs even if the slot can't be mapped PRIVATE. KVM x86 only + * supports KVM_MEMORY_ATTRIBUTE_PRIVATE, and so it *seems* like KVM + * can simply ignore such slots. But if userspace is making memory + * PRIVATE, then KVM must prevent the guest from accessing the memory + * as shared. And if userspace is making memory SHARED and this point + * is reached, then at least one page within the range was previously + * PRIVATE, i.e. the slot's possible hugepage ranges are changing. + * Zapping SPTEs in this case ensures KVM will reassess whether or not + * a hugepage can be used for affected ranges. + */ + if (WARN_ON_ONCE(!kvm_arch_has_private_mem(kvm))) + return false; + + if (WARN_ON_ONCE(range->end <= range->start)) + return false; + + /* + * If the head and tail pages of the range currently allow a hugepage, + * i.e. reside fully in the slot and don't have mixed attributes, then + * add each corresponding hugepage range to the ongoing invalidation, + * e.g. to prevent KVM from creating a hugepage in response to a fault + * for a gfn whose attributes aren't changing. Note, only the range + * of gfns whose attributes are being modified needs to be explicitly + * unmapped, as that will unmap any existing hugepages. + */ + for (level = PG_LEVEL_2M; level <= KVM_MAX_HUGEPAGE_LEVEL; level++) { + gfn_t start = gfn_round_for_level(range->start, level); + gfn_t end = gfn_round_for_level(range->end - 1, level); + gfn_t nr_pages = KVM_PAGES_PER_HPAGE(level); + + if ((start != range->start || start + nr_pages > range->end) && + start >= slot->base_gfn && + start + nr_pages <= slot->base_gfn + slot->npages && + !hugepage_test_mixed(slot, start, level)) + kvm_mmu_invalidate_range_add(kvm, start, start + nr_pages); + + if (end == start) + continue; + + if ((end + nr_pages) > range->end && + (end + nr_pages) <= (slot->base_gfn + slot->npages) && + !hugepage_test_mixed(slot, end, level)) + kvm_mmu_invalidate_range_add(kvm, end, end + nr_pages); + } + + /* Unmap the old attribute page. */ + if (range->arg.attributes & KVM_MEMORY_ATTRIBUTE_PRIVATE) + range->attr_filter = KVM_FILTER_SHARED; + else + range->attr_filter = KVM_FILTER_PRIVATE; + + return kvm_unmap_gfn_range(kvm, range); +} + + + static bool hugepage_has_attrs(struct kvm *kvm, struct kvm_memory_slot *slot, gfn_t gfn, int level, unsigned long attrs) { From b1525d0a8d3ad98487775232c3a861f3f4d894ea Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Micka=C3=ABl=20Sala=C3=BCn?= Date: Sat, 3 May 2025 08:53:58 +0200 Subject: [PATCH 143/555] landlock: Remove KUnit test that triggers a warning MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit A KUnit test checking boundaries triggers a canary warning, which may be disturbing. Let's remove this test for now. Hopefully, KUnit will soon get support for suppressing warning backtraces [1]. Cc: Alessandro Carminati Cc: Andrew Morton Cc: Günther Noack Reported-by: Tingmao Wang Closes: https://lore.kernel.org/r/20250327213807.12964-1-m@maowtm.org Link: https://lore.kernel.org/r/20250425193249.78b45d2589575c15f483c3d8@linux-foundation.org [1] Link: https://lore.kernel.org/r/20250503065359.3625407-1-mic@digikod.net Signed-off-by: Mickaël Salaün --- security/landlock/audit.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/security/landlock/audit.c b/security/landlock/audit.c index 7e5e0ed0e4e5..58d5c40d4d0e 100644 --- a/security/landlock/audit.c +++ b/security/landlock/audit.c @@ -175,7 +175,7 @@ static void test_get_hierarchy(struct kunit *const test) KUNIT_EXPECT_EQ(test, 10, get_hierarchy(&dom2, 0)->id); KUNIT_EXPECT_EQ(test, 20, get_hierarchy(&dom2, 1)->id); KUNIT_EXPECT_EQ(test, 30, get_hierarchy(&dom2, 2)->id); - KUNIT_EXPECT_EQ(test, 30, get_hierarchy(&dom2, -1)->id); + /* KUNIT_EXPECT_EQ(test, 30, get_hierarchy(&dom2, -1)->id); */ } #endif /* CONFIG_SECURITY_LANDLOCK_KUNIT_TEST */ From b53e523261bf058ea4a518b482222e7a277b186b Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Sun, 4 May 2025 08:06:28 -0600 Subject: [PATCH 144/555] io_uring: always arm linked timeouts prior to issue There are a few spots where linked timeouts are armed, and not all of them adhere to the pre-arm, attempt issue, post-arm pattern. This can be problematic if the linked request returns that it will trigger a callback later, and does so before the linked timeout is fully armed. Consolidate all the linked timeout handling into __io_issue_sqe(), rather than have it spread throughout the various issue entry points. Cc: stable@vger.kernel.org Link: https://github.com/axboe/liburing/issues/1390 Reported-by: Chase Hiltz Signed-off-by: Jens Axboe --- io_uring/io_uring.c | 50 ++++++++++++++------------------------------- 1 file changed, 15 insertions(+), 35 deletions(-) diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c index a2b256e96d5d..769814d71153 100644 --- a/io_uring/io_uring.c +++ b/io_uring/io_uring.c @@ -448,24 +448,6 @@ static struct io_kiocb *__io_prep_linked_timeout(struct io_kiocb *req) return req->link; } -static inline struct io_kiocb *io_prep_linked_timeout(struct io_kiocb *req) -{ - if (likely(!(req->flags & REQ_F_ARM_LTIMEOUT))) - return NULL; - return __io_prep_linked_timeout(req); -} - -static noinline void __io_arm_ltimeout(struct io_kiocb *req) -{ - io_queue_linked_timeout(__io_prep_linked_timeout(req)); -} - -static inline void io_arm_ltimeout(struct io_kiocb *req) -{ - if (unlikely(req->flags & REQ_F_ARM_LTIMEOUT)) - __io_arm_ltimeout(req); -} - static void io_prep_async_work(struct io_kiocb *req) { const struct io_issue_def *def = &io_issue_defs[req->opcode]; @@ -518,7 +500,6 @@ static void io_prep_async_link(struct io_kiocb *req) static void io_queue_iowq(struct io_kiocb *req) { - struct io_kiocb *link = io_prep_linked_timeout(req); struct io_uring_task *tctx = req->tctx; BUG_ON(!tctx); @@ -543,8 +524,6 @@ static void io_queue_iowq(struct io_kiocb *req) trace_io_uring_queue_async_work(req, io_wq_is_hashed(&req->work)); io_wq_enqueue(tctx->io_wq, &req->work); - if (link) - io_queue_linked_timeout(link); } static void io_req_queue_iowq_tw(struct io_kiocb *req, io_tw_token_t tw) @@ -1724,15 +1703,22 @@ static bool io_assign_file(struct io_kiocb *req, const struct io_issue_def *def, return !!req->file; } +#define REQ_ISSUE_SLOW_FLAGS (REQ_F_CREDS | REQ_F_ARM_LTIMEOUT) + static inline int __io_issue_sqe(struct io_kiocb *req, unsigned int issue_flags, const struct io_issue_def *def) { const struct cred *creds = NULL; + struct io_kiocb *link = NULL; int ret; - if (unlikely((req->flags & REQ_F_CREDS) && req->creds != current_cred())) - creds = override_creds(req->creds); + if (unlikely(req->flags & REQ_ISSUE_SLOW_FLAGS)) { + if ((req->flags & REQ_F_CREDS) && req->creds != current_cred()) + creds = override_creds(req->creds); + if (req->flags & REQ_F_ARM_LTIMEOUT) + link = __io_prep_linked_timeout(req); + } if (!def->audit_skip) audit_uring_entry(req->opcode); @@ -1742,8 +1728,12 @@ static inline int __io_issue_sqe(struct io_kiocb *req, if (!def->audit_skip) audit_uring_exit(!ret, ret); - if (creds) - revert_creds(creds); + if (unlikely(creds || link)) { + if (creds) + revert_creds(creds); + if (link) + io_queue_linked_timeout(link); + } return ret; } @@ -1769,7 +1759,6 @@ static int io_issue_sqe(struct io_kiocb *req, unsigned int issue_flags) if (ret == IOU_ISSUE_SKIP_COMPLETE) { ret = 0; - io_arm_ltimeout(req); /* If the op doesn't have a file, we're not polling for it */ if ((req->ctx->flags & IORING_SETUP_IOPOLL) && def->iopoll_queue) @@ -1824,8 +1813,6 @@ void io_wq_submit_work(struct io_wq_work *work) else req_ref_get(req); - io_arm_ltimeout(req); - /* either cancelled or io-wq is dying, so don't touch tctx->iowq */ if (atomic_read(&work->flags) & IO_WQ_WORK_CANCEL) { fail: @@ -1941,15 +1928,11 @@ struct file *io_file_get_normal(struct io_kiocb *req, int fd) static void io_queue_async(struct io_kiocb *req, int ret) __must_hold(&req->ctx->uring_lock) { - struct io_kiocb *linked_timeout; - if (ret != -EAGAIN || (req->flags & REQ_F_NOWAIT)) { io_req_defer_failed(req, ret); return; } - linked_timeout = io_prep_linked_timeout(req); - switch (io_arm_poll_handler(req, 0)) { case IO_APOLL_READY: io_kbuf_recycle(req, 0); @@ -1962,9 +1945,6 @@ static void io_queue_async(struct io_kiocb *req, int ret) case IO_APOLL_OK: break; } - - if (linked_timeout) - io_queue_linked_timeout(linked_timeout); } static inline void io_queue_sqe(struct io_kiocb *req) From df2e19a883fdea698cdbed4987987db999b19d58 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 4 May 2025 13:50:09 -0400 Subject: [PATCH 145/555] bcachefs: thread_with_stdio: fix spinning instead of exiting bch2_stdio_redirect_vprintf() was missing a check for stdio->done, i.e. exiting. This caused the thread attempting to print to spin, and since it was being called from the kthread ran by thread_with_stdio, the userspace side hung as well. Change it to return -EPIPE - i.e. writing to a pipe that's been closed. Reported-by: Jan Solanti Signed-off-by: Kent Overstreet --- fs/bcachefs/thread_with_file.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/fs/bcachefs/thread_with_file.c b/fs/bcachefs/thread_with_file.c index dea73bc1cb51..314a24d15d4e 100644 --- a/fs/bcachefs/thread_with_file.c +++ b/fs/bcachefs/thread_with_file.c @@ -455,8 +455,10 @@ ssize_t bch2_stdio_redirect_vprintf(struct stdio_redirect *stdio, bool nonblocki struct stdio_buf *buf = &stdio->output; unsigned long flags; ssize_t ret; - again: + if (stdio->done) + return -EPIPE; + spin_lock_irqsave(&buf->lock, flags); ret = bch2_darray_vprintf(&buf->buf, GFP_NOWAIT, fmt, args); spin_unlock_irqrestore(&buf->lock, flags); From 5214a9f6c0f56644acb9d2cbb58facf1856d322b Mon Sep 17 00:00:00 2001 From: "Borislav Petkov (AMD)" Date: Mon, 14 Apr 2025 11:59:33 +0200 Subject: [PATCH 146/555] x86/microcode: Consolidate the loader enablement checking MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Consolidate the whole logic which determines whether the microcode loader should be enabled or not into a single function and call it everywhere. Well, almost everywhere - not in mk_early_pgtbl_32() because there the kernel is running without paging enabled and checking dis_ucode_ldr et al would require physical addresses and uglification of the code. But since this is 32-bit, the easier thing to do is to simply map the initrd unconditionally especially since that mapping is getting removed later anyway by zap_early_initrd_mapping() and avoid the uglification. In doing so, address the issue of old 486er machines without CPUID support, not booting current kernels. [ mingo: Fix no previous prototype for ‘microcode_loader_disabled’ [-Wmissing-prototypes] ] Fixes: 4c585af7180c1 ("x86/boot/32: Temporarily map initrd for microcode loading") Signed-off-by: Borislav Petkov (AMD) Signed-off-by: Ingo Molnar Signed-off-by: Borislav Petkov (AMD) Cc: Link: https://lore.kernel.org/r/CANpbe9Wm3z8fy9HbgS8cuhoj0TREYEEkBipDuhgkWFvqX0UoVQ@mail.gmail.com --- arch/x86/include/asm/microcode.h | 2 + arch/x86/kernel/cpu/microcode/amd.c | 6 ++- arch/x86/kernel/cpu/microcode/core.c | 60 ++++++++++++++---------- arch/x86/kernel/cpu/microcode/intel.c | 2 +- arch/x86/kernel/cpu/microcode/internal.h | 1 - arch/x86/kernel/head32.c | 4 -- 6 files changed, 42 insertions(+), 33 deletions(-) diff --git a/arch/x86/include/asm/microcode.h b/arch/x86/include/asm/microcode.h index 695e569159c1..be7cddc414e4 100644 --- a/arch/x86/include/asm/microcode.h +++ b/arch/x86/include/asm/microcode.h @@ -17,10 +17,12 @@ struct ucode_cpu_info { void load_ucode_bsp(void); void load_ucode_ap(void); void microcode_bsp_resume(void); +bool __init microcode_loader_disabled(void); #else static inline void load_ucode_bsp(void) { } static inline void load_ucode_ap(void) { } static inline void microcode_bsp_resume(void) { } +static inline bool __init microcode_loader_disabled(void) { return false; } #endif extern unsigned long initrd_start_early; diff --git a/arch/x86/kernel/cpu/microcode/amd.c b/arch/x86/kernel/cpu/microcode/amd.c index 4a10d35e70aa..96cb992d50ef 100644 --- a/arch/x86/kernel/cpu/microcode/amd.c +++ b/arch/x86/kernel/cpu/microcode/amd.c @@ -1098,15 +1098,17 @@ static enum ucode_state load_microcode_amd(u8 family, const u8 *data, size_t siz static int __init save_microcode_in_initrd(void) { - unsigned int cpuid_1_eax = native_cpuid_eax(1); struct cpuinfo_x86 *c = &boot_cpu_data; struct cont_desc desc = { 0 }; + unsigned int cpuid_1_eax; enum ucode_state ret; struct cpio_data cp; - if (dis_ucode_ldr || c->x86_vendor != X86_VENDOR_AMD || c->x86 < 0x10) + if (microcode_loader_disabled() || c->x86_vendor != X86_VENDOR_AMD || c->x86 < 0x10) return 0; + cpuid_1_eax = native_cpuid_eax(1); + if (!find_blobs_in_containers(&cp)) return -EINVAL; diff --git a/arch/x86/kernel/cpu/microcode/core.c b/arch/x86/kernel/cpu/microcode/core.c index b3658d11e7b6..079f046ee26d 100644 --- a/arch/x86/kernel/cpu/microcode/core.c +++ b/arch/x86/kernel/cpu/microcode/core.c @@ -41,8 +41,8 @@ #include "internal.h" -static struct microcode_ops *microcode_ops; -bool dis_ucode_ldr = true; +static struct microcode_ops *microcode_ops; +static bool dis_ucode_ldr = false; bool force_minrev = IS_ENABLED(CONFIG_MICROCODE_LATE_FORCE_MINREV); module_param(force_minrev, bool, S_IRUSR | S_IWUSR); @@ -84,6 +84,9 @@ static bool amd_check_current_patch_level(void) u32 lvl, dummy, i; u32 *levels; + if (x86_cpuid_vendor() != X86_VENDOR_AMD) + return false; + native_rdmsr(MSR_AMD64_PATCH_LEVEL, lvl, dummy); levels = final_levels; @@ -95,27 +98,29 @@ static bool amd_check_current_patch_level(void) return false; } -static bool __init check_loader_disabled_bsp(void) +bool __init microcode_loader_disabled(void) { - static const char *__dis_opt_str = "dis_ucode_ldr"; - const char *cmdline = boot_command_line; - const char *option = __dis_opt_str; - - /* - * CPUID(1).ECX[31]: reserved for hypervisor use. This is still not - * completely accurate as xen pv guests don't see that CPUID bit set but - * that's good enough as they don't land on the BSP path anyway. - */ - if (native_cpuid_ecx(1) & BIT(31)) + if (dis_ucode_ldr) return true; - if (x86_cpuid_vendor() == X86_VENDOR_AMD) { - if (amd_check_current_patch_level()) - return true; - } - - if (cmdline_find_option_bool(cmdline, option) <= 0) - dis_ucode_ldr = false; + /* + * Disable when: + * + * 1) The CPU does not support CPUID. + * + * 2) Bit 31 in CPUID[1]:ECX is clear + * The bit is reserved for hypervisor use. This is still not + * completely accurate as XEN PV guests don't see that CPUID bit + * set, but that's good enough as they don't land on the BSP + * path anyway. + * + * 3) Certain AMD patch levels are not allowed to be + * overwritten. + */ + if (!have_cpuid_p() || + native_cpuid_ecx(1) & BIT(31) || + amd_check_current_patch_level()) + dis_ucode_ldr = true; return dis_ucode_ldr; } @@ -125,7 +130,10 @@ void __init load_ucode_bsp(void) unsigned int cpuid_1_eax; bool intel = true; - if (!have_cpuid_p()) + if (cmdline_find_option_bool(boot_command_line, "dis_ucode_ldr") > 0) + dis_ucode_ldr = true; + + if (microcode_loader_disabled()) return; cpuid_1_eax = native_cpuid_eax(1); @@ -146,9 +154,6 @@ void __init load_ucode_bsp(void) return; } - if (check_loader_disabled_bsp()) - return; - if (intel) load_ucode_intel_bsp(&early_data); else @@ -159,6 +164,11 @@ void load_ucode_ap(void) { unsigned int cpuid_1_eax; + /* + * Can't use microcode_loader_disabled() here - .init section + * hell. It doesn't have to either - the BSP variant must've + * parsed cmdline already anyway. + */ if (dis_ucode_ldr) return; @@ -810,7 +820,7 @@ static int __init microcode_init(void) struct cpuinfo_x86 *c = &boot_cpu_data; int error; - if (dis_ucode_ldr) + if (microcode_loader_disabled()) return -EINVAL; if (c->x86_vendor == X86_VENDOR_INTEL) diff --git a/arch/x86/kernel/cpu/microcode/intel.c b/arch/x86/kernel/cpu/microcode/intel.c index 819199bc0119..2a397da43923 100644 --- a/arch/x86/kernel/cpu/microcode/intel.c +++ b/arch/x86/kernel/cpu/microcode/intel.c @@ -389,7 +389,7 @@ static int __init save_builtin_microcode(void) if (xchg(&ucode_patch_va, NULL) != UCODE_BSP_LOADED) return 0; - if (dis_ucode_ldr || boot_cpu_data.x86_vendor != X86_VENDOR_INTEL) + if (microcode_loader_disabled() || boot_cpu_data.x86_vendor != X86_VENDOR_INTEL) return 0; uci.mc = get_microcode_blob(&uci, true); diff --git a/arch/x86/kernel/cpu/microcode/internal.h b/arch/x86/kernel/cpu/microcode/internal.h index 5df621752fef..50a9702ae4e2 100644 --- a/arch/x86/kernel/cpu/microcode/internal.h +++ b/arch/x86/kernel/cpu/microcode/internal.h @@ -94,7 +94,6 @@ static inline unsigned int x86_cpuid_family(void) return x86_family(eax); } -extern bool dis_ucode_ldr; extern bool force_minrev; #ifdef CONFIG_CPU_SUP_AMD diff --git a/arch/x86/kernel/head32.c b/arch/x86/kernel/head32.c index de001b2146ab..375f2d7f1762 100644 --- a/arch/x86/kernel/head32.c +++ b/arch/x86/kernel/head32.c @@ -145,10 +145,6 @@ void __init __no_stack_protector mk_early_pgtbl_32(void) *ptr = (unsigned long)ptep + PAGE_OFFSET; #ifdef CONFIG_MICROCODE_INITRD32 - /* Running on a hypervisor? */ - if (native_cpuid_ecx(1) & BIT(31)) - return; - params = (struct boot_params *)__pa_nodebug(&boot_params); if (!params->hdr.ramdisk_size || !params->hdr.ramdisk_image) return; From 936b73feab5cd7eae8fe3d08a7ac9b1f8ac68042 Mon Sep 17 00:00:00 2001 From: Vinay Belgaumkar Date: Mon, 28 Apr 2025 11:35:55 -0700 Subject: [PATCH 147/555] drm/i915/slpc: Balance the inc/dec for num_waiters As seen in some recent failures, SLPC num_waiters value is < 0. This happens because the inc/dec are not balanced. We should skip decrement for the same conditions as the increment. Currently, we do that for power saving profile mode. This patch also ensures that num_waiters is incremented in the case min_softlimit is at boost freq. It ensures that we don't reduce the frequency while this request is in flight. v2: Add Fixes tags Closes: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/13598 Fixes: f864a29afc32 ("drm/i915/slpc: Optmize waitboost for SLPC") Fixes: 4a82ceb04ad4 ("drm/i915/slpc: Add sysfs for SLPC power profiles") Cc: Sk Anirban Reviewed-by: Sk Anirban Signed-off-by: Vinay Belgaumkar Signed-off-by: Daniele Ceraolo Spurio Link: https://lore.kernel.org/r/20250428183555.3250021-1-vinay.belgaumkar@intel.com (cherry picked from commit d26e55085f4b7a63677670db827541209257b313) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/gt/intel_rps.c | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/i915/gt/intel_rps.c b/drivers/gpu/drm/i915/gt/intel_rps.c index 64e9317f58fb..71ee01d9ef64 100644 --- a/drivers/gpu/drm/i915/gt/intel_rps.c +++ b/drivers/gpu/drm/i915/gt/intel_rps.c @@ -1001,6 +1001,10 @@ void intel_rps_dec_waiters(struct intel_rps *rps) if (rps_uses_slpc(rps)) { slpc = rps_to_slpc(rps); + /* Don't decrement num_waiters for req where increment was skipped */ + if (slpc->power_profile == SLPC_POWER_PROFILES_POWER_SAVING) + return; + intel_guc_slpc_dec_waiters(slpc); } else { atomic_dec(&rps->num_waiters); @@ -1029,11 +1033,15 @@ void intel_rps_boost(struct i915_request *rq) if (slpc->power_profile == SLPC_POWER_PROFILES_POWER_SAVING) return; - if (slpc->min_freq_softlimit >= slpc->boost_freq) - return; - /* Return if old value is non zero */ if (!atomic_fetch_inc(&slpc->num_waiters)) { + /* + * Skip queuing boost work if frequency is already boosted, + * but still increment num_waiters. + */ + if (slpc->min_freq_softlimit >= slpc->boost_freq) + return; + GT_TRACE(rps_to_gt(rps), "boost fence:%llx:%llx\n", rq->fence.context, rq->fence.seqno); queue_work(rps_to_gt(rps)->i915->unordered_wq, From a73fa3690a1f3014d6677e368dce4e70767a6ba2 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Fri, 2 May 2025 13:10:35 +0200 Subject: [PATCH 148/555] spi: loopback-test: Do not split 1024-byte hexdumps spi_test_print_hex_dump() prints buffers holding less than 1024 bytes in full. Larger buffers are truncated: only the first 512 and the last 512 bytes are printed, separated by a truncation message. The latter is confusing in case the buffer holds exactly 1024 bytes, as all data is printed anyway. Fix this by printing buffers holding up to and including 1024 bytes in full. Fixes: 84e0c4e5e2c4ef42 ("spi: add loopback test driver to allow for spi_master regression tests") Signed-off-by: Geert Uytterhoeven Link: https://patch.msgid.link/37ee1bc90c6554c9347040adabf04188c8f704aa.1746184171.git.geert+renesas@glider.be Signed-off-by: Mark Brown --- drivers/spi/spi-loopback-test.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/spi/spi-loopback-test.c b/drivers/spi/spi-loopback-test.c index 31a878d9458d..7740f94847a8 100644 --- a/drivers/spi/spi-loopback-test.c +++ b/drivers/spi/spi-loopback-test.c @@ -420,7 +420,7 @@ MODULE_LICENSE("GPL"); static void spi_test_print_hex_dump(char *pre, const void *ptr, size_t len) { /* limit the hex_dump */ - if (len < 1024) { + if (len <= 1024) { print_hex_dump(KERN_INFO, pre, DUMP_PREFIX_OFFSET, 16, 1, ptr, len, 0); From 0f67578587bb9e5a8eecfcdf6b8a501b5bd90526 Mon Sep 17 00:00:00 2001 From: Christian Hewitt Date: Sat, 3 May 2025 08:44:43 +0000 Subject: [PATCH 149/555] arm64: dts: amlogic: dreambox: fix missing clkc_audio node Add the clkc_audio node to fix audio support on Dreambox One/Two. Fixes: 83a6f4c62cb1 ("arm64: dts: meson: add initial support for Dreambox One/Two") CC: stable@vger.kernel.org Suggested-by: Emanuel Strobel Signed-off-by: Christian Hewitt Reviewed-by: Martin Blumenstingl Link: https://lore.kernel.org/r/20250503084443.3704866-1-christianshewitt@gmail.com Signed-off-by: Neil Armstrong --- arch/arm64/boot/dts/amlogic/meson-g12b-dreambox.dtsi | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/arch/arm64/boot/dts/amlogic/meson-g12b-dreambox.dtsi b/arch/arm64/boot/dts/amlogic/meson-g12b-dreambox.dtsi index de35fa2d7a6d..8e3e3354ed67 100644 --- a/arch/arm64/boot/dts/amlogic/meson-g12b-dreambox.dtsi +++ b/arch/arm64/boot/dts/amlogic/meson-g12b-dreambox.dtsi @@ -116,6 +116,10 @@ status = "okay"; }; +&clkc_audio { + status = "okay"; +}; + &frddr_a { status = "okay"; }; From eb16b3727c05ed36420c90eca1e8f0e279514c1c Mon Sep 17 00:00:00 2001 From: Nylon Chen Date: Fri, 11 Apr 2025 15:38:49 +0800 Subject: [PATCH 150/555] riscv: misaligned: Add handling for ZCB instructions Add support for the Zcb extension's compressed half-word instructions (C.LHU, C.LH, and C.SH) in the RISC-V misaligned access trap handler. Signed-off-by: Zong Li Signed-off-by: Nylon Chen Fixes: 956d705dd279 ("riscv: Unaligned load/store handling for M_MODE") Reviewed-by: Alexandre Ghiti Link: https://lore.kernel.org/r/20250411073850.3699180-2-nylon.chen@sifive.com Signed-off-by: Alexandre Ghiti --- arch/riscv/kernel/traps_misaligned.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/arch/riscv/kernel/traps_misaligned.c b/arch/riscv/kernel/traps_misaligned.c index 4354c87c0376..dde5d11dc1b5 100644 --- a/arch/riscv/kernel/traps_misaligned.c +++ b/arch/riscv/kernel/traps_misaligned.c @@ -88,6 +88,13 @@ #define INSN_MATCH_C_FSWSP 0xe002 #define INSN_MASK_C_FSWSP 0xe003 +#define INSN_MATCH_C_LHU 0x8400 +#define INSN_MASK_C_LHU 0xfc43 +#define INSN_MATCH_C_LH 0x8440 +#define INSN_MASK_C_LH 0xfc43 +#define INSN_MATCH_C_SH 0x8c00 +#define INSN_MASK_C_SH 0xfc43 + #define INSN_LEN(insn) ((((insn) & 0x3) < 0x3) ? 2 : 4) #if defined(CONFIG_64BIT) @@ -431,6 +438,13 @@ static int handle_scalar_misaligned_load(struct pt_regs *regs) fp = 1; len = 4; #endif + } else if ((insn & INSN_MASK_C_LHU) == INSN_MATCH_C_LHU) { + len = 2; + insn = RVC_RS2S(insn) << SH_RD; + } else if ((insn & INSN_MASK_C_LH) == INSN_MATCH_C_LH) { + len = 2; + shift = 8 * (sizeof(ulong) - len); + insn = RVC_RS2S(insn) << SH_RD; } else { regs->epc = epc; return -1; @@ -530,6 +544,9 @@ static int handle_scalar_misaligned_store(struct pt_regs *regs) len = 4; val.data_ulong = GET_F32_RS2C(insn, regs); #endif + } else if ((insn & INSN_MASK_C_SH) == INSN_MATCH_C_SH) { + len = 2; + val.data_ulong = GET_RS2S(insn, regs); } else { regs->epc = epc; return -1; From f5c84eff634ba003326aa034c414e2a9dcb7c6a7 Mon Sep 17 00:00:00 2001 From: Lizhi Xu Date: Mon, 28 Apr 2025 22:36:26 +0800 Subject: [PATCH 151/555] loop: Add sanity check for read/write_iter Some file systems do not support read_iter/write_iter, such as selinuxfs in this issue. So before calling them, first confirm that the interface is supported and then call it. It is releavant in that vfs_iter_read/write have the check, and removal of their used caused szybot to be able to hit this issue. Fixes: f2fed441c69b ("loop: stop using vfs_iter__{read,write} for buffered I/O") Reported-by: syzbot+6af973a3b8dfd2faefdc@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=6af973a3b8dfd2faefdc Signed-off-by: Lizhi Xu Reviewed-by: Christoph Hellwig Link: https://lore.kernel.org/r/20250428143626.3318717-1-lizhi.xu@windriver.com Signed-off-by: Jens Axboe --- drivers/block/loop.c | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+) diff --git a/drivers/block/loop.c b/drivers/block/loop.c index 46cba261075f..b8ba7de08753 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -505,6 +505,17 @@ static void loop_assign_backing_file(struct loop_device *lo, struct file *file) lo->lo_min_dio_size = loop_query_min_dio_size(lo); } +static int loop_check_backing_file(struct file *file) +{ + if (!file->f_op->read_iter) + return -EINVAL; + + if ((file->f_mode & FMODE_WRITE) && !file->f_op->write_iter) + return -EINVAL; + + return 0; +} + /* * loop_change_fd switched the backing store of a loopback device to * a new file. This is useful for operating system installers to free up @@ -526,6 +537,10 @@ static int loop_change_fd(struct loop_device *lo, struct block_device *bdev, if (!file) return -EBADF; + error = loop_check_backing_file(file); + if (error) + return error; + /* suppress uevents while reconfiguring the device */ dev_set_uevent_suppress(disk_to_dev(lo->lo_disk), 1); @@ -963,6 +978,14 @@ static int loop_configure(struct loop_device *lo, blk_mode_t mode, if (!file) return -EBADF; + + if ((mode & BLK_OPEN_WRITE) && !file->f_op->write_iter) + return -EINVAL; + + error = loop_check_backing_file(file); + if (error) + return error; + is_loop = is_loop_device(file); /* This is safe, since we have a reference from open(). */ From 94cff94634e506a4a44684bee1875d2dbf782722 Mon Sep 17 00:00:00 2001 From: Sebastian Andrzej Siewior Date: Fri, 4 Apr 2025 15:31:16 +0200 Subject: [PATCH 152/555] clocksource/i8253: Use raw_spinlock_irqsave() in clockevent_i8253_disable() On x86 during boot, clockevent_i8253_disable() can be invoked via x86_late_time_init -> hpet_time_init() -> pit_timer_init() which happens with enabled interrupts. If some of the old i8253 hardware is actually used then lockdep will notice that i8253_lock is used in hard interrupt context. This causes lockdep to complain because it observed the lock being acquired with interrupts enabled and in hard interrupt context. Make clockevent_i8253_disable() acquire the lock with raw_spinlock_irqsave() to cure this. [ tglx: Massage change log and use guard() ] Fixes: c8c4076723dac ("x86/timer: Skip PIT initialization on modern chipsets") Signed-off-by: Sebastian Andrzej Siewior Signed-off-by: Thomas Gleixner Cc: stable@vger.kernel.org Link: https://lore.kernel.org/all/20250404133116.p-XRWJXf@linutronix.de --- drivers/clocksource/i8253.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/clocksource/i8253.c b/drivers/clocksource/i8253.c index 39f7c2d736d1..b603c25f3dfa 100644 --- a/drivers/clocksource/i8253.c +++ b/drivers/clocksource/i8253.c @@ -103,7 +103,7 @@ int __init clocksource_i8253_init(void) #ifdef CONFIG_CLKEVT_I8253 void clockevent_i8253_disable(void) { - raw_spin_lock(&i8253_lock); + guard(raw_spinlock_irqsave)(&i8253_lock); /* * Writing the MODE register should stop the counter, according to @@ -132,8 +132,6 @@ void clockevent_i8253_disable(void) outb_p(0, PIT_CH0); outb_p(0x30, PIT_MODE); - - raw_spin_unlock(&i8253_lock); } static int pit_shutdown(struct clock_event_device *evt) From 0db8a9a943e9b651952a62e6e985effb4161ac5e Mon Sep 17 00:00:00 2001 From: Konstantin Shkolnyy Date: Fri, 18 Apr 2025 10:20:14 -0500 Subject: [PATCH 153/555] s390/configs: Enable VDPA on Nvidia ConnectX-6 network card ConnectX-6 is the first VDPA-capable NIC. For earlier NICs, Nvidia implements a VDPA emulation in s/w, which hasn't been validated on s390. Add options necessary for VDPA to work. These options are also added because Fedora has them: CONFIG_VDPA_SIM CONFIG_VDPA_SIM_NET CONFIG_VDPA_SIM_BLOCK CONFIG_VDPA_USER CONFIG_VP_VDPA Signed-off-by: Konstantin Shkolnyy Signed-off-by: Heiko Carstens --- arch/s390/configs/debug_defconfig | 8 ++++++++ arch/s390/configs/defconfig | 8 ++++++++ 2 files changed, 16 insertions(+) diff --git a/arch/s390/configs/debug_defconfig b/arch/s390/configs/debug_defconfig index 6f2c9ce1b154..68ba3bedfe00 100644 --- a/arch/s390/configs/debug_defconfig +++ b/arch/s390/configs/debug_defconfig @@ -628,8 +628,16 @@ CONFIG_VIRTIO_PCI=m CONFIG_VIRTIO_BALLOON=m CONFIG_VIRTIO_MEM=m CONFIG_VIRTIO_INPUT=y +CONFIG_VDPA=m +CONFIG_VDPA_SIM=m +CONFIG_VDPA_SIM_NET=m +CONFIG_VDPA_SIM_BLOCK=m +CONFIG_VDPA_USER=m +CONFIG_MLX5_VDPA_NET=m +CONFIG_VP_VDPA=m CONFIG_VHOST_NET=m CONFIG_VHOST_VSOCK=m +CONFIG_VHOST_VDPA=m CONFIG_EXT4_FS=y CONFIG_EXT4_FS_POSIX_ACL=y CONFIG_EXT4_FS_SECURITY=y diff --git a/arch/s390/configs/defconfig b/arch/s390/configs/defconfig index f18a7d97ac21..57b34f9fabf5 100644 --- a/arch/s390/configs/defconfig +++ b/arch/s390/configs/defconfig @@ -618,8 +618,16 @@ CONFIG_VIRTIO_PCI=m CONFIG_VIRTIO_BALLOON=m CONFIG_VIRTIO_MEM=m CONFIG_VIRTIO_INPUT=y +CONFIG_VDPA=m +CONFIG_VDPA_SIM=m +CONFIG_VDPA_SIM_NET=m +CONFIG_VDPA_SIM_BLOCK=m +CONFIG_VDPA_USER=m +CONFIG_MLX5_VDPA_NET=m +CONFIG_VP_VDPA=m CONFIG_VHOST_NET=m CONFIG_VHOST_VSOCK=m +CONFIG_VHOST_VDPA=m CONFIG_EXT4_FS=y CONFIG_EXT4_FS_POSIX_ACL=y CONFIG_EXT4_FS_SECURITY=y From d2b8111c22d7d82f28df574acc54aec44ce3d45c Mon Sep 17 00:00:00 2001 From: Konstantin Shkolnyy Date: Fri, 18 Apr 2025 10:20:15 -0500 Subject: [PATCH 154/555] s390/configs: Enable options required for TC flow offload While testing Open vSwitch with Nvidia ConnectX-6 NIC, it was noticed that it didn't offload TC flows into the NIC, and its log contained many messages such as: "failed to offload flow: No such file or directory: " and, upon enabling more versose logging, additionally: "received NAK error=2 - TC classifier not found" The options enabled here are listed as requirements in Nvidia online documentation, among other options that were already enabled. Now all options listed by Nvidia are enabled.. This option is also added because Fedora has it: CONFIG_NET_EMATCH Signed-off-by: Konstantin Shkolnyy Signed-off-by: Heiko Carstens --- arch/s390/configs/debug_defconfig | 6 ++++++ arch/s390/configs/defconfig | 6 ++++++ 2 files changed, 12 insertions(+) diff --git a/arch/s390/configs/debug_defconfig b/arch/s390/configs/debug_defconfig index 68ba3bedfe00..193d251b4c8f 100644 --- a/arch/s390/configs/debug_defconfig +++ b/arch/s390/configs/debug_defconfig @@ -395,6 +395,9 @@ CONFIG_CLS_U32_MARK=y CONFIG_NET_CLS_FLOW=m CONFIG_NET_CLS_CGROUP=y CONFIG_NET_CLS_BPF=m +CONFIG_NET_CLS_FLOWER=m +CONFIG_NET_CLS_MATCHALL=m +CONFIG_NET_EMATCH=y CONFIG_NET_CLS_ACT=y CONFIG_NET_ACT_POLICE=m CONFIG_NET_ACT_GACT=m @@ -405,6 +408,9 @@ CONFIG_NET_ACT_PEDIT=m CONFIG_NET_ACT_SIMP=m CONFIG_NET_ACT_SKBEDIT=m CONFIG_NET_ACT_CSUM=m +CONFIG_NET_ACT_VLAN=m +CONFIG_NET_ACT_TUNNEL_KEY=m +CONFIG_NET_ACT_CT=m CONFIG_NET_ACT_GATE=m CONFIG_NET_TC_SKB_EXT=y CONFIG_DNS_RESOLVER=y diff --git a/arch/s390/configs/defconfig b/arch/s390/configs/defconfig index 57b34f9fabf5..ec8fdeba4f3e 100644 --- a/arch/s390/configs/defconfig +++ b/arch/s390/configs/defconfig @@ -385,6 +385,9 @@ CONFIG_CLS_U32_MARK=y CONFIG_NET_CLS_FLOW=m CONFIG_NET_CLS_CGROUP=y CONFIG_NET_CLS_BPF=m +CONFIG_NET_CLS_FLOWER=m +CONFIG_NET_CLS_MATCHALL=m +CONFIG_NET_EMATCH=y CONFIG_NET_CLS_ACT=y CONFIG_NET_ACT_POLICE=m CONFIG_NET_ACT_GACT=m @@ -395,6 +398,9 @@ CONFIG_NET_ACT_PEDIT=m CONFIG_NET_ACT_SIMP=m CONFIG_NET_ACT_SKBEDIT=m CONFIG_NET_ACT_CSUM=m +CONFIG_NET_ACT_VLAN=m +CONFIG_NET_ACT_TUNNEL_KEY=m +CONFIG_NET_ACT_CT=m CONFIG_NET_ACT_GATE=m CONFIG_NET_TC_SKB_EXT=y CONFIG_DNS_RESOLVER=y From ae952eea6f4a7e2193f8721a5366049946e012e7 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Thu, 24 Apr 2025 17:07:01 +0200 Subject: [PATCH 155/555] s390/entry: Fix last breaking event handling in case of stack corruption In case of stack corruption stack_invalid() is called and the expectation is that register r10 contains the last breaking event address. This dependency is quite subtle and broke a couple of years ago without that anybody noticed. Fix this by getting rid of the dependency and read the last breaking event address from lowcore. Fixes: 56e62a737028 ("s390: convert to generic entry") Acked-by: Ilya Leoshkevich Reviewed-by: Alexander Gordeev Signed-off-by: Heiko Carstens --- arch/s390/kernel/entry.S | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S index dd291c9ad6a6..9980c17ba22d 100644 --- a/arch/s390/kernel/entry.S +++ b/arch/s390/kernel/entry.S @@ -602,7 +602,8 @@ SYM_CODE_START(stack_invalid) stmg %r0,%r7,__PT_R0(%r11) stmg %r8,%r9,__PT_PSW(%r11) mvc __PT_R8(64,%r11),0(%r14) - stg %r10,__PT_ORIG_GPR2(%r11) # store last break to orig_gpr2 + GET_LC %r2 + mvc __PT_ORIG_GPR2(8,%r11),__LC_PGM_LAST_BREAK(%r2) xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) lgr %r2,%r11 # pass pointer to pt_regs jg kernel_stack_invalid From 833542b3e3d23f640aef6569ba1fd641bc195fa0 Mon Sep 17 00:00:00 2001 From: Gerald Schaefer Date: Tue, 29 Apr 2025 18:21:01 +0200 Subject: [PATCH 156/555] s390/dcssblk: Fix build error with CONFIG_DAX=m and CONFIG_DCSSBLK=y After commit 653d7825c149 ("dcssblk: mark DAX broken, remove FS_DAX_LIMITED support") moved the "select DAX" from config DCSSBLK to the new config DCSSBLK_DAX, randconfig tests could result in build errors like this: s390-linux-ld: drivers/s390/block/dcssblk.o: in function `dcssblk_shared_store': drivers/s390/block/dcssblk.c:417: undefined reference to `kill_dax' s390-linux-ld: drivers/s390/block/dcssblk.c:418: undefined reference to `put_dax' This is because it's now possible to have CONFIG_DCSSBLK=y, but CONFIG_DAX=m. Fix this by adding "depends on DAX || DAX=n" to config DCSSBLK, to make it explicit that we want either no DAX, or the same "y/m" for both config DAX and DCSSBLK, similar to config BLK_DEV_DM. This also requires removing the "select DAX" from config DCSSBLK_DAX, or else there would be a recursive dependency detected. DCSSBLK_DAX is marked as BROKEN at the moment, and won't work well with DAX anyway, so it doesn't really matter if it is selected. Fixes: 653d7825c149 ("dcssblk: mark DAX broken, remove FS_DAX_LIMITED support") Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202504291604.pvjonhWX-lkp@intel.com/ Signed-off-by: Gerald Schaefer Signed-off-by: Heiko Carstens --- drivers/s390/block/Kconfig | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/s390/block/Kconfig b/drivers/s390/block/Kconfig index 4bfe469c04aa..8c1c908d2c6e 100644 --- a/drivers/s390/block/Kconfig +++ b/drivers/s390/block/Kconfig @@ -5,7 +5,7 @@ comment "S/390 block device drivers" config DCSSBLK def_tristate m prompt "DCSSBLK support" - depends on S390 && BLOCK + depends on S390 && BLOCK && (DAX || DAX=n) help Support for dcss block device @@ -14,7 +14,6 @@ config DCSSBLK_DAX depends on DCSSBLK # requires S390 ZONE_DEVICE support depends on BROKEN - select DAX prompt "DCSSBLK DAX support" help Enable DAX operation for the dcss block device From 3a47b1e3cea247857aa48cfe3d0a07e989e6c57d Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Wed, 30 Apr 2025 16:41:01 +0200 Subject: [PATCH 157/555] s390: Update defconfigs Just the regular update of all defconfigs. Signed-off-by: Heiko Carstens --- arch/s390/configs/debug_defconfig | 14 +++++--------- arch/s390/configs/defconfig | 10 +++------- arch/s390/configs/zfcpdump_defconfig | 1 - 3 files changed, 8 insertions(+), 17 deletions(-) diff --git a/arch/s390/configs/debug_defconfig b/arch/s390/configs/debug_defconfig index 193d251b4c8f..24b22f6a9e99 100644 --- a/arch/s390/configs/debug_defconfig +++ b/arch/s390/configs/debug_defconfig @@ -38,7 +38,6 @@ CONFIG_USER_NS=y CONFIG_CHECKPOINT_RESTORE=y CONFIG_SCHED_AUTOGROUP=y CONFIG_EXPERT=y -# CONFIG_SYSFS_SYSCALL is not set CONFIG_PROFILING=y CONFIG_KEXEC=y CONFIG_KEXEC_FILE=y @@ -92,7 +91,6 @@ CONFIG_UNIXWARE_DISKLABEL=y CONFIG_IOSCHED_BFQ=y CONFIG_BINFMT_MISC=m CONFIG_ZSWAP=y -CONFIG_ZSMALLOC=y CONFIG_ZSMALLOC_STAT=y CONFIG_SLAB_BUCKETS=y CONFIG_SLUB_STATS=y @@ -668,7 +666,6 @@ CONFIG_NILFS2_FS=m CONFIG_BCACHEFS_FS=y CONFIG_BCACHEFS_QUOTA=y CONFIG_BCACHEFS_POSIX_ACL=y -CONFIG_FS_DAX=y CONFIG_EXPORTFS_BLOCK_OPS=y CONFIG_FS_ENCRYPTION=y CONFIG_FS_VERITY=y @@ -738,11 +735,10 @@ CONFIG_NLS_UTF8=m CONFIG_DLM=m CONFIG_UNICODE=y CONFIG_PERSISTENT_KEYRINGS=y +CONFIG_BIG_KEYS=y CONFIG_ENCRYPTED_KEYS=m CONFIG_KEY_NOTIFICATIONS=y CONFIG_SECURITY=y -CONFIG_HARDENED_USERCOPY=y -CONFIG_FORTIFY_SOURCE=y CONFIG_SECURITY_SELINUX=y CONFIG_SECURITY_SELINUX_BOOTPARAM=y CONFIG_SECURITY_LOCKDOWN_LSM=y @@ -755,6 +751,8 @@ CONFIG_IMA=y CONFIG_IMA_DEFAULT_HASH_SHA256=y CONFIG_IMA_WRITE_POLICY=y CONFIG_IMA_APPRAISE=y +CONFIG_FORTIFY_SOURCE=y +CONFIG_HARDENED_USERCOPY=y CONFIG_BUG_ON_DATA_CORRUPTION=y CONFIG_CRYPTO_USER=m # CONFIG_CRYPTO_MANAGER_DISABLE_TESTS is not set @@ -770,7 +768,6 @@ CONFIG_CRYPTO_AES_TI=m CONFIG_CRYPTO_ANUBIS=m CONFIG_CRYPTO_ARIA=m CONFIG_CRYPTO_BLOWFISH=m -CONFIG_CRYPTO_CAMELLIA=m CONFIG_CRYPTO_CAST5=m CONFIG_CRYPTO_CAST6=m CONFIG_CRYPTO_DES=m @@ -815,7 +812,6 @@ CONFIG_CRYPTO_SHA3_512_S390=m CONFIG_CRYPTO_GHASH_S390=m CONFIG_CRYPTO_AES_S390=m CONFIG_CRYPTO_DES_S390=m -CONFIG_CRYPTO_CHACHA_S390=m CONFIG_CRYPTO_HMAC_S390=m CONFIG_ZCRYPT=m CONFIG_PKEY=m @@ -826,9 +822,9 @@ CONFIG_PKEY_UV=m CONFIG_CRYPTO_PAES_S390=m CONFIG_CRYPTO_DEV_VIRTIO=m CONFIG_SYSTEM_BLACKLIST_KEYRING=y +CONFIG_CRYPTO_KRB5=m +CONFIG_CRYPTO_KRB5_SELFTESTS=y CONFIG_CORDIC=m -CONFIG_CRYPTO_LIB_CURVE25519=m -CONFIG_CRYPTO_LIB_CHACHA20POLY1305=m CONFIG_RANDOM32_SELFTEST=y CONFIG_XZ_DEC_MICROLZMA=y CONFIG_DMA_CMA=y diff --git a/arch/s390/configs/defconfig b/arch/s390/configs/defconfig index ec8fdeba4f3e..2b8b42d569bc 100644 --- a/arch/s390/configs/defconfig +++ b/arch/s390/configs/defconfig @@ -36,7 +36,6 @@ CONFIG_USER_NS=y CONFIG_CHECKPOINT_RESTORE=y CONFIG_SCHED_AUTOGROUP=y CONFIG_EXPERT=y -# CONFIG_SYSFS_SYSCALL is not set CONFIG_PROFILING=y CONFIG_KEXEC=y CONFIG_KEXEC_FILE=y @@ -86,7 +85,6 @@ CONFIG_UNIXWARE_DISKLABEL=y CONFIG_IOSCHED_BFQ=y CONFIG_BINFMT_MISC=m CONFIG_ZSWAP=y -CONFIG_ZSMALLOC=y CONFIG_ZSMALLOC_STAT=y CONFIG_SLAB_BUCKETS=y # CONFIG_COMPAT_BRK is not set @@ -655,7 +653,6 @@ CONFIG_NILFS2_FS=m CONFIG_BCACHEFS_FS=m CONFIG_BCACHEFS_QUOTA=y CONFIG_BCACHEFS_POSIX_ACL=y -CONFIG_FS_DAX=y CONFIG_EXPORTFS_BLOCK_OPS=y CONFIG_FS_ENCRYPTION=y CONFIG_FS_VERITY=y @@ -725,6 +722,7 @@ CONFIG_NLS_UTF8=m CONFIG_DLM=m CONFIG_UNICODE=y CONFIG_PERSISTENT_KEYRINGS=y +CONFIG_BIG_KEYS=y CONFIG_ENCRYPTED_KEYS=m CONFIG_KEY_NOTIFICATIONS=y CONFIG_SECURITY=y @@ -756,7 +754,6 @@ CONFIG_CRYPTO_AES_TI=m CONFIG_CRYPTO_ANUBIS=m CONFIG_CRYPTO_ARIA=m CONFIG_CRYPTO_BLOWFISH=m -CONFIG_CRYPTO_CAMELLIA=m CONFIG_CRYPTO_CAST5=m CONFIG_CRYPTO_CAST6=m CONFIG_CRYPTO_DES=m @@ -802,7 +799,6 @@ CONFIG_CRYPTO_SHA3_512_S390=m CONFIG_CRYPTO_GHASH_S390=m CONFIG_CRYPTO_AES_S390=m CONFIG_CRYPTO_DES_S390=m -CONFIG_CRYPTO_CHACHA_S390=m CONFIG_CRYPTO_HMAC_S390=m CONFIG_ZCRYPT=m CONFIG_PKEY=m @@ -813,10 +809,10 @@ CONFIG_PKEY_UV=m CONFIG_CRYPTO_PAES_S390=m CONFIG_CRYPTO_DEV_VIRTIO=m CONFIG_SYSTEM_BLACKLIST_KEYRING=y +CONFIG_CRYPTO_KRB5=m +CONFIG_CRYPTO_KRB5_SELFTESTS=y CONFIG_CORDIC=m CONFIG_PRIME_NUMBERS=m -CONFIG_CRYPTO_LIB_CURVE25519=m -CONFIG_CRYPTO_LIB_CHACHA20POLY1305=m CONFIG_XZ_DEC_MICROLZMA=y CONFIG_DMA_CMA=y CONFIG_CMA_SIZE_MBYTES=0 diff --git a/arch/s390/configs/zfcpdump_defconfig b/arch/s390/configs/zfcpdump_defconfig index 853b2326a171..8163c1702720 100644 --- a/arch/s390/configs/zfcpdump_defconfig +++ b/arch/s390/configs/zfcpdump_defconfig @@ -70,7 +70,6 @@ CONFIG_DEBUG_KERNEL=y CONFIG_DEBUG_INFO_DWARF4=y CONFIG_DEBUG_FS=y CONFIG_PANIC_ON_OOPS=y -# CONFIG_SCHED_DEBUG is not set CONFIG_RCU_CPU_STALL_TIMEOUT=60 # CONFIG_RCU_TRACE is not set # CONFIG_FTRACE is not set From 6328bdc988d23201c700e1e7e04eb05a1149ac1e Mon Sep 17 00:00:00 2001 From: Michal Pecio Date: Mon, 5 May 2025 15:56:29 +0300 Subject: [PATCH 158/555] usb: xhci: Don't trust the EP Context cycle bit when moving HW dequeue VIA VL805 doesn't bother updating the EP Context cycle bit when the endpoint halts. This is seen by patching xhci_move_dequeue_past_td() to print the cycle bits of the EP Context and the TRB at hw_dequeue and then disconnecting a flash drive while reading it. Actual cycle state is random as expected, but the EP Context bit is always 1. This means that the cycle state produced by this function is wrong half the time, and then the endpoint stops working. Work around it by looking at the cycle bit of TD's end_trb instead of believing the Endpoint or Stream Context. Specifically: - rename cycle_found to hw_dequeue_found to avoid confusion - initialize new_cycle from td->end_trb instead of hw_dequeue - switch new_cycle toggling to happen after end_trb is found Now a workload which regularly stalls the device works normally for a few hours and clearly demonstrates the HW bug - the EP Context bit is not updated in a new cycle until Set TR Dequeue overwrites it: [ +0,000298] sd 10:0:0:0: [sdc] Attached SCSI disk [ +0,011758] cycle bits: TRB 1 EP Ctx 1 [ +5,947138] cycle bits: TRB 1 EP Ctx 1 [ +0,065731] cycle bits: TRB 0 EP Ctx 1 [ +0,064022] cycle bits: TRB 0 EP Ctx 0 [ +0,063297] cycle bits: TRB 0 EP Ctx 0 [ +0,069823] cycle bits: TRB 0 EP Ctx 0 [ +0,063390] cycle bits: TRB 1 EP Ctx 0 [ +0,063064] cycle bits: TRB 1 EP Ctx 1 [ +0,062293] cycle bits: TRB 1 EP Ctx 1 [ +0,066087] cycle bits: TRB 0 EP Ctx 1 [ +0,063636] cycle bits: TRB 0 EP Ctx 0 [ +0,066360] cycle bits: TRB 0 EP Ctx 0 Also tested on the buggy ASM1042 which moves EP Context dequeue to the next TRB after errors, one problem case addressed by the rework that implemented this loop. In this case hw_dequeue can be enqueue, so simply picking the cycle bit of TRB at hw_dequeue wouldn't work. Commit 5255660b208a ("xhci: add quirk for host controllers that don't update endpoint DCS") tried to solve the stale cycle problem, but it was more complex and got reverted due to a reported issue. Cc: Jonathan Bell Cc: Oliver Neukum Signed-off-by: Michal Pecio Signed-off-by: Mathias Nyman Link: https://lore.kernel.org/r/20250505125630.561699-2-mathias.nyman@linux.intel.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/host/xhci-ring.c | 19 +++++++++---------- 1 file changed, 9 insertions(+), 10 deletions(-) diff --git a/drivers/usb/host/xhci-ring.c b/drivers/usb/host/xhci-ring.c index b906bc2eea5f..423bf3649570 100644 --- a/drivers/usb/host/xhci-ring.c +++ b/drivers/usb/host/xhci-ring.c @@ -699,7 +699,7 @@ static int xhci_move_dequeue_past_td(struct xhci_hcd *xhci, int new_cycle; dma_addr_t addr; u64 hw_dequeue; - bool cycle_found = false; + bool hw_dequeue_found = false; bool td_last_trb_found = false; u32 trb_sct = 0; int ret; @@ -715,25 +715,24 @@ static int xhci_move_dequeue_past_td(struct xhci_hcd *xhci, hw_dequeue = xhci_get_hw_deq(xhci, dev, ep_index, stream_id); new_seg = ep_ring->deq_seg; new_deq = ep_ring->dequeue; - new_cycle = hw_dequeue & 0x1; + new_cycle = le32_to_cpu(td->end_trb->generic.field[3]) & TRB_CYCLE; /* - * We want to find the pointer, segment and cycle state of the new trb - * (the one after current TD's end_trb). We know the cycle state at - * hw_dequeue, so walk the ring until both hw_dequeue and end_trb are - * found. + * Walk the ring until both the next TRB and hw_dequeue are found (don't + * move hw_dequeue back if it went forward due to a HW bug). Cycle state + * is loaded from a known good TRB, track later toggles to maintain it. */ do { - if (!cycle_found && xhci_trb_virt_to_dma(new_seg, new_deq) + if (!hw_dequeue_found && xhci_trb_virt_to_dma(new_seg, new_deq) == (dma_addr_t)(hw_dequeue & ~0xf)) { - cycle_found = true; + hw_dequeue_found = true; if (td_last_trb_found) break; } if (new_deq == td->end_trb) td_last_trb_found = true; - if (cycle_found && trb_is_link(new_deq) && + if (td_last_trb_found && trb_is_link(new_deq) && link_trb_toggles_cycle(new_deq)) new_cycle ^= 0x1; @@ -745,7 +744,7 @@ static int xhci_move_dequeue_past_td(struct xhci_hcd *xhci, return -EINVAL; } - } while (!cycle_found || !td_last_trb_found); + } while (!hw_dequeue_found || !td_last_trb_found); /* Don't update the ring cycle state for the producer (us). */ addr = xhci_trb_virt_to_dma(new_seg, new_deq); From cab63934c33b12c0d1e9f4da7450928057f2c142 Mon Sep 17 00:00:00 2001 From: Mathias Nyman Date: Mon, 5 May 2025 15:56:30 +0300 Subject: [PATCH 159/555] xhci: dbc: Avoid event polling busyloop if pending rx transfers are inactive. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Event polling delay is set to 0 if there are any pending requests in either rx or tx requests lists. Checking for pending requests does not work well for "IN" transfers as the tty driver always queues requests to the list and TRBs to the ring, preparing to receive data from the host. This causes unnecessary busylooping and cpu hogging. Only set the event polling delay to 0 if there are pending tx "write" transfers, or if it was less than 10ms since last active data transfer in any direction. Cc: Łukasz Bartosik Fixes: fb18e5bb9660 ("xhci: dbc: poll at different rate depending on data transfer activity") Cc: stable@vger.kernel.org Signed-off-by: Mathias Nyman Link: https://lore.kernel.org/r/20250505125630.561699-3-mathias.nyman@linux.intel.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/host/xhci-dbgcap.c | 19 ++++++++++++++++--- drivers/usb/host/xhci-dbgcap.h | 3 +++ 2 files changed, 19 insertions(+), 3 deletions(-) diff --git a/drivers/usb/host/xhci-dbgcap.c b/drivers/usb/host/xhci-dbgcap.c index fd7895b24367..0d4ce5734165 100644 --- a/drivers/usb/host/xhci-dbgcap.c +++ b/drivers/usb/host/xhci-dbgcap.c @@ -823,6 +823,7 @@ static enum evtreturn xhci_dbc_do_handle_events(struct xhci_dbc *dbc) { dma_addr_t deq; union xhci_trb *evt; + enum evtreturn ret = EVT_DONE; u32 ctrl, portsc; bool update_erdp = false; @@ -909,6 +910,7 @@ static enum evtreturn xhci_dbc_do_handle_events(struct xhci_dbc *dbc) break; case TRB_TYPE(TRB_TRANSFER): dbc_handle_xfer_event(dbc, evt); + ret = EVT_XFER_DONE; break; default: break; @@ -927,7 +929,7 @@ static enum evtreturn xhci_dbc_do_handle_events(struct xhci_dbc *dbc) lo_hi_writeq(deq, &dbc->regs->erdp); } - return EVT_DONE; + return ret; } static void xhci_dbc_handle_events(struct work_struct *work) @@ -936,6 +938,7 @@ static void xhci_dbc_handle_events(struct work_struct *work) struct xhci_dbc *dbc; unsigned long flags; unsigned int poll_interval; + unsigned long busypoll_timelimit; dbc = container_of(to_delayed_work(work), struct xhci_dbc, event_work); poll_interval = dbc->poll_interval; @@ -954,11 +957,21 @@ static void xhci_dbc_handle_events(struct work_struct *work) dbc->driver->disconnect(dbc); break; case EVT_DONE: - /* set fast poll rate if there are pending data transfers */ + /* + * Set fast poll rate if there are pending out transfers, or + * a transfer was recently processed + */ + busypoll_timelimit = dbc->xfer_timestamp + + msecs_to_jiffies(DBC_XFER_INACTIVITY_TIMEOUT); + if (!list_empty(&dbc->eps[BULK_OUT].list_pending) || - !list_empty(&dbc->eps[BULK_IN].list_pending)) + time_is_after_jiffies(busypoll_timelimit)) poll_interval = 0; break; + case EVT_XFER_DONE: + dbc->xfer_timestamp = jiffies; + poll_interval = 0; + break; default: dev_info(dbc->dev, "stop handling dbc events\n"); return; diff --git a/drivers/usb/host/xhci-dbgcap.h b/drivers/usb/host/xhci-dbgcap.h index 9dc8f4d8077c..47ac72c2286d 100644 --- a/drivers/usb/host/xhci-dbgcap.h +++ b/drivers/usb/host/xhci-dbgcap.h @@ -96,6 +96,7 @@ struct dbc_ep { #define DBC_WRITE_BUF_SIZE 8192 #define DBC_POLL_INTERVAL_DEFAULT 64 /* milliseconds */ #define DBC_POLL_INTERVAL_MAX 5000 /* milliseconds */ +#define DBC_XFER_INACTIVITY_TIMEOUT 10 /* milliseconds */ /* * Private structure for DbC hardware state: */ @@ -142,6 +143,7 @@ struct xhci_dbc { enum dbc_state state; struct delayed_work event_work; unsigned int poll_interval; /* ms */ + unsigned long xfer_timestamp; unsigned resume_required:1; struct dbc_ep eps[2]; @@ -187,6 +189,7 @@ struct dbc_request { enum evtreturn { EVT_ERR = -1, EVT_DONE, + EVT_XFER_DONE, EVT_GSER, EVT_DISC, }; From 844f766e02d0aba973d78f3ade38ad8bae399347 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 4 May 2025 15:01:34 -0400 Subject: [PATCH 160/555] bcachefs: Improve want_cached_ptr() If promote target isn't set, rebalance should still leave a cached copy on the faster device. Fall back to foreground_target if it's set, or allow a cached copy on any device if neither are set. Signed-off-by: Kent Overstreet --- fs/bcachefs/extents.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/fs/bcachefs/extents.c b/fs/bcachefs/extents.c index dca2b8425cc0..e597fb9c9823 100644 --- a/fs/bcachefs/extents.c +++ b/fs/bcachefs/extents.c @@ -1056,8 +1056,9 @@ bch2_extent_has_ptr(struct bkey_s_c k1, struct extent_ptr_decoded p1, struct bke static bool want_cached_ptr(struct bch_fs *c, struct bch_io_opts *opts, struct bch_extent_ptr *ptr) { - if (!opts->promote_target || - !bch2_dev_in_target(c, ptr->dev, opts->promote_target)) + unsigned target = opts->promote_target ?: opts->foreground_target; + + if (target && !bch2_dev_in_target(c, ptr->dev, target)) return false; struct bch_dev *ca = bch2_dev_rcu_noerror(c, ptr->dev); From 50a7b899a0d806f961edadfc3357cb6679826795 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 4 May 2025 16:31:40 -0400 Subject: [PATCH 161/555] bcachefs: Ensure proper write alignment There was a buggy version of bcachefs-tools which picked misaligned bucket sizes when formatting, and we're also about to do dynamic block sizes - which will allow picking logical block size or physical block size of the device per-write, allowing for better compression ratios at the cost of slightly worse write performance (i.e. forcing the device to do RMW or extra buffering). To account for this, tweak bch2_alloc_sectors_start() to properly align open_buckets to the blocksize of the write we're about to do. Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_foreground.c | 22 +++++++++++++++++++++- 1 file changed, 21 insertions(+), 1 deletion(-) diff --git a/fs/bcachefs/alloc_foreground.c b/fs/bcachefs/alloc_foreground.c index effafc3e0ced..7ec022e9361a 100644 --- a/fs/bcachefs/alloc_foreground.c +++ b/fs/bcachefs/alloc_foreground.c @@ -1422,11 +1422,31 @@ alloc_done: wp->sectors_free = UINT_MAX; - open_bucket_for_each(c, &wp->ptrs, ob, i) + open_bucket_for_each(c, &wp->ptrs, ob, i) { + /* + * Ensure proper write alignment - either due to misaligned + * bucket sizes (from buggy bcachefs-tools), or writes that mix + * logical/physical alignment: + */ + struct bch_dev *ca = ob_dev(c, ob); + u64 offset = bucket_to_sector(ca, ob->bucket) + + ca->mi.bucket_size - + ob->sectors_free; + unsigned align = round_up(offset, block_sectors(c)) - offset; + + ob->sectors_free = max_t(int, 0, ob->sectors_free - align); + wp->sectors_free = min(wp->sectors_free, ob->sectors_free); + } wp->sectors_free = rounddown(wp->sectors_free, block_sectors(c)); + /* Did alignment use up space in an open_bucket? */ + if (unlikely(!wp->sectors_free)) { + bch2_alloc_sectors_done(c, wp); + goto retry; + } + BUG_ON(!wp->sectors_free || wp->sectors_free == UINT_MAX); return 0; From 7a69fa65718a7258aa89fca06b975f4357daeac3 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 5 May 2025 14:13:21 -0400 Subject: [PATCH 162/555] bcachefs: Add missing barriers before wake_up_bit() wake_up() doesn't require a barrier - but wake_up_bit() does. This only affected non x86, and primarily lead to lost wakeups after btree node reads. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_io.c | 9 ++++++++- fs/bcachefs/buckets.h | 1 + fs/bcachefs/ec.h | 1 + 3 files changed, 10 insertions(+), 1 deletion(-) diff --git a/fs/bcachefs/btree_io.c b/fs/bcachefs/btree_io.c index 5fd4a58d2ad2..60782f3e5aec 100644 --- a/fs/bcachefs/btree_io.c +++ b/fs/bcachefs/btree_io.c @@ -41,6 +41,7 @@ void bch2_btree_node_io_unlock(struct btree *b) clear_btree_node_write_in_flight_inner(b); clear_btree_node_write_in_flight(b); + smp_mb__after_atomic(); wake_up_bit(&b->flags, BTREE_NODE_write_in_flight); } @@ -1400,6 +1401,7 @@ start: printbuf_exit(&buf); clear_btree_node_read_in_flight(b); + smp_mb__after_atomic(); wake_up_bit(&b->flags, BTREE_NODE_read_in_flight); } @@ -1595,6 +1597,7 @@ fsck_err: printbuf_exit(&buf); clear_btree_node_read_in_flight(b); + smp_mb__after_atomic(); wake_up_bit(&b->flags, BTREE_NODE_read_in_flight); } @@ -1721,6 +1724,7 @@ void bch2_btree_node_read(struct btree_trans *trans, struct btree *b, set_btree_node_read_error(b); bch2_btree_lost_data(c, b->c.btree_id); clear_btree_node_read_in_flight(b); + smp_mb__after_atomic(); wake_up_bit(&b->flags, BTREE_NODE_read_in_flight); printbuf_exit(&buf); return; @@ -2061,8 +2065,10 @@ static void __btree_node_write_done(struct bch_fs *c, struct btree *b, u64 start if (new & (1U << BTREE_NODE_write_in_flight)) __bch2_btree_node_write(c, b, BTREE_WRITE_ALREADY_STARTED|type); - else + else { + smp_mb__after_atomic(); wake_up_bit(&b->flags, BTREE_NODE_write_in_flight); + } } static void btree_node_write_done(struct bch_fs *c, struct btree *b, u64 start_time) @@ -2175,6 +2181,7 @@ static void btree_node_write_endio(struct bio *bio) } clear_btree_node_write_in_flight_inner(b); + smp_mb__after_atomic(); wake_up_bit(&b->flags, BTREE_NODE_write_in_flight_inner); INIT_WORK(&wb->work, btree_node_write_work); queue_work(c->btree_io_complete_wq, &wb->work); diff --git a/fs/bcachefs/buckets.h b/fs/bcachefs/buckets.h index 8d75b27a1418..af1532de4a37 100644 --- a/fs/bcachefs/buckets.h +++ b/fs/bcachefs/buckets.h @@ -44,6 +44,7 @@ static inline void bucket_unlock(struct bucket *b) BUILD_BUG_ON(!((union ulong_byte_assert) { .ulong = 1UL << BUCKET_LOCK_BITNR }).byte); clear_bit_unlock(BUCKET_LOCK_BITNR, (void *) &b->lock); + smp_mb__after_atomic(); wake_up_bit((void *) &b->lock, BUCKET_LOCK_BITNR); } diff --git a/fs/bcachefs/ec.h b/fs/bcachefs/ec.h index 62d27e04d763..51893e1ee874 100644 --- a/fs/bcachefs/ec.h +++ b/fs/bcachefs/ec.h @@ -160,6 +160,7 @@ static inline void gc_stripe_unlock(struct gc_stripe *s) BUILD_BUG_ON(!((union ulong_byte_assert) { .ulong = 1UL << BUCKET_LOCK_BITNR }).byte); clear_bit_unlock(BUCKET_LOCK_BITNR, (void *) &s->lock); + smp_mb__after_atomic(); wake_up_bit((void *) &s->lock, BUCKET_LOCK_BITNR); } From aed4ccbf4595e08f3fa80c7faaf1d2188d61bc70 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sun, 4 May 2025 18:46:16 -0400 Subject: [PATCH 163/555] bcachefs: fix hung task timeout in journal read Signed-off-by: Kent Overstreet --- fs/bcachefs/journal_io.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/fs/bcachefs/journal_io.c b/fs/bcachefs/journal_io.c index 63cdf885c9e2..ded18a94ed02 100644 --- a/fs/bcachefs/journal_io.c +++ b/fs/bcachefs/journal_io.c @@ -19,6 +19,7 @@ #include #include +#include void bch2_journal_pos_from_member_info_set(struct bch_fs *c) { @@ -1262,7 +1263,8 @@ int bch2_journal_read(struct bch_fs *c, degraded = true; } - closure_sync(&jlist.cl); + while (closure_sync_timeout(&jlist.cl, sysctl_hung_task_timeout_secs * HZ / 2)) + ; if (jlist.ret) return jlist.ret; From 157dbc4a321f5bb6f8b6c724d12ba720a90f1a7c Mon Sep 17 00:00:00 2001 From: Sebastian Ott Date: Mon, 5 May 2025 19:31:48 +0200 Subject: [PATCH 164/555] KVM: arm64: Fix uninitialized memcache pointer in user_mem_abort() Commit fce886a60207 ("KVM: arm64: Plumb the pKVM MMU in KVM") made the initialization of the local memcache variable in user_mem_abort() conditional, leaving a codepath where it is used uninitialized via kvm_pgtable_stage2_map(). This can fail on any path that requires a stage-2 allocation without transition via a permission fault or dirty logging. Fix this by making sure that memcache is always valid. Fixes: fce886a60207 ("KVM: arm64: Plumb the pKVM MMU in KVM") Signed-off-by: Sebastian Ott Reviewed-by: Marc Zyngier Cc: stable@vger.kernel.org Link: https://lore.kernel.org/kvmarm/3f5db4c7-ccce-fb95-595c-692fa7aad227@redhat.com/ Link: https://lore.kernel.org/r/20250505173148.33900-1-sebott@redhat.com Signed-off-by: Oliver Upton --- arch/arm64/kvm/mmu.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/arch/arm64/kvm/mmu.c b/arch/arm64/kvm/mmu.c index 754f2fe0cc67..eeda92330ade 100644 --- a/arch/arm64/kvm/mmu.c +++ b/arch/arm64/kvm/mmu.c @@ -1501,6 +1501,11 @@ static int user_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa, return -EFAULT; } + if (!is_protected_kvm_enabled()) + memcache = &vcpu->arch.mmu_page_cache; + else + memcache = &vcpu->arch.pkvm_memcache; + /* * Permission faults just need to update the existing leaf entry, * and so normally don't require allocations from the memcache. The @@ -1510,13 +1515,11 @@ static int user_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa, if (!fault_is_perm || (logging_active && write_fault)) { int min_pages = kvm_mmu_cache_min_pages(vcpu->arch.hw_mmu); - if (!is_protected_kvm_enabled()) { - memcache = &vcpu->arch.mmu_page_cache; + if (!is_protected_kvm_enabled()) ret = kvm_mmu_topup_memory_cache(memcache, min_pages); - } else { - memcache = &vcpu->arch.pkvm_memcache; + else ret = topup_hyp_memcache(memcache, min_pages); - } + if (ret) return ret; } From 859c60276e12a3a18c65a3f9325e656a068c9b62 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Tue, 29 Apr 2025 12:43:26 +0100 Subject: [PATCH 165/555] KVM: arm64: Force HCR_EL2.xMO to 1 at all times in VHE mode We keep setting and clearing these bits depending on the role of the host kernel, mimicking what we do for nVHE. But that's actually pretty pointless, as we always want physical interrupts to make it to the host, at EL2. This has also two problems: - it prevents IRQs from being taken when these bits are cleared if the implementation has chosen to implement these bits as masks when HCR_EL2.{TGE,xMO}=={0,0} - it triggers a bad erratum on the AmpereOne HW, which catches fire on clearing these bits while an interrupt is being taken (AC03_CPU_36). Let's kill these two birds with a single stone, and permanently set the xMO bits when running VHE. This involves a bit of surgery on code paths that rely on flipping these bits on and off for other purposes. Note that the earliest setting of hcr_el2 (in the init_hcr_el2 macro) is left untouched as is runs extremely early, with interrupts disabled, and soon enough overwritten with the final value containing the xMO bits. Reported-by: D Scott Phillips Signed-off-by: Marc Zyngier Link: https://lore.kernel.org/r/20250429114326.3618875-1-maz@kernel.org Signed-off-by: Oliver Upton --- arch/arm64/include/asm/kvm_arm.h | 2 +- arch/arm64/kvm/hyp/vgic-v3-sr.c | 36 +++++++++++++++++++------------- 2 files changed, 22 insertions(+), 16 deletions(-) diff --git a/arch/arm64/include/asm/kvm_arm.h b/arch/arm64/include/asm/kvm_arm.h index 974d72b5905b..bba4b0e93091 100644 --- a/arch/arm64/include/asm/kvm_arm.h +++ b/arch/arm64/include/asm/kvm_arm.h @@ -100,7 +100,7 @@ HCR_FMO | HCR_IMO | HCR_PTW | HCR_TID3 | HCR_TID1) #define HCR_HOST_NVHE_FLAGS (HCR_RW | HCR_API | HCR_APK | HCR_ATA) #define HCR_HOST_NVHE_PROTECTED_FLAGS (HCR_HOST_NVHE_FLAGS | HCR_TSC) -#define HCR_HOST_VHE_FLAGS (HCR_RW | HCR_TGE | HCR_E2H) +#define HCR_HOST_VHE_FLAGS (HCR_RW | HCR_TGE | HCR_E2H | HCR_AMO | HCR_IMO | HCR_FMO) #define HCRX_HOST_FLAGS (HCRX_EL2_MSCEn | HCRX_EL2_TCR2En | HCRX_EL2_EnFPM) #define MPAMHCR_HOST_FLAGS 0 diff --git a/arch/arm64/kvm/hyp/vgic-v3-sr.c b/arch/arm64/kvm/hyp/vgic-v3-sr.c index ed363aa3027e..50aa8dbcae75 100644 --- a/arch/arm64/kvm/hyp/vgic-v3-sr.c +++ b/arch/arm64/kvm/hyp/vgic-v3-sr.c @@ -429,23 +429,27 @@ u64 __vgic_v3_get_gic_config(void) /* * To check whether we have a MMIO-based (GICv2 compatible) * CPU interface, we need to disable the system register - * view. To do that safely, we have to prevent any interrupt - * from firing (which would be deadly). + * view. * - * Note that this only makes sense on VHE, as interrupts are - * already masked for nVHE as part of the exception entry to - * EL2. - */ - if (has_vhe()) - flags = local_daif_save(); - - /* * Table 11-2 "Permitted ICC_SRE_ELx.SRE settings" indicates * that to be able to set ICC_SRE_EL1.SRE to 0, all the * interrupt overrides must be set. You've got to love this. + * + * As we always run VHE with HCR_xMO set, no extra xMO + * manipulation is required in that case. + * + * To safely disable SRE, we have to prevent any interrupt + * from firing (which would be deadly). This only makes sense + * on VHE, as interrupts are already masked for nVHE as part + * of the exception entry to EL2. */ - sysreg_clear_set(hcr_el2, 0, HCR_AMO | HCR_FMO | HCR_IMO); - isb(); + if (has_vhe()) { + flags = local_daif_save(); + } else { + sysreg_clear_set(hcr_el2, 0, HCR_AMO | HCR_FMO | HCR_IMO); + isb(); + } + write_gicreg(0, ICC_SRE_EL1); isb(); @@ -453,11 +457,13 @@ u64 __vgic_v3_get_gic_config(void) write_gicreg(sre, ICC_SRE_EL1); isb(); - sysreg_clear_set(hcr_el2, HCR_AMO | HCR_FMO | HCR_IMO, 0); - isb(); - if (has_vhe()) + if (has_vhe()) { local_daif_restore(flags); + } else { + sysreg_clear_set(hcr_el2, HCR_AMO | HCR_FMO | HCR_IMO, 0); + isb(); + } val = (val & ICC_SRE_EL1_SRE) ? 0 : (1ULL << 63); val |= read_gicreg(ICH_VTR_EL2); From 7af7cfbe78e23af0ac3cbe9b2b16da69f1412222 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Tue, 29 Apr 2025 12:41:16 +0100 Subject: [PATCH 166/555] KVM: arm64: Prevent userspace from disabling AArch64 support at any virtualisable EL A sorry excuse for a selftest is trying to disable AArch64 support. And yes, this goes as well as you can imagine. Let's forbid this sort of things. Normal userspace shouldn't get caught doing that. Signed-off-by: Marc Zyngier Reviewed-by: Ganapatrao Kulkarni Link: https://lore.kernel.org/r/20250429114117.3618800-2-maz@kernel.org Signed-off-by: Oliver Upton --- arch/arm64/kvm/sys_regs.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c index 005ad28f7306..5dde9285afc8 100644 --- a/arch/arm64/kvm/sys_regs.c +++ b/arch/arm64/kvm/sys_regs.c @@ -1945,6 +1945,12 @@ static int set_id_aa64pfr0_el1(struct kvm_vcpu *vcpu, if ((hw_val & mpam_mask) == (user_val & mpam_mask)) user_val &= ~ID_AA64PFR0_EL1_MPAM_MASK; + /* Fail the guest's request to disable the AA64 ISA at EL{0,1,2} */ + if (!FIELD_GET(ID_AA64PFR0_EL1_EL0, user_val) || + !FIELD_GET(ID_AA64PFR0_EL1_EL1, user_val) || + (vcpu_has_nv(vcpu) && !FIELD_GET(ID_AA64PFR0_EL1_EL2, user_val))) + return -EINVAL; + return set_id_reg(vcpu, rd, user_val); } From b60e285b6acdec6d24fbde36d59f6d806f91cdc6 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Tue, 29 Apr 2025 12:41:17 +0100 Subject: [PATCH 167/555] KVM: arm64: selftest: Don't try to disable AArch64 support Trying to cut the branch you are sat on is pretty dumb. And so is trying to disable the instruction set you are executing on. Signed-off-by: Marc Zyngier Reviewed-by: Ganapatrao Kulkarni Link: https://lore.kernel.org/r/20250429114117.3618800-3-maz@kernel.org Signed-off-by: Oliver Upton --- tools/testing/selftests/kvm/arm64/set_id_regs.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/tools/testing/selftests/kvm/arm64/set_id_regs.c b/tools/testing/selftests/kvm/arm64/set_id_regs.c index 322b9d3b0125..57708de2075d 100644 --- a/tools/testing/selftests/kvm/arm64/set_id_regs.c +++ b/tools/testing/selftests/kvm/arm64/set_id_regs.c @@ -129,10 +129,10 @@ static const struct reg_ftr_bits ftr_id_aa64pfr0_el1[] = { REG_FTR_BITS(FTR_LOWER_SAFE, ID_AA64PFR0_EL1, DIT, 0), REG_FTR_BITS(FTR_LOWER_SAFE, ID_AA64PFR0_EL1, SEL2, 0), REG_FTR_BITS(FTR_EXACT, ID_AA64PFR0_EL1, GIC, 0), - REG_FTR_BITS(FTR_LOWER_SAFE, ID_AA64PFR0_EL1, EL3, 0), - REG_FTR_BITS(FTR_LOWER_SAFE, ID_AA64PFR0_EL1, EL2, 0), - REG_FTR_BITS(FTR_LOWER_SAFE, ID_AA64PFR0_EL1, EL1, 0), - REG_FTR_BITS(FTR_LOWER_SAFE, ID_AA64PFR0_EL1, EL0, 0), + REG_FTR_BITS(FTR_LOWER_SAFE, ID_AA64PFR0_EL1, EL3, 1), + REG_FTR_BITS(FTR_LOWER_SAFE, ID_AA64PFR0_EL1, EL2, 1), + REG_FTR_BITS(FTR_LOWER_SAFE, ID_AA64PFR0_EL1, EL1, 1), + REG_FTR_BITS(FTR_LOWER_SAFE, ID_AA64PFR0_EL1, EL0, 1), REG_FTR_END, }; From 9c618560994794d6f477e81f3b695fe799feec8a Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 5 May 2025 15:52:57 -0400 Subject: [PATCH 168/555] bcachefs: Call bch2_fs_start before getting vfs superblock This reverts 1fdbe0b184c8 bcachefs: Make sure c->vfs_sb is set before starting fs switched up bch2_fs_get_tree() so that we got a superblock before calling bch2_fs_start, so that c->vfs_sb would always be initialized while the filesystem was active. This turned out not to be necessary, because blk_holder_ops were implemented using our own locking, not vfs locking. And this had the side effect of creating a super_block and doing our full recovery (including potentially fsck) before setting SB_BORN, which causes things like sync calls to hang until our recovery is finished. Signed-off-by: Kent Overstreet --- fs/bcachefs/fs.c | 11 +++-------- 1 file changed, 3 insertions(+), 8 deletions(-) diff --git a/fs/bcachefs/fs.c b/fs/bcachefs/fs.c index 113db85b6ef9..b6801861c66f 100644 --- a/fs/bcachefs/fs.c +++ b/fs/bcachefs/fs.c @@ -2502,10 +2502,9 @@ static int bch2_fs_get_tree(struct fs_context *fc) bch2_opts_apply(&c->opts, opts); - /* - * need to initialise sb and set c->vfs_sb _before_ starting fs, - * for blk_holder_ops - */ + ret = bch2_fs_start(c); + if (ret) + goto err_stop_fs; sb = sget(fc->fs_type, NULL, bch2_set_super, fc->sb_flags|SB_NOSEC, c); ret = PTR_ERR_OR_ZERO(sb); @@ -2567,10 +2566,6 @@ got_sb: sb->s_shrink->seeks = 0; - ret = bch2_fs_start(c); - if (ret) - goto err_put_super; - #ifdef CONFIG_UNICODE sb->s_encoding = c->cf_encoding; #endif From 3769478610135e82b262640252d90f6efb05be71 Mon Sep 17 00:00:00 2001 From: Cong Wang Date: Mon, 28 Apr 2025 16:29:54 -0700 Subject: [PATCH 169/555] sch_htb: make htb_deactivate() idempotent Alan reported a NULL pointer dereference in htb_next_rb_node() after we made htb_qlen_notify() idempotent. It turns out in the following case it introduced some regression: htb_dequeue_tree(): |-> fq_codel_dequeue() |-> qdisc_tree_reduce_backlog() |-> htb_qlen_notify() |-> htb_deactivate() |-> htb_next_rb_node() |-> htb_deactivate() For htb_next_rb_node(), after calling the 1st htb_deactivate(), the clprio[prio]->ptr could be already set to NULL, which means htb_next_rb_node() is vulnerable here. For htb_deactivate(), although we checked qlen before calling it, in case of qlen==0 after qdisc_tree_reduce_backlog(), we may call it again which triggers the warning inside. To fix the issues here, we need to: 1) Make htb_deactivate() idempotent, that is, simply return if we already call it before. 2) Make htb_next_rb_node() safe against ptr==NULL. Many thanks to Alan for testing and for the reproducer. Fixes: 5ba8b837b522 ("sch_htb: make htb_qlen_notify() idempotent") Reported-by: Alan J. Wylie Signed-off-by: Cong Wang Link: https://patch.msgid.link/20250428232955.1740419-2-xiyou.wangcong@gmail.com Signed-off-by: Jakub Kicinski --- net/sched/sch_htb.c | 15 ++++++--------- 1 file changed, 6 insertions(+), 9 deletions(-) diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c index 4b9a639b642e..14bf71f57057 100644 --- a/net/sched/sch_htb.c +++ b/net/sched/sch_htb.c @@ -348,7 +348,8 @@ static void htb_add_to_wait_tree(struct htb_sched *q, */ static inline void htb_next_rb_node(struct rb_node **n) { - *n = rb_next(*n); + if (*n) + *n = rb_next(*n); } /** @@ -609,8 +610,8 @@ static inline void htb_activate(struct htb_sched *q, struct htb_class *cl) */ static inline void htb_deactivate(struct htb_sched *q, struct htb_class *cl) { - WARN_ON(!cl->prio_activity); - + if (!cl->prio_activity) + return; htb_deactivate_prios(q, cl); cl->prio_activity = 0; } @@ -1485,8 +1486,6 @@ static void htb_qlen_notify(struct Qdisc *sch, unsigned long arg) { struct htb_class *cl = (struct htb_class *)arg; - if (!cl->prio_activity) - return; htb_deactivate(qdisc_priv(sch), cl); } @@ -1740,8 +1739,7 @@ static int htb_delete(struct Qdisc *sch, unsigned long arg, if (cl->parent) cl->parent->children--; - if (cl->prio_activity) - htb_deactivate(q, cl); + htb_deactivate(q, cl); if (cl->cmode != HTB_CAN_SEND) htb_safe_rb_erase(&cl->pq_node, @@ -1949,8 +1947,7 @@ static int htb_change_class(struct Qdisc *sch, u32 classid, /* turn parent into inner node */ qdisc_purge_queue(parent->leaf.q); parent_qdisc = parent->leaf.q; - if (parent->prio_activity) - htb_deactivate(q, parent); + htb_deactivate(q, parent); /* remove from evt list because of level change */ if (parent->cmode != HTB_CAN_SEND) { From 63890286f557aa5c4eed7e90a5a31658de8fdb4d Mon Sep 17 00:00:00 2001 From: Cong Wang Date: Mon, 28 Apr 2025 16:29:55 -0700 Subject: [PATCH 170/555] selftests/tc-testing: Add a test case to cover basic HTB+FQ_CODEL case Integrate the reproducer from Alan into TC selftests and use scapy to generate TCP traffic instead of relying on ping command. Cc: Alan J. Wylie Signed-off-by: Cong Wang Link: https://patch.msgid.link/20250428232955.1740419-3-xiyou.wangcong@gmail.com Signed-off-by: Jakub Kicinski --- .../tc-testing/tc-tests/infra/qdiscs.json | 35 +++++++++++++++++++ 1 file changed, 35 insertions(+) diff --git a/tools/testing/selftests/tc-testing/tc-tests/infra/qdiscs.json b/tools/testing/selftests/tc-testing/tc-tests/infra/qdiscs.json index 0843f6d37e9c..a951c0d33cd2 100644 --- a/tools/testing/selftests/tc-testing/tc-tests/infra/qdiscs.json +++ b/tools/testing/selftests/tc-testing/tc-tests/infra/qdiscs.json @@ -538,5 +538,40 @@ "$TC qdisc del dev $DUMMY handle 1:0 root", "$IP addr del 10.10.10.10/24 dev $DUMMY || true" ] + }, + { + "id": "62c4", + "name": "Test HTB with FQ_CODEL - basic functionality", + "category": [ + "qdisc", + "htb", + "fq_codel" + ], + "plugins": { + "requires": [ + "nsPlugin", + "scapyPlugin" + ] + }, + "setup": [ + "$TC qdisc add dev $DEV1 root handle 1: htb default 11", + "$TC class add dev $DEV1 parent 1: classid 1:1 htb rate 10kbit", + "$TC class add dev $DEV1 parent 1:1 classid 1:11 htb rate 10kbit prio 0 quantum 1486", + "$TC qdisc add dev $DEV1 parent 1:11 fq_codel quantum 300 noecn", + "sleep 0.5" + ], + "scapy": { + "iface": "$DEV0", + "count": 5, + "packet": "Ether()/IP(dst='10.10.10.1', src='10.10.10.10')/TCP(sport=12345, dport=80)" + }, + "cmdUnderTest": "$TC -s qdisc show dev $DEV1", + "expExitCode": "0", + "verifyCmd": "$TC -s qdisc show dev $DEV1 | grep -A 5 'qdisc fq_codel'", + "matchPattern": "Sent [0-9]+ bytes [0-9]+ pkt", + "matchCount": "1", + "teardown": [ + "$TC qdisc del dev $DEV1 handle 1: root" + ] } ] From 1e20324b23f0afba27997434fb978f1e4a1dbcb6 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 30 Apr 2025 09:37:58 -0700 Subject: [PATCH 171/555] virtio-net: don't re-enable refill work too early when NAPI is disabled Commit 4bc12818b363 ("virtio-net: disable delayed refill when pausing rx") fixed a deadlock between reconfig paths and refill work trying to disable the same NAPI instance. The refill work can't run in parallel with reconfig because trying to double-disable a NAPI instance causes a stall under the instance lock, which the reconfig path needs to re-enable the NAPI and therefore unblock the stalled thread. There are two cases where we re-enable refill too early. One is in the virtnet_set_queues() handler. We call it when installing XDP: virtnet_rx_pause_all(vi); ... virtnet_napi_tx_disable(..); ... virtnet_set_queues(..); ... virtnet_rx_resume_all(..); We want the work to be disabled until we call virtnet_rx_resume_all(), but virtnet_set_queues() kicks it before NAPIs were re-enabled. The other case is a more trivial case of mis-ordering in __virtnet_rx_resume() found by code inspection. Taking the spin lock in virtnet_set_queues() (requested during review) may be unnecessary as we are under rtnl_lock and so are all paths writing to ->refill_enabled. Acked-by: Michael S. Tsirkin Reviewed-by: Bui Quang Minh Fixes: 4bc12818b363 ("virtio-net: disable delayed refill when pausing rx") Fixes: 413f0271f396 ("net: protect NAPI enablement with netdev_lock()") Link: https://patch.msgid.link/20250430163758.3029367-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- drivers/net/virtio_net.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index 848fab51dfa1..b5549d542c02 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -3383,12 +3383,15 @@ static void __virtnet_rx_resume(struct virtnet_info *vi, bool refill) { bool running = netif_running(vi->dev); + bool schedule_refill = false; if (refill && !try_fill_recv(vi, rq, GFP_KERNEL)) - schedule_delayed_work(&vi->refill, 0); - + schedule_refill = true; if (running) virtnet_napi_enable(rq); + + if (schedule_refill) + schedule_delayed_work(&vi->refill, 0); } static void virtnet_rx_resume_all(struct virtnet_info *vi) @@ -3728,8 +3731,10 @@ static int virtnet_set_queues(struct virtnet_info *vi, u16 queue_pairs) succ: vi->curr_queue_pairs = queue_pairs; /* virtnet_open() will refill when device is going to up. */ - if (dev->flags & IFF_UP) + spin_lock_bh(&vi->refill_lock); + if (dev->flags & IFF_UP && vi->refill_enabled) schedule_delayed_work(&vi->refill, 0); + spin_unlock_bh(&vi->refill_lock); return 0; } From 4397684a292a71fbc1e815c3e283f7490ddce5ae Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 30 Apr 2025 09:38:36 -0700 Subject: [PATCH 172/555] virtio-net: free xsk_buffs on error in virtnet_xsk_pool_enable() The selftests added to our CI by Bui Quang Minh recently reveals that there is a mem leak on the error path of virtnet_xsk_pool_enable(): unreferenced object 0xffff88800a68a000 (size 2048): comm "xdp_helper", pid 318, jiffies 4294692778 hex dump (first 32 bytes): 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ backtrace (crc 0): __kvmalloc_node_noprof+0x402/0x570 virtnet_xsk_pool_enable+0x293/0x6a0 (drivers/net/virtio_net.c:5882) xp_assign_dev+0x369/0x670 (net/xdp/xsk_buff_pool.c:226) xsk_bind+0x6a5/0x1ae0 __sys_bind+0x15e/0x230 __x64_sys_bind+0x72/0xb0 do_syscall_64+0xc1/0x1d0 entry_SYSCALL_64_after_hwframe+0x77/0x7f Acked-by: Jason Wang Fixes: e9f3962441c0 ("virtio_net: xsk: rx: support fill with xsk buffer") Link: https://patch.msgid.link/20250430163836.3029761-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- drivers/net/virtio_net.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index b5549d542c02..f9e3e628ec4d 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -5890,8 +5890,10 @@ static int virtnet_xsk_pool_enable(struct net_device *dev, hdr_dma = virtqueue_dma_map_single_attrs(sq->vq, &xsk_hdr, vi->hdr_len, DMA_TO_DEVICE, 0); - if (virtqueue_dma_mapping_error(sq->vq, hdr_dma)) - return -ENOMEM; + if (virtqueue_dma_mapping_error(sq->vq, hdr_dma)) { + err = -ENOMEM; + goto err_free_buffs; + } err = xsk_pool_dma_map(pool, dma_dev, 0); if (err) @@ -5919,6 +5921,8 @@ err_rq: err_xsk_map: virtqueue_dma_unmap_single_attrs(rq->vq, hdr_dma, vi->hdr_len, DMA_TO_DEVICE, 0); +err_free_buffs: + kvfree(rq->xsk_buffs); return err; } From e66b0a8f048bc8590eb1047480f946898a3f80c9 Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Tue, 15 Apr 2025 09:52:30 +0200 Subject: [PATCH 173/555] i2c: omap: fix deprecated of_property_read_bool() use Using of_property_read_bool() for non-boolean properties is deprecated and results in a warning during runtime since commit c141ecc3cecd ("of: Warn when of_property_read_bool() is used on non-boolean properties"). Fixes: b6ef830c60b6 ("i2c: omap: Add support for setting mux") Cc: Jayesh Choudhary Signed-off-by: Johan Hovold Acked-by: Mukesh Kumar Savaliya Link: https://lore.kernel.org/r/20250415075230.16235-1-johan+linaro@kernel.org Signed-off-by: Andi Shyti --- drivers/i2c/busses/i2c-omap.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/i2c/busses/i2c-omap.c b/drivers/i2c/busses/i2c-omap.c index 16afb9ca19bb..876791d20ed5 100644 --- a/drivers/i2c/busses/i2c-omap.c +++ b/drivers/i2c/busses/i2c-omap.c @@ -1454,7 +1454,7 @@ omap_i2c_probe(struct platform_device *pdev) (1000 * omap->speed / 8); } - if (of_property_read_bool(node, "mux-states")) { + if (of_property_present(node, "mux-states")) { struct mux_state *mux_state; mux_state = devm_mux_state_get(&pdev->dev, NULL); From c360eb0c3ccb95306704fd221442283ee82f1f58 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Wed, 30 Apr 2025 11:21:35 -0500 Subject: [PATCH 174/555] dt-bindings: net: ethernet-controller: Add informative text about RGMII delays Device Tree and Ethernet MAC driver writers often misunderstand RGMII delays. Rewrite the Normative section in terms of the PCB, is the PCB adding the 2ns delay. This meaning was previous implied by the definition, but often wrongly interpreted due to the ambiguous wording and looking at the definition from the wrong perspective. The new definition concentrates clearly on the hardware, and should be less ambiguous. Add an Informative section to the end of the binding describing in detail what the four RGMII delays mean. This expands on just the PCB meaning, adding in the implications for the MAC and PHY. Additionally, when the MAC or PHY needs to add a delay, which is software configuration, describe how Linux does this, in the hope of reducing errors. Make it clear other users of device tree binding may implement the software configuration in other ways while still conforming to the binding. Fixes: 9d3de3c58347 ("dt-bindings: net: Add YAML schemas for the generic Ethernet options") Signed-off-by: Andrew Lunn Acked-by: Conor Dooley Link: https://patch.msgid.link/20250430-v6-15-rc3-net-rgmii-delays-v2-1-099ae651d5e5@lunn.ch Signed-off-by: Jakub Kicinski --- .../bindings/net/ethernet-controller.yaml | 97 +++++++++++++++++-- 1 file changed, 90 insertions(+), 7 deletions(-) diff --git a/Documentation/devicetree/bindings/net/ethernet-controller.yaml b/Documentation/devicetree/bindings/net/ethernet-controller.yaml index 45819b235800..a2d4c626f659 100644 --- a/Documentation/devicetree/bindings/net/ethernet-controller.yaml +++ b/Documentation/devicetree/bindings/net/ethernet-controller.yaml @@ -74,19 +74,17 @@ properties: - rev-rmii - moca - # RX and TX delays are added by the MAC when required + # RX and TX delays are provided by the PCB. See below - rgmii - # RGMII with internal RX and TX delays provided by the PHY, - # the MAC should not add the RX or TX delays in this case + # RX and TX delays are not provided by the PCB. This is the most + # frequent case. See below - rgmii-id - # RGMII with internal RX delay provided by the PHY, the MAC - # should not add an RX delay in this case + # TX delay is provided by the PCB. See below - rgmii-rxid - # RGMII with internal TX delay provided by the PHY, the MAC - # should not add an TX delay in this case + # RX delay is provided by the PCB. See below - rgmii-txid - rtbi - smii @@ -286,4 +284,89 @@ allOf: additionalProperties: true +# Informative +# =========== +# +# 'phy-modes' & 'phy-connection-type' properties 'rgmii', 'rgmii-id', +# 'rgmii-rxid', and 'rgmii-txid' are frequently used wrongly by +# developers. This informative section clarifies their usage. +# +# The RGMII specification requires a 2ns delay between the data and +# clock signals on the RGMII bus. How this delay is implemented is not +# specified. +# +# One option is to make the clock traces on the PCB longer than the +# data traces. A sufficiently difference in length can provide the 2ns +# delay. If both the RX and TX delays are implemented in this manner, +# 'rgmii' should be used, so indicating the PCB adds the delays. +# +# If the PCB does not add these delays via extra long traces, +# 'rgmii-id' should be used. Here, 'id' refers to 'internal delay', +# where either the MAC or PHY adds the delay. +# +# If only one of the two delays are implemented via extra long clock +# lines, either 'rgmii-rxid' or 'rgmii-txid' should be used, +# indicating the MAC or PHY should implement one of the delays +# internally, while the PCB implements the other delay. +# +# Device Tree describes hardware, and in this case, it describes the +# PCB between the MAC and the PHY, if the PCB implements delays or +# not. +# +# In practice, very few PCBs make use of extra long clock lines. Hence +# any RGMII phy mode other than 'rgmii-id' is probably wrong, and is +# unlikely to be accepted during review without details provided in +# the commit description and comments in the .dts file. +# +# When the PCB does not implement the delays, the MAC or PHY must. As +# such, this is software configuration, and so not described in Device +# Tree. +# +# The following describes how Linux implements the configuration of +# the MAC and PHY to add these delays when the PCB does not. As stated +# above, developers often get this wrong, and the aim of this section +# is reduce the frequency of these errors by Linux developers. Other +# users of the Device Tree may implement it differently, and still be +# consistent with both the normative and informative description +# above. +# +# By default in Linux, when using phylib/phylink, the MAC is expected +# to read the 'phy-mode' from Device Tree, not implement any delays, +# and pass the value to the PHY. The PHY will then implement delays as +# specified by the 'phy-mode'. The PHY should always be reconfigured +# to implement the needed delays, replacing any setting performed by +# strapping or the bootloader, etc. +# +# Experience to date is that all PHYs which implement RGMII also +# implement the ability to add or not add the needed delays. Hence +# this default is expected to work in all cases. Ignoring this default +# is likely to be questioned by Reviews, and require a strong argument +# to be accepted. +# +# There are a small number of cases where the MAC has hard coded +# delays which cannot be disabled. The 'phy-mode' only describes the +# PCB. The inability to disable the delays in the MAC does not change +# the meaning of 'phy-mode'. It does however mean that a 'phy-mode' of +# 'rgmii' is now invalid, it cannot be supported, since both the PCB +# and the MAC and PHY adding delays cannot result in a functional +# link. Thus the MAC should report a fatal error for any modes which +# cannot be supported. When the MAC implements the delay, it must +# ensure that the PHY does not also implement the same delay. So it +# must modify the phy-mode it passes to the PHY, removing the delay it +# has added. Failure to remove the delay will result in a +# non-functioning link. +# +# Sometimes there is a need to fine tune the delays. Often the MAC or +# PHY can perform this fine tuning. In the MAC node, the Device Tree +# properties 'rx-internal-delay-ps' and 'tx-internal-delay-ps' should +# be used to indicate fine tuning performed by the MAC. The values +# expected here are small. A value of 2000ps, i.e 2ns, and a phy-mode +# of 'rgmii' will not be accepted by Reviewers. +# +# If the PHY is to perform fine tuning, the properties +# 'rx-internal-delay-ps' and 'tx-internal-delay-ps' in the PHY node +# should be used. When the PHY is implementing delays, e.g. 'rgmii-id' +# these properties should have a value near to 2000ps. If the PCB is +# implementing delays, e.g. 'rgmii', a small value can be used to fine +# tune the delay added by the PCB. ... From 3e6a0243ff002ddbd7ee18a8974ae61d2e6ed00d Mon Sep 17 00:00:00 2001 From: Guillaume Nault Date: Sat, 3 May 2025 00:57:52 +0200 Subject: [PATCH 175/555] gre: Fix again IPv6 link-local address generation. Use addrconf_addr_gen() to generate IPv6 link-local addresses on GRE devices in most cases and fall back to using add_v4_addrs() only in case the GRE configuration is incompatible with addrconf_addr_gen(). GRE used to use addrconf_addr_gen() until commit e5dd729460ca ("ip/ip6_gre: use the same logic as SIT interfaces when computing v6LL address") restricted this use to gretap and ip6gretap devices, and created add_v4_addrs() (borrowed from SIT) for non-Ethernet GRE ones. The original problem came when commit 9af28511be10 ("addrconf: refuse isatap eui64 for INADDR_ANY") made __ipv6_isatap_ifid() fail when its addr parameter was 0. The commit says that this would create an invalid address, however, I couldn't find any RFC saying that the generated interface identifier would be wrong. Anyway, since gre over IPv4 devices pass their local tunnel address to __ipv6_isatap_ifid(), that commit broke their IPv6 link-local address generation when the local address was unspecified. Then commit e5dd729460ca ("ip/ip6_gre: use the same logic as SIT interfaces when computing v6LL address") tried to fix that case by defining add_v4_addrs() and calling it to generate the IPv6 link-local address instead of using addrconf_addr_gen() (apart for gretap and ip6gretap devices, which would still use the regular addrconf_addr_gen(), since they have a MAC address). That broke several use cases because add_v4_addrs() isn't properly integrated into the rest of IPv6 Neighbor Discovery code. Several of these shortcomings have been fixed over time, but add_v4_addrs() remains broken on several aspects. In particular, it doesn't send any Router Sollicitations, so the SLAAC process doesn't start until the interface receives a Router Advertisement. Also, add_v4_addrs() mostly ignores the address generation mode of the interface (/proc/sys/net/ipv6/conf/*/addr_gen_mode), thus breaking the IN6_ADDR_GEN_MODE_RANDOM and IN6_ADDR_GEN_MODE_STABLE_PRIVACY cases. Fix the situation by using add_v4_addrs() only in the specific scenario where the normal method would fail. That is, for interfaces that have all of the following characteristics: * run over IPv4, * transport IP packets directly, not Ethernet (that is, not gretap interfaces), * tunnel endpoint is INADDR_ANY (that is, 0), * device address generation mode is EUI64. In all other cases, revert back to the regular addrconf_addr_gen(). Also, remove the special case for ip6gre interfaces in add_v4_addrs(), since ip6gre devices now always use addrconf_addr_gen() instead. Note: This patch was originally applied as commit 183185a18ff9 ("gre: Fix IPv6 link-local address generation."). However, it was then reverted by commit fc486c2d060f ("Revert "gre: Fix IPv6 link-local address generation."") because it uncovered another bug that ended up breaking net/forwarding/ip6gre_custom_multipath_hash.sh. That other bug has now been fixed by commit 4d0ab3a6885e ("ipv6: Start path selection from the first nexthop"). Therefore we can now revive this GRE patch (no changes since original commit 183185a18ff9 ("gre: Fix IPv6 link-local address generation."). Fixes: e5dd729460ca ("ip/ip6_gre: use the same logic as SIT interfaces when computing v6LL address") Signed-off-by: Guillaume Nault Reviewed-by: Ido Schimmel Link: https://patch.msgid.link/a88cc5c4811af36007645d610c95102dccb360a6.1746225214.git.gnault@redhat.com Signed-off-by: Jakub Kicinski --- net/ipv6/addrconf.c | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 9ba83f0c9928..c6b22170dc49 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -3214,16 +3214,13 @@ static void add_v4_addrs(struct inet6_dev *idev) struct in6_addr addr; struct net_device *dev; struct net *net = dev_net(idev->dev); - int scope, plen, offset = 0; + int scope, plen; u32 pflags = 0; ASSERT_RTNL(); memset(&addr, 0, sizeof(struct in6_addr)); - /* in case of IP6GRE the dev_addr is an IPv6 and therefore we use only the last 4 bytes */ - if (idev->dev->addr_len == sizeof(struct in6_addr)) - offset = sizeof(struct in6_addr) - 4; - memcpy(&addr.s6_addr32[3], idev->dev->dev_addr + offset, 4); + memcpy(&addr.s6_addr32[3], idev->dev->dev_addr, 4); if (!(idev->dev->flags & IFF_POINTOPOINT) && idev->dev->type == ARPHRD_SIT) { scope = IPV6_ADDR_COMPATv4; @@ -3534,7 +3531,13 @@ static void addrconf_gre_config(struct net_device *dev) return; } - if (dev->type == ARPHRD_ETHER) { + /* Generate the IPv6 link-local address using addrconf_addr_gen(), + * unless we have an IPv4 GRE device not bound to an IP address and + * which is in EUI64 mode (as __ipv6_isatap_ifid() would fail in this + * case). Such devices fall back to add_v4_addrs() instead. + */ + if (!(dev->type == ARPHRD_IPGRE && *(__be32 *)dev->dev_addr == 0 && + idev->cnf.addr_gen_mode == IN6_ADDR_GEN_MODE_EUI64)) { addrconf_addr_gen(idev, true); return; } From b6a6006b0e3d10e62c465613f07ff49268baedb1 Mon Sep 17 00:00:00 2001 From: Guillaume Nault Date: Sat, 3 May 2025 00:57:59 +0200 Subject: [PATCH 176/555] selftests: Add IPv6 link-local address generation tests for GRE devices. GRE devices have their special code for IPv6 link-local address generation that has been the source of several regressions in the past. Add selftest to check that all gre, ip6gre, gretap and ip6gretap get an IPv6 link-link local address in accordance with the net.ipv6.conf..addr_gen_mode sysctl. Note: This patch was originally applied as commit 6f50175ccad4 ("selftests: Add IPv6 link-local address generation tests for GRE devices."). However, it was then reverted by commit 355d940f4d5a ("Revert "selftests: Add IPv6 link-local address generation tests for GRE devices."") because the commit it depended on was going to be reverted. Now that the situation is resolved, we can add this selftest again (no changes since original patch, appart from context update in tools/testing/selftests/net/Makefile). Signed-off-by: Guillaume Nault Reviewed-by: Ido Schimmel Tested-by: Ido Schimmel Link: https://patch.msgid.link/2c3a5733cb3a6e3119504361a9b9f89fda570a2d.1746225214.git.gnault@redhat.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/Makefile | 1 + .../testing/selftests/net/gre_ipv6_lladdr.sh | 177 ++++++++++++++++++ 2 files changed, 178 insertions(+) create mode 100755 tools/testing/selftests/net/gre_ipv6_lladdr.sh diff --git a/tools/testing/selftests/net/Makefile b/tools/testing/selftests/net/Makefile index 124078b56fa4..70a38f485d4d 100644 --- a/tools/testing/selftests/net/Makefile +++ b/tools/testing/selftests/net/Makefile @@ -31,6 +31,7 @@ TEST_PROGS += veth.sh TEST_PROGS += ioam6.sh TEST_PROGS += gro.sh TEST_PROGS += gre_gso.sh +TEST_PROGS += gre_ipv6_lladdr.sh TEST_PROGS += cmsg_so_mark.sh TEST_PROGS += cmsg_so_priority.sh TEST_PROGS += test_so_rcv.sh diff --git a/tools/testing/selftests/net/gre_ipv6_lladdr.sh b/tools/testing/selftests/net/gre_ipv6_lladdr.sh new file mode 100755 index 000000000000..5b34f6e1f831 --- /dev/null +++ b/tools/testing/selftests/net/gre_ipv6_lladdr.sh @@ -0,0 +1,177 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 + +source ./lib.sh + +PAUSE_ON_FAIL="no" + +# The trap function handler +# +exit_cleanup_all() +{ + cleanup_all_ns + + exit "${EXIT_STATUS}" +} + +# Add fake IPv4 and IPv6 networks on the loopback device, to be used as +# underlay by future GRE devices. +# +setup_basenet() +{ + ip -netns "${NS0}" link set dev lo up + ip -netns "${NS0}" address add dev lo 192.0.2.10/24 + ip -netns "${NS0}" address add dev lo 2001:db8::10/64 nodad +} + +# Check if network device has an IPv6 link-local address assigned. +# +# Parameters: +# +# * $1: The network device to test +# * $2: An extra regular expression that should be matched (to verify the +# presence of extra attributes) +# * $3: The expected return code from grep (to allow checking the absence of +# a link-local address) +# * $4: The user visible name for the scenario being tested +# +check_ipv6_ll_addr() +{ + local DEV="$1" + local EXTRA_MATCH="$2" + local XRET="$3" + local MSG="$4" + + RET=0 + set +e + ip -netns "${NS0}" -6 address show dev "${DEV}" scope link | grep "fe80::" | grep -q "${EXTRA_MATCH}" + check_err_fail "${XRET}" $? "" + log_test "${MSG}" + set -e +} + +# Create a GRE device and verify that it gets an IPv6 link-local address as +# expected. +# +# Parameters: +# +# * $1: The device type (gre, ip6gre, gretap or ip6gretap) +# * $2: The local underlay IP address (can be an IPv4, an IPv6 or "any") +# * $3: The remote underlay IP address (can be an IPv4, an IPv6 or "any") +# * $4: The IPv6 interface identifier generation mode to use for the GRE +# device (eui64, none, stable-privacy or random). +# +test_gre_device() +{ + local GRE_TYPE="$1" + local LOCAL_IP="$2" + local REMOTE_IP="$3" + local MODE="$4" + local ADDR_GEN_MODE + local MATCH_REGEXP + local MSG + + ip link add netns "${NS0}" name gretest type "${GRE_TYPE}" local "${LOCAL_IP}" remote "${REMOTE_IP}" + + case "${MODE}" in + "eui64") + ADDR_GEN_MODE=0 + MATCH_REGEXP="" + MSG="${GRE_TYPE}, mode: 0 (EUI64), ${LOCAL_IP} -> ${REMOTE_IP}" + XRET=0 + ;; + "none") + ADDR_GEN_MODE=1 + MATCH_REGEXP="" + MSG="${GRE_TYPE}, mode: 1 (none), ${LOCAL_IP} -> ${REMOTE_IP}" + XRET=1 # No link-local address should be generated + ;; + "stable-privacy") + ADDR_GEN_MODE=2 + MATCH_REGEXP="stable-privacy" + MSG="${GRE_TYPE}, mode: 2 (stable privacy), ${LOCAL_IP} -> ${REMOTE_IP}" + XRET=0 + # Initialise stable_secret (required for stable-privacy mode) + ip netns exec "${NS0}" sysctl -qw net.ipv6.conf.gretest.stable_secret="2001:db8::abcd" + ;; + "random") + ADDR_GEN_MODE=3 + MATCH_REGEXP="stable-privacy" + MSG="${GRE_TYPE}, mode: 3 (random), ${LOCAL_IP} -> ${REMOTE_IP}" + XRET=0 + ;; + esac + + # Check that IPv6 link-local address is generated when device goes up + ip netns exec "${NS0}" sysctl -qw net.ipv6.conf.gretest.addr_gen_mode="${ADDR_GEN_MODE}" + ip -netns "${NS0}" link set dev gretest up + check_ipv6_ll_addr gretest "${MATCH_REGEXP}" "${XRET}" "config: ${MSG}" + + # Now disable link-local address generation + ip -netns "${NS0}" link set dev gretest down + ip netns exec "${NS0}" sysctl -qw net.ipv6.conf.gretest.addr_gen_mode=1 + ip -netns "${NS0}" link set dev gretest up + + # Check that link-local address generation works when re-enabled while + # the device is already up + ip netns exec "${NS0}" sysctl -qw net.ipv6.conf.gretest.addr_gen_mode="${ADDR_GEN_MODE}" + check_ipv6_ll_addr gretest "${MATCH_REGEXP}" "${XRET}" "update: ${MSG}" + + ip -netns "${NS0}" link del dev gretest +} + +test_gre4() +{ + local GRE_TYPE + local MODE + + for GRE_TYPE in "gre" "gretap"; do + printf "\n####\nTesting IPv6 link-local address generation on ${GRE_TYPE} devices\n####\n\n" + + for MODE in "eui64" "none" "stable-privacy" "random"; do + test_gre_device "${GRE_TYPE}" 192.0.2.10 192.0.2.11 "${MODE}" + test_gre_device "${GRE_TYPE}" any 192.0.2.11 "${MODE}" + test_gre_device "${GRE_TYPE}" 192.0.2.10 any "${MODE}" + done + done +} + +test_gre6() +{ + local GRE_TYPE + local MODE + + for GRE_TYPE in "ip6gre" "ip6gretap"; do + printf "\n####\nTesting IPv6 link-local address generation on ${GRE_TYPE} devices\n####\n\n" + + for MODE in "eui64" "none" "stable-privacy" "random"; do + test_gre_device "${GRE_TYPE}" 2001:db8::10 2001:db8::11 "${MODE}" + test_gre_device "${GRE_TYPE}" any 2001:db8::11 "${MODE}" + test_gre_device "${GRE_TYPE}" 2001:db8::10 any "${MODE}" + done + done +} + +usage() +{ + echo "Usage: $0 [-p]" + exit 1 +} + +while getopts :p o +do + case $o in + p) PAUSE_ON_FAIL="yes";; + *) usage;; + esac +done + +setup_ns NS0 + +set -e +trap exit_cleanup_all EXIT + +setup_basenet + +test_gre4 +test_gre6 From b344a48cbe5fb24697addc6afbb7358b938a7d31 Mon Sep 17 00:00:00 2001 From: Mohsin Bashir Date: Fri, 2 May 2025 18:35:16 -0700 Subject: [PATCH 177/555] selftests: drv: net: fix test failure on ipv6 sys The `get_interface_info` call has ip version hard-coded which leads to failures on an IPV6 system. The NetDrvEnv class already gathers information about remote interface, so instead of fixing the local implementation switch to using cfg.remote_ifname. Before: ./drivers/net/ping.py Traceback (most recent call last): File "/new_tests/./drivers/net/ping.py", line 217, in main() File "/new_tests/./drivers/net/ping.py", line 204, in main get_interface_info(cfg) File "/new_tests/./drivers/net/ping.py", line 128, in get_interface_info raise KsftFailEx('Can not get remote interface') net.lib.py.ksft.KsftFailEx: Can not get remote interface After: ./drivers/net/ping.py TAP version 13 1..6 ok 1 ping.test_default # SKIP Test requires IPv4 connectivity ok 2 ping.test_xdp_generic_sb # SKIP Test requires IPv4 connectivity ok 3 ping.test_xdp_generic_mb # SKIP Test requires IPv4 connectivity ok 4 ping.test_xdp_native_sb # SKIP Test requires IPv4 connectivity ok 5 ping.test_xdp_native_mb # SKIP Test requires IPv4 connectivity ok 6 ping.test_xdp_offload # SKIP device does not support offloaded XDP Totals: pass:0 fail:0 xfail:0 xpass:0 skip:6 error:0 Fixes: 75cc19c8ff89 ("selftests: drv-net: add xdp cases for ping.py") Signed-off-by: Mohsin Bashir Reviewed-by: David Wei Link: https://patch.msgid.link/20250503013518.1722913-2-mohsin.bashr@gmail.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/drivers/net/ping.py | 22 +++++++++------------ 1 file changed, 9 insertions(+), 13 deletions(-) diff --git a/tools/testing/selftests/drivers/net/ping.py b/tools/testing/selftests/drivers/net/ping.py index 4b6822866066..5272e8b3536d 100755 --- a/tools/testing/selftests/drivers/net/ping.py +++ b/tools/testing/selftests/drivers/net/ping.py @@ -9,7 +9,6 @@ from lib.py import EthtoolFamily, NetDrvEpEnv from lib.py import bkg, cmd, wait_port_listen, rand_port from lib.py import defer, ethtool, ip -remote_ifname="" no_sleep=False def _test_v4(cfg) -> None: @@ -57,7 +56,7 @@ def _set_offload_checksum(cfg, netnl, on) -> None: def _set_xdp_generic_sb_on(cfg) -> None: prog = cfg.net_lib_dir / "xdp_dummy.bpf.o" - cmd(f"ip link set dev {remote_ifname} mtu 1500", shell=True, host=cfg.remote) + cmd(f"ip link set dev {cfg.remote_ifname} mtu 1500", shell=True, host=cfg.remote) cmd(f"ip link set dev {cfg.ifname} mtu 1500 xdpgeneric obj {prog} sec xdp", shell=True) defer(cmd, f"ip link set dev {cfg.ifname} xdpgeneric off") @@ -66,8 +65,8 @@ def _set_xdp_generic_sb_on(cfg) -> None: def _set_xdp_generic_mb_on(cfg) -> None: prog = cfg.net_lib_dir / "xdp_dummy.bpf.o" - cmd(f"ip link set dev {remote_ifname} mtu 9000", shell=True, host=cfg.remote) - defer(ip, f"link set dev {remote_ifname} mtu 1500", host=cfg.remote) + cmd(f"ip link set dev {cfg.remote_ifname} mtu 9000", shell=True, host=cfg.remote) + defer(ip, f"link set dev {cfg.remote_ifname} mtu 1500", host=cfg.remote) ip("link set dev %s mtu 9000 xdpgeneric obj %s sec xdp.frags" % (cfg.ifname, prog)) defer(ip, f"link set dev {cfg.ifname} mtu 1500 xdpgeneric off") @@ -76,7 +75,7 @@ def _set_xdp_generic_mb_on(cfg) -> None: def _set_xdp_native_sb_on(cfg) -> None: prog = cfg.net_lib_dir / "xdp_dummy.bpf.o" - cmd(f"ip link set dev {remote_ifname} mtu 1500", shell=True, host=cfg.remote) + cmd(f"ip link set dev {cfg.remote_ifname} mtu 1500", shell=True, host=cfg.remote) cmd(f"ip -j link set dev {cfg.ifname} mtu 1500 xdp obj {prog} sec xdp", shell=True) defer(ip, f"link set dev {cfg.ifname} mtu 1500 xdp off") xdp_info = ip("-d link show %s" % (cfg.ifname), json=True)[0] @@ -93,8 +92,8 @@ def _set_xdp_native_sb_on(cfg) -> None: def _set_xdp_native_mb_on(cfg) -> None: prog = cfg.net_lib_dir / "xdp_dummy.bpf.o" - cmd(f"ip link set dev {remote_ifname} mtu 9000", shell=True, host=cfg.remote) - defer(ip, f"link set dev {remote_ifname} mtu 1500", host=cfg.remote) + cmd(f"ip link set dev {cfg.remote_ifname} mtu 9000", shell=True, host=cfg.remote) + defer(ip, f"link set dev {cfg.remote_ifname} mtu 1500", host=cfg.remote) try: cmd(f"ip link set dev {cfg.ifname} mtu 9000 xdp obj {prog} sec xdp.frags", shell=True) defer(ip, f"link set dev {cfg.ifname} mtu 1500 xdp off") @@ -112,18 +111,15 @@ def _set_xdp_offload_on(cfg) -> None: except Exception as e: raise KsftSkipEx('device does not support offloaded XDP') defer(ip, f"link set dev {cfg.ifname} xdpoffload off") - cmd(f"ip link set dev {remote_ifname} mtu 1500", shell=True, host=cfg.remote) + cmd(f"ip link set dev {cfg.remote_ifname} mtu 1500", shell=True, host=cfg.remote) if no_sleep != True: time.sleep(10) def get_interface_info(cfg) -> None: - global remote_ifname global no_sleep - remote_info = cmd(f"ip -4 -o addr show to {cfg.remote_addr_v['4']} | awk '{{print $2}}'", shell=True, host=cfg.remote).stdout - remote_ifname = remote_info.rstrip('\n') - if remote_ifname == "": + if cfg.remote_ifname == "": raise KsftFailEx('Can not get remote interface') local_info = ip("-d link show %s" % (cfg.ifname), json=True)[0] if 'parentbus' in local_info and local_info['parentbus'] == "netdevsim": @@ -136,7 +132,7 @@ def set_interface_init(cfg) -> None: cmd(f"ip link set dev {cfg.ifname} xdp off ", shell=True) cmd(f"ip link set dev {cfg.ifname} xdpgeneric off ", shell=True) cmd(f"ip link set dev {cfg.ifname} xdpoffload off", shell=True) - cmd(f"ip link set dev {remote_ifname} mtu 1500", shell=True, host=cfg.remote) + cmd(f"ip link set dev {cfg.remote_ifname} mtu 1500", shell=True, host=cfg.remote) def test_default(cfg, netnl) -> None: _set_offload_checksum(cfg, netnl, "off") From 8bb7d8e5cf7f44ea89a445975cbd78888324f234 Mon Sep 17 00:00:00 2001 From: Mohsin Bashir Date: Fri, 2 May 2025 18:35:17 -0700 Subject: [PATCH 178/555] selftests: drv: net: avoid skipping tests On a system with either of the ipv4 or ipv6 information missing, tests are currently skipped. Ideally, the test should run as long as at least one address family is present. This patch make test run whenever possible. Before: ./drivers/net/ping.py TAP version 13 1..6 ok 1 ping.test_default # SKIP Test requires IPv4 connectivity ok 2 ping.test_xdp_generic_sb # SKIP Test requires IPv4 connectivity ok 3 ping.test_xdp_generic_mb # SKIP Test requires IPv4 connectivity ok 4 ping.test_xdp_native_sb # SKIP Test requires IPv4 connectivity ok 5 ping.test_xdp_native_mb # SKIP Test requires IPv4 connectivity ok 6 ping.test_xdp_offload # SKIP device does not support offloaded XDP Totals: pass:0 fail:0 xfail:0 xpass:0 skip:6 error:0 After: ./drivers/net/ping.py TAP version 13 1..6 ok 1 ping.test_default ok 2 ping.test_xdp_generic_sb ok 3 ping.test_xdp_generic_mb ok 4 ping.test_xdp_native_sb ok 5 ping.test_xdp_native_mb ok 6 ping.test_xdp_offload # SKIP device does not support offloaded XDP Totals: pass:5 fail:0 xfail:0 xpass:0 skip:1 error:0 Fixes: 75cc19c8ff89 ("selftests: drv-net: add xdp cases for ping.py") Signed-off-by: Mohsin Bashir Link: https://patch.msgid.link/20250503013518.1722913-3-mohsin.bashr@gmail.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/drivers/net/ping.py | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/drivers/net/ping.py b/tools/testing/selftests/drivers/net/ping.py index 5272e8b3536d..16b7d3ab0fc8 100755 --- a/tools/testing/selftests/drivers/net/ping.py +++ b/tools/testing/selftests/drivers/net/ping.py @@ -12,7 +12,8 @@ from lib.py import defer, ethtool, ip no_sleep=False def _test_v4(cfg) -> None: - cfg.require_ipver("4") + if not cfg.addr_v["4"]: + return cmd("ping -c 1 -W0.5 " + cfg.remote_addr_v["4"]) cmd("ping -c 1 -W0.5 " + cfg.addr_v["4"], host=cfg.remote) @@ -20,7 +21,8 @@ def _test_v4(cfg) -> None: cmd("ping -s 65000 -c 1 -W0.5 " + cfg.addr_v["4"], host=cfg.remote) def _test_v6(cfg) -> None: - cfg.require_ipver("6") + if not cfg.addr_v["6"]: + return cmd("ping -c 1 -W5 " + cfg.remote_addr_v["6"]) cmd("ping -c 1 -W5 " + cfg.addr_v["6"], host=cfg.remote) From 4a9d494ca24b7fd509b3953fcb43d580cb05097b Mon Sep 17 00:00:00 2001 From: Mohsin Bashir Date: Fri, 2 May 2025 18:35:18 -0700 Subject: [PATCH 179/555] selftests: drv: net: add version indicator Currently, the test result does not differentiate between the cases when either one of the address families are configured or if both the address families are configured. Ideally, the result should report if a particular case was skipped. ./drivers/net/ping.py TAP version 13 1..7 ok 1 ping.test_default_v4 # SKIP Test requires IPv4 connectivity ok 2 ping.test_default_v6 ok 3 ping.test_xdp_generic_sb ok 4 ping.test_xdp_generic_mb ok 5 ping.test_xdp_native_sb ok 6 ping.test_xdp_native_mb ok 7 ping.test_xdp_offload # SKIP device does not support offloaded XDP Totals: pass:5 fail:0 xfail:0 xpass:0 skip:2 error:0 Fixes: 75cc19c8ff89 ("selftests: drv-net: add xdp cases for ping.py") Signed-off-by: Mohsin Bashir Reviewed-by: David Wei Link: https://patch.msgid.link/20250503013518.1722913-4-mohsin.bashr@gmail.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/drivers/net/ping.py | 17 ++++++++++++++--- 1 file changed, 14 insertions(+), 3 deletions(-) diff --git a/tools/testing/selftests/drivers/net/ping.py b/tools/testing/selftests/drivers/net/ping.py index 16b7d3ab0fc8..af8df2313a3b 100755 --- a/tools/testing/selftests/drivers/net/ping.py +++ b/tools/testing/selftests/drivers/net/ping.py @@ -136,13 +136,23 @@ def set_interface_init(cfg) -> None: cmd(f"ip link set dev {cfg.ifname} xdpoffload off", shell=True) cmd(f"ip link set dev {cfg.remote_ifname} mtu 1500", shell=True, host=cfg.remote) -def test_default(cfg, netnl) -> None: +def test_default_v4(cfg, netnl) -> None: + cfg.require_ipver("4") + _set_offload_checksum(cfg, netnl, "off") _test_v4(cfg) - _test_v6(cfg) _test_tcp(cfg) _set_offload_checksum(cfg, netnl, "on") _test_v4(cfg) + _test_tcp(cfg) + +def test_default_v6(cfg, netnl) -> None: + cfg.require_ipver("6") + + _set_offload_checksum(cfg, netnl, "off") + _test_v6(cfg) + _test_tcp(cfg) + _set_offload_checksum(cfg, netnl, "on") _test_v6(cfg) _test_tcp(cfg) @@ -200,7 +210,8 @@ def main() -> None: with NetDrvEpEnv(__file__) as cfg: get_interface_info(cfg) set_interface_init(cfg) - ksft_run([test_default, + ksft_run([test_default_v4, + test_default_v6, test_xdp_generic_sb, test_xdp_generic_mb, test_xdp_native_sb, From 4720f9707c783f642332dee3d56dccaefa850e42 Mon Sep 17 00:00:00 2001 From: David Wei Date: Fri, 2 May 2025 21:30:50 -0700 Subject: [PATCH 180/555] tools: ynl-gen: validate 0 len strings from kernel Strings from the kernel are guaranteed to be null terminated and ynl_attr_validate() checks for this. But it doesn't check if the string has a len of 0, which would cause problems when trying to access data[len - 1]. Fix this by checking that len is positive. Signed-off-by: David Wei Link: https://patch.msgid.link/20250503043050.861238-1-dw@davidwei.uk Signed-off-by: Jakub Kicinski --- tools/net/ynl/lib/ynl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/net/ynl/lib/ynl.c b/tools/net/ynl/lib/ynl.c index ce32cb35007d..c4da34048ef8 100644 --- a/tools/net/ynl/lib/ynl.c +++ b/tools/net/ynl/lib/ynl.c @@ -364,7 +364,7 @@ int ynl_attr_validate(struct ynl_parse_arg *yarg, const struct nlattr *attr) "Invalid attribute (binary %s)", policy->name); return -1; case YNL_PT_NUL_STR: - if ((!policy->len || len <= policy->len) && !data[len - 1]) + if (len && (!policy->len || len <= policy->len) && !data[len - 1]) break; yerr(yarg->ys, YNL_ERROR_ATTR_INVALID, "Invalid attribute (string %s)", policy->name); From 6ba0982c3235a047c953bf73a7b014af7840c4de Mon Sep 17 00:00:00 2001 From: Luis Chamberlain Date: Mon, 5 May 2025 12:35:19 -0700 Subject: [PATCH 181/555] swapfile: disable swapon for bs > ps devices Devices which have a requirement for bs > ps cannot be supported for swap as swap still needs work. Now that the block device cache sets the min order for block devices we need this stop gap otherwise all swap operations are rejected. Without this you'll end up with errors on these devices as the swap code still needs much love to support min order. With this we at least now put a stop gap of its use, until the swap subsystem completes its major overhaul: mkswap: /dev/nvme3n1: warning: wiping old swap signature. Setting up swapspace version 1, size = 100 GiB (107374178304 bytes) no label, UUID=6af76b5c-7e7b-4902-b7f7-4c24dde6fa36 swapon: /dev/nvme3n1: swapon failed: Invalid argument Reviewed-by: Davidlohr Bueso Signed-off-by: Luis Chamberlain Link: https://lore.kernel.org/aBkS926thy9zvdZb@bombadil.infradead.org Signed-off-by: Christian Brauner --- mm/swapfile.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/mm/swapfile.c b/mm/swapfile.c index 2eff8b51a945..74a6c580b00d 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -3322,6 +3322,15 @@ SYSCALL_DEFINE2(swapon, const char __user *, specialfile, int, swap_flags) goto bad_swap_unlock_inode; } + /* + * The swap subsystem needs a major overhaul to support this. + * It doesn't work yet so just disable it for now. + */ + if (mapping_min_folio_order(mapping) > 0) { + error = -EINVAL; + goto bad_swap_unlock_inode; + } + /* * Read the swap header. */ From 2bb04ea9e5b7a2ef583c042ed5f8111804606e9d Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= Date: Fri, 2 May 2025 15:01:01 +0200 Subject: [PATCH 182/555] drm/ttm: Fix ttm_backup kerneldoc MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The docs were not properly updated from an earlier version of the code. Fixes: e7b5d23e5d47 ("drm/ttm: Provide a shmem backup implementation") Cc: Christian König Cc: Matthew Brost Cc: Matthew Auld Cc: dri-devel@lists.freedesktop.org Signed-off-by: Thomas Hellström Reviewed-by: Matthew Brost Link: https://lore.kernel.org/r/20250502130101.3185-1-thomas.hellstrom@linux.intel.com --- drivers/gpu/drm/ttm/ttm_backup.c | 2 +- include/drm/ttm/ttm_backup.h | 6 +++--- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/ttm/ttm_backup.c b/drivers/gpu/drm/ttm/ttm_backup.c index 93c007f18855..f58e7393888f 100644 --- a/drivers/gpu/drm/ttm/ttm_backup.c +++ b/drivers/gpu/drm/ttm/ttm_backup.c @@ -55,7 +55,7 @@ void ttm_backup_drop(struct ttm_backup *backup, pgoff_t handle) * @backup: The struct backup pointer used to back up the page. * @dst: The struct page to copy into. * @handle: The handle returned when the page was backed up. - * @intr: Try to perform waits interruptable or at least killable. + * @intr: Try to perform waits interruptible or at least killable. * * Return: 0 on success, Negative error code on failure, notably * -EINTR if @intr was set to true and a signal is pending. diff --git a/include/drm/ttm/ttm_backup.h b/include/drm/ttm/ttm_backup.h index 24ad120b8827..574b932177cc 100644 --- a/include/drm/ttm/ttm_backup.h +++ b/include/drm/ttm/ttm_backup.h @@ -16,7 +16,7 @@ struct ttm_backup; * @handle: The handle to convert. * * Converts an opaque handle received from the - * struct ttm_backoup_ops::backup_page() function to an (invalid) + * ttm_backup_backup_page() function to an (invalid) * struct page pointer suitable for a struct page array. * * Return: An (invalid) struct page pointer. @@ -45,8 +45,8 @@ static inline bool ttm_backup_page_ptr_is_handle(const struct page *page) * * Return: The handle that was previously used in * ttm_backup_handle_to_page_ptr() to obtain a struct page pointer, suitable - * for use as argument in the struct ttm_backup_ops drop() or - * copy_backed_up_page() functions. + * for use as argument in the struct ttm_backup_drop() or + * ttm_backup_copy_page() functions. */ static inline unsigned long ttm_backup_page_ptr_to_handle(const struct page *page) From d4ad53adfe21df1464bae5ed916085a69d6ffb3d Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= Date: Fri, 2 May 2025 15:00:14 +0200 Subject: [PATCH 183/555] drm/ttm: Remove the struct ttm_backup abstraction MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The abstraction was previously added to support separate ttm_backup implementations. However with the current implementation casting from a struct file to a struct ttm_backup, we run into trouble since struct file may have randomized the layout and gcc complains. Remove the struct ttm_backup abstraction Cc: dri-devel@lists.freedesktop.org Cc: Matthew Brost Cc: Dave Airlie Cc: Christian König Cc: Matthew Auld Cc: Al Viro Reported-by: Kees Cook Closes: https://lore.kernel.org/dri-devel/9c8dbbafdaf9f3f089da2cde5a772d69579b3795.camel@linux.intel.com/T/#mb153ab9216cb813b92bdeb36f391ad4808c2ba29 Suggested-by: Christian König Fixes: 70d645deac98 ("drm/ttm: Add helpers for shrinking") Signed-off-by: Thomas Hellström Reviewed-by: Christian König Link: https://lore.kernel.org/r/20250502130014.3156-1-thomas.hellstrom@linux.intel.com --- drivers/gpu/drm/ttm/ttm_backup.c | 42 +++++++++----------------------- drivers/gpu/drm/ttm/ttm_pool.c | 6 ++--- drivers/gpu/drm/ttm/ttm_tt.c | 2 +- include/drm/ttm/ttm_backup.h | 12 ++++----- include/drm/ttm/ttm_tt.h | 2 +- 5 files changed, 21 insertions(+), 43 deletions(-) diff --git a/drivers/gpu/drm/ttm/ttm_backup.c b/drivers/gpu/drm/ttm/ttm_backup.c index f58e7393888f..9e2d72c447ee 100644 --- a/drivers/gpu/drm/ttm/ttm_backup.c +++ b/drivers/gpu/drm/ttm/ttm_backup.c @@ -7,20 +7,6 @@ #include #include -/* - * Casting from randomized struct file * to struct ttm_backup * is fine since - * struct ttm_backup is never defined nor dereferenced. - */ -static struct file *ttm_backup_to_file(struct ttm_backup *backup) -{ - return (void *)backup; -} - -static struct ttm_backup *ttm_file_to_backup(struct file *file) -{ - return (void *)file; -} - /* * Need to map shmem indices to handle since a handle value * of 0 means error, following the swp_entry_t convention. @@ -40,12 +26,12 @@ static pgoff_t ttm_backup_handle_to_shmem_idx(pgoff_t handle) * @backup: The struct backup pointer used to obtain the handle * @handle: The handle obtained from the @backup_page function. */ -void ttm_backup_drop(struct ttm_backup *backup, pgoff_t handle) +void ttm_backup_drop(struct file *backup, pgoff_t handle) { loff_t start = ttm_backup_handle_to_shmem_idx(handle); start <<= PAGE_SHIFT; - shmem_truncate_range(file_inode(ttm_backup_to_file(backup)), start, + shmem_truncate_range(file_inode(backup), start, start + PAGE_SIZE - 1); } @@ -60,11 +46,10 @@ void ttm_backup_drop(struct ttm_backup *backup, pgoff_t handle) * Return: 0 on success, Negative error code on failure, notably * -EINTR if @intr was set to true and a signal is pending. */ -int ttm_backup_copy_page(struct ttm_backup *backup, struct page *dst, +int ttm_backup_copy_page(struct file *backup, struct page *dst, pgoff_t handle, bool intr) { - struct file *filp = ttm_backup_to_file(backup); - struct address_space *mapping = filp->f_mapping; + struct address_space *mapping = backup->f_mapping; struct folio *from_folio; pgoff_t idx = ttm_backup_handle_to_shmem_idx(handle); @@ -106,12 +91,11 @@ int ttm_backup_copy_page(struct ttm_backup *backup, struct page *dst, * the folio size- and usage. */ s64 -ttm_backup_backup_page(struct ttm_backup *backup, struct page *page, +ttm_backup_backup_page(struct file *backup, struct page *page, bool writeback, pgoff_t idx, gfp_t page_gfp, gfp_t alloc_gfp) { - struct file *filp = ttm_backup_to_file(backup); - struct address_space *mapping = filp->f_mapping; + struct address_space *mapping = backup->f_mapping; unsigned long handle = 0; struct folio *to_folio; int ret; @@ -161,9 +145,9 @@ ttm_backup_backup_page(struct ttm_backup *backup, struct page *page, * * After a call to this function, it's illegal to use the @backup pointer. */ -void ttm_backup_fini(struct ttm_backup *backup) +void ttm_backup_fini(struct file *backup) { - fput(ttm_backup_to_file(backup)); + fput(backup); } /** @@ -194,14 +178,10 @@ EXPORT_SYMBOL_GPL(ttm_backup_bytes_avail); * * Create a backup utilizing shmem objects. * - * Return: A pointer to a struct ttm_backup on success, + * Return: A pointer to a struct file on success, * an error pointer on error. */ -struct ttm_backup *ttm_backup_shmem_create(loff_t size) +struct file *ttm_backup_shmem_create(loff_t size) { - struct file *filp; - - filp = shmem_file_setup("ttm shmem backup", size, 0); - - return ttm_file_to_backup(filp); + return shmem_file_setup("ttm shmem backup", size, 0); } diff --git a/drivers/gpu/drm/ttm/ttm_pool.c b/drivers/gpu/drm/ttm/ttm_pool.c index 83b10706ba89..c2ea865be657 100644 --- a/drivers/gpu/drm/ttm/ttm_pool.c +++ b/drivers/gpu/drm/ttm/ttm_pool.c @@ -506,7 +506,7 @@ static void ttm_pool_allocated_page_commit(struct page *allocated, * if successful, populate the page-table and dma-address arrays. */ static int ttm_pool_restore_commit(struct ttm_pool_tt_restore *restore, - struct ttm_backup *backup, + struct file *backup, const struct ttm_operation_ctx *ctx, struct ttm_pool_alloc_state *alloc) @@ -655,7 +655,7 @@ static void ttm_pool_free_range(struct ttm_pool *pool, struct ttm_tt *tt, pgoff_t start_page, pgoff_t end_page) { struct page **pages = &tt->pages[start_page]; - struct ttm_backup *backup = tt->backup; + struct file *backup = tt->backup; pgoff_t i, nr; for (i = start_page; i < end_page; i += nr, pages += nr) { @@ -963,7 +963,7 @@ void ttm_pool_drop_backed_up(struct ttm_tt *tt) long ttm_pool_backup(struct ttm_pool *pool, struct ttm_tt *tt, const struct ttm_backup_flags *flags) { - struct ttm_backup *backup = tt->backup; + struct file *backup = tt->backup; struct page *page; unsigned long handle; gfp_t alloc_gfp; diff --git a/drivers/gpu/drm/ttm/ttm_tt.c b/drivers/gpu/drm/ttm/ttm_tt.c index df0aa6c4b8b8..698cd4bf5e46 100644 --- a/drivers/gpu/drm/ttm/ttm_tt.c +++ b/drivers/gpu/drm/ttm/ttm_tt.c @@ -544,7 +544,7 @@ EXPORT_SYMBOL(ttm_tt_pages_limit); */ int ttm_tt_setup_backup(struct ttm_tt *tt) { - struct ttm_backup *backup = + struct file *backup = ttm_backup_shmem_create(((loff_t)tt->num_pages) << PAGE_SHIFT); if (WARN_ON_ONCE(!(tt->page_flags & TTM_TT_FLAG_EXTERNAL_MAPPABLE))) diff --git a/include/drm/ttm/ttm_backup.h b/include/drm/ttm/ttm_backup.h index 574b932177cc..c33cba111171 100644 --- a/include/drm/ttm/ttm_backup.h +++ b/include/drm/ttm/ttm_backup.h @@ -9,8 +9,6 @@ #include #include -struct ttm_backup; - /** * ttm_backup_handle_to_page_ptr() - Convert handle to struct page pointer * @handle: The handle to convert. @@ -55,20 +53,20 @@ ttm_backup_page_ptr_to_handle(const struct page *page) return (unsigned long)page >> 1; } -void ttm_backup_drop(struct ttm_backup *backup, pgoff_t handle); +void ttm_backup_drop(struct file *backup, pgoff_t handle); -int ttm_backup_copy_page(struct ttm_backup *backup, struct page *dst, +int ttm_backup_copy_page(struct file *backup, struct page *dst, pgoff_t handle, bool intr); s64 -ttm_backup_backup_page(struct ttm_backup *backup, struct page *page, +ttm_backup_backup_page(struct file *backup, struct page *page, bool writeback, pgoff_t idx, gfp_t page_gfp, gfp_t alloc_gfp); -void ttm_backup_fini(struct ttm_backup *backup); +void ttm_backup_fini(struct file *backup); u64 ttm_backup_bytes_avail(void); -struct ttm_backup *ttm_backup_shmem_create(loff_t size); +struct file *ttm_backup_shmem_create(loff_t size); #endif diff --git a/include/drm/ttm/ttm_tt.h b/include/drm/ttm/ttm_tt.h index 13cf47f3322f..406437ad674b 100644 --- a/include/drm/ttm/ttm_tt.h +++ b/include/drm/ttm/ttm_tt.h @@ -118,7 +118,7 @@ struct ttm_tt { * ttm_tt_create() callback is responsible for assigning * this field. */ - struct ttm_backup *backup; + struct file *backup; /** * @caching: The current caching state of the pages, see enum * ttm_caching. From 363cd2b81cfdf706bbfc9ec78db000c9b1ecc552 Mon Sep 17 00:00:00 2001 From: Yeoreum Yun Date: Fri, 2 May 2025 19:04:12 +0100 Subject: [PATCH 184/555] arm64: cpufeature: Move arm64_use_ng_mappings to the .data section to prevent wrong idmap generation The PTE_MAYBE_NG macro sets the nG page table bit according to the value of "arm64_use_ng_mappings". This variable is currently placed in the .bss section. create_init_idmap() is called before the .bss section initialisation which is done in early_map_kernel(). Therefore, data/test_prot in create_init_idmap() could be set incorrectly through the PAGE_KERNEL -> PROT_DEFAULT -> PTE_MAYBE_NG macros. # llvm-objdump-21 --syms vmlinux-gcc | grep arm64_use_ng_mappings ffff800082f242a8 g O .bss 0000000000000001 arm64_use_ng_mappings The create_init_idmap() function disassembly compiled with llvm-21: // create_init_idmap() ffff80008255c058: d10103ff sub sp, sp, #0x40 ffff80008255c05c: a9017bfd stp x29, x30, [sp, #0x10] ffff80008255c060: a90257f6 stp x22, x21, [sp, #0x20] ffff80008255c064: a9034ff4 stp x20, x19, [sp, #0x30] ffff80008255c068: 910043fd add x29, sp, #0x10 ffff80008255c06c: 90003fc8 adrp x8, 0xffff800082d54000 ffff80008255c070: d280e06a mov x10, #0x703 // =1795 ffff80008255c074: 91400409 add x9, x0, #0x1, lsl #12 // =0x1000 ffff80008255c078: 394a4108 ldrb w8, [x8, #0x290] ------------- (1) ffff80008255c07c: f2e00d0a movk x10, #0x68, lsl #48 ffff80008255c080: f90007e9 str x9, [sp, #0x8] ffff80008255c084: aa0103f3 mov x19, x1 ffff80008255c088: aa0003f4 mov x20, x0 ffff80008255c08c: 14000000 b 0xffff80008255c08c <__pi_create_init_idmap+0x34> ffff80008255c090: aa082d56 orr x22, x10, x8, lsl #11 -------- (2) Note (1) is loading the arm64_use_ng_mappings value in w8 and (2) is set the text or data prot with the w8 value to set PTE_NG bit. If the .bss section isn't initialized, x8 could include a garbage value and generate an incorrect mapping. Annotate arm64_use_ng_mappings as __read_mostly so that it is placed in the .data section. Fixes: 84b04d3e6bdb ("arm64: kernel: Create initial ID map from C code") Cc: stable@vger.kernel.org # 6.9.x Tested-by: Nathan Chancellor Signed-off-by: Yeoreum Yun Link: https://lore.kernel.org/r/20250502180412.3774883-1-yeoreum.yun@arm.com [catalin.marinas@arm.com: use __read_mostly instead of __ro_after_init] [catalin.marinas@arm.com: slight tweaking of the code comment] Signed-off-by: Catalin Marinas --- arch/arm64/kernel/cpufeature.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/arch/arm64/kernel/cpufeature.c b/arch/arm64/kernel/cpufeature.c index 9c4d6d552b25..4c46d80aa64b 100644 --- a/arch/arm64/kernel/cpufeature.c +++ b/arch/arm64/kernel/cpufeature.c @@ -114,7 +114,14 @@ static struct arm64_cpu_capabilities const __ro_after_init *cpucap_ptrs[ARM64_NC DECLARE_BITMAP(boot_cpucaps, ARM64_NCAPS); -bool arm64_use_ng_mappings = false; +/* + * arm64_use_ng_mappings must be placed in the .data section, otherwise it + * ends up in the .bss section where it is initialized in early_map_kernel() + * after the MMU (with the idmap) was enabled. create_init_idmap() - which + * runs before early_map_kernel() and reads the variable via PTE_MAYBE_NG - + * may end up generating an incorrect idmap page table attributes. + */ +bool arm64_use_ng_mappings __read_mostly = false; EXPORT_SYMBOL(arm64_use_ng_mappings); DEFINE_PER_CPU_READ_MOSTLY(const char *, this_cpu_vector) = vectors; From 4db6c75124d871fbabf8243f947d34cc7e0697fc Mon Sep 17 00:00:00 2001 From: Daniel Golle Date: Mon, 5 May 2025 02:07:32 +0100 Subject: [PATCH 185/555] net: ethernet: mtk_eth_soc: reset all TX queues on DMA free The purpose of resetting the TX queue is to reset the byte and packet count as well as to clear the software flow control XOFF bit. MediaTek developers pointed out that netdev_reset_queue would only resets queue 0 of the network device. Queues that are not reset may cause unexpected issues. Packets may stop being sent after reset and "transmit timeout" log may be displayed. Import fix from MediaTek's SDK to resolve this issue. Link: https://git01.mediatek.com/plugins/gitiles/openwrt/feeds/mtk-openwrt-feeds/+/319c0d9905579a46dc448579f892f364f1f84818 Fixes: f63959c7eec31 ("net: ethernet: mtk_eth_soc: implement multi-queue support for per-port queues") Signed-off-by: Daniel Golle Link: https://patch.msgid.link/c9ff9adceac4f152239a0f65c397f13547639175.1746406763.git.daniel@makrotopia.org Signed-off-by: Paolo Abeni --- drivers/net/ethernet/mediatek/mtk_eth_soc.c | 16 ++++++++++++---- 1 file changed, 12 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index 8fda4ce80d81..53c39561b6d9 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -3186,11 +3186,19 @@ static int mtk_dma_init(struct mtk_eth *eth) static void mtk_dma_free(struct mtk_eth *eth) { const struct mtk_soc_data *soc = eth->soc; - int i; + int i, j, txqs = 1; + + if (MTK_HAS_CAPS(eth->soc->caps, MTK_QDMA)) + txqs = MTK_QDMA_NUM_QUEUES; + + for (i = 0; i < MTK_MAX_DEVS; i++) { + if (!eth->netdev[i]) + continue; + + for (j = 0; j < txqs; j++) + netdev_tx_reset_subqueue(eth->netdev[i], j); + } - for (i = 0; i < MTK_MAX_DEVS; i++) - if (eth->netdev[i]) - netdev_reset_queue(eth->netdev[i]); if (!MTK_HAS_CAPS(soc->caps, MTK_SRAM) && eth->scratch_ring) { dma_free_coherent(eth->dma_dev, MTK_QDMA_RING_SIZE * soc->tx.desc_size, From e8716b5b0dff1b3d523b4a83fd5e94d57b887c5c Mon Sep 17 00:00:00 2001 From: Frank Wunderlich Date: Mon, 5 May 2025 02:07:58 +0100 Subject: [PATCH 186/555] net: ethernet: mtk_eth_soc: do not reset PSE when setting FE Remove redundant PSE reset. When setting FE register there is no need to reset PSE, doing so may cause FE to work abnormal. Link: https://git01.mediatek.com/plugins/gitiles/openwrt/feeds/mtk-openwrt-feeds/+/3a5223473e086a4b54a2b9a44df7d9ddcc2bc75a Fixes: dee4dd10c79aa ("net: ethernet: mtk_eth_soc: ppe: add support for multiple PPEs") Signed-off-by: Frank Wunderlich Link: https://patch.msgid.link/18f0ac7d83f82defa3342c11ef0d1362f6b81e88.1746406763.git.daniel@makrotopia.org Signed-off-by: Paolo Abeni --- drivers/net/ethernet/mediatek/mtk_eth_soc.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index 53c39561b6d9..22a532695fb0 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -3473,9 +3473,6 @@ static int mtk_open(struct net_device *dev) } mtk_gdm_config(eth, target_mac->id, gdm_config); } - /* Reset and enable PSE */ - mtk_w32(eth, RST_GL_PSE, MTK_RST_GL); - mtk_w32(eth, 0, MTK_RST_GL); napi_enable(ð->tx_napi); napi_enable(ð->rx_napi); From 7c6fa1797a725732981f2d77711c867166737719 Mon Sep 17 00:00:00 2001 From: Kevin Baker Date: Mon, 5 May 2025 12:02:56 -0500 Subject: [PATCH 187/555] drm/panel: simple: Update timings for AUO G101EVN010 Switch to panel timings based on datasheet for the AUO G101EVN01.0 LVDS panel. Default timings were tested on the panel. Previous mode-based timings resulted in horizontal display shift. Signed-off-by: Kevin Baker Fixes: 4fb86404a977 ("drm/panel: simple: Add AUO G101EVN010 panel support") Reviewed-by: Neil Armstrong Link: https://lore.kernel.org/r/20250505170256.1385113-1-kevinb@ventureresearch.com Signed-off-by: Neil Armstrong Link: https://lore.kernel.org/r/20250505170256.1385113-1-kevinb@ventureresearch.com --- drivers/gpu/drm/panel/panel-simple.c | 25 +++++++++++++------------ 1 file changed, 13 insertions(+), 12 deletions(-) diff --git a/drivers/gpu/drm/panel/panel-simple.c b/drivers/gpu/drm/panel/panel-simple.c index 232b03c1a259..33a37539de57 100644 --- a/drivers/gpu/drm/panel/panel-simple.c +++ b/drivers/gpu/drm/panel/panel-simple.c @@ -1027,27 +1027,28 @@ static const struct panel_desc auo_g070vvn01 = { }, }; -static const struct drm_display_mode auo_g101evn010_mode = { - .clock = 68930, - .hdisplay = 1280, - .hsync_start = 1280 + 82, - .hsync_end = 1280 + 82 + 2, - .htotal = 1280 + 82 + 2 + 84, - .vdisplay = 800, - .vsync_start = 800 + 8, - .vsync_end = 800 + 8 + 2, - .vtotal = 800 + 8 + 2 + 6, +static const struct display_timing auo_g101evn010_timing = { + .pixelclock = { 64000000, 68930000, 85000000 }, + .hactive = { 1280, 1280, 1280 }, + .hfront_porch = { 8, 64, 256 }, + .hback_porch = { 8, 64, 256 }, + .hsync_len = { 40, 168, 767 }, + .vactive = { 800, 800, 800 }, + .vfront_porch = { 4, 8, 100 }, + .vback_porch = { 4, 8, 100 }, + .vsync_len = { 8, 16, 223 }, }; static const struct panel_desc auo_g101evn010 = { - .modes = &auo_g101evn010_mode, - .num_modes = 1, + .timings = &auo_g101evn010_timing, + .num_timings = 1, .bpc = 6, .size = { .width = 216, .height = 135, }, .bus_format = MEDIA_BUS_FMT_RGB666_1X7X3_SPWG, + .bus_flags = DRM_BUS_FLAG_DE_HIGH, .connector_type = DRM_MODE_CONNECTOR_LVDS, }; From 0ca6df4f40cf4c32487944aaf48319cb6c25accc Mon Sep 17 00:00:00 2001 From: Norbert Szetei Date: Fri, 2 May 2025 08:21:58 +0900 Subject: [PATCH 188/555] ksmbd: prevent out-of-bounds stream writes by validating *pos ksmbd_vfs_stream_write() did not validate whether the write offset (*pos) was within the bounds of the existing stream data length (v_len). If *pos was greater than or equal to v_len, this could lead to an out-of-bounds memory write. This patch adds a check to ensure *pos is less than v_len before proceeding. If the condition fails, -EINVAL is returned. Cc: stable@vger.kernel.org Signed-off-by: Norbert Szetei Acked-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/server/vfs.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/fs/smb/server/vfs.c b/fs/smb/server/vfs.c index 391d07da586c..482eba0f4dc1 100644 --- a/fs/smb/server/vfs.c +++ b/fs/smb/server/vfs.c @@ -426,6 +426,13 @@ static int ksmbd_vfs_stream_write(struct ksmbd_file *fp, char *buf, loff_t *pos, goto out; } + if (v_len <= *pos) { + pr_err("stream write position %lld is out of bounds (stream length: %zd)\n", + *pos, v_len); + err = -EINVAL; + goto out; + } + if (v_len < size) { wbuf = kvzalloc(size, KSMBD_DEFAULT_GFP); if (!wbuf) { From 36991c1ccde2d5a521577c448ffe07fcccfe104d Mon Sep 17 00:00:00 2001 From: Sean Heelan Date: Tue, 6 May 2025 22:04:52 +0900 Subject: [PATCH 189/555] ksmbd: Fix UAF in __close_file_table_ids A use-after-free is possible if one thread destroys the file via __ksmbd_close_fd while another thread holds a reference to it. The existing checks on fp->refcount are not sufficient to prevent this. The fix takes ft->lock around the section which removes the file from the file table. This prevents two threads acquiring the same file pointer via __close_file_table_ids, as well as the other functions which retrieve a file from the IDR and which already use this same lock. Cc: stable@vger.kernel.org Signed-off-by: Sean Heelan Acked-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/server/vfs_cache.c | 33 ++++++++++++++++++++++++++------- 1 file changed, 26 insertions(+), 7 deletions(-) diff --git a/fs/smb/server/vfs_cache.c b/fs/smb/server/vfs_cache.c index 1f8fa3468173..dfed6fce8904 100644 --- a/fs/smb/server/vfs_cache.c +++ b/fs/smb/server/vfs_cache.c @@ -661,21 +661,40 @@ __close_file_table_ids(struct ksmbd_file_table *ft, bool (*skip)(struct ksmbd_tree_connect *tcon, struct ksmbd_file *fp)) { - unsigned int id; - struct ksmbd_file *fp; - int num = 0; + struct ksmbd_file *fp; + unsigned int id = 0; + int num = 0; - idr_for_each_entry(ft->idr, fp, id) { - if (skip(tcon, fp)) + while (1) { + write_lock(&ft->lock); + fp = idr_get_next(ft->idr, &id); + if (!fp) { + write_unlock(&ft->lock); + break; + } + + if (skip(tcon, fp) || + !atomic_dec_and_test(&fp->refcount)) { + id++; + write_unlock(&ft->lock); continue; + } set_close_state_blocked_works(fp); + idr_remove(ft->idr, fp->volatile_id); + fp->volatile_id = KSMBD_NO_FID; + write_unlock(&ft->lock); + + down_write(&fp->f_ci->m_lock); + list_del_init(&fp->node); + up_write(&fp->f_ci->m_lock); - if (!atomic_dec_and_test(&fp->refcount)) - continue; __ksmbd_close_fd(ft, fp); + num++; + id++; } + return num; } From dcaeeb8ae84c5506ebc574732838264f3887738c Mon Sep 17 00:00:00 2001 From: Antonios Salios Date: Fri, 25 Apr 2025 13:17:45 +0200 Subject: [PATCH 190/555] can: m_can: m_can_class_allocate_dev(): initialize spin lock on device probe The spin lock tx_handling_spinlock in struct m_can_classdev is not being initialized. This leads the following spinlock bad magic complaint from the kernel, eg. when trying to send CAN frames with cansend from can-utils: | BUG: spinlock bad magic on CPU#0, cansend/95 | lock: 0xff60000002ec1010, .magic: 00000000, .owner: /-1, .owner_cpu: 0 | CPU: 0 UID: 0 PID: 95 Comm: cansend Not tainted 6.15.0-rc3-00032-ga79be02bba5c #5 NONE | Hardware name: MachineWare SIM-V (DT) | Call Trace: | [] dump_backtrace+0x1c/0x24 | [] show_stack+0x28/0x34 | [] dump_stack_lvl+0x4a/0x68 | [] dump_stack+0x14/0x1c | [] spin_dump+0x62/0x6e | [] do_raw_spin_lock+0xd0/0x142 | [] _raw_spin_lock_irqsave+0x20/0x2c | [] m_can_start_xmit+0x90/0x34a | [] dev_hard_start_xmit+0xa6/0xee | [] sch_direct_xmit+0x114/0x292 | [] __dev_queue_xmit+0x3b0/0xaa8 | [] can_send+0xc6/0x242 | [] raw_sendmsg+0x1a8/0x36c | [] sock_write_iter+0x9a/0xee | [] vfs_write+0x184/0x3a6 | [] ksys_write+0xa0/0xc0 | [] __riscv_sys_write+0x14/0x1c | [] do_trap_ecall_u+0x168/0x212 | [] handle_exception+0x146/0x152 Initializing the spin lock in m_can_class_allocate_dev solves that problem. Fixes: 1fa80e23c150 ("can: m_can: Introduce a tx_fifo_in_flight counter") Signed-off-by: Antonios Salios Reviewed-by: Vincent Mailhol Link: https://patch.msgid.link/20250425111744.37604-2-antonios@mwa.re Reviewed-by: Markus Schneider-Pargmann Signed-off-by: Marc Kleine-Budde --- drivers/net/can/m_can/m_can.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/can/m_can/m_can.c b/drivers/net/can/m_can/m_can.c index 884a6352c42b..326ede9d400f 100644 --- a/drivers/net/can/m_can/m_can.c +++ b/drivers/net/can/m_can/m_can.c @@ -2379,6 +2379,7 @@ struct m_can_classdev *m_can_class_allocate_dev(struct device *dev, SET_NETDEV_DEV(net_dev, dev); m_can_of_parse_mram(class_dev, mram_config_vals); + spin_lock_init(&class_dev->tx_handling_spinlock); out: return class_dev; } From 5e1663810e11c64956aa7e280cf74b2f3284d816 Mon Sep 17 00:00:00 2001 From: Kelsey Maes Date: Wed, 30 Apr 2025 09:15:01 -0700 Subject: [PATCH 191/555] can: mcp251xfd: fix TDC setting for low data bit rates The TDC is currently hardcoded enabled. This means that even for lower CAN-FD data bitrates (with a DBRP (data bitrate prescaler) > 2) a TDC is configured. This leads to a bus-off condition. ISO 11898-1 section 11.3.3 says "Transmitter delay compensation" (TDC) is only applicable if DBRP is 1 or 2. To fix the problem, switch the driver to use the TDC calculation provided by the CAN driver framework (which respects ISO 11898-1 section 11.3.3). This has the positive side effect that userspace can control TDC as needed. Demonstration of the feature in action: | $ ip link set can0 up type can bitrate 125000 dbitrate 500000 fd on | $ ip -details link show can0 | 3: can0: mtu 72 qdisc pfifo_fast state UP mode DEFAULT group default qlen 10 | link/can promiscuity 0 allmulti 0 minmtu 0 maxmtu 0 | can state ERROR-ACTIVE (berr-counter tx 0 rx 0) restart-ms 0 | bitrate 125000 sample-point 0.875 | tq 50 prop-seg 69 phase-seg1 70 phase-seg2 20 sjw 10 brp 2 | mcp251xfd: tseg1 2..256 tseg2 1..128 sjw 1..128 brp 1..256 brp_inc 1 | dbitrate 500000 dsample-point 0.875 | dtq 125 dprop-seg 6 dphase-seg1 7 dphase-seg2 2 dsjw 1 dbrp 5 | mcp251xfd: dtseg1 1..32 dtseg2 1..16 dsjw 1..16 dbrp 1..256 dbrp_inc 1 | tdcv 0..63 tdco 0..63 | clock 40000000 numtxqueues 1 numrxqueues 1 gso_max_size 65536 gso_max_segs 65535 tso_max_size 65536 tso_max_segs 65535 gro_max_size 65536 parentbus spi parentdev spi0.0 | $ ip link set can0 up type can bitrate 1000000 dbitrate 4000000 fd on | $ ip -details link show can0 | 3: can0: mtu 72 qdisc pfifo_fast state UP mode DEFAULT group default qlen 10 | link/can promiscuity 0 allmulti 0 minmtu 0 maxmtu 0 | can state ERROR-ACTIVE (berr-counter tx 0 rx 0) restart-ms 0 | bitrate 1000000 sample-point 0.750 | tq 25 prop-seg 14 phase-seg1 15 phase-seg2 10 sjw 5 brp 1 | mcp251xfd: tseg1 2..256 tseg2 1..128 sjw 1..128 brp 1..256 brp_inc 1 | dbitrate 4000000 dsample-point 0.700 | dtq 25 dprop-seg 3 dphase-seg1 3 dphase-seg2 3 dsjw 1 dbrp 1 | tdco 7 | mcp251xfd: dtseg1 1..32 dtseg2 1..16 dsjw 1..16 dbrp 1..256 dbrp_inc 1 | tdcv 0..63 tdco 0..63 | clock 40000000 numtxqueues 1 numrxqueues 1 gso_max_size 65536 gso_max_segs 65535 tso_max_size 65536 tso_max_segs 65535 gro_max_size 65536 parentbus spi parentdev spi0.0 There has been some confusion about the MCP2518FD using a relative or absolute TDCO due to the datasheet specifying a range of [-64,63]. I have a custom board with a 40 MHz clock and an estimated loop delay of 100 to 216 ns. During testing at a data bit rate of 4 Mbit/s I found that using can_get_relative_tdco() resulted in bus-off errors. The final TDCO value was 1 which corresponds to a 10% SSP in an absolute configuration. This behavior is expected if the TDCO value is really absolute and not relative. Using priv->can.tdc.tdco instead results in a final TDCO of 8, setting the SSP at exactly 80%. This configuration works. The automatic, manual, and off TDC modes were tested at speeds up to, and including, 8 Mbit/s on real hardware and behave as expected. Fixes: 55e5b97f003e ("can: mcp25xxfd: add driver for Microchip MCP25xxFD SPI CAN") Reported-by: Kelsey Maes Closes: https://lore.kernel.org/all/C2121586-C87F-4B23-A933-845362C29CA1@vpprocess.com Reviewed-by: Vincent Mailhol Signed-off-by: Kelsey Maes Link: https://patch.msgid.link/20250430161501.79370-1-kelsey@vpprocess.com [mkl: add comment] Signed-off-by: Marc Kleine-Budde --- .../net/can/spi/mcp251xfd/mcp251xfd-core.c | 40 +++++++++++++++---- 1 file changed, 32 insertions(+), 8 deletions(-) diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c index 3bc56517fe7a..064d81c724f4 100644 --- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c +++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c @@ -75,6 +75,24 @@ static const struct can_bittiming_const mcp251xfd_data_bittiming_const = { .brp_inc = 1, }; +/* The datasheet of the mcp2518fd (DS20006027B) specifies a range of + * [-64,63] for TDCO, indicating a relative TDCO. + * + * Manual tests have shown, that using a relative TDCO configuration + * results in bus off, while an absolute configuration works. + * + * For TDCO use the max value (63) from the data sheet, but 0 as the + * minimum. + */ +static const struct can_tdc_const mcp251xfd_tdc_const = { + .tdcv_min = 0, + .tdcv_max = 63, + .tdco_min = 0, + .tdco_max = 63, + .tdcf_min = 0, + .tdcf_max = 0, +}; + static const char *__mcp251xfd_get_model_str(enum mcp251xfd_model model) { switch (model) { @@ -510,8 +528,7 @@ static int mcp251xfd_set_bittiming(const struct mcp251xfd_priv *priv) { const struct can_bittiming *bt = &priv->can.bittiming; const struct can_bittiming *dbt = &priv->can.data_bittiming; - u32 val = 0; - s8 tdco; + u32 tdcmod, val = 0; int err; /* CAN Control Register @@ -575,11 +592,16 @@ static int mcp251xfd_set_bittiming(const struct mcp251xfd_priv *priv) return err; /* Transmitter Delay Compensation */ - tdco = clamp_t(int, dbt->brp * (dbt->prop_seg + dbt->phase_seg1), - -64, 63); - val = FIELD_PREP(MCP251XFD_REG_TDC_TDCMOD_MASK, - MCP251XFD_REG_TDC_TDCMOD_AUTO) | - FIELD_PREP(MCP251XFD_REG_TDC_TDCO_MASK, tdco); + if (priv->can.ctrlmode & CAN_CTRLMODE_TDC_AUTO) + tdcmod = MCP251XFD_REG_TDC_TDCMOD_AUTO; + else if (priv->can.ctrlmode & CAN_CTRLMODE_TDC_MANUAL) + tdcmod = MCP251XFD_REG_TDC_TDCMOD_MANUAL; + else + tdcmod = MCP251XFD_REG_TDC_TDCMOD_DISABLED; + + val = FIELD_PREP(MCP251XFD_REG_TDC_TDCMOD_MASK, tdcmod) | + FIELD_PREP(MCP251XFD_REG_TDC_TDCV_MASK, priv->can.tdc.tdcv) | + FIELD_PREP(MCP251XFD_REG_TDC_TDCO_MASK, priv->can.tdc.tdco); return regmap_write(priv->map_reg, MCP251XFD_REG_TDC, val); } @@ -2083,10 +2105,12 @@ static int mcp251xfd_probe(struct spi_device *spi) priv->can.do_get_berr_counter = mcp251xfd_get_berr_counter; priv->can.bittiming_const = &mcp251xfd_bittiming_const; priv->can.data_bittiming_const = &mcp251xfd_data_bittiming_const; + priv->can.tdc_const = &mcp251xfd_tdc_const; priv->can.ctrlmode_supported = CAN_CTRLMODE_LOOPBACK | CAN_CTRLMODE_LISTENONLY | CAN_CTRLMODE_BERR_REPORTING | CAN_CTRLMODE_FD | CAN_CTRLMODE_FD_NON_ISO | - CAN_CTRLMODE_CC_LEN8_DLC; + CAN_CTRLMODE_CC_LEN8_DLC | CAN_CTRLMODE_TDC_AUTO | + CAN_CTRLMODE_TDC_MANUAL; set_bit(MCP251XFD_FLAGS_DOWN, priv->flags); priv->ndev = ndev; priv->spi = spi; From 84f5eb833f53ae192baed4cfb8d9eaab43481fc9 Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Fri, 2 May 2025 16:13:44 +0200 Subject: [PATCH 192/555] can: mcp251xfd: mcp251xfd_remove(): fix order of unregistration calls If a driver is removed, the driver framework invokes the driver's remove callback. A CAN driver's remove function calls unregister_candev(), which calls net_device_ops::ndo_stop further down in the call stack for interfaces which are in the "up" state. With the mcp251xfd driver the removal of the module causes the following warning: | WARNING: CPU: 0 PID: 352 at net/core/dev.c:7342 __netif_napi_del_locked+0xc8/0xd8 as can_rx_offload_del() deletes the NAPI, while it is still active, because the interface is still up. To fix the warning, first unregister the network interface, which calls net_device_ops::ndo_stop, which disables the NAPI, and then call can_rx_offload_del(). Fixes: 55e5b97f003e ("can: mcp25xxfd: add driver for Microchip MCP25xxFD SPI CAN") Cc: stable@vger.kernel.org Link: https://patch.msgid.link/20250502-can-rx-offload-del-v1-1-59a9b131589d@pengutronix.de Signed-off-by: Marc Kleine-Budde --- drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c index 064d81c724f4..c30b04f8fc0d 100644 --- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c +++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c @@ -2198,8 +2198,8 @@ static void mcp251xfd_remove(struct spi_device *spi) struct mcp251xfd_priv *priv = spi_get_drvdata(spi); struct net_device *ndev = priv->ndev; - can_rx_offload_del(&priv->offload); mcp251xfd_unregister(priv); + can_rx_offload_del(&priv->offload); spi->max_speed_hz = priv->spi_max_speed_hz_orig; free_candev(ndev); } From db492e24f9b05547ba12b4783f09c9d943cf42fe Mon Sep 17 00:00:00 2001 From: Johannes Thumshirn Date: Tue, 6 May 2025 13:27:30 +0200 Subject: [PATCH 193/555] block: only update request sector if needed In case of a ZONE APPEND write, regardless of native ZONE APPEND or the emulation layer in the zone write plugging code, the sector the data got written to by the device needs to be updated in the bio. At the moment, this is done for every native ZONE APPEND write and every request that is flagged with 'BIO_ZONE_WRITE_PLUGGING'. But thus superfluously updates the sector for regular writes to a zoned block device. Check if a bio is a native ZONE APPEND write or if the bio is flagged as 'BIO_EMULATES_ZONE_APPEND', meaning the block layer's zone write plugging code handles the ZONE APPEND and translates it into a regular write and back. Only if one of these two criterion is met, update the sector in the bio upon completion. Signed-off-by: Johannes Thumshirn Reviewed-by: Christoph Hellwig Link: https://lore.kernel.org/r/dea089581cb6b777c1cd1500b38ac0b61df4b2d1.1746530748.git.jth@kernel.org Signed-off-by: Jens Axboe --- block/blk.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/block/blk.h b/block/blk.h index 328075787814..594eeba7b949 100644 --- a/block/blk.h +++ b/block/blk.h @@ -480,7 +480,8 @@ static inline void blk_zone_update_request_bio(struct request *rq, * the original BIO sector so that blk_zone_write_plug_bio_endio() can * lookup the zone write plug. */ - if (req_op(rq) == REQ_OP_ZONE_APPEND || bio_zone_write_plugging(bio)) + if (req_op(rq) == REQ_OP_ZONE_APPEND || + bio_flagged(bio, BIO_EMULATES_ZONE_APPEND)) bio->bi_iter.bi_sector = rq->__sector; } void blk_zone_write_plug_bio_endio(struct bio *bio); From 037ada7a3181300218e4fd78bef6a741cfa7f808 Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Fri, 2 May 2025 16:13:45 +0200 Subject: [PATCH 194/555] can: rockchip_canfd: rkcanfd_remove(): fix order of unregistration calls If a driver is removed, the driver framework invokes the driver's remove callback. A CAN driver's remove function calls unregister_candev(), which calls net_device_ops::ndo_stop further down in the call stack for interfaces which are in the "up" state. The removal of the module causes a warning, as can_rx_offload_del() deletes the NAPI, while it is still active, because the interface is still up. To fix the warning, first unregister the network interface, which calls net_device_ops::ndo_stop, which disables the NAPI, and then call can_rx_offload_del(). Fixes: ff60bfbaf67f ("can: rockchip_canfd: add driver for Rockchip CAN-FD controller") Cc: stable@vger.kernel.org Link: https://patch.msgid.link/20250502-can-rx-offload-del-v1-2-59a9b131589d@pengutronix.de Reviewed-by: Markus Schneider-Pargmann Signed-off-by: Marc Kleine-Budde --- drivers/net/can/rockchip/rockchip_canfd-core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/can/rockchip/rockchip_canfd-core.c b/drivers/net/can/rockchip/rockchip_canfd-core.c index 7107a37da36c..c3fb3176ce42 100644 --- a/drivers/net/can/rockchip/rockchip_canfd-core.c +++ b/drivers/net/can/rockchip/rockchip_canfd-core.c @@ -937,8 +937,8 @@ static void rkcanfd_remove(struct platform_device *pdev) struct rkcanfd_priv *priv = platform_get_drvdata(pdev); struct net_device *ndev = priv->ndev; - can_rx_offload_del(&priv->offload); rkcanfd_unregister(priv); + can_rx_offload_del(&priv->offload); free_candev(ndev); } From 0713a1b3276b98c7dafbeefef00d7bc3a9119a84 Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Fri, 2 May 2025 16:13:46 +0200 Subject: [PATCH 195/555] can: mcan: m_can_class_unregister(): fix order of unregistration calls If a driver is removed, the driver framework invokes the driver's remove callback. A CAN driver's remove function calls unregister_candev(), which calls net_device_ops::ndo_stop further down in the call stack for interfaces which are in the "up" state. The removal of the module causes a warning, as can_rx_offload_del() deletes the NAPI, while it is still active, because the interface is still up. To fix the warning, first unregister the network interface, which calls net_device_ops::ndo_stop, which disables the NAPI, and then call can_rx_offload_del(). Fixes: 1be37d3b0414 ("can: m_can: fix periph RX path: use rx-offload to ensure skbs are sent from softirq context") Cc: stable@vger.kernel.org Link: https://patch.msgid.link/20250502-can-rx-offload-del-v1-3-59a9b131589d@pengutronix.de Reviewed-by: Markus Schneider-Pargmann Signed-off-by: Marc Kleine-Budde --- drivers/net/can/m_can/m_can.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/can/m_can/m_can.c b/drivers/net/can/m_can/m_can.c index 326ede9d400f..c2c116ce1087 100644 --- a/drivers/net/can/m_can/m_can.c +++ b/drivers/net/can/m_can/m_can.c @@ -2463,9 +2463,9 @@ EXPORT_SYMBOL_GPL(m_can_class_register); void m_can_class_unregister(struct m_can_classdev *cdev) { + unregister_candev(cdev->net); if (cdev->is_peripheral) can_rx_offload_del(&cdev->offload); - unregister_candev(cdev->net); } EXPORT_SYMBOL_GPL(m_can_class_unregister); From 511e64e13d8cc72853275832e3f372607466c18c Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Tue, 29 Apr 2025 09:05:55 +0200 Subject: [PATCH 196/555] can: gw: fix RCU/BH usage in cgw_create_job() As reported by Sebastian Andrzej Siewior the use of local_bh_disable() is only feasible in uni processor systems to update the modification rules. The usual use-case to update the modification rules is to update the data of the modifications but not the modification types (AND/OR/XOR/SET) or the checksum functions itself. To omit additional memory allocations to maintain fast modification switching times, the modification description space is doubled at gw-job creation time so that only the reference to the active modification description is changed under rcu protection. Rename cgw_job::mod to cf_mod and make it a RCU pointer. Allocate in cgw_create_job() and free it together with cgw_job in cgw_job_free_rcu(). Update all users to dereference cgw_job::cf_mod with a RCU accessor and if possible once. [bigeasy: Replace mod1/mod2 from the Oliver's original patch with dynamic allocation, use RCU annotation and accessor] Reported-by: Sebastian Andrzej Siewior Closes: https://lore.kernel.org/linux-can/20231031112349.y0aLoBrz@linutronix.de/ Fixes: dd895d7f21b2 ("can: cangw: introduce optional uid to reference created routing jobs") Tested-by: Oliver Hartkopp Signed-off-by: Oliver Hartkopp Signed-off-by: Sebastian Andrzej Siewior Link: https://patch.msgid.link/20250429070555.cs-7b_eZ@linutronix.de Signed-off-by: Marc Kleine-Budde --- net/can/gw.c | 151 +++++++++++++++++++++++++++++++-------------------- 1 file changed, 91 insertions(+), 60 deletions(-) diff --git a/net/can/gw.c b/net/can/gw.c index ef93293c1fae..55eccb1c7620 100644 --- a/net/can/gw.c +++ b/net/can/gw.c @@ -130,7 +130,7 @@ struct cgw_job { u32 handled_frames; u32 dropped_frames; u32 deleted_frames; - struct cf_mod mod; + struct cf_mod __rcu *cf_mod; union { /* CAN frame data source */ struct net_device *dev; @@ -459,6 +459,7 @@ static void can_can_gw_rcv(struct sk_buff *skb, void *data) struct cgw_job *gwj = (struct cgw_job *)data; struct canfd_frame *cf; struct sk_buff *nskb; + struct cf_mod *mod; int modidx = 0; /* process strictly Classic CAN or CAN FD frames */ @@ -506,7 +507,8 @@ static void can_can_gw_rcv(struct sk_buff *skb, void *data) * When there is at least one modification function activated, * we need to copy the skb as we want to modify skb->data. */ - if (gwj->mod.modfunc[0]) + mod = rcu_dereference(gwj->cf_mod); + if (mod->modfunc[0]) nskb = skb_copy(skb, GFP_ATOMIC); else nskb = skb_clone(skb, GFP_ATOMIC); @@ -529,8 +531,8 @@ static void can_can_gw_rcv(struct sk_buff *skb, void *data) cf = (struct canfd_frame *)nskb->data; /* perform preprocessed modification functions if there are any */ - while (modidx < MAX_MODFUNCTIONS && gwj->mod.modfunc[modidx]) - (*gwj->mod.modfunc[modidx++])(cf, &gwj->mod); + while (modidx < MAX_MODFUNCTIONS && mod->modfunc[modidx]) + (*mod->modfunc[modidx++])(cf, mod); /* Has the CAN frame been modified? */ if (modidx) { @@ -546,11 +548,11 @@ static void can_can_gw_rcv(struct sk_buff *skb, void *data) } /* check for checksum updates */ - if (gwj->mod.csumfunc.crc8) - (*gwj->mod.csumfunc.crc8)(cf, &gwj->mod.csum.crc8); + if (mod->csumfunc.crc8) + (*mod->csumfunc.crc8)(cf, &mod->csum.crc8); - if (gwj->mod.csumfunc.xor) - (*gwj->mod.csumfunc.xor)(cf, &gwj->mod.csum.xor); + if (mod->csumfunc.xor) + (*mod->csumfunc.xor)(cf, &mod->csum.xor); } /* clear the skb timestamp if not configured the other way */ @@ -581,9 +583,20 @@ static void cgw_job_free_rcu(struct rcu_head *rcu_head) { struct cgw_job *gwj = container_of(rcu_head, struct cgw_job, rcu); + /* cgw_job::cf_mod is always accessed from the same cgw_job object within + * the same RCU read section. Once cgw_job is scheduled for removal, + * cf_mod can also be removed without mandating an additional grace period. + */ + kfree(rcu_access_pointer(gwj->cf_mod)); kmem_cache_free(cgw_cache, gwj); } +/* Return cgw_job::cf_mod with RTNL protected section */ +static struct cf_mod *cgw_job_cf_mod(struct cgw_job *gwj) +{ + return rcu_dereference_protected(gwj->cf_mod, rtnl_is_locked()); +} + static int cgw_notifier(struct notifier_block *nb, unsigned long msg, void *ptr) { @@ -616,6 +629,7 @@ static int cgw_put_job(struct sk_buff *skb, struct cgw_job *gwj, int type, { struct rtcanmsg *rtcan; struct nlmsghdr *nlh; + struct cf_mod *mod; nlh = nlmsg_put(skb, pid, seq, type, sizeof(*rtcan), flags); if (!nlh) @@ -650,82 +664,83 @@ static int cgw_put_job(struct sk_buff *skb, struct cgw_job *gwj, int type, goto cancel; } + mod = cgw_job_cf_mod(gwj); if (gwj->flags & CGW_FLAGS_CAN_FD) { struct cgw_fdframe_mod mb; - if (gwj->mod.modtype.and) { - memcpy(&mb.cf, &gwj->mod.modframe.and, sizeof(mb.cf)); - mb.modtype = gwj->mod.modtype.and; + if (mod->modtype.and) { + memcpy(&mb.cf, &mod->modframe.and, sizeof(mb.cf)); + mb.modtype = mod->modtype.and; if (nla_put(skb, CGW_FDMOD_AND, sizeof(mb), &mb) < 0) goto cancel; } - if (gwj->mod.modtype.or) { - memcpy(&mb.cf, &gwj->mod.modframe.or, sizeof(mb.cf)); - mb.modtype = gwj->mod.modtype.or; + if (mod->modtype.or) { + memcpy(&mb.cf, &mod->modframe.or, sizeof(mb.cf)); + mb.modtype = mod->modtype.or; if (nla_put(skb, CGW_FDMOD_OR, sizeof(mb), &mb) < 0) goto cancel; } - if (gwj->mod.modtype.xor) { - memcpy(&mb.cf, &gwj->mod.modframe.xor, sizeof(mb.cf)); - mb.modtype = gwj->mod.modtype.xor; + if (mod->modtype.xor) { + memcpy(&mb.cf, &mod->modframe.xor, sizeof(mb.cf)); + mb.modtype = mod->modtype.xor; if (nla_put(skb, CGW_FDMOD_XOR, sizeof(mb), &mb) < 0) goto cancel; } - if (gwj->mod.modtype.set) { - memcpy(&mb.cf, &gwj->mod.modframe.set, sizeof(mb.cf)); - mb.modtype = gwj->mod.modtype.set; + if (mod->modtype.set) { + memcpy(&mb.cf, &mod->modframe.set, sizeof(mb.cf)); + mb.modtype = mod->modtype.set; if (nla_put(skb, CGW_FDMOD_SET, sizeof(mb), &mb) < 0) goto cancel; } } else { struct cgw_frame_mod mb; - if (gwj->mod.modtype.and) { - memcpy(&mb.cf, &gwj->mod.modframe.and, sizeof(mb.cf)); - mb.modtype = gwj->mod.modtype.and; + if (mod->modtype.and) { + memcpy(&mb.cf, &mod->modframe.and, sizeof(mb.cf)); + mb.modtype = mod->modtype.and; if (nla_put(skb, CGW_MOD_AND, sizeof(mb), &mb) < 0) goto cancel; } - if (gwj->mod.modtype.or) { - memcpy(&mb.cf, &gwj->mod.modframe.or, sizeof(mb.cf)); - mb.modtype = gwj->mod.modtype.or; + if (mod->modtype.or) { + memcpy(&mb.cf, &mod->modframe.or, sizeof(mb.cf)); + mb.modtype = mod->modtype.or; if (nla_put(skb, CGW_MOD_OR, sizeof(mb), &mb) < 0) goto cancel; } - if (gwj->mod.modtype.xor) { - memcpy(&mb.cf, &gwj->mod.modframe.xor, sizeof(mb.cf)); - mb.modtype = gwj->mod.modtype.xor; + if (mod->modtype.xor) { + memcpy(&mb.cf, &mod->modframe.xor, sizeof(mb.cf)); + mb.modtype = mod->modtype.xor; if (nla_put(skb, CGW_MOD_XOR, sizeof(mb), &mb) < 0) goto cancel; } - if (gwj->mod.modtype.set) { - memcpy(&mb.cf, &gwj->mod.modframe.set, sizeof(mb.cf)); - mb.modtype = gwj->mod.modtype.set; + if (mod->modtype.set) { + memcpy(&mb.cf, &mod->modframe.set, sizeof(mb.cf)); + mb.modtype = mod->modtype.set; if (nla_put(skb, CGW_MOD_SET, sizeof(mb), &mb) < 0) goto cancel; } } - if (gwj->mod.uid) { - if (nla_put_u32(skb, CGW_MOD_UID, gwj->mod.uid) < 0) + if (mod->uid) { + if (nla_put_u32(skb, CGW_MOD_UID, mod->uid) < 0) goto cancel; } - if (gwj->mod.csumfunc.crc8) { + if (mod->csumfunc.crc8) { if (nla_put(skb, CGW_CS_CRC8, CGW_CS_CRC8_LEN, - &gwj->mod.csum.crc8) < 0) + &mod->csum.crc8) < 0) goto cancel; } - if (gwj->mod.csumfunc.xor) { + if (mod->csumfunc.xor) { if (nla_put(skb, CGW_CS_XOR, CGW_CS_XOR_LEN, - &gwj->mod.csum.xor) < 0) + &mod->csum.xor) < 0) goto cancel; } @@ -1059,7 +1074,7 @@ static int cgw_create_job(struct sk_buff *skb, struct nlmsghdr *nlh, struct net *net = sock_net(skb->sk); struct rtcanmsg *r; struct cgw_job *gwj; - struct cf_mod mod; + struct cf_mod *mod; struct can_can_gw ccgw; u8 limhops = 0; int err = 0; @@ -1078,37 +1093,48 @@ static int cgw_create_job(struct sk_buff *skb, struct nlmsghdr *nlh, if (r->gwtype != CGW_TYPE_CAN_CAN) return -EINVAL; - err = cgw_parse_attr(nlh, &mod, CGW_TYPE_CAN_CAN, &ccgw, &limhops); - if (err < 0) - return err; + mod = kmalloc(sizeof(*mod), GFP_KERNEL); + if (!mod) + return -ENOMEM; - if (mod.uid) { + err = cgw_parse_attr(nlh, mod, CGW_TYPE_CAN_CAN, &ccgw, &limhops); + if (err < 0) + goto out_free_cf; + + if (mod->uid) { ASSERT_RTNL(); /* check for updating an existing job with identical uid */ hlist_for_each_entry(gwj, &net->can.cgw_list, list) { - if (gwj->mod.uid != mod.uid) + struct cf_mod *old_cf; + + old_cf = cgw_job_cf_mod(gwj); + if (old_cf->uid != mod->uid) continue; /* interfaces & filters must be identical */ - if (memcmp(&gwj->ccgw, &ccgw, sizeof(ccgw))) - return -EINVAL; + if (memcmp(&gwj->ccgw, &ccgw, sizeof(ccgw))) { + err = -EINVAL; + goto out_free_cf; + } - /* update modifications with disabled softirq & quit */ - local_bh_disable(); - memcpy(&gwj->mod, &mod, sizeof(mod)); - local_bh_enable(); + rcu_assign_pointer(gwj->cf_mod, mod); + kfree_rcu_mightsleep(old_cf); return 0; } } /* ifindex == 0 is not allowed for job creation */ - if (!ccgw.src_idx || !ccgw.dst_idx) - return -ENODEV; + if (!ccgw.src_idx || !ccgw.dst_idx) { + err = -ENODEV; + goto out_free_cf; + } gwj = kmem_cache_alloc(cgw_cache, GFP_KERNEL); - if (!gwj) - return -ENOMEM; + if (!gwj) { + err = -ENOMEM; + goto out_free_cf; + } gwj->handled_frames = 0; gwj->dropped_frames = 0; @@ -1118,7 +1144,7 @@ static int cgw_create_job(struct sk_buff *skb, struct nlmsghdr *nlh, gwj->limit_hops = limhops; /* insert already parsed information */ - memcpy(&gwj->mod, &mod, sizeof(mod)); + RCU_INIT_POINTER(gwj->cf_mod, mod); memcpy(&gwj->ccgw, &ccgw, sizeof(ccgw)); err = -ENODEV; @@ -1152,9 +1178,11 @@ static int cgw_create_job(struct sk_buff *skb, struct nlmsghdr *nlh, if (!err) hlist_add_head_rcu(&gwj->list, &net->can.cgw_list); out: - if (err) + if (err) { kmem_cache_free(cgw_cache, gwj); - +out_free_cf: + kfree(mod); + } return err; } @@ -1214,19 +1242,22 @@ static int cgw_remove_job(struct sk_buff *skb, struct nlmsghdr *nlh, /* remove only the first matching entry */ hlist_for_each_entry_safe(gwj, nx, &net->can.cgw_list, list) { + struct cf_mod *cf_mod; + if (gwj->flags != r->flags) continue; if (gwj->limit_hops != limhops) continue; + cf_mod = cgw_job_cf_mod(gwj); /* we have a match when uid is enabled and identical */ - if (gwj->mod.uid || mod.uid) { - if (gwj->mod.uid != mod.uid) + if (cf_mod->uid || mod.uid) { + if (cf_mod->uid != mod.uid) continue; } else { /* no uid => check for identical modifications */ - if (memcmp(&gwj->mod, &mod, sizeof(mod))) + if (memcmp(cf_mod, &mod, sizeof(mod))) continue; } From d90b023718a17d308d831fde36b3bb6fa3b511e0 Mon Sep 17 00:00:00 2001 From: Steve French Date: Sun, 4 May 2025 18:26:45 -0500 Subject: [PATCH 197/555] smb3 client: warn when parse contexts returns error on compounded operation Coverity noticed that the rc on smb2_parse_contexts() was not being checked in the case of compounded operations. Since we don't want to stop parsing the following compounded responses which are likely valid, we can't easily error out here, but at least print a warning message if server has a bug causing us to skip parsing the open response contexts. Addresses-Coverity: 1639191 Acked-by: Paulo Alcantara (Red Hat) Signed-off-by: Steve French --- fs/smb/client/smb2inode.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/fs/smb/client/smb2inode.c b/fs/smb/client/smb2inode.c index 57d9bfbadd97..2a3e46b8e15a 100644 --- a/fs/smb/client/smb2inode.c +++ b/fs/smb/client/smb2inode.c @@ -666,6 +666,8 @@ finished: /* smb2_parse_contexts() fills idata->fi.IndexNumber */ rc = smb2_parse_contexts(server, &rsp_iov[0], &oparms->fid->epoch, oparms->fid->lease_key, &oplock, &idata->fi, NULL); + if (rc) + cifs_dbg(VFS, "rc: %d parsing context of compound op\n", rc); } for (i = 0; i < num_cmds; i++) { From d4e89d212d401672e9cdfe825d947ee3a9fbe3f5 Mon Sep 17 00:00:00 2001 From: Daniel Sneddon Date: Mon, 5 May 2025 14:35:12 -0700 Subject: [PATCH 198/555] x86/bpf: Call branch history clearing sequence on exit Classic BPF programs have been identified as potential vectors for intra-mode Branch Target Injection (BTI) attacks. Classic BPF programs can be run by unprivileged users. They allow unprivileged code to execute inside the kernel. Attackers can use unprivileged cBPF to craft branch history in kernel mode that can influence the target of indirect branches. Introduce a branch history buffer (BHB) clearing sequence during the JIT compilation of classic BPF programs. The clearing sequence is the same as is used in previous mitigations to protect syscalls. Since eBPF programs already have their own mitigations in place, only insert the call on classic programs that aren't run by privileged users. Signed-off-by: Daniel Sneddon Signed-off-by: Pawan Gupta Signed-off-by: Dave Hansen Acked-by: Daniel Borkmann Reviewed-by: Alexandre Chartre --- arch/x86/net/bpf_jit_comp.c | 31 +++++++++++++++++++++++++++++++ 1 file changed, 31 insertions(+) diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c index 9e5fe2ba858f..6fb786c8b2aa 100644 --- a/arch/x86/net/bpf_jit_comp.c +++ b/arch/x86/net/bpf_jit_comp.c @@ -1502,6 +1502,30 @@ static void emit_priv_frame_ptr(u8 **pprog, void __percpu *priv_frame_ptr) #define PRIV_STACK_GUARD_SZ 8 #define PRIV_STACK_GUARD_VAL 0xEB9F12345678eb9fULL +static int emit_spectre_bhb_barrier(u8 **pprog, u8 *ip, + struct bpf_prog *bpf_prog) +{ + u8 *prog = *pprog; + u8 *func; + + if (cpu_feature_enabled(X86_FEATURE_CLEAR_BHB_LOOP)) { + /* The clearing sequence clobbers eax and ecx. */ + EMIT1(0x50); /* push rax */ + EMIT1(0x51); /* push rcx */ + ip += 2; + + func = (u8 *)clear_bhb_loop; + ip += x86_call_depth_emit_accounting(&prog, func, ip); + + if (emit_call(&prog, func, ip)) + return -EINVAL; + EMIT1(0x59); /* pop rcx */ + EMIT1(0x58); /* pop rax */ + } + *pprog = prog; + return 0; +} + static int do_jit(struct bpf_prog *bpf_prog, int *addrs, u8 *image, u8 *rw_image, int oldproglen, struct jit_context *ctx, bool jmp_padding) { @@ -2544,6 +2568,13 @@ emit_jmp: seen_exit = true; /* Update cleanup_addr */ ctx->cleanup_addr = proglen; + if (bpf_prog_was_classic(bpf_prog) && + !capable(CAP_SYS_ADMIN)) { + u8 *ip = image + addrs[i - 1]; + + if (emit_spectre_bhb_barrier(&prog, ip, bpf_prog)) + return -EINVAL; + } if (bpf_prog->aux->exception_boundary) { pop_callee_regs(&prog, all_callee_regs_used); pop_r12(&prog); From 9f725eec8fc0b39bdc07dcc8897283c367c1a163 Mon Sep 17 00:00:00 2001 From: Daniel Sneddon Date: Mon, 5 May 2025 14:35:12 -0700 Subject: [PATCH 199/555] x86/bpf: Add IBHF call at end of classic BPF Classic BPF programs can be run by unprivileged users, allowing unprivileged code to execute inside the kernel. Attackers can use this to craft branch history in kernel mode that can influence the target of indirect branches. BHI_DIS_S provides user-kernel isolation of branch history, but cBPF can be used to bypass this protection by crafting branch history in kernel mode. To stop intra-mode attacks via cBPF programs, Intel created a new instruction Indirect Branch History Fence (IBHF). IBHF prevents the predicted targets of subsequent indirect branches from being influenced by branch history prior to the IBHF. IBHF is only effective while BHI_DIS_S is enabled. Add the IBHF instruction to cBPF jitted code's exit path. Add the new fence when the hardware mitigation is enabled (i.e., X86_FEATURE_CLEAR_BHB_HW is set) or after the software sequence (X86_FEATURE_CLEAR_BHB_LOOP) is being used in a virtual machine. Note that X86_FEATURE_CLEAR_BHB_HW and X86_FEATURE_CLEAR_BHB_LOOP are mutually exclusive, so the JIT compiler will only emit the new fence, not the SW sequence, when X86_FEATURE_CLEAR_BHB_HW is set. Hardware that enumerates BHI_NO basically has BHI_DIS_S protections always enabled, regardless of the value of BHI_DIS_S. Since BHI_DIS_S doesn't protect against intra-mode attacks, enumerate BHI bug on BHI_NO hardware as well. Signed-off-by: Daniel Sneddon Signed-off-by: Pawan Gupta Signed-off-by: Dave Hansen Acked-by: Daniel Borkmann Reviewed-by: Alexandre Chartre --- arch/x86/kernel/cpu/common.c | 9 ++++++--- arch/x86/net/bpf_jit_comp.c | 19 +++++++++++++++++++ 2 files changed, 25 insertions(+), 3 deletions(-) diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 12126adbc3a9..5ab13d9241c0 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -1439,9 +1439,12 @@ static void __init cpu_set_bug_bits(struct cpuinfo_x86 *c) if (vulnerable_to_rfds(x86_arch_cap_msr)) setup_force_cpu_bug(X86_BUG_RFDS); - /* When virtualized, eIBRS could be hidden, assume vulnerable */ - if (!(x86_arch_cap_msr & ARCH_CAP_BHI_NO) && - !cpu_matches(cpu_vuln_whitelist, NO_BHI) && + /* + * Intel parts with eIBRS are vulnerable to BHI attacks. Parts with + * BHI_NO still need to use the BHI mitigation to prevent Intra-mode + * attacks. When virtualized, eIBRS could be hidden, assume vulnerable. + */ + if (!cpu_matches(cpu_vuln_whitelist, NO_BHI) && (boot_cpu_has(X86_FEATURE_IBRS_ENHANCED) || boot_cpu_has(X86_FEATURE_HYPERVISOR))) setup_force_cpu_bug(X86_BUG_BHI); diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c index 6fb786c8b2aa..e472572392ef 100644 --- a/arch/x86/net/bpf_jit_comp.c +++ b/arch/x86/net/bpf_jit_comp.c @@ -41,6 +41,8 @@ static u8 *emit_code(u8 *ptr, u32 bytes, unsigned int len) #define EMIT2(b1, b2) EMIT((b1) + ((b2) << 8), 2) #define EMIT3(b1, b2, b3) EMIT((b1) + ((b2) << 8) + ((b3) << 16), 3) #define EMIT4(b1, b2, b3, b4) EMIT((b1) + ((b2) << 8) + ((b3) << 16) + ((b4) << 24), 4) +#define EMIT5(b1, b2, b3, b4, b5) \ + do { EMIT1(b1); EMIT4(b2, b3, b4, b5); } while (0) #define EMIT1_off32(b1, off) \ do { EMIT1(b1); EMIT(off, 4); } while (0) @@ -1522,6 +1524,23 @@ static int emit_spectre_bhb_barrier(u8 **pprog, u8 *ip, EMIT1(0x59); /* pop rcx */ EMIT1(0x58); /* pop rax */ } + /* Insert IBHF instruction */ + if ((cpu_feature_enabled(X86_FEATURE_CLEAR_BHB_LOOP) && + cpu_feature_enabled(X86_FEATURE_HYPERVISOR)) || + (cpu_feature_enabled(X86_FEATURE_CLEAR_BHB_HW) && + IS_ENABLED(CONFIG_X86_64))) { + /* + * Add an Indirect Branch History Fence (IBHF). IBHF acts as a + * fence preventing branch history from before the fence from + * affecting indirect branches after the fence. This is + * specifically used in cBPF jitted code to prevent Intra-mode + * BHI attacks. The IBHF instruction is designed to be a NOP on + * hardware that doesn't need or support it. The REP and REX.W + * prefixes are required by the microcode, and they also ensure + * that the NOP is unlikely to be used in existing code. + */ + EMIT5(0xF3, 0x48, 0x0F, 0x1E, 0xF8); /* ibhf */ + } *pprog = prog; return 0; } From 073fdbe02c69c43fb7c0d547ec265c7747d4a646 Mon Sep 17 00:00:00 2001 From: Pawan Gupta Date: Mon, 5 May 2025 14:35:12 -0700 Subject: [PATCH 200/555] x86/bhi: Do not set BHI_DIS_S in 32-bit mode With the possibility of intra-mode BHI via cBPF, complete mitigation for BHI is to use IBHF (history fence) instruction with BHI_DIS_S set. Since this new instruction is only available in 64-bit mode, setting BHI_DIS_S in 32-bit mode is only a partial mitigation. Do not set BHI_DIS_S in 32-bit mode so as to avoid reporting misleading mitigated status. With this change IBHF won't be used in 32-bit mode, also remove the CONFIG_X86_64 check from emit_spectre_bhb_barrier(). Suggested-by: Josh Poimboeuf Signed-off-by: Pawan Gupta Signed-off-by: Dave Hansen Reviewed-by: Josh Poimboeuf Reviewed-by: Alexandre Chartre --- arch/x86/kernel/cpu/bugs.c | 6 +++--- arch/x86/net/bpf_jit_comp.c | 5 +++-- 2 files changed, 6 insertions(+), 5 deletions(-) diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 362602b705cc..f219f0f4f2d1 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -1697,11 +1697,11 @@ static void __init bhi_select_mitigation(void) return; } - /* Mitigate in hardware if supported */ - if (spec_ctrl_bhi_dis()) + if (!IS_ENABLED(CONFIG_X86_64)) return; - if (!IS_ENABLED(CONFIG_X86_64)) + /* Mitigate in hardware if supported */ + if (spec_ctrl_bhi_dis()) return; if (bhi_mitigation == BHI_MITIGATION_VMEXIT_ONLY) { diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c index e472572392ef..8a0fabb850b7 100644 --- a/arch/x86/net/bpf_jit_comp.c +++ b/arch/x86/net/bpf_jit_comp.c @@ -1527,8 +1527,7 @@ static int emit_spectre_bhb_barrier(u8 **pprog, u8 *ip, /* Insert IBHF instruction */ if ((cpu_feature_enabled(X86_FEATURE_CLEAR_BHB_LOOP) && cpu_feature_enabled(X86_FEATURE_HYPERVISOR)) || - (cpu_feature_enabled(X86_FEATURE_CLEAR_BHB_HW) && - IS_ENABLED(CONFIG_X86_64))) { + cpu_feature_enabled(X86_FEATURE_CLEAR_BHB_HW)) { /* * Add an Indirect Branch History Fence (IBHF). IBHF acts as a * fence preventing branch history from before the fence from @@ -1538,6 +1537,8 @@ static int emit_spectre_bhb_barrier(u8 **pprog, u8 *ip, * hardware that doesn't need or support it. The REP and REX.W * prefixes are required by the microcode, and they also ensure * that the NOP is unlikely to be used in existing code. + * + * IBHF is not a valid instruction in 32-bit mode. */ EMIT5(0xF3, 0x48, 0x0F, 0x1E, 0xF8); /* ibhf */ } From 42e31f0daf80d9f7bc4ab4000f2795ec3ddf5206 Mon Sep 17 00:00:00 2001 From: Oscar Salvador Date: Tue, 6 May 2025 13:10:12 +0200 Subject: [PATCH 201/555] mm,mm_init: Mark set_high_memory as __init set_high_memory() touches arch_zone_lowest_possible_pfn which is marked as __initdata, which creates a section mismatch. Since the only user of the function is free_area_init() which is also marked as __init, mark set_high_memory() as __init as well. Signed-off-by: Oscar Salvador Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202505060901.Qcs06UoB-lkp@intel.com/ Link: https://lore.kernel.org/r/20250506111012.108743-1-osalvador@suse.de Signed-off-by: Mike Rapoport (Microsoft) --- mm/mm_init.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/mm_init.c b/mm/mm_init.c index 9659689b8ace..327764ca0ee4 100644 --- a/mm/mm_init.c +++ b/mm/mm_init.c @@ -1786,7 +1786,7 @@ static bool arch_has_descending_max_zone_pfns(void) return IS_ENABLED(CONFIG_ARC) && !IS_ENABLED(CONFIG_ARC_HAS_PAE40); } -static void set_high_memory(void) +static void __init set_high_memory(void) { phys_addr_t highmem = memblock_end_of_DRAM(); From d0706bfd3ee40923c001c6827b786a309e2a8713 Mon Sep 17 00:00:00 2001 From: Zhu Yanjun Date: Tue, 6 May 2025 17:10:08 +0200 Subject: [PATCH 202/555] RDMA/core: Fix "KASAN: slab-use-after-free Read in ib_register_device" problem Call Trace: __dump_stack lib/dump_stack.c:94 [inline] dump_stack_lvl+0x116/0x1f0 lib/dump_stack.c:120 print_address_description mm/kasan/report.c:408 [inline] print_report+0xc3/0x670 mm/kasan/report.c:521 kasan_report+0xe0/0x110 mm/kasan/report.c:634 strlen+0x93/0xa0 lib/string.c:420 __fortify_strlen include/linux/fortify-string.h:268 [inline] get_kobj_path_length lib/kobject.c:118 [inline] kobject_get_path+0x3f/0x2a0 lib/kobject.c:158 kobject_uevent_env+0x289/0x1870 lib/kobject_uevent.c:545 ib_register_device drivers/infiniband/core/device.c:1472 [inline] ib_register_device+0x8cf/0xe00 drivers/infiniband/core/device.c:1393 rxe_register_device+0x275/0x320 drivers/infiniband/sw/rxe/rxe_verbs.c:1552 rxe_net_add+0x8e/0xe0 drivers/infiniband/sw/rxe/rxe_net.c:550 rxe_newlink+0x70/0x190 drivers/infiniband/sw/rxe/rxe.c:225 nldev_newlink+0x3a3/0x680 drivers/infiniband/core/nldev.c:1796 rdma_nl_rcv_msg+0x387/0x6e0 drivers/infiniband/core/netlink.c:195 rdma_nl_rcv_skb.constprop.0.isra.0+0x2e5/0x450 netlink_unicast_kernel net/netlink/af_netlink.c:1313 [inline] netlink_unicast+0x53a/0x7f0 net/netlink/af_netlink.c:1339 netlink_sendmsg+0x8d1/0xdd0 net/netlink/af_netlink.c:1883 sock_sendmsg_nosec net/socket.c:712 [inline] __sock_sendmsg net/socket.c:727 [inline] ____sys_sendmsg+0xa95/0xc70 net/socket.c:2566 ___sys_sendmsg+0x134/0x1d0 net/socket.c:2620 __sys_sendmsg+0x16d/0x220 net/socket.c:2652 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline] do_syscall_64+0xcd/0x260 arch/x86/entry/syscall_64.c:94 entry_SYSCALL_64_after_hwframe+0x77/0x7f This problem is similar to the problem that the commit 1d6a9e7449e2 ("RDMA/core: Fix use-after-free when rename device name") fixes. The root cause is: the function ib_device_rename() renames the name with lock. But in the function kobject_uevent(), this name is accessed without lock protection at the same time. The solution is to add the lock protection when this name is accessed in the function kobject_uevent(). Fixes: 779e0bf47632 ("RDMA/core: Do not indicate device ready when device enablement fails") Link: https://patch.msgid.link/r/20250506151008.75701-1-yanjun.zhu@linux.dev Reported-by: syzbot+e2ce9e275ecc70a30b72@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=e2ce9e275ecc70a30b72 Signed-off-by: Zhu Yanjun Signed-off-by: Jason Gunthorpe --- drivers/infiniband/core/device.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/core/device.c b/drivers/infiniband/core/device.c index b4e3e4beb7f4..d4263385850a 100644 --- a/drivers/infiniband/core/device.c +++ b/drivers/infiniband/core/device.c @@ -1352,6 +1352,9 @@ static void ib_device_notify_register(struct ib_device *device) down_read(&devices_rwsem); + /* Mark for userspace that device is ready */ + kobject_uevent(&device->dev.kobj, KOBJ_ADD); + ret = rdma_nl_notify_event(device, 0, RDMA_REGISTER_EVENT); if (ret) goto out; @@ -1468,10 +1471,9 @@ int ib_register_device(struct ib_device *device, const char *name, return ret; } dev_set_uevent_suppress(&device->dev, false); - /* Mark for userspace that device is ready */ - kobject_uevent(&device->dev.kobj, KOBJ_ADD); ib_device_notify_register(device); + ib_device_put(device); return 0; From c1d9dac0db168198b6f63f460665256dedad9b6e Mon Sep 17 00:00:00 2001 From: Alex Williamson Date: Fri, 2 May 2025 16:40:31 -0600 Subject: [PATCH 203/555] vfio/pci: Align huge faults to order The vfio-pci huge_fault handler doesn't make any attempt to insert a mapping containing the faulting address, it only inserts mappings if the faulting address and resulting pfn are aligned. This works in a lot of cases, particularly in conjunction with QEMU where DMA mappings linearly fault the mmap. However, there are configurations where we don't get that linear faulting and pages are faulted on-demand. The scenario reported in the bug below is such a case, where the physical address width of the CPU is greater than that of the IOMMU, resulting in a VM where guest firmware has mapped device MMIO beyond the address width of the IOMMU. In this configuration, the MMIO is faulted on demand and tracing indicates that occasionally the faults generate a VM_FAULT_OOM. Given the use case, this results in a "error: kvm run failed Bad address", killing the VM. The host is not under memory pressure in this test, therefore it's suspected that VM_FAULT_OOM is actually the result of a NULL return from __pte_offset_map_lock() in the get_locked_pte() path from insert_pfn(). This suggests a potential race inserting a pte concurrent to a pmd, and maybe indicates some deficiency in the mm layer properly handling such a case. Nevertheless, Peter noted the inconsistency of vfio-pci's huge_fault handler where our mapping granularity depends on the alignment of the faulting address relative to the order rather than aligning the faulting address to the order to more consistently insert huge mappings. This change not only uses the page tables more consistently and efficiently, but as any fault to an aligned page results in the same mapping, the race condition suspected in the VM_FAULT_OOM is avoided. Reported-by: Adolfo Closes: https://bugzilla.kernel.org/show_bug.cgi?id=220057 Fixes: 09dfc8a5f2ce ("vfio/pci: Fallback huge faults for unaligned pfn") Cc: stable@vger.kernel.org Tested-by: Adolfo Co-developed-by: Peter Xu Signed-off-by: Peter Xu Link: https://lore.kernel.org/r/20250502224035.3183451-1-alex.williamson@redhat.com Signed-off-by: Alex Williamson --- drivers/vfio/pci/vfio_pci_core.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/vfio/pci/vfio_pci_core.c b/drivers/vfio/pci/vfio_pci_core.c index 35f9046af315..6328c3a05bcd 100644 --- a/drivers/vfio/pci/vfio_pci_core.c +++ b/drivers/vfio/pci/vfio_pci_core.c @@ -1646,14 +1646,14 @@ static vm_fault_t vfio_pci_mmap_huge_fault(struct vm_fault *vmf, { struct vm_area_struct *vma = vmf->vma; struct vfio_pci_core_device *vdev = vma->vm_private_data; - unsigned long pfn, pgoff = vmf->pgoff - vma->vm_pgoff; + unsigned long addr = vmf->address & ~((PAGE_SIZE << order) - 1); + unsigned long pgoff = (addr - vma->vm_start) >> PAGE_SHIFT; + unsigned long pfn = vma_to_pfn(vma) + pgoff; vm_fault_t ret = VM_FAULT_SIGBUS; - pfn = vma_to_pfn(vma) + pgoff; - - if (order && (pfn & ((1 << order) - 1) || - vmf->address & ((PAGE_SIZE << order) - 1) || - vmf->address + (PAGE_SIZE << order) > vma->vm_end)) { + if (order && (addr < vma->vm_start || + addr + (PAGE_SIZE << order) > vma->vm_end || + pfn & ((1 << order) - 1))) { ret = VM_FAULT_FALLBACK; goto out; } From 023c1f2f0609218103cbcb48e0104b144d4a16dc Mon Sep 17 00:00:00 2001 From: Veerendranath Jakkam Date: Thu, 24 Apr 2025 18:01:42 +0530 Subject: [PATCH 204/555] wifi: cfg80211: fix out-of-bounds access during multi-link element defragmentation Currently during the multi-link element defragmentation process, the multi-link element length added to the total IEs length when calculating the length of remaining IEs after the multi-link element in cfg80211_defrag_mle(). This could lead to out-of-bounds access if the multi-link element or its corresponding fragment elements are the last elements in the IEs buffer. To address this issue, correctly calculate the remaining IEs length by deducting the multi-link element end offset from total IEs end offset. Cc: stable@vger.kernel.org Fixes: 2481b5da9c6b ("wifi: cfg80211: handle BSS data contained in ML probe responses") Signed-off-by: Veerendranath Jakkam Link: https://patch.msgid.link/20250424-fix_mle_defragmentation_oob_access-v1-1-84412a1743fa@quicinc.com Signed-off-by: Johannes Berg --- net/wireless/scan.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/wireless/scan.c b/net/wireless/scan.c index 9865f305275d..ddd3a97f6609 100644 --- a/net/wireless/scan.c +++ b/net/wireless/scan.c @@ -2681,7 +2681,7 @@ cfg80211_defrag_mle(const struct element *mle, const u8 *ie, size_t ielen, /* Required length for first defragmentation */ buf_len = mle->datalen - 1; for_each_element(elem, mle->data + mle->datalen, - ielen - sizeof(*mle) + mle->datalen) { + ie + ielen - mle->data - mle->datalen) { if (elem->id != WLAN_EID_FRAGMENT) break; From e12a42f64fc3d74872b349eedd47f90c6676b78a Mon Sep 17 00:00:00 2001 From: Michael-CY Lee Date: Mon, 5 May 2025 16:19:46 +0800 Subject: [PATCH 205/555] wifi: mac80211: fix the type of status_code for negotiated TID to Link Mapping The status code should be type of __le16. Fixes: 83e897a961b8 ("wifi: ieee80211: add definitions for negotiated TID to Link map") Fixes: 8f500fbc6c65 ("wifi: mac80211: process and save negotiated TID to Link mapping request") Signed-off-by: Michael-CY Lee Link: https://patch.msgid.link/20250505081946.3927214-1-michael-cy.lee@mediatek.com Signed-off-by: Johannes Berg --- include/linux/ieee80211.h | 2 +- net/mac80211/mlme.c | 12 ++++++------ 2 files changed, 7 insertions(+), 7 deletions(-) diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h index 508d466de1cc..457b4fba88bd 100644 --- a/include/linux/ieee80211.h +++ b/include/linux/ieee80211.h @@ -1526,7 +1526,7 @@ struct ieee80211_mgmt { struct { u8 action_code; u8 dialog_token; - u8 status_code; + __le16 status_code; u8 variable[]; } __packed ttlm_res; struct { diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index 5d1f2d6d09ad..35eaf0812c5b 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -7675,6 +7675,7 @@ ieee80211_send_neg_ttlm_res(struct ieee80211_sub_if_data *sdata, int hdr_len = offsetofend(struct ieee80211_mgmt, u.action.u.ttlm_res); int ttlm_max_len = 2 + 1 + sizeof(struct ieee80211_ttlm_elem) + 1 + 2 * 2 * IEEE80211_TTLM_NUM_TIDS; + u16 status_code; skb = dev_alloc_skb(local->tx_headroom + hdr_len + ttlm_max_len); if (!skb) @@ -7697,19 +7698,18 @@ ieee80211_send_neg_ttlm_res(struct ieee80211_sub_if_data *sdata, WARN_ON(1); fallthrough; case NEG_TTLM_RES_REJECT: - mgmt->u.action.u.ttlm_res.status_code = - WLAN_STATUS_DENIED_TID_TO_LINK_MAPPING; + status_code = WLAN_STATUS_DENIED_TID_TO_LINK_MAPPING; break; case NEG_TTLM_RES_ACCEPT: - mgmt->u.action.u.ttlm_res.status_code = WLAN_STATUS_SUCCESS; + status_code = WLAN_STATUS_SUCCESS; break; case NEG_TTLM_RES_SUGGEST_PREFERRED: - mgmt->u.action.u.ttlm_res.status_code = - WLAN_STATUS_PREF_TID_TO_LINK_MAPPING_SUGGESTED; + status_code = WLAN_STATUS_PREF_TID_TO_LINK_MAPPING_SUGGESTED; ieee80211_neg_ttlm_add_suggested_map(skb, neg_ttlm); break; } + mgmt->u.action.u.ttlm_res.status_code = cpu_to_le16(status_code); ieee80211_tx_skb(sdata, skb); } @@ -7875,7 +7875,7 @@ void ieee80211_process_neg_ttlm_res(struct ieee80211_sub_if_data *sdata, * This can be better implemented in the future, to handle request * rejections. */ - if (mgmt->u.action.u.ttlm_res.status_code != WLAN_STATUS_SUCCESS) + if (le16_to_cpu(mgmt->u.action.u.ttlm_res.status_code) != WLAN_STATUS_SUCCESS) __ieee80211_disconnect(sdata); } From ebedf8b7f05b9c886d68d63025db8d1b12343157 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Tue, 6 May 2025 21:42:59 +0200 Subject: [PATCH 206/555] wifi: iwlwifi: add support for Killer on MTL For now, we need another entry for these devices, this will be changed completely for 6.16. Closes: https://bugzilla.kernel.org/show_bug.cgi?id=219926 Link: https://patch.msgid.link/20250506214258.2efbdc9e9a82.I31915ec252bd1c74bd53b89a0e214e42a74b6f2e@changeid Signed-off-by: Johannes Berg --- drivers/net/wireless/intel/iwlwifi/pcie/drv.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c index debeea2b3ae5..00056e76ea3d 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c @@ -588,6 +588,8 @@ VISIBLE_IF_IWLWIFI_KUNIT const struct iwl_dev_info iwl_dev_info_table[] = { IWL_DEV_INFO(0x7A70, 0x1692, iwlax411_2ax_cfg_so_gf4_a0, iwl_ax411_killer_1690i_name), IWL_DEV_INFO(0x7AF0, 0x1691, iwlax411_2ax_cfg_so_gf4_a0, iwl_ax411_killer_1690s_name), IWL_DEV_INFO(0x7AF0, 0x1692, iwlax411_2ax_cfg_so_gf4_a0, iwl_ax411_killer_1690i_name), + IWL_DEV_INFO(0x7F70, 0x1691, iwlax411_2ax_cfg_so_gf4_a0, iwl_ax411_killer_1690s_name), + IWL_DEV_INFO(0x7F70, 0x1692, iwlax411_2ax_cfg_so_gf4_a0, iwl_ax411_killer_1690i_name), IWL_DEV_INFO(0x271C, 0x0214, iwl9260_2ac_cfg, iwl9260_1_name), IWL_DEV_INFO(0x7E40, 0x1691, iwl_cfg_ma, iwl_ax411_killer_1690s_name), From 19f5ca461d5fc09bdf93a9f8e4bd78ed3a49dc71 Mon Sep 17 00:00:00 2001 From: Miguel Ojeda Date: Fri, 2 May 2025 16:02:33 +0200 Subject: [PATCH 207/555] objtool/rust: add one more `noreturn` Rust function for Rust 1.87.0 Starting with Rust 1.87.0 (expected 2025-05-15), `objtool` may report: rust/core.o: warning: objtool: _R..._4core9panicking9panic_fmt() falls through to next function _R..._4core9panicking18panic_nounwind_fmt() rust/core.o: warning: objtool: _R..._4core9panicking18panic_nounwind_fmt() falls through to next function _R..._4core9panicking5panic() The reason is that `rust_begin_unwind` is now mangled: _R..._7___rustc17rust_begin_unwind Thus add the mangled one to the list so that `objtool` knows it is actually `noreturn`. See commit 56d680dd23c3 ("objtool/rust: list `noreturn` Rust functions") for more details. Alternatively, we could remove the fixed one in `noreturn.h` and relax this test to cover both, but it seems best to be strict as long as we can. Cc: stable@vger.kernel.org # Needed in 6.12.y and later (Rust is pinned in older LTSs). Cc: Josh Poimboeuf Cc: Peter Zijlstra Reviewed-by: Alice Ryhl Link: https://lore.kernel.org/r/20250502140237.1659624-2-ojeda@kernel.org Signed-off-by: Miguel Ojeda --- tools/objtool/check.c | 1 + 1 file changed, 1 insertion(+) diff --git a/tools/objtool/check.c b/tools/objtool/check.c index 3a411064fa34..b21b12ec88d9 100644 --- a/tools/objtool/check.c +++ b/tools/objtool/check.c @@ -227,6 +227,7 @@ static bool is_rust_noreturn(const struct symbol *func) str_ends_with(func->name, "_4core9panicking19assert_failed_inner") || str_ends_with(func->name, "_4core9panicking30panic_null_pointer_dereference") || str_ends_with(func->name, "_4core9panicking36panic_misaligned_pointer_dereference") || + str_ends_with(func->name, "_7___rustc17rust_begin_unwind") || strstr(func->name, "_4core9panicking13assert_failed") || strstr(func->name, "_4core9panicking11panic_const24panic_const_") || (strstr(func->name, "_4core5slice5index24slice_") && From a39f3087092716f2bd531d6fdc20403c3dc2a879 Mon Sep 17 00:00:00 2001 From: Miguel Ojeda Date: Fri, 2 May 2025 16:02:34 +0200 Subject: [PATCH 208/555] rust: allow Rust 1.87.0's `clippy::ptr_eq` lint Starting with Rust 1.87.0 (expected 2025-05-15) [1], Clippy may expand the `ptr_eq` lint, e.g.: error: use `core::ptr::eq` when comparing raw pointers --> rust/kernel/list.rs:438:12 | 438 | if self.first == item { | ^^^^^^^^^^^^^^^^^^ help: try: `core::ptr::eq(self.first, item)` | = help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#ptr_eq = note: `-D clippy::ptr-eq` implied by `-D warnings` = help: to override `-D warnings` add `#[allow(clippy::ptr_eq)]` It is expected that a PR to relax the lint will be backported [2] by the time Rust 1.87.0 releases, since the lint was considered too eager (at least by default) [3]. Thus allow the lint temporarily just in case. Cc: stable@vger.kernel.org # Needed in 6.12.y and later (Rust is pinned in older LTSs). Link: https://github.com/rust-lang/rust-clippy/pull/14339 [1] Link: https://github.com/rust-lang/rust-clippy/pull/14526 [2] Link: https://github.com/rust-lang/rust-clippy/issues/14525 [3] Link: https://lore.kernel.org/r/20250502140237.1659624-3-ojeda@kernel.org [ Converted to `allow`s since backport was confirmed. - Miguel ] Signed-off-by: Miguel Ojeda --- rust/kernel/alloc/kvec.rs | 3 +++ rust/kernel/list.rs | 3 +++ 2 files changed, 6 insertions(+) diff --git a/rust/kernel/alloc/kvec.rs b/rust/kernel/alloc/kvec.rs index ae9d072741ce..87a71fd40c3c 100644 --- a/rust/kernel/alloc/kvec.rs +++ b/rust/kernel/alloc/kvec.rs @@ -2,6 +2,9 @@ //! Implementation of [`Vec`]. +// May not be needed in Rust 1.87.0 (pending beta backport). +#![allow(clippy::ptr_eq)] + use super::{ allocator::{KVmalloc, Kmalloc, Vmalloc}, layout::ArrayLayout, diff --git a/rust/kernel/list.rs b/rust/kernel/list.rs index a335c3b1ff5e..2054682c5724 100644 --- a/rust/kernel/list.rs +++ b/rust/kernel/list.rs @@ -4,6 +4,9 @@ //! A linked list implementation. +// May not be needed in Rust 1.87.0 (pending beta backport). +#![allow(clippy::ptr_eq)] + use crate::sync::ArcBorrow; use crate::types::Opaque; use core::iter::{DoubleEndedIterator, FusedIterator}; From 7129ea6e242b00938532537da41ddf5fa3e21471 Mon Sep 17 00:00:00 2001 From: Miguel Ojeda Date: Fri, 2 May 2025 16:02:35 +0200 Subject: [PATCH 209/555] rust: clean Rust 1.88.0's `unnecessary_transmutes` lint Starting with Rust 1.88.0 (expected 2025-06-26) [1][2], `rustc` may introduce a new lint that catches unnecessary transmutes, e.g.: error: unnecessary transmute --> rust/uapi/uapi_generated.rs:23242:18 | 23242 | unsafe { ::core::mem::transmute(self._bitfield_1.get(0usize, 1u8) as u8) } | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ help: replace this with: `(self._bitfield_1.get(0usize, 1u8) as u8 == 1)` | = note: `-D unnecessary-transmutes` implied by `-D warnings` = help: to override `-D warnings` add `#[allow(unnecessary_transmutes)]` There are a lot of them (at least 300), but luckily they are all in `bindgen`-generated code. Thus clean all up by allowing it there. Since unknown lints trigger a lint itself in older compilers, do it conditionally so that we can keep the `unknown_lints` lint enabled. Cc: stable@vger.kernel.org # Needed in 6.12.y and later (Rust is pinned in older LTSs). Link: https://github.com/rust-lang/rust/pull/136083 [1] Link: https://github.com/rust-lang/rust/issues/136067 [2] Reviewed-by: Alice Ryhl Link: https://lore.kernel.org/r/20250502140237.1659624-4-ojeda@kernel.org Signed-off-by: Miguel Ojeda --- init/Kconfig | 3 +++ rust/bindings/lib.rs | 1 + rust/uapi/lib.rs | 1 + 3 files changed, 5 insertions(+) diff --git a/init/Kconfig b/init/Kconfig index 63f5974b9fa6..4cdd1049283c 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -140,6 +140,9 @@ config LD_CAN_USE_KEEP_IN_OVERLAY config RUSTC_HAS_COERCE_POINTEE def_bool RUSTC_VERSION >= 108400 +config RUSTC_HAS_UNNECESSARY_TRANSMUTES + def_bool RUSTC_VERSION >= 108800 + config PAHOLE_VERSION int default $(shell,$(srctree)/scripts/pahole-version.sh $(PAHOLE)) diff --git a/rust/bindings/lib.rs b/rust/bindings/lib.rs index 014af0d1fc70..a08eb5518cac 100644 --- a/rust/bindings/lib.rs +++ b/rust/bindings/lib.rs @@ -26,6 +26,7 @@ #[allow(dead_code)] #[allow(clippy::undocumented_unsafe_blocks)] +#[cfg_attr(CONFIG_RUSTC_HAS_UNNECESSARY_TRANSMUTES, allow(unnecessary_transmutes))] mod bindings_raw { // Manual definition for blocklisted types. type __kernel_size_t = usize; diff --git a/rust/uapi/lib.rs b/rust/uapi/lib.rs index 13495910271f..c98d7a8cde77 100644 --- a/rust/uapi/lib.rs +++ b/rust/uapi/lib.rs @@ -24,6 +24,7 @@ unreachable_pub, unsafe_op_in_unsafe_fn )] +#![cfg_attr(CONFIG_RUSTC_HAS_UNNECESSARY_TRANSMUTES, allow(unnecessary_transmutes))] // Manual definition of blocklisted types. type __kernel_size_t = usize; From c016722fd57551f8a6fcf472c9d2bcf2130ea0ec Mon Sep 17 00:00:00 2001 From: Miguel Ojeda Date: Fri, 2 May 2025 16:02:36 +0200 Subject: [PATCH 210/555] rust: clean Rust 1.88.0's warning about `clippy::disallowed_macros` configuration Starting with Rust 1.88.0 (expected 2025-06-26) [1], Clippy may start warning about paths that do not resolve in the `disallowed_macros` configuration: warning: `kernel::dbg` does not refer to an existing macro --> .clippy.toml:10:5 | 10 | { path = "kernel::dbg", reason = "the `dbg!` macro is intended as a debugging tool" }, | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ This is a lint we requested at [2], due to the trouble debugging the lint due to false negatives (e.g. [3]), which we use to emulate `clippy::dbg_macro` [4]. See commit 8577c9dca799 ("rust: replace `clippy::dbg_macro` with `disallowed_macros`") for more details. Given the false negatives are not resolved yet, it is expected that Clippy complains about not finding this macro. Thus, until the false negatives are fixed (and, even then, probably we will need to wait for the MSRV to raise enough), use the escape hatch to allow an invalid path. Cc: stable@vger.kernel.org # Needed in 6.12.y and later (Rust is pinned in older LTSs). Link: https://github.com/rust-lang/rust-clippy/pull/14397 [1] Link: https://github.com/rust-lang/rust-clippy/issues/11432 [2] Link: https://github.com/rust-lang/rust-clippy/issues/11431 [3] Link: https://github.com/rust-lang/rust-clippy/issues/11303 [4] Reviewed-by: Alice Ryhl Link: https://lore.kernel.org/r/20250502140237.1659624-5-ojeda@kernel.org Signed-off-by: Miguel Ojeda --- .clippy.toml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.clippy.toml b/.clippy.toml index 815c94732ed7..137f41d203de 100644 --- a/.clippy.toml +++ b/.clippy.toml @@ -7,5 +7,5 @@ check-private-items = true disallowed-macros = [ # The `clippy::dbg_macro` lint only works with `std::dbg!`, thus we simulate # it here, see: https://github.com/rust-lang/rust-clippy/issues/11303. - { path = "kernel::dbg", reason = "the `dbg!` macro is intended as a debugging tool" }, + { path = "kernel::dbg", reason = "the `dbg!` macro is intended as a debugging tool", allow-invalid = true }, ] From 211dcf77856db64c73e0c3b9ce0c624ec855daca Mon Sep 17 00:00:00 2001 From: Miguel Ojeda Date: Fri, 2 May 2025 16:02:37 +0200 Subject: [PATCH 211/555] rust: clean Rust 1.88.0's `clippy::uninlined_format_args` lint Starting with Rust 1.88.0 (expected 2025-06-26) [1], `rustc` may move back the `uninlined_format_args` to `style` from `pedantic` (it was there waiting for rust-analyzer suppotr), and thus we will start to see lints like: warning: variables can be used directly in the `format!` string --> rust/macros/kunit.rs:105:37 | 105 | let kunit_wrapper_fn_name = format!("kunit_rust_wrapper_{}", test); | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ | = help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#uninlined_format_args help: change this to | 105 - let kunit_wrapper_fn_name = format!("kunit_rust_wrapper_{}", test); 105 + let kunit_wrapper_fn_name = format!("kunit_rust_wrapper_{test}"); There is even a case that is a pure removal: warning: variables can be used directly in the `format!` string --> rust/macros/module.rs:51:13 | 51 | format!("{field}={content}\0", field = field, content = content) | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ | = help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#uninlined_format_args help: change this to | 51 - format!("{field}={content}\0", field = field, content = content) 51 + format!("{field}={content}\0") The lints all seem like nice cleanups, thus just apply them. We may want to disable `allow-mixed-uninlined-format-args` in the future. Cc: stable@vger.kernel.org # Needed in 6.12.y and later (Rust is pinned in older LTSs). Link: https://github.com/rust-lang/rust-clippy/pull/14160 [1] Acked-by: Benno Lossin Reviewed-by: Tamir Duberstein Reviewed-by: Alice Ryhl Link: https://lore.kernel.org/r/20250502140237.1659624-6-ojeda@kernel.org Signed-off-by: Miguel Ojeda --- drivers/gpu/nova-core/gpu.rs | 2 +- rust/kernel/str.rs | 46 +++++++++++------------ rust/macros/kunit.rs | 13 ++----- rust/macros/module.rs | 19 +++------- rust/macros/paste.rs | 2 +- rust/pin-init/internal/src/pinned_drop.rs | 3 +- 6 files changed, 35 insertions(+), 50 deletions(-) diff --git a/drivers/gpu/nova-core/gpu.rs b/drivers/gpu/nova-core/gpu.rs index 17c9660da450..ab0e5a72a059 100644 --- a/drivers/gpu/nova-core/gpu.rs +++ b/drivers/gpu/nova-core/gpu.rs @@ -93,7 +93,7 @@ impl Chipset { // For now, redirect to fmt::Debug for convenience. impl fmt::Display for Chipset { fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result { - write!(f, "{:?}", self) + write!(f, "{self:?}") } } diff --git a/rust/kernel/str.rs b/rust/kernel/str.rs index 878111cb77bc..fb61ce81ea28 100644 --- a/rust/kernel/str.rs +++ b/rust/kernel/str.rs @@ -73,7 +73,7 @@ impl fmt::Display for BStr { b'\r' => f.write_str("\\r")?, // Printable characters. 0x20..=0x7e => f.write_char(b as char)?, - _ => write!(f, "\\x{:02x}", b)?, + _ => write!(f, "\\x{b:02x}")?, } } Ok(()) @@ -109,7 +109,7 @@ impl fmt::Debug for BStr { b'\\' => f.write_str("\\\\")?, // Printable characters. 0x20..=0x7e => f.write_char(b as char)?, - _ => write!(f, "\\x{:02x}", b)?, + _ => write!(f, "\\x{b:02x}")?, } } f.write_char('"') @@ -447,7 +447,7 @@ impl fmt::Display for CStr { // Printable character. f.write_char(c as char)?; } else { - write!(f, "\\x{:02x}", c)?; + write!(f, "\\x{c:02x}")?; } } Ok(()) @@ -479,7 +479,7 @@ impl fmt::Debug for CStr { // Printable characters. b'\"' => f.write_str("\\\"")?, 0x20..=0x7e => f.write_char(c as char)?, - _ => write!(f, "\\x{:02x}", c)?, + _ => write!(f, "\\x{c:02x}")?, } } f.write_str("\"") @@ -641,13 +641,13 @@ mod tests { #[test] fn test_cstr_display() { let hello_world = CStr::from_bytes_with_nul(b"hello, world!\0").unwrap(); - assert_eq!(format!("{}", hello_world), "hello, world!"); + assert_eq!(format!("{hello_world}"), "hello, world!"); let non_printables = CStr::from_bytes_with_nul(b"\x01\x09\x0a\0").unwrap(); - assert_eq!(format!("{}", non_printables), "\\x01\\x09\\x0a"); + assert_eq!(format!("{non_printables}"), "\\x01\\x09\\x0a"); let non_ascii = CStr::from_bytes_with_nul(b"d\xe9j\xe0 vu\0").unwrap(); - assert_eq!(format!("{}", non_ascii), "d\\xe9j\\xe0 vu"); + assert_eq!(format!("{non_ascii}"), "d\\xe9j\\xe0 vu"); let good_bytes = CStr::from_bytes_with_nul(b"\xf0\x9f\xa6\x80\0").unwrap(); - assert_eq!(format!("{}", good_bytes), "\\xf0\\x9f\\xa6\\x80"); + assert_eq!(format!("{good_bytes}"), "\\xf0\\x9f\\xa6\\x80"); } #[test] @@ -658,47 +658,47 @@ mod tests { bytes[i as usize] = i.wrapping_add(1); } let cstr = CStr::from_bytes_with_nul(&bytes).unwrap(); - assert_eq!(format!("{}", cstr), ALL_ASCII_CHARS); + assert_eq!(format!("{cstr}"), ALL_ASCII_CHARS); } #[test] fn test_cstr_debug() { let hello_world = CStr::from_bytes_with_nul(b"hello, world!\0").unwrap(); - assert_eq!(format!("{:?}", hello_world), "\"hello, world!\""); + assert_eq!(format!("{hello_world:?}"), "\"hello, world!\""); let non_printables = CStr::from_bytes_with_nul(b"\x01\x09\x0a\0").unwrap(); - assert_eq!(format!("{:?}", non_printables), "\"\\x01\\x09\\x0a\""); + assert_eq!(format!("{non_printables:?}"), "\"\\x01\\x09\\x0a\""); let non_ascii = CStr::from_bytes_with_nul(b"d\xe9j\xe0 vu\0").unwrap(); - assert_eq!(format!("{:?}", non_ascii), "\"d\\xe9j\\xe0 vu\""); + assert_eq!(format!("{non_ascii:?}"), "\"d\\xe9j\\xe0 vu\""); let good_bytes = CStr::from_bytes_with_nul(b"\xf0\x9f\xa6\x80\0").unwrap(); - assert_eq!(format!("{:?}", good_bytes), "\"\\xf0\\x9f\\xa6\\x80\""); + assert_eq!(format!("{good_bytes:?}"), "\"\\xf0\\x9f\\xa6\\x80\""); } #[test] fn test_bstr_display() { let hello_world = BStr::from_bytes(b"hello, world!"); - assert_eq!(format!("{}", hello_world), "hello, world!"); + assert_eq!(format!("{hello_world}"), "hello, world!"); let escapes = BStr::from_bytes(b"_\t_\n_\r_\\_\'_\"_"); - assert_eq!(format!("{}", escapes), "_\\t_\\n_\\r_\\_'_\"_"); + assert_eq!(format!("{escapes}"), "_\\t_\\n_\\r_\\_'_\"_"); let others = BStr::from_bytes(b"\x01"); - assert_eq!(format!("{}", others), "\\x01"); + assert_eq!(format!("{others}"), "\\x01"); let non_ascii = BStr::from_bytes(b"d\xe9j\xe0 vu"); - assert_eq!(format!("{}", non_ascii), "d\\xe9j\\xe0 vu"); + assert_eq!(format!("{non_ascii}"), "d\\xe9j\\xe0 vu"); let good_bytes = BStr::from_bytes(b"\xf0\x9f\xa6\x80"); - assert_eq!(format!("{}", good_bytes), "\\xf0\\x9f\\xa6\\x80"); + assert_eq!(format!("{good_bytes}"), "\\xf0\\x9f\\xa6\\x80"); } #[test] fn test_bstr_debug() { let hello_world = BStr::from_bytes(b"hello, world!"); - assert_eq!(format!("{:?}", hello_world), "\"hello, world!\""); + assert_eq!(format!("{hello_world:?}"), "\"hello, world!\""); let escapes = BStr::from_bytes(b"_\t_\n_\r_\\_\'_\"_"); - assert_eq!(format!("{:?}", escapes), "\"_\\t_\\n_\\r_\\\\_'_\\\"_\""); + assert_eq!(format!("{escapes:?}"), "\"_\\t_\\n_\\r_\\\\_'_\\\"_\""); let others = BStr::from_bytes(b"\x01"); - assert_eq!(format!("{:?}", others), "\"\\x01\""); + assert_eq!(format!("{others:?}"), "\"\\x01\""); let non_ascii = BStr::from_bytes(b"d\xe9j\xe0 vu"); - assert_eq!(format!("{:?}", non_ascii), "\"d\\xe9j\\xe0 vu\""); + assert_eq!(format!("{non_ascii:?}"), "\"d\\xe9j\\xe0 vu\""); let good_bytes = BStr::from_bytes(b"\xf0\x9f\xa6\x80"); - assert_eq!(format!("{:?}", good_bytes), "\"\\xf0\\x9f\\xa6\\x80\""); + assert_eq!(format!("{good_bytes:?}"), "\"\\xf0\\x9f\\xa6\\x80\""); } } diff --git a/rust/macros/kunit.rs b/rust/macros/kunit.rs index 4f553ecf40c0..99ccac82edde 100644 --- a/rust/macros/kunit.rs +++ b/rust/macros/kunit.rs @@ -15,10 +15,7 @@ pub(crate) fn kunit_tests(attr: TokenStream, ts: TokenStream) -> TokenStream { } if attr.len() > 255 { - panic!( - "The test suite name `{}` exceeds the maximum length of 255 bytes", - attr - ) + panic!("The test suite name `{attr}` exceeds the maximum length of 255 bytes") } let mut tokens: Vec<_> = ts.into_iter().collect(); @@ -102,16 +99,14 @@ pub(crate) fn kunit_tests(attr: TokenStream, ts: TokenStream) -> TokenStream { let mut kunit_macros = "".to_owned(); let mut test_cases = "".to_owned(); for test in &tests { - let kunit_wrapper_fn_name = format!("kunit_rust_wrapper_{}", test); + let kunit_wrapper_fn_name = format!("kunit_rust_wrapper_{test}"); let kunit_wrapper = format!( - "unsafe extern \"C\" fn {}(_test: *mut kernel::bindings::kunit) {{ {}(); }}", - kunit_wrapper_fn_name, test + "unsafe extern \"C\" fn {kunit_wrapper_fn_name}(_test: *mut kernel::bindings::kunit) {{ {test}(); }}" ); writeln!(kunit_macros, "{kunit_wrapper}").unwrap(); writeln!( test_cases, - " kernel::kunit::kunit_case(kernel::c_str!(\"{}\"), {}),", - test, kunit_wrapper_fn_name + " kernel::kunit::kunit_case(kernel::c_str!(\"{test}\"), {kunit_wrapper_fn_name})," ) .unwrap(); } diff --git a/rust/macros/module.rs b/rust/macros/module.rs index a9418fbc9b44..2f66107847f7 100644 --- a/rust/macros/module.rs +++ b/rust/macros/module.rs @@ -48,7 +48,7 @@ impl<'a> ModInfoBuilder<'a> { ) } else { // Loadable modules' modinfo strings go as-is. - format!("{field}={content}\0", field = field, content = content) + format!("{field}={content}\0") }; write!( @@ -126,10 +126,7 @@ impl ModuleInfo { }; if seen_keys.contains(&key) { - panic!( - "Duplicated key \"{}\". Keys can only be specified once.", - key - ); + panic!("Duplicated key \"{key}\". Keys can only be specified once."); } assert_eq!(expect_punct(it), ':'); @@ -143,10 +140,7 @@ impl ModuleInfo { "license" => info.license = expect_string_ascii(it), "alias" => info.alias = Some(expect_string_array(it)), "firmware" => info.firmware = Some(expect_string_array(it)), - _ => panic!( - "Unknown key \"{}\". Valid keys are: {:?}.", - key, EXPECTED_KEYS - ), + _ => panic!("Unknown key \"{key}\". Valid keys are: {EXPECTED_KEYS:?}."), } assert_eq!(expect_punct(it), ','); @@ -158,7 +152,7 @@ impl ModuleInfo { for key in REQUIRED_KEYS { if !seen_keys.iter().any(|e| e == key) { - panic!("Missing required key \"{}\".", key); + panic!("Missing required key \"{key}\"."); } } @@ -170,10 +164,7 @@ impl ModuleInfo { } if seen_keys != ordered_keys { - panic!( - "Keys are not ordered as expected. Order them like: {:?}.", - ordered_keys - ); + panic!("Keys are not ordered as expected. Order them like: {ordered_keys:?}."); } info diff --git a/rust/macros/paste.rs b/rust/macros/paste.rs index 6529a387673f..cce712d19855 100644 --- a/rust/macros/paste.rs +++ b/rust/macros/paste.rs @@ -50,7 +50,7 @@ fn concat_helper(tokens: &[TokenTree]) -> Vec<(String, Span)> { let tokens = group.stream().into_iter().collect::>(); segments.append(&mut concat_helper(tokens.as_slice())); } - token => panic!("unexpected token in paste segments: {:?}", token), + token => panic!("unexpected token in paste segments: {token:?}"), }; } diff --git a/rust/pin-init/internal/src/pinned_drop.rs b/rust/pin-init/internal/src/pinned_drop.rs index c824dd8b436d..c4ca7a70b726 100644 --- a/rust/pin-init/internal/src/pinned_drop.rs +++ b/rust/pin-init/internal/src/pinned_drop.rs @@ -28,8 +28,7 @@ pub(crate) fn pinned_drop(_args: TokenStream, input: TokenStream) -> TokenStream // Found the end of the generics, this should be `PinnedDrop`. assert!( matches!(tt, TokenTree::Ident(i) if i.to_string() == "PinnedDrop"), - "expected 'PinnedDrop', found: '{:?}'", - tt + "expected 'PinnedDrop', found: '{tt:?}'" ); pinned_drop_idx = Some(i); break; From 5595c31c370957aabe739ac3996aedba8267603f Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Pawe=C5=82=20Anikiel?= Date: Thu, 10 Apr 2025 11:54:20 +0000 Subject: [PATCH 212/555] x86/Kconfig: make CFI_AUTO_DEFAULT depend on !RUST or Rust >= 1.88 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Calling core::fmt::write() from rust code while FineIBT is enabled results in a kernel panic: [ 4614.199779] kernel BUG at arch/x86/kernel/cet.c:132! [ 4614.205343] Oops: invalid opcode: 0000 [#1] PREEMPT SMP NOPTI [ 4614.211781] CPU: 2 UID: 0 PID: 6057 Comm: dmabuf_dump Tainted: G U O 6.12.17-android16-0-g6ab38c534a43 #1 9da040f27673ec3945e23b998a0f8bd64c846599 [ 4614.227832] Tainted: [U]=USER, [O]=OOT_MODULE [ 4614.241247] RIP: 0010:do_kernel_cp_fault+0xea/0xf0 ... [ 4614.398144] RIP: 0010:_RNvXs5_NtNtNtCs3o2tGsuHyou_4core3fmt3num3impyNtB9_7Display3fmt+0x0/0x20 [ 4614.407792] Code: 48 f7 df 48 0f 48 f9 48 89 f2 89 c6 5d e9 18 fd ff ff 0f 1f 84 00 00 00 00 00 f3 0f 1e fa 41 81 ea 14 61 af 2c 74 03 0f 0b 90 <66> 0f 1f 00 55 48 89 e5 48 89 f2 48 8b 3f be 01 00 00 00 5d e9 e7 [ 4614.428775] RSP: 0018:ffffb95acfa4ba68 EFLAGS: 00010246 [ 4614.434609] RAX: 0000000000000000 RBX: 0000000000000010 RCX: 0000000000000000 [ 4614.442587] RDX: 0000000000000007 RSI: ffffb95acfa4ba70 RDI: ffffb95acfa4bc88 [ 4614.450557] RBP: ffffb95acfa4bae0 R08: ffff0a00ffffff05 R09: 0000000000000070 [ 4614.458527] R10: 0000000000000000 R11: ffffffffab67eaf0 R12: ffffb95acfa4bcc8 [ 4614.466493] R13: ffffffffac5d50f0 R14: 0000000000000000 R15: 0000000000000000 [ 4614.474473] ? __cfi__RNvXs5_NtNtNtCs3o2tGsuHyou_4core3fmt3num3impyNtB9_7Display3fmt+0x10/0x10 [ 4614.484118] ? _RNvNtCs3o2tGsuHyou_4core3fmt5write+0x1d2/0x250 This happens because core::fmt::write() calls core::fmt::rt::Argument::fmt(), which currently has CFI disabled: library/core/src/fmt/rt.rs: 171 // FIXME: Transmuting formatter in new and indirectly branching to/calling 172 // it here is an explicit CFI violation. 173 #[allow(inline_no_sanitize)] 174 #[no_sanitize(cfi, kcfi)] 175 #[inline] 176 pub(super) unsafe fn fmt(&self, f: &mut Formatter<'_>) -> Result { This causes a Control Protection exception, because FineIBT has sealed off the original function's endbr64. This makes rust currently incompatible with FineIBT. Add a Kconfig dependency that prevents FineIBT from getting turned on by default if rust is enabled. [ Rust 1.88.0 (scheduled for 2025-06-26) should have this fixed [1], and thus we relaxed the condition with Rust >= 1.88. When `objtool` lands checking for this with e.g. [2], the plan is to ideally run that in upstream Rust's CI to prevent regressions early [3], since we do not control `core`'s source code. Alice tested the Rust PR backported to an older compiler. Peter would like that Rust provides a stable `core` which can be pulled into the kernel: "Relying on that much out of tree code is 'unfortunate'". - Miguel ] Signed-off-by: Paweł Anikiel Reviewed-by: Alice Ryhl Acked-by: Peter Zijlstra Link: https://github.com/rust-lang/rust/pull/139632 [1] Link: https://lore.kernel.org/rust-for-linux/20250410154556.GB9003@noisy.programming.kicks-ass.net/ [2] Link: https://github.com/rust-lang/rust/pull/139632#issuecomment-2801950873 [3] Link: https://lore.kernel.org/r/20250410115420.366349-1-panikiel@google.com Link: https://lore.kernel.org/r/att0-CANiq72kjDM0cKALVy4POEzhfdT4nO7tqz0Pm7xM+3=_0+L1t=A@mail.gmail.com [ Reduced splat. - Miguel ] Signed-off-by: Miguel Ojeda --- arch/x86/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 4b9f378e05f6..5873c9e39919 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -2368,6 +2368,7 @@ config STRICT_SIGALTSTACK_SIZE config CFI_AUTO_DEFAULT bool "Attempt to use FineIBT by default at boot time" depends on FINEIBT + depends on !RUST || RUSTC_VERSION >= 108800 default y help Attempt to use FineIBT by default at boot time. If enabled, From 0093cb194a7511d1e68865fa35b763c72e44c2f0 Mon Sep 17 00:00:00 2001 From: Przemek Kitszel Date: Mon, 5 May 2025 09:19:38 -0700 Subject: [PATCH 213/555] ice: use DSN instead of PCI BDF for ice_adapter index Use Device Serial Number instead of PCI bus/device/function for the index of struct ice_adapter. Functions on the same physical device should point to the very same ice_adapter instance, but with two PFs, when at least one of them is PCI-e passed-through to a VM, it is no longer the case - PFs will get seemingly random PCI BDF values, and thus indices, what finally leds to each of them being on their own instance of ice_adapter. That causes them to don't attempt any synchronization of the PTP HW clock usage, or any other future resources. DSN works nicely in place of the index, as it is "immutable" in terms of virtualization. Fixes: 0e2bddf9e5f9 ("ice: add ice_adapter for shared data across PFs on the same NIC") Suggested-by: Jacob Keller Suggested-by: Jakub Kicinski Suggested-by: Jiri Pirko Reviewed-by: Aleksandr Loktionov Signed-off-by: Przemek Kitszel Reviewed-by: Simon Horman Tested-by: Rinitha S (A Contingent worker at Intel) Signed-off-by: Tony Nguyen Reviewed-by: Jiri Pirko Link: https://patch.msgid.link/20250505161939.2083581-1-anthony.l.nguyen@intel.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/intel/ice/ice_adapter.c | 47 ++++++++------------ drivers/net/ethernet/intel/ice/ice_adapter.h | 6 ++- 2 files changed, 22 insertions(+), 31 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/ice_adapter.c b/drivers/net/ethernet/intel/ice/ice_adapter.c index 01a08cfd0090..66e070095d1b 100644 --- a/drivers/net/ethernet/intel/ice/ice_adapter.c +++ b/drivers/net/ethernet/intel/ice/ice_adapter.c @@ -1,7 +1,6 @@ // SPDX-License-Identifier: GPL-2.0-only // SPDX-FileCopyrightText: Copyright Red Hat -#include #include #include #include @@ -14,32 +13,16 @@ static DEFINE_XARRAY(ice_adapters); static DEFINE_MUTEX(ice_adapters_mutex); -/* PCI bus number is 8 bits. Slot is 5 bits. Domain can have the rest. */ -#define INDEX_FIELD_DOMAIN GENMASK(BITS_PER_LONG - 1, 13) -#define INDEX_FIELD_DEV GENMASK(31, 16) -#define INDEX_FIELD_BUS GENMASK(12, 5) -#define INDEX_FIELD_SLOT GENMASK(4, 0) - -static unsigned long ice_adapter_index(const struct pci_dev *pdev) +static unsigned long ice_adapter_index(u64 dsn) { - unsigned int domain = pci_domain_nr(pdev->bus); - - WARN_ON(domain > FIELD_MAX(INDEX_FIELD_DOMAIN)); - - switch (pdev->device) { - case ICE_DEV_ID_E825C_BACKPLANE: - case ICE_DEV_ID_E825C_QSFP: - case ICE_DEV_ID_E825C_SFP: - case ICE_DEV_ID_E825C_SGMII: - return FIELD_PREP(INDEX_FIELD_DEV, pdev->device); - default: - return FIELD_PREP(INDEX_FIELD_DOMAIN, domain) | - FIELD_PREP(INDEX_FIELD_BUS, pdev->bus->number) | - FIELD_PREP(INDEX_FIELD_SLOT, PCI_SLOT(pdev->devfn)); - } +#if BITS_PER_LONG == 64 + return dsn; +#else + return (u32)dsn ^ (u32)(dsn >> 32); +#endif } -static struct ice_adapter *ice_adapter_new(void) +static struct ice_adapter *ice_adapter_new(u64 dsn) { struct ice_adapter *adapter; @@ -47,6 +30,7 @@ static struct ice_adapter *ice_adapter_new(void) if (!adapter) return NULL; + adapter->device_serial_number = dsn; spin_lock_init(&adapter->ptp_gltsyn_time_lock); refcount_set(&adapter->refcount, 1); @@ -77,23 +61,26 @@ static void ice_adapter_free(struct ice_adapter *adapter) * Return: Pointer to ice_adapter on success. * ERR_PTR() on error. -ENOMEM is the only possible error. */ -struct ice_adapter *ice_adapter_get(const struct pci_dev *pdev) +struct ice_adapter *ice_adapter_get(struct pci_dev *pdev) { - unsigned long index = ice_adapter_index(pdev); + u64 dsn = pci_get_dsn(pdev); struct ice_adapter *adapter; + unsigned long index; int err; + index = ice_adapter_index(dsn); scoped_guard(mutex, &ice_adapters_mutex) { err = xa_insert(&ice_adapters, index, NULL, GFP_KERNEL); if (err == -EBUSY) { adapter = xa_load(&ice_adapters, index); refcount_inc(&adapter->refcount); + WARN_ON_ONCE(adapter->device_serial_number != dsn); return adapter; } if (err) return ERR_PTR(err); - adapter = ice_adapter_new(); + adapter = ice_adapter_new(dsn); if (!adapter) return ERR_PTR(-ENOMEM); xa_store(&ice_adapters, index, adapter, GFP_KERNEL); @@ -110,11 +97,13 @@ struct ice_adapter *ice_adapter_get(const struct pci_dev *pdev) * * Context: Process, may sleep. */ -void ice_adapter_put(const struct pci_dev *pdev) +void ice_adapter_put(struct pci_dev *pdev) { - unsigned long index = ice_adapter_index(pdev); + u64 dsn = pci_get_dsn(pdev); struct ice_adapter *adapter; + unsigned long index; + index = ice_adapter_index(dsn); scoped_guard(mutex, &ice_adapters_mutex) { adapter = xa_load(&ice_adapters, index); if (WARN_ON(!adapter)) diff --git a/drivers/net/ethernet/intel/ice/ice_adapter.h b/drivers/net/ethernet/intel/ice/ice_adapter.h index e233225848b3..ac15c0d2bc1a 100644 --- a/drivers/net/ethernet/intel/ice/ice_adapter.h +++ b/drivers/net/ethernet/intel/ice/ice_adapter.h @@ -32,6 +32,7 @@ struct ice_port_list { * @refcount: Reference count. struct ice_pf objects hold the references. * @ctrl_pf: Control PF of the adapter * @ports: Ports list + * @device_serial_number: DSN cached for collision detection on 32bit systems */ struct ice_adapter { refcount_t refcount; @@ -40,9 +41,10 @@ struct ice_adapter { struct ice_pf *ctrl_pf; struct ice_port_list ports; + u64 device_serial_number; }; -struct ice_adapter *ice_adapter_get(const struct pci_dev *pdev); -void ice_adapter_put(const struct pci_dev *pdev); +struct ice_adapter *ice_adapter_get(struct pci_dev *pdev); +void ice_adapter_put(struct pci_dev *pdev); #endif /* _ICE_ADAPTER_H */ From 08e9f2d584c4732180edee4cb2dbfa7586d7d5a3 Mon Sep 17 00:00:00 2001 From: Cosmin Ratiu Date: Mon, 5 May 2025 22:47:13 +0300 Subject: [PATCH 214/555] net: Lock netdevices during dev_shutdown __qdisc_destroy() calls into various qdiscs .destroy() op, which in turn can call .ndo_setup_tc(), which requires the netdev instance lock. This commit extends the critical section in unregister_netdevice_many_notify() to cover dev_shutdown() (and dev_tcx_uninstall() as a side-effect) and acquires the netdev instance lock in __dev_change_net_namespace() for the other dev_shutdown() call. This should now guarantee that for all qdisc ops, the netdev instance lock is held during .ndo_setup_tc(). Fixes: a0527ee2df3f ("net: hold netdev instance lock during qdisc ndo_setup_tc") Signed-off-by: Cosmin Ratiu Acked-by: Stanislav Fomichev Link: https://patch.msgid.link/20250505194713.1723399-1-cratiu@nvidia.com Signed-off-by: Jakub Kicinski --- net/core/dev.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/net/core/dev.c b/net/core/dev.c index 1be7cb73a602..92e004c354ea 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -11966,9 +11966,9 @@ void unregister_netdevice_many_notify(struct list_head *head, struct sk_buff *skb = NULL; /* Shutdown queueing discipline. */ + netdev_lock_ops(dev); dev_shutdown(dev); dev_tcx_uninstall(dev); - netdev_lock_ops(dev); dev_xdp_uninstall(dev); dev_memory_provider_uninstall(dev); netdev_unlock_ops(dev); @@ -12161,7 +12161,9 @@ int __dev_change_net_namespace(struct net_device *dev, struct net *net, synchronize_net(); /* Shutdown queueing discipline. */ + netdev_lock_ops(dev); dev_shutdown(dev); + netdev_unlock_ops(dev); /* Notify protocols, that we are about to destroy * this device. They should clean all the things. From 35076d2223c731f7be75af61e67f90807384d030 Mon Sep 17 00:00:00 2001 From: Gao Xiang Date: Tue, 6 May 2025 18:18:50 +0800 Subject: [PATCH 215/555] erofs: ensure the extra temporary copy is valid for shortened bvecs When compressed data deduplication is enabled, multiple logical extents may reference the same compressed physical cluster. The previous commit 94c43de73521 ("erofs: fix wrong primary bvec selection on deduplicated extents") already avoids using shortened bvecs. However, in such cases, the extra temporary buffers also need to be preserved for later use in z_erofs_fill_other_copies() to to prevent data corruption. IOWs, extra temporary buffers have to be retained not only due to varying start relative offsets (`pageofs_out`, as indicated by `pcl->multibases`) but also because of shortened bvecs. android.hardware.graphics.composer@2.1.so : 270696 bytes 0: 0.. 204185 | 204185 : 628019200.. 628084736 | 65536 -> 1: 204185.. 225536 | 21351 : 544063488.. 544129024 | 65536 2: 225536.. 270696 | 45160 : 0.. 0 | 0 com.android.vndk.v28.apex : 93814897 bytes ... 364: 53869896..54095257 | 225361 : 543997952.. 544063488 | 65536 -> 365: 54095257..54309344 | 214087 : 544063488.. 544129024 | 65536 366: 54309344..54514557 | 205213 : 544129024.. 544194560 | 65536 ... Both 204185 and 54095257 have the same start relative offset of 3481, but the logical page 55 of `android.hardware.graphics.composer@2.1.so` ranges from 225280 to 229632, forming a shortened bvec [225280, 225536) that cannot be used for decompressing the range from 54095257 to 54309344 of `com.android.vndk.v28.apex`. Since `pcl->multibases` is already meaningless, just mark `be->keepxcpy` on demand for simplicity. Again, this issue can only lead to data corruption if `-Ededupe` is on. Fixes: 94c43de73521 ("erofs: fix wrong primary bvec selection on deduplicated extents") Reviewed-by: Hongbo Li Signed-off-by: Gao Xiang Link: https://lore.kernel.org/r/20250506101850.191506-1-hsiangkao@linux.alibaba.com --- fs/erofs/zdata.c | 29 +++++++++++++---------------- 1 file changed, 13 insertions(+), 16 deletions(-) diff --git a/fs/erofs/zdata.c b/fs/erofs/zdata.c index 5c061aaeeb45..b8e6b76c23d5 100644 --- a/fs/erofs/zdata.c +++ b/fs/erofs/zdata.c @@ -79,9 +79,6 @@ struct z_erofs_pcluster { /* L: whether partial decompression or not */ bool partial; - /* L: indicate several pageofs_outs or not */ - bool multibases; - /* L: whether extra buffer allocations are best-effort */ bool besteffort; @@ -1046,8 +1043,6 @@ static int z_erofs_scan_folio(struct z_erofs_frontend *f, break; erofs_onlinefolio_split(folio); - if (f->pcl->pageofs_out != (map->m_la & ~PAGE_MASK)) - f->pcl->multibases = true; if (f->pcl->length < offset + end - map->m_la) { f->pcl->length = offset + end - map->m_la; f->pcl->pageofs_out = map->m_la & ~PAGE_MASK; @@ -1093,7 +1088,6 @@ struct z_erofs_backend { struct page *onstack_pages[Z_EROFS_ONSTACK_PAGES]; struct super_block *sb; struct z_erofs_pcluster *pcl; - /* pages with the longest decompressed length for deduplication */ struct page **decompressed_pages; /* pages to keep the compressed data */ @@ -1102,6 +1096,8 @@ struct z_erofs_backend { struct list_head decompressed_secondary_bvecs; struct page **pagepool; unsigned int onstack_used, nr_pages; + /* indicate if temporary copies should be preserved for later use */ + bool keepxcpy; }; struct z_erofs_bvec_item { @@ -1112,18 +1108,20 @@ struct z_erofs_bvec_item { static void z_erofs_do_decompressed_bvec(struct z_erofs_backend *be, struct z_erofs_bvec *bvec) { + int poff = bvec->offset + be->pcl->pageofs_out; struct z_erofs_bvec_item *item; - unsigned int pgnr; + struct page **page; - if (!((bvec->offset + be->pcl->pageofs_out) & ~PAGE_MASK) && - (bvec->end == PAGE_SIZE || - bvec->offset + bvec->end == be->pcl->length)) { - pgnr = (bvec->offset + be->pcl->pageofs_out) >> PAGE_SHIFT; - DBG_BUGON(pgnr >= be->nr_pages); - if (!be->decompressed_pages[pgnr]) { - be->decompressed_pages[pgnr] = bvec->page; + if (!(poff & ~PAGE_MASK) && (bvec->end == PAGE_SIZE || + bvec->offset + bvec->end == be->pcl->length)) { + DBG_BUGON((poff >> PAGE_SHIFT) >= be->nr_pages); + page = be->decompressed_pages + (poff >> PAGE_SHIFT); + if (!*page) { + *page = bvec->page; return; } + } else { + be->keepxcpy = true; } /* (cold path) one pcluster is requested multiple times */ @@ -1289,7 +1287,7 @@ static int z_erofs_decompress_pcluster(struct z_erofs_backend *be, int err) .alg = pcl->algorithmformat, .inplace_io = overlapped, .partial_decoding = pcl->partial, - .fillgaps = pcl->multibases, + .fillgaps = be->keepxcpy, .gfp = pcl->besteffort ? GFP_KERNEL : GFP_NOWAIT | __GFP_NORETRY }, be->pagepool); @@ -1346,7 +1344,6 @@ static int z_erofs_decompress_pcluster(struct z_erofs_backend *be, int err) pcl->length = 0; pcl->partial = true; - pcl->multibases = false; pcl->besteffort = false; pcl->bvset.nextpage = NULL; pcl->vcnt = 0; From 78cd408356fe3edbac66598772fd347bf3e32c1f Mon Sep 17 00:00:00 2001 From: Stanislav Fomichev Date: Mon, 5 May 2025 18:19:19 -0700 Subject: [PATCH 216/555] net: add missing instance lock to dev_set_promiscuity Accidentally spotted while trying to understand what else needs to be renamed to netif_ prefix. Most of the calls to dev_set_promiscuity are adjacent to dev_set_allmulti or dev_disable_lro so it should be safe to add the lock. Note that new netif_set_promiscuity is currently unused, the locked paths call __dev_set_promiscuity directly. Fixes: ad7c7b2172c3 ("net: hold netdev instance lock during sysfs operations") Signed-off-by: Stanislav Fomichev Link: https://patch.msgid.link/20250506011919.2882313-1-sdf@fomichev.me Signed-off-by: Jakub Kicinski --- include/linux/netdevice.h | 1 + net/core/dev.c | 14 +------------- net/core/dev_api.c | 23 +++++++++++++++++++++++ 3 files changed, 25 insertions(+), 13 deletions(-) diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 2d11d013cabe..7ea022750e4e 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -4972,6 +4972,7 @@ static inline void __dev_mc_unsync(struct net_device *dev, /* Functions used for secondary unicast and multicast support */ void dev_set_rx_mode(struct net_device *dev); +int netif_set_promiscuity(struct net_device *dev, int inc); int dev_set_promiscuity(struct net_device *dev, int inc); int netif_set_allmulti(struct net_device *dev, int inc, bool notify); int dev_set_allmulti(struct net_device *dev, int inc); diff --git a/net/core/dev.c b/net/core/dev.c index 92e004c354ea..11da1e272ec2 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -9193,18 +9193,7 @@ static int __dev_set_promiscuity(struct net_device *dev, int inc, bool notify) return 0; } -/** - * dev_set_promiscuity - update promiscuity count on a device - * @dev: device - * @inc: modifier - * - * Add or remove promiscuity from a device. While the count in the device - * remains above zero the interface remains promiscuous. Once it hits zero - * the device reverts back to normal filtering operation. A negative inc - * value is used to drop promiscuity on the device. - * Return 0 if successful or a negative errno code on error. - */ -int dev_set_promiscuity(struct net_device *dev, int inc) +int netif_set_promiscuity(struct net_device *dev, int inc) { unsigned int old_flags = dev->flags; int err; @@ -9216,7 +9205,6 @@ int dev_set_promiscuity(struct net_device *dev, int inc) dev_set_rx_mode(dev); return err; } -EXPORT_SYMBOL(dev_set_promiscuity); int netif_set_allmulti(struct net_device *dev, int inc, bool notify) { diff --git a/net/core/dev_api.c b/net/core/dev_api.c index 90898cd540ce..f9a160ab596f 100644 --- a/net/core/dev_api.c +++ b/net/core/dev_api.c @@ -267,6 +267,29 @@ void dev_disable_lro(struct net_device *dev) } EXPORT_SYMBOL(dev_disable_lro); +/** + * dev_set_promiscuity() - update promiscuity count on a device + * @dev: device + * @inc: modifier + * + * Add or remove promiscuity from a device. While the count in the device + * remains above zero the interface remains promiscuous. Once it hits zero + * the device reverts back to normal filtering operation. A negative inc + * value is used to drop promiscuity on the device. + * Return 0 if successful or a negative errno code on error. + */ +int dev_set_promiscuity(struct net_device *dev, int inc) +{ + int ret; + + netdev_lock_ops(dev); + ret = netif_set_promiscuity(dev, inc); + netdev_unlock_ops(dev); + + return ret; +} +EXPORT_SYMBOL(dev_set_promiscuity); + /** * dev_set_allmulti() - update allmulti count on a device * @dev: device From 650415fca0a97472fdd79725e35152614d1aad76 Mon Sep 17 00:00:00 2001 From: Daniel Wagner Date: Fri, 2 May 2025 10:58:00 +0200 Subject: [PATCH 217/555] nvme: unblock ctrl state transition for firmware update The original nvme subsystem design didn't have a CONNECTING state; the state machine allowed transitions from RESETTING to LIVE directly. With the introduction of nvme fabrics the CONNECTING state was introduce. Over time the nvme-pci started to use the CONNECTING state as well. Eventually, a bug fix for the nvme-fc started to depend that the only valid transition to LIVE was from CONNECTING. Though this change didn't update the firmware update handler which was still depending on RESETTING to LIVE transition. The simplest way to address it for the time being is to switch into CONNECTING state before going to LIVE state. Fixes: d2fe192348f9 ("nvme: only allow entering LIVE from CONNECTING state") Reported-by: Guenter Roeck Signed-off-by: Daniel Wagner Closes: https://lore.kernel.org/all/0134ea15-8d5f-41f7-9e9a-d7e6d82accaa@roeck-us.net Reviewed-by: Keith Busch Reviewed-by: Sagi Grimberg Reviewed-by: Guenter Roeck --- drivers/nvme/host/core.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c index eb6ea8acb3cc..ac53629fce68 100644 --- a/drivers/nvme/host/core.c +++ b/drivers/nvme/host/core.c @@ -4493,7 +4493,8 @@ static void nvme_fw_act_work(struct work_struct *work) msleep(100); } - if (!nvme_change_ctrl_state(ctrl, NVME_CTRL_LIVE)) + if (!nvme_change_ctrl_state(ctrl, NVME_CTRL_CONNECTING) || + !nvme_change_ctrl_state(ctrl, NVME_CTRL_LIVE)) return; nvme_unquiesce_io_queues(ctrl); From ffea7c73d181db273be8b306be6bc573dfe2257b Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Wed, 30 Apr 2025 11:59:15 +0100 Subject: [PATCH 218/555] KVM: arm64: Properly save/restore HCRX_EL2 Rather than restoring HCRX_EL2 to a fixed value on vcpu exit, perform a full save/restore of the register, ensuring that we don't lose bits that would have been set at some point in the host kernel lifetime, such as the GCSEn bit. Fixes: ff5181d8a2a82 ("arm64/gcs: Provide basic EL2 setup to allow GCS usage at EL0 and EL1") Signed-off-by: Marc Zyngier Link: https://lore.kernel.org/r/20250430105916.3815157-2-maz@kernel.org Signed-off-by: Oliver Upton --- arch/arm64/kvm/hyp/include/hyp/switch.h | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-) diff --git a/arch/arm64/kvm/hyp/include/hyp/switch.h b/arch/arm64/kvm/hyp/include/hyp/switch.h index b741ea6aefa5..96f625dc7256 100644 --- a/arch/arm64/kvm/hyp/include/hyp/switch.h +++ b/arch/arm64/kvm/hyp/include/hyp/switch.h @@ -235,6 +235,8 @@ static inline void __deactivate_traps_mpam(void) static inline void __activate_traps_common(struct kvm_vcpu *vcpu) { + struct kvm_cpu_context *hctxt = host_data_ptr(host_ctxt); + /* Trap on AArch32 cp15 c15 (impdef sysregs) accesses (EL1 or EL0) */ write_sysreg(1 << 15, hstr_el2); @@ -245,11 +247,8 @@ static inline void __activate_traps_common(struct kvm_vcpu *vcpu) * EL1 instead of being trapped to EL2. */ if (system_supports_pmuv3()) { - struct kvm_cpu_context *hctxt; - write_sysreg(0, pmselr_el0); - hctxt = host_data_ptr(host_ctxt); ctxt_sys_reg(hctxt, PMUSERENR_EL0) = read_sysreg(pmuserenr_el0); write_sysreg(ARMV8_PMU_USERENR_MASK, pmuserenr_el0); vcpu_set_flag(vcpu, PMUSERENR_ON_CPU); @@ -269,6 +268,7 @@ static inline void __activate_traps_common(struct kvm_vcpu *vcpu) hcrx &= ~clr; } + ctxt_sys_reg(hctxt, HCRX_EL2) = read_sysreg_s(SYS_HCRX_EL2); write_sysreg_s(hcrx, SYS_HCRX_EL2); } @@ -278,19 +278,18 @@ static inline void __activate_traps_common(struct kvm_vcpu *vcpu) static inline void __deactivate_traps_common(struct kvm_vcpu *vcpu) { + struct kvm_cpu_context *hctxt = host_data_ptr(host_ctxt); + write_sysreg(*host_data_ptr(host_debug_state.mdcr_el2), mdcr_el2); write_sysreg(0, hstr_el2); if (system_supports_pmuv3()) { - struct kvm_cpu_context *hctxt; - - hctxt = host_data_ptr(host_ctxt); write_sysreg(ctxt_sys_reg(hctxt, PMUSERENR_EL0), pmuserenr_el0); vcpu_clear_flag(vcpu, PMUSERENR_ON_CPU); } if (cpus_have_final_cap(ARM64_HAS_HCX)) - write_sysreg_s(HCRX_HOST_FLAGS, SYS_HCRX_EL2); + write_sysreg_s(ctxt_sys_reg(hctxt, HCRX_EL2), SYS_HCRX_EL2); __deactivate_traps_hfgxtr(vcpu); __deactivate_traps_mpam(); From ef296ee98bb19ee2a6dbf76155184c99da0d7c71 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Wed, 30 Apr 2025 11:59:16 +0100 Subject: [PATCH 219/555] KVM: arm64: Kill HCRX_HOST_FLAGS HCRX_HOST_FLAGS, like most of these hardcoded setups, are not a good match for options that can be selectively enabled or disabled. Nothing but the early setup is relying on it now, so kill the macro and move the bag of bits where they belong. Signed-off-by: Marc Zyngier Link: https://lore.kernel.org/r/20250430105916.3815157-3-maz@kernel.org Signed-off-by: Oliver Upton --- arch/arm64/include/asm/el2_setup.h | 2 +- arch/arm64/include/asm/kvm_arm.h | 1 - 2 files changed, 1 insertion(+), 2 deletions(-) diff --git a/arch/arm64/include/asm/el2_setup.h b/arch/arm64/include/asm/el2_setup.h index ebceaae3c749..d40e427ddad9 100644 --- a/arch/arm64/include/asm/el2_setup.h +++ b/arch/arm64/include/asm/el2_setup.h @@ -52,7 +52,7 @@ mrs x0, id_aa64mmfr1_el1 ubfx x0, x0, #ID_AA64MMFR1_EL1_HCX_SHIFT, #4 cbz x0, .Lskip_hcrx_\@ - mov_q x0, HCRX_HOST_FLAGS + mov_q x0, (HCRX_EL2_MSCEn | HCRX_EL2_TCR2En | HCRX_EL2_EnFPM) /* Enable GCS if supported */ mrs_s x1, SYS_ID_AA64PFR1_EL1 diff --git a/arch/arm64/include/asm/kvm_arm.h b/arch/arm64/include/asm/kvm_arm.h index bba4b0e93091..e9c8a581e16f 100644 --- a/arch/arm64/include/asm/kvm_arm.h +++ b/arch/arm64/include/asm/kvm_arm.h @@ -102,7 +102,6 @@ #define HCR_HOST_NVHE_PROTECTED_FLAGS (HCR_HOST_NVHE_FLAGS | HCR_TSC) #define HCR_HOST_VHE_FLAGS (HCR_RW | HCR_TGE | HCR_E2H | HCR_AMO | HCR_IMO | HCR_FMO) -#define HCRX_HOST_FLAGS (HCRX_EL2_MSCEn | HCRX_EL2_TCR2En | HCRX_EL2_EnFPM) #define MPAMHCR_HOST_FLAGS 0 /* TCR_EL2 Registers bits */ From 3949e28786cd0afcd96a46ce6629245203f629e5 Mon Sep 17 00:00:00 2001 From: Mostafa Saleh Date: Thu, 1 May 2025 16:24:50 +0000 Subject: [PATCH 220/555] KVM: arm64: Fix memory check in host_stage2_set_owner_locked() I found this simple bug while preparing some patches for pKVM. AFAICT, it should be harmless (besides crashing the kernel if it was misbehaving) Fixes: e94a7dea2972 ("KVM: arm64: Move host page ownership tracking to the hyp vmemmap") Signed-off-by: Mostafa Saleh Link: https://lore.kernel.org/r/20250501162450.2784043-1-smostafa@google.com Signed-off-by: Oliver Upton --- arch/arm64/kvm/hyp/nvhe/mem_protect.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm64/kvm/hyp/nvhe/mem_protect.c b/arch/arm64/kvm/hyp/nvhe/mem_protect.c index 2a5284f749b4..e80f3ebd3e2a 100644 --- a/arch/arm64/kvm/hyp/nvhe/mem_protect.c +++ b/arch/arm64/kvm/hyp/nvhe/mem_protect.c @@ -503,7 +503,7 @@ int host_stage2_set_owner_locked(phys_addr_t addr, u64 size, u8 owner_id) { int ret; - if (!addr_is_memory(addr)) + if (!range_is_memory(addr, addr + size)) return -EPERM; ret = host_stage2_try(kvm_pgtable_stage2_set_owner, &host_mmu.pgt, From 55dd5b4db3bf04cf077a8d1712f6295d4517c337 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Wed, 7 May 2025 11:49:41 +0200 Subject: [PATCH 221/555] udf: Make sure i_lenExtents is uptodate on inode eviction UDF maintains total length of all extents in i_lenExtents. Generally we keep extent lengths (and thus i_lenExtents) block aligned because it makes the file appending logic simpler. However the standard mandates that the inode size must match the length of all extents and thus we trim the last extent when closing the file. To catch possible bugs we also verify that i_lenExtents matches i_size when evicting inode from memory. Commit b405c1e58b73 ("udf: refactor udf_next_aext() to handle error") however broke the code updating i_lenExtents and thus udf_evict_inode() ended up spewing lots of errors about incorrectly sized extents although the extents were actually sized properly. Fix the updating of i_lenExtents to silence the errors. Fixes: b405c1e58b73 ("udf: refactor udf_next_aext() to handle error") CC: stable@vger.kernel.org Signed-off-by: Jan Kara --- fs/udf/truncate.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/udf/truncate.c b/fs/udf/truncate.c index 4f33a4a48886..b4071c9cf8c9 100644 --- a/fs/udf/truncate.c +++ b/fs/udf/truncate.c @@ -115,7 +115,7 @@ void udf_truncate_tail_extent(struct inode *inode) } /* This inode entry is in-memory only and thus we don't have to mark * the inode dirty */ - if (ret == 0) + if (ret >= 0) iinfo->i_lenExtents = inode->i_size; brelse(epos.bh); } From 42420c50c68f3e95e90de2479464f420602229fc Mon Sep 17 00:00:00 2001 From: Niklas Schnelle Date: Wed, 30 Apr 2025 15:26:18 +0200 Subject: [PATCH 222/555] s390/pci: Fix missing check for zpci_create_device() error return The zpci_create_device() function returns an error pointer that needs to be checked before dereferencing it as a struct zpci_dev pointer. Add the missing check in __clp_add() where it was missed when adding the scan_list in the fixed commit. Simply not adding the device to the scan list results in the previous behavior. Cc: stable@vger.kernel.org Fixes: 0467cdde8c43 ("s390/pci: Sort PCI functions prior to creating virtual busses") Signed-off-by: Niklas Schnelle Reviewed-by: Gerd Bayer Signed-off-by: Heiko Carstens --- arch/s390/pci/pci_clp.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/s390/pci/pci_clp.c b/arch/s390/pci/pci_clp.c index 9a929bbcc397..241f7251c873 100644 --- a/arch/s390/pci/pci_clp.c +++ b/arch/s390/pci/pci_clp.c @@ -428,6 +428,8 @@ static void __clp_add(struct clp_fh_list_entry *entry, void *data) return; } zdev = zpci_create_device(entry->fid, entry->fh, entry->config_state); + if (IS_ERR(zdev)) + return; list_add_tail(&zdev->entry, scan_list); } From 05a2538f2b48500cf4e8a0a0ce76623cc5bafcf1 Mon Sep 17 00:00:00 2001 From: Niklas Schnelle Date: Wed, 30 Apr 2025 15:26:19 +0200 Subject: [PATCH 223/555] s390/pci: Fix duplicate pci_dev_put() in disable_slot() when PF has child VFs With commit bcb5d6c76903 ("s390/pci: introduce lock to synchronize state of zpci_dev's") the code to ignore power off of a PF that has child VFs was changed from a direct return to a goto to the unlock and pci_dev_put() section. The change however left the existing pci_dev_put() untouched resulting in a doubple put. This can subsequently cause a use after free if the struct pci_dev is released in an unexpected state. Fix this by removing the extra pci_dev_put(). Cc: stable@vger.kernel.org Fixes: bcb5d6c76903 ("s390/pci: introduce lock to synchronize state of zpci_dev's") Signed-off-by: Niklas Schnelle Reviewed-by: Gerd Bayer Signed-off-by: Heiko Carstens --- drivers/pci/hotplug/s390_pci_hpc.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/pci/hotplug/s390_pci_hpc.c b/drivers/pci/hotplug/s390_pci_hpc.c index 055518ee354d..e9e9aaa91770 100644 --- a/drivers/pci/hotplug/s390_pci_hpc.c +++ b/drivers/pci/hotplug/s390_pci_hpc.c @@ -59,7 +59,6 @@ static int disable_slot(struct hotplug_slot *hotplug_slot) pdev = pci_get_slot(zdev->zbus->bus, zdev->devfn); if (pdev && pci_num_vf(pdev)) { - pci_dev_put(pdev); rc = -EBUSY; goto out; } From a032f29a15412fab9f4352e0032836d51420a338 Mon Sep 17 00:00:00 2001 From: John Chau Date: Mon, 5 May 2025 01:55:13 +0900 Subject: [PATCH 224/555] platform/x86: thinkpad_acpi: Support also NEC Lavie X1475JAS MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Change get_thinkpad_model_data() to check for additional vendor name "NEC" in order to support NEC Lavie X1475JAS notebook (and perhaps more). The reason of this works with minimal changes is because NEC Lavie X1475JAS is a Thinkpad inside. ACPI dumps reveals its OEM ID to be "LENOVO", BIOS version "R2PET30W" matches typical Lenovo BIOS version, the existence of HKEY of LEN0268, with DMI fw string is "R2PHT24W". I compiled and tested with my own machine, attached the dmesg below as proof of work: [ 6.288932] thinkpad_acpi: ThinkPad ACPI Extras v0.26 [ 6.288937] thinkpad_acpi: http://ibm-acpi.sf.net/ [ 6.288938] thinkpad_acpi: ThinkPad BIOS R2PET30W (1.11 ), EC R2PHT24W [ 6.307000] thinkpad_acpi: radio switch found; radios are enabled [ 6.307030] thinkpad_acpi: This ThinkPad has standard ACPI backlight brightness control, supported by the ACPI video driver [ 6.307033] thinkpad_acpi: Disabling thinkpad-acpi brightness events by default... [ 6.320322] thinkpad_acpi: rfkill switch tpacpi_bluetooth_sw: radio is unblocked [ 6.371963] thinkpad_acpi: secondary fan control detected & enabled [ 6.391922] thinkpad_acpi: battery 1 registered (start 0, stop 85, behaviours: 0x7) [ 6.398375] input: ThinkPad Extra Buttons as /devices/platform/thinkpad_acpi/input/input13 Signed-off-by: John Chau Link: https://lore.kernel.org/r/20250504165513.295135-1-johnchau@0atlas.com Reviewed-by: Ilpo Järvinen Signed-off-by: Ilpo Järvinen --- drivers/platform/x86/thinkpad_acpi.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/platform/x86/thinkpad_acpi.c b/drivers/platform/x86/thinkpad_acpi.c index 5790095c175e..92b21e49faf6 100644 --- a/drivers/platform/x86/thinkpad_acpi.c +++ b/drivers/platform/x86/thinkpad_acpi.c @@ -11478,6 +11478,8 @@ static int __must_check __init get_thinkpad_model_data( tp->vendor = PCI_VENDOR_ID_IBM; else if (dmi_name_in_vendors("LENOVO")) tp->vendor = PCI_VENDOR_ID_LENOVO; + else if (dmi_name_in_vendors("NEC")) + tp->vendor = PCI_VENDOR_ID_LENOVO; else return 0; From 0887817e4953885fbd6a5c1bec2fdd339261eb19 Mon Sep 17 00:00:00 2001 From: Runhua He Date: Wed, 7 May 2025 18:01:03 +0800 Subject: [PATCH 225/555] platform/x86/amd/pmc: Declare quirk_spurious_8042 for MECHREVO Wujie 14XA (GX4HRXL) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit MECHREVO Wujie 14XA (GX4HRXL) wakes up immediately after s2idle entry. This happens regardless of whether the laptop is plugged into AC power, or whether any peripheral is plugged into the laptop. Similar to commit a55bdad5dfd1 ("platform/x86/amd/pmc: Disable keyboard wakeup on AMD Framework 13"), the MECHREVO Wujie 14XA wakes up almost instantly after s2idle suspend entry (IRQ1 is the keyboard): 2025-04-18 17:23:57,588 DEBUG: PM: Triggering wakeup from IRQ 9 2025-04-18 17:23:57,588 DEBUG: PM: Triggering wakeup from IRQ 1 Add this model to the spurious_8042 quirk to workaround this. This patch does not affect the wake-up function of the built-in keyboard. Because the firmware of this machine adds an insurance for keyboard wake-up events, as it always triggers an additional IRQ 9 to wake up the system. Suggested-by: Mingcong Bai Suggested-by: Xinhui Yang Suggested-by: Rong Zhang Fixes: a55bdad5dfd1 ("platform/x86/amd/pmc: Disable keyboard wakeup on AMD Framework 13") Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4166 Cc: Mario Limonciello Link: https://zhuanldan.zhihu.com/p/730538041 Tested-by: Yemu Lu Signed-off-by: Runhua He Link: https://lore.kernel.org/r/20250507100103.995395-1-hua@aosc.io Reviewed-by: Ilpo Järvinen Signed-off-by: Ilpo Järvinen --- drivers/platform/x86/amd/pmc/pmc-quirks.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/platform/x86/amd/pmc/pmc-quirks.c b/drivers/platform/x86/amd/pmc/pmc-quirks.c index b4f49720c87f..2e3f6fc67c56 100644 --- a/drivers/platform/x86/amd/pmc/pmc-quirks.c +++ b/drivers/platform/x86/amd/pmc/pmc-quirks.c @@ -217,6 +217,13 @@ static const struct dmi_system_id fwbug_list[] = { DMI_MATCH(DMI_BIOS_VERSION, "03.05"), } }, + { + .ident = "MECHREVO Wujie 14X (GX4HRXL)", + .driver_data = &quirk_spurious_8042, + .matches = { + DMI_MATCH(DMI_BOARD_NAME, "WUJIE14-GX4HRXL"), + } + }, {} }; From bfcfe6d335a967f8ea0c1980960e6f0205b5de6e Mon Sep 17 00:00:00 2001 From: Hans de Goede Date: Thu, 1 May 2025 15:17:02 +0200 Subject: [PATCH 226/555] platform/x86: asus-wmi: Fix wlan_ctrl_by_user detection MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The wlan_ctrl_by_user detection was introduced by commit a50bd128f28c ("asus-wmi: record wlan status while controlled by userapp"). Quoting from that commit's commit message: """ When you call WMIMethod(DSTS, 0x00010011) to get WLAN status, it may return (1) 0x00050001 (On) (2) 0x00050000 (Off) (3) 0x00030001 (On) (4) 0x00030000 (Off) (5) 0x00000002 (Unknown) (1), (2) means that the model has hardware GPIO for WLAN, you can call WMIMethod(DEVS, 0x00010011, 1 or 0) to turn WLAN on/off. (3), (4) means that the model doesn’t have hardware GPIO, you need to use API or driver library to turn WLAN on/off, and call WMIMethod(DEVS, 0x00010012, 1 or 0) to set WLAN LED status. After you set WLAN LED status, you can see the WLAN status is changed with WMIMethod(DSTS, 0x00010011). Because the status is recorded lastly (ex: Windows), you can use it for synchronization. (5) means that the model doesn’t have WLAN device. WLAN is the ONLY special case with upper rule. """ The wlan_ctrl_by_user flag should be set on 0x0003000? ((3), (4) above) return values, but the flag mistakenly also gets set on laptops with 0x0005000? ((1), (2)) return values. This is causing rfkill problems on laptops where 0x0005000? is returned. Fix the check to only set the wlan_ctrl_by_user flag for 0x0003000? return values. Fixes: a50bd128f28c ("asus-wmi: record wlan status while controlled by userapp") Link: https://bugzilla.kernel.org/show_bug.cgi?id=219786 Signed-off-by: Hans de Goede Reviewed-by: Armin Wolf Link: https://lore.kernel.org/r/20250501131702.103360-2-hdegoede@redhat.com Reviewed-by: Ilpo Järvinen Signed-off-by: Ilpo Järvinen --- drivers/platform/x86/asus-wmi.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/platform/x86/asus-wmi.c b/drivers/platform/x86/asus-wmi.c index 0c697b46f436..47cc766624d7 100644 --- a/drivers/platform/x86/asus-wmi.c +++ b/drivers/platform/x86/asus-wmi.c @@ -4779,7 +4779,8 @@ static int asus_wmi_add(struct platform_device *pdev) goto fail_leds; asus_wmi_get_devstate(asus, ASUS_WMI_DEVID_WLAN, &result); - if (result & (ASUS_WMI_DSTS_PRESENCE_BIT | ASUS_WMI_DSTS_USER_BIT)) + if ((result & (ASUS_WMI_DSTS_PRESENCE_BIT | ASUS_WMI_DSTS_USER_BIT)) == + (ASUS_WMI_DSTS_PRESENCE_BIT | ASUS_WMI_DSTS_USER_BIT)) asus->driver->wlan_ctrl_by_user = 1; if (!(asus->driver->wlan_ctrl_by_user && ashs_present())) { From cd9c058489053e172a6654cad82ee936d1b09fab Mon Sep 17 00:00:00 2001 From: John Ernberg Date: Fri, 2 May 2025 11:40:55 +0000 Subject: [PATCH 227/555] xen: swiotlb: Use swiotlb bouncing if kmalloc allocation demands it Xen swiotlb support was missed when the patch set starting with 4ab5f8ec7d71 ("mm/slab: decouple ARCH_KMALLOC_MINALIGN from ARCH_DMA_MINALIGN") was merged. When running Xen on iMX8QXP, a SoC without IOMMU, the effect was that USB transfers ended up corrupted when there was more than one URB inflight at the same time. Add a call to dma_kmalloc_needs_bounce() to make sure that allocations too small for DMA get bounced via swiotlb. Closes: https://lore.kernel.org/linux-usb/ab2776f0-b838-4cf6-a12a-c208eb6aad59@actia.se/ Fixes: 4ab5f8ec7d71 ("mm/slab: decouple ARCH_KMALLOC_MINALIGN from ARCH_DMA_MINALIGN") Cc: stable@kernel.org # v6.5+ Signed-off-by: John Ernberg Reviewed-by: Stefano Stabellini Signed-off-by: Juergen Gross Message-ID: <20250502114043.1968976-2-john.ernberg@actia.se> --- drivers/xen/swiotlb-xen.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/xen/swiotlb-xen.c b/drivers/xen/swiotlb-xen.c index 1f65795cf5d7..ef56a2500ed6 100644 --- a/drivers/xen/swiotlb-xen.c +++ b/drivers/xen/swiotlb-xen.c @@ -217,6 +217,7 @@ static dma_addr_t xen_swiotlb_map_page(struct device *dev, struct page *page, * buffering it. */ if (dma_capable(dev, dev_addr, size, true) && + !dma_kmalloc_needs_bounce(dev, size, dir) && !range_straddles_page_boundary(phys, size) && !xen_arch_need_swiotlb(dev, phys, dev_addr) && !is_swiotlb_force_bounce(dev)) From 687b2bae0efff9b25e071737d6af5004e6e35af5 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Wed, 7 May 2025 07:34:24 -0600 Subject: [PATCH 228/555] io_uring: ensure deferred completions are flushed for multishot Multishot normally uses io_req_post_cqe() to post completions, but when stopping it, it may finish up with a deferred completion. This is fine, except if another multishot event triggers before the deferred completions get flushed. If this occurs, then CQEs may get reordered in the CQ ring, as new multishot completions get posted before the deferred ones are flushed. This can cause confusion on the application side, if strict ordering is required for the use case. When multishot posting via io_req_post_cqe(), flush any pending deferred completions first, if any. Cc: stable@vger.kernel.org # 6.1+ Reported-by: Norman Maurer Reported-by: Christian Mazakas Signed-off-by: Jens Axboe --- io_uring/io_uring.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c index 769814d71153..541e65a1eebf 100644 --- a/io_uring/io_uring.c +++ b/io_uring/io_uring.c @@ -848,6 +848,14 @@ bool io_req_post_cqe(struct io_kiocb *req, s32 res, u32 cflags) struct io_ring_ctx *ctx = req->ctx; bool posted; + /* + * If multishot has already posted deferred completions, ensure that + * those are flushed first before posting this one. If not, CQEs + * could get reordered. + */ + if (!wq_list_empty(&ctx->submit_state.compl_reqs)) + __io_submit_flush_completions(ctx); + lockdep_assert(!io_wq_current_is_worker()); lockdep_assert_held(&ctx->uring_lock); From 90989869baae47ee2aa3bcb6f6eb9fbbe4287958 Mon Sep 17 00:00:00 2001 From: Jason Andryuk Date: Tue, 6 May 2025 16:44:56 -0400 Subject: [PATCH 229/555] xenbus: Allow PVH dom0 a non-local xenstore Make xenbus_init() allow a non-local xenstore for a PVH dom0 - it is currently forced to XS_LOCAL. With Hyperlaunch booting dom0 and a xenstore stubdom, dom0 can be handled as a regular XS_HVM following the late init path. Ideally we'd drop the use of xen_initial_domain() and just check for the event channel instead. However, ARM has a xen,enhanced no-xenstore mode, where the event channel and PFN would both be 0. Retain the xen_initial_domain() check, and use that for an additional check when the event channel is 0. Check the full 64bit HVM_PARAM_STORE_EVTCHN value to catch the off chance that high bits are set for the 32bit event channel. Signed-off-by: Jason Andryuk Change-Id: I5506da42e4c6b8e85079fefb2f193c8de17c7437 Reviewed-by: Stefano Stabellini Signed-off-by: Juergen Gross Message-ID: <20250506204456.5220-1-jason.andryuk@amd.com> --- drivers/xen/xenbus/xenbus_probe.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) diff --git a/drivers/xen/xenbus/xenbus_probe.c b/drivers/xen/xenbus/xenbus_probe.c index 6d32ffb01136..86fe6e779056 100644 --- a/drivers/xen/xenbus/xenbus_probe.c +++ b/drivers/xen/xenbus/xenbus_probe.c @@ -966,9 +966,15 @@ static int __init xenbus_init(void) if (xen_pv_domain()) xen_store_domain_type = XS_PV; if (xen_hvm_domain()) + { xen_store_domain_type = XS_HVM; - if (xen_hvm_domain() && xen_initial_domain()) - xen_store_domain_type = XS_LOCAL; + err = hvm_get_parameter(HVM_PARAM_STORE_EVTCHN, &v); + if (err) + goto out_error; + xen_store_evtchn = (int)v; + if (!v && xen_initial_domain()) + xen_store_domain_type = XS_LOCAL; + } if (xen_pv_domain() && !xen_start_info->store_evtchn) xen_store_domain_type = XS_LOCAL; if (xen_pv_domain() && xen_start_info->store_evtchn) @@ -987,10 +993,6 @@ static int __init xenbus_init(void) xen_store_interface = gfn_to_virt(xen_store_gfn); break; case XS_HVM: - err = hvm_get_parameter(HVM_PARAM_STORE_EVTCHN, &v); - if (err) - goto out_error; - xen_store_evtchn = (int)v; err = hvm_get_parameter(HVM_PARAM_STORE_PFN, &v); if (err) goto out_error; From 1f0304dfd9d217c2f8b04a9ef4b3258a66eedd27 Mon Sep 17 00:00:00 2001 From: Jason Andryuk Date: Tue, 6 May 2025 17:09:33 -0400 Subject: [PATCH 230/555] xenbus: Use kref to track req lifetime MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Marek reported seeing a NULL pointer fault in the xenbus_thread callstack: BUG: kernel NULL pointer dereference, address: 0000000000000000 RIP: e030:__wake_up_common+0x4c/0x180 Call Trace: __wake_up_common_lock+0x82/0xd0 process_msg+0x18e/0x2f0 xenbus_thread+0x165/0x1c0 process_msg+0x18e is req->cb(req). req->cb is set to xs_wake_up(), a thin wrapper around wake_up(), or xenbus_dev_queue_reply(). It seems like it was xs_wake_up() in this case. It seems like req may have woken up the xs_wait_for_reply(), which kfree()ed the req. When xenbus_thread resumes, it faults on the zero-ed data. Linux Device Drivers 2nd edition states: "Normally, a wake_up call can cause an immediate reschedule to happen, meaning that other processes might run before wake_up returns." ... which would match the behaviour observed. Change to keeping two krefs on each request. One for the caller, and one for xenbus_thread. Each will kref_put() when finished, and the last will free it. This use of kref matches the description in Documentation/core-api/kref.rst Link: https://lore.kernel.org/xen-devel/ZO0WrR5J0xuwDIxW@mail-itl/ Reported-by: Marek Marczykowski-Górecki Fixes: fd8aa9095a95 ("xen: optimize xenbus driver for multiple concurrent xenstore accesses") Cc: stable@vger.kernel.org Signed-off-by: Jason Andryuk Reviewed-by: Juergen Gross Signed-off-by: Juergen Gross Message-ID: <20250506210935.5607-1-jason.andryuk@amd.com> --- drivers/xen/xenbus/xenbus.h | 2 ++ drivers/xen/xenbus/xenbus_comms.c | 9 ++++----- drivers/xen/xenbus/xenbus_dev_frontend.c | 2 +- drivers/xen/xenbus/xenbus_xs.c | 18 ++++++++++++++++-- 4 files changed, 23 insertions(+), 8 deletions(-) diff --git a/drivers/xen/xenbus/xenbus.h b/drivers/xen/xenbus/xenbus.h index 13821e7e825e..9ac0427724a3 100644 --- a/drivers/xen/xenbus/xenbus.h +++ b/drivers/xen/xenbus/xenbus.h @@ -77,6 +77,7 @@ enum xb_req_state { struct xb_req_data { struct list_head list; wait_queue_head_t wq; + struct kref kref; struct xsd_sockmsg msg; uint32_t caller_req_id; enum xsd_sockmsg_type type; @@ -103,6 +104,7 @@ int xb_init_comms(void); void xb_deinit_comms(void); int xs_watch_msg(struct xs_watch_event *event); void xs_request_exit(struct xb_req_data *req); +void xs_free_req(struct kref *kref); int xenbus_match(struct device *_dev, const struct device_driver *_drv); int xenbus_dev_probe(struct device *_dev); diff --git a/drivers/xen/xenbus/xenbus_comms.c b/drivers/xen/xenbus/xenbus_comms.c index e5fda0256feb..82df2da1b880 100644 --- a/drivers/xen/xenbus/xenbus_comms.c +++ b/drivers/xen/xenbus/xenbus_comms.c @@ -309,8 +309,8 @@ static int process_msg(void) virt_wmb(); req->state = xb_req_state_got_reply; req->cb(req); - } else - kfree(req); + } + kref_put(&req->kref, xs_free_req); } mutex_unlock(&xs_response_mutex); @@ -386,14 +386,13 @@ static int process_writes(void) state.req->msg.type = XS_ERROR; state.req->err = err; list_del(&state.req->list); - if (state.req->state == xb_req_state_aborted) - kfree(state.req); - else { + if (state.req->state != xb_req_state_aborted) { /* write err, then update state */ virt_wmb(); state.req->state = xb_req_state_got_reply; wake_up(&state.req->wq); } + kref_put(&state.req->kref, xs_free_req); mutex_unlock(&xb_write_mutex); diff --git a/drivers/xen/xenbus/xenbus_dev_frontend.c b/drivers/xen/xenbus/xenbus_dev_frontend.c index 46f8916597e5..f5c21ba64df5 100644 --- a/drivers/xen/xenbus/xenbus_dev_frontend.c +++ b/drivers/xen/xenbus/xenbus_dev_frontend.c @@ -406,7 +406,7 @@ void xenbus_dev_queue_reply(struct xb_req_data *req) mutex_unlock(&u->reply_mutex); kfree(req->body); - kfree(req); + kref_put(&req->kref, xs_free_req); kref_put(&u->kref, xenbus_file_free); diff --git a/drivers/xen/xenbus/xenbus_xs.c b/drivers/xen/xenbus/xenbus_xs.c index d32c726f7a12..dcf9182c8451 100644 --- a/drivers/xen/xenbus/xenbus_xs.c +++ b/drivers/xen/xenbus/xenbus_xs.c @@ -112,6 +112,12 @@ static void xs_suspend_exit(void) wake_up_all(&xs_state_enter_wq); } +void xs_free_req(struct kref *kref) +{ + struct xb_req_data *req = container_of(kref, struct xb_req_data, kref); + kfree(req); +} + static uint32_t xs_request_enter(struct xb_req_data *req) { uint32_t rq_id; @@ -237,6 +243,12 @@ static void xs_send(struct xb_req_data *req, struct xsd_sockmsg *msg) req->caller_req_id = req->msg.req_id; req->msg.req_id = xs_request_enter(req); + /* + * Take 2nd ref. One for this thread, and the second for the + * xenbus_thread. + */ + kref_get(&req->kref); + mutex_lock(&xb_write_mutex); list_add_tail(&req->list, &xb_write_list); notify = list_is_singular(&xb_write_list); @@ -261,8 +273,8 @@ static void *xs_wait_for_reply(struct xb_req_data *req, struct xsd_sockmsg *msg) if (req->state == xb_req_state_queued || req->state == xb_req_state_wait_reply) req->state = xb_req_state_aborted; - else - kfree(req); + + kref_put(&req->kref, xs_free_req); mutex_unlock(&xb_write_mutex); return ret; @@ -291,6 +303,7 @@ int xenbus_dev_request_and_reply(struct xsd_sockmsg *msg, void *par) req->cb = xenbus_dev_queue_reply; req->par = par; req->user_req = true; + kref_init(&req->kref); xs_send(req, msg); @@ -319,6 +332,7 @@ static void *xs_talkv(struct xenbus_transaction t, req->num_vecs = num_vecs; req->cb = xs_wake_up; req->user_req = false; + kref_init(&req->kref); msg.req_id = 0; msg.tx_id = t.id; From 428dc9fc0873989d73918d4a9cc22745b7bbc799 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 5 May 2025 11:30:39 -1000 Subject: [PATCH 231/555] sched_ext: bpf_iter_scx_dsq_new() should always initialize iterator BPF programs may call next() and destroy() on BPF iterators even after new() returns an error value (e.g. bpf_for_each() macro ignores error returns from new()). bpf_iter_scx_dsq_new() could leave the iterator in an uninitialized state after an error return causing bpf_iter_scx_dsq_next() to dereference garbage data. Make bpf_iter_scx_dsq_new() always clear $kit->dsq so that next() and destroy() become noops. Signed-off-by: Tejun Heo Fixes: 650ba21b131e ("sched_ext: Implement DSQ iterator") Cc: stable@vger.kernel.org # v6.12+ Acked-by: Andrea Righi --- kernel/sched/ext.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c index 4e37b40ce280..f5133249fd4d 100644 --- a/kernel/sched/ext.c +++ b/kernel/sched/ext.c @@ -6827,6 +6827,12 @@ __bpf_kfunc int bpf_iter_scx_dsq_new(struct bpf_iter_scx_dsq *it, u64 dsq_id, BUILD_BUG_ON(__alignof__(struct bpf_iter_scx_dsq_kern) != __alignof__(struct bpf_iter_scx_dsq)); + /* + * next() and destroy() will be called regardless of the return value. + * Always clear $kit->dsq. + */ + kit->dsq = NULL; + if (flags & ~__SCX_DSQ_ITER_USER_FLAGS) return -EINVAL; From 1e2e3044c1bc64a64aa0eaf7c17f7832c26c9775 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Tue, 29 Apr 2025 15:05:59 -0400 Subject: [PATCH 232/555] Bluetooth: MGMT: Fix MGMT_OP_ADD_DEVICE invalid device flags Device flags could be updated in the meantime while MGMT_OP_ADD_DEVICE is pending on hci_update_passive_scan_sync so instead of setting the current_flags as cmd->user_data just do a lookup using hci_conn_params_lookup and use the latest stored flags. Fixes: a182d9c84f9c ("Bluetooth: MGMT: Fix Add Device to responding before completing") Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/mgmt.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index c1e1e529e26c..46b22708dfbd 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -7506,11 +7506,16 @@ static void add_device_complete(struct hci_dev *hdev, void *data, int err) struct mgmt_cp_add_device *cp = cmd->param; if (!err) { + struct hci_conn_params *params; + + params = hci_conn_params_lookup(hdev, &cp->addr.bdaddr, + le_addr_type(cp->addr.type)); + device_added(cmd->sk, hdev, &cp->addr.bdaddr, cp->addr.type, cp->action); device_flags_changed(NULL, hdev, &cp->addr.bdaddr, cp->addr.type, hdev->conn_flags, - PTR_UINT(cmd->user_data)); + params ? params->flags : 0); } mgmt_cmd_complete(cmd->sk, hdev->id, MGMT_OP_ADD_DEVICE, @@ -7613,8 +7618,6 @@ static int add_device(struct sock *sk, struct hci_dev *hdev, goto unlock; } - cmd->user_data = UINT_PTR(current_flags); - err = hci_cmd_sync_queue(hdev, add_device_sync, cmd, add_device_complete); if (err < 0) { From a6aeb739974ec73e5217c75a7c008a688d3d5cf1 Mon Sep 17 00:00:00 2001 From: Dmitry Antipov Date: Wed, 7 May 2025 09:50:44 +0300 Subject: [PATCH 233/555] module: ensure that kobject_put() is safe for module type kobjects In 'lookup_or_create_module_kobject()', an internal kobject is created using 'module_ktype'. So call to 'kobject_put()' on error handling path causes an attempt to use an uninitialized completion pointer in 'module_kobject_release()'. In this scenario, we just want to release kobject without an extra synchronization required for a regular module unloading process, so adding an extra check whether 'complete()' is actually required makes 'kobject_put()' safe. Reported-by: syzbot+7fb8a372e1f6add936dd@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=7fb8a372e1f6add936dd Fixes: 942e443127e9 ("module: Fix mod->mkobj.kobj potentially freed too early") Cc: stable@vger.kernel.org Suggested-by: Petr Pavlu Signed-off-by: Dmitry Antipov Link: https://lore.kernel.org/r/20250507065044.86529-1-dmantipov@yandex.ru Signed-off-by: Petr Pavlu --- kernel/params.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/kernel/params.c b/kernel/params.c index e668fc90b83e..b92d64161b75 100644 --- a/kernel/params.c +++ b/kernel/params.c @@ -943,7 +943,9 @@ struct kset *module_kset; static void module_kobj_release(struct kobject *kobj) { struct module_kobject *mk = to_module_kobject(kobj); - complete(mk->kobj_completion); + + if (mk->kobj_completion) + complete(mk->kobj_completion); } const struct kobj_type module_ktype = { From 3ca02e63edccb78ef3659bebc68579c7224a6ca2 Mon Sep 17 00:00:00 2001 From: Paul Aurich Date: Tue, 6 May 2025 22:28:09 -0700 Subject: [PATCH 234/555] smb: client: Avoid race in open_cached_dir with lease breaks A pre-existing valid cfid returned from find_or_create_cached_dir might race with a lease break, meaning open_cached_dir doesn't consider it valid, and thinks it's newly-constructed. This leaks a dentry reference if the allocation occurs before the queued lease break work runs. Avoid the race by extending holding the cfid_list_lock across find_or_create_cached_dir and when the result is checked. Cc: stable@vger.kernel.org Reviewed-by: Henrique Carvalho Signed-off-by: Paul Aurich Signed-off-by: Steve French --- fs/smb/client/cached_dir.c | 10 ++-------- 1 file changed, 2 insertions(+), 8 deletions(-) diff --git a/fs/smb/client/cached_dir.c b/fs/smb/client/cached_dir.c index fe738623cf1b..240d82c6f908 100644 --- a/fs/smb/client/cached_dir.c +++ b/fs/smb/client/cached_dir.c @@ -29,7 +29,6 @@ static struct cached_fid *find_or_create_cached_dir(struct cached_fids *cfids, { struct cached_fid *cfid; - spin_lock(&cfids->cfid_list_lock); list_for_each_entry(cfid, &cfids->entries, entry) { if (!strcmp(cfid->path, path)) { /* @@ -38,25 +37,20 @@ static struct cached_fid *find_or_create_cached_dir(struct cached_fids *cfids, * being deleted due to a lease break. */ if (!cfid->time || !cfid->has_lease) { - spin_unlock(&cfids->cfid_list_lock); return NULL; } kref_get(&cfid->refcount); - spin_unlock(&cfids->cfid_list_lock); return cfid; } } if (lookup_only) { - spin_unlock(&cfids->cfid_list_lock); return NULL; } if (cfids->num_entries >= max_cached_dirs) { - spin_unlock(&cfids->cfid_list_lock); return NULL; } cfid = init_cached_dir(path); if (cfid == NULL) { - spin_unlock(&cfids->cfid_list_lock); return NULL; } cfid->cfids = cfids; @@ -74,7 +68,6 @@ static struct cached_fid *find_or_create_cached_dir(struct cached_fids *cfids, */ cfid->has_lease = true; - spin_unlock(&cfids->cfid_list_lock); return cfid; } @@ -187,8 +180,10 @@ replay_again: if (!utf16_path) return -ENOMEM; + spin_lock(&cfids->cfid_list_lock); cfid = find_or_create_cached_dir(cfids, path, lookup_only, tcon->max_cached_dirs); if (cfid == NULL) { + spin_unlock(&cfids->cfid_list_lock); kfree(utf16_path); return -ENOENT; } @@ -197,7 +192,6 @@ replay_again: * Otherwise, it is either a new entry or laundromat worker removed it * from @cfids->entries. Caller will put last reference if the latter. */ - spin_lock(&cfids->cfid_list_lock); if (cfid->has_lease && cfid->time) { spin_unlock(&cfids->cfid_list_lock); *ret_cfid = cfid; From 473f09f362e5979efbff40c9bbce58892ad39d66 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 6 May 2025 00:22:26 -0400 Subject: [PATCH 235/555] bcachefs: journal_shutdown is EROFS, not EIO We often filter out EROFS errors to avoid log spew after an emergency shutdown - journal_shutdown is just another emergency shutdown error. Signed-off-by: Kent Overstreet --- fs/bcachefs/errcode.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/bcachefs/errcode.h b/fs/bcachefs/errcode.h index a615e4852ded..d9ebffa5b3a2 100644 --- a/fs/bcachefs/errcode.h +++ b/fs/bcachefs/errcode.h @@ -269,7 +269,7 @@ x(BCH_ERR_invalid_sb, invalid_sb_downgrade) \ x(BCH_ERR_invalid, invalid_bkey) \ x(BCH_ERR_operation_blocked, nocow_lock_blocked) \ - x(EIO, journal_shutdown) \ + x(EROFS, journal_shutdown) \ x(EIO, journal_flush_err) \ x(EIO, journal_write_err) \ x(EIO, btree_node_read_err) \ From 2fea3aa76e352cd071bee71e5c43da339639b310 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 7 May 2025 13:50:00 -0400 Subject: [PATCH 236/555] bcachefs: Filter out harmless EROFS error messages These just indicate that we're shutting down. Signed-off-by: Kent Overstreet --- fs/bcachefs/move.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/bcachefs/move.c b/fs/bcachefs/move.c index fc396b9fa754..dfdbb9259985 100644 --- a/fs/bcachefs/move.c +++ b/fs/bcachefs/move.c @@ -784,7 +784,8 @@ static int __bch2_move_data_phys(struct moving_context *ctxt, goto err; ret = bch2_btree_write_buffer_tryflush(trans); - bch_err_msg(c, ret, "flushing btree write buffer"); + if (!bch2_err_matches(ret, EROFS)) + bch_err_msg(c, ret, "flushing btree write buffer"); if (ret) goto err; From da18dabc3784663a088943e613c36cd17aeb52d3 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 7 May 2025 16:54:25 -0400 Subject: [PATCH 237/555] bcachefs: Ensure superblock gets written when we go ERO When we go emergency read-only, make sure we do a final write_super() to persist counters and error counts - this can be critical for piecing together what fsck was doing. Signed-off-by: Kent Overstreet --- fs/bcachefs/super.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c index 27943082c093..84a37d971ffd 100644 --- a/fs/bcachefs/super.c +++ b/fs/bcachefs/super.c @@ -377,6 +377,11 @@ void bch2_fs_read_only(struct bch_fs *c) bch_verbose(c, "marking filesystem clean"); bch2_fs_mark_clean(c); } else { + /* Make sure error counts/counters are persisted */ + mutex_lock(&c->sb_lock); + bch2_write_super(c); + mutex_unlock(&c->sb_lock); + bch_verbose(c, "done going read-only, filesystem not clean"); } } From 8e4d28036c293241b312b1fceafb32b994f80fcc Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 7 May 2025 13:32:15 -0400 Subject: [PATCH 238/555] bcachefs: Don't aggressively discard the journal We frequently use 'bcachefs list_journal -a' for debugging, as it provides a record of all btree transactions, and a history of what happened. But it's not so useful if we immediately discard journal buckets right after they're no longer dirty. This tweaks journal reclaim to only discard when we're low on space, keeping the journal mostly un-discarded. Signed-off-by: Kent Overstreet --- fs/bcachefs/journal_reclaim.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/fs/bcachefs/journal_reclaim.c b/fs/bcachefs/journal_reclaim.c index ea670c3c43d8..976464d8a695 100644 --- a/fs/bcachefs/journal_reclaim.c +++ b/fs/bcachefs/journal_reclaim.c @@ -266,10 +266,11 @@ out: static bool should_discard_bucket(struct journal *j, struct journal_device *ja) { - bool ret; - spin_lock(&j->lock); - ret = ja->discard_idx != ja->dirty_idx_ondisk; + unsigned min_free = max(4, ja->nr / 8); + + bool ret = bch2_journal_dev_buckets_available(j, ja, journal_space_discarded) < min_free && + ja->discard_idx != ja->dirty_idx_ondisk; spin_unlock(&j->lock); return ret; From f04f03d3e99bc8f89b6af5debf07ff67d961bc23 Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Wed, 7 May 2025 14:52:55 -0700 Subject: [PATCH 239/555] Input: synaptics - enable SMBus for HP Elitebook 850 G1 The kernel reports that the touchpad for this device can support SMBus mode. Reported-by: jt Link: https://lore.kernel.org/r/iys5dbv3ldddsgobfkxldazxyp54kay4bozzmagga6emy45jop@2ebvuxgaui4u Cc: stable@vger.kernel.org Signed-off-by: Dmitry Torokhov --- drivers/input/mouse/synaptics.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/input/mouse/synaptics.c b/drivers/input/mouse/synaptics.c index 309c360aab55..fdb471060118 100644 --- a/drivers/input/mouse/synaptics.c +++ b/drivers/input/mouse/synaptics.c @@ -190,6 +190,7 @@ static const char * const smbus_pnp_ids[] = { "LEN2054", /* E480 */ "LEN2055", /* E580 */ "LEN2068", /* T14 Gen 1 */ + "SYN3003", /* HP EliteBook 850 G1 */ "SYN3015", /* HP EliteBook 840 G2 */ "SYN3052", /* HP EliteBook 840 G4 */ "SYN3221", /* HP 15-ay000 */ From 6d7ea0881000966607772451b789b5fb5766f11d Mon Sep 17 00:00:00 2001 From: Manuel Fombuena Date: Wed, 7 May 2025 12:05:26 -0700 Subject: [PATCH 240/555] Input: synaptics - enable InterTouch on Dynabook Portege X30-D [ 5.989588] psmouse serio1: synaptics: Your touchpad (PNP: TOS0213 PNP0f03) says it can support a different bus. If i2c-hid and hid-rmi are not used, you might want to try setting psmouse.synaptics_intertouch to 1 and report this to linux-input@vger.kernel.org. [ 6.039923] psmouse serio1: synaptics: Touchpad model: 1, fw: 9.32, id: 0x1e2a1, caps: 0xf00223/0x840300/0x12e800/0x52d884, board id: 3322, fw id: 2658004 The board is labelled TM3322. Present on the Toshiba / Dynabook Portege X30-D and possibly others. Confirmed working well with psmouse.synaptics_intertouch=1 and local build. Signed-off-by: Manuel Fombuena Signed-off-by: Aditya Garg Link: https://lore.kernel.org/r/PN3PR01MB9597711E7933A08389FEC31DB888A@PN3PR01MB9597.INDPRD01.PROD.OUTLOOK.COM Cc: stable@vger.kernel.org Signed-off-by: Dmitry Torokhov --- drivers/input/mouse/synaptics.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/input/mouse/synaptics.c b/drivers/input/mouse/synaptics.c index fdb471060118..9c63e3c31fa0 100644 --- a/drivers/input/mouse/synaptics.c +++ b/drivers/input/mouse/synaptics.c @@ -196,6 +196,7 @@ static const char * const smbus_pnp_ids[] = { "SYN3221", /* HP 15-ay000 */ "SYN323d", /* HP Spectre X360 13-w013dx */ "SYN3257", /* HP Envy 13-ad105ng */ + "TOS0213", /* Dynabook Portege X30-D */ NULL }; #endif From 47d768b32e644b56901bb4bbbdb1feb01ea86c85 Mon Sep 17 00:00:00 2001 From: Aditya Garg Date: Wed, 7 May 2025 12:06:32 -0700 Subject: [PATCH 241/555] Input: synaptics - enable InterTouch on Dynabook Portege X30L-G Enable InterTouch mode on Dynabook Portege X30L-G by adding "TOS01f6" to the list of SMBus-enabled variants. Reported-by: Xuntao Chi Tested-by: Xuntao Chi Signed-off-by: Aditya Garg Link: https://lore.kernel.org/r/PN3PR01MB959786E4AC797160CDA93012B888A@PN3PR01MB9597.INDPRD01.PROD.OUTLOOK.COM Cc: stable@vger.kernel.org Signed-off-by: Dmitry Torokhov --- drivers/input/mouse/synaptics.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/input/mouse/synaptics.c b/drivers/input/mouse/synaptics.c index 9c63e3c31fa0..9a1592378c1b 100644 --- a/drivers/input/mouse/synaptics.c +++ b/drivers/input/mouse/synaptics.c @@ -196,6 +196,7 @@ static const char * const smbus_pnp_ids[] = { "SYN3221", /* HP 15-ay000 */ "SYN323d", /* HP Spectre X360 13-w013dx */ "SYN3257", /* HP Envy 13-ad105ng */ + "TOS01f6", /* Dynabook Portege X30L-G */ "TOS0213", /* Dynabook Portege X30-D */ NULL }; From 2abc698ac77314e0de5b33a6d96a39c5159d88e4 Mon Sep 17 00:00:00 2001 From: Aditya Garg Date: Wed, 7 May 2025 12:09:00 -0700 Subject: [PATCH 242/555] Input: synaptics - enable InterTouch on TUXEDO InfinityBook Pro 14 v5 Enable InterTouch mode on TUXEDO InfinityBook Pro 14 v5 by adding "SYN1221" to the list of SMBus-enabled variants. Add support for InterTouch on SYN1221 by adding it to the list of SMBus-enabled variants. Reported-by: Matthias Eilert Tested-by: Matthias Eilert Signed-off-by: Aditya Garg Link: https://lore.kernel.org/r/PN3PR01MB9597C033C4BC20EE2A0C4543B888A@PN3PR01MB9597.INDPRD01.PROD.OUTLOOK.COM Cc: stable@vger.kernel.org Signed-off-by: Dmitry Torokhov --- drivers/input/mouse/synaptics.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/input/mouse/synaptics.c b/drivers/input/mouse/synaptics.c index 9a1592378c1b..a582bf6540c5 100644 --- a/drivers/input/mouse/synaptics.c +++ b/drivers/input/mouse/synaptics.c @@ -190,6 +190,7 @@ static const char * const smbus_pnp_ids[] = { "LEN2054", /* E480 */ "LEN2055", /* E580 */ "LEN2068", /* T14 Gen 1 */ + "SYN1221", /* TUXEDO InfinityBook Pro 14 v5 */ "SYN3003", /* HP EliteBook 850 G1 */ "SYN3015", /* HP EliteBook 840 G2 */ "SYN3052", /* HP EliteBook 840 G4 */ From a609cb4cc07aa9ab8f50466622814356c06f2c17 Mon Sep 17 00:00:00 2001 From: Aditya Garg Date: Wed, 7 May 2025 12:12:15 -0700 Subject: [PATCH 243/555] Input: synaptics - enable InterTouch on Dell Precision M3800 Enable InterTouch mode on Dell Precision M3800 by adding "DLL060d" to the list of SMBus-enabled variants. Reported-by: Markus Rathgeb Signed-off-by: Aditya Garg Link: https://lore.kernel.org/r/PN3PR01MB959789DD6D574E16141E5DC4B888A@PN3PR01MB9597.INDPRD01.PROD.OUTLOOK.COM Cc: stable@vger.kernel.org Signed-off-by: Dmitry Torokhov --- drivers/input/mouse/synaptics.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/input/mouse/synaptics.c b/drivers/input/mouse/synaptics.c index a582bf6540c5..c5c88a75a019 100644 --- a/drivers/input/mouse/synaptics.c +++ b/drivers/input/mouse/synaptics.c @@ -164,6 +164,7 @@ static const char * const topbuttonpad_pnp_ids[] = { #ifdef CONFIG_MOUSE_PS2_SYNAPTICS_SMBUS static const char * const smbus_pnp_ids[] = { /* all of the topbuttonpad_pnp_ids are valid, we just add some extras */ + "DLL060d", /* Dell Precision M3800 */ "LEN0048", /* X1 Carbon 3 */ "LEN0046", /* X250 */ "LEN0049", /* Yoga 11e */ From e34090d7214e0516eb8722aee295cb2507317c07 Mon Sep 17 00:00:00 2001 From: Julian Anastasov Date: Sat, 3 May 2025 01:01:18 +0300 Subject: [PATCH 244/555] ipvs: fix uninit-value for saddr in do_output_route4 syzbot reports for uninit-value for the saddr argument [1]. commit 4754957f04f5 ("ipvs: do not use random local source address for tunnels") already implies that the input value of saddr should be ignored but the code is still reading it which can prevent to connect the route. Fix it by changing the argument to ret_saddr. [1] BUG: KMSAN: uninit-value in do_output_route4+0x42c/0x4d0 net/netfilter/ipvs/ip_vs_xmit.c:147 do_output_route4+0x42c/0x4d0 net/netfilter/ipvs/ip_vs_xmit.c:147 __ip_vs_get_out_rt+0x403/0x21d0 net/netfilter/ipvs/ip_vs_xmit.c:330 ip_vs_tunnel_xmit+0x205/0x2380 net/netfilter/ipvs/ip_vs_xmit.c:1136 ip_vs_in_hook+0x1aa5/0x35b0 net/netfilter/ipvs/ip_vs_core.c:2063 nf_hook_entry_hookfn include/linux/netfilter.h:154 [inline] nf_hook_slow+0xf7/0x400 net/netfilter/core.c:626 nf_hook include/linux/netfilter.h:269 [inline] __ip_local_out+0x758/0x7e0 net/ipv4/ip_output.c:118 ip_local_out net/ipv4/ip_output.c:127 [inline] ip_send_skb+0x6a/0x3c0 net/ipv4/ip_output.c:1501 udp_send_skb+0xfda/0x1b70 net/ipv4/udp.c:1195 udp_sendmsg+0x2fe3/0x33c0 net/ipv4/udp.c:1483 inet_sendmsg+0x1fc/0x280 net/ipv4/af_inet.c:851 sock_sendmsg_nosec net/socket.c:712 [inline] __sock_sendmsg+0x267/0x380 net/socket.c:727 ____sys_sendmsg+0x91b/0xda0 net/socket.c:2566 ___sys_sendmsg+0x28d/0x3c0 net/socket.c:2620 __sys_sendmmsg+0x41d/0x880 net/socket.c:2702 __compat_sys_sendmmsg net/compat.c:360 [inline] __do_compat_sys_sendmmsg net/compat.c:367 [inline] __se_compat_sys_sendmmsg net/compat.c:364 [inline] __ia32_compat_sys_sendmmsg+0xc8/0x140 net/compat.c:364 ia32_sys_call+0x3ffa/0x41f0 arch/x86/include/generated/asm/syscalls_32.h:346 do_syscall_32_irqs_on arch/x86/entry/syscall_32.c:83 [inline] __do_fast_syscall_32+0xb0/0x110 arch/x86/entry/syscall_32.c:306 do_fast_syscall_32+0x38/0x80 arch/x86/entry/syscall_32.c:331 do_SYSENTER_32+0x1f/0x30 arch/x86/entry/syscall_32.c:369 entry_SYSENTER_compat_after_hwframe+0x84/0x8e Uninit was created at: slab_post_alloc_hook mm/slub.c:4167 [inline] slab_alloc_node mm/slub.c:4210 [inline] __kmalloc_cache_noprof+0x8fa/0xe00 mm/slub.c:4367 kmalloc_noprof include/linux/slab.h:905 [inline] ip_vs_dest_dst_alloc net/netfilter/ipvs/ip_vs_xmit.c:61 [inline] __ip_vs_get_out_rt+0x35d/0x21d0 net/netfilter/ipvs/ip_vs_xmit.c:323 ip_vs_tunnel_xmit+0x205/0x2380 net/netfilter/ipvs/ip_vs_xmit.c:1136 ip_vs_in_hook+0x1aa5/0x35b0 net/netfilter/ipvs/ip_vs_core.c:2063 nf_hook_entry_hookfn include/linux/netfilter.h:154 [inline] nf_hook_slow+0xf7/0x400 net/netfilter/core.c:626 nf_hook include/linux/netfilter.h:269 [inline] __ip_local_out+0x758/0x7e0 net/ipv4/ip_output.c:118 ip_local_out net/ipv4/ip_output.c:127 [inline] ip_send_skb+0x6a/0x3c0 net/ipv4/ip_output.c:1501 udp_send_skb+0xfda/0x1b70 net/ipv4/udp.c:1195 udp_sendmsg+0x2fe3/0x33c0 net/ipv4/udp.c:1483 inet_sendmsg+0x1fc/0x280 net/ipv4/af_inet.c:851 sock_sendmsg_nosec net/socket.c:712 [inline] __sock_sendmsg+0x267/0x380 net/socket.c:727 ____sys_sendmsg+0x91b/0xda0 net/socket.c:2566 ___sys_sendmsg+0x28d/0x3c0 net/socket.c:2620 __sys_sendmmsg+0x41d/0x880 net/socket.c:2702 __compat_sys_sendmmsg net/compat.c:360 [inline] __do_compat_sys_sendmmsg net/compat.c:367 [inline] __se_compat_sys_sendmmsg net/compat.c:364 [inline] __ia32_compat_sys_sendmmsg+0xc8/0x140 net/compat.c:364 ia32_sys_call+0x3ffa/0x41f0 arch/x86/include/generated/asm/syscalls_32.h:346 do_syscall_32_irqs_on arch/x86/entry/syscall_32.c:83 [inline] __do_fast_syscall_32+0xb0/0x110 arch/x86/entry/syscall_32.c:306 do_fast_syscall_32+0x38/0x80 arch/x86/entry/syscall_32.c:331 do_SYSENTER_32+0x1f/0x30 arch/x86/entry/syscall_32.c:369 entry_SYSENTER_compat_after_hwframe+0x84/0x8e CPU: 0 UID: 0 PID: 22408 Comm: syz.4.5165 Not tainted 6.15.0-rc3-syzkaller-00019-gbc3372351d0c #0 PREEMPT(undef) Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 02/12/2025 Reported-by: syzbot+04b9a82855c8aed20860@syzkaller.appspotmail.com Closes: https://lore.kernel.org/netdev/68138dfa.050a0220.14dd7d.0017.GAE@google.com/ Fixes: 4754957f04f5 ("ipvs: do not use random local source address for tunnels") Signed-off-by: Julian Anastasov Acked-by: Simon Horman Signed-off-by: Pablo Neira Ayuso --- net/netfilter/ipvs/ip_vs_xmit.c | 27 ++++++++------------------- 1 file changed, 8 insertions(+), 19 deletions(-) diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index 3313bceb6cc9..014f07740369 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -119,13 +119,12 @@ __mtu_check_toobig_v6(const struct sk_buff *skb, u32 mtu) return false; } -/* Get route to daddr, update *saddr, optionally bind route to saddr */ +/* Get route to daddr, optionally bind route to saddr */ static struct rtable *do_output_route4(struct net *net, __be32 daddr, - int rt_mode, __be32 *saddr) + int rt_mode, __be32 *ret_saddr) { struct flowi4 fl4; struct rtable *rt; - bool loop = false; memset(&fl4, 0, sizeof(fl4)); fl4.daddr = daddr; @@ -135,23 +134,17 @@ static struct rtable *do_output_route4(struct net *net, __be32 daddr, retry: rt = ip_route_output_key(net, &fl4); if (IS_ERR(rt)) { - /* Invalid saddr ? */ - if (PTR_ERR(rt) == -EINVAL && *saddr && - rt_mode & IP_VS_RT_MODE_CONNECT && !loop) { - *saddr = 0; - flowi4_update_output(&fl4, 0, daddr, 0); - goto retry; - } IP_VS_DBG_RL("ip_route_output error, dest: %pI4\n", &daddr); return NULL; - } else if (!*saddr && rt_mode & IP_VS_RT_MODE_CONNECT && fl4.saddr) { + } + if (rt_mode & IP_VS_RT_MODE_CONNECT && fl4.saddr) { ip_rt_put(rt); - *saddr = fl4.saddr; flowi4_update_output(&fl4, 0, daddr, fl4.saddr); - loop = true; + rt_mode = 0; goto retry; } - *saddr = fl4.saddr; + if (ret_saddr) + *ret_saddr = fl4.saddr; return rt; } @@ -344,19 +337,15 @@ __ip_vs_get_out_rt(struct netns_ipvs *ipvs, int skb_af, struct sk_buff *skb, if (ret_saddr) *ret_saddr = dest_dst->dst_saddr.ip; } else { - __be32 saddr = htonl(INADDR_ANY); - noref = 0; /* For such unconfigured boxes avoid many route lookups * for performance reasons because we do not remember saddr */ rt_mode &= ~IP_VS_RT_MODE_CONNECT; - rt = do_output_route4(net, daddr, rt_mode, &saddr); + rt = do_output_route4(net, daddr, rt_mode, ret_saddr); if (!rt) goto err_unreach; - if (ret_saddr) - *ret_saddr = saddr; } local = (rt->rt_flags & RTCF_LOCAL) ? 1 : 0; From 8478a729c0462273188263136880480729e9efca Mon Sep 17 00:00:00 2001 From: Jozsef Kadlecsik Date: Wed, 7 May 2025 17:01:59 +0200 Subject: [PATCH 245/555] netfilter: ipset: fix region locking in hash types Region locking introduced in v5.6-rc4 contained three macros to handle the region locks: ahash_bucket_start(), ahash_bucket_end() which gave back the start and end hash bucket values belonging to a given region lock and ahash_region() which should give back the region lock belonging to a given hash bucket. The latter was incorrect which can lead to a race condition between the garbage collector and adding new elements when a hash type of set is defined with timeouts. Fixes: f66ee0410b1c ("netfilter: ipset: Fix "INFO: rcu detected stall in hash_xxx" reports") Reported-by: Kota Toda Signed-off-by: Jozsef Kadlecsik Signed-off-by: Pablo Neira Ayuso --- net/netfilter/ipset/ip_set_hash_gen.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/netfilter/ipset/ip_set_hash_gen.h b/net/netfilter/ipset/ip_set_hash_gen.h index cf3ce72c3de6..5251524b96af 100644 --- a/net/netfilter/ipset/ip_set_hash_gen.h +++ b/net/netfilter/ipset/ip_set_hash_gen.h @@ -64,7 +64,7 @@ struct hbucket { #define ahash_sizeof_regions(htable_bits) \ (ahash_numof_locks(htable_bits) * sizeof(struct ip_set_region)) #define ahash_region(n, htable_bits) \ - ((n) % ahash_numof_locks(htable_bits)) + ((n) / jhash_size(HTABLE_REGION_BITS)) #define ahash_bucket_start(h, htable_bits) \ ((htable_bits) < HTABLE_REGION_BITS ? 0 \ : (h) * jhash_size(HTABLE_REGION_BITS)) From 9984db63742099ee3f3cff35cf71306d10e64356 Mon Sep 17 00:00:00 2001 From: Roman Li Date: Mon, 14 Apr 2025 12:56:48 -0400 Subject: [PATCH 246/555] drm/amd/display: Fix invalid context error in dml helper [Why] "BUG: sleeping function called from invalid context" error. after: "drm/amd/display: Protect FPU in dml2_validate()/dml21_validate()" The populate_dml_plane_cfg_from_plane_state() uses the GFP_KERNEL flag for memory allocation, which shouldn't be used in atomic contexts. The allocation is needed only for using another helper function get_scaler_data_for_plane(). [How] Modify helpers to pass a pointer to scaler_data within existing context, eliminating the need for dynamic memory allocation/deallocation and copying. Fixes: 366e77cd4923 ("drm/amd/display: Protect FPU in dml2_validate()/dml21_validate()") Reviewed-by: Aurabindo Pillai Signed-off-by: Roman Li Signed-off-by: Ray Wu Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher (cherry picked from commit bd3e84bc98f81b44f2c43936bdadc3241d654259) Cc: stable@vger.kernel.org --- .../amd/display/dc/dml2/dml2_translation_helper.c | 14 +++++--------- 1 file changed, 5 insertions(+), 9 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c index 2061d43b92e1..ab6baf269801 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c @@ -973,7 +973,9 @@ static void populate_dml_surface_cfg_from_plane_state(enum dml_project_id dml2_p } } -static void get_scaler_data_for_plane(const struct dc_plane_state *in, struct dc_state *context, struct scaler_data *out) +static struct scaler_data *get_scaler_data_for_plane( + const struct dc_plane_state *in, + struct dc_state *context) { int i; struct pipe_ctx *temp_pipe = &context->res_ctx.temp_pipe; @@ -994,7 +996,7 @@ static void get_scaler_data_for_plane(const struct dc_plane_state *in, struct dc } ASSERT(i < MAX_PIPES); - memcpy(out, &temp_pipe->plane_res.scl_data, sizeof(*out)); + return &temp_pipe->plane_res.scl_data; } static void populate_dummy_dml_plane_cfg(struct dml_plane_cfg_st *out, unsigned int location, @@ -1057,11 +1059,7 @@ static void populate_dml_plane_cfg_from_plane_state(struct dml_plane_cfg_st *out const struct dc_plane_state *in, struct dc_state *context, const struct soc_bounding_box_st *soc) { - struct scaler_data *scaler_data = kzalloc(sizeof(*scaler_data), GFP_KERNEL); - if (!scaler_data) - return; - - get_scaler_data_for_plane(in, context, scaler_data); + struct scaler_data *scaler_data = get_scaler_data_for_plane(in, context); out->CursorBPP[location] = dml_cur_32bit; out->CursorWidth[location] = 256; @@ -1126,8 +1124,6 @@ static void populate_dml_plane_cfg_from_plane_state(struct dml_plane_cfg_st *out out->DynamicMetadataTransmittedBytes[location] = 0; out->NumberOfCursors[location] = 1; - - kfree(scaler_data); } static unsigned int map_stream_to_dml_display_cfg(const struct dml2_context *dml2, From f1c6be3999d2be2673a51a9be0caf9348e254e52 Mon Sep 17 00:00:00 2001 From: Aurabindo Pillai Date: Wed, 16 Apr 2025 11:26:54 -0400 Subject: [PATCH 247/555] drm/amd/display: more liberal vmin/vmax update for freesync [Why] FAMS2 expects vmin/vmax to be updated in the case when freesync is off, but supported. But we only update it when freesync is enabled. [How] Change the vsync handler such that dc_stream_adjust_vmin_vmax() its called irrespective of whether freesync is enabled. If freesync is supported, then there is no harm in updating vmin/vmax registers. Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3546 Reviewed-by: ChiaHsuan Chung Signed-off-by: Aurabindo Pillai Signed-off-by: Ray Wu Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher (cherry picked from commit cfb2d41831ee5647a4ae0ea7c24971a92d5dfa0d) Cc: stable@vger.kernel.org --- .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index 536f73131c2d..93dd9e273c65 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -673,15 +673,21 @@ static void dm_crtc_high_irq(void *interrupt_params) spin_lock_irqsave(&adev_to_drm(adev)->event_lock, flags); if (acrtc->dm_irq_params.stream && - acrtc->dm_irq_params.vrr_params.supported && - acrtc->dm_irq_params.freesync_config.state == - VRR_STATE_ACTIVE_VARIABLE) { + acrtc->dm_irq_params.vrr_params.supported) { + bool replay_en = acrtc->dm_irq_params.stream->link->replay_settings.replay_feature_enabled; + bool psr_en = acrtc->dm_irq_params.stream->link->psr_settings.psr_feature_enabled; + bool fs_active_var_en = acrtc->dm_irq_params.freesync_config.state == VRR_STATE_ACTIVE_VARIABLE; + mod_freesync_handle_v_update(adev->dm.freesync_module, acrtc->dm_irq_params.stream, &acrtc->dm_irq_params.vrr_params); - dc_stream_adjust_vmin_vmax(adev->dm.dc, acrtc->dm_irq_params.stream, - &acrtc->dm_irq_params.vrr_params.adjust); + /* update vmin_vmax only if freesync is enabled, or only if PSR and REPLAY are disabled */ + if (fs_active_var_en || (!fs_active_var_en && !replay_en && !psr_en)) { + dc_stream_adjust_vmin_vmax(adev->dm.dc, + acrtc->dm_irq_params.stream, + &acrtc->dm_irq_params.vrr_params.adjust); + } } /* From 2a24755774ef8db33139e2d9a968cc585c6488da Mon Sep 17 00:00:00 2001 From: Alex Hung Date: Wed, 23 Apr 2025 10:02:11 -0600 Subject: [PATCH 248/555] drm/amd/display: Remove unnecessary DC_FP_START/DC_FP_END [WHY & HOW] Remove the unnecessary DC_FP_START/DC_FP_END pair to reduce time in preempt_disable. It also fixes "BUG: sleeping function called from invalid context" error messages because of calling kzalloc with GFP_KERNEL which can sleep. Reviewed-by: Aurabindo Pillai Signed-off-by: Alex Hung Signed-off-by: Ray Wu Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher (cherry picked from commit 94da0735b67b3ada90a3e2e017d306f070306f1d) --- .../gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c | 6 ------ 1 file changed, 6 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c index 2a59cc61ed8c..944650cb13de 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c @@ -2114,8 +2114,6 @@ static bool dcn32_resource_construct( #define REG_STRUCT dccg_regs dccg_regs_init(); - DC_FP_START(); - ctx->dc_bios->regs = &bios_regs; pool->base.res_cap = &res_cap_dcn32; @@ -2501,14 +2499,10 @@ static bool dcn32_resource_construct( if (ASICREV_IS_GC_11_0_3(dc->ctx->asic_id.hw_internal_rev) && (dc->config.sdpif_request_limit_words_per_umc == 0)) dc->config.sdpif_request_limit_words_per_umc = 16; - DC_FP_END(); - return true; create_fail: - DC_FP_END(); - dcn32_resource_destruct(pool); return false; From eba692ca3abca258b3214a6e4126afefad1822f0 Mon Sep 17 00:00:00 2001 From: Austin Zheng Date: Thu, 17 Apr 2025 10:24:29 -0400 Subject: [PATCH 249/555] drm/amd/display: Call FP Protect Before Mode Programming/Mode Support [Why] Memory allocation occurs within dml21_validate() for adding phantom planes. May cause kernel to be tainted due to usage of FP Start. [How] Move FP start from dml21_validate to before mode programming/mode support. Calculations requiring floating point are all done within mode programming or mode support. Reviewed-by: Alvin Lee Signed-off-by: Austin Zheng Signed-off-by: Ray Wu Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher (cherry picked from commit fe3250f10819b411808ab9ae1d824c5fc9b59170) --- drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c index 5d16f36ec95c..ed6584535e89 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c @@ -234,7 +234,9 @@ static bool dml21_mode_check_and_programming(const struct dc *in_dc, struct dc_s if (!result) return false; + DC_FP_START(); result = dml2_build_mode_programming(mode_programming); + DC_FP_END(); if (!result) return false; @@ -277,7 +279,9 @@ static bool dml21_check_mode_support(const struct dc *in_dc, struct dc_state *co mode_support->dml2_instance = dml_init->dml2_instance; dml21_map_dc_state_into_dml_display_cfg(in_dc, context, dml_ctx); dml_ctx->v21.mode_programming.dml2_instance->scratch.build_mode_programming_locals.mode_programming_params.programming = dml_ctx->v21.mode_programming.programming; + DC_FP_START(); is_supported = dml2_check_mode_supported(mode_support); + DC_FP_END(); if (!is_supported) return false; @@ -288,16 +292,12 @@ bool dml21_validate(const struct dc *in_dc, struct dc_state *context, struct dml { bool out = false; - DC_FP_START(); - /* Use dml_validate_only for fast_validate path */ if (fast_validate) out = dml21_check_mode_support(in_dc, context, dml_ctx); else out = dml21_mode_check_and_programming(in_dc, context, dml_ctx); - DC_FP_END(); - return out; } From 5a3846648c0523fd850b7f0aec78c0139453ab8b Mon Sep 17 00:00:00 2001 From: Wayne Lin Date: Fri, 18 Apr 2025 16:31:59 +0800 Subject: [PATCH 250/555] drm/amd/display: Shift DMUB AUX reply command if necessary [Why] Defined value of dmub AUX reply command field get updated but didn't adjust dm receiving side accordingly. [How] Check the received reply command value to see if it's updated version or not. Adjust it if necessary. Fixes: ead08b95fa50 ("drm/amd/display: Fix race condition in DPIA AUX transfer") Cc: Mario Limonciello Cc: Alex Deucher Reviewed-by: Ray Wu Signed-off-by: Wayne Lin Signed-off-by: Ray Wu Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher (cherry picked from commit d5c9ade755a9afa210840708a12a8f44c0d532f4) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index 93dd9e273c65..d9292da56948 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -12758,8 +12758,11 @@ int amdgpu_dm_process_dmub_aux_transfer_sync( goto out; } + payload->reply[0] = adev->dm.dmub_notify->aux_reply.command & 0xF; + if (adev->dm.dmub_notify->aux_reply.command & 0xF0) + /* The reply is stored in the top nibble of the command. */ + payload->reply[0] = (adev->dm.dmub_notify->aux_reply.command >> 4) & 0xF; - payload->reply[0] = adev->dm.dmub_notify->aux_reply.command; if (!payload->write && p_notify->aux_reply.length && (payload->reply[0] == AUX_TRANSACTION_REPLY_AUX_ACK)) { From bc70e11b550d37fbd9eaed0f113ba560894f1609 Mon Sep 17 00:00:00 2001 From: Wayne Lin Date: Sun, 20 Apr 2025 16:29:07 +0800 Subject: [PATCH 251/555] drm/amd/display: Fix the checking condition in dmub aux handling [Why & How] Fix the checking condition for detecting AUX_RET_ERROR_PROTOCOL_ERROR. It was wrongly checking by "not equals to" Reviewed-by: Ray Wu Signed-off-by: Wayne Lin Signed-off-by: Ray Wu Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher (cherry picked from commit 1db6c9e9b62e1a8912f0a281c941099fca678da3) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index d9292da56948..68bbe7d57773 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -12749,7 +12749,7 @@ int amdgpu_dm_process_dmub_aux_transfer_sync( * Transient states before tunneling is enabled could * lead to this error. We can ignore this for now. */ - if (p_notify->result != AUX_RET_ERROR_PROTOCOL_ERROR) { + if (p_notify->result == AUX_RET_ERROR_PROTOCOL_ERROR) { DRM_WARN("DPIA AUX failed on 0x%x(%d), error %d\n", payload->address, payload->length, p_notify->result); From 396dc51b3b7ea524bf8061f478332d0039e96d5d Mon Sep 17 00:00:00 2001 From: Wayne Lin Date: Sun, 20 Apr 2025 16:56:54 +0800 Subject: [PATCH 252/555] drm/amd/display: Remove incorrect checking in dmub aux handler [Why & How] "Request length != reply length" is expected behavior defined in spec. It's not an invalid reply. Besides, replied data handling logic is not designed to be written in amdgpu_dm_process_dmub_aux_transfer_sync(). Remove the incorrectly handling section. Fixes: ead08b95fa50 ("drm/amd/display: Fix race condition in DPIA AUX transfer") Cc: Mario Limonciello Cc: Alex Deucher Reviewed-by: Ray Wu Signed-off-by: Wayne Lin Signed-off-by: Ray Wu Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher (cherry picked from commit 81b5c6fa62af62fe89ae9576f41aae37830b94cb) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 12 +----------- 1 file changed, 1 insertion(+), 11 deletions(-) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index 68bbe7d57773..d63f219de88e 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -12764,19 +12764,9 @@ int amdgpu_dm_process_dmub_aux_transfer_sync( payload->reply[0] = (adev->dm.dmub_notify->aux_reply.command >> 4) & 0xF; if (!payload->write && p_notify->aux_reply.length && - (payload->reply[0] == AUX_TRANSACTION_REPLY_AUX_ACK)) { - - if (payload->length != p_notify->aux_reply.length) { - DRM_WARN("invalid read length %d from DPIA AUX 0x%x(%d)!\n", - p_notify->aux_reply.length, - payload->address, payload->length); - *operation_result = AUX_RET_ERROR_INVALID_REPLY; - goto out; - } - + (payload->reply[0] == AUX_TRANSACTION_REPLY_AUX_ACK)) memcpy(payload->data, p_notify->aux_reply.data, p_notify->aux_reply.length); - } /* success */ ret = p_notify->aux_reply.length; From 3924f45d4de7250a603fd7b50379237a6a0e5adf Mon Sep 17 00:00:00 2001 From: Wayne Lin Date: Sun, 20 Apr 2025 17:50:03 +0800 Subject: [PATCH 253/555] drm/amd/display: Copy AUX read reply data whenever length > 0 [Why] amdgpu_dm_process_dmub_aux_transfer_sync() should return all exact data reply from the sink side. Don't do the analysis job in it. [How] Remove unnecessary check condition AUX_TRANSACTION_REPLY_AUX_ACK. Fixes: ead08b95fa50 ("drm/amd/display: Fix race condition in DPIA AUX transfer") Cc: Mario Limonciello Cc: Alex Deucher Reviewed-by: Ray Wu Signed-off-by: Wayne Lin Signed-off-by: Ray Wu Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher (cherry picked from commit 9b540e3fe6796fec4fb1344f3be8952fc2f084d4) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index d63f219de88e..64df8ca448b3 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -12763,8 +12763,7 @@ int amdgpu_dm_process_dmub_aux_transfer_sync( /* The reply is stored in the top nibble of the command. */ payload->reply[0] = (adev->dm.dmub_notify->aux_reply.command >> 4) & 0xF; - if (!payload->write && p_notify->aux_reply.length && - (payload->reply[0] == AUX_TRANSACTION_REPLY_AUX_ACK)) + if (!payload->write && p_notify->aux_reply.length) memcpy(payload->data, p_notify->aux_reply.data, p_notify->aux_reply.length); From 65924ec69b29296845c7f628112353438e63ea56 Mon Sep 17 00:00:00 2001 From: Wayne Lin Date: Sun, 20 Apr 2025 19:22:14 +0800 Subject: [PATCH 254/555] drm/amd/display: Fix wrong handling for AUX_DEFER case [Why] We incorrectly ack all bytes get written when the reply actually is defer. When it's defer, means sink is not ready for the request. We should retry the request. [How] Only reply all data get written when receive I2C_ACK|AUX_ACK. Otherwise, reply the number of actual written bytes received from the sink. Add some messages to facilitate debugging as well. Fixes: ad6756b4d773 ("drm/amd/display: Shift dc link aux to aux_payload") Cc: Mario Limonciello Cc: Alex Deucher Reviewed-by: Ray Wu Signed-off-by: Wayne Lin Signed-off-by: Ray Wu Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher (cherry picked from commit 3637e457eb0000bc37d8bbbec95964aad2fb29fd) Cc: stable@vger.kernel.org --- .../display/amdgpu_dm/amdgpu_dm_mst_types.c | 28 ++++++++++++++++--- 1 file changed, 24 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c index 7ceedf626d23..074b79fd5822 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c @@ -51,6 +51,9 @@ #define PEAK_FACTOR_X1000 1006 +/* + * This function handles both native AUX and I2C-Over-AUX transactions. + */ static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux, struct drm_dp_aux_msg *msg) { @@ -87,15 +90,25 @@ static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux, if (adev->dm.aux_hpd_discon_quirk) { if (msg->address == DP_SIDEBAND_MSG_DOWN_REQ_BASE && operation_result == AUX_RET_ERROR_HPD_DISCON) { - result = 0; + result = msg->size; operation_result = AUX_RET_SUCCESS; } } - if (payload.write && result >= 0) - result = msg->size; + /* + * result equals to 0 includes the cases of AUX_DEFER/I2C_DEFER + */ + if (payload.write && result >= 0) { + if (result) { + /*one byte indicating partially written bytes. Force 0 to retry*/ + drm_info(adev_to_drm(adev), "amdgpu: AUX partially written\n"); + result = 0; + } else if (!payload.reply[0]) + /*I2C_ACK|AUX_ACK*/ + result = msg->size; + } - if (result < 0) + if (result < 0) { switch (operation_result) { case AUX_RET_SUCCESS: break; @@ -114,6 +127,13 @@ static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux, break; } + drm_info(adev_to_drm(adev), "amdgpu: DP AUX transfer fail:%d\n", operation_result); + } + + if (payload.reply[0]) + drm_info(adev_to_drm(adev), "amdgpu: AUX reply command not ACK: 0x%02x.", + payload.reply[0]); + return result; } From b7e84fb708392b37e5dbb2a95db9b94a0e3f0aa2 Mon Sep 17 00:00:00 2001 From: Ruijing Dong Date: Fri, 2 May 2025 11:19:26 -0400 Subject: [PATCH 255/555] drm/amdgpu/vcn: using separate VCN1_AON_SOC offset VCN1_AON_SOC_ADDRESS_3_0 offset varies on different VCN generations, the issue in vcn4.0.5 is caused by a different VCN1_AON_SOC_ADDRESS_3_0 offset. This patch does the following: 1. use the same offset for other VCN generations. 2. use the vcn4.0.5 special offset 3. update vcn_4_0 and vcn_5_0 Acked-by: Saleemkhan Jamadar Reviewed-by: Leo Liu Signed-off-by: Ruijing Dong Signed-off-by: Alex Deucher (cherry picked from commit 5c89ceda9984498b28716944633a9a01cbb2c90d) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h | 1 - drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c | 1 + drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c | 1 + drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c | 1 + drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c | 4 +++- drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 1 + drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c | 1 + drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c | 3 ++- 8 files changed, 10 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h index cdcdae7f71ce..83adf81defc7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h @@ -66,7 +66,6 @@ #define VCN_ENC_CMD_REG_WAIT 0x0000000c #define VCN_AON_SOC_ADDRESS_2_0 0x1f800 -#define VCN1_AON_SOC_ADDRESS_3_0 0x48000 #define VCN_VID_IP_ADDRESS_2_0 0x0 #define VCN_AON_IP_ADDRESS_2_0 0x30000 diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c index 8e7a36f26e9c..b8d835c9e17e 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c @@ -39,6 +39,7 @@ #define VCN_VID_SOC_ADDRESS_2_0 0x1fa00 #define VCN1_VID_SOC_ADDRESS_3_0 0x48200 +#define VCN1_AON_SOC_ADDRESS_3_0 0x48000 #define mmUVD_CONTEXT_ID_INTERNAL_OFFSET 0x1fd #define mmUVD_GPCOM_VCPU_CMD_INTERNAL_OFFSET 0x503 diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c index d716510b8dd6..3eec1b8feaee 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c @@ -39,6 +39,7 @@ #define VCN_VID_SOC_ADDRESS_2_0 0x1fa00 #define VCN1_VID_SOC_ADDRESS_3_0 0x48200 +#define VCN1_AON_SOC_ADDRESS_3_0 0x48000 #define mmUVD_CONTEXT_ID_INTERNAL_OFFSET 0x27 #define mmUVD_GPCOM_VCPU_CMD_INTERNAL_OFFSET 0x0f diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c index 22ae1939476f..0b19f0ab4480 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c @@ -40,6 +40,7 @@ #define VCN_VID_SOC_ADDRESS_2_0 0x1fa00 #define VCN1_VID_SOC_ADDRESS_3_0 0x48200 +#define VCN1_AON_SOC_ADDRESS_3_0 0x48000 #define mmUVD_CONTEXT_ID_INTERNAL_OFFSET 0x27 #define mmUVD_GPCOM_VCPU_CMD_INTERNAL_OFFSET 0x0f diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c index c6f6392c1c20..1f777c125b00 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c @@ -46,6 +46,7 @@ #define VCN_VID_SOC_ADDRESS_2_0 0x1fb00 #define VCN1_VID_SOC_ADDRESS_3_0 0x48300 +#define VCN1_AON_SOC_ADDRESS_3_0 0x48000 #define VCN_HARVEST_MMSCH 0 @@ -614,7 +615,8 @@ static void vcn_v4_0_mc_resume_dpg_mode(struct amdgpu_vcn_inst *vinst, /* VCN global tiling registers */ WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( - VCN, 0, regUVD_GFX10_ADDR_CONFIG), adev->gfx.config.gb_addr_config, 0, indirect); + VCN, inst_idx, regUVD_GFX10_ADDR_CONFIG), + adev->gfx.config.gb_addr_config, 0, indirect); } /** diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c index 3e176b4b7c69..012f6ea928ec 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c @@ -45,6 +45,7 @@ #define VCN_VID_SOC_ADDRESS_2_0 0x1fb00 #define VCN1_VID_SOC_ADDRESS_3_0 0x48300 +#define VCN1_AON_SOC_ADDRESS_3_0 0x48000 static const struct amdgpu_hwip_reg_entry vcn_reg_list_4_0_3[] = { SOC15_REG_ENTRY_STR(VCN, 0, regUVD_POWER_STATUS), diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c index ba603b2246e2..a1171e6152ed 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c @@ -46,6 +46,7 @@ #define VCN_VID_SOC_ADDRESS_2_0 0x1fb00 #define VCN1_VID_SOC_ADDRESS_3_0 (0x48300 + 0x38000) +#define VCN1_AON_SOC_ADDRESS_3_0 (0x48000 + 0x38000) #define VCN_HARVEST_MMSCH 0 diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c index d99d05f42f1d..b90da3d3e140 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c @@ -533,7 +533,8 @@ static void vcn_v5_0_0_mc_resume_dpg_mode(struct amdgpu_vcn_inst *vinst, /* VCN global tiling registers */ WREG32_SOC24_DPG_MODE(inst_idx, SOC24_DPG_MODE_OFFSET( - VCN, 0, regUVD_GFX10_ADDR_CONFIG), adev->gfx.config.gb_addr_config, 0, indirect); + VCN, inst_idx, regUVD_GFX10_ADDR_CONFIG), + adev->gfx.config.gb_addr_config, 0, indirect); return; } From d0ce1aaa8531a4a4707711cab5721374751c51b0 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Thu, 1 May 2025 13:00:16 -0400 Subject: [PATCH 256/555] Revert "drm/amd: Stop evicting resources on APUs in suspend" This reverts commit 3a9626c816db901def438dc2513622e281186d39. This breaks S4 because we end up setting the s3/s0ix flags even when we are entering s4 since prepare is used by both flows. The causes both the S3/s0ix and s4 flags to be set which breaks several checks in the driver which assume they are mutually exclusive. Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3634 Cc: Mario Limonciello Reviewed-by: Mario Limonciello Signed-off-by: Alex Deucher (cherry picked from commit ce8f7d95899c2869b47ea6ce0b3e5bf304b2fff4) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/amdgpu.h | 2 -- drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c | 18 ------------------ drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 11 ++--------- 3 files changed, 2 insertions(+), 29 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index ef6e78224fdf..c3641331d4de 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -1614,11 +1614,9 @@ static inline void amdgpu_acpi_get_backlight_caps(struct amdgpu_dm_backlight_cap #if defined(CONFIG_ACPI) && defined(CONFIG_SUSPEND) bool amdgpu_acpi_is_s3_active(struct amdgpu_device *adev); bool amdgpu_acpi_is_s0ix_active(struct amdgpu_device *adev); -void amdgpu_choose_low_power_state(struct amdgpu_device *adev); #else static inline bool amdgpu_acpi_is_s0ix_active(struct amdgpu_device *adev) { return false; } static inline bool amdgpu_acpi_is_s3_active(struct amdgpu_device *adev) { return false; } -static inline void amdgpu_choose_low_power_state(struct amdgpu_device *adev) { } #endif void amdgpu_register_gpu_instance(struct amdgpu_device *adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c index b7f8f2ff143d..707e131f89d2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c @@ -1533,22 +1533,4 @@ bool amdgpu_acpi_is_s0ix_active(struct amdgpu_device *adev) #endif /* CONFIG_AMD_PMC */ } -/** - * amdgpu_choose_low_power_state - * - * @adev: amdgpu_device_pointer - * - * Choose the target low power state for the GPU - */ -void amdgpu_choose_low_power_state(struct amdgpu_device *adev) -{ - if (adev->in_runpm) - return; - - if (amdgpu_acpi_is_s0ix_active(adev)) - adev->in_s0ix = true; - else if (amdgpu_acpi_is_s3_active(adev)) - adev->in_s3 = true; -} - #endif /* CONFIG_SUSPEND */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 7f354cd532dc..5ac7bd5942d0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -4949,15 +4949,13 @@ int amdgpu_device_prepare(struct drm_device *dev) struct amdgpu_device *adev = drm_to_adev(dev); int i, r; - amdgpu_choose_low_power_state(adev); - if (dev->switch_power_state == DRM_SWITCH_POWER_OFF) return 0; /* Evict the majority of BOs before starting suspend sequence */ r = amdgpu_device_evict_resources(adev); if (r) - goto unprepare; + return r; flush_delayed_work(&adev->gfx.gfx_off_delay_work); @@ -4968,15 +4966,10 @@ int amdgpu_device_prepare(struct drm_device *dev) continue; r = adev->ip_blocks[i].version->funcs->prepare_suspend(&adev->ip_blocks[i]); if (r) - goto unprepare; + return r; } return 0; - -unprepare: - adev->in_s0ix = adev->in_s3 = adev->in_s4 = false; - - return r; } /** From 4aaffc85751da5722e858e4333e8cf0aa4b6c78f Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Thu, 1 May 2025 13:46:46 -0400 Subject: [PATCH 257/555] drm/amdgpu: fix pm notifier handling Set the s3/s0ix and s4 flags in the pm notifier so that we can skip the resource evictions properly in pm prepare based on whether we are suspending or hibernating. Drop the eviction as processes are not frozen at this time, we we can end up getting stuck trying to evict VRAM while applications continue to submit work which causes the buffers to get pulled back into VRAM. v2: Move suspend flags out of pm notifier (Mario) Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4178 Fixes: 2965e6355dcd ("drm/amd: Add Suspend/Hibernate notification callback support") Cc: Mario Limonciello Reviewed-by: Mario Limonciello Signed-off-by: Alex Deucher (cherry picked from commit 06f2dcc241e7e5c681f81fbc46cacdf4bfd7d6d7) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 18 +++++------------- drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 10 +--------- 2 files changed, 6 insertions(+), 22 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 5ac7bd5942d0..f8b3e04d71ed 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -4907,28 +4907,20 @@ static int amdgpu_device_evict_resources(struct amdgpu_device *adev) * @data: data * * This function is called when the system is about to suspend or hibernate. - * It is used to evict resources from the device before the system goes to - * sleep while there is still access to swap. + * It is used to set the appropriate flags so that eviction can be optimized + * in the pm prepare callback. */ static int amdgpu_device_pm_notifier(struct notifier_block *nb, unsigned long mode, void *data) { struct amdgpu_device *adev = container_of(nb, struct amdgpu_device, pm_nb); - int r; switch (mode) { case PM_HIBERNATION_PREPARE: adev->in_s4 = true; - fallthrough; - case PM_SUSPEND_PREPARE: - r = amdgpu_device_evict_resources(adev); - /* - * This is considered non-fatal at this time because - * amdgpu_device_prepare() will also fatally evict resources. - * See https://gitlab.freedesktop.org/drm/amd/-/issues/3781 - */ - if (r) - drm_warn(adev_to_drm(adev), "Failed to evict resources, freeze active processes if problems occur: %d\n", r); + break; + case PM_POST_HIBERNATION: + adev->in_s4 = false; break; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 24ee4710f807..72c807f5822e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -2615,13 +2615,8 @@ static int amdgpu_pmops_freeze(struct device *dev) static int amdgpu_pmops_thaw(struct device *dev) { struct drm_device *drm_dev = dev_get_drvdata(dev); - struct amdgpu_device *adev = drm_to_adev(drm_dev); - int r; - r = amdgpu_device_resume(drm_dev, true); - adev->in_s4 = false; - - return r; + return amdgpu_device_resume(drm_dev, true); } static int amdgpu_pmops_poweroff(struct device *dev) @@ -2634,9 +2629,6 @@ static int amdgpu_pmops_poweroff(struct device *dev) static int amdgpu_pmops_restore(struct device *dev) { struct drm_device *drm_dev = dev_get_drvdata(dev); - struct amdgpu_device *adev = drm_to_adev(drm_dev); - - adev->in_s4 = false; return amdgpu_device_resume(drm_dev, true); } From f690e3974755a650259a45d71456decc9c96a282 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Wed, 30 Apr 2025 12:45:04 -0400 Subject: [PATCH 258/555] drm/amdgpu/hdp4: use memcfg register to post the write for HDP flush Reading back the remapped HDP flush register seems to cause problems on some platforms. All we need is a read, so read back the memcfg register. Fixes: c9b8dcabb52a ("drm/amdgpu/hdp4.0: do a posting read when flushing HDP") Reported-by: Alexey Klimov Link: https://lists.freedesktop.org/archives/amd-gfx/2025-April/123150.html Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4119 Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3908 Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher (cherry picked from commit 5c937b4a6050316af37ef214825b6340b5e9e391) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c b/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c index f1dc13b3ab38..cbbeadeb53f7 100644 --- a/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c @@ -41,7 +41,12 @@ static void hdp_v4_0_flush_hdp(struct amdgpu_device *adev, { if (!ring || !ring->funcs->emit_wreg) { WREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0); - RREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2); + /* We just need to read back a register to post the write. + * Reading back the remapped register causes problems on + * some platforms so just read back the memory size register. + */ + if (adev->nbio.funcs->get_memsize) + adev->nbio.funcs->get_memsize(adev); } else { amdgpu_ring_emit_wreg(ring, (adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0); } From 6beb6835c1fbb3f676aebb51a5fee6b77fed9308 Mon Sep 17 00:00:00 2001 From: Eelco Chaudron Date: Tue, 6 May 2025 16:28:54 +0200 Subject: [PATCH 259/555] openvswitch: Fix unsafe attribute parsing in output_userspace() This patch replaces the manual Netlink attribute iteration in output_userspace() with nla_for_each_nested(), which ensures that only well-formed attributes are processed. Fixes: ccb1352e76cf ("net: Add Open vSwitch kernel components.") Signed-off-by: Eelco Chaudron Acked-by: Ilya Maximets Acked-by: Aaron Conole Link: https://patch.msgid.link/0bd65949df61591d9171c0dc13e42cea8941da10.1746541734.git.echaudro@redhat.com Signed-off-by: Jakub Kicinski --- net/openvswitch/actions.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c index 61fea7baae5d..2f22ca59586f 100644 --- a/net/openvswitch/actions.c +++ b/net/openvswitch/actions.c @@ -975,8 +975,7 @@ static int output_userspace(struct datapath *dp, struct sk_buff *skb, upcall.cmd = OVS_PACKET_CMD_ACTION; upcall.mru = OVS_CB(skb)->mru; - for (a = nla_data(attr), rem = nla_len(attr); rem > 0; - a = nla_next(a, &rem)) { + nla_for_each_nested(a, attr, rem) { switch (nla_type(a)) { case OVS_USERSPACE_ATTR_USERDATA: upcall.userdata = a; From 4a7843cc8a41b9612becccc07715ed017770eb89 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Tue, 6 May 2025 18:56:47 +0200 Subject: [PATCH 260/555] net: airoha: Add missing field to ppe_mbox_data struct The official Airoha EN7581 firmware requires adding max_packet field in ppe_mbox_data struct while the unofficial one used to develop the Airoha EN7581 flowtable support does not require this field. This patch does not introduce any real backwards compatible issue since EN7581 fw is not publicly available in linux-firmware or other repositories (e.g. OpenWrt) yet and the official fw version will use this new layout. For this reason this change needs to be backported. Moreover, make explicit the padding added by the compiler introducing the rsv array in init_info struct. At the same time use u32 instead of int for init_info and set_info struct definitions in ppe_mbox_data struct. Fixes: 23290c7bc190d ("net: airoha: Introduce Airoha NPU support") Reviewed-by: Simon Horman Reviewed-by: Jacob Keller Signed-off-by: Lorenzo Bianconi Link: https://patch.msgid.link/20250506-airoha-en7581-fix-ppe_mbox_data-v5-1-29cabed6864d@kernel.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/airoha/airoha_npu.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/airoha/airoha_npu.c b/drivers/net/ethernet/airoha/airoha_npu.c index 7a5710f9ccf6..ead0625e781f 100644 --- a/drivers/net/ethernet/airoha/airoha_npu.c +++ b/drivers/net/ethernet/airoha/airoha_npu.c @@ -104,12 +104,14 @@ struct ppe_mbox_data { u8 xpon_hal_api; u8 wan_xsi; u8 ct_joyme4; - int ppe_type; - int wan_mode; - int wan_sel; + u8 max_packet; + u8 rsv[3]; + u32 ppe_type; + u32 wan_mode; + u32 wan_sel; } init_info; struct { - int func_id; + u32 func_id; u32 size; u32 data; } set_info; From c4327229948879814229b46aa26a750718888503 Mon Sep 17 00:00:00 2001 From: Paul Chaignon Date: Mon, 5 May 2025 21:58:04 +0200 Subject: [PATCH 261/555] bpf: Scrub packet on bpf_redirect_peer When bpf_redirect_peer is used to redirect packets to a device in another network namespace, the skb isn't scrubbed. That can lead skb information from one namespace to be "misused" in another namespace. As one example, this is causing Cilium to drop traffic when using bpf_redirect_peer to redirect packets that just went through IPsec decryption to a container namespace. The following pwru trace shows (1) the packet path from the host's XFRM layer to the container's XFRM layer where it's dropped and (2) the number of active skb extensions at each function. NETNS MARK IFACE TUPLE FUNC 4026533547 d00 eth0 10.244.3.124:35473->10.244.2.158:53 xfrm_rcv_cb .active_extensions = (__u8)2, 4026533547 d00 eth0 10.244.3.124:35473->10.244.2.158:53 xfrm4_rcv_cb .active_extensions = (__u8)2, 4026533547 d00 eth0 10.244.3.124:35473->10.244.2.158:53 gro_cells_receive .active_extensions = (__u8)2, [...] 4026533547 0 eth0 10.244.3.124:35473->10.244.2.158:53 skb_do_redirect .active_extensions = (__u8)2, 4026534999 0 eth0 10.244.3.124:35473->10.244.2.158:53 ip_rcv .active_extensions = (__u8)2, 4026534999 0 eth0 10.244.3.124:35473->10.244.2.158:53 ip_rcv_core .active_extensions = (__u8)2, [...] 4026534999 0 eth0 10.244.3.124:35473->10.244.2.158:53 udp_queue_rcv_one_skb .active_extensions = (__u8)2, 4026534999 0 eth0 10.244.3.124:35473->10.244.2.158:53 __xfrm_policy_check .active_extensions = (__u8)2, 4026534999 0 eth0 10.244.3.124:35473->10.244.2.158:53 __xfrm_decode_session .active_extensions = (__u8)2, 4026534999 0 eth0 10.244.3.124:35473->10.244.2.158:53 security_xfrm_decode_session .active_extensions = (__u8)2, 4026534999 0 eth0 10.244.3.124:35473->10.244.2.158:53 kfree_skb_reason(SKB_DROP_REASON_XFRM_POLICY) .active_extensions = (__u8)2, In this case, there are no XFRM policies in the container's network namespace so the drop is unexpected. When we decrypt the IPsec packet, the XFRM state used for decryption is set in the skb extensions. This information is preserved across the netns switch. When we reach the XFRM policy check in the container's netns, __xfrm_policy_check drops the packet with LINUX_MIB_XFRMINNOPOLS because a (container-side) XFRM policy can't be found that matches the (host-side) XFRM state used for decryption. This patch fixes this by scrubbing the packet when using bpf_redirect_peer, as is done on typical netns switches via veth devices except skb->mark and skb->tstamp are not zeroed. Fixes: 9aa1206e8f482 ("bpf: Add redirect_peer helper") Signed-off-by: Paul Chaignon Acked-by: Daniel Borkmann Acked-by: Martin KaFai Lau Link: https://patch.msgid.link/1728ead5e0fe45e7a6542c36bd4e3ca07a73b7d6.1746460653.git.paul.chaignon@gmail.com Signed-off-by: Jakub Kicinski --- net/core/filter.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/core/filter.c b/net/core/filter.c index 79cab4d78dc3..577a4504e26f 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -2509,6 +2509,7 @@ int skb_do_redirect(struct sk_buff *skb) goto out_drop; skb->dev = dev; dev_sw_netstats_rx_add(dev, skb->len); + skb_scrub_packet(skb, false); return -EAGAIN; } return flags & BPF_F_NEIGH ? From f5c79ffdc250bc8c90fd4fdf1e5d7ac4647912d5 Mon Sep 17 00:00:00 2001 From: Paul Chaignon Date: Mon, 5 May 2025 21:58:39 +0200 Subject: [PATCH 262/555] bpf: Clarify handling of mark and tstamp by redirect_peer When switching network namespaces with the bpf_redirect_peer helper, the skb->mark and skb->tstamp fields are not zeroed out like they can be on a typical netns switch. This patch clarifies that in the helper description. Signed-off-by: Paul Chaignon Acked-by: Daniel Borkmann Acked-by: Martin KaFai Lau Link: https://patch.msgid.link/ccc86af26d43c5c0b776bcba2601b7479c0d46d0.1746460653.git.paul.chaignon@gmail.com Signed-off-by: Jakub Kicinski --- include/uapi/linux/bpf.h | 3 +++ tools/include/uapi/linux/bpf.h | 3 +++ 2 files changed, 6 insertions(+) diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index 28705ae67784..fd404729b115 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -4968,6 +4968,9 @@ union bpf_attr { * the netns switch takes place from ingress to ingress without * going through the CPU's backlog queue. * + * *skb*\ **->mark** and *skb*\ **->tstamp** are not cleared during + * the netns switch. + * * The *flags* argument is reserved and must be 0. The helper is * currently only supported for tc BPF program types at the * ingress hook and for veth and netkit target device types. The diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h index 28705ae67784..fd404729b115 100644 --- a/tools/include/uapi/linux/bpf.h +++ b/tools/include/uapi/linux/bpf.h @@ -4968,6 +4968,9 @@ union bpf_attr { * the netns switch takes place from ingress to ingress without * going through the CPU's backlog queue. * + * *skb*\ **->mark** and *skb*\ **->tstamp** are not cleared during + * the netns switch. + * * The *flags* argument is reserved and must be 0. The helper is * currently only supported for tc BPF program types at the * ingress hook and for veth and netkit target device types. The From e5641daa0ea1932e2b0fa7f27843e712244f6538 Mon Sep 17 00:00:00 2001 From: Meghana Malladi Date: Tue, 6 May 2025 16:35:44 +0530 Subject: [PATCH 263/555] net: ti: icssg-prueth: Set XDP feature flags for ndev xdp_features demonstrates what all XDP capabilities are supported on a given network device. The driver needs to set these xdp_features flag to let the network stack know what XDP features a given driver is supporting. These flags need to be set for a given ndev irrespective of any XDP program being loaded or not. Fixes: 62aa3246f462 ("net: ti: icssg-prueth: Add XDP support") Signed-off-by: Meghana Malladi Reviewed-by: Jacob Keller Link: https://patch.msgid.link/20250506110546.4065715-2-m-malladi@ti.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/ti/icssg/icssg_prueth.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth.c b/drivers/net/ethernet/ti/icssg/icssg_prueth.c index 443f90fa6557..ee35fecf61e7 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_prueth.c +++ b/drivers/net/ethernet/ti/icssg/icssg_prueth.c @@ -1109,11 +1109,6 @@ static int emac_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frame static int emac_xdp_setup(struct prueth_emac *emac, struct netdev_bpf *bpf) { struct bpf_prog *prog = bpf->prog; - xdp_features_t val; - - val = NETDEV_XDP_ACT_BASIC | NETDEV_XDP_ACT_REDIRECT | - NETDEV_XDP_ACT_NDO_XMIT; - xdp_set_features_flag(emac->ndev, val); if (!emac->xdpi.prog && !prog) return 0; @@ -1291,6 +1286,10 @@ static int prueth_netdev_init(struct prueth *prueth, ndev->hw_features = NETIF_F_SG; ndev->features = ndev->hw_features | NETIF_F_HW_VLAN_CTAG_FILTER; ndev->hw_features |= NETIF_PRUETH_HSR_OFFLOAD_FEATURES; + xdp_set_features_flag(ndev, + NETDEV_XDP_ACT_BASIC | + NETDEV_XDP_ACT_REDIRECT | + NETDEV_XDP_ACT_NDO_XMIT); netif_napi_add(ndev, &emac->napi_rx, icssg_napi_rx_poll); hrtimer_setup(&emac->rx_hrtimer, &emac_rx_timer_callback, CLOCK_MONOTONIC, From 8b3fae3e2376b70b7a76005263bc34d034b9c7bf Mon Sep 17 00:00:00 2001 From: Meghana Malladi Date: Tue, 6 May 2025 16:35:45 +0530 Subject: [PATCH 264/555] net: ti: icssg-prueth: Fix kernel panic during concurrent Tx queue access Add __netif_tx_lock() to ensure that only one packet is being transmitted at a time to avoid race conditions in the netif_txq struct and prevent packet data corruption. Failing to do so causes kernel panic with the following error: [ 2184.746764] ------------[ cut here ]------------ [ 2184.751412] kernel BUG at lib/dynamic_queue_limits.c:99! [ 2184.756728] Internal error: Oops - BUG: 00000000f2000800 [#1] PREEMPT SMP logs: https://gist.github.com/MeghanaMalladiTI/9c7aa5fc3b7fb03f87c74aad487956e9 The lock is acquired before calling emac_xmit_xdp_frame() and released after the call returns. This ensures that the TX queue is protected from concurrent access during the transmission of XDP frames. Fixes: 62aa3246f462 ("net: ti: icssg-prueth: Add XDP support") Signed-off-by: Meghana Malladi Reviewed-by: Jacob Keller Link: https://patch.msgid.link/20250506110546.4065715-3-m-malladi@ti.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/ti/icssg/icssg_common.c | 7 ++++++- drivers/net/ethernet/ti/icssg/icssg_prueth.c | 7 ++++++- 2 files changed, 12 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/ti/icssg/icssg_common.c b/drivers/net/ethernet/ti/icssg/icssg_common.c index b4be76e13a2f..38a80e004d1c 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_common.c +++ b/drivers/net/ethernet/ti/icssg/icssg_common.c @@ -650,6 +650,8 @@ static u32 emac_run_xdp(struct prueth_emac *emac, struct xdp_buff *xdp, struct page *page, u32 *len) { struct net_device *ndev = emac->ndev; + struct netdev_queue *netif_txq; + int cpu = smp_processor_id(); struct bpf_prog *xdp_prog; struct xdp_frame *xdpf; u32 pkt_len = *len; @@ -669,8 +671,11 @@ static u32 emac_run_xdp(struct prueth_emac *emac, struct xdp_buff *xdp, goto drop; } - q_idx = smp_processor_id() % emac->tx_ch_num; + q_idx = cpu % emac->tx_ch_num; + netif_txq = netdev_get_tx_queue(ndev, q_idx); + __netif_tx_lock(netif_txq, cpu); result = emac_xmit_xdp_frame(emac, xdpf, page, q_idx); + __netif_tx_unlock(netif_txq); if (result == ICSSG_XDP_CONSUMED) { ndev->stats.tx_dropped++; goto drop; diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth.c b/drivers/net/ethernet/ti/icssg/icssg_prueth.c index ee35fecf61e7..86fc1278127c 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_prueth.c +++ b/drivers/net/ethernet/ti/icssg/icssg_prueth.c @@ -1075,17 +1075,21 @@ static int emac_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frame { struct prueth_emac *emac = netdev_priv(dev); struct net_device *ndev = emac->ndev; + struct netdev_queue *netif_txq; + int cpu = smp_processor_id(); struct xdp_frame *xdpf; unsigned int q_idx; int nxmit = 0; u32 err; int i; - q_idx = smp_processor_id() % emac->tx_ch_num; + q_idx = cpu % emac->tx_ch_num; + netif_txq = netdev_get_tx_queue(ndev, q_idx); if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK)) return -EINVAL; + __netif_tx_lock(netif_txq, cpu); for (i = 0; i < n; i++) { xdpf = frames[i]; err = emac_xmit_xdp_frame(emac, xdpf, NULL, q_idx); @@ -1095,6 +1099,7 @@ static int emac_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frame } nxmit++; } + __netif_tx_unlock(netif_txq); return nxmit; } From 1884fc85ae6ed0652fa324c66b1d89e25174e73b Mon Sep 17 00:00:00 2001 From: Meghana Malladi Date: Tue, 6 May 2025 16:35:46 +0530 Subject: [PATCH 265/555] net: ti: icssg-prueth: Report BQL before sending XDP packets When sending out any kind of traffic, it is essential that the driver keeps reporting BQL of the number of bytes that have been sent so that BQL can track the amount of data in the queue and prevents it from overflowing. If BQL is not reported, the driver may continue sending packets even when the queue is full, leading to packet loss, congestion and decreased network performance. Currently this is missing in emac_xmit_xdp_frame() and this patch fixes it. Fixes: 62aa3246f462 ("net: ti: icssg-prueth: Add XDP support") Signed-off-by: Meghana Malladi Reviewed-by: Jacob Keller Link: https://patch.msgid.link/20250506110546.4065715-4-m-malladi@ti.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/ti/icssg/icssg_common.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/ti/icssg/icssg_common.c b/drivers/net/ethernet/ti/icssg/icssg_common.c index 38a80e004d1c..d88a0180294e 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_common.c +++ b/drivers/net/ethernet/ti/icssg/icssg_common.c @@ -187,7 +187,6 @@ int emac_tx_complete_packets(struct prueth_emac *emac, int chn, xdp_return_frame(xdpf); break; default: - netdev_err(ndev, "tx_complete: invalid swdata type %d\n", swdata->type); prueth_xmit_free(tx_chn, desc_tx); ndev->stats.tx_dropped++; continue; @@ -567,6 +566,7 @@ u32 emac_xmit_xdp_frame(struct prueth_emac *emac, { struct cppi5_host_desc_t *first_desc; struct net_device *ndev = emac->ndev; + struct netdev_queue *netif_txq; struct prueth_tx_chn *tx_chn; dma_addr_t desc_dma, buf_dma; struct prueth_swdata *swdata; @@ -620,12 +620,17 @@ u32 emac_xmit_xdp_frame(struct prueth_emac *emac, swdata->data.xdpf = xdpf; } + /* Report BQL before sending the packet */ + netif_txq = netdev_get_tx_queue(ndev, tx_chn->id); + netdev_tx_sent_queue(netif_txq, xdpf->len); + cppi5_hdesc_set_pktlen(first_desc, xdpf->len); desc_dma = k3_cppi_desc_pool_virt2dma(tx_chn->desc_pool, first_desc); ret = k3_udma_glue_push_tx_chn(tx_chn->tx_chn, first_desc, desc_dma); if (ret) { netdev_err(ndev, "xdp tx: push failed: %d\n", ret); + netdev_tx_completed_queue(netif_txq, 1, xdpf->len); goto drop_free_descs; } @@ -984,6 +989,7 @@ enum netdev_tx icssg_ndo_start_xmit(struct sk_buff *skb, struct net_device *ndev ret = k3_udma_glue_push_tx_chn(tx_chn->tx_chn, first_desc, desc_dma); if (ret) { netdev_err(ndev, "tx: push failed: %d\n", ret); + netdev_tx_completed_queue(netif_txq, 1, pkt_len); goto drop_free_descs; } From e979a7c79fbc706f6dac913af379ef4caa04d3d5 Mon Sep 17 00:00:00 2001 From: Aaron Kling Date: Tue, 6 May 2025 13:36:59 -0500 Subject: [PATCH 266/555] spi: tegra114: Use value to check for invalid delays A delay unit of 0 is a valid entry, thus it is not valid to check for unused delays. Instead, check the value field; if that is zero, the given delay is unset. Fixes: 4426e6b4ecf6 ("spi: tegra114: Don't fail set_cs_timing when delays are zero") Cc: stable@vger.kernel.org Signed-off-by: Aaron Kling Reviewed-by: Jon Hunter Link: https://patch.msgid.link/20250506-spi-tegra114-fixup-v1-1-136dc2f732f3@gmail.com Signed-off-by: Mark Brown --- drivers/spi/spi-tegra114.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/spi/spi-tegra114.c b/drivers/spi/spi-tegra114.c index 2a8bb798e95b..795a8482c2c7 100644 --- a/drivers/spi/spi-tegra114.c +++ b/drivers/spi/spi-tegra114.c @@ -728,9 +728,9 @@ static int tegra_spi_set_hw_cs_timing(struct spi_device *spi) u32 inactive_cycles; u8 cs_state; - if ((setup->unit && setup->unit != SPI_DELAY_UNIT_SCK) || - (hold->unit && hold->unit != SPI_DELAY_UNIT_SCK) || - (inactive->unit && inactive->unit != SPI_DELAY_UNIT_SCK)) { + if ((setup->value && setup->unit != SPI_DELAY_UNIT_SCK) || + (hold->value && hold->unit != SPI_DELAY_UNIT_SCK) || + (inactive->value && inactive->unit != SPI_DELAY_UNIT_SCK)) { dev_err(&spi->dev, "Invalid delay unit %d, should be SPI_DELAY_UNIT_SCK\n", SPI_DELAY_UNIT_SCK); From 5f93185a757ff38b36f849c659aeef368db15a68 Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Tue, 29 Apr 2025 22:17:00 +0200 Subject: [PATCH 267/555] net: dsa: b53: allow leaky reserved multicast Allow reserved multicast to ignore VLAN membership so STP and other management protocols work without a PVID VLAN configured when using a vlan aware bridge. Fixes: 967dd82ffc52 ("net: dsa: b53: Add support for Broadcom RoboSwitch") Signed-off-by: Jonas Gorski Tested-by: Florian Fainelli Reviewed-by: Florian Fainelli Link: https://patch.msgid.link/20250429201710.330937-2-jonas.gorski@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/b53/b53_common.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index e5ba71897906..62866165ad03 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -373,9 +373,11 @@ static void b53_enable_vlan(struct b53_device *dev, int port, bool enable, b53_read8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL5, &vc5); } + vc1 &= ~VC1_RX_MCST_FWD_EN; + if (enable) { vc0 |= VC0_VLAN_EN | VC0_VID_CHK_EN | VC0_VID_HASH_VID; - vc1 |= VC1_RX_MCST_UNTAG_EN | VC1_RX_MCST_FWD_EN; + vc1 |= VC1_RX_MCST_UNTAG_EN; vc4 &= ~VC4_ING_VID_CHECK_MASK; if (enable_filtering) { vc4 |= VC4_ING_VID_VIO_DROP << VC4_ING_VID_CHECK_S; @@ -393,7 +395,7 @@ static void b53_enable_vlan(struct b53_device *dev, int port, bool enable, } else { vc0 &= ~(VC0_VLAN_EN | VC0_VID_CHK_EN | VC0_VID_HASH_VID); - vc1 &= ~(VC1_RX_MCST_UNTAG_EN | VC1_RX_MCST_FWD_EN); + vc1 &= ~VC1_RX_MCST_UNTAG_EN; vc4 &= ~VC4_ING_VID_CHECK_MASK; vc5 &= ~VC5_DROP_VTABLE_MISS; From 425f11d4cc9bd9e97e6825d9abb2c51a068ca7b5 Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Tue, 29 Apr 2025 22:17:01 +0200 Subject: [PATCH 268/555] net: dsa: b53: keep CPU port always tagged again The Broadcom management header does not carry the original VLAN tag state information, just the ingress port, so for untagged frames we do not know from which VLAN they originated. Therefore keep the CPU port always tagged except for VLAN 0. Fixes the following setup: $ ip link add br0 type bridge vlan_filtering 1 $ ip link set sw1p1 master br0 $ bridge vlan add dev br0 pvid untagged self $ ip link add sw1p2.10 link sw1p2 type vlan id 10 Where VID 10 would stay untagged on the CPU port. Fixes: 2c32a3d3c233 ("net: dsa: b53: Do not force CPU to be always tagged") Signed-off-by: Jonas Gorski Tested-by: Florian Fainelli Reviewed-by: Florian Fainelli Link: https://patch.msgid.link/20250429201710.330937-3-jonas.gorski@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/b53/b53_common.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index 62866165ad03..9d4fb54b4ced 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -1135,6 +1135,11 @@ static int b53_setup(struct dsa_switch *ds) */ ds->untag_bridge_pvid = dev->tag_protocol == DSA_TAG_PROTO_NONE; + /* The switch does not tell us the original VLAN for untagged + * packets, so keep the CPU port always tagged. + */ + ds->untag_vlan_aware_bridge_pvid = true; + ret = b53_reset_switch(dev); if (ret) { dev_err(ds->dev, "failed to reset switch\n"); @@ -1545,6 +1550,9 @@ int b53_vlan_add(struct dsa_switch *ds, int port, if (vlan->vid == 0 && vlan->vid == b53_default_pvid(dev)) untagged = true; + if (vlan->vid > 0 && dsa_is_cpu_port(ds, port)) + untagged = false; + vl->members |= BIT(port); if (untagged && !b53_vlan_port_needs_forced_tagged(ds, port)) vl->untag |= BIT(port); From f480851981043d9bb6447ca9883ade9247b9a0ad Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Tue, 29 Apr 2025 22:17:02 +0200 Subject: [PATCH 269/555] net: dsa: b53: fix clearing PVID of a port Currently the PVID of ports are only set when adding/updating VLANs with PVID set or removing VLANs, but not when clearing the PVID flag of a VLAN. E.g. the following flow $ ip link add br0 type bridge vlan_filtering 1 $ ip link set sw1p1 master bridge $ bridge vlan add dev sw1p1 vid 10 pvid untagged $ bridge vlan add dev sw1p1 vid 10 untagged Would keep the PVID set as 10, despite the flag being cleared. Fix this by checking if we need to unset the PVID on vlan updates. Fixes: a2482d2ce349 ("net: dsa: b53: Plug in VLAN support") Signed-off-by: Jonas Gorski Tested-by: Florian Fainelli Reviewed-by: Florian Fainelli Link: https://patch.msgid.link/20250429201710.330937-4-jonas.gorski@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/b53/b53_common.c | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index 9d4fb54b4ced..65d74c455c57 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -1537,12 +1537,21 @@ int b53_vlan_add(struct dsa_switch *ds, int port, bool untagged = vlan->flags & BRIDGE_VLAN_INFO_UNTAGGED; bool pvid = vlan->flags & BRIDGE_VLAN_INFO_PVID; struct b53_vlan *vl; + u16 old_pvid, new_pvid; int err; err = b53_vlan_prepare(ds, port, vlan); if (err) return err; + b53_read16(dev, B53_VLAN_PAGE, B53_VLAN_PORT_DEF_TAG(port), &old_pvid); + if (pvid) + new_pvid = vlan->vid; + else if (!pvid && vlan->vid == old_pvid) + new_pvid = b53_default_pvid(dev); + else + new_pvid = old_pvid; + vl = &dev->vlans[vlan->vid]; b53_get_vlan_entry(dev, vlan->vid, vl); @@ -1562,9 +1571,9 @@ int b53_vlan_add(struct dsa_switch *ds, int port, b53_set_vlan_entry(dev, vlan->vid, vl); b53_fast_age_vlan(dev, vlan->vid); - if (pvid && !dsa_is_cpu_port(ds, port)) { + if (!dsa_is_cpu_port(ds, port) && new_pvid != old_pvid) { b53_write16(dev, B53_VLAN_PAGE, B53_VLAN_PORT_DEF_TAG(port), - vlan->vid); + new_pvid); b53_fast_age_vlan(dev, vlan->vid); } From 083c6b28c0cbcd83b6af1a10f2c82937129b3438 Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Tue, 29 Apr 2025 22:17:03 +0200 Subject: [PATCH 270/555] net: dsa: b53: fix flushing old pvid VLAN on pvid change Presumably the intention here was to flush the VLAN of the old pvid, not the added VLAN again, which we already flushed before. Fixes: a2482d2ce349 ("net: dsa: b53: Plug in VLAN support") Signed-off-by: Jonas Gorski Tested-by: Florian Fainelli Reviewed-by: Florian Fainelli Link: https://patch.msgid.link/20250429201710.330937-5-jonas.gorski@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/b53/b53_common.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index 65d74c455c57..c67c0b5fbc1b 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -1574,7 +1574,7 @@ int b53_vlan_add(struct dsa_switch *ds, int port, if (!dsa_is_cpu_port(ds, port) && new_pvid != old_pvid) { b53_write16(dev, B53_VLAN_PAGE, B53_VLAN_PORT_DEF_TAG(port), new_pvid); - b53_fast_age_vlan(dev, vlan->vid); + b53_fast_age_vlan(dev, old_pvid); } return 0; From a1c1901c5cc881425cc45992ab6c5418174e9e5a Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Tue, 29 Apr 2025 22:17:04 +0200 Subject: [PATCH 271/555] net: dsa: b53: fix VLAN ID for untagged vlan on bridge leave The untagged default VLAN is added to the default vlan, which may be one, but we modify the VLAN 0 entry on bridge leave. Fix this to use the correct VLAN entry for the default pvid. Fixes: fea83353177a ("net: dsa: b53: Fix default VLAN ID") Signed-off-by: Jonas Gorski Tested-by: Florian Fainelli Reviewed-by: Florian Fainelli Link: https://patch.msgid.link/20250429201710.330937-6-jonas.gorski@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/b53/b53_common.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index c67c0b5fbc1b..c60b552b945c 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -1986,7 +1986,7 @@ EXPORT_SYMBOL(b53_br_join); void b53_br_leave(struct dsa_switch *ds, int port, struct dsa_bridge bridge) { struct b53_device *dev = ds->priv; - struct b53_vlan *vl = &dev->vlans[0]; + struct b53_vlan *vl; s8 cpu_port = dsa_to_port(ds, port)->cpu_dp->index; unsigned int i; u16 pvlan, reg, pvid; @@ -2012,6 +2012,7 @@ void b53_br_leave(struct dsa_switch *ds, int port, struct dsa_bridge bridge) dev->ports[port].vlan_ctl_mask = pvlan; pvid = b53_default_pvid(dev); + vl = &dev->vlans[pvid]; /* Make this port join all VLANs without VLAN entries */ if (is58xx(dev)) { From 13b152ae40495966501697693f048f47430c50fd Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Tue, 29 Apr 2025 22:17:05 +0200 Subject: [PATCH 272/555] net: dsa: b53: always rejoin default untagged VLAN on bridge leave While JOIN_ALL_VLAN allows to join all VLANs, we still need to keep the default VLAN enabled so that untagged traffic stays untagged. So rejoin the default VLAN even for switches with JOIN_ALL_VLAN support. Fixes: 48aea33a77ab ("net: dsa: b53: Add JOIN_ALL_VLAN support") Signed-off-by: Jonas Gorski Tested-by: Florian Fainelli Reviewed-by: Florian Fainelli Link: https://patch.msgid.link/20250429201710.330937-7-jonas.gorski@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/b53/b53_common.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index c60b552b945c..4871e117f5ef 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -2021,12 +2021,12 @@ void b53_br_leave(struct dsa_switch *ds, int port, struct dsa_bridge bridge) if (!(reg & BIT(cpu_port))) reg |= BIT(cpu_port); b53_write16(dev, B53_VLAN_PAGE, B53_JOIN_ALL_VLAN_EN, reg); - } else { - b53_get_vlan_entry(dev, pvid, vl); - vl->members |= BIT(port) | BIT(cpu_port); - vl->untag |= BIT(port) | BIT(cpu_port); - b53_set_vlan_entry(dev, pvid, vl); } + + b53_get_vlan_entry(dev, pvid, vl); + vl->members |= BIT(port) | BIT(cpu_port); + vl->untag |= BIT(port) | BIT(cpu_port); + b53_set_vlan_entry(dev, pvid, vl); } EXPORT_SYMBOL(b53_br_leave); From 45e9d59d39503bb3e6ab4d258caea4ba6496e2dc Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Tue, 29 Apr 2025 22:17:06 +0200 Subject: [PATCH 273/555] net: dsa: b53: do not allow to configure VLAN 0 Since we cannot set forwarding destinations per VLAN, we should not have a VLAN 0 configured, as it would allow untagged traffic to work across ports on VLAN aware bridges regardless if a PVID untagged VLAN exists. So remove the VLAN 0 on join, an re-add it on leave. But only do so if we have a VLAN aware bridge, as without it, untagged traffic would become tagged with VID 0 on a VLAN unaware bridge. Fixes: a2482d2ce349 ("net: dsa: b53: Plug in VLAN support") Signed-off-by: Jonas Gorski Tested-by: Florian Fainelli Reviewed-by: Florian Fainelli Link: https://patch.msgid.link/20250429201710.330937-8-jonas.gorski@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/b53/b53_common.c | 36 ++++++++++++++++++++++++-------- 1 file changed, 27 insertions(+), 9 deletions(-) diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index 4871e117f5ef..0b28791cca52 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -1544,6 +1544,9 @@ int b53_vlan_add(struct dsa_switch *ds, int port, if (err) return err; + if (vlan->vid == 0) + return 0; + b53_read16(dev, B53_VLAN_PAGE, B53_VLAN_PORT_DEF_TAG(port), &old_pvid); if (pvid) new_pvid = vlan->vid; @@ -1556,10 +1559,7 @@ int b53_vlan_add(struct dsa_switch *ds, int port, b53_get_vlan_entry(dev, vlan->vid, vl); - if (vlan->vid == 0 && vlan->vid == b53_default_pvid(dev)) - untagged = true; - - if (vlan->vid > 0 && dsa_is_cpu_port(ds, port)) + if (dsa_is_cpu_port(ds, port)) untagged = false; vl->members |= BIT(port); @@ -1589,6 +1589,9 @@ int b53_vlan_del(struct dsa_switch *ds, int port, struct b53_vlan *vl; u16 pvid; + if (vlan->vid == 0) + return 0; + b53_read16(dev, B53_VLAN_PAGE, B53_VLAN_PORT_DEF_TAG(port), &pvid); vl = &dev->vlans[vlan->vid]; @@ -1935,8 +1938,9 @@ int b53_br_join(struct dsa_switch *ds, int port, struct dsa_bridge bridge, bool *tx_fwd_offload, struct netlink_ext_ack *extack) { struct b53_device *dev = ds->priv; + struct b53_vlan *vl; s8 cpu_port = dsa_to_port(ds, port)->cpu_dp->index; - u16 pvlan, reg; + u16 pvlan, reg, pvid; unsigned int i; /* On 7278, port 7 which connects to the ASP should only receive @@ -1945,6 +1949,9 @@ int b53_br_join(struct dsa_switch *ds, int port, struct dsa_bridge bridge, if (dev->chip_id == BCM7278_DEVICE_ID && port == 7) return -EINVAL; + pvid = b53_default_pvid(dev); + vl = &dev->vlans[pvid]; + /* Make this port leave the all VLANs join since we will have proper * VLAN entries from now on */ @@ -1956,6 +1963,15 @@ int b53_br_join(struct dsa_switch *ds, int port, struct dsa_bridge bridge, b53_write16(dev, B53_VLAN_PAGE, B53_JOIN_ALL_VLAN_EN, reg); } + if (ds->vlan_filtering) { + b53_get_vlan_entry(dev, pvid, vl); + vl->members &= ~BIT(port); + if (vl->members == BIT(cpu_port)) + vl->members &= ~BIT(cpu_port); + vl->untag = vl->members; + b53_set_vlan_entry(dev, pvid, vl); + } + b53_read16(dev, B53_PVLAN_PAGE, B53_PVLAN_PORT_MASK(port), &pvlan); b53_for_each_port(dev, i) { @@ -2023,10 +2039,12 @@ void b53_br_leave(struct dsa_switch *ds, int port, struct dsa_bridge bridge) b53_write16(dev, B53_VLAN_PAGE, B53_JOIN_ALL_VLAN_EN, reg); } - b53_get_vlan_entry(dev, pvid, vl); - vl->members |= BIT(port) | BIT(cpu_port); - vl->untag |= BIT(port) | BIT(cpu_port); - b53_set_vlan_entry(dev, pvid, vl); + if (ds->vlan_filtering) { + b53_get_vlan_entry(dev, pvid, vl); + vl->members |= BIT(port) | BIT(cpu_port); + vl->untag |= BIT(port) | BIT(cpu_port); + b53_set_vlan_entry(dev, pvid, vl); + } } EXPORT_SYMBOL(b53_br_leave); From f089652b6b16452535dcc5cbaa6e2bb05acd3f93 Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Tue, 29 Apr 2025 22:17:07 +0200 Subject: [PATCH 274/555] net: dsa: b53: do not program vlans when vlan filtering is off Documentation/networking/switchdev.rst says: - with VLAN filtering turned off: the bridge is strictly VLAN unaware and its data path will process all Ethernet frames as if they are VLAN-untagged. The bridge VLAN database can still be modified, but the modifications should have no effect while VLAN filtering is turned off. This breaks if we immediately apply the VLAN configuration, so skip writing it when vlan_filtering is off. Fixes: 0ee2af4ebbe3 ("net: dsa: set configure_vlan_while_not_filtering to true by default") Signed-off-by: Jonas Gorski Tested-by: Florian Fainelli Reviewed-by: Florian Fainelli Link: https://patch.msgid.link/20250429201710.330937-9-jonas.gorski@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/b53/b53_common.c | 48 +++++++++++++++++++------------- 1 file changed, 28 insertions(+), 20 deletions(-) diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index 0b28791cca52..ee2f1be62618 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -1547,6 +1547,9 @@ int b53_vlan_add(struct dsa_switch *ds, int port, if (vlan->vid == 0) return 0; + if (!ds->vlan_filtering) + return 0; + b53_read16(dev, B53_VLAN_PAGE, B53_VLAN_PORT_DEF_TAG(port), &old_pvid); if (pvid) new_pvid = vlan->vid; @@ -1592,6 +1595,9 @@ int b53_vlan_del(struct dsa_switch *ds, int port, if (vlan->vid == 0) return 0; + if (!ds->vlan_filtering) + return 0; + b53_read16(dev, B53_VLAN_PAGE, B53_VLAN_PORT_DEF_TAG(port), &pvid); vl = &dev->vlans[vlan->vid]; @@ -1952,18 +1958,20 @@ int b53_br_join(struct dsa_switch *ds, int port, struct dsa_bridge bridge, pvid = b53_default_pvid(dev); vl = &dev->vlans[pvid]; - /* Make this port leave the all VLANs join since we will have proper - * VLAN entries from now on - */ - if (is58xx(dev)) { - b53_read16(dev, B53_VLAN_PAGE, B53_JOIN_ALL_VLAN_EN, ®); - reg &= ~BIT(port); - if ((reg & BIT(cpu_port)) == BIT(cpu_port)) - reg &= ~BIT(cpu_port); - b53_write16(dev, B53_VLAN_PAGE, B53_JOIN_ALL_VLAN_EN, reg); - } - if (ds->vlan_filtering) { + /* Make this port leave the all VLANs join since we will have + * proper VLAN entries from now on + */ + if (is58xx(dev)) { + b53_read16(dev, B53_VLAN_PAGE, B53_JOIN_ALL_VLAN_EN, + ®); + reg &= ~BIT(port); + if ((reg & BIT(cpu_port)) == BIT(cpu_port)) + reg &= ~BIT(cpu_port); + b53_write16(dev, B53_VLAN_PAGE, B53_JOIN_ALL_VLAN_EN, + reg); + } + b53_get_vlan_entry(dev, pvid, vl); vl->members &= ~BIT(port); if (vl->members == BIT(cpu_port)) @@ -2030,16 +2038,16 @@ void b53_br_leave(struct dsa_switch *ds, int port, struct dsa_bridge bridge) pvid = b53_default_pvid(dev); vl = &dev->vlans[pvid]; - /* Make this port join all VLANs without VLAN entries */ - if (is58xx(dev)) { - b53_read16(dev, B53_VLAN_PAGE, B53_JOIN_ALL_VLAN_EN, ®); - reg |= BIT(port); - if (!(reg & BIT(cpu_port))) - reg |= BIT(cpu_port); - b53_write16(dev, B53_VLAN_PAGE, B53_JOIN_ALL_VLAN_EN, reg); - } - if (ds->vlan_filtering) { + /* Make this port join all VLANs without VLAN entries */ + if (is58xx(dev)) { + b53_read16(dev, B53_VLAN_PAGE, B53_JOIN_ALL_VLAN_EN, ®); + reg |= BIT(port); + if (!(reg & BIT(cpu_port))) + reg |= BIT(cpu_port); + b53_write16(dev, B53_VLAN_PAGE, B53_JOIN_ALL_VLAN_EN, reg); + } + b53_get_vlan_entry(dev, pvid, vl); vl->members |= BIT(port) | BIT(cpu_port); vl->untag |= BIT(port) | BIT(cpu_port); From 2dc2bd57111582895e10f54ea380329c89873f1c Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Tue, 29 Apr 2025 22:17:08 +0200 Subject: [PATCH 275/555] net: dsa: b53: fix toggling vlan_filtering To allow runtime switching between vlan aware and vlan non-aware mode, we need to properly keep track of any bridge VLAN configuration. Likewise, we need to know when we actually switch between both modes, to not have to rewrite the full VLAN table every time we update the VLANs. So keep track of the current vlan_filtering mode, and on changes, apply the appropriate VLAN configuration. Fixes: 0ee2af4ebbe3 ("net: dsa: set configure_vlan_while_not_filtering to true by default") Signed-off-by: Jonas Gorski Tested-by: Florian Fainelli Reviewed-by: Florian Fainelli Link: https://patch.msgid.link/20250429201710.330937-10-jonas.gorski@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/b53/b53_common.c | 110 +++++++++++++++++++++---------- drivers/net/dsa/b53/b53_priv.h | 2 + 2 files changed, 78 insertions(+), 34 deletions(-) diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index ee2f1be62618..0a7749b416f7 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -763,6 +763,22 @@ static bool b53_vlan_port_needs_forced_tagged(struct dsa_switch *ds, int port) return dev->tag_protocol == DSA_TAG_PROTO_NONE && dsa_is_cpu_port(ds, port); } +static bool b53_vlan_port_may_join_untagged(struct dsa_switch *ds, int port) +{ + struct b53_device *dev = ds->priv; + struct dsa_port *dp; + + if (!dev->vlan_filtering) + return true; + + dp = dsa_to_port(ds, port); + + if (dsa_port_is_cpu(dp)) + return true; + + return dp->bridge == NULL; +} + int b53_configure_vlan(struct dsa_switch *ds) { struct b53_device *dev = ds->priv; @@ -781,7 +797,7 @@ int b53_configure_vlan(struct dsa_switch *ds) b53_do_vlan_op(dev, VTA_CMD_CLEAR); } - b53_enable_vlan(dev, -1, dev->vlan_enabled, ds->vlan_filtering); + b53_enable_vlan(dev, -1, dev->vlan_enabled, dev->vlan_filtering); /* Create an untagged VLAN entry for the default PVID in case * CONFIG_VLAN_8021Q is disabled and there are no calls to @@ -789,26 +805,39 @@ int b53_configure_vlan(struct dsa_switch *ds) * entry. Do this only when the tagging protocol is not * DSA_TAG_PROTO_NONE */ + v = &dev->vlans[def_vid]; b53_for_each_port(dev, i) { - v = &dev->vlans[def_vid]; - v->members |= BIT(i); - if (!b53_vlan_port_needs_forced_tagged(ds, i)) - v->untag = v->members; - b53_write16(dev, B53_VLAN_PAGE, - B53_VLAN_PORT_DEF_TAG(i), def_vid); - } - - /* Upon initial call we have not set-up any VLANs, but upon - * system resume, we need to restore all VLAN entries. - */ - for (vid = def_vid; vid < dev->num_vlans; vid++) { - v = &dev->vlans[vid]; - - if (!v->members) + if (!b53_vlan_port_may_join_untagged(ds, i)) continue; - b53_set_vlan_entry(dev, vid, v); - b53_fast_age_vlan(dev, vid); + vl.members |= BIT(i); + if (!b53_vlan_port_needs_forced_tagged(ds, i)) + vl.untag = vl.members; + b53_write16(dev, B53_VLAN_PAGE, B53_VLAN_PORT_DEF_TAG(i), + def_vid); + } + b53_set_vlan_entry(dev, def_vid, &vl); + + if (dev->vlan_filtering) { + /* Upon initial call we have not set-up any VLANs, but upon + * system resume, we need to restore all VLAN entries. + */ + for (vid = def_vid + 1; vid < dev->num_vlans; vid++) { + v = &dev->vlans[vid]; + + if (!v->members) + continue; + + b53_set_vlan_entry(dev, vid, v); + b53_fast_age_vlan(dev, vid); + } + + b53_for_each_port(dev, i) { + if (!dsa_is_cpu_port(ds, i)) + b53_write16(dev, B53_VLAN_PAGE, + B53_VLAN_PORT_DEF_TAG(i), + dev->ports[i].pvid); + } } return 0; @@ -1127,7 +1156,9 @@ EXPORT_SYMBOL(b53_setup_devlink_resources); static int b53_setup(struct dsa_switch *ds) { struct b53_device *dev = ds->priv; + struct b53_vlan *vl; unsigned int port; + u16 pvid; int ret; /* Request bridge PVID untagged when DSA_TAG_PROTO_NONE is set @@ -1146,6 +1177,15 @@ static int b53_setup(struct dsa_switch *ds) return ret; } + /* setup default vlan for filtering mode */ + pvid = b53_default_pvid(dev); + vl = &dev->vlans[pvid]; + b53_for_each_port(dev, port) { + vl->members |= BIT(port); + if (!b53_vlan_port_needs_forced_tagged(ds, port)) + vl->untag |= BIT(port); + } + b53_reset_mib(dev); ret = b53_apply_config(dev); @@ -1499,7 +1539,10 @@ int b53_vlan_filtering(struct dsa_switch *ds, int port, bool vlan_filtering, { struct b53_device *dev = ds->priv; - b53_enable_vlan(dev, port, dev->vlan_enabled, vlan_filtering); + if (dev->vlan_filtering != vlan_filtering) { + dev->vlan_filtering = vlan_filtering; + b53_apply_config(dev); + } return 0; } @@ -1524,7 +1567,7 @@ static int b53_vlan_prepare(struct dsa_switch *ds, int port, if (vlan->vid >= dev->num_vlans) return -ERANGE; - b53_enable_vlan(dev, port, true, ds->vlan_filtering); + b53_enable_vlan(dev, port, true, dev->vlan_filtering); return 0; } @@ -1547,21 +1590,17 @@ int b53_vlan_add(struct dsa_switch *ds, int port, if (vlan->vid == 0) return 0; - if (!ds->vlan_filtering) - return 0; - - b53_read16(dev, B53_VLAN_PAGE, B53_VLAN_PORT_DEF_TAG(port), &old_pvid); + old_pvid = dev->ports[port].pvid; if (pvid) new_pvid = vlan->vid; else if (!pvid && vlan->vid == old_pvid) new_pvid = b53_default_pvid(dev); else new_pvid = old_pvid; + dev->ports[port].pvid = new_pvid; vl = &dev->vlans[vlan->vid]; - b53_get_vlan_entry(dev, vlan->vid, vl); - if (dsa_is_cpu_port(ds, port)) untagged = false; @@ -1571,6 +1610,9 @@ int b53_vlan_add(struct dsa_switch *ds, int port, else vl->untag &= ~BIT(port); + if (!dev->vlan_filtering) + return 0; + b53_set_vlan_entry(dev, vlan->vid, vl); b53_fast_age_vlan(dev, vlan->vid); @@ -1595,23 +1637,22 @@ int b53_vlan_del(struct dsa_switch *ds, int port, if (vlan->vid == 0) return 0; - if (!ds->vlan_filtering) - return 0; - - b53_read16(dev, B53_VLAN_PAGE, B53_VLAN_PORT_DEF_TAG(port), &pvid); + pvid = dev->ports[port].pvid; vl = &dev->vlans[vlan->vid]; - b53_get_vlan_entry(dev, vlan->vid, vl); - vl->members &= ~BIT(port); if (pvid == vlan->vid) pvid = b53_default_pvid(dev); + dev->ports[port].pvid = pvid; if (untagged && !b53_vlan_port_needs_forced_tagged(ds, port)) vl->untag &= ~(BIT(port)); + if (!dev->vlan_filtering) + return 0; + b53_set_vlan_entry(dev, vlan->vid, vl); b53_fast_age_vlan(dev, vlan->vid); @@ -1958,7 +1999,7 @@ int b53_br_join(struct dsa_switch *ds, int port, struct dsa_bridge bridge, pvid = b53_default_pvid(dev); vl = &dev->vlans[pvid]; - if (ds->vlan_filtering) { + if (dev->vlan_filtering) { /* Make this port leave the all VLANs join since we will have * proper VLAN entries from now on */ @@ -2038,7 +2079,7 @@ void b53_br_leave(struct dsa_switch *ds, int port, struct dsa_bridge bridge) pvid = b53_default_pvid(dev); vl = &dev->vlans[pvid]; - if (ds->vlan_filtering) { + if (dev->vlan_filtering) { /* Make this port join all VLANs without VLAN entries */ if (is58xx(dev)) { b53_read16(dev, B53_VLAN_PAGE, B53_JOIN_ALL_VLAN_EN, ®); @@ -2803,6 +2844,7 @@ struct b53_device *b53_switch_alloc(struct device *base, ds->ops = &b53_switch_ops; ds->phylink_mac_ops = &b53_phylink_mac_ops; dev->vlan_enabled = true; + dev->vlan_filtering = false; /* Let DSA handle the case were multiple bridges span the same switch * device and different VLAN awareness settings are requested, which * would be breaking filtering semantics for any of the other bridge diff --git a/drivers/net/dsa/b53/b53_priv.h b/drivers/net/dsa/b53/b53_priv.h index 0166c37a13a7..4636e27fd1ee 100644 --- a/drivers/net/dsa/b53/b53_priv.h +++ b/drivers/net/dsa/b53/b53_priv.h @@ -96,6 +96,7 @@ struct b53_pcs { struct b53_port { u16 vlan_ctl_mask; + u16 pvid; struct ethtool_keee eee; }; @@ -147,6 +148,7 @@ struct b53_device { unsigned int num_vlans; struct b53_vlan *vlans; bool vlan_enabled; + bool vlan_filtering; unsigned int num_ports; struct b53_port *ports; From 9f34ad89bcf0e6df6f8b01f1bdab211493fc66d1 Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Tue, 29 Apr 2025 22:17:09 +0200 Subject: [PATCH 276/555] net: dsa: b53: fix learning on VLAN unaware bridges When VLAN filtering is off, we configure the switch to forward, but not learn on VLAN table misses. This effectively disables learning while not filtering. Fix this by switching to forward and learn. Setting the learning disable register will still control whether learning actually happens. Fixes: dad8d7c6452b ("net: dsa: b53: Properly account for VLAN filtering") Signed-off-by: Jonas Gorski Tested-by: Florian Fainelli Reviewed-by: Florian Fainelli Link: https://patch.msgid.link/20250429201710.330937-11-jonas.gorski@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/b53/b53_common.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index 0a7749b416f7..a2c0b44fc6be 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -383,7 +383,7 @@ static void b53_enable_vlan(struct b53_device *dev, int port, bool enable, vc4 |= VC4_ING_VID_VIO_DROP << VC4_ING_VID_CHECK_S; vc5 |= VC5_DROP_VTABLE_MISS; } else { - vc4 |= VC4_ING_VID_VIO_FWD << VC4_ING_VID_CHECK_S; + vc4 |= VC4_NO_ING_VID_CHK << VC4_ING_VID_CHECK_S; vc5 &= ~VC5_DROP_VTABLE_MISS; } From 2e7179c628d3cb9aee75e412473813b099e11ed4 Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Tue, 29 Apr 2025 22:17:10 +0200 Subject: [PATCH 277/555] net: dsa: b53: do not set learning and unicast/multicast on up When a port gets set up, b53 disables learning and enables the port for flooding. This can undo any bridge configuration on the port. E.g. the following flow would disable learning on a port: $ ip link add br0 type bridge $ ip link set sw1p1 master br0 <- enables learning for sw1p1 $ ip link set br0 up $ ip link set sw1p1 up <- disables learning again Fix this by populating dsa_switch_ops::port_setup(), and set up initial config there. Fixes: f9b3827ee66c ("net: dsa: b53: Support setting learning on port") Signed-off-by: Jonas Gorski Tested-by: Florian Fainelli Reviewed-by: Florian Fainelli Link: https://patch.msgid.link/20250429201710.330937-12-jonas.gorski@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/b53/b53_common.c | 21 +++++++++++++-------- drivers/net/dsa/b53/b53_priv.h | 1 + drivers/net/dsa/bcm_sf2.c | 1 + 3 files changed, 15 insertions(+), 8 deletions(-) diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index a2c0b44fc6be..9eb39cfa5fb2 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -578,6 +578,18 @@ static void b53_eee_enable_set(struct dsa_switch *ds, int port, bool enable) b53_write16(dev, B53_EEE_PAGE, B53_EEE_EN_CTRL, reg); } +int b53_setup_port(struct dsa_switch *ds, int port) +{ + struct b53_device *dev = ds->priv; + + b53_port_set_ucast_flood(dev, port, true); + b53_port_set_mcast_flood(dev, port, true); + b53_port_set_learning(dev, port, false); + + return 0; +} +EXPORT_SYMBOL(b53_setup_port); + int b53_enable_port(struct dsa_switch *ds, int port, struct phy_device *phy) { struct b53_device *dev = ds->priv; @@ -590,10 +602,6 @@ int b53_enable_port(struct dsa_switch *ds, int port, struct phy_device *phy) cpu_port = dsa_to_port(ds, port)->cpu_dp->index; - b53_port_set_ucast_flood(dev, port, true); - b53_port_set_mcast_flood(dev, port, true); - b53_port_set_learning(dev, port, false); - if (dev->ops->irq_enable) ret = dev->ops->irq_enable(dev, port); if (ret) @@ -724,10 +732,6 @@ static void b53_enable_cpu_port(struct b53_device *dev, int port) b53_write8(dev, B53_CTRL_PAGE, B53_PORT_CTRL(port), port_ctrl); b53_brcm_hdr_setup(dev->ds, port); - - b53_port_set_ucast_flood(dev, port, true); - b53_port_set_mcast_flood(dev, port, true); - b53_port_set_learning(dev, port, false); } static void b53_enable_mib(struct b53_device *dev) @@ -2387,6 +2391,7 @@ static const struct dsa_switch_ops b53_switch_ops = { .phy_read = b53_phy_read16, .phy_write = b53_phy_write16, .phylink_get_caps = b53_phylink_get_caps, + .port_setup = b53_setup_port, .port_enable = b53_enable_port, .port_disable = b53_disable_port, .support_eee = b53_support_eee, diff --git a/drivers/net/dsa/b53/b53_priv.h b/drivers/net/dsa/b53/b53_priv.h index 4636e27fd1ee..2cf3e6a81e37 100644 --- a/drivers/net/dsa/b53/b53_priv.h +++ b/drivers/net/dsa/b53/b53_priv.h @@ -384,6 +384,7 @@ enum dsa_tag_protocol b53_get_tag_protocol(struct dsa_switch *ds, int port, enum dsa_tag_protocol mprot); void b53_mirror_del(struct dsa_switch *ds, int port, struct dsa_mall_mirror_tc_entry *mirror); +int b53_setup_port(struct dsa_switch *ds, int port); int b53_enable_port(struct dsa_switch *ds, int port, struct phy_device *phy); void b53_disable_port(struct dsa_switch *ds, int port); void b53_brcm_hdr_setup(struct dsa_switch *ds, int port); diff --git a/drivers/net/dsa/bcm_sf2.c b/drivers/net/dsa/bcm_sf2.c index fa2bf3fa9019..454a8c7fd7ee 100644 --- a/drivers/net/dsa/bcm_sf2.c +++ b/drivers/net/dsa/bcm_sf2.c @@ -1230,6 +1230,7 @@ static const struct dsa_switch_ops bcm_sf2_ops = { .resume = bcm_sf2_sw_resume, .get_wol = bcm_sf2_sw_get_wol, .set_wol = bcm_sf2_sw_set_wol, + .port_setup = b53_setup_port, .port_enable = bcm_sf2_port_setup, .port_disable = bcm_sf2_port_disable, .support_eee = b53_support_eee, From df84d2fd35c6f2d6b2241e47f26e1829eab26186 Mon Sep 17 00:00:00 2001 From: Lance Yang Date: Thu, 17 Apr 2025 23:30:41 +0800 Subject: [PATCH 278/555] mailmap: add entries for Lance Yang I'm moving to @linux.dev and mapping my old addresses to it. Link: https://lkml.kernel.org/r/20250417153041.38977-1-lance.yang@linux.dev Signed-off-by: Lance Yang Signed-off-by: Andrew Morton --- .mailmap | 2 ++ 1 file changed, 2 insertions(+) diff --git a/.mailmap b/.mailmap index 9afde79e1936..6f0a4e965d6a 100644 --- a/.mailmap +++ b/.mailmap @@ -447,6 +447,8 @@ Luca Ceresoli Luca Weiss Lukasz Luba Luo Jie +Lance Yang +Lance Yang Maciej W. Rozycki Maciej W. Rozycki Maharaja Kennadyrajan From bd1261b16d9131d79723d982d54295e7f309797a Mon Sep 17 00:00:00 2001 From: Heming Zhao Date: Mon, 14 Apr 2025 14:01:23 +0800 Subject: [PATCH 279/555] ocfs2: fix the issue with discontiguous allocation in the global_bitmap commit 4eb7b93e0310 ("ocfs2: improve write IO performance when fragmentation is high") introduced another regression. The following ocfs2-test case can trigger this issue: > discontig_runner.sh => activate_discontig_bg.sh => resv_unwritten: > ${RESV_UNWRITTEN_BIN} -f ${WORK_PLACE}/large_testfile -s 0 -l \ > $((${FILE_MAJOR_SIZE_M}*1024*1024)) In my env, test disk size (by "fdisk -l "): > 53687091200 bytes, 104857600 sectors. Above command is: > /usr/local/ocfs2-test/bin/resv_unwritten -f \ > /mnt/ocfs2/ocfs2-activate-discontig-bg-dir/large_testfile -s 0 -l \ > 53187969024 Error log: > [*] Reserve 50724M space for a LARGE file, reserve 200M space for future test. > ioctl error 28: "No space left on device" > resv allocation failed Unknown error -1 > reserve unwritten region from 0 to 53187969024. Call flow: __ocfs2_change_file_space //by ioctl OCFS2_IOC_RESVSP64 ocfs2_allocate_unwritten_extents //start:0 len:53187969024 while() + ocfs2_get_clusters //cpos:0, alloc_size:1623168 (cluster number) + ocfs2_extend_allocation + ocfs2_lock_allocators | + choose OCFS2_AC_USE_MAIN & ocfs2_cluster_group_search | + ocfs2_add_inode_data ocfs2_add_clusters_in_btree __ocfs2_claim_clusters ocfs2_claim_suballoc_bits + During the allocation of the final part of the large file (after ~47GB), no chain had the required contiguous bits_wanted. Consequently, the allocation failed. How to fix: When OCFS2 is encountering fragmented allocation, the file system should stop attempting bits_wanted contiguous allocation and instead provide the largest available contiguous free bits from the cluster groups. Link: https://lkml.kernel.org/r/20250414060125.19938-2-heming.zhao@suse.com Fixes: 4eb7b93e0310 ("ocfs2: improve write IO performance when fragmentation is high") Signed-off-by: Heming Zhao Reported-by: Gautham Ananthakrishna Reviewed-by: Joseph Qi Cc: Mark Fasheh Cc: Joel Becker Cc: Junxiao Bi Cc: Changwei Ge Cc: Jun Piao Cc: Signed-off-by: Andrew Morton --- fs/ocfs2/suballoc.c | 38 ++++++++++++++++++++++++++++++++------ fs/ocfs2/suballoc.h | 1 + 2 files changed, 33 insertions(+), 6 deletions(-) diff --git a/fs/ocfs2/suballoc.c b/fs/ocfs2/suballoc.c index f7b483f0de2a..6ac4dcd54588 100644 --- a/fs/ocfs2/suballoc.c +++ b/fs/ocfs2/suballoc.c @@ -698,10 +698,12 @@ static int ocfs2_block_group_alloc(struct ocfs2_super *osb, bg_bh = ocfs2_block_group_alloc_contig(osb, handle, alloc_inode, ac, cl); - if (PTR_ERR(bg_bh) == -ENOSPC) + if (PTR_ERR(bg_bh) == -ENOSPC) { + ac->ac_which = OCFS2_AC_USE_MAIN_DISCONTIG; bg_bh = ocfs2_block_group_alloc_discontig(handle, alloc_inode, ac, cl); + } if (IS_ERR(bg_bh)) { status = PTR_ERR(bg_bh); bg_bh = NULL; @@ -1794,6 +1796,7 @@ static int ocfs2_search_chain(struct ocfs2_alloc_context *ac, { int status; u16 chain; + u32 contig_bits; u64 next_group; struct inode *alloc_inode = ac->ac_inode; struct buffer_head *group_bh = NULL; @@ -1819,10 +1822,21 @@ static int ocfs2_search_chain(struct ocfs2_alloc_context *ac, status = -ENOSPC; /* for now, the chain search is a bit simplistic. We just use * the 1st group with any empty bits. */ - while ((status = ac->ac_group_search(alloc_inode, group_bh, - bits_wanted, min_bits, - ac->ac_max_block, - res)) == -ENOSPC) { + while (1) { + if (ac->ac_which == OCFS2_AC_USE_MAIN_DISCONTIG) { + contig_bits = le16_to_cpu(bg->bg_contig_free_bits); + if (!contig_bits) + contig_bits = ocfs2_find_max_contig_free_bits(bg->bg_bitmap, + le16_to_cpu(bg->bg_bits), 0); + if (bits_wanted > contig_bits && contig_bits >= min_bits) + bits_wanted = contig_bits; + } + + status = ac->ac_group_search(alloc_inode, group_bh, + bits_wanted, min_bits, + ac->ac_max_block, res); + if (status != -ENOSPC) + break; if (!bg->bg_next_group) break; @@ -1982,6 +1996,7 @@ static int ocfs2_claim_suballoc_bits(struct ocfs2_alloc_context *ac, victim = ocfs2_find_victim_chain(cl); ac->ac_chain = victim; +search: status = ocfs2_search_chain(ac, handle, bits_wanted, min_bits, res, &bits_left); if (!status) { @@ -2022,6 +2037,16 @@ static int ocfs2_claim_suballoc_bits(struct ocfs2_alloc_context *ac, } } + /* Chains can't supply the bits_wanted contiguous space. + * We should switch to using every single bit when allocating + * from the global bitmap. */ + if (i == le16_to_cpu(cl->cl_next_free_rec) && + status == -ENOSPC && ac->ac_which == OCFS2_AC_USE_MAIN) { + ac->ac_which = OCFS2_AC_USE_MAIN_DISCONTIG; + ac->ac_chain = victim; + goto search; + } + set_hint: if (status != -ENOSPC) { /* If the next search of this group is not likely to @@ -2365,7 +2390,8 @@ int __ocfs2_claim_clusters(handle_t *handle, BUG_ON(ac->ac_bits_given >= ac->ac_bits_wanted); BUG_ON(ac->ac_which != OCFS2_AC_USE_LOCAL - && ac->ac_which != OCFS2_AC_USE_MAIN); + && ac->ac_which != OCFS2_AC_USE_MAIN + && ac->ac_which != OCFS2_AC_USE_MAIN_DISCONTIG); if (ac->ac_which == OCFS2_AC_USE_LOCAL) { WARN_ON(min_clusters > 1); diff --git a/fs/ocfs2/suballoc.h b/fs/ocfs2/suballoc.h index b481b834857d..bcf2ed4a8631 100644 --- a/fs/ocfs2/suballoc.h +++ b/fs/ocfs2/suballoc.h @@ -29,6 +29,7 @@ struct ocfs2_alloc_context { #define OCFS2_AC_USE_MAIN 2 #define OCFS2_AC_USE_INODE 3 #define OCFS2_AC_USE_META 4 +#define OCFS2_AC_USE_MAIN_DISCONTIG 5 u32 ac_which; /* these are used by the chain search */ From 00a241f528427b63c415a410293b86e66098888e Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Wed, 16 Apr 2025 18:09:50 -0700 Subject: [PATCH 280/555] x86: disable image size check for test builds 64-bit allyesconfig builds fail with x86_64-linux-ld: kernel image bigger than KERNEL_IMAGE_SIZE Bisect points to commit 6f110a5e4f99 ("Disable SLUB_TINY for build testing") as the responsible commit. Reverting that patch does indeed fix the problem. Further analysis shows that disabling SLUB_TINY enables KASAN, and that KASAN is responsible for the image size increase. Solve the build problem by disabling the image size check for test builds. [akpm@linux-foundation.org: add comment, fix nearby typo (sink->sync)] [akpm@linux-foundation.org: fix comment snafu Link: https://lore.kernel.org/oe-kbuild-all/202504191813.4r9H6Glt-lkp@intel.com/ Link: https://lkml.kernel.org/r/20250417010950.2203847-1-linux@roeck-us.net Fixes: 6f110a5e4f99 ("Disable SLUB_TINY for build testing") Signed-off-by: Guenter Roeck Cc: Linus Torvalds Cc: Alexander Potapenko Cc: Andrey Konovalov Cc: Andrey Ryabinin Cc: Borislav Betkov Cc: Dmitriy Vyukov Cc: "H. Peter Anvin" Cc: Ingo Molnar Cc: Thomas Gleinxer Cc: Vincenzo Frascino Cc: Signed-off-by: Andrew Morton --- arch/x86/kernel/vmlinux.lds.S | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/arch/x86/kernel/vmlinux.lds.S b/arch/x86/kernel/vmlinux.lds.S index ccdc45e5b759..aa4d0221583c 100644 --- a/arch/x86/kernel/vmlinux.lds.S +++ b/arch/x86/kernel/vmlinux.lds.S @@ -466,10 +466,18 @@ SECTIONS } /* - * The ASSERT() sink to . is intentional, for binutils 2.14 compatibility: + * COMPILE_TEST kernels can be large - CONFIG_KASAN, for example, can cause + * this. Let's assume that nobody will be running a COMPILE_TEST kernel and + * let's assert that fuller build coverage is more valuable than being able to + * run a COMPILE_TEST kernel. + */ +#ifndef CONFIG_COMPILE_TEST +/* + * The ASSERT() sync to . is intentional, for binutils 2.14 compatibility: */ . = ASSERT((_end - LOAD_OFFSET <= KERNEL_IMAGE_SIZE), "kernel image bigger than KERNEL_IMAGE_SIZE"); +#endif /* needed for Clang - see arch/x86/entry/entry.S */ PROVIDE(__ref_stack_chk_guard = __stack_chk_guard); From e81224f0ba22e13038381d41a8356ab256d56f23 Mon Sep 17 00:00:00 2001 From: Lorenzo Stoakes Date: Fri, 18 Apr 2025 16:00:52 +0100 Subject: [PATCH 281/555] MAINTAINERS: add reverse mapping section Separate out the reverse mapping part of memory management and assign appropriate maintainers and reviewers. David has long been invovled in work with the reverse mapping and continues to do so, so is well suited to maintain this area of the kernel. I have a lot of experience working with the anonymous reverse mapping and continue to work in this area, and also have good knowledge of the walking code and code related to VMAs. This helps people identify who to ask for help, and also additionally makes life easier in review. Link: https://lkml.kernel.org/r/20250418150052.299220-1-lorenzo.stoakes@oracle.com Signed-off-by: Lorenzo Stoakes Acked-by: David Hildenbrand Acked-by: Liam R. Howlett Acked-by: Harry Yoo Acked-by: Vlastimil Babka Signed-off-by: Andrew Morton --- MAINTAINERS | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 69511c3b2b76..c90842b4239e 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -15505,7 +15505,6 @@ F: include/linux/mm_*.h F: include/linux/mmzone.h F: include/linux/mmu_notifier.h F: include/linux/pagewalk.h -F: include/linux/rmap.h F: include/trace/events/ksm.h F: mm/ F: tools/mm/ @@ -15545,6 +15544,19 @@ F: mm/page_alloc.c F: include/linux/gfp.h F: include/linux/compaction.h +MEMORY MANAGEMENT - RMAP (REVERSE MAPPING) +M: Andrew Morton +M: David Hildenbrand +M: Lorenzo Stoakes +R: Rik van Riel +R: Liam R. Howlett +R: Vlastimil Babka +R: Harry Yoo +L: linux-mm@kvack.org +S: Maintained +F: include/linux/rmap.h +F: mm/rmap.c + MEMORY MANAGEMENT - SECRETMEM M: Andrew Morton M: Mike Rapoport From be6e843fc51a584672dfd9c4a6a24c8cb81d5fb7 Mon Sep 17 00:00:00 2001 From: Gavin Guo Date: Mon, 21 Apr 2025 19:35:36 +0800 Subject: [PATCH 282/555] mm/huge_memory: fix dereferencing invalid pmd migration entry When migrating a THP, concurrent access to the PMD migration entry during a deferred split scan can lead to an invalid address access, as illustrated below. To prevent this invalid access, it is necessary to check the PMD migration entry and return early. In this context, there is no need to use pmd_to_swp_entry and pfn_swap_entry_to_page to verify the equality of the target folio. Since the PMD migration entry is locked, it cannot be served as the target. Mailing list discussion and explanation from Hugh Dickins: "An anon_vma lookup points to a location which may contain the folio of interest, but might instead contain another folio: and weeding out those other folios is precisely what the "folio != pmd_folio((*pmd)" check (and the "risk of replacing the wrong folio" comment a few lines above it) is for." BUG: unable to handle page fault for address: ffffea60001db008 CPU: 0 UID: 0 PID: 2199114 Comm: tee Not tainted 6.14.0+ #4 NONE Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2 04/01/2014 RIP: 0010:split_huge_pmd_locked+0x3b5/0x2b60 Call Trace: try_to_migrate_one+0x28c/0x3730 rmap_walk_anon+0x4f6/0x770 unmap_folio+0x196/0x1f0 split_huge_page_to_list_to_order+0x9f6/0x1560 deferred_split_scan+0xac5/0x12a0 shrinker_debugfs_scan_write+0x376/0x470 full_proxy_write+0x15c/0x220 vfs_write+0x2fc/0xcb0 ksys_write+0x146/0x250 do_syscall_64+0x6a/0x120 entry_SYSCALL_64_after_hwframe+0x76/0x7e The bug is found by syzkaller on an internal kernel, then confirmed on upstream. Link: https://lkml.kernel.org/r/20250421113536.3682201-1-gavinguo@igalia.com Link: https://lore.kernel.org/all/20250414072737.1698513-1-gavinguo@igalia.com/ Link: https://lore.kernel.org/all/20250418085802.2973519-1-gavinguo@igalia.com/ Fixes: 84c3fc4e9c56 ("mm: thp: check pmd migration entry in common path") Signed-off-by: Gavin Guo Acked-by: David Hildenbrand Acked-by: Hugh Dickins Acked-by: Zi Yan Reviewed-by: Gavin Shan Cc: Florent Revest Cc: Matthew Wilcox (Oracle) Cc: Miaohe Lin Cc: Signed-off-by: Andrew Morton --- mm/huge_memory.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 2a47682d1ab7..47d76d03ce30 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -3075,6 +3075,8 @@ static void __split_huge_pmd_locked(struct vm_area_struct *vma, pmd_t *pmd, void split_huge_pmd_locked(struct vm_area_struct *vma, unsigned long address, pmd_t *pmd, bool freeze, struct folio *folio) { + bool pmd_migration = is_pmd_migration_entry(*pmd); + VM_WARN_ON_ONCE(folio && !folio_test_pmd_mappable(folio)); VM_WARN_ON_ONCE(!IS_ALIGNED(address, HPAGE_PMD_SIZE)); VM_WARN_ON_ONCE(folio && !folio_test_locked(folio)); @@ -3085,9 +3087,12 @@ void split_huge_pmd_locked(struct vm_area_struct *vma, unsigned long address, * require a folio to check the PMD against. Otherwise, there * is a risk of replacing the wrong folio. */ - if (pmd_trans_huge(*pmd) || pmd_devmap(*pmd) || - is_pmd_migration_entry(*pmd)) { - if (folio && folio != pmd_folio(*pmd)) + if (pmd_trans_huge(*pmd) || pmd_devmap(*pmd) || pmd_migration) { + /* + * Do not apply pmd_folio() to a migration entry; and folio lock + * guarantees that it must be of the wrong folio anyway. + */ + if (folio && (pmd_migration || folio != pmd_folio(*pmd))) return; __split_huge_pmd_locked(vma, pmd, address, freeze); } From 31d4cd4eb2f8d9b87ebfa6a5e443a59e3b3d7b8c Mon Sep 17 00:00:00 2001 From: Mark Tinguely Date: Fri, 11 Apr 2025 11:31:24 -0500 Subject: [PATCH 283/555] ocfs2: fix panic in failed foilio allocation commit 7e119cff9d0a ("ocfs2: convert w_pages to w_folios") and commit 9a5e08652dc4b ("ocfs2: use an array of folios instead of an array of pages") save -ENOMEM in the folio array upon allocation failure and call the folio array free code. The folio array free code expects either valid folio pointers or NULL. Finding the -ENOMEM will result in a panic. Fix by NULLing the error folio entry. Link: https://lkml.kernel.org/r/c879a52b-835c-4fa0-902b-8b2e9196dcbd@oracle.com Fixes: 7e119cff9d0a ("ocfs2: convert w_pages to w_folios") Fixes: 9a5e08652dc4b ("ocfs2: use an array of folios instead of an array of pages") Signed-off-by: Mark Tinguely Reviewed-by: Matthew Wilcox (Oracle) Cc: Changwei Ge Cc: Joel Becker Cc: Junxiao Bi Cc: Mark Fasheh Cc: Nathan Chancellor Cc: Signed-off-by: Andrew Morton --- fs/ocfs2/alloc.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c index b8ac85b548c7..821cb7874685 100644 --- a/fs/ocfs2/alloc.c +++ b/fs/ocfs2/alloc.c @@ -6918,6 +6918,7 @@ static int ocfs2_grab_folios(struct inode *inode, loff_t start, loff_t end, if (IS_ERR(folios[numfolios])) { ret = PTR_ERR(folios[numfolios]); mlog_errno(ret); + folios[numfolios] = NULL; goto out; } From a47694ecb8bcf2d60d665eaade913f4f59956c44 Mon Sep 17 00:00:00 2001 From: Lorenzo Stoakes Date: Wed, 23 Apr 2025 13:30:42 +0100 Subject: [PATCH 284/555] MAINTAINERS: add core mm section In furtherance of ongoing efforts to ensure people are aware of who de-facto maintains/has an interest in specific parts of mm, as well trying to avoid get_maintainers.pl listing only Andrew and the mailing list for mm files - establish a 'core' memory management section establishing David as co-maintainer alongside Andrew (thanks David for volunteering!) along with a number of relevant reviewers. We try to keep things as fine-grained as possible, so we place only obviously 'general' mm things here. For files which are specific to a particular part of mm, we prefer new entries. Link: https://lkml.kernel.org/r/20250423123042.59082-1-lorenzo.stoakes@oracle.com Signed-off-by: Lorenzo Stoakes Acked-by: Michal Hocko Acked-by: Vlastimil Babka Acked-by: Mike Rapoport (Microsoft) Acked-by: Liam R. Howlett Acked-by: David Hildenbrand Cc: Lorenzo Stoakes Cc: Suren Baghdasaryan Signed-off-by: Andrew Morton --- MAINTAINERS | 28 +++++++++++++++++++++++++--- 1 file changed, 25 insertions(+), 3 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index c90842b4239e..5b5d07abdf32 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -15494,14 +15494,11 @@ F: Documentation/mm/ F: include/linux/gfp.h F: include/linux/gfp_types.h F: include/linux/memfd.h -F: include/linux/memory.h F: include/linux/memory_hotplug.h F: include/linux/memory-tiers.h F: include/linux/mempolicy.h F: include/linux/mempool.h F: include/linux/memremap.h -F: include/linux/mm.h -F: include/linux/mm_*.h F: include/linux/mmzone.h F: include/linux/mmu_notifier.h F: include/linux/pagewalk.h @@ -15511,6 +15508,31 @@ F: tools/mm/ F: tools/testing/selftests/mm/ N: include/linux/page[-_]* +MEMORY MANAGEMENT - CORE +M: Andrew Morton +M: David Hildenbrand +R: Lorenzo Stoakes +R: Liam R. Howlett +R: Vlastimil Babka +R: Mike Rapoport +R: Suren Baghdasaryan +R: Michal Hocko +L: linux-mm@kvack.org +S: Maintained +W: http://www.linux-mm.org +T: git git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm +F: include/linux/memory.h +F: include/linux/mm.h +F: include/linux/mm_*.h +F: include/linux/mmdebug.h +F: include/linux/pagewalk.h +F: mm/Kconfig +F: mm/debug.c +F: mm/init-mm.c +F: mm/memory.c +F: mm/pagewalk.c +F: mm/util.c + MEMORY MANAGEMENT - EXECMEM M: Andrew Morton M: Mike Rapoport From ab00ddd802f80e31fc9639c652d736fe3913feae Mon Sep 17 00:00:00 2001 From: Feng Tang Date: Wed, 23 Apr 2025 18:36:45 +0800 Subject: [PATCH 285/555] selftests/mm: compaction_test: support platform with huge mount of memory When running mm selftest to verify mm patches, 'compaction_test' case failed on an x86 server with 1TB memory. And the root cause is that it has too much free memory than what the test supports. The test case tries to allocate 100000 huge pages, which is about 200 GB for that x86 server, and when it succeeds, it expects it's large than 1/3 of 80% of the free memory in system. This logic only works for platform with 750 GB ( 200 / (1/3) / 80% ) or less free memory, and may raise false alarm for others. Fix it by changing the fixed page number to self-adjustable number according to the real number of free memory. Link: https://lkml.kernel.org/r/20250423103645.2758-1-feng.tang@linux.alibaba.com Fixes: bd67d5c15cc1 ("Test compaction of mlocked memory") Signed-off-by: Feng Tang Acked-by: Dev Jain Reviewed-by: Baolin Wang Tested-by: Baolin Wang Cc: Shuah Khan Cc: Sri Jayaramappa Cc: Signed-off-by: Andrew Morton --- tools/testing/selftests/mm/compaction_test.c | 19 ++++++++++++++----- 1 file changed, 14 insertions(+), 5 deletions(-) diff --git a/tools/testing/selftests/mm/compaction_test.c b/tools/testing/selftests/mm/compaction_test.c index 2c3a0eb6b22d..9bc4591c7b16 100644 --- a/tools/testing/selftests/mm/compaction_test.c +++ b/tools/testing/selftests/mm/compaction_test.c @@ -90,6 +90,8 @@ int check_compaction(unsigned long mem_free, unsigned long hugepage_size, int compaction_index = 0; char nr_hugepages[20] = {0}; char init_nr_hugepages[24] = {0}; + char target_nr_hugepages[24] = {0}; + int slen; snprintf(init_nr_hugepages, sizeof(init_nr_hugepages), "%lu", initial_nr_hugepages); @@ -106,11 +108,18 @@ int check_compaction(unsigned long mem_free, unsigned long hugepage_size, goto out; } - /* Request a large number of huge pages. The Kernel will allocate - as much as it can */ - if (write(fd, "100000", (6*sizeof(char))) != (6*sizeof(char))) { - ksft_print_msg("Failed to write 100000 to /proc/sys/vm/nr_hugepages: %s\n", - strerror(errno)); + /* + * Request huge pages for about half of the free memory. The Kernel + * will allocate as much as it can, and we expect it will get at least 1/3 + */ + nr_hugepages_ul = mem_free / hugepage_size / 2; + snprintf(target_nr_hugepages, sizeof(target_nr_hugepages), + "%lu", nr_hugepages_ul); + + slen = strlen(target_nr_hugepages); + if (write(fd, target_nr_hugepages, slen) != slen) { + ksft_print_msg("Failed to write %lu to /proc/sys/vm/nr_hugepages: %s\n", + nr_hugepages_ul, strerror(errno)); goto close_fd; } From 95567729173e62e0e60a1f8ad9eb2e1320a8ccac Mon Sep 17 00:00:00 2001 From: Peter Xu Date: Thu, 24 Apr 2025 17:57:28 -0400 Subject: [PATCH 286/555] mm/userfaultfd: fix uninitialized output field for -EAGAIN race While discussing some userfaultfd relevant issues recently, Andrea noticed a potential ABI breakage with -EAGAIN on almost all userfaultfd ioctl()s. Quote from Andrea, explaining how -EAGAIN was processed, and how this should fix it (taking example of UFFDIO_COPY ioctl): The "mmap_changing" and "stale pmd" conditions are already reported as -EAGAIN written in the copy field, this does not change it. This change removes the subnormal case that left copy.copy uninitialized and required apps to explicitly set the copy field to get deterministic behavior (which is a requirement contrary to the documentation in both the manpage and source code). In turn there's no alteration to backwards compatibility as result of this change because userland will find the copy field consistently set to -EAGAIN, and not anymore sometime -EAGAIN and sometime uninitialized. Even then the change only can make a difference to non cooperative users of userfaultfd, so when UFFD_FEATURE_EVENT_* is enabled, which is not true for the vast majority of apps using userfaultfd or this unintended uninitialized field may have been noticed sooner. Meanwhile, since this bug existed for years, it also almost affects all ioctl()s that was introduced later. Besides UFFDIO_ZEROPAGE, these also get affected in the same way: - UFFDIO_CONTINUE - UFFDIO_POISON - UFFDIO_MOVE This patch should have fixed all of them. Link: https://lkml.kernel.org/r/20250424215729.194656-2-peterx@redhat.com Fixes: df2cc96e7701 ("userfaultfd: prevent non-cooperative events vs mcopy_atomic races") Fixes: f619147104c8 ("userfaultfd: add UFFDIO_CONTINUE ioctl") Fixes: fc71884a5f59 ("mm: userfaultfd: add new UFFDIO_POISON ioctl") Fixes: adef440691ba ("userfaultfd: UFFDIO_MOVE uABI") Signed-off-by: Peter Xu Reported-by: Andrea Arcangeli Suggested-by: Andrea Arcangeli Reviewed-by: David Hildenbrand Cc: Mike Rapoport Cc: Axel Rasmussen Cc: Suren Baghdasaryan Cc: Signed-off-by: Andrew Morton --- fs/userfaultfd.c | 28 ++++++++++++++++++++++------ 1 file changed, 22 insertions(+), 6 deletions(-) diff --git a/fs/userfaultfd.c b/fs/userfaultfd.c index d80f94346199..22f4bf956ba1 100644 --- a/fs/userfaultfd.c +++ b/fs/userfaultfd.c @@ -1585,8 +1585,11 @@ static int userfaultfd_copy(struct userfaultfd_ctx *ctx, user_uffdio_copy = (struct uffdio_copy __user *) arg; ret = -EAGAIN; - if (atomic_read(&ctx->mmap_changing)) + if (unlikely(atomic_read(&ctx->mmap_changing))) { + if (unlikely(put_user(ret, &user_uffdio_copy->copy))) + return -EFAULT; goto out; + } ret = -EFAULT; if (copy_from_user(&uffdio_copy, user_uffdio_copy, @@ -1641,8 +1644,11 @@ static int userfaultfd_zeropage(struct userfaultfd_ctx *ctx, user_uffdio_zeropage = (struct uffdio_zeropage __user *) arg; ret = -EAGAIN; - if (atomic_read(&ctx->mmap_changing)) + if (unlikely(atomic_read(&ctx->mmap_changing))) { + if (unlikely(put_user(ret, &user_uffdio_zeropage->zeropage))) + return -EFAULT; goto out; + } ret = -EFAULT; if (copy_from_user(&uffdio_zeropage, user_uffdio_zeropage, @@ -1744,8 +1750,11 @@ static int userfaultfd_continue(struct userfaultfd_ctx *ctx, unsigned long arg) user_uffdio_continue = (struct uffdio_continue __user *)arg; ret = -EAGAIN; - if (atomic_read(&ctx->mmap_changing)) + if (unlikely(atomic_read(&ctx->mmap_changing))) { + if (unlikely(put_user(ret, &user_uffdio_continue->mapped))) + return -EFAULT; goto out; + } ret = -EFAULT; if (copy_from_user(&uffdio_continue, user_uffdio_continue, @@ -1801,8 +1810,11 @@ static inline int userfaultfd_poison(struct userfaultfd_ctx *ctx, unsigned long user_uffdio_poison = (struct uffdio_poison __user *)arg; ret = -EAGAIN; - if (atomic_read(&ctx->mmap_changing)) + if (unlikely(atomic_read(&ctx->mmap_changing))) { + if (unlikely(put_user(ret, &user_uffdio_poison->updated))) + return -EFAULT; goto out; + } ret = -EFAULT; if (copy_from_user(&uffdio_poison, user_uffdio_poison, @@ -1870,8 +1882,12 @@ static int userfaultfd_move(struct userfaultfd_ctx *ctx, user_uffdio_move = (struct uffdio_move __user *) arg; - if (atomic_read(&ctx->mmap_changing)) - return -EAGAIN; + ret = -EAGAIN; + if (unlikely(atomic_read(&ctx->mmap_changing))) { + if (unlikely(put_user(ret, &user_uffdio_move->move))) + return -EFAULT; + goto out; + } if (copy_from_user(&uffdio_move, user_uffdio_move, /* don't copy "move" last field */ From 09fc97b3abe9e72c799a1c757acc47e746844619 Mon Sep 17 00:00:00 2001 From: Lorenzo Stoakes Date: Thu, 24 Apr 2025 12:16:32 +0100 Subject: [PATCH 287/555] MAINTAINERS: add mm THP section As part of the ongoing efforts to sub-divide memory management maintainership and reviewership, establish a section for Transparent Huge Page support and add appropriate maintainers and reviewers. [lorenzo.stoakes@oracle.com: add Dev Jain as THP reviewer] Link: https://lkml.kernel.org/r/327e6f2f-0f0f-48af-9ca2-3f8cadf0d8bf@lucifer.local Link: https://lkml.kernel.org/r/20250424111632.103637-1-lorenzo.stoakes@oracle.com Signed-off-by: Lorenzo Stoakes Acked-by: David Hildenbrand Reviewed-by: Ryan Roberts Reviewed-by: Baolin Wang Acked-by: Zi Yan Cc: Dev Jain Cc: Liam Howlett Cc: Mariano Pache Cc: Matthew Wilcox Signed-off-by: Andrew Morton --- MAINTAINERS | 24 ++++++++++++++++++++++++ 1 file changed, 24 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index 5b5d07abdf32..f62b266630cf 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -15587,6 +15587,30 @@ S: Maintained F: include/linux/secretmem.h F: mm/secretmem.c +MEMORY MANAGEMENT - THP (TRANSPARENT HUGE PAGE) +M: Andrew Morton +M: David Hildenbrand +R: Zi Yan +R: Baolin Wang +R: Lorenzo Stoakes +R: Liam R. Howlett +R: Nico Pache +R: Ryan Roberts +R: Dev Jain +L: linux-mm@kvack.org +S: Maintained +W: http://www.linux-mm.org +T: git git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm +F: Documentation/admin-guide/mm/transhuge.rst +F: include/linux/huge_mm.h +F: include/linux/khugepaged.h +F: include/trace/events/huge_memory.h +F: mm/huge_memory.c +F: mm/khugepaged.c +F: tools/testing/selftests/mm/khugepaged.c +F: tools/testing/selftests/mm/split_huge_page_test.c +F: tools/testing/selftests/mm/transhuge-stress.c + MEMORY MANAGEMENT - USERFAULTFD M: Andrew Morton R: Peter Xu From 80fbee76ebbdf92acd9f293178850bf0b103a6b6 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= Date: Thu, 24 Apr 2025 08:02:51 +0200 Subject: [PATCH 288/555] mailmap: map Uwe's BayLibre addresses to a single one MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When I started working for BayLibre I wasn't aware that the mailserver rewrote the sender address and so a few commits entered kernel history with a working but unexpected address. Map the unexpected to the intended one. This also makes the author of those commits (e.g. 32b4f1a4f07f ("pwm: jz4740: Another few conversions to regmap_{set,clear}_bits()")) match the address used in the sign-off line. Link: https://lkml.kernel.org/r/20250424060250.3085683-2-u.kleine-koenig@baylibre.com Signed-off-by: Uwe Kleine-König Signed-off-by: Andrew Morton --- .mailmap | 1 + 1 file changed, 1 insertion(+) diff --git a/.mailmap b/.mailmap index 6f0a4e965d6a..4f661e434eaa 100644 --- a/.mailmap +++ b/.mailmap @@ -751,6 +751,7 @@ Tvrtko Ursulin Tycho Andersen Tzung-Bi Shih Uwe Kleine-König +Uwe Kleine-König Uwe Kleine-König Uwe Kleine-König Uwe Kleine-König From c0fb83088f0cc4ee4706e0495ee8b06f49daa716 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Thu, 24 Apr 2025 15:45:11 +0200 Subject: [PATCH 289/555] ocfs2: switch osb->disable_recovery to enum Patch series "ocfs2: Fix deadlocks in quota recovery", v3. This implements another approach to fixing quota recovery deadlocks. We avoid grabbing sb->s_umount semaphore from ocfs2_finish_quota_recovery() and instead stop quota recovery early in ocfs2_dismount_volume(). This patch (of 3): We will need more recovery states than just pure enable / disable to fix deadlocks with quota recovery. Switch osb->disable_recovery to enum. Link: https://lkml.kernel.org/r/20250424134301.1392-1-jack@suse.cz Link: https://lkml.kernel.org/r/20250424134515.18933-4-jack@suse.cz Fixes: 5f530de63cfc ("ocfs2: Use s_umount for quota recovery protection") Signed-off-by: Jan Kara Reviewed-by: Heming Zhao Tested-by: Heming Zhao Acked-by: Joseph Qi Cc: Mark Fasheh Cc: Joel Becker Cc: Junxiao Bi Cc: Changwei Ge Cc: Jun Piao Cc: Murad Masimov Cc: Shichangkuo Cc: Signed-off-by: Andrew Morton --- fs/ocfs2/journal.c | 14 ++++++++------ fs/ocfs2/ocfs2.h | 7 ++++++- 2 files changed, 14 insertions(+), 7 deletions(-) diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c index c7a9729dc9d0..e79f10cc43d9 100644 --- a/fs/ocfs2/journal.c +++ b/fs/ocfs2/journal.c @@ -174,7 +174,7 @@ int ocfs2_recovery_init(struct ocfs2_super *osb) struct ocfs2_recovery_map *rm; mutex_init(&osb->recovery_lock); - osb->disable_recovery = 0; + osb->recovery_state = OCFS2_REC_ENABLED; osb->recovery_thread_task = NULL; init_waitqueue_head(&osb->recovery_event); @@ -206,7 +206,7 @@ void ocfs2_recovery_exit(struct ocfs2_super *osb) /* disable any new recovery threads and wait for any currently * running ones to exit. Do this before setting the vol_state. */ mutex_lock(&osb->recovery_lock); - osb->disable_recovery = 1; + osb->recovery_state = OCFS2_REC_DISABLED; mutex_unlock(&osb->recovery_lock); wait_event(osb->recovery_event, !ocfs2_recovery_thread_running(osb)); @@ -1582,14 +1582,16 @@ bail: void ocfs2_recovery_thread(struct ocfs2_super *osb, int node_num) { + int was_set = -1; + mutex_lock(&osb->recovery_lock); + if (osb->recovery_state < OCFS2_REC_DISABLED) + was_set = ocfs2_recovery_map_set(osb, node_num); trace_ocfs2_recovery_thread(node_num, osb->node_num, - osb->disable_recovery, osb->recovery_thread_task, - osb->disable_recovery ? - -1 : ocfs2_recovery_map_set(osb, node_num)); + osb->recovery_state, osb->recovery_thread_task, was_set); - if (osb->disable_recovery) + if (osb->recovery_state == OCFS2_REC_DISABLED) goto out; if (osb->recovery_thread_task) diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h index 51c52768132d..e713361939f0 100644 --- a/fs/ocfs2/ocfs2.h +++ b/fs/ocfs2/ocfs2.h @@ -308,6 +308,11 @@ enum ocfs2_journal_trigger_type { void ocfs2_initialize_journal_triggers(struct super_block *sb, struct ocfs2_triggers triggers[]); +enum ocfs2_recovery_state { + OCFS2_REC_ENABLED = 0, + OCFS2_REC_DISABLED, +}; + struct ocfs2_journal; struct ocfs2_slot_info; struct ocfs2_recovery_map; @@ -370,7 +375,7 @@ struct ocfs2_super struct ocfs2_recovery_map *recovery_map; struct ocfs2_replay_map *replay_map; struct task_struct *recovery_thread_task; - int disable_recovery; + enum ocfs2_recovery_state recovery_state; wait_queue_head_t checkpoint_event; struct ocfs2_journal *journal; unsigned long osb_commit_interval; From 8f947e0fd595951460f5a6e1ac29baa82fa02eab Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Thu, 24 Apr 2025 15:45:12 +0200 Subject: [PATCH 290/555] ocfs2: implement handshaking with ocfs2 recovery thread We will need ocfs2 recovery thread to acknowledge transitions of recovery_state when disabling particular types of recovery. This is similar to what currently happens when disabling recovery completely, just more general. Implement the handshake and use it for exit from recovery. Link: https://lkml.kernel.org/r/20250424134515.18933-5-jack@suse.cz Fixes: 5f530de63cfc ("ocfs2: Use s_umount for quota recovery protection") Signed-off-by: Jan Kara Reviewed-by: Heming Zhao Tested-by: Heming Zhao Acked-by: Joseph Qi Cc: Changwei Ge Cc: Joel Becker Cc: Jun Piao Cc: Junxiao Bi Cc: Mark Fasheh Cc: Murad Masimov Cc: Shichangkuo Cc: Signed-off-by: Andrew Morton --- fs/ocfs2/journal.c | 52 +++++++++++++++++++++++++++++++--------------- fs/ocfs2/ocfs2.h | 4 ++++ 2 files changed, 39 insertions(+), 17 deletions(-) diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c index e79f10cc43d9..2987df396ad8 100644 --- a/fs/ocfs2/journal.c +++ b/fs/ocfs2/journal.c @@ -190,31 +190,48 @@ int ocfs2_recovery_init(struct ocfs2_super *osb) return 0; } -/* we can't grab the goofy sem lock from inside wait_event, so we use - * memory barriers to make sure that we'll see the null task before - * being woken up */ static int ocfs2_recovery_thread_running(struct ocfs2_super *osb) { - mb(); return osb->recovery_thread_task != NULL; } +static void ocfs2_recovery_disable(struct ocfs2_super *osb, + enum ocfs2_recovery_state state) +{ + mutex_lock(&osb->recovery_lock); + /* + * If recovery thread is not running, we can directly transition to + * final state. + */ + if (!ocfs2_recovery_thread_running(osb)) { + osb->recovery_state = state + 1; + goto out_lock; + } + osb->recovery_state = state; + /* Wait for recovery thread to acknowledge state transition */ + wait_event_cmd(osb->recovery_event, + !ocfs2_recovery_thread_running(osb) || + osb->recovery_state >= state + 1, + mutex_unlock(&osb->recovery_lock), + mutex_lock(&osb->recovery_lock)); +out_lock: + mutex_unlock(&osb->recovery_lock); + + /* + * At this point we know that no more recovery work can be queued so + * wait for any recovery completion work to complete. + */ + if (osb->ocfs2_wq) + flush_workqueue(osb->ocfs2_wq); +} + void ocfs2_recovery_exit(struct ocfs2_super *osb) { struct ocfs2_recovery_map *rm; /* disable any new recovery threads and wait for any currently * running ones to exit. Do this before setting the vol_state. */ - mutex_lock(&osb->recovery_lock); - osb->recovery_state = OCFS2_REC_DISABLED; - mutex_unlock(&osb->recovery_lock); - wait_event(osb->recovery_event, !ocfs2_recovery_thread_running(osb)); - - /* At this point, we know that no more recovery threads can be - * launched, so wait for any recovery completion work to - * complete. */ - if (osb->ocfs2_wq) - flush_workqueue(osb->ocfs2_wq); + ocfs2_recovery_disable(osb, OCFS2_REC_WANT_DISABLE); /* * Now that recovery is shut down, and the osb is about to be @@ -1569,7 +1586,8 @@ bail: ocfs2_free_replay_slots(osb); osb->recovery_thread_task = NULL; - mb(); /* sync with ocfs2_recovery_thread_running */ + if (osb->recovery_state == OCFS2_REC_WANT_DISABLE) + osb->recovery_state = OCFS2_REC_DISABLED; wake_up(&osb->recovery_event); mutex_unlock(&osb->recovery_lock); @@ -1585,13 +1603,13 @@ void ocfs2_recovery_thread(struct ocfs2_super *osb, int node_num) int was_set = -1; mutex_lock(&osb->recovery_lock); - if (osb->recovery_state < OCFS2_REC_DISABLED) + if (osb->recovery_state < OCFS2_REC_WANT_DISABLE) was_set = ocfs2_recovery_map_set(osb, node_num); trace_ocfs2_recovery_thread(node_num, osb->node_num, osb->recovery_state, osb->recovery_thread_task, was_set); - if (osb->recovery_state == OCFS2_REC_DISABLED) + if (osb->recovery_state >= OCFS2_REC_WANT_DISABLE) goto out; if (osb->recovery_thread_task) diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h index e713361939f0..70b2d5c8c228 100644 --- a/fs/ocfs2/ocfs2.h +++ b/fs/ocfs2/ocfs2.h @@ -310,6 +310,10 @@ void ocfs2_initialize_journal_triggers(struct super_block *sb, enum ocfs2_recovery_state { OCFS2_REC_ENABLED = 0, + OCFS2_REC_WANT_DISABLE, + /* + * Must be OCFS2_REC_WANT_DISABLE + 1 for ocfs2_recovery_exit() to work + */ OCFS2_REC_DISABLED, }; From fcaf3b2683b05a9684acdebda706a12025a6927a Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Thu, 24 Apr 2025 15:45:13 +0200 Subject: [PATCH 291/555] ocfs2: stop quota recovery before disabling quotas Currently quota recovery is synchronized with unmount using sb->s_umount semaphore. That is however prone to deadlocks because flush_workqueue(osb->ocfs2_wq) called from umount code can wait for quota recovery to complete while ocfs2_finish_quota_recovery() waits for sb->s_umount semaphore. Grabbing of sb->s_umount semaphore in ocfs2_finish_quota_recovery() is only needed to protect that function from disabling of quotas from ocfs2_dismount_volume(). Handle this problem by disabling quota recovery early during unmount in ocfs2_dismount_volume() instead so that we can drop acquisition of sb->s_umount from ocfs2_finish_quota_recovery(). Link: https://lkml.kernel.org/r/20250424134515.18933-6-jack@suse.cz Fixes: 5f530de63cfc ("ocfs2: Use s_umount for quota recovery protection") Signed-off-by: Jan Kara Reported-by: Shichangkuo Reported-by: Murad Masimov Reviewed-by: Heming Zhao Tested-by: Heming Zhao Acked-by: Joseph Qi Cc: Changwei Ge Cc: Joel Becker Cc: Jun Piao Cc: Junxiao Bi Cc: Mark Fasheh Cc: Signed-off-by: Andrew Morton --- fs/ocfs2/journal.c | 20 ++++++++++++++++++-- fs/ocfs2/journal.h | 1 + fs/ocfs2/ocfs2.h | 6 ++++++ fs/ocfs2/quota_local.c | 9 ++------- fs/ocfs2/super.c | 3 +++ 5 files changed, 30 insertions(+), 9 deletions(-) diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c index 2987df396ad8..e5f58ff2175f 100644 --- a/fs/ocfs2/journal.c +++ b/fs/ocfs2/journal.c @@ -225,6 +225,11 @@ out_lock: flush_workqueue(osb->ocfs2_wq); } +void ocfs2_recovery_disable_quota(struct ocfs2_super *osb) +{ + ocfs2_recovery_disable(osb, OCFS2_REC_QUOTA_WANT_DISABLE); +} + void ocfs2_recovery_exit(struct ocfs2_super *osb) { struct ocfs2_recovery_map *rm; @@ -1489,6 +1494,18 @@ static int __ocfs2_recovery_thread(void *arg) } } restart: + if (quota_enabled) { + mutex_lock(&osb->recovery_lock); + /* Confirm that recovery thread will no longer recover quotas */ + if (osb->recovery_state == OCFS2_REC_QUOTA_WANT_DISABLE) { + osb->recovery_state = OCFS2_REC_QUOTA_DISABLED; + wake_up(&osb->recovery_event); + } + if (osb->recovery_state >= OCFS2_REC_QUOTA_DISABLED) + quota_enabled = 0; + mutex_unlock(&osb->recovery_lock); + } + status = ocfs2_super_lock(osb, 1); if (status < 0) { mlog_errno(status); @@ -1592,8 +1609,7 @@ bail: mutex_unlock(&osb->recovery_lock); - if (quota_enabled) - kfree(rm_quota); + kfree(rm_quota); return status; } diff --git a/fs/ocfs2/journal.h b/fs/ocfs2/journal.h index e3c3a35dc5e0..6397170f302f 100644 --- a/fs/ocfs2/journal.h +++ b/fs/ocfs2/journal.h @@ -148,6 +148,7 @@ void ocfs2_wait_for_recovery(struct ocfs2_super *osb); int ocfs2_recovery_init(struct ocfs2_super *osb); void ocfs2_recovery_exit(struct ocfs2_super *osb); +void ocfs2_recovery_disable_quota(struct ocfs2_super *osb); int ocfs2_compute_replay_slots(struct ocfs2_super *osb); void ocfs2_free_replay_slots(struct ocfs2_super *osb); diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h index 70b2d5c8c228..6aaa94c554c1 100644 --- a/fs/ocfs2/ocfs2.h +++ b/fs/ocfs2/ocfs2.h @@ -310,6 +310,12 @@ void ocfs2_initialize_journal_triggers(struct super_block *sb, enum ocfs2_recovery_state { OCFS2_REC_ENABLED = 0, + OCFS2_REC_QUOTA_WANT_DISABLE, + /* + * Must be OCFS2_REC_QUOTA_WANT_DISABLE + 1 for + * ocfs2_recovery_disable_quota() to work. + */ + OCFS2_REC_QUOTA_DISABLED, OCFS2_REC_WANT_DISABLE, /* * Must be OCFS2_REC_WANT_DISABLE + 1 for ocfs2_recovery_exit() to work diff --git a/fs/ocfs2/quota_local.c b/fs/ocfs2/quota_local.c index 2956d888c131..e272429da3db 100644 --- a/fs/ocfs2/quota_local.c +++ b/fs/ocfs2/quota_local.c @@ -453,8 +453,7 @@ out: /* Sync changes in local quota file into global quota file and * reinitialize local quota file. - * The function expects local quota file to be already locked and - * s_umount locked in shared mode. */ + * The function expects local quota file to be already locked. */ static int ocfs2_recover_local_quota_file(struct inode *lqinode, int type, struct ocfs2_quota_recovery *rec) @@ -588,7 +587,6 @@ int ocfs2_finish_quota_recovery(struct ocfs2_super *osb, { unsigned int ino[OCFS2_MAXQUOTAS] = { LOCAL_USER_QUOTA_SYSTEM_INODE, LOCAL_GROUP_QUOTA_SYSTEM_INODE }; - struct super_block *sb = osb->sb; struct ocfs2_local_disk_dqinfo *ldinfo; struct buffer_head *bh; handle_t *handle; @@ -600,7 +598,6 @@ int ocfs2_finish_quota_recovery(struct ocfs2_super *osb, printk(KERN_NOTICE "ocfs2: Finishing quota recovery on device (%s) for " "slot %u\n", osb->dev_str, slot_num); - down_read(&sb->s_umount); for (type = 0; type < OCFS2_MAXQUOTAS; type++) { if (list_empty(&(rec->r_list[type]))) continue; @@ -677,7 +674,6 @@ out_put: break; } out: - up_read(&sb->s_umount); kfree(rec); return status; } @@ -843,8 +839,7 @@ static int ocfs2_local_free_info(struct super_block *sb, int type) ocfs2_release_local_quota_bitmaps(&oinfo->dqi_chunk); /* - * s_umount held in exclusive mode protects us against racing with - * recovery thread... + * ocfs2_dismount_volume() has already aborted quota recovery... */ if (oinfo->dqi_rec) { ocfs2_free_quota_recovery(oinfo->dqi_rec); diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index 8bb5022f3082..3d2533950bae 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -1812,6 +1812,9 @@ static void ocfs2_dismount_volume(struct super_block *sb, int mnt_err) /* Orphan scan should be stopped as early as possible */ ocfs2_orphan_scan_stop(osb); + /* Stop quota recovery so that we can disable quotas */ + ocfs2_recovery_disable_quota(osb); + ocfs2_disable_quotas(osb); /* All dquots should be freed by now */ From a8efadda8649506e80d256cc09656acc0783df2b Mon Sep 17 00:00:00 2001 From: Lorenzo Stoakes Date: Fri, 25 Apr 2025 17:24:36 +0100 Subject: [PATCH 292/555] tools/testing/selftests: fix guard region test tmpfs assumption The current implementation of the guard region tests assume that /tmp is mounted as tmpfs, that is shmem. This isn't always the case, and at least one instance of a spurious test failure has been reported as a result. This assumption is unsafe, rushed and silly - and easily remedied by simply using memfd, so do so. We also have to fixup the readonly_file test to explicitly only be applicable to file-backed cases. Link: https://lkml.kernel.org/r/20250425162436.564002-1-lorenzo.stoakes@oracle.com Fixes: 272f37d3e99a ("tools/selftests: expand all guard region tests to file-backed") Signed-off-by: Lorenzo Stoakes Reported-by: Ryan Roberts Closes: https://lore.kernel.org/linux-mm/a2d2766b-0ab4-437b-951a-8595a7506fe9@arm.com/ Reviewed-by: Ryan Roberts Cc: Signed-off-by: Andrew Morton --- tools/testing/selftests/mm/guard-regions.c | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) diff --git a/tools/testing/selftests/mm/guard-regions.c b/tools/testing/selftests/mm/guard-regions.c index b3d0e2771096..eba43ead13ae 100644 --- a/tools/testing/selftests/mm/guard-regions.c +++ b/tools/testing/selftests/mm/guard-regions.c @@ -271,12 +271,16 @@ FIXTURE_SETUP(guard_regions) self->page_size = (unsigned long)sysconf(_SC_PAGESIZE); setup_sighandler(); - if (variant->backing == ANON_BACKED) + switch (variant->backing) { + case ANON_BACKED: return; - - self->fd = open_file( - variant->backing == SHMEM_BACKED ? "/tmp/" : "", - self->path); + case LOCAL_FILE_BACKED: + self->fd = open_file("", self->path); + break; + case SHMEM_BACKED: + self->fd = memfd_create(self->path, 0); + break; + } /* We truncate file to at least 100 pages, tests can modify as needed. */ ASSERT_EQ(ftruncate(self->fd, 100 * self->page_size), 0); @@ -1696,7 +1700,7 @@ TEST_F(guard_regions, readonly_file) char *ptr; int i; - if (variant->backing == ANON_BACKED) + if (variant->backing != LOCAL_FILE_BACKED) SKIP(return, "Read-only test specific to file-backed"); /* Map shared so we can populate with pattern, populate it, unmap. */ From a0309faf1cb0622cac7c820150b7abf2024acff5 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Fri, 25 Apr 2025 17:11:07 -0700 Subject: [PATCH 293/555] mm: vmalloc: support more granular vrealloc() sizing Introduce struct vm_struct::requested_size so that the requested (re)allocation size is retained separately from the allocated area size. This means that KASAN will correctly poison the correct spans of requested bytes. This also means we can support growing the usable portion of an allocation that can already be supported by the existing area's existing allocation. Link: https://lkml.kernel.org/r/20250426001105.it.679-kees@kernel.org Fixes: 3ddc2fefe6f3 ("mm: vmalloc: implement vrealloc()") Signed-off-by: Kees Cook Reported-by: Erhard Furtner Closes: https://lore.kernel.org/all/20250408192503.6149a816@outsider.home/ Reviewed-by: Danilo Krummrich Cc: Michal Hocko Cc: "Uladzislau Rezki (Sony)" Cc: Vlastimil Babka Cc: Signed-off-by: Andrew Morton --- include/linux/vmalloc.h | 1 + mm/vmalloc.c | 31 ++++++++++++++++++++++++------- 2 files changed, 25 insertions(+), 7 deletions(-) diff --git a/include/linux/vmalloc.h b/include/linux/vmalloc.h index 31e9ffd936e3..5ca8d4dd149d 100644 --- a/include/linux/vmalloc.h +++ b/include/linux/vmalloc.h @@ -61,6 +61,7 @@ struct vm_struct { unsigned int nr_pages; phys_addr_t phys_addr; const void *caller; + unsigned long requested_size; }; struct vmap_area { diff --git a/mm/vmalloc.c b/mm/vmalloc.c index 3ed720a787ec..2d7511654831 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -1940,7 +1940,7 @@ static inline void setup_vmalloc_vm(struct vm_struct *vm, { vm->flags = flags; vm->addr = (void *)va->va_start; - vm->size = va_size(va); + vm->size = vm->requested_size = va_size(va); vm->caller = caller; va->vm = vm; } @@ -3133,6 +3133,7 @@ struct vm_struct *__get_vm_area_node(unsigned long size, area->flags = flags; area->caller = caller; + area->requested_size = requested_size; va = alloc_vmap_area(size, align, start, end, node, gfp_mask, 0, area); if (IS_ERR(va)) { @@ -4063,6 +4064,8 @@ EXPORT_SYMBOL(vzalloc_node_noprof); */ void *vrealloc_noprof(const void *p, size_t size, gfp_t flags) { + struct vm_struct *vm = NULL; + size_t alloced_size = 0; size_t old_size = 0; void *n; @@ -4072,15 +4075,17 @@ void *vrealloc_noprof(const void *p, size_t size, gfp_t flags) } if (p) { - struct vm_struct *vm; - vm = find_vm_area(p); if (unlikely(!vm)) { WARN(1, "Trying to vrealloc() nonexistent vm area (%p)\n", p); return NULL; } - old_size = get_vm_area_size(vm); + alloced_size = get_vm_area_size(vm); + old_size = vm->requested_size; + if (WARN(alloced_size < old_size, + "vrealloc() has mismatched area vs requested sizes (%p)\n", p)) + return NULL; } /* @@ -4088,14 +4093,26 @@ void *vrealloc_noprof(const void *p, size_t size, gfp_t flags) * would be a good heuristic for when to shrink the vm_area? */ if (size <= old_size) { - /* Zero out spare memory. */ - if (want_init_on_alloc(flags)) + /* Zero out "freed" memory. */ + if (want_init_on_free()) memset((void *)p + size, 0, old_size - size); + vm->requested_size = size; kasan_poison_vmalloc(p + size, old_size - size); - kasan_unpoison_vmalloc(p, size, KASAN_VMALLOC_PROT_NORMAL); return (void *)p; } + /* + * We already have the bytes available in the allocation; use them. + */ + if (size <= alloced_size) { + kasan_unpoison_vmalloc(p + old_size, size - old_size, + KASAN_VMALLOC_PROT_NORMAL); + /* Zero out "alloced" memory. */ + if (want_init_on_alloc(flags)) + memset((void *)p + old_size, 0, size - old_size); + vm->requested_size = size; + } + /* TODO: Grow the vm_area, i.e. allocate and map additional pages. */ n = __vmalloc_noprof(size, flags); if (!n) From 22adb528621ddc92f887882a658507fbf88a5214 Mon Sep 17 00:00:00 2001 From: Madhavan Srinivasan Date: Mon, 28 Apr 2025 18:49:34 +0530 Subject: [PATCH 294/555] selftests/mm: fix build break when compiling pkey_util.c Commit 50910acd6f615 ("selftests/mm: use sys_pkey helpers consistently") added a pkey_util.c to refactor some of the protection_keys functions accessible by other tests. But this broken the build in powerpc in two ways, pkey-powerpc.h: In function `arch_is_powervm': pkey-powerpc.h:73:21: error: storage size of `buf' isn't known 73 | struct stat buf; | ^~~ pkey-powerpc.h:75:14: error: implicit declaration of function `stat'; did you mean `strcat'? [-Wimplicit-function-declaration] 75 | if ((stat("/sys/firmware/devicetree/base/ibm,partition-name", &buf) == 0) && | ^~~~ | strcat Since pkey_util.c includes pkeys-helper.h, which in turn includes pkeys-powerpc.h, stat.h including is missing for "struct stat". This is fixed by adding "sys/stat.h" in pkeys-powerpc.h Secondly, pkey-powerpc.h:55:18: warning: format `%llx' expects argument of type `long long unsigned int', but argument 3 has type `u64' {aka `long unsigned int'} [-Wformat=] 55 | dprintf4("%s() changing %016llx to %016llx\n", | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 56 | __func__, __read_pkey_reg(), pkey_reg); | ~~~~~~~~~~~~~~~~~ | | | u64 {aka long unsigned int} pkey-helpers.h:63:32: note: in definition of macro `dprintf_level' 63 | sigsafe_printf(args); \ | ^~~~ These format specifier related warning are removed by adding "__SANE_USERSPACE_TYPES__" to pkeys_utils.c. Link: https://lkml.kernel.org/r/20250428131937.641989-1-nysal@linux.ibm.com Fixes: 50910acd6f61 ("selftests/mm: use sys_pkey helpers consistently") Signed-off-by: Madhavan Srinivasan Signed-off-by: Nysal Jan K.A. Tested-by: Venkat Rao Bagalkote Cc: Signed-off-by: Andrew Morton --- tools/testing/selftests/mm/pkey-powerpc.h | 2 ++ tools/testing/selftests/mm/pkey_util.c | 1 + 2 files changed, 3 insertions(+) diff --git a/tools/testing/selftests/mm/pkey-powerpc.h b/tools/testing/selftests/mm/pkey-powerpc.h index 1bad310d282a..d8ec906b8120 100644 --- a/tools/testing/selftests/mm/pkey-powerpc.h +++ b/tools/testing/selftests/mm/pkey-powerpc.h @@ -3,6 +3,8 @@ #ifndef _PKEYS_POWERPC_H #define _PKEYS_POWERPC_H +#include + #ifndef SYS_pkey_alloc # define SYS_pkey_alloc 384 # define SYS_pkey_free 385 diff --git a/tools/testing/selftests/mm/pkey_util.c b/tools/testing/selftests/mm/pkey_util.c index ca4ad0d44ab2..255b332f7a08 100644 --- a/tools/testing/selftests/mm/pkey_util.c +++ b/tools/testing/selftests/mm/pkey_util.c @@ -1,4 +1,5 @@ // SPDX-License-Identifier: GPL-2.0-only +#define __SANE_USERSPACE_TYPES__ #include #include From 8cf6ecb18baac867585fe1cba5dde6dbf3b6d29a Mon Sep 17 00:00:00 2001 From: "Nysal Jan K.A." Date: Mon, 28 Apr 2025 18:49:35 +0530 Subject: [PATCH 295/555] selftests/mm: fix a build failure on powerpc The compiler is unaware of the size of code generated by the ".rept" assembler directive. This results in the compiler emitting branch instructions where the offset to branch to exceeds the maximum allowed value, resulting in build failures like the following: CC protection_keys /tmp/ccypKWAE.s: Assembler messages: /tmp/ccypKWAE.s:2073: Error: operand out of range (0x0000000000020158 is not between 0xffffffffffff8000 and 0x0000000000007ffc) /tmp/ccypKWAE.s:2509: Error: operand out of range (0x0000000000020130 is not between 0xffffffffffff8000 and 0x0000000000007ffc) Fix the issue by manually adding nop instructions using the preprocessor. Link: https://lkml.kernel.org/r/20250428131937.641989-2-nysal@linux.ibm.com Fixes: 46036188ea1f ("selftests/mm: build with -O2") Reported-by: Madhavan Srinivasan Signed-off-by: Nysal Jan K.A. Tested-by: Venkat Rao Bagalkote Reviewed-by: Donet Tom Tested-by: Donet Tom Cc: Signed-off-by: Andrew Morton --- tools/testing/selftests/mm/pkey-powerpc.h | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/mm/pkey-powerpc.h b/tools/testing/selftests/mm/pkey-powerpc.h index d8ec906b8120..17bf2d1b0192 100644 --- a/tools/testing/selftests/mm/pkey-powerpc.h +++ b/tools/testing/selftests/mm/pkey-powerpc.h @@ -104,8 +104,18 @@ static inline void expect_fault_on_read_execonly_key(void *p1, int pkey) return; } +#define REPEAT_8(s) s s s s s s s s +#define REPEAT_64(s) REPEAT_8(s) REPEAT_8(s) REPEAT_8(s) REPEAT_8(s) \ + REPEAT_8(s) REPEAT_8(s) REPEAT_8(s) REPEAT_8(s) +#define REPEAT_512(s) REPEAT_64(s) REPEAT_64(s) REPEAT_64(s) REPEAT_64(s) \ + REPEAT_64(s) REPEAT_64(s) REPEAT_64(s) REPEAT_64(s) +#define REPEAT_4096(s) REPEAT_512(s) REPEAT_512(s) REPEAT_512(s) REPEAT_512(s) \ + REPEAT_512(s) REPEAT_512(s) REPEAT_512(s) REPEAT_512(s) +#define REPEAT_16384(s) REPEAT_4096(s) REPEAT_4096(s) \ + REPEAT_4096(s) REPEAT_4096(s) + /* 4-byte instructions * 16384 = 64K page */ -#define __page_o_noops() asm(".rept 16384 ; nop; .endr") +#define __page_o_noops() asm(REPEAT_16384("nop\n")) static inline void *malloc_pkey_with_mprotect_subpage(long size, int prot, u16 pkey) { From 9a9794a81a8a1be8670aa673ec0f0fbfbddeccae Mon Sep 17 00:00:00 2001 From: Kairui Song Date: Tue, 29 Apr 2025 17:48:03 +0800 Subject: [PATCH 296/555] mm, swap: fix false warning for large allocation with !THP_SWAP The !CONFIG_THP_SWAP check existed before just fine because slot cache would reject high order allocation and let the caller split all folios and try again. But slot cache is gone, so large allocation will directly go to the allocator, and the allocator should just fail silently to inform caller to do the folio split, this is totally fine and expected. Remove this meaningless warning. Link: https://lkml.kernel.org/r/20250429094803.85518-1-ryncsn@gmail.com Fixes: 0ff67f990bd4 ("mm, swap: remove swap slot cache") Signed-off-by: Kairui Song Reported-by: Heiko Carstens Closes: https://lore.kernel.org/linux-mm/20250428135252.25453B17-hca@linux.ibm.com/ Tested-by: Heiko Carstens Cc: Baoquan He Signed-off-by: Andrew Morton --- mm/swapfile.c | 23 ++++++++++++++++------- 1 file changed, 16 insertions(+), 7 deletions(-) diff --git a/mm/swapfile.c b/mm/swapfile.c index 2eff8b51a945..f214843612dc 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -1272,13 +1272,22 @@ int folio_alloc_swap(struct folio *folio, gfp_t gfp) VM_BUG_ON_FOLIO(!folio_test_locked(folio), folio); VM_BUG_ON_FOLIO(!folio_test_uptodate(folio), folio); - /* - * Should not even be attempting large allocations when huge - * page swap is disabled. Warn and fail the allocation. - */ - if (order && (!IS_ENABLED(CONFIG_THP_SWAP) || size > SWAPFILE_CLUSTER)) { - VM_WARN_ON_ONCE(1); - return -EINVAL; + if (order) { + /* + * Reject large allocation when THP_SWAP is disabled, + * the caller should split the folio and try again. + */ + if (!IS_ENABLED(CONFIG_THP_SWAP)) + return -EAGAIN; + + /* + * Allocation size should never exceed cluster size + * (HPAGE_PMD_SIZE). + */ + if (size > SWAPFILE_CLUSTER) { + VM_WARN_ON_ONCE(1); + return -EINVAL; + } } local_lock(&percpu_swap_cluster.lock); From dac2a4f663c4bcd75add07aedf9d30c9f8e5ead5 Mon Sep 17 00:00:00 2001 From: Frank van der Linden Date: Thu, 1 May 2025 04:43:24 +0000 Subject: [PATCH 297/555] mm/hugetlb: copy the CMA flag when demoting Since commit d2d786714080 ("mm/hugetlb: enable bootmem allocation from CMA areas"), a flag is used to mark hugetlb folios as allocated from CMA. This flag is also used to decide if it should be freed to CMA. However, the flag isn't copied to the smaller folios when a hugetlb folio is broken up for demotion, which would cause it to be freed incorrectly. Fix this by copying the flag to the smaller order hugetlb pages created from the original one. Link: https://lkml.kernel.org/r/20250501044325.20365-1-fvdl@google.com Fixes: d2d786714080 ("mm/hugetlb: enable bootmem allocation from CMA areas") Signed-off-by: Frank van der Linden Reviewed-by: Anshuman Khandual Reviewed-by: Oscar Salvador Reviewed-by: Jane Chu Cc: Muchun Song Signed-off-by: Andrew Morton --- mm/hugetlb.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/mm/hugetlb.c b/mm/hugetlb.c index e3e6ac991b9c..6ea1be71aa42 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -4034,10 +4034,13 @@ static long demote_free_hugetlb_folios(struct hstate *src, struct hstate *dst, list_for_each_entry_safe(folio, next, src_list, lru) { int i; + bool cma; if (folio_test_hugetlb_vmemmap_optimized(folio)) continue; + cma = folio_test_hugetlb_cma(folio); + list_del(&folio->lru); split_page_owner(&folio->page, huge_page_order(src), huge_page_order(dst)); @@ -4053,6 +4056,9 @@ static long demote_free_hugetlb_folios(struct hstate *src, struct hstate *dst, new_folio->mapping = NULL; init_new_hugetlb_folio(dst, new_folio); + /* Copy the CMA flag so that it is freed correctly */ + if (cma) + folio_set_hugetlb_cma(new_folio); list_add(&new_folio->lru, &dst_list); } } From fb881cd7604536b17a1927fb0533f9a6982ffcc5 Mon Sep 17 00:00:00 2001 From: Ryusuke Konishi Date: Sat, 3 May 2025 14:33:14 +0900 Subject: [PATCH 298/555] nilfs2: fix deadlock warnings caused by lock dependency in init_nilfs() After commit c0e473a0d226 ("block: fix race between set_blocksize and read paths") was merged, set_blocksize() called by sb_set_blocksize() now locks the inode of the backing device file. As a result of this change, syzbot started reporting deadlock warnings due to a circular dependency involving the semaphore "ns_sem" of the nilfs object, the inode lock of the backing device file, and the locks that this inode lock is transitively dependent on. This is caused by a new lock dependency added by the above change, since init_nilfs() calls sb_set_blocksize() in the lock section of "ns_sem". However, these warnings are false positives because init_nilfs() is called in the early stage of the mount operation and the filesystem has not yet started. The reason why "ns_sem" is locked in init_nilfs() was to avoid a race condition in nilfs_fill_super() caused by sharing a nilfs object among multiple filesystem instances (super block structures) in the early implementation. However, nilfs objects and super block structures have long ago become one-to-one, and there is no longer any need to use the semaphore there. So, fix this issue by removing the use of the semaphore "ns_sem" in init_nilfs(). Link: https://lkml.kernel.org/r/20250503053327.12294-1-konishi.ryusuke@gmail.com Fixes: c0e473a0d226 ("block: fix race between set_blocksize and read paths") Signed-off-by: Ryusuke Konishi Reported-by: syzbot+00f7f5b884b117ee6773@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=00f7f5b884b117ee6773 Tested-by: syzbot+00f7f5b884b117ee6773@syzkaller.appspotmail.com Reported-by: syzbot+f30591e72bfc24d4715b@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=f30591e72bfc24d4715b Tested-by: syzbot+f30591e72bfc24d4715b@syzkaller.appspotmail.com> Signed-off-by: Andrew Morton --- fs/nilfs2/the_nilfs.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/fs/nilfs2/the_nilfs.c b/fs/nilfs2/the_nilfs.c index cb01ea81724d..d0bcf744c553 100644 --- a/fs/nilfs2/the_nilfs.c +++ b/fs/nilfs2/the_nilfs.c @@ -705,8 +705,6 @@ int init_nilfs(struct the_nilfs *nilfs, struct super_block *sb) int blocksize; int err; - down_write(&nilfs->ns_sem); - blocksize = sb_min_blocksize(sb, NILFS_MIN_BLOCK_SIZE); if (!blocksize) { nilfs_err(sb, "unable to set blocksize"); @@ -779,7 +777,6 @@ int init_nilfs(struct the_nilfs *nilfs, struct super_block *sb) set_nilfs_init(nilfs); err = 0; out: - up_write(&nilfs->ns_sem); return err; failed_sbh: From 7b08b74f3d99f6b801250683c751d391128799ec Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Petr=20Van=C4=9Bk?= Date: Fri, 2 May 2025 23:50:19 +0200 Subject: [PATCH 299/555] mm: fix folio_pte_batch() on XEN PV MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit On XEN PV, folio_pte_batch() can incorrectly batch beyond the end of a folio due to a corner case in pte_advance_pfn(). Specifically, when the PFN following the folio maps to an invalidated MFN, expected_pte = pte_advance_pfn(expected_pte, nr); produces a pte_none(). If the actual next PTE in memory is also pte_none(), the pte_same() succeeds, if (!pte_same(pte, expected_pte)) break; the loop is not broken, and batching continues into unrelated memory. For example, with a 4-page folio, the PTE layout might look like this: [ 53.465673] [ T2552] folio_pte_batch: printing PTE values at addr=0x7f1ac9dc5000 [ 53.465674] [ T2552] PTE[453] = 000000010085c125 [ 53.465679] [ T2552] PTE[454] = 000000010085d125 [ 53.465682] [ T2552] PTE[455] = 000000010085e125 [ 53.465684] [ T2552] PTE[456] = 000000010085f125 [ 53.465686] [ T2552] PTE[457] = 0000000000000000 <-- not present [ 53.465689] [ T2552] PTE[458] = 0000000101da7125 pte_advance_pfn(PTE[456]) returns a pte_none() due to invalid PFN->MFN mapping. The next actual PTE (PTE[457]) is also pte_none(), so the loop continues and includes PTE[457] in the batch, resulting in 5 batched entries for a 4-page folio. This triggers the following warning: [ 53.465751] [ T2552] page: refcount:85 mapcount:20 mapping:ffff88813ff4f6a8 index:0x110 pfn:0x10085c [ 53.465754] [ T2552] head: order:2 mapcount:80 entire_mapcount:0 nr_pages_mapped:4 pincount:0 [ 53.465756] [ T2552] memcg:ffff888003573000 [ 53.465758] [ T2552] aops:0xffffffff8226fd20 ino:82467c dentry name(?):"libc.so.6" [ 53.465761] [ T2552] flags: 0x2000000000416c(referenced|uptodate|lru|active|private|head|node=0|zone=2) [ 53.465764] [ T2552] raw: 002000000000416c ffffea0004021f08 ffffea0004021908 ffff88813ff4f6a8 [ 53.465767] [ T2552] raw: 0000000000000110 ffff888133d8bd40 0000005500000013 ffff888003573000 [ 53.465768] [ T2552] head: 002000000000416c ffffea0004021f08 ffffea0004021908 ffff88813ff4f6a8 [ 53.465770] [ T2552] head: 0000000000000110 ffff888133d8bd40 0000005500000013 ffff888003573000 [ 53.465772] [ T2552] head: 0020000000000202 ffffea0004021701 000000040000004f 00000000ffffffff [ 53.465774] [ T2552] head: 0000000300000003 8000000300000002 0000000000000013 0000000000000004 [ 53.465775] [ T2552] page dumped because: VM_WARN_ON_FOLIO((_Generic((page + nr_pages - 1), const struct page *: (const struct folio *)_compound_head(page + nr_pages - 1), struct page *: (struct folio *)_compound_head(page + nr_pages - 1))) != folio) Original code works as expected everywhere, except on XEN PV, where pte_advance_pfn() can yield a pte_none() after balloon inflation due to MFNs invalidation. In XEN, pte_advance_pfn() ends up calling __pte()->xen_make_pte()->pte_pfn_to_mfn(), which returns pte_none() when mfn == INVALID_P2M_ENTRY. The pte_pfn_to_mfn() documents that nastiness: If there's no mfn for the pfn, then just create an empty non-present pte. Unfortunately this loses information about the original pfn, so pte_mfn_to_pfn is asymmetric. While such hacks should certainly be removed, we can do better in folio_pte_batch() and simply check ahead of time how many PTEs we can possibly batch in our folio. This way, we can not only fix the issue but cleanup the code: removing the pte_pfn() check inside the loop body and avoiding end_ptr comparison + arithmetic. Link: https://lkml.kernel.org/r/20250502215019.822-2-arkamar@atlas.cz Fixes: f8d937761d65 ("mm/memory: optimize fork() with PTE-mapped THP") Co-developed-by: David Hildenbrand Signed-off-by: David Hildenbrand Signed-off-by: Petr Vaněk Cc: Ryan Roberts Cc: Signed-off-by: Andrew Morton --- mm/internal.h | 27 +++++++++++---------------- 1 file changed, 11 insertions(+), 16 deletions(-) diff --git a/mm/internal.h b/mm/internal.h index e9695baa5922..25a29872c634 100644 --- a/mm/internal.h +++ b/mm/internal.h @@ -248,11 +248,9 @@ static inline int folio_pte_batch(struct folio *folio, unsigned long addr, pte_t *start_ptep, pte_t pte, int max_nr, fpb_t flags, bool *any_writable, bool *any_young, bool *any_dirty) { - unsigned long folio_end_pfn = folio_pfn(folio) + folio_nr_pages(folio); - const pte_t *end_ptep = start_ptep + max_nr; pte_t expected_pte, *ptep; bool writable, young, dirty; - int nr; + int nr, cur_nr; if (any_writable) *any_writable = false; @@ -265,11 +263,15 @@ static inline int folio_pte_batch(struct folio *folio, unsigned long addr, VM_WARN_ON_FOLIO(!folio_test_large(folio) || max_nr < 1, folio); VM_WARN_ON_FOLIO(page_folio(pfn_to_page(pte_pfn(pte))) != folio, folio); + /* Limit max_nr to the actual remaining PFNs in the folio we could batch. */ + max_nr = min_t(unsigned long, max_nr, + folio_pfn(folio) + folio_nr_pages(folio) - pte_pfn(pte)); + nr = pte_batch_hint(start_ptep, pte); expected_pte = __pte_batch_clear_ignored(pte_advance_pfn(pte, nr), flags); ptep = start_ptep + nr; - while (ptep < end_ptep) { + while (nr < max_nr) { pte = ptep_get(ptep); if (any_writable) writable = !!pte_write(pte); @@ -282,14 +284,6 @@ static inline int folio_pte_batch(struct folio *folio, unsigned long addr, if (!pte_same(pte, expected_pte)) break; - /* - * Stop immediately once we reached the end of the folio. In - * corner cases the next PFN might fall into a different - * folio. - */ - if (pte_pfn(pte) >= folio_end_pfn) - break; - if (any_writable) *any_writable |= writable; if (any_young) @@ -297,12 +291,13 @@ static inline int folio_pte_batch(struct folio *folio, unsigned long addr, if (any_dirty) *any_dirty |= dirty; - nr = pte_batch_hint(ptep, pte); - expected_pte = pte_advance_pfn(expected_pte, nr); - ptep += nr; + cur_nr = pte_batch_hint(ptep, pte); + expected_pte = pte_advance_pfn(expected_pte, cur_nr); + ptep += cur_nr; + nr += cur_nr; } - return min(ptep - start_ptep, max_nr); + return min(nr, max_nr); } /** From f34343cc11afc7bb1f881c3492bee3484016bf71 Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Tue, 6 May 2025 08:59:39 -0700 Subject: [PATCH 300/555] fbnic: Fix initialization of mailbox descriptor rings Address to issues with the FW mailbox descriptor initialization. We need to reverse the order of accesses when we invalidate an entry versus writing an entry. When writing an entry we write upper and then lower as the lower 32b contain the valid bit that makes the entire address valid. However for invalidation we should write it in the reverse order so that the upper is marked invalid before we update it. Without this change we may see FW attempt to access pages with the upper 32b of the address set to 0 which will likely result in DMAR faults due to write access failures on mailbox shutdown. Fixes: da3cde08209e ("eth: fbnic: Add FW communication mechanism") Signed-off-by: Alexander Duyck Reviewed-by: Simon Horman Reviewed-by: Jacob Keller Link: https://patch.msgid.link/174654717972.499179.8083789731819297034.stgit@ahduyck-xeon-server.home.arpa Reviewed-by: Jakub Kicinski Signed-off-by: Paolo Abeni --- drivers/net/ethernet/meta/fbnic/fbnic_fw.c | 32 ++++++++++++++++------ 1 file changed, 23 insertions(+), 9 deletions(-) diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_fw.c b/drivers/net/ethernet/meta/fbnic/fbnic_fw.c index 88db3dacb940..c4956f0a741e 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_fw.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_fw.c @@ -17,11 +17,29 @@ static void __fbnic_mbx_wr_desc(struct fbnic_dev *fbd, int mbx_idx, { u32 desc_offset = FBNIC_IPC_MBX(mbx_idx, desc_idx); + /* Write the upper 32b and then the lower 32b. Doing this the + * FW can then read lower, upper, lower to verify that the state + * of the descriptor wasn't changed mid-transaction. + */ fw_wr32(fbd, desc_offset + 1, upper_32_bits(desc)); fw_wrfl(fbd); fw_wr32(fbd, desc_offset, lower_32_bits(desc)); } +static void __fbnic_mbx_invalidate_desc(struct fbnic_dev *fbd, int mbx_idx, + int desc_idx, u32 desc) +{ + u32 desc_offset = FBNIC_IPC_MBX(mbx_idx, desc_idx); + + /* For initialization we write the lower 32b of the descriptor first. + * This way we can set the state to mark it invalid before we clear the + * upper 32b. + */ + fw_wr32(fbd, desc_offset, desc); + fw_wrfl(fbd); + fw_wr32(fbd, desc_offset + 1, 0); +} + static u64 __fbnic_mbx_rd_desc(struct fbnic_dev *fbd, int mbx_idx, int desc_idx) { u32 desc_offset = FBNIC_IPC_MBX(mbx_idx, desc_idx); @@ -41,21 +59,17 @@ static void fbnic_mbx_init_desc_ring(struct fbnic_dev *fbd, int mbx_idx) * solid stop for the firmware to hit when it is done looping * through the ring. */ - __fbnic_mbx_wr_desc(fbd, mbx_idx, 0, 0); - - fw_wrfl(fbd); + __fbnic_mbx_invalidate_desc(fbd, mbx_idx, 0, 0); /* We then fill the rest of the ring starting at the end and moving * back toward descriptor 0 with skip descriptors that have no * length nor address, and tell the firmware that they can skip * them and just move past them to the one we initialized to 0. */ - for (desc_idx = FBNIC_IPC_MBX_DESC_LEN; --desc_idx;) { - __fbnic_mbx_wr_desc(fbd, mbx_idx, desc_idx, - FBNIC_IPC_MBX_DESC_FW_CMPL | - FBNIC_IPC_MBX_DESC_HOST_CMPL); - fw_wrfl(fbd); - } + for (desc_idx = FBNIC_IPC_MBX_DESC_LEN; --desc_idx;) + __fbnic_mbx_invalidate_desc(fbd, mbx_idx, desc_idx, + FBNIC_IPC_MBX_DESC_FW_CMPL | + FBNIC_IPC_MBX_DESC_HOST_CMPL); } void fbnic_mbx_init(struct fbnic_dev *fbd) From 3b12f00ddd08e888273b2ac0488d396d90a836fc Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Tue, 6 May 2025 08:59:46 -0700 Subject: [PATCH 301/555] fbnic: Gate AXI read/write enabling on FW mailbox In order to prevent the device from throwing spurious writes and/or reads at us we need to gate the AXI fabric interface to the PCIe until such time as we know the FW is in a known good state. To accomplish this we use the mailbox as a mechanism for us to recognize that the FW has acknowledged our presence and is no longer sending any stale message data to us. We start in fbnic_mbx_init by calling fbnic_mbx_reset_desc_ring function, disabling the DMA in both directions, and then invalidating all the descriptors in each ring. We then poll the mailbox in fbnic_mbx_poll_tx_ready and when the interrupt is set by the FW we pick it up and mark the mailboxes as ready, while also enabling the DMA. Once we have completed all the transactions and need to shut down we call into fbnic_mbx_clean which will in turn call fbnic_mbx_reset_desc_ring for each ring and shut down the DMA and once again invalidate the descriptors. Fixes: 3646153161f1 ("eth: fbnic: Add register init to set PCIe/Ethernet device config") Fixes: da3cde08209e ("eth: fbnic: Add FW communication mechanism") Signed-off-by: Alexander Duyck Reviewed-by: Simon Horman Reviewed-by: Jacob Keller Link: https://patch.msgid.link/174654718623.499179.7445197308109347982.stgit@ahduyck-xeon-server.home.arpa Reviewed-by: Jakub Kicinski Signed-off-by: Paolo Abeni --- drivers/net/ethernet/meta/fbnic/fbnic_csr.h | 2 ++ drivers/net/ethernet/meta/fbnic/fbnic_fw.c | 38 +++++++++++++++++---- drivers/net/ethernet/meta/fbnic/fbnic_mac.c | 6 ---- 3 files changed, 33 insertions(+), 13 deletions(-) diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_csr.h b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h index 3b12a0ab5906..51bee8072420 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_csr.h +++ b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h @@ -796,8 +796,10 @@ enum { /* PUL User Registers */ #define FBNIC_CSR_START_PUL_USER 0x31000 /* CSR section delimiter */ #define FBNIC_PUL_OB_TLP_HDR_AW_CFG 0x3103d /* 0xc40f4 */ +#define FBNIC_PUL_OB_TLP_HDR_AW_CFG_FLUSH CSR_BIT(19) #define FBNIC_PUL_OB_TLP_HDR_AW_CFG_BME CSR_BIT(18) #define FBNIC_PUL_OB_TLP_HDR_AR_CFG 0x3103e /* 0xc40f8 */ +#define FBNIC_PUL_OB_TLP_HDR_AR_CFG_FLUSH CSR_BIT(19) #define FBNIC_PUL_OB_TLP_HDR_AR_CFG_BME CSR_BIT(18) #define FBNIC_PUL_USER_OB_RD_TLP_CNT_31_0 \ 0x3106e /* 0xc41b8 */ diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_fw.c b/drivers/net/ethernet/meta/fbnic/fbnic_fw.c index c4956f0a741e..f4749bfd840c 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_fw.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_fw.c @@ -51,10 +51,26 @@ static u64 __fbnic_mbx_rd_desc(struct fbnic_dev *fbd, int mbx_idx, int desc_idx) return desc; } -static void fbnic_mbx_init_desc_ring(struct fbnic_dev *fbd, int mbx_idx) +static void fbnic_mbx_reset_desc_ring(struct fbnic_dev *fbd, int mbx_idx) { int desc_idx; + /* Disable DMA transactions from the device, + * and flush any transactions triggered during cleaning + */ + switch (mbx_idx) { + case FBNIC_IPC_MBX_RX_IDX: + wr32(fbd, FBNIC_PUL_OB_TLP_HDR_AW_CFG, + FBNIC_PUL_OB_TLP_HDR_AW_CFG_FLUSH); + break; + case FBNIC_IPC_MBX_TX_IDX: + wr32(fbd, FBNIC_PUL_OB_TLP_HDR_AR_CFG, + FBNIC_PUL_OB_TLP_HDR_AR_CFG_FLUSH); + break; + } + + wrfl(fbd); + /* Initialize first descriptor to all 0s. Doing this gives us a * solid stop for the firmware to hit when it is done looping * through the ring. @@ -90,7 +106,7 @@ void fbnic_mbx_init(struct fbnic_dev *fbd) wr32(fbd, FBNIC_INTR_CLEAR(0), 1u << FBNIC_FW_MSIX_ENTRY); for (i = 0; i < FBNIC_IPC_MBX_INDICES; i++) - fbnic_mbx_init_desc_ring(fbd, i); + fbnic_mbx_reset_desc_ring(fbd, i); } static int fbnic_mbx_map_msg(struct fbnic_dev *fbd, int mbx_idx, @@ -155,7 +171,7 @@ static void fbnic_mbx_clean_desc_ring(struct fbnic_dev *fbd, int mbx_idx) { int i; - fbnic_mbx_init_desc_ring(fbd, mbx_idx); + fbnic_mbx_reset_desc_ring(fbd, mbx_idx); for (i = FBNIC_IPC_MBX_DESC_LEN; i--;) fbnic_mbx_unmap_and_free_msg(fbd, mbx_idx, i); @@ -354,7 +370,7 @@ static int fbnic_fw_xmit_cap_msg(struct fbnic_dev *fbd) return (err == -EOPNOTSUPP) ? 0 : err; } -static void fbnic_mbx_postinit_desc_ring(struct fbnic_dev *fbd, int mbx_idx) +static void fbnic_mbx_init_desc_ring(struct fbnic_dev *fbd, int mbx_idx) { struct fbnic_fw_mbx *mbx = &fbd->mbx[mbx_idx]; @@ -366,10 +382,18 @@ static void fbnic_mbx_postinit_desc_ring(struct fbnic_dev *fbd, int mbx_idx) switch (mbx_idx) { case FBNIC_IPC_MBX_RX_IDX: + /* Enable DMA writes from the device */ + wr32(fbd, FBNIC_PUL_OB_TLP_HDR_AW_CFG, + FBNIC_PUL_OB_TLP_HDR_AW_CFG_BME); + /* Make sure we have a page for the FW to write to */ fbnic_mbx_alloc_rx_msgs(fbd); break; case FBNIC_IPC_MBX_TX_IDX: + /* Enable DMA reads from the device */ + wr32(fbd, FBNIC_PUL_OB_TLP_HDR_AR_CFG, + FBNIC_PUL_OB_TLP_HDR_AR_CFG_BME); + /* Force version to 1 if we successfully requested an update * from the firmware. This should be overwritten once we get * the actual version from the firmware in the capabilities @@ -386,7 +410,7 @@ static void fbnic_mbx_postinit(struct fbnic_dev *fbd) { int i; - /* We only need to do this on the first interrupt following init. + /* We only need to do this on the first interrupt following reset. * this primes the mailbox so that we will have cleared all the * skip descriptors. */ @@ -396,7 +420,7 @@ static void fbnic_mbx_postinit(struct fbnic_dev *fbd) wr32(fbd, FBNIC_INTR_CLEAR(0), 1u << FBNIC_FW_MSIX_ENTRY); for (i = 0; i < FBNIC_IPC_MBX_INDICES; i++) - fbnic_mbx_postinit_desc_ring(fbd, i); + fbnic_mbx_init_desc_ring(fbd, i); } /** @@ -894,7 +918,7 @@ int fbnic_mbx_poll_tx_ready(struct fbnic_dev *fbd) * avoid the mailbox getting stuck closed if the interrupt * is reset. */ - fbnic_mbx_init_desc_ring(fbd, FBNIC_IPC_MBX_TX_IDX); + fbnic_mbx_reset_desc_ring(fbd, FBNIC_IPC_MBX_TX_IDX); msleep(200); diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_mac.c b/drivers/net/ethernet/meta/fbnic/fbnic_mac.c index 14291401f463..dde4a37116e2 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_mac.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_mac.c @@ -79,12 +79,6 @@ static void fbnic_mac_init_axi(struct fbnic_dev *fbd) fbnic_init_readrq(fbd, FBNIC_QM_RNI_RBP_CTL, cls, readrq); fbnic_init_mps(fbd, FBNIC_QM_RNI_RDE_CTL, cls, mps); fbnic_init_mps(fbd, FBNIC_QM_RNI_RCM_CTL, cls, mps); - - /* Enable XALI AR/AW outbound */ - wr32(fbd, FBNIC_PUL_OB_TLP_HDR_AW_CFG, - FBNIC_PUL_OB_TLP_HDR_AW_CFG_BME); - wr32(fbd, FBNIC_PUL_OB_TLP_HDR_AR_CFG, - FBNIC_PUL_OB_TLP_HDR_AR_CFG_BME); } static void fbnic_mac_init_qm(struct fbnic_dev *fbd) From 682a61281d1036962b68d651994cc407371daef5 Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Tue, 6 May 2025 08:59:52 -0700 Subject: [PATCH 302/555] fbnic: Add additional handling of IRQs We have two issues that need to be addressed in our IRQ handling. One is the fact that we can end up double-freeing IRQs in the event of an exception handling error such as a PCIe reset/recovery that fails. To prevent that from becoming an issue we can use the msix_vector values to indicate that we have successfully requested/freed the IRQ by only setting or clearing them when we have completed the given action. The other issue is that we have several potential races in our IRQ path due to us manipulating the mask before the vector has been truly disabled. In order to handle that in the case of the FW mailbox we need to not auto-enable the IRQ and instead will be enabling/disabling it separately. In the case of the PCS vector we can mitigate this by unmapping it and synchronizing the IRQ before we clear the mask. The general order of operations after this change is now to request the interrupt, poll the FW mailbox to ready, and then enable the interrupt. For the shutdown we do the reverse where we disable the interrupt, flush any pending Tx, and then free the IRQ. I am renaming the enable/disable to request/free to be equivilent with the IRQ calls being used. We may see additions in the future to enable/disable the IRQs versus request/free them for certain use cases. Fixes: da3cde08209e ("eth: fbnic: Add FW communication mechanism") Fixes: 69684376eed5 ("eth: fbnic: Add link detection") Signed-off-by: Alexander Duyck Reviewed-by: Simon Horman Reviewed-by: Jacob Keller Link: https://patch.msgid.link/174654719271.499179.3634535105127848325.stgit@ahduyck-xeon-server.home.arpa Reviewed-by: Jakub Kicinski Signed-off-by: Paolo Abeni --- drivers/net/ethernet/meta/fbnic/fbnic.h | 8 +- drivers/net/ethernet/meta/fbnic/fbnic_irq.c | 148 ++++++++++++------ .../net/ethernet/meta/fbnic/fbnic_netdev.c | 5 +- drivers/net/ethernet/meta/fbnic/fbnic_pci.c | 14 +- 4 files changed, 113 insertions(+), 62 deletions(-) diff --git a/drivers/net/ethernet/meta/fbnic/fbnic.h b/drivers/net/ethernet/meta/fbnic/fbnic.h index 4ca7b99ef131..de6b1a340f55 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic.h +++ b/drivers/net/ethernet/meta/fbnic/fbnic.h @@ -154,14 +154,14 @@ struct fbnic_dev *fbnic_devlink_alloc(struct pci_dev *pdev); void fbnic_devlink_register(struct fbnic_dev *fbd); void fbnic_devlink_unregister(struct fbnic_dev *fbd); -int fbnic_fw_enable_mbx(struct fbnic_dev *fbd); -void fbnic_fw_disable_mbx(struct fbnic_dev *fbd); +int fbnic_fw_request_mbx(struct fbnic_dev *fbd); +void fbnic_fw_free_mbx(struct fbnic_dev *fbd); void fbnic_hwmon_register(struct fbnic_dev *fbd); void fbnic_hwmon_unregister(struct fbnic_dev *fbd); -int fbnic_pcs_irq_enable(struct fbnic_dev *fbd); -void fbnic_pcs_irq_disable(struct fbnic_dev *fbd); +int fbnic_pcs_request_irq(struct fbnic_dev *fbd); +void fbnic_pcs_free_irq(struct fbnic_dev *fbd); void fbnic_napi_name_irqs(struct fbnic_dev *fbd); int fbnic_napi_request_irq(struct fbnic_dev *fbd, diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_irq.c b/drivers/net/ethernet/meta/fbnic/fbnic_irq.c index 1bbc0e56f3a0..1c88a2bf3a7a 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_irq.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_irq.c @@ -19,69 +19,105 @@ static irqreturn_t fbnic_fw_msix_intr(int __always_unused irq, void *data) return IRQ_HANDLED; } -/** - * fbnic_fw_enable_mbx - Configure and initialize Firmware Mailbox - * @fbd: Pointer to device to initialize - * - * This function will initialize the firmware mailbox rings, enable the IRQ - * and initialize the communication between the Firmware and the host. The - * firmware is expected to respond to the initialization by sending an - * interrupt essentially notifying the host that it has seen the - * initialization and is now synced up. - * - * Return: non-zero on failure. - **/ -int fbnic_fw_enable_mbx(struct fbnic_dev *fbd) +static int __fbnic_fw_enable_mbx(struct fbnic_dev *fbd, int vector) { - u32 vector = fbd->fw_msix_vector; int err; - /* Request the IRQ for FW Mailbox vector. */ - err = request_threaded_irq(vector, NULL, &fbnic_fw_msix_intr, - IRQF_ONESHOT, dev_name(fbd->dev), fbd); - if (err) - return err; - /* Initialize mailbox and attempt to poll it into ready state */ fbnic_mbx_init(fbd); err = fbnic_mbx_poll_tx_ready(fbd); if (err) { dev_warn(fbd->dev, "FW mailbox did not enter ready state\n"); - free_irq(vector, fbd); return err; } - /* Enable interrupts */ + /* Enable interrupt and unmask the vector */ + enable_irq(vector); fbnic_wr32(fbd, FBNIC_INTR_MASK_CLEAR(0), 1u << FBNIC_FW_MSIX_ENTRY); return 0; } /** - * fbnic_fw_disable_mbx - Disable mailbox and place it in standby state - * @fbd: Pointer to device to disable + * fbnic_fw_request_mbx - Configure and initialize Firmware Mailbox + * @fbd: Pointer to device to initialize * - * This function will disable the mailbox interrupt, free any messages still - * in the mailbox and place it into a standby state. The firmware is - * expected to see the update and assume that the host is in the reset state. + * This function will allocate the IRQ and then reinitialize the mailbox + * starting communication between the host and firmware. + * + * Return: non-zero on failure. **/ -void fbnic_fw_disable_mbx(struct fbnic_dev *fbd) +int fbnic_fw_request_mbx(struct fbnic_dev *fbd) { - /* Disable interrupt and free vector */ - fbnic_wr32(fbd, FBNIC_INTR_MASK_SET(0), 1u << FBNIC_FW_MSIX_ENTRY); + struct pci_dev *pdev = to_pci_dev(fbd->dev); + int vector, err; - /* Free the vector */ - free_irq(fbd->fw_msix_vector, fbd); + WARN_ON(fbd->fw_msix_vector); + + vector = pci_irq_vector(pdev, FBNIC_FW_MSIX_ENTRY); + if (vector < 0) + return vector; + + /* Request the IRQ for FW Mailbox vector. */ + err = request_threaded_irq(vector, NULL, &fbnic_fw_msix_intr, + IRQF_ONESHOT | IRQF_NO_AUTOEN, + dev_name(fbd->dev), fbd); + if (err) + return err; + + /* Initialize mailbox and attempt to poll it into ready state */ + err = __fbnic_fw_enable_mbx(fbd, vector); + if (err) + free_irq(vector, fbd); + + fbd->fw_msix_vector = vector; + + return err; +} + +/** + * fbnic_fw_disable_mbx - Temporarily place mailbox in standby state + * @fbd: Pointer to device + * + * Shutdown the mailbox by notifying the firmware to stop sending us logs, mask + * and synchronize the IRQ, and then clean up the rings. + **/ +static void fbnic_fw_disable_mbx(struct fbnic_dev *fbd) +{ + /* Disable interrupt and synchronize the IRQ */ + disable_irq(fbd->fw_msix_vector); + + /* Mask the vector */ + fbnic_wr32(fbd, FBNIC_INTR_MASK_SET(0), 1u << FBNIC_FW_MSIX_ENTRY); /* Make sure disabling logs message is sent, must be done here to * avoid risk of completing without a running interrupt. */ fbnic_mbx_flush_tx(fbd); - - /* Reset the mailboxes to the initialized state */ fbnic_mbx_clean(fbd); } +/** + * fbnic_fw_free_mbx - Disable mailbox and place it in standby state + * @fbd: Pointer to device to disable + * + * This function will disable the mailbox interrupt, free any messages still + * in the mailbox and place it into a disabled state. The firmware is + * expected to see the update and assume that the host is in the reset state. + **/ +void fbnic_fw_free_mbx(struct fbnic_dev *fbd) +{ + /* Vector has already been freed */ + if (!fbd->fw_msix_vector) + return; + + fbnic_fw_disable_mbx(fbd); + + /* Free the vector */ + free_irq(fbd->fw_msix_vector, fbd); + fbd->fw_msix_vector = 0; +} + static irqreturn_t fbnic_pcs_msix_intr(int __always_unused irq, void *data) { struct fbnic_dev *fbd = data; @@ -101,7 +137,7 @@ static irqreturn_t fbnic_pcs_msix_intr(int __always_unused irq, void *data) } /** - * fbnic_pcs_irq_enable - Configure the MAC to enable it to advertise link + * fbnic_pcs_request_irq - Configure the PCS to enable it to advertise link * @fbd: Pointer to device to initialize * * This function provides basic bringup for the MAC/PCS IRQ. For now the IRQ @@ -109,41 +145,61 @@ static irqreturn_t fbnic_pcs_msix_intr(int __always_unused irq, void *data) * * Return: non-zero on failure. **/ -int fbnic_pcs_irq_enable(struct fbnic_dev *fbd) +int fbnic_pcs_request_irq(struct fbnic_dev *fbd) { - u32 vector = fbd->pcs_msix_vector; - int err; + struct pci_dev *pdev = to_pci_dev(fbd->dev); + int vector, err; - /* Request the IRQ for MAC link vector. - * Map MAC cause to it, and unmask it + WARN_ON(fbd->pcs_msix_vector); + + vector = pci_irq_vector(pdev, FBNIC_PCS_MSIX_ENTRY); + if (vector < 0) + return vector; + + /* Request the IRQ for PCS link vector. + * Map PCS cause to it, and unmask it */ err = request_irq(vector, &fbnic_pcs_msix_intr, 0, fbd->netdev->name, fbd); if (err) return err; + /* Map and enable interrupt, unmask vector after link is configured */ fbnic_wr32(fbd, FBNIC_INTR_MSIX_CTRL(FBNIC_INTR_MSIX_CTRL_PCS_IDX), FBNIC_PCS_MSIX_ENTRY | FBNIC_INTR_MSIX_CTRL_ENABLE); + fbd->pcs_msix_vector = vector; + return 0; } /** - * fbnic_pcs_irq_disable - Teardown the MAC IRQ to prepare for stopping + * fbnic_pcs_free_irq - Teardown the PCS IRQ to prepare for stopping * @fbd: Pointer to device that is stopping * - * This function undoes the work done in fbnic_pcs_irq_enable and prepares + * This function undoes the work done in fbnic_pcs_request_irq and prepares * the device to no longer receive traffic on the host interface. **/ -void fbnic_pcs_irq_disable(struct fbnic_dev *fbd) +void fbnic_pcs_free_irq(struct fbnic_dev *fbd) { + /* Vector has already been freed */ + if (!fbd->pcs_msix_vector) + return; + /* Disable interrupt */ fbnic_wr32(fbd, FBNIC_INTR_MSIX_CTRL(FBNIC_INTR_MSIX_CTRL_PCS_IDX), FBNIC_PCS_MSIX_ENTRY); + fbnic_wrfl(fbd); + + /* Synchronize IRQ to prevent race that would unmask vector */ + synchronize_irq(fbd->pcs_msix_vector); + + /* Mask the vector */ fbnic_wr32(fbd, FBNIC_INTR_MASK_SET(0), 1u << FBNIC_PCS_MSIX_ENTRY); /* Free the vector */ free_irq(fbd->pcs_msix_vector, fbd); + fbd->pcs_msix_vector = 0; } void fbnic_synchronize_irq(struct fbnic_dev *fbd, int nr) @@ -226,9 +282,6 @@ void fbnic_free_irqs(struct fbnic_dev *fbd) { struct pci_dev *pdev = to_pci_dev(fbd->dev); - fbd->pcs_msix_vector = 0; - fbd->fw_msix_vector = 0; - fbd->num_irqs = 0; pci_free_irq_vectors(pdev); @@ -254,8 +307,5 @@ int fbnic_alloc_irqs(struct fbnic_dev *fbd) fbd->num_irqs = num_irqs; - fbd->pcs_msix_vector = pci_irq_vector(pdev, FBNIC_PCS_MSIX_ENTRY); - fbd->fw_msix_vector = pci_irq_vector(pdev, FBNIC_FW_MSIX_ENTRY); - return 0; } diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c index 79a01fdd1dd1..2524d9b88d59 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c @@ -44,9 +44,10 @@ int __fbnic_open(struct fbnic_net *fbn) if (err) goto time_stop; - err = fbnic_pcs_irq_enable(fbd); + err = fbnic_pcs_request_irq(fbd); if (err) goto time_stop; + /* Pull the BMC config and initialize the RPC */ fbnic_bmc_rpc_init(fbd); fbnic_rss_reinit(fbd, fbn); @@ -82,7 +83,7 @@ static int fbnic_stop(struct net_device *netdev) struct fbnic_net *fbn = netdev_priv(netdev); fbnic_down(fbn); - fbnic_pcs_irq_disable(fbn->fbd); + fbnic_pcs_free_irq(fbn->fbd); fbnic_time_stop(fbn); fbnic_fw_xmit_ownership_msg(fbn->fbd, false); diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c index 6cbbc2ee3e1f..4e8595239c0f 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c @@ -283,7 +283,7 @@ static int fbnic_probe(struct pci_dev *pdev, const struct pci_device_id *ent) goto free_irqs; } - err = fbnic_fw_enable_mbx(fbd); + err = fbnic_fw_request_mbx(fbd); if (err) { dev_err(&pdev->dev, "Firmware mailbox initialization failure\n"); @@ -363,7 +363,7 @@ static void fbnic_remove(struct pci_dev *pdev) fbnic_hwmon_unregister(fbd); fbnic_dbg_fbd_exit(fbd); fbnic_devlink_unregister(fbd); - fbnic_fw_disable_mbx(fbd); + fbnic_fw_free_mbx(fbd); fbnic_free_irqs(fbd); fbnic_devlink_free(fbd); @@ -387,7 +387,7 @@ static int fbnic_pm_suspend(struct device *dev) rtnl_unlock(); null_uc_addr: - fbnic_fw_disable_mbx(fbd); + fbnic_fw_free_mbx(fbd); /* Free the IRQs so they aren't trying to occupy sleeping CPUs */ fbnic_free_irqs(fbd); @@ -420,7 +420,7 @@ static int __fbnic_pm_resume(struct device *dev) fbd->mac->init_regs(fbd); /* Re-enable mailbox */ - err = fbnic_fw_enable_mbx(fbd); + err = fbnic_fw_request_mbx(fbd); if (err) goto err_free_irqs; @@ -438,15 +438,15 @@ static int __fbnic_pm_resume(struct device *dev) if (netif_running(netdev)) { err = __fbnic_open(fbn); if (err) - goto err_disable_mbx; + goto err_free_mbx; } rtnl_unlock(); return 0; -err_disable_mbx: +err_free_mbx: rtnl_unlock(); - fbnic_fw_disable_mbx(fbd); + fbnic_fw_free_mbx(fbd); err_free_irqs: fbnic_free_irqs(fbd); err_invalidate_uc_addr: From 0f9a959a0addd9bbc47e5d16c36b3a7f97981915 Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Tue, 6 May 2025 08:59:59 -0700 Subject: [PATCH 303/555] fbnic: Actually flush_tx instead of stalling out The fbnic_mbx_flush_tx function had a number of issues. First, we were waiting 200ms for the firmware to process the packets. We can drop this to 20ms and in almost all cases this should be more than enough time. So by changing this we can significantly reduce shutdown time. Second, we were not making sure that the Tx path was actually shut off. As such we could still have packets added while we were flushing the mailbox. To prevent that we can now clear the ready flag for the Tx side and it should stay down since the interrupt is disabled. Third, we kept re-reading the tail due to the second issue. The tail should not move after we have started the flush so we can just read it once while we are holding the mailbox Tx lock. By doing that we are guaranteed that the value should be consistent. Fourth, we were keeping a count of descriptors cleaned due to the second and third issues called out. That count is not a valid reason to be exiting the cleanup, and with the tail only being read once we shouldn't see any cases where the tail moves after the disable so the tracking of count can be dropped. Fifth, we were using attempts * sleep time to determine how long we would wait in our polling loop to flush out the Tx. This can be very imprecise. In order to tighten up the timing we are shifting over to using a jiffies value of jiffies + 10 * HZ + 1 to determine the jiffies value we should stop polling at as this should be accurate within once sleep cycle for the total amount of time spent polling. Fixes: da3cde08209e ("eth: fbnic: Add FW communication mechanism") Signed-off-by: Alexander Duyck Reviewed-by: Simon Horman Reviewed-by: Jacob Keller Link: https://patch.msgid.link/174654719929.499179.16406653096197423749.stgit@ahduyck-xeon-server.home.arpa Reviewed-by: Jakub Kicinski Signed-off-by: Paolo Abeni --- drivers/net/ethernet/meta/fbnic/fbnic_fw.c | 31 +++++++++++----------- 1 file changed, 16 insertions(+), 15 deletions(-) diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_fw.c b/drivers/net/ethernet/meta/fbnic/fbnic_fw.c index f4749bfd840c..d019191d6ae9 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_fw.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_fw.c @@ -930,35 +930,36 @@ int fbnic_mbx_poll_tx_ready(struct fbnic_dev *fbd) void fbnic_mbx_flush_tx(struct fbnic_dev *fbd) { + unsigned long timeout = jiffies + 10 * HZ + 1; struct fbnic_fw_mbx *tx_mbx; - int attempts = 50; - u8 count = 0; - - /* Nothing to do if there is no mailbox */ - if (!fbnic_fw_present(fbd)) - return; + u8 tail; /* Record current Rx stats */ tx_mbx = &fbd->mbx[FBNIC_IPC_MBX_TX_IDX]; - /* Nothing to do if mailbox never got to ready */ - if (!tx_mbx->ready) - return; + spin_lock_irq(&fbd->fw_tx_lock); + + /* Clear ready to prevent any further attempts to transmit */ + tx_mbx->ready = false; + + /* Read tail to determine the last tail state for the ring */ + tail = tx_mbx->tail; + + spin_unlock_irq(&fbd->fw_tx_lock); /* Give firmware time to process packet, - * we will wait up to 10 seconds which is 50 waits of 200ms. + * we will wait up to 10 seconds which is 500 waits of 20ms. */ do { u8 head = tx_mbx->head; - if (head == tx_mbx->tail) + /* Tx ring is empty once head == tail */ + if (head == tail) break; - msleep(200); + msleep(20); fbnic_mbx_process_tx_msgs(fbd); - - count += (tx_mbx->head - head) % FBNIC_IPC_MBX_DESC_LEN; - } while (count < FBNIC_IPC_MBX_DESC_LEN && --attempts); + } while (time_is_after_jiffies(timeout)); } void fbnic_get_fw_ver_commit_str(struct fbnic_dev *fbd, char *fw_version, From cdbb2dc3996a60ed3d7431c1239a8ca98c778e04 Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Tue, 6 May 2025 09:00:05 -0700 Subject: [PATCH 304/555] fbnic: Cleanup handling of completions There was an issue in that if we were to shutdown we could be left with a completion in flight as the mailbox went away. To address that I have added an fbnic_mbx_evict_all_cmpl function that is meant to essentially create a "broken pipe" type response so that all callers will receive an error indicating that the connection has been broken as a result of us shutting down the mailbox. Fixes: 378e5cc1c6c6 ("eth: fbnic: hwmon: Add completion infrastructure for firmware requests") Signed-off-by: Alexander Duyck Reviewed-by: Jacob Keller Link: https://patch.msgid.link/174654720578.499179.380252598204530873.stgit@ahduyck-xeon-server.home.arpa Reviewed-by: Jakub Kicinski Signed-off-by: Paolo Abeni --- drivers/net/ethernet/meta/fbnic/fbnic_fw.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_fw.c b/drivers/net/ethernet/meta/fbnic/fbnic_fw.c index d019191d6ae9..732875aae46c 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_fw.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_fw.c @@ -928,6 +928,20 @@ int fbnic_mbx_poll_tx_ready(struct fbnic_dev *fbd) return attempts ? 0 : -ETIMEDOUT; } +static void __fbnic_fw_evict_cmpl(struct fbnic_fw_completion *cmpl_data) +{ + cmpl_data->result = -EPIPE; + complete(&cmpl_data->done); +} + +static void fbnic_mbx_evict_all_cmpl(struct fbnic_dev *fbd) +{ + if (fbd->cmpl_data) { + __fbnic_fw_evict_cmpl(fbd->cmpl_data); + fbd->cmpl_data = NULL; + } +} + void fbnic_mbx_flush_tx(struct fbnic_dev *fbd) { unsigned long timeout = jiffies + 10 * HZ + 1; @@ -945,6 +959,9 @@ void fbnic_mbx_flush_tx(struct fbnic_dev *fbd) /* Read tail to determine the last tail state for the ring */ tail = tx_mbx->tail; + /* Flush any completions as we are no longer processing Rx */ + fbnic_mbx_evict_all_cmpl(fbd); + spin_unlock_irq(&fbd->fw_tx_lock); /* Give firmware time to process packet, From ab064f6005973d456f95ae99cd9ea0d8ab676cce Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Tue, 6 May 2025 09:00:12 -0700 Subject: [PATCH 305/555] fbnic: Improve responsiveness of fbnic_mbx_poll_tx_ready There were a couple different issues found in fbnic_mbx_poll_tx_ready. Among them were the fact that we were sleeping much longer than we actually needed to as the actual FW could respond in under 20ms. The other issue was that we would just keep polling the mailbox even if the device itself had gone away. To address the responsiveness issues we can decrease the sleeps to 20ms and use a jiffies based timeout value rather than just counting the number of times we slept and then polled. To address the hardware going away we can move the check for the firmware BAR being present from where it was and place it inside the loop after the mailbox descriptor ring is initialized and before we sleep so that we just abort and return an error if the device went away during initialization. With these two changes we see a significant improvement in boot times for the driver. Fixes: da3cde08209e ("eth: fbnic: Add FW communication mechanism") Signed-off-by: Alexander Duyck Reviewed-by: Jacob Keller Link: https://patch.msgid.link/174654721224.499179.2698616208976624755.stgit@ahduyck-xeon-server.home.arpa Reviewed-by: Jakub Kicinski Signed-off-by: Paolo Abeni --- drivers/net/ethernet/meta/fbnic/fbnic_fw.c | 19 +++++++++++-------- 1 file changed, 11 insertions(+), 8 deletions(-) diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_fw.c b/drivers/net/ethernet/meta/fbnic/fbnic_fw.c index 732875aae46c..d344b454f28b 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_fw.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_fw.c @@ -905,27 +905,30 @@ void fbnic_mbx_poll(struct fbnic_dev *fbd) int fbnic_mbx_poll_tx_ready(struct fbnic_dev *fbd) { + unsigned long timeout = jiffies + 10 * HZ + 1; struct fbnic_fw_mbx *tx_mbx; - int attempts = 50; - - /* Immediate fail if BAR4 isn't there */ - if (!fbnic_fw_present(fbd)) - return -ENODEV; tx_mbx = &fbd->mbx[FBNIC_IPC_MBX_TX_IDX]; - while (!tx_mbx->ready && --attempts) { + while (!tx_mbx->ready) { + if (!time_is_after_jiffies(timeout)) + return -ETIMEDOUT; + /* Force the firmware to trigger an interrupt response to * avoid the mailbox getting stuck closed if the interrupt * is reset. */ fbnic_mbx_reset_desc_ring(fbd, FBNIC_IPC_MBX_TX_IDX); - msleep(200); + /* Immediate fail if BAR4 went away */ + if (!fbnic_fw_present(fbd)) + return -ENODEV; + + msleep(20); fbnic_mbx_poll(fbd); } - return attempts ? 0 : -ETIMEDOUT; + return 0; } static void __fbnic_fw_evict_cmpl(struct fbnic_fw_completion *cmpl_data) From 1b34d1c1dc8384884febd83140c9afbc7c4b9eb8 Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Tue, 6 May 2025 09:00:18 -0700 Subject: [PATCH 306/555] fbnic: Pull fbnic_fw_xmit_cap_msg use out of interrupt context This change pulls the call to fbnic_fw_xmit_cap_msg out of fbnic_mbx_init_desc_ring and instead places it in the polling function for getting the Tx ready. Doing that we can avoid the potential issue with an interrupt coming in later from the firmware that causes it to get fired in interrupt context. Fixes: 20d2e88cc746 ("eth: fbnic: Add initial messaging to notify FW of our presence") Signed-off-by: Alexander Duyck Reviewed-by: Jacob Keller Link: https://patch.msgid.link/174654721876.499179.9839651602256668493.stgit@ahduyck-xeon-server.home.arpa Reviewed-by: Jakub Kicinski Signed-off-by: Paolo Abeni --- drivers/net/ethernet/meta/fbnic/fbnic_fw.c | 43 ++++++++-------------- 1 file changed, 16 insertions(+), 27 deletions(-) diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_fw.c b/drivers/net/ethernet/meta/fbnic/fbnic_fw.c index d344b454f28b..90a45c701543 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_fw.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_fw.c @@ -352,24 +352,6 @@ static int fbnic_fw_xmit_simple_msg(struct fbnic_dev *fbd, u32 msg_type) return err; } -/** - * fbnic_fw_xmit_cap_msg - Allocate and populate a FW capabilities message - * @fbd: FBNIC device structure - * - * Return: NULL on failure to allocate, error pointer on error, or pointer - * to new TLV test message. - * - * Sends a single TLV header indicating the host wants the firmware to - * confirm the capabilities and version. - **/ -static int fbnic_fw_xmit_cap_msg(struct fbnic_dev *fbd) -{ - int err = fbnic_fw_xmit_simple_msg(fbd, FBNIC_TLV_MSG_ID_HOST_CAP_REQ); - - /* Return 0 if we are not calling this on ASIC */ - return (err == -EOPNOTSUPP) ? 0 : err; -} - static void fbnic_mbx_init_desc_ring(struct fbnic_dev *fbd, int mbx_idx) { struct fbnic_fw_mbx *mbx = &fbd->mbx[mbx_idx]; @@ -393,15 +375,6 @@ static void fbnic_mbx_init_desc_ring(struct fbnic_dev *fbd, int mbx_idx) /* Enable DMA reads from the device */ wr32(fbd, FBNIC_PUL_OB_TLP_HDR_AR_CFG, FBNIC_PUL_OB_TLP_HDR_AR_CFG_BME); - - /* Force version to 1 if we successfully requested an update - * from the firmware. This should be overwritten once we get - * the actual version from the firmware in the capabilities - * request message. - */ - if (!fbnic_fw_xmit_cap_msg(fbd) && - !fbd->fw_cap.running.mgmt.version) - fbd->fw_cap.running.mgmt.version = 1; break; } } @@ -907,6 +880,7 @@ int fbnic_mbx_poll_tx_ready(struct fbnic_dev *fbd) { unsigned long timeout = jiffies + 10 * HZ + 1; struct fbnic_fw_mbx *tx_mbx; + int err; tx_mbx = &fbd->mbx[FBNIC_IPC_MBX_TX_IDX]; while (!tx_mbx->ready) { @@ -928,7 +902,22 @@ int fbnic_mbx_poll_tx_ready(struct fbnic_dev *fbd) fbnic_mbx_poll(fbd); } + /* Request an update from the firmware. This should overwrite + * mgmt.version once we get the actual version from the firmware + * in the capabilities request message. + */ + err = fbnic_fw_xmit_simple_msg(fbd, FBNIC_TLV_MSG_ID_HOST_CAP_REQ); + if (err) + goto clean_mbx; + + /* Use "1" to indicate we entered the state waiting for a response */ + fbd->fw_cap.running.mgmt.version = 1; + return 0; +clean_mbx: + /* Cleanup Rx buffers and disable mailbox */ + fbnic_mbx_clean(fbd); + return err; } static void __fbnic_fw_evict_cmpl(struct fbnic_fw_completion *cmpl_data) From ce2fa1dba204c761582674cf2eb9cbe0b949b5c7 Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Tue, 6 May 2025 09:00:25 -0700 Subject: [PATCH 307/555] fbnic: Do not allow mailbox to toggle to ready outside fbnic_mbx_poll_tx_ready We had originally thought to have the mailbox go to ready in the background while we were doing other things. One issue with this though is that we can't disable it by clearing the ready state without also blocking interrupts or calls to mbx_poll as it will just pop back to life during an interrupt. In order to prevent that from happening we can pull the code for toggling to ready out of the interrupt path and instead place it in the fbnic_mbx_poll_tx_ready path so that it becomes the only spot where the Rx/Tx can toggle to the ready state. By doing this we can prevent races where we disable the DMA and/or free buffers only to have an interrupt fire and undo what we have done. Fixes: da3cde08209e ("eth: fbnic: Add FW communication mechanism") Signed-off-by: Alexander Duyck Reviewed-by: Jacob Keller Link: https://patch.msgid.link/174654722518.499179.11612865740376848478.stgit@ahduyck-xeon-server.home.arpa Reviewed-by: Jakub Kicinski Signed-off-by: Paolo Abeni --- drivers/net/ethernet/meta/fbnic/fbnic_fw.c | 27 ++++++++-------------- 1 file changed, 10 insertions(+), 17 deletions(-) diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_fw.c b/drivers/net/ethernet/meta/fbnic/fbnic_fw.c index 90a45c701543..3d9636a6c968 100644 --- a/drivers/net/ethernet/meta/fbnic/fbnic_fw.c +++ b/drivers/net/ethernet/meta/fbnic/fbnic_fw.c @@ -356,10 +356,6 @@ static void fbnic_mbx_init_desc_ring(struct fbnic_dev *fbd, int mbx_idx) { struct fbnic_fw_mbx *mbx = &fbd->mbx[mbx_idx]; - /* This is a one time init, so just exit if it is completed */ - if (mbx->ready) - return; - mbx->ready = true; switch (mbx_idx) { @@ -379,21 +375,18 @@ static void fbnic_mbx_init_desc_ring(struct fbnic_dev *fbd, int mbx_idx) } } -static void fbnic_mbx_postinit(struct fbnic_dev *fbd) +static bool fbnic_mbx_event(struct fbnic_dev *fbd) { - int i; - /* We only need to do this on the first interrupt following reset. * this primes the mailbox so that we will have cleared all the * skip descriptors. */ if (!(rd32(fbd, FBNIC_INTR_STATUS(0)) & (1u << FBNIC_FW_MSIX_ENTRY))) - return; + return false; wr32(fbd, FBNIC_INTR_CLEAR(0), 1u << FBNIC_FW_MSIX_ENTRY); - for (i = 0; i < FBNIC_IPC_MBX_INDICES; i++) - fbnic_mbx_init_desc_ring(fbd, i); + return true; } /** @@ -870,7 +863,7 @@ next_page: void fbnic_mbx_poll(struct fbnic_dev *fbd) { - fbnic_mbx_postinit(fbd); + fbnic_mbx_event(fbd); fbnic_mbx_process_tx_msgs(fbd); fbnic_mbx_process_rx_msgs(fbd); @@ -879,11 +872,9 @@ void fbnic_mbx_poll(struct fbnic_dev *fbd) int fbnic_mbx_poll_tx_ready(struct fbnic_dev *fbd) { unsigned long timeout = jiffies + 10 * HZ + 1; - struct fbnic_fw_mbx *tx_mbx; - int err; + int err, i; - tx_mbx = &fbd->mbx[FBNIC_IPC_MBX_TX_IDX]; - while (!tx_mbx->ready) { + do { if (!time_is_after_jiffies(timeout)) return -ETIMEDOUT; @@ -898,9 +889,11 @@ int fbnic_mbx_poll_tx_ready(struct fbnic_dev *fbd) return -ENODEV; msleep(20); + } while (!fbnic_mbx_event(fbd)); - fbnic_mbx_poll(fbd); - } + /* FW has shown signs of life. Enable DMA and start Tx/Rx */ + for (i = 0; i < FBNIC_IPC_MBX_INDICES; i++) + fbnic_mbx_init_desc_ring(fbd, i); /* Request an update from the firmware. This should overwrite * mgmt.version once we get the actual version from the firmware From 23fa6a23d97182d36ca3c71e43c804fa91e46a03 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 6 May 2025 17:32:20 -0700 Subject: [PATCH 308/555] net: export a helper for adding up queue stats Older drivers and drivers with lower queue counts often have a static array of queues, rather than allocating structs for each queue on demand. Add a helper for adding up qstats from a queue range. Expectation is that driver will pass a queue range [netdev->real_num_*x_queues, MAX). It was tempting to always use num_*x_queues as the end, but virtio seems to clamp its queue count after allocating the netdev. And this way we can trivaly reuse the helper for [0, real_..). Signed-off-by: Jakub Kicinski Link: https://patch.msgid.link/20250507003221.823267-2-kuba@kernel.org Signed-off-by: Paolo Abeni --- include/net/netdev_queues.h | 6 ++++ net/core/netdev-genl.c | 69 +++++++++++++++++++++++++++---------- 2 files changed, 56 insertions(+), 19 deletions(-) diff --git a/include/net/netdev_queues.h b/include/net/netdev_queues.h index 825141d675e5..d01c82983b4d 100644 --- a/include/net/netdev_queues.h +++ b/include/net/netdev_queues.h @@ -103,6 +103,12 @@ struct netdev_stat_ops { struct netdev_queue_stats_tx *tx); }; +void netdev_stat_queue_sum(struct net_device *netdev, + int rx_start, int rx_end, + struct netdev_queue_stats_rx *rx_sum, + int tx_start, int tx_end, + struct netdev_queue_stats_tx *tx_sum); + /** * struct netdev_queue_mgmt_ops - netdev ops for queue management * diff --git a/net/core/netdev-genl.c b/net/core/netdev-genl.c index 230743bdbb14..dae9f0d432fb 100644 --- a/net/core/netdev-genl.c +++ b/net/core/netdev-genl.c @@ -708,25 +708,66 @@ netdev_nl_stats_by_queue(struct net_device *netdev, struct sk_buff *rsp, return 0; } +/** + * netdev_stat_queue_sum() - add up queue stats from range of queues + * @netdev: net_device + * @rx_start: index of the first Rx queue to query + * @rx_end: index after the last Rx queue (first *not* to query) + * @rx_sum: output Rx stats, should be already initialized + * @tx_start: index of the first Tx queue to query + * @tx_end: index after the last Tx queue (first *not* to query) + * @tx_sum: output Tx stats, should be already initialized + * + * Add stats from [start, end) range of queue IDs to *x_sum structs. + * The sum structs must be already initialized. Usually this + * helper is invoked from the .get_base_stats callbacks of drivers + * to account for stats of disabled queues. In that case the ranges + * are usually [netdev->real_num_*x_queues, netdev->num_*x_queues). + */ +void netdev_stat_queue_sum(struct net_device *netdev, + int rx_start, int rx_end, + struct netdev_queue_stats_rx *rx_sum, + int tx_start, int tx_end, + struct netdev_queue_stats_tx *tx_sum) +{ + const struct netdev_stat_ops *ops; + struct netdev_queue_stats_rx rx; + struct netdev_queue_stats_tx tx; + int i; + + ops = netdev->stat_ops; + + for (i = rx_start; i < rx_end; i++) { + memset(&rx, 0xff, sizeof(rx)); + if (ops->get_queue_stats_rx) + ops->get_queue_stats_rx(netdev, i, &rx); + netdev_nl_stats_add(rx_sum, &rx, sizeof(rx)); + } + for (i = tx_start; i < tx_end; i++) { + memset(&tx, 0xff, sizeof(tx)); + if (ops->get_queue_stats_tx) + ops->get_queue_stats_tx(netdev, i, &tx); + netdev_nl_stats_add(tx_sum, &tx, sizeof(tx)); + } +} +EXPORT_SYMBOL(netdev_stat_queue_sum); + static int netdev_nl_stats_by_netdev(struct net_device *netdev, struct sk_buff *rsp, const struct genl_info *info) { - struct netdev_queue_stats_rx rx_sum, rx; - struct netdev_queue_stats_tx tx_sum, tx; - const struct netdev_stat_ops *ops; + struct netdev_queue_stats_rx rx_sum; + struct netdev_queue_stats_tx tx_sum; void *hdr; - int i; - ops = netdev->stat_ops; /* Netdev can't guarantee any complete counters */ - if (!ops->get_base_stats) + if (!netdev->stat_ops->get_base_stats) return 0; memset(&rx_sum, 0xff, sizeof(rx_sum)); memset(&tx_sum, 0xff, sizeof(tx_sum)); - ops->get_base_stats(netdev, &rx_sum, &tx_sum); + netdev->stat_ops->get_base_stats(netdev, &rx_sum, &tx_sum); /* The op was there, but nothing reported, don't bother */ if (!memchr_inv(&rx_sum, 0xff, sizeof(rx_sum)) && @@ -739,18 +780,8 @@ netdev_nl_stats_by_netdev(struct net_device *netdev, struct sk_buff *rsp, if (nla_put_u32(rsp, NETDEV_A_QSTATS_IFINDEX, netdev->ifindex)) goto nla_put_failure; - for (i = 0; i < netdev->real_num_rx_queues; i++) { - memset(&rx, 0xff, sizeof(rx)); - if (ops->get_queue_stats_rx) - ops->get_queue_stats_rx(netdev, i, &rx); - netdev_nl_stats_add(&rx_sum, &rx, sizeof(rx)); - } - for (i = 0; i < netdev->real_num_tx_queues; i++) { - memset(&tx, 0xff, sizeof(tx)); - if (ops->get_queue_stats_tx) - ops->get_queue_stats_tx(netdev, i, &tx); - netdev_nl_stats_add(&tx_sum, &tx, sizeof(tx)); - } + netdev_stat_queue_sum(netdev, 0, netdev->real_num_rx_queues, &rx_sum, + 0, netdev->real_num_tx_queues, &tx_sum); if (netdev_nl_stats_write_rx(rsp, &rx_sum) || netdev_nl_stats_write_tx(rsp, &tx_sum)) From 001160ec8c59115efc39e197d40829bdafd4d7f5 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 6 May 2025 17:32:21 -0700 Subject: [PATCH 309/555] virtio-net: fix total qstat values NIPA tests report that the interface statistics reported via qstat are lower than those reported via ip link. Looks like this is because some tests flip the queue count up and down, and we end up with some of the traffic accounted on disabled queues. Add up counters from disabled queues. Fixes: d888f04c09bb ("virtio-net: support queue stat") Signed-off-by: Jakub Kicinski Link: https://patch.msgid.link/20250507003221.823267-3-kuba@kernel.org Signed-off-by: Paolo Abeni --- drivers/net/virtio_net.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index f9e3e628ec4d..e53ba600605a 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -5678,6 +5678,10 @@ static void virtnet_get_base_stats(struct net_device *dev, if (vi->device_stats_cap & VIRTIO_NET_STATS_TYPE_TX_SPEED) tx->hw_drop_ratelimits = 0; + + netdev_stat_queue_sum(dev, + dev->real_num_rx_queues, vi->max_queue_pairs, rx, + dev->real_num_tx_queues, vi->max_queue_pairs, tx); } static const struct netdev_stat_ops virtnet_stat_ops = { From 10aba126bc86904e2f5afeaa26354e877a593c95 Mon Sep 17 00:00:00 2001 From: Andi Shyti Date: Tue, 6 May 2025 01:15:11 +0200 Subject: [PATCH 310/555] MAINTAINERS: Remove entry for Seth Heasley Seth's mails bounce back, remove his maintainership. Signed-off-by: Andi Shyti Link: https://lore.kernel.org/r/20250505231511.3175151-1-andi.shyti@kernel.org --- MAINTAINERS | 1 - 1 file changed, 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 69511c3b2b76..70229ed0c55a 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -11235,7 +11235,6 @@ S: Maintained F: drivers/i2c/busses/i2c-cht-wc.c I2C/SMBUS ISMT DRIVER -M: Seth Heasley M: Neil Horman L: linux-i2c@vger.kernel.org F: Documentation/i2c/busses/i2c-ismt.rst From fd94de9f9e7aac11ec659e386b9db1203d502023 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Cl=C3=A9ment=20L=C3=A9ger?= Date: Tue, 22 Apr 2025 18:23:08 +0200 Subject: [PATCH 311/555] riscv: misaligned: factorize trap handling MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Since both load/store and user/kernel should use almost the same path and that we are going to add some code around that, factorize it. Signed-off-by: Clément Léger Reviewed-by: Alexandre Ghiti Link: https://lore.kernel.org/r/20250422162324.956065-2-cleger@rivosinc.com Signed-off-by: Alexandre Ghiti --- arch/riscv/kernel/traps.c | 74 +++++++++++++++++++++------------------ 1 file changed, 40 insertions(+), 34 deletions(-) diff --git a/arch/riscv/kernel/traps.c b/arch/riscv/kernel/traps.c index 8ff8e8b36524..b1d991c78a23 100644 --- a/arch/riscv/kernel/traps.c +++ b/arch/riscv/kernel/traps.c @@ -198,47 +198,53 @@ asmlinkage __visible __trap_section void do_trap_insn_illegal(struct pt_regs *re DO_ERROR_INFO(do_trap_load_fault, SIGSEGV, SEGV_ACCERR, "load access fault"); +enum misaligned_access_type { + MISALIGNED_STORE, + MISALIGNED_LOAD, +}; +static const struct { + const char *type_str; + int (*handler)(struct pt_regs *regs); +} misaligned_handler[] = { + [MISALIGNED_STORE] = { + .type_str = "Oops - store (or AMO) address misaligned", + .handler = handle_misaligned_store, + }, + [MISALIGNED_LOAD] = { + .type_str = "Oops - load address misaligned", + .handler = handle_misaligned_load, + }, +}; + +static void do_trap_misaligned(struct pt_regs *regs, enum misaligned_access_type type) +{ + irqentry_state_t state; + + if (user_mode(regs)) + irqentry_enter_from_user_mode(regs); + else + state = irqentry_nmi_enter(regs); + + if (misaligned_handler[type].handler(regs)) + do_trap_error(regs, SIGBUS, BUS_ADRALN, regs->epc, + misaligned_handler[type].type_str); + + if (user_mode(regs)) + irqentry_exit_to_user_mode(regs); + else + irqentry_nmi_exit(regs, state); +} + asmlinkage __visible __trap_section void do_trap_load_misaligned(struct pt_regs *regs) { - if (user_mode(regs)) { - irqentry_enter_from_user_mode(regs); - - if (handle_misaligned_load(regs)) - do_trap_error(regs, SIGBUS, BUS_ADRALN, regs->epc, - "Oops - load address misaligned"); - - irqentry_exit_to_user_mode(regs); - } else { - irqentry_state_t state = irqentry_nmi_enter(regs); - - if (handle_misaligned_load(regs)) - do_trap_error(regs, SIGBUS, BUS_ADRALN, regs->epc, - "Oops - load address misaligned"); - - irqentry_nmi_exit(regs, state); - } + do_trap_misaligned(regs, MISALIGNED_LOAD); } asmlinkage __visible __trap_section void do_trap_store_misaligned(struct pt_regs *regs) { - if (user_mode(regs)) { - irqentry_enter_from_user_mode(regs); - - if (handle_misaligned_store(regs)) - do_trap_error(regs, SIGBUS, BUS_ADRALN, regs->epc, - "Oops - store (or AMO) address misaligned"); - - irqentry_exit_to_user_mode(regs); - } else { - irqentry_state_t state = irqentry_nmi_enter(regs); - - if (handle_misaligned_store(regs)) - do_trap_error(regs, SIGBUS, BUS_ADRALN, regs->epc, - "Oops - store (or AMO) address misaligned"); - - irqentry_nmi_exit(regs, state); - } + do_trap_misaligned(regs, MISALIGNED_STORE); } + DO_ERROR_INFO(do_trap_store_fault, SIGSEGV, SEGV_ACCERR, "store (or AMO) access fault"); DO_ERROR_INFO(do_trap_ecall_s, From 453805f0a28fc5091e46145e6560c776f7c7a611 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Cl=C3=A9ment=20L=C3=A9ger?= Date: Tue, 22 Apr 2025 18:23:09 +0200 Subject: [PATCH 312/555] riscv: misaligned: enable IRQs while handling misaligned accesses MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit We can safely reenable IRQs if coming from userspace. This allows to access user memory that could potentially trigger a page fault. Fixes: b686ecdeacf6 ("riscv: misaligned: Restrict user access to kernel memory") Signed-off-by: Clément Léger Reviewed-by: Alexandre Ghiti Link: https://lore.kernel.org/r/20250422162324.956065-3-cleger@rivosinc.com Signed-off-by: Alexandre Ghiti --- arch/riscv/kernel/traps.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/arch/riscv/kernel/traps.c b/arch/riscv/kernel/traps.c index b1d991c78a23..9c83848797a7 100644 --- a/arch/riscv/kernel/traps.c +++ b/arch/riscv/kernel/traps.c @@ -220,19 +220,23 @@ static void do_trap_misaligned(struct pt_regs *regs, enum misaligned_access_type { irqentry_state_t state; - if (user_mode(regs)) + if (user_mode(regs)) { irqentry_enter_from_user_mode(regs); - else + local_irq_enable(); + } else { state = irqentry_nmi_enter(regs); + } if (misaligned_handler[type].handler(regs)) do_trap_error(regs, SIGBUS, BUS_ADRALN, regs->epc, misaligned_handler[type].type_str); - if (user_mode(regs)) + if (user_mode(regs)) { + local_irq_disable(); irqentry_exit_to_user_mode(regs); - else + } else { irqentry_nmi_exit(regs, state); + } } asmlinkage __visible __trap_section void do_trap_load_misaligned(struct pt_regs *regs) From 897e8aece3c8a1a66b3eae58df1832a524d45319 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Cl=C3=A9ment=20L=C3=A9ger?= Date: Tue, 22 Apr 2025 18:23:10 +0200 Subject: [PATCH 313/555] riscv: misaligned: use get_user() instead of __get_user() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Now that we can safely handle user memory accesses while in the misaligned access handlers, use get_user() instead of __get_user() to have user memory access checks. Signed-off-by: Clément Léger Reviewed-by: Alexandre Ghiti Link: https://lore.kernel.org/r/20250422162324.956065-4-cleger@rivosinc.com Signed-off-by: Alexandre Ghiti --- arch/riscv/kernel/traps_misaligned.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/riscv/kernel/traps_misaligned.c b/arch/riscv/kernel/traps_misaligned.c index dde5d11dc1b5..77c788660223 100644 --- a/arch/riscv/kernel/traps_misaligned.c +++ b/arch/riscv/kernel/traps_misaligned.c @@ -275,7 +275,7 @@ static unsigned long get_f32_rs(unsigned long insn, u8 fp_reg_offset, int __ret; \ \ if (user_mode(regs)) { \ - __ret = __get_user(insn, (type __user *) insn_addr); \ + __ret = get_user(insn, (type __user *) insn_addr); \ } else { \ insn = *(type *)insn_addr; \ __ret = 0; \ From ae08d55807c099357c047dba17624b09414635dd Mon Sep 17 00:00:00 2001 From: Nam Cao Date: Sun, 4 May 2025 12:19:20 +0200 Subject: [PATCH 314/555] riscv: Fix kernel crash due to PR_SET_TAGGED_ADDR_CTRL When userspace does PR_SET_TAGGED_ADDR_CTRL, but Supm extension is not available, the kernel crashes: Oops - illegal instruction [#1] [snip] epc : set_tagged_addr_ctrl+0x112/0x15a ra : set_tagged_addr_ctrl+0x74/0x15a epc : ffffffff80011ace ra : ffffffff80011a30 sp : ffffffc60039be10 [snip] status: 0000000200000120 badaddr: 0000000010a79073 cause: 0000000000000002 set_tagged_addr_ctrl+0x112/0x15a __riscv_sys_prctl+0x352/0x73c do_trap_ecall_u+0x17c/0x20c andle_exception+0x150/0x15c Fix it by checking if Supm is available. Fixes: 09d6775f503b ("riscv: Add support for userspace pointer masking") Signed-off-by: Nam Cao Cc: stable@vger.kernel.org Reviewed-by: Samuel Holland Link: https://lore.kernel.org/r/20250504101920.3393053-1-namcao@linutronix.de Signed-off-by: Alexandre Ghiti --- arch/riscv/kernel/process.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/arch/riscv/kernel/process.c b/arch/riscv/kernel/process.c index 7c244de77180..3db2c0c07acd 100644 --- a/arch/riscv/kernel/process.c +++ b/arch/riscv/kernel/process.c @@ -275,6 +275,9 @@ long set_tagged_addr_ctrl(struct task_struct *task, unsigned long arg) unsigned long pmm; u8 pmlen; + if (!riscv_has_extension_unlikely(RISCV_ISA_EXT_SUPM)) + return -EINVAL; + if (is_compat_thread(ti)) return -EINVAL; From e9d86b8e17e725ee6088970981ab99f29abd1997 Mon Sep 17 00:00:00 2001 From: Alexandre Ghiti Date: Tue, 8 Apr 2025 09:28:51 +0200 Subject: [PATCH 315/555] scripts: Do not strip .rela.dyn section MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The .rela.dyn section contains runtime relocations and is only emitted for a relocatable kernel. riscv uses this section to relocate the kernel at runtime but that section is stripped from vmlinux. That prevents kexec to successfully load vmlinux since it does not contain the relocations info needed. Fixes: 559d1e45a16d ("riscv: Use --emit-relocs in order to move .rela.dyn in init") Tested-by: Björn Töpel Reviewed-by: Björn Töpel Acked-by: Ard Biesheuvel Link: https://lore.kernel.org/r/20250408072851.90275-1-alexghiti@rivosinc.com Signed-off-by: Alexandre Ghiti --- scripts/Makefile.vmlinux | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/Makefile.vmlinux b/scripts/Makefile.vmlinux index b0a6cd5b818c..85d60d986401 100644 --- a/scripts/Makefile.vmlinux +++ b/scripts/Makefile.vmlinux @@ -13,7 +13,7 @@ ifdef CONFIG_ARCH_VMLINUX_NEEDS_RELOCS vmlinux-final := vmlinux.unstripped quiet_cmd_strip_relocs = RSTRIP $@ - cmd_strip_relocs = $(OBJCOPY) --remove-section='.rel*' $< $@ + cmd_strip_relocs = $(OBJCOPY) --remove-section='.rel*' --remove-section=!'.rel*.dyn' $< $@ vmlinux: $(vmlinux-final) FORCE $(call if_changed,strip_relocs) From 7f1c3de1370bc6a8ad5157336b258067dac0ae9c Mon Sep 17 00:00:00 2001 From: Samuel Holland Date: Wed, 7 May 2025 07:52:18 -0700 Subject: [PATCH 316/555] riscv: Disallow PR_GET_TAGGED_ADDR_CTRL without Supm When the prctl() interface for pointer masking was added, it did not check that the pointer masking ISA extension was supported, only the individual submodes. Userspace could still attempt to disable pointer masking and query the pointer masking state. commit 81de1afb2dd1 ("riscv: Fix kernel crash due to PR_SET_TAGGED_ADDR_CTRL") disallowed the former, as the senvcfg write could crash on older systems. PR_GET_TAGGED_ADDR_CTRL state does not crash, because it reads only kernel-internal state and not senvcfg, but it should still be disallowed for consistency. Fixes: 09d6775f503b ("riscv: Add support for userspace pointer masking") Signed-off-by: Samuel Holland Reviewed-by: Nam Cao Link: https://lore.kernel.org/r/20250507145230.2272871-1-samuel.holland@sifive.com Signed-off-by: Alexandre Ghiti --- arch/riscv/kernel/process.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/arch/riscv/kernel/process.c b/arch/riscv/kernel/process.c index 3db2c0c07acd..15d8f75902f8 100644 --- a/arch/riscv/kernel/process.c +++ b/arch/riscv/kernel/process.c @@ -333,6 +333,9 @@ long get_tagged_addr_ctrl(struct task_struct *task) struct thread_info *ti = task_thread_info(task); long ret = 0; + if (!riscv_has_extension_unlikely(RISCV_ISA_EXT_SUPM)) + return -EINVAL; + if (is_compat_thread(ti)) return -EINVAL; From e3417ab75ab2e7dca6372a1bfa26b1be3ac5889e Mon Sep 17 00:00:00 2001 From: Sean Christopherson Date: Mon, 5 May 2025 11:03:00 -0700 Subject: [PATCH 317/555] KVM: SVM: Set/clear SRSO's BP_SPEC_REDUCE on 0 <=> 1 VM count transitions Set the magic BP_SPEC_REDUCE bit to mitigate SRSO when running VMs if and only if KVM has at least one active VM. Leaving the bit set at all times unfortunately degrades performance by a wee bit more than expected. Use a dedicated spinlock and counter instead of hooking virtualization enablement, as changing the behavior of kvm.enable_virt_at_load based on SRSO_BP_SPEC_REDUCE is painful, and has its own drawbacks, e.g. could result in performance issues for flows that are sensitive to VM creation latency. Defer setting BP_SPEC_REDUCE until VMRUN is imminent to avoid impacting performance on CPUs that aren't running VMs, e.g. if a setup is using housekeeping CPUs. Setting BP_SPEC_REDUCE in task context, i.e. without blasting IPIs to all CPUs, also helps avoid serializing 1<=>N transitions without incurring a gross amount of complexity (see the Link for details on how ugly coordinating via IPIs gets). Link: https://lore.kernel.org/all/aBOnzNCngyS_pQIW@google.com Fixes: 8442df2b49ed ("x86/bugs: KVM: Add support for SRSO_MSR_FIX") Reported-by: Michael Larabel Closes: https://www.phoronix.com/review/linux-615-amd-regression Cc: Borislav Petkov Tested-by: Borislav Petkov (AMD) Link: https://lore.kernel.org/r/20250505180300.973137-1-seanjc@google.com Signed-off-by: Sean Christopherson --- arch/x86/kvm/svm/svm.c | 71 ++++++++++++++++++++++++++++++++++++++---- arch/x86/kvm/svm/svm.h | 2 ++ 2 files changed, 67 insertions(+), 6 deletions(-) diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c index c5470d842aed..a89c271a1951 100644 --- a/arch/x86/kvm/svm/svm.c +++ b/arch/x86/kvm/svm/svm.c @@ -607,9 +607,6 @@ static void svm_disable_virtualization_cpu(void) kvm_cpu_svm_disable(); amd_pmu_disable_virt(); - - if (cpu_feature_enabled(X86_FEATURE_SRSO_BP_SPEC_REDUCE)) - msr_clear_bit(MSR_ZEN4_BP_CFG, MSR_ZEN4_BP_CFG_BP_SPEC_REDUCE_BIT); } static int svm_enable_virtualization_cpu(void) @@ -687,9 +684,6 @@ static int svm_enable_virtualization_cpu(void) rdmsr(MSR_TSC_AUX, sev_es_host_save_area(sd)->tsc_aux, msr_hi); } - if (cpu_feature_enabled(X86_FEATURE_SRSO_BP_SPEC_REDUCE)) - msr_set_bit(MSR_ZEN4_BP_CFG, MSR_ZEN4_BP_CFG_BP_SPEC_REDUCE_BIT); - return 0; } @@ -1518,6 +1512,63 @@ static void svm_vcpu_free(struct kvm_vcpu *vcpu) __free_pages(virt_to_page(svm->msrpm), get_order(MSRPM_SIZE)); } +#ifdef CONFIG_CPU_MITIGATIONS +static DEFINE_SPINLOCK(srso_lock); +static atomic_t srso_nr_vms; + +static void svm_srso_clear_bp_spec_reduce(void *ign) +{ + struct svm_cpu_data *sd = this_cpu_ptr(&svm_data); + + if (!sd->bp_spec_reduce_set) + return; + + msr_clear_bit(MSR_ZEN4_BP_CFG, MSR_ZEN4_BP_CFG_BP_SPEC_REDUCE_BIT); + sd->bp_spec_reduce_set = false; +} + +static void svm_srso_vm_destroy(void) +{ + if (!cpu_feature_enabled(X86_FEATURE_SRSO_BP_SPEC_REDUCE)) + return; + + if (atomic_dec_return(&srso_nr_vms)) + return; + + guard(spinlock)(&srso_lock); + + /* + * Verify a new VM didn't come along, acquire the lock, and increment + * the count before this task acquired the lock. + */ + if (atomic_read(&srso_nr_vms)) + return; + + on_each_cpu(svm_srso_clear_bp_spec_reduce, NULL, 1); +} + +static void svm_srso_vm_init(void) +{ + if (!cpu_feature_enabled(X86_FEATURE_SRSO_BP_SPEC_REDUCE)) + return; + + /* + * Acquire the lock on 0 => 1 transitions to ensure a potential 1 => 0 + * transition, i.e. destroying the last VM, is fully complete, e.g. so + * that a delayed IPI doesn't clear BP_SPEC_REDUCE after a vCPU runs. + */ + if (atomic_inc_not_zero(&srso_nr_vms)) + return; + + guard(spinlock)(&srso_lock); + + atomic_inc(&srso_nr_vms); +} +#else +static void svm_srso_vm_init(void) { } +static void svm_srso_vm_destroy(void) { } +#endif + static void svm_prepare_switch_to_guest(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); @@ -1550,6 +1601,11 @@ static void svm_prepare_switch_to_guest(struct kvm_vcpu *vcpu) (!boot_cpu_has(X86_FEATURE_V_TSC_AUX) || !sev_es_guest(vcpu->kvm))) kvm_set_user_return_msr(tsc_aux_uret_slot, svm->tsc_aux, -1ull); + if (cpu_feature_enabled(X86_FEATURE_SRSO_BP_SPEC_REDUCE) && + !sd->bp_spec_reduce_set) { + sd->bp_spec_reduce_set = true; + msr_set_bit(MSR_ZEN4_BP_CFG, MSR_ZEN4_BP_CFG_BP_SPEC_REDUCE_BIT); + } svm->guest_state_loaded = true; } @@ -5040,6 +5096,8 @@ static void svm_vm_destroy(struct kvm *kvm) { avic_vm_destroy(kvm); sev_vm_destroy(kvm); + + svm_srso_vm_destroy(); } static int svm_vm_init(struct kvm *kvm) @@ -5065,6 +5123,7 @@ static int svm_vm_init(struct kvm *kvm) return ret; } + svm_srso_vm_init(); return 0; } diff --git a/arch/x86/kvm/svm/svm.h b/arch/x86/kvm/svm/svm.h index d4490eaed55d..f16b068c4228 100644 --- a/arch/x86/kvm/svm/svm.h +++ b/arch/x86/kvm/svm/svm.h @@ -335,6 +335,8 @@ struct svm_cpu_data { u32 next_asid; u32 min_asid; + bool bp_spec_reduce_set; + struct vmcb *save_area; unsigned long save_area_pa; From c82b6357a5465a3222780ac5d3edcdfb02208cc3 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Wed, 30 Apr 2025 15:07:03 -0400 Subject: [PATCH 318/555] Bluetooth: hci_event: Fix not using key encryption size when its known This fixes the regression introduced by 50c1241e6a8a ("Bluetooth: l2cap: Check encryption key size on incoming connection") introduced a check for l2cap_check_enc_key_size which checks for hcon->enc_key_size which may not be initialized if HCI_OP_READ_ENC_KEY_SIZE is still pending. If the key encryption size is known, due previously reading it using HCI_OP_READ_ENC_KEY_SIZE, then store it as part of link_key/smp_ltk structures so the next time the encryption is changed their values are used as conn->enc_key_size thus avoiding the racing against HCI_OP_READ_ENC_KEY_SIZE. Now that the enc_size is stored as part of key the information the code then attempts to check that there is no downgrade of security if HCI_OP_READ_ENC_KEY_SIZE returns a value smaller than what has been previously stored. Link: https://bugzilla.kernel.org/show_bug.cgi?id=220061 Link: https://bugzilla.kernel.org/show_bug.cgi?id=220063 Fixes: 522e9ed157e3 ("Bluetooth: l2cap: Check encryption key size on incoming connection") Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/hci_core.h | 1 + net/bluetooth/hci_conn.c | 24 +++++++++++ net/bluetooth/hci_event.c | 73 ++++++++++++++++++-------------- 3 files changed, 67 insertions(+), 31 deletions(-) diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 522d837a23fa..54bfeeaa0995 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -1798,6 +1798,7 @@ struct hci_conn_params *hci_pend_le_action_lookup(struct list_head *list, void hci_uuids_clear(struct hci_dev *hdev); void hci_link_keys_clear(struct hci_dev *hdev); +u8 *hci_conn_key_enc_size(struct hci_conn *conn); struct link_key *hci_find_link_key(struct hci_dev *hdev, bdaddr_t *bdaddr); struct link_key *hci_add_link_key(struct hci_dev *hdev, struct hci_conn *conn, bdaddr_t *bdaddr, u8 *val, u8 type, diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index 6533e281ada3..946d2ae551f8 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -3023,3 +3023,27 @@ void hci_conn_tx_dequeue(struct hci_conn *conn) kfree_skb(skb); } + +u8 *hci_conn_key_enc_size(struct hci_conn *conn) +{ + if (conn->type == ACL_LINK) { + struct link_key *key; + + key = hci_find_link_key(conn->hdev, &conn->dst); + if (!key) + return NULL; + + return &key->pin_len; + } else if (conn->type == LE_LINK) { + struct smp_ltk *ltk; + + ltk = hci_find_ltk(conn->hdev, &conn->dst, conn->dst_type, + conn->role); + if (!ltk) + return NULL; + + return <k->enc_size; + } + + return NULL; +} diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index 6d6061111ac5..c38ada69c3d7 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -739,10 +739,17 @@ static u8 hci_cc_read_enc_key_size(struct hci_dev *hdev, void *data, handle); conn->enc_key_size = 0; } else { + u8 *key_enc_size = hci_conn_key_enc_size(conn); + conn->enc_key_size = rp->key_size; status = 0; - if (conn->enc_key_size < hdev->min_enc_key_size) { + /* Attempt to check if the key size is too small or if it has + * been downgraded from the last time it was stored as part of + * the link_key. + */ + if (conn->enc_key_size < hdev->min_enc_key_size || + (key_enc_size && conn->enc_key_size < *key_enc_size)) { /* As slave role, the conn->state has been set to * BT_CONNECTED and l2cap conn req might not be received * yet, at this moment the l2cap layer almost does @@ -755,6 +762,10 @@ static u8 hci_cc_read_enc_key_size(struct hci_dev *hdev, void *data, clear_bit(HCI_CONN_ENCRYPT, &conn->flags); clear_bit(HCI_CONN_AES_CCM, &conn->flags); } + + /* Update the key encryption size with the connection one */ + if (key_enc_size && *key_enc_size != conn->enc_key_size) + *key_enc_size = conn->enc_key_size; } hci_encrypt_cfm(conn, status); @@ -3065,6 +3076,34 @@ static void hci_inquiry_result_evt(struct hci_dev *hdev, void *edata, hci_dev_unlock(hdev); } +static int hci_read_enc_key_size(struct hci_dev *hdev, struct hci_conn *conn) +{ + struct hci_cp_read_enc_key_size cp; + u8 *key_enc_size = hci_conn_key_enc_size(conn); + + if (!read_key_size_capable(hdev)) { + conn->enc_key_size = HCI_LINK_KEY_SIZE; + return -EOPNOTSUPP; + } + + bt_dev_dbg(hdev, "hcon %p", conn); + + memset(&cp, 0, sizeof(cp)); + cp.handle = cpu_to_le16(conn->handle); + + /* If the key enc_size is already known, use it as conn->enc_key_size, + * otherwise use hdev->min_enc_key_size so the likes of + * l2cap_check_enc_key_size don't fail while waiting for + * HCI_OP_READ_ENC_KEY_SIZE response. + */ + if (key_enc_size && *key_enc_size) + conn->enc_key_size = *key_enc_size; + else + conn->enc_key_size = hdev->min_enc_key_size; + + return hci_send_cmd(hdev, HCI_OP_READ_ENC_KEY_SIZE, sizeof(cp), &cp); +} + static void hci_conn_complete_evt(struct hci_dev *hdev, void *data, struct sk_buff *skb) { @@ -3157,23 +3196,11 @@ static void hci_conn_complete_evt(struct hci_dev *hdev, void *data, if (ev->encr_mode == 1 && !test_bit(HCI_CONN_ENCRYPT, &conn->flags) && ev->link_type == ACL_LINK) { struct link_key *key; - struct hci_cp_read_enc_key_size cp; key = hci_find_link_key(hdev, &ev->bdaddr); if (key) { set_bit(HCI_CONN_ENCRYPT, &conn->flags); - - if (!read_key_size_capable(hdev)) { - conn->enc_key_size = HCI_LINK_KEY_SIZE; - } else { - cp.handle = cpu_to_le16(conn->handle); - if (hci_send_cmd(hdev, HCI_OP_READ_ENC_KEY_SIZE, - sizeof(cp), &cp)) { - bt_dev_err(hdev, "sending read key size failed"); - conn->enc_key_size = HCI_LINK_KEY_SIZE; - } - } - + hci_read_enc_key_size(hdev, conn); hci_encrypt_cfm(conn, ev->status); } } @@ -3612,24 +3639,8 @@ static void hci_encrypt_change_evt(struct hci_dev *hdev, void *data, /* Try reading the encryption key size for encrypted ACL links */ if (!ev->status && ev->encrypt && conn->type == ACL_LINK) { - struct hci_cp_read_enc_key_size cp; - - /* Only send HCI_Read_Encryption_Key_Size if the - * controller really supports it. If it doesn't, assume - * the default size (16). - */ - if (!read_key_size_capable(hdev)) { - conn->enc_key_size = HCI_LINK_KEY_SIZE; + if (hci_read_enc_key_size(hdev, conn)) goto notify; - } - - cp.handle = cpu_to_le16(conn->handle); - if (hci_send_cmd(hdev, HCI_OP_READ_ENC_KEY_SIZE, - sizeof(cp), &cp)) { - bt_dev_err(hdev, "sending read key size failed"); - conn->enc_key_size = HCI_LINK_KEY_SIZE; - goto notify; - } goto unlock; } From 63de8abd97ddb9b758bd8f915ecbd18e1f1a87a0 Mon Sep 17 00:00:00 2001 From: James Morse Date: Thu, 9 Dec 2021 15:12:19 +0000 Subject: [PATCH 319/555] arm64: insn: Add support for encoding DSB To generate code in the eBPF epilogue that uses the DSB instruction, insn.c needs a heler to encode the type and domain. Re-use the crm encoding logic from the DMB instruction. Signed-off-by: James Morse Reviewed-by: Catalin Marinas --- arch/arm64/include/asm/insn.h | 1 + arch/arm64/lib/insn.c | 76 +++++++++++++++++++++-------------- 2 files changed, 46 insertions(+), 31 deletions(-) diff --git a/arch/arm64/include/asm/insn.h b/arch/arm64/include/asm/insn.h index 39577f1d079a..18c7811774d3 100644 --- a/arch/arm64/include/asm/insn.h +++ b/arch/arm64/include/asm/insn.h @@ -706,6 +706,7 @@ u32 aarch64_insn_gen_cas(enum aarch64_insn_register result, } #endif u32 aarch64_insn_gen_dmb(enum aarch64_insn_mb_type type); +u32 aarch64_insn_gen_dsb(enum aarch64_insn_mb_type type); u32 aarch64_insn_gen_mrs(enum aarch64_insn_register result, enum aarch64_insn_system_register sysreg); diff --git a/arch/arm64/lib/insn.c b/arch/arm64/lib/insn.c index 9bef696e2230..4e298baddc2e 100644 --- a/arch/arm64/lib/insn.c +++ b/arch/arm64/lib/insn.c @@ -5,6 +5,7 @@ * * Copyright (C) 2014-2016 Zi Shen Lim */ +#include #include #include #include @@ -1500,43 +1501,41 @@ u32 aarch64_insn_gen_extr(enum aarch64_insn_variant variant, return aarch64_insn_encode_register(AARCH64_INSN_REGTYPE_RM, insn, Rm); } +static u32 __get_barrier_crm_val(enum aarch64_insn_mb_type type) +{ + switch (type) { + case AARCH64_INSN_MB_SY: + return 0xf; + case AARCH64_INSN_MB_ST: + return 0xe; + case AARCH64_INSN_MB_LD: + return 0xd; + case AARCH64_INSN_MB_ISH: + return 0xb; + case AARCH64_INSN_MB_ISHST: + return 0xa; + case AARCH64_INSN_MB_ISHLD: + return 0x9; + case AARCH64_INSN_MB_NSH: + return 0x7; + case AARCH64_INSN_MB_NSHST: + return 0x6; + case AARCH64_INSN_MB_NSHLD: + return 0x5; + default: + pr_err("%s: unknown barrier type %d\n", __func__, type); + return AARCH64_BREAK_FAULT; + } +} + u32 aarch64_insn_gen_dmb(enum aarch64_insn_mb_type type) { u32 opt; u32 insn; - switch (type) { - case AARCH64_INSN_MB_SY: - opt = 0xf; - break; - case AARCH64_INSN_MB_ST: - opt = 0xe; - break; - case AARCH64_INSN_MB_LD: - opt = 0xd; - break; - case AARCH64_INSN_MB_ISH: - opt = 0xb; - break; - case AARCH64_INSN_MB_ISHST: - opt = 0xa; - break; - case AARCH64_INSN_MB_ISHLD: - opt = 0x9; - break; - case AARCH64_INSN_MB_NSH: - opt = 0x7; - break; - case AARCH64_INSN_MB_NSHST: - opt = 0x6; - break; - case AARCH64_INSN_MB_NSHLD: - opt = 0x5; - break; - default: - pr_err("%s: unknown dmb type %d\n", __func__, type); + opt = __get_barrier_crm_val(type); + if (opt == AARCH64_BREAK_FAULT) return AARCH64_BREAK_FAULT; - } insn = aarch64_insn_get_dmb_value(); insn &= ~GENMASK(11, 8); @@ -1545,6 +1544,21 @@ u32 aarch64_insn_gen_dmb(enum aarch64_insn_mb_type type) return insn; } +u32 aarch64_insn_gen_dsb(enum aarch64_insn_mb_type type) +{ + u32 opt, insn; + + opt = __get_barrier_crm_val(type); + if (opt == AARCH64_BREAK_FAULT) + return AARCH64_BREAK_FAULT; + + insn = aarch64_insn_get_dsb_base_value(); + insn &= ~GENMASK(11, 8); + insn |= (opt << 8); + + return insn; +} + u32 aarch64_insn_gen_mrs(enum aarch64_insn_register result, enum aarch64_insn_system_register sysreg) { From e7956c92f396a44eeeb6eaf7a5b5e1ad24db6748 Mon Sep 17 00:00:00 2001 From: James Morse Date: Mon, 19 Aug 2024 14:15:53 +0100 Subject: [PATCH 320/555] arm64: proton-pack: Expose whether the platform is mitigated by firmware is_spectre_bhb_fw_affected() allows the caller to determine if the CPU is known to need a firmware mitigation. CPUs are either on the list of CPUs we know about, or firmware has been queried and reported that the platform is affected - and mitigated by firmware. This helper is not useful to determine if the platform is mitigated by firmware. A CPU could be on the know list, but the firmware may not be implemented. Its affected but not mitigated. spectre_bhb_enable_mitigation() handles this distinction by checking the firmware state before enabling the mitigation. Add a helper to expose this state. This will be used by the BPF JIT to determine if calling firmware for a mitigation is necessary and supported. Signed-off-by: James Morse Reviewed-by: Catalin Marinas --- arch/arm64/include/asm/spectre.h | 1 + arch/arm64/kernel/proton-pack.c | 5 +++++ 2 files changed, 6 insertions(+) diff --git a/arch/arm64/include/asm/spectre.h b/arch/arm64/include/asm/spectre.h index f1524cdeacf1..78fb720987fb 100644 --- a/arch/arm64/include/asm/spectre.h +++ b/arch/arm64/include/asm/spectre.h @@ -97,6 +97,7 @@ enum mitigation_state arm64_get_meltdown_state(void); enum mitigation_state arm64_get_spectre_bhb_state(void); bool is_spectre_bhb_affected(const struct arm64_cpu_capabilities *entry, int scope); +bool is_spectre_bhb_fw_mitigated(void); void spectre_bhb_enable_mitigation(const struct arm64_cpu_capabilities *__unused); bool try_emulate_el1_ssbs(struct pt_regs *regs, u32 instr); diff --git a/arch/arm64/kernel/proton-pack.c b/arch/arm64/kernel/proton-pack.c index b607f6dfc5e6..355f1f3098a2 100644 --- a/arch/arm64/kernel/proton-pack.c +++ b/arch/arm64/kernel/proton-pack.c @@ -1094,6 +1094,11 @@ void spectre_bhb_enable_mitigation(const struct arm64_cpu_capabilities *entry) update_mitigation_state(&spectre_bhb_state, state); } +bool is_spectre_bhb_fw_mitigated(void) +{ + return test_bit(BHB_FW, &system_bhb_mitigations); +} + /* Patched to NOP when enabled */ void noinstr spectre_bhb_patch_loop_mitigation_enable(struct alt_instr *alt, __le32 *origptr, From a1152be30a043d2d4dcb1683415f328bf3c51978 Mon Sep 17 00:00:00 2001 From: James Morse Date: Tue, 29 Apr 2025 13:55:17 +0100 Subject: [PATCH 321/555] arm64: proton-pack: Expose whether the branchy loop k value Add a helper to expose the k value of the branchy loop. This is needed by the BPF JIT to generate the mitigation sequence in BPF programs. Signed-off-by: James Morse Reviewed-by: Catalin Marinas --- arch/arm64/include/asm/spectre.h | 1 + arch/arm64/kernel/proton-pack.c | 5 +++++ 2 files changed, 6 insertions(+) diff --git a/arch/arm64/include/asm/spectre.h b/arch/arm64/include/asm/spectre.h index 78fb720987fb..bca12134245c 100644 --- a/arch/arm64/include/asm/spectre.h +++ b/arch/arm64/include/asm/spectre.h @@ -97,6 +97,7 @@ enum mitigation_state arm64_get_meltdown_state(void); enum mitigation_state arm64_get_spectre_bhb_state(void); bool is_spectre_bhb_affected(const struct arm64_cpu_capabilities *entry, int scope); +u8 get_spectre_bhb_loop_value(void); bool is_spectre_bhb_fw_mitigated(void); void spectre_bhb_enable_mitigation(const struct arm64_cpu_capabilities *__unused); bool try_emulate_el1_ssbs(struct pt_regs *regs, u32 instr); diff --git a/arch/arm64/kernel/proton-pack.c b/arch/arm64/kernel/proton-pack.c index 355f1f3098a2..3154094a9e33 100644 --- a/arch/arm64/kernel/proton-pack.c +++ b/arch/arm64/kernel/proton-pack.c @@ -999,6 +999,11 @@ bool is_spectre_bhb_affected(const struct arm64_cpu_capabilities *entry, return true; } +u8 get_spectre_bhb_loop_value(void) +{ + return max_bhb_k; +} + static void this_cpu_set_vectors(enum arm64_bp_harden_el1_vectors slot) { const char *v = arm64_get_bp_hardening_vector(slot); From 0dfefc2ea2f29ced2416017d7e5b1253a54c2735 Mon Sep 17 00:00:00 2001 From: James Morse Date: Thu, 9 Dec 2021 15:13:24 +0000 Subject: [PATCH 322/555] arm64: bpf: Add BHB mitigation to the epilogue for cBPF programs A malicious BPF program may manipulate the branch history to influence what the hardware speculates will happen next. On exit from a BPF program, emit the BHB mititgation sequence. This is only applied for 'classic' cBPF programs that are loaded by seccomp. Signed-off-by: James Morse Reviewed-by: Catalin Marinas Acked-by: Daniel Borkmann --- arch/arm64/include/asm/spectre.h | 1 + arch/arm64/kernel/proton-pack.c | 2 +- arch/arm64/net/bpf_jit_comp.c | 54 +++++++++++++++++++++++++++++--- 3 files changed, 52 insertions(+), 5 deletions(-) diff --git a/arch/arm64/include/asm/spectre.h b/arch/arm64/include/asm/spectre.h index bca12134245c..8fef12626090 100644 --- a/arch/arm64/include/asm/spectre.h +++ b/arch/arm64/include/asm/spectre.h @@ -97,6 +97,7 @@ enum mitigation_state arm64_get_meltdown_state(void); enum mitigation_state arm64_get_spectre_bhb_state(void); bool is_spectre_bhb_affected(const struct arm64_cpu_capabilities *entry, int scope); +extern bool __nospectre_bhb; u8 get_spectre_bhb_loop_value(void); bool is_spectre_bhb_fw_mitigated(void); void spectre_bhb_enable_mitigation(const struct arm64_cpu_capabilities *__unused); diff --git a/arch/arm64/kernel/proton-pack.c b/arch/arm64/kernel/proton-pack.c index 3154094a9e33..4459b613077e 100644 --- a/arch/arm64/kernel/proton-pack.c +++ b/arch/arm64/kernel/proton-pack.c @@ -1021,7 +1021,7 @@ static void this_cpu_set_vectors(enum arm64_bp_harden_el1_vectors slot) isb(); } -static bool __read_mostly __nospectre_bhb; +bool __read_mostly __nospectre_bhb; static int __init parse_spectre_bhb_param(char *str) { __nospectre_bhb = true; diff --git a/arch/arm64/net/bpf_jit_comp.c b/arch/arm64/net/bpf_jit_comp.c index 70d7c89d3ac9..0ab8e47062d9 100644 --- a/arch/arm64/net/bpf_jit_comp.c +++ b/arch/arm64/net/bpf_jit_comp.c @@ -7,6 +7,7 @@ #define pr_fmt(fmt) "bpf_jit: " fmt +#include #include #include #include @@ -17,6 +18,7 @@ #include #include #include +#include #include #include #include @@ -939,7 +941,48 @@ static void build_plt(struct jit_ctx *ctx) plt->target = (u64)&dummy_tramp; } -static void build_epilogue(struct jit_ctx *ctx) +/* Clobbers BPF registers 1-4, aka x0-x3 */ +static void __maybe_unused build_bhb_mitigation(struct jit_ctx *ctx) +{ + const u8 r1 = bpf2a64[BPF_REG_1]; /* aka x0 */ + u8 k = get_spectre_bhb_loop_value(); + + if (!IS_ENABLED(CONFIG_MITIGATE_SPECTRE_BRANCH_HISTORY) || + cpu_mitigations_off() || __nospectre_bhb || + arm64_get_spectre_v2_state() == SPECTRE_VULNERABLE) + return; + + if (supports_clearbhb(SCOPE_SYSTEM)) { + emit(aarch64_insn_gen_hint(AARCH64_INSN_HINT_CLEARBHB), ctx); + return; + } + + if (k) { + emit_a64_mov_i64(r1, k, ctx); + emit(A64_B(1), ctx); + emit(A64_SUBS_I(true, r1, r1, 1), ctx); + emit(A64_B_(A64_COND_NE, -2), ctx); + emit(aarch64_insn_gen_dsb(AARCH64_INSN_MB_ISH), ctx); + emit(aarch64_insn_get_isb_value(), ctx); + } + + if (is_spectre_bhb_fw_mitigated()) { + emit(A64_ORR_I(false, r1, AARCH64_INSN_REG_ZR, + ARM_SMCCC_ARCH_WORKAROUND_3), ctx); + switch (arm_smccc_1_1_get_conduit()) { + case SMCCC_CONDUIT_HVC: + emit(aarch64_insn_get_hvc_value(), ctx); + break; + case SMCCC_CONDUIT_SMC: + emit(aarch64_insn_get_smc_value(), ctx); + break; + default: + pr_err_once("Firmware mitigation enabled with unknown conduit\n"); + } + } +} + +static void build_epilogue(struct jit_ctx *ctx, bool was_classic) { const u8 r0 = bpf2a64[BPF_REG_0]; const u8 ptr = bpf2a64[TCCNT_PTR]; @@ -952,10 +995,13 @@ static void build_epilogue(struct jit_ctx *ctx) emit(A64_POP(A64_ZR, ptr, A64_SP), ctx); + if (was_classic) + build_bhb_mitigation(ctx); + /* Restore FP/LR registers */ emit(A64_POP(A64_FP, A64_LR, A64_SP), ctx); - /* Set return value */ + /* Move the return value from bpf:r0 (aka x7) to x0 */ emit(A64_MOV(1, A64_R(0), r0), ctx); /* Authenticate lr */ @@ -1898,7 +1944,7 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *prog) } ctx.epilogue_offset = ctx.idx; - build_epilogue(&ctx); + build_epilogue(&ctx, was_classic); build_plt(&ctx); extable_align = __alignof__(struct exception_table_entry); @@ -1961,7 +2007,7 @@ skip_init_ctx: goto out_free_hdr; } - build_epilogue(&ctx); + build_epilogue(&ctx, was_classic); build_plt(&ctx); /* Extra pass to validate JITed code. */ From f300769ead032513a68e4a02e806393402e626f8 Mon Sep 17 00:00:00 2001 From: James Morse Date: Tue, 29 Apr 2025 16:03:38 +0100 Subject: [PATCH 323/555] arm64: bpf: Only mitigate cBPF programs loaded by unprivileged users Support for eBPF programs loaded by unprivileged users is typically disabled. This means only cBPF programs need to be mitigated for BHB. In addition, only mitigate cBPF programs that were loaded by an unprivileged user. Privileged users can also load the same program via eBPF, making the mitigation pointless. Signed-off-by: James Morse Reviewed-by: Catalin Marinas Acked-by: Daniel Borkmann --- arch/arm64/net/bpf_jit_comp.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/arch/arm64/net/bpf_jit_comp.c b/arch/arm64/net/bpf_jit_comp.c index 0ab8e47062d9..634d78422adb 100644 --- a/arch/arm64/net/bpf_jit_comp.c +++ b/arch/arm64/net/bpf_jit_comp.c @@ -952,6 +952,9 @@ static void __maybe_unused build_bhb_mitigation(struct jit_ctx *ctx) arm64_get_spectre_v2_state() == SPECTRE_VULNERABLE) return; + if (capable(CAP_SYS_ADMIN)) + return; + if (supports_clearbhb(SCOPE_SYSTEM)) { emit(aarch64_insn_gen_hint(AARCH64_INSN_HINT_CLEARBHB), ctx); return; From efe676a1a7554219eae0b0dcfe1e0cdcc9ef9aef Mon Sep 17 00:00:00 2001 From: James Morse Date: Mon, 12 Aug 2024 17:50:22 +0100 Subject: [PATCH 324/555] arm64: proton-pack: Add new CPUs 'k' values for branch mitigation Update the list of 'k' values for the branch mitigation from arm's website. Add the values for Cortex-X1C. The MIDR_EL1 value can be found here: https://developer.arm.com/documentation/101968/0002/Register-descriptions/AArch> Link: https://developer.arm.com/documentation/110280/2-0/?lang=en Signed-off-by: James Morse Reviewed-by: Catalin Marinas --- arch/arm64/include/asm/cputype.h | 2 ++ arch/arm64/kernel/proton-pack.c | 1 + 2 files changed, 3 insertions(+) diff --git a/arch/arm64/include/asm/cputype.h b/arch/arm64/include/asm/cputype.h index d1cc0571798b..dffff6763812 100644 --- a/arch/arm64/include/asm/cputype.h +++ b/arch/arm64/include/asm/cputype.h @@ -81,6 +81,7 @@ #define ARM_CPU_PART_CORTEX_A78AE 0xD42 #define ARM_CPU_PART_CORTEX_X1 0xD44 #define ARM_CPU_PART_CORTEX_A510 0xD46 +#define ARM_CPU_PART_CORTEX_X1C 0xD4C #define ARM_CPU_PART_CORTEX_A520 0xD80 #define ARM_CPU_PART_CORTEX_A710 0xD47 #define ARM_CPU_PART_CORTEX_A715 0xD4D @@ -168,6 +169,7 @@ #define MIDR_CORTEX_A78AE MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_A78AE) #define MIDR_CORTEX_X1 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_X1) #define MIDR_CORTEX_A510 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_A510) +#define MIDR_CORTEX_X1C MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_X1C) #define MIDR_CORTEX_A520 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_A520) #define MIDR_CORTEX_A710 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_A710) #define MIDR_CORTEX_A715 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_A715) diff --git a/arch/arm64/kernel/proton-pack.c b/arch/arm64/kernel/proton-pack.c index 4459b613077e..edf1783ffc81 100644 --- a/arch/arm64/kernel/proton-pack.c +++ b/arch/arm64/kernel/proton-pack.c @@ -891,6 +891,7 @@ static u8 spectre_bhb_loop_affected(void) MIDR_ALL_VERSIONS(MIDR_CORTEX_A78AE), MIDR_ALL_VERSIONS(MIDR_CORTEX_A78C), MIDR_ALL_VERSIONS(MIDR_CORTEX_X1), + MIDR_ALL_VERSIONS(MIDR_CORTEX_X1C), MIDR_ALL_VERSIONS(MIDR_CORTEX_A710), MIDR_ALL_VERSIONS(MIDR_CORTEX_X2), MIDR_ALL_VERSIONS(MIDR_NEOVERSE_N2), From c0d0a9ff6d5b5b23ddabde8bcbafb28fa454ae00 Mon Sep 17 00:00:00 2001 From: Aaron Lu Date: Thu, 8 May 2025 16:30:36 +0800 Subject: [PATCH 325/555] block: remove test of incorrect io priority level Ever since commit eca2040972b4("scsi: block: ioprio: Clean up interface definition"), the macro IOPRIO_PRIO_LEVEL() will mask the level value to something between 0 and 7 so necessarily, level will always be lower than IOPRIO_NR_LEVELS(8). Remove this obsolete check. Reported-by: Kexin Wei Cc: Damien Le Moal Signed-off-by: Aaron Lu Reviewed-by: Damien Le Moal Link: https://lore.kernel.org/r/20250508083018.GA769554@bytedance Signed-off-by: Jens Axboe --- block/ioprio.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/block/ioprio.c b/block/ioprio.c index 73301a261429..f0ee2798539c 100644 --- a/block/ioprio.c +++ b/block/ioprio.c @@ -46,12 +46,8 @@ int ioprio_check_cap(int ioprio) */ if (!capable(CAP_SYS_ADMIN) && !capable(CAP_SYS_NICE)) return -EPERM; - fallthrough; - /* rt has prio field too */ - case IOPRIO_CLASS_BE: - if (level >= IOPRIO_NR_LEVELS) - return -EINVAL; break; + case IOPRIO_CLASS_BE: case IOPRIO_CLASS_IDLE: break; case IOPRIO_CLASS_NONE: From 0e33e0f339b91eecd9558311449a3d1e728722d4 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Wed, 30 Apr 2025 12:46:56 -0400 Subject: [PATCH 326/555] drm/amdgpu/hdp5: use memcfg register to post the write for HDP flush Reading back the remapped HDP flush register seems to cause problems on some platforms. All we need is a read, so read back the memcfg register. Fixes: cf424020e040 ("drm/amdgpu/hdp5.0: do a posting read when flushing HDP") Reported-by: Alexey Klimov Link: https://lists.freedesktop.org/archives/amd-gfx/2025-April/123150.html Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4119 Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3908 Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher (cherry picked from commit a5cb344033c7598762e89255e8ff52827abb57a4) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/hdp_v5_0.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/hdp_v5_0.c b/drivers/gpu/drm/amd/amdgpu/hdp_v5_0.c index 43195c079748..086a647308df 100644 --- a/drivers/gpu/drm/amd/amdgpu/hdp_v5_0.c +++ b/drivers/gpu/drm/amd/amdgpu/hdp_v5_0.c @@ -32,7 +32,12 @@ static void hdp_v5_0_flush_hdp(struct amdgpu_device *adev, { if (!ring || !ring->funcs->emit_wreg) { WREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0); - RREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2); + /* We just need to read back a register to post the write. + * Reading back the remapped register causes problems on + * some platforms so just read back the memory size register. + */ + if (adev->nbio.funcs->get_memsize) + adev->nbio.funcs->get_memsize(adev); } else { amdgpu_ring_emit_wreg(ring, (adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0); } From dbc988c689333faeeed44d5561f372ff20395304 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Wed, 30 Apr 2025 12:47:37 -0400 Subject: [PATCH 327/555] drm/amdgpu/hdp5.2: use memcfg register to post the write for HDP flush Reading back the remapped HDP flush register seems to cause problems on some platforms. All we need is a read, so read back the memcfg register. Fixes: f756dbac1ce1 ("drm/amdgpu/hdp5.2: do a posting read when flushing HDP") Reported-by: Alexey Klimov Link: https://lists.freedesktop.org/archives/amd-gfx/2025-April/123150.html Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4119 Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3908 Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher (cherry picked from commit 4a89b7698e771914b4d5b571600c76e2fdcbe2a9) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/hdp_v5_2.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/hdp_v5_2.c b/drivers/gpu/drm/amd/amdgpu/hdp_v5_2.c index fcb8dd2876bc..40940b4ab400 100644 --- a/drivers/gpu/drm/amd/amdgpu/hdp_v5_2.c +++ b/drivers/gpu/drm/amd/amdgpu/hdp_v5_2.c @@ -33,7 +33,17 @@ static void hdp_v5_2_flush_hdp(struct amdgpu_device *adev, if (!ring || !ring->funcs->emit_wreg) { WREG32_NO_KIQ((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0); - RREG32_NO_KIQ((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2); + if (amdgpu_sriov_vf(adev)) { + /* this is fine because SR_IOV doesn't remap the register */ + RREG32_NO_KIQ((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2); + } else { + /* We just need to read back a register to post the write. + * Reading back the remapped register causes problems on + * some platforms so just read back the memory size register. + */ + if (adev->nbio.funcs->get_memsize) + adev->nbio.funcs->get_memsize(adev); + } } else { amdgpu_ring_emit_wreg(ring, (adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, From ca28e80abe4219c8f1a2961ae05102d70af6dc87 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Wed, 30 Apr 2025 12:48:51 -0400 Subject: [PATCH 328/555] drm/amdgpu/hdp6: use memcfg register to post the write for HDP flush Reading back the remapped HDP flush register seems to cause problems on some platforms. All we need is a read, so read back the memcfg register. Fixes: abe1cbaec6cf ("drm/amdgpu/hdp6.0: do a posting read when flushing HDP") Reported-by: Alexey Klimov Link: https://lists.freedesktop.org/archives/amd-gfx/2025-April/123150.html Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4119 Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3908 Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher (cherry picked from commit 84141ff615951359c9a99696fd79a36c465ed847) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/hdp_v6_0.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/hdp_v6_0.c b/drivers/gpu/drm/amd/amdgpu/hdp_v6_0.c index a88d25a06c29..6ccd31c8bc69 100644 --- a/drivers/gpu/drm/amd/amdgpu/hdp_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/hdp_v6_0.c @@ -35,7 +35,12 @@ static void hdp_v6_0_flush_hdp(struct amdgpu_device *adev, { if (!ring || !ring->funcs->emit_wreg) { WREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0); - RREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2); + /* We just need to read back a register to post the write. + * Reading back the remapped register causes problems on + * some platforms so just read back the memory size register. + */ + if (adev->nbio.funcs->get_memsize) + adev->nbio.funcs->get_memsize(adev); } else { amdgpu_ring_emit_wreg(ring, (adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0); } From 5a11a2767731139bf87e667331aa2209e33a1d19 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Wed, 30 Apr 2025 12:50:02 -0400 Subject: [PATCH 329/555] drm/amdgpu/hdp7: use memcfg register to post the write for HDP flush Reading back the remapped HDP flush register seems to cause problems on some platforms. All we need is a read, so read back the memcfg register. Fixes: 689275140cb8 ("drm/amdgpu/hdp7.0: do a posting read when flushing HDP") Reported-by: Alexey Klimov Link: https://lists.freedesktop.org/archives/amd-gfx/2025-April/123150.html Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4119 Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3908 Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher (cherry picked from commit dbc064adfcf9095e7d895bea87b2f75c1ab23236) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/hdp_v7_0.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/hdp_v7_0.c b/drivers/gpu/drm/amd/amdgpu/hdp_v7_0.c index 49f7eb4fbd11..2c9239a22f39 100644 --- a/drivers/gpu/drm/amd/amdgpu/hdp_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/hdp_v7_0.c @@ -32,7 +32,12 @@ static void hdp_v7_0_flush_hdp(struct amdgpu_device *adev, { if (!ring || !ring->funcs->emit_wreg) { WREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0); - RREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2); + /* We just need to read back a register to post the write. + * Reading back the remapped register causes problems on + * some platforms so just read back the memory size register. + */ + if (adev->nbio.funcs->get_memsize) + adev->nbio.funcs->get_memsize(adev); } else { amdgpu_ring_emit_wreg(ring, (adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0); } From adfab6b39202481bb43286fff94def4953793fdb Mon Sep 17 00:00:00 2001 From: Jeremy Linton Date: Wed, 7 May 2025 21:30:25 -0500 Subject: [PATCH 330/555] ACPI: PPTT: Fix processor subtable walk The original PPTT code had a bug where the processor subtable length was not correctly validated when encountering a truncated acpi_pptt_processor node. Commit 7ab4f0e37a0f4 ("ACPI PPTT: Fix coding mistakes in a couple of sizeof() calls") attempted to fix this by validating the size is as large as the acpi_pptt_processor node structure. This introduced a regression where the last processor node in the PPTT table is ignored if it doesn't contain any private resources. That results errors like: ACPI PPTT: PPTT table found, but unable to locate core XX (XX) ACPI: SPE must be homogeneous Furthermore, it fails in a common case where the node length isn't equal to the acpi_pptt_processor structure size, leaving the original bug in a modified form. Correct the regression by adjusting the loop termination conditions as suggested by the bug reporters. An additional check performed after the subtable node type is detected, validates the acpi_pptt_processor node is fully contained in the PPTT table. Repeating the check in acpi_pptt_leaf_node() is largely redundant as the node is already known to be fully contained in the table. The case where a final truncated node's parent property is accepted, but the node itself is rejected should not be considered a bug. Fixes: 7ab4f0e37a0f4 ("ACPI PPTT: Fix coding mistakes in a couple of sizeof() calls") Reported-by: Maximilian Heyne Closes: https://lore.kernel.org/linux-acpi/20250506-draco-taped-15f475cd@mheyne-amazon/ Reported-by: Yicong Yang Closes: https://lore.kernel.org/linux-acpi/20250507035124.28071-1-yangyicong@huawei.com/ Signed-off-by: Jeremy Linton Tested-by: Yicong Yang Reviewed-by: Sudeep Holla Tested-by: Maximilian Heyne Cc: All applicable # 7ab4f0e37a0f4: ACPI PPTT: Fix coding mistakes ... Link: https://patch.msgid.link/20250508023025.1301030-1-jeremy.linton@arm.com Signed-off-by: Rafael J. Wysocki --- drivers/acpi/pptt.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/drivers/acpi/pptt.c b/drivers/acpi/pptt.c index f73ce6e13065..54676e3d82dd 100644 --- a/drivers/acpi/pptt.c +++ b/drivers/acpi/pptt.c @@ -231,16 +231,18 @@ static int acpi_pptt_leaf_node(struct acpi_table_header *table_hdr, sizeof(struct acpi_table_pptt)); proc_sz = sizeof(struct acpi_pptt_processor); - while ((unsigned long)entry + proc_sz < table_end) { + /* ignore subtable types that are smaller than a processor node */ + while ((unsigned long)entry + proc_sz <= table_end) { cpu_node = (struct acpi_pptt_processor *)entry; + if (entry->type == ACPI_PPTT_TYPE_PROCESSOR && cpu_node->parent == node_entry) return 0; if (entry->length == 0) return 0; + entry = ACPI_ADD_PTR(struct acpi_subtable_header, entry, entry->length); - } return 1; } @@ -273,15 +275,18 @@ static struct acpi_pptt_processor *acpi_find_processor_node(struct acpi_table_he proc_sz = sizeof(struct acpi_pptt_processor); /* find the processor structure associated with this cpuid */ - while ((unsigned long)entry + proc_sz < table_end) { + while ((unsigned long)entry + proc_sz <= table_end) { cpu_node = (struct acpi_pptt_processor *)entry; if (entry->length == 0) { pr_warn("Invalid zero length subtable\n"); break; } + /* entry->length may not equal proc_sz, revalidate the processor structure length */ if (entry->type == ACPI_PPTT_TYPE_PROCESSOR && acpi_cpu_id == cpu_node->acpi_processor_id && + (unsigned long)entry + entry->length <= table_end && + entry->length == proc_sz + cpu_node->number_of_priv_resources * sizeof(u32) && acpi_pptt_leaf_node(table_hdr, cpu_node)) { return (struct acpi_pptt_processor *)entry; } From 391008f34e711253c5983b0bf52277cc43723127 Mon Sep 17 00:00:00 2001 From: Matthew Brost Date: Tue, 8 Apr 2025 08:59:15 -0700 Subject: [PATCH 331/555] drm/xe: Add page queue multiplier For an unknown reason the math to determine the PF queue size does is not correct - compute UMD applications are overflowing the PF queue which is fatal. A multippier of 8 fixes the problem. Fixes: 3338e4f90c14 ("drm/xe: Use topology to determine page fault queue size") Cc: stable@vger.kernel.org Signed-off-by: Matthew Brost Reviewed-by: Jagmeet Randhawa Link: https://lore.kernel.org/r/20250408155915.78770-1-matthew.brost@intel.com (cherry picked from commit 29582e0ea75c95668d168b12406e3c56cf5a73c4) Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/xe_gt_pagefault.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_gt_pagefault.c b/drivers/gpu/drm/xe/xe_gt_pagefault.c index c5ad9a0a89c2..0c22b3a36655 100644 --- a/drivers/gpu/drm/xe/xe_gt_pagefault.c +++ b/drivers/gpu/drm/xe/xe_gt_pagefault.c @@ -435,9 +435,16 @@ static int xe_alloc_pf_queue(struct xe_gt *gt, struct pf_queue *pf_queue) num_eus = bitmap_weight(gt->fuse_topo.eu_mask_per_dss, XE_MAX_EU_FUSE_BITS) * num_dss; - /* user can issue separate page faults per EU and per CS */ + /* + * user can issue separate page faults per EU and per CS + * + * XXX: Multiplier required as compute UMD are getting PF queue errors + * without it. Follow on why this multiplier is required. + */ +#define PF_MULTIPLIER 8 pf_queue->num_dw = - (num_eus + XE_NUM_HW_ENGINES) * PF_MSG_LEN_DW; + (num_eus + XE_NUM_HW_ENGINES) * PF_MSG_LEN_DW * PF_MULTIPLIER; +#undef PF_MULTIPLIER pf_queue->gt = gt; pf_queue->data = devm_kcalloc(xe->drm.dev, pf_queue->num_dw, From 51c0ee84e4dc339287b2d7335f2b54d747794c83 Mon Sep 17 00:00:00 2001 From: Tejas Upadhyay Date: Mon, 28 Apr 2025 13:53:57 +0530 Subject: [PATCH 332/555] drm/xe/tests/mocs: Hold XE_FORCEWAKE_ALL for LNCF regs LNCF registers report wrong values when XE_FORCEWAKE_GT only is held. Holding XE_FORCEWAKE_ALL ensures correct operations on LNCF regs. V2(Himal): - Use xe_force_wake_ref_has_domain Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/1999 Fixes: a6a4ea6d7d37 ("drm/xe: Add mocs kunit") Reviewed-by: Himal Prasad Ghimiray Link: https://patchwork.freedesktop.org/patch/msgid/20250428082357.1730068-1-tejas.upadhyay@intel.com Signed-off-by: Tejas Upadhyay (cherry picked from commit 70a2585e582058e94fe4381a337be42dec800337) Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/tests/xe_mocs.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/xe/tests/xe_mocs.c b/drivers/gpu/drm/xe/tests/xe_mocs.c index ef1e5256c56a..0e502feaca81 100644 --- a/drivers/gpu/drm/xe/tests/xe_mocs.c +++ b/drivers/gpu/drm/xe/tests/xe_mocs.c @@ -46,8 +46,11 @@ static void read_l3cc_table(struct xe_gt *gt, unsigned int fw_ref, i; u32 reg_val; - fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FW_GT); - KUNIT_ASSERT_NE_MSG(test, fw_ref, 0, "Forcewake Failed.\n"); + fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FORCEWAKE_ALL); + if (!xe_force_wake_ref_has_domain(fw_ref, XE_FORCEWAKE_ALL)) { + xe_force_wake_put(gt_to_fw(gt), fw_ref); + KUNIT_ASSERT_TRUE_MSG(test, true, "Forcewake Failed.\n"); + } for (i = 0; i < info->num_mocs_regs; i++) { if (!(i & 1)) { From 03552d8ac0afcc080c339faa0b726e2c0e9361cb Mon Sep 17 00:00:00 2001 From: Daniele Ceraolo Spurio Date: Fri, 2 May 2025 08:51:04 -0700 Subject: [PATCH 333/555] drm/xe/gsc: do not flush the GSC worker from the reset path The workqueue used for the reset worker is marked as WQ_MEM_RECLAIM, while the GSC one isn't (and can't be as we need to do memory allocations in the gsc worker). Therefore, we can't flush the latter from the former. The reason why we had such a flush was to avoid interrupting either the GSC FW load or in progress GSC proxy operations. GSC proxy operations fall into 2 categories: 1) GSC proxy init: this only happens once immediately after GSC FW load and does not support being interrupted. The only way to recover from an interruption of the proxy init is to do an FLR and re-load the GSC. 2) GSC proxy request: this can happen in response to a request that the driver sends to the GSC. If this is interrupted, the GSC FW will timeout and the driver request will be failed, but overall the GSC will keep working fine. Flushing the work allowed us to avoid interruption in both cases (unless the hang came from the GSC engine itself, in which case we're toast anyway). However, a failure on a proxy request is tolerable if we're in a scenario where we're triggering a GT reset (i.e., something is already gone pretty wrong), so what we really need to avoid is interrupting the init flow, which we can do by polling on the register that reports when the proxy init is complete (as that ensure us that all the load and init operations have been completed). Note that during suspend we still want to do a flush of the worker to make sure it completes any operations involving the HW before the power is cut. v2: fix spelling in commit msg, rename waiter function (Julia) Fixes: dd0e89e5edc2 ("drm/xe/gsc: GSC FW load") Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/4830 Signed-off-by: Daniele Ceraolo Spurio Cc: John Harrison Cc: Alan Previn Cc: # v6.8+ Reviewed-by: Julia Filipchuk Link: https://lore.kernel.org/r/20250502155104.2201469-1-daniele.ceraolospurio@intel.com (cherry picked from commit 12370bfcc4f0bdf70279ec5b570eb298963422b5) Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/xe_gsc.c | 22 ++++++++++++++++++++++ drivers/gpu/drm/xe/xe_gsc.h | 1 + drivers/gpu/drm/xe/xe_gsc_proxy.c | 11 +++++++++++ drivers/gpu/drm/xe/xe_gsc_proxy.h | 1 + drivers/gpu/drm/xe/xe_gt.c | 2 +- drivers/gpu/drm/xe/xe_uc.c | 8 +++++++- drivers/gpu/drm/xe/xe_uc.h | 1 + 7 files changed, 44 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_gsc.c b/drivers/gpu/drm/xe/xe_gsc.c index fd41113f8572..0bcf97063ff6 100644 --- a/drivers/gpu/drm/xe/xe_gsc.c +++ b/drivers/gpu/drm/xe/xe_gsc.c @@ -555,6 +555,28 @@ void xe_gsc_wait_for_worker_completion(struct xe_gsc *gsc) flush_work(&gsc->work); } +void xe_gsc_stop_prepare(struct xe_gsc *gsc) +{ + struct xe_gt *gt = gsc_to_gt(gsc); + int ret; + + if (!xe_uc_fw_is_loadable(&gsc->fw) || xe_uc_fw_is_in_error_state(&gsc->fw)) + return; + + xe_force_wake_assert_held(gt_to_fw(gt), XE_FW_GSC); + + /* + * If the GSC FW load or the proxy init are interrupted, the only way + * to recover it is to do an FLR and reload the GSC from scratch. + * Therefore, let's wait for the init to complete before stopping + * operations. The proxy init is the last step, so we can just wait on + * that + */ + ret = xe_gsc_wait_for_proxy_init_done(gsc); + if (ret) + xe_gt_err(gt, "failed to wait for GSC init completion before uc stop\n"); +} + /* * wa_14015076503: if the GSC FW is loaded, we need to alert it before doing a * GSC engine reset by writing a notification bit in the GS1 register and then diff --git a/drivers/gpu/drm/xe/xe_gsc.h b/drivers/gpu/drm/xe/xe_gsc.h index d99f66c38075..b8b8e0810ad9 100644 --- a/drivers/gpu/drm/xe/xe_gsc.h +++ b/drivers/gpu/drm/xe/xe_gsc.h @@ -16,6 +16,7 @@ struct xe_hw_engine; int xe_gsc_init(struct xe_gsc *gsc); int xe_gsc_init_post_hwconfig(struct xe_gsc *gsc); void xe_gsc_wait_for_worker_completion(struct xe_gsc *gsc); +void xe_gsc_stop_prepare(struct xe_gsc *gsc); void xe_gsc_load_start(struct xe_gsc *gsc); void xe_gsc_hwe_irq_handler(struct xe_hw_engine *hwe, u16 intr_vec); diff --git a/drivers/gpu/drm/xe/xe_gsc_proxy.c b/drivers/gpu/drm/xe/xe_gsc_proxy.c index 8cf70b228ff3..d0519cd6704a 100644 --- a/drivers/gpu/drm/xe/xe_gsc_proxy.c +++ b/drivers/gpu/drm/xe/xe_gsc_proxy.c @@ -71,6 +71,17 @@ bool xe_gsc_proxy_init_done(struct xe_gsc *gsc) HECI1_FWSTS1_PROXY_STATE_NORMAL; } +int xe_gsc_wait_for_proxy_init_done(struct xe_gsc *gsc) +{ + struct xe_gt *gt = gsc_to_gt(gsc); + + /* Proxy init can take up to 500ms, so wait double that for safety */ + return xe_mmio_wait32(>->mmio, HECI_FWSTS1(MTL_GSC_HECI1_BASE), + HECI1_FWSTS1_CURRENT_STATE, + HECI1_FWSTS1_PROXY_STATE_NORMAL, + USEC_PER_SEC, NULL, false); +} + static void __gsc_proxy_irq_rmw(struct xe_gsc *gsc, u32 clr, u32 set) { struct xe_gt *gt = gsc_to_gt(gsc); diff --git a/drivers/gpu/drm/xe/xe_gsc_proxy.h b/drivers/gpu/drm/xe/xe_gsc_proxy.h index fdef56995cd4..765602221dbc 100644 --- a/drivers/gpu/drm/xe/xe_gsc_proxy.h +++ b/drivers/gpu/drm/xe/xe_gsc_proxy.h @@ -12,6 +12,7 @@ struct xe_gsc; int xe_gsc_proxy_init(struct xe_gsc *gsc); bool xe_gsc_proxy_init_done(struct xe_gsc *gsc); +int xe_gsc_wait_for_proxy_init_done(struct xe_gsc *gsc); int xe_gsc_proxy_start(struct xe_gsc *gsc); int xe_gsc_proxy_request_handler(struct xe_gsc *gsc); diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c index 10a9e3c72b36..66198cf2662c 100644 --- a/drivers/gpu/drm/xe/xe_gt.c +++ b/drivers/gpu/drm/xe/xe_gt.c @@ -857,7 +857,7 @@ void xe_gt_suspend_prepare(struct xe_gt *gt) fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FORCEWAKE_ALL); - xe_uc_stop_prepare(>->uc); + xe_uc_suspend_prepare(>->uc); xe_force_wake_put(gt_to_fw(gt), fw_ref); } diff --git a/drivers/gpu/drm/xe/xe_uc.c b/drivers/gpu/drm/xe/xe_uc.c index c14bd2282044..3a8751a8b92d 100644 --- a/drivers/gpu/drm/xe/xe_uc.c +++ b/drivers/gpu/drm/xe/xe_uc.c @@ -244,7 +244,7 @@ void xe_uc_gucrc_disable(struct xe_uc *uc) void xe_uc_stop_prepare(struct xe_uc *uc) { - xe_gsc_wait_for_worker_completion(&uc->gsc); + xe_gsc_stop_prepare(&uc->gsc); xe_guc_stop_prepare(&uc->guc); } @@ -278,6 +278,12 @@ again: goto again; } +void xe_uc_suspend_prepare(struct xe_uc *uc) +{ + xe_gsc_wait_for_worker_completion(&uc->gsc); + xe_guc_stop_prepare(&uc->guc); +} + int xe_uc_suspend(struct xe_uc *uc) { /* GuC submission not enabled, nothing to do */ diff --git a/drivers/gpu/drm/xe/xe_uc.h b/drivers/gpu/drm/xe/xe_uc.h index 3813c1ede450..c23e6f5e2514 100644 --- a/drivers/gpu/drm/xe/xe_uc.h +++ b/drivers/gpu/drm/xe/xe_uc.h @@ -18,6 +18,7 @@ int xe_uc_reset_prepare(struct xe_uc *uc); void xe_uc_stop_prepare(struct xe_uc *uc); void xe_uc_stop(struct xe_uc *uc); int xe_uc_start(struct xe_uc *uc); +void xe_uc_suspend_prepare(struct xe_uc *uc); int xe_uc_suspend(struct xe_uc *uc); int xe_uc_sanitize_reset(struct xe_uc *uc); void xe_uc_declare_wedged(struct xe_uc *uc); From 9d271a4f5ba52520e448ab223b1a91c6e35f17c7 Mon Sep 17 00:00:00 2001 From: Shuicheng Lin Date: Wed, 7 May 2025 02:23:02 +0000 Subject: [PATCH 334/555] drm/xe: Release force wake first then runtime power xe_force_wake_get() is dependent on xe_pm_runtime_get(), so for the release path, xe_force_wake_put() should be called first then xe_pm_runtime_put(). Combine the error path and normal path together with goto. Fixes: 85d547608ef5 ("drm/xe/xe_gt_debugfs: Update handling of xe_force_wake_get return") Cc: Himal Prasad Ghimiray Cc: Rodrigo Vivi Signed-off-by: Shuicheng Lin Reviewed-by: Himal Prasad Ghimiray Link: https://lore.kernel.org/r/20250507022302.2187527-1-shuicheng.lin@intel.com Signed-off-by: Rodrigo Vivi (cherry picked from commit 432cd94efdca06296cc5e76d673546f58aa90ee1) Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/xe_gt_debugfs.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_gt_debugfs.c b/drivers/gpu/drm/xe/xe_gt_debugfs.c index 2d63a69cbfa3..f7005a3643e6 100644 --- a/drivers/gpu/drm/xe/xe_gt_debugfs.c +++ b/drivers/gpu/drm/xe/xe_gt_debugfs.c @@ -92,22 +92,23 @@ static int hw_engines(struct xe_gt *gt, struct drm_printer *p) struct xe_hw_engine *hwe; enum xe_hw_engine_id id; unsigned int fw_ref; + int ret = 0; xe_pm_runtime_get(xe); fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FORCEWAKE_ALL); if (!xe_force_wake_ref_has_domain(fw_ref, XE_FORCEWAKE_ALL)) { - xe_pm_runtime_put(xe); - xe_force_wake_put(gt_to_fw(gt), fw_ref); - return -ETIMEDOUT; + ret = -ETIMEDOUT; + goto fw_put; } for_each_hw_engine(hwe, gt, id) xe_hw_engine_print(hwe, p); +fw_put: xe_force_wake_put(gt_to_fw(gt), fw_ref); xe_pm_runtime_put(xe); - return 0; + return ret; } static int powergate_info(struct xe_gt *gt, struct drm_printer *p) From 564467e9d06c6352fac9100e8957d40a1a50234c Mon Sep 17 00:00:00 2001 From: Shuicheng Lin Date: Fri, 2 May 2025 17:00:52 +0000 Subject: [PATCH 335/555] drm/xe: Add config control for svm flush work Without CONFIG_DRM_XE_GPUSVM set, GPU SVM is not initialized thus below warning pops. Refine the flush work code to be controlled by the config to avoid below warning: " [ 453.132028] ------------[ cut here ]------------ [ 453.132527] WARNING: CPU: 9 PID: 4491 at kernel/workqueue.c:4205 __flush_work+0x379/0x3a0 [ 453.133355] Modules linked in: xe drm_ttm_helper ttm gpu_sched drm_buddy drm_suballoc_helper drm_gpuvm drm_exec [ 453.134352] CPU: 9 UID: 0 PID: 4491 Comm: xe_exec_mix_mod Tainted: G U W 6.15.0-rc3+ #7 PREEMPT(full) [ 453.135405] Tainted: [U]=USER, [W]=WARN ... [ 453.136921] RIP: 0010:__flush_work+0x379/0x3a0 [ 453.137417] Code: 8b 45 00 48 8b 55 08 89 c7 48 c1 e8 04 83 e7 08 83 e0 0f 83 cf 02 89 c6 48 0f ba 6d 00 03 e9 d5 fe ff ff 0f 0b e9 db fd ff ff <0f> 0b 45 31 e4 e9 d1 fd ff ff 0f 0b e9 03 ff ff ff 0f 0b e9 d6 fe [ 453.139250] RSP: 0018:ffffc90000c67b18 EFLAGS: 00010246 [ 453.139782] RAX: 0000000000000000 RBX: ffff888108a24000 RCX: 0000000000002000 [ 453.140521] RDX: 0000000000000001 RSI: 0000000000000000 RDI: ffff8881016d61c8 [ 453.141253] RBP: ffff8881016d61c8 R08: 0000000000000000 R09: 0000000000000000 [ 453.141985] R10: 0000000000000000 R11: 0000000008a24000 R12: 0000000000000001 [ 453.142709] R13: 0000000000000002 R14: 0000000000000000 R15: ffff888107db8c00 [ 453.143450] FS: 00007f44853d4c80(0000) GS:ffff8882f469b000(0000) knlGS:0000000000000000 [ 453.144276] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 453.144853] CR2: 00007f4487629228 CR3: 00000001016aa000 CR4: 00000000000406f0 [ 453.145594] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 453.146320] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 453.147061] Call Trace: [ 453.147336] [ 453.147579] ? tick_nohz_tick_stopped+0xd/0x30 [ 453.148067] ? xas_load+0x9/0xb0 [ 453.148435] ? xa_load+0x6f/0xb0 [ 453.148781] __xe_vm_bind_ioctl+0xbd5/0x1500 [xe] [ 453.149338] ? dev_printk_emit+0x48/0x70 [ 453.149762] ? _dev_printk+0x57/0x80 [ 453.150148] ? drm_ioctl+0x17c/0x440 [ 453.150544] ? __drm_dev_vprintk+0x36/0x90 [ 453.150983] ? __pfx_xe_vm_bind_ioctl+0x10/0x10 [xe] [ 453.151575] ? drm_ioctl_kernel+0x9f/0xf0 [ 453.151998] ? __pfx_xe_vm_bind_ioctl+0x10/0x10 [xe] [ 453.152560] drm_ioctl_kernel+0x9f/0xf0 [ 453.152968] drm_ioctl+0x20f/0x440 [ 453.153332] ? __pfx_xe_vm_bind_ioctl+0x10/0x10 [xe] [ 453.153893] ? ioctl_has_perm.constprop.0.isra.0+0xae/0x100 [ 453.154489] ? memory_bm_test_bit+0x5/0x60 [ 453.154935] xe_drm_ioctl+0x47/0x70 [xe] [ 453.155419] __x64_sys_ioctl+0x8d/0xc0 [ 453.155824] do_syscall_64+0x47/0x110 [ 453.156228] entry_SYSCALL_64_after_hwframe+0x76/0x7e " v2 (Matt): refine commit message to have more details add Fixes tag move the code to xe_svm.h which already have the config remove a blank line per codestyle suggestion Fixes: 63f6e480d115 ("drm/xe: Add SVM garbage collector") Cc: Matthew Brost Signed-off-by: Shuicheng Lin Reviewed-by: Matthew Brost Signed-off-by: Matthew Brost Link: https://lore.kernel.org/r/20250502170052.1787973-1-shuicheng.lin@intel.com (cherry picked from commit 9d80698bcd97a5ad1088bcbb055e73fd068895e2) Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/xe_svm.c | 12 ++++++++++++ drivers/gpu/drm/xe/xe_svm.h | 8 ++++++++ drivers/gpu/drm/xe/xe_vm.c | 3 +-- 3 files changed, 21 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c index 0b6547c06961..24c578e1170e 100644 --- a/drivers/gpu/drm/xe/xe_svm.c +++ b/drivers/gpu/drm/xe/xe_svm.c @@ -947,3 +947,15 @@ int xe_devm_add(struct xe_tile *tile, struct xe_vram_region *vr) return 0; } #endif + +/** + * xe_svm_flush() - SVM flush + * @vm: The VM. + * + * Flush all SVM actions. + */ +void xe_svm_flush(struct xe_vm *vm) +{ + if (xe_vm_in_fault_mode(vm)) + flush_work(&vm->svm.garbage_collector.work); +} diff --git a/drivers/gpu/drm/xe/xe_svm.h b/drivers/gpu/drm/xe/xe_svm.h index e059590e5076..be306fe7aaa4 100644 --- a/drivers/gpu/drm/xe/xe_svm.h +++ b/drivers/gpu/drm/xe/xe_svm.h @@ -72,6 +72,9 @@ bool xe_svm_has_mapping(struct xe_vm *vm, u64 start, u64 end); int xe_svm_bo_evict(struct xe_bo *bo); void xe_svm_range_debug(struct xe_svm_range *range, const char *operation); + +void xe_svm_flush(struct xe_vm *vm); + #else static inline bool xe_svm_range_pages_valid(struct xe_svm_range *range) { @@ -124,6 +127,11 @@ static inline void xe_svm_range_debug(struct xe_svm_range *range, const char *operation) { } + +static inline void xe_svm_flush(struct xe_vm *vm) +{ +} + #endif /** diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c index 60303998bd61..367c84b90e9e 100644 --- a/drivers/gpu/drm/xe/xe_vm.c +++ b/drivers/gpu/drm/xe/xe_vm.c @@ -3312,8 +3312,7 @@ int xe_vm_bind_ioctl(struct drm_device *dev, void *data, struct drm_file *file) } /* Ensure all UNMAPs visible */ - if (xe_vm_in_fault_mode(vm)) - flush_work(&vm->svm.garbage_collector.work); + xe_svm_flush(vm); err = down_write_killable(&vm->lock); if (err) From 6b3ab7f2cbfaeb6580709cd8ef4d72cfd01bfde4 Mon Sep 17 00:00:00 2001 From: Nathan Chancellor Date: Wed, 7 May 2025 21:47:45 +0100 Subject: [PATCH 336/555] net: qede: Initialize qede_ll_ops with designated initializer After a recent change [1] in clang's randstruct implementation to randomize structures that only contain function pointers, there is an error because qede_ll_ops get randomized but does not use a designated initializer for the first member: drivers/net/ethernet/qlogic/qede/qede_main.c:206:2: error: a randomized struct can only be initialized with a designated initializer 206 | { | ^ Explicitly initialize the common member using a designated initializer to fix the build. Cc: stable@vger.kernel.org Fixes: 035f7f87b729 ("randstruct: Enable Clang support") Link: https://github.com/llvm/llvm-project/commit/04364fb888eea6db9811510607bed4b200bcb082 [1] Signed-off-by: Nathan Chancellor Link: https://patch.msgid.link/20250507-qede-fix-clang-randstruct-v1-1-5ccc15626fba@kernel.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/qlogic/qede/qede_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c index 99df00c30b8c..b5d744d2586f 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_main.c +++ b/drivers/net/ethernet/qlogic/qede/qede_main.c @@ -203,7 +203,7 @@ static struct pci_driver qede_pci_driver = { }; static struct qed_eth_cb_ops qede_ll_ops = { - { + .common = { #ifdef CONFIG_RFS_ACCEL .arfs_filter_op = qede_arfs_filter_op, #endif From da8bf5daa5e55a6af2b285ecda460d6454712ff4 Mon Sep 17 00:00:00 2001 From: Tom Lendacky Date: Thu, 8 May 2025 12:24:10 -0500 Subject: [PATCH 337/555] memblock: Accept allocated memory before use in memblock_double_array() When increasing the array size in memblock_double_array() and the slab is not yet available, a call to memblock_find_in_range() is used to reserve/allocate memory. However, the range returned may not have been accepted, which can result in a crash when booting an SNP guest: RIP: 0010:memcpy_orig+0x68/0x130 Code: ... RSP: 0000:ffffffff9cc03ce8 EFLAGS: 00010006 RAX: ff11001ff83e5000 RBX: 0000000000000000 RCX: fffffffffffff000 RDX: 0000000000000bc0 RSI: ffffffff9dba8860 RDI: ff11001ff83e5c00 RBP: 0000000000002000 R08: 0000000000000000 R09: 0000000000002000 R10: 000000207fffe000 R11: 0000040000000000 R12: ffffffff9d06ef78 R13: ff11001ff83e5000 R14: ffffffff9dba7c60 R15: 0000000000000c00 memblock_double_array+0xff/0x310 memblock_add_range+0x1fb/0x2f0 memblock_reserve+0x4f/0xa0 memblock_alloc_range_nid+0xac/0x130 memblock_alloc_internal+0x53/0xc0 memblock_alloc_try_nid+0x3d/0xa0 swiotlb_init_remap+0x149/0x2f0 mem_init+0xb/0xb0 mm_core_init+0x8f/0x350 start_kernel+0x17e/0x5d0 x86_64_start_reservations+0x14/0x30 x86_64_start_kernel+0x92/0xa0 secondary_startup_64_no_verify+0x194/0x19b Mitigate this by calling accept_memory() on the memory range returned before the slab is available. Prior to v6.12, the accept_memory() interface used a 'start' and 'end' parameter instead of 'start' and 'size', therefore the accept_memory() call must be adjusted to specify 'start + size' for 'end' when applying to kernels prior to v6.12. Cc: stable@vger.kernel.org # see patch description, needs adjustments for <= 6.11 Fixes: dcdfdd40fa82 ("mm: Add support for unaccepted memory") Signed-off-by: Tom Lendacky Link: https://lore.kernel.org/r/da1ac73bf4ded761e21b4e4bb5178382a580cd73.1746725050.git.thomas.lendacky@amd.com Signed-off-by: Mike Rapoport (Microsoft) --- mm/memblock.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/mm/memblock.c b/mm/memblock.c index d3509414b8c3..0e9ebb8aa7fe 100644 --- a/mm/memblock.c +++ b/mm/memblock.c @@ -457,7 +457,14 @@ static int __init_memblock memblock_double_array(struct memblock_type *type, min(new_area_start, memblock.current_limit), new_alloc_size, PAGE_SIZE); - new_array = addr ? __va(addr) : NULL; + if (addr) { + /* The memory may not have been accepted, yet. */ + accept_memory(addr, new_alloc_size); + + new_array = __va(addr); + } else { + new_array = NULL; + } } if (!addr) { pr_err("memblock: Failed to double %s array from %ld to %ld entries !\n", From 732b87a409667a370b87955c518e5d004de740b5 Mon Sep 17 00:00:00 2001 From: Imre Deak Date: Wed, 7 May 2025 18:19:53 +0300 Subject: [PATCH 338/555] drm/i915/dp: Fix determining SST/MST mode during MTP TU state computation Determining the SST/MST mode during state computation must be done based on the output type stored in the CRTC state, which in turn is set once based on the modeset connector's SST vs. MST type and will not change as long as the connector is using the CRTC. OTOH the MST mode indicated by the given connector's intel_dp::is_mst flag can change independently of the above output type, based on what sink is at any moment plugged to the connector. Fix the state computation accordingly. Cc: Jani Nikula Fixes: f6971d7427c2 ("drm/i915/mst: adapt intel_dp_mtp_tu_compute_config() for 128b/132b SST") Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/4607 Reviewed-by: Jani Nikula Signed-off-by: Imre Deak Link: https://lore.kernel.org/r/20250507151953.251846-1-imre.deak@intel.com (cherry picked from commit 0f45696ddb2b901fbf15cb8d2e89767be481d59f) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/display/intel_dp_mst.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/display/intel_dp_mst.c b/drivers/gpu/drm/i915/display/intel_dp_mst.c index 02f95108c637..6dc2d31ccb5a 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_mst.c +++ b/drivers/gpu/drm/i915/display/intel_dp_mst.c @@ -242,7 +242,7 @@ int intel_dp_mtp_tu_compute_config(struct intel_dp *intel_dp, to_intel_connector(conn_state->connector); const struct drm_display_mode *adjusted_mode = &crtc_state->hw.adjusted_mode; - bool is_mst = intel_dp->is_mst; + bool is_mst = intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DP_MST); int bpp_x16, slots = -EINVAL; int dsc_slice_count = 0; int max_dpt_bpp_x16; From 9379508f0661a5042ab7e0234e604038c2d63f83 Mon Sep 17 00:00:00 2001 From: Ahmad Fatoum Date: Tue, 22 Apr 2025 09:12:35 +0200 Subject: [PATCH 339/555] arm64: dts: imx8mp: use 800MHz NoC OPP for nominal drive mode When running in nominal drive mode, the maximum allowed frequency for the NoC is 800MHz, but the OPP table for the i.MX8MP interconnect device listed the 1GHz operating point for the NoC, regardless of the active mode. The newly introduced imx8mp-nominal.dtsi header reconfigures the clock controller to observe nominal drive mode limits, so have it modify the maximum NoC OPP as well. Fixes: 255fbd9eabe7 ("arm64: dts: imx8mp: Add optional nominal drive mode DTSI") Signed-off-by: Ahmad Fatoum Signed-off-by: Shawn Guo --- arch/arm64/boot/dts/freescale/imx8mp-nominal.dtsi | 2 ++ arch/arm64/boot/dts/freescale/imx8mp.dtsi | 6 ++++++ 2 files changed, 8 insertions(+) diff --git a/arch/arm64/boot/dts/freescale/imx8mp-nominal.dtsi b/arch/arm64/boot/dts/freescale/imx8mp-nominal.dtsi index dc0ccd723c6d..2ce1860b244d 100644 --- a/arch/arm64/boot/dts/freescale/imx8mp-nominal.dtsi +++ b/arch/arm64/boot/dts/freescale/imx8mp-nominal.dtsi @@ -88,3 +88,5 @@ <0>, <0>, <400000000>, <1039500000>; }; + +/delete-node/ &{noc_opp_table/opp-1000000000}; diff --git a/arch/arm64/boot/dts/freescale/imx8mp.dtsi b/arch/arm64/boot/dts/freescale/imx8mp.dtsi index ce6793b2d57e..7c1c87eab54c 100644 --- a/arch/arm64/boot/dts/freescale/imx8mp.dtsi +++ b/arch/arm64/boot/dts/freescale/imx8mp.dtsi @@ -1645,6 +1645,12 @@ opp-hz = /bits/ 64 <200000000>; }; + /* Nominal drive mode maximum */ + opp-800000000 { + opp-hz = /bits/ 64 <800000000>; + }; + + /* Overdrive mode maximum */ opp-1000000000 { opp-hz = /bits/ 64 <1000000000>; }; From 04679f3c27e132c1a2d3881de2f0c5d7128de7c1 Mon Sep 17 00:00:00 2001 From: Jeremy Bongio Date: Wed, 7 May 2025 12:30:10 +0000 Subject: [PATCH 340/555] fs: Remove redundant errseq_set call in mark_buffer_write_io_error. mark_buffer_write_io_error sets sb->s_wb_err to -EIO twice. Once in mapping_set_error and once in errseq_set. Only mapping_set_error checks if bh->b_assoc_map->host is NULL. Discovered during null pointer dereference during writeback to a failing device: [] ? mark_buffer_write_io_error+0x98/0xc0 [] ? mark_buffer_write_io_error+0x8e/0xc0 [] end_buffer_async_write+0x90/0xd0 [] end_bio_bh_io_sync+0x2b/0x40 [] blk_update_request+0x1b6/0x480 [] blk_mq_end_request+0x18/0x30 [] blk_mq_dispatch_rq_list+0x4da/0x8e0 [] __blk_mq_sched_dispatch_requests+0x218/0x6a0 [] blk_mq_sched_dispatch_requests+0x3a/0x80 [] blk_mq_run_hw_queue+0x108/0x330 [] blk_mq_flush_plug_list+0x178/0x5f0 [] __blk_flush_plug+0x41/0x120 [] blk_finish_plug+0x22/0x40 [] wb_writeback+0x150/0x280 [] ? set_worker_desc+0x9f/0xc0 [] wb_workfn+0x24e/0x4a0 Fixes: 485e9605c0573 ("fs/buffer.c: record blockdev write errors in super_block that it backs") Signed-off-by: Jeremy Bongio Link: https://lore.kernel.org/20250507123010.1228243-1-jbongio@google.com Reviewed-by: Christoph Hellwig Signed-off-by: Christian Brauner --- fs/buffer.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/fs/buffer.c b/fs/buffer.c index 7be23ff20b27..7ba1807145aa 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -1220,10 +1220,8 @@ void mark_buffer_write_io_error(struct buffer_head *bh) /* FIXME: do we need to set this in both places? */ if (bh->b_folio && bh->b_folio->mapping) mapping_set_error(bh->b_folio->mapping, -EIO); - if (bh->b_assoc_map) { + if (bh->b_assoc_map) mapping_set_error(bh->b_assoc_map, -EIO); - errseq_set(&bh->b_assoc_map->host->i_sb->s_wb_err, -EIO); - } } EXPORT_SYMBOL(mark_buffer_write_io_error); From 2d3cbfd6d54a2c39ce3244f33f85c595844bd7b8 Mon Sep 17 00:00:00 2001 From: Cong Wang Date: Tue, 6 May 2025 21:35:58 -0700 Subject: [PATCH 341/555] net_sched: Flush gso_skb list too during ->change() Previously, when reducing a qdisc's limit via the ->change() operation, only the main skb queue was trimmed, potentially leaving packets in the gso_skb list. This could result in NULL pointer dereference when we only check sch->limit against sch->q.qlen. This patch introduces a new helper, qdisc_dequeue_internal(), which ensures both the gso_skb list and the main queue are properly flushed when trimming excess packets. All relevant qdiscs (codel, fq, fq_codel, fq_pie, hhf, pie) are updated to use this helper in their ->change() routines. Fixes: 76e3cc126bb2 ("codel: Controlled Delay AQM") Fixes: 4b549a2ef4be ("fq_codel: Fair Queue Codel AQM") Fixes: afe4fd062416 ("pkt_sched: fq: Fair Queue packet scheduler") Fixes: ec97ecf1ebe4 ("net: sched: add Flow Queue PIE packet scheduler") Fixes: 10239edf86f1 ("net-qdisc-hhf: Heavy-Hitter Filter (HHF) qdisc") Fixes: d4b36210c2e6 ("net: pkt_sched: PIE AQM scheme") Reported-by: Will Reported-by: Savy Signed-off-by: Cong Wang Signed-off-by: David S. Miller --- include/net/sch_generic.h | 15 +++++++++++++++ net/sched/sch_codel.c | 2 +- net/sched/sch_fq.c | 2 +- net/sched/sch_fq_codel.c | 2 +- net/sched/sch_fq_pie.c | 2 +- net/sched/sch_hhf.c | 2 +- net/sched/sch_pie.c | 2 +- 7 files changed, 21 insertions(+), 6 deletions(-) diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index d48c657191cd..1c05fed05f2b 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -1031,6 +1031,21 @@ static inline struct sk_buff *__qdisc_dequeue_head(struct qdisc_skb_head *qh) return skb; } +static inline struct sk_buff *qdisc_dequeue_internal(struct Qdisc *sch, bool direct) +{ + struct sk_buff *skb; + + skb = __skb_dequeue(&sch->gso_skb); + if (skb) { + sch->q.qlen--; + return skb; + } + if (direct) + return __qdisc_dequeue_head(&sch->q); + else + return sch->dequeue(sch); +} + static inline struct sk_buff *qdisc_dequeue_head(struct Qdisc *sch) { struct sk_buff *skb = __qdisc_dequeue_head(&sch->q); diff --git a/net/sched/sch_codel.c b/net/sched/sch_codel.c index 12dd71139da3..c93761040c6e 100644 --- a/net/sched/sch_codel.c +++ b/net/sched/sch_codel.c @@ -144,7 +144,7 @@ static int codel_change(struct Qdisc *sch, struct nlattr *opt, qlen = sch->q.qlen; while (sch->q.qlen > sch->limit) { - struct sk_buff *skb = __qdisc_dequeue_head(&sch->q); + struct sk_buff *skb = qdisc_dequeue_internal(sch, true); dropped += qdisc_pkt_len(skb); qdisc_qstats_backlog_dec(sch, skb); diff --git a/net/sched/sch_fq.c b/net/sched/sch_fq.c index 2ca5332cfcc5..902ff5470607 100644 --- a/net/sched/sch_fq.c +++ b/net/sched/sch_fq.c @@ -1136,7 +1136,7 @@ static int fq_change(struct Qdisc *sch, struct nlattr *opt, sch_tree_lock(sch); } while (sch->q.qlen > sch->limit) { - struct sk_buff *skb = fq_dequeue(sch); + struct sk_buff *skb = qdisc_dequeue_internal(sch, false); if (!skb) break; diff --git a/net/sched/sch_fq_codel.c b/net/sched/sch_fq_codel.c index 6c9029f71e88..2a0f3a513bfa 100644 --- a/net/sched/sch_fq_codel.c +++ b/net/sched/sch_fq_codel.c @@ -441,7 +441,7 @@ static int fq_codel_change(struct Qdisc *sch, struct nlattr *opt, while (sch->q.qlen > sch->limit || q->memory_usage > q->memory_limit) { - struct sk_buff *skb = fq_codel_dequeue(sch); + struct sk_buff *skb = qdisc_dequeue_internal(sch, false); q->cstats.drop_len += qdisc_pkt_len(skb); rtnl_kfree_skbs(skb, skb); diff --git a/net/sched/sch_fq_pie.c b/net/sched/sch_fq_pie.c index f3b8203d3e85..df7fac95ab15 100644 --- a/net/sched/sch_fq_pie.c +++ b/net/sched/sch_fq_pie.c @@ -366,7 +366,7 @@ static int fq_pie_change(struct Qdisc *sch, struct nlattr *opt, /* Drop excess packets if new limit is lower */ while (sch->q.qlen > sch->limit) { - struct sk_buff *skb = fq_pie_qdisc_dequeue(sch); + struct sk_buff *skb = qdisc_dequeue_internal(sch, false); len_dropped += qdisc_pkt_len(skb); num_dropped += 1; diff --git a/net/sched/sch_hhf.c b/net/sched/sch_hhf.c index 44d9efe1a96a..5aa434b46707 100644 --- a/net/sched/sch_hhf.c +++ b/net/sched/sch_hhf.c @@ -564,7 +564,7 @@ static int hhf_change(struct Qdisc *sch, struct nlattr *opt, qlen = sch->q.qlen; prev_backlog = sch->qstats.backlog; while (sch->q.qlen > sch->limit) { - struct sk_buff *skb = hhf_dequeue(sch); + struct sk_buff *skb = qdisc_dequeue_internal(sch, false); rtnl_kfree_skbs(skb, skb); } diff --git a/net/sched/sch_pie.c b/net/sched/sch_pie.c index 3771d000b30d..ff49a6c97033 100644 --- a/net/sched/sch_pie.c +++ b/net/sched/sch_pie.c @@ -195,7 +195,7 @@ static int pie_change(struct Qdisc *sch, struct nlattr *opt, /* Drop excess packets if new limit is lower */ qlen = sch->q.qlen; while (sch->q.qlen > sch->limit) { - struct sk_buff *skb = __qdisc_dequeue_head(&sch->q); + struct sk_buff *skb = qdisc_dequeue_internal(sch, true); dropped += qdisc_pkt_len(skb); qdisc_qstats_backlog_dec(sch, skb); From 16ce349b15069334710faa10f2e09866f8391f26 Mon Sep 17 00:00:00 2001 From: Cong Wang Date: Tue, 6 May 2025 21:35:59 -0700 Subject: [PATCH 342/555] selftests/tc-testing: Add qdisc limit trimming tests Added new test cases for FQ, FQ_CODEL, FQ_PIE, and HHF qdiscs to verify queue trimming behavior when the qdisc limit is dynamically reduced. Each test injects packets, reduces the qdisc limit, and checks that the new limit is enforced. This is still best effort since timing qdisc backlog is not easy. Signed-off-by: Cong Wang Signed-off-by: David S. Miller --- .../tc-testing/tc-tests/qdiscs/codel.json | 24 +++++++++++++++++++ .../tc-testing/tc-tests/qdiscs/fq.json | 22 +++++++++++++++++ .../tc-testing/tc-tests/qdiscs/fq_codel.json | 22 +++++++++++++++++ .../tc-testing/tc-tests/qdiscs/fq_pie.json | 22 +++++++++++++++++ .../tc-testing/tc-tests/qdiscs/hhf.json | 22 +++++++++++++++++ .../tc-testing/tc-tests/qdiscs/pie.json | 24 +++++++++++++++++++ 6 files changed, 136 insertions(+) create mode 100644 tools/testing/selftests/tc-testing/tc-tests/qdiscs/pie.json diff --git a/tools/testing/selftests/tc-testing/tc-tests/qdiscs/codel.json b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/codel.json index e9469ee71e6f..6d515d0e5ed6 100644 --- a/tools/testing/selftests/tc-testing/tc-tests/qdiscs/codel.json +++ b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/codel.json @@ -189,5 +189,29 @@ "teardown": [ "$TC qdisc del dev $DUMMY handle 1: root" ] + }, + { + "id": "deb1", + "name": "CODEL test qdisc limit trimming", + "category": ["qdisc", "codel"], + "plugins": { + "requires": ["nsPlugin", "scapyPlugin"] + }, + "setup": [ + "$TC qdisc add dev $DEV1 handle 1: root codel limit 10" + ], + "scapy": [ + { + "iface": "$DEV0", + "count": 10, + "packet": "Ether(type=0x800)/IP(src='10.0.0.10',dst='10.0.0.20')/TCP(sport=5000,dport=10)" + } + ], + "cmdUnderTest": "$TC qdisc change dev $DEV1 handle 1: root codel limit 1", + "expExitCode": "0", + "verifyCmd": "$TC qdisc show dev $DEV1", + "matchPattern": "qdisc codel 1: root refcnt [0-9]+ limit 1p target 5ms interval 100ms", + "matchCount": "1", + "teardown": ["$TC qdisc del dev $DEV1 handle 1: root"] } ] diff --git a/tools/testing/selftests/tc-testing/tc-tests/qdiscs/fq.json b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/fq.json index 3a537b2ec4c9..24faf4e12dfa 100644 --- a/tools/testing/selftests/tc-testing/tc-tests/qdiscs/fq.json +++ b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/fq.json @@ -377,5 +377,27 @@ "teardown": [ "$TC qdisc del dev $DUMMY handle 1: root" ] + }, + { + "id": "9479", + "name": "FQ test qdisc limit trimming", + "category": ["qdisc", "fq"], + "plugins": {"requires": ["nsPlugin", "scapyPlugin"]}, + "setup": [ + "$TC qdisc add dev $DEV1 handle 1: root fq limit 10" + ], + "scapy": [ + { + "iface": "$DEV0", + "count": 10, + "packet": "Ether(type=0x800)/IP(src='10.0.0.10',dst='10.0.0.20')/TCP(sport=5000,dport=10)" + } + ], + "cmdUnderTest": "$TC qdisc change dev $DEV1 handle 1: root fq limit 1", + "expExitCode": "0", + "verifyCmd": "$TC qdisc show dev $DEV1", + "matchPattern": "qdisc fq 1: root refcnt [0-9]+ limit 1p", + "matchCount": "1", + "teardown": ["$TC qdisc del dev $DEV1 handle 1: root"] } ] diff --git a/tools/testing/selftests/tc-testing/tc-tests/qdiscs/fq_codel.json b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/fq_codel.json index 9774b1e8801b..4ce62b857fd7 100644 --- a/tools/testing/selftests/tc-testing/tc-tests/qdiscs/fq_codel.json +++ b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/fq_codel.json @@ -294,5 +294,27 @@ "teardown": [ "$TC qdisc del dev $DUMMY handle 1: root" ] + }, + { + "id": "0436", + "name": "FQ_CODEL test qdisc limit trimming", + "category": ["qdisc", "fq_codel"], + "plugins": {"requires": ["nsPlugin", "scapyPlugin"]}, + "setup": [ + "$TC qdisc add dev $DEV1 handle 1: root fq_codel limit 10" + ], + "scapy": [ + { + "iface": "$DEV0", + "count": 10, + "packet": "Ether(type=0x800)/IP(src='10.0.0.10',dst='10.0.0.20')/TCP(sport=5000,dport=10)" + } + ], + "cmdUnderTest": "$TC qdisc change dev $DEV1 handle 1: root fq_codel limit 1", + "expExitCode": "0", + "verifyCmd": "$TC qdisc show dev $DEV1", + "matchPattern": "qdisc fq_codel 1: root refcnt [0-9]+ limit 1p flows 1024 quantum.*target 5ms interval 100ms memory_limit 32Mb ecn drop_batch 64", + "matchCount": "1", + "teardown": ["$TC qdisc del dev $DEV1 handle 1: root"] } ] diff --git a/tools/testing/selftests/tc-testing/tc-tests/qdiscs/fq_pie.json b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/fq_pie.json index d012d88d67fe..229fe1bf4a90 100644 --- a/tools/testing/selftests/tc-testing/tc-tests/qdiscs/fq_pie.json +++ b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/fq_pie.json @@ -18,5 +18,27 @@ "matchCount": "1", "teardown": [ ] + }, + { + "id": "83bf", + "name": "FQ_PIE test qdisc limit trimming", + "category": ["qdisc", "fq_pie"], + "plugins": {"requires": ["nsPlugin", "scapyPlugin"]}, + "setup": [ + "$TC qdisc add dev $DEV1 handle 1: root fq_pie limit 10" + ], + "scapy": [ + { + "iface": "$DEV0", + "count": 10, + "packet": "Ether(type=0x800)/IP(src='10.0.0.10',dst='10.0.0.20')/TCP(sport=5000,dport=10)" + } + ], + "cmdUnderTest": "$TC qdisc change dev $DEV1 handle 1: root fq_pie limit 1", + "expExitCode": "0", + "verifyCmd": "$TC qdisc show dev $DEV1", + "matchPattern": "qdisc fq_pie 1: root refcnt [0-9]+ limit 1p", + "matchCount": "1", + "teardown": ["$TC qdisc del dev $DEV1 handle 1: root"] } ] diff --git a/tools/testing/selftests/tc-testing/tc-tests/qdiscs/hhf.json b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/hhf.json index dbef5474b26b..0ca19fac54a5 100644 --- a/tools/testing/selftests/tc-testing/tc-tests/qdiscs/hhf.json +++ b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/hhf.json @@ -188,5 +188,27 @@ "teardown": [ "$TC qdisc del dev $DUMMY handle 1: root" ] + }, + { + "id": "385f", + "name": "HHF test qdisc limit trimming", + "category": ["qdisc", "hhf"], + "plugins": {"requires": ["nsPlugin", "scapyPlugin"]}, + "setup": [ + "$TC qdisc add dev $DEV1 handle 1: root hhf limit 10" + ], + "scapy": [ + { + "iface": "$DEV0", + "count": 10, + "packet": "Ether(type=0x800)/IP(src='10.0.0.10',dst='10.0.0.20')/TCP(sport=5000,dport=10)" + } + ], + "cmdUnderTest": "$TC qdisc change dev $DEV1 handle 1: root hhf limit 1", + "expExitCode": "0", + "verifyCmd": "$TC qdisc show dev $DEV1", + "matchPattern": "qdisc hhf 1: root refcnt [0-9]+ limit 1p.*hh_limit 2048 reset_timeout 40ms admit_bytes 128Kb evict_timeout 1s non_hh_weight 2", + "matchCount": "1", + "teardown": ["$TC qdisc del dev $DEV1 handle 1: root"] } ] diff --git a/tools/testing/selftests/tc-testing/tc-tests/qdiscs/pie.json b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/pie.json new file mode 100644 index 000000000000..1a98b66e8030 --- /dev/null +++ b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/pie.json @@ -0,0 +1,24 @@ +[ + { + "id": "6158", + "name": "PIE test qdisc limit trimming", + "category": ["qdisc", "pie"], + "plugins": {"requires": ["nsPlugin", "scapyPlugin"]}, + "setup": [ + "$TC qdisc add dev $DEV1 handle 1: root pie limit 10" + ], + "scapy": [ + { + "iface": "$DEV0", + "count": 10, + "packet": "Ether(type=0x800)/IP(src='10.0.0.10',dst='10.0.0.20')/TCP(sport=5000,dport=10)" + } + ], + "cmdUnderTest": "$TC qdisc change dev $DEV1 handle 1: root pie limit 1", + "expExitCode": "0", + "verifyCmd": "$TC qdisc show dev $DEV1", + "matchPattern": "qdisc pie 1: root refcnt [0-9]+ limit 1p", + "matchCount": "1", + "teardown": ["$TC qdisc del dev $DEV1 handle 1: root"] + } +] From 92835cebab120f8a5f023a26a792a2ac3f816c4f Mon Sep 17 00:00:00 2001 From: Gabriel Krisman Bertazi Date: Thu, 8 May 2025 14:12:03 -0400 Subject: [PATCH 343/555] io_uring/sqpoll: Increase task_work submission batch size Our QA team reported a 10%-23%, throughput reduction on an io_uring sqpoll testcase doing IO to a null_blk, that I traced back to a reduction of the device submission queue depth utilization. It turns out that, after commit af5d68f8892f ("io_uring/sqpoll: manage task_work privately"), we capped the number of task_work entries that can be completed from a single spin of sqpoll to only 8 entries, before the sqpoll goes around to (potentially) sleep. While this cap doesn't drive the submission side directly, it impacts the completion behavior, which affects the number of IO queued by fio per sqpoll cycle on the submission side, and io_uring ends up seeing less ios per sqpoll cycle. As a result, block layer plugging is less effective, and we see more time spent inside the block layer in profilings charts, and increased submission latency measured by fio. There are other places that have increased overhead once sqpoll sleeps more often, such as the sqpoll utilization calculation. But, in this microbenchmark, those were not representative enough in perf charts, and their removal didn't yield measurable changes in throughput. The major overhead comes from the fact we plug less, and less often, when submitting to the block layer. My benchmark is: fio --ioengine=io_uring --direct=1 --iodepth=128 --runtime=300 --bs=4k \ --invalidate=1 --time_based --ramp_time=10 --group_reporting=1 \ --filename=/dev/nullb0 --name=RandomReads-direct-nullb-sqpoll-4k-1 \ --rw=randread --numjobs=1 --sqthread_poll In one machine, tested on top of Linux 6.15-rc1, we have the following baseline: READ: bw=4994MiB/s (5236MB/s), 4994MiB/s-4994MiB/s (5236MB/s-5236MB/s), io=439GiB (471GB), run=90001-90001msec With this patch: READ: bw=5762MiB/s (6042MB/s), 5762MiB/s-5762MiB/s (6042MB/s-6042MB/s), io=506GiB (544GB), run=90001-90001msec which is a 15% improvement in measured bandwidth. The average submission latency is noticeably lowered too. As measured by fio: Baseline: lat (usec): min=20, max=241, avg=99.81, stdev=3.38 Patched: lat (usec): min=26, max=226, avg=86.48, stdev=4.82 If we look at blktrace, we can also see the plugging behavior is improved. In the baseline, we end up limited to plugging 8 requests in the block layer regardless of the device queue depth size, while after patching we can drive more io, and we manage to utilize the full device queue. In the baseline, after a stabilization phase, an ordinary submission looks like: 254,0 1 49942 0.016028795 5977 U N [iou-sqp-5976] 7 After patching, I see consistently more requests per unplug. 254,0 1 4996 0.001432872 3145 U N [iou-sqp-3144] 32 Ideally, the cap size would at least be the deep enough to fill the device queue, but we can't predict that behavior, or assume all IO goes to a single device, and thus can't guess the ideal batch size. We also don't want to let the tw run unbounded, though I'm not sure it would really be a problem. Instead, let's just give it a more sensible value that will allow for more efficient batching. I've tested with different cap values, and initially proposed to increase the cap to 1024. Jens argued it is too big of a bump and I observed that, with 32, I'm no longer able to observe this bottleneck in any of my machines. Fixes: af5d68f8892f ("io_uring/sqpoll: manage task_work privately") Signed-off-by: Gabriel Krisman Bertazi Link: https://lore.kernel.org/r/20250508181203.3785544-1-krisman@suse.de Signed-off-by: Jens Axboe --- io_uring/sqpoll.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/io_uring/sqpoll.c b/io_uring/sqpoll.c index d037cc68e9d3..03c699493b5a 100644 --- a/io_uring/sqpoll.c +++ b/io_uring/sqpoll.c @@ -20,7 +20,7 @@ #include "sqpoll.h" #define IORING_SQPOLL_CAP_ENTRIES_VALUE 8 -#define IORING_TW_CAP_ENTRIES_VALUE 8 +#define IORING_TW_CAP_ENTRIES_VALUE 32 enum { IO_SQ_THREAD_SHOULD_STOP = 0, From c6888983134e2ccc2db8ffd2720b0d4826d952e4 Mon Sep 17 00:00:00 2001 From: Himanshu Bhavani Date: Mon, 5 May 2025 11:28:27 +0530 Subject: [PATCH 344/555] arm64: dts: imx8mp-var-som: Fix LDO5 shutdown causing SD card timeout Fix SD card timeout issue caused by LDO5 regulator getting disabled after boot. The kernel log shows LDO5 being disabled, which leads to a timeout on USDHC2: [ 33.760561] LDO5: disabling [ 81.119861] mmc1: Timeout waiting for hardware interrupt. To prevent this, set regulator-boot-on and regulator-always-on for LDO5. Also add the vqmmc regulator to properly support 1.8V/3.3V signaling for USDHC2 using a GPIO-controlled regulator. Fixes: 6c2a1f4f71258 ("arm64: dts: imx8mp-var-som-symphony: Add Variscite Symphony board and VAR-SOM-MX8MP SoM") Signed-off-by: Himanshu Bhavani Acked-by: Tarang Raval Signed-off-by: Shawn Guo --- arch/arm64/boot/dts/freescale/imx8mp-var-som.dtsi | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/arch/arm64/boot/dts/freescale/imx8mp-var-som.dtsi b/arch/arm64/boot/dts/freescale/imx8mp-var-som.dtsi index b2ac2583a592..b59da91fdd04 100644 --- a/arch/arm64/boot/dts/freescale/imx8mp-var-som.dtsi +++ b/arch/arm64/boot/dts/freescale/imx8mp-var-som.dtsi @@ -35,7 +35,6 @@ <0x1 0x00000000 0 0xc0000000>; }; - reg_usdhc2_vmmc: regulator-usdhc2-vmmc { compatible = "regulator-fixed"; regulator-name = "VSD_3V3"; @@ -46,6 +45,16 @@ startup-delay-us = <100>; off-on-delay-us = <12000>; }; + + reg_usdhc2_vqmmc: regulator-usdhc2-vqmmc { + compatible = "regulator-gpio"; + regulator-name = "VSD_VSEL"; + regulator-min-microvolt = <1800000>; + regulator-max-microvolt = <3300000>; + gpios = <&gpio2 12 GPIO_ACTIVE_HIGH>; + states = <3300000 0x0 1800000 0x1>; + vin-supply = <&ldo5>; + }; }; &A53_0 { @@ -205,6 +214,7 @@ pinctrl-2 = <&pinctrl_usdhc2_200mhz>, <&pinctrl_usdhc2_gpio>; cd-gpios = <&gpio1 14 GPIO_ACTIVE_LOW>; vmmc-supply = <®_usdhc2_vmmc>; + vqmmc-supply = <®_usdhc2_vqmmc>; bus-width = <4>; status = "okay"; }; From fea4e317f9e7e1f449ce90dedc27a2d2a95bee5a Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Thu, 8 May 2025 15:41:32 -0700 Subject: [PATCH 345/555] x86/mm: Eliminate window where TLB flushes may be inadvertently skipped tl;dr: There is a window in the mm switching code where the new CR3 is set and the CPU should be getting TLB flushes for the new mm. But should_flush_tlb() has a bug and suppresses the flush. Fix it by widening the window where should_flush_tlb() sends an IPI. Long Version: === History === There were a few things leading up to this. First, updating mm_cpumask() was observed to be too expensive, so it was made lazier. But being lazy caused too many unnecessary IPIs to CPUs due to the now-lazy mm_cpumask(). So code was added to cull mm_cpumask() periodically[2]. But that culling was a bit too aggressive and skipped sending TLB flushes to CPUs that need them. So here we are again. === Problem === The too-aggressive code in should_flush_tlb() strikes in this window: // Turn on IPIs for this CPU/mm combination, but only // if should_flush_tlb() agrees: cpumask_set_cpu(cpu, mm_cpumask(next)); next_tlb_gen = atomic64_read(&next->context.tlb_gen); choose_new_asid(next, next_tlb_gen, &new_asid, &need_flush); load_new_mm_cr3(need_flush); // ^ After 'need_flush' is set to false, IPIs *MUST* // be sent to this CPU and not be ignored. this_cpu_write(cpu_tlbstate.loaded_mm, next); // ^ Not until this point does should_flush_tlb() // become true! should_flush_tlb() will suppress TLB flushes between load_new_mm_cr3() and writing to 'loaded_mm', which is a window where they should not be suppressed. Whoops. === Solution === Thankfully, the fuzzy "just about to write CR3" window is already marked with loaded_mm==LOADED_MM_SWITCHING. Simply checking for that state in should_flush_tlb() is sufficient to ensure that the CPU is targeted with an IPI. This will cause more TLB flush IPIs. But the window is relatively small and I do not expect this to cause any kind of measurable performance impact. Update the comment where LOADED_MM_SWITCHING is written since it grew yet another user. Peter Z also raised a concern that should_flush_tlb() might not observe 'loaded_mm' and 'is_lazy' in the same order that switch_mm_irqs_off() writes them. Add a barrier to ensure that they are observed in the order they are written. Signed-off-by: Dave Hansen Acked-by: Rik van Riel Link: https://lore.kernel.org/oe-lkp/202411282207.6bd28eae-lkp@intel.com/ [1] Fixes: 6db2526c1d69 ("x86/mm/tlb: Only trim the mm_cpumask once a second") [2] Reported-by: Stephen Dolan Cc: stable@vger.kernel.org Acked-by: Ingo Molnar Acked-by: Peter Zijlstra (Intel) Signed-off-by: Linus Torvalds --- arch/x86/mm/tlb.c | 22 +++++++++++++++++++--- 1 file changed, 19 insertions(+), 3 deletions(-) diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c index eb83348f9305..b6d6750e4bd1 100644 --- a/arch/x86/mm/tlb.c +++ b/arch/x86/mm/tlb.c @@ -899,8 +899,9 @@ void switch_mm_irqs_off(struct mm_struct *unused, struct mm_struct *next, cond_mitigation(tsk); /* - * Let nmi_uaccess_okay() and finish_asid_transition() - * know that CR3 is changing. + * Indicate that CR3 is about to change. nmi_uaccess_okay() + * and others are sensitive to the window where mm_cpumask(), + * CR3 and cpu_tlbstate.loaded_mm are not all in sync. */ this_cpu_write(cpu_tlbstate.loaded_mm, LOADED_MM_SWITCHING); barrier(); @@ -1204,8 +1205,16 @@ done: static bool should_flush_tlb(int cpu, void *data) { + struct mm_struct *loaded_mm = per_cpu(cpu_tlbstate.loaded_mm, cpu); struct flush_tlb_info *info = data; + /* + * Order the 'loaded_mm' and 'is_lazy' against their + * write ordering in switch_mm_irqs_off(). Ensure + * 'is_lazy' is at least as new as 'loaded_mm'. + */ + smp_rmb(); + /* Lazy TLB will get flushed at the next context switch. */ if (per_cpu(cpu_tlbstate_shared.is_lazy, cpu)) return false; @@ -1214,8 +1223,15 @@ static bool should_flush_tlb(int cpu, void *data) if (!info->mm) return true; + /* + * While switching, the remote CPU could have state from + * either the prev or next mm. Assume the worst and flush. + */ + if (loaded_mm == LOADED_MM_SWITCHING) + return true; + /* The target mm is loaded, and the CPU is not lazy. */ - if (per_cpu(cpu_tlbstate.loaded_mm, cpu) == info->mm) + if (loaded_mm == info->mm) return true; /* In cpumask, but not the loaded mm? Periodically remove by flushing. */ From 39b5ef791d109dd54c7c2e6e87933edfcc0ad1ac Mon Sep 17 00:00:00 2001 From: Waiman Long Date: Thu, 8 May 2025 15:24:13 -0400 Subject: [PATCH 346/555] cgroup/cpuset: Extend kthread_is_per_cpu() check to all PF_NO_SETAFFINITY tasks Commit ec5fbdfb99d1 ("cgroup/cpuset: Enable update_tasks_cpumask() on top_cpuset") enabled us to pull CPUs dedicated to child partitions from tasks in top_cpuset by ignoring per cpu kthreads. However, there can be other kthreads that are not per cpu but have PF_NO_SETAFFINITY flag set to indicate that we shouldn't mess with their CPU affinity. For other kthreads, their affinity will be changed to skip CPUs dedicated to child partitions whether it is an isolating or a scheduling one. As all the per cpu kthreads have PF_NO_SETAFFINITY set, the PF_NO_SETAFFINITY tasks are essentially a superset of per cpu kthreads. Fix this issue by dropping the kthread_is_per_cpu() check and checking the PF_NO_SETAFFINITY flag instead. Fixes: ec5fbdfb99d1 ("cgroup/cpuset: Enable update_tasks_cpumask() on top_cpuset") Signed-off-by: Waiman Long Acked-by: Frederic Weisbecker Signed-off-by: Tejun Heo --- kernel/cgroup/cpuset.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c index 306b60430091..24b70ea3e6ce 100644 --- a/kernel/cgroup/cpuset.c +++ b/kernel/cgroup/cpuset.c @@ -1116,9 +1116,11 @@ void cpuset_update_tasks_cpumask(struct cpuset *cs, struct cpumask *new_cpus) if (top_cs) { /* - * Percpu kthreads in top_cpuset are ignored + * PF_NO_SETAFFINITY tasks are ignored. + * All per cpu kthreads should have PF_NO_SETAFFINITY + * flag set, see kthread_set_per_cpu(). */ - if (kthread_is_per_cpu(task)) + if (task->flags & PF_NO_SETAFFINITY) continue; cpumask_andnot(new_cpus, possible_mask, subpartitions_cpus); } else { From eb0851e14432f3b87c77b704c835ac376deda03a Mon Sep 17 00:00:00 2001 From: I Hsin Cheng Date: Tue, 6 May 2025 02:43:38 +0800 Subject: [PATCH 347/555] drm/meson: Use 1000ULL when operating with mode->clock Coverity scan reported the usage of "mode->clock * 1000" may lead to integer overflow. Use "1000ULL" instead of "1000" when utilizing it to avoid potential integer overflow issue. Link: https://scan5.scan.coverity.com/#/project-view/10074/10063?selectedIssue=1646759 Signed-off-by: I Hsin Cheng Reviewed-by: Martin Blumenstingl Fixes: 1017560164b6 ("drm/meson: use unsigned long long / Hz for frequency types") Link: https://lore.kernel.org/r/20250505184338.678540-1-richard120310@gmail.com Signed-off-by: Neil Armstrong --- drivers/gpu/drm/meson/meson_encoder_hdmi.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/meson/meson_encoder_hdmi.c b/drivers/gpu/drm/meson/meson_encoder_hdmi.c index 7752d8ac85f0..c08fa93e50a3 100644 --- a/drivers/gpu/drm/meson/meson_encoder_hdmi.c +++ b/drivers/gpu/drm/meson/meson_encoder_hdmi.c @@ -75,7 +75,7 @@ static void meson_encoder_hdmi_set_vclk(struct meson_encoder_hdmi *encoder_hdmi, unsigned long long venc_freq; unsigned long long hdmi_freq; - vclk_freq = mode->clock * 1000; + vclk_freq = mode->clock * 1000ULL; /* For 420, pixel clock is half unlike venc clock */ if (encoder_hdmi->output_bus_fmt == MEDIA_BUS_FMT_UYYVYY8_0_5X24) @@ -123,7 +123,7 @@ static enum drm_mode_status meson_encoder_hdmi_mode_valid(struct drm_bridge *bri struct meson_encoder_hdmi *encoder_hdmi = bridge_to_meson_encoder_hdmi(bridge); struct meson_drm *priv = encoder_hdmi->priv; bool is_hdmi2_sink = display_info->hdmi.scdc.supported; - unsigned long long clock = mode->clock * 1000; + unsigned long long clock = mode->clock * 1000ULL; unsigned long long phy_freq; unsigned long long vclk_freq; unsigned long long venc_freq; From 1ac116ce6468670eeda39345a5585df308243dca Mon Sep 17 00:00:00 2001 From: Pawan Gupta Date: Fri, 11 Apr 2025 15:36:38 -0700 Subject: [PATCH 348/555] Documentation: x86/bugs/its: Add ITS documentation Add the admin-guide for Indirect Target Selection (ITS). Signed-off-by: Pawan Gupta Signed-off-by: Dave Hansen Reviewed-by: Josh Poimboeuf Reviewed-by: Alexandre Chartre --- Documentation/admin-guide/hw-vuln/index.rst | 1 + .../hw-vuln/indirect-target-selection.rst | 168 ++++++++++++++++++ 2 files changed, 169 insertions(+) create mode 100644 Documentation/admin-guide/hw-vuln/indirect-target-selection.rst diff --git a/Documentation/admin-guide/hw-vuln/index.rst b/Documentation/admin-guide/hw-vuln/index.rst index 451874b8135d..ce296b8430fc 100644 --- a/Documentation/admin-guide/hw-vuln/index.rst +++ b/Documentation/admin-guide/hw-vuln/index.rst @@ -23,3 +23,4 @@ are configurable at compile, boot or run time. gather_data_sampling reg-file-data-sampling rsb + indirect-target-selection diff --git a/Documentation/admin-guide/hw-vuln/indirect-target-selection.rst b/Documentation/admin-guide/hw-vuln/indirect-target-selection.rst new file mode 100644 index 000000000000..d9ca64108d23 --- /dev/null +++ b/Documentation/admin-guide/hw-vuln/indirect-target-selection.rst @@ -0,0 +1,168 @@ +.. SPDX-License-Identifier: GPL-2.0 + +Indirect Target Selection (ITS) +=============================== + +ITS is a vulnerability in some Intel CPUs that support Enhanced IBRS and were +released before Alder Lake. ITS may allow an attacker to control the prediction +of indirect branches and RETs located in the lower half of a cacheline. + +ITS is assigned CVE-2024-28956 with a CVSS score of 4.7 (Medium). + +Scope of Impact +--------------- +- **eIBRS Guest/Host Isolation**: Indirect branches in KVM/kernel may still be + predicted with unintended target corresponding to a branch in the guest. + +- **Intra-Mode BTI**: In-kernel training such as through cBPF or other native + gadgets. + +- **Indirect Branch Prediction Barrier (IBPB)**: After an IBPB, indirect + branches may still be predicted with targets corresponding to direct branches + executed prior to the IBPB. This is fixed by the IPU 2025.1 microcode, which + should be available via distro updates. Alternatively microcode can be + obtained from Intel's github repository [#f1]_. + +Affected CPUs +------------- +Below is the list of ITS affected CPUs [#f2]_ [#f3]_: + + ======================== ============ ==================== =============== + Common name Family_Model eIBRS Intra-mode BTI + Guest/Host Isolation + ======================== ============ ==================== =============== + SKYLAKE_X (step >= 6) 06_55H Affected Affected + ICELAKE_X 06_6AH Not affected Affected + ICELAKE_D 06_6CH Not affected Affected + ICELAKE_L 06_7EH Not affected Affected + TIGERLAKE_L 06_8CH Not affected Affected + TIGERLAKE 06_8DH Not affected Affected + KABYLAKE_L (step >= 12) 06_8EH Affected Affected + KABYLAKE (step >= 13) 06_9EH Affected Affected + COMETLAKE 06_A5H Affected Affected + COMETLAKE_L 06_A6H Affected Affected + ROCKETLAKE 06_A7H Not affected Affected + ======================== ============ ==================== =============== + +- All affected CPUs enumerate Enhanced IBRS feature. +- IBPB isolation is affected on all ITS affected CPUs, and need a microcode + update for mitigation. +- None of the affected CPUs enumerate BHI_CTRL which was introduced in Golden + Cove (Alder Lake and Sapphire Rapids). This can help guests to determine the + host's affected status. +- Intel Atom CPUs are not affected by ITS. + +Mitigation +---------- +As only the indirect branches and RETs that have their last byte of instruction +in the lower half of the cacheline are vulnerable to ITS, the basic idea behind +the mitigation is to not allow indirect branches in the lower half. + +This is achieved by relying on existing retpoline support in the kernel, and in +compilers. ITS-vulnerable retpoline sites are runtime patched to point to newly +added ITS-safe thunks. These safe thunks consists of indirect branch in the +second half of the cacheline. Not all retpoline sites are patched to thunks, if +a retpoline site is evaluated to be ITS-safe, it is replaced with an inline +indirect branch. + +Dynamic thunks +~~~~~~~~~~~~~~ +From a dynamically allocated pool of safe-thunks, each vulnerable site is +replaced with a new thunk, such that they get a unique address. This could +improve the branch prediction accuracy. Also, it is a defense-in-depth measure +against aliasing. + +Note, for simplicity, indirect branches in eBPF programs are always replaced +with a jump to a static thunk in __x86_indirect_its_thunk_array. If required, +in future this can be changed to use dynamic thunks. + +All vulnerable RETs are replaced with a static thunk, they do not use dynamic +thunks. This is because RETs get their prediction from RSB mostly that does not +depend on source address. RETs that underflow RSB may benefit from dynamic +thunks. But, RETs significantly outnumber indirect branches, and any benefit +from a unique source address could be outweighed by the increased icache +footprint and iTLB pressure. + +Retpoline +~~~~~~~~~ +Retpoline sequence also mitigates ITS-unsafe indirect branches. For this +reason, when retpoline is enabled, ITS mitigation only relocates the RETs to +safe thunks. Unless user requested the RSB-stuffing mitigation. + +RSB Stuffing +~~~~~~~~~~~~ +RSB-stuffing via Call Depth Tracking is a mitigation for Retbleed RSB-underflow +attacks. And it also mitigates RETs that are vulnerable to ITS. + +Mitigation in guests +^^^^^^^^^^^^^^^^^^^^ +All guests deploy ITS mitigation by default, irrespective of eIBRS enumeration +and Family/Model of the guest. This is because eIBRS feature could be hidden +from a guest. One exception to this is when a guest enumerates BHI_DIS_S, which +indicates that the guest is running on an unaffected host. + +To prevent guests from unnecessarily deploying the mitigation on unaffected +platforms, Intel has defined ITS_NO bit(62) in MSR IA32_ARCH_CAPABILITIES. When +a guest sees this bit set, it should not enumerate the ITS bug. Note, this bit +is not set by any hardware, but is **intended for VMMs to synthesize** it for +guests as per the host's affected status. + +Mitigation options +^^^^^^^^^^^^^^^^^^ +The ITS mitigation can be controlled using the "indirect_target_selection" +kernel parameter. The available options are: + + ======== =================================================================== + on (default) Deploy the "Aligned branch/return thunks" mitigation. + If spectre_v2 mitigation enables retpoline, aligned-thunks are only + deployed for the affected RET instructions. Retpoline mitigates + indirect branches. + + off Disable ITS mitigation. + + vmexit Equivalent to "=on" if the CPU is affected by guest/host isolation + part of ITS. Otherwise, mitigation is not deployed. This option is + useful when host userspace is not in the threat model, and only + attacks from guest to host are considered. + + stuff Deploy RSB-fill mitigation when retpoline is also deployed. + Otherwise, deploy the default mitigation. When retpoline mitigation + is enabled, RSB-stuffing via Call-Depth-Tracking also mitigates + ITS. + + force Force the ITS bug and deploy the default mitigation. + ======== =================================================================== + +Sysfs reporting +--------------- + +The sysfs file showing ITS mitigation status is: + + /sys/devices/system/cpu/vulnerabilities/indirect_target_selection + +Note, microcode mitigation status is not reported in this file. + +The possible values in this file are: + +.. list-table:: + + * - Not affected + - The processor is not vulnerable. + * - Vulnerable + - System is vulnerable and no mitigation has been applied. + * - Vulnerable, KVM: Not affected + - System is vulnerable to intra-mode BTI, but not affected by eIBRS + guest/host isolation. + * - Mitigation: Aligned branch/return thunks + - The mitigation is enabled, affected indirect branches and RETs are + relocated to safe thunks. + * - Mitigation: Retpolines, Stuffing RSB + - The mitigation is enabled using retpoline and RSB stuffing. + +References +---------- +.. [#f1] Microcode repository - https://github.com/intel/Intel-Linux-Processor-Microcode-Data-Files + +.. [#f2] Affected Processors list - https://www.intel.com/content/www/us/en/developer/topic-technology/software-security-guidance/processors-affected-consolidated-product-cpu-model.html + +.. [#f3] Affected Processors list (machine readable) - https://github.com/intel/Intel-affected-processor-list From 159013a7ca18c271ff64192deb62a689b622d860 Mon Sep 17 00:00:00 2001 From: Pawan Gupta Date: Fri, 21 Jun 2024 17:40:41 -0700 Subject: [PATCH 349/555] x86/its: Enumerate Indirect Target Selection (ITS) bug ITS bug in some pre-Alderlake Intel CPUs may allow indirect branches in the first half of a cache line get predicted to a target of a branch located in the second half of the cache line. Set X86_BUG_ITS on affected CPUs. Mitigation to follow in later commits. Signed-off-by: Pawan Gupta Signed-off-by: Dave Hansen Reviewed-by: Josh Poimboeuf Reviewed-by: Alexandre Chartre --- arch/x86/include/asm/cpufeatures.h | 1 + arch/x86/include/asm/msr-index.h | 8 +++++ arch/x86/kernel/cpu/common.c | 58 +++++++++++++++++++++++------- arch/x86/kvm/x86.c | 4 ++- 4 files changed, 58 insertions(+), 13 deletions(-) diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpufeatures.h index 6c2c152d8a67..2ff978fa414b 100644 --- a/arch/x86/include/asm/cpufeatures.h +++ b/arch/x86/include/asm/cpufeatures.h @@ -533,4 +533,5 @@ #define X86_BUG_BHI X86_BUG(1*32 + 3) /* "bhi" CPU is affected by Branch History Injection */ #define X86_BUG_IBPB_NO_RET X86_BUG(1*32 + 4) /* "ibpb_no_ret" IBPB omits return target predictions */ #define X86_BUG_SPECTRE_V2_USER X86_BUG(1*32 + 5) /* "spectre_v2_user" CPU is affected by Spectre variant 2 attack between user processes */ +#define X86_BUG_ITS X86_BUG(1*32 + 6) /* "its" CPU is affected by Indirect Target Selection */ #endif /* _ASM_X86_CPUFEATURES_H */ diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h index e6134ef2263d..e7d2f460fcc6 100644 --- a/arch/x86/include/asm/msr-index.h +++ b/arch/x86/include/asm/msr-index.h @@ -211,6 +211,14 @@ * VERW clears CPU Register * File. */ +#define ARCH_CAP_ITS_NO BIT_ULL(62) /* + * Not susceptible to + * Indirect Target Selection. + * This bit is not set by + * HW, but is synthesized by + * VMMs for guests to know + * their affected status. + */ #define MSR_IA32_FLUSH_CMD 0x0000010b #define L1D_FLUSH BIT(0) /* diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 12126adbc3a9..accdee7ab1f7 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -1227,6 +1227,8 @@ static const __initconst struct x86_cpu_id cpu_vuln_whitelist[] = { #define GDS BIT(6) /* CPU is affected by Register File Data Sampling */ #define RFDS BIT(7) +/* CPU is affected by Indirect Target Selection */ +#define ITS BIT(8) static const struct x86_cpu_id cpu_vuln_blacklist[] __initconst = { VULNBL_INTEL_STEPS(INTEL_IVYBRIDGE, X86_STEP_MAX, SRBDS), @@ -1238,22 +1240,25 @@ static const struct x86_cpu_id cpu_vuln_blacklist[] __initconst = { VULNBL_INTEL_STEPS(INTEL_BROADWELL_G, X86_STEP_MAX, SRBDS), VULNBL_INTEL_STEPS(INTEL_BROADWELL_X, X86_STEP_MAX, MMIO), VULNBL_INTEL_STEPS(INTEL_BROADWELL, X86_STEP_MAX, SRBDS), - VULNBL_INTEL_STEPS(INTEL_SKYLAKE_X, X86_STEP_MAX, MMIO | RETBLEED | GDS), + VULNBL_INTEL_STEPS(INTEL_SKYLAKE_X, 0x5, MMIO | RETBLEED | GDS), + VULNBL_INTEL_STEPS(INTEL_SKYLAKE_X, X86_STEP_MAX, MMIO | RETBLEED | GDS | ITS), VULNBL_INTEL_STEPS(INTEL_SKYLAKE_L, X86_STEP_MAX, MMIO | RETBLEED | GDS | SRBDS), VULNBL_INTEL_STEPS(INTEL_SKYLAKE, X86_STEP_MAX, MMIO | RETBLEED | GDS | SRBDS), - VULNBL_INTEL_STEPS(INTEL_KABYLAKE_L, X86_STEP_MAX, MMIO | RETBLEED | GDS | SRBDS), - VULNBL_INTEL_STEPS(INTEL_KABYLAKE, X86_STEP_MAX, MMIO | RETBLEED | GDS | SRBDS), + VULNBL_INTEL_STEPS(INTEL_KABYLAKE_L, 0xb, MMIO | RETBLEED | GDS | SRBDS), + VULNBL_INTEL_STEPS(INTEL_KABYLAKE_L, X86_STEP_MAX, MMIO | RETBLEED | GDS | SRBDS | ITS), + VULNBL_INTEL_STEPS(INTEL_KABYLAKE, 0xc, MMIO | RETBLEED | GDS | SRBDS), + VULNBL_INTEL_STEPS(INTEL_KABYLAKE, X86_STEP_MAX, MMIO | RETBLEED | GDS | SRBDS | ITS), VULNBL_INTEL_STEPS(INTEL_CANNONLAKE_L, X86_STEP_MAX, RETBLEED), - VULNBL_INTEL_STEPS(INTEL_ICELAKE_L, X86_STEP_MAX, MMIO | MMIO_SBDS | RETBLEED | GDS), - VULNBL_INTEL_STEPS(INTEL_ICELAKE_D, X86_STEP_MAX, MMIO | GDS), - VULNBL_INTEL_STEPS(INTEL_ICELAKE_X, X86_STEP_MAX, MMIO | GDS), - VULNBL_INTEL_STEPS(INTEL_COMETLAKE, X86_STEP_MAX, MMIO | MMIO_SBDS | RETBLEED | GDS), - VULNBL_INTEL_STEPS(INTEL_COMETLAKE_L, 0x0, MMIO | RETBLEED), - VULNBL_INTEL_STEPS(INTEL_COMETLAKE_L, X86_STEP_MAX, MMIO | MMIO_SBDS | RETBLEED | GDS), - VULNBL_INTEL_STEPS(INTEL_TIGERLAKE_L, X86_STEP_MAX, GDS), - VULNBL_INTEL_STEPS(INTEL_TIGERLAKE, X86_STEP_MAX, GDS), + VULNBL_INTEL_STEPS(INTEL_ICELAKE_L, X86_STEP_MAX, MMIO | MMIO_SBDS | RETBLEED | GDS | ITS), + VULNBL_INTEL_STEPS(INTEL_ICELAKE_D, X86_STEP_MAX, MMIO | GDS | ITS), + VULNBL_INTEL_STEPS(INTEL_ICELAKE_X, X86_STEP_MAX, MMIO | GDS | ITS), + VULNBL_INTEL_STEPS(INTEL_COMETLAKE, X86_STEP_MAX, MMIO | MMIO_SBDS | RETBLEED | GDS | ITS), + VULNBL_INTEL_STEPS(INTEL_COMETLAKE_L, 0x0, MMIO | RETBLEED | ITS), + VULNBL_INTEL_STEPS(INTEL_COMETLAKE_L, X86_STEP_MAX, MMIO | MMIO_SBDS | RETBLEED | GDS | ITS), + VULNBL_INTEL_STEPS(INTEL_TIGERLAKE_L, X86_STEP_MAX, GDS | ITS), + VULNBL_INTEL_STEPS(INTEL_TIGERLAKE, X86_STEP_MAX, GDS | ITS), VULNBL_INTEL_STEPS(INTEL_LAKEFIELD, X86_STEP_MAX, MMIO | MMIO_SBDS | RETBLEED), - VULNBL_INTEL_STEPS(INTEL_ROCKETLAKE, X86_STEP_MAX, MMIO | RETBLEED | GDS), + VULNBL_INTEL_STEPS(INTEL_ROCKETLAKE, X86_STEP_MAX, MMIO | RETBLEED | GDS | ITS), VULNBL_INTEL_TYPE(INTEL_ALDERLAKE, ATOM, RFDS), VULNBL_INTEL_STEPS(INTEL_ALDERLAKE_L, X86_STEP_MAX, RFDS), VULNBL_INTEL_TYPE(INTEL_RAPTORLAKE, ATOM, RFDS), @@ -1318,6 +1323,32 @@ static bool __init vulnerable_to_rfds(u64 x86_arch_cap_msr) return cpu_matches(cpu_vuln_blacklist, RFDS); } +static bool __init vulnerable_to_its(u64 x86_arch_cap_msr) +{ + /* The "immunity" bit trumps everything else: */ + if (x86_arch_cap_msr & ARCH_CAP_ITS_NO) + return false; + if (boot_cpu_data.x86_vendor != X86_VENDOR_INTEL) + return false; + + /* None of the affected CPUs have BHI_CTRL */ + if (boot_cpu_has(X86_FEATURE_BHI_CTRL)) + return false; + + /* + * If a VMM did not expose ITS_NO, assume that a guest could + * be running on a vulnerable hardware or may migrate to such + * hardware. + */ + if (boot_cpu_has(X86_FEATURE_HYPERVISOR)) + return true; + + if (cpu_matches(cpu_vuln_blacklist, ITS)) + return true; + + return false; +} + static void __init cpu_set_bug_bits(struct cpuinfo_x86 *c) { u64 x86_arch_cap_msr = x86_read_arch_cap_msr(); @@ -1449,6 +1480,9 @@ static void __init cpu_set_bug_bits(struct cpuinfo_x86 *c) if (cpu_has(c, X86_FEATURE_AMD_IBPB) && !cpu_has(c, X86_FEATURE_AMD_IBPB_RET)) setup_force_cpu_bug(X86_BUG_IBPB_NO_RET); + if (vulnerable_to_its(x86_arch_cap_msr)) + setup_force_cpu_bug(X86_BUG_ITS); + if (cpu_matches(cpu_vuln_whitelist, NO_MELTDOWN)) return; diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index df5b99ea1f18..2626567c35df 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -1584,7 +1584,7 @@ EXPORT_SYMBOL_GPL(kvm_emulate_rdpmc); ARCH_CAP_PSCHANGE_MC_NO | ARCH_CAP_TSX_CTRL_MSR | ARCH_CAP_TAA_NO | \ ARCH_CAP_SBDR_SSDP_NO | ARCH_CAP_FBSDP_NO | ARCH_CAP_PSDP_NO | \ ARCH_CAP_FB_CLEAR | ARCH_CAP_RRSBA | ARCH_CAP_PBRSB_NO | ARCH_CAP_GDS_NO | \ - ARCH_CAP_RFDS_NO | ARCH_CAP_RFDS_CLEAR | ARCH_CAP_BHI_NO) + ARCH_CAP_RFDS_NO | ARCH_CAP_RFDS_CLEAR | ARCH_CAP_BHI_NO | ARCH_CAP_ITS_NO) static u64 kvm_get_arch_capabilities(void) { @@ -1618,6 +1618,8 @@ static u64 kvm_get_arch_capabilities(void) data |= ARCH_CAP_MDS_NO; if (!boot_cpu_has_bug(X86_BUG_RFDS)) data |= ARCH_CAP_RFDS_NO; + if (!boot_cpu_has_bug(X86_BUG_ITS)) + data |= ARCH_CAP_ITS_NO; if (!boot_cpu_has(X86_FEATURE_RTM)) { /* From 8754e67ad4ac692c67ff1f99c0d07156f04ae40c Mon Sep 17 00:00:00 2001 From: Pawan Gupta Date: Fri, 21 Jun 2024 21:17:21 -0700 Subject: [PATCH 350/555] x86/its: Add support for ITS-safe indirect thunk Due to ITS, indirect branches in the lower half of a cacheline may be vulnerable to branch target injection attack. Introduce ITS-safe thunks to patch indirect branches in the lower half of cacheline with the thunk. Also thunk any eBPF generated indirect branches in emit_indirect_jump(). Below category of indirect branches are not mitigated: - Indirect branches in the .init section are not mitigated because they are discarded after boot. - Indirect branches that are explicitly marked retpoline-safe. Note that retpoline also mitigates the indirect branches against ITS. This is because the retpoline sequence fills an RSB entry before RET, and it does not suffer from RSB-underflow part of the ITS. Signed-off-by: Pawan Gupta Signed-off-by: Dave Hansen Reviewed-by: Josh Poimboeuf Reviewed-by: Alexandre Chartre --- arch/x86/Kconfig | 11 +++++++ arch/x86/include/asm/cpufeatures.h | 1 + arch/x86/include/asm/nospec-branch.h | 4 +++ arch/x86/kernel/alternative.c | 45 ++++++++++++++++++++++++++-- arch/x86/kernel/vmlinux.lds.S | 6 ++++ arch/x86/lib/retpoline.S | 28 +++++++++++++++++ arch/x86/net/bpf_jit_comp.c | 5 +++- 7 files changed, 96 insertions(+), 4 deletions(-) diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 4b9f378e05f6..a6bd0590c437 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -2710,6 +2710,17 @@ config MITIGATION_SSB of speculative execution in a similar way to the Meltdown and Spectre security vulnerabilities. +config MITIGATION_ITS + bool "Enable Indirect Target Selection mitigation" + depends on CPU_SUP_INTEL && X86_64 + depends on MITIGATION_RETPOLINE && MITIGATION_RETHUNK + default y + help + Enable Indirect Target Selection (ITS) mitigation. ITS is a bug in + BPU on some Intel CPUs that may allow Spectre V2 style attacks. If + disabled, mitigation cannot be enabled via cmdline. + See + endif config ARCH_HAS_ADD_PAGES diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpufeatures.h index 2ff978fa414b..03af8191073a 100644 --- a/arch/x86/include/asm/cpufeatures.h +++ b/arch/x86/include/asm/cpufeatures.h @@ -481,6 +481,7 @@ #define X86_FEATURE_AMD_HETEROGENEOUS_CORES (21*32 + 6) /* Heterogeneous Core Topology */ #define X86_FEATURE_AMD_WORKLOAD_CLASS (21*32 + 7) /* Workload Classification */ #define X86_FEATURE_PREFER_YMM (21*32 + 8) /* Avoid ZMM registers due to downclocking */ +#define X86_FEATURE_INDIRECT_THUNK_ITS (21*32 + 9) /* Use thunk for indirect branches in lower half of cacheline */ /* * BUG word(s) diff --git a/arch/x86/include/asm/nospec-branch.h b/arch/x86/include/asm/nospec-branch.h index 5c43f145454d..a5eb0cd93c92 100644 --- a/arch/x86/include/asm/nospec-branch.h +++ b/arch/x86/include/asm/nospec-branch.h @@ -336,10 +336,14 @@ #else /* __ASSEMBLER__ */ +#define ITS_THUNK_SIZE 64 + typedef u8 retpoline_thunk_t[RETPOLINE_THUNK_SIZE]; +typedef u8 its_thunk_t[ITS_THUNK_SIZE]; extern retpoline_thunk_t __x86_indirect_thunk_array[]; extern retpoline_thunk_t __x86_indirect_call_thunk_array[]; extern retpoline_thunk_t __x86_indirect_jump_thunk_array[]; +extern its_thunk_t __x86_indirect_its_thunk_array[]; #ifdef CONFIG_MITIGATION_RETHUNK extern void __x86_return_thunk(void); diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c index bf82c6f7d690..333c0295b0a5 100644 --- a/arch/x86/kernel/alternative.c +++ b/arch/x86/kernel/alternative.c @@ -581,7 +581,8 @@ static int emit_indirect(int op, int reg, u8 *bytes) return i; } -static int emit_call_track_retpoline(void *addr, struct insn *insn, int reg, u8 *bytes) +static int __emit_trampoline(void *addr, struct insn *insn, u8 *bytes, + void *call_dest, void *jmp_dest) { u8 op = insn->opcode.bytes[0]; int i = 0; @@ -602,7 +603,7 @@ static int emit_call_track_retpoline(void *addr, struct insn *insn, int reg, u8 switch (op) { case CALL_INSN_OPCODE: __text_gen_insn(bytes+i, op, addr+i, - __x86_indirect_call_thunk_array[reg], + call_dest, CALL_INSN_SIZE); i += CALL_INSN_SIZE; break; @@ -610,7 +611,7 @@ static int emit_call_track_retpoline(void *addr, struct insn *insn, int reg, u8 case JMP32_INSN_OPCODE: clang_jcc: __text_gen_insn(bytes+i, op, addr+i, - __x86_indirect_jump_thunk_array[reg], + jmp_dest, JMP32_INSN_SIZE); i += JMP32_INSN_SIZE; break; @@ -625,6 +626,35 @@ clang_jcc: return i; } +static int emit_call_track_retpoline(void *addr, struct insn *insn, int reg, u8 *bytes) +{ + return __emit_trampoline(addr, insn, bytes, + __x86_indirect_call_thunk_array[reg], + __x86_indirect_jump_thunk_array[reg]); +} + +#ifdef CONFIG_MITIGATION_ITS +static int emit_its_trampoline(void *addr, struct insn *insn, int reg, u8 *bytes) +{ + return __emit_trampoline(addr, insn, bytes, + __x86_indirect_its_thunk_array[reg], + __x86_indirect_its_thunk_array[reg]); +} + +/* Check if an indirect branch is at ITS-unsafe address */ +static bool cpu_wants_indirect_its_thunk_at(unsigned long addr, int reg) +{ + if (!cpu_feature_enabled(X86_FEATURE_INDIRECT_THUNK_ITS)) + return false; + + /* Indirect branch opcode is 2 or 3 bytes depending on reg */ + addr += 1 + reg / 8; + + /* Lower-half of the cacheline? */ + return !(addr & 0x20); +} +#endif + /* * Rewrite the compiler generated retpoline thunk calls. * @@ -699,6 +729,15 @@ static int patch_retpoline(void *addr, struct insn *insn, u8 *bytes) bytes[i++] = 0xe8; /* LFENCE */ } +#ifdef CONFIG_MITIGATION_ITS + /* + * Check if the address of last byte of emitted-indirect is in + * lower-half of the cacheline. Such branches need ITS mitigation. + */ + if (cpu_wants_indirect_its_thunk_at((unsigned long)addr + i, reg)) + return emit_its_trampoline(addr, insn, reg, bytes); +#endif + ret = emit_indirect(op, reg, bytes + i); if (ret < 0) return ret; diff --git a/arch/x86/kernel/vmlinux.lds.S b/arch/x86/kernel/vmlinux.lds.S index ccdc45e5b759..c3e048e7e161 100644 --- a/arch/x86/kernel/vmlinux.lds.S +++ b/arch/x86/kernel/vmlinux.lds.S @@ -497,6 +497,12 @@ PROVIDE(__ref_stack_chk_guard = __stack_chk_guard); "SRSO function pair won't alias"); #endif +#if defined(CONFIG_MITIGATION_ITS) && !defined(CONFIG_DEBUG_FORCE_FUNCTION_ALIGN_64B) +. = ASSERT(__x86_indirect_its_thunk_rax & 0x20, "__x86_indirect_thunk_rax not in second half of cacheline"); +. = ASSERT(((__x86_indirect_its_thunk_rcx - __x86_indirect_its_thunk_rax) % 64) == 0, "Indirect thunks are not cacheline apart"); +. = ASSERT(__x86_indirect_its_thunk_array == __x86_indirect_its_thunk_rax, "Gap in ITS thunk array"); +#endif + #endif /* CONFIG_X86_64 */ /* diff --git a/arch/x86/lib/retpoline.S b/arch/x86/lib/retpoline.S index a26c43abd47d..a06891892853 100644 --- a/arch/x86/lib/retpoline.S +++ b/arch/x86/lib/retpoline.S @@ -367,6 +367,34 @@ SYM_FUNC_END(call_depth_return_thunk) #endif /* CONFIG_MITIGATION_CALL_DEPTH_TRACKING */ +#ifdef CONFIG_MITIGATION_ITS + +.macro ITS_THUNK reg + +SYM_INNER_LABEL(__x86_indirect_its_thunk_\reg, SYM_L_GLOBAL) + UNWIND_HINT_UNDEFINED + ANNOTATE_NOENDBR + ANNOTATE_RETPOLINE_SAFE + jmp *%\reg + int3 + .align 32, 0xcc /* fill to the end of the line */ + .skip 32, 0xcc /* skip to the next upper half */ +.endm + +/* ITS mitigation requires thunks be aligned to upper half of cacheline */ +.align 64, 0xcc +.skip 32, 0xcc +SYM_CODE_START(__x86_indirect_its_thunk_array) + +#define GEN(reg) ITS_THUNK reg +#include +#undef GEN + + .align 64, 0xcc +SYM_CODE_END(__x86_indirect_its_thunk_array) + +#endif + /* * This function name is magical and is used by -mfunction-return=thunk-extern * for the compiler to generate JMPs to it. diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c index 9e5fe2ba858f..01e8119db5db 100644 --- a/arch/x86/net/bpf_jit_comp.c +++ b/arch/x86/net/bpf_jit_comp.c @@ -661,7 +661,10 @@ static void emit_indirect_jump(u8 **pprog, int reg, u8 *ip) { u8 *prog = *pprog; - if (cpu_feature_enabled(X86_FEATURE_RETPOLINE_LFENCE)) { + if (cpu_feature_enabled(X86_FEATURE_INDIRECT_THUNK_ITS)) { + OPTIMIZER_HIDE_VAR(reg); + emit_jump(&prog, &__x86_indirect_its_thunk_array[reg], ip); + } else if (cpu_feature_enabled(X86_FEATURE_RETPOLINE_LFENCE)) { EMIT_LFENCE(); EMIT2(0xFF, 0xE0 + reg); } else if (cpu_feature_enabled(X86_FEATURE_RETPOLINE)) { From a75bf27fe41abe658c53276a0c486c4bf9adecfc Mon Sep 17 00:00:00 2001 From: Pawan Gupta Date: Fri, 21 Jun 2024 21:17:21 -0700 Subject: [PATCH 351/555] x86/its: Add support for ITS-safe return thunk RETs in the lower half of cacheline may be affected by ITS bug, specifically when the RSB-underflows. Use ITS-safe return thunk for such RETs. RETs that are not patched: - RET in retpoline sequence does not need to be patched, because the sequence itself fills an RSB before RET. - RET in Call Depth Tracking (CDT) thunks __x86_indirect_{call|jump}_thunk and call_depth_return_thunk are not patched because CDT by design prevents RSB-underflow. - RETs in .init section are not reachable after init. - RETs that are explicitly marked safe with ANNOTATE_UNRET_SAFE. Signed-off-by: Pawan Gupta Signed-off-by: Dave Hansen Reviewed-by: Josh Poimboeuf Reviewed-by: Alexandre Chartre --- arch/x86/include/asm/alternative.h | 14 ++++++++++++++ arch/x86/include/asm/nospec-branch.h | 6 ++++++ arch/x86/kernel/alternative.c | 19 +++++++++++++++++-- arch/x86/kernel/ftrace.c | 2 +- arch/x86/kernel/static_call.c | 4 ++-- arch/x86/kernel/vmlinux.lds.S | 4 ++++ arch/x86/lib/retpoline.S | 13 ++++++++++++- arch/x86/net/bpf_jit_comp.c | 2 +- 8 files changed, 57 insertions(+), 7 deletions(-) diff --git a/arch/x86/include/asm/alternative.h b/arch/x86/include/asm/alternative.h index 4a37a8bd87fd..4037c611537c 100644 --- a/arch/x86/include/asm/alternative.h +++ b/arch/x86/include/asm/alternative.h @@ -124,6 +124,20 @@ static __always_inline int x86_call_depth_emit_accounting(u8 **pprog, } #endif +#if defined(CONFIG_MITIGATION_RETHUNK) && defined(CONFIG_OBJTOOL) +extern bool cpu_wants_rethunk(void); +extern bool cpu_wants_rethunk_at(void *addr); +#else +static __always_inline bool cpu_wants_rethunk(void) +{ + return false; +} +static __always_inline bool cpu_wants_rethunk_at(void *addr) +{ + return false; +} +#endif + #ifdef CONFIG_SMP extern void alternatives_smp_module_add(struct module *mod, char *name, void *locks, void *locks_end, diff --git a/arch/x86/include/asm/nospec-branch.h b/arch/x86/include/asm/nospec-branch.h index a5eb0cd93c92..7d04ade33541 100644 --- a/arch/x86/include/asm/nospec-branch.h +++ b/arch/x86/include/asm/nospec-branch.h @@ -367,6 +367,12 @@ static inline void srso_return_thunk(void) {} static inline void srso_alias_return_thunk(void) {} #endif +#ifdef CONFIG_MITIGATION_ITS +extern void its_return_thunk(void); +#else +static inline void its_return_thunk(void) {} +#endif + extern void retbleed_return_thunk(void); extern void srso_return_thunk(void); extern void srso_alias_return_thunk(void); diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c index 333c0295b0a5..3e67e72bffdc 100644 --- a/arch/x86/kernel/alternative.c +++ b/arch/x86/kernel/alternative.c @@ -814,6 +814,21 @@ void __init_or_module noinline apply_retpolines(s32 *start, s32 *end) #ifdef CONFIG_MITIGATION_RETHUNK +bool cpu_wants_rethunk(void) +{ + return cpu_feature_enabled(X86_FEATURE_RETHUNK); +} + +bool cpu_wants_rethunk_at(void *addr) +{ + if (!cpu_feature_enabled(X86_FEATURE_RETHUNK)) + return false; + if (x86_return_thunk != its_return_thunk) + return true; + + return !((unsigned long)addr & 0x20); +} + /* * Rewrite the compiler generated return thunk tail-calls. * @@ -830,7 +845,7 @@ static int patch_return(void *addr, struct insn *insn, u8 *bytes) int i = 0; /* Patch the custom return thunks... */ - if (cpu_feature_enabled(X86_FEATURE_RETHUNK)) { + if (cpu_wants_rethunk_at(addr)) { i = JMP32_INSN_SIZE; __text_gen_insn(bytes, JMP32_INSN_OPCODE, addr, x86_return_thunk, i); } else { @@ -847,7 +862,7 @@ void __init_or_module noinline apply_returns(s32 *start, s32 *end) { s32 *s; - if (cpu_feature_enabled(X86_FEATURE_RETHUNK)) + if (cpu_wants_rethunk()) static_call_force_reinit(); for (s = start; s < end; s++) { diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c index cace6e8d7cc7..5eb1514af559 100644 --- a/arch/x86/kernel/ftrace.c +++ b/arch/x86/kernel/ftrace.c @@ -354,7 +354,7 @@ create_trampoline(struct ftrace_ops *ops, unsigned int *tramp_size) goto fail; ip = trampoline + size; - if (cpu_feature_enabled(X86_FEATURE_RETHUNK)) + if (cpu_wants_rethunk_at(ip)) __text_gen_insn(ip, JMP32_INSN_OPCODE, ip, x86_return_thunk, JMP32_INSN_SIZE); else memcpy(ip, retq, sizeof(retq)); diff --git a/arch/x86/kernel/static_call.c b/arch/x86/kernel/static_call.c index a59c72e77645..c3d7ff44b29a 100644 --- a/arch/x86/kernel/static_call.c +++ b/arch/x86/kernel/static_call.c @@ -81,7 +81,7 @@ static void __ref __static_call_transform(void *insn, enum insn_type type, break; case RET: - if (cpu_feature_enabled(X86_FEATURE_RETHUNK)) + if (cpu_wants_rethunk_at(insn)) code = text_gen_insn(JMP32_INSN_OPCODE, insn, x86_return_thunk); else code = &retinsn; @@ -90,7 +90,7 @@ static void __ref __static_call_transform(void *insn, enum insn_type type, case JCC: if (!func) { func = __static_call_return; - if (cpu_feature_enabled(X86_FEATURE_RETHUNK)) + if (cpu_wants_rethunk()) func = x86_return_thunk; } diff --git a/arch/x86/kernel/vmlinux.lds.S b/arch/x86/kernel/vmlinux.lds.S index c3e048e7e161..e97f5773c8df 100644 --- a/arch/x86/kernel/vmlinux.lds.S +++ b/arch/x86/kernel/vmlinux.lds.S @@ -503,6 +503,10 @@ PROVIDE(__ref_stack_chk_guard = __stack_chk_guard); . = ASSERT(__x86_indirect_its_thunk_array == __x86_indirect_its_thunk_rax, "Gap in ITS thunk array"); #endif +#if defined(CONFIG_MITIGATION_ITS) && !defined(CONFIG_DEBUG_FORCE_FUNCTION_ALIGN_64B) +. = ASSERT(its_return_thunk & 0x20, "its_return_thunk not in second half of cacheline"); +#endif + #endif /* CONFIG_X86_64 */ /* diff --git a/arch/x86/lib/retpoline.S b/arch/x86/lib/retpoline.S index a06891892853..ebca28fe7e31 100644 --- a/arch/x86/lib/retpoline.S +++ b/arch/x86/lib/retpoline.S @@ -393,7 +393,18 @@ SYM_CODE_START(__x86_indirect_its_thunk_array) .align 64, 0xcc SYM_CODE_END(__x86_indirect_its_thunk_array) -#endif +.align 64, 0xcc +.skip 32, 0xcc +SYM_CODE_START(its_return_thunk) + UNWIND_HINT_FUNC + ANNOTATE_NOENDBR + ANNOTATE_UNRET_SAFE + ret + int3 +SYM_CODE_END(its_return_thunk) +EXPORT_SYMBOL(its_return_thunk) + +#endif /* CONFIG_MITIGATION_ITS */ /* * This function name is magical and is used by -mfunction-return=thunk-extern diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c index 01e8119db5db..a5b65c09910b 100644 --- a/arch/x86/net/bpf_jit_comp.c +++ b/arch/x86/net/bpf_jit_comp.c @@ -686,7 +686,7 @@ static void emit_return(u8 **pprog, u8 *ip) { u8 *prog = *pprog; - if (cpu_feature_enabled(X86_FEATURE_RETHUNK)) { + if (cpu_wants_rethunk()) { emit_jump(&prog, x86_return_thunk, ip); } else { EMIT1(0xC3); /* ret */ From f4818881c47fd91fcb6d62373c57c7844e3de1c0 Mon Sep 17 00:00:00 2001 From: Pawan Gupta Date: Fri, 21 Jun 2024 20:23:23 -0700 Subject: [PATCH 352/555] x86/its: Enable Indirect Target Selection mitigation Indirect Target Selection (ITS) is a bug in some pre-ADL Intel CPUs with eIBRS. It affects prediction of indirect branch and RETs in the lower half of cacheline. Due to ITS such branches may get wrongly predicted to a target of (direct or indirect) branch that is located in the upper half of the cacheline. Scope of impact =============== Guest/host isolation -------------------- When eIBRS is used for guest/host isolation, the indirect branches in the VMM may still be predicted with targets corresponding to branches in the guest. Intra-mode ---------- cBPF or other native gadgets can be used for intra-mode training and disclosure using ITS. User/kernel isolation --------------------- When eIBRS is enabled user/kernel isolation is not impacted. Indirect Branch Prediction Barrier (IBPB) ----------------------------------------- After an IBPB, indirect branches may be predicted with targets corresponding to direct branches which were executed prior to IBPB. This is mitigated by a microcode update. Add cmdline parameter indirect_target_selection=off|on|force to control the mitigation to relocate the affected branches to an ITS-safe thunk i.e. located in the upper half of cacheline. Also add the sysfs reporting. When retpoline mitigation is deployed, ITS safe-thunks are not needed, because retpoline sequence is already ITS-safe. Similarly, when call depth tracking (CDT) mitigation is deployed (retbleed=stuff), ITS safe return thunk is not used, as CDT prevents RSB-underflow. To not overcomplicate things, ITS mitigation is not supported with spectre-v2 lfence;jmp mitigation. Moreover, it is less practical to deploy lfence;jmp mitigation on ITS affected parts anyways. Signed-off-by: Pawan Gupta Signed-off-by: Dave Hansen Reviewed-by: Josh Poimboeuf Reviewed-by: Alexandre Chartre --- .../ABI/testing/sysfs-devices-system-cpu | 1 + .../admin-guide/kernel-parameters.txt | 13 ++ arch/x86/kernel/cpu/bugs.c | 140 +++++++++++++++++- drivers/base/cpu.c | 3 + include/linux/cpu.h | 2 + 5 files changed, 155 insertions(+), 4 deletions(-) diff --git a/Documentation/ABI/testing/sysfs-devices-system-cpu b/Documentation/ABI/testing/sysfs-devices-system-cpu index 206079d3bd5b..6a1acabb29d8 100644 --- a/Documentation/ABI/testing/sysfs-devices-system-cpu +++ b/Documentation/ABI/testing/sysfs-devices-system-cpu @@ -511,6 +511,7 @@ Description: information about CPUs heterogeneity. What: /sys/devices/system/cpu/vulnerabilities /sys/devices/system/cpu/vulnerabilities/gather_data_sampling + /sys/devices/system/cpu/vulnerabilities/indirect_target_selection /sys/devices/system/cpu/vulnerabilities/itlb_multihit /sys/devices/system/cpu/vulnerabilities/l1tf /sys/devices/system/cpu/vulnerabilities/mds diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index d9fd26b95b34..d3e78e795e0d 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -2202,6 +2202,18 @@ different crypto accelerators. This option can be used to achieve best performance for particular HW. + indirect_target_selection= [X86,Intel] Mitigation control for Indirect + Target Selection(ITS) bug in Intel CPUs. Updated + microcode is also required for a fix in IBPB. + + on: Enable mitigation (default). + off: Disable mitigation. + force: Force the ITS bug and deploy default + mitigation. + + For details see: + Documentation/admin-guide/hw-vuln/indirect-target-selection.rst + init= [KNL] Format: Run specified binary instead of /sbin/init as init @@ -3693,6 +3705,7 @@ expose users to several CPU vulnerabilities. Equivalent to: if nokaslr then kpti=0 [ARM64] gather_data_sampling=off [X86] + indirect_target_selection=off [X86] kvm.nx_huge_pages=off [X86] l1tf=off [X86] mds=off [X86] diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 362602b705cc..417596ce5352 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -49,6 +49,7 @@ static void __init srbds_select_mitigation(void); static void __init l1d_flush_select_mitigation(void); static void __init srso_select_mitigation(void); static void __init gds_select_mitigation(void); +static void __init its_select_mitigation(void); /* The base value of the SPEC_CTRL MSR without task-specific bits set */ u64 x86_spec_ctrl_base; @@ -66,6 +67,14 @@ static DEFINE_MUTEX(spec_ctrl_mutex); void (*x86_return_thunk)(void) __ro_after_init = __x86_return_thunk; +static void __init set_return_thunk(void *thunk) +{ + if (x86_return_thunk != __x86_return_thunk) + pr_warn("x86/bugs: return thunk changed\n"); + + x86_return_thunk = thunk; +} + /* Update SPEC_CTRL MSR and its cached copy unconditionally */ static void update_spec_ctrl(u64 val) { @@ -178,6 +187,7 @@ void __init cpu_select_mitigations(void) */ srso_select_mitigation(); gds_select_mitigation(); + its_select_mitigation(); } /* @@ -1118,7 +1128,7 @@ do_cmd_auto: setup_force_cpu_cap(X86_FEATURE_RETHUNK); setup_force_cpu_cap(X86_FEATURE_UNRET); - x86_return_thunk = retbleed_return_thunk; + set_return_thunk(retbleed_return_thunk); if (boot_cpu_data.x86_vendor != X86_VENDOR_AMD && boot_cpu_data.x86_vendor != X86_VENDOR_HYGON) @@ -1153,7 +1163,7 @@ do_cmd_auto: setup_force_cpu_cap(X86_FEATURE_RETHUNK); setup_force_cpu_cap(X86_FEATURE_CALL_DEPTH); - x86_return_thunk = call_depth_return_thunk; + set_return_thunk(call_depth_return_thunk); break; default: @@ -1187,6 +1197,115 @@ do_cmd_auto: pr_info("%s\n", retbleed_strings[retbleed_mitigation]); } +#undef pr_fmt +#define pr_fmt(fmt) "ITS: " fmt + +enum its_mitigation_cmd { + ITS_CMD_OFF, + ITS_CMD_ON, +}; + +enum its_mitigation { + ITS_MITIGATION_OFF, + ITS_MITIGATION_ALIGNED_THUNKS, + ITS_MITIGATION_RETPOLINE_STUFF, +}; + +static const char * const its_strings[] = { + [ITS_MITIGATION_OFF] = "Vulnerable", + [ITS_MITIGATION_ALIGNED_THUNKS] = "Mitigation: Aligned branch/return thunks", + [ITS_MITIGATION_RETPOLINE_STUFF] = "Mitigation: Retpolines, Stuffing RSB", +}; + +static enum its_mitigation its_mitigation __ro_after_init = ITS_MITIGATION_ALIGNED_THUNKS; + +static enum its_mitigation_cmd its_cmd __ro_after_init = + IS_ENABLED(CONFIG_MITIGATION_ITS) ? ITS_CMD_ON : ITS_CMD_OFF; + +static int __init its_parse_cmdline(char *str) +{ + if (!str) + return -EINVAL; + + if (!IS_ENABLED(CONFIG_MITIGATION_ITS)) { + pr_err("Mitigation disabled at compile time, ignoring option (%s)", str); + return 0; + } + + if (!strcmp(str, "off")) { + its_cmd = ITS_CMD_OFF; + } else if (!strcmp(str, "on")) { + its_cmd = ITS_CMD_ON; + } else if (!strcmp(str, "force")) { + its_cmd = ITS_CMD_ON; + setup_force_cpu_bug(X86_BUG_ITS); + } else { + pr_err("Ignoring unknown indirect_target_selection option (%s).", str); + } + + return 0; +} +early_param("indirect_target_selection", its_parse_cmdline); + +static void __init its_select_mitigation(void) +{ + enum its_mitigation_cmd cmd = its_cmd; + + if (!boot_cpu_has_bug(X86_BUG_ITS) || cpu_mitigations_off()) { + its_mitigation = ITS_MITIGATION_OFF; + return; + } + + /* Retpoline+CDT mitigates ITS, bail out */ + if (boot_cpu_has(X86_FEATURE_RETPOLINE) && + boot_cpu_has(X86_FEATURE_CALL_DEPTH)) { + its_mitigation = ITS_MITIGATION_RETPOLINE_STUFF; + goto out; + } + + /* Exit early to avoid irrelevant warnings */ + if (cmd == ITS_CMD_OFF) { + its_mitigation = ITS_MITIGATION_OFF; + goto out; + } + if (spectre_v2_enabled == SPECTRE_V2_NONE) { + pr_err("WARNING: Spectre-v2 mitigation is off, disabling ITS\n"); + its_mitigation = ITS_MITIGATION_OFF; + goto out; + } + if (!IS_ENABLED(CONFIG_MITIGATION_RETPOLINE) || + !IS_ENABLED(CONFIG_MITIGATION_RETHUNK)) { + pr_err("WARNING: ITS mitigation depends on retpoline and rethunk support\n"); + its_mitigation = ITS_MITIGATION_OFF; + goto out; + } + if (IS_ENABLED(CONFIG_DEBUG_FORCE_FUNCTION_ALIGN_64B)) { + pr_err("WARNING: ITS mitigation is not compatible with CONFIG_DEBUG_FORCE_FUNCTION_ALIGN_64B\n"); + its_mitigation = ITS_MITIGATION_OFF; + goto out; + } + if (boot_cpu_has(X86_FEATURE_RETPOLINE_LFENCE)) { + pr_err("WARNING: ITS mitigation is not compatible with lfence mitigation\n"); + its_mitigation = ITS_MITIGATION_OFF; + goto out; + } + + switch (cmd) { + case ITS_CMD_OFF: + its_mitigation = ITS_MITIGATION_OFF; + break; + case ITS_CMD_ON: + its_mitigation = ITS_MITIGATION_ALIGNED_THUNKS; + if (!boot_cpu_has(X86_FEATURE_RETPOLINE)) + setup_force_cpu_cap(X86_FEATURE_INDIRECT_THUNK_ITS); + setup_force_cpu_cap(X86_FEATURE_RETHUNK); + set_return_thunk(its_return_thunk); + break; + } +out: + pr_info("%s\n", its_strings[its_mitigation]); +} + #undef pr_fmt #define pr_fmt(fmt) "Spectre V2 : " fmt @@ -2607,10 +2726,10 @@ static void __init srso_select_mitigation(void) if (boot_cpu_data.x86 == 0x19) { setup_force_cpu_cap(X86_FEATURE_SRSO_ALIAS); - x86_return_thunk = srso_alias_return_thunk; + set_return_thunk(srso_alias_return_thunk); } else { setup_force_cpu_cap(X86_FEATURE_SRSO); - x86_return_thunk = srso_return_thunk; + set_return_thunk(srso_return_thunk); } if (has_microcode) srso_mitigation = SRSO_MITIGATION_SAFE_RET; @@ -2800,6 +2919,11 @@ static ssize_t rfds_show_state(char *buf) return sysfs_emit(buf, "%s\n", rfds_strings[rfds_mitigation]); } +static ssize_t its_show_state(char *buf) +{ + return sysfs_emit(buf, "%s\n", its_strings[its_mitigation]); +} + static char *stibp_state(void) { if (spectre_v2_in_eibrs_mode(spectre_v2_enabled) && @@ -2982,6 +3106,9 @@ static ssize_t cpu_show_common(struct device *dev, struct device_attribute *attr case X86_BUG_RFDS: return rfds_show_state(buf); + case X86_BUG_ITS: + return its_show_state(buf); + default: break; } @@ -3061,6 +3188,11 @@ ssize_t cpu_show_reg_file_data_sampling(struct device *dev, struct device_attrib { return cpu_show_common(dev, attr, buf, X86_BUG_RFDS); } + +ssize_t cpu_show_indirect_target_selection(struct device *dev, struct device_attribute *attr, char *buf) +{ + return cpu_show_common(dev, attr, buf, X86_BUG_ITS); +} #endif void __warn_thunk(void) diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c index a7e511849875..50651435577c 100644 --- a/drivers/base/cpu.c +++ b/drivers/base/cpu.c @@ -600,6 +600,7 @@ CPU_SHOW_VULN_FALLBACK(spec_rstack_overflow); CPU_SHOW_VULN_FALLBACK(gds); CPU_SHOW_VULN_FALLBACK(reg_file_data_sampling); CPU_SHOW_VULN_FALLBACK(ghostwrite); +CPU_SHOW_VULN_FALLBACK(indirect_target_selection); static DEVICE_ATTR(meltdown, 0444, cpu_show_meltdown, NULL); static DEVICE_ATTR(spectre_v1, 0444, cpu_show_spectre_v1, NULL); @@ -616,6 +617,7 @@ static DEVICE_ATTR(spec_rstack_overflow, 0444, cpu_show_spec_rstack_overflow, NU static DEVICE_ATTR(gather_data_sampling, 0444, cpu_show_gds, NULL); static DEVICE_ATTR(reg_file_data_sampling, 0444, cpu_show_reg_file_data_sampling, NULL); static DEVICE_ATTR(ghostwrite, 0444, cpu_show_ghostwrite, NULL); +static DEVICE_ATTR(indirect_target_selection, 0444, cpu_show_indirect_target_selection, NULL); static struct attribute *cpu_root_vulnerabilities_attrs[] = { &dev_attr_meltdown.attr, @@ -633,6 +635,7 @@ static struct attribute *cpu_root_vulnerabilities_attrs[] = { &dev_attr_gather_data_sampling.attr, &dev_attr_reg_file_data_sampling.attr, &dev_attr_ghostwrite.attr, + &dev_attr_indirect_target_selection.attr, NULL }; diff --git a/include/linux/cpu.h b/include/linux/cpu.h index e3049543008b..3aa955102b34 100644 --- a/include/linux/cpu.h +++ b/include/linux/cpu.h @@ -78,6 +78,8 @@ extern ssize_t cpu_show_gds(struct device *dev, extern ssize_t cpu_show_reg_file_data_sampling(struct device *dev, struct device_attribute *attr, char *buf); extern ssize_t cpu_show_ghostwrite(struct device *dev, struct device_attribute *attr, char *buf); +extern ssize_t cpu_show_indirect_target_selection(struct device *dev, + struct device_attribute *attr, char *buf); extern __printf(4, 5) struct device *cpu_device_create(struct device *parent, void *drvdata, From 2665281a07e19550944e8354a2024635a7b2714a Mon Sep 17 00:00:00 2001 From: Pawan Gupta Date: Mon, 18 Nov 2024 09:53:12 -0800 Subject: [PATCH 353/555] x86/its: Add "vmexit" option to skip mitigation on some CPUs Ice Lake generation CPUs are not affected by guest/host isolation part of ITS. If a user is only concerned about KVM guests, they can now choose a new cmdline option "vmexit" that will not deploy the ITS mitigation when CPU is not affected by guest/host isolation. This saves the performance overhead of ITS mitigation on Ice Lake gen CPUs. When "vmexit" option selected, if the CPU is affected by ITS guest/host isolation, the default ITS mitigation is deployed. Signed-off-by: Pawan Gupta Signed-off-by: Dave Hansen Reviewed-by: Josh Poimboeuf Reviewed-by: Alexandre Chartre --- .../admin-guide/kernel-parameters.txt | 2 ++ arch/x86/include/asm/cpufeatures.h | 1 + arch/x86/kernel/cpu/bugs.c | 11 +++++++++++ arch/x86/kernel/cpu/common.c | 19 ++++++++++++------- 4 files changed, 26 insertions(+), 7 deletions(-) diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index d3e78e795e0d..e70d15095f28 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -2210,6 +2210,8 @@ off: Disable mitigation. force: Force the ITS bug and deploy default mitigation. + vmexit: Only deploy mitigation if CPU is affected by + guest/host isolation part of ITS. For details see: Documentation/admin-guide/hw-vuln/indirect-target-selection.rst diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpufeatures.h index 03af8191073a..39e61212ac9a 100644 --- a/arch/x86/include/asm/cpufeatures.h +++ b/arch/x86/include/asm/cpufeatures.h @@ -535,4 +535,5 @@ #define X86_BUG_IBPB_NO_RET X86_BUG(1*32 + 4) /* "ibpb_no_ret" IBPB omits return target predictions */ #define X86_BUG_SPECTRE_V2_USER X86_BUG(1*32 + 5) /* "spectre_v2_user" CPU is affected by Spectre variant 2 attack between user processes */ #define X86_BUG_ITS X86_BUG(1*32 + 6) /* "its" CPU is affected by Indirect Target Selection */ +#define X86_BUG_ITS_NATIVE_ONLY X86_BUG(1*32 + 7) /* "its_native_only" CPU is affected by ITS, VMX is not affected */ #endif /* _ASM_X86_CPUFEATURES_H */ diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 417596ce5352..e919d645d830 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -1203,16 +1203,19 @@ do_cmd_auto: enum its_mitigation_cmd { ITS_CMD_OFF, ITS_CMD_ON, + ITS_CMD_VMEXIT, }; enum its_mitigation { ITS_MITIGATION_OFF, + ITS_MITIGATION_VMEXIT_ONLY, ITS_MITIGATION_ALIGNED_THUNKS, ITS_MITIGATION_RETPOLINE_STUFF, }; static const char * const its_strings[] = { [ITS_MITIGATION_OFF] = "Vulnerable", + [ITS_MITIGATION_VMEXIT_ONLY] = "Mitigation: Vulnerable, KVM: Not affected", [ITS_MITIGATION_ALIGNED_THUNKS] = "Mitigation: Aligned branch/return thunks", [ITS_MITIGATION_RETPOLINE_STUFF] = "Mitigation: Retpolines, Stuffing RSB", }; @@ -1239,6 +1242,8 @@ static int __init its_parse_cmdline(char *str) } else if (!strcmp(str, "force")) { its_cmd = ITS_CMD_ON; setup_force_cpu_bug(X86_BUG_ITS); + } else if (!strcmp(str, "vmexit")) { + its_cmd = ITS_CMD_VMEXIT; } else { pr_err("Ignoring unknown indirect_target_selection option (%s).", str); } @@ -1294,6 +1299,12 @@ static void __init its_select_mitigation(void) case ITS_CMD_OFF: its_mitigation = ITS_MITIGATION_OFF; break; + case ITS_CMD_VMEXIT: + if (boot_cpu_has_bug(X86_BUG_ITS_NATIVE_ONLY)) { + its_mitigation = ITS_MITIGATION_VMEXIT_ONLY; + goto out; + } + fallthrough; case ITS_CMD_ON: its_mitigation = ITS_MITIGATION_ALIGNED_THUNKS; if (!boot_cpu_has(X86_FEATURE_RETPOLINE)) diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index accdee7ab1f7..6a6c3be5af6a 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -1229,6 +1229,8 @@ static const __initconst struct x86_cpu_id cpu_vuln_whitelist[] = { #define RFDS BIT(7) /* CPU is affected by Indirect Target Selection */ #define ITS BIT(8) +/* CPU is affected by Indirect Target Selection, but guest-host isolation is not affected */ +#define ITS_NATIVE_ONLY BIT(9) static const struct x86_cpu_id cpu_vuln_blacklist[] __initconst = { VULNBL_INTEL_STEPS(INTEL_IVYBRIDGE, X86_STEP_MAX, SRBDS), @@ -1249,16 +1251,16 @@ static const struct x86_cpu_id cpu_vuln_blacklist[] __initconst = { VULNBL_INTEL_STEPS(INTEL_KABYLAKE, 0xc, MMIO | RETBLEED | GDS | SRBDS), VULNBL_INTEL_STEPS(INTEL_KABYLAKE, X86_STEP_MAX, MMIO | RETBLEED | GDS | SRBDS | ITS), VULNBL_INTEL_STEPS(INTEL_CANNONLAKE_L, X86_STEP_MAX, RETBLEED), - VULNBL_INTEL_STEPS(INTEL_ICELAKE_L, X86_STEP_MAX, MMIO | MMIO_SBDS | RETBLEED | GDS | ITS), - VULNBL_INTEL_STEPS(INTEL_ICELAKE_D, X86_STEP_MAX, MMIO | GDS | ITS), - VULNBL_INTEL_STEPS(INTEL_ICELAKE_X, X86_STEP_MAX, MMIO | GDS | ITS), + VULNBL_INTEL_STEPS(INTEL_ICELAKE_L, X86_STEP_MAX, MMIO | MMIO_SBDS | RETBLEED | GDS | ITS | ITS_NATIVE_ONLY), + VULNBL_INTEL_STEPS(INTEL_ICELAKE_D, X86_STEP_MAX, MMIO | GDS | ITS | ITS_NATIVE_ONLY), + VULNBL_INTEL_STEPS(INTEL_ICELAKE_X, X86_STEP_MAX, MMIO | GDS | ITS | ITS_NATIVE_ONLY), VULNBL_INTEL_STEPS(INTEL_COMETLAKE, X86_STEP_MAX, MMIO | MMIO_SBDS | RETBLEED | GDS | ITS), VULNBL_INTEL_STEPS(INTEL_COMETLAKE_L, 0x0, MMIO | RETBLEED | ITS), VULNBL_INTEL_STEPS(INTEL_COMETLAKE_L, X86_STEP_MAX, MMIO | MMIO_SBDS | RETBLEED | GDS | ITS), - VULNBL_INTEL_STEPS(INTEL_TIGERLAKE_L, X86_STEP_MAX, GDS | ITS), - VULNBL_INTEL_STEPS(INTEL_TIGERLAKE, X86_STEP_MAX, GDS | ITS), + VULNBL_INTEL_STEPS(INTEL_TIGERLAKE_L, X86_STEP_MAX, GDS | ITS | ITS_NATIVE_ONLY), + VULNBL_INTEL_STEPS(INTEL_TIGERLAKE, X86_STEP_MAX, GDS | ITS | ITS_NATIVE_ONLY), VULNBL_INTEL_STEPS(INTEL_LAKEFIELD, X86_STEP_MAX, MMIO | MMIO_SBDS | RETBLEED), - VULNBL_INTEL_STEPS(INTEL_ROCKETLAKE, X86_STEP_MAX, MMIO | RETBLEED | GDS | ITS), + VULNBL_INTEL_STEPS(INTEL_ROCKETLAKE, X86_STEP_MAX, MMIO | RETBLEED | GDS | ITS | ITS_NATIVE_ONLY), VULNBL_INTEL_TYPE(INTEL_ALDERLAKE, ATOM, RFDS), VULNBL_INTEL_STEPS(INTEL_ALDERLAKE_L, X86_STEP_MAX, RFDS), VULNBL_INTEL_TYPE(INTEL_RAPTORLAKE, ATOM, RFDS), @@ -1480,8 +1482,11 @@ static void __init cpu_set_bug_bits(struct cpuinfo_x86 *c) if (cpu_has(c, X86_FEATURE_AMD_IBPB) && !cpu_has(c, X86_FEATURE_AMD_IBPB_RET)) setup_force_cpu_bug(X86_BUG_IBPB_NO_RET); - if (vulnerable_to_its(x86_arch_cap_msr)) + if (vulnerable_to_its(x86_arch_cap_msr)) { setup_force_cpu_bug(X86_BUG_ITS); + if (cpu_matches(cpu_vuln_blacklist, ITS_NATIVE_ONLY)) + setup_force_cpu_bug(X86_BUG_ITS_NATIVE_ONLY); + } if (cpu_matches(cpu_vuln_whitelist, NO_MELTDOWN)) return; From facd226f7e0c8ca936ac114aba43cb3e8b94e41e Mon Sep 17 00:00:00 2001 From: Pawan Gupta Date: Mon, 2 Dec 2024 12:07:08 -0800 Subject: [PATCH 354/555] x86/its: Add support for RSB stuffing mitigation When retpoline mitigation is enabled for spectre-v2, enabling call-depth-tracking and RSB stuffing also mitigates ITS. Add cmdline option indirect_target_selection=stuff to allow enabling RSB stuffing mitigation. When retpoline mitigation is not enabled, =stuff option is ignored, and default mitigation for ITS is deployed. Signed-off-by: Pawan Gupta Signed-off-by: Dave Hansen Reviewed-by: Josh Poimboeuf Reviewed-by: Alexandre Chartre --- .../admin-guide/kernel-parameters.txt | 3 +++ arch/x86/kernel/cpu/bugs.c | 19 +++++++++++++++++++ 2 files changed, 22 insertions(+) diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index e70d15095f28..8f75ec177399 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -2212,6 +2212,9 @@ mitigation. vmexit: Only deploy mitigation if CPU is affected by guest/host isolation part of ITS. + stuff: Deploy RSB-fill mitigation when retpoline is + also deployed. Otherwise, deploy the default + mitigation. For details see: Documentation/admin-guide/hw-vuln/indirect-target-selection.rst diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index e919d645d830..c6dcc03a43da 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -1204,6 +1204,7 @@ enum its_mitigation_cmd { ITS_CMD_OFF, ITS_CMD_ON, ITS_CMD_VMEXIT, + ITS_CMD_RSB_STUFF, }; enum its_mitigation { @@ -1244,6 +1245,8 @@ static int __init its_parse_cmdline(char *str) setup_force_cpu_bug(X86_BUG_ITS); } else if (!strcmp(str, "vmexit")) { its_cmd = ITS_CMD_VMEXIT; + } else if (!strcmp(str, "stuff")) { + its_cmd = ITS_CMD_RSB_STUFF; } else { pr_err("Ignoring unknown indirect_target_selection option (%s).", str); } @@ -1295,6 +1298,12 @@ static void __init its_select_mitigation(void) goto out; } + if (cmd == ITS_CMD_RSB_STUFF && + (!boot_cpu_has(X86_FEATURE_RETPOLINE) || !IS_ENABLED(CONFIG_MITIGATION_CALL_DEPTH_TRACKING))) { + pr_err("RSB stuff mitigation not supported, using default\n"); + cmd = ITS_CMD_ON; + } + switch (cmd) { case ITS_CMD_OFF: its_mitigation = ITS_MITIGATION_OFF; @@ -1312,6 +1321,16 @@ static void __init its_select_mitigation(void) setup_force_cpu_cap(X86_FEATURE_RETHUNK); set_return_thunk(its_return_thunk); break; + case ITS_CMD_RSB_STUFF: + its_mitigation = ITS_MITIGATION_RETPOLINE_STUFF; + setup_force_cpu_cap(X86_FEATURE_RETHUNK); + setup_force_cpu_cap(X86_FEATURE_CALL_DEPTH); + set_return_thunk(call_depth_return_thunk); + if (retbleed_mitigation == RETBLEED_MITIGATION_NONE) { + retbleed_mitigation = RETBLEED_MITIGATION_STUFF; + pr_info("Retbleed mitigation updated to stuffing\n"); + } + break; } out: pr_info("%s\n", its_strings[its_mitigation]); From f0cd7091cc5a032c8870b4285305d9172569d126 Mon Sep 17 00:00:00 2001 From: Pawan Gupta Date: Fri, 2 May 2025 06:25:19 -0700 Subject: [PATCH 355/555] x86/its: Align RETs in BHB clear sequence to avoid thunking The software mitigation for BHI is to execute BHB clear sequence at syscall entry, and possibly after a cBPF program. ITS mitigation thunks RETs in the lower half of the cacheline. This causes the RETs in the BHB clear sequence to be thunked as well, adding unnecessary branches to the BHB clear sequence. Since the sequence is in hot path, align the RET instructions in the sequence to avoid thunking. This is how disassembly clear_bhb_loop() looks like after this change: 0x44 <+4>: mov $0x5,%ecx 0x49 <+9>: call 0xffffffff81001d9b 0x4e <+14>: jmp 0xffffffff81001de5 0x53 <+19>: int3 ... 0x9b <+91>: call 0xffffffff81001dce 0xa0 <+96>: ret 0xa1 <+97>: int3 ... 0xce <+142>: mov $0x5,%eax 0xd3 <+147>: jmp 0xffffffff81001dd6 0xd5 <+149>: nop 0xd6 <+150>: sub $0x1,%eax 0xd9 <+153>: jne 0xffffffff81001dd3 0xdb <+155>: sub $0x1,%ecx 0xde <+158>: jne 0xffffffff81001d9b 0xe0 <+160>: ret 0xe1 <+161>: int3 0xe2 <+162>: int3 0xe3 <+163>: int3 0xe4 <+164>: int3 0xe5 <+165>: lfence 0xe8 <+168>: pop %rbp 0xe9 <+169>: ret Suggested-by: Andrew Cooper Signed-off-by: Pawan Gupta Signed-off-by: Dave Hansen Reviewed-by: Alexandre Chartre --- arch/x86/entry/entry_64.S | 20 +++++++++++++++++--- 1 file changed, 17 insertions(+), 3 deletions(-) diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S index f40bdf97d390..ed04a968cc7d 100644 --- a/arch/x86/entry/entry_64.S +++ b/arch/x86/entry/entry_64.S @@ -1525,7 +1525,9 @@ SYM_CODE_END(rewind_stack_and_make_dead) * ORC to unwind properly. * * The alignment is for performance and not for safety, and may be safely - * refactored in the future if needed. + * refactored in the future if needed. The .skips are for safety, to ensure + * that all RETs are in the second half of a cacheline to mitigate Indirect + * Target Selection, rather than taking the slowpath via its_return_thunk. */ SYM_FUNC_START(clear_bhb_loop) ANNOTATE_NOENDBR @@ -1536,10 +1538,22 @@ SYM_FUNC_START(clear_bhb_loop) call 1f jmp 5f .align 64, 0xcc + /* + * Shift instructions so that the RET is in the upper half of the + * cacheline and don't take the slowpath to its_return_thunk. + */ + .skip 32 - (.Lret1 - 1f), 0xcc ANNOTATE_INTRA_FUNCTION_CALL 1: call 2f - RET +.Lret1: RET .align 64, 0xcc + /* + * As above shift instructions for RET at .Lret2 as well. + * + * This should be ideally be: .skip 32 - (.Lret2 - 2f), 0xcc + * but some Clang versions (e.g. 18) don't like this. + */ + .skip 32 - 18, 0xcc 2: movl $5, %eax 3: jmp 4f nop @@ -1547,7 +1561,7 @@ SYM_FUNC_START(clear_bhb_loop) jnz 3b sub $1, %ecx jnz 1b - RET +.Lret2: RET 5: lfence pop %rbp RET From 3238532ba55c9bf9b63ae17ebfc5deb320c737c3 Mon Sep 17 00:00:00 2001 From: Wolfram Sang Date: Mon, 31 Mar 2025 21:06:42 +0200 Subject: [PATCH 356/555] MAINTAINERS: delete email for Shiraz Hashim The email address bounced. I couldn't find a newer one in recent git history (last activity 9 years ago), so delete this email entry. Signed-off-by: Wolfram Sang Link: https://lore.kernel.org/r/20250331190731.5094-2-wsa+renesas@sang-engineering.com Signed-off-by: Arnd Bergmann --- MAINTAINERS | 1 - 1 file changed, 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 53ea11c7faf0..786f12b623c7 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -22826,7 +22826,6 @@ F: drivers/accessibility/speakup/ SPEAR PLATFORM/CLOCK/PINCTRL SUPPORT M: Viresh Kumar -M: Shiraz Hashim L: linux-arm-kernel@lists.infradead.org (moderated for non-subscribers) L: soc@lists.linux.dev S: Maintained From 09acc3266cdd8e32a62dd6c5052f44724111120b Mon Sep 17 00:00:00 2001 From: "Rob Herring (Arm)" Date: Wed, 9 Apr 2025 16:02:54 -0500 Subject: [PATCH 357/555] arm64: dts: amazon: Fix simple-bus node name schema warnings Fix a couple of node name warnings from the schema checks: arch/arm64/boot/dts/amazon/alpine-v2-evp.dt.yaml: io-fabric: $nodename:0: 'io-fabric' does not match '^(bus|soc|axi|ahb|apb)(@[0-9a-f]+)?$' arch/arm64/boot/dts/amazon/alpine-v3-evp.dt.yaml: io-fabric: $nodename:0: 'io-fabric' does not match '^(bus|soc|axi|ahb|apb)(@[0-9a-f]+)?$' Signed-off-by: Rob Herring (Arm) Link: https://lore.kernel.org/r/20250409210255.1541298-1-robh@kernel.org Signed-off-by: Arnd Bergmann --- arch/arm64/boot/dts/amazon/alpine-v2.dtsi | 2 +- arch/arm64/boot/dts/amazon/alpine-v3.dtsi | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm64/boot/dts/amazon/alpine-v2.dtsi b/arch/arm64/boot/dts/amazon/alpine-v2.dtsi index da9de4986660..5a72f0b64247 100644 --- a/arch/arm64/boot/dts/amazon/alpine-v2.dtsi +++ b/arch/arm64/boot/dts/amazon/alpine-v2.dtsi @@ -151,7 +151,7 @@ al,msi-num-spis = <160>; }; - io-fabric@fc000000 { + io-bus@fc000000 { compatible = "simple-bus"; #address-cells = <1>; #size-cells = <1>; diff --git a/arch/arm64/boot/dts/amazon/alpine-v3.dtsi b/arch/arm64/boot/dts/amazon/alpine-v3.dtsi index 8b6156b5af65..dea60d136c2e 100644 --- a/arch/arm64/boot/dts/amazon/alpine-v3.dtsi +++ b/arch/arm64/boot/dts/amazon/alpine-v3.dtsi @@ -361,7 +361,7 @@ interrupt-parent = <&gic>; }; - io-fabric@fc000000 { + io-bus@fc000000 { compatible = "simple-bus"; #address-cells = <1>; #size-cells = <1>; From d6d1e3e6580ca35071ad474381f053cbf1fb6414 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 23 Apr 2025 16:30:11 +0200 Subject: [PATCH 358/555] mm/execmem: Unify early execmem_cache behaviour Early kernel memory is RWX, only at the end of early boot (before SMP) do we mark things ROX. Have execmem_cache mirror this behaviour for early users. This avoids having to remember what code is execmem and what is not -- we can poke everything with impunity ;-) Also performance for not having to do endless text_poke_mm switches. Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Pawan Gupta Signed-off-by: Dave Hansen Reviewed-by: Alexandre Chartre --- arch/x86/mm/init_32.c | 3 +++ arch/x86/mm/init_64.c | 3 +++ include/linux/execmem.h | 8 +++++++- mm/execmem.c | 40 +++++++++++++++++++++++++++++++++++++--- 4 files changed, 50 insertions(+), 4 deletions(-) diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c index ad662cc4605c..7fa3965dcef1 100644 --- a/arch/x86/mm/init_32.c +++ b/arch/x86/mm/init_32.c @@ -30,6 +30,7 @@ #include #include #include +#include #include #include @@ -755,6 +756,8 @@ void mark_rodata_ro(void) pr_info("Write protecting kernel text and read-only data: %luk\n", size >> 10); + execmem_cache_make_ro(); + kernel_set_to_readonly = 1; #ifdef CONFIG_CPA_DEBUG diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index 7c4f6f591f2b..949a447f75ec 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -34,6 +34,7 @@ #include #include #include +#include #include #include @@ -1391,6 +1392,8 @@ void mark_rodata_ro(void) (end - start) >> 10); set_memory_ro(start, (end - start) >> PAGE_SHIFT); + execmem_cache_make_ro(); + kernel_set_to_readonly = 1; /* diff --git a/include/linux/execmem.h b/include/linux/execmem.h index 65655a5d1be2..089c71cc8ddf 100644 --- a/include/linux/execmem.h +++ b/include/linux/execmem.h @@ -53,7 +53,7 @@ enum execmem_range_flags { EXECMEM_ROX_CACHE = (1 << 1), }; -#ifdef CONFIG_ARCH_HAS_EXECMEM_ROX +#if defined(CONFIG_ARCH_HAS_EXECMEM_ROX) && defined(CONFIG_EXECMEM) /** * execmem_fill_trapping_insns - set memory to contain instructions that * will trap @@ -93,9 +93,15 @@ int execmem_make_temp_rw(void *ptr, size_t size); * Return: 0 on success or negative error code on failure. */ int execmem_restore_rox(void *ptr, size_t size); + +/* + * Called from mark_readonly(), where the system transitions to ROX. + */ +void execmem_cache_make_ro(void); #else static inline int execmem_make_temp_rw(void *ptr, size_t size) { return 0; } static inline int execmem_restore_rox(void *ptr, size_t size) { return 0; } +static inline void execmem_cache_make_ro(void) { } #endif /** diff --git a/mm/execmem.c b/mm/execmem.c index e6c4f5076ca8..6f7a2653b280 100644 --- a/mm/execmem.c +++ b/mm/execmem.c @@ -254,6 +254,34 @@ out_unlock: return ptr; } +static bool execmem_cache_rox = false; + +void execmem_cache_make_ro(void) +{ + struct maple_tree *free_areas = &execmem_cache.free_areas; + struct maple_tree *busy_areas = &execmem_cache.busy_areas; + MA_STATE(mas_free, free_areas, 0, ULONG_MAX); + MA_STATE(mas_busy, busy_areas, 0, ULONG_MAX); + struct mutex *mutex = &execmem_cache.mutex; + void *area; + + execmem_cache_rox = true; + + mutex_lock(mutex); + + mas_for_each(&mas_free, area, ULONG_MAX) { + unsigned long pages = mas_range_len(&mas_free) >> PAGE_SHIFT; + set_memory_ro(mas_free.index, pages); + } + + mas_for_each(&mas_busy, area, ULONG_MAX) { + unsigned long pages = mas_range_len(&mas_busy) >> PAGE_SHIFT; + set_memory_ro(mas_busy.index, pages); + } + + mutex_unlock(mutex); +} + static int execmem_cache_populate(struct execmem_range *range, size_t size) { unsigned long vm_flags = VM_ALLOW_HUGE_VMAP; @@ -274,9 +302,15 @@ static int execmem_cache_populate(struct execmem_range *range, size_t size) /* fill memory with instructions that will trap */ execmem_fill_trapping_insns(p, alloc_size, /* writable = */ true); - err = set_memory_rox((unsigned long)p, vm->nr_pages); - if (err) - goto err_free_mem; + if (execmem_cache_rox) { + err = set_memory_rox((unsigned long)p, vm->nr_pages); + if (err) + goto err_free_mem; + } else { + err = set_memory_x((unsigned long)p, vm->nr_pages); + if (err) + goto err_free_mem; + } err = execmem_cache_add(p, alloc_size); if (err) From ebebe30794d38c51f71fe4951ba6af4159d9837d Mon Sep 17 00:00:00 2001 From: Pawan Gupta Date: Sat, 3 May 2025 09:46:31 -0700 Subject: [PATCH 359/555] x86/ibt: Keep IBT disabled during alternative patching cfi_rewrite_callers() updates the fineIBT hash matching at the caller side, but except for paranoid-mode it relies on apply_retpoline() and friends for any ENDBR relocation. This could temporarily cause an indirect branch to land on a poisoned ENDBR. For instance, with para-virtualization enabled, a simple wrmsrl() could have an indirect branch pointing to native_write_msr() who's ENDBR has been relocated due to fineIBT: : push %rbp mov %rsp,%rbp mov %esi,%eax mov %rsi,%rdx shr $0x20,%rdx mov %edi,%edi mov %rax,%rsi call *0x21e65d0(%rip) # ^^^^^^^^^^^^^^^^^^^^^^^ Such an indirect call during the alternative patching could #CP if the caller is not *yet* adjusted for the new target ENDBR. To prevent a false #CP, keep CET-IBT disabled until all callers are patched. Patching during the module load does not need to be guarded by IBT-disable because the module code is not executed until the patching is complete. Signed-off-by: Pawan Gupta Signed-off-by: Dave Hansen Reviewed-by: Alexandre Chartre --- arch/x86/kernel/alternative.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c index 3e67e72bffdc..6e0b09b2c972 100644 --- a/arch/x86/kernel/alternative.c +++ b/arch/x86/kernel/alternative.c @@ -31,6 +31,7 @@ #include #include #include +#include int __read_mostly alternatives_patched; @@ -2085,6 +2086,8 @@ static noinline void __init alt_reloc_selftest(void) void __init alternative_instructions(void) { + u64 ibt; + int3_selftest(); /* @@ -2111,6 +2114,9 @@ void __init alternative_instructions(void) */ paravirt_set_cap(); + /* Keep CET-IBT disabled until caller/callee are patched */ + ibt = ibt_save(/*disable*/ true); + __apply_fineibt(__retpoline_sites, __retpoline_sites_end, __cfi_sites, __cfi_sites_end, true); @@ -2134,6 +2140,8 @@ void __init alternative_instructions(void) */ apply_seal_endbr(__ibt_endbr_seal, __ibt_endbr_seal_end); + ibt_restore(ibt); + #ifdef CONFIG_SMP /* Patch to UP if other cpus not imminent. */ if (!noreplace_smp && (num_present_cpus() == 1 || setup_max_cpus <= 1)) { From 872df34d7c51a79523820ea6a14860398c639b87 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 14 Oct 2024 10:05:48 -0700 Subject: [PATCH 360/555] x86/its: Use dynamic thunks for indirect branches ITS mitigation moves the unsafe indirect branches to a safe thunk. This could degrade the prediction accuracy as the source address of indirect branches becomes same for different execution paths. To improve the predictions, and hence the performance, assign a separate thunk for each indirect callsite. This is also a defense-in-depth measure to avoid indirect branches aliasing with each other. As an example, 5000 dynamic thunks would utilize around 16 bits of the address space, thereby gaining entropy. For a BTB that uses 32 bits for indexing, dynamic thunks could provide better prediction accuracy over fixed thunks. Have ITS thunks be variable sized and use EXECMEM_MODULE_TEXT such that they are both more flexible (got to extend them later) and live in 2M TLBs, just like kernel code, avoiding undue TLB pressure. Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Pawan Gupta Signed-off-by: Dave Hansen Reviewed-by: Alexandre Chartre --- arch/x86/Kconfig | 1 + arch/x86/include/asm/alternative.h | 10 +++ arch/x86/kernel/alternative.c | 127 ++++++++++++++++++++++++++++- arch/x86/kernel/module.c | 6 ++ include/linux/execmem.h | 3 + include/linux/module.h | 5 ++ 6 files changed, 149 insertions(+), 3 deletions(-) diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index a6bd0590c437..a1156a554f6d 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -2714,6 +2714,7 @@ config MITIGATION_ITS bool "Enable Indirect Target Selection mitigation" depends on CPU_SUP_INTEL && X86_64 depends on MITIGATION_RETPOLINE && MITIGATION_RETHUNK + select EXECMEM default y help Enable Indirect Target Selection (ITS) mitigation. ITS is a bug in diff --git a/arch/x86/include/asm/alternative.h b/arch/x86/include/asm/alternative.h index 4037c611537c..47948ebbb409 100644 --- a/arch/x86/include/asm/alternative.h +++ b/arch/x86/include/asm/alternative.h @@ -124,6 +124,16 @@ static __always_inline int x86_call_depth_emit_accounting(u8 **pprog, } #endif +#ifdef CONFIG_MITIGATION_ITS +extern void its_init_mod(struct module *mod); +extern void its_fini_mod(struct module *mod); +extern void its_free_mod(struct module *mod); +#else /* CONFIG_MITIGATION_ITS */ +static inline void its_init_mod(struct module *mod) { } +static inline void its_fini_mod(struct module *mod) { } +static inline void its_free_mod(struct module *mod) { } +#endif + #if defined(CONFIG_MITIGATION_RETHUNK) && defined(CONFIG_OBJTOOL) extern bool cpu_wants_rethunk(void); extern bool cpu_wants_rethunk_at(void *addr); diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c index 6e0b09b2c972..7a10e3ed5d0b 100644 --- a/arch/x86/kernel/alternative.c +++ b/arch/x86/kernel/alternative.c @@ -18,6 +18,7 @@ #include #include #include +#include #include #include #include @@ -32,6 +33,7 @@ #include #include #include +#include int __read_mostly alternatives_patched; @@ -125,6 +127,121 @@ const unsigned char * const x86_nops[ASM_NOP_MAX+1] = #endif }; +#ifdef CONFIG_MITIGATION_ITS + +static struct module *its_mod; +static void *its_page; +static unsigned int its_offset; + +/* Initialize a thunk with the "jmp *reg; int3" instructions. */ +static void *its_init_thunk(void *thunk, int reg) +{ + u8 *bytes = thunk; + int i = 0; + + if (reg >= 8) { + bytes[i++] = 0x41; /* REX.B prefix */ + reg -= 8; + } + bytes[i++] = 0xff; + bytes[i++] = 0xe0 + reg; /* jmp *reg */ + bytes[i++] = 0xcc; + + return thunk; +} + +void its_init_mod(struct module *mod) +{ + if (!cpu_feature_enabled(X86_FEATURE_INDIRECT_THUNK_ITS)) + return; + + mutex_lock(&text_mutex); + its_mod = mod; + its_page = NULL; +} + +void its_fini_mod(struct module *mod) +{ + if (!cpu_feature_enabled(X86_FEATURE_INDIRECT_THUNK_ITS)) + return; + + WARN_ON_ONCE(its_mod != mod); + + its_mod = NULL; + its_page = NULL; + mutex_unlock(&text_mutex); + + for (int i = 0; i < mod->its_num_pages; i++) { + void *page = mod->its_page_array[i]; + execmem_restore_rox(page, PAGE_SIZE); + } +} + +void its_free_mod(struct module *mod) +{ + if (!cpu_feature_enabled(X86_FEATURE_INDIRECT_THUNK_ITS)) + return; + + for (int i = 0; i < mod->its_num_pages; i++) { + void *page = mod->its_page_array[i]; + execmem_free(page); + } + kfree(mod->its_page_array); +} + +static void *its_alloc(void) +{ + void *page __free(execmem) = execmem_alloc(EXECMEM_MODULE_TEXT, PAGE_SIZE); + + if (!page) + return NULL; + + if (its_mod) { + void *tmp = krealloc(its_mod->its_page_array, + (its_mod->its_num_pages+1) * sizeof(void *), + GFP_KERNEL); + if (!tmp) + return NULL; + + its_mod->its_page_array = tmp; + its_mod->its_page_array[its_mod->its_num_pages++] = page; + + execmem_make_temp_rw(page, PAGE_SIZE); + } + + return no_free_ptr(page); +} + +static void *its_allocate_thunk(int reg) +{ + int size = 3 + (reg / 8); + void *thunk; + + if (!its_page || (its_offset + size - 1) >= PAGE_SIZE) { + its_page = its_alloc(); + if (!its_page) { + pr_err("ITS page allocation failed\n"); + return NULL; + } + memset(its_page, INT3_INSN_OPCODE, PAGE_SIZE); + its_offset = 32; + } + + /* + * If the indirect branch instruction will be in the lower half + * of a cacheline, then update the offset to reach the upper half. + */ + if ((its_offset + size - 1) % 64 < 32) + its_offset = ((its_offset - 1) | 0x3F) + 33; + + thunk = its_page + its_offset; + its_offset += size; + + return its_init_thunk(thunk, reg); +} + +#endif + /* * Nomenclature for variable names to simplify and clarify this code and ease * any potential staring at it: @@ -637,9 +754,13 @@ static int emit_call_track_retpoline(void *addr, struct insn *insn, int reg, u8 #ifdef CONFIG_MITIGATION_ITS static int emit_its_trampoline(void *addr, struct insn *insn, int reg, u8 *bytes) { - return __emit_trampoline(addr, insn, bytes, - __x86_indirect_its_thunk_array[reg], - __x86_indirect_its_thunk_array[reg]); + u8 *thunk = __x86_indirect_its_thunk_array[reg]; + u8 *tmp = its_allocate_thunk(reg); + + if (tmp) + thunk = tmp; + + return __emit_trampoline(addr, insn, bytes, thunk, thunk); } /* Check if an indirect branch is at ITS-unsafe address */ diff --git a/arch/x86/kernel/module.c b/arch/x86/kernel/module.c index a7998f351701..ff07558b7ebc 100644 --- a/arch/x86/kernel/module.c +++ b/arch/x86/kernel/module.c @@ -266,6 +266,8 @@ int module_finalize(const Elf_Ehdr *hdr, ibt_endbr = s; } + its_init_mod(me); + if (retpolines || cfi) { void *rseg = NULL, *cseg = NULL; unsigned int rsize = 0, csize = 0; @@ -286,6 +288,9 @@ int module_finalize(const Elf_Ehdr *hdr, void *rseg = (void *)retpolines->sh_addr; apply_retpolines(rseg, rseg + retpolines->sh_size); } + + its_fini_mod(me); + if (returns) { void *rseg = (void *)returns->sh_addr; apply_returns(rseg, rseg + returns->sh_size); @@ -326,4 +331,5 @@ int module_finalize(const Elf_Ehdr *hdr, void module_arch_cleanup(struct module *mod) { alternatives_smp_module_del(mod); + its_free_mod(mod); } diff --git a/include/linux/execmem.h b/include/linux/execmem.h index 089c71cc8ddf..ca42d5e46ccc 100644 --- a/include/linux/execmem.h +++ b/include/linux/execmem.h @@ -4,6 +4,7 @@ #include #include +#include #if (defined(CONFIG_KASAN_GENERIC) || defined(CONFIG_KASAN_SW_TAGS)) && \ !defined(CONFIG_KASAN_VMALLOC) @@ -176,6 +177,8 @@ void *execmem_alloc(enum execmem_type type, size_t size); */ void execmem_free(void *ptr); +DEFINE_FREE(execmem, void *, if (_T) execmem_free(_T)); + #ifdef CONFIG_MMU /** * execmem_vmap - create virtual mapping for EXECMEM_MODULE_DATA memory diff --git a/include/linux/module.h b/include/linux/module.h index b3329110d668..8050f77c3b64 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -586,6 +586,11 @@ struct module { atomic_t refcnt; #endif +#ifdef CONFIG_MITIGATION_ITS + int its_num_pages; + void **its_page_array; +#endif + #ifdef CONFIG_CONSTRUCTORS /* Constructor functions. */ ctor_fn_t *ctors; From e52c1dc7455d32c8a55f9949d300e5e87d011fa6 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 23 Apr 2025 09:57:31 +0200 Subject: [PATCH 361/555] x86/its: FineIBT-paranoid vs ITS FineIBT-paranoid was using the retpoline bytes for the paranoid check, disabling retpolines, because all parts that have IBT also have eIBRS and thus don't need no stinking retpolines. Except... ITS needs the retpolines for indirect calls must not be in the first half of a cacheline :-/ So what was the paranoid call sequence: : 0: 41 ba 78 56 34 12 mov $0x12345678, %r10d 6: 45 3b 53 f7 cmp -0x9(%r11), %r10d a: 4d 8d 5b lea -0x10(%r11), %r11 e: 75 fd jne d 10: 41 ff d3 call *%r11 13: 90 nop Now becomes: : 0: 41 ba 78 56 34 12 mov $0x12345678, %r10d 6: 45 3b 53 f7 cmp -0x9(%r11), %r10d a: 4d 8d 5b f0 lea -0x10(%r11), %r11 e: 2e e8 XX XX XX XX cs call __x86_indirect_paranoid_thunk_r11 Where the paranoid_thunk looks like: 1d: (bad) __x86_indirect_paranoid_thunk_r11: 1e: 75 fd jne 1d __x86_indirect_its_thunk_r11: 20: 41 ff eb jmp *%r11 23: cc int3 [ dhansen: remove initialization to false ] Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Pawan Gupta Signed-off-by: Dave Hansen Reviewed-by: Alexandre Chartre --- arch/x86/include/asm/alternative.h | 8 ++ arch/x86/kernel/alternative.c | 145 +++++++++++++++++++++++++---- arch/x86/lib/retpoline.S | 15 ++- arch/x86/net/bpf_jit_comp.c | 2 +- tools/objtool/arch/x86/decode.c | 9 ++ 5 files changed, 159 insertions(+), 20 deletions(-) diff --git a/arch/x86/include/asm/alternative.h b/arch/x86/include/asm/alternative.h index 47948ebbb409..f2294784babc 100644 --- a/arch/x86/include/asm/alternative.h +++ b/arch/x86/include/asm/alternative.h @@ -6,6 +6,7 @@ #include #include #include +#include #define ALT_FLAGS_SHIFT 16 @@ -128,10 +129,17 @@ static __always_inline int x86_call_depth_emit_accounting(u8 **pprog, extern void its_init_mod(struct module *mod); extern void its_fini_mod(struct module *mod); extern void its_free_mod(struct module *mod); +extern u8 *its_static_thunk(int reg); #else /* CONFIG_MITIGATION_ITS */ static inline void its_init_mod(struct module *mod) { } static inline void its_fini_mod(struct module *mod) { } static inline void its_free_mod(struct module *mod) { } +static inline u8 *its_static_thunk(int reg) +{ + WARN_ONCE(1, "ITS not compiled in"); + + return NULL; +} #endif #if defined(CONFIG_MITIGATION_RETHUNK) && defined(CONFIG_OBJTOOL) diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c index 7a10e3ed5d0b..48fd04e90114 100644 --- a/arch/x86/kernel/alternative.c +++ b/arch/x86/kernel/alternative.c @@ -127,6 +127,10 @@ const unsigned char * const x86_nops[ASM_NOP_MAX+1] = #endif }; +#ifdef CONFIG_FINEIBT +static bool cfi_paranoid __ro_after_init; +#endif + #ifdef CONFIG_MITIGATION_ITS static struct module *its_mod; @@ -137,8 +141,25 @@ static unsigned int its_offset; static void *its_init_thunk(void *thunk, int reg) { u8 *bytes = thunk; + int offset = 0; int i = 0; +#ifdef CONFIG_FINEIBT + if (cfi_paranoid) { + /* + * When ITS uses indirect branch thunk the fineibt_paranoid + * caller sequence doesn't fit in the caller site. So put the + * remaining part of the sequence ( + JNE) into the ITS + * thunk. + */ + bytes[i++] = 0xea; /* invalid instruction */ + bytes[i++] = 0x75; /* JNE */ + bytes[i++] = 0xfd; + + offset = 1; + } +#endif + if (reg >= 8) { bytes[i++] = 0x41; /* REX.B prefix */ reg -= 8; @@ -147,7 +168,7 @@ static void *its_init_thunk(void *thunk, int reg) bytes[i++] = 0xe0 + reg; /* jmp *reg */ bytes[i++] = 0xcc; - return thunk; + return thunk + offset; } void its_init_mod(struct module *mod) @@ -217,6 +238,17 @@ static void *its_allocate_thunk(int reg) int size = 3 + (reg / 8); void *thunk; +#ifdef CONFIG_FINEIBT + /* + * The ITS thunk contains an indirect jump and an int3 instruction so + * its size is 3 or 4 bytes depending on the register used. If CFI + * paranoid is used then 3 extra bytes are added in the ITS thunk to + * complete the fineibt_paranoid caller sequence. + */ + if (cfi_paranoid) + size += 3; +#endif + if (!its_page || (its_offset + size - 1) >= PAGE_SIZE) { its_page = its_alloc(); if (!its_page) { @@ -240,6 +272,18 @@ static void *its_allocate_thunk(int reg) return its_init_thunk(thunk, reg); } +u8 *its_static_thunk(int reg) +{ + u8 *thunk = __x86_indirect_its_thunk_array[reg]; + +#ifdef CONFIG_FINEIBT + /* Paranoid thunk starts 2 bytes before */ + if (cfi_paranoid) + return thunk - 2; +#endif + return thunk; +} + #endif /* @@ -775,8 +819,17 @@ static bool cpu_wants_indirect_its_thunk_at(unsigned long addr, int reg) /* Lower-half of the cacheline? */ return !(addr & 0x20); } +#else /* CONFIG_MITIGATION_ITS */ + +#ifdef CONFIG_FINEIBT +static bool cpu_wants_indirect_its_thunk_at(unsigned long addr, int reg) +{ + return false; +} #endif +#endif /* CONFIG_MITIGATION_ITS */ + /* * Rewrite the compiler generated retpoline thunk calls. * @@ -893,6 +946,7 @@ void __init_or_module noinline apply_retpolines(s32 *start, s32 *end) int len, ret; u8 bytes[16]; u8 op1, op2; + u8 *dest; ret = insn_decode_kernel(&insn, addr); if (WARN_ON_ONCE(ret < 0)) @@ -909,6 +963,12 @@ void __init_or_module noinline apply_retpolines(s32 *start, s32 *end) case CALL_INSN_OPCODE: case JMP32_INSN_OPCODE: + /* Check for cfi_paranoid + ITS */ + dest = addr + insn.length + insn.immediate.value; + if (dest[-1] == 0xea && (dest[0] & 0xf0) == 0x70) { + WARN_ON_ONCE(cfi_mode != CFI_FINEIBT); + continue; + } break; case 0x0f: /* escape */ @@ -1198,8 +1258,6 @@ int cfi_get_func_arity(void *func) static bool cfi_rand __ro_after_init = true; static u32 cfi_seed __ro_after_init; -static bool cfi_paranoid __ro_after_init = false; - /* * Re-hash the CFI hash with a boot-time seed while making sure the result is * not a valid ENDBR instruction. @@ -1612,6 +1670,19 @@ static int cfi_rand_callers(s32 *start, s32 *end) return 0; } +static int emit_paranoid_trampoline(void *addr, struct insn *insn, int reg, u8 *bytes) +{ + u8 *thunk = (void *)__x86_indirect_its_thunk_array[reg] - 2; + +#ifdef CONFIG_MITIGATION_ITS + u8 *tmp = its_allocate_thunk(reg); + if (tmp) + thunk = tmp; +#endif + + return __emit_trampoline(addr, insn, bytes, thunk, thunk); +} + static int cfi_rewrite_callers(s32 *start, s32 *end) { s32 *s; @@ -1653,9 +1724,14 @@ static int cfi_rewrite_callers(s32 *start, s32 *end) memcpy(bytes, fineibt_paranoid_start, fineibt_paranoid_size); memcpy(bytes + fineibt_caller_hash, &hash, 4); - ret = emit_indirect(op, 11, bytes + fineibt_paranoid_ind); - if (WARN_ON_ONCE(ret != 3)) - continue; + if (cpu_wants_indirect_its_thunk_at((unsigned long)addr + fineibt_paranoid_ind, 11)) { + emit_paranoid_trampoline(addr + fineibt_caller_size, + &insn, 11, bytes + fineibt_caller_size); + } else { + ret = emit_indirect(op, 11, bytes + fineibt_paranoid_ind); + if (WARN_ON_ONCE(ret != 3)) + continue; + } text_poke_early(addr, bytes, fineibt_paranoid_size); } @@ -1882,29 +1958,66 @@ Efault: return false; } +static bool is_paranoid_thunk(unsigned long addr) +{ + u32 thunk; + + __get_kernel_nofault(&thunk, (u32 *)addr, u32, Efault); + return (thunk & 0x00FFFFFF) == 0xfd75ea; + +Efault: + return false; +} + /* * regs->ip points to a LOCK Jcc.d8 instruction from the fineibt_paranoid_start[] - * sequence. + * sequence, or to an invalid instruction (0xea) + Jcc.d8 for cfi_paranoid + ITS + * thunk. */ static bool decode_fineibt_paranoid(struct pt_regs *regs, unsigned long *target, u32 *type) { unsigned long addr = regs->ip - fineibt_paranoid_ud; - u32 hash; - if (!cfi_paranoid || !is_cfi_trap(addr + fineibt_caller_size - LEN_UD2)) + if (!cfi_paranoid) return false; - __get_kernel_nofault(&hash, addr + fineibt_caller_hash, u32, Efault); - *target = regs->r11 + fineibt_preamble_size; - *type = regs->r10; + if (is_cfi_trap(addr + fineibt_caller_size - LEN_UD2)) { + *target = regs->r11 + fineibt_preamble_size; + *type = regs->r10; + + /* + * Since the trapping instruction is the exact, but LOCK prefixed, + * Jcc.d8 that got us here, the normal fixup will work. + */ + return true; + } /* - * Since the trapping instruction is the exact, but LOCK prefixed, - * Jcc.d8 that got us here, the normal fixup will work. + * The cfi_paranoid + ITS thunk combination results in: + * + * 0: 41 ba 78 56 34 12 mov $0x12345678, %r10d + * 6: 45 3b 53 f7 cmp -0x9(%r11), %r10d + * a: 4d 8d 5b f0 lea -0x10(%r11), %r11 + * e: 2e e8 XX XX XX XX cs call __x86_indirect_paranoid_thunk_r11 + * + * Where the paranoid_thunk looks like: + * + * 1d: (bad) + * __x86_indirect_paranoid_thunk_r11: + * 1e: 75 fd jne 1d + * __x86_indirect_its_thunk_r11: + * 20: 41 ff eb jmp *%r11 + * 23: cc int3 + * */ - return true; + if (is_paranoid_thunk(regs->ip)) { + *target = regs->r11 + fineibt_preamble_size; + *type = regs->r10; + + regs->ip = *target; + return true; + } -Efault: return false; } diff --git a/arch/x86/lib/retpoline.S b/arch/x86/lib/retpoline.S index ebca28fe7e31..39374949daa2 100644 --- a/arch/x86/lib/retpoline.S +++ b/arch/x86/lib/retpoline.S @@ -371,6 +371,15 @@ SYM_FUNC_END(call_depth_return_thunk) .macro ITS_THUNK reg +/* + * If CFI paranoid is used then the ITS thunk starts with opcodes (0xea; jne 1b) + * that complete the fineibt_paranoid caller sequence. + */ +1: .byte 0xea +SYM_INNER_LABEL(__x86_indirect_paranoid_thunk_\reg, SYM_L_GLOBAL) + UNWIND_HINT_UNDEFINED + ANNOTATE_NOENDBR + jne 1b SYM_INNER_LABEL(__x86_indirect_its_thunk_\reg, SYM_L_GLOBAL) UNWIND_HINT_UNDEFINED ANNOTATE_NOENDBR @@ -378,19 +387,19 @@ SYM_INNER_LABEL(__x86_indirect_its_thunk_\reg, SYM_L_GLOBAL) jmp *%\reg int3 .align 32, 0xcc /* fill to the end of the line */ - .skip 32, 0xcc /* skip to the next upper half */ + .skip 32 - (__x86_indirect_its_thunk_\reg - 1b), 0xcc /* skip to the next upper half */ .endm /* ITS mitigation requires thunks be aligned to upper half of cacheline */ .align 64, 0xcc -.skip 32, 0xcc -SYM_CODE_START(__x86_indirect_its_thunk_array) +.skip 29, 0xcc #define GEN(reg) ITS_THUNK reg #include #undef GEN .align 64, 0xcc +SYM_FUNC_ALIAS(__x86_indirect_its_thunk_array, __x86_indirect_its_thunk_rax) SYM_CODE_END(__x86_indirect_its_thunk_array) .align 64, 0xcc diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c index a5b65c09910b..a31e58c6d89e 100644 --- a/arch/x86/net/bpf_jit_comp.c +++ b/arch/x86/net/bpf_jit_comp.c @@ -663,7 +663,7 @@ static void emit_indirect_jump(u8 **pprog, int reg, u8 *ip) if (cpu_feature_enabled(X86_FEATURE_INDIRECT_THUNK_ITS)) { OPTIMIZER_HIDE_VAR(reg); - emit_jump(&prog, &__x86_indirect_its_thunk_array[reg], ip); + emit_jump(&prog, its_static_thunk(reg), ip); } else if (cpu_feature_enabled(X86_FEATURE_RETPOLINE_LFENCE)) { EMIT_LFENCE(); EMIT2(0xFF, 0xE0 + reg); diff --git a/tools/objtool/arch/x86/decode.c b/tools/objtool/arch/x86/decode.c index 3ce7b54003c2..687c5eafb49a 100644 --- a/tools/objtool/arch/x86/decode.c +++ b/tools/objtool/arch/x86/decode.c @@ -189,6 +189,15 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec op2 = ins.opcode.bytes[1]; op3 = ins.opcode.bytes[2]; + /* + * XXX hack, decoder is buggered and thinks 0xea is 7 bytes long. + */ + if (op1 == 0xea) { + insn->len = 1; + insn->type = INSN_BUG; + return 0; + } + if (ins.rex_prefix.nbytes) { rex = ins.rex_prefix.bytes[0]; rex_w = X86_REX_W(rex) >> 3; From 7a9b709e7cc5ce1ffb84ce07bf6d157e1de758df Mon Sep 17 00:00:00 2001 From: Pawan Gupta Date: Tue, 24 Dec 2024 16:09:28 -0800 Subject: [PATCH 362/555] selftest/x86/bugs: Add selftests for ITS Below are the tests added for Indirect Target Selection (ITS): - its_sysfs.py - Check if sysfs reflects the correct mitigation status for the mitigation selected via the kernel cmdline. - its_permutations.py - tests mitigation selection with cmdline permutations with other bugs like spectre_v2 and retbleed. - its_indirect_alignment.py - verifies that for addresses in .retpoline_sites section that belong to lower half of cacheline are patched to ITS-safe thunk. Typical output looks like below: Site 49: function symbol: __x64_sys_restart_syscall+0x1f <0xffffffffbb1509af> # vmlinux: 0xffffffff813509af: jmp 0xffffffff81f5a8e0 # kcore: 0xffffffffbb1509af: jmpq *%rax # ITS thunk NOT expected for site 49 # PASSED: Found *%rax # Site 50: function symbol: __resched_curr+0xb0 <0xffffffffbb181910> # vmlinux: 0xffffffff81381910: jmp 0xffffffff81f5a8e0 # kcore: 0xffffffffbb181910: jmp 0xffffffffc02000fc # ITS thunk expected for site 50 # PASSED: Found 0xffffffffc02000fc -> jmpq *%rax - its_ret_alignment.py - verifies that for addresses in .return_sites section that belong to lower half of cacheline are patched to its_return_thunk. Typical output looks like below: Site 97: function symbol: collect_event+0x48 <0xffffffffbb007f18> # vmlinux: 0xffffffff81207f18: jmp 0xffffffff81f5b500 # kcore: 0xffffffffbb007f18: jmp 0xffffffffbbd5b560 # PASSED: Found jmp 0xffffffffbbd5b560 # Site 98: function symbol: collect_event+0xa4 <0xffffffffbb007f74> # vmlinux: 0xffffffff81207f74: jmp 0xffffffff81f5b500 # kcore: 0xffffffffbb007f74: retq # PASSED: Found retq Some of these tests have dependency on tools like virtme-ng[1] and drgn[2]. When the dependencies are not met, the test will be skipped. [1] https://github.com/arighi/virtme-ng [2] https://github.com/osandov/drgn Co-developed-by: Tao Zhang Signed-off-by: Tao Zhang Signed-off-by: Pawan Gupta Signed-off-by: Dave Hansen --- tools/testing/selftests/Makefile | 1 + tools/testing/selftests/x86/bugs/Makefile | 3 + tools/testing/selftests/x86/bugs/common.py | 164 ++++++++++++++++++ .../x86/bugs/its_indirect_alignment.py | 150 ++++++++++++++++ .../selftests/x86/bugs/its_permutations.py | 109 ++++++++++++ .../selftests/x86/bugs/its_ret_alignment.py | 139 +++++++++++++++ tools/testing/selftests/x86/bugs/its_sysfs.py | 65 +++++++ 7 files changed, 631 insertions(+) create mode 100644 tools/testing/selftests/x86/bugs/Makefile create mode 100755 tools/testing/selftests/x86/bugs/common.py create mode 100755 tools/testing/selftests/x86/bugs/its_indirect_alignment.py create mode 100755 tools/testing/selftests/x86/bugs/its_permutations.py create mode 100755 tools/testing/selftests/x86/bugs/its_ret_alignment.py create mode 100755 tools/testing/selftests/x86/bugs/its_sysfs.py diff --git a/tools/testing/selftests/Makefile b/tools/testing/selftests/Makefile index c77c8c8e3d9b..80fb84fa3cfc 100644 --- a/tools/testing/selftests/Makefile +++ b/tools/testing/selftests/Makefile @@ -121,6 +121,7 @@ TARGETS += user_events TARGETS += vDSO TARGETS += mm TARGETS += x86 +TARGETS += x86/bugs TARGETS += zram #Please keep the TARGETS list alphabetically sorted # Run "make quicktest=1 run_tests" or diff --git a/tools/testing/selftests/x86/bugs/Makefile b/tools/testing/selftests/x86/bugs/Makefile new file mode 100644 index 000000000000..8ff2d7226c7f --- /dev/null +++ b/tools/testing/selftests/x86/bugs/Makefile @@ -0,0 +1,3 @@ +TEST_PROGS := its_sysfs.py its_permutations.py its_indirect_alignment.py its_ret_alignment.py +TEST_FILES := common.py +include ../../lib.mk diff --git a/tools/testing/selftests/x86/bugs/common.py b/tools/testing/selftests/x86/bugs/common.py new file mode 100755 index 000000000000..2f9664a80617 --- /dev/null +++ b/tools/testing/selftests/x86/bugs/common.py @@ -0,0 +1,164 @@ +#!/usr/bin/env python3 +# SPDX-License-Identifier: GPL-2.0 +# +# Copyright (c) 2025 Intel Corporation +# +# This contains kselftest framework adapted common functions for testing +# mitigation for x86 bugs. + +import os, sys, re, shutil + +sys.path.insert(0, '../../kselftest') +import ksft + +def read_file(path): + if not os.path.exists(path): + return None + with open(path, 'r') as file: + return file.read().strip() + +def cpuinfo_has(arg): + cpuinfo = read_file('/proc/cpuinfo') + if arg in cpuinfo: + return True + return False + +def cmdline_has(arg): + cmdline = read_file('/proc/cmdline') + if arg in cmdline: + return True + return False + +def cmdline_has_either(args): + cmdline = read_file('/proc/cmdline') + for arg in args: + if arg in cmdline: + return True + return False + +def cmdline_has_none(args): + return not cmdline_has_either(args) + +def cmdline_has_all(args): + cmdline = read_file('/proc/cmdline') + for arg in args: + if arg not in cmdline: + return False + return True + +def get_sysfs(bug): + return read_file("/sys/devices/system/cpu/vulnerabilities/" + bug) + +def sysfs_has(bug, mitigation): + status = get_sysfs(bug) + if mitigation in status: + return True + return False + +def sysfs_has_either(bugs, mitigations): + for bug in bugs: + for mitigation in mitigations: + if sysfs_has(bug, mitigation): + return True + return False + +def sysfs_has_none(bugs, mitigations): + return not sysfs_has_either(bugs, mitigations) + +def sysfs_has_all(bugs, mitigations): + for bug in bugs: + for mitigation in mitigations: + if not sysfs_has(bug, mitigation): + return False + return True + +def bug_check_pass(bug, found): + ksft.print_msg(f"\nFound: {found}") + # ksft.print_msg(f"\ncmdline: {read_file('/proc/cmdline')}") + ksft.test_result_pass(f'{bug}: {found}') + +def bug_check_fail(bug, found, expected): + ksft.print_msg(f'\nFound:\t {found}') + ksft.print_msg(f'Expected:\t {expected}') + ksft.print_msg(f"\ncmdline: {read_file('/proc/cmdline')}") + ksft.test_result_fail(f'{bug}: {found}') + +def bug_status_unknown(bug, found): + ksft.print_msg(f'\nUnknown status: {found}') + ksft.print_msg(f"\ncmdline: {read_file('/proc/cmdline')}") + ksft.test_result_fail(f'{bug}: {found}') + +def basic_checks_sufficient(bug, mitigation): + if not mitigation: + bug_status_unknown(bug, "None") + return True + elif mitigation == "Not affected": + ksft.test_result_pass(bug) + return True + elif mitigation == "Vulnerable": + if cmdline_has_either([f'{bug}=off', 'mitigations=off']): + bug_check_pass(bug, mitigation) + return True + return False + +def get_section_info(vmlinux, section_name): + from elftools.elf.elffile import ELFFile + with open(vmlinux, 'rb') as f: + elffile = ELFFile(f) + section = elffile.get_section_by_name(section_name) + if section is None: + ksft.print_msg("Available sections in vmlinux:") + for sec in elffile.iter_sections(): + ksft.print_msg(sec.name) + raise ValueError(f"Section {section_name} not found in {vmlinux}") + return section['sh_addr'], section['sh_offset'], section['sh_size'] + +def get_patch_sites(vmlinux, offset, size): + import struct + output = [] + with open(vmlinux, 'rb') as f: + f.seek(offset) + i = 0 + while i < size: + data = f.read(4) # s32 + if not data: + break + sym_offset = struct.unpack(' 1: + arg_vmlinux = os.sys.argv[1] + if not os.path.exists(arg_vmlinux): + ksft.test_result_fail(f"its_indirect_alignment.py: vmlinux not found at argument path: {arg_vmlinux}") + ksft.exit_fail() + os.makedirs(f"/usr/lib/debug/lib/modules/{os.uname().release}", exist_ok=True) + os.system(f'cp {arg_vmlinux} /usr/lib/debug/lib/modules/$(uname -r)/vmlinux') + +vmlinux = f"/usr/lib/debug/lib/modules/{os.uname().release}/vmlinux" +if not os.path.exists(vmlinux): + ksft.test_result_fail(f"its_indirect_alignment.py: vmlinux not found at {vmlinux}") + ksft.exit_fail() + +ksft.print_msg(f"Using vmlinux: {vmlinux}") + +retpolines_start_vmlinux, retpolines_sec_offset, size = c.get_section_info(vmlinux, '.retpoline_sites') +ksft.print_msg(f"vmlinux: Section .retpoline_sites (0x{retpolines_start_vmlinux:x}) found at 0x{retpolines_sec_offset:x} with size 0x{size:x}") + +sites_offset = c.get_patch_sites(vmlinux, retpolines_sec_offset, size) +total_retpoline_tests = len(sites_offset) +ksft.print_msg(f"Found {total_retpoline_tests} retpoline sites") + +prog = c.get_runtime_kernel() +retpolines_start_kcore = prog.symbol('__retpoline_sites').address +ksft.print_msg(f'kcore: __retpoline_sites: 0x{retpolines_start_kcore:x}') + +x86_indirect_its_thunk_r15 = prog.symbol('__x86_indirect_its_thunk_r15').address +ksft.print_msg(f'kcore: __x86_indirect_its_thunk_r15: 0x{x86_indirect_its_thunk_r15:x}') + +tests_passed = 0 +tests_failed = 0 +tests_unknown = 0 + +with open(vmlinux, 'rb') as f: + elffile = ELFFile(f) + text_section = elffile.get_section_by_name('.text') + + for i in range(0, len(sites_offset)): + site = retpolines_start_kcore + sites_offset[i] + vmlinux_site = retpolines_start_vmlinux + sites_offset[i] + passed = unknown = failed = False + try: + vmlinux_insn = c.get_instruction_from_vmlinux(elffile, text_section, text_section['sh_addr'], vmlinux_site) + kcore_insn = list(cap.disasm(prog.read(site, 16), site))[0] + operand = kcore_insn.op_str + insn_end = site + kcore_insn.size - 1 # TODO handle Jcc.32 __x86_indirect_thunk_\reg + safe_site = insn_end & 0x20 + site_status = "" if safe_site else "(unsafe)" + + ksft.print_msg(f"\nSite {i}: {identify_address(prog, site)} <0x{site:x}> {site_status}") + ksft.print_msg(f"\tvmlinux: 0x{vmlinux_insn.address:x}:\t{vmlinux_insn.mnemonic}\t{vmlinux_insn.op_str}") + ksft.print_msg(f"\tkcore: 0x{kcore_insn.address:x}:\t{kcore_insn.mnemonic}\t{kcore_insn.op_str}") + + if (site & 0x20) ^ (insn_end & 0x20): + ksft.print_msg(f"\tSite at safe/unsafe boundary: {str(kcore_insn.bytes)} {kcore_insn.mnemonic} {operand}") + if safe_site: + tests_passed += 1 + passed = True + ksft.print_msg(f"\tPASSED: At safe address") + continue + + if operand.startswith('0xffffffff'): + thunk = int(operand, 16) + if thunk > x86_indirect_its_thunk_r15: + insn_at_thunk = list(cap.disasm(prog.read(thunk, 16), thunk))[0] + operand += ' -> ' + insn_at_thunk.mnemonic + ' ' + insn_at_thunk.op_str + ' ' + if 'jmp' in insn_at_thunk.mnemonic and thunk & 0x20: + ksft.print_msg(f"\tPASSED: Found {operand} at safe address") + passed = True + if not passed: + if kcore_insn.operands[0].type == capstone.CS_OP_IMM: + operand += ' <' + prog.symbol(int(operand, 16)) + '>' + if '__x86_indirect_its_thunk_' in operand: + ksft.print_msg(f"\tPASSED: Found {operand}") + else: + ksft.print_msg(f"\tPASSED: Found direct branch: {kcore_insn}, ITS thunk not required.") + passed = True + else: + unknown = True + if passed: + tests_passed += 1 + elif unknown: + ksft.print_msg(f"UNKNOWN: unexpected operand: {kcore_insn}") + tests_unknown += 1 + else: + ksft.print_msg(f'\t************* FAILED *************') + ksft.print_msg(f"\tFound {kcore_insn.bytes} {kcore_insn.mnemonic} {operand}") + ksft.print_msg(f'\t**********************************') + tests_failed += 1 + except Exception as e: + ksft.print_msg(f"UNKNOWN: An unexpected error occurred: {e}") + tests_unknown += 1 + +ksft.print_msg(f"\n\nSummary:") +ksft.print_msg(f"PASS: \t{tests_passed} \t/ {total_retpoline_tests}") +ksft.print_msg(f"FAIL: \t{tests_failed} \t/ {total_retpoline_tests}") +ksft.print_msg(f"UNKNOWN: \t{tests_unknown} \t/ {total_retpoline_tests}") + +if tests_failed == 0: + ksft.test_result_pass("All ITS return thunk sites passed") +else: + ksft.test_result_fail(f"{tests_failed} ITS return thunk sites failed") +ksft.finished() diff --git a/tools/testing/selftests/x86/bugs/its_permutations.py b/tools/testing/selftests/x86/bugs/its_permutations.py new file mode 100755 index 000000000000..3204f4728c62 --- /dev/null +++ b/tools/testing/selftests/x86/bugs/its_permutations.py @@ -0,0 +1,109 @@ +#!/usr/bin/env python3 +# SPDX-License-Identifier: GPL-2.0 +# +# Copyright (c) 2025 Intel Corporation +# +# Test for indirect target selection (ITS) cmdline permutations with other bugs +# like spectre_v2 and retbleed. + +import os, sys, subprocess, itertools, re, shutil + +test_dir = os.path.dirname(os.path.realpath(__file__)) +sys.path.insert(0, test_dir + '/../../kselftest') +import ksft +import common as c + +bug = "indirect_target_selection" +mitigation = c.get_sysfs(bug) + +if not mitigation or "Not affected" in mitigation: + ksft.test_result_skip("Skipping its_permutations.py: not applicable") + ksft.finished() + +if shutil.which('vng') is None: + ksft.test_result_skip("Skipping its_permutations.py: virtme-ng ('vng') not found in PATH.") + ksft.finished() + +TEST = f"{test_dir}/its_sysfs.py" +default_kparam = ['clearcpuid=hypervisor', 'panic=5', 'panic_on_warn=1', 'oops=panic', 'nmi_watchdog=1', 'hung_task_panic=1'] + +DEBUG = " -v " + +# Install dependencies +# https://github.com/arighi/virtme-ng +# apt install virtme-ng +BOOT_CMD = f"vng --run {test_dir}/../../../../../arch/x86/boot/bzImage " +#BOOT_CMD += DEBUG + +bug = "indirect_target_selection" + +input_options = { + 'indirect_target_selection' : ['off', 'on', 'stuff', 'vmexit'], + 'retbleed' : ['off', 'stuff', 'auto'], + 'spectre_v2' : ['off', 'on', 'eibrs', 'retpoline', 'ibrs', 'eibrs,retpoline'], +} + +def pretty_print(output): + OKBLUE = '\033[94m' + OKGREEN = '\033[92m' + WARNING = '\033[93m' + FAIL = '\033[91m' + ENDC = '\033[0m' + BOLD = '\033[1m' + + # Define patterns and their corresponding colors + patterns = { + r"^ok \d+": OKGREEN, + r"^not ok \d+": FAIL, + r"^# Testing .*": OKBLUE, + r"^# Found: .*": WARNING, + r"^# Totals: .*": BOLD, + r"pass:([1-9]\d*)": OKGREEN, + r"fail:([1-9]\d*)": FAIL, + r"skip:([1-9]\d*)": WARNING, + } + + # Apply colors based on patterns + for pattern, color in patterns.items(): + output = re.sub(pattern, lambda match: f"{color}{match.group(0)}{ENDC}", output, flags=re.MULTILINE) + + print(output) + +combinations = list(itertools.product(*input_options.values())) +ksft.print_header() +ksft.set_plan(len(combinations)) + +logs = "" + +for combination in combinations: + append = "" + log = "" + for p in default_kparam: + append += f' --append={p}' + command = BOOT_CMD + append + test_params = "" + for i, key in enumerate(input_options.keys()): + param = f'{key}={combination[i]}' + test_params += f' {param}' + command += f" --append={param}" + command += f" -- {TEST}" + test_name = f"{bug} {test_params}" + pretty_print(f'# Testing {test_name}') + t = subprocess.Popen(command, shell=True, stdout=subprocess.PIPE, stderr=subprocess.STDOUT) + t.wait() + output, _ = t.communicate() + if t.returncode == 0: + ksft.test_result_pass(test_name) + else: + ksft.test_result_fail(test_name) + output = output.decode() + log += f" {output}" + pretty_print(log) + logs += output + "\n" + +# Optionally use tappy to parse the output +# apt install python3-tappy +with open("logs.txt", "w") as f: + f.write(logs) + +ksft.finished() diff --git a/tools/testing/selftests/x86/bugs/its_ret_alignment.py b/tools/testing/selftests/x86/bugs/its_ret_alignment.py new file mode 100755 index 000000000000..f40078d9f6ff --- /dev/null +++ b/tools/testing/selftests/x86/bugs/its_ret_alignment.py @@ -0,0 +1,139 @@ +#!/usr/bin/env python3 +# SPDX-License-Identifier: GPL-2.0 +# +# Copyright (c) 2025 Intel Corporation +# +# Test for indirect target selection (ITS) mitigation. +# +# Tests if the RETs are correctly patched by evaluating the +# vmlinux .return_sites in /proc/kcore. +# +# Install dependencies +# add-apt-repository ppa:michel-slm/kernel-utils +# apt update +# apt install -y python3-drgn python3-pyelftools python3-capstone +# +# Run on target machine +# mkdir -p /usr/lib/debug/lib/modules/$(uname -r) +# cp $VMLINUX /usr/lib/debug/lib/modules/$(uname -r)/vmlinux +# +# Usage: ./its_ret_alignment.py + +import os, sys, argparse +from pathlib import Path + +this_dir = os.path.dirname(os.path.realpath(__file__)) +sys.path.insert(0, this_dir + '/../../kselftest') +import ksft +import common as c + +bug = "indirect_target_selection" +mitigation = c.get_sysfs(bug) +if not mitigation or "Aligned branch/return thunks" not in mitigation: + ksft.test_result_skip("Skipping its_ret_alignment.py: Aligned branch/return thunks not enabled") + ksft.finished() + +c.check_dependencies_or_skip(['drgn', 'elftools', 'capstone'], script_name="its_ret_alignment.py") + +from elftools.elf.elffile import ELFFile +from drgn.helpers.common.memory import identify_address + +cap = c.init_capstone() + +if len(os.sys.argv) > 1: + arg_vmlinux = os.sys.argv[1] + if not os.path.exists(arg_vmlinux): + ksft.test_result_fail(f"its_ret_alignment.py: vmlinux not found at user-supplied path: {arg_vmlinux}") + ksft.exit_fail() + os.makedirs(f"/usr/lib/debug/lib/modules/{os.uname().release}", exist_ok=True) + os.system(f'cp {arg_vmlinux} /usr/lib/debug/lib/modules/$(uname -r)/vmlinux') + +vmlinux = f"/usr/lib/debug/lib/modules/{os.uname().release}/vmlinux" +if not os.path.exists(vmlinux): + ksft.test_result_fail(f"its_ret_alignment.py: vmlinux not found at {vmlinux}") + ksft.exit_fail() + +ksft.print_msg(f"Using vmlinux: {vmlinux}") + +rethunks_start_vmlinux, rethunks_sec_offset, size = c.get_section_info(vmlinux, '.return_sites') +ksft.print_msg(f"vmlinux: Section .return_sites (0x{rethunks_start_vmlinux:x}) found at 0x{rethunks_sec_offset:x} with size 0x{size:x}") + +sites_offset = c.get_patch_sites(vmlinux, rethunks_sec_offset, size) +total_rethunk_tests = len(sites_offset) +ksft.print_msg(f"Found {total_rethunk_tests} rethunk sites") + +prog = c.get_runtime_kernel() +rethunks_start_kcore = prog.symbol('__return_sites').address +ksft.print_msg(f'kcore: __rethunk_sites: 0x{rethunks_start_kcore:x}') + +its_return_thunk = prog.symbol('its_return_thunk').address +ksft.print_msg(f'kcore: its_return_thunk: 0x{its_return_thunk:x}') + +tests_passed = 0 +tests_failed = 0 +tests_unknown = 0 +tests_skipped = 0 + +with open(vmlinux, 'rb') as f: + elffile = ELFFile(f) + text_section = elffile.get_section_by_name('.text') + + for i in range(len(sites_offset)): + site = rethunks_start_kcore + sites_offset[i] + vmlinux_site = rethunks_start_vmlinux + sites_offset[i] + try: + passed = unknown = failed = skipped = False + + symbol = identify_address(prog, site) + vmlinux_insn = c.get_instruction_from_vmlinux(elffile, text_section, text_section['sh_addr'], vmlinux_site) + kcore_insn = list(cap.disasm(prog.read(site, 16), site))[0] + + insn_end = site + kcore_insn.size - 1 + + safe_site = insn_end & 0x20 + site_status = "" if safe_site else "(unsafe)" + + ksft.print_msg(f"\nSite {i}: {symbol} <0x{site:x}> {site_status}") + ksft.print_msg(f"\tvmlinux: 0x{vmlinux_insn.address:x}:\t{vmlinux_insn.mnemonic}\t{vmlinux_insn.op_str}") + ksft.print_msg(f"\tkcore: 0x{kcore_insn.address:x}:\t{kcore_insn.mnemonic}\t{kcore_insn.op_str}") + + if safe_site: + tests_passed += 1 + passed = True + ksft.print_msg(f"\tPASSED: At safe address") + continue + + if "jmp" in kcore_insn.mnemonic: + passed = True + elif "ret" not in kcore_insn.mnemonic: + skipped = True + + if passed: + ksft.print_msg(f"\tPASSED: Found {kcore_insn.mnemonic} {kcore_insn.op_str}") + tests_passed += 1 + elif skipped: + ksft.print_msg(f"\tSKIPPED: Found '{kcore_insn.mnemonic}'") + tests_skipped += 1 + elif unknown: + ksft.print_msg(f"UNKNOWN: An unknown instruction: {kcore_insn}") + tests_unknown += 1 + else: + ksft.print_msg(f'\t************* FAILED *************') + ksft.print_msg(f"\tFound {kcore_insn.mnemonic} {kcore_insn.op_str}") + ksft.print_msg(f'\t**********************************') + tests_failed += 1 + except Exception as e: + ksft.print_msg(f"UNKNOWN: An unexpected error occurred: {e}") + tests_unknown += 1 + +ksft.print_msg(f"\n\nSummary:") +ksft.print_msg(f"PASSED: \t{tests_passed} \t/ {total_rethunk_tests}") +ksft.print_msg(f"FAILED: \t{tests_failed} \t/ {total_rethunk_tests}") +ksft.print_msg(f"SKIPPED: \t{tests_skipped} \t/ {total_rethunk_tests}") +ksft.print_msg(f"UNKNOWN: \t{tests_unknown} \t/ {total_rethunk_tests}") + +if tests_failed == 0: + ksft.test_result_pass("All ITS return thunk sites passed.") +else: + ksft.test_result_fail(f"{tests_failed} failed sites need ITS return thunks.") +ksft.finished() diff --git a/tools/testing/selftests/x86/bugs/its_sysfs.py b/tools/testing/selftests/x86/bugs/its_sysfs.py new file mode 100755 index 000000000000..7bca81f2f606 --- /dev/null +++ b/tools/testing/selftests/x86/bugs/its_sysfs.py @@ -0,0 +1,65 @@ +#!/usr/bin/env python3 +# SPDX-License-Identifier: GPL-2.0 +# +# Copyright (c) 2025 Intel Corporation +# +# Test for Indirect Target Selection(ITS) mitigation sysfs status. + +import sys, os, re +this_dir = os.path.dirname(os.path.realpath(__file__)) +sys.path.insert(0, this_dir + '/../../kselftest') +import ksft + +from common import * + +bug = "indirect_target_selection" +mitigation = get_sysfs(bug) + +ITS_MITIGATION_ALIGNED_THUNKS = "Mitigation: Aligned branch/return thunks" +ITS_MITIGATION_RETPOLINE_STUFF = "Mitigation: Retpolines, Stuffing RSB" +ITS_MITIGATION_VMEXIT_ONLY = "Mitigation: Vulnerable, KVM: Not affected" +ITS_MITIGATION_VULNERABLE = "Vulnerable" + +def check_mitigation(): + if mitigation == ITS_MITIGATION_ALIGNED_THUNKS: + if cmdline_has(f'{bug}=stuff') and sysfs_has("spectre_v2", "Retpolines"): + bug_check_fail(bug, ITS_MITIGATION_ALIGNED_THUNKS, ITS_MITIGATION_RETPOLINE_STUFF) + return + if cmdline_has(f'{bug}=vmexit') and cpuinfo_has('its_native_only'): + bug_check_fail(bug, ITS_MITIGATION_ALIGNED_THUNKS, ITS_MITIGATION_VMEXIT_ONLY) + return + bug_check_pass(bug, ITS_MITIGATION_ALIGNED_THUNKS) + return + + if mitigation == ITS_MITIGATION_RETPOLINE_STUFF: + if cmdline_has(f'{bug}=stuff') and sysfs_has("spectre_v2", "Retpolines"): + bug_check_pass(bug, ITS_MITIGATION_RETPOLINE_STUFF) + return + if sysfs_has('retbleed', 'Stuffing'): + bug_check_pass(bug, ITS_MITIGATION_RETPOLINE_STUFF) + return + bug_check_fail(bug, ITS_MITIGATION_RETPOLINE_STUFF, ITS_MITIGATION_ALIGNED_THUNKS) + + if mitigation == ITS_MITIGATION_VMEXIT_ONLY: + if cmdline_has(f'{bug}=vmexit') and cpuinfo_has('its_native_only'): + bug_check_pass(bug, ITS_MITIGATION_VMEXIT_ONLY) + return + bug_check_fail(bug, ITS_MITIGATION_VMEXIT_ONLY, ITS_MITIGATION_ALIGNED_THUNKS) + + if mitigation == ITS_MITIGATION_VULNERABLE: + if sysfs_has("spectre_v2", "Vulnerable"): + bug_check_pass(bug, ITS_MITIGATION_VULNERABLE) + else: + bug_check_fail(bug, "Mitigation", ITS_MITIGATION_VULNERABLE) + + bug_status_unknown(bug, mitigation) + return + +ksft.print_header() +ksft.set_plan(1) +ksft.print_msg(f'{bug}: {mitigation} ...') + +if not basic_checks_sufficient(bug, mitigation): + check_mitigation() + +ksft.finished() From 250cf3693060a5f803c5f1ddc082bb06b16112a9 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 27 Apr 2025 15:41:51 -0400 Subject: [PATCH 363/555] __legitimize_mnt(): check for MNT_SYNC_UMOUNT should be under mount_lock ... or we risk stealing final mntput from sync umount - raising mnt_count after umount(2) has verified that victim is not busy, but before it has set MNT_SYNC_UMOUNT; in that case __legitimize_mnt() doesn't see that it's safe to quietly undo mnt_count increment and leaves dropping the reference to caller, where it'll be a full-blown mntput(). Check under mount_lock is needed; leaving the current one done before taking that makes no sense - it's nowhere near common enough to bother with. Reviewed-by: Christian Brauner Signed-off-by: Al Viro --- fs/namespace.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/fs/namespace.c b/fs/namespace.c index 98a5cd756e9a..eba4748388b1 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -790,12 +790,8 @@ int __legitimize_mnt(struct vfsmount *bastard, unsigned seq) smp_mb(); // see mntput_no_expire() if (likely(!read_seqretry(&mount_lock, seq))) return 0; - if (bastard->mnt_flags & MNT_SYNC_UMOUNT) { - mnt_add_count(mnt, -1); - return 1; - } lock_mount_hash(); - if (unlikely(bastard->mnt_flags & MNT_DOOMED)) { + if (unlikely(bastard->mnt_flags & (MNT_SYNC_UMOUNT | MNT_DOOMED))) { mnt_add_count(mnt, -1); unlock_mount_hash(); return 1; From 45375814eb3f4245956c0c85092a4eee4441d167 Mon Sep 17 00:00:00 2001 From: Hangbin Liu Date: Thu, 8 May 2025 03:54:14 +0000 Subject: [PATCH 364/555] tools/net/ynl: ethtool: fix crash when Hardware Clock info is missing Fix a crash in the ethtool YNL implementation when Hardware Clock information is not present in the response. This ensures graceful handling of devices or drivers that do not provide this optional field. e.g. Traceback (most recent call last): File "/net/tools/net/ynl/pyynl/./ethtool.py", line 438, in main() ~~~~^^ File "/net/tools/net/ynl/pyynl/./ethtool.py", line 341, in main print(f'PTP Hardware Clock: {tsinfo["phc-index"]}') ~~~~~~^^^^^^^^^^^^^ KeyError: 'phc-index' Fixes: f3d07b02b2b8 ("tools: ynl: ethtool testing tool") Signed-off-by: Hangbin Liu Acked-by: Stanislav Fomichev Link: https://patch.msgid.link/20250508035414.82974-1-liuhangbin@gmail.com Signed-off-by: Jakub Kicinski --- tools/net/ynl/pyynl/ethtool.py | 22 +++++++++++++++------- 1 file changed, 15 insertions(+), 7 deletions(-) diff --git a/tools/net/ynl/pyynl/ethtool.py b/tools/net/ynl/pyynl/ethtool.py index af7fddd7b085..cab6b576c876 100755 --- a/tools/net/ynl/pyynl/ethtool.py +++ b/tools/net/ynl/pyynl/ethtool.py @@ -338,16 +338,24 @@ def main(): print('Capabilities:') [print(f'\t{v}') for v in bits_to_dict(tsinfo['timestamping'])] - print(f'PTP Hardware Clock: {tsinfo["phc-index"]}') + print(f'PTP Hardware Clock: {tsinfo.get("phc-index", "none")}') - print('Hardware Transmit Timestamp Modes:') - [print(f'\t{v}') for v in bits_to_dict(tsinfo['tx-types'])] + if 'tx-types' in tsinfo: + print('Hardware Transmit Timestamp Modes:') + [print(f'\t{v}') for v in bits_to_dict(tsinfo['tx-types'])] + else: + print('Hardware Transmit Timestamp Modes: none') - print('Hardware Receive Filter Modes:') - [print(f'\t{v}') for v in bits_to_dict(tsinfo['rx-filters'])] + if 'rx-filters' in tsinfo: + print('Hardware Receive Filter Modes:') + [print(f'\t{v}') for v in bits_to_dict(tsinfo['rx-filters'])] + else: + print('Hardware Receive Filter Modes: none') + + if 'stats' in tsinfo and tsinfo['stats']: + print('Statistics:') + [print(f'\t{k}: {v}') for k, v in tsinfo['stats'].items()] - print('Statistics:') - [print(f'\t{k}: {v}') for k, v in tsinfo['stats'].items()] return print(f'Settings for {args.device}:') From f11cf946c0a92c560a890d68e4775723353599e1 Mon Sep 17 00:00:00 2001 From: Matt Johnston Date: Thu, 8 May 2025 13:18:32 +0800 Subject: [PATCH 365/555] net: mctp: Don't access ifa_index when missing In mctp_dump_addrinfo, ifa_index can be used to filter interfaces, but only when the struct ifaddrmsg is provided. Otherwise it will be comparing to uninitialised memory - reproducible in the syzkaller case from dhcpd, or busybox "ip addr show". The kernel MCTP implementation has always filtered by ifa_index, so existing userspace programs expecting to dump MCTP addresses must already be passing a valid ifa_index value (either 0 or a real index). BUG: KMSAN: uninit-value in mctp_dump_addrinfo+0x208/0xac0 net/mctp/device.c:128 mctp_dump_addrinfo+0x208/0xac0 net/mctp/device.c:128 rtnl_dump_all+0x3ec/0x5b0 net/core/rtnetlink.c:4380 rtnl_dumpit+0xd5/0x2f0 net/core/rtnetlink.c:6824 netlink_dump+0x97b/0x1690 net/netlink/af_netlink.c:2309 Fixes: 583be982d934 ("mctp: Add device handling and netlink interface") Reported-by: syzbot+e76d52dadc089b9d197f@syzkaller.appspotmail.com Closes: https://lore.kernel.org/all/68135815.050a0220.3a872c.000e.GAE@google.com/ Reported-by: syzbot+1065a199625a388fce60@syzkaller.appspotmail.com Closes: https://lore.kernel.org/all/681357d6.050a0220.14dd7d.000d.GAE@google.com/ Signed-off-by: Matt Johnston Link: https://patch.msgid.link/20250508-mctp-addr-dump-v2-1-c8a53fd2dd66@codeconstruct.com.au Signed-off-by: Jakub Kicinski --- net/mctp/device.c | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) diff --git a/net/mctp/device.c b/net/mctp/device.c index 8e0724c56723..7c0dcf3df319 100644 --- a/net/mctp/device.c +++ b/net/mctp/device.c @@ -117,11 +117,18 @@ static int mctp_dump_addrinfo(struct sk_buff *skb, struct netlink_callback *cb) struct net_device *dev; struct ifaddrmsg *hdr; struct mctp_dev *mdev; - int ifindex, rc; + int ifindex = 0, rc; - hdr = nlmsg_data(cb->nlh); - // filter by ifindex if requested - ifindex = hdr->ifa_index; + /* Filter by ifindex if a header is provided */ + if (cb->nlh->nlmsg_len >= nlmsg_msg_size(sizeof(*hdr))) { + hdr = nlmsg_data(cb->nlh); + ifindex = hdr->ifa_index; + } else { + if (cb->strict_check) { + NL_SET_ERR_MSG(cb->extack, "mctp: Invalid header for addr dump request"); + return -EINVAL; + } + } rcu_read_lock(); for_each_netdev_dump(net, dev, mcb->ifindex) { From 97c4e094a4b2edbb4fffeda718f8e806f825a18f Mon Sep 17 00:00:00 2001 From: Cosmin Ratiu Date: Thu, 8 May 2025 11:44:34 +0300 Subject: [PATCH 366/555] tests/ncdevmem: Fix double-free of queue array netdev_bind_rx takes ownership of the queue array passed as parameter and frees it, so a queue array buffer cannot be reused across multiple netdev_bind_rx calls. This commit fixes that by always passing in a newly created queue array to all netdev_bind_rx calls in ncdevmem. Fixes: 85585b4bc8d8 ("selftests: add ncdevmem, netcat for devmem TCP") Signed-off-by: Cosmin Ratiu Acked-by: Stanislav Fomichev Reviewed-by: Joe Damato Reviewed-by: Mina Almasry Link: https://patch.msgid.link/20250508084434.1933069-1-cratiu@nvidia.com Signed-off-by: Jakub Kicinski --- .../selftests/drivers/net/hw/ncdevmem.c | 55 ++++++++----------- 1 file changed, 22 insertions(+), 33 deletions(-) diff --git a/tools/testing/selftests/drivers/net/hw/ncdevmem.c b/tools/testing/selftests/drivers/net/hw/ncdevmem.c index 2bf14ac2b8c6..9d48004ff1a1 100644 --- a/tools/testing/selftests/drivers/net/hw/ncdevmem.c +++ b/tools/testing/selftests/drivers/net/hw/ncdevmem.c @@ -431,6 +431,22 @@ static int parse_address(const char *str, int port, struct sockaddr_in6 *sin6) return 0; } +static struct netdev_queue_id *create_queues(void) +{ + struct netdev_queue_id *queues; + size_t i = 0; + + queues = calloc(num_queues, sizeof(*queues)); + for (i = 0; i < num_queues; i++) { + queues[i]._present.type = 1; + queues[i]._present.id = 1; + queues[i].type = NETDEV_QUEUE_TYPE_RX; + queues[i].id = start_queue + i; + } + + return queues; +} + int do_server(struct memory_buffer *mem) { char ctrl_data[sizeof(int) * 20000]; @@ -448,7 +464,6 @@ int do_server(struct memory_buffer *mem) char buffer[256]; int socket_fd; int client_fd; - size_t i = 0; int ret; ret = parse_address(server_ip, atoi(port), &server_sin); @@ -471,16 +486,7 @@ int do_server(struct memory_buffer *mem) sleep(1); - queues = malloc(sizeof(*queues) * num_queues); - - for (i = 0; i < num_queues; i++) { - queues[i]._present.type = 1; - queues[i]._present.id = 1; - queues[i].type = NETDEV_QUEUE_TYPE_RX; - queues[i].id = start_queue + i; - } - - if (bind_rx_queue(ifindex, mem->fd, queues, num_queues, &ys)) + if (bind_rx_queue(ifindex, mem->fd, create_queues(), num_queues, &ys)) error(1, 0, "Failed to bind\n"); tmp_mem = malloc(mem->size); @@ -545,7 +551,6 @@ int do_server(struct memory_buffer *mem) goto cleanup; } - i++; for (cm = CMSG_FIRSTHDR(&msg); cm; cm = CMSG_NXTHDR(&msg, cm)) { if (cm->cmsg_level != SOL_SOCKET || (cm->cmsg_type != SCM_DEVMEM_DMABUF && @@ -630,10 +635,8 @@ cleanup: void run_devmem_tests(void) { - struct netdev_queue_id *queues; struct memory_buffer *mem; struct ynl_sock *ys; - size_t i = 0; mem = provider->alloc(getpagesize() * NUM_PAGES); @@ -641,38 +644,24 @@ void run_devmem_tests(void) if (configure_rss()) error(1, 0, "rss error\n"); - queues = calloc(num_queues, sizeof(*queues)); - if (configure_headersplit(1)) error(1, 0, "Failed to configure header split\n"); - if (!bind_rx_queue(ifindex, mem->fd, queues, num_queues, &ys)) + if (!bind_rx_queue(ifindex, mem->fd, + calloc(num_queues, sizeof(struct netdev_queue_id)), + num_queues, &ys)) error(1, 0, "Binding empty queues array should have failed\n"); - for (i = 0; i < num_queues; i++) { - queues[i]._present.type = 1; - queues[i]._present.id = 1; - queues[i].type = NETDEV_QUEUE_TYPE_RX; - queues[i].id = start_queue + i; - } - if (configure_headersplit(0)) error(1, 0, "Failed to configure header split\n"); - if (!bind_rx_queue(ifindex, mem->fd, queues, num_queues, &ys)) + if (!bind_rx_queue(ifindex, mem->fd, create_queues(), num_queues, &ys)) error(1, 0, "Configure dmabuf with header split off should have failed\n"); if (configure_headersplit(1)) error(1, 0, "Failed to configure header split\n"); - for (i = 0; i < num_queues; i++) { - queues[i]._present.type = 1; - queues[i]._present.id = 1; - queues[i].type = NETDEV_QUEUE_TYPE_RX; - queues[i].id = start_queue + i; - } - - if (bind_rx_queue(ifindex, mem->fd, queues, num_queues, &ys)) + if (bind_rx_queue(ifindex, mem->fd, create_queues(), num_queues, &ys)) error(1, 0, "Failed to bind\n"); /* Deactivating a bound queue should not be legal */ From 65781e19dcfcb4aed1167d87a3ffcc2a0c071d47 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 28 Apr 2025 23:56:14 -0400 Subject: [PATCH 367/555] do_umount(): add missing barrier before refcount checks in sync case do_umount() analogue of the race fixed in 119e1ef80ecf "fix __legitimize_mnt()/mntput() race". Here we want to make sure that if __legitimize_mnt() doesn't notice our lock_mount_hash(), we will notice their refcount increment. Harder to hit than mntput_no_expire() one, fortunately, and consequences are milder (sync umount acting like umount -l on a rare race with RCU pathwalk hitting at just the wrong time instead of use-after-free galore mntput_no_expire() counterpart used to be hit). Still a bug... Fixes: 48a066e72d97 ("RCU'd vfsmounts") Reviewed-by: Christian Brauner Signed-off-by: Al Viro --- fs/namespace.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/namespace.c b/fs/namespace.c index eba4748388b1..d8a344d0a80a 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -787,7 +787,7 @@ int __legitimize_mnt(struct vfsmount *bastard, unsigned seq) return 0; mnt = real_mount(bastard); mnt_add_count(mnt, 1); - smp_mb(); // see mntput_no_expire() + smp_mb(); // see mntput_no_expire() and do_umount() if (likely(!read_seqretry(&mount_lock, seq))) return 0; lock_mount_hash(); @@ -2044,6 +2044,7 @@ static int do_umount(struct mount *mnt, int flags) umount_tree(mnt, UMOUNT_PROPAGATE); retval = 0; } else { + smp_mb(); // paired with __legitimize_mnt() shrink_submounts(mnt); retval = -EBUSY; if (!propagate_mount_busy(mnt, 2)) { From 267fc3a06a37bec30cc5b4d97fb8409102bc7a9d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 28 Apr 2025 21:43:23 -0400 Subject: [PATCH 368/555] do_move_mount(): don't leak MNTNS_PROPAGATING on failures as it is, a failed move_mount(2) from anon namespace breaks all further propagation into that namespace, including normal mounts in non-anon namespaces that would otherwise propagate there. Fixes: 064fe6e233e8 ("mount: handle mount propagation for detached mount trees") Reviewed-by: Christian Brauner Signed-off-by: Al Viro --- fs/namespace.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/fs/namespace.c b/fs/namespace.c index d8a344d0a80a..04a9bb9f31fa 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -3715,15 +3715,14 @@ static int do_move_mount(struct path *old_path, if (err) goto out; - if (is_anon_ns(ns)) - ns->mntns_flags &= ~MNTNS_PROPAGATING; - /* if the mount is moved, it should no longer be expire * automatically */ list_del_init(&old->mnt_expire); if (attached) put_mountpoint(old_mp); out: + if (is_anon_ns(ns)) + ns->mntns_flags &= ~MNTNS_PROPAGATING; unlock_mount(mp); if (!err) { if (attached) { From d1ddc6f1d9f0cf887834eb54a5a68bbfeec1bb77 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 8 May 2025 15:35:51 -0400 Subject: [PATCH 369/555] fix IS_MNT_PROPAGATING uses propagate_mnt() does not attach anything to mounts created during propagate_mnt() itself. What's more, anything on ->mnt_slave_list of such new mount must also be new, so we don't need to even look there. When move_mount() had been introduced, we've got an additional class of mounts to skip - if we are moving from anon namespace, we do not want to propagate to mounts we are moving (i.e. all mounts in that anon namespace). Unfortunately, the part about "everything on their ->mnt_slave_list will also be ignorable" is not true - if we have propagation graph A -> B -> C and do OPEN_TREE_CLONE open_tree() of B, we get A -> [B <-> B'] -> C as propagation graph, where B' is a clone of B in our detached tree. Making B private will result in A -> B' -> C C still gets propagation from A, as it would after making B private if we hadn't done that open_tree(), but now the propagation goes through B'. Trying to move_mount() our detached tree on subdirectory in A should have * moved B' on that subdirectory in A * skipped the corresponding subdirectory in B' itself * copied B' on the corresponding subdirectory in C. As it is, the logics in propagation_next() and friends ends up skipping propagation into C, since it doesn't consider anything downstream of B'. IOW, walking the propagation graph should only skip the ->mnt_slave_list of new mounts; the only places where the check for "in that one anon namespace" are applicable are propagate_one() (where we should treat that as the same kind of thing as "mountpoint we are looking at is not visible in the mount we are looking at") and propagation_would_overmount(). The latter is better dealt with in the caller (can_move_mount_beneath()); on the first call of propagation_would_overmount() the test is always false, on the second it is always true in "move from anon namespace" case and always false in "move within our namespace" one, so it's easier to just use check_mnt() before bothering with the second call and be done with that. Fixes: 064fe6e233e8 ("mount: handle mount propagation for detached mount trees") Reviewed-by: Christian Brauner Signed-off-by: Al Viro --- fs/namespace.c | 3 ++- fs/pnode.c | 17 +++++++++-------- fs/pnode.h | 2 +- 3 files changed, 12 insertions(+), 10 deletions(-) diff --git a/fs/namespace.c b/fs/namespace.c index 04a9bb9f31fa..1b466c54a357 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -3557,7 +3557,8 @@ static int can_move_mount_beneath(const struct path *from, * @mnt_from itself. This defeats the whole purpose of mounting * @mnt_from beneath @mnt_to. */ - if (propagation_would_overmount(parent_mnt_to, mnt_from, mp)) + if (check_mnt(mnt_from) && + propagation_would_overmount(parent_mnt_to, mnt_from, mp)) return -EINVAL; return 0; diff --git a/fs/pnode.c b/fs/pnode.c index 7a062a5de10e..fb77427df39e 100644 --- a/fs/pnode.c +++ b/fs/pnode.c @@ -150,7 +150,7 @@ static struct mount *propagation_next(struct mount *m, struct mount *origin) { /* are there any slaves of this mount? */ - if (!IS_MNT_PROPAGATED(m) && !list_empty(&m->mnt_slave_list)) + if (!IS_MNT_NEW(m) && !list_empty(&m->mnt_slave_list)) return first_slave(m); while (1) { @@ -174,7 +174,7 @@ static struct mount *skip_propagation_subtree(struct mount *m, * Advance m such that propagation_next will not return * the slaves of m. */ - if (!IS_MNT_PROPAGATED(m) && !list_empty(&m->mnt_slave_list)) + if (!IS_MNT_NEW(m) && !list_empty(&m->mnt_slave_list)) m = last_slave(m); return m; @@ -185,7 +185,7 @@ static struct mount *next_group(struct mount *m, struct mount *origin) while (1) { while (1) { struct mount *next; - if (!IS_MNT_PROPAGATED(m) && !list_empty(&m->mnt_slave_list)) + if (!IS_MNT_NEW(m) && !list_empty(&m->mnt_slave_list)) return first_slave(m); next = next_peer(m); if (m->mnt_group_id == origin->mnt_group_id) { @@ -226,11 +226,15 @@ static int propagate_one(struct mount *m, struct mountpoint *dest_mp) struct mount *child; int type; /* skip ones added by this propagate_mnt() */ - if (IS_MNT_PROPAGATED(m)) + if (IS_MNT_NEW(m)) return 0; - /* skip if mountpoint isn't covered by it */ + /* skip if mountpoint isn't visible in m */ if (!is_subdir(dest_mp->m_dentry, m->mnt.mnt_root)) return 0; + /* skip if m is in the anon_ns we are emptying */ + if (m->mnt_ns->mntns_flags & MNTNS_PROPAGATING) + return 0; + if (peers(m, last_dest)) { type = CL_MAKE_SHARED; } else { @@ -380,9 +384,6 @@ bool propagation_would_overmount(const struct mount *from, if (!IS_MNT_SHARED(from)) return false; - if (IS_MNT_PROPAGATED(to)) - return false; - if (to->mnt.mnt_root != mp->m_dentry) return false; diff --git a/fs/pnode.h b/fs/pnode.h index ddafe0d087ca..34b6247af01d 100644 --- a/fs/pnode.h +++ b/fs/pnode.h @@ -12,7 +12,7 @@ #define IS_MNT_SHARED(m) ((m)->mnt.mnt_flags & MNT_SHARED) #define IS_MNT_SLAVE(m) ((m)->mnt_master) -#define IS_MNT_PROPAGATED(m) (!(m)->mnt_ns || ((m)->mnt_ns->mntns_flags & MNTNS_PROPAGATING)) +#define IS_MNT_NEW(m) (!(m)->mnt_ns) #define CLEAR_MNT_SHARED(m) ((m)->mnt.mnt_flags &= ~MNT_SHARED) #define IS_MNT_UNBINDABLE(m) ((m)->mnt.mnt_flags & MNT_UNBINDABLE) #define IS_MNT_MARKED(m) ((m)->mnt.mnt_flags & MNT_MARKED) From e4f349bd6e58051df698b82f94721f18a02a293d Mon Sep 17 00:00:00 2001 From: Andrew Jeffery Date: Thu, 8 May 2025 14:16:00 +0930 Subject: [PATCH 370/555] net: mctp: Ensure keys maintain only one ref to corresponding dev mctp_flow_prepare_output() is called in mctp_route_output(), which places outbound packets onto a given interface. The packet may represent a message fragment, in which case we provoke an unbalanced reference count to the underlying device. This causes trouble if we ever attempt to remove the interface: [ 48.702195] usb 1-1: USB disconnect, device number 2 [ 58.883056] unregister_netdevice: waiting for mctpusb0 to become free. Usage count = 2 [ 69.022548] unregister_netdevice: waiting for mctpusb0 to become free. Usage count = 2 [ 79.172568] unregister_netdevice: waiting for mctpusb0 to become free. Usage count = 2 ... Predicate the invocation of mctp_dev_set_key() in mctp_flow_prepare_output() on not already having associated the device with the key. It's not yet realistic to uphold the property that the key maintains only one device reference earlier in the transmission sequence as the route (and therefore the device) may not be known at the time the key is associated with the socket. Fixes: 67737c457281 ("mctp: Pass flow data & flow release events to drivers") Acked-by: Jeremy Kerr Signed-off-by: Andrew Jeffery Link: https://patch.msgid.link/20250508-mctp-dev-refcount-v1-1-d4f965c67bb5@codeconstruct.com.au Signed-off-by: Jakub Kicinski --- net/mctp/route.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/net/mctp/route.c b/net/mctp/route.c index 4c460160914f..d9c8e5a5f9ce 100644 --- a/net/mctp/route.c +++ b/net/mctp/route.c @@ -313,8 +313,10 @@ static void mctp_flow_prepare_output(struct sk_buff *skb, struct mctp_dev *dev) key = flow->key; - if (WARN_ON(key->dev && key->dev != dev)) + if (key->dev) { + WARN_ON(key->dev != dev); return; + } mctp_dev_set_key(dev, key); } From 9dda18a32b4a6693fccd3f7c0738af646147b1cf Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Thu, 10 Apr 2025 05:22:21 -0700 Subject: [PATCH 371/555] tracing: fprobe: Fix RCU warning message in list traversal When CONFIG_PROVE_RCU_LIST is enabled, fprobe triggers the following warning: WARNING: suspicious RCU usage kernel/trace/fprobe.c:457 RCU-list traversed in non-reader section!! other info that might help us debug this: #1: ffffffff863c4e08 (fprobe_mutex){+.+.}-{4:4}, at: fprobe_module_callback+0x7b/0x8c0 Call Trace: fprobe_module_callback notifier_call_chain blocking_notifier_call_chain This warning occurs because fprobe_remove_node_in_module() traverses an RCU list using RCU primitives without holding an RCU read lock. However, the function is only called from fprobe_module_callback(), which holds the fprobe_mutex lock that provides sufficient protection for safely traversing the list. Fix the warning by specifying the locking design to the CONFIG_PROVE_RCU_LIST mechanism. Add the lockdep_is_held() argument to hlist_for_each_entry_rcu() to inform the RCU checker that fprobe_mutex provides the required protection. Link: https://lore.kernel.org/all/20250410-fprobe-v1-1-068ef5f41436@debian.org/ Fixes: a3dc2983ca7b90 ("tracing: fprobe: Cleanup fprobe hash when module unloading") Signed-off-by: Breno Leitao Tested-by: Antonio Quartulli Tested-by: Matthieu Baerts (NGI0) Signed-off-by: Masami Hiramatsu (Google) --- kernel/trace/fprobe.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/kernel/trace/fprobe.c b/kernel/trace/fprobe.c index 95c6e3473a76..ba7ff14f5339 100644 --- a/kernel/trace/fprobe.c +++ b/kernel/trace/fprobe.c @@ -454,7 +454,8 @@ static void fprobe_remove_node_in_module(struct module *mod, struct hlist_head * struct fprobe_hlist_node *node; int ret = 0; - hlist_for_each_entry_rcu(node, head, hlist) { + hlist_for_each_entry_rcu(node, head, hlist, + lockdep_is_held(&fprobe_mutex)) { if (!within_module(node->addr, mod)) continue; if (delete_fprobe_node(node)) From e41b5af4519f90f9a751805ede2102ae36caf5d0 Mon Sep 17 00:00:00 2001 From: Paul Cacheux Date: Sun, 4 May 2025 20:27:52 +0200 Subject: [PATCH 372/555] tracing: add missing trace_probe_log_clear for eprobes Make sure trace_probe_log_clear is called in the tracing eprobe code path, matching the trace_probe_log_init call. Link: https://lore.kernel.org/all/20250504-fix-trace-probe-log-race-v3-1-9e99fec7eddc@gmail.com/ Signed-off-by: Paul Cacheux Acked-by: Steven Rostedt (Google) Signed-off-by: Masami Hiramatsu (Google) --- kernel/trace/trace_eprobe.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/kernel/trace/trace_eprobe.c b/kernel/trace/trace_eprobe.c index c08355c3ef32..916555f0de81 100644 --- a/kernel/trace/trace_eprobe.c +++ b/kernel/trace/trace_eprobe.c @@ -969,10 +969,13 @@ static int __trace_eprobe_create(int argc, const char *argv[]) goto error; } } + trace_probe_log_clear(); return ret; + parse_error: ret = -EINVAL; error: + trace_probe_log_clear(); trace_event_probe_cleanup(ep); return ret; } From e98960bc4df931a2f4fa88f7e55370699ca4dd82 Mon Sep 17 00:00:00 2001 From: Ulf Hansson Date: Thu, 6 Mar 2025 12:50:21 +0100 Subject: [PATCH 373/555] Input: hisi_powerkey - enable system-wakeup for s2idle To wake up the system from s2idle when pressing the power-button, let's convert from using pm_wakeup_event() to pm_wakeup_dev_event(), as it allows us to specify the "hard" in-parameter, which needs to be set for s2idle. Signed-off-by: Ulf Hansson Link: https://lore.kernel.org/r/20250306115021.797426-1-ulf.hansson@linaro.org Signed-off-by: Dmitry Torokhov --- drivers/input/misc/hisi_powerkey.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/input/misc/hisi_powerkey.c b/drivers/input/misc/hisi_powerkey.c index d3c293a95d32..d315017324d9 100644 --- a/drivers/input/misc/hisi_powerkey.c +++ b/drivers/input/misc/hisi_powerkey.c @@ -30,7 +30,7 @@ static irqreturn_t hi65xx_power_press_isr(int irq, void *q) { struct input_dev *input = q; - pm_wakeup_event(input->dev.parent, MAX_HELD_TIME); + pm_wakeup_dev_event(input->dev.parent, MAX_HELD_TIME, true); input_report_key(input, KEY_POWER, 1); input_sync(input); From d05a424bea9aa3435009d5c462055008cc1545d8 Mon Sep 17 00:00:00 2001 From: Vicki Pfau Date: Fri, 28 Mar 2025 16:43:36 -0700 Subject: [PATCH 374/555] Input: xpad - fix two controller table values Two controllers -- Mad Catz JOYTECH NEO SE Advanced and PDP Mirror's Edge Official -- were missing the value of the mapping field, and thus wouldn't detect properly. Signed-off-by: Vicki Pfau Link: https://lore.kernel.org/r/20250328234345.989761-1-vi@endrift.com Fixes: 540602a43ae5 ("Input: xpad - add a few new VID/PID combinations") Fixes: 3492321e2e60 ("Input: xpad - add multiple supported devices") Cc: stable@vger.kernel.org Signed-off-by: Dmitry Torokhov --- drivers/input/joystick/xpad.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/input/joystick/xpad.c b/drivers/input/joystick/xpad.c index c933e47173bd..2c935cf79ddc 100644 --- a/drivers/input/joystick/xpad.c +++ b/drivers/input/joystick/xpad.c @@ -205,7 +205,7 @@ static const struct xpad_device { { 0x0738, 0x9871, "Mad Catz Portable Drum", 0, XTYPE_XBOX360 }, { 0x0738, 0xb726, "Mad Catz Xbox controller - MW2", 0, XTYPE_XBOX360 }, { 0x0738, 0xb738, "Mad Catz MVC2TE Stick 2", MAP_TRIGGERS_TO_BUTTONS, XTYPE_XBOX360 }, - { 0x0738, 0xbeef, "Mad Catz JOYTECH NEO SE Advanced GamePad", XTYPE_XBOX360 }, + { 0x0738, 0xbeef, "Mad Catz JOYTECH NEO SE Advanced GamePad", 0, XTYPE_XBOX360 }, { 0x0738, 0xcb02, "Saitek Cyborg Rumble Pad - PC/Xbox 360", 0, XTYPE_XBOX360 }, { 0x0738, 0xcb03, "Saitek P3200 Rumble Pad - PC/Xbox 360", 0, XTYPE_XBOX360 }, { 0x0738, 0xcb29, "Saitek Aviator Stick AV8R02", 0, XTYPE_XBOX360 }, @@ -240,7 +240,7 @@ static const struct xpad_device { { 0x0e6f, 0x0146, "Rock Candy Wired Controller for Xbox One", 0, XTYPE_XBOXONE }, { 0x0e6f, 0x0147, "PDP Marvel Xbox One Controller", 0, XTYPE_XBOXONE }, { 0x0e6f, 0x015c, "PDP Xbox One Arcade Stick", MAP_TRIGGERS_TO_BUTTONS, XTYPE_XBOXONE }, - { 0x0e6f, 0x015d, "PDP Mirror's Edge Official Wired Controller for Xbox One", XTYPE_XBOXONE }, + { 0x0e6f, 0x015d, "PDP Mirror's Edge Official Wired Controller for Xbox One", 0, XTYPE_XBOXONE }, { 0x0e6f, 0x0161, "PDP Xbox One Controller", 0, XTYPE_XBOXONE }, { 0x0e6f, 0x0162, "PDP Xbox One Controller", 0, XTYPE_XBOXONE }, { 0x0e6f, 0x0163, "PDP Xbox One Controller", 0, XTYPE_XBOXONE }, From 4ef46367073b107ec22f46fe5f12176e87c238e8 Mon Sep 17 00:00:00 2001 From: Vicki Pfau Date: Sat, 10 May 2025 22:59:25 -0700 Subject: [PATCH 375/555] Input: xpad - fix Share button on Xbox One controllers The Share button, if present, is always one of two offsets from the end of the file, depending on the presence of a specific interface. As we lack parsing for the identify packet we can't automatically determine the presence of that interface, but we can hardcode which of these offsets is correct for a given controller. More controllers are probably fixable by adding the MAP_SHARE_BUTTON in the future, but for now I only added the ones that I have the ability to test directly. Signed-off-by: Vicki Pfau Link: https://lore.kernel.org/r/20250328234345.989761-2-vi@endrift.com Cc: stable@vger.kernel.org Signed-off-by: Dmitry Torokhov --- drivers/input/joystick/xpad.c | 35 ++++++++++++++++++++--------------- 1 file changed, 20 insertions(+), 15 deletions(-) diff --git a/drivers/input/joystick/xpad.c b/drivers/input/joystick/xpad.c index 2c935cf79ddc..8ee7d8e5d1c7 100644 --- a/drivers/input/joystick/xpad.c +++ b/drivers/input/joystick/xpad.c @@ -77,12 +77,13 @@ * xbox d-pads should map to buttons, as is required for DDR pads * but we map them to axes when possible to simplify things */ -#define MAP_DPAD_TO_BUTTONS (1 << 0) -#define MAP_TRIGGERS_TO_BUTTONS (1 << 1) -#define MAP_STICKS_TO_NULL (1 << 2) -#define MAP_SELECT_BUTTON (1 << 3) -#define MAP_PADDLES (1 << 4) -#define MAP_PROFILE_BUTTON (1 << 5) +#define MAP_DPAD_TO_BUTTONS BIT(0) +#define MAP_TRIGGERS_TO_BUTTONS BIT(1) +#define MAP_STICKS_TO_NULL BIT(2) +#define MAP_SHARE_BUTTON BIT(3) +#define MAP_PADDLES BIT(4) +#define MAP_PROFILE_BUTTON BIT(5) +#define MAP_SHARE_OFFSET BIT(6) #define DANCEPAD_MAP_CONFIG (MAP_DPAD_TO_BUTTONS | \ MAP_TRIGGERS_TO_BUTTONS | MAP_STICKS_TO_NULL) @@ -135,7 +136,7 @@ static const struct xpad_device { { 0x03f0, 0x048D, "HyperX Clutch", 0, XTYPE_XBOX360 }, /* wireless */ { 0x03f0, 0x0495, "HyperX Clutch Gladiate", 0, XTYPE_XBOXONE }, { 0x03f0, 0x07A0, "HyperX Clutch Gladiate RGB", 0, XTYPE_XBOXONE }, - { 0x03f0, 0x08B6, "HyperX Clutch Gladiate", 0, XTYPE_XBOXONE }, /* v2 */ + { 0x03f0, 0x08B6, "HyperX Clutch Gladiate", MAP_SHARE_BUTTON, XTYPE_XBOXONE }, /* v2 */ { 0x03f0, 0x09B4, "HyperX Clutch Tanto", 0, XTYPE_XBOXONE }, { 0x044f, 0x0f00, "Thrustmaster Wheel", 0, XTYPE_XBOX }, { 0x044f, 0x0f03, "Thrustmaster Wheel", 0, XTYPE_XBOX }, @@ -159,7 +160,7 @@ static const struct xpad_device { { 0x045e, 0x0719, "Xbox 360 Wireless Receiver", MAP_DPAD_TO_BUTTONS, XTYPE_XBOX360W }, { 0x045e, 0x0b00, "Microsoft X-Box One Elite 2 pad", MAP_PADDLES, XTYPE_XBOXONE }, { 0x045e, 0x0b0a, "Microsoft X-Box Adaptive Controller", MAP_PROFILE_BUTTON, XTYPE_XBOXONE }, - { 0x045e, 0x0b12, "Microsoft Xbox Series S|X Controller", MAP_SELECT_BUTTON, XTYPE_XBOXONE }, + { 0x045e, 0x0b12, "Microsoft Xbox Series S|X Controller", MAP_SHARE_BUTTON | MAP_SHARE_OFFSET, XTYPE_XBOXONE }, { 0x046d, 0xc21d, "Logitech Gamepad F310", 0, XTYPE_XBOX360 }, { 0x046d, 0xc21e, "Logitech Gamepad F510", 0, XTYPE_XBOX360 }, { 0x046d, 0xc21f, "Logitech Gamepad F710", 0, XTYPE_XBOX360 }, @@ -211,7 +212,7 @@ static const struct xpad_device { { 0x0738, 0xcb29, "Saitek Aviator Stick AV8R02", 0, XTYPE_XBOX360 }, { 0x0738, 0xf738, "Super SFIV FightStick TE S", 0, XTYPE_XBOX360 }, { 0x07ff, 0xffff, "Mad Catz GamePad", 0, XTYPE_XBOX360 }, - { 0x0b05, 0x1a38, "ASUS ROG RAIKIRI", 0, XTYPE_XBOXONE }, + { 0x0b05, 0x1a38, "ASUS ROG RAIKIRI", MAP_SHARE_BUTTON, XTYPE_XBOXONE }, { 0x0b05, 0x1abb, "ASUS ROG RAIKIRI PRO", 0, XTYPE_XBOXONE }, { 0x0c12, 0x0005, "Intec wireless", 0, XTYPE_XBOX }, { 0x0c12, 0x8801, "Nyko Xbox Controller", 0, XTYPE_XBOX }, @@ -391,7 +392,7 @@ static const struct xpad_device { { 0x2dc8, 0x6001, "8BitDo SN30 Pro", 0, XTYPE_XBOX360 }, { 0x2e24, 0x0652, "Hyperkin Duke X-Box One pad", 0, XTYPE_XBOXONE }, { 0x2e24, 0x1688, "Hyperkin X91 X-Box One pad", 0, XTYPE_XBOXONE }, - { 0x2e95, 0x0504, "SCUF Gaming Controller", MAP_SELECT_BUTTON, XTYPE_XBOXONE }, + { 0x2e95, 0x0504, "SCUF Gaming Controller", MAP_SHARE_BUTTON, XTYPE_XBOXONE }, { 0x31e3, 0x1100, "Wooting One", 0, XTYPE_XBOX360 }, { 0x31e3, 0x1200, "Wooting Two", 0, XTYPE_XBOX360 }, { 0x31e3, 0x1210, "Wooting Lekker", 0, XTYPE_XBOX360 }, @@ -1028,7 +1029,7 @@ static void xpad360w_process_packet(struct usb_xpad *xpad, u16 cmd, unsigned cha * The report format was gleaned from * https://github.com/kylelemons/xbox/blob/master/xbox.go */ -static void xpadone_process_packet(struct usb_xpad *xpad, u16 cmd, unsigned char *data) +static void xpadone_process_packet(struct usb_xpad *xpad, u16 cmd, unsigned char *data, u32 len) { struct input_dev *dev = xpad->dev; bool do_sync = false; @@ -1069,8 +1070,12 @@ static void xpadone_process_packet(struct usb_xpad *xpad, u16 cmd, unsigned char /* menu/view buttons */ input_report_key(dev, BTN_START, data[4] & BIT(2)); input_report_key(dev, BTN_SELECT, data[4] & BIT(3)); - if (xpad->mapping & MAP_SELECT_BUTTON) - input_report_key(dev, KEY_RECORD, data[22] & BIT(0)); + if (xpad->mapping & MAP_SHARE_BUTTON) { + if (xpad->mapping & MAP_SHARE_OFFSET) + input_report_key(dev, KEY_RECORD, data[len - 26] & BIT(0)); + else + input_report_key(dev, KEY_RECORD, data[len - 18] & BIT(0)); + } /* buttons A,B,X,Y */ input_report_key(dev, BTN_A, data[4] & BIT(4)); @@ -1218,7 +1223,7 @@ static void xpad_irq_in(struct urb *urb) xpad360w_process_packet(xpad, 0, xpad->idata); break; case XTYPE_XBOXONE: - xpadone_process_packet(xpad, 0, xpad->idata); + xpadone_process_packet(xpad, 0, xpad->idata, urb->actual_length); break; default: xpad_process_packet(xpad, 0, xpad->idata); @@ -1945,7 +1950,7 @@ static int xpad_init_input(struct usb_xpad *xpad) xpad->xtype == XTYPE_XBOXONE) { for (i = 0; xpad360_btn[i] >= 0; i++) input_set_capability(input_dev, EV_KEY, xpad360_btn[i]); - if (xpad->mapping & MAP_SELECT_BUTTON) + if (xpad->mapping & MAP_SHARE_BUTTON) input_set_capability(input_dev, EV_KEY, KEY_RECORD); } else { for (i = 0; xpad_btn[i] >= 0; i++) From 7026d23cb383120712f3a214b0b33ca349cd21a0 Mon Sep 17 00:00:00 2001 From: Vicki Pfau Date: Sat, 10 May 2025 23:00:10 -0700 Subject: [PATCH 376/555] Input: xpad - add support for several more controllers This adds support for several new controllers, all of which include Share buttons: - HORI Drum controller - PowerA Fusion Pro 4 - 8BitDo Ultimate 3-mode Controller - Hyperkin DuchesS Xbox One controller - PowerA MOGA XP-Ultra controller Signed-off-by: Vicki Pfau Link: https://lore.kernel.org/r/20250328234345.989761-4-vi@endrift.com Cc: stable@vger.kernel.org Signed-off-by: Dmitry Torokhov --- drivers/input/joystick/xpad.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/input/joystick/xpad.c b/drivers/input/joystick/xpad.c index 8ee7d8e5d1c7..418553cadc03 100644 --- a/drivers/input/joystick/xpad.c +++ b/drivers/input/joystick/xpad.c @@ -282,6 +282,7 @@ static const struct xpad_device { { 0x0f0d, 0x00dc, "HORIPAD FPS for Nintendo Switch", MAP_TRIGGERS_TO_BUTTONS, XTYPE_XBOX360 }, { 0x0f0d, 0x0151, "Hori Racing Wheel Overdrive for Xbox Series X", 0, XTYPE_XBOXONE }, { 0x0f0d, 0x0152, "Hori Racing Wheel Overdrive for Xbox Series X", 0, XTYPE_XBOXONE }, + { 0x0f0d, 0x01b2, "HORI Taiko No Tatsujin Drum Controller", MAP_SHARE_BUTTON, XTYPE_XBOXONE }, { 0x0f30, 0x010b, "Philips Recoil", 0, XTYPE_XBOX }, { 0x0f30, 0x0202, "Joytech Advanced Controller", 0, XTYPE_XBOX }, { 0x0f30, 0x8888, "BigBen XBMiniPad Controller", 0, XTYPE_XBOX }, @@ -354,6 +355,8 @@ static const struct xpad_device { { 0x20d6, 0x2001, "BDA Xbox Series X Wired Controller", 0, XTYPE_XBOXONE }, { 0x20d6, 0x2009, "PowerA Enhanced Wired Controller for Xbox Series X|S", 0, XTYPE_XBOXONE }, { 0x20d6, 0x281f, "PowerA Wired Controller For Xbox 360", 0, XTYPE_XBOX360 }, + { 0x20d6, 0x400b, "PowerA FUSION Pro 4 Wired Controller", MAP_SHARE_BUTTON, XTYPE_XBOXONE }, + { 0x20d6, 0x890b, "PowerA MOGA XP-Ultra Controller", MAP_SHARE_BUTTON, XTYPE_XBOXONE }, { 0x2345, 0xe00b, "Machenike G5 Pro Controller", 0, XTYPE_XBOX360 }, { 0x24c6, 0x5000, "Razer Atrox Arcade Stick", MAP_TRIGGERS_TO_BUTTONS, XTYPE_XBOX360 }, { 0x24c6, 0x5300, "PowerA MINI PROEX Controller", 0, XTYPE_XBOX360 }, @@ -385,11 +388,13 @@ static const struct xpad_device { { 0x294b, 0x3404, "Snakebyte GAMEPAD RGB X", 0, XTYPE_XBOXONE }, { 0x2993, 0x2001, "TECNO Pocket Go", 0, XTYPE_XBOX360 }, { 0x2dc8, 0x2000, "8BitDo Pro 2 Wired Controller fox Xbox", 0, XTYPE_XBOXONE }, + { 0x2dc8, 0x200f, "8BitDo Ultimate 3-mode Controller for Xbox", MAP_SHARE_BUTTON, XTYPE_XBOXONE }, { 0x2dc8, 0x3106, "8BitDo Ultimate Wireless / Pro 2 Wired Controller", 0, XTYPE_XBOX360 }, { 0x2dc8, 0x3109, "8BitDo Ultimate Wireless Bluetooth", 0, XTYPE_XBOX360 }, { 0x2dc8, 0x310a, "8BitDo Ultimate 2C Wireless Controller", 0, XTYPE_XBOX360 }, { 0x2dc8, 0x310b, "8BitDo Ultimate 2 Wireless Controller", 0, XTYPE_XBOX360 }, { 0x2dc8, 0x6001, "8BitDo SN30 Pro", 0, XTYPE_XBOX360 }, + { 0x2e24, 0x0423, "Hyperkin DuchesS Xbox One pad", MAP_SHARE_BUTTON, XTYPE_XBOXONE }, { 0x2e24, 0x0652, "Hyperkin Duke X-Box One pad", 0, XTYPE_XBOXONE }, { 0x2e24, 0x1688, "Hyperkin X91 X-Box One pad", 0, XTYPE_XBOXONE }, { 0x2e95, 0x0504, "SCUF Gaming Controller", MAP_SHARE_BUTTON, XTYPE_XBOXONE }, @@ -716,8 +721,10 @@ static const struct xboxone_init_packet xboxone_init_packets[] = { XBOXONE_INIT_PKT(0x045e, 0x0b00, xboxone_s_init), XBOXONE_INIT_PKT(0x045e, 0x0b00, extra_input_packet_init), XBOXONE_INIT_PKT(0x0e6f, 0x0000, xboxone_pdp_led_on), + XBOXONE_INIT_PKT(0x0f0d, 0x01b2, xboxone_pdp_led_on), XBOXONE_INIT_PKT(0x20d6, 0xa01a, xboxone_pdp_led_on), XBOXONE_INIT_PKT(0x0e6f, 0x0000, xboxone_pdp_auth), + XBOXONE_INIT_PKT(0x0f0d, 0x01b2, xboxone_pdp_auth), XBOXONE_INIT_PKT(0x20d6, 0xa01a, xboxone_pdp_auth), XBOXONE_INIT_PKT(0x24c6, 0x541a, xboxone_rumblebegin_init), XBOXONE_INIT_PKT(0x24c6, 0x542a, xboxone_rumblebegin_init), From 93406e9d024058b3bf487656ddd0ac552e5a366e Mon Sep 17 00:00:00 2001 From: Vicki Pfau Date: Sat, 10 May 2025 23:06:34 -0700 Subject: [PATCH 377/555] Input: xpad - fix xpad_device sorting A recent commit put one entry in the wrong place. This just moves it to the right place. Signed-off-by: Vicki Pfau Link: https://lore.kernel.org/r/20250328234345.989761-5-vi@endrift.com Signed-off-by: Dmitry Torokhov --- drivers/input/joystick/xpad.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/input/joystick/xpad.c b/drivers/input/joystick/xpad.c index 418553cadc03..57a5ff3d1992 100644 --- a/drivers/input/joystick/xpad.c +++ b/drivers/input/joystick/xpad.c @@ -141,9 +141,9 @@ static const struct xpad_device { { 0x044f, 0x0f00, "Thrustmaster Wheel", 0, XTYPE_XBOX }, { 0x044f, 0x0f03, "Thrustmaster Wheel", 0, XTYPE_XBOX }, { 0x044f, 0x0f07, "Thrustmaster, Inc. Controller", 0, XTYPE_XBOX }, - { 0x044f, 0xd01e, "ThrustMaster, Inc. ESWAP X 2 ELDEN RING EDITION", 0, XTYPE_XBOXONE }, { 0x044f, 0x0f10, "Thrustmaster Modena GT Wheel", 0, XTYPE_XBOX }, { 0x044f, 0xb326, "Thrustmaster Gamepad GP XID", 0, XTYPE_XBOX360 }, + { 0x044f, 0xd01e, "ThrustMaster, Inc. ESWAP X 2 ELDEN RING EDITION", 0, XTYPE_XBOXONE }, { 0x045e, 0x0202, "Microsoft X-Box pad v1 (US)", 0, XTYPE_XBOX }, { 0x045e, 0x0285, "Microsoft X-Box pad (Japan)", 0, XTYPE_XBOX }, { 0x045e, 0x0287, "Microsoft Xbox Controller S", 0, XTYPE_XBOX }, From f7387eff4bad33d12719c66c43541c095556ae4e Mon Sep 17 00:00:00 2001 From: Seongman Lee Date: Sun, 11 May 2025 18:23:28 +0900 Subject: [PATCH 378/555] x86/sev: Fix operator precedence in GHCB_MSR_VMPL_REQ_LEVEL macro The GHCB_MSR_VMPL_REQ_LEVEL macro lacked parentheses around the bitmask expression, causing the shift operation to bind too early. As a result, when requesting VMPL1 (e.g., GHCB_MSR_VMPL_REQ_LEVEL(1)), incorrect values such as 0x000000016 were generated instead of the intended 0x100000016 (the requested VMPL level is specified in GHCBData[39:32]). Fix the precedence issue by grouping the masked value before applying the shift. [ bp: Massage commit message. ] Fixes: 34ff65901735 ("x86/sev: Use kernel provided SVSM Calling Areas") Signed-off-by: Seongman Lee Signed-off-by: Borislav Petkov (AMD) Link: https://lore.kernel.org/20250511092329.12680-1-cloudlee1719@gmail.com --- arch/x86/include/asm/sev-common.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/include/asm/sev-common.h b/arch/x86/include/asm/sev-common.h index acb85b9346d8..0020d77a0800 100644 --- a/arch/x86/include/asm/sev-common.h +++ b/arch/x86/include/asm/sev-common.h @@ -116,7 +116,7 @@ enum psc_op { #define GHCB_MSR_VMPL_REQ 0x016 #define GHCB_MSR_VMPL_REQ_LEVEL(v) \ /* GHCBData[39:32] */ \ - (((u64)(v) & GENMASK_ULL(7, 0) << 32) | \ + ((((u64)(v) & GENMASK_ULL(7, 0)) << 32) | \ /* GHCBDdata[11:0] */ \ GHCB_MSR_VMPL_REQ) From ff7b190aef6cccdb6f14d20c5753081fe6420e0b Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Sun, 11 May 2025 15:45:27 +0200 Subject: [PATCH 379/555] ALSA: seq: Fix delivery of UMP events to group ports When an event with UMP message is sent to a UMP client, the EP port receives always no matter where the event is sent to, as it's a catch-all port. OTOH, if an event is sent to EP port, and if the event has a certain UMP Group, it should have been delivered to the associated UMP Group port, too, but this was ignored, so far. This patch addresses the behavior. Now a UMP event sent to the Endpoint port will be delivered to the subscribers of the UMP group port the event is associated with. The patch also does a bit of refactoring to simplify the code about __deliver_to_subscribers(). Fixes: 177ccf811df4 ("ALSA: seq: Support MIDI 2.0 UMP Endpoint port") Link: https://patch.msgid.link/20250511134528.6314-1-tiwai@suse.de Signed-off-by: Takashi Iwai --- sound/core/seq/seq_clientmgr.c | 52 ++++++++++++++++++++------------ sound/core/seq/seq_ump_convert.c | 18 +++++++++++ sound/core/seq/seq_ump_convert.h | 1 + 3 files changed, 52 insertions(+), 19 deletions(-) diff --git a/sound/core/seq/seq_clientmgr.c b/sound/core/seq/seq_clientmgr.c index 198c598a5393..880240924bfd 100644 --- a/sound/core/seq/seq_clientmgr.c +++ b/sound/core/seq/seq_clientmgr.c @@ -732,15 +732,21 @@ static int snd_seq_deliver_single_event(struct snd_seq_client *client, */ static int __deliver_to_subscribers(struct snd_seq_client *client, struct snd_seq_event *event, - struct snd_seq_client_port *src_port, - int atomic, int hop) + int port, int atomic, int hop) { + struct snd_seq_client_port *src_port; struct snd_seq_subscribers *subs; int err, result = 0, num_ev = 0; union __snd_seq_event event_saved; size_t saved_size; struct snd_seq_port_subs_info *grp; + if (port < 0) + return 0; + src_port = snd_seq_port_use_ptr(client, port); + if (!src_port) + return 0; + /* save original event record */ saved_size = snd_seq_event_packet_size(event); memcpy(&event_saved, event, saved_size); @@ -775,6 +781,7 @@ static int __deliver_to_subscribers(struct snd_seq_client *client, read_unlock(&grp->list_lock); else up_read(&grp->list_mutex); + snd_seq_port_unlock(src_port); memcpy(event, &event_saved, saved_size); return (result < 0) ? result : num_ev; } @@ -783,25 +790,32 @@ static int deliver_to_subscribers(struct snd_seq_client *client, struct snd_seq_event *event, int atomic, int hop) { - struct snd_seq_client_port *src_port; - int ret = 0, ret2; + int ret; +#if IS_ENABLED(CONFIG_SND_SEQ_UMP) + int ret2; +#endif - src_port = snd_seq_port_use_ptr(client, event->source.port); - if (src_port) { - ret = __deliver_to_subscribers(client, event, src_port, atomic, hop); - snd_seq_port_unlock(src_port); - } - - if (client->ump_endpoint_port < 0 || - event->source.port == client->ump_endpoint_port) + ret = __deliver_to_subscribers(client, event, + event->source.port, atomic, hop); +#if IS_ENABLED(CONFIG_SND_SEQ_UMP) + if (!snd_seq_client_is_ump(client) || client->ump_endpoint_port < 0) return ret; - - src_port = snd_seq_port_use_ptr(client, client->ump_endpoint_port); - if (!src_port) - return ret; - ret2 = __deliver_to_subscribers(client, event, src_port, atomic, hop); - snd_seq_port_unlock(src_port); - return ret2 < 0 ? ret2 : ret; + /* If it's an event from EP port (and with a UMP group), + * deliver to subscribers of the corresponding UMP group port, too. + * Or, if it's from non-EP port, deliver to subscribers of EP port, too. + */ + if (event->source.port == client->ump_endpoint_port) + ret2 = __deliver_to_subscribers(client, event, + snd_seq_ump_group_port(event), + atomic, hop); + else + ret2 = __deliver_to_subscribers(client, event, + client->ump_endpoint_port, + atomic, hop); + if (ret2 < 0) + return ret2; +#endif + return ret; } /* deliver an event to the destination port(s). diff --git a/sound/core/seq/seq_ump_convert.c b/sound/core/seq/seq_ump_convert.c index ff7e558b4d51..db2f169cae11 100644 --- a/sound/core/seq/seq_ump_convert.c +++ b/sound/core/seq/seq_ump_convert.c @@ -1285,3 +1285,21 @@ int snd_seq_deliver_to_ump(struct snd_seq_client *source, else return cvt_to_ump_midi1(dest, dest_port, event, atomic, hop); } + +/* return the UMP group-port number of the event; + * return -1 if groupless or non-UMP event + */ +int snd_seq_ump_group_port(const struct snd_seq_event *event) +{ + const struct snd_seq_ump_event *ump_ev = + (const struct snd_seq_ump_event *)event; + unsigned char type; + + if (!snd_seq_ev_is_ump(event)) + return -1; + type = ump_message_type(ump_ev->ump[0]); + if (ump_is_groupless_msg(type)) + return -1; + /* group-port number starts from 1 */ + return ump_message_group(ump_ev->ump[0]) + 1; +} diff --git a/sound/core/seq/seq_ump_convert.h b/sound/core/seq/seq_ump_convert.h index 6c146d803280..4abf0a7637d7 100644 --- a/sound/core/seq/seq_ump_convert.h +++ b/sound/core/seq/seq_ump_convert.h @@ -18,5 +18,6 @@ int snd_seq_deliver_to_ump(struct snd_seq_client *source, struct snd_seq_client_port *dest_port, struct snd_seq_event *event, int atomic, int hop); +int snd_seq_ump_group_port(const struct snd_seq_event *event); #endif /* __SEQ_UMP_CONVERT_H */ From 1f93d877f09d987f08baedd50597aaaa72a37be4 Mon Sep 17 00:00:00 2001 From: Peter Ujfalusi Date: Thu, 8 May 2025 21:12:07 +0300 Subject: [PATCH 380/555] ALSA/hda: intel-sdw-acpi: Correct sdw_intel_acpi_scan() function parameter The acpi_handle should be just a handle and not a pointer in sdw_intel_acpi_scan() parameter list. It is called with 'acpi_handle handle' as parameter and it is passing it to acpi_walk_namespace, which also expects acpi_handle and not acpi_handle* Signed-off-by: Peter Ujfalusi Reviewed-by: Bard Liao Reviewed-by: Liam Girdwood Reviewed-by: Ranjani Sridharan Link: https://patch.msgid.link/20250508181207.22113-1-peter.ujfalusi@linux.intel.com Signed-off-by: Takashi Iwai --- include/linux/soundwire/sdw_intel.h | 2 +- sound/hda/intel-sdw-acpi.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/include/linux/soundwire/sdw_intel.h b/include/linux/soundwire/sdw_intel.h index 493d9de4e472..dc6ebaee3d18 100644 --- a/include/linux/soundwire/sdw_intel.h +++ b/include/linux/soundwire/sdw_intel.h @@ -365,7 +365,7 @@ struct sdw_intel_res { * on e.g. which machine driver to select (I2S mode, HDaudio or * SoundWire). */ -int sdw_intel_acpi_scan(acpi_handle *parent_handle, +int sdw_intel_acpi_scan(acpi_handle parent_handle, struct sdw_intel_acpi_info *info); void sdw_intel_process_wakeen_event(struct sdw_intel_ctx *ctx); diff --git a/sound/hda/intel-sdw-acpi.c b/sound/hda/intel-sdw-acpi.c index 8686adaf4531..d3511135f7d3 100644 --- a/sound/hda/intel-sdw-acpi.c +++ b/sound/hda/intel-sdw-acpi.c @@ -177,7 +177,7 @@ static acpi_status sdw_intel_acpi_cb(acpi_handle handle, u32 level, * sdw_intel_startup() is required for creation of devices and bus * startup */ -int sdw_intel_acpi_scan(acpi_handle *parent_handle, +int sdw_intel_acpi_scan(acpi_handle parent_handle, struct sdw_intel_acpi_info *info) { acpi_status status; From dd33993a9721ab1dae38bd37c9f665987d554239 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Sun, 11 May 2025 16:11:45 +0200 Subject: [PATCH 381/555] ALSA: ump: Fix a typo of snd_ump_stream_msg_device_info s/devince/device/ It's used only internally, so no any behavior changes. Fixes: 37e0e14128e0 ("ALSA: ump: Support UMP Endpoint and Function Block parsing") Acked-by: Greg Kroah-Hartman Link: https://patch.msgid.link/20250511141147.10246-1-tiwai@suse.de Signed-off-by: Takashi Iwai --- drivers/usb/gadget/function/f_midi2.c | 2 +- include/sound/ump_msg.h | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/usb/gadget/function/f_midi2.c b/drivers/usb/gadget/function/f_midi2.c index 12e866fb311d..0a800ba53816 100644 --- a/drivers/usb/gadget/function/f_midi2.c +++ b/drivers/usb/gadget/function/f_midi2.c @@ -475,7 +475,7 @@ static void reply_ump_stream_ep_info(struct f_midi2_ep *ep) /* reply a UMP EP device info */ static void reply_ump_stream_ep_device(struct f_midi2_ep *ep) { - struct snd_ump_stream_msg_devince_info rep = { + struct snd_ump_stream_msg_device_info rep = { .type = UMP_MSG_TYPE_STREAM, .status = UMP_STREAM_MSG_STATUS_DEVICE_INFO, .manufacture_id = ep->info.manufacturer, diff --git a/include/sound/ump_msg.h b/include/sound/ump_msg.h index 72f60ddfea75..9556b4755a1e 100644 --- a/include/sound/ump_msg.h +++ b/include/sound/ump_msg.h @@ -604,7 +604,7 @@ struct snd_ump_stream_msg_ep_info { } __packed; /* UMP Stream Message: Device Info Notification (128bit) */ -struct snd_ump_stream_msg_devince_info { +struct snd_ump_stream_msg_device_info { #ifdef __BIG_ENDIAN_BITFIELD /* 0 */ u32 type:4; @@ -754,7 +754,7 @@ struct snd_ump_stream_msg_fb_name { union snd_ump_stream_msg { struct snd_ump_stream_msg_ep_discovery ep_discovery; struct snd_ump_stream_msg_ep_info ep_info; - struct snd_ump_stream_msg_devince_info device_info; + struct snd_ump_stream_msg_device_info device_info; struct snd_ump_stream_msg_stream_cfg stream_cfg; struct snd_ump_stream_msg_fb_discovery fb_discovery; struct snd_ump_stream_msg_fb_info fb_info; From 82f2b0b97b36ee3fcddf0f0780a9a0825d52fec3 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 11 May 2025 14:54:11 -0700 Subject: [PATCH 382/555] Linux 6.15-rc6 --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index b29cc321ffd9..64c514f4bc19 100644 --- a/Makefile +++ b/Makefile @@ -2,7 +2,7 @@ VERSION = 6 PATCHLEVEL = 15 SUBLEVEL = 0 -EXTRAVERSION = -rc5 +EXTRAVERSION = -rc6 NAME = Baby Opossum Posse # *DOCUMENTATION* From a833a693a490ecff8ba377654c6d4d333718b6b1 Mon Sep 17 00:00:00 2001 From: Wupeng Ma Date: Thu, 10 Apr 2025 14:26:33 +0800 Subject: [PATCH 383/555] mm: hugetlb: fix incorrect fallback for subpool During our testing with hugetlb subpool enabled, we observe that hstate->resv_huge_pages may underflow into negative values. Root cause analysis reveals a race condition in subpool reservation fallback handling as follow: hugetlb_reserve_pages() /* Attempt subpool reservation */ gbl_reserve = hugepage_subpool_get_pages(spool, chg); /* Global reservation may fail after subpool allocation */ if (hugetlb_acct_memory(h, gbl_reserve) < 0) goto out_put_pages; out_put_pages: /* This incorrectly restores reservation to subpool */ hugepage_subpool_put_pages(spool, chg); When hugetlb_acct_memory() fails after subpool allocation, the current implementation over-commits subpool reservations by returning the full 'chg' value instead of the actual allocated 'gbl_reserve' amount. This discrepancy propagates to global reservations during subsequent releases, eventually causing resv_huge_pages underflow. This problem can be trigger easily with the following steps: 1. reverse hugepage for hugeltb allocation 2. mount hugetlbfs with min_size to enable hugetlb subpool 3. alloc hugepages with two task(make sure the second will fail due to insufficient amount of hugepages) 4. with for a few seconds and repeat step 3 which will make hstate->resv_huge_pages to go below zero. To fix this problem, return corrent amount of pages to subpool during the fallback after hugepage_subpool_get_pages is called. Link: https://lkml.kernel.org/r/20250410062633.3102457-1-mawupeng1@huawei.com Fixes: 1c5ecae3a93f ("hugetlbfs: add minimum size accounting to subpools") Signed-off-by: Wupeng Ma Tested-by: Joshua Hahn Reviewed-by: Oscar Salvador Cc: David Hildenbrand Cc: Ma Wupeng Cc: Muchun Song Cc: Signed-off-by: Andrew Morton --- mm/hugetlb.c | 28 ++++++++++++++++++++++------ 1 file changed, 22 insertions(+), 6 deletions(-) diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 6ea1be71aa42..7ae38bfb9096 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -3010,7 +3010,7 @@ struct folio *alloc_hugetlb_folio(struct vm_area_struct *vma, struct hugepage_subpool *spool = subpool_vma(vma); struct hstate *h = hstate_vma(vma); struct folio *folio; - long retval, gbl_chg; + long retval, gbl_chg, gbl_reserve; map_chg_state map_chg; int ret, idx; struct hugetlb_cgroup *h_cg = NULL; @@ -3163,8 +3163,16 @@ out_uncharge_cgroup_reservation: hugetlb_cgroup_uncharge_cgroup_rsvd(idx, pages_per_huge_page(h), h_cg); out_subpool_put: - if (map_chg) - hugepage_subpool_put_pages(spool, 1); + /* + * put page to subpool iff the quota of subpool's rsv_hpages is used + * during hugepage_subpool_get_pages. + */ + if (map_chg && !gbl_chg) { + gbl_reserve = hugepage_subpool_put_pages(spool, 1); + hugetlb_acct_memory(h, -gbl_reserve); + } + + out_end_reservation: if (map_chg != MAP_CHG_ENFORCED) vma_end_reservation(h, vma, addr); @@ -7239,7 +7247,7 @@ bool hugetlb_reserve_pages(struct inode *inode, struct vm_area_struct *vma, vm_flags_t vm_flags) { - long chg = -1, add = -1; + long chg = -1, add = -1, spool_resv, gbl_resv; struct hstate *h = hstate_inode(inode); struct hugepage_subpool *spool = subpool_inode(inode); struct resv_map *resv_map; @@ -7374,8 +7382,16 @@ bool hugetlb_reserve_pages(struct inode *inode, return true; out_put_pages: - /* put back original number of pages, chg */ - (void)hugepage_subpool_put_pages(spool, chg); + spool_resv = chg - gbl_reserve; + if (spool_resv) { + /* put sub pool's reservation back, chg - gbl_reserve */ + gbl_resv = hugepage_subpool_put_pages(spool, spool_resv); + /* + * subpool's reserved pages can not be put back due to race, + * return to hstate. + */ + hugetlb_acct_memory(h, -gbl_resv); + } out_uncharge_cgroup: hugetlb_cgroup_uncharge_cgroup_rsvd(hstate_index(h), chg * pages_per_huge_page(h), h_cg); From e9f180d7cfde23b9f8eebd60272465176373ab2c Mon Sep 17 00:00:00 2001 From: David Hildenbrand Date: Tue, 22 Apr 2025 16:49:42 +0200 Subject: [PATCH 384/555] kernel/fork: only call untrack_pfn_clear() on VMAs duplicated for fork() Not intuitive, but vm_area_dup() located in kernel/fork.c is not only used for duplicating VMAs during fork(), but also for duplicating VMAs when splitting VMAs or when mremap()'ing them. VM_PFNMAP mappings can at least get ordinarily mremap()'ed (no change in size) and apparently also shrunk during mremap(), which implies duplicating the VMA in __split_vma() first. In case of ordinary mremap() (no change in size), we first duplicate the VMA in copy_vma_and_data()->copy_vma() to then call untrack_pfn_clear() on the old VMA: we effectively move the VM_PAT reservation. So the untrack_pfn_clear() call on the new VMA duplicating is wrong in that context. Splitting of VMAs seems problematic, because we don't duplicate/adjust the reservation when splitting the VMA. Instead, in memtype_erase() -- called during zapping/munmap -- we shrink a reservation in case only the end address matches: Assume we split a VMA into A and B, both would share a reservation until B is unmapped. So when unmapping B, the reservation would be updated to cover only A. When unmapping A, we would properly remove the now-shrunk reservation. That scenario describes the mremap() shrinking (old_size > new_size), where we split + unmap B, and the untrack_pfn_clear() on the new VMA when is wrong. What if we manage to split a VM_PFNMAP VMA into A and B and unmap A first? It would be broken because we would never free the reservation. Likely, there are ways to trigger such a VMA split outside of mremap(). Affecting other VMA duplication was not intended, vm_area_dup() being used outside of kernel/fork.c was an oversight. So let's fix that for; how to handle VMA splits better should be investigated separately. With a simple reproducer that uses mprotect() to split such a VMA I can trigger x86/PAT: pat_mremap:26448 freeing invalid memtype [mem 0x00000000-0x00000fff] Link: https://lkml.kernel.org/r/20250422144942.2871395-1-david@redhat.com Fixes: dc84bc2aba85 ("x86/mm/pat: Fix VM_PAT handling when fork() fails in copy_page_range()") Signed-off-by: David Hildenbrand Reviewed-by: Lorenzo Stoakes Cc: Ingo Molnar Cc: Dave Hansen Cc: Andy Lutomirski Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Borislav Petkov Cc: Rik van Riel Cc: "H. Peter Anvin" Cc: Linus Torvalds Signed-off-by: Andrew Morton --- kernel/fork.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/kernel/fork.c b/kernel/fork.c index c4b26cd8998b..168681fc4b25 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -498,10 +498,6 @@ struct vm_area_struct *vm_area_dup(struct vm_area_struct *orig) vma_numab_state_init(new); dup_anon_vma_name(orig, new); - /* track_pfn_copy() will later take care of copying internal state. */ - if (unlikely(new->vm_flags & VM_PFNMAP)) - untrack_pfn_clear(new); - return new; } @@ -672,6 +668,11 @@ static __latent_entropy int dup_mmap(struct mm_struct *mm, tmp = vm_area_dup(mpnt); if (!tmp) goto fail_nomem; + + /* track_pfn_copy() will later take care of copying internal state. */ + if (unlikely(tmp->vm_flags & VM_PFNMAP)) + untrack_pfn_clear(tmp); + retval = vma_dup_policy(mpnt, tmp); if (retval) goto fail_nomem_policy; From 4b7c0857f87a4c56e6e0a774939c2504b7afdc00 Mon Sep 17 00:00:00 2001 From: Kairui Song Date: Fri, 25 Apr 2025 15:43:25 +0800 Subject: [PATCH 385/555] mm/memory: fix mapcount / refcount sanity check for mTHP reuse The following WARNING was triggered during swap stress test with mTHP enabled: [ 6609.335758] ------------[ cut here ]------------ [ 6609.337758] WARNING: CPU: 82 PID: 755116 at mm/memory.c:3794 do_wp_page+0x1084/0x10e0 [ 6609.340922] Modules linked in: zram virtiofs [ 6609.342699] CPU: 82 UID: 0 PID: 755116 Comm: sh Kdump: loaded Not tainted 6.15.0-rc1+ #1429 PREEMPT(voluntary) [ 6609.347620] Hardware name: Red Hat KVM/RHEL-AV, BIOS 0.0.0 02/06/2015 [ 6609.349909] RIP: 0010:do_wp_page+0x1084/0x10e0 [ 6609.351532] Code: ff ff 48 c7 c6 80 ba 49 82 4c 89 ef e8 95 fd fe ff 0f 0b bd f5 ff ff ff e9 43 fb ff ff 41 83 a9 bc 12 00 00 01 e9 5c fb ff ff <0f> 0b e9 a6 fc ff ff 65 ff 00 f0 48 0f b a 6d 00 1f 0f 83 82 fc ff [ 6609.357959] RSP: 0000:ffffc90002273d40 EFLAGS: 00010287 [ 6609.359915] RAX: 000000000000000f RBX: 0000000000000000 RCX: 000fffffffe00000 [ 6609.362606] RDX: 0000000000000010 RSI: 000055a119ac1000 RDI: ffffea000ae6ec00 [ 6609.365143] RBP: ffffea000ae6ec68 R08: 84000002b9bb1025 R09: 000055a119ab6000 [ 6609.367569] R10: ffff8881caa2ad80 R11: 0000000000000000 R12: ffff8881caa2ad80 [ 6609.370255] R13: ffffea000ae6ec00 R14: 000055a119ac1c9c R15: ffffc90002273dd8 [ 6609.373007] FS: 00007f08e467f740(0000) GS:ffff88a07c214000(0000) knlGS:0000000000000000 [ 6609.375999] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 6609.377946] CR2: 000055a119ac1c9c CR3: 00000001adfd6005 CR4: 0000000000770eb0 [ 6609.380376] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 6609.382853] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 6609.385216] PKRU: 55555554 [ 6609.386141] Call Trace: [ 6609.387017] [ 6609.387718] ? ___pte_offset_map+0x1b/0x110 [ 6609.389056] __handle_mm_fault+0xa51/0xf00 [ 6609.390363] ? exc_page_fault+0x6a/0x140 [ 6609.391629] handle_mm_fault+0x13d/0x360 [ 6609.392856] do_user_addr_fault+0x2f2/0x7f0 [ 6609.394160] ? sigprocmask+0x77/0xa0 [ 6609.395375] exc_page_fault+0x6a/0x140 [ 6609.396735] asm_exc_page_fault+0x26/0x30 [ 6609.398224] RIP: 0033:0x55a1050bc18b [ 6609.399567] Code: 8b 3f 4d 85 ff 74 40 41 39 5f 18 75 f2 49 8b 7f 08 44 38 27 75 e9 4c 89 c6 4c 89 45 c8 e8 bd 83 fa ff 4c 8b 45 c8 85 c0 75 d5 <41> 83 47 1c 01 48 83 c4 28 4c 89 f8 5b 4 1 5c 41 5d 41 5e 41 5f 5d [ 6609.405971] RSP: 002b:00007ffcf5f37d90 EFLAGS: 00010246 [ 6609.407737] RAX: 0000000000000000 RBX: 00000000182768fa RCX: 0000000000000000 [ 6609.410151] RDX: 00000000000000fa RSI: 000055a105175c7b RDI: 000055a119ac1c60 [ 6609.412606] RBP: 00007ffcf5f37de0 R08: 000055a105175c7b R09: 0000000000000000 [ 6609.414998] R10: 000000004d2dfb5a R11: 0000000000000246 R12: 0000000000000050 [ 6609.417193] R13: 00000000000000fa R14: 000055a119abaf60 R15: 000055a119ac1c80 [ 6609.419268] [ 6609.419928] ---[ end trace 0000000000000000 ]--- The WARN_ON here is simply incorrect. The refcount here must be at least the mapcount, not the opposite. Each mapcount must have a corresponding refcount, but the refcount may increase if other components grab the folio, which is acceptable. Meanwhile, having a mapcount larger than refcount is a real problem. So fix the WARN_ON condition. Link: https://lkml.kernel.org/r/20250425074325.61833-1-ryncsn@gmail.com Fixes: 1da190f4d0a6 ("mm: Copy-on-Write (COW) reuse support for PTE-mapped THP") Signed-off-by: Kairui Song Reported-by: Kairui Song Closes: https://lore.kernel.org/all/CAMgjq7D+ea3eg9gRCVvRnto3Sv3_H3WVhupX4e=k8T5QAfBHbw@mail.gmail.com/ Suggested-by: David Hildenbrand Acked-by: David Hildenbrand Reviewed-by: Anshuman Khandual Reviewed-by: Oscar Salvador Signed-off-by: Andrew Morton --- mm/memory.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/memory.c b/mm/memory.c index ba3ea0a82f7f..49199410805c 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -3751,7 +3751,7 @@ static bool __wp_can_reuse_large_anon_folio(struct folio *folio, /* Stabilize the mapcount vs. refcount and recheck. */ folio_lock_large_mapcount(folio); - VM_WARN_ON_ONCE(folio_large_mapcount(folio) < folio_ref_count(folio)); + VM_WARN_ON_ONCE_FOLIO(folio_large_mapcount(folio) > folio_ref_count(folio), folio); if (folio_test_large_maybe_mapped_shared(folio)) goto unlock; From 0ae0227fa31dda5bfc6b5a0145952d46fe57408b Mon Sep 17 00:00:00 2001 From: David Wang <00107082@163.com> Date: Tue, 6 May 2025 03:30:34 +0800 Subject: [PATCH 386/555] mm/codetag: move tag retrieval back upfront in __free_pages() Commit 51ff4d7486f0 ("mm: avoid extra mem_alloc_profiling_enabled() checks") introduces a possible use-after-free scenario, when page is non-compound, page[0] could be released by other thread right after put_page_testzero failed in current thread, pgalloc_tag_sub_pages afterwards would manipulate an invalid page for accounting remaining pages: [timeline] [thread1] [thread2] | alloc_page non-compound V | get_page, rf counter inc V | in ___free_pages | put_page_testzero fails V | put_page, page released V | in ___free_pages, | pgalloc_tag_sub_pages | manipulate an invalid page V Restore __free_pages() to its state before, retrieve alloc tag beforehand. Link: https://lkml.kernel.org/r/20250505193034.91682-1-00107082@163.com Fixes: 51ff4d7486f0 ("mm: avoid extra mem_alloc_profiling_enabled() checks") Signed-off-by: David Wang <00107082@163.com> Acked-by: Suren Baghdasaryan Reviewed-by: Vlastimil Babka Cc: Brendan Jackman Cc: Johannes Weiner Cc: Michal Hocko Cc: Shakeel Butt Cc: Vlastimil Babka Cc: Zi Yan Signed-off-by: Andrew Morton --- include/linux/pgalloc_tag.h | 8 ++++++++ mm/page_alloc.c | 15 ++++++--------- 2 files changed, 14 insertions(+), 9 deletions(-) diff --git a/include/linux/pgalloc_tag.h b/include/linux/pgalloc_tag.h index c74077977830..8a7f4f802c57 100644 --- a/include/linux/pgalloc_tag.h +++ b/include/linux/pgalloc_tag.h @@ -188,6 +188,13 @@ static inline struct alloc_tag *__pgalloc_tag_get(struct page *page) return tag; } +static inline struct alloc_tag *pgalloc_tag_get(struct page *page) +{ + if (mem_alloc_profiling_enabled()) + return __pgalloc_tag_get(page); + return NULL; +} + void pgalloc_tag_split(struct folio *folio, int old_order, int new_order); void pgalloc_tag_swap(struct folio *new, struct folio *old); @@ -199,6 +206,7 @@ static inline void clear_page_tag_ref(struct page *page) {} static inline void alloc_tag_sec_init(void) {} static inline void pgalloc_tag_split(struct folio *folio, int old_order, int new_order) {} static inline void pgalloc_tag_swap(struct folio *new, struct folio *old) {} +static inline struct alloc_tag *pgalloc_tag_get(struct page *page) { return NULL; } #endif /* CONFIG_MEM_ALLOC_PROFILING */ diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 5669baf2a6fe..1b00e14a9780 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1151,14 +1151,9 @@ static inline void pgalloc_tag_sub(struct page *page, unsigned int nr) __pgalloc_tag_sub(page, nr); } -static inline void pgalloc_tag_sub_pages(struct page *page, unsigned int nr) +/* When tag is not NULL, assuming mem_alloc_profiling_enabled */ +static inline void pgalloc_tag_sub_pages(struct alloc_tag *tag, unsigned int nr) { - struct alloc_tag *tag; - - if (!mem_alloc_profiling_enabled()) - return; - - tag = __pgalloc_tag_get(page); if (tag) this_cpu_sub(tag->counters->bytes, PAGE_SIZE * nr); } @@ -1168,7 +1163,7 @@ static inline void pgalloc_tag_sub_pages(struct page *page, unsigned int nr) static inline void pgalloc_tag_add(struct page *page, struct task_struct *task, unsigned int nr) {} static inline void pgalloc_tag_sub(struct page *page, unsigned int nr) {} -static inline void pgalloc_tag_sub_pages(struct page *page, unsigned int nr) {} +static inline void pgalloc_tag_sub_pages(struct alloc_tag *tag, unsigned int nr) {} #endif /* CONFIG_MEM_ALLOC_PROFILING */ @@ -5065,11 +5060,13 @@ static void ___free_pages(struct page *page, unsigned int order, { /* get PageHead before we drop reference */ int head = PageHead(page); + /* get alloc tag in case the page is released by others */ + struct alloc_tag *tag = pgalloc_tag_get(page); if (put_page_testzero(page)) __free_frozen_pages(page, order, fpi_flags); else if (!head) { - pgalloc_tag_sub_pages(page, (1 << order) - 1); + pgalloc_tag_sub_pages(tag, (1 << order) - 1); while (order-- > 0) __free_frozen_pages(page + (1 << order), order, fpi_flags); From d55582d6c947a195ed48a1893ba23d2f077bf224 Mon Sep 17 00:00:00 2001 From: Lorenzo Stoakes Date: Tue, 6 May 2025 18:36:01 +0100 Subject: [PATCH 387/555] MAINTAINERS: add mm GUP section As part of the ongoing efforts to sub-divide memory management maintainership and reviewership, establish a section for GUP (Get User Pages) support and add appropriate maintainers and reviewers. Link: https://lkml.kernel.org/r/20250506173601.97562-1-lorenzo.stoakes@oracle.com Signed-off-by: Lorenzo Stoakes Reviewed-by: John Hubbard Acked-by: David Hildenbrand Cc: Jason Gunthorpe Cc: Peter Xu Signed-off-by: Andrew Morton --- MAINTAINERS | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index f21f1dabb5fe..88102bb64dfe 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -15542,6 +15542,18 @@ S: Maintained F: include/linux/execmem.h F: mm/execmem.c +MEMORY MANAGEMENT - GUP (GET USER PAGES) +M: Andrew Morton +M: David Hildenbrand +R: Jason Gunthorpe +R: John Hubbard +R: Peter Xu +L: linux-mm@kvack.org +S: Maintained +W: http://www.linux-mm.org +T: git git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm +F: mm/gup.c + MEMORY MANAGEMENT - NUMA MEMBLOCKS AND NUMA EMULATION M: Andrew Morton M: Mike Rapoport From 23fa022a07555a9cd2dcfe827c769a89c4c2e21e Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Tue, 6 May 2025 14:25:08 +0300 Subject: [PATCH 388/555] mm/page_alloc: ensure try_alloc_pages() plays well with unaccepted memory try_alloc_pages() will not attempt to allocate memory if the system has *any* unaccepted memory. Memory is accepted as needed and can remain in the system indefinitely, causing the interface to always fail. Rather than immediately giving up, attempt to use already accepted memory on free lists. Pass 'alloc_flags' to cond_accept_memory() and do not accept new memory for ALLOC_TRYLOCK requests. Found via code inspection - only BPF uses this at present and the runtime effects are unclear. Link: https://lkml.kernel.org/r/20250506112509.905147-2-kirill.shutemov@linux.intel.com Signed-off-by: Kirill A. Shutemov Fixes: 97769a53f117 ("mm, bpf: Introduce try_alloc_pages() for opportunistic page allocation") Cc: Alexei Starovoitov Cc: Vlastimil Babka Cc: Suren Baghdasaryan Cc: Michal Hocko Cc: Brendan Jackman Cc: Johannes Weiner Cc: Signed-off-by: Andrew Morton --- mm/page_alloc.c | 28 +++++++++++++++------------- 1 file changed, 15 insertions(+), 13 deletions(-) diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 1b00e14a9780..7248e300d36e 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -290,7 +290,8 @@ EXPORT_SYMBOL(nr_online_nodes); #endif static bool page_contains_unaccepted(struct page *page, unsigned int order); -static bool cond_accept_memory(struct zone *zone, unsigned int order); +static bool cond_accept_memory(struct zone *zone, unsigned int order, + int alloc_flags); static bool __free_unaccepted(struct page *page); int page_group_by_mobility_disabled __read_mostly; @@ -3611,7 +3612,7 @@ retry: } } - cond_accept_memory(zone, order); + cond_accept_memory(zone, order, alloc_flags); /* * Detect whether the number of free pages is below high @@ -3638,7 +3639,7 @@ check_alloc_wmark: gfp_mask)) { int ret; - if (cond_accept_memory(zone, order)) + if (cond_accept_memory(zone, order, alloc_flags)) goto try_this_zone; /* @@ -3691,7 +3692,7 @@ try_this_zone: return page; } else { - if (cond_accept_memory(zone, order)) + if (cond_accept_memory(zone, order, alloc_flags)) goto try_this_zone; /* Try again if zone has deferred pages */ @@ -4844,7 +4845,7 @@ unsigned long alloc_pages_bulk_noprof(gfp_t gfp, int preferred_nid, goto failed; } - cond_accept_memory(zone, 0); + cond_accept_memory(zone, 0, alloc_flags); retry_this_zone: mark = wmark_pages(zone, alloc_flags & ALLOC_WMARK_MASK) + nr_pages; if (zone_watermark_fast(zone, 0, mark, @@ -4853,7 +4854,7 @@ retry_this_zone: break; } - if (cond_accept_memory(zone, 0)) + if (cond_accept_memory(zone, 0, alloc_flags)) goto retry_this_zone; /* Try again if zone has deferred pages */ @@ -7281,7 +7282,8 @@ static inline bool has_unaccepted_memory(void) return static_branch_unlikely(&zones_with_unaccepted_pages); } -static bool cond_accept_memory(struct zone *zone, unsigned int order) +static bool cond_accept_memory(struct zone *zone, unsigned int order, + int alloc_flags) { long to_accept, wmark; bool ret = false; @@ -7292,6 +7294,10 @@ static bool cond_accept_memory(struct zone *zone, unsigned int order) if (list_empty(&zone->unaccepted_pages)) return false; + /* Bailout, since try_to_accept_memory_one() needs to take a lock */ + if (alloc_flags & ALLOC_TRYLOCK) + return false; + wmark = promo_wmark_pages(zone); /* @@ -7348,7 +7354,8 @@ static bool page_contains_unaccepted(struct page *page, unsigned int order) return false; } -static bool cond_accept_memory(struct zone *zone, unsigned int order) +static bool cond_accept_memory(struct zone *zone, unsigned int order, + int alloc_flags) { return false; } @@ -7419,11 +7426,6 @@ struct page *try_alloc_pages_noprof(int nid, unsigned int order) if (!pcp_allowed_order(order)) return NULL; -#ifdef CONFIG_UNACCEPTED_MEMORY - /* Bailout, since try_to_accept_memory_one() needs to take a lock */ - if (has_unaccepted_memory()) - return NULL; -#endif /* Bailout, since _deferred_grow_zone() needs to take a lock */ if (deferred_pages_enabled()) return NULL; From fefc075182275057ce607effaa3daa9e6e3bdc73 Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Tue, 6 May 2025 16:32:07 +0300 Subject: [PATCH 389/555] mm/page_alloc: fix race condition in unaccepted memory handling The page allocator tracks the number of zones that have unaccepted memory using static_branch_enc/dec() and uses that static branch in hot paths to determine if it needs to deal with unaccepted memory. Borislav and Thomas pointed out that the tracking is racy: operations on static_branch are not serialized against adding/removing unaccepted pages to/from the zone. Sanity checks inside static_branch machinery detects it: WARNING: CPU: 0 PID: 10 at kernel/jump_label.c:276 __static_key_slow_dec_cpuslocked+0x8e/0xa0 The comment around the WARN() explains the problem: /* * Warn about the '-1' case though; since that means a * decrement is concurrent with a first (0->1) increment. IOW * people are trying to disable something that wasn't yet fully * enabled. This suggests an ordering problem on the user side. */ The effect of this static_branch optimization is only visible on microbenchmark. Instead of adding more complexity around it, remove it altogether. Link: https://lkml.kernel.org/r/20250506133207.1009676-1-kirill.shutemov@linux.intel.com Signed-off-by: Kirill A. Shutemov Fixes: dcdfdd40fa82 ("mm: Add support for unaccepted memory") Link: https://lore.kernel.org/all/20250506092445.GBaBnVXXyvnazly6iF@fat_crate.local Reported-by: Borislav Petkov Tested-by: Borislav Petkov (AMD) Reported-by: Thomas Gleixner Cc: Vlastimil Babka Cc: Suren Baghdasaryan Cc: Michal Hocko Cc: Brendan Jackman Cc: Johannes Weiner Cc: [6.5+] Signed-off-by: Andrew Morton --- mm/internal.h | 1 - mm/mm_init.c | 1 - mm/page_alloc.c | 47 ----------------------------------------------- 3 files changed, 49 deletions(-) diff --git a/mm/internal.h b/mm/internal.h index 25a29872c634..5c7a2b43ad76 100644 --- a/mm/internal.h +++ b/mm/internal.h @@ -1590,7 +1590,6 @@ unsigned long move_page_tables(struct pagetable_move_control *pmc); #ifdef CONFIG_UNACCEPTED_MEMORY void accept_page(struct page *page); -void unaccepted_cleanup_work(struct work_struct *work); #else /* CONFIG_UNACCEPTED_MEMORY */ static inline void accept_page(struct page *page) { diff --git a/mm/mm_init.c b/mm/mm_init.c index 327764ca0ee4..eedce9321e13 100644 --- a/mm/mm_init.c +++ b/mm/mm_init.c @@ -1441,7 +1441,6 @@ static void __meminit zone_init_free_lists(struct zone *zone) #ifdef CONFIG_UNACCEPTED_MEMORY INIT_LIST_HEAD(&zone->unaccepted_pages); - INIT_WORK(&zone->unaccepted_cleanup, unaccepted_cleanup_work); #endif } diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 7248e300d36e..8258349e49ac 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -7172,16 +7172,8 @@ bool has_managed_dma(void) #ifdef CONFIG_UNACCEPTED_MEMORY -/* Counts number of zones with unaccepted pages. */ -static DEFINE_STATIC_KEY_FALSE(zones_with_unaccepted_pages); - static bool lazy_accept = true; -void unaccepted_cleanup_work(struct work_struct *work) -{ - static_branch_dec(&zones_with_unaccepted_pages); -} - static int __init accept_memory_parse(char *p) { if (!strcmp(p, "lazy")) { @@ -7206,11 +7198,7 @@ static bool page_contains_unaccepted(struct page *page, unsigned int order) static void __accept_page(struct zone *zone, unsigned long *flags, struct page *page) { - bool last; - list_del(&page->lru); - last = list_empty(&zone->unaccepted_pages); - account_freepages(zone, -MAX_ORDER_NR_PAGES, MIGRATE_MOVABLE); __mod_zone_page_state(zone, NR_UNACCEPTED, -MAX_ORDER_NR_PAGES); __ClearPageUnaccepted(page); @@ -7219,28 +7207,6 @@ static void __accept_page(struct zone *zone, unsigned long *flags, accept_memory(page_to_phys(page), PAGE_SIZE << MAX_PAGE_ORDER); __free_pages_ok(page, MAX_PAGE_ORDER, FPI_TO_TAIL); - - if (last) { - /* - * There are two corner cases: - * - * - If allocation occurs during the CPU bring up, - * static_branch_dec() cannot be used directly as - * it causes a deadlock on cpu_hotplug_lock. - * - * Instead, use schedule_work() to prevent deadlock. - * - * - If allocation occurs before workqueues are initialized, - * static_branch_dec() should be called directly. - * - * Workqueues are initialized before CPU bring up, so this - * will not conflict with the first scenario. - */ - if (system_wq) - schedule_work(&zone->unaccepted_cleanup); - else - unaccepted_cleanup_work(&zone->unaccepted_cleanup); - } } void accept_page(struct page *page) @@ -7277,20 +7243,12 @@ static bool try_to_accept_memory_one(struct zone *zone) return true; } -static inline bool has_unaccepted_memory(void) -{ - return static_branch_unlikely(&zones_with_unaccepted_pages); -} - static bool cond_accept_memory(struct zone *zone, unsigned int order, int alloc_flags) { long to_accept, wmark; bool ret = false; - if (!has_unaccepted_memory()) - return false; - if (list_empty(&zone->unaccepted_pages)) return false; @@ -7328,22 +7286,17 @@ static bool __free_unaccepted(struct page *page) { struct zone *zone = page_zone(page); unsigned long flags; - bool first = false; if (!lazy_accept) return false; spin_lock_irqsave(&zone->lock, flags); - first = list_empty(&zone->unaccepted_pages); list_add_tail(&page->lru, &zone->unaccepted_pages); account_freepages(zone, MAX_ORDER_NR_PAGES, MIGRATE_MOVABLE); __mod_zone_page_state(zone, NR_UNACCEPTED, MAX_ORDER_NR_PAGES); __SetPageUnaccepted(page); spin_unlock_irqrestore(&zone->lock, flags); - if (first) - static_branch_inc(&zones_with_unaccepted_pages); - return true; } From 02f5bf89f0b0a50f821425932a3590eeb9f193ac Mon Sep 17 00:00:00 2001 From: Sergey Senozhatsky Date: Wed, 7 May 2025 14:42:24 +0900 Subject: [PATCH 390/555] zsmalloc: don't underflow size calculation in zs_obj_write() Do not mix class->size and object size during offsets/sizes calculation in zs_obj_write(). Size classes can merge into clusters, based on objects-per-zspage and pages-per-zspage characteristics, so some size classes can store objects smaller than class->size. This becomes problematic when object size is much smaller than class->size. zsmalloc can falsely decide that object spans two physical pages, because a larger class->size value is used for that check, while the actual object is much smaller and fits the free space of the first physical page, so there is nothing to write to the second page and memcpy() size calculation underflows. Unable to handle kernel paging request at virtual address ffffc00081ff4000 pc : __memcpy+0x10/0x24 lr : zs_obj_write+0x1b0/0x1d0 [zsmalloc] Call trace: __memcpy+0x10/0x24 (P) zram_write_page+0x150/0x4fc [zram] zram_submit_bio+0x5e0/0x6a4 [zram] __submit_bio+0x168/0x220 submit_bio_noacct_nocheck+0x128/0x2c8 submit_bio_noacct+0x19c/0x2f8 This is mostly seen on system with larger page-sizes, because size class cluters of such systems hold wider size ranges than on 4K PAGE_SIZE systems. Assume a 16K PAGE_SIZE system, a write of 820 bytes object to a 864-bytes size class at offset 15560. 15560 + 864 is more than 16384 so zsmalloc attempts to memcpy() it to two physical pages. However, 16384 - 15560 = 824 which is more than 820, so the object in fact doesn't span two physical pages, and there is no data to write to the second physical page. We always know the exact size in bytes of the object that we are about to write (store), so use it instead of class->size. Link: https://lkml.kernel.org/r/20250507054312.4135983-1-senozhatsky@chromium.org Fixes: 44f76413496e ("zsmalloc: introduce new object mapping API") Signed-off-by: Sergey Senozhatsky Reported-by: Igor Belousov Tested-by: Igor Belousov Acked-by: Johannes Weiner Cc: Minchan Kim Signed-off-by: Andrew Morton --- mm/zsmalloc.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/mm/zsmalloc.c b/mm/zsmalloc.c index 961b270f023c..d14a7e317ac8 100644 --- a/mm/zsmalloc.c +++ b/mm/zsmalloc.c @@ -1243,19 +1243,19 @@ void zs_obj_write(struct zs_pool *pool, unsigned long handle, class = zspage_class(pool, zspage); off = offset_in_page(class->size * obj_idx); - if (off + class->size <= PAGE_SIZE) { + if (!ZsHugePage(zspage)) + off += ZS_HANDLE_SIZE; + + if (off + mem_len <= PAGE_SIZE) { /* this object is contained entirely within a page */ void *dst = kmap_local_zpdesc(zpdesc); - if (!ZsHugePage(zspage)) - off += ZS_HANDLE_SIZE; memcpy(dst + off, handle_mem, mem_len); kunmap_local(dst); } else { /* this object spans two pages */ size_t sizes[2]; - off += ZS_HANDLE_SIZE; sizes[0] = PAGE_SIZE - off; sizes[1] = mem_len - sizes[0]; From 75cb1cca2c880179a11c7dd9380b6f14e41a06a4 Mon Sep 17 00:00:00 2001 From: Barry Song Date: Fri, 9 May 2025 10:09:12 +1200 Subject: [PATCH 391/555] mm: userfaultfd: correct dirty flags set for both present and swap pte MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit As David pointed out, what truly matters for mremap and userfaultfd move operations is the soft dirty bit. The current comment and implementation—which always sets the dirty bit for present PTEs and fails to set the soft dirty bit for swap PTEs—are incorrect. This could break features like Checkpoint-Restore in Userspace (CRIU). This patch updates the behavior to correctly set the soft dirty bit for both present and swap PTEs in accordance with mremap. Link: https://lkml.kernel.org/r/20250508220912.7275-1-21cnbao@gmail.com Fixes: adef440691ba ("userfaultfd: UFFDIO_MOVE uABI") Signed-off-by: Barry Song Reported-by: David Hildenbrand Closes: https://lore.kernel.org/linux-mm/02f14ee1-923f-47e3-a994-4950afb9afcc@redhat.com/ Acked-by: Peter Xu Reviewed-by: Suren Baghdasaryan Cc: Lokesh Gidra Cc: Andrea Arcangeli Cc: Signed-off-by: Andrew Morton --- mm/userfaultfd.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/mm/userfaultfd.c b/mm/userfaultfd.c index 7d5d709cc838..e0db855c89b4 100644 --- a/mm/userfaultfd.c +++ b/mm/userfaultfd.c @@ -1064,8 +1064,13 @@ static int move_present_pte(struct mm_struct *mm, src_folio->index = linear_page_index(dst_vma, dst_addr); orig_dst_pte = mk_pte(&src_folio->page, dst_vma->vm_page_prot); - /* Follow mremap() behavior and treat the entry dirty after the move */ - orig_dst_pte = pte_mkwrite(pte_mkdirty(orig_dst_pte), dst_vma); + /* Set soft dirty bit so userspace can notice the pte was moved */ +#ifdef CONFIG_MEM_SOFT_DIRTY + orig_dst_pte = pte_mksoft_dirty(orig_dst_pte); +#endif + if (pte_dirty(orig_src_pte)) + orig_dst_pte = pte_mkdirty(orig_dst_pte); + orig_dst_pte = pte_mkwrite(orig_dst_pte, dst_vma); set_pte_at(mm, dst_addr, dst_pte, orig_dst_pte); out: @@ -1100,6 +1105,9 @@ static int move_swap_pte(struct mm_struct *mm, struct vm_area_struct *dst_vma, } orig_src_pte = ptep_get_and_clear(mm, src_addr, src_pte); +#ifdef CONFIG_MEM_SOFT_DIRTY + orig_src_pte = pte_swp_mksoft_dirty(orig_src_pte); +#endif set_pte_at(mm, dst_addr, dst_pte, orig_src_pte); double_pt_unlock(dst_ptl, src_ptl); From 9520a2b3f0b5e182f73410e45b9b92ea51d9b828 Mon Sep 17 00:00:00 2001 From: Sami Tolvanen Date: Mon, 7 Apr 2025 23:08:44 +0000 Subject: [PATCH 392/555] kbuild: Require pahole v1.29 with GENDWARFKSYMS on X86 With CONFIG_GENDWARFKSYMS, __gendwarfksyms_ptr variables are added to the kernel in EXPORT_SYMBOL() to ensure DWARF type information is available for exported symbols in the TUs where they're actually exported. These symbols are dropped when linking vmlinux, but dangling references to them remain in DWARF. With CONFIG_DEBUG_INFO_BTF enabled on X86, pahole versions after commit 47dcb534e253 ("btf_encoder: Stop indexing symbols for VARs") and before commit 9810758003ce ("btf_encoder: Verify 0 address DWARF variables are in ELF section") place these symbols in the .data..percpu section, which results in an "Invalid offset" error in btf_datasec_check_meta() during boot, as all the variables are at zero offset and have non-zero size. If CONFIG_DEBUG_INFO_BTF_MODULES is enabled, this also results in a failure to load modules with: failed to validate module [$module] BTF: -22 As the issue occurs in pahole v1.28 and the fix was merged after v1.29 was released, require pahole v1.29 when GENDWARFKSYMS is enabled with DEBUG_INFO_BTF on X86. Reported-by: Paolo Pisati Signed-off-by: Sami Tolvanen Signed-off-by: Masahiro Yamada --- kernel/module/Kconfig | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/kernel/module/Kconfig b/kernel/module/Kconfig index d7762ef5949a..39278737bb68 100644 --- a/kernel/module/Kconfig +++ b/kernel/module/Kconfig @@ -192,6 +192,11 @@ config GENDWARFKSYMS depends on !DEBUG_INFO_REDUCED && !DEBUG_INFO_SPLIT # Requires ELF object files. depends on !LTO + # To avoid conflicts with the discarded __gendwarfksyms_ptr symbols on + # X86, requires pahole before commit 47dcb534e253 ("btf_encoder: Stop + # indexing symbols for VARs") or after commit 9810758003ce ("btf_encoder: + # Verify 0 address DWARF variables are in ELF section"). + depends on !X86 || !DEBUG_INFO_BTF || PAHOLE_VERSION < 128 || PAHOLE_VERSION > 129 help Calculate symbol versions from DWARF debugging information using gendwarfksyms. Requires DEBUG_INFO to be enabled. From 56045757accf5a51c9146585acd5f76fa4fbdb97 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Fri, 25 Apr 2025 20:08:15 -0700 Subject: [PATCH 393/555] usr/include: openrisc: don't HDRTEST bpf_perf_event.h Since openrisc does not support PERF_EVENTS, omit the HDRTEST of bpf_perf_event.h for arch/openrisc/. Fixes a build error: usr/include/linux/bpf_perf_event.h:14:28: error: field 'regs' has incomplete type Signed-off-by: Randy Dunlap Acked-by: Stafford Horne Signed-off-by: Masahiro Yamada --- usr/include/Makefile | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/usr/include/Makefile b/usr/include/Makefile index e3d6b03527fe..f02f41941b60 100644 --- a/usr/include/Makefile +++ b/usr/include/Makefile @@ -59,6 +59,10 @@ ifeq ($(SRCARCH),arc) no-header-test += linux/bpf_perf_event.h endif +ifeq ($(SRCARCH),openrisc) +no-header-test += linux/bpf_perf_event.h +endif + ifeq ($(SRCARCH),powerpc) no-header-test += linux/bpf_perf_event.h endif From 657f96cb7c06a2ef85d166f1b055baeb6511c324 Mon Sep 17 00:00:00 2001 From: WangYuli Date: Tue, 22 Apr 2025 18:54:01 +0800 Subject: [PATCH 394/555] kbuild: deb-pkg: Add libdw-dev:native to Build-Depends-Arch The dwarf.h header, which is included by scripts/gendwarfksyms/gendwarfksyms.h, resides within the libdw-dev package. This portion of the code is compiled under the condition that CONFIG_GENDWARFKSYMS is enabled. Consequently, add libdw-dev to Build-Depends-Arch to prevent unforeseen compilation failures. Fix follow possible error: In file included from scripts/gendwarfksyms/symbols.c:6: scripts/gendwarfksyms/gendwarfksyms.h:6:10: fatal error: 'dwarf.h' file not found 6 | #include | ^~~~~~~~~ Fixes: f28568841ae0 ("tools: Add gendwarfksyms") Reviewed-by: Sami Tolvanen Signed-off-by: WangYuli Reviewed-by: Nicolas Schier Tested-by: Nicolas Schier Signed-off-by: Masahiro Yamada --- scripts/package/mkdebian | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/package/mkdebian b/scripts/package/mkdebian index 744ddba01d93..d4b007b38a47 100755 --- a/scripts/package/mkdebian +++ b/scripts/package/mkdebian @@ -210,7 +210,7 @@ Rules-Requires-Root: no Build-Depends: debhelper-compat (= 12) Build-Depends-Arch: bc, bison, flex, gcc-${host_gnu} , - kmod, libelf-dev:native, + kmod, libdw-dev:native, libelf-dev:native, libssl-dev:native, libssl-dev , python3:native, rsync Homepage: https://www.kernel.org/ From 5bd6bdd0f76e257c029ff34100f2959cd4992486 Mon Sep 17 00:00:00 2001 From: WangYuli Date: Tue, 22 Apr 2025 18:54:02 +0800 Subject: [PATCH 395/555] kbuild: rpm-pkg: Add (elfutils-devel or libdw-devel) to BuildRequires The dwarf.h header, which is included by scripts/gendwarfksyms/gendwarfksyms.h, resides within elfutils-devel or libdw-devel package. This portion of the code is compiled under the condition that CONFIG_GENDWARFKSYMS is enabled. Consequently, add (elfutils-devel or libdw-devel) to BuildRequires to prevent unforeseen compilation failures. Fix follow possible error: In file included from scripts/gendwarfksyms/cache.c:6: scripts/gendwarfksyms/gendwarfksyms.h:6:10: fatal error: 'dwarf.h' file not found 6 | #include | ^~~~~~~~~ Link: https://lore.kernel.org/all/3e52d80d-0c60-4df5-8cb5-21d4b1fce7b7@suse.com/ Fixes: f28568841ae0 ("tools: Add gendwarfksyms") Suggested-by: Petr Pavlu Signed-off-by: WangYuli Reviewed-by: Nicolas Schier Signed-off-by: Masahiro Yamada --- scripts/package/kernel.spec | 1 + 1 file changed, 1 insertion(+) diff --git a/scripts/package/kernel.spec b/scripts/package/kernel.spec index 726f34e11960..98f206cb7c60 100644 --- a/scripts/package/kernel.spec +++ b/scripts/package/kernel.spec @@ -16,6 +16,7 @@ Source1: config Source2: diff.patch Provides: kernel-%{KERNELRELEASE} BuildRequires: bc binutils bison dwarves +BuildRequires: (elfutils-devel or libdw-devel) BuildRequires: (elfutils-libelf-devel or libelf-devel) flex BuildRequires: gcc make openssl openssl-devel perl python3 rsync From d0afcfeb9e3810ec89d1ffde1a0e36621bb75dca Mon Sep 17 00:00:00 2001 From: Nathan Chancellor Date: Tue, 6 May 2025 14:02:01 -0700 Subject: [PATCH 396/555] kbuild: Disable -Wdefault-const-init-unsafe A new on by default warning in clang [1] aims to flags instances where const variables without static or thread local storage or const members in aggregate types are not initialized because it can lead to an indeterminate value. This is quite noisy for the kernel due to instances originating from header files such as: drivers/gpu/drm/i915/gt/intel_ring.h:62:2: error: default initialization of an object of type 'typeof (ring->size)' (aka 'const unsigned int') leaves the object uninitialized [-Werror,-Wdefault-const-init-var-unsafe] 62 | typecheck(typeof(ring->size), next); | ^ include/linux/typecheck.h:10:9: note: expanded from macro 'typecheck' 10 | ({ type __dummy; \ | ^ include/net/ip.h:478:14: error: default initialization of an object of type 'typeof (rt->dst.expires)' (aka 'const unsigned long') leaves the object uninitialized [-Werror,-Wdefault-const-init-var-unsafe] 478 | if (mtu && time_before(jiffies, rt->dst.expires)) | ^ include/linux/jiffies.h:138:26: note: expanded from macro 'time_before' 138 | #define time_before(a,b) time_after(b,a) | ^ include/linux/jiffies.h:128:3: note: expanded from macro 'time_after' 128 | (typecheck(unsigned long, a) && \ | ^ include/linux/typecheck.h:11:12: note: expanded from macro 'typecheck' 11 | typeof(x) __dummy2; \ | ^ include/linux/list.h:409:27: warning: default initialization of an object of type 'union (unnamed union at include/linux/list.h:409:27)' with const member leaves the object uninitialized [-Wdefault-const-init-field-unsafe] 409 | struct list_head *next = smp_load_acquire(&head->next); | ^ include/asm-generic/barrier.h:176:29: note: expanded from macro 'smp_load_acquire' 176 | #define smp_load_acquire(p) __smp_load_acquire(p) | ^ arch/arm64/include/asm/barrier.h:164:59: note: expanded from macro '__smp_load_acquire' 164 | union { __unqual_scalar_typeof(*p) __val; char __c[1]; } __u; \ | ^ include/linux/list.h:409:27: note: member '__val' declared 'const' here crypto/scatterwalk.c:66:22: error: default initialization of an object of type 'struct scatter_walk' with const member leaves the object uninitialized [-Werror,-Wdefault-const-init-field-unsafe] 66 | struct scatter_walk walk; | ^ include/crypto/algapi.h:112:15: note: member 'addr' declared 'const' here 112 | void *const addr; | ^ fs/hugetlbfs/inode.c:733:24: error: default initialization of an object of type 'struct vm_area_struct' with const member leaves the object uninitialized [-Werror,-Wdefault-const-init-field-unsafe] 733 | struct vm_area_struct pseudo_vma; | ^ include/linux/mm_types.h:803:20: note: member 'vm_flags' declared 'const' here 803 | const vm_flags_t vm_flags; | ^ Silencing the instances from typecheck.h is difficult because '= {}' is not available in older but supported compilers and '= {0}' would cause warnings about a literal 0 being treated as NULL. While it might be possible to come up with a local hack to silence the warning for clang-21+, it may not be worth it since -Wuninitialized will still trigger if an uninitialized const variable is actually used. In all audited cases of the "field" variant of the warning, the members are either not used in the particular call path, modified through other means such as memset() / memcpy() because the containing object is not const, or are within a union with other non-const members. Since this warning does not appear to have a high signal to noise ratio, just disable it. Cc: stable@vger.kernel.org Link: https://github.com/llvm/llvm-project/commit/576161cb6069e2c7656a8ef530727a0f4aefff30 [1] Reported-by: Linux Kernel Functional Testing Closes: https://lore.kernel.org/CA+G9fYuNjKcxFKS_MKPRuga32XbndkLGcY-PVuoSwzv6VWbY=w@mail.gmail.com/ Reported-by: Marcus Seyfarth Closes: https://github.com/ClangBuiltLinux/linux/issues/2088 Signed-off-by: Nathan Chancellor Signed-off-by: Masahiro Yamada --- scripts/Makefile.extrawarn | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/scripts/Makefile.extrawarn b/scripts/Makefile.extrawarn index d88acdf40855..fd649c68e198 100644 --- a/scripts/Makefile.extrawarn +++ b/scripts/Makefile.extrawarn @@ -37,6 +37,18 @@ KBUILD_CFLAGS += -Wno-gnu # https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111219 KBUILD_CFLAGS += $(call cc-disable-warning, format-overflow-non-kprintf) KBUILD_CFLAGS += $(call cc-disable-warning, format-truncation-non-kprintf) + +# Clang may emit a warning when a const variable, such as the dummy variables +# in typecheck(), or const member of an aggregate type are not initialized, +# which can result in unexpected behavior. However, in many audited cases of +# the "field" variant of the warning, this is intentional because the field is +# never used within a particular call path, the field is within a union with +# other non-const members, or the containing object is not const so the field +# can be modified via memcpy() / memset(). While the variable warning also gets +# disabled with this same switch, there should not be too much coverage lost +# because -Wuninitialized will still flag when an uninitialized const variable +# is used. +KBUILD_CFLAGS += $(call cc-disable-warning, default-const-init-unsafe) else # gcc inanely warns about local variables called 'main' From ab09da75700e9d25c7dfbc7f7934920beb5e39b9 Mon Sep 17 00:00:00 2001 From: Masahiro Yamada Date: Wed, 7 May 2025 16:49:33 +0900 Subject: [PATCH 397/555] um: let 'make clean' properly clean underlying SUBARCH as well Building the kernel with O= is affected by stale in-tree build artifacts. So, if the source tree is not clean, Kbuild displays the following: $ make ARCH=um O=build defconfig make[1]: Entering directory '/.../linux/build' *** *** The source tree is not clean, please run 'make ARCH=um mrproper' *** in /.../linux *** make[2]: *** [/.../linux/Makefile:673: outputmakefile] Error 1 make[1]: *** [/.../linux/Makefile:248: __sub-make] Error 2 make[1]: Leaving directory '/.../linux/build' make: *** [Makefile:248: __sub-make] Error 2 Usually, running 'make mrproper' is sufficient for cleaning the source tree for out-of-tree builds. However, building UML generates build artifacts not only in arch/um/, but also in the SUBARCH directory (i.e., arch/x86/). If in-tree stale files remain under arch/x86/, Kbuild will reuse them instead of creating new ones under the specified build directory. This commit makes 'make ARCH=um clean' recurse into the SUBARCH directory. Reported-by: Shuah Khan Closes: https://lore.kernel.org/lkml/20250502172459.14175-1-skhan@linuxfoundation.org/ Signed-off-by: Masahiro Yamada Acked-by: Johannes Berg Reviewed-by: David Gow Reviewed-by: Shuah Khan --- arch/um/Makefile | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/um/Makefile b/arch/um/Makefile index 1d36a613aad8..9ed792e565c9 100644 --- a/arch/um/Makefile +++ b/arch/um/Makefile @@ -154,5 +154,6 @@ MRPROPER_FILES += $(HOST_DIR)/include/generated archclean: @find . \( -name '*.bb' -o -name '*.bbg' -o -name '*.da' \ -o -name '*.gcov' \) -type f -print | xargs rm -f + $(Q)$(MAKE) -f $(srctree)/Makefile ARCH=$(HEADER_ARCH) clean export HEADER_ARCH SUBARCH USER_CFLAGS CFLAGS_NO_HARDENING DEV_NULL_PATH From d1b99cdf22e0416440265166824ebabfcb5f1afa Mon Sep 17 00:00:00 2001 From: Masahiro Yamada Date: Fri, 9 May 2025 22:23:59 +0900 Subject: [PATCH 398/555] init: remove unused CONFIG_CC_CAN_LINK_STATIC This is a leftover from commit 98e20e5e13d2 ("bpfilter: remove bpfilter"). Signed-off-by: Masahiro Yamada --- init/Kconfig | 5 ----- 1 file changed, 5 deletions(-) diff --git a/init/Kconfig b/init/Kconfig index 4cdd1049283c..bf3a920064be 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -87,11 +87,6 @@ config CC_CAN_LINK default $(success,$(srctree)/scripts/cc-can-link.sh $(CC) $(CLANG_FLAGS) $(USERCFLAGS) $(USERLDFLAGS) $(m64-flag)) if 64BIT default $(success,$(srctree)/scripts/cc-can-link.sh $(CC) $(CLANG_FLAGS) $(USERCFLAGS) $(USERLDFLAGS) $(m32-flag)) -config CC_CAN_LINK_STATIC - bool - default $(success,$(srctree)/scripts/cc-can-link.sh $(CC) $(CLANG_FLAGS) $(USERCFLAGS) $(USERLDFLAGS) $(m64-flag) -static) if 64BIT - default $(success,$(srctree)/scripts/cc-can-link.sh $(CC) $(CLANG_FLAGS) $(USERCFLAGS) $(USERLDFLAGS) $(m32-flag) -static) - # Fixed in GCC 14, 13.3, 12.4 and 11.5 # https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113921 config GCC_ASM_GOTO_OUTPUT_BROKEN From f0e4b333cf67b3d5da56bd01a125f45c102f7d27 Mon Sep 17 00:00:00 2001 From: Masahiro Yamada Date: Sun, 11 May 2025 12:55:19 +0900 Subject: [PATCH 399/555] kbuild: fix dependency on sorttable Commit ac4f06789b4f ("kbuild: Create intermediate vmlinux build with relocations preserved") missed replacing one occurrence of "vmlinux" that was added during the same development cycle. Fixes: ac4f06789b4f ("kbuild: Create intermediate vmlinux build with relocations preserved") Signed-off-by: Masahiro Yamada Acked-by: Ard Biesheuvel --- scripts/Makefile.vmlinux | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/Makefile.vmlinux b/scripts/Makefile.vmlinux index 85d60d986401..0c4bb1274a08 100644 --- a/scripts/Makefile.vmlinux +++ b/scripts/Makefile.vmlinux @@ -94,7 +94,7 @@ $(vmlinux-final): $(RESOLVE_BTFIDS) endif ifdef CONFIG_BUILDTIME_TABLE_SORT -vmlinux: scripts/sorttable +$(vmlinux-final): scripts/sorttable endif # module.builtin.ranges From 020d7f14489b8bc38c6bd4c5b3c25262e8b6de63 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Thomas=20Wei=C3=9Fschuh?= Date: Sun, 11 May 2025 08:02:27 +0200 Subject: [PATCH 400/555] Revert "kbuild: make all file references relative to source root" MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This reverts commit cacd22ce69585a91c386243cd662ada962431e63. -ffile-prefix-map breaks the ability of debuggers to find the source file corresponding to object files. As there is no simple or uniform way to specify the source directory explicitly, this breaks developers workflows. Revert the unconditional usage of -ffile-prefix-map. Reported-by: Matthieu Baerts Closes: https://lore.kernel.org/lkml/edc50aa7-0740-4942-8c15-96f12f2acc7e@kernel.org/ Reported-by: Ville Syrjälä Closes: https://lore.kernel.org/lkml/aBEttQH4kimHFScx@intel.com/ Fixes: cacd22ce6958 ("kbuild: make all file references relative to source root") Signed-off-by: Thomas Weißschuh Signed-off-by: Masahiro Yamada --- Documentation/kbuild/reproducible-builds.rst | 17 +++++++++++++++++ Makefile | 2 +- 2 files changed, 18 insertions(+), 1 deletion(-) diff --git a/Documentation/kbuild/reproducible-builds.rst b/Documentation/kbuild/reproducible-builds.rst index a7762486c93f..f2dcc39044e6 100644 --- a/Documentation/kbuild/reproducible-builds.rst +++ b/Documentation/kbuild/reproducible-builds.rst @@ -46,6 +46,21 @@ The kernel embeds the building user and host names in `KBUILD_BUILD_USER and KBUILD_BUILD_HOST`_ variables. If you are building from a git commit, you could use its committer address. +Absolute filenames +------------------ + +When the kernel is built out-of-tree, debug information may include +absolute filenames for the source files. This must be overridden by +including the ``-fdebug-prefix-map`` option in the `KCFLAGS`_ variable. + +Depending on the compiler used, the ``__FILE__`` macro may also expand +to an absolute filename in an out-of-tree build. Kbuild automatically +uses the ``-fmacro-prefix-map`` option to prevent this, if it is +supported. + +The Reproducible Builds web site has more information about these +`prefix-map options`_. + Generated files in source packages ---------------------------------- @@ -116,5 +131,7 @@ See ``scripts/setlocalversion`` for details. .. _KBUILD_BUILD_TIMESTAMP: kbuild.html#kbuild-build-timestamp .. _KBUILD_BUILD_USER and KBUILD_BUILD_HOST: kbuild.html#kbuild-build-user-kbuild-build-host +.. _KCFLAGS: kbuild.html#kcflags +.. _prefix-map options: https://reproducible-builds.org/docs/build-path/ .. _Reproducible Builds project: https://reproducible-builds.org/ .. _SOURCE_DATE_EPOCH: https://reproducible-builds.org/docs/source-date-epoch/ diff --git a/Makefile b/Makefile index 64c514f4bc19..52f84aaf1025 100644 --- a/Makefile +++ b/Makefile @@ -1068,7 +1068,7 @@ KBUILD_CFLAGS += -fno-builtin-wcslen # change __FILE__ to the relative path to the source directory ifdef building_out_of_srctree -KBUILD_CPPFLAGS += $(call cc-option,-ffile-prefix-map=$(srcroot)/=) +KBUILD_CPPFLAGS += $(call cc-option,-fmacro-prefix-map=$(srcroot)/=) KBUILD_RUSTFLAGS += --remap-path-prefix=$(srcroot)/= endif From 8cf5b3f836147d8d4e7c6eb4c01945b97dab8297 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Thomas=20Wei=C3=9Fschuh?= Date: Sun, 11 May 2025 08:02:28 +0200 Subject: [PATCH 401/555] Revert "kbuild, rust: use -fremap-path-prefix to make paths relative" MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This reverts commit dbdffaf50ff9cee3259a7cef8a7bd9e0f0ba9f13. --remap-path-prefix breaks the ability of debuggers to find the source file corresponding to object files. As there is no simple or uniform way to specify the source directory explicitly, this breaks developers workflows. Revert the unconditional usage of --remap-path-prefix, equivalent to the same change for -ffile-prefix-map in KBUILD_CPPFLAGS. Fixes: dbdffaf50ff9 ("kbuild, rust: use -fremap-path-prefix to make paths relative") Signed-off-by: Thomas Weißschuh Acked-by: Miguel Ojeda Signed-off-by: Masahiro Yamada --- Makefile | 1 - 1 file changed, 1 deletion(-) diff --git a/Makefile b/Makefile index 52f84aaf1025..0eb9e6c49b32 100644 --- a/Makefile +++ b/Makefile @@ -1069,7 +1069,6 @@ KBUILD_CFLAGS += -fno-builtin-wcslen # change __FILE__ to the relative path to the source directory ifdef building_out_of_srctree KBUILD_CPPFLAGS += $(call cc-option,-fmacro-prefix-map=$(srcroot)/=) -KBUILD_RUSTFLAGS += --remap-path-prefix=$(srcroot)/= endif # include additional Makefiles when needed From e0cd396d899805d56df91b989f8efad3a36df0da Mon Sep 17 00:00:00 2001 From: Masahiro Yamada Date: Mon, 12 May 2025 14:36:58 +0900 Subject: [PATCH 402/555] kbuild: fix typos "module.builtin" to "modules.builtin" The filenames in the comments do not match the actual generated files. Signed-off-by: Masahiro Yamada --- scripts/Makefile.vmlinux | 2 +- scripts/Makefile.vmlinux_o | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/scripts/Makefile.vmlinux b/scripts/Makefile.vmlinux index 0c4bb1274a08..b64862dc6f08 100644 --- a/scripts/Makefile.vmlinux +++ b/scripts/Makefile.vmlinux @@ -97,7 +97,7 @@ ifdef CONFIG_BUILDTIME_TABLE_SORT $(vmlinux-final): scripts/sorttable endif -# module.builtin.ranges +# modules.builtin.ranges # --------------------------------------------------------------------------- ifdef CONFIG_BUILTIN_MODULE_RANGES __default: modules.builtin.ranges diff --git a/scripts/Makefile.vmlinux_o b/scripts/Makefile.vmlinux_o index 938c7457717e..b024ffb3e201 100644 --- a/scripts/Makefile.vmlinux_o +++ b/scripts/Makefile.vmlinux_o @@ -73,7 +73,7 @@ vmlinux.o: $(initcalls-lds) vmlinux.a $(KBUILD_VMLINUX_LIBS) FORCE targets += vmlinux.o -# module.builtin.modinfo +# modules.builtin.modinfo # --------------------------------------------------------------------------- OBJCOPYFLAGS_modules.builtin.modinfo := -j .modinfo -O binary @@ -82,7 +82,7 @@ targets += modules.builtin.modinfo modules.builtin.modinfo: vmlinux.o FORCE $(call if_changed,objcopy) -# module.builtin +# modules.builtin # --------------------------------------------------------------------------- # The second line aids cases where multiple modules share the same object. From 9c1798259b9420f38f1fa1b83e3d864c3eb1a83e Mon Sep 17 00:00:00 2001 From: Fabio Estevam Date: Thu, 17 Apr 2025 07:34:58 -0300 Subject: [PATCH 403/555] drm/tiny: panel-mipi-dbi: Use drm_client_setup_with_fourcc() Since commit 559358282e5b ("drm/fb-helper: Don't use the preferred depth for the BPP default"), RGB565 displays such as the CFAF240320X no longer render correctly: colors are distorted and the content is shown twice horizontally. This regression is due to the fbdev emulation layer defaulting to 32 bits per pixel, whereas the display expects 16 bpp (RGB565). As a result, the framebuffer data is incorrectly interpreted by the panel. Fix the issue by calling drm_client_setup_with_fourcc() with a format explicitly selected based on the display's bits-per-pixel value. For 16 bpp, use DRM_FORMAT_RGB565; for other values, fall back to the previous behavior. This ensures that the allocated framebuffer format matches the hardware expectations, avoiding color and layout corruption. Tested on a CFAF240320X display with an RGB565 configuration, confirming correct colors and layout after applying this patch. Cc: stable@vger.kernel.org Fixes: 559358282e5b ("drm/fb-helper: Don't use the preferred depth for the BPP default") Signed-off-by: Fabio Estevam Reviewed-by: Thomas Zimmermann Reviewed-by: Javier Martinez Canillas Signed-off-by: Thomas Zimmermann Link: https://lore.kernel.org/r/20250417103458.2496790-1-festevam@gmail.com --- drivers/gpu/drm/tiny/panel-mipi-dbi.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/tiny/panel-mipi-dbi.c b/drivers/gpu/drm/tiny/panel-mipi-dbi.c index 0460ecaef4bd..23914a9f7fd3 100644 --- a/drivers/gpu/drm/tiny/panel-mipi-dbi.c +++ b/drivers/gpu/drm/tiny/panel-mipi-dbi.c @@ -390,7 +390,10 @@ static int panel_mipi_dbi_spi_probe(struct spi_device *spi) spi_set_drvdata(spi, drm); - drm_client_setup(drm, NULL); + if (bpp == 16) + drm_client_setup_with_fourcc(drm, DRM_FORMAT_RGB565); + else + drm_client_setup_with_fourcc(drm, DRM_FORMAT_RGB888); return 0; } From 3039ed432745f8fdf5cbb43fdc60b2e1aad624c1 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Micka=C3=ABl=20Sala=C3=BCn?= Date: Mon, 12 May 2025 11:37:30 +0200 Subject: [PATCH 404/555] landlock: Improve bit operations in audit code MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Use the BIT() and BIT_ULL() macros in the new audit code instead of explicit shifts to improve readability. Use bitmask instead of modulo operation to simplify code. Add test_range1_rand15() and test_range2_rand15() KUnit tests to improve get_id_range() coverage. Signed-off-by: Günther Noack Link: https://lore.kernel.org/r/20250512093732.1408485-1-mic@digikod.net Signed-off-by: Mickaël Salaün --- security/landlock/audit.c | 2 +- security/landlock/id.c | 33 +++++++++++++++++++++++++++++++-- security/landlock/syscalls.c | 3 ++- 3 files changed, 34 insertions(+), 4 deletions(-) diff --git a/security/landlock/audit.c b/security/landlock/audit.c index 58d5c40d4d0e..c52d079cdb77 100644 --- a/security/landlock/audit.c +++ b/security/landlock/audit.c @@ -437,7 +437,7 @@ void landlock_log_denial(const struct landlock_cred_security *const subject, return; /* Checks if the current exec was restricting itself. */ - if (subject->domain_exec & (1 << youngest_layer)) { + if (subject->domain_exec & BIT(youngest_layer)) { /* Ignores denials for the same execution. */ if (!youngest_denied->log_same_exec) return; diff --git a/security/landlock/id.c b/security/landlock/id.c index 11fab9259c15..56f7cc0fc744 100644 --- a/security/landlock/id.c +++ b/security/landlock/id.c @@ -7,6 +7,7 @@ #include #include +#include #include #include @@ -25,7 +26,7 @@ static void __init init_id(atomic64_t *const counter, const u32 random_32bits) * Ensures sure 64-bit values are always used by user space (or may * fail with -EOVERFLOW), and makes this testable. */ - init = 1ULL << 32; + init = BIT_ULL(32); /* * Makes a large (2^32) boot-time value to limit ID collision in logs @@ -105,7 +106,7 @@ static u64 get_id_range(size_t number_of_ids, atomic64_t *const counter, * to get a new ID (e.g. a full landlock_restrict_self() call), and the * cost of draining all available IDs during the system's uptime. */ - random_4bits = random_4bits % (1 << 4); + random_4bits &= 0b1111; step = number_of_ids + random_4bits; /* It is safe to cast a signed atomic to an unsigned value. */ @@ -144,6 +145,19 @@ static void test_range1_rand1(struct kunit *const test) init + 2); } +static void test_range1_rand15(struct kunit *const test) +{ + atomic64_t counter; + u64 init; + + init = get_random_u32(); + atomic64_set(&counter, init); + KUNIT_EXPECT_EQ(test, get_id_range(1, &counter, 15), init); + KUNIT_EXPECT_EQ( + test, get_id_range(get_random_u8(), &counter, get_random_u8()), + init + 16); +} + static void test_range1_rand16(struct kunit *const test) { atomic64_t counter; @@ -196,6 +210,19 @@ static void test_range2_rand2(struct kunit *const test) init + 4); } +static void test_range2_rand15(struct kunit *const test) +{ + atomic64_t counter; + u64 init; + + init = get_random_u32(); + atomic64_set(&counter, init); + KUNIT_EXPECT_EQ(test, get_id_range(2, &counter, 15), init); + KUNIT_EXPECT_EQ( + test, get_id_range(get_random_u8(), &counter, get_random_u8()), + init + 17); +} + static void test_range2_rand16(struct kunit *const test) { atomic64_t counter; @@ -232,10 +259,12 @@ static struct kunit_case __refdata test_cases[] = { KUNIT_CASE(test_init_once), KUNIT_CASE(test_range1_rand0), KUNIT_CASE(test_range1_rand1), + KUNIT_CASE(test_range1_rand15), KUNIT_CASE(test_range1_rand16), KUNIT_CASE(test_range2_rand0), KUNIT_CASE(test_range2_rand1), KUNIT_CASE(test_range2_rand2), + KUNIT_CASE(test_range2_rand15), KUNIT_CASE(test_range2_rand16), {} /* clang-format on */ diff --git a/security/landlock/syscalls.c b/security/landlock/syscalls.c index b9561e3417ae..33eafb71e4f3 100644 --- a/security/landlock/syscalls.c +++ b/security/landlock/syscalls.c @@ -9,6 +9,7 @@ #include #include +#include #include #include #include @@ -563,7 +564,7 @@ SYSCALL_DEFINE2(landlock_restrict_self, const int, ruleset_fd, const __u32, new_llcred->domain = new_dom; #ifdef CONFIG_AUDIT - new_llcred->domain_exec |= 1 << (new_dom->num_layers - 1); + new_llcred->domain_exec |= BIT(new_dom->num_layers - 1); #endif /* CONFIG_AUDIT */ return commit_creds(new_cred); From 386cd3dcfd63491619b4034b818737fc0219e128 Mon Sep 17 00:00:00 2001 From: Alexey Makhalov Date: Tue, 18 Mar 2025 00:40:31 +0000 Subject: [PATCH 405/555] MAINTAINERS: Update Alexey Makhalov's email address Fix a typo in an email address. Closes: https://lore.kernel.org/all/20240925-rational-succinct-vulture-cca9fb@lemur/T/ Reported-by: Konstantin Ryabitsev Reported-by: Juergen Gross Signed-off-by: Alexey Makhalov Signed-off-by: Borislav Petkov (AMD) Cc: stable@vger.kernel.org Link: https://lore.kernel.org/20250318004031.2703923-1-alexey.makhalov@broadcom.com --- MAINTAINERS | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index 69511c3b2b76..0d3a252db2b6 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -18373,7 +18373,7 @@ F: include/uapi/linux/ppdev.h PARAVIRT_OPS INTERFACE M: Juergen Gross R: Ajay Kaher -R: Alexey Makhalov +R: Alexey Makhalov R: Broadcom internal kernel review list L: virtualization@lists.linux.dev L: x86@kernel.org @@ -25859,7 +25859,7 @@ F: drivers/misc/vmw_balloon.c VMWARE HYPERVISOR INTERFACE M: Ajay Kaher -M: Alexey Makhalov +M: Alexey Makhalov R: Broadcom internal kernel review list L: virtualization@lists.linux.dev L: x86@kernel.org @@ -25887,7 +25887,7 @@ F: drivers/scsi/vmw_pvscsi.h VMWARE VIRTUAL PTP CLOCK DRIVER M: Nick Shi R: Ajay Kaher -R: Alexey Makhalov +R: Alexey Makhalov R: Broadcom internal kernel review list L: netdev@vger.kernel.org S: Supported From 2f0c6fceae0df59a9c0edb5d41529b90ee9f2fcb Mon Sep 17 00:00:00 2001 From: Markus Burri Date: Thu, 8 May 2025 15:06:08 +0200 Subject: [PATCH 406/555] accel/ivpu: Use effective buffer size for zero terminator Use the effective written size instead of original size as index for zero termination. If the input from user-space is to larger and the input is truncated, the original size is out-of-bound. Since there is an upfront size check here, the change is for consistency. Signed-off-by: Markus Burri Reviewed-by: Jacek Lawrynowicz Signed-off-by: Jacek Lawrynowicz Link: https://lore.kernel.org/r/20250508130612.82270-3-markus.burri@mt.com --- drivers/accel/ivpu/ivpu_debugfs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/accel/ivpu/ivpu_debugfs.c b/drivers/accel/ivpu/ivpu_debugfs.c index f0dad0c9ce33..cd24ccd20ba6 100644 --- a/drivers/accel/ivpu/ivpu_debugfs.c +++ b/drivers/accel/ivpu/ivpu_debugfs.c @@ -455,7 +455,7 @@ priority_bands_fops_write(struct file *file, const char __user *user_buf, size_t if (ret < 0) return ret; - buf[size] = '\0'; + buf[ret] = '\0'; ret = sscanf(buf, "%u %u %u %u", &band, &grace_period, &process_grace_period, &process_quantum); if (ret != 4) From 8098514bd5ca98beca6ec725751d82d0d5b492d8 Mon Sep 17 00:00:00 2001 From: Keith Busch Date: Fri, 9 May 2025 08:38:02 -0700 Subject: [PATCH 407/555] block: always allocate integrity buffer when required Many nvme metadata formats can not strip or generate the metadata on the controller side. For these formats, a host provided integrity buffer is mandatory even if it isn't checked. The block integrity read_verify and write_generate attributes prevent allocating the metadata buffer, but we need it when the format requires it, otherwise reads and writes will be rejected by the driver with IO errors. Assume the integrity buffer can be offloaded to the controller if the metadata size is the same as the protection information size. Otherwise provide an unchecked host buffer when the read verify or write generation attributes are disabled. This fixes the following nvme warning: ------------[ cut here ]------------ WARNING: CPU: 1 PID: 371 at drivers/nvme/host/core.c:1036 nvme_setup_rw+0x122/0x210 ... RIP: 0010:nvme_setup_rw+0x122/0x210 ... Call Trace: nvme_setup_cmd+0x1b4/0x280 nvme_queue_rqs+0xc4/0x1f0 [nvme] blk_mq_dispatch_queue_requests+0x24a/0x430 blk_mq_flush_plug_list+0x50/0x140 __blk_flush_plug+0xc1/0x100 __submit_bio+0x1c1/0x360 ? submit_bio_noacct_nocheck+0x2d6/0x3c0 submit_bio_noacct_nocheck+0x2d6/0x3c0 ? submit_bio_noacct+0x47/0x4c0 submit_bio_wait+0x48/0xa0 __blkdev_direct_IO_simple+0xee/0x210 ? current_time+0x1d/0x100 ? current_time+0x1d/0x100 ? __bio_clone+0xb0/0xb0 blkdev_read_iter+0xbb/0x140 vfs_read+0x239/0x310 ksys_read+0x58/0xc0 do_syscall_64+0x6c/0x180 entry_SYSCALL_64_after_hwframe+0x4b/0x53 Signed-off-by: Keith Busch Reviewed-by: Martin K. Petersen Reviewed-by: Christoph Hellwig Link: https://lore.kernel.org/r/20250509153802.3482493-1-kbusch@meta.com Signed-off-by: Jens Axboe --- block/bio-integrity-auto.c | 62 +++++++++++++++++++++++++++++--------- 1 file changed, 47 insertions(+), 15 deletions(-) diff --git a/block/bio-integrity-auto.c b/block/bio-integrity-auto.c index e524c609be50..9c6657664792 100644 --- a/block/bio-integrity-auto.c +++ b/block/bio-integrity-auto.c @@ -9,6 +9,7 @@ * not aware of PI. */ #include +#include #include #include "blk.h" @@ -43,6 +44,29 @@ static void bio_integrity_verify_fn(struct work_struct *work) bio_endio(bio); } +#define BIP_CHECK_FLAGS (BIP_CHECK_GUARD | BIP_CHECK_REFTAG | BIP_CHECK_APPTAG) +static bool bip_should_check(struct bio_integrity_payload *bip) +{ + return bip->bip_flags & BIP_CHECK_FLAGS; +} + +static bool bi_offload_capable(struct blk_integrity *bi) +{ + switch (bi->csum_type) { + case BLK_INTEGRITY_CSUM_CRC64: + return bi->tuple_size == sizeof(struct crc64_pi_tuple); + case BLK_INTEGRITY_CSUM_CRC: + case BLK_INTEGRITY_CSUM_IP: + return bi->tuple_size == sizeof(struct t10_pi_tuple); + default: + pr_warn_once("%s: unknown integrity checksum type:%d\n", + __func__, bi->csum_type); + fallthrough; + case BLK_INTEGRITY_CSUM_NONE: + return false; + } +} + /** * __bio_integrity_endio - Integrity I/O completion function * @bio: Protected bio @@ -54,12 +78,12 @@ static void bio_integrity_verify_fn(struct work_struct *work) */ bool __bio_integrity_endio(struct bio *bio) { - struct blk_integrity *bi = blk_get_integrity(bio->bi_bdev->bd_disk); struct bio_integrity_payload *bip = bio_integrity(bio); struct bio_integrity_data *bid = container_of(bip, struct bio_integrity_data, bip); - if (bio_op(bio) == REQ_OP_READ && !bio->bi_status && bi->csum_type) { + if (bio_op(bio) == REQ_OP_READ && !bio->bi_status && + bip_should_check(bip)) { INIT_WORK(&bid->work, bio_integrity_verify_fn); queue_work(kintegrityd_wq, &bid->work); return false; @@ -84,6 +108,7 @@ bool bio_integrity_prep(struct bio *bio) { struct blk_integrity *bi = blk_get_integrity(bio->bi_bdev->bd_disk); struct bio_integrity_data *bid; + bool set_flags = true; gfp_t gfp = GFP_NOIO; unsigned int len; void *buf; @@ -100,19 +125,24 @@ bool bio_integrity_prep(struct bio *bio) switch (bio_op(bio)) { case REQ_OP_READ: - if (bi->flags & BLK_INTEGRITY_NOVERIFY) - return true; + if (bi->flags & BLK_INTEGRITY_NOVERIFY) { + if (bi_offload_capable(bi)) + return true; + set_flags = false; + } break; case REQ_OP_WRITE: - if (bi->flags & BLK_INTEGRITY_NOGENERATE) - return true; - /* * Zero the memory allocated to not leak uninitialized kernel * memory to disk for non-integrity metadata where nothing else * initializes the memory. */ - if (bi->csum_type == BLK_INTEGRITY_CSUM_NONE) + if (bi->flags & BLK_INTEGRITY_NOGENERATE) { + if (bi_offload_capable(bi)) + return true; + set_flags = false; + gfp |= __GFP_ZERO; + } else if (bi->csum_type == BLK_INTEGRITY_CSUM_NONE) gfp |= __GFP_ZERO; break; default: @@ -137,19 +167,21 @@ bool bio_integrity_prep(struct bio *bio) bid->bip.bip_flags |= BIP_BLOCK_INTEGRITY; bip_set_seed(&bid->bip, bio->bi_iter.bi_sector); - if (bi->csum_type == BLK_INTEGRITY_CSUM_IP) - bid->bip.bip_flags |= BIP_IP_CHECKSUM; - if (bi->csum_type) - bid->bip.bip_flags |= BIP_CHECK_GUARD; - if (bi->flags & BLK_INTEGRITY_REF_TAG) - bid->bip.bip_flags |= BIP_CHECK_REFTAG; + if (set_flags) { + if (bi->csum_type == BLK_INTEGRITY_CSUM_IP) + bid->bip.bip_flags |= BIP_IP_CHECKSUM; + if (bi->csum_type) + bid->bip.bip_flags |= BIP_CHECK_GUARD; + if (bi->flags & BLK_INTEGRITY_REF_TAG) + bid->bip.bip_flags |= BIP_CHECK_REFTAG; + } if (bio_integrity_add_page(bio, virt_to_page(buf), len, offset_in_page(buf)) < len) goto err_end_io; /* Auto-generate integrity metadata if this is a write */ - if (bio_data_dir(bio) == WRITE) + if (bio_data_dir(bio) == WRITE && bip_should_check(&bid->bip)) blk_integrity_generate(bio); else bid->saved_bio_iter = bio->bi_iter; From 63166b815dc163b2e46426cecf707dc5923d6d13 Mon Sep 17 00:00:00 2001 From: hexue Date: Mon, 12 May 2025 13:20:25 +0800 Subject: [PATCH 408/555] io_uring/uring_cmd: fix hybrid polling initialization issue Modify the check for whether the timer is initialized during IO transfer when passthrough is used with hybrid polling, to ensure that it's always setup correctly. Cc: stable@vger.kernel.org Fixes: 01ee194d1aba ("io_uring: add support for hybrid IOPOLL") Signed-off-by: hexue Link: https://lore.kernel.org/r/20250512052025.293031-1-xue01.he@samsung.com Signed-off-by: Jens Axboe --- io_uring/uring_cmd.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/io_uring/uring_cmd.c b/io_uring/uring_cmd.c index a9ea7d29cdd9..430ed620ddfe 100644 --- a/io_uring/uring_cmd.c +++ b/io_uring/uring_cmd.c @@ -254,6 +254,11 @@ int io_uring_cmd(struct io_kiocb *req, unsigned int issue_flags) return -EOPNOTSUPP; issue_flags |= IO_URING_F_IOPOLL; req->iopoll_completed = 0; + if (ctx->flags & IORING_SETUP_HYBRID_IOPOLL) { + /* make sure every req only blocks once */ + req->flags &= ~REQ_F_IOPOLL_STATE; + req->iopoll_start = ktime_get_ns(); + } } ret = file->f_op->uring_cmd(ioucmd, issue_flags); From f446c6311e86618a1f81eb576b56a6266307238f Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Mon, 12 May 2025 09:06:06 -0600 Subject: [PATCH 409/555] io_uring/memmap: don't use page_address() on a highmem page For older/32-bit systems with highmem, don't assume that the pages in a mapped region are always going to be mapped. If io_region_init_ptr() finds that the pages are coalescable, also check if the first page is a HighMem page or not. If it is, fall through to the usual vmap() mapping rather than attempt to get the unmapped page address. Cc: stable@vger.kernel.org Fixes: c4d0ac1c1567 ("io_uring/memmap: optimise single folio regions") Link: https://lore.kernel.org/all/681fe2fb.050a0220.f2294.001a.GAE@google.com/ Reported-by: syzbot+5b8c4abafcb1d791ccfc@syzkaller.appspotmail.com Link: https://lore.kernel.org/all/681fed0a.050a0220.f2294.001c.GAE@google.com/ Reported-by: syzbot+6456a99dfdc2e78c4feb@syzkaller.appspotmail.com Tested-by: syzbot+6456a99dfdc2e78c4feb@syzkaller.appspotmail.com Signed-off-by: Jens Axboe --- io_uring/memmap.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/io_uring/memmap.c b/io_uring/memmap.c index 76fcc79656b0..07f8a5cbd37e 100644 --- a/io_uring/memmap.c +++ b/io_uring/memmap.c @@ -116,7 +116,7 @@ static int io_region_init_ptr(struct io_mapped_region *mr) void *ptr; if (io_check_coalesce_buffer(mr->pages, mr->nr_pages, &ifd)) { - if (ifd.nr_folios == 1) { + if (ifd.nr_folios == 1 && !PageHighMem(mr->pages[0])) { mr->ptr = page_address(mr->pages[0]); return 0; } From 579cb52ecd46ce0351fc3d40882ebdb60332a0bc Mon Sep 17 00:00:00 2001 From: Wolfram Sang Date: Thu, 24 Apr 2025 23:22:23 +0200 Subject: [PATCH 410/555] MAINTAINERS: add me as maintainer for the gpio sloppy logic analyzer This was forgotten when the analyzer went upstream. Signed-off-by: Wolfram Sang Link: https://lore.kernel.org/r/20250424212234.5313-2-wsa+renesas@sang-engineering.com Signed-off-by: Bartosz Golaszewski --- MAINTAINERS | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index f21f1dabb5fe..a634c1c7188c 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -10147,6 +10147,13 @@ F: drivers/gpio/gpio-regmap.c F: include/linux/gpio/regmap.h K: (devm_)?gpio_regmap_(un)?register +GPIO SLOPPY LOGIC ANALYZER +M: Wolfram Sang +S: Supported +F: Documentation/dev-tools/gpio-sloppy-logic-analyzer.rst +F: drivers/gpio/gpio-sloppy-logic-analyzer.c +F: tools/gpio/gpio-sloppy-logic-analyzer.sh + GPIO SUBSYSTEM M: Linus Walleij M: Bartosz Golaszewski From 1b0c192c92ea1fe2dcb178f84adf15fe37c3e7c8 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Fri, 9 May 2025 15:26:57 -0400 Subject: [PATCH 411/555] tracing: samples: Initialize trace_array_printk() with the correct function When using trace_array_printk() on a created instance, the correct function to use to initialize it is: trace_array_init_printk() Not trace_printk_init_buffer() The former is a proper function to use, the latter is for initializing trace_printk() and causes the NOTICE banner to be displayed. Cc: stable@vger.kernel.org Cc: Masami Hiramatsu Cc: Mathieu Desnoyers Cc: Divya Indi Link: https://lore.kernel.org/20250509152657.0f6744d9@gandalf.local.home Fixes: 89ed42495ef4a ("tracing: Sample module to demonstrate kernel access to Ftrace instances.") Fixes: 38ce2a9e33db6 ("tracing: Add trace_array_init_printk() to initialize instance trace_printk() buffers") Signed-off-by: Steven Rostedt (Google) --- samples/ftrace/sample-trace-array.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/samples/ftrace/sample-trace-array.c b/samples/ftrace/sample-trace-array.c index dac67c367457..4147616102f9 100644 --- a/samples/ftrace/sample-trace-array.c +++ b/samples/ftrace/sample-trace-array.c @@ -112,7 +112,7 @@ static int __init sample_trace_array_init(void) /* * If context specific per-cpu buffers havent already been allocated. */ - trace_printk_init_buffers(); + trace_array_init_printk(tr); simple_tsk = kthread_run(simple_thread, NULL, "sample-instance"); if (IS_ERR(simple_tsk)) { From 54db6d1bdd71fa90172a2a6aca3308bbf7fa7eb5 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Mon, 5 May 2025 16:03:16 +0100 Subject: [PATCH 412/555] btrfs: fix discard worker infinite loop after disabling discard If the discard worker is running and there's currently only one block group, that block group is a data block group, it's in the unused block groups discard list and is being used (it got an extent allocated from it after becoming unused), the worker can end up in an infinite loop if a transaction abort happens or the async discard is disabled (during remount or unmount for example). This happens like this: 1) Task A, the discard worker, is at peek_discard_list() and find_next_block_group() returns block group X; 2) Block group X is in the unused block groups discard list (its discard index is BTRFS_DISCARD_INDEX_UNUSED) since at some point in the past it become an unused block group and was added to that list, but then later it got an extent allocated from it, so its ->used counter is not zero anymore; 3) The current transaction is aborted by task B and we end up at __btrfs_handle_fs_error() in the transaction abort path, where we call btrfs_discard_stop(), which clears BTRFS_FS_DISCARD_RUNNING from fs_info, and then at __btrfs_handle_fs_error() we set the fs to RO mode (setting SB_RDONLY in the super block's s_flags field); 4) Task A calls __add_to_discard_list() with the goal of moving the block group from the unused block groups discard list into another discard list, but at __add_to_discard_list() we end up doing nothing because btrfs_run_discard_work() returns false, since the super block has SB_RDONLY set in its flags and BTRFS_FS_DISCARD_RUNNING is not set anymore in fs_info->flags. So block group X remains in the unused block groups discard list; 5) Task A then does a goto into the 'again' label, calls find_next_block_group() again we gets block group X again. Then it repeats the previous steps over and over since there are not other block groups in the discard lists and block group X is never moved out of the unused block groups discard list since btrfs_run_discard_work() keeps returning false and therefore __add_to_discard_list() doesn't move block group X out of that discard list. When this happens we can get a soft lockup report like this: [71.957] watchdog: BUG: soft lockup - CPU#0 stuck for 27s! [kworker/u4:3:97] [71.957] Modules linked in: xfs af_packet rfkill (...) [71.957] CPU: 0 UID: 0 PID: 97 Comm: kworker/u4:3 Tainted: G W 6.14.2-1-default #1 openSUSE Tumbleweed 968795ef2b1407352128b466fe887416c33af6fa [71.957] Tainted: [W]=WARN [71.957] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.2-3-gd478f380-rebuilt.opensuse.org 04/01/2014 [71.957] Workqueue: btrfs_discard btrfs_discard_workfn [btrfs] [71.957] RIP: 0010:btrfs_discard_workfn+0xc4/0x400 [btrfs] [71.957] Code: c1 01 48 83 (...) [71.957] RSP: 0018:ffffafaec03efe08 EFLAGS: 00000246 [71.957] RAX: ffff897045500000 RBX: ffff8970413ed8d0 RCX: 0000000000000000 [71.957] RDX: 0000000000000001 RSI: ffff8970413ed8d0 RDI: 0000000a8f1272ad [71.957] RBP: 0000000a9d61c60e R08: ffff897045500140 R09: 8080808080808080 [71.957] R10: ffff897040276800 R11: fefefefefefefeff R12: ffff8970413ed860 [71.957] R13: ffff897045500000 R14: ffff8970413ed868 R15: 0000000000000000 [71.957] FS: 0000000000000000(0000) GS:ffff89707bc00000(0000) knlGS:0000000000000000 [71.957] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [71.957] CR2: 00005605bcc8d2f0 CR3: 000000010376a001 CR4: 0000000000770ef0 [71.957] PKRU: 55555554 [71.957] Call Trace: [71.957] [71.957] process_one_work+0x17e/0x330 [71.957] worker_thread+0x2ce/0x3f0 [71.957] ? __pfx_worker_thread+0x10/0x10 [71.957] kthread+0xef/0x220 [71.957] ? __pfx_kthread+0x10/0x10 [71.957] ret_from_fork+0x34/0x50 [71.957] ? __pfx_kthread+0x10/0x10 [71.957] ret_from_fork_asm+0x1a/0x30 [71.957] [71.957] Kernel panic - not syncing: softlockup: hung tasks [71.987] CPU: 0 UID: 0 PID: 97 Comm: kworker/u4:3 Tainted: G W L 6.14.2-1-default #1 openSUSE Tumbleweed 968795ef2b1407352128b466fe887416c33af6fa [71.989] Tainted: [W]=WARN, [L]=SOFTLOCKUP [71.989] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.2-3-gd478f380-rebuilt.opensuse.org 04/01/2014 [71.991] Workqueue: btrfs_discard btrfs_discard_workfn [btrfs] [71.992] Call Trace: [71.993] [71.994] dump_stack_lvl+0x5a/0x80 [71.994] panic+0x10b/0x2da [71.995] watchdog_timer_fn.cold+0x9a/0xa1 [71.996] ? __pfx_watchdog_timer_fn+0x10/0x10 [71.997] __hrtimer_run_queues+0x132/0x2a0 [71.997] hrtimer_interrupt+0xff/0x230 [71.998] __sysvec_apic_timer_interrupt+0x55/0x100 [71.999] sysvec_apic_timer_interrupt+0x6c/0x90 [72.000] [72.000] [72.001] asm_sysvec_apic_timer_interrupt+0x1a/0x20 [72.002] RIP: 0010:btrfs_discard_workfn+0xc4/0x400 [btrfs] [72.002] Code: c1 01 48 83 (...) [72.005] RSP: 0018:ffffafaec03efe08 EFLAGS: 00000246 [72.006] RAX: ffff897045500000 RBX: ffff8970413ed8d0 RCX: 0000000000000000 [72.006] RDX: 0000000000000001 RSI: ffff8970413ed8d0 RDI: 0000000a8f1272ad [72.007] RBP: 0000000a9d61c60e R08: ffff897045500140 R09: 8080808080808080 [72.008] R10: ffff897040276800 R11: fefefefefefefeff R12: ffff8970413ed860 [72.009] R13: ffff897045500000 R14: ffff8970413ed868 R15: 0000000000000000 [72.010] ? btrfs_discard_workfn+0x51/0x400 [btrfs 23b01089228eb964071fb7ca156eee8cd3bf996f] [72.011] process_one_work+0x17e/0x330 [72.012] worker_thread+0x2ce/0x3f0 [72.013] ? __pfx_worker_thread+0x10/0x10 [72.014] kthread+0xef/0x220 [72.014] ? __pfx_kthread+0x10/0x10 [72.015] ret_from_fork+0x34/0x50 [72.015] ? __pfx_kthread+0x10/0x10 [72.016] ret_from_fork_asm+0x1a/0x30 [72.017] [72.017] Kernel Offset: 0x15000000 from 0xffffffff81000000 (relocation range: 0xffffffff80000000-0xffffffffbfffffff) [72.019] Rebooting in 90 seconds.. So fix this by making sure we move a block group out of the unused block groups discard list when calling __add_to_discard_list(). Fixes: 2bee7eb8bb81 ("btrfs: discard one region at a time in async discard") Link: https://bugzilla.suse.com/show_bug.cgi?id=1242012 CC: stable@vger.kernel.org # 5.10+ Reviewed-by: Boris Burkov Reviewed-by: Daniel Vacek Signed-off-by: Filipe Manana Signed-off-by: David Sterba --- fs/btrfs/discard.c | 17 +++++++++++++++-- 1 file changed, 15 insertions(+), 2 deletions(-) diff --git a/fs/btrfs/discard.c b/fs/btrfs/discard.c index d6eef4bd9e9d..de23c4b3515e 100644 --- a/fs/btrfs/discard.c +++ b/fs/btrfs/discard.c @@ -94,8 +94,6 @@ static void __add_to_discard_list(struct btrfs_discard_ctl *discard_ctl, struct btrfs_block_group *block_group) { lockdep_assert_held(&discard_ctl->lock); - if (!btrfs_run_discard_work(discard_ctl)) - return; if (list_empty(&block_group->discard_list) || block_group->discard_index == BTRFS_DISCARD_INDEX_UNUSED) { @@ -118,6 +116,9 @@ static void add_to_discard_list(struct btrfs_discard_ctl *discard_ctl, if (!btrfs_is_block_group_data_only(block_group)) return; + if (!btrfs_run_discard_work(discard_ctl)) + return; + spin_lock(&discard_ctl->lock); __add_to_discard_list(discard_ctl, block_group); spin_unlock(&discard_ctl->lock); @@ -244,6 +245,18 @@ again: block_group->used != 0) { if (btrfs_is_block_group_data_only(block_group)) { __add_to_discard_list(discard_ctl, block_group); + /* + * The block group must have been moved to other + * discard list even if discard was disabled in + * the meantime or a transaction abort happened, + * otherwise we can end up in an infinite loop, + * always jumping into the 'again' label and + * keep getting this block group over and over + * in case there are no other block groups in + * the discard lists. + */ + ASSERT(block_group->discard_index != + BTRFS_DISCARD_INDEX_UNUSED); } else { list_del_init(&block_group->discard_list); btrfs_put_block_group(block_group); From a0fd1c6098633f9a95fc2f636383546c82b704c3 Mon Sep 17 00:00:00 2001 From: Boris Burkov Date: Wed, 7 May 2025 12:42:24 -0700 Subject: [PATCH 413/555] btrfs: fix folio leak in submit_one_async_extent() If btrfs_reserve_extent() fails while submitting an async_extent for a compressed write, then we fail to call free_async_extent_pages() on the async_extent and leak its folios. A likely cause for such a failure would be btrfs_reserve_extent() failing to find a large enough contiguous free extent for the compressed extent. I was able to reproduce this by: 1. mount with compress-force=zstd:3 2. fallocating most of a filesystem to a big file 3. fragmenting the remaining free space 4. trying to copy in a file which zstd would generate large compressed extents for (vmlinux worked well for this) Step 4. hits the memory leak and can be repeated ad nauseam to eventually exhaust the system memory. Fix this by detecting the case where we fallback to uncompressed submission for a compressed async_extent and ensuring that we call free_async_extent_pages(). Fixes: 131a821a243f ("btrfs: fallback if compressed IO fails for ENOSPC") CC: stable@vger.kernel.org # 6.1+ Reviewed-by: Filipe Manana Co-developed-by: Josef Bacik Signed-off-by: Boris Burkov Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/inode.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index d295a37fa049..c1bd17915f81 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1109,6 +1109,7 @@ static void submit_one_async_extent(struct async_chunk *async_chunk, struct extent_state *cached = NULL; struct extent_map *em; int ret = 0; + bool free_pages = false; u64 start = async_extent->start; u64 end = async_extent->start + async_extent->ram_size - 1; @@ -1129,7 +1130,10 @@ static void submit_one_async_extent(struct async_chunk *async_chunk, } if (async_extent->compress_type == BTRFS_COMPRESS_NONE) { + ASSERT(!async_extent->folios); + ASSERT(async_extent->nr_folios == 0); submit_uncompressed_range(inode, async_extent, locked_folio); + free_pages = true; goto done; } @@ -1145,6 +1149,7 @@ static void submit_one_async_extent(struct async_chunk *async_chunk, * fall back to uncompressed. */ submit_uncompressed_range(inode, async_extent, locked_folio); + free_pages = true; goto done; } @@ -1186,6 +1191,8 @@ static void submit_one_async_extent(struct async_chunk *async_chunk, done: if (async_chunk->blkcg_css) kthread_associate_blkcg(NULL); + if (free_pages) + free_async_extent_pages(async_extent); kfree(async_extent); return; From 4ce2affc6ef9f84b4aebbf18bd5c57397b6024eb Mon Sep 17 00:00:00 2001 From: Kyoji Ogasawara Date: Fri, 9 May 2025 19:26:31 +0900 Subject: [PATCH 414/555] btrfs: add back warning for mount option commit values exceeding 300 The Btrfs documentation states that if the commit value is greater than 300 a warning should be issued. The warning was accidentally lost in the new mount API update. Fixes: 6941823cc878 ("btrfs: remove old mount API code") CC: stable@vger.kernel.org # 6.12+ Reviewed-by: Qu Wenruo Reviewed-by: Anand Jain Signed-off-by: Kyoji Ogasawara Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/fs.h | 1 + fs/btrfs/super.c | 4 ++++ 2 files changed, 5 insertions(+) diff --git a/fs/btrfs/fs.h b/fs/btrfs/fs.h index bcca43046064..7baa2ed45198 100644 --- a/fs/btrfs/fs.h +++ b/fs/btrfs/fs.h @@ -300,6 +300,7 @@ enum { #define BTRFS_FEATURE_INCOMPAT_SAFE_CLEAR 0ULL #define BTRFS_DEFAULT_COMMIT_INTERVAL (30) +#define BTRFS_WARNING_COMMIT_INTERVAL (300) #define BTRFS_DEFAULT_MAX_INLINE (2048) struct btrfs_dev_replace { diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index 7121d8c7a318..7310e2fa8526 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -569,6 +569,10 @@ static int btrfs_parse_param(struct fs_context *fc, struct fs_parameter *param) break; case Opt_commit_interval: ctx->commit_interval = result.uint_32; + if (ctx->commit_interval > BTRFS_WARNING_COMMIT_INTERVAL) { + btrfs_warn(NULL, "excessive commit interval %u, use with care", + ctx->commit_interval); + } if (ctx->commit_interval == 0) ctx->commit_interval = BTRFS_DEFAULT_COMMIT_INTERVAL; break; From c92d6089d8ad7d4d815ebcedee3f3907b539ff1f Mon Sep 17 00:00:00 2001 From: Mathieu Othacehe Date: Fri, 9 May 2025 14:19:35 +0200 Subject: [PATCH 415/555] net: cadence: macb: Fix a possible deadlock in macb_halt_tx. There is a situation where after THALT is set high, TGO stays high as well. Because jiffies are never updated, as we are in a context with interrupts disabled, we never exit that loop and have a deadlock. That deadlock was noticed on a sama5d4 device that stayed locked for days. Use retries instead of jiffies so that the timeout really works and we do not have a deadlock anymore. Fixes: e86cd53afc590 ("net/macb: better manage tx errors") Signed-off-by: Mathieu Othacehe Reviewed-by: Simon Horman Link: https://patch.msgid.link/20250509121935.16282-1-othacehe@gnu.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/cadence/macb_main.c | 19 ++++++------------- 1 file changed, 6 insertions(+), 13 deletions(-) diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index 1fe8ec37491b..e1e8bd2ec155 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -997,22 +997,15 @@ static void macb_update_stats(struct macb *bp) static int macb_halt_tx(struct macb *bp) { - unsigned long halt_time, timeout; - u32 status; + u32 status; macb_writel(bp, NCR, macb_readl(bp, NCR) | MACB_BIT(THALT)); - timeout = jiffies + usecs_to_jiffies(MACB_HALT_TIMEOUT); - do { - halt_time = jiffies; - status = macb_readl(bp, TSR); - if (!(status & MACB_BIT(TGO))) - return 0; - - udelay(250); - } while (time_before(halt_time, timeout)); - - return -ETIMEDOUT; + /* Poll TSR until TGO is cleared or timeout. */ + return read_poll_timeout_atomic(macb_readl, status, + !(status & MACB_BIT(TGO)), + 250, MACB_HALT_TIMEOUT, false, + bp, TSR); } static void macb_tx_unmap(struct macb *bp, struct macb_tx_skb *tx_skb, int budget) From af5f54b0ef9ef72d5fc7d57f0406372e8f317958 Mon Sep 17 00:00:00 2001 From: Cosmin Ratiu Date: Fri, 9 May 2025 10:28:50 +0300 Subject: [PATCH 416/555] net: Lock lower level devices when updating features __netdev_update_features() expects the netdevice to be ops-locked, but it gets called recursively on the lower level netdevices to sync their features, and nothing locks those. This commit fixes that, with the assumption that it shouldn't be possible for both higher-level and lover-level netdevices to require the instance lock, because that would lead to lock dependency warnings. Without this, playing with higher level (e.g. vxlan) netdevices on top of netdevices with instance locking enabled can run into issues: WARNING: CPU: 59 PID: 206496 at ./include/net/netdev_lock.h:17 netif_napi_add_weight_locked+0x753/0xa60 [...] Call Trace: mlx5e_open_channel+0xc09/0x3740 [mlx5_core] mlx5e_open_channels+0x1f0/0x770 [mlx5_core] mlx5e_safe_switch_params+0x1b5/0x2e0 [mlx5_core] set_feature_lro+0x1c2/0x330 [mlx5_core] mlx5e_handle_feature+0xc8/0x140 [mlx5_core] mlx5e_set_features+0x233/0x2e0 [mlx5_core] __netdev_update_features+0x5be/0x1670 __netdev_update_features+0x71f/0x1670 dev_ethtool+0x21c5/0x4aa0 dev_ioctl+0x438/0xae0 sock_ioctl+0x2ba/0x690 __x64_sys_ioctl+0xa78/0x1700 do_syscall_64+0x6d/0x140 entry_SYSCALL_64_after_hwframe+0x4b/0x53 Fixes: 7e4d784f5810 ("net: hold netdev instance lock during rtnetlink operations") Signed-off-by: Cosmin Ratiu Acked-by: Stanislav Fomichev Link: https://patch.msgid.link/20250509072850.2002821-1-cratiu@nvidia.com Signed-off-by: Jakub Kicinski --- net/core/dev.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/core/dev.c b/net/core/dev.c index 11da1e272ec2..0d891634c692 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -10441,6 +10441,7 @@ static void netdev_sync_lower_features(struct net_device *upper, if (!(features & feature) && (lower->features & feature)) { netdev_dbg(upper, "Disabling feature %pNF on lower dev %s.\n", &feature, lower->name); + netdev_lock_ops(lower); lower->wanted_features &= ~feature; __netdev_update_features(lower); @@ -10449,6 +10450,7 @@ static void netdev_sync_lower_features(struct net_device *upper, &feature, lower->name); else netdev_features_change(lower); + netdev_unlock_ops(lower); } } } From 498625a8ab2c8e1c9ab5105744310e8d6952cc01 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Fri, 9 May 2025 14:38:16 +0300 Subject: [PATCH 417/555] net: dsa: sja1105: discard incoming frames in BR_STATE_LISTENING It has been reported that when under a bridge with stp_state=1, the logs get spammed with this message: [ 251.734607] fsl_dpaa2_eth dpni.5 eth0: Couldn't decode source port Further debugging shows the following info associated with packets: source_port=-1, switch_id=-1, vid=-1, vbid=1 In other words, they are data plane packets which are supposed to be decoded by dsa_tag_8021q_find_port_by_vbid(), but the latter (correctly) refuses to do so, because no switch port is currently in BR_STATE_LEARNING or BR_STATE_FORWARDING - so the packet is effectively unexpected. The error goes away after the port progresses to BR_STATE_LEARNING in 15 seconds (the default forward_time of the bridge), because then, dsa_tag_8021q_find_port_by_vbid() can correctly associate the data plane packets with a plausible bridge port in a plausible STP state. Re-reading IEEE 802.1D-1990, I see the following: "4.4.2 Learning: (...) The Forwarding Process shall discard received frames." IEEE 802.1D-2004 further clarifies: "DISABLED, BLOCKING, LISTENING, and BROKEN all correspond to the DISCARDING port state. While those dot1dStpPortStates serve to distinguish reasons for discarding frames, the operation of the Forwarding and Learning processes is the same for all of them. (...) LISTENING represents a port that the spanning tree algorithm has selected to be part of the active topology (computing a Root Port or Designated Port role) but is temporarily discarding frames to guard against loops or incorrect learning." Well, this is not what the driver does - instead it sets mac[port].ingress = true. To get rid of the log spam, prevent unexpected data plane packets to be received by software by discarding them on ingress in the LISTENING state. In terms of blame attribution: the prints only date back to commit d7f9787a763f ("net: dsa: tag_8021q: add support for imprecise RX based on the VBID"). However, the settings would permit a LISTENING port to forward to a FORWARDING port, and the standard suggests that's not OK. Fixes: 640f763f98c2 ("net: dsa: sja1105: Add support for Spanning Tree Protocol") Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20250509113816.2221992-1-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/dsa/sja1105/sja1105_main.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/drivers/net/dsa/sja1105/sja1105_main.c b/drivers/net/dsa/sja1105/sja1105_main.c index f8454f3b6f9c..f674c400f05b 100644 --- a/drivers/net/dsa/sja1105/sja1105_main.c +++ b/drivers/net/dsa/sja1105/sja1105_main.c @@ -2081,6 +2081,7 @@ static void sja1105_bridge_stp_state_set(struct dsa_switch *ds, int port, switch (state) { case BR_STATE_DISABLED: case BR_STATE_BLOCKING: + case BR_STATE_LISTENING: /* From UM10944 description of DRPDTAG (why put this there?): * "Management traffic flows to the port regardless of the state * of the INGRESS flag". So BPDUs are still be allowed to pass. @@ -2090,11 +2091,6 @@ static void sja1105_bridge_stp_state_set(struct dsa_switch *ds, int port, mac[port].egress = false; mac[port].dyn_learn = false; break; - case BR_STATE_LISTENING: - mac[port].ingress = true; - mac[port].egress = false; - mac[port].dyn_learn = false; - break; case BR_STATE_LEARNING: mac[port].ingress = true; mac[port].egress = false; From e8007fad5457ea547ca63bb011fdb03213571c7e Mon Sep 17 00:00:00 2001 From: Steve Siwinski Date: Thu, 8 May 2025 16:01:22 -0400 Subject: [PATCH 418/555] scsi: sd_zbc: block: Respect bio vector limits for REPORT ZONES buffer The REPORT ZONES buffer size is currently limited by the HBA's maximum segment count to ensure the buffer can be mapped. However, the block layer further limits the number of iovec entries to 1024 when allocating a bio. To avoid allocation of buffers too large to be mapped, further restrict the maximum buffer size to BIO_MAX_INLINE_VECS. Replace the UIO_MAXIOV symbolic name with the more contextually appropriate BIO_MAX_INLINE_VECS. Fixes: b091ac616846 ("sd_zbc: Fix report zones buffer allocation") Cc: stable@vger.kernel.org Signed-off-by: Steve Siwinski Link: https://lore.kernel.org/r/20250508200122.243129-1-ssiwinski@atto.com Reviewed-by: Damien Le Moal Signed-off-by: Martin K. Petersen --- block/bio.c | 2 +- drivers/scsi/sd_zbc.c | 6 +++++- include/linux/bio.h | 1 + 3 files changed, 7 insertions(+), 2 deletions(-) diff --git a/block/bio.c b/block/bio.c index 4e6c85a33d74..4be592d37fb6 100644 --- a/block/bio.c +++ b/block/bio.c @@ -611,7 +611,7 @@ struct bio *bio_kmalloc(unsigned short nr_vecs, gfp_t gfp_mask) { struct bio *bio; - if (nr_vecs > UIO_MAXIOV) + if (nr_vecs > BIO_MAX_INLINE_VECS) return NULL; return kmalloc(struct_size(bio, bi_inline_vecs, nr_vecs), gfp_mask); } diff --git a/drivers/scsi/sd_zbc.c b/drivers/scsi/sd_zbc.c index 7a447ff600d2..a8db66428f80 100644 --- a/drivers/scsi/sd_zbc.c +++ b/drivers/scsi/sd_zbc.c @@ -169,6 +169,7 @@ static void *sd_zbc_alloc_report_buffer(struct scsi_disk *sdkp, unsigned int nr_zones, size_t *buflen) { struct request_queue *q = sdkp->disk->queue; + unsigned int max_segments; size_t bufsize; void *buf; @@ -180,12 +181,15 @@ static void *sd_zbc_alloc_report_buffer(struct scsi_disk *sdkp, * Furthermore, since the report zone command cannot be split, make * sure that the allocated buffer can always be mapped by limiting the * number of pages allocated to the HBA max segments limit. + * Since max segments can be larger than the max inline bio vectors, + * further limit the allocated buffer to BIO_MAX_INLINE_VECS. */ nr_zones = min(nr_zones, sdkp->zone_info.nr_zones); bufsize = roundup((nr_zones + 1) * 64, SECTOR_SIZE); bufsize = min_t(size_t, bufsize, queue_max_hw_sectors(q) << SECTOR_SHIFT); - bufsize = min_t(size_t, bufsize, queue_max_segments(q) << PAGE_SHIFT); + max_segments = min(BIO_MAX_INLINE_VECS, queue_max_segments(q)); + bufsize = min_t(size_t, bufsize, max_segments << PAGE_SHIFT); while (bufsize >= SECTOR_SIZE) { buf = kvzalloc(bufsize, GFP_KERNEL | __GFP_NORETRY); diff --git a/include/linux/bio.h b/include/linux/bio.h index cafc7c215de8..b786ec5bcc81 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -11,6 +11,7 @@ #include #define BIO_MAX_VECS 256U +#define BIO_MAX_INLINE_VECS UIO_MAXIOV struct queue_limits; From 40696426b8c8c4f13cf6ac52f0470eed144be4b2 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Tue, 6 May 2025 20:35:40 -0700 Subject: [PATCH 419/555] nvme-pci: make nvme_pci_npages_prp() __always_inline The only reason nvme_pci_npages_prp() could be used as a compile-time known result in BUILD_BUG_ON() is because the compiler was always choosing to inline the function. Under special circumstances (sanitizer coverage functions disabled for __init functions on ARCH=um), the compiler decided to stop inlining it: drivers/nvme/host/pci.c: In function 'nvme_init': include/linux/compiler_types.h:557:45: error: call to '__compiletime_assert_678' declared with attribute error: BUILD_BUG_ON failed: nvme_pci_npages_prp() > NVME_MAX_NR_ALLOCATIONS 557 | _compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__) | ^ include/linux/compiler_types.h:538:25: note: in definition of macro '__compiletime_assert' 538 | prefix ## suffix(); \ | ^~~~~~ include/linux/compiler_types.h:557:9: note: in expansion of macro '_compiletime_assert' 557 | _compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__) | ^~~~~~~~~~~~~~~~~~~ include/linux/build_bug.h:39:37: note: in expansion of macro 'compiletime_assert' 39 | #define BUILD_BUG_ON_MSG(cond, msg) compiletime_assert(!(cond), msg) | ^~~~~~~~~~~~~~~~~~ include/linux/build_bug.h:50:9: note: in expansion of macro 'BUILD_BUG_ON_MSG' 50 | BUILD_BUG_ON_MSG(condition, "BUILD_BUG_ON failed: " #condition) | ^~~~~~~~~~~~~~~~ drivers/nvme/host/pci.c:3804:9: note: in expansion of macro 'BUILD_BUG_ON' 3804 | BUILD_BUG_ON(nvme_pci_npages_prp() > NVME_MAX_NR_ALLOCATIONS); | ^~~~~~~~~~~~ Force it to be __always_inline to make sure it is always available for use with BUILD_BUG_ON(). Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202505061846.12FMyRjj-lkp@intel.com/ Fixes: c372cdd1efdf ("nvme-pci: iod npages fits in s8") Signed-off-by: Kees Cook Signed-off-by: Christoph Hellwig --- drivers/nvme/host/pci.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c index 2e30e9be7408..de084ece2136 100644 --- a/drivers/nvme/host/pci.c +++ b/drivers/nvme/host/pci.c @@ -390,7 +390,7 @@ static bool nvme_dbbuf_update_and_check_event(u16 value, __le32 *dbbuf_db, * as it only leads to a small amount of wasted memory for the lifetime of * the I/O. */ -static int nvme_pci_npages_prp(void) +static __always_inline int nvme_pci_npages_prp(void) { unsigned max_bytes = (NVME_MAX_KB_SZ * 1024) + NVME_CTRL_PAGE_SIZE; unsigned nprps = DIV_ROUND_UP(max_bytes, NVME_CTRL_PAGE_SIZE); From 3d8932133dcecbd9bef1559533c1089601006f45 Mon Sep 17 00:00:00 2001 From: Keith Busch Date: Thu, 8 May 2025 16:57:06 +0200 Subject: [PATCH 420/555] nvme-pci: acquire cq_poll_lock in nvme_poll_irqdisable We need to lock this queue for that condition because the timeout work executes per-namespace and can poll the poll CQ. Reported-by: Hannes Reinecke Closes: https://lore.kernel.org/all/20240902130728.1999-1-hare@kernel.org/ Fixes: a0fa9647a54e ("NVMe: add blk polling support") Signed-off-by: Keith Busch Signed-off-by: Daniel Wagner Signed-off-by: Christoph Hellwig --- drivers/nvme/host/pci.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c index de084ece2136..a9390ac7211e 100644 --- a/drivers/nvme/host/pci.c +++ b/drivers/nvme/host/pci.c @@ -1202,7 +1202,9 @@ static void nvme_poll_irqdisable(struct nvme_queue *nvmeq) WARN_ON_ONCE(test_bit(NVMEQ_POLLED, &nvmeq->flags)); disable_irq(pci_irq_vector(pdev, nvmeq->cq_vector)); + spin_lock(&nvmeq->cq_poll_lock); nvme_poll_cq(nvmeq, NULL); + spin_unlock(&nvmeq->cq_poll_lock); enable_irq(pci_irq_vector(pdev, nvmeq->cq_vector)); } From 85adf2094abb9084770dc4ab302aaa9c5d26dd2d Mon Sep 17 00:00:00 2001 From: Damien Le Moal Date: Fri, 9 May 2025 08:25:00 +0900 Subject: [PATCH 421/555] nvmet: pci-epf: clear completion queue IRQ flag on delete The function nvmet_pci_epf_delete_cq() unconditionally calls nvmet_pci_epf_remove_irq_vector() even for completion queues that do not have interrupts enabled. Furthermore, for completion queues that do have IRQ enabled, deleting and re-creating the completion queue leaves the flag NVMET_PCI_EPF_Q_IRQ_ENABLED set, even if the completion queue is being re-created with IRQ disabled. Fix these issues by calling nvmet_pci_epf_remove_irq_vector() only if NVMET_PCI_EPF_Q_IRQ_ENABLED is set and make sure to always clear that flag. Fixes: 0faa0fe6f90e ("nvmet: New NVMe PCI endpoint function target driver") Cc: stable@vger.kernel.org Signed-off-by: Damien Le Moal Signed-off-by: Christoph Hellwig --- drivers/nvme/target/pci-epf.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/nvme/target/pci-epf.c b/drivers/nvme/target/pci-epf.c index 7fab7f3d79b7..d5442991f2fb 100644 --- a/drivers/nvme/target/pci-epf.c +++ b/drivers/nvme/target/pci-epf.c @@ -1344,7 +1344,8 @@ static u16 nvmet_pci_epf_delete_cq(struct nvmet_ctrl *tctrl, u16 cqid) cancel_delayed_work_sync(&cq->work); nvmet_pci_epf_drain_queue(cq); - nvmet_pci_epf_remove_irq_vector(ctrl, cq->vector); + if (test_and_clear_bit(NVMET_PCI_EPF_Q_IRQ_ENABLED, &cq->flags)) + nvmet_pci_epf_remove_irq_vector(ctrl, cq->vector); nvmet_pci_epf_mem_unmap(ctrl->nvme_epf, &cq->pci_map); return NVME_SC_SUCCESS; From 2c3a6f6a28051f323baf19b48af86e48b812831d Mon Sep 17 00:00:00 2001 From: Damien Le Moal Date: Fri, 9 May 2025 08:25:01 +0900 Subject: [PATCH 422/555] nvmet: pci-epf: do not fall back to using INTX if not supported Some endpoint PCIe controllers do not support raising legacy INTX interrupts. This support is indicated by the intx_capable field of struct pci_epc_features. Modify nvmet_pci_epf_raise_irq() to not automatically fallback to trying raising an INTX interrupt after an MSI or MSI-X error if the controller does not support INTX. Signed-off-by: Damien Le Moal Signed-off-by: Christoph Hellwig --- drivers/nvme/target/pci-epf.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/drivers/nvme/target/pci-epf.c b/drivers/nvme/target/pci-epf.c index d5442991f2fb..bde7818c673d 100644 --- a/drivers/nvme/target/pci-epf.c +++ b/drivers/nvme/target/pci-epf.c @@ -636,14 +636,16 @@ static void nvmet_pci_epf_raise_irq(struct nvmet_pci_epf_ctrl *ctrl, switch (nvme_epf->irq_type) { case PCI_IRQ_MSIX: case PCI_IRQ_MSI: + /* + * If we fail to raise an MSI or MSI-X interrupt, it is likely + * because the host is using legacy INTX IRQs (e.g. BIOS, + * grub), but we can fallback to the INTX type only if the + * endpoint controller supports this type. + */ ret = pci_epc_raise_irq(epf->epc, epf->func_no, epf->vfunc_no, nvme_epf->irq_type, cq->vector + 1); - if (!ret) + if (!ret || !nvme_epf->epc_features->intx_capable) break; - /* - * If we got an error, it is likely because the host is using - * legacy IRQs (e.g. BIOS, grub). - */ fallthrough; case PCI_IRQ_INTX: ret = pci_epc_raise_irq(epf->epc, epf->func_no, epf->vfunc_no, From 4236e600bf902202214aa6277e84c4738c56f762 Mon Sep 17 00:00:00 2001 From: Damien Le Moal Date: Fri, 9 May 2025 08:25:02 +0900 Subject: [PATCH 423/555] nvmet: pci-epf: cleanup nvmet_pci_epf_raise_irq() There is no point in taking the controller irq_lock and calling nvmet_pci_epf_should_raise_irq() for a completion queue which does not have IRQ enabled (NVMET_PCI_EPF_Q_IRQ_ENABLED flag is not set). Move the test for the NVMET_PCI_EPF_Q_IRQ_ENABLED flag out of nvmet_pci_epf_should_raise_irq() to the top of nvmet_pci_epf_raise_irq() to return early when no IRQ should be raised. Also, use dev_err_ratelimited() to avoid a message storm under load when raising IRQs is failing. Signed-off-by: Damien Le Moal Reviewed-by: Niklas Cassel Signed-off-by: Christoph Hellwig --- drivers/nvme/target/pci-epf.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/nvme/target/pci-epf.c b/drivers/nvme/target/pci-epf.c index bde7818c673d..a6ccf3fcccc2 100644 --- a/drivers/nvme/target/pci-epf.c +++ b/drivers/nvme/target/pci-epf.c @@ -596,9 +596,6 @@ static bool nvmet_pci_epf_should_raise_irq(struct nvmet_pci_epf_ctrl *ctrl, struct nvmet_pci_epf_irq_vector *iv = cq->iv; bool ret; - if (!test_bit(NVMET_PCI_EPF_Q_IRQ_ENABLED, &cq->flags)) - return false; - /* IRQ coalescing for the admin queue is not allowed. */ if (!cq->qid) return true; @@ -625,7 +622,8 @@ static void nvmet_pci_epf_raise_irq(struct nvmet_pci_epf_ctrl *ctrl, struct pci_epf *epf = nvme_epf->epf; int ret = 0; - if (!test_bit(NVMET_PCI_EPF_Q_LIVE, &cq->flags)) + if (!test_bit(NVMET_PCI_EPF_Q_LIVE, &cq->flags) || + !test_bit(NVMET_PCI_EPF_Q_IRQ_ENABLED, &cq->flags)) return; mutex_lock(&ctrl->irq_lock); @@ -658,7 +656,9 @@ static void nvmet_pci_epf_raise_irq(struct nvmet_pci_epf_ctrl *ctrl, } if (ret) - dev_err(ctrl->dev, "Failed to raise IRQ (err=%d)\n", ret); + dev_err_ratelimited(ctrl->dev, + "CQ[%u]: Failed to raise IRQ (err=%d)\n", + cq->qid, ret); unlock: mutex_unlock(&ctrl->irq_lock); From 4f6f3f4fe31695cd69874457573a2236c05fef02 Mon Sep 17 00:00:00 2001 From: Damien Le Moal Date: Fri, 9 May 2025 08:25:03 +0900 Subject: [PATCH 424/555] nvmet: pci-epf: improve debug message Improve the debug message of nvmet_pci_epf_create_cq() to indicate if a completion queue IRQ is disabled. Signed-off-by: Damien Le Moal Reviewed-by: Niklas Cassel Signed-off-by: Christoph Hellwig --- drivers/nvme/target/pci-epf.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/drivers/nvme/target/pci-epf.c b/drivers/nvme/target/pci-epf.c index a6ccf3fcccc2..94a908b2340e 100644 --- a/drivers/nvme/target/pci-epf.c +++ b/drivers/nvme/target/pci-epf.c @@ -1321,8 +1321,14 @@ static u16 nvmet_pci_epf_create_cq(struct nvmet_ctrl *tctrl, set_bit(NVMET_PCI_EPF_Q_LIVE, &cq->flags); - dev_dbg(ctrl->dev, "CQ[%u]: %u entries of %zu B, IRQ vector %u\n", - cqid, qsize, cq->qes, cq->vector); + if (test_bit(NVMET_PCI_EPF_Q_IRQ_ENABLED, &cq->flags)) + dev_dbg(ctrl->dev, + "CQ[%u]: %u entries of %zu B, IRQ vector %u\n", + cqid, qsize, cq->qes, cq->vector); + else + dev_dbg(ctrl->dev, + "CQ[%u]: %u entries of %zu B, IRQ disabled\n", + cqid, qsize, cq->qes); return NVME_SC_SUCCESS; From 8113d610a79885db5d53b5e01138ed4159e05ce9 Mon Sep 17 00:00:00 2001 From: Damien Le Moal Date: Fri, 9 May 2025 08:25:04 +0900 Subject: [PATCH 425/555] nvmet: pci-epf: remove NVMET_PCI_EPF_Q_IS_SQ The flag NVMET_PCI_EPF_Q_IS_SQ is set but never used. Remove it. Signed-off-by: Damien Le Moal Signed-off-by: Christoph Hellwig --- drivers/nvme/target/pci-epf.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/nvme/target/pci-epf.c b/drivers/nvme/target/pci-epf.c index 94a908b2340e..7123c855b5a6 100644 --- a/drivers/nvme/target/pci-epf.c +++ b/drivers/nvme/target/pci-epf.c @@ -62,8 +62,7 @@ static DEFINE_MUTEX(nvmet_pci_epf_ports_mutex); #define NVMET_PCI_EPF_CQ_RETRY_INTERVAL msecs_to_jiffies(1) enum nvmet_pci_epf_queue_flags { - NVMET_PCI_EPF_Q_IS_SQ = 0, /* The queue is a submission queue */ - NVMET_PCI_EPF_Q_LIVE, /* The queue is live */ + NVMET_PCI_EPF_Q_LIVE = 0, /* The queue is live */ NVMET_PCI_EPF_Q_IRQ_ENABLED, /* IRQ is enabled for this queue */ }; @@ -1542,7 +1541,6 @@ static void nvmet_pci_epf_init_queue(struct nvmet_pci_epf_ctrl *ctrl, if (sq) { queue = &ctrl->sq[qid]; - set_bit(NVMET_PCI_EPF_Q_IS_SQ, &queue->flags); } else { queue = &ctrl->cq[qid]; INIT_DELAYED_WORK(&queue->work, nvmet_pci_epf_cq_work); From 76ca05e0abe31a4f47a5b5a85041b5a22c03baf8 Mon Sep 17 00:00:00 2001 From: Oleksij Rempel Date: Sun, 4 May 2025 10:14:33 +0200 Subject: [PATCH 426/555] net: dsa: microchip: let phylink manage PHY EEE configuration on KSZ switches Phylink expects MAC drivers to provide LPI callbacks to properly manage Energy Efficient Ethernet (EEE) configuration. On KSZ switches with integrated PHYs, LPI is internally handled by hardware, while ports without integrated PHYs have no documented MAC-level LPI support. Provide dummy mac_disable_tx_lpi() and mac_enable_tx_lpi() callbacks to satisfy phylink requirements. Also, set default EEE capabilities during phylink initialization where applicable. Since phylink can now gracefully handle optional EEE configuration, remove the need for the MICREL_NO_EEE PHY flag. This change addresses issues caused by incomplete EEE refactoring introduced in commit fe0d4fd9285e ("net: phy: Keep track of EEE configuration"). It is not easily possible to fix all older kernels, but this patch ensures proper behavior on latest kernels and can be considered for backporting to stable kernels starting from v6.14. Fixes: fe0d4fd9285e ("net: phy: Keep track of EEE configuration") Signed-off-by: Oleksij Rempel Cc: stable@vger.kernel.org # v6.14+ Link: https://patch.msgid.link/20250504081434.424489-2-o.rempel@pengutronix.de Signed-off-by: Paolo Abeni --- drivers/net/dsa/microchip/ksz_common.c | 137 +++++++++++++++++++------ 1 file changed, 108 insertions(+), 29 deletions(-) diff --git a/drivers/net/dsa/microchip/ksz_common.c b/drivers/net/dsa/microchip/ksz_common.c index 89f0796894af..f95a9aac56ee 100644 --- a/drivers/net/dsa/microchip/ksz_common.c +++ b/drivers/net/dsa/microchip/ksz_common.c @@ -265,16 +265,70 @@ static void ksz_phylink_mac_link_down(struct phylink_config *config, unsigned int mode, phy_interface_t interface); +/** + * ksz_phylink_mac_disable_tx_lpi() - Callback to signal LPI support (Dummy) + * @config: phylink config structure + * + * This function is a dummy handler. See ksz_phylink_mac_enable_tx_lpi() for + * a detailed explanation of EEE/LPI handling in KSZ switches. + */ +static void ksz_phylink_mac_disable_tx_lpi(struct phylink_config *config) +{ +} + +/** + * ksz_phylink_mac_enable_tx_lpi() - Callback to signal LPI support (Dummy) + * @config: phylink config structure + * @timer: timer value before entering LPI (unused) + * @tx_clock_stop: whether to stop the TX clock in LPI mode (unused) + * + * This function signals to phylink that the driver architecture supports + * LPI management, enabling phylink to control EEE advertisement during + * negotiation according to IEEE Std 802.3 (Clause 78). + * + * Hardware Management of EEE/LPI State: + * For KSZ switch ports with integrated PHYs (e.g., KSZ9893R ports 1-2), + * observation and testing suggest that the actual EEE / Low Power Idle (LPI) + * state transitions are managed autonomously by the hardware based on + * the auto-negotiation results. (Note: While the datasheet describes EEE + * operation based on negotiation, it doesn't explicitly detail the internal + * MAC/PHY interaction, so autonomous hardware management of the MAC state + * for LPI is inferred from observed behavior). + * This hardware control, consistent with the switch's ability to operate + * autonomously via strapping, means MAC-level software intervention is not + * required or exposed for managing the LPI state once EEE is negotiated. + * (Ref: KSZ9893R Data Sheet DS00002420D, primarily Section 4.7.5 explaining + * EEE, also Sections 4.1.7 on Auto-Negotiation and 3.2.1 on Configuration + * Straps). + * + * Additionally, ports configured as MAC interfaces (e.g., KSZ9893R port 3) + * lack documented MAC-level LPI control. + * + * Therefore, this callback performs no action and serves primarily to inform + * phylink of LPI awareness and to document the inferred hardware behavior. + * + * Returns: 0 (Always success) + */ +static int ksz_phylink_mac_enable_tx_lpi(struct phylink_config *config, + u32 timer, bool tx_clock_stop) +{ + return 0; +} + static const struct phylink_mac_ops ksz88x3_phylink_mac_ops = { .mac_config = ksz88x3_phylink_mac_config, .mac_link_down = ksz_phylink_mac_link_down, .mac_link_up = ksz8_phylink_mac_link_up, + .mac_disable_tx_lpi = ksz_phylink_mac_disable_tx_lpi, + .mac_enable_tx_lpi = ksz_phylink_mac_enable_tx_lpi, }; static const struct phylink_mac_ops ksz8_phylink_mac_ops = { .mac_config = ksz_phylink_mac_config, .mac_link_down = ksz_phylink_mac_link_down, .mac_link_up = ksz8_phylink_mac_link_up, + .mac_disable_tx_lpi = ksz_phylink_mac_disable_tx_lpi, + .mac_enable_tx_lpi = ksz_phylink_mac_enable_tx_lpi, }; static const struct ksz_dev_ops ksz88xx_dev_ops = { @@ -358,6 +412,8 @@ static const struct phylink_mac_ops ksz9477_phylink_mac_ops = { .mac_config = ksz_phylink_mac_config, .mac_link_down = ksz_phylink_mac_link_down, .mac_link_up = ksz9477_phylink_mac_link_up, + .mac_disable_tx_lpi = ksz_phylink_mac_disable_tx_lpi, + .mac_enable_tx_lpi = ksz_phylink_mac_enable_tx_lpi, }; static const struct ksz_dev_ops ksz9477_dev_ops = { @@ -401,6 +457,8 @@ static const struct phylink_mac_ops lan937x_phylink_mac_ops = { .mac_config = ksz_phylink_mac_config, .mac_link_down = ksz_phylink_mac_link_down, .mac_link_up = ksz9477_phylink_mac_link_up, + .mac_disable_tx_lpi = ksz_phylink_mac_disable_tx_lpi, + .mac_enable_tx_lpi = ksz_phylink_mac_enable_tx_lpi, }; static const struct ksz_dev_ops lan937x_dev_ops = { @@ -2016,6 +2074,18 @@ static void ksz_phylink_get_caps(struct dsa_switch *ds, int port, if (dev->dev_ops->get_caps) dev->dev_ops->get_caps(dev, port, config); + + if (ds->ops->support_eee && ds->ops->support_eee(ds, port)) { + memcpy(config->lpi_interfaces, config->supported_interfaces, + sizeof(config->lpi_interfaces)); + + config->lpi_capabilities = MAC_100FD; + if (dev->info->gbit_capable[port]) + config->lpi_capabilities |= MAC_1000FD; + + /* EEE is fully operational */ + config->eee_enabled_default = true; + } } void ksz_r_mib_stats64(struct ksz_device *dev, int port) @@ -3008,31 +3078,6 @@ static u32 ksz_get_phy_flags(struct dsa_switch *ds, int port) if (!port) return MICREL_KSZ8_P1_ERRATA; break; - case KSZ8567_CHIP_ID: - /* KSZ8567R Errata DS80000752C Module 4 */ - case KSZ8765_CHIP_ID: - case KSZ8794_CHIP_ID: - case KSZ8795_CHIP_ID: - /* KSZ879x/KSZ877x/KSZ876x Errata DS80000687C Module 2 */ - case KSZ9477_CHIP_ID: - /* KSZ9477S Errata DS80000754A Module 4 */ - case KSZ9567_CHIP_ID: - /* KSZ9567S Errata DS80000756A Module 4 */ - case KSZ9896_CHIP_ID: - /* KSZ9896C Errata DS80000757A Module 3 */ - case KSZ9897_CHIP_ID: - case LAN9646_CHIP_ID: - /* KSZ9897R Errata DS80000758C Module 4 */ - /* Energy Efficient Ethernet (EEE) feature select must be manually disabled - * The EEE feature is enabled by default, but it is not fully - * operational. It must be manually disabled through register - * controls. If not disabled, the PHY ports can auto-negotiate - * to enable EEE, and this feature can cause link drops when - * linked to another device supporting EEE. - * - * The same item appears in the errata for all switches above. - */ - return MICREL_NO_EEE; } return 0; @@ -3466,6 +3511,20 @@ static int ksz_max_mtu(struct dsa_switch *ds, int port) return -EOPNOTSUPP; } +/** + * ksz_support_eee - Determine Energy Efficient Ethernet (EEE) support for a + * port + * @ds: Pointer to the DSA switch structure + * @port: Port number to check + * + * This function also documents devices where EEE was initially advertised but + * later withdrawn due to reliability issues, as described in official errata + * documents. These devices are explicitly listed to record known limitations, + * even if there is no technical necessity for runtime checks. + * + * Returns: true if the internal PHY on the given port supports fully + * operational EEE, false otherwise. + */ static bool ksz_support_eee(struct dsa_switch *ds, int port) { struct ksz_device *dev = ds->priv; @@ -3475,15 +3534,35 @@ static bool ksz_support_eee(struct dsa_switch *ds, int port) switch (dev->chip_id) { case KSZ8563_CHIP_ID: - case KSZ8567_CHIP_ID: - case KSZ9477_CHIP_ID: case KSZ9563_CHIP_ID: - case KSZ9567_CHIP_ID: case KSZ9893_CHIP_ID: + return true; + case KSZ8567_CHIP_ID: + /* KSZ8567R Errata DS80000752C Module 4 */ + case KSZ8765_CHIP_ID: + case KSZ8794_CHIP_ID: + case KSZ8795_CHIP_ID: + /* KSZ879x/KSZ877x/KSZ876x Errata DS80000687C Module 2 */ + case KSZ9477_CHIP_ID: + /* KSZ9477S Errata DS80000754A Module 4 */ + case KSZ9567_CHIP_ID: + /* KSZ9567S Errata DS80000756A Module 4 */ case KSZ9896_CHIP_ID: + /* KSZ9896C Errata DS80000757A Module 3 */ case KSZ9897_CHIP_ID: case LAN9646_CHIP_ID: - return true; + /* KSZ9897R Errata DS80000758C Module 4 */ + /* Energy Efficient Ethernet (EEE) feature select must be + * manually disabled + * The EEE feature is enabled by default, but it is not fully + * operational. It must be manually disabled through register + * controls. If not disabled, the PHY ports can auto-negotiate + * to enable EEE, and this feature can cause link drops when + * linked to another device supporting EEE. + * + * The same item appears in the errata for all switches above. + */ + break; } return false; From 8c619eb21b8e87ae95877e9cca9fcb0e3115776e Mon Sep 17 00:00:00 2001 From: Oleksij Rempel Date: Sun, 4 May 2025 10:14:34 +0200 Subject: [PATCH 427/555] net: phy: micrel: remove KSZ9477 EEE quirks now handled by phylink The KSZ9477 PHY driver contained workarounds for broken EEE capability advertisements by manually masking supported EEE modes and forcibly disabling EEE if MICREL_NO_EEE was set. With proper MAC-side EEE handling implemented via phylink, these quirks are no longer necessary. Remove MICREL_NO_EEE handling and the use of ksz9477_get_features(). This simplifies the PHY driver and avoids duplicated EEE management logic. Signed-off-by: Oleksij Rempel Cc: stable@vger.kernel.org # v6.14+ Link: https://patch.msgid.link/20250504081434.424489-3-o.rempel@pengutronix.de Signed-off-by: Paolo Abeni --- drivers/net/phy/micrel.c | 7 ------- include/linux/micrel_phy.h | 1 - 2 files changed, 8 deletions(-) diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c index 24882d30f685..e2c6569d8c45 100644 --- a/drivers/net/phy/micrel.c +++ b/drivers/net/phy/micrel.c @@ -2027,12 +2027,6 @@ static int ksz9477_config_init(struct phy_device *phydev) return err; } - /* According to KSZ9477 Errata DS80000754C (Module 4) all EEE modes - * in this switch shall be regarded as broken. - */ - if (phydev->dev_flags & MICREL_NO_EEE) - phy_disable_eee(phydev); - return kszphy_config_init(phydev); } @@ -5705,7 +5699,6 @@ static struct phy_driver ksphy_driver[] = { .handle_interrupt = kszphy_handle_interrupt, .suspend = genphy_suspend, .resume = ksz9477_resume, - .get_features = ksz9477_get_features, } }; module_phy_driver(ksphy_driver); diff --git a/include/linux/micrel_phy.h b/include/linux/micrel_phy.h index 591bf5b5e8dc..9af01bdd86d2 100644 --- a/include/linux/micrel_phy.h +++ b/include/linux/micrel_phy.h @@ -44,7 +44,6 @@ #define MICREL_PHY_50MHZ_CLK BIT(0) #define MICREL_PHY_FXEN BIT(1) #define MICREL_KSZ8_P1_ERRATA BIT(2) -#define MICREL_NO_EEE BIT(3) #define MICREL_KSZ9021_EXTREG_CTRL 0xB #define MICREL_KSZ9021_EXTREG_DATA_WRITE 0xC From 2b24eb060c2bb9ef79e1d3bcf633ba1bc95215d6 Mon Sep 17 00:00:00 2001 From: Christian Heusel Date: Mon, 12 May 2025 22:23:37 +0200 Subject: [PATCH 428/555] ALSA: usb-audio: Add sample rate quirk for Audioengine D1 A user reported on the Arch Linux Forums that their device is emitting the following message in the kernel journal, which is fixed by adding the quirk as submitted in this patch: > kernel: usb 1-2: current rate 8436480 is different from the runtime rate 48000 There also is an entry for this product line added long time ago. Their specific device has the following ID: $ lsusb | grep Audio Bus 001 Device 002: ID 1101:0003 EasyPass Industrial Co., Ltd Audioengine D1 Link: https://bbs.archlinux.org/viewtopic.php?id=305494 Fixes: 93f9d1a4ac593 ("ALSA: usb-audio: Apply sample rate quirk for Audioengine D1") Cc: stable@vger.kernel.org Signed-off-by: Christian Heusel Link: https://patch.msgid.link/20250512-audioengine-quirk-addition-v1-1-4c370af6eff7@heusel.eu Signed-off-by: Takashi Iwai --- sound/usb/quirks.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/sound/usb/quirks.c b/sound/usb/quirks.c index 9112313a9dbc..eb192834db68 100644 --- a/sound/usb/quirks.c +++ b/sound/usb/quirks.c @@ -2250,6 +2250,8 @@ static const struct usb_audio_quirk_flags_table quirk_flags_table[] = { QUIRK_FLAG_FIXED_RATE), DEVICE_FLG(0x0fd9, 0x0008, /* Hauppauge HVR-950Q */ QUIRK_FLAG_SHARE_MEDIA_DEVICE | QUIRK_FLAG_ALIGN_TRANSFER), + DEVICE_FLG(0x1101, 0x0003, /* Audioengine D1 */ + QUIRK_FLAG_GET_SAMPLE_RATE), DEVICE_FLG(0x1224, 0x2a25, /* Jieli Technology USB PHY 2.0 */ QUIRK_FLAG_GET_SAMPLE_RATE | QUIRK_FLAG_MIC_RES_16), DEVICE_FLG(0x1395, 0x740a, /* Sennheiser DECT */ From 66e48ef6ef506c89ec1b3851c6f9f5f80b5835ff Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Tue, 13 May 2025 09:31:04 +0200 Subject: [PATCH 429/555] ALSA: sh: SND_AICA should depend on SH_DMA_API If CONFIG_SH_DMA_API=n: WARNING: unmet direct dependencies detected for G2_DMA Depends on [n]: SH_DREAMCAST [=y] && SH_DMA_API [=n] Selected by [y]: - SND_AICA [=y] && SOUND [=y] && SND [=y] && SND_SUPERH [=y] && SH_DREAMCAST [=y] SND_AICA selects G2_DMA. As the latter depends on SH_DMA_API, the former should depend on SH_DMA_API, too. Fixes: f477a538c14d07f8 ("sh: dma: fix kconfig dependency for G2_DMA") Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202505131320.PzgTtl9H-lkp@intel.com/ Signed-off-by: Geert Uytterhoeven Link: https://patch.msgid.link/b90625f8a9078d0d304bafe862cbe3a3fab40082.1747121335.git.geert+renesas@glider.be Signed-off-by: Takashi Iwai --- sound/sh/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/sh/Kconfig b/sound/sh/Kconfig index b75fbb3236a7..f5fa09d740b4 100644 --- a/sound/sh/Kconfig +++ b/sound/sh/Kconfig @@ -14,7 +14,7 @@ if SND_SUPERH config SND_AICA tristate "Dreamcast Yamaha AICA sound" - depends on SH_DREAMCAST + depends on SH_DREAMCAST && SH_DMA_API select SND_PCM select G2_DMA help From 3e38f946062b4845961ab86b726651b4457b2af8 Mon Sep 17 00:00:00 2001 From: Emanuele Ghidoli Date: Mon, 12 May 2025 11:54:41 +0200 Subject: [PATCH 430/555] gpio: pca953x: fix IRQ storm on system wake up If an input changes state during wake-up and is used as an interrupt source, the IRQ handler reads the volatile input register to clear the interrupt mask and deassert the IRQ line. However, the IRQ handler is triggered before access to the register is granted, causing the read operation to fail. As a result, the IRQ handler enters a loop, repeatedly printing the "failed reading register" message, until `pca953x_resume()` is eventually called, which restores the driver context and enables access to registers. Fix by disabling the IRQ line before entering suspend mode, and re-enabling it after the driver context is restored in `pca953x_resume()`. An IRQ can be disabled with disable_irq() and still wake the system as long as the IRQ has wake enabled, so the wake-up functionality is preserved. Fixes: b76574300504 ("gpio: pca953x: Restore registers after suspend/resume cycle") Cc: stable@vger.kernel.org Signed-off-by: Emanuele Ghidoli Signed-off-by: Francesco Dolcini Reviewed-by: Andy Shevchenko Tested-by: Geert Uytterhoeven Link: https://lore.kernel.org/r/20250512095441.31645-1-francesco@dolcini.it Signed-off-by: Bartosz Golaszewski --- drivers/gpio/gpio-pca953x.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/gpio/gpio-pca953x.c b/drivers/gpio/gpio-pca953x.c index 442435ded020..13cc120cf11f 100644 --- a/drivers/gpio/gpio-pca953x.c +++ b/drivers/gpio/gpio-pca953x.c @@ -1204,6 +1204,8 @@ static int pca953x_restore_context(struct pca953x_chip *chip) guard(mutex)(&chip->i2c_lock); + if (chip->client->irq > 0) + enable_irq(chip->client->irq); regcache_cache_only(chip->regmap, false); regcache_mark_dirty(chip->regmap); ret = pca953x_regcache_sync(chip); @@ -1216,6 +1218,10 @@ static int pca953x_restore_context(struct pca953x_chip *chip) static void pca953x_save_context(struct pca953x_chip *chip) { guard(mutex)(&chip->i2c_lock); + + /* Disable IRQ to prevent early triggering while regmap "cache only" is on */ + if (chip->client->irq > 0) + disable_irq(chip->client->irq); regcache_cache_only(chip->regmap, true); } From 7118be7c6072f40391923543fdd1563b8d56377c Mon Sep 17 00:00:00 2001 From: Markus Burri Date: Fri, 9 May 2025 17:04:59 +0200 Subject: [PATCH 431/555] gpio: virtuser: fix potential out-of-bound write If the caller wrote more characters, count is truncated to the max available space in "simple_write_to_buffer". Check that the input size does not exceed the buffer size. Write a zero termination afterwards. Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202505091754.285hHbr2-lkp@intel.com/ Signed-off-by: Markus Burri Link: https://lore.kernel.org/r/20250509150459.115489-1-markus.burri@mt.com Signed-off-by: Bartosz Golaszewski --- drivers/gpio/gpio-virtuser.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/drivers/gpio/gpio-virtuser.c b/drivers/gpio/gpio-virtuser.c index 13407fd4f0eb..eab6726953b4 100644 --- a/drivers/gpio/gpio-virtuser.c +++ b/drivers/gpio/gpio-virtuser.c @@ -401,10 +401,15 @@ static ssize_t gpio_virtuser_direction_do_write(struct file *file, char buf[32], *trimmed; int ret, dir, val = 0; - ret = simple_write_to_buffer(buf, sizeof(buf), ppos, user_buf, count); + if (count >= sizeof(buf)) + return -EINVAL; + + ret = simple_write_to_buffer(buf, sizeof(buf) - 1, ppos, user_buf, count); if (ret < 0) return ret; + buf[ret] = '\0'; + trimmed = strim(buf); if (strcmp(trimmed, "input") == 0) { @@ -623,12 +628,15 @@ static ssize_t gpio_virtuser_consumer_write(struct file *file, char buf[GPIO_VIRTUSER_NAME_BUF_LEN + 2]; int ret; + if (count >= sizeof(buf)) + return -EINVAL; + ret = simple_write_to_buffer(buf, GPIO_VIRTUSER_NAME_BUF_LEN, ppos, user_buf, count); if (ret < 0) return ret; - buf[strlen(buf) - 1] = '\0'; + buf[ret] = '\0'; ret = gpiod_set_consumer_name(data->ad.desc, buf); if (ret) From fd837de3c9cb1a162c69bc1fb1f438467fe7f2f5 Mon Sep 17 00:00:00 2001 From: "Masami Hiramatsu (Google)" Date: Sat, 10 May 2025 12:44:41 +0900 Subject: [PATCH 432/555] tracing: probes: Fix a possible race in trace_probe_log APIs Since the shared trace_probe_log variable can be accessed and modified via probe event create operation of kprobe_events, uprobe_events, and dynamic_events, it should be protected. In the dynamic_events, all operations are serialized by `dyn_event_ops_mutex`. But kprobe_events and uprobe_events interfaces are not serialized. To solve this issue, introduces dyn_event_create(), which runs create() operation under the mutex, for kprobe_events and uprobe_events. This also uses lockdep to check the mutex is held when using trace_probe_log* APIs. Link: https://lore.kernel.org/all/174684868120.551552.3068655787654268804.stgit@devnote2/ Reported-by: Paul Cacheux Closes: https://lore.kernel.org/all/20250510074456.805a16872b591e2971a4d221@kernel.org/ Fixes: ab105a4fb894 ("tracing: Use tracing error_log with probe events") Signed-off-by: Masami Hiramatsu (Google) --- kernel/trace/trace_dynevent.c | 16 +++++++++++++++- kernel/trace/trace_dynevent.h | 1 + kernel/trace/trace_kprobe.c | 2 +- kernel/trace/trace_probe.c | 9 +++++++++ kernel/trace/trace_uprobe.c | 2 +- 5 files changed, 27 insertions(+), 3 deletions(-) diff --git a/kernel/trace/trace_dynevent.c b/kernel/trace/trace_dynevent.c index a322e4f249a5..5d64a18cacac 100644 --- a/kernel/trace/trace_dynevent.c +++ b/kernel/trace/trace_dynevent.c @@ -16,7 +16,7 @@ #include "trace_output.h" /* for trace_event_sem */ #include "trace_dynevent.h" -static DEFINE_MUTEX(dyn_event_ops_mutex); +DEFINE_MUTEX(dyn_event_ops_mutex); static LIST_HEAD(dyn_event_ops_list); bool trace_event_dyn_try_get_ref(struct trace_event_call *dyn_call) @@ -116,6 +116,20 @@ int dyn_event_release(const char *raw_command, struct dyn_event_operations *type return ret; } +/* + * Locked version of event creation. The event creation must be protected by + * dyn_event_ops_mutex because of protecting trace_probe_log. + */ +int dyn_event_create(const char *raw_command, struct dyn_event_operations *type) +{ + int ret; + + mutex_lock(&dyn_event_ops_mutex); + ret = type->create(raw_command); + mutex_unlock(&dyn_event_ops_mutex); + return ret; +} + static int create_dyn_event(const char *raw_command) { struct dyn_event_operations *ops; diff --git a/kernel/trace/trace_dynevent.h b/kernel/trace/trace_dynevent.h index 936477a111d3..beee3f8d7544 100644 --- a/kernel/trace/trace_dynevent.h +++ b/kernel/trace/trace_dynevent.h @@ -100,6 +100,7 @@ void *dyn_event_seq_next(struct seq_file *m, void *v, loff_t *pos); void dyn_event_seq_stop(struct seq_file *m, void *v); int dyn_events_release_all(struct dyn_event_operations *type); int dyn_event_release(const char *raw_command, struct dyn_event_operations *type); +int dyn_event_create(const char *raw_command, struct dyn_event_operations *type); /* * for_each_dyn_event - iterate over the dyn_event list diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index 2703b96d8990..3e5c47b6d7b2 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -1089,7 +1089,7 @@ static int create_or_delete_trace_kprobe(const char *raw_command) if (raw_command[0] == '-') return dyn_event_release(raw_command, &trace_kprobe_ops); - ret = trace_kprobe_create(raw_command); + ret = dyn_event_create(raw_command, &trace_kprobe_ops); return ret == -ECANCELED ? -EINVAL : ret; } diff --git a/kernel/trace/trace_probe.c b/kernel/trace/trace_probe.c index 2eeecb6c95ee..424751cdf31f 100644 --- a/kernel/trace/trace_probe.c +++ b/kernel/trace/trace_probe.c @@ -154,9 +154,12 @@ fail: } static struct trace_probe_log trace_probe_log; +extern struct mutex dyn_event_ops_mutex; void trace_probe_log_init(const char *subsystem, int argc, const char **argv) { + lockdep_assert_held(&dyn_event_ops_mutex); + trace_probe_log.subsystem = subsystem; trace_probe_log.argc = argc; trace_probe_log.argv = argv; @@ -165,11 +168,15 @@ void trace_probe_log_init(const char *subsystem, int argc, const char **argv) void trace_probe_log_clear(void) { + lockdep_assert_held(&dyn_event_ops_mutex); + memset(&trace_probe_log, 0, sizeof(trace_probe_log)); } void trace_probe_log_set_index(int index) { + lockdep_assert_held(&dyn_event_ops_mutex); + trace_probe_log.index = index; } @@ -178,6 +185,8 @@ void __trace_probe_log_err(int offset, int err_type) char *command, *p; int i, len = 0, pos = 0; + lockdep_assert_held(&dyn_event_ops_mutex); + if (!trace_probe_log.argv) return; diff --git a/kernel/trace/trace_uprobe.c b/kernel/trace/trace_uprobe.c index 3386439ec9f6..35cf76c75dd7 100644 --- a/kernel/trace/trace_uprobe.c +++ b/kernel/trace/trace_uprobe.c @@ -741,7 +741,7 @@ static int create_or_delete_trace_uprobe(const char *raw_command) if (raw_command[0] == '-') return dyn_event_release(raw_command, &trace_uprobe_ops); - ret = trace_uprobe_create(raw_command); + ret = dyn_event_create(raw_command, &trace_uprobe_ops); return ret == -ECANCELED ? -EINVAL : ret; } From 4227ea91e2657f7965e34313448e9d0a2b67712e Mon Sep 17 00:00:00 2001 From: Jonas Gorski Date: Thu, 8 May 2025 11:14:24 +0200 Subject: [PATCH 433/555] net: dsa: b53: prevent standalone from trying to forward to other ports When bridged ports and standalone ports share a VLAN, e.g. via VLAN uppers, or untagged traffic with a vlan unaware bridge, the ASIC will still try to forward traffic to known FDB entries on standalone ports. But since the port VLAN masks prevent forwarding to bridged ports, this traffic will be dropped. This e.g. can be observed in the bridge_vlan_unaware ping tests, where this breaks pinging with learning on. Work around this by enabling the simplified EAP mode on switches supporting it for standalone ports, which causes the ASIC to redirect traffic of unknown source MAC addresses to the CPU port. Since standalone ports do not learn, there are no known source MAC addresses, so effectively this redirects all incoming traffic to the CPU port. Fixes: ff39c2d68679 ("net: dsa: b53: Add bridge support") Signed-off-by: Jonas Gorski Reviewed-by: Florian Fainelli Reviewed-by: Vladimir Oltean Link: https://patch.msgid.link/20250508091424.26870-1-jonas.gorski@gmail.com Signed-off-by: Paolo Abeni --- drivers/net/dsa/b53/b53_common.c | 33 ++++++++++++++++++++++++++++++++ drivers/net/dsa/b53/b53_regs.h | 14 ++++++++++++++ 2 files changed, 47 insertions(+) diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index 9eb39cfa5fb2..7216eb8f9493 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -326,6 +326,26 @@ static void b53_get_vlan_entry(struct b53_device *dev, u16 vid, } } +static void b53_set_eap_mode(struct b53_device *dev, int port, int mode) +{ + u64 eap_conf; + + if (is5325(dev) || is5365(dev) || dev->chip_id == BCM5389_DEVICE_ID) + return; + + b53_read64(dev, B53_EAP_PAGE, B53_PORT_EAP_CONF(port), &eap_conf); + + if (is63xx(dev)) { + eap_conf &= ~EAP_MODE_MASK_63XX; + eap_conf |= (u64)mode << EAP_MODE_SHIFT_63XX; + } else { + eap_conf &= ~EAP_MODE_MASK; + eap_conf |= (u64)mode << EAP_MODE_SHIFT; + } + + b53_write64(dev, B53_EAP_PAGE, B53_PORT_EAP_CONF(port), eap_conf); +} + static void b53_set_forwarding(struct b53_device *dev, int enable) { u8 mgmt; @@ -586,6 +606,13 @@ int b53_setup_port(struct dsa_switch *ds, int port) b53_port_set_mcast_flood(dev, port, true); b53_port_set_learning(dev, port, false); + /* Force all traffic to go to the CPU port to prevent the ASIC from + * trying to forward to bridged ports on matching FDB entries, then + * dropping frames because it isn't allowed to forward there. + */ + if (dsa_is_user_port(ds, port)) + b53_set_eap_mode(dev, port, EAP_MODE_SIMPLIFIED); + return 0; } EXPORT_SYMBOL(b53_setup_port); @@ -2042,6 +2069,9 @@ int b53_br_join(struct dsa_switch *ds, int port, struct dsa_bridge bridge, pvlan |= BIT(i); } + /* Disable redirection of unknown SA to the CPU port */ + b53_set_eap_mode(dev, port, EAP_MODE_BASIC); + /* Configure the local port VLAN control membership to include * remote ports and update the local port bitmask */ @@ -2077,6 +2107,9 @@ void b53_br_leave(struct dsa_switch *ds, int port, struct dsa_bridge bridge) pvlan &= ~BIT(i); } + /* Enable redirection of unknown SA to the CPU port */ + b53_set_eap_mode(dev, port, EAP_MODE_SIMPLIFIED); + b53_write16(dev, B53_PVLAN_PAGE, B53_PVLAN_PORT_MASK(port), pvlan); dev->ports[port].vlan_ctl_mask = pvlan; diff --git a/drivers/net/dsa/b53/b53_regs.h b/drivers/net/dsa/b53/b53_regs.h index bfbcb66bef66..5f7a0e5c5709 100644 --- a/drivers/net/dsa/b53/b53_regs.h +++ b/drivers/net/dsa/b53/b53_regs.h @@ -50,6 +50,9 @@ /* Jumbo Frame Registers */ #define B53_JUMBO_PAGE 0x40 +/* EAP Registers */ +#define B53_EAP_PAGE 0x42 + /* EEE Control Registers Page */ #define B53_EEE_PAGE 0x92 @@ -480,6 +483,17 @@ #define JMS_MIN_SIZE 1518 #define JMS_MAX_SIZE 9724 +/************************************************************************* + * EAP Page Registers + *************************************************************************/ +#define B53_PORT_EAP_CONF(i) (0x20 + 8 * (i)) +#define EAP_MODE_SHIFT 51 +#define EAP_MODE_SHIFT_63XX 50 +#define EAP_MODE_MASK (0x3ull << EAP_MODE_SHIFT) +#define EAP_MODE_MASK_63XX (0x3ull << EAP_MODE_SHIFT_63XX) +#define EAP_MODE_BASIC 0 +#define EAP_MODE_SIMPLIFIED 3 + /************************************************************************* * EEE Configuration Page Registers *************************************************************************/ From 578e1b96fad7402ff7e9c7648c8f1ad0225147c8 Mon Sep 17 00:00:00 2001 From: Rong Zhang Date: Mon, 12 May 2025 23:24:19 +0800 Subject: [PATCH 434/555] HID: bpf: abort dispatch if device destroyed The current HID bpf implementation assumes no output report/request will go through it after hid_bpf_destroy_device() has been called. This leads to a bug that unplugging certain types of HID devices causes a cleaned- up SRCU to be accessed. The bug was previously a hidden failure until a recent x86 percpu change [1] made it access not-present pages. The bug will be triggered if the conditions below are met: A) a device under the driver has some LEDs on B) hid_ll_driver->request() is uninplemented (e.g., logitech-djreceiver) If condition A is met, hidinput_led_worker() is always scheduled *after* hid_bpf_destroy_device(). hid_destroy_device ` hid_bpf_destroy_device ` cleanup_srcu_struct(&hdev->bpf.srcu) ` hid_remove_device ` ... ` led_classdev_unregister ` led_trigger_set(led_cdev, NULL) ` led_set_brightness(led_cdev, LED_OFF) ` ... ` input_inject_event ` input_event_dispose ` hidinput_input_event ` schedule_work(&hid->led_work) [hidinput_led_worker] This is fine when condition B is not met, where hidinput_led_worker() calls hid_ll_driver->request(). This is the case for most HID drivers, which implement it or use the generic one from usbhid. The driver itself or an underlying driver will then abort processing the request. Otherwise, hidinput_led_worker() tries hid_hw_output_report() and leads to the bug. hidinput_led_worker ` hid_hw_output_report ` dispatch_hid_bpf_output_report ` srcu_read_lock(&hdev->bpf.srcu) ` srcu_read_unlock(&hdev->bpf.srcu, idx) The bug has existed since the introduction [2] of dispatch_hid_bpf_output_report(). However, the same bug also exists in dispatch_hid_bpf_raw_requests(), and I've reproduced (no visible effect because of the lack of [1], but confirmed bpf.destroyed == 1) the bug against the commit (i.e., the Fixes:) introducing the function. This is because hidinput_led_worker() falls back to hid_hw_raw_request() when hid_ll_driver->output_report() is uninplemented (e.g., logitech- djreceiver). hidinput_led_worker ` hid_hw_output_report: -ENOSYS ` hid_hw_raw_request ` dispatch_hid_bpf_raw_requests ` srcu_read_lock(&hdev->bpf.srcu) ` srcu_read_unlock(&hdev->bpf.srcu, idx) Fix the issue by returning early in the two mentioned functions if hid_bpf has been marked as destroyed. Though dispatch_hid_bpf_device_event() handles input events, and there is no evidence that it may be called after the destruction, the same check, as a safety net, is also added to it to maintain the consistency among all dispatch functions. The impact of the bug on other architectures is unclear. Even if it acts as a hidden failure, this is still dangerous because it corrupts whatever is on the address calculated by SRCU. Thus, CC'ing the stable list. [1]: commit 9d7de2aa8b41 ("x86/percpu/64: Use relative percpu offsets") [2]: commit 9286675a2aed ("HID: bpf: add HID-BPF hooks for hid_hw_output_report") Closes: https://lore.kernel.org/all/20250506145548.GGaBoi9Jzp3aeJizTR@fat_crate.local/ Fixes: 8bd0488b5ea5 ("HID: bpf: add HID-BPF hooks for hid_hw_raw_requests") Cc: stable@vger.kernel.org Signed-off-by: Rong Zhang Tested-by: Petr Tesarik Link: https://patch.msgid.link/20250512152420.87441-1-i@rong.moe Signed-off-by: Benjamin Tissoires --- drivers/hid/bpf/hid_bpf_dispatch.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/drivers/hid/bpf/hid_bpf_dispatch.c b/drivers/hid/bpf/hid_bpf_dispatch.c index 2e96ec6a3073..9a06f9b0e4ef 100644 --- a/drivers/hid/bpf/hid_bpf_dispatch.c +++ b/drivers/hid/bpf/hid_bpf_dispatch.c @@ -38,6 +38,9 @@ dispatch_hid_bpf_device_event(struct hid_device *hdev, enum hid_report_type type struct hid_bpf_ops *e; int ret; + if (unlikely(hdev->bpf.destroyed)) + return ERR_PTR(-ENODEV); + if (type >= HID_REPORT_TYPES) return ERR_PTR(-EINVAL); @@ -93,6 +96,9 @@ int dispatch_hid_bpf_raw_requests(struct hid_device *hdev, struct hid_bpf_ops *e; int ret, idx; + if (unlikely(hdev->bpf.destroyed)) + return -ENODEV; + if (rtype >= HID_REPORT_TYPES) return -EINVAL; @@ -130,6 +136,9 @@ int dispatch_hid_bpf_output_report(struct hid_device *hdev, struct hid_bpf_ops *e; int ret, idx; + if (unlikely(hdev->bpf.destroyed)) + return -ENODEV; + idx = srcu_read_lock(&hdev->bpf.srcu); list_for_each_entry_srcu(e, &hdev->bpf.prog_list, list, srcu_read_lock_held(&hdev->bpf.srcu)) { From d2062cc1b1c367d5d019f595ef860159e1301351 Mon Sep 17 00:00:00 2001 From: Ashish Kalra Date: Mon, 28 Apr 2025 21:41:51 +0000 Subject: [PATCH 435/555] x86/sev: Do not touch VMSA pages during SNP guest memory kdump When kdump is running makedumpfile to generate vmcore and dump SNP guest memory it touches the VMSA page of the vCPU executing kdump. It then results in unrecoverable #NPF/RMP faults as the VMSA page is marked busy/in-use when the vCPU is running and subsequently a causes guest softlockup/hang. Additionally, other APs may be halted in guest mode and their VMSA pages are marked busy and touching these VMSA pages during guest memory dump will also cause #NPF. Issue AP_DESTROY GHCB calls on other APs to ensure they are kicked out of guest mode and then clear the VMSA bit on their VMSA pages. If the vCPU running kdump is an AP, mark it's VMSA page as offline to ensure that makedumpfile excludes that page while dumping guest memory. Fixes: 3074152e56c9 ("x86/sev: Convert shared memory back to private on kexec") Signed-off-by: Ashish Kalra Signed-off-by: Borislav Petkov (AMD) Reviewed-by: Pankaj Gupta Reviewed-by: Tom Lendacky Tested-by: Srikanth Aithal Cc: stable@vger.kernel.org Link: https://lore.kernel.org/20250428214151.155464-1-Ashish.Kalra@amd.com --- arch/x86/coco/sev/core.c | 244 +++++++++++++++++++++++++-------------- 1 file changed, 158 insertions(+), 86 deletions(-) diff --git a/arch/x86/coco/sev/core.c b/arch/x86/coco/sev/core.c index b0c1a7a57497..41060ba41b5c 100644 --- a/arch/x86/coco/sev/core.c +++ b/arch/x86/coco/sev/core.c @@ -959,6 +959,102 @@ void snp_accept_memory(phys_addr_t start, phys_addr_t end) set_pages_state(vaddr, npages, SNP_PAGE_STATE_PRIVATE); } +static int vmgexit_ap_control(u64 event, struct sev_es_save_area *vmsa, u32 apic_id) +{ + bool create = event != SVM_VMGEXIT_AP_DESTROY; + struct ghcb_state state; + unsigned long flags; + struct ghcb *ghcb; + int ret = 0; + + local_irq_save(flags); + + ghcb = __sev_get_ghcb(&state); + + vc_ghcb_invalidate(ghcb); + + if (create) + ghcb_set_rax(ghcb, vmsa->sev_features); + + ghcb_set_sw_exit_code(ghcb, SVM_VMGEXIT_AP_CREATION); + ghcb_set_sw_exit_info_1(ghcb, + ((u64)apic_id << 32) | + ((u64)snp_vmpl << 16) | + event); + ghcb_set_sw_exit_info_2(ghcb, __pa(vmsa)); + + sev_es_wr_ghcb_msr(__pa(ghcb)); + VMGEXIT(); + + if (!ghcb_sw_exit_info_1_is_valid(ghcb) || + lower_32_bits(ghcb->save.sw_exit_info_1)) { + pr_err("SNP AP %s error\n", (create ? "CREATE" : "DESTROY")); + ret = -EINVAL; + } + + __sev_put_ghcb(&state); + + local_irq_restore(flags); + + return ret; +} + +static int snp_set_vmsa(void *va, void *caa, int apic_id, bool make_vmsa) +{ + int ret; + + if (snp_vmpl) { + struct svsm_call call = {}; + unsigned long flags; + + local_irq_save(flags); + + call.caa = this_cpu_read(svsm_caa); + call.rcx = __pa(va); + + if (make_vmsa) { + /* Protocol 0, Call ID 2 */ + call.rax = SVSM_CORE_CALL(SVSM_CORE_CREATE_VCPU); + call.rdx = __pa(caa); + call.r8 = apic_id; + } else { + /* Protocol 0, Call ID 3 */ + call.rax = SVSM_CORE_CALL(SVSM_CORE_DELETE_VCPU); + } + + ret = svsm_perform_call_protocol(&call); + + local_irq_restore(flags); + } else { + /* + * If the kernel runs at VMPL0, it can change the VMSA + * bit for a page using the RMPADJUST instruction. + * However, for the instruction to succeed it must + * target the permissions of a lesser privileged (higher + * numbered) VMPL level, so use VMPL1. + */ + u64 attrs = 1; + + if (make_vmsa) + attrs |= RMPADJUST_VMSA_PAGE_BIT; + + ret = rmpadjust((unsigned long)va, RMP_PG_SIZE_4K, attrs); + } + + return ret; +} + +static void snp_cleanup_vmsa(struct sev_es_save_area *vmsa, int apic_id) +{ + int err; + + err = snp_set_vmsa(vmsa, NULL, apic_id, false); + if (err) + pr_err("clear VMSA page failed (%u), leaking page\n", err); + else + free_page((unsigned long)vmsa); +} + static void set_pte_enc(pte_t *kpte, int level, void *va) { struct pte_enc_desc d = { @@ -1055,6 +1151,65 @@ void snp_kexec_begin(void) pr_warn("Failed to stop shared<->private conversions\n"); } +/* + * Shutdown all APs except the one handling kexec/kdump and clearing + * the VMSA tag on AP's VMSA pages as they are not being used as + * VMSA page anymore. + */ +static void shutdown_all_aps(void) +{ + struct sev_es_save_area *vmsa; + int apic_id, this_cpu, cpu; + + this_cpu = get_cpu(); + + /* + * APs are already in HLT loop when enc_kexec_finish() callback + * is invoked. + */ + for_each_present_cpu(cpu) { + vmsa = per_cpu(sev_vmsa, cpu); + + /* + * The BSP or offlined APs do not have guest allocated VMSA + * and there is no need to clear the VMSA tag for this page. + */ + if (!vmsa) + continue; + + /* + * Cannot clear the VMSA tag for the currently running vCPU. + */ + if (this_cpu == cpu) { + unsigned long pa; + struct page *p; + + pa = __pa(vmsa); + /* + * Mark the VMSA page of the running vCPU as offline + * so that is excluded and not touched by makedumpfile + * while generating vmcore during kdump. + */ + p = pfn_to_online_page(pa >> PAGE_SHIFT); + if (p) + __SetPageOffline(p); + continue; + } + + apic_id = cpuid_to_apicid[cpu]; + + /* + * Issue AP destroy to ensure AP gets kicked out of guest mode + * to allow using RMPADJUST to remove the VMSA tag on it's + * VMSA page. + */ + vmgexit_ap_control(SVM_VMGEXIT_AP_DESTROY, vmsa, apic_id); + snp_cleanup_vmsa(vmsa, apic_id); + } + + put_cpu(); +} + void snp_kexec_finish(void) { struct sev_es_runtime_data *data; @@ -1069,6 +1224,8 @@ void snp_kexec_finish(void) if (!IS_ENABLED(CONFIG_KEXEC_CORE)) return; + shutdown_all_aps(); + unshare_all_memory(); /* @@ -1090,51 +1247,6 @@ void snp_kexec_finish(void) } } -static int snp_set_vmsa(void *va, void *caa, int apic_id, bool make_vmsa) -{ - int ret; - - if (snp_vmpl) { - struct svsm_call call = {}; - unsigned long flags; - - local_irq_save(flags); - - call.caa = this_cpu_read(svsm_caa); - call.rcx = __pa(va); - - if (make_vmsa) { - /* Protocol 0, Call ID 2 */ - call.rax = SVSM_CORE_CALL(SVSM_CORE_CREATE_VCPU); - call.rdx = __pa(caa); - call.r8 = apic_id; - } else { - /* Protocol 0, Call ID 3 */ - call.rax = SVSM_CORE_CALL(SVSM_CORE_DELETE_VCPU); - } - - ret = svsm_perform_call_protocol(&call); - - local_irq_restore(flags); - } else { - /* - * If the kernel runs at VMPL0, it can change the VMSA - * bit for a page using the RMPADJUST instruction. - * However, for the instruction to succeed it must - * target the permissions of a lesser privileged (higher - * numbered) VMPL level, so use VMPL1. - */ - u64 attrs = 1; - - if (make_vmsa) - attrs |= RMPADJUST_VMSA_PAGE_BIT; - - ret = rmpadjust((unsigned long)va, RMP_PG_SIZE_4K, attrs); - } - - return ret; -} - #define __ATTR_BASE (SVM_SELECTOR_P_MASK | SVM_SELECTOR_S_MASK) #define INIT_CS_ATTRIBS (__ATTR_BASE | SVM_SELECTOR_READ_MASK | SVM_SELECTOR_CODE_MASK) #define INIT_DS_ATTRIBS (__ATTR_BASE | SVM_SELECTOR_WRITE_MASK) @@ -1166,24 +1278,10 @@ static void *snp_alloc_vmsa_page(int cpu) return page_address(p + 1); } -static void snp_cleanup_vmsa(struct sev_es_save_area *vmsa, int apic_id) -{ - int err; - - err = snp_set_vmsa(vmsa, NULL, apic_id, false); - if (err) - pr_err("clear VMSA page failed (%u), leaking page\n", err); - else - free_page((unsigned long)vmsa); -} - static int wakeup_cpu_via_vmgexit(u32 apic_id, unsigned long start_ip) { struct sev_es_save_area *cur_vmsa, *vmsa; - struct ghcb_state state; struct svsm_ca *caa; - unsigned long flags; - struct ghcb *ghcb; u8 sipi_vector; int cpu, ret; u64 cr4; @@ -1297,33 +1395,7 @@ static int wakeup_cpu_via_vmgexit(u32 apic_id, unsigned long start_ip) } /* Issue VMGEXIT AP Creation NAE event */ - local_irq_save(flags); - - ghcb = __sev_get_ghcb(&state); - - vc_ghcb_invalidate(ghcb); - ghcb_set_rax(ghcb, vmsa->sev_features); - ghcb_set_sw_exit_code(ghcb, SVM_VMGEXIT_AP_CREATION); - ghcb_set_sw_exit_info_1(ghcb, - ((u64)apic_id << 32) | - ((u64)snp_vmpl << 16) | - SVM_VMGEXIT_AP_CREATE); - ghcb_set_sw_exit_info_2(ghcb, __pa(vmsa)); - - sev_es_wr_ghcb_msr(__pa(ghcb)); - VMGEXIT(); - - if (!ghcb_sw_exit_info_1_is_valid(ghcb) || - lower_32_bits(ghcb->save.sw_exit_info_1)) { - pr_err("SNP AP Creation error\n"); - ret = -EINVAL; - } - - __sev_put_ghcb(&state); - - local_irq_restore(flags); - - /* Perform cleanup if there was an error */ + ret = vmgexit_ap_control(SVM_VMGEXIT_AP_CREATE, vmsa, apic_id); if (ret) { snp_cleanup_vmsa(vmsa, apic_id); vmsa = NULL; From 82b7f88f2316c5442708daeb0b5ec5aa54c8ff7f Mon Sep 17 00:00:00 2001 From: Ashish Kalra Date: Tue, 6 May 2025 18:35:29 +0000 Subject: [PATCH 436/555] x86/sev: Make sure pages are not skipped during kdump When shared pages are being converted to private during kdump, additional checks are performed. They include handling the case of a GHCB page being contained within a huge page. Currently, this check incorrectly skips a page just below the GHCB page from being transitioned back to private during kdump preparation. This skipped page causes a 0x404 #VC exception when it is accessed later while dumping guest memory for vmcore generation. Correct the range to be checked for GHCB contained in a huge page. Also, ensure that the skipped huge page containing the GHCB page is transitioned back to private by applying the correct address mask later when changing GHCBs to private at end of kdump preparation. [ bp: Massage commit message. ] Fixes: 3074152e56c9 ("x86/sev: Convert shared memory back to private on kexec") Signed-off-by: Ashish Kalra Signed-off-by: Borislav Petkov (AMD) Reviewed-by: Tom Lendacky Tested-by: Srikanth Aithal Cc: stable@vger.kernel.org Link: https://lore.kernel.org/20250506183529.289549-1-Ashish.Kalra@amd.com --- arch/x86/coco/sev/core.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/arch/x86/coco/sev/core.c b/arch/x86/coco/sev/core.c index 41060ba41b5c..36beaac713c1 100644 --- a/arch/x86/coco/sev/core.c +++ b/arch/x86/coco/sev/core.c @@ -1101,7 +1101,8 @@ static void unshare_all_memory(void) data = per_cpu(runtime_data, cpu); ghcb = (unsigned long)&data->ghcb_page; - if (addr <= ghcb && ghcb <= addr + size) { + /* Handle the case of a huge page containing the GHCB page */ + if (addr <= ghcb && ghcb < addr + size) { skipped_addr = true; break; } @@ -1213,8 +1214,8 @@ static void shutdown_all_aps(void) void snp_kexec_finish(void) { struct sev_es_runtime_data *data; + unsigned long size, addr; unsigned int level, cpu; - unsigned long size; struct ghcb *ghcb; pte_t *pte; @@ -1242,8 +1243,10 @@ void snp_kexec_finish(void) ghcb = &data->ghcb_page; pte = lookup_address((unsigned long)ghcb, &level); size = page_level_size(level); - set_pte_enc(pte, level, (void *)ghcb); - snp_set_memory_private((unsigned long)ghcb, (size / PAGE_SIZE)); + /* Handle the case of a huge page containing the GHCB page */ + addr = (unsigned long)ghcb & page_level_mask(level); + set_pte_enc(pte, level, (void *)addr); + snp_set_memory_private(addr, (size / PAGE_SIZE)); } } From a0fa7873f2f869087b1e7793f7fac3713a1e3afe Mon Sep 17 00:00:00 2001 From: Philip Yang Date: Wed, 7 May 2025 11:04:32 -0400 Subject: [PATCH 437/555] drm/amdgpu: csa unmap use uninterruptible lock MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit After process exit to unmap csa and free GPU vm, if signal is accepted and then waiting to take vm lock is interrupted and return, it causes memory leaking and below warning backtrace. Change to use uninterruptible wait lock fix the issue. WARNING: CPU: 69 PID: 167800 at amd/amdgpu/amdgpu_kms.c:1525 amdgpu_driver_postclose_kms+0x294/0x2a0 [amdgpu] Call Trace: drm_file_free.part.0+0x1da/0x230 [drm] drm_close_helper.isra.0+0x65/0x70 [drm] drm_release+0x6a/0x120 [drm] amdgpu_drm_release+0x51/0x60 [amdgpu] __fput+0x9f/0x280 ____fput+0xe/0x20 task_work_run+0x67/0xa0 do_exit+0x217/0x3c0 do_group_exit+0x3b/0xb0 get_signal+0x14a/0x8d0 arch_do_signal_or_restart+0xde/0x100 exit_to_user_mode_loop+0xc1/0x1a0 exit_to_user_mode_prepare+0xf4/0x100 syscall_exit_to_user_mode+0x17/0x40 do_syscall_64+0x69/0xc0 Signed-off-by: Philip Yang Reviewed-by: Christian König Signed-off-by: Alex Deucher (cherry picked from commit 7dbbfb3c171a6f63b01165958629c9c26abf38ab) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c index cfdf558b48b6..02138aa55793 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c @@ -109,7 +109,7 @@ int amdgpu_unmap_static_csa(struct amdgpu_device *adev, struct amdgpu_vm *vm, struct drm_exec exec; int r; - drm_exec_init(&exec, DRM_EXEC_INTERRUPTIBLE_WAIT, 0); + drm_exec_init(&exec, 0, 0); drm_exec_until_all_locked(&exec) { r = amdgpu_vm_lock_pd(vm, &exec, 0); if (likely(!r)) From 2d73b0845ab3963856e857b810600e5594bc29f4 Mon Sep 17 00:00:00 2001 From: Tim Huang Date: Thu, 8 May 2025 13:37:35 +0800 Subject: [PATCH 438/555] drm/amdgpu: fix incorrect MALL size for GFX1151 On GFX1151, the reported MALL cache size reflects only half of its actual size; this adjustment corrects the discrepancy. Signed-off-by: Tim Huang Acked-by: Alex Deucher Reviewed-by: Yifan Zhang Signed-off-by: Alex Deucher (cherry picked from commit 0a5c060b593ad152318f89e5564bfdfcff8a6ac0) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c index e74e26b6a4f2..fec9a007533a 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c @@ -752,6 +752,18 @@ static int gmc_v11_0_sw_init(struct amdgpu_ip_block *ip_block) adev->gmc.vram_type = vram_type; adev->gmc.vram_vendor = vram_vendor; + /* The mall_size is already calculated as mall_size_per_umc * num_umc. + * However, for gfx1151, which features a 2-to-1 UMC mapping, + * the result must be multiplied by 2 to determine the actual mall size. + */ + switch (amdgpu_ip_version(adev, GC_HWIP, 0)) { + case IP_VERSION(11, 5, 1): + adev->gmc.mall_size *= 2; + break; + default: + break; + } + switch (amdgpu_ip_version(adev, GC_HWIP, 0)) { case IP_VERSION(11, 0, 0): case IP_VERSION(11, 0, 1): From d433981385c62c72080e26f1c00a961d18b233be Mon Sep 17 00:00:00 2001 From: Wayne Lin Date: Fri, 25 Apr 2025 14:44:02 +0800 Subject: [PATCH 439/555] drm/amd/display: Correct the reply value when AUX write incomplete [Why] Now forcing aux->transfer to return 0 when incomplete AUX write is inappropriate. It should return bytes have been transferred. [How] aux->transfer is asked not to change original msg except reply field of drm_dp_aux_msg structure. Copy the msg->buffer when it's write request, and overwrite the first byte when sink reply 1 byte indicating partially written byte number. Then we can return the correct value without changing the original msg. Fixes: 3637e457eb00 ("drm/amd/display: Fix wrong handling for AUX_DEFER case") Cc: Mario Limonciello Cc: Alex Deucher Reviewed-by: Ray Wu Signed-off-by: Wayne Lin Signed-off-by: Ray Wu Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher (cherry picked from commit 7ac37f0dcd2e0b729fa7b5513908dc8ab802b540) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 3 ++- .../drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c | 10 ++++++++-- 2 files changed, 10 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index 64df8ca448b3..1525c408d452 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -12763,7 +12763,8 @@ int amdgpu_dm_process_dmub_aux_transfer_sync( /* The reply is stored in the top nibble of the command. */ payload->reply[0] = (adev->dm.dmub_notify->aux_reply.command >> 4) & 0xF; - if (!payload->write && p_notify->aux_reply.length) + /*write req may receive a byte indicating partially written number as well*/ + if (p_notify->aux_reply.length) memcpy(payload->data, p_notify->aux_reply.data, p_notify->aux_reply.length); diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c index 074b79fd5822..c93eef6f7a6b 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c @@ -62,6 +62,7 @@ static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux, enum aux_return_code_type operation_result; struct amdgpu_device *adev; struct ddc_service *ddc; + uint8_t copy[16]; if (WARN_ON(msg->size > 16)) return -E2BIG; @@ -77,6 +78,11 @@ static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux, (msg->request & DP_AUX_I2C_WRITE_STATUS_UPDATE) != 0; payload.defer_delay = 0; + if (payload.write) { + memcpy(copy, msg->buffer, msg->size); + payload.data = copy; + } + result = dc_link_aux_transfer_raw(TO_DM_AUX(aux)->ddc_service, &payload, &operation_result); @@ -100,9 +106,9 @@ static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux, */ if (payload.write && result >= 0) { if (result) { - /*one byte indicating partially written bytes. Force 0 to retry*/ + /*one byte indicating partially written bytes*/ drm_info(adev_to_drm(adev), "amdgpu: AUX partially written\n"); - result = 0; + result = payload.data[0]; } else if (!payload.reply[0]) /*I2C_ACK|AUX_ACK*/ result = msg->size; From 190818d1b653fbf16138286345e67c8c1a83f29f Mon Sep 17 00:00:00 2001 From: Gabe Teeger Date: Fri, 25 Apr 2025 10:31:39 -0400 Subject: [PATCH 440/555] Revert: "drm/amd/display: Enable urgent latency adjustment on DCN35" This reverts commit 756c85e4d0dd ("drm/amd/display: Enable urgent latency adjustment on DCN35") Reason for revert: Negative power impact. Reviewed-by: Nicholas Kazlauskas Signed-off-by: Gabe Teeger Signed-off-by: Ray Wu Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher (cherry picked from commit 9334c491cd8f388232b9a187bf0ddb728482bd6f) --- drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c index d9159ca55412..92f0a099d089 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c @@ -195,9 +195,9 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_5_soc = { .dcn_downspread_percent = 0.5, .gpuvm_min_page_size_bytes = 4096, .hostvm_min_page_size_bytes = 4096, - .do_urgent_latency_adjustment = 1, + .do_urgent_latency_adjustment = 0, .urgent_latency_adjustment_fabric_clock_component_us = 0, - .urgent_latency_adjustment_fabric_clock_reference_mhz = 3000, + .urgent_latency_adjustment_fabric_clock_reference_mhz = 0, }; void dcn35_build_wm_range_table_fpu(struct clk_mgr *clk_mgr) From 874697e127931bf50a37ce9d96ee80f3a08a0c38 Mon Sep 17 00:00:00 2001 From: John Olender Date: Wed, 16 Apr 2025 02:54:26 -0400 Subject: [PATCH 441/555] drm/amd/display: Defer BW-optimization-blocked DRR adjustments [Why & How] Instead of dropping DRR updates, defer them. This fixes issues where monitor continues to see incorrect refresh rate after VRR was turned off by userspace. Fixes: 32953485c558 ("drm/amd/display: Do not update DRR while BW optimizations pending") Link: https://gitlab.freedesktop.org/drm/amd/-/issues/3546 Reviewed-by: Sun peng Li Signed-off-by: John Olender Signed-off-by: Aurabindo Pillai Signed-off-by: Ray Wu Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher (cherry picked from commit 53761b7ecd83e6fbb9f2206f8c980a6aa308c844) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 2 ++ drivers/gpu/drm/amd/display/dc/core/dc.c | 10 +++++++--- 2 files changed, 9 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index 1525c408d452..cc01b9c68b47 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -372,6 +372,8 @@ get_crtc_by_otg_inst(struct amdgpu_device *adev, static inline bool is_dc_timing_adjust_needed(struct dm_crtc_state *old_state, struct dm_crtc_state *new_state) { + if (new_state->stream->adjust.timing_adjust_pending) + return true; if (new_state->freesync_config.state == VRR_STATE_ACTIVE_FIXED) return true; else if (amdgpu_dm_crtc_vrr_active(old_state) != amdgpu_dm_crtc_vrr_active(new_state)) diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index 28d1353f403d..ba4ce8a63158 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -439,9 +439,12 @@ bool dc_stream_adjust_vmin_vmax(struct dc *dc, * Don't adjust DRR while there's bandwidth optimizations pending to * avoid conflicting with firmware updates. */ - if (dc->ctx->dce_version > DCE_VERSION_MAX) - if (dc->optimized_required || dc->wm_optimized_required) + if (dc->ctx->dce_version > DCE_VERSION_MAX) { + if (dc->optimized_required || dc->wm_optimized_required) { + stream->adjust.timing_adjust_pending = true; return false; + } + } dc_exit_ips_for_hw_access(dc); @@ -3168,7 +3171,8 @@ static void copy_stream_update_to_stream(struct dc *dc, if (update->crtc_timing_adjust) { if (stream->adjust.v_total_min != update->crtc_timing_adjust->v_total_min || - stream->adjust.v_total_max != update->crtc_timing_adjust->v_total_max) + stream->adjust.v_total_max != update->crtc_timing_adjust->v_total_max || + stream->adjust.timing_adjust_pending) update->crtc_timing_adjust->timing_adjust_pending = true; stream->adjust = *update->crtc_timing_adjust; update->crtc_timing_adjust->timing_adjust_pending = false; From 3c1a467372e0c356b1d3c59f6d199ed5a6612dd1 Mon Sep 17 00:00:00 2001 From: George Shen Date: Thu, 24 Apr 2025 10:02:59 -0400 Subject: [PATCH 442/555] drm/amd/display: fix link_set_dpms_off multi-display MST corner case [Why & How] When MST config is unplugged/replugged too quickly, it can potentially result in a scenario where previous DC state has not been reset before the HPD link detection sequence begins. In this case, driver will disable the streams/link prior to re-enabling the link for link training. There is a bug in the current logic that does not account for the fact that current_state can be released and cleared prior to swapping to a new state (resulting in the pipe_ctx stream pointers to be cleared) in between disabling streams. To resolve this, cache the original streams prior to committing any stream updates. Reviewed-by: Wenjing Liu Signed-off-by: George Shen Signed-off-by: Ray Wu Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher (cherry picked from commit 1561782686ccc36af844d55d31b44c938dd412dc) --- drivers/gpu/drm/amd/display/dc/link/link_dpms.c | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c index 268626e73c54..53c961f86d43 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c @@ -148,6 +148,7 @@ void link_blank_dp_stream(struct dc_link *link, bool hw_init) void link_set_all_streams_dpms_off_for_link(struct dc_link *link) { struct pipe_ctx *pipes[MAX_PIPES]; + struct dc_stream_state *streams[MAX_PIPES]; struct dc_state *state = link->dc->current_state; uint8_t count; int i; @@ -160,10 +161,18 @@ void link_set_all_streams_dpms_off_for_link(struct dc_link *link) link_get_master_pipes_with_dpms_on(link, state, &count, pipes); + /* The subsequent call to dc_commit_updates_for_stream for a full update + * will release the current state and swap to a new state. Releasing the + * current state results in the stream pointers in the pipe_ctx structs + * to be zero'd. Hence, cache all streams prior to dc_commit_updates_for_stream. + */ + for (i = 0; i < count; i++) + streams[i] = pipes[i]->stream; + for (i = 0; i < count; i++) { - stream_update.stream = pipes[i]->stream; + stream_update.stream = streams[i]; dc_commit_updates_for_stream(link->ctx->dc, NULL, 0, - pipes[i]->stream, &stream_update, + streams[i], &stream_update, state); } From 2ddac70fed50485aa4ae49cdb7478ce41d8d4715 Mon Sep 17 00:00:00 2001 From: Aurabindo Pillai Date: Tue, 29 Apr 2025 05:03:30 +0800 Subject: [PATCH 443/555] drm/amd/display: check stream id dml21 wrapper to get plane_id [Why & How] Fix a false positive warning which occurs due to lack of correct checks when querying plane_id in DML21. This fixes the warning when performing a mode1 reset (cat /sys/kernel/debug/dri/1/amdgpu_gpu_recover): [ 35.751250] WARNING: CPU: 11 PID: 326 at /tmp/amd.PHpyAl7v/amd/amdgpu/../display/dc/dml2/dml2_dc_resource_mgmt.c:91 dml2_map_dc_pipes+0x243d/0x3f40 [amdgpu] [ 35.751434] Modules linked in: amdgpu(OE) amddrm_ttm_helper(OE) amdttm(OE) amddrm_buddy(OE) amdxcp(OE) amddrm_exec(OE) amd_sched(OE) amdkcl(OE) drm_suballoc_helper drm_ttm_helper ttm drm_display_helper cec rc_core i2c_algo_bit rfcomm qrtr cmac algif_hash algif_skcipher af_alg bnep amd_atl intel_rapl_msr intel_rapl_common snd_hda_codec_hdmi snd_hda_intel edac_mce_amd snd_intel_dspcfg snd_intel_sdw_acpi snd_hda_codec kvm_amd snd_hda_core snd_hwdep snd_pcm kvm snd_seq_midi snd_seq_midi_event snd_rawmidi crct10dif_pclmul polyval_clmulni polyval_generic btusb ghash_clmulni_intel sha256_ssse3 btrtl sha1_ssse3 snd_seq btintel aesni_intel btbcm btmtk snd_seq_device crypto_simd sunrpc cryptd bluetooth snd_timer ccp binfmt_misc rapl snd i2c_piix4 wmi_bmof gigabyte_wmi k10temp i2c_smbus soundcore gpio_amdpt mac_hid sch_fq_codel msr parport_pc ppdev lp parport efi_pstore nfnetlink dmi_sysfs ip_tables x_tables autofs4 hid_generic usbhid hid crc32_pclmul igc ahci xhci_pci libahci xhci_pci_renesas video wmi [ 35.751501] CPU: 11 UID: 0 PID: 326 Comm: kworker/u64:9 Tainted: G OE 6.11.0-21-generic #21~24.04.1-Ubuntu [ 35.751504] Tainted: [O]=OOT_MODULE, [E]=UNSIGNED_MODULE [ 35.751505] Hardware name: Gigabyte Technology Co., Ltd. X670E AORUS PRO X/X670E AORUS PRO X, BIOS F30 05/22/2024 [ 35.751506] Workqueue: amdgpu-reset-dev amdgpu_debugfs_reset_work [amdgpu] [ 35.751638] RIP: 0010:dml2_map_dc_pipes+0x243d/0x3f40 [amdgpu] [ 35.751794] Code: 6d 0c 00 00 8b 84 24 88 00 00 00 41 3b 44 9c 20 0f 84 fc 07 00 00 48 83 c3 01 48 83 fb 06 75 b3 4c 8b 64 24 68 4c 8b 6c 24 40 <0f> 0b b8 06 00 00 00 49 8b 94 24 a0 49 00 00 89 c3 83 f8 07 0f 87 [ 35.751796] RSP: 0018:ffffbfa3805d7680 EFLAGS: 00010246 [ 35.751798] RAX: 0000000000010000 RBX: 0000000000000006 RCX: 0000000000000000 [ 35.751799] RDX: 0000000000000000 RSI: 0000000000000005 RDI: 0000000000000000 [ 35.751800] RBP: ffffbfa3805d78f0 R08: 0000000000000000 R09: 0000000000000000 [ 35.751801] R10: 0000000000000000 R11: 0000000000000000 R12: ffffbfa383249000 [ 35.751802] R13: ffffa0e68f280000 R14: ffffbfa383249658 R15: 0000000000000000 [ 35.751803] FS: 0000000000000000(0000) GS:ffffa0edbe580000(0000) knlGS:0000000000000000 [ 35.751804] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 35.751805] CR2: 00005d847ef96c58 CR3: 000000041de3e000 CR4: 0000000000f50ef0 [ 35.751806] PKRU: 55555554 [ 35.751807] Call Trace: [ 35.751810] [ 35.751816] ? show_regs+0x6c/0x80 [ 35.751820] ? __warn+0x88/0x140 [ 35.751822] ? dml2_map_dc_pipes+0x243d/0x3f40 [amdgpu] [ 35.751964] ? report_bug+0x182/0x1b0 [ 35.751969] ? handle_bug+0x6e/0xb0 [ 35.751972] ? exc_invalid_op+0x18/0x80 [ 35.751974] ? asm_exc_invalid_op+0x1b/0x20 [ 35.751978] ? dml2_map_dc_pipes+0x243d/0x3f40 [amdgpu] [ 35.752117] ? math_pow+0x48/0xa0 [amdgpu] [ 35.752256] ? srso_alias_return_thunk+0x5/0xfbef5 [ 35.752260] ? math_pow+0x48/0xa0 [amdgpu] [ 35.752400] ? srso_alias_return_thunk+0x5/0xfbef5 [ 35.752403] ? math_pow+0x11/0xa0 [amdgpu] [ 35.752524] ? srso_alias_return_thunk+0x5/0xfbef5 [ 35.752526] ? core_dcn4_mode_programming+0xe4d/0x20d0 [amdgpu] [ 35.752663] ? srso_alias_return_thunk+0x5/0xfbef5 [ 35.752669] dml21_validate+0x3d4/0x980 [amdgpu] Reviewed-by: Austin Zheng Signed-off-by: Aurabindo Pillai Signed-off-by: Ray Wu Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher (cherry picked from commit f8ad62c0a93e5dd94243e10f1b742232e4d6411e) --- .../dc/dml2/dml21/dml21_translation_helper.c | 20 ++++++++++--------- 1 file changed, 11 insertions(+), 9 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c index 0c8ec30ea672..731fbd4bc600 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c @@ -910,7 +910,7 @@ static void populate_dml21_plane_config_from_plane_state(struct dml2_context *dm } //TODO : Could be possibly moved to a common helper layer. -static bool dml21_wrapper_get_plane_id(const struct dc_state *context, const struct dc_plane_state *plane, unsigned int *plane_id) +static bool dml21_wrapper_get_plane_id(const struct dc_state *context, unsigned int stream_id, const struct dc_plane_state *plane, unsigned int *plane_id) { int i, j; @@ -918,10 +918,12 @@ static bool dml21_wrapper_get_plane_id(const struct dc_state *context, const str return false; for (i = 0; i < context->stream_count; i++) { - for (j = 0; j < context->stream_status[i].plane_count; j++) { - if (context->stream_status[i].plane_states[j] == plane) { - *plane_id = (i << 16) | j; - return true; + if (context->streams[i]->stream_id == stream_id) { + for (j = 0; j < context->stream_status[i].plane_count; j++) { + if (context->stream_status[i].plane_states[j] == plane) { + *plane_id = (i << 16) | j; + return true; + } } } } @@ -944,14 +946,14 @@ static unsigned int map_stream_to_dml21_display_cfg(const struct dml2_context *d return location; } -static unsigned int map_plane_to_dml21_display_cfg(const struct dml2_context *dml_ctx, +static unsigned int map_plane_to_dml21_display_cfg(const struct dml2_context *dml_ctx, unsigned int stream_id, const struct dc_plane_state *plane, const struct dc_state *context) { unsigned int plane_id; int i = 0; int location = -1; - if (!dml21_wrapper_get_plane_id(context, plane, &plane_id)) { + if (!dml21_wrapper_get_plane_id(context, stream_id, plane, &plane_id)) { ASSERT(false); return -1; } @@ -1037,7 +1039,7 @@ bool dml21_map_dc_state_into_dml_display_cfg(const struct dc *in_dc, struct dc_s dml_dispcfg->plane_descriptors[disp_cfg_plane_location].stream_index = disp_cfg_stream_location; } else { for (plane_index = 0; plane_index < context->stream_status[stream_index].plane_count; plane_index++) { - disp_cfg_plane_location = map_plane_to_dml21_display_cfg(dml_ctx, context->stream_status[stream_index].plane_states[plane_index], context); + disp_cfg_plane_location = map_plane_to_dml21_display_cfg(dml_ctx, context->streams[stream_index]->stream_id, context->stream_status[stream_index].plane_states[plane_index], context); if (disp_cfg_plane_location < 0) disp_cfg_plane_location = dml_dispcfg->num_planes++; @@ -1048,7 +1050,7 @@ bool dml21_map_dc_state_into_dml_display_cfg(const struct dc *in_dc, struct dc_s populate_dml21_plane_config_from_plane_state(dml_ctx, &dml_dispcfg->plane_descriptors[disp_cfg_plane_location], context->stream_status[stream_index].plane_states[plane_index], context, stream_index); dml_dispcfg->plane_descriptors[disp_cfg_plane_location].stream_index = disp_cfg_stream_location; - if (dml21_wrapper_get_plane_id(context, context->stream_status[stream_index].plane_states[plane_index], &dml_ctx->v21.dml_to_dc_pipe_mapping.disp_cfg_to_plane_id[disp_cfg_plane_location])) + if (dml21_wrapper_get_plane_id(context, context->streams[stream_index]->stream_id, context->stream_status[stream_index].plane_states[plane_index], &dml_ctx->v21.dml_to_dc_pipe_mapping.disp_cfg_to_plane_id[disp_cfg_plane_location])) dml_ctx->v21.dml_to_dc_pipe_mapping.disp_cfg_to_plane_id_valid[disp_cfg_plane_location] = true; /* apply forced pstate policy */ From a3b7e65b6be59e686e163fa1ceb0922f996897c2 Mon Sep 17 00:00:00 2001 From: Melissa Wen Date: Wed, 30 Apr 2025 11:11:47 -0300 Subject: [PATCH 444/555] drm/amd/display: Fix null check of pipe_ctx->plane_state for update_dchubp_dpp Similar to commit 6a057072ddd1 ("drm/amd/display: Fix null check for pipe_ctx->plane_state in dcn20_program_pipe") that addresses a null pointer dereference on dcn20_update_dchubp_dpp. This is the same function hooked for update_dchubp_dpp in dcn401, with the same issue. Fix possible null pointer deference on dcn401_program_pipe too. Fixes: 63ab80d9ac0a ("drm/amd/display: DML2.1 Post-Si Cleanup") Signed-off-by: Melissa Wen Reviewed-by: Alex Hung Signed-off-by: Alex Deucher (cherry picked from commit d8d47f739752227957d8efc0cb894761bfe1d879) --- drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c index 5489f3d431f6..3af6a3402b89 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c @@ -1980,9 +1980,9 @@ void dcn401_program_pipe( dc->res_pool->hubbub, pipe_ctx->plane_res.hubp->inst, pipe_ctx->hubp_regs.det_size); } - if (pipe_ctx->update_flags.raw || - (pipe_ctx->plane_state && pipe_ctx->plane_state->update_flags.raw) || - pipe_ctx->stream->update_flags.raw) + if (pipe_ctx->plane_state && (pipe_ctx->update_flags.raw || + pipe_ctx->plane_state->update_flags.raw || + pipe_ctx->stream->update_flags.raw)) dc->hwss.update_dchubp_dpp(dc, pipe_ctx, context); if (pipe_ctx->plane_state && (pipe_ctx->update_flags.bits.enable || From d33724ffb743d3d2698bd969e29253ae0cff9739 Mon Sep 17 00:00:00 2001 From: Wayne Lin Date: Tue, 13 May 2025 11:20:24 +0800 Subject: [PATCH 445/555] drm/amd/display: Avoid flooding unnecessary info messages It's expected that we'll encounter temporary exceptions during aux transactions. Adjust logging from drm_info to drm_dbg_dp to prevent flooding with unnecessary log messages. Fixes: 3637e457eb00 ("drm/amd/display: Fix wrong handling for AUX_DEFER case") Cc: Mario Limonciello Cc: Alex Deucher Signed-off-by: Wayne Lin Acked-by: Alex Deucher Link: https://lore.kernel.org/r/20250513032026.838036-1-Wayne.Lin@amd.com Signed-off-by: Mario Limonciello Signed-off-by: Alex Deucher (cherry picked from commit 9a9c3e1fe5256da14a0a307dff0478f90c55fc8c) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c index c93eef6f7a6b..5cdbc86ef8f5 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c @@ -107,7 +107,7 @@ static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux, if (payload.write && result >= 0) { if (result) { /*one byte indicating partially written bytes*/ - drm_info(adev_to_drm(adev), "amdgpu: AUX partially written\n"); + drm_dbg_dp(adev_to_drm(adev), "amdgpu: AUX partially written\n"); result = payload.data[0]; } else if (!payload.reply[0]) /*I2C_ACK|AUX_ACK*/ @@ -133,11 +133,11 @@ static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux, break; } - drm_info(adev_to_drm(adev), "amdgpu: DP AUX transfer fail:%d\n", operation_result); + drm_dbg_dp(adev_to_drm(adev), "amdgpu: DP AUX transfer fail:%d\n", operation_result); } if (payload.reply[0]) - drm_info(adev_to_drm(adev), "amdgpu: AUX reply command not ACK: 0x%02x.", + drm_dbg_dp(adev_to_drm(adev), "amdgpu: AUX reply command not ACK: 0x%02x.", payload.reply[0]); return result; From 2d2f82e1a8a118d26c2489674b6f6cbb41d04376 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= Date: Thu, 8 May 2025 13:29:31 +0200 Subject: [PATCH 446/555] drm/xe: Fix the gem shrinker name MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The xe buffer object shrinker name is visible in the /shrinker directory and most if not all other shinkers follow a naming convention that looks like -_: Follow the same convention for xe, changing the name to drm-xe_gem:. Other shrinkers typically use the device node for but since drm drivers typically don't have a single unique device- node, instead use the unique name in the drm device. Fixes: 00c8efc3180f ("drm/xe: Add a shrinker for xe bos") Cc: Matthew Brost Signed-off-by: Thomas Hellström Reviewed-by: Francois Dugast Link: https://lore.kernel.org/r/20250508112931.3347-1-thomas.hellstrom@linux.intel.com (cherry picked from commit 243bf99e2fe75edf8df1711c1377b6fc020b806c) Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/xe_shrinker.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/xe/xe_shrinker.c b/drivers/gpu/drm/xe/xe_shrinker.c index 8184390f9c7b..86d47aaf0358 100644 --- a/drivers/gpu/drm/xe/xe_shrinker.c +++ b/drivers/gpu/drm/xe/xe_shrinker.c @@ -227,7 +227,7 @@ struct xe_shrinker *xe_shrinker_create(struct xe_device *xe) if (!shrinker) return ERR_PTR(-ENOMEM); - shrinker->shrink = shrinker_alloc(0, "xe system shrinker"); + shrinker->shrink = shrinker_alloc(0, "drm-xe_gem:%s", xe->drm.unique); if (!shrinker->shrink) { kfree(shrinker); return ERR_PTR(-ENOMEM); From 24ee8d9432b5744fce090af3d829a39aa4abf63f Mon Sep 17 00:00:00 2001 From: Yazen Ghannam Date: Tue, 13 May 2025 20:48:57 +0000 Subject: [PATCH 447/555] x86/CPU/AMD: Add X86_FEATURE_ZEN6 Add a synthetic feature flag for Zen6. [ bp: Move the feature flag to a free slot and avoid future merge conflicts from incoming stuff. ] Signed-off-by: Yazen Ghannam Signed-off-by: Borislav Petkov (AMD) Link: https://lore.kernel.org/20250513204857.3376577-1-yazen.ghannam@amd.com --- arch/x86/include/asm/cpufeatures.h | 2 +- arch/x86/kernel/cpu/amd.c | 5 +++++ 2 files changed, 6 insertions(+), 1 deletion(-) diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpufeatures.h index 6c2c152d8a67..9bb17c714f4a 100644 --- a/arch/x86/include/asm/cpufeatures.h +++ b/arch/x86/include/asm/cpufeatures.h @@ -75,7 +75,7 @@ #define X86_FEATURE_CENTAUR_MCR ( 3*32+ 3) /* "centaur_mcr" Centaur MCRs (= MTRRs) */ #define X86_FEATURE_K8 ( 3*32+ 4) /* Opteron, Athlon64 */ #define X86_FEATURE_ZEN5 ( 3*32+ 5) /* CPU based on Zen5 microarchitecture */ -/* Free ( 3*32+ 6) */ +#define X86_FEATURE_ZEN6 ( 3*32+ 6) /* CPU based on Zen6 microarchitecture */ /* Free ( 3*32+ 7) */ #define X86_FEATURE_CONSTANT_TSC ( 3*32+ 8) /* "constant_tsc" TSC ticks at a constant rate */ #define X86_FEATURE_UP ( 3*32+ 9) /* "up" SMP kernel running on UP */ diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 2b36379ff675..4e06baab40bb 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -472,6 +472,11 @@ static void bsp_init_amd(struct cpuinfo_x86 *c) case 0x60 ... 0x7f: setup_force_cpu_cap(X86_FEATURE_ZEN5); break; + case 0x50 ... 0x5f: + case 0x90 ... 0xaf: + case 0xc0 ... 0xcf: + setup_force_cpu_cap(X86_FEATURE_ZEN6); + break; default: goto warn; } From 9f35e33144ae5377d6a8de86dd3bd4d995c6ac65 Mon Sep 17 00:00:00 2001 From: Eric Biggers Date: Mon, 12 May 2025 19:58:39 -0700 Subject: [PATCH 448/555] x86/its: Fix build errors when CONFIG_MODULES=n Fix several build errors when CONFIG_MODULES=n, including the following: ../arch/x86/kernel/alternative.c:195:25: error: incomplete definition of type 'struct module' 195 | for (int i = 0; i < mod->its_num_pages; i++) { Fixes: 872df34d7c51 ("x86/its: Use dynamic thunks for indirect branches") Cc: stable@vger.kernel.org Signed-off-by: Eric Biggers Acked-by: Dave Hansen Tested-by: Steven Rostedt (Google) Reviewed-by: Alexandre Chartre Signed-off-by: Linus Torvalds --- arch/x86/kernel/alternative.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c index 48fd04e90114..45bcff181cba 100644 --- a/arch/x86/kernel/alternative.c +++ b/arch/x86/kernel/alternative.c @@ -133,7 +133,9 @@ static bool cfi_paranoid __ro_after_init; #ifdef CONFIG_MITIGATION_ITS +#ifdef CONFIG_MODULES static struct module *its_mod; +#endif static void *its_page; static unsigned int its_offset; @@ -171,6 +173,7 @@ static void *its_init_thunk(void *thunk, int reg) return thunk + offset; } +#ifdef CONFIG_MODULES void its_init_mod(struct module *mod) { if (!cpu_feature_enabled(X86_FEATURE_INDIRECT_THUNK_ITS)) @@ -209,6 +212,7 @@ void its_free_mod(struct module *mod) } kfree(mod->its_page_array); } +#endif /* CONFIG_MODULES */ static void *its_alloc(void) { @@ -217,6 +221,7 @@ static void *its_alloc(void) if (!page) return NULL; +#ifdef CONFIG_MODULES if (its_mod) { void *tmp = krealloc(its_mod->its_page_array, (its_mod->its_num_pages+1) * sizeof(void *), @@ -229,6 +234,7 @@ static void *its_alloc(void) execmem_make_temp_rw(page, PAGE_SIZE); } +#endif /* CONFIG_MODULES */ return no_free_ptr(page); } From 396786af1cea5ca1c041f113a6002d62f277ccb0 Mon Sep 17 00:00:00 2001 From: Lukas Wunner Date: Sun, 11 May 2025 07:12:00 +0200 Subject: [PATCH 449/555] tools: ynl-gen: Allow multi-attr without nested-attributes again Since commit ce6cb8113c84 ("tools: ynl-gen: individually free previous values on double set"), specifying the "multi-attr" property raises an error unless the "nested-attributes" property is specified as well: File "tools/net/ynl/./pyynl/ynl_gen_c.py", line 1147, in _load_nested_sets child = self.pure_nested_structs.get(nested) ^^^^^^ UnboundLocalError: cannot access local variable 'nested' where it is not associated with a value This appears to be a bug since there are existing specs which omit "nested-attributes" on "multi-attr" attributes. Also, according to Documentation/userspace-api/netlink/specs.rst, multi-attr "is the recommended way of implementing arrays (no extra nesting)", suggesting that nesting should even be avoided in favor of multi-attr. Fix the indentation of the if-block introduced by the commit to avoid the error. Fixes: ce6cb8113c84 ("tools: ynl-gen: individually free previous values on double set") Signed-off-by: Lukas Wunner Link: https://patch.msgid.link/d6b58684b7e5bfb628f7313e6893d0097904e1d1.1746940107.git.lukas@wunner.de Signed-off-by: Jakub Kicinski --- tools/net/ynl/pyynl/ynl_gen_c.py | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/tools/net/ynl/pyynl/ynl_gen_c.py b/tools/net/ynl/pyynl/ynl_gen_c.py index 30c0a34b2784..a7f08edbc235 100755 --- a/tools/net/ynl/pyynl/ynl_gen_c.py +++ b/tools/net/ynl/pyynl/ynl_gen_c.py @@ -1143,10 +1143,9 @@ class Family(SpecFamily): self.pure_nested_structs[nested].request = True if attr in rs_members['reply']: self.pure_nested_structs[nested].reply = True - - if spec.is_multi_val(): - child = self.pure_nested_structs.get(nested) - child.in_multi_val = True + if spec.is_multi_val(): + child = self.pure_nested_structs.get(nested) + child.in_multi_val = True self._sort_pure_types() From 7fd7ad6f36af36f30a06d165eff3780cb139fa79 Mon Sep 17 00:00:00 2001 From: Konstantin Shkolnyy Date: Wed, 7 May 2025 10:14:56 -0500 Subject: [PATCH 450/555] vsock/test: Fix occasional failure in SIOCOUTQ tests These tests: "SOCK_STREAM ioctl(SIOCOUTQ) 0 unsent bytes" "SOCK_SEQPACKET ioctl(SIOCOUTQ) 0 unsent bytes" output: "Unexpected 'SIOCOUTQ' value, expected 0, got 64 (CLIENT)". They test that the SIOCOUTQ ioctl reports 0 unsent bytes after the data have been received by the other side. However, sometimes there is a delay in updating this "unsent bytes" counter, and the test fails even though the counter properly goes to 0 several milliseconds later. The delay occurs in the kernel because the used buffer notification callback virtio_vsock_tx_done(), called upon receipt of the data by the other side, doesn't update the counter itself. It delegates that to a kernel thread (via vsock->tx_work). Sometimes that thread is delayed more than the test expects. Change the test to poll SIOCOUTQ until it returns 0 or a timeout occurs. Signed-off-by: Konstantin Shkolnyy Reviewed-by: Stefano Garzarella Fixes: 18ee44ce97c1 ("test/vsock: add ioctl unsent bytes test") Link: https://patch.msgid.link/20250507151456.2577061-1-kshk@linux.ibm.com Signed-off-by: Jakub Kicinski --- tools/testing/vsock/vsock_test.c | 28 ++++++++++++++++------------ 1 file changed, 16 insertions(+), 12 deletions(-) diff --git a/tools/testing/vsock/vsock_test.c b/tools/testing/vsock/vsock_test.c index d0f6d253ac72..613551132a96 100644 --- a/tools/testing/vsock/vsock_test.c +++ b/tools/testing/vsock/vsock_test.c @@ -1264,21 +1264,25 @@ static void test_unsent_bytes_client(const struct test_opts *opts, int type) send_buf(fd, buf, sizeof(buf), 0, sizeof(buf)); control_expectln("RECEIVED"); - ret = ioctl(fd, SIOCOUTQ, &sock_bytes_unsent); - if (ret < 0) { - if (errno == EOPNOTSUPP) { - fprintf(stderr, "Test skipped, SIOCOUTQ not supported.\n"); - } else { + /* SIOCOUTQ isn't guaranteed to instantly track sent data. Even though + * the "RECEIVED" message means that the other side has received the + * data, there can be a delay in our kernel before updating the "unsent + * bytes" counter. Repeat SIOCOUTQ until it returns 0. + */ + timeout_begin(TIMEOUT); + do { + ret = ioctl(fd, SIOCOUTQ, &sock_bytes_unsent); + if (ret < 0) { + if (errno == EOPNOTSUPP) { + fprintf(stderr, "Test skipped, SIOCOUTQ not supported.\n"); + break; + } perror("ioctl"); exit(EXIT_FAILURE); } - } else if (ret == 0 && sock_bytes_unsent != 0) { - fprintf(stderr, - "Unexpected 'SIOCOUTQ' value, expected 0, got %i\n", - sock_bytes_unsent); - exit(EXIT_FAILURE); - } - + timeout_check("SIOCOUTQ"); + } while (sock_bytes_unsent != 0); + timeout_end(); close(fd); } From 588431474eb7572e57a927fa8558c9ba2f8af143 Mon Sep 17 00:00:00 2001 From: Carolina Jubran Date: Sun, 11 May 2025 13:15:52 +0300 Subject: [PATCH 451/555] net/mlx5e: Disable MACsec offload for uplink representor profile MACsec offload is not supported in switchdev mode for uplink representors. When switching to the uplink representor profile, the MACsec offload feature must be cleared from the netdevice's features. If left enabled, attempts to add offloads result in a null pointer dereference, as the uplink representor does not support MACsec offload even though the feature bit remains set. Clear NETIF_F_HW_MACSEC in mlx5e_fix_uplink_rep_features(). Kernel log: Oops: general protection fault, probably for non-canonical address 0xdffffc000000000f: 0000 [#1] SMP KASAN KASAN: null-ptr-deref in range [0x0000000000000078-0x000000000000007f] CPU: 29 UID: 0 PID: 4714 Comm: ip Not tainted 6.14.0-rc4_for_upstream_debug_2025_03_02_17_35 #1 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.16.0-0-gd239552ce722-prebuilt.qemu.org 04/01/2014 RIP: 0010:__mutex_lock+0x128/0x1dd0 Code: d0 7c 08 84 d2 0f 85 ad 15 00 00 8b 35 91 5c fe 03 85 f6 75 29 49 8d 7e 60 48 b8 00 00 00 00 00 fc ff df 48 89 fa 48 c1 ea 03 <80> 3c 02 00 0f 85 a6 15 00 00 4d 3b 76 60 0f 85 fd 0b 00 00 65 ff RSP: 0018:ffff888147a4f160 EFLAGS: 00010206 RAX: dffffc0000000000 RBX: 0000000000000000 RCX: 0000000000000001 RDX: 000000000000000f RSI: 0000000000000000 RDI: 0000000000000078 RBP: ffff888147a4f2e0 R08: ffffffffa05d2c19 R09: 0000000000000000 R10: 0000000000000001 R11: 0000000000000000 R12: 0000000000000000 R13: dffffc0000000000 R14: 0000000000000018 R15: ffff888152de0000 FS: 00007f855e27d800(0000) GS:ffff88881ee80000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00000000004e5768 CR3: 000000013ae7c005 CR4: 0000000000372eb0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe07f0 DR7: 0000000000000400 Call Trace: ? die_addr+0x3d/0xa0 ? exc_general_protection+0x144/0x220 ? asm_exc_general_protection+0x22/0x30 ? mlx5e_macsec_add_secy+0xf9/0x700 [mlx5_core] ? __mutex_lock+0x128/0x1dd0 ? lockdep_set_lock_cmp_fn+0x190/0x190 ? mlx5e_macsec_add_secy+0xf9/0x700 [mlx5_core] ? mutex_lock_io_nested+0x1ae0/0x1ae0 ? lock_acquire+0x1c2/0x530 ? macsec_upd_offload+0x145/0x380 ? lockdep_hardirqs_on_prepare+0x400/0x400 ? kasan_save_stack+0x30/0x40 ? kasan_save_stack+0x20/0x40 ? kasan_save_track+0x10/0x30 ? __kasan_kmalloc+0x77/0x90 ? __kmalloc_noprof+0x249/0x6b0 ? genl_family_rcv_msg_attrs_parse.constprop.0+0xb5/0x240 ? mlx5e_macsec_add_secy+0xf9/0x700 [mlx5_core] mlx5e_macsec_add_secy+0xf9/0x700 [mlx5_core] ? mlx5e_macsec_add_rxsa+0x11a0/0x11a0 [mlx5_core] macsec_update_offload+0x26c/0x820 ? macsec_set_mac_address+0x4b0/0x4b0 ? lockdep_hardirqs_on_prepare+0x284/0x400 ? _raw_spin_unlock_irqrestore+0x47/0x50 macsec_upd_offload+0x2c8/0x380 ? macsec_update_offload+0x820/0x820 ? __nla_parse+0x22/0x30 ? genl_family_rcv_msg_attrs_parse.constprop.0+0x15e/0x240 genl_family_rcv_msg_doit+0x1cc/0x2a0 ? genl_family_rcv_msg_attrs_parse.constprop.0+0x240/0x240 ? cap_capable+0xd4/0x330 genl_rcv_msg+0x3ea/0x670 ? genl_family_rcv_msg_dumpit+0x2a0/0x2a0 ? lockdep_set_lock_cmp_fn+0x190/0x190 ? macsec_update_offload+0x820/0x820 netlink_rcv_skb+0x12b/0x390 ? genl_family_rcv_msg_dumpit+0x2a0/0x2a0 ? netlink_ack+0xd80/0xd80 ? rwsem_down_read_slowpath+0xf90/0xf90 ? netlink_deliver_tap+0xcd/0xac0 ? netlink_deliver_tap+0x155/0xac0 ? _copy_from_iter+0x1bb/0x12c0 genl_rcv+0x24/0x40 netlink_unicast+0x440/0x700 ? netlink_attachskb+0x760/0x760 ? lock_acquire+0x1c2/0x530 ? __might_fault+0xbb/0x170 netlink_sendmsg+0x749/0xc10 ? netlink_unicast+0x700/0x700 ? __might_fault+0xbb/0x170 ? netlink_unicast+0x700/0x700 __sock_sendmsg+0xc5/0x190 ____sys_sendmsg+0x53f/0x760 ? import_iovec+0x7/0x10 ? kernel_sendmsg+0x30/0x30 ? __copy_msghdr+0x3c0/0x3c0 ? filter_irq_stacks+0x90/0x90 ? stack_depot_save_flags+0x28/0xa30 ___sys_sendmsg+0xeb/0x170 ? kasan_save_stack+0x30/0x40 ? copy_msghdr_from_user+0x110/0x110 ? do_syscall_64+0x6d/0x140 ? lock_acquire+0x1c2/0x530 ? __virt_addr_valid+0x116/0x3b0 ? __virt_addr_valid+0x1da/0x3b0 ? lock_downgrade+0x680/0x680 ? __delete_object+0x21/0x50 __sys_sendmsg+0xf7/0x180 ? __sys_sendmsg_sock+0x20/0x20 ? kmem_cache_free+0x14c/0x4e0 ? __x64_sys_close+0x78/0xd0 do_syscall_64+0x6d/0x140 entry_SYSCALL_64_after_hwframe+0x4b/0x53 RIP: 0033:0x7f855e113367 Code: 0e 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff eb b9 0f 1f 00 f3 0f 1e fa 64 8b 04 25 18 00 00 00 85 c0 75 10 b8 2e 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 51 c3 48 83 ec 28 89 54 24 1c 48 89 74 24 10 RSP: 002b:00007ffd15e90c88 EFLAGS: 00000246 ORIG_RAX: 000000000000002e RAX: ffffffffffffffda RBX: 0000000000000002 RCX: 00007f855e113367 RDX: 0000000000000000 RSI: 00007ffd15e90cf0 RDI: 0000000000000004 RBP: 00007ffd15e90dbc R08: 0000000000000028 R09: 000000000045d100 R10: 00007f855e011dd8 R11: 0000000000000246 R12: 0000000000000019 R13: 0000000067c6b785 R14: 00000000004a1e80 R15: 0000000000000000 Modules linked in: 8021q garp mrp sch_ingress openvswitch nsh mlx5_ib mlx5_fwctl mlx5_dpll mlx5_core rpcrdma rdma_ucm ib_iser libiscsi scsi_transport_iscsi ib_umad rdma_cm ib_ipoib iw_cm ib_cm ib_uverbs ib_core xt_conntrack xt_MASQUERADE nf_conntrack_netlink nfnetlink xt_addrtype iptable_nat nf_nat br_netfilter rpcsec_gss_krb5 auth_rpcgss oid_registry overlay zram zsmalloc fuse [last unloaded: mlx5_core] ---[ end trace 0000000000000000 ]--- Fixes: 8ff0ac5be144 ("net/mlx5: Add MACsec offload Tx command support") Signed-off-by: Carolina Jubran Reviewed-by: Shahar Shitrit Reviewed-by: Dragos Tatulea Signed-off-by: Tariq Toukan Reviewed-by: Simon Horman Link: https://patch.msgid.link/1746958552-561295-1-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 3506024c2453..9bd166f489e7 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -4349,6 +4349,10 @@ static netdev_features_t mlx5e_fix_uplink_rep_features(struct net_device *netdev if (netdev->features & NETIF_F_HW_VLAN_CTAG_FILTER) netdev_warn(netdev, "Disabling HW_VLAN CTAG FILTERING, not supported in switchdev mode\n"); + features &= ~NETIF_F_HW_MACSEC; + if (netdev->features & NETIF_F_HW_MACSEC) + netdev_warn(netdev, "Disabling HW MACsec offload, not supported in switchdev mode\n"); + return features; } From d5c17e36549cb417f249ecbd20c9ab008aaf45d3 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Mon, 12 May 2025 16:17:51 +0300 Subject: [PATCH 452/555] docs: networking: timestamping: improve stacked PHC sentence The first paragraph makes no grammatical sense. I suppose a portion of the intended sentece is missing: "[The challenge with ] stacked PHCs (...) is that they uncover bugs". Rephrase, and at the same time simplify the structure of the sentence a little bit, it is not easy to follow. Fixes: 94d9f78f4d64 ("docs: networking: timestamping: add section for stacked PHC devices") Signed-off-by: Vladimir Oltean Acked-by: Richard Cochran Link: https://patch.msgid.link/20250512131751.320283-1-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- Documentation/networking/timestamping.rst | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/Documentation/networking/timestamping.rst b/Documentation/networking/timestamping.rst index b8fef8101176..7aabead90648 100644 --- a/Documentation/networking/timestamping.rst +++ b/Documentation/networking/timestamping.rst @@ -811,11 +811,9 @@ Documentation/devicetree/bindings/ptp/timestamper.txt for more details. 3.2.4 Other caveats for MAC drivers ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ -Stacked PHCs, especially DSA (but not only) - since that doesn't require any -modification to MAC drivers, so it is more difficult to ensure correctness of -all possible code paths - is that they uncover bugs which were impossible to -trigger before the existence of stacked PTP clocks. One example has to do with -this line of code, already presented earlier:: +The use of stacked PHCs may uncover MAC driver bugs which were impossible to +trigger without them. One example has to do with this line of code, already +presented earlier:: skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS; From 9d8a99c5a7c7f4f7eca2c168a4ec254409670035 Mon Sep 17 00:00:00 2001 From: Abdun Nihaal Date: Mon, 12 May 2025 10:18:27 +0530 Subject: [PATCH 453/555] qlcnic: fix memory leak in qlcnic_sriov_channel_cfg_cmd() In one of the error paths in qlcnic_sriov_channel_cfg_cmd(), the memory allocated in qlcnic_sriov_alloc_bc_mbx_args() for mailbox arguments is not freed. Fix that by jumping to the error path that frees them, by calling qlcnic_free_mbx_args(). This was found using static analysis. Fixes: f197a7aa6288 ("qlcnic: VF-PF communication channel implementation") Signed-off-by: Abdun Nihaal Reviewed-by: Simon Horman Link: https://patch.msgid.link/20250512044829.36400-1-abdun.nihaal@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov_common.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov_common.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov_common.c index 28d24d59efb8..d57b976b9040 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov_common.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov_common.c @@ -1484,8 +1484,11 @@ static int qlcnic_sriov_channel_cfg_cmd(struct qlcnic_adapter *adapter, u8 cmd_o } cmd_op = (cmd.rsp.arg[0] & 0xff); - if (cmd.rsp.arg[0] >> 25 == 2) - return 2; + if (cmd.rsp.arg[0] >> 25 == 2) { + ret = 2; + goto out; + } + if (cmd_op == QLCNIC_BC_CMD_CHANNEL_INIT) set_bit(QLC_BC_VF_STATE, &vf->state); else From a21675ee3b1ba094e229ae4cd8bddf7d215ab1b9 Mon Sep 17 00:00:00 2001 From: Alan Adamson Date: Thu, 8 May 2025 15:38:00 -0700 Subject: [PATCH 454/555] nvme: multipath: enable BLK_FEAT_ATOMIC_WRITES for multipathing A change to QEMU resulted in all nvme controllers (single and multi-controller subsystems) to have its CMIC.MCTRS bit set which indicates the subsystem supports multiple controllers and it is possible a namespace can be shared between those multiple controllers in a multipath configuration. When a namespace of a CMIC.MCTRS enabled subsystem is allocated, a multipath node is created. The queue limits for this node are inherited from the namespace being allocated. When inheriting queue limits, the features being inherited need to be specified. The atomic write feature (BLK_FEAT_ATOMIC_WRITES) was not specified so the atomic queue limits were not inherited by the multipath disk node which resulted in the sysfs atomic write attributes being zeroed. The fix is to include BLK_FEAT_ATOMIC_WRITES in the list of features to be inherited. Signed-off-by: Alan Adamson Reviewed-by: John Garry Signed-off-by: Christoph Hellwig --- drivers/nvme/host/multipath.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/nvme/host/multipath.c b/drivers/nvme/host/multipath.c index 61b1d267ffda..537a6271a06e 100644 --- a/drivers/nvme/host/multipath.c +++ b/drivers/nvme/host/multipath.c @@ -638,7 +638,8 @@ int nvme_mpath_alloc_disk(struct nvme_ctrl *ctrl, struct nvme_ns_head *head) blk_set_stacking_limits(&lim); lim.dma_alignment = 3; - lim.features |= BLK_FEAT_IO_STAT | BLK_FEAT_NOWAIT | BLK_FEAT_POLL; + lim.features |= BLK_FEAT_IO_STAT | BLK_FEAT_NOWAIT | + BLK_FEAT_POLL | BLK_FEAT_ATOMIC_WRITES; if (head->ids.csi == NVME_CSI_ZNS) lim.features |= BLK_FEAT_ZONED; From 3f097adb9b6c804636bcf8d01e0e7bc037bee0d3 Mon Sep 17 00:00:00 2001 From: Hal Feng Date: Tue, 22 Apr 2025 18:12:44 +0800 Subject: [PATCH 455/555] phy: starfive: jh7110-usb: Fix USB 2.0 host occasional detection failure JH7110 USB 2.0 host fails to detect USB 2.0 devices occasionally. With a long time of debugging and testing, we found that setting Rx clock gating control signal to normal power consumption mode can solve this problem. Signed-off-by: Hal Feng Link: https://lore.kernel.org/r/20250422101244.51686-1-hal.feng@starfivetech.com Signed-off-by: Vinod Koul --- drivers/phy/starfive/phy-jh7110-usb.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/phy/starfive/phy-jh7110-usb.c b/drivers/phy/starfive/phy-jh7110-usb.c index cb5454fbe2c8..b505d89860b4 100644 --- a/drivers/phy/starfive/phy-jh7110-usb.c +++ b/drivers/phy/starfive/phy-jh7110-usb.c @@ -18,6 +18,8 @@ #include #define USB_125M_CLK_RATE 125000000 +#define USB_CLK_MODE_OFF 0x0 +#define USB_CLK_MODE_RX_NORMAL_PWR BIT(1) #define USB_LS_KEEPALIVE_OFF 0x4 #define USB_LS_KEEPALIVE_ENABLE BIT(4) @@ -78,6 +80,7 @@ static int jh7110_usb2_phy_init(struct phy *_phy) { struct jh7110_usb2_phy *phy = phy_get_drvdata(_phy); int ret; + unsigned int val; ret = clk_set_rate(phy->usb_125m_clk, USB_125M_CLK_RATE); if (ret) @@ -87,6 +90,10 @@ static int jh7110_usb2_phy_init(struct phy *_phy) if (ret) return ret; + val = readl(phy->regs + USB_CLK_MODE_OFF); + val |= USB_CLK_MODE_RX_NORMAL_PWR; + writel(val, phy->regs + USB_CLK_MODE_OFF); + return 0; } From f9475055b11c0c70979bd1667a76b2ebae638eb7 Mon Sep 17 00:00:00 2001 From: Algea Cao Date: Sun, 27 Apr 2025 17:51:24 +0800 Subject: [PATCH 456/555] phy: phy-rockchip-samsung-hdptx: Fix PHY PLL output 50.25MHz error When using HDMI PLL frequency division coefficient at 50.25MHz that is calculated by rk_hdptx_phy_clk_pll_calc(), it fails to get PHY LANE lock. Although the calculated values are within the allowable range of PHY PLL configuration. In order to fix the PHY LANE lock error and provide the expected 50.25MHz output, manually compute the required PHY PLL frequency division coefficient and add it to ropll_tmds_cfg configuration table. Signed-off-by: Algea Cao Reviewed-by: Cristian Ciocaltea Acked-by: Heiko Stuebner Link: https://lore.kernel.org/r/20250427095124.3354439-1-algea.cao@rock-chips.com Signed-off-by: Vinod Koul --- drivers/phy/rockchip/phy-rockchip-samsung-hdptx.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/phy/rockchip/phy-rockchip-samsung-hdptx.c b/drivers/phy/rockchip/phy-rockchip-samsung-hdptx.c index fe7c05748356..77236f012a1f 100644 --- a/drivers/phy/rockchip/phy-rockchip-samsung-hdptx.c +++ b/drivers/phy/rockchip/phy-rockchip-samsung-hdptx.c @@ -476,6 +476,8 @@ static const struct ropll_config ropll_tmds_cfg[] = { 1, 1, 0, 0x20, 0x0c, 1, 0x0e, 0, 0, }, { 650000, 162, 162, 1, 1, 11, 1, 1, 1, 1, 1, 1, 1, 54, 0, 16, 4, 1, 1, 1, 0, 0x20, 0x0c, 1, 0x0e, 0, 0, }, + { 502500, 84, 84, 1, 1, 7, 1, 1, 1, 1, 1, 1, 1, 11, 1, 4, 5, + 4, 11, 1, 0, 0x20, 0x0c, 1, 0x0e, 0, 0, }, { 337500, 0x70, 0x70, 1, 1, 0xf, 1, 1, 1, 1, 1, 1, 1, 0x2, 0, 0x01, 5, 1, 1, 1, 0, 0x20, 0x0c, 1, 0x0e, 0, 0, }, { 400000, 100, 100, 1, 1, 11, 1, 1, 0, 1, 0, 1, 1, 0x9, 0, 0x05, 0, From fb98bd0a13de2c9d96cb5c00c81b5ca118ac9d71 Mon Sep 17 00:00:00 2001 From: Alessandro Grassi Date: Fri, 2 May 2025 11:55:20 +0200 Subject: [PATCH 457/555] spi: spi-sun4i: fix early activation The SPI interface is activated before the CPOL setting is applied. In that moment, the clock idles high and CS goes low. After a short delay, CPOL and other settings are applied, which may cause the clock to change state and idle low. This transition is not part of a clock cycle, and it can confuse the receiving device. To prevent this unexpected transition, activate the interface while CPOL and the other settings are being applied. Signed-off-by: Alessandro Grassi Link: https://patch.msgid.link/20250502095520.13825-1-alessandro.grassi@mailbox.org Signed-off-by: Mark Brown --- drivers/spi/spi-sun4i.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/spi/spi-sun4i.c b/drivers/spi/spi-sun4i.c index f89826d7dc49..aa92fd5a35a9 100644 --- a/drivers/spi/spi-sun4i.c +++ b/drivers/spi/spi-sun4i.c @@ -264,6 +264,9 @@ static int sun4i_spi_transfer_one(struct spi_controller *host, else reg |= SUN4I_CTL_DHB; + /* Now that the settings are correct, enable the interface */ + reg |= SUN4I_CTL_ENABLE; + sun4i_spi_write(sspi, SUN4I_CTL_REG, reg); /* Ensure that we have a parent clock fast enough */ @@ -404,7 +407,7 @@ static int sun4i_spi_runtime_resume(struct device *dev) } sun4i_spi_write(sspi, SUN4I_CTL_REG, - SUN4I_CTL_ENABLE | SUN4I_CTL_MASTER | SUN4I_CTL_TP); + SUN4I_CTL_MASTER | SUN4I_CTL_TP); return 0; From 6b0cd72757c69bc2d45da42b41023e288d02e772 Mon Sep 17 00:00:00 2001 From: Cosmin Tanislav Date: Thu, 8 May 2025 09:49:43 +0300 Subject: [PATCH 458/555] regulator: max20086: fix invalid memory access max20086_parse_regulators_dt() calls of_regulator_match() using an array of struct of_regulator_match allocated on the stack for the matches argument. of_regulator_match() calls devm_of_regulator_put_matches(), which calls devres_alloc() to allocate a struct devm_of_regulator_matches which will be de-allocated using devm_of_regulator_put_matches(). struct devm_of_regulator_matches is populated with the stack allocated matches array. If the device fails to probe, devm_of_regulator_put_matches() will be called and will try to call of_node_put() on that stack pointer, generating the following dmesg entries: max20086 6-0028: Failed to read DEVICE_ID reg: -121 kobject: '\xc0$\xa5\x03' (000000002cebcb7a): is not initialized, yet kobject_put() is being called. Followed by a stack trace matching the call flow described above. Switch to allocating the matches array using devm_kcalloc() to avoid accessing the stack pointer long after it's out of scope. This also has the advantage of allowing multiple max20086 to probe without overriding the data stored inside the global of_regulator_match. Fixes: bfff546aae50 ("regulator: Add MAX20086-MAX20089 driver") Signed-off-by: Cosmin Tanislav Link: https://patch.msgid.link/20250508064947.2567255-1-demonsingur@gmail.com Signed-off-by: Mark Brown --- drivers/regulator/max20086-regulator.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/regulator/max20086-regulator.c b/drivers/regulator/max20086-regulator.c index 59eb23d467ec..198d45f8e884 100644 --- a/drivers/regulator/max20086-regulator.c +++ b/drivers/regulator/max20086-regulator.c @@ -132,7 +132,7 @@ static int max20086_regulators_register(struct max20086 *chip) static int max20086_parse_regulators_dt(struct max20086 *chip, bool *boot_on) { - struct of_regulator_match matches[MAX20086_MAX_REGULATORS] = { }; + struct of_regulator_match *matches; struct device_node *node; unsigned int i; int ret; @@ -143,6 +143,11 @@ static int max20086_parse_regulators_dt(struct max20086 *chip, bool *boot_on) return -ENODEV; } + matches = devm_kcalloc(chip->dev, chip->info->num_outputs, + sizeof(*matches), GFP_KERNEL); + if (!matches) + return -ENOMEM; + for (i = 0; i < chip->info->num_outputs; ++i) matches[i].name = max20086_output_names[i]; From 314007549d89adebdd1e214a743d7e26edbd075e Mon Sep 17 00:00:00 2001 From: Hariprasad Kelam Date: Mon, 12 May 2025 11:59:01 +0530 Subject: [PATCH 459/555] octeontx2-pf: Fix ethtool support for SDP representors The hardware supports multiple MAC types, including RPM, SDP, and LBK. However, features such as link settings and pause frames are only available on RPM MAC, and not supported on SDP or LBK. This patch updates the ethtool operations logic accordingly to reflect this behavior. Fixes: 2f7f33a09516 ("octeontx2-pf: Add representors for sdp MAC") Signed-off-by: Hariprasad Kelam Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- .../net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c index 010385b29988..45b8c9230184 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c @@ -315,7 +315,7 @@ static void otx2_get_pauseparam(struct net_device *netdev, struct otx2_nic *pfvf = netdev_priv(netdev); struct cgx_pause_frm_cfg *req, *rsp; - if (is_otx2_lbkvf(pfvf->pdev)) + if (is_otx2_lbkvf(pfvf->pdev) || is_otx2_sdp_rep(pfvf->pdev)) return; mutex_lock(&pfvf->mbox.lock); @@ -347,7 +347,7 @@ static int otx2_set_pauseparam(struct net_device *netdev, if (pause->autoneg) return -EOPNOTSUPP; - if (is_otx2_lbkvf(pfvf->pdev)) + if (is_otx2_lbkvf(pfvf->pdev) || is_otx2_sdp_rep(pfvf->pdev)) return -EOPNOTSUPP; if (pause->rx_pause) @@ -941,8 +941,8 @@ static u32 otx2_get_link(struct net_device *netdev) { struct otx2_nic *pfvf = netdev_priv(netdev); - /* LBK link is internal and always UP */ - if (is_otx2_lbkvf(pfvf->pdev)) + /* LBK and SDP links are internal and always UP */ + if (is_otx2_lbkvf(pfvf->pdev) || is_otx2_sdp_rep(pfvf->pdev)) return 1; return pfvf->linfo.link_up; } @@ -1413,7 +1413,7 @@ static int otx2vf_get_link_ksettings(struct net_device *netdev, { struct otx2_nic *pfvf = netdev_priv(netdev); - if (is_otx2_lbkvf(pfvf->pdev)) { + if (is_otx2_lbkvf(pfvf->pdev) || is_otx2_sdp_rep(pfvf->pdev)) { cmd->base.duplex = DUPLEX_FULL; cmd->base.speed = SPEED_100000; } else { From 9e000f1b7f31684cc5927e034360b87ac7919593 Mon Sep 17 00:00:00 2001 From: Wentao Liang Date: Wed, 14 May 2025 17:24:44 +0800 Subject: [PATCH 460/555] ALSA: es1968: Add error handling for snd_pcm_hw_constraint_pow2() The function snd_es1968_capture_open() calls the function snd_pcm_hw_constraint_pow2(), but does not check its return value. A proper implementation can be found in snd_cx25821_pcm_open(). Add error handling for snd_pcm_hw_constraint_pow2() and propagate its error code. Fixes: b942cf815b57 ("[ALSA] es1968 - Fix stuttering capture") Cc: stable@vger.kernel.org # v2.6.22 Signed-off-by: Wentao Liang Link: https://patch.msgid.link/20250514092444.331-1-vulab@iscas.ac.cn Signed-off-by: Takashi Iwai --- sound/pci/es1968.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/sound/pci/es1968.c b/sound/pci/es1968.c index c6c018b40c69..4e0693f0ab0f 100644 --- a/sound/pci/es1968.c +++ b/sound/pci/es1968.c @@ -1561,7 +1561,7 @@ static int snd_es1968_capture_open(struct snd_pcm_substream *substream) struct snd_pcm_runtime *runtime = substream->runtime; struct es1968 *chip = snd_pcm_substream_chip(substream); struct esschan *es; - int apu1, apu2; + int err, apu1, apu2; apu1 = snd_es1968_alloc_apu_pair(chip, ESM_APU_PCM_CAPTURE); if (apu1 < 0) @@ -1605,7 +1605,9 @@ static int snd_es1968_capture_open(struct snd_pcm_substream *substream) runtime->hw = snd_es1968_capture; runtime->hw.buffer_bytes_max = runtime->hw.period_bytes_max = calc_available_memory_size(chip) - 1024; /* keep MIXBUF size */ - snd_pcm_hw_constraint_pow2(runtime, 0, SNDRV_PCM_HW_PARAM_BUFFER_BYTES); + err = snd_pcm_hw_constraint_pow2(runtime, 0, SNDRV_PCM_HW_PARAM_BUFFER_BYTES); + if (err < 0) + return err; spin_lock_irq(&chip->substream_lock); list_add(&es->list, &chip->substream_list); From 83c178470e0bf690d34c8c08440f2421b82e881c Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 23 Apr 2025 16:08:23 +0300 Subject: [PATCH 461/555] phy: tegra: xusb: remove a stray unlock We used to take a lock in tegra186_utmi_bias_pad_power_on() but now we have moved the lock into the caller. Unfortunately, when we moved the lock this unlock was left behind and it results in a double unlock. Delete it now. Fixes: b47158fb4295 ("phy: tegra: xusb: Use a bitmask for UTMI pad power state tracking") Signed-off-by: Dan Carpenter Reviewed-by: Jon Hunter Link: https://lore.kernel.org/r/aAjmR6To4EnvRl4G@stanley.mountain Signed-off-by: Vinod Koul --- drivers/phy/tegra/xusb-tegra186.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/phy/tegra/xusb-tegra186.c b/drivers/phy/tegra/xusb-tegra186.c index cc7b8a6a999f..23a23f2d64e5 100644 --- a/drivers/phy/tegra/xusb-tegra186.c +++ b/drivers/phy/tegra/xusb-tegra186.c @@ -656,8 +656,6 @@ static void tegra186_utmi_bias_pad_power_on(struct tegra_xusb_padctl *padctl) } else { clk_disable_unprepare(priv->usb2_trk_clk); } - - mutex_unlock(&padctl->lock); } static void tegra186_utmi_bias_pad_power_off(struct tegra_xusb_padctl *padctl) From 54c4c58713aaff76c2422ff5750e557ab3b100d7 Mon Sep 17 00:00:00 2001 From: Claudiu Beznea Date: Wed, 7 May 2025 15:50:28 +0300 Subject: [PATCH 462/555] phy: renesas: rcar-gen3-usb2: Fix role detection on unbind/bind It has been observed on the Renesas RZ/G3S SoC that unbinding and binding the PHY driver leads to role autodetection failures. This issue occurs when PHY 3 is the first initialized PHY. PHY 3 does not have an interrupt associated with the USB2_INT_ENABLE register (as rcar_gen3_int_enable[3] = 0). As a result, rcar_gen3_init_otg() is called to initialize OTG without enabling PHY interrupts. To resolve this, add rcar_gen3_is_any_otg_rphy_initialized() and call it in role_store(), role_show(), and rcar_gen3_init_otg(). At the same time, rcar_gen3_init_otg() is only called when initialization for a PHY with interrupt bits is in progress. As a result, the struct rcar_gen3_phy::otg_initialized is no longer needed. Fixes: 549b6b55b005 ("phy: renesas: rcar-gen3-usb2: enable/disable independent irqs") Cc: stable@vger.kernel.org Reviewed-by: Yoshihiro Shimoda Tested-by: Yoshihiro Shimoda Reviewed-by: Lad Prabhakar Signed-off-by: Claudiu Beznea Link: https://lore.kernel.org/r/20250507125032.565017-2-claudiu.beznea.uj@bp.renesas.com Signed-off-by: Vinod Koul --- drivers/phy/renesas/phy-rcar-gen3-usb2.c | 33 ++++++++++-------------- 1 file changed, 14 insertions(+), 19 deletions(-) diff --git a/drivers/phy/renesas/phy-rcar-gen3-usb2.c b/drivers/phy/renesas/phy-rcar-gen3-usb2.c index 775f4f973a6c..46afba2fe0dc 100644 --- a/drivers/phy/renesas/phy-rcar-gen3-usb2.c +++ b/drivers/phy/renesas/phy-rcar-gen3-usb2.c @@ -107,7 +107,6 @@ struct rcar_gen3_phy { struct rcar_gen3_chan *ch; u32 int_enable_bits; bool initialized; - bool otg_initialized; bool powered; }; @@ -320,16 +319,15 @@ static bool rcar_gen3_is_any_rphy_initialized(struct rcar_gen3_chan *ch) return false; } -static bool rcar_gen3_needs_init_otg(struct rcar_gen3_chan *ch) +static bool rcar_gen3_is_any_otg_rphy_initialized(struct rcar_gen3_chan *ch) { - int i; - - for (i = 0; i < NUM_OF_PHYS; i++) { - if (ch->rphys[i].otg_initialized) - return false; + for (enum rcar_gen3_phy_index i = PHY_INDEX_BOTH_HC; i <= PHY_INDEX_EHCI; + i++) { + if (ch->rphys[i].initialized) + return true; } - return true; + return false; } static bool rcar_gen3_are_all_rphys_power_off(struct rcar_gen3_chan *ch) @@ -351,7 +349,7 @@ static ssize_t role_store(struct device *dev, struct device_attribute *attr, bool is_b_device; enum phy_mode cur_mode, new_mode; - if (!ch->is_otg_channel || !rcar_gen3_is_any_rphy_initialized(ch)) + if (!ch->is_otg_channel || !rcar_gen3_is_any_otg_rphy_initialized(ch)) return -EIO; if (sysfs_streq(buf, "host")) @@ -389,7 +387,7 @@ static ssize_t role_show(struct device *dev, struct device_attribute *attr, { struct rcar_gen3_chan *ch = dev_get_drvdata(dev); - if (!ch->is_otg_channel || !rcar_gen3_is_any_rphy_initialized(ch)) + if (!ch->is_otg_channel || !rcar_gen3_is_any_otg_rphy_initialized(ch)) return -EIO; return sprintf(buf, "%s\n", rcar_gen3_is_host(ch) ? "host" : @@ -402,6 +400,9 @@ static void rcar_gen3_init_otg(struct rcar_gen3_chan *ch) void __iomem *usb2_base = ch->base; u32 val; + if (!ch->is_otg_channel || rcar_gen3_is_any_otg_rphy_initialized(ch)) + return; + /* Should not use functions of read-modify-write a register */ val = readl(usb2_base + USB2_LINECTRL1); val = (val & ~USB2_LINECTRL1_DP_RPD) | USB2_LINECTRL1_DPRPD_EN | @@ -465,12 +466,9 @@ static int rcar_gen3_phy_usb2_init(struct phy *p) writel(USB2_SPD_RSM_TIMSET_INIT, usb2_base + USB2_SPD_RSM_TIMSET); writel(USB2_OC_TIMSET_INIT, usb2_base + USB2_OC_TIMSET); - /* Initialize otg part */ - if (channel->is_otg_channel) { - if (rcar_gen3_needs_init_otg(channel)) - rcar_gen3_init_otg(channel); - rphy->otg_initialized = true; - } + /* Initialize otg part (only if we initialize a PHY with IRQs). */ + if (rphy->int_enable_bits) + rcar_gen3_init_otg(channel); rphy->initialized = true; @@ -486,9 +484,6 @@ static int rcar_gen3_phy_usb2_exit(struct phy *p) rphy->initialized = false; - if (channel->is_otg_channel) - rphy->otg_initialized = false; - val = readl(usb2_base + USB2_INT_ENABLE); val &= ~rphy->int_enable_bits; if (!rcar_gen3_is_any_rphy_initialized(channel)) From de76809f60cc938d3580bbbd5b04b7d12af6ce3a Mon Sep 17 00:00:00 2001 From: Claudiu Beznea Date: Wed, 7 May 2025 15:50:29 +0300 Subject: [PATCH 463/555] phy: renesas: rcar-gen3-usb2: Move IRQ request in probe Commit 08b0ad375ca6 ("phy: renesas: rcar-gen3-usb2: move IRQ registration to init") moved the IRQ request operation from probe to struct phy_ops::phy_init API to avoid triggering interrupts (which lead to register accesses) while the PHY clocks (enabled through runtime PM APIs) are not active. If this happens, it results in a synchronous abort. One way to reproduce this issue is by enabling CONFIG_DEBUG_SHIRQ, which calls free_irq() on driver removal. Move the IRQ request and free operations back to probe, and take the runtime PM state into account in IRQ handler. This commit is preparatory for the subsequent fixes in this series. Reviewed-by: Yoshihiro Shimoda Tested-by: Yoshihiro Shimoda Reviewed-by: Lad Prabhakar Signed-off-by: Claudiu Beznea Link: https://lore.kernel.org/r/20250507125032.565017-3-claudiu.beznea.uj@bp.renesas.com Signed-off-by: Vinod Koul --- drivers/phy/renesas/phy-rcar-gen3-usb2.c | 46 +++++++++++++----------- 1 file changed, 26 insertions(+), 20 deletions(-) diff --git a/drivers/phy/renesas/phy-rcar-gen3-usb2.c b/drivers/phy/renesas/phy-rcar-gen3-usb2.c index 46afba2fe0dc..bb05fd26eb7f 100644 --- a/drivers/phy/renesas/phy-rcar-gen3-usb2.c +++ b/drivers/phy/renesas/phy-rcar-gen3-usb2.c @@ -120,7 +120,6 @@ struct rcar_gen3_chan { struct work_struct work; struct mutex lock; /* protects rphys[...].powered */ enum usb_dr_mode dr_mode; - int irq; u32 obint_enable_bits; bool extcon_host; bool is_otg_channel; @@ -428,16 +427,25 @@ static irqreturn_t rcar_gen3_phy_usb2_irq(int irq, void *_ch) { struct rcar_gen3_chan *ch = _ch; void __iomem *usb2_base = ch->base; - u32 status = readl(usb2_base + USB2_OBINTSTA); + struct device *dev = ch->dev; irqreturn_t ret = IRQ_NONE; + u32 status; + pm_runtime_get_noresume(dev); + + if (pm_runtime_suspended(dev)) + goto rpm_put; + + status = readl(usb2_base + USB2_OBINTSTA); if (status & ch->obint_enable_bits) { - dev_vdbg(ch->dev, "%s: %08x\n", __func__, status); + dev_vdbg(dev, "%s: %08x\n", __func__, status); writel(ch->obint_enable_bits, usb2_base + USB2_OBINTSTA); rcar_gen3_device_recognition(ch); ret = IRQ_HANDLED; } +rpm_put: + pm_runtime_put_noidle(dev); return ret; } @@ -447,17 +455,6 @@ static int rcar_gen3_phy_usb2_init(struct phy *p) struct rcar_gen3_chan *channel = rphy->ch; void __iomem *usb2_base = channel->base; u32 val; - int ret; - - if (!rcar_gen3_is_any_rphy_initialized(channel) && channel->irq >= 0) { - INIT_WORK(&channel->work, rcar_gen3_phy_usb2_work); - ret = request_irq(channel->irq, rcar_gen3_phy_usb2_irq, - IRQF_SHARED, dev_name(channel->dev), channel); - if (ret < 0) { - dev_err(channel->dev, "No irq handler (%d)\n", channel->irq); - return ret; - } - } /* Initialize USB2 part */ val = readl(usb2_base + USB2_INT_ENABLE); @@ -490,9 +487,6 @@ static int rcar_gen3_phy_usb2_exit(struct phy *p) val &= ~USB2_INT_ENABLE_UCOM_INTEN; writel(val, usb2_base + USB2_INT_ENABLE); - if (channel->irq >= 0 && !rcar_gen3_is_any_rphy_initialized(channel)) - free_irq(channel->irq, channel); - return 0; } @@ -698,7 +692,7 @@ static int rcar_gen3_phy_usb2_probe(struct platform_device *pdev) struct device *dev = &pdev->dev; struct rcar_gen3_chan *channel; struct phy_provider *provider; - int ret = 0, i; + int ret = 0, i, irq; if (!dev->of_node) { dev_err(dev, "This driver needs device tree\n"); @@ -714,8 +708,6 @@ static int rcar_gen3_phy_usb2_probe(struct platform_device *pdev) return PTR_ERR(channel->base); channel->obint_enable_bits = USB2_OBINT_BITS; - /* get irq number here and request_irq for OTG in phy_init */ - channel->irq = platform_get_irq_optional(pdev, 0); channel->dr_mode = rcar_gen3_get_dr_mode(dev->of_node); if (channel->dr_mode != USB_DR_MODE_UNKNOWN) { channel->is_otg_channel = true; @@ -784,6 +776,20 @@ static int rcar_gen3_phy_usb2_probe(struct platform_device *pdev) channel->vbus = NULL; } + irq = platform_get_irq_optional(pdev, 0); + if (irq < 0 && irq != -ENXIO) { + ret = irq; + goto error; + } else if (irq > 0) { + INIT_WORK(&channel->work, rcar_gen3_phy_usb2_work); + ret = devm_request_irq(dev, irq, rcar_gen3_phy_usb2_irq, + IRQF_SHARED, dev_name(dev), channel); + if (ret < 0) { + dev_err(dev, "Failed to request irq (%d)\n", irq); + goto error; + } + } + provider = devm_of_phy_provider_register(dev, rcar_gen3_phy_usb2_xlate); if (IS_ERR(provider)) { dev_err(dev, "Failed to register PHY provider\n"); From 55a387ebb9219cbe4edfa8ba9996ccb0e7ad4932 Mon Sep 17 00:00:00 2001 From: Claudiu Beznea Date: Wed, 7 May 2025 15:50:30 +0300 Subject: [PATCH 464/555] phy: renesas: rcar-gen3-usb2: Lock around hardware registers and driver data The phy-rcar-gen3-usb2 driver exposes four individual PHYs that are requested and configured by PHY users. The struct phy_ops APIs access the same set of registers to configure all PHYs. Additionally, PHY settings can be modified through sysfs or an IRQ handler. While some struct phy_ops APIs are protected by a driver-wide mutex, others rely on individual PHY-specific mutexes. This approach can lead to various issues, including: 1/ the IRQ handler may interrupt PHY settings in progress, racing with hardware configuration protected by a mutex lock 2/ due to msleep(20) in rcar_gen3_init_otg(), while a configuration thread suspends to wait for the delay, another thread may try to configure another PHY (with phy_init() + phy_power_on()); re-running the phy_init() goes to the exact same configuration code, re-running the same hardware configuration on the same set of registers (and bits) which might impact the result of the msleep for the 1st configuring thread 3/ sysfs can configure the hardware (though role_store()) and it can still race with the phy_init()/phy_power_on() APIs calling into the drivers struct phy_ops To address these issues, add a spinlock to protect hardware register access and driver private data structures (e.g., calls to rcar_gen3_is_any_rphy_initialized()). Checking driver-specific data remains necessary as all PHY instances share common settings. With this change, the existing mutex protection is removed and the cleanup.h helpers are used. While at it, to keep the code simpler, do not skip regulator_enable()/regulator_disable() APIs in rcar_gen3_phy_usb2_power_on()/rcar_gen3_phy_usb2_power_off() as the regulators enable/disable operations are reference counted anyway. Fixes: f3b5a8d9b50d ("phy: rcar-gen3-usb2: Add R-Car Gen3 USB2 PHY driver") Cc: stable@vger.kernel.org Reviewed-by: Yoshihiro Shimoda Tested-by: Yoshihiro Shimoda Reviewed-by: Lad Prabhakar Signed-off-by: Claudiu Beznea Link: https://lore.kernel.org/r/20250507125032.565017-4-claudiu.beznea.uj@bp.renesas.com Signed-off-by: Vinod Koul --- drivers/phy/renesas/phy-rcar-gen3-usb2.c | 49 +++++++++++++----------- 1 file changed, 26 insertions(+), 23 deletions(-) diff --git a/drivers/phy/renesas/phy-rcar-gen3-usb2.c b/drivers/phy/renesas/phy-rcar-gen3-usb2.c index bb05fd26eb7f..00ce564463de 100644 --- a/drivers/phy/renesas/phy-rcar-gen3-usb2.c +++ b/drivers/phy/renesas/phy-rcar-gen3-usb2.c @@ -9,6 +9,7 @@ * Copyright (C) 2014 Cogent Embedded, Inc. */ +#include #include #include #include @@ -118,7 +119,7 @@ struct rcar_gen3_chan { struct regulator *vbus; struct reset_control *rstc; struct work_struct work; - struct mutex lock; /* protects rphys[...].powered */ + spinlock_t lock; /* protects access to hardware and driver data structure. */ enum usb_dr_mode dr_mode; u32 obint_enable_bits; bool extcon_host; @@ -348,6 +349,8 @@ static ssize_t role_store(struct device *dev, struct device_attribute *attr, bool is_b_device; enum phy_mode cur_mode, new_mode; + guard(spinlock_irqsave)(&ch->lock); + if (!ch->is_otg_channel || !rcar_gen3_is_any_otg_rphy_initialized(ch)) return -EIO; @@ -415,7 +418,7 @@ static void rcar_gen3_init_otg(struct rcar_gen3_chan *ch) val = readl(usb2_base + USB2_ADPCTRL); writel(val | USB2_ADPCTRL_IDPULLUP, usb2_base + USB2_ADPCTRL); } - msleep(20); + mdelay(20); writel(0xffffffff, usb2_base + USB2_OBINTSTA); writel(ch->obint_enable_bits, usb2_base + USB2_OBINTEN); @@ -436,12 +439,14 @@ static irqreturn_t rcar_gen3_phy_usb2_irq(int irq, void *_ch) if (pm_runtime_suspended(dev)) goto rpm_put; - status = readl(usb2_base + USB2_OBINTSTA); - if (status & ch->obint_enable_bits) { - dev_vdbg(dev, "%s: %08x\n", __func__, status); - writel(ch->obint_enable_bits, usb2_base + USB2_OBINTSTA); - rcar_gen3_device_recognition(ch); - ret = IRQ_HANDLED; + scoped_guard(spinlock, &ch->lock) { + status = readl(usb2_base + USB2_OBINTSTA); + if (status & ch->obint_enable_bits) { + dev_vdbg(dev, "%s: %08x\n", __func__, status); + writel(ch->obint_enable_bits, usb2_base + USB2_OBINTSTA); + rcar_gen3_device_recognition(ch); + ret = IRQ_HANDLED; + } } rpm_put: @@ -456,6 +461,8 @@ static int rcar_gen3_phy_usb2_init(struct phy *p) void __iomem *usb2_base = channel->base; u32 val; + guard(spinlock_irqsave)(&channel->lock); + /* Initialize USB2 part */ val = readl(usb2_base + USB2_INT_ENABLE); val |= USB2_INT_ENABLE_UCOM_INTEN | rphy->int_enable_bits; @@ -479,6 +486,8 @@ static int rcar_gen3_phy_usb2_exit(struct phy *p) void __iomem *usb2_base = channel->base; u32 val; + guard(spinlock_irqsave)(&channel->lock); + rphy->initialized = false; val = readl(usb2_base + USB2_INT_ENABLE); @@ -498,16 +507,17 @@ static int rcar_gen3_phy_usb2_power_on(struct phy *p) u32 val; int ret = 0; - mutex_lock(&channel->lock); - if (!rcar_gen3_are_all_rphys_power_off(channel)) - goto out; - if (channel->vbus) { ret = regulator_enable(channel->vbus); if (ret) - goto out; + return ret; } + guard(spinlock_irqsave)(&channel->lock); + + if (!rcar_gen3_are_all_rphys_power_off(channel)) + goto out; + val = readl(usb2_base + USB2_USBCTR); val |= USB2_USBCTR_PLL_RST; writel(val, usb2_base + USB2_USBCTR); @@ -517,7 +527,6 @@ static int rcar_gen3_phy_usb2_power_on(struct phy *p) out: /* The powered flag should be set for any other phys anyway */ rphy->powered = true; - mutex_unlock(&channel->lock); return 0; } @@ -528,18 +537,12 @@ static int rcar_gen3_phy_usb2_power_off(struct phy *p) struct rcar_gen3_chan *channel = rphy->ch; int ret = 0; - mutex_lock(&channel->lock); - rphy->powered = false; - - if (!rcar_gen3_are_all_rphys_power_off(channel)) - goto out; + scoped_guard(spinlock_irqsave, &channel->lock) + rphy->powered = false; if (channel->vbus) ret = regulator_disable(channel->vbus); -out: - mutex_unlock(&channel->lock); - return ret; } @@ -750,7 +753,7 @@ static int rcar_gen3_phy_usb2_probe(struct platform_device *pdev) if (phy_data->no_adp_ctrl) channel->obint_enable_bits = USB2_OBINT_IDCHG_EN; - mutex_init(&channel->lock); + spin_lock_init(&channel->lock); for (i = 0; i < NUM_OF_PHYS; i++) { channel->rphys[i].phy = devm_phy_create(dev, NULL, phy_data->phy_usb2_ops); From 9ce71e85b29eb63e48e294479742e670513f03a0 Mon Sep 17 00:00:00 2001 From: Claudiu Beznea Date: Wed, 7 May 2025 15:50:31 +0300 Subject: [PATCH 465/555] phy: renesas: rcar-gen3-usb2: Assert PLL reset on PHY power off Assert PLL reset on PHY power off. This saves power. Fixes: f3b5a8d9b50d ("phy: rcar-gen3-usb2: Add R-Car Gen3 USB2 PHY driver") Cc: stable@vger.kernel.org Reviewed-by: Yoshihiro Shimoda Tested-by: Yoshihiro Shimoda Reviewed-by: Lad Prabhakar Signed-off-by: Claudiu Beznea Link: https://lore.kernel.org/r/20250507125032.565017-5-claudiu.beznea.uj@bp.renesas.com Signed-off-by: Vinod Koul --- drivers/phy/renesas/phy-rcar-gen3-usb2.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/drivers/phy/renesas/phy-rcar-gen3-usb2.c b/drivers/phy/renesas/phy-rcar-gen3-usb2.c index 00ce564463de..118899efda70 100644 --- a/drivers/phy/renesas/phy-rcar-gen3-usb2.c +++ b/drivers/phy/renesas/phy-rcar-gen3-usb2.c @@ -537,9 +537,17 @@ static int rcar_gen3_phy_usb2_power_off(struct phy *p) struct rcar_gen3_chan *channel = rphy->ch; int ret = 0; - scoped_guard(spinlock_irqsave, &channel->lock) + scoped_guard(spinlock_irqsave, &channel->lock) { rphy->powered = false; + if (rcar_gen3_are_all_rphys_power_off(channel)) { + u32 val = readl(channel->base + USB2_USBCTR); + + val |= USB2_USBCTR_PLL_RST; + writel(val, channel->base + USB2_USBCTR); + } + } + if (channel->vbus) ret = regulator_disable(channel->vbus); From 86e70849f4b2b4597ac9f7c7931f2a363774be25 Mon Sep 17 00:00:00 2001 From: Claudiu Beznea Date: Wed, 7 May 2025 15:50:32 +0300 Subject: [PATCH 466/555] phy: renesas: rcar-gen3-usb2: Set timing registers only once phy-rcar-gen3-usb2 driver exports 4 PHYs. The timing registers are common to all PHYs. There is no need to set them every time a PHY is initialized. Set timing register only when the 1st PHY is initialized. Fixes: f3b5a8d9b50d ("phy: rcar-gen3-usb2: Add R-Car Gen3 USB2 PHY driver") Cc: stable@vger.kernel.org Reviewed-by: Yoshihiro Shimoda Tested-by: Yoshihiro Shimoda Reviewed-by: Lad Prabhakar Signed-off-by: Claudiu Beznea Link: https://lore.kernel.org/r/20250507125032.565017-6-claudiu.beznea.uj@bp.renesas.com Signed-off-by: Vinod Koul --- drivers/phy/renesas/phy-rcar-gen3-usb2.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/phy/renesas/phy-rcar-gen3-usb2.c b/drivers/phy/renesas/phy-rcar-gen3-usb2.c index 118899efda70..9fdf17e0848a 100644 --- a/drivers/phy/renesas/phy-rcar-gen3-usb2.c +++ b/drivers/phy/renesas/phy-rcar-gen3-usb2.c @@ -467,8 +467,11 @@ static int rcar_gen3_phy_usb2_init(struct phy *p) val = readl(usb2_base + USB2_INT_ENABLE); val |= USB2_INT_ENABLE_UCOM_INTEN | rphy->int_enable_bits; writel(val, usb2_base + USB2_INT_ENABLE); - writel(USB2_SPD_RSM_TIMSET_INIT, usb2_base + USB2_SPD_RSM_TIMSET); - writel(USB2_OC_TIMSET_INIT, usb2_base + USB2_OC_TIMSET); + + if (!rcar_gen3_is_any_rphy_initialized(channel)) { + writel(USB2_SPD_RSM_TIMSET_INIT, usb2_base + USB2_SPD_RSM_TIMSET); + writel(USB2_OC_TIMSET_INIT, usb2_base + USB2_OC_TIMSET); + } /* Initialize otg part (only if we initialize a PHY with IRQs). */ if (rphy->int_enable_bits) From b2ea5f49580c0762d17d80d8083cb89bc3acf74f Mon Sep 17 00:00:00 2001 From: Ma Ke Date: Mon, 3 Mar 2025 15:27:39 +0800 Subject: [PATCH 467/555] phy: Fix error handling in tegra_xusb_port_init If device_add() fails, do not use device_unregister() for error handling. device_unregister() consists two functions: device_del() and put_device(). device_unregister() should only be called after device_add() succeeded because device_del() undoes what device_add() does if successful. Change device_unregister() to put_device() call before returning from the function. As comment of device_add() says, 'if device_add() succeeds, you should call device_del() when you want to get rid of it. If device_add() has not succeeded, use only put_device() to drop the reference count'. Found by code review. Cc: stable@vger.kernel.org Fixes: 53d2a715c240 ("phy: Add Tegra XUSB pad controller support") Signed-off-by: Ma Ke Acked-by: Thierry Reding Link: https://lore.kernel.org/r/20250303072739.3874987-1-make24@iscas.ac.cn Signed-off-by: Vinod Koul --- drivers/phy/tegra/xusb.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/phy/tegra/xusb.c b/drivers/phy/tegra/xusb.c index 79d4814d758d..c89df95aa6ca 100644 --- a/drivers/phy/tegra/xusb.c +++ b/drivers/phy/tegra/xusb.c @@ -548,16 +548,16 @@ static int tegra_xusb_port_init(struct tegra_xusb_port *port, err = dev_set_name(&port->dev, "%s-%u", name, index); if (err < 0) - goto unregister; + goto put_device; err = device_add(&port->dev); if (err < 0) - goto unregister; + goto put_device; return 0; -unregister: - device_unregister(&port->dev); +put_device: + put_device(&port->dev); return err; } From d871198ee431d90f5308d53998c1ba1d5db5619a Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Tue, 13 May 2025 15:02:23 -0600 Subject: [PATCH 468/555] io_uring/fdinfo: grab ctx->uring_lock around io_uring_show_fdinfo() Not everything requires locking in there, which is why the 'has_lock' variable exists. But enough does that it's a bit unwieldy to manage. Wrap the whole thing in a ->uring_lock trylock, and just return with no output if we fail to grab it. The existing trylock() will already have greatly diminished utility/output for the failure case. This fixes an issue with reading the SQE fields, if the ring is being actively resized at the same time. Reported-by: Jann Horn Fixes: 79cfe9e59c2a ("io_uring/register: add IORING_REGISTER_RESIZE_RINGS") Signed-off-by: Jens Axboe --- io_uring/fdinfo.c | 48 ++++++++++++++++++++++++----------------------- 1 file changed, 25 insertions(+), 23 deletions(-) diff --git a/io_uring/fdinfo.c b/io_uring/fdinfo.c index 9414ca6d101c..e0d6a59a89fa 100644 --- a/io_uring/fdinfo.c +++ b/io_uring/fdinfo.c @@ -86,13 +86,8 @@ static inline void napi_show_fdinfo(struct io_ring_ctx *ctx, } #endif -/* - * Caller holds a reference to the file already, we don't need to do - * anything else to get an extra reference. - */ -__cold void io_uring_show_fdinfo(struct seq_file *m, struct file *file) +static void __io_uring_show_fdinfo(struct io_ring_ctx *ctx, struct seq_file *m) { - struct io_ring_ctx *ctx = file->private_data; struct io_overflow_cqe *ocqe; struct io_rings *r = ctx->rings; struct rusage sq_usage; @@ -106,7 +101,6 @@ __cold void io_uring_show_fdinfo(struct seq_file *m, struct file *file) unsigned int sq_entries, cq_entries; int sq_pid = -1, sq_cpu = -1; u64 sq_total_time = 0, sq_work_time = 0; - bool has_lock; unsigned int i; if (ctx->flags & IORING_SETUP_CQE32) @@ -176,15 +170,7 @@ __cold void io_uring_show_fdinfo(struct seq_file *m, struct file *file) seq_printf(m, "\n"); } - /* - * Avoid ABBA deadlock between the seq lock and the io_uring mutex, - * since fdinfo case grabs it in the opposite direction of normal use - * cases. If we fail to get the lock, we just don't iterate any - * structures that could be going away outside the io_uring mutex. - */ - has_lock = mutex_trylock(&ctx->uring_lock); - - if (has_lock && (ctx->flags & IORING_SETUP_SQPOLL)) { + if (ctx->flags & IORING_SETUP_SQPOLL) { struct io_sq_data *sq = ctx->sq_data; /* @@ -206,7 +192,7 @@ __cold void io_uring_show_fdinfo(struct seq_file *m, struct file *file) seq_printf(m, "SqTotalTime:\t%llu\n", sq_total_time); seq_printf(m, "SqWorkTime:\t%llu\n", sq_work_time); seq_printf(m, "UserFiles:\t%u\n", ctx->file_table.data.nr); - for (i = 0; has_lock && i < ctx->file_table.data.nr; i++) { + for (i = 0; i < ctx->file_table.data.nr; i++) { struct file *f = NULL; if (ctx->file_table.data.nodes[i]) @@ -218,7 +204,7 @@ __cold void io_uring_show_fdinfo(struct seq_file *m, struct file *file) } } seq_printf(m, "UserBufs:\t%u\n", ctx->buf_table.nr); - for (i = 0; has_lock && i < ctx->buf_table.nr; i++) { + for (i = 0; i < ctx->buf_table.nr; i++) { struct io_mapped_ubuf *buf = NULL; if (ctx->buf_table.nodes[i]) @@ -228,7 +214,7 @@ __cold void io_uring_show_fdinfo(struct seq_file *m, struct file *file) else seq_printf(m, "%5u: \n", i); } - if (has_lock && !xa_empty(&ctx->personalities)) { + if (!xa_empty(&ctx->personalities)) { unsigned long index; const struct cred *cred; @@ -238,7 +224,7 @@ __cold void io_uring_show_fdinfo(struct seq_file *m, struct file *file) } seq_puts(m, "PollList:\n"); - for (i = 0; has_lock && i < (1U << ctx->cancel_table.hash_bits); i++) { + for (i = 0; i < (1U << ctx->cancel_table.hash_bits); i++) { struct io_hash_bucket *hb = &ctx->cancel_table.hbs[i]; struct io_kiocb *req; @@ -247,9 +233,6 @@ __cold void io_uring_show_fdinfo(struct seq_file *m, struct file *file) task_work_pending(req->tctx->task)); } - if (has_lock) - mutex_unlock(&ctx->uring_lock); - seq_puts(m, "CqOverflowList:\n"); spin_lock(&ctx->completion_lock); list_for_each_entry(ocqe, &ctx->cq_overflow_list, list) { @@ -262,4 +245,23 @@ __cold void io_uring_show_fdinfo(struct seq_file *m, struct file *file) spin_unlock(&ctx->completion_lock); napi_show_fdinfo(ctx, m); } + +/* + * Caller holds a reference to the file already, we don't need to do + * anything else to get an extra reference. + */ +__cold void io_uring_show_fdinfo(struct seq_file *m, struct file *file) +{ + struct io_ring_ctx *ctx = file->private_data; + + /* + * Avoid ABBA deadlock between the seq lock and the io_uring mutex, + * since fdinfo case grabs it in the opposite direction of normal use + * cases. + */ + if (mutex_trylock(&ctx->uring_lock)) { + __io_uring_show_fdinfo(ctx, m); + mutex_unlock(&ctx->uring_lock); + } +} #endif From 8695f060a02953b33ac6240895dcb9c7ce16c91c Mon Sep 17 00:00:00 2001 From: Alan Adamson Date: Thu, 8 May 2025 15:38:01 -0700 Subject: [PATCH 469/555] nvme: all namespaces in a subsystem must adhere to a common atomic write size The first namespace configured in a subsystem sets the subsystem's atomic write size based on its AWUPF or NAWUPF. Subsequent namespaces must have an atomic write size (per their AWUPF or NAWUPF) less than or equal to the subsystem's atomic write size, or their probing will be rejected. Signed-off-by: Alan Adamson [hch: fold in review comments from John Garry] Signed-off-by: Christoph Hellwig Reviewed-by: John Garry --- drivers/nvme/host/core.c | 30 +++++++++++++++++++++++++++--- drivers/nvme/host/nvme.h | 3 ++- 2 files changed, 29 insertions(+), 4 deletions(-) diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c index ac53629fce68..6b04473c0ab7 100644 --- a/drivers/nvme/host/core.c +++ b/drivers/nvme/host/core.c @@ -2059,7 +2059,21 @@ static bool nvme_update_disk_info(struct nvme_ns *ns, struct nvme_id_ns *id, if (id->nsfeat & NVME_NS_FEAT_ATOMICS && id->nawupf) atomic_bs = (1 + le16_to_cpu(id->nawupf)) * bs; else - atomic_bs = (1 + ns->ctrl->subsys->awupf) * bs; + atomic_bs = (1 + ns->ctrl->awupf) * bs; + + /* + * Set subsystem atomic bs. + */ + if (ns->ctrl->subsys->atomic_bs) { + if (atomic_bs != ns->ctrl->subsys->atomic_bs) { + dev_err_ratelimited(ns->ctrl->device, + "%s: Inconsistent Atomic Write Size, Namespace will not be added: Subsystem=%d bytes, Controller/Namespace=%d bytes\n", + ns->disk ? ns->disk->disk_name : "?", + ns->ctrl->subsys->atomic_bs, + atomic_bs); + } + } else + ns->ctrl->subsys->atomic_bs = atomic_bs; nvme_update_atomic_write_disk_info(ns, id, lim, bs, atomic_bs); } @@ -2201,6 +2215,17 @@ static int nvme_update_ns_info_block(struct nvme_ns *ns, nvme_set_chunk_sectors(ns, id, &lim); if (!nvme_update_disk_info(ns, id, &lim)) capacity = 0; + + /* + * Validate the max atomic write size fits within the subsystem's + * atomic write capabilities. + */ + if (lim.atomic_write_hw_max > ns->ctrl->subsys->atomic_bs) { + blk_mq_unfreeze_queue(ns->disk->queue, memflags); + ret = -ENXIO; + goto out; + } + nvme_config_discard(ns, &lim); if (IS_ENABLED(CONFIG_BLK_DEV_ZONED) && ns->head->ids.csi == NVME_CSI_ZNS) @@ -3031,7 +3056,6 @@ static int nvme_init_subsystem(struct nvme_ctrl *ctrl, struct nvme_id_ctrl *id) kfree(subsys); return -EINVAL; } - subsys->awupf = le16_to_cpu(id->awupf); nvme_mpath_default_iopolicy(subsys); subsys->dev.class = &nvme_subsys_class; @@ -3441,7 +3465,7 @@ static int nvme_init_identify(struct nvme_ctrl *ctrl) dev_pm_qos_expose_latency_tolerance(ctrl->device); else if (!ctrl->apst_enabled && prev_apst_enabled) dev_pm_qos_hide_latency_tolerance(ctrl->device); - + ctrl->awupf = le16_to_cpu(id->awupf); out_free: kfree(id); return ret; diff --git a/drivers/nvme/host/nvme.h b/drivers/nvme/host/nvme.h index 51e078642127..8fc4683418a3 100644 --- a/drivers/nvme/host/nvme.h +++ b/drivers/nvme/host/nvme.h @@ -410,6 +410,7 @@ struct nvme_ctrl { enum nvme_ctrl_type cntrltype; enum nvme_dctype dctype; + u16 awupf; /* 0's based value. */ }; static inline enum nvme_ctrl_state nvme_ctrl_state(struct nvme_ctrl *ctrl) @@ -442,11 +443,11 @@ struct nvme_subsystem { u8 cmic; enum nvme_subsys_type subtype; u16 vendor_id; - u16 awupf; /* 0's based awupf value. */ struct ida ns_ida; #ifdef CONFIG_NVME_MULTIPATH enum nvme_iopolicy iopolicy; #endif + u32 atomic_bs; }; /* From 72c7d62583ebce7baeb61acce6057c361f73be4a Mon Sep 17 00:00:00 2001 From: Hyejeong Choi Date: Mon, 12 May 2025 21:06:38 -0500 Subject: [PATCH 470/555] dma-buf: insert memory barrier before updating num_fences MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit smp_store_mb() inserts memory barrier after storing operation. It is different with what the comment is originally aiming so Null pointer dereference can be happened if memory update is reordered. Signed-off-by: Hyejeong Choi Fixes: a590d0fdbaa5 ("dma-buf: Update reservation shared_count after adding the new fence") CC: stable@vger.kernel.org Reviewed-by: Christian König Link: https://lore.kernel.org/r/20250513020638.GA2329653@au1-maretx-p37.eng.sarc.samsung.com Signed-off-by: Christian König --- drivers/dma-buf/dma-resv.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/dma-buf/dma-resv.c b/drivers/dma-buf/dma-resv.c index 5f8d010516f0..b1ef4546346d 100644 --- a/drivers/dma-buf/dma-resv.c +++ b/drivers/dma-buf/dma-resv.c @@ -320,8 +320,9 @@ void dma_resv_add_fence(struct dma_resv *obj, struct dma_fence *fence, count++; dma_resv_list_set(fobj, i, fence, usage); - /* pointer update must be visible before we extend the num_fences */ - smp_store_mb(fobj->num_fences, count); + /* fence update must be visible before we extend the num_fences */ + smp_wmb(); + fobj->num_fences = count; } EXPORT_SYMBOL(dma_resv_add_fence); From 09dab6ce0243bc1939bd1f77a066ccdd72d44efe Mon Sep 17 00:00:00 2001 From: Wengang Wang Date: Mon, 5 May 2025 16:35:49 -0700 Subject: [PATCH 471/555] xfs: free up mp->m_free[0].count in error case In xfs_init_percpu_counters(), memory for mp->m_free[0].count wasn't freed in error case. Free it up in this patch. Signed-off-by: Wengang Wang Fixes: 712bae96631852 ("xfs: generalize the freespace and reserved blocks handling") Reviewed-by: Darrick J. Wong Reviewed-by: Christoph Hellwig Signed-off-by: Carlos Maiolino --- fs/xfs/xfs_super.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index b2dd0c0bf509..3be041647ec1 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -1149,7 +1149,7 @@ xfs_init_percpu_counters( return 0; free_freecounters: - while (--i > 0) + while (--i >= 0) percpu_counter_destroy(&mp->m_free[i].count); percpu_counter_destroy(&mp->m_delalloc_rtextents); free_delalloc: From fbecd731de05707a73a3d86c6cbe6b9441cdbedc Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 12 May 2025 16:43:05 +0200 Subject: [PATCH 472/555] xfs: fix zoned GC data corruption due to wrong bv_offset xfs_zone_gc_write_chunk writes out the data buffer read in earlier using the same bio, and currenly looks at bv_offset for the offset into the scratch folio for that. But commit 26064d3e2b4d ("block: fix adding folio to bio") changed how bv_page and bv_offset are calculated for adding larger folios, breaking this fragile logic. Switch to extracting the full physical address from the old bio_vec, and calculate the offset into the folio from that instead. This fixes data corruption during garbage collection with heavy rockdsb workloads. Thanks to Hans for tracking down the culprit commit during long bisection sessions. Fixes: 26064d3e2b4d ("block: fix adding folio to bio") Fixes: 080d01c41d44 ("xfs: implement zoned garbage collection") Reported-by: Hans Holmberg Signed-off-by: Christoph Hellwig Reviewed-by: Hans Holmberg Tested-by: Hans Holmberg Reviewed-by: Darrick J. Wong Signed-off-by: Carlos Maiolino --- fs/xfs/xfs_zone_gc.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/fs/xfs/xfs_zone_gc.c b/fs/xfs/xfs_zone_gc.c index 81c94dd1d596..d613a4094db6 100644 --- a/fs/xfs/xfs_zone_gc.c +++ b/fs/xfs/xfs_zone_gc.c @@ -807,7 +807,8 @@ xfs_zone_gc_write_chunk( { struct xfs_zone_gc_data *data = chunk->data; struct xfs_mount *mp = chunk->ip->i_mount; - unsigned int folio_offset = chunk->bio.bi_io_vec->bv_offset; + phys_addr_t bvec_paddr = + bvec_phys(bio_first_bvec_all(&chunk->bio)); struct xfs_gc_bio *split_chunk; if (chunk->bio.bi_status) @@ -822,7 +823,7 @@ xfs_zone_gc_write_chunk( bio_reset(&chunk->bio, mp->m_rtdev_targp->bt_bdev, REQ_OP_WRITE); bio_add_folio_nofail(&chunk->bio, chunk->scratch->folio, chunk->len, - folio_offset); + offset_in_folio(chunk->scratch->folio, bvec_paddr)); while ((split_chunk = xfs_zone_gc_split_write(data, chunk))) xfs_zone_gc_submit_write(data, split_chunk); From 95b613339c0e5fe651a3ef7605708478bc34a5af Mon Sep 17 00:00:00 2001 From: "Nirjhar Roy (IBM)" Date: Mon, 12 May 2025 22:00:32 +0530 Subject: [PATCH 473/555] xfs: Fail remount with noattr2 on a v5 with v4 enabled Bug: When we compile the kernel with CONFIG_XFS_SUPPORT_V4=y, remount with "-o remount,noattr2" on a v5 XFS does not fail explicitly. Reproduction: mkfs.xfs -f /dev/loop0 mount /dev/loop0 /mnt/scratch mount -o remount,noattr2 /dev/loop0 /mnt/scratch However, with CONFIG_XFS_SUPPORT_V4=n, the remount correctly fails explicitly. This is because the way the following 2 functions are defined: static inline bool xfs_has_attr2 (struct xfs_mount *mp) { return !IS_ENABLED(CONFIG_XFS_SUPPORT_V4) || (mp->m_features & XFS_FEAT_ATTR2); } static inline bool xfs_has_noattr2 (const struct xfs_mount *mp) { return mp->m_features & XFS_FEAT_NOATTR2; } xfs_has_attr2() returns true when CONFIG_XFS_SUPPORT_V4=n and hence, the following if condition in xfs_fs_validate_params() succeeds and returns -EINVAL: /* * We have not read the superblock at this point, so only the attr2 * mount option can set the attr2 feature by this stage. */ if (xfs_has_attr2(mp) && xfs_has_noattr2(mp)) { xfs_warn(mp, "attr2 and noattr2 cannot both be specified."); return -EINVAL; } With CONFIG_XFS_SUPPORT_V4=y, xfs_has_attr2() always return false and hence no error is returned. Fix: Check if the existing mount has crc enabled(i.e, of type v5 and has attr2 enabled) and the remount has noattr2, if yes, return -EINVAL. I have tested xfs/{189,539} in fstests with v4 and v5 XFS with both CONFIG_XFS_SUPPORT_V4=y/n and they both behave as expected. This patch also fixes remount from noattr2 -> attr2 (on a v4 xfs). Related discussion in [1] [1] https://lore.kernel.org/all/Z65o6nWxT00MaUrW@dread.disaster.area/ Signed-off-by: Nirjhar Roy (IBM) Reviewed-by: Christoph Hellwig Reviewed-by: Carlos Maiolino Signed-off-by: Carlos Maiolino --- fs/xfs/xfs_super.c | 26 ++++++++++++++++++++++++++ 1 file changed, 26 insertions(+) diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index 3be041647ec1..4a11ddccc563 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -2114,6 +2114,21 @@ xfs_fs_reconfigure( if (error) return error; + /* attr2 -> noattr2 */ + if (xfs_has_noattr2(new_mp)) { + if (xfs_has_crc(mp)) { + xfs_warn(mp, + "attr2 is always enabled for a V5 filesystem - can't be changed."); + return -EINVAL; + } + mp->m_features &= ~XFS_FEAT_ATTR2; + mp->m_features |= XFS_FEAT_NOATTR2; + } else if (xfs_has_attr2(new_mp)) { + /* noattr2 -> attr2 */ + mp->m_features &= ~XFS_FEAT_NOATTR2; + mp->m_features |= XFS_FEAT_ATTR2; + } + /* inode32 -> inode64 */ if (xfs_has_small_inums(mp) && !xfs_has_small_inums(new_mp)) { mp->m_features &= ~XFS_FEAT_SMALL_INUMS; @@ -2126,6 +2141,17 @@ xfs_fs_reconfigure( mp->m_maxagi = xfs_set_inode_alloc(mp, mp->m_sb.sb_agcount); } + /* + * Now that mp has been modified according to the remount options, we + * do a final option validation with xfs_finish_flags() just like it is + * just like it is done during mount. We cannot use + * done during mount. We cannot use xfs_finish_flags() on new_mp as it + * contains only the user given options. + */ + error = xfs_finish_flags(mp); + if (error) + return error; + /* ro -> rw */ if (xfs_is_readonly(mp) && !(flags & SB_RDONLY)) { error = xfs_remount_rw(mp); From fa8deae92f473a2ebc0e1c7cfa316f5c083e1880 Mon Sep 17 00:00:00 2001 From: Carlos Maiolino Date: Mon, 12 May 2025 13:42:55 +0200 Subject: [PATCH 474/555] xfs: Fix a comment on xfs_ail_delete It doesn't return anything. Reviewed-by: Christoph Hellwig Reviewed-by: Darrick J. Wong Signed-off-by: Carlos Maiolino Reviewed-by: Chandan Babu R Signed-off-by: Carlos Maiolino --- fs/xfs/xfs_trans_ail.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/xfs/xfs_trans_ail.c b/fs/xfs/xfs_trans_ail.c index 85a649fec6ac..7d327a3e5a73 100644 --- a/fs/xfs/xfs_trans_ail.c +++ b/fs/xfs/xfs_trans_ail.c @@ -315,7 +315,7 @@ xfs_ail_splice( } /* - * Delete the given item from the AIL. Return a pointer to the item. + * Delete the given item from the AIL. */ static void xfs_ail_delete( From 08c73a4b2e3cd2ff9db0ecb3c5c0dd7e23edc770 Mon Sep 17 00:00:00 2001 From: Carlos Maiolino Date: Mon, 12 May 2025 13:42:56 +0200 Subject: [PATCH 475/555] xfs: Fix comment on xfs_trans_ail_update_bulk() This function doesn't take the AIL lock, but should be called with AIL lock held. Also (hopefuly) simplify the comment. Signed-off-by: Carlos Maiolino Reviewed-by: Christoph Hellwig Reviewed-by: Chandan Babu R Signed-off-by: Carlos Maiolino --- fs/xfs/xfs_trans_ail.c | 32 +++++++++++++++++--------------- 1 file changed, 17 insertions(+), 15 deletions(-) diff --git a/fs/xfs/xfs_trans_ail.c b/fs/xfs/xfs_trans_ail.c index 7d327a3e5a73..67c328d23e4a 100644 --- a/fs/xfs/xfs_trans_ail.c +++ b/fs/xfs/xfs_trans_ail.c @@ -777,26 +777,28 @@ xfs_ail_update_finish( } /* - * xfs_trans_ail_update - bulk AIL insertion operation. + * xfs_trans_ail_update_bulk - bulk AIL insertion operation. * - * @xfs_trans_ail_update takes an array of log items that all need to be + * @xfs_trans_ail_update_bulk takes an array of log items that all need to be * positioned at the same LSN in the AIL. If an item is not in the AIL, it will - * be added. Otherwise, it will be repositioned by removing it and re-adding - * it to the AIL. If we move the first item in the AIL, update the log tail to - * match the new minimum LSN in the AIL. + * be added. Otherwise, it will be repositioned by removing it and re-adding + * it to the AIL. * - * This function takes the AIL lock once to execute the update operations on - * all the items in the array, and as such should not be called with the AIL - * lock held. As a result, once we have the AIL lock, we need to check each log - * item LSN to confirm it needs to be moved forward in the AIL. + * If we move the first item in the AIL, update the log tail to match the new + * minimum LSN in the AIL. * - * To optimise the insert operation, we delete all the items from the AIL in - * the first pass, moving them into a temporary list, then splice the temporary - * list into the correct position in the AIL. This avoids needing to do an - * insert operation on every item. + * This function should be called with the AIL lock held. * - * This function must be called with the AIL lock held. The lock is dropped - * before returning. + * To optimise the insert operation, we add all items to a temporary list, then + * splice this list into the correct position in the AIL. + * + * Items that are already in the AIL are first deleted from their current + * location before being added to the temporary list. + * + * This avoids needing to do an insert operation on every item. + * + * The AIL lock is dropped by xfs_ail_update_finish() before returning to + * the caller. */ void xfs_trans_ail_update_bulk( From 3fd2f4bc010cdfbc07dd21018dc65bd9370eb7a4 Mon Sep 17 00:00:00 2001 From: Shuai Xue Date: Fri, 4 Apr 2025 20:02:09 +0800 Subject: [PATCH 476/555] dmaengine: idxd: fix memory leak in error handling path of idxd_setup_wqs Memory allocated for wqs is not freed if an error occurs during idxd_setup_wqs(). To fix it, free the allocated memory in the reverse order of allocation before exiting the function in case of an error. Fixes: 7c5dd23e57c1 ("dmaengine: idxd: fix wq conf_dev 'struct device' lifetime") Fixes: 700af3a0a26c ("dmaengine: idxd: add 'struct idxd_dev' as wrapper for conf_dev") Fixes: de5819b99489 ("dmaengine: idxd: track enabled workqueues in bitmap") Fixes: b0325aefd398 ("dmaengine: idxd: add WQ operation cap restriction support") Cc: stable@vger.kernel.org Signed-off-by: Shuai Xue Reviewed-by: Dave Jiang Reviewed-by: Fenghua Yu Link: https://lore.kernel.org/r/20250404120217.48772-2-xueshuai@linux.alibaba.com Signed-off-by: Vinod Koul --- drivers/dma/idxd/init.c | 30 +++++++++++++++++++++--------- 1 file changed, 21 insertions(+), 9 deletions(-) diff --git a/drivers/dma/idxd/init.c b/drivers/dma/idxd/init.c index fca1d2924999..80fb189c9624 100644 --- a/drivers/dma/idxd/init.c +++ b/drivers/dma/idxd/init.c @@ -169,8 +169,8 @@ static int idxd_setup_wqs(struct idxd_device *idxd) idxd->wq_enable_map = bitmap_zalloc_node(idxd->max_wqs, GFP_KERNEL, dev_to_node(dev)); if (!idxd->wq_enable_map) { - kfree(idxd->wqs); - return -ENOMEM; + rc = -ENOMEM; + goto err_bitmap; } for (i = 0; i < idxd->max_wqs; i++) { @@ -189,10 +189,8 @@ static int idxd_setup_wqs(struct idxd_device *idxd) conf_dev->bus = &dsa_bus_type; conf_dev->type = &idxd_wq_device_type; rc = dev_set_name(conf_dev, "wq%d.%d", idxd->id, wq->id); - if (rc < 0) { - put_device(conf_dev); + if (rc < 0) goto err; - } mutex_init(&wq->wq_lock); init_waitqueue_head(&wq->err_queue); @@ -203,7 +201,6 @@ static int idxd_setup_wqs(struct idxd_device *idxd) wq->enqcmds_retries = IDXD_ENQCMDS_RETRIES; wq->wqcfg = kzalloc_node(idxd->wqcfg_size, GFP_KERNEL, dev_to_node(dev)); if (!wq->wqcfg) { - put_device(conf_dev); rc = -ENOMEM; goto err; } @@ -211,9 +208,8 @@ static int idxd_setup_wqs(struct idxd_device *idxd) if (idxd->hw.wq_cap.op_config) { wq->opcap_bmap = bitmap_zalloc(IDXD_MAX_OPCAP_BITS, GFP_KERNEL); if (!wq->opcap_bmap) { - put_device(conf_dev); rc = -ENOMEM; - goto err; + goto err_opcap_bmap; } bitmap_copy(wq->opcap_bmap, idxd->opcap_bmap, IDXD_MAX_OPCAP_BITS); } @@ -224,12 +220,28 @@ static int idxd_setup_wqs(struct idxd_device *idxd) return 0; - err: +err_opcap_bmap: + kfree(wq->wqcfg); + +err: + put_device(conf_dev); + kfree(wq); + while (--i >= 0) { wq = idxd->wqs[i]; + if (idxd->hw.wq_cap.op_config) + bitmap_free(wq->opcap_bmap); + kfree(wq->wqcfg); conf_dev = wq_confdev(wq); put_device(conf_dev); + kfree(wq); + } + bitmap_free(idxd->wq_enable_map); + +err_bitmap: + kfree(idxd->wqs); + return rc; } From 817bced19d1dbdd0b473580d026dc0983e30e17b Mon Sep 17 00:00:00 2001 From: Shuai Xue Date: Fri, 4 Apr 2025 20:02:10 +0800 Subject: [PATCH 477/555] dmaengine: idxd: fix memory leak in error handling path of idxd_setup_engines Memory allocated for engines is not freed if an error occurs during idxd_setup_engines(). To fix it, free the allocated memory in the reverse order of allocation before exiting the function in case of an error. Fixes: 75b911309060 ("dmaengine: idxd: fix engine conf_dev lifetime") Cc: stable@vger.kernel.org Signed-off-by: Shuai Xue Reviewed-by: Dave Jiang Reviewed-by: Fenghua Yu Link: https://lore.kernel.org/r/20250404120217.48772-3-xueshuai@linux.alibaba.com Signed-off-by: Vinod Koul --- drivers/dma/idxd/init.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/dma/idxd/init.c b/drivers/dma/idxd/init.c index 80fb189c9624..ff6ec3c0f604 100644 --- a/drivers/dma/idxd/init.c +++ b/drivers/dma/idxd/init.c @@ -275,6 +275,7 @@ static int idxd_setup_engines(struct idxd_device *idxd) rc = dev_set_name(conf_dev, "engine%d.%d", idxd->id, engine->id); if (rc < 0) { put_device(conf_dev); + kfree(engine); goto err; } @@ -288,7 +289,10 @@ static int idxd_setup_engines(struct idxd_device *idxd) engine = idxd->engines[i]; conf_dev = engine_confdev(engine); put_device(conf_dev); + kfree(engine); } + kfree(idxd->engines); + return rc; } From aa6f4f945b10eac57aed46154ae7d6fada7fccc7 Mon Sep 17 00:00:00 2001 From: Shuai Xue Date: Fri, 4 Apr 2025 20:02:11 +0800 Subject: [PATCH 478/555] dmaengine: idxd: fix memory leak in error handling path of idxd_setup_groups Memory allocated for groups is not freed if an error occurs during idxd_setup_groups(). To fix it, free the allocated memory in the reverse order of allocation before exiting the function in case of an error. Fixes: defe49f96012 ("dmaengine: idxd: fix group conf_dev lifetime") Cc: stable@vger.kernel.org Signed-off-by: Shuai Xue Reviewed-by: Dave Jiang Reviewed-by: Fenghua Yu Link: https://lore.kernel.org/r/20250404120217.48772-4-xueshuai@linux.alibaba.com Signed-off-by: Vinod Koul --- drivers/dma/idxd/init.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/dma/idxd/init.c b/drivers/dma/idxd/init.c index ff6ec3c0f604..7f0a26e2e0a5 100644 --- a/drivers/dma/idxd/init.c +++ b/drivers/dma/idxd/init.c @@ -326,6 +326,7 @@ static int idxd_setup_groups(struct idxd_device *idxd) rc = dev_set_name(conf_dev, "group%d.%d", idxd->id, group->id); if (rc < 0) { put_device(conf_dev); + kfree(group); goto err; } @@ -350,7 +351,10 @@ static int idxd_setup_groups(struct idxd_device *idxd) while (--i >= 0) { group = idxd->groups[i]; put_device(group_confdev(group)); + kfree(group); } + kfree(idxd->groups); + return rc; } From 61259fb96e023f7299c442c48b13e72c441fc0f2 Mon Sep 17 00:00:00 2001 From: Shuai Xue Date: Fri, 4 Apr 2025 20:02:12 +0800 Subject: [PATCH 479/555] dmaengine: idxd: Add missing cleanup for early error out in idxd_setup_internals The idxd_setup_internals() is missing some cleanup when things fail in the middle. Add the appropriate cleanup routines: - cleanup groups - cleanup enginces - cleanup wqs to make sure it exits gracefully. Fixes: defe49f96012 ("dmaengine: idxd: fix group conf_dev lifetime") Cc: stable@vger.kernel.org Suggested-by: Fenghua Yu Signed-off-by: Shuai Xue Reviewed-by: Fenghua Yu Reviewed-by: Dave Jiang Link: https://lore.kernel.org/r/20250404120217.48772-5-xueshuai@linux.alibaba.com Signed-off-by: Vinod Koul --- drivers/dma/idxd/init.c | 58 ++++++++++++++++++++++++++++++++++++----- 1 file changed, 51 insertions(+), 7 deletions(-) diff --git a/drivers/dma/idxd/init.c b/drivers/dma/idxd/init.c index 7f0a26e2e0a5..a40fb2fd5006 100644 --- a/drivers/dma/idxd/init.c +++ b/drivers/dma/idxd/init.c @@ -155,6 +155,25 @@ static void idxd_cleanup_interrupts(struct idxd_device *idxd) pci_free_irq_vectors(pdev); } +static void idxd_clean_wqs(struct idxd_device *idxd) +{ + struct idxd_wq *wq; + struct device *conf_dev; + int i; + + for (i = 0; i < idxd->max_wqs; i++) { + wq = idxd->wqs[i]; + if (idxd->hw.wq_cap.op_config) + bitmap_free(wq->opcap_bmap); + kfree(wq->wqcfg); + conf_dev = wq_confdev(wq); + put_device(conf_dev); + kfree(wq); + } + bitmap_free(idxd->wq_enable_map); + kfree(idxd->wqs); +} + static int idxd_setup_wqs(struct idxd_device *idxd) { struct device *dev = &idxd->pdev->dev; @@ -245,6 +264,21 @@ err_bitmap: return rc; } +static void idxd_clean_engines(struct idxd_device *idxd) +{ + struct idxd_engine *engine; + struct device *conf_dev; + int i; + + for (i = 0; i < idxd->max_engines; i++) { + engine = idxd->engines[i]; + conf_dev = engine_confdev(engine); + put_device(conf_dev); + kfree(engine); + } + kfree(idxd->engines); +} + static int idxd_setup_engines(struct idxd_device *idxd) { struct idxd_engine *engine; @@ -296,6 +330,19 @@ static int idxd_setup_engines(struct idxd_device *idxd) return rc; } +static void idxd_clean_groups(struct idxd_device *idxd) +{ + struct idxd_group *group; + int i; + + for (i = 0; i < idxd->max_groups; i++) { + group = idxd->groups[i]; + put_device(group_confdev(group)); + kfree(group); + } + kfree(idxd->groups); +} + static int idxd_setup_groups(struct idxd_device *idxd) { struct device *dev = &idxd->pdev->dev; @@ -410,7 +457,7 @@ static int idxd_init_evl(struct idxd_device *idxd) static int idxd_setup_internals(struct idxd_device *idxd) { struct device *dev = &idxd->pdev->dev; - int rc, i; + int rc; init_waitqueue_head(&idxd->cmd_waitq); @@ -441,14 +488,11 @@ static int idxd_setup_internals(struct idxd_device *idxd) err_evl: destroy_workqueue(idxd->wq); err_wkq_create: - for (i = 0; i < idxd->max_groups; i++) - put_device(group_confdev(idxd->groups[i])); + idxd_clean_groups(idxd); err_group: - for (i = 0; i < idxd->max_engines; i++) - put_device(engine_confdev(idxd->engines[i])); + idxd_clean_engines(idxd); err_engine: - for (i = 0; i < idxd->max_wqs; i++) - put_device(wq_confdev(idxd->wqs[i])); + idxd_clean_wqs(idxd); err_wqs: return rc; } From 61d651572b6c4fe50c7b39a390760f3a910c7ccf Mon Sep 17 00:00:00 2001 From: Shuai Xue Date: Fri, 4 Apr 2025 20:02:13 +0800 Subject: [PATCH 480/555] dmaengine: idxd: Add missing cleanups in cleanup internals The idxd_cleanup_internals() function only decreases the reference count of groups, engines, and wqs but is missing the step to release memory resources. To fix this, use the cleanup helper to properly release the memory resources. Fixes: ddf742d4f3f1 ("dmaengine: idxd: Add missing cleanup for early error out in probe call") Cc: stable@vger.kernel.org Signed-off-by: Shuai Xue Reviewed-by: Fenghua Yu Reviewed-by: Dave Jiang Link: https://lore.kernel.org/r/20250404120217.48772-6-xueshuai@linux.alibaba.com Signed-off-by: Vinod Koul --- drivers/dma/idxd/init.c | 11 +++-------- 1 file changed, 3 insertions(+), 8 deletions(-) diff --git a/drivers/dma/idxd/init.c b/drivers/dma/idxd/init.c index a40fb2fd5006..f8129d2d53f1 100644 --- a/drivers/dma/idxd/init.c +++ b/drivers/dma/idxd/init.c @@ -407,14 +407,9 @@ static int idxd_setup_groups(struct idxd_device *idxd) static void idxd_cleanup_internals(struct idxd_device *idxd) { - int i; - - for (i = 0; i < idxd->max_groups; i++) - put_device(group_confdev(idxd->groups[i])); - for (i = 0; i < idxd->max_engines; i++) - put_device(engine_confdev(idxd->engines[i])); - for (i = 0; i < idxd->max_wqs; i++) - put_device(wq_confdev(idxd->wqs[i])); + idxd_clean_groups(idxd); + idxd_clean_engines(idxd); + idxd_clean_wqs(idxd); destroy_workqueue(idxd->wq); } From 46a5cca76c76c86063000a12936f8e7875295838 Mon Sep 17 00:00:00 2001 From: Shuai Xue Date: Fri, 4 Apr 2025 20:02:14 +0800 Subject: [PATCH 481/555] dmaengine: idxd: fix memory leak in error handling path of idxd_alloc Memory allocated for idxd is not freed if an error occurs during idxd_alloc(). To fix it, free the allocated memory in the reverse order of allocation before exiting the function in case of an error. Fixes: a8563a33a5e2 ("dmanegine: idxd: reformat opcap output to match bitmap_parse() input") Cc: stable@vger.kernel.org Signed-off-by: Shuai Xue Reviewed-by: Dave Jiang Reviewed-by: Fenghua Yu Link: https://lore.kernel.org/r/20250404120217.48772-7-xueshuai@linux.alibaba.com Signed-off-by: Vinod Koul --- drivers/dma/idxd/init.c | 24 +++++++++++++++--------- 1 file changed, 15 insertions(+), 9 deletions(-) diff --git a/drivers/dma/idxd/init.c b/drivers/dma/idxd/init.c index f8129d2d53f1..302d8983ed8c 100644 --- a/drivers/dma/idxd/init.c +++ b/drivers/dma/idxd/init.c @@ -604,28 +604,34 @@ static struct idxd_device *idxd_alloc(struct pci_dev *pdev, struct idxd_driver_d idxd_dev_set_type(&idxd->idxd_dev, idxd->data->type); idxd->id = ida_alloc(&idxd_ida, GFP_KERNEL); if (idxd->id < 0) - return NULL; + goto err_ida; idxd->opcap_bmap = bitmap_zalloc_node(IDXD_MAX_OPCAP_BITS, GFP_KERNEL, dev_to_node(dev)); - if (!idxd->opcap_bmap) { - ida_free(&idxd_ida, idxd->id); - return NULL; - } + if (!idxd->opcap_bmap) + goto err_opcap; device_initialize(conf_dev); conf_dev->parent = dev; conf_dev->bus = &dsa_bus_type; conf_dev->type = idxd->data->dev_type; rc = dev_set_name(conf_dev, "%s%d", idxd->data->name_prefix, idxd->id); - if (rc < 0) { - put_device(conf_dev); - return NULL; - } + if (rc < 0) + goto err_name; spin_lock_init(&idxd->dev_lock); spin_lock_init(&idxd->cmd_lock); return idxd; + +err_name: + put_device(conf_dev); + bitmap_free(idxd->opcap_bmap); +err_opcap: + ida_free(&idxd_ida, idxd->id); +err_ida: + kfree(idxd); + + return NULL; } static int idxd_enable_system_pasid(struct idxd_device *idxd) From 90022b3a6981ec234902be5dbf0f983a12c759fc Mon Sep 17 00:00:00 2001 From: Shuai Xue Date: Fri, 4 Apr 2025 20:02:15 +0800 Subject: [PATCH 482/555] dmaengine: idxd: fix memory leak in error handling path of idxd_pci_probe Memory allocated for idxd is not freed if an error occurs during idxd_pci_probe(). To fix it, free the allocated memory in the reverse order of allocation before exiting the function in case of an error. Fixes: bfe1d56091c1 ("dmaengine: idxd: Init and probe for Intel data accelerators") Cc: stable@vger.kernel.org Signed-off-by: Shuai Xue Reviewed-by: Dave Jiang Reviewed-by: Fenghua Yu Link: https://lore.kernel.org/r/20250404120217.48772-8-xueshuai@linux.alibaba.com Signed-off-by: Vinod Koul --- drivers/dma/idxd/init.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) diff --git a/drivers/dma/idxd/init.c b/drivers/dma/idxd/init.c index 302d8983ed8c..f2b5b17538c0 100644 --- a/drivers/dma/idxd/init.c +++ b/drivers/dma/idxd/init.c @@ -587,6 +587,17 @@ static void idxd_read_caps(struct idxd_device *idxd) idxd->hw.iaa_cap.bits = ioread64(idxd->reg_base + IDXD_IAACAP_OFFSET); } +static void idxd_free(struct idxd_device *idxd) +{ + if (!idxd) + return; + + put_device(idxd_confdev(idxd)); + bitmap_free(idxd->opcap_bmap); + ida_free(&idxd_ida, idxd->id); + kfree(idxd); +} + static struct idxd_device *idxd_alloc(struct pci_dev *pdev, struct idxd_driver_data *data) { struct device *dev = &pdev->dev; @@ -1255,7 +1266,7 @@ int idxd_pci_probe_alloc(struct idxd_device *idxd, struct pci_dev *pdev, err: pci_iounmap(pdev, idxd->reg_base); err_iomap: - put_device(idxd_confdev(idxd)); + idxd_free(idxd); err_idxd_alloc: pci_disable_device(pdev); return rc; From d5449ff1b04dfe9ed8e455769aa01e4c2ccf6805 Mon Sep 17 00:00:00 2001 From: Shuai Xue Date: Fri, 4 Apr 2025 20:02:16 +0800 Subject: [PATCH 483/555] dmaengine: idxd: Add missing idxd cleanup to fix memory leak in remove call The remove call stack is missing idxd cleanup to free bitmap, ida and the idxd_device. Call idxd_free() helper routines to make sure we exit gracefully. Fixes: bfe1d56091c1 ("dmaengine: idxd: Init and probe for Intel data accelerators") Cc: stable@vger.kernel.org Suggested-by: Vinicius Costa Gomes Signed-off-by: Shuai Xue Reviewed-by: Fenghua Yu Reviewed-by: Dave Jiang Link: https://lore.kernel.org/r/20250404120217.48772-9-xueshuai@linux.alibaba.com Signed-off-by: Vinod Koul --- drivers/dma/idxd/init.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/dma/idxd/init.c b/drivers/dma/idxd/init.c index f2b5b17538c0..974b926bd930 100644 --- a/drivers/dma/idxd/init.c +++ b/drivers/dma/idxd/init.c @@ -1335,6 +1335,7 @@ static void idxd_remove(struct pci_dev *pdev) destroy_workqueue(idxd->wq); perfmon_pmu_remove(idxd); put_device(idxd_confdev(idxd)); + idxd_free(idxd); } static struct pci_driver idxd_pci_driver = { From a409e919ca321cc0e28f8abf96fde299f0072a81 Mon Sep 17 00:00:00 2001 From: Shuai Xue Date: Fri, 4 Apr 2025 20:02:17 +0800 Subject: [PATCH 484/555] dmaengine: idxd: Refactor remove call with idxd_cleanup() helper The idxd_cleanup() helper cleans up perfmon, interrupts, internals and so on. Refactor remove call with the idxd_cleanup() helper to avoid code duplication. Note, this also fixes the missing put_device() for idxd groups, enginces and wqs. Fixes: bfe1d56091c1 ("dmaengine: idxd: Init and probe for Intel data accelerators") Cc: stable@vger.kernel.org Suggested-by: Vinicius Costa Gomes Signed-off-by: Shuai Xue Reviewed-by: Fenghua Yu Reviewed-by: Dave Jiang Link: https://lore.kernel.org/r/20250404120217.48772-10-xueshuai@linux.alibaba.com Signed-off-by: Vinod Koul --- drivers/dma/idxd/init.c | 14 ++------------ 1 file changed, 2 insertions(+), 12 deletions(-) diff --git a/drivers/dma/idxd/init.c b/drivers/dma/idxd/init.c index 974b926bd930..760b7d81fcd8 100644 --- a/drivers/dma/idxd/init.c +++ b/drivers/dma/idxd/init.c @@ -1308,7 +1308,6 @@ static void idxd_shutdown(struct pci_dev *pdev) static void idxd_remove(struct pci_dev *pdev) { struct idxd_device *idxd = pci_get_drvdata(pdev); - struct idxd_irq_entry *irq_entry; idxd_unregister_devices(idxd); /* @@ -1321,21 +1320,12 @@ static void idxd_remove(struct pci_dev *pdev) get_device(idxd_confdev(idxd)); device_unregister(idxd_confdev(idxd)); idxd_shutdown(pdev); - if (device_pasid_enabled(idxd)) - idxd_disable_system_pasid(idxd); idxd_device_remove_debugfs(idxd); - - irq_entry = idxd_get_ie(idxd, 0); - free_irq(irq_entry->vector, irq_entry); - pci_free_irq_vectors(pdev); + idxd_cleanup(idxd); pci_iounmap(pdev, idxd->reg_base); - if (device_user_pasid_enabled(idxd)) - idxd_disable_sva(pdev); - pci_disable_device(pdev); - destroy_workqueue(idxd->wq); - perfmon_pmu_remove(idxd); put_device(idxd_confdev(idxd)); idxd_free(idxd); + pci_disable_device(pdev); } static struct pci_driver idxd_pci_driver = { From ae74cd15ade833adc289279b5c6f12e78f64d4d7 Mon Sep 17 00:00:00 2001 From: Dave Jiang Date: Thu, 8 May 2025 10:05:48 -0700 Subject: [PATCH 485/555] dmaengine: idxd: Fix ->poll() return value The fix to block access from different address space did not return a correct value for ->poll() change. kernel test bot reported that a return value of type __poll_t is expected rather than int. Fix to return POLLNVAL to indicate invalid request. Fixes: 8dfa57aabff6 ("dmaengine: idxd: Fix allowing write() from different address spaces") Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202505081851.rwD7jVxg-lkp@intel.com/ Signed-off-by: Dave Jiang Link: https://lore.kernel.org/r/20250508170548.2747425-1-dave.jiang@intel.com Signed-off-by: Vinod Koul --- drivers/dma/idxd/cdev.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/dma/idxd/cdev.c b/drivers/dma/idxd/cdev.c index a2fb2b847752..6d12033649f8 100644 --- a/drivers/dma/idxd/cdev.c +++ b/drivers/dma/idxd/cdev.c @@ -500,7 +500,7 @@ static __poll_t idxd_cdev_poll(struct file *filp, __poll_t out = 0; if (current->mm != ctx->mm) - return -EPERM; + return POLLNVAL; poll_wait(filp, &wq->err_queue, wait); spin_lock(&idxd->dev_lock); From 157ae5ffd76a2857ccb4b7ce40bc5a344ca00395 Mon Sep 17 00:00:00 2001 From: Qiu-ji Chen Date: Thu, 8 May 2025 15:36:33 +0800 Subject: [PATCH 486/555] dmaengine: mediatek: Fix a possible deadlock error in mtk_cqdma_tx_status() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fix a potential deadlock bug. Observe that in the mtk-cqdma.c file, functions like mtk_cqdma_issue_pending() and mtk_cqdma_free_active_desc() properly acquire the pc lock before the vc lock when handling pc and vc fields. However, mtk_cqdma_tx_status() violates this order by first acquiring the vc lock before invoking mtk_cqdma_find_active_desc(), which subsequently takes the pc lock. This reversed locking sequence (vc → pc) contradicts the established pc → vc order and creates deadlock risks. Fix the issue by moving the vc lock acquisition code from mtk_cqdma_find_active_desc() to mtk_cqdma_tx_status(). Ensure the pc lock is acquired before the vc lock in the calling function to maintain correct locking hierarchy. Note that since mtk_cqdma_find_active_desc() is a static function with only one caller (mtk_cqdma_tx_status()), this modification safely eliminates the deadlock possibility without affecting other components. This possible bug is found by an experimental static analysis tool developed by our team. This tool analyzes the locking APIs to extract function pairs that can be concurrently executed, and then analyzes the instructions in the paired functions to identify possible concurrency bugs including deadlocks, data races and atomicity violations. Fixes: b1f01e48df5a ("dmaengine: mediatek: Add MediaTek Command-Queue DMA controller for MT6765 SoC") Cc: stable@vger.kernel.org Signed-off-by: Qiu-ji Chen Reviewed-by: AngeloGioacchino Del Regno Link: https://lore.kernel.org/r/20250508073634.3719-1-chenqiuji666@gmail.com Signed-off-by: Vinod Koul --- drivers/dma/mediatek/mtk-cqdma.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/drivers/dma/mediatek/mtk-cqdma.c b/drivers/dma/mediatek/mtk-cqdma.c index d5ddb4e30e71..e35271ac1eed 100644 --- a/drivers/dma/mediatek/mtk-cqdma.c +++ b/drivers/dma/mediatek/mtk-cqdma.c @@ -422,13 +422,10 @@ static struct virt_dma_desc *mtk_cqdma_find_active_desc(struct dma_chan *c, struct virt_dma_desc *vd; unsigned long flags; - spin_lock_irqsave(&cvc->pc->lock, flags); list_for_each_entry(vd, &cvc->pc->queue, node) if (vd->tx.cookie == cookie) { - spin_unlock_irqrestore(&cvc->pc->lock, flags); return vd; } - spin_unlock_irqrestore(&cvc->pc->lock, flags); list_for_each_entry(vd, &cvc->vc.desc_issued, node) if (vd->tx.cookie == cookie) @@ -452,9 +449,11 @@ static enum dma_status mtk_cqdma_tx_status(struct dma_chan *c, if (ret == DMA_COMPLETE || !txstate) return ret; + spin_lock_irqsave(&cvc->pc->lock, flags); spin_lock_irqsave(&cvc->vc.lock, flags); vd = mtk_cqdma_find_active_desc(c, cookie); spin_unlock_irqrestore(&cvc->vc.lock, flags); + spin_unlock_irqrestore(&cvc->pc->lock, flags); if (vd) { cvd = to_cqdma_vdesc(vd); From 5e27af0514e2249a9ccc9a762abd3b74e03a1f90 Mon Sep 17 00:00:00 2001 From: Stefan Wahren Date: Thu, 24 Apr 2025 13:48:29 +0200 Subject: [PATCH 487/555] dmaengine: fsl-edma: Fix return code for unhandled interrupts For fsl,imx93-edma4 two DMA channels share the same interrupt. So in case fsl_edma3_tx_handler is called for the "wrong" channel, the return code must be IRQ_NONE. This signalize that the interrupt wasn't handled. Fixes: 72f5801a4e2b ("dmaengine: fsl-edma: integrate v3 support") Signed-off-by: Stefan Wahren Reviewed-by: Joy Zou Link: https://lore.kernel.org/r/20250424114829.9055-1-wahrenst@gmx.net Signed-off-by: Vinod Koul --- drivers/dma/fsl-edma-main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/dma/fsl-edma-main.c b/drivers/dma/fsl-edma-main.c index 756d67325db5..66bfa28d984e 100644 --- a/drivers/dma/fsl-edma-main.c +++ b/drivers/dma/fsl-edma-main.c @@ -57,7 +57,7 @@ static irqreturn_t fsl_edma3_tx_handler(int irq, void *dev_id) intr = edma_readl_chreg(fsl_chan, ch_int); if (!intr) - return IRQ_HANDLED; + return IRQ_NONE; edma_writel_chreg(fsl_chan, 1, ch_int); From 2468b0e3d5659dfde77f081f266e1111a981efb8 Mon Sep 17 00:00:00 2001 From: Tianyang Zhang Date: Wed, 14 May 2025 22:17:43 +0800 Subject: [PATCH 488/555] LoongArch: Prevent cond_resched() occurring within kernel-fpu When CONFIG_PREEMPT_COUNT is not configured (i.e. CONFIG_PREEMPT_NONE/ CONFIG_PREEMPT_VOLUNTARY), preempt_disable() / preempt_enable() merely acts as a barrier(). However, in these cases cond_resched() can still trigger a context switch and modify the CSR.EUEN, resulting in do_fpu() exception being activated within the kernel-fpu critical sections, as demonstrated in the following path: dcn32_calculate_wm_and_dlg() DC_FP_START() dcn32_calculate_wm_and_dlg_fpu() dcn32_find_dummy_latency_index_for_fw_based_mclk_switch() dcn32_internal_validate_bw() dcn32_enable_phantom_stream() dc_create_stream_for_sink() kzalloc(GFP_KERNEL) __kmem_cache_alloc_node() __cond_resched() DC_FP_END() This patch is similar to commit d02198550423a0b (x86/fpu: Improve crypto performance by making kernel-mode FPU reliably usable in softirqs). It uses local_bh_disable() instead of preempt_disable() for non-RT kernels so it can avoid the cond_resched() issue, and also extend the kernel-fpu application scenarios to the softirq context. Cc: stable@vger.kernel.org Signed-off-by: Tianyang Zhang Signed-off-by: Huacai Chen --- arch/loongarch/kernel/kfpu.c | 22 ++++++++++++++++++++-- 1 file changed, 20 insertions(+), 2 deletions(-) diff --git a/arch/loongarch/kernel/kfpu.c b/arch/loongarch/kernel/kfpu.c index ec5b28e570c9..4c476904227f 100644 --- a/arch/loongarch/kernel/kfpu.c +++ b/arch/loongarch/kernel/kfpu.c @@ -18,11 +18,28 @@ static unsigned int euen_mask = CSR_EUEN_FPEN; static DEFINE_PER_CPU(bool, in_kernel_fpu); static DEFINE_PER_CPU(unsigned int, euen_current); +static inline void fpregs_lock(void) +{ + if (IS_ENABLED(CONFIG_PREEMPT_RT)) + preempt_disable(); + else + local_bh_disable(); +} + +static inline void fpregs_unlock(void) +{ + if (IS_ENABLED(CONFIG_PREEMPT_RT)) + preempt_enable(); + else + local_bh_enable(); +} + void kernel_fpu_begin(void) { unsigned int *euen_curr; - preempt_disable(); + if (!irqs_disabled()) + fpregs_lock(); WARN_ON(this_cpu_read(in_kernel_fpu)); @@ -73,7 +90,8 @@ void kernel_fpu_end(void) this_cpu_write(in_kernel_fpu, false); - preempt_enable(); + if (!irqs_disabled()) + fpregs_unlock(); } EXPORT_SYMBOL_GPL(kernel_fpu_end); From 90436d234230e9a950ccd87831108b688b27a234 Mon Sep 17 00:00:00 2001 From: Huacai Chen Date: Wed, 14 May 2025 22:17:43 +0800 Subject: [PATCH 489/555] LoongArch: Fix MAX_REG_OFFSET calculation Fix MAX_REG_OFFSET calculation, make it point to the last register in 'struct pt_regs' and not to the marker itself, which could allow regs_get_register() to return an invalid offset. Cc: stable@vger.kernel.org Fixes: 803b0fc5c3f2baa6e5 ("LoongArch: Add process management") Signed-off-by: Huacai Chen --- arch/loongarch/include/asm/ptrace.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/loongarch/include/asm/ptrace.h b/arch/loongarch/include/asm/ptrace.h index a5b63c84f854..e5d21e836d99 100644 --- a/arch/loongarch/include/asm/ptrace.h +++ b/arch/loongarch/include/asm/ptrace.h @@ -55,7 +55,7 @@ static inline void instruction_pointer_set(struct pt_regs *regs, unsigned long v /* Query offset/name of register from its name/offset */ extern int regs_query_register_offset(const char *name); -#define MAX_REG_OFFSET (offsetof(struct pt_regs, __last)) +#define MAX_REG_OFFSET (offsetof(struct pt_regs, __last) - sizeof(unsigned long)) /** * regs_get_register() - get register value from its offset From 3e245b7b74c3a2ead5fa4bad27cc275284c75189 Mon Sep 17 00:00:00 2001 From: Huacai Chen Date: Wed, 14 May 2025 22:17:52 +0800 Subject: [PATCH 490/555] LoongArch: Move __arch_cpu_idle() to .cpuidle.text section Now arch_cpu_idle() is annotated with __cpuidle which means it is in the .cpuidle.text section, but __arch_cpu_idle() isn't. Thus, fix the missing .cpuidle.text section assignment for __arch_cpu_idle() in order to correct backtracing with nmi_backtrace(). The principle is similar to the commit 97c8580e85cf81c ("MIPS: Annotate cpu_wait implementations with __cpuidle") Cc: stable@vger.kernel.org Signed-off-by: Huacai Chen --- arch/loongarch/kernel/genex.S | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/arch/loongarch/kernel/genex.S b/arch/loongarch/kernel/genex.S index 4f0912141781..733a7665e434 100644 --- a/arch/loongarch/kernel/genex.S +++ b/arch/loongarch/kernel/genex.S @@ -16,6 +16,7 @@ #include #include + .section .cpuidle.text, "ax" .align 5 SYM_FUNC_START(__arch_cpu_idle) /* start of idle interrupt region */ @@ -31,14 +32,16 @@ SYM_FUNC_START(__arch_cpu_idle) */ idle 0 /* end of idle interrupt region */ -1: jr ra +idle_exit: + jr ra SYM_FUNC_END(__arch_cpu_idle) + .previous SYM_CODE_START(handle_vint) UNWIND_HINT_UNDEFINED BACKUP_T0T1 SAVE_ALL - la_abs t1, 1b + la_abs t1, idle_exit LONG_L t0, sp, PT_ERA /* 3 instructions idle interrupt region */ ori t0, t0, 0b1100 From ceb9155d058a11242aa0572875c44e9713b1a2be Mon Sep 17 00:00:00 2001 From: Huacai Chen Date: Wed, 14 May 2025 22:17:52 +0800 Subject: [PATCH 491/555] LoongArch: Save and restore CSR.CNTC for hibernation Save and restore CSR.CNTC for hibernation which is similar to suspend. For host this is unnecessary because sched clock is ensured continuous, but for kvm guest sched clock isn't enough because rdtime.d should also be continuous. Host::rdtime.d = Host::CSR.CNTC + counter Guest::rdtime.d = Host::CSR.CNTC + Host::CSR.GCNTC + Guest::CSR.CNTC + counter so, Guest::rdtime.d = Host::rdtime.d + Host::CSR.GCNTC + Guest::CSR.CNTC To ensure Guest::rdtime.d continuous, Host::rdtime.d should be at first continuous, while Host::CSR.GCNTC / Guest::CSR.CNTC is maintained by KVM. Cc: stable@vger.kernel.org Signed-off-by: Xianglai Li Signed-off-by: Huacai Chen --- arch/loongarch/kernel/time.c | 2 +- arch/loongarch/power/hibernate.c | 3 +++ 2 files changed, 4 insertions(+), 1 deletion(-) diff --git a/arch/loongarch/kernel/time.c b/arch/loongarch/kernel/time.c index e2d3bfeb6366..bc75a3a69fc8 100644 --- a/arch/loongarch/kernel/time.c +++ b/arch/loongarch/kernel/time.c @@ -111,7 +111,7 @@ static unsigned long __init get_loops_per_jiffy(void) return lpj; } -static long init_offset __nosavedata; +static long init_offset; void save_counter(void) { diff --git a/arch/loongarch/power/hibernate.c b/arch/loongarch/power/hibernate.c index 1e0590542f98..e7b7346592cb 100644 --- a/arch/loongarch/power/hibernate.c +++ b/arch/loongarch/power/hibernate.c @@ -2,6 +2,7 @@ #include #include #include +#include #include #include @@ -14,6 +15,7 @@ struct pt_regs saved_regs; void save_processor_state(void) { + save_counter(); saved_crmd = csr_read32(LOONGARCH_CSR_CRMD); saved_prmd = csr_read32(LOONGARCH_CSR_PRMD); saved_euen = csr_read32(LOONGARCH_CSR_EUEN); @@ -26,6 +28,7 @@ void save_processor_state(void) void restore_processor_state(void) { + sync_counter(); csr_write32(saved_crmd, LOONGARCH_CSR_CRMD); csr_write32(saved_prmd, LOONGARCH_CSR_PRMD); csr_write32(saved_euen, LOONGARCH_CSR_EUEN); From 0b326b2371f94e798137cc1a3c5c2eef2bc69061 Mon Sep 17 00:00:00 2001 From: Tiezhu Yang Date: Wed, 14 May 2025 22:18:10 +0800 Subject: [PATCH 492/555] LoongArch: uprobes: Remove user_{en,dis}able_single_step() When executing the "perf probe" and "perf stat" test cases about some cryptographic algorithm, the output shows that "Trace/breakpoint trap". This is because it uses the software singlestep breakpoint for uprobes on LoongArch, and no need to use the hardware singlestep. So just remove the related function call to user_{en,dis}able_single_step() for uprobes on LoongArch. How to reproduce: Please make sure CONFIG_UPROBE_EVENTS is set and openssl supports sm2 algorithm, then execute the following command. cd tools/perf && make ./perf probe -x /usr/lib64/libcrypto.so BN_mod_mul_montgomery ./perf stat -e probe_libcrypto:BN_mod_mul_montgomery openssl speed sm2 Cc: stable@vger.kernel.org Fixes: 19bc6cb64092 ("LoongArch: Add uprobes support") Signed-off-by: Tiezhu Yang Signed-off-by: Huacai Chen --- arch/loongarch/kernel/uprobes.c | 4 ---- 1 file changed, 4 deletions(-) diff --git a/arch/loongarch/kernel/uprobes.c b/arch/loongarch/kernel/uprobes.c index 87abc7137b73..0ab9d8d631c4 100644 --- a/arch/loongarch/kernel/uprobes.c +++ b/arch/loongarch/kernel/uprobes.c @@ -42,7 +42,6 @@ int arch_uprobe_pre_xol(struct arch_uprobe *auprobe, struct pt_regs *regs) utask->autask.saved_trap_nr = current->thread.trap_nr; current->thread.trap_nr = UPROBE_TRAP_NR; instruction_pointer_set(regs, utask->xol_vaddr); - user_enable_single_step(current); return 0; } @@ -59,8 +58,6 @@ int arch_uprobe_post_xol(struct arch_uprobe *auprobe, struct pt_regs *regs) else instruction_pointer_set(regs, utask->vaddr + LOONGARCH_INSN_SIZE); - user_disable_single_step(current); - return 0; } @@ -70,7 +67,6 @@ void arch_uprobe_abort_xol(struct arch_uprobe *auprobe, struct pt_regs *regs) current->thread.trap_nr = utask->autask.saved_trap_nr; instruction_pointer_set(regs, utask->vaddr); - user_disable_single_step(current); } bool arch_uprobe_xol_was_trapped(struct task_struct *t) From 12614f794274f63fbdfe76771b2b332077d63848 Mon Sep 17 00:00:00 2001 From: Tiezhu Yang Date: Wed, 14 May 2025 22:18:10 +0800 Subject: [PATCH 493/555] LoongArch: uprobes: Remove redundant code about resume_era arch_uprobe_skip_sstep() returns true if instruction was emulated, that is to say, there is no need to single step for the emulated instructions. regs->csr_era will point to the destination address directly after the exception, so the resume_era related code is redundant, just remove them. Cc: stable@vger.kernel.org Fixes: 19bc6cb64092 ("LoongArch: Add uprobes support") Signed-off-by: Tiezhu Yang Signed-off-by: Huacai Chen --- arch/loongarch/include/asm/uprobes.h | 1 - arch/loongarch/kernel/uprobes.c | 7 +------ 2 files changed, 1 insertion(+), 7 deletions(-) diff --git a/arch/loongarch/include/asm/uprobes.h b/arch/loongarch/include/asm/uprobes.h index 99a0d198927f..025fc3f0a102 100644 --- a/arch/loongarch/include/asm/uprobes.h +++ b/arch/loongarch/include/asm/uprobes.h @@ -15,7 +15,6 @@ typedef u32 uprobe_opcode_t; #define UPROBE_XOLBP_INSN __emit_break(BRK_UPROBE_XOLBP) struct arch_uprobe { - unsigned long resume_era; u32 insn[2]; u32 ixol[2]; bool simulate; diff --git a/arch/loongarch/kernel/uprobes.c b/arch/loongarch/kernel/uprobes.c index 0ab9d8d631c4..6022eb0f71db 100644 --- a/arch/loongarch/kernel/uprobes.c +++ b/arch/loongarch/kernel/uprobes.c @@ -52,11 +52,7 @@ int arch_uprobe_post_xol(struct arch_uprobe *auprobe, struct pt_regs *regs) WARN_ON_ONCE(current->thread.trap_nr != UPROBE_TRAP_NR); current->thread.trap_nr = utask->autask.saved_trap_nr; - - if (auprobe->simulate) - instruction_pointer_set(regs, auprobe->resume_era); - else - instruction_pointer_set(regs, utask->vaddr + LOONGARCH_INSN_SIZE); + instruction_pointer_set(regs, utask->vaddr + LOONGARCH_INSN_SIZE); return 0; } @@ -86,7 +82,6 @@ bool arch_uprobe_skip_sstep(struct arch_uprobe *auprobe, struct pt_regs *regs) insn.word = auprobe->insn[0]; arch_simulate_insn(insn, regs); - auprobe->resume_era = regs->csr_era; return true; } From fb0ea6e4878a45b1ac81972027907fc424a792e6 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Tue, 13 May 2025 18:28:15 +0100 Subject: [PATCH 494/555] irqchip: Drop MSI_CHIP_FLAG_SET_ACK from unsuspecting MSI drivers Commit 1c000dcaad2be ("irqchip/irq-msi-lib: Optionally set default irq_eoi()/irq_ack()") added blanket MSI_CHIP_FLAG_SET_ACK flags, irrespective of whether the underlying irqchip required it or not. Drop it from a number of drivers that do not require it. Fixes: 1c000dcaad2be ("irqchip/irq-msi-lib: Optionally set default irq_eoi()/irq_ack()") Signed-off-by: Marc Zyngier Signed-off-by: Thomas Gleixner Link: https://lore.kernel.org/all/20250513172819.2216709-6-maz@kernel.org --- drivers/irqchip/irq-gic-v2m.c | 2 +- drivers/irqchip/irq-gic-v3-its-msi-parent.c | 2 +- drivers/irqchip/irq-gic-v3-mbi.c | 2 +- drivers/irqchip/irq-mvebu-gicp.c | 2 +- drivers/irqchip/irq-mvebu-odmi.c | 2 +- 5 files changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/irqchip/irq-gic-v2m.c b/drivers/irqchip/irq-gic-v2m.c index dc98c39d2b20..cc6a6c1585d2 100644 --- a/drivers/irqchip/irq-gic-v2m.c +++ b/drivers/irqchip/irq-gic-v2m.c @@ -252,7 +252,7 @@ static void __init gicv2m_teardown(void) static struct msi_parent_ops gicv2m_msi_parent_ops = { .supported_flags = GICV2M_MSI_FLAGS_SUPPORTED, .required_flags = GICV2M_MSI_FLAGS_REQUIRED, - .chip_flags = MSI_CHIP_FLAG_SET_EOI | MSI_CHIP_FLAG_SET_ACK, + .chip_flags = MSI_CHIP_FLAG_SET_EOI, .bus_select_token = DOMAIN_BUS_NEXUS, .bus_select_mask = MATCH_PCI_MSI | MATCH_PLATFORM_MSI, .prefix = "GICv2m-", diff --git a/drivers/irqchip/irq-gic-v3-its-msi-parent.c b/drivers/irqchip/irq-gic-v3-its-msi-parent.c index bdb04c808148..c5a7eb1c0419 100644 --- a/drivers/irqchip/irq-gic-v3-its-msi-parent.c +++ b/drivers/irqchip/irq-gic-v3-its-msi-parent.c @@ -203,7 +203,7 @@ static bool its_init_dev_msi_info(struct device *dev, struct irq_domain *domain, const struct msi_parent_ops gic_v3_its_msi_parent_ops = { .supported_flags = ITS_MSI_FLAGS_SUPPORTED, .required_flags = ITS_MSI_FLAGS_REQUIRED, - .chip_flags = MSI_CHIP_FLAG_SET_EOI | MSI_CHIP_FLAG_SET_ACK, + .chip_flags = MSI_CHIP_FLAG_SET_EOI, .bus_select_token = DOMAIN_BUS_NEXUS, .bus_select_mask = MATCH_PCI_MSI | MATCH_PLATFORM_MSI, .prefix = "ITS-", diff --git a/drivers/irqchip/irq-gic-v3-mbi.c b/drivers/irqchip/irq-gic-v3-mbi.c index 34e9ca77a8c3..647b18e24e0c 100644 --- a/drivers/irqchip/irq-gic-v3-mbi.c +++ b/drivers/irqchip/irq-gic-v3-mbi.c @@ -197,7 +197,7 @@ static bool mbi_init_dev_msi_info(struct device *dev, struct irq_domain *domain, static const struct msi_parent_ops gic_v3_mbi_msi_parent_ops = { .supported_flags = MBI_MSI_FLAGS_SUPPORTED, .required_flags = MBI_MSI_FLAGS_REQUIRED, - .chip_flags = MSI_CHIP_FLAG_SET_EOI | MSI_CHIP_FLAG_SET_ACK, + .chip_flags = MSI_CHIP_FLAG_SET_EOI, .bus_select_token = DOMAIN_BUS_NEXUS, .bus_select_mask = MATCH_PCI_MSI | MATCH_PLATFORM_MSI, .prefix = "MBI-", diff --git a/drivers/irqchip/irq-mvebu-gicp.c b/drivers/irqchip/irq-mvebu-gicp.c index d67f93f6d750..60b976286636 100644 --- a/drivers/irqchip/irq-mvebu-gicp.c +++ b/drivers/irqchip/irq-mvebu-gicp.c @@ -161,7 +161,7 @@ static const struct irq_domain_ops gicp_domain_ops = { static const struct msi_parent_ops gicp_msi_parent_ops = { .supported_flags = GICP_MSI_FLAGS_SUPPORTED, .required_flags = GICP_MSI_FLAGS_REQUIRED, - .chip_flags = MSI_CHIP_FLAG_SET_EOI | MSI_CHIP_FLAG_SET_ACK, + .chip_flags = MSI_CHIP_FLAG_SET_EOI, .bus_select_token = DOMAIN_BUS_GENERIC_MSI, .bus_select_mask = MATCH_PLATFORM_MSI, .prefix = "GICP-", diff --git a/drivers/irqchip/irq-mvebu-odmi.c b/drivers/irqchip/irq-mvebu-odmi.c index 28f7e81df94f..54f6f0811573 100644 --- a/drivers/irqchip/irq-mvebu-odmi.c +++ b/drivers/irqchip/irq-mvebu-odmi.c @@ -157,7 +157,7 @@ static const struct irq_domain_ops odmi_domain_ops = { static const struct msi_parent_ops odmi_msi_parent_ops = { .supported_flags = ODMI_MSI_FLAGS_SUPPORTED, .required_flags = ODMI_MSI_FLAGS_REQUIRED, - .chip_flags = MSI_CHIP_FLAG_SET_EOI | MSI_CHIP_FLAG_SET_ACK, + .chip_flags = MSI_CHIP_FLAG_SET_EOI, .bus_select_token = DOMAIN_BUS_GENERIC_MSI, .bus_select_mask = MATCH_PLATFORM_MSI, .prefix = "ODMI-", From e765bf89f42b5c82132a556b630affeb82b2a21f Mon Sep 17 00:00:00 2001 From: Ilya Guterman Date: Sat, 10 May 2025 19:21:30 +0900 Subject: [PATCH 495/555] nvme-pci: add NVME_QUIRK_NO_DEEPEST_PS quirk for SOLIDIGM P44 Pro This commit adds the NVME_QUIRK_NO_DEEPEST_PS quirk for device [126f:2262], which belongs to device SOLIDIGM P44 Pro SSDPFKKW020X7 The device frequently have trouble exiting the deepest power state (5), resulting in the entire disk being unresponsive. Verified by setting nvme_core.default_ps_max_latency_us=10000 and observing the expected behavior. Signed-off-by: Ilya Guterman Signed-off-by: Christoph Hellwig --- drivers/nvme/host/pci.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c index a9390ac7211e..f1dd804151b1 100644 --- a/drivers/nvme/host/pci.c +++ b/drivers/nvme/host/pci.c @@ -3739,6 +3739,8 @@ static const struct pci_device_id nvme_id_table[] = { .driver_data = NVME_QUIRK_NO_DEEPEST_PS, }, { PCI_DEVICE(0x1e49, 0x0041), /* ZHITAI TiPro7000 NVMe SSD */ .driver_data = NVME_QUIRK_NO_DEEPEST_PS, }, + { PCI_DEVICE(0x025e, 0xf1ac), /* SOLIDIGM P44 pro SSDPFKKW020X7 */ + .driver_data = NVME_QUIRK_NO_DEEPEST_PS, }, { PCI_DEVICE(0xc0a9, 0x540a), /* Crucial P2 */ .driver_data = NVME_QUIRK_BOGUS_NID, }, { PCI_DEVICE(0x1d97, 0x2263), /* Lexar NM610 */ From 1cfe51ef07ca3286581d612debfb0430eeccbb65 Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Tue, 13 May 2025 19:56:41 +0200 Subject: [PATCH 496/555] i2c: designware: Fix an error handling path in i2c_dw_pci_probe() If navi_amd_register_client() fails, the previous i2c_dw_probe() call should be undone by a corresponding i2c_del_adapter() call, as already done in the remove function. Fixes: 17631e8ca2d3 ("i2c: designware: Add driver support for AMD NAVI GPU") Signed-off-by: Christophe JAILLET Cc: # v5.13+ Acked-by: Jarkko Nikula Signed-off-by: Andi Shyti Link: https://lore.kernel.org/r/fcd9651835a32979df8802b2db9504c523a8ebbb.1747158983.git.christophe.jaillet@wanadoo.fr --- drivers/i2c/busses/i2c-designware-pcidrv.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/i2c/busses/i2c-designware-pcidrv.c b/drivers/i2c/busses/i2c-designware-pcidrv.c index 8e0267c7cc29..f21f9877c040 100644 --- a/drivers/i2c/busses/i2c-designware-pcidrv.c +++ b/drivers/i2c/busses/i2c-designware-pcidrv.c @@ -278,9 +278,11 @@ static int i2c_dw_pci_probe(struct pci_dev *pdev, if ((dev->flags & MODEL_MASK) == MODEL_AMD_NAVI_GPU) { dev->slave = i2c_new_ccgx_ucsi(&dev->adapter, dev->irq, &dgpu_node); - if (IS_ERR(dev->slave)) + if (IS_ERR(dev->slave)) { + i2c_del_adapter(&dev->adapter); return dev_err_probe(device, PTR_ERR(dev->slave), "register UCSI failed\n"); + } } pm_runtime_set_autosuspend_delay(device, 1000); From fe14c0f096f58d2569e587e9f4b05d772272bbb4 Mon Sep 17 00:00:00 2001 From: Melissa Wen Date: Tue, 22 Apr 2025 11:58:11 -0300 Subject: [PATCH 497/555] Revert "drm/amd/display: Hardware cursor changes color when switched to software cursor" MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This reverts commit 272e6aab14bbf98d7a06b2b1cd6308a02d4a10a1. Applying degamma curve to the cursor by default breaks Linux userspace expectation. On Linux, AMD display manager enables cursor degamma ROM just for implict sRGB on HW versions where degamma is split into two blocks: degamma ROM for pre-defined TFs and `gamma correction` for user/custom curves, and degamma ROM settings doesn't apply to cursor plane. Link: https://gitlab.freedesktop.org/drm/amd/-/issues/1513 Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2803 Reported-by: Michel Dänzer Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4144 Signed-off-by: Melissa Wen Reviewed-by: Alex Hung Signed-off-by: Alex Deucher (cherry picked from commit f6a305d4748801a6c799ae9375b2ecff3aed094b) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp_cm.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp_cm.c b/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp_cm.c index 1236e0f9a256..712aff7e17f7 100644 --- a/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp_cm.c +++ b/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp_cm.c @@ -120,10 +120,11 @@ void dpp401_set_cursor_attributes( enum dc_cursor_color_format color_format = cursor_attributes->color_format; int cur_rom_en = 0; - // DCN4 should always do Cursor degamma for Cursor Color modes if (color_format == CURSOR_MODE_COLOR_PRE_MULTIPLIED_ALPHA || color_format == CURSOR_MODE_COLOR_UN_PRE_MULTIPLIED_ALPHA) { - cur_rom_en = 1; + if (cursor_attributes->attribute_flags.bits.ENABLE_CURSOR_DEGAMMA) { + cur_rom_en = 1; + } } REG_UPDATE_3(CURSOR0_CONTROL, From ee7360fc27d6045510f8fe459b5649b2af27811a Mon Sep 17 00:00:00 2001 From: "David (Ming Qiang) Wu" Date: Mon, 12 May 2025 15:14:43 -0400 Subject: [PATCH 498/555] drm/amdgpu: read back register after written for VCN v4.0.5 On VCN v4.0.5 there is a race condition where the WPTR is not updated after starting from idle when doorbell is used. Adding register read-back after written at function end is to ensure all register writes are done before they can be used. Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/12528 Signed-off-by: David (Ming Qiang) Wu Reviewed-by: Mario Limonciello Tested-by: Mario Limonciello Reviewed-by: Alex Deucher Reviewed-by: Ruijing Dong Signed-off-by: Alex Deucher (cherry picked from commit 07c9db090b86e5211188e1b351303fbc673378cf) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c index a1171e6152ed..f11df9c2ec13 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c @@ -1023,6 +1023,10 @@ static int vcn_v4_0_5_start_dpg_mode(struct amdgpu_vcn_inst *vinst, ring->doorbell_index << VCN_RB1_DB_CTRL__OFFSET__SHIFT | VCN_RB1_DB_CTRL__EN_MASK); + /* Keeping one read-back to ensure all register writes are done, otherwise + * it may introduce race conditions */ + RREG32_SOC15(VCN, inst_idx, regVCN_RB1_DB_CTRL); + return 0; } @@ -1205,6 +1209,10 @@ static int vcn_v4_0_5_start(struct amdgpu_vcn_inst *vinst) WREG32_SOC15(VCN, i, regVCN_RB_ENABLE, tmp); fw_shared->sq.queue_mode &= ~(FW_QUEUE_RING_RESET | FW_QUEUE_DPG_HOLD_OFF); + /* Keeping one read-back to ensure all register writes are done, otherwise + * it may introduce race conditions */ + RREG32_SOC15(VCN, i, regVCN_RB_ENABLE); + return 0; } From b1f704107cf27906a9cea542b626b96019104663 Mon Sep 17 00:00:00 2001 From: Aradhya Bhatia Date: Mon, 12 May 2025 06:50:04 +0000 Subject: [PATCH 499/555] drm/xe/xe2hpg: Add Wa_22021007897 Add Wa_22021007897 for the Xe2_HPG (graphics version: 20.01) IP. It is a permanent workaround, and applicable on all the steppings. Reviewed-by: Gustavo Sousa Reviewed-by: Tejas Upadhyay Signed-off-by: Aradhya Bhatia Link: https://lore.kernel.org/r/20250512065004.2576-1-aradhya.bhatia@intel.com Signed-off-by: Matt Roper (cherry picked from commit e5c13e2c505b73a8667ef9a0fd5cbd4227e483e6) Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/regs/xe_gt_regs.h | 1 + drivers/gpu/drm/xe/xe_wa.c | 4 ++++ 2 files changed, 5 insertions(+) diff --git a/drivers/gpu/drm/xe/regs/xe_gt_regs.h b/drivers/gpu/drm/xe/regs/xe_gt_regs.h index da1f198ac107..181913967ac9 100644 --- a/drivers/gpu/drm/xe/regs/xe_gt_regs.h +++ b/drivers/gpu/drm/xe/regs/xe_gt_regs.h @@ -157,6 +157,7 @@ #define XEHPG_SC_INSTDONE_EXTRA2 XE_REG_MCR(0x7108) #define COMMON_SLICE_CHICKEN4 XE_REG(0x7300, XE_REG_OPTION_MASKED) +#define SBE_PUSH_CONSTANT_BEHIND_FIX_ENABLE REG_BIT(12) #define DISABLE_TDC_LOAD_BALANCING_CALC REG_BIT(6) #define COMMON_SLICE_CHICKEN3 XE_REG(0x7304, XE_REG_OPTION_MASKED) diff --git a/drivers/gpu/drm/xe/xe_wa.c b/drivers/gpu/drm/xe/xe_wa.c index 24f644c0a673..2f833f0d575f 100644 --- a/drivers/gpu/drm/xe/xe_wa.c +++ b/drivers/gpu/drm/xe/xe_wa.c @@ -815,6 +815,10 @@ static const struct xe_rtp_entry_sr lrc_was[] = { XE_RTP_RULES(GRAPHICS_VERSION(2001), ENGINE_CLASS(RENDER)), XE_RTP_ACTIONS(SET(CHICKEN_RASTER_1, DIS_CLIP_NEGATIVE_BOUNDING_BOX)) }, + { XE_RTP_NAME("22021007897"), + XE_RTP_RULES(GRAPHICS_VERSION(2001), ENGINE_CLASS(RENDER)), + XE_RTP_ACTIONS(SET(COMMON_SLICE_CHICKEN4, SBE_PUSH_CONSTANT_BEHIND_FIX_ENABLE)) + }, /* Xe3_LPG */ { XE_RTP_NAME("14021490052"), From 7bd68ce21d39150d80806233965326359f517b78 Mon Sep 17 00:00:00 2001 From: Himal Prasad Ghimiray Date: Mon, 12 May 2025 06:54:55 -0700 Subject: [PATCH 500/555] drm/gpusvm: Introduce devmem_only flag for allocation This commit adds a new flag, devmem_only, to the drm_gpusvm structure. The purpose of this flag is to ensure that the get_pages function allocates memory exclusively from the device's memory. If the allocation from device memory fails, the function will return an -EFAULT error. Required for shared CPU and GPU atomics on certain devices. v3: - s/vram_only/devmem_only/ Fixes: 99624bdff867 ("drm/gpusvm: Add support for GPU Shared Virtual Memory") Cc: stable@vger.kernel.org Signed-off-by: Matthew Brost Signed-off-by: Himal Prasad Ghimiray Reviewed-by: Matthew Brost Link: https://lore.kernel.org/r/20250512135500.1405019-2-matthew.brost@intel.com (cherry picked from commit 8a9b978ebd47df9e0694c34748c2d6fa0c31eb4d) Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/drm_gpusvm.c | 5 +++++ include/drm/drm_gpusvm.h | 2 ++ 2 files changed, 7 insertions(+) diff --git a/drivers/gpu/drm/drm_gpusvm.c b/drivers/gpu/drm/drm_gpusvm.c index de424e670995..a58d03e6cac2 100644 --- a/drivers/gpu/drm/drm_gpusvm.c +++ b/drivers/gpu/drm/drm_gpusvm.c @@ -1454,6 +1454,11 @@ map_pages: goto err_unmap; } + if (ctx->devmem_only) { + err = -EFAULT; + goto err_unmap; + } + addr = dma_map_page(gpusvm->drm->dev, page, 0, PAGE_SIZE << order, diff --git a/include/drm/drm_gpusvm.h b/include/drm/drm_gpusvm.h index df120b4d1f83..9fd25fc880a4 100644 --- a/include/drm/drm_gpusvm.h +++ b/include/drm/drm_gpusvm.h @@ -286,6 +286,7 @@ struct drm_gpusvm { * @in_notifier: entering from a MMU notifier * @read_only: operating on read-only memory * @devmem_possible: possible to use device memory + * @devmem_only: use only device memory * * Context that is DRM GPUSVM is operating in (i.e. user arguments). */ @@ -294,6 +295,7 @@ struct drm_gpusvm_ctx { unsigned int in_notifier :1; unsigned int read_only :1; unsigned int devmem_possible :1; + unsigned int devmem_only :1; }; int drm_gpusvm_init(struct drm_gpusvm *gpusvm, From 794f5493f518916380578f14d999de92b930b609 Mon Sep 17 00:00:00 2001 From: Matthew Brost Date: Mon, 12 May 2025 06:54:56 -0700 Subject: [PATCH 501/555] drm/xe: Strict migration policy for atomic SVM faults MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Mixing GPU and CPU atomics does not work unless a strict migration policy of GPU atomics must be device memory. Enforce a policy of must be in VRAM with a retry loop of 3 attempts, if retry loop fails abort fault. Removing always_migrate_to_vram modparam as we now have real migration policy. v2: - Only retry migration on atomics - Drop alway migrate modparam v3: - Only set vram_only on DGFX (Himal) - Bail on get_pages failure if vram_only and retry count exceeded (Himal) - s/vram_only/devmem_only - Update xe_svm_range_is_valid to accept devmem_only argument v4: - Fix logic bug get_pages failure v5: - Fix commit message (Himal) - Mention removing always_migrate_to_vram in commit message (Lucas) - Fix xe_svm_range_is_valid to check for devmem pages - Bail on devmem_only && !migrate_devmem (Thomas) v6: - Add READ_ONCE barriers for opportunistic checks (Thomas) - Pair READ_ONCE with WRITE_ONCE (Thomas) v7: - Adjust comments (Thomas) Fixes: 2f118c949160 ("drm/xe: Add SVM VRAM migration") Cc: stable@vger.kernel.org Signed-off-by: Himal Prasad Ghimiray Signed-off-by: Matthew Brost Acked-by: Himal Prasad Ghimiray Reviewed-by: Thomas Hellström Link: https://lore.kernel.org/r/20250512135500.1405019-3-matthew.brost@intel.com (cherry picked from commit a9ac0fa455b050d03e3032501368048fb284d318) Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/drm_gpusvm.c | 23 +++++-- drivers/gpu/drm/xe/xe_module.c | 3 - drivers/gpu/drm/xe/xe_module.h | 1 - drivers/gpu/drm/xe/xe_pt.c | 14 ++++- drivers/gpu/drm/xe/xe_svm.c | 111 +++++++++++++++++++++++++-------- drivers/gpu/drm/xe/xe_svm.h | 5 -- include/drm/drm_gpusvm.h | 40 +++++++----- 7 files changed, 140 insertions(+), 57 deletions(-) diff --git a/drivers/gpu/drm/drm_gpusvm.c b/drivers/gpu/drm/drm_gpusvm.c index a58d03e6cac2..41f6616bcf76 100644 --- a/drivers/gpu/drm/drm_gpusvm.c +++ b/drivers/gpu/drm/drm_gpusvm.c @@ -1118,6 +1118,10 @@ static void __drm_gpusvm_range_unmap_pages(struct drm_gpusvm *gpusvm, lockdep_assert_held(&gpusvm->notifier_lock); if (range->flags.has_dma_mapping) { + struct drm_gpusvm_range_flags flags = { + .__flags = range->flags.__flags, + }; + for (i = 0, j = 0; i < npages; j++) { struct drm_pagemap_device_addr *addr = &range->dma_addr[j]; @@ -1131,8 +1135,12 @@ static void __drm_gpusvm_range_unmap_pages(struct drm_gpusvm *gpusvm, dev, *addr); i += 1 << addr->order; } - range->flags.has_devmem_pages = false; - range->flags.has_dma_mapping = false; + + /* WRITE_ONCE pairs with READ_ONCE for opportunistic checks */ + flags.has_devmem_pages = false; + flags.has_dma_mapping = false; + WRITE_ONCE(range->flags.__flags, flags.__flags); + range->dpagemap = NULL; } } @@ -1334,6 +1342,7 @@ int drm_gpusvm_range_get_pages(struct drm_gpusvm *gpusvm, int err = 0; struct dev_pagemap *pagemap; struct drm_pagemap *dpagemap; + struct drm_gpusvm_range_flags flags; retry: hmm_range.notifier_seq = mmu_interval_read_begin(notifier); @@ -1378,7 +1387,8 @@ map_pages: */ drm_gpusvm_notifier_lock(gpusvm); - if (range->flags.unmapped) { + flags.__flags = range->flags.__flags; + if (flags.unmapped) { drm_gpusvm_notifier_unlock(gpusvm); err = -EFAULT; goto err_free; @@ -1474,14 +1484,17 @@ map_pages: } i += 1 << order; num_dma_mapped = i; - range->flags.has_dma_mapping = true; + flags.has_dma_mapping = true; } if (zdd) { - range->flags.has_devmem_pages = true; + flags.has_devmem_pages = true; range->dpagemap = dpagemap; } + /* WRITE_ONCE pairs with READ_ONCE for opportunistic checks */ + WRITE_ONCE(range->flags.__flags, flags.__flags); + drm_gpusvm_notifier_unlock(gpusvm); kvfree(pfns); set_seqno: diff --git a/drivers/gpu/drm/xe/xe_module.c b/drivers/gpu/drm/xe/xe_module.c index 9f4632e39a1a..e861c694f336 100644 --- a/drivers/gpu/drm/xe/xe_module.c +++ b/drivers/gpu/drm/xe/xe_module.c @@ -29,9 +29,6 @@ struct xe_modparam xe_modparam = { module_param_named(svm_notifier_size, xe_modparam.svm_notifier_size, uint, 0600); MODULE_PARM_DESC(svm_notifier_size, "Set the svm notifier size(in MiB), must be power of 2"); -module_param_named(always_migrate_to_vram, xe_modparam.always_migrate_to_vram, bool, 0444); -MODULE_PARM_DESC(always_migrate_to_vram, "Always migrate to VRAM on GPU fault"); - module_param_named_unsafe(force_execlist, xe_modparam.force_execlist, bool, 0444); MODULE_PARM_DESC(force_execlist, "Force Execlist submission"); diff --git a/drivers/gpu/drm/xe/xe_module.h b/drivers/gpu/drm/xe/xe_module.h index 84339e509c80..5a3bfea8b7b4 100644 --- a/drivers/gpu/drm/xe/xe_module.h +++ b/drivers/gpu/drm/xe/xe_module.h @@ -12,7 +12,6 @@ struct xe_modparam { bool force_execlist; bool probe_display; - bool always_migrate_to_vram; u32 force_vram_bar_size; int guc_log_level; char *guc_firmware_path; diff --git a/drivers/gpu/drm/xe/xe_pt.c b/drivers/gpu/drm/xe/xe_pt.c index ffaf0d02dc7d..856038553b81 100644 --- a/drivers/gpu/drm/xe/xe_pt.c +++ b/drivers/gpu/drm/xe/xe_pt.c @@ -2232,11 +2232,19 @@ static void op_commit(struct xe_vm *vm, } case DRM_GPUVA_OP_DRIVER: { + /* WRITE_ONCE pairs with READ_ONCE in xe_svm.c */ + if (op->subop == XE_VMA_SUBOP_MAP_RANGE) { - op->map_range.range->tile_present |= BIT(tile->id); - op->map_range.range->tile_invalidated &= ~BIT(tile->id); + WRITE_ONCE(op->map_range.range->tile_present, + op->map_range.range->tile_present | + BIT(tile->id)); + WRITE_ONCE(op->map_range.range->tile_invalidated, + op->map_range.range->tile_invalidated & + ~BIT(tile->id)); } else if (op->subop == XE_VMA_SUBOP_UNMAP_RANGE) { - op->unmap_range.range->tile_present &= ~BIT(tile->id); + WRITE_ONCE(op->unmap_range.range->tile_present, + op->unmap_range.range->tile_present & + ~BIT(tile->id)); } break; } diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c index 24c578e1170e..e65d3b820ee0 100644 --- a/drivers/gpu/drm/xe/xe_svm.c +++ b/drivers/gpu/drm/xe/xe_svm.c @@ -15,8 +15,17 @@ static bool xe_svm_range_in_vram(struct xe_svm_range *range) { - /* Not reliable without notifier lock */ - return range->base.flags.has_devmem_pages; + /* + * Advisory only check whether the range is currently backed by VRAM + * memory. + */ + + struct drm_gpusvm_range_flags flags = { + /* Pairs with WRITE_ONCE in drm_gpusvm.c */ + .__flags = READ_ONCE(range->base.flags.__flags), + }; + + return flags.has_devmem_pages; } static bool xe_svm_range_has_vram_binding(struct xe_svm_range *range) @@ -645,9 +654,16 @@ void xe_svm_fini(struct xe_vm *vm) } static bool xe_svm_range_is_valid(struct xe_svm_range *range, - struct xe_tile *tile) + struct xe_tile *tile, + bool devmem_only) { - return (range->tile_present & ~range->tile_invalidated) & BIT(tile->id); + /* + * Advisory only check whether the range currently has a valid mapping, + * READ_ONCE pairs with WRITE_ONCE in xe_pt.c + */ + return ((READ_ONCE(range->tile_present) & + ~READ_ONCE(range->tile_invalidated)) & BIT(tile->id)) && + (!devmem_only || xe_svm_range_in_vram(range)); } static struct xe_vram_region *tile_to_vr(struct xe_tile *tile) @@ -712,6 +728,36 @@ unlock: return err; } +static bool supports_4K_migration(struct xe_device *xe) +{ + if (xe->info.vram_flags & XE_VRAM_FLAGS_NEED64K) + return false; + + return true; +} + +static bool xe_svm_range_needs_migrate_to_vram(struct xe_svm_range *range, + struct xe_vma *vma) +{ + struct xe_vm *vm = range_to_vm(&range->base); + u64 range_size = xe_svm_range_size(range); + + if (!range->base.flags.migrate_devmem) + return false; + + if (xe_svm_range_in_vram(range)) { + drm_dbg(&vm->xe->drm, "Range is already in VRAM\n"); + return false; + } + + if (range_size <= SZ_64K && !supports_4K_migration(vm->xe)) { + drm_dbg(&vm->xe->drm, "Platform doesn't support SZ_4K range migration\n"); + return false; + } + + return true; +} + /** * xe_svm_handle_pagefault() - SVM handle page fault * @vm: The VM. @@ -735,11 +781,14 @@ int xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma, IS_ENABLED(CONFIG_DRM_XE_DEVMEM_MIRROR), .check_pages_threshold = IS_DGFX(vm->xe) && IS_ENABLED(CONFIG_DRM_XE_DEVMEM_MIRROR) ? SZ_64K : 0, + .devmem_only = atomic && IS_DGFX(vm->xe) && + IS_ENABLED(CONFIG_DRM_XE_DEVMEM_MIRROR), }; struct xe_svm_range *range; struct drm_gpusvm_range *r; struct drm_exec exec; struct dma_fence *fence; + int migrate_try_count = ctx.devmem_only ? 3 : 1; ktime_t end = 0; int err; @@ -758,24 +807,30 @@ retry: if (IS_ERR(r)) return PTR_ERR(r); + if (ctx.devmem_only && !r->flags.migrate_devmem) + return -EACCES; + range = to_xe_range(r); - if (xe_svm_range_is_valid(range, tile)) + if (xe_svm_range_is_valid(range, tile, ctx.devmem_only)) return 0; range_debug(range, "PAGE FAULT"); - /* XXX: Add migration policy, for now migrate range once */ - if (!range->skip_migrate && range->base.flags.migrate_devmem && - xe_svm_range_size(range) >= SZ_64K) { - range->skip_migrate = true; - + if (--migrate_try_count >= 0 && + xe_svm_range_needs_migrate_to_vram(range, vma)) { err = xe_svm_alloc_vram(vm, tile, range, &ctx); if (err) { - drm_dbg(&vm->xe->drm, - "VRAM allocation failed, falling back to " - "retrying fault, asid=%u, errno=%pe\n", - vm->usm.asid, ERR_PTR(err)); - goto retry; + if (migrate_try_count || !ctx.devmem_only) { + drm_dbg(&vm->xe->drm, + "VRAM allocation failed, falling back to retrying fault, asid=%u, errno=%pe\n", + vm->usm.asid, ERR_PTR(err)); + goto retry; + } else { + drm_err(&vm->xe->drm, + "VRAM allocation failed, retry count exceeded, asid=%u, errno=%pe\n", + vm->usm.asid, ERR_PTR(err)); + return err; + } } } @@ -783,15 +838,22 @@ retry: err = drm_gpusvm_range_get_pages(&vm->svm.gpusvm, r, &ctx); /* Corner where CPU mappings have changed */ if (err == -EOPNOTSUPP || err == -EFAULT || err == -EPERM) { - if (err == -EOPNOTSUPP) { - range_debug(range, "PAGE FAULT - EVICT PAGES"); - drm_gpusvm_range_evict(&vm->svm.gpusvm, &range->base); + if (migrate_try_count > 0 || !ctx.devmem_only) { + if (err == -EOPNOTSUPP) { + range_debug(range, "PAGE FAULT - EVICT PAGES"); + drm_gpusvm_range_evict(&vm->svm.gpusvm, + &range->base); + } + drm_dbg(&vm->xe->drm, + "Get pages failed, falling back to retrying, asid=%u, gpusvm=%p, errno=%pe\n", + vm->usm.asid, &vm->svm.gpusvm, ERR_PTR(err)); + range_debug(range, "PAGE FAULT - RETRY PAGES"); + goto retry; + } else { + drm_err(&vm->xe->drm, + "Get pages failed, retry count exceeded, asid=%u, gpusvm=%p, errno=%pe\n", + vm->usm.asid, &vm->svm.gpusvm, ERR_PTR(err)); } - drm_dbg(&vm->xe->drm, - "Get pages failed, falling back to retrying, asid=%u, gpusvm=%p, errno=%pe\n", - vm->usm.asid, &vm->svm.gpusvm, ERR_PTR(err)); - range_debug(range, "PAGE FAULT - RETRY PAGES"); - goto retry; } if (err) { range_debug(range, "PAGE FAULT - FAIL PAGE COLLECT"); @@ -825,9 +887,6 @@ retry_bind: } drm_exec_fini(&exec); - if (xe_modparam.always_migrate_to_vram) - range->skip_migrate = false; - dma_fence_wait(fence, false); dma_fence_put(fence); diff --git a/drivers/gpu/drm/xe/xe_svm.h b/drivers/gpu/drm/xe/xe_svm.h index be306fe7aaa4..fe58ac2f4baa 100644 --- a/drivers/gpu/drm/xe/xe_svm.h +++ b/drivers/gpu/drm/xe/xe_svm.h @@ -36,11 +36,6 @@ struct xe_svm_range { * range. Protected by GPU SVM notifier lock. */ u8 tile_invalidated; - /** - * @skip_migrate: Skip migration to VRAM, protected by GPU fault handler - * locking. - */ - u8 skip_migrate :1; }; #if IS_ENABLED(CONFIG_DRM_GPUSVM) diff --git a/include/drm/drm_gpusvm.h b/include/drm/drm_gpusvm.h index 9fd25fc880a4..653d48dbe1c1 100644 --- a/include/drm/drm_gpusvm.h +++ b/include/drm/drm_gpusvm.h @@ -185,6 +185,31 @@ struct drm_gpusvm_notifier { } flags; }; +/** + * struct drm_gpusvm_range_flags - Structure representing a GPU SVM range flags + * + * @migrate_devmem: Flag indicating whether the range can be migrated to device memory + * @unmapped: Flag indicating if the range has been unmapped + * @partial_unmap: Flag indicating if the range has been partially unmapped + * @has_devmem_pages: Flag indicating if the range has devmem pages + * @has_dma_mapping: Flag indicating if the range has a DMA mapping + * @__flags: Flags for range in u16 form (used for READ_ONCE) + */ +struct drm_gpusvm_range_flags { + union { + struct { + /* All flags below must be set upon creation */ + u16 migrate_devmem : 1; + /* All flags below must be set / cleared under notifier lock */ + u16 unmapped : 1; + u16 partial_unmap : 1; + u16 has_devmem_pages : 1; + u16 has_dma_mapping : 1; + }; + u16 __flags; + }; +}; + /** * struct drm_gpusvm_range - Structure representing a GPU SVM range * @@ -198,11 +223,6 @@ struct drm_gpusvm_notifier { * @dpagemap: The struct drm_pagemap of the device pages we're dma-mapping. * Note this is assuming only one drm_pagemap per range is allowed. * @flags: Flags for range - * @flags.migrate_devmem: Flag indicating whether the range can be migrated to device memory - * @flags.unmapped: Flag indicating if the range has been unmapped - * @flags.partial_unmap: Flag indicating if the range has been partially unmapped - * @flags.has_devmem_pages: Flag indicating if the range has devmem pages - * @flags.has_dma_mapping: Flag indicating if the range has a DMA mapping * * This structure represents a GPU SVM range used for tracking memory ranges * mapped in a DRM device. @@ -216,15 +236,7 @@ struct drm_gpusvm_range { unsigned long notifier_seq; struct drm_pagemap_device_addr *dma_addr; struct drm_pagemap *dpagemap; - struct { - /* All flags below must be set upon creation */ - u16 migrate_devmem : 1; - /* All flags below must be set / cleared under notifier lock */ - u16 unmapped : 1; - u16 partial_unmap : 1; - u16 has_devmem_pages : 1; - u16 has_dma_mapping : 1; - } flags; + struct drm_gpusvm_range_flags flags; }; /** From df8c37810b5a4a5c8bff1f880f59163c800a4985 Mon Sep 17 00:00:00 2001 From: Matthew Brost Date: Mon, 12 May 2025 06:54:57 -0700 Subject: [PATCH 502/555] drm/gpusvm: Add timeslicing support to GPU SVM Add timeslicing support to GPU SVM which will guarantee the GPU a minimum execution time on piece of physical memory before migration back to CPU. Intended to implement strict migration policies which require memory to be in a certain placement for correct execution. Required for shared CPU and GPU atomics on certain devices. Fixes: 99624bdff867 ("drm/gpusvm: Add support for GPU Shared Virtual Memory") Cc: stable@vger.kernel.org Signed-off-by: Matthew Brost Reviewed-by: Himal Prasad Ghimiray Link: https://lore.kernel.org/r/20250512135500.1405019-4-matthew.brost@intel.com (cherry picked from commit 8dc1812b5b3a42311d28eb385eed88e2053ad3cb) Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/drm_gpusvm.c | 9 +++++++++ include/drm/drm_gpusvm.h | 5 +++++ 2 files changed, 14 insertions(+) diff --git a/drivers/gpu/drm/drm_gpusvm.c b/drivers/gpu/drm/drm_gpusvm.c index 41f6616bcf76..4b2f32889f00 100644 --- a/drivers/gpu/drm/drm_gpusvm.c +++ b/drivers/gpu/drm/drm_gpusvm.c @@ -1783,6 +1783,8 @@ int drm_gpusvm_migrate_to_devmem(struct drm_gpusvm *gpusvm, goto err_finalize; /* Upon success bind devmem allocation to range and zdd */ + devmem_allocation->timeslice_expiration = get_jiffies_64() + + msecs_to_jiffies(ctx->timeslice_ms); zdd->devmem_allocation = devmem_allocation; /* Owns ref */ err_finalize: @@ -2003,6 +2005,13 @@ static int __drm_gpusvm_migrate_to_ram(struct vm_area_struct *vas, void *buf; int i, err = 0; + if (page) { + zdd = page->zone_device_data; + if (time_before64(get_jiffies_64(), + zdd->devmem_allocation->timeslice_expiration)) + return 0; + } + start = ALIGN_DOWN(fault_addr, size); end = ALIGN(fault_addr + 1, size); diff --git a/include/drm/drm_gpusvm.h b/include/drm/drm_gpusvm.h index 653d48dbe1c1..eaf704d3d05e 100644 --- a/include/drm/drm_gpusvm.h +++ b/include/drm/drm_gpusvm.h @@ -89,6 +89,7 @@ struct drm_gpusvm_devmem_ops { * @ops: Pointer to the operations structure for GPU SVM device memory * @dpagemap: The struct drm_pagemap of the pages this allocation belongs to. * @size: Size of device memory allocation + * @timeslice_expiration: Timeslice expiration in jiffies */ struct drm_gpusvm_devmem { struct device *dev; @@ -97,6 +98,7 @@ struct drm_gpusvm_devmem { const struct drm_gpusvm_devmem_ops *ops; struct drm_pagemap *dpagemap; size_t size; + u64 timeslice_expiration; }; /** @@ -295,6 +297,8 @@ struct drm_gpusvm { * @check_pages_threshold: Check CPU pages for present if chunk is less than or * equal to threshold. If not present, reduce chunk * size. + * @timeslice_ms: The timeslice MS which in minimum time a piece of memory + * remains with either exclusive GPU or CPU access. * @in_notifier: entering from a MMU notifier * @read_only: operating on read-only memory * @devmem_possible: possible to use device memory @@ -304,6 +308,7 @@ struct drm_gpusvm { */ struct drm_gpusvm_ctx { unsigned long check_pages_threshold; + unsigned long timeslice_ms; unsigned int in_notifier :1; unsigned int read_only :1; unsigned int devmem_possible :1; From 1b36ea2fc6879fed02b675e86867bc422b76f50e Mon Sep 17 00:00:00 2001 From: Matthew Brost Date: Mon, 12 May 2025 06:54:58 -0700 Subject: [PATCH 503/555] drm/xe: Timeslice GPU on atomic SVM fault Ensure GPU can make forward progress on an atomic SVM GPU fault by giving the GPU a timeslice of 5ms v2: - Reduce timeslice to 5ms - Double timeslice on retry - Split out GPU SVM changes into independent patch v5: - Double timeslice in a few more places Fixes: 2f118c949160 ("drm/xe: Add SVM VRAM migration") Cc: stable@vger.kernel.org Signed-off-by: Matthew Brost Reviewed-by: Himal Prasad Ghimiray Link: https://lore.kernel.org/r/20250512135500.1405019-5-matthew.brost@intel.com (cherry picked from commit a5d8d3be1dea8154edbbea481081469627665659) Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/xe_svm.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c index e65d3b820ee0..975094c1a582 100644 --- a/drivers/gpu/drm/xe/xe_svm.c +++ b/drivers/gpu/drm/xe/xe_svm.c @@ -783,6 +783,8 @@ int xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma, IS_ENABLED(CONFIG_DRM_XE_DEVMEM_MIRROR) ? SZ_64K : 0, .devmem_only = atomic && IS_DGFX(vm->xe) && IS_ENABLED(CONFIG_DRM_XE_DEVMEM_MIRROR), + .timeslice_ms = atomic && IS_DGFX(vm->xe) && + IS_ENABLED(CONFIG_DRM_XE_DEVMEM_MIRROR) ? 5 : 0, }; struct xe_svm_range *range; struct drm_gpusvm_range *r; @@ -819,6 +821,7 @@ retry: if (--migrate_try_count >= 0 && xe_svm_range_needs_migrate_to_vram(range, vma)) { err = xe_svm_alloc_vram(vm, tile, range, &ctx); + ctx.timeslice_ms <<= 1; /* Double timeslice if we have to retry */ if (err) { if (migrate_try_count || !ctx.devmem_only) { drm_dbg(&vm->xe->drm, @@ -838,6 +841,7 @@ retry: err = drm_gpusvm_range_get_pages(&vm->svm.gpusvm, r, &ctx); /* Corner where CPU mappings have changed */ if (err == -EOPNOTSUPP || err == -EFAULT || err == -EPERM) { + ctx.timeslice_ms <<= 1; /* Double timeslice if we have to retry */ if (migrate_try_count > 0 || !ctx.devmem_only) { if (err == -EOPNOTSUPP) { range_debug(range, "PAGE FAULT - EVICT PAGES"); @@ -877,6 +881,7 @@ retry_bind: drm_exec_fini(&exec); err = PTR_ERR(fence); if (err == -EAGAIN) { + ctx.timeslice_ms <<= 1; /* Double timeslice if we have to retry */ range_debug(range, "PAGE FAULT - RETRY BIND"); goto retry; } From 66c8f7b435bddb7d8577ac8a57e175a6cb147227 Mon Sep 17 00:00:00 2001 From: Umesh Nerlige Ramappa Date: Fri, 9 May 2025 09:12:01 -0700 Subject: [PATCH 504/555] drm/xe: Save CTX_TIMESTAMP mmio value instead of LRC value For determining actual job execution time, save the current value of the CTX_TIMESTAMP register rather than the value saved in LRC since the current register value is the closest to the start time of the job. v2: Define MI_STORE_REGISTER_MEM to fix compile error v3: Place MI_STORE_REGISTER_MEM sorted by MI_INSTR (Lucas) Fixes: 65921374c48f ("drm/xe: Emit ctx timestamp copy in ring ops") Signed-off-by: Umesh Nerlige Ramappa Reviewed-by: Matthew Brost Reviewed-by: Lucas De Marchi Link: https://lore.kernel.org/r/20250509161159.2173069-6-umesh.nerlige.ramappa@intel.com (cherry picked from commit 38b14233e5deff51db8faec287b4acd227152246) Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/instructions/xe_mi_commands.h | 4 ++++ drivers/gpu/drm/xe/xe_lrc.c | 2 +- drivers/gpu/drm/xe/xe_ring_ops.c | 7 ++----- 3 files changed, 7 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/xe/instructions/xe_mi_commands.h b/drivers/gpu/drm/xe/instructions/xe_mi_commands.h index 167fb0f742de..5a47991b4b81 100644 --- a/drivers/gpu/drm/xe/instructions/xe_mi_commands.h +++ b/drivers/gpu/drm/xe/instructions/xe_mi_commands.h @@ -47,6 +47,10 @@ #define MI_LRI_FORCE_POSTED REG_BIT(12) #define MI_LRI_LEN(x) (((x) & 0xff) + 1) +#define MI_STORE_REGISTER_MEM (__MI_INSTR(0x24) | XE_INSTR_NUM_DW(4)) +#define MI_SRM_USE_GGTT REG_BIT(22) +#define MI_SRM_ADD_CS_OFFSET REG_BIT(19) + #define MI_FLUSH_DW __MI_INSTR(0x26) #define MI_FLUSH_DW_PROTECTED_MEM_EN REG_BIT(22) #define MI_FLUSH_DW_STORE_INDEX REG_BIT(21) diff --git a/drivers/gpu/drm/xe/xe_lrc.c b/drivers/gpu/drm/xe/xe_lrc.c index df3ceddede07..48e8fd85eabe 100644 --- a/drivers/gpu/drm/xe/xe_lrc.c +++ b/drivers/gpu/drm/xe/xe_lrc.c @@ -684,7 +684,7 @@ static inline u32 __xe_lrc_start_seqno_offset(struct xe_lrc *lrc) static u32 __xe_lrc_ctx_job_timestamp_offset(struct xe_lrc *lrc) { - /* The start seqno is stored in the driver-defined portion of PPHWSP */ + /* This is stored in the driver-defined portion of PPHWSP */ return xe_lrc_pphwsp_offset(lrc) + LRC_CTX_JOB_TIMESTAMP_OFFSET; } diff --git a/drivers/gpu/drm/xe/xe_ring_ops.c b/drivers/gpu/drm/xe/xe_ring_ops.c index a7582b097ae6..bc1689db4cd7 100644 --- a/drivers/gpu/drm/xe/xe_ring_ops.c +++ b/drivers/gpu/drm/xe/xe_ring_ops.c @@ -234,13 +234,10 @@ static u32 get_ppgtt_flag(struct xe_sched_job *job) static int emit_copy_timestamp(struct xe_lrc *lrc, u32 *dw, int i) { - dw[i++] = MI_COPY_MEM_MEM | MI_COPY_MEM_MEM_SRC_GGTT | - MI_COPY_MEM_MEM_DST_GGTT; + dw[i++] = MI_STORE_REGISTER_MEM | MI_SRM_USE_GGTT | MI_SRM_ADD_CS_OFFSET; + dw[i++] = RING_CTX_TIMESTAMP(0).addr; dw[i++] = xe_lrc_ctx_job_timestamp_ggtt_addr(lrc); dw[i++] = 0; - dw[i++] = xe_lrc_ctx_timestamp_ggtt_addr(lrc); - dw[i++] = 0; - dw[i++] = MI_NOOP; return i; } From ce15563e49fb0b5c802564433ff8468acd1339eb Mon Sep 17 00:00:00 2001 From: Umesh Nerlige Ramappa Date: Fri, 9 May 2025 09:12:02 -0700 Subject: [PATCH 505/555] drm/xe: Save the gt pointer in lrc and drop the tile Save the gt pointer in the lrc so that it can used for gt based helpers. Signed-off-by: Umesh Nerlige Ramappa Reviewed-by: Matthew Brost Reviewed-by: Lucas De Marchi Link: https://lore.kernel.org/r/20250509161159.2173069-7-umesh.nerlige.ramappa@intel.com (cherry picked from commit 741d3ef8b8b88fab2729ca89de1180e49bc9cef0) Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/xe_lrc.c | 4 ++-- drivers/gpu/drm/xe/xe_lrc_types.h | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_lrc.c b/drivers/gpu/drm/xe/xe_lrc.c index 48e8fd85eabe..79b43fec4f79 100644 --- a/drivers/gpu/drm/xe/xe_lrc.c +++ b/drivers/gpu/drm/xe/xe_lrc.c @@ -864,7 +864,7 @@ static void *empty_lrc_data(struct xe_hw_engine *hwe) static void xe_lrc_set_ppgtt(struct xe_lrc *lrc, struct xe_vm *vm) { - u64 desc = xe_vm_pdp4_descriptor(vm, lrc->tile); + u64 desc = xe_vm_pdp4_descriptor(vm, gt_to_tile(lrc->gt)); xe_lrc_write_ctx_reg(lrc, CTX_PDP0_UDW, upper_32_bits(desc)); xe_lrc_write_ctx_reg(lrc, CTX_PDP0_LDW, lower_32_bits(desc)); @@ -896,6 +896,7 @@ static int xe_lrc_init(struct xe_lrc *lrc, struct xe_hw_engine *hwe, int err; kref_init(&lrc->refcount); + lrc->gt = gt; lrc->flags = 0; lrc_size = ring_size + xe_gt_lrc_size(gt, hwe->class); if (xe_gt_has_indirect_ring_state(gt)) @@ -914,7 +915,6 @@ static int xe_lrc_init(struct xe_lrc *lrc, struct xe_hw_engine *hwe, return PTR_ERR(lrc->bo); lrc->size = lrc_size; - lrc->tile = gt_to_tile(hwe->gt); lrc->ring.size = ring_size; lrc->ring.tail = 0; lrc->ctx_timestamp = 0; diff --git a/drivers/gpu/drm/xe/xe_lrc_types.h b/drivers/gpu/drm/xe/xe_lrc_types.h index 71ecb453f811..cd38586ae989 100644 --- a/drivers/gpu/drm/xe/xe_lrc_types.h +++ b/drivers/gpu/drm/xe/xe_lrc_types.h @@ -25,8 +25,8 @@ struct xe_lrc { /** @size: size of lrc including any indirect ring state page */ u32 size; - /** @tile: tile which this LRC belongs to */ - struct xe_tile *tile; + /** @gt: gt which this LRC belongs to */ + struct xe_gt *gt; /** @flags: LRC flags */ #define XE_LRC_FLAG_INDIRECT_RING_STATE 0x1 From 617d824c5323b8474b3665ae6c410c98b839e0b0 Mon Sep 17 00:00:00 2001 From: Umesh Nerlige Ramappa Date: Fri, 9 May 2025 09:12:03 -0700 Subject: [PATCH 506/555] drm/xe: Add WA BB to capture active context utilization Context Timestamp (CTX_TIMESTAMP) in the LRC accumulates the run ticks of the context, but only gets updated when the context switches out. In order to check how long a context has been active before it switches out, two things are required: (1) Determine if the context is running: To do so, we program the WA BB to set an initial value for CTX_TIMESTAMP in the LRC. The value chosen is 1 since 0 is the initial value when the LRC is initialized. During a query, we just check for this value to determine if the context is active. If the context switched out, it would overwrite this location with the actual CTX_TIMESTAMP MMIO value. Note that WA BB runs as the last part of the context restore, so reusing this LRC location will not clobber anything. (2) Calculate the time that the context has been active for: The CTX_TIMESTAMP ticks only when the context is active. If a context is active, we just use the CTX_TIMESTAMP MMIO as the new value of utilization. While doing so, we need to read the CTX_TIMESTAMP MMIO for the specific engine instance. Since we do not know which instance the context is running on until it is scheduled, we also read the ENGINE_ID MMIO in the WA BB and store it in the PPHSWP. Using the above 2 instructions in a WA BB, capture active context utilization. v2: (Matt Brost) - This breaks TDR, fix it by saving the CTX_TIMESTAMP register "drm/xe: Save CTX_TIMESTAMP mmio value instead of LRC value" - Drop tile from LRC if using gt "drm/xe: Save the gt pointer in LRC and drop the tile" v3: - Remove helpers for bb_per_ctx_ptr (Matt) - Add define for context active value (Matt) - Use 64 bit CTX TIMESTAMP for platforms that support it. For platforms that don't, live with the rare race. (Matt, Lucas) - Convert engine id to hwe and get the MMIO value (Lucas) - Correct commit message on when WA BB runs (Lucas) v4: - s/GRAPHICS_VER(...)/xe->info.has_64bit_timestamp/ (Matt) - Drop support for active utilization on a VF (CI failure) - In xe_lrc_init ensure the lrc value is 0 to begin with (CI regression) v5: - Minor checkpatch fix - Squash into previous commit and make TDR use 32-bit time - Update code comment to match commit msg Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/4532 Cc: # v6.13+ Suggested-by: Lucas De Marchi Signed-off-by: Umesh Nerlige Ramappa Reviewed-by: Lucas De Marchi Link: https://lore.kernel.org/r/20250509161159.2173069-8-umesh.nerlige.ramappa@intel.com (cherry picked from commit 82b98cadb01f63cdb159e596ec06866d00f8e8c7) Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/regs/xe_engine_regs.h | 5 + drivers/gpu/drm/xe/regs/xe_lrc_layout.h | 2 + drivers/gpu/drm/xe/xe_device_types.h | 2 + drivers/gpu/drm/xe/xe_exec_queue.c | 2 +- drivers/gpu/drm/xe/xe_guc_submit.c | 2 +- drivers/gpu/drm/xe/xe_lrc.c | 193 +++++++++++++++++++++-- drivers/gpu/drm/xe/xe_lrc.h | 5 +- drivers/gpu/drm/xe/xe_lrc_types.h | 5 +- drivers/gpu/drm/xe/xe_pci.c | 2 + drivers/gpu/drm/xe/xe_pci_types.h | 1 + drivers/gpu/drm/xe/xe_trace_lrc.h | 8 +- 11 files changed, 208 insertions(+), 19 deletions(-) diff --git a/drivers/gpu/drm/xe/regs/xe_engine_regs.h b/drivers/gpu/drm/xe/regs/xe_engine_regs.h index fb8ec317b6ee..891f928d80ce 100644 --- a/drivers/gpu/drm/xe/regs/xe_engine_regs.h +++ b/drivers/gpu/drm/xe/regs/xe_engine_regs.h @@ -43,6 +43,10 @@ #define XEHPC_BCS8_RING_BASE 0x3ee000 #define GSCCS_RING_BASE 0x11a000 +#define ENGINE_ID(base) XE_REG((base) + 0x8c) +#define ENGINE_INSTANCE_ID REG_GENMASK(9, 4) +#define ENGINE_CLASS_ID REG_GENMASK(2, 0) + #define RING_TAIL(base) XE_REG((base) + 0x30) #define TAIL_ADDR REG_GENMASK(20, 3) @@ -154,6 +158,7 @@ #define STOP_RING REG_BIT(8) #define RING_CTX_TIMESTAMP(base) XE_REG((base) + 0x3a8) +#define RING_CTX_TIMESTAMP_UDW(base) XE_REG((base) + 0x3ac) #define CSBE_DEBUG_STATUS(base) XE_REG((base) + 0x3fc) #define RING_FORCE_TO_NONPRIV(base, i) XE_REG(((base) + 0x4d0) + (i) * 4) diff --git a/drivers/gpu/drm/xe/regs/xe_lrc_layout.h b/drivers/gpu/drm/xe/regs/xe_lrc_layout.h index 57944f90bbf6..994af591a2e8 100644 --- a/drivers/gpu/drm/xe/regs/xe_lrc_layout.h +++ b/drivers/gpu/drm/xe/regs/xe_lrc_layout.h @@ -11,7 +11,9 @@ #define CTX_RING_TAIL (0x06 + 1) #define CTX_RING_START (0x08 + 1) #define CTX_RING_CTL (0x0a + 1) +#define CTX_BB_PER_CTX_PTR (0x12 + 1) #define CTX_TIMESTAMP (0x22 + 1) +#define CTX_TIMESTAMP_UDW (0x24 + 1) #define CTX_INDIRECT_RING_STATE (0x26 + 1) #define CTX_PDP0_UDW (0x30 + 1) #define CTX_PDP0_LDW (0x32 + 1) diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h index 9f8667ebba85..0482f26aa480 100644 --- a/drivers/gpu/drm/xe/xe_device_types.h +++ b/drivers/gpu/drm/xe/xe_device_types.h @@ -330,6 +330,8 @@ struct xe_device { u8 has_sriov:1; /** @info.has_usm: Device has unified shared memory support */ u8 has_usm:1; + /** @info.has_64bit_timestamp: Device supports 64-bit timestamps */ + u8 has_64bit_timestamp:1; /** @info.is_dgfx: is discrete device */ u8 is_dgfx:1; /** diff --git a/drivers/gpu/drm/xe/xe_exec_queue.c b/drivers/gpu/drm/xe/xe_exec_queue.c index 606922d9dd73..cd9b1c32f30f 100644 --- a/drivers/gpu/drm/xe/xe_exec_queue.c +++ b/drivers/gpu/drm/xe/xe_exec_queue.c @@ -830,7 +830,7 @@ void xe_exec_queue_update_run_ticks(struct xe_exec_queue *q) { struct xe_device *xe = gt_to_xe(q->gt); struct xe_lrc *lrc; - u32 old_ts, new_ts; + u64 old_ts, new_ts; int idx; /* diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c index 31bc2022bfc2..769781d577df 100644 --- a/drivers/gpu/drm/xe/xe_guc_submit.c +++ b/drivers/gpu/drm/xe/xe_guc_submit.c @@ -941,7 +941,7 @@ static bool check_timeout(struct xe_exec_queue *q, struct xe_sched_job *job) return xe_sched_invalidate_job(job, 2); } - ctx_timestamp = xe_lrc_ctx_timestamp(q->lrc[0]); + ctx_timestamp = lower_32_bits(xe_lrc_ctx_timestamp(q->lrc[0])); ctx_job_timestamp = xe_lrc_ctx_job_timestamp(q->lrc[0]); /* diff --git a/drivers/gpu/drm/xe/xe_lrc.c b/drivers/gpu/drm/xe/xe_lrc.c index 79b43fec4f79..03bfba696b37 100644 --- a/drivers/gpu/drm/xe/xe_lrc.c +++ b/drivers/gpu/drm/xe/xe_lrc.c @@ -24,6 +24,7 @@ #include "xe_hw_fence.h" #include "xe_map.h" #include "xe_memirq.h" +#include "xe_mmio.h" #include "xe_sriov.h" #include "xe_trace_lrc.h" #include "xe_vm.h" @@ -650,6 +651,7 @@ u32 xe_lrc_pphwsp_offset(struct xe_lrc *lrc) #define LRC_START_SEQNO_PPHWSP_OFFSET (LRC_SEQNO_PPHWSP_OFFSET + 8) #define LRC_CTX_JOB_TIMESTAMP_OFFSET (LRC_START_SEQNO_PPHWSP_OFFSET + 8) #define LRC_PARALLEL_PPHWSP_OFFSET 2048 +#define LRC_ENGINE_ID_PPHWSP_OFFSET 2096 #define LRC_PPHWSP_SIZE SZ_4K u32 xe_lrc_regs_offset(struct xe_lrc *lrc) @@ -694,11 +696,21 @@ static inline u32 __xe_lrc_parallel_offset(struct xe_lrc *lrc) return xe_lrc_pphwsp_offset(lrc) + LRC_PARALLEL_PPHWSP_OFFSET; } +static inline u32 __xe_lrc_engine_id_offset(struct xe_lrc *lrc) +{ + return xe_lrc_pphwsp_offset(lrc) + LRC_ENGINE_ID_PPHWSP_OFFSET; +} + static u32 __xe_lrc_ctx_timestamp_offset(struct xe_lrc *lrc) { return __xe_lrc_regs_offset(lrc) + CTX_TIMESTAMP * sizeof(u32); } +static u32 __xe_lrc_ctx_timestamp_udw_offset(struct xe_lrc *lrc) +{ + return __xe_lrc_regs_offset(lrc) + CTX_TIMESTAMP_UDW * sizeof(u32); +} + static inline u32 __xe_lrc_indirect_ring_offset(struct xe_lrc *lrc) { /* Indirect ring state page is at the very end of LRC */ @@ -726,8 +738,10 @@ DECL_MAP_ADDR_HELPERS(regs) DECL_MAP_ADDR_HELPERS(start_seqno) DECL_MAP_ADDR_HELPERS(ctx_job_timestamp) DECL_MAP_ADDR_HELPERS(ctx_timestamp) +DECL_MAP_ADDR_HELPERS(ctx_timestamp_udw) DECL_MAP_ADDR_HELPERS(parallel) DECL_MAP_ADDR_HELPERS(indirect_ring) +DECL_MAP_ADDR_HELPERS(engine_id) #undef DECL_MAP_ADDR_HELPERS @@ -742,19 +756,38 @@ u32 xe_lrc_ctx_timestamp_ggtt_addr(struct xe_lrc *lrc) return __xe_lrc_ctx_timestamp_ggtt_addr(lrc); } +/** + * xe_lrc_ctx_timestamp_udw_ggtt_addr() - Get ctx timestamp udw GGTT address + * @lrc: Pointer to the lrc. + * + * Returns: ctx timestamp udw GGTT address + */ +u32 xe_lrc_ctx_timestamp_udw_ggtt_addr(struct xe_lrc *lrc) +{ + return __xe_lrc_ctx_timestamp_udw_ggtt_addr(lrc); +} + /** * xe_lrc_ctx_timestamp() - Read ctx timestamp value * @lrc: Pointer to the lrc. * * Returns: ctx timestamp value */ -u32 xe_lrc_ctx_timestamp(struct xe_lrc *lrc) +u64 xe_lrc_ctx_timestamp(struct xe_lrc *lrc) { struct xe_device *xe = lrc_to_xe(lrc); struct iosys_map map; + u32 ldw, udw = 0; map = __xe_lrc_ctx_timestamp_map(lrc); - return xe_map_read32(xe, &map); + ldw = xe_map_read32(xe, &map); + + if (xe->info.has_64bit_timestamp) { + map = __xe_lrc_ctx_timestamp_udw_map(lrc); + udw = xe_map_read32(xe, &map); + } + + return (u64)udw << 32 | ldw; } /** @@ -877,6 +910,65 @@ static void xe_lrc_finish(struct xe_lrc *lrc) xe_bo_unpin(lrc->bo); xe_bo_unlock(lrc->bo); xe_bo_put(lrc->bo); + xe_bo_unpin_map_no_vm(lrc->bb_per_ctx_bo); +} + +/* + * xe_lrc_setup_utilization() - Setup wa bb to assist in calculating active + * context run ticks. + * @lrc: Pointer to the lrc. + * + * Context Timestamp (CTX_TIMESTAMP) in the LRC accumulates the run ticks of the + * context, but only gets updated when the context switches out. In order to + * check how long a context has been active before it switches out, two things + * are required: + * + * (1) Determine if the context is running: + * To do so, we program the WA BB to set an initial value for CTX_TIMESTAMP in + * the LRC. The value chosen is 1 since 0 is the initial value when the LRC is + * initialized. During a query, we just check for this value to determine if the + * context is active. If the context switched out, it would overwrite this + * location with the actual CTX_TIMESTAMP MMIO value. Note that WA BB runs as + * the last part of context restore, so reusing this LRC location will not + * clobber anything. + * + * (2) Calculate the time that the context has been active for: + * The CTX_TIMESTAMP ticks only when the context is active. If a context is + * active, we just use the CTX_TIMESTAMP MMIO as the new value of utilization. + * While doing so, we need to read the CTX_TIMESTAMP MMIO for the specific + * engine instance. Since we do not know which instance the context is running + * on until it is scheduled, we also read the ENGINE_ID MMIO in the WA BB and + * store it in the PPHSWP. + */ +#define CONTEXT_ACTIVE 1ULL +static void xe_lrc_setup_utilization(struct xe_lrc *lrc) +{ + u32 *cmd; + + cmd = lrc->bb_per_ctx_bo->vmap.vaddr; + + *cmd++ = MI_STORE_REGISTER_MEM | MI_SRM_USE_GGTT | MI_SRM_ADD_CS_OFFSET; + *cmd++ = ENGINE_ID(0).addr; + *cmd++ = __xe_lrc_engine_id_ggtt_addr(lrc); + *cmd++ = 0; + + *cmd++ = MI_STORE_DATA_IMM | MI_SDI_GGTT | MI_SDI_NUM_DW(1); + *cmd++ = __xe_lrc_ctx_timestamp_ggtt_addr(lrc); + *cmd++ = 0; + *cmd++ = lower_32_bits(CONTEXT_ACTIVE); + + if (lrc_to_xe(lrc)->info.has_64bit_timestamp) { + *cmd++ = MI_STORE_DATA_IMM | MI_SDI_GGTT | MI_SDI_NUM_DW(1); + *cmd++ = __xe_lrc_ctx_timestamp_udw_ggtt_addr(lrc); + *cmd++ = 0; + *cmd++ = upper_32_bits(CONTEXT_ACTIVE); + } + + *cmd++ = MI_BATCH_BUFFER_END; + + xe_lrc_write_ctx_reg(lrc, CTX_BB_PER_CTX_PTR, + xe_bo_ggtt_addr(lrc->bb_per_ctx_bo) | 1); + } #define PVC_CTX_ASID (0x2e + 1) @@ -893,6 +985,7 @@ static int xe_lrc_init(struct xe_lrc *lrc, struct xe_hw_engine *hwe, void *init_data = NULL; u32 arb_enable; u32 lrc_size; + u32 bo_flags; int err; kref_init(&lrc->refcount); @@ -902,22 +995,30 @@ static int xe_lrc_init(struct xe_lrc *lrc, struct xe_hw_engine *hwe, if (xe_gt_has_indirect_ring_state(gt)) lrc->flags |= XE_LRC_FLAG_INDIRECT_RING_STATE; + bo_flags = XE_BO_FLAG_VRAM_IF_DGFX(tile) | XE_BO_FLAG_GGTT | + XE_BO_FLAG_GGTT_INVALIDATE; + /* * FIXME: Perma-pinning LRC as we don't yet support moving GGTT address * via VM bind calls. */ lrc->bo = xe_bo_create_pin_map(xe, tile, vm, lrc_size, ttm_bo_type_kernel, - XE_BO_FLAG_VRAM_IF_DGFX(tile) | - XE_BO_FLAG_GGTT | - XE_BO_FLAG_GGTT_INVALIDATE); + bo_flags); if (IS_ERR(lrc->bo)) return PTR_ERR(lrc->bo); + lrc->bb_per_ctx_bo = xe_bo_create_pin_map(xe, tile, NULL, SZ_4K, + ttm_bo_type_kernel, + bo_flags); + if (IS_ERR(lrc->bb_per_ctx_bo)) { + err = PTR_ERR(lrc->bb_per_ctx_bo); + goto err_lrc_finish; + } + lrc->size = lrc_size; lrc->ring.size = ring_size; lrc->ring.tail = 0; - lrc->ctx_timestamp = 0; xe_hw_fence_ctx_init(&lrc->fence_ctx, hwe->gt, hwe->fence_irq, hwe->name); @@ -990,7 +1091,10 @@ static int xe_lrc_init(struct xe_lrc *lrc, struct xe_hw_engine *hwe, xe_lrc_read_ctx_reg(lrc, CTX_CONTEXT_CONTROL) | _MASKED_BIT_ENABLE(CTX_CTRL_PXP_ENABLE)); + lrc->ctx_timestamp = 0; xe_lrc_write_ctx_reg(lrc, CTX_TIMESTAMP, 0); + if (lrc_to_xe(lrc)->info.has_64bit_timestamp) + xe_lrc_write_ctx_reg(lrc, CTX_TIMESTAMP_UDW, 0); if (xe->info.has_asid && vm) xe_lrc_write_ctx_reg(lrc, PVC_CTX_ASID, vm->usm.asid); @@ -1019,6 +1123,8 @@ static int xe_lrc_init(struct xe_lrc *lrc, struct xe_hw_engine *hwe, map = __xe_lrc_start_seqno_map(lrc); xe_map_write32(lrc_to_xe(lrc), &map, lrc->fence_ctx.next_seqno - 1); + xe_lrc_setup_utilization(lrc); + return 0; err_lrc_finish: @@ -1238,6 +1344,21 @@ struct iosys_map xe_lrc_parallel_map(struct xe_lrc *lrc) return __xe_lrc_parallel_map(lrc); } +/** + * xe_lrc_engine_id() - Read engine id value + * @lrc: Pointer to the lrc. + * + * Returns: context id value + */ +static u32 xe_lrc_engine_id(struct xe_lrc *lrc) +{ + struct xe_device *xe = lrc_to_xe(lrc); + struct iosys_map map; + + map = __xe_lrc_engine_id_map(lrc); + return xe_map_read32(xe, &map); +} + static int instr_dw(u32 cmd_header) { /* GFXPIPE "SINGLE_DW" opcodes are a single dword */ @@ -1684,7 +1805,7 @@ struct xe_lrc_snapshot *xe_lrc_snapshot_capture(struct xe_lrc *lrc) snapshot->lrc_offset = xe_lrc_pphwsp_offset(lrc); snapshot->lrc_size = lrc->bo->size - snapshot->lrc_offset; snapshot->lrc_snapshot = NULL; - snapshot->ctx_timestamp = xe_lrc_ctx_timestamp(lrc); + snapshot->ctx_timestamp = lower_32_bits(xe_lrc_ctx_timestamp(lrc)); snapshot->ctx_job_timestamp = xe_lrc_ctx_job_timestamp(lrc); return snapshot; } @@ -1784,22 +1905,74 @@ void xe_lrc_snapshot_free(struct xe_lrc_snapshot *snapshot) kfree(snapshot); } +static int get_ctx_timestamp(struct xe_lrc *lrc, u32 engine_id, u64 *reg_ctx_ts) +{ + u16 class = REG_FIELD_GET(ENGINE_CLASS_ID, engine_id); + u16 instance = REG_FIELD_GET(ENGINE_INSTANCE_ID, engine_id); + struct xe_hw_engine *hwe; + u64 val; + + hwe = xe_gt_hw_engine(lrc->gt, class, instance, false); + if (xe_gt_WARN_ONCE(lrc->gt, !hwe || xe_hw_engine_is_reserved(hwe), + "Unexpected engine class:instance %d:%d for context utilization\n", + class, instance)) + return -1; + + if (lrc_to_xe(lrc)->info.has_64bit_timestamp) + val = xe_mmio_read64_2x32(&hwe->gt->mmio, + RING_CTX_TIMESTAMP(hwe->mmio_base)); + else + val = xe_mmio_read32(&hwe->gt->mmio, + RING_CTX_TIMESTAMP(hwe->mmio_base)); + + *reg_ctx_ts = val; + + return 0; +} + /** * xe_lrc_update_timestamp() - Update ctx timestamp * @lrc: Pointer to the lrc. * @old_ts: Old timestamp value * * Populate @old_ts current saved ctx timestamp, read new ctx timestamp and - * update saved value. + * update saved value. With support for active contexts, the calculation may be + * slightly racy, so follow a read-again logic to ensure that the context is + * still active before returning the right timestamp. * * Returns: New ctx timestamp value */ -u32 xe_lrc_update_timestamp(struct xe_lrc *lrc, u32 *old_ts) +u64 xe_lrc_update_timestamp(struct xe_lrc *lrc, u64 *old_ts) { + u64 lrc_ts, reg_ts; + u32 engine_id; + *old_ts = lrc->ctx_timestamp; - lrc->ctx_timestamp = xe_lrc_ctx_timestamp(lrc); + lrc_ts = xe_lrc_ctx_timestamp(lrc); + /* CTX_TIMESTAMP mmio read is invalid on VF, so return the LRC value */ + if (IS_SRIOV_VF(lrc_to_xe(lrc))) { + lrc->ctx_timestamp = lrc_ts; + goto done; + } + if (lrc_ts == CONTEXT_ACTIVE) { + engine_id = xe_lrc_engine_id(lrc); + if (!get_ctx_timestamp(lrc, engine_id, ®_ts)) + lrc->ctx_timestamp = reg_ts; + + /* read lrc again to ensure context is still active */ + lrc_ts = xe_lrc_ctx_timestamp(lrc); + } + + /* + * If context switched out, just use the lrc_ts. Note that this needs to + * be a separate if condition. + */ + if (lrc_ts != CONTEXT_ACTIVE) + lrc->ctx_timestamp = lrc_ts; + +done: trace_xe_lrc_update_timestamp(lrc, *old_ts); return lrc->ctx_timestamp; diff --git a/drivers/gpu/drm/xe/xe_lrc.h b/drivers/gpu/drm/xe/xe_lrc.h index 0b40f349ab95..eb6e8de8c939 100644 --- a/drivers/gpu/drm/xe/xe_lrc.h +++ b/drivers/gpu/drm/xe/xe_lrc.h @@ -120,7 +120,8 @@ void xe_lrc_snapshot_print(struct xe_lrc_snapshot *snapshot, struct drm_printer void xe_lrc_snapshot_free(struct xe_lrc_snapshot *snapshot); u32 xe_lrc_ctx_timestamp_ggtt_addr(struct xe_lrc *lrc); -u32 xe_lrc_ctx_timestamp(struct xe_lrc *lrc); +u32 xe_lrc_ctx_timestamp_udw_ggtt_addr(struct xe_lrc *lrc); +u64 xe_lrc_ctx_timestamp(struct xe_lrc *lrc); u32 xe_lrc_ctx_job_timestamp_ggtt_addr(struct xe_lrc *lrc); u32 xe_lrc_ctx_job_timestamp(struct xe_lrc *lrc); @@ -136,6 +137,6 @@ u32 xe_lrc_ctx_job_timestamp(struct xe_lrc *lrc); * * Returns the current LRC timestamp */ -u32 xe_lrc_update_timestamp(struct xe_lrc *lrc, u32 *old_ts); +u64 xe_lrc_update_timestamp(struct xe_lrc *lrc, u64 *old_ts); #endif diff --git a/drivers/gpu/drm/xe/xe_lrc_types.h b/drivers/gpu/drm/xe/xe_lrc_types.h index cd38586ae989..ae24cf6f8dd9 100644 --- a/drivers/gpu/drm/xe/xe_lrc_types.h +++ b/drivers/gpu/drm/xe/xe_lrc_types.h @@ -52,7 +52,10 @@ struct xe_lrc { struct xe_hw_fence_ctx fence_ctx; /** @ctx_timestamp: readout value of CTX_TIMESTAMP on last update */ - u32 ctx_timestamp; + u64 ctx_timestamp; + + /** @bb_per_ctx_bo: buffer object for per context batch wa buffer */ + struct xe_bo *bb_per_ctx_bo; }; struct xe_lrc_snapshot; diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c index 818f023166d5..f4d108dc49b1 100644 --- a/drivers/gpu/drm/xe/xe_pci.c +++ b/drivers/gpu/drm/xe/xe_pci.c @@ -140,6 +140,7 @@ static const struct xe_graphics_desc graphics_xelpg = { .has_indirect_ring_state = 1, \ .has_range_tlb_invalidation = 1, \ .has_usm = 1, \ + .has_64bit_timestamp = 1, \ .va_bits = 48, \ .vm_max_level = 4, \ .hw_engine_mask = \ @@ -668,6 +669,7 @@ static int xe_info_init(struct xe_device *xe, xe->info.has_range_tlb_invalidation = graphics_desc->has_range_tlb_invalidation; xe->info.has_usm = graphics_desc->has_usm; + xe->info.has_64bit_timestamp = graphics_desc->has_64bit_timestamp; for_each_remote_tile(tile, xe, id) { int err; diff --git a/drivers/gpu/drm/xe/xe_pci_types.h b/drivers/gpu/drm/xe/xe_pci_types.h index e9b9bbc138d3..ca6b10d35573 100644 --- a/drivers/gpu/drm/xe/xe_pci_types.h +++ b/drivers/gpu/drm/xe/xe_pci_types.h @@ -21,6 +21,7 @@ struct xe_graphics_desc { u8 has_indirect_ring_state:1; u8 has_range_tlb_invalidation:1; u8 has_usm:1; + u8 has_64bit_timestamp:1; }; struct xe_media_desc { diff --git a/drivers/gpu/drm/xe/xe_trace_lrc.h b/drivers/gpu/drm/xe/xe_trace_lrc.h index 5c669a0b2180..d525cbee1e34 100644 --- a/drivers/gpu/drm/xe/xe_trace_lrc.h +++ b/drivers/gpu/drm/xe/xe_trace_lrc.h @@ -19,12 +19,12 @@ #define __dev_name_lrc(lrc) dev_name(gt_to_xe((lrc)->fence_ctx.gt)->drm.dev) TRACE_EVENT(xe_lrc_update_timestamp, - TP_PROTO(struct xe_lrc *lrc, uint32_t old), + TP_PROTO(struct xe_lrc *lrc, uint64_t old), TP_ARGS(lrc, old), TP_STRUCT__entry( __field(struct xe_lrc *, lrc) - __field(u32, old) - __field(u32, new) + __field(u64, old) + __field(u64, new) __string(name, lrc->fence_ctx.name) __string(device_id, __dev_name_lrc(lrc)) ), @@ -36,7 +36,7 @@ TRACE_EVENT(xe_lrc_update_timestamp, __assign_str(name); __assign_str(device_id); ), - TP_printk("lrc=:%p lrc->name=%s old=%u new=%u device_id:%s", + TP_printk("lrc=:%p lrc->name=%s old=%llu new=%llu device_id:%s", __entry->lrc, __get_str(name), __entry->old, __entry->new, __get_str(device_id)) From e333332657f615ac2b55aa35565c4a882018bbe9 Mon Sep 17 00:00:00 2001 From: pengdonglin Date: Mon, 12 May 2025 17:42:45 +0800 Subject: [PATCH 507/555] ftrace: Fix preemption accounting for stacktrace trigger command When using the stacktrace trigger command to trace syscalls, the preemption count was consistently reported as 1 when the system call event itself had 0 ("."). For example: root@ubuntu22-vm:/sys/kernel/tracing/events/syscalls/sys_enter_read $ echo stacktrace > trigger $ echo 1 > enable sshd-416 [002] ..... 232.864910: sys_read(fd: a, buf: 556b1f3221d0, count: 8000) sshd-416 [002] ...1. 232.864913: => ftrace_syscall_enter => syscall_trace_enter => do_syscall_64 => entry_SYSCALL_64_after_hwframe The root cause is that the trace framework disables preemption in __DO_TRACE before invoking the trigger callback. Use the tracing_gen_ctx_dec() that will accommodate for the increase of the preemption count in __DO_TRACE when calling the callback. The result is the accurate reporting of: sshd-410 [004] ..... 210.117660: sys_read(fd: 4, buf: 559b725ba130, count: 40000) sshd-410 [004] ..... 210.117662: => ftrace_syscall_enter => syscall_trace_enter => do_syscall_64 => entry_SYSCALL_64_after_hwframe Cc: stable@vger.kernel.org Fixes: ce33c845b030c ("tracing: Dump stacktrace trigger to the corresponding instance") Link: https://lore.kernel.org/20250512094246.1167956-1-dolinux.peng@gmail.com Signed-off-by: pengdonglin Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace_events_trigger.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/trace/trace_events_trigger.c b/kernel/trace/trace_events_trigger.c index b66b6d235d91..6e87ae2a1a66 100644 --- a/kernel/trace/trace_events_trigger.c +++ b/kernel/trace/trace_events_trigger.c @@ -1560,7 +1560,7 @@ stacktrace_trigger(struct event_trigger_data *data, struct trace_event_file *file = data->private_data; if (file) - __trace_stack(file->tr, tracing_gen_ctx(), STACK_SKIP); + __trace_stack(file->tr, tracing_gen_ctx_dec(), STACK_SKIP); else trace_dump_stack(STACK_SKIP); } From 11aff32439df6ca5b3b891b43032faf88f4a6a29 Mon Sep 17 00:00:00 2001 From: pengdonglin Date: Mon, 12 May 2025 17:42:46 +0800 Subject: [PATCH 508/555] ftrace: Fix preemption accounting for stacktrace filter command The preemption count of the stacktrace filter command to trace ksys_read is consistently incorrect: $ echo ksys_read:stacktrace > set_ftrace_filter <...>-453 [004] ...1. 38.308956: => ksys_read => do_syscall_64 => entry_SYSCALL_64_after_hwframe The root cause is that the trace framework disables preemption when invoking the filter command callback in function_trace_probe_call: preempt_disable_notrace(); probe_ops->func(ip, parent_ip, probe_opsbe->tr, probe_ops, probe->data); preempt_enable_notrace(); Use tracing_gen_ctx_dec() to account for the preempt_disable_notrace(), which will output the correct preemption count: $ echo ksys_read:stacktrace > set_ftrace_filter <...>-410 [006] ..... 31.420396: => ksys_read => do_syscall_64 => entry_SYSCALL_64_after_hwframe Cc: stable@vger.kernel.org Fixes: 36590c50b2d07 ("tracing: Merge irqflags + preempt counter.") Link: https://lore.kernel.org/20250512094246.1167956-2-dolinux.peng@gmail.com Signed-off-by: pengdonglin Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace_functions.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/kernel/trace/trace_functions.c b/kernel/trace/trace_functions.c index 98ccf3f00c51..4e37a0f6aaa3 100644 --- a/kernel/trace/trace_functions.c +++ b/kernel/trace/trace_functions.c @@ -633,11 +633,7 @@ ftrace_traceoff(unsigned long ip, unsigned long parent_ip, static __always_inline void trace_stack(struct trace_array *tr) { - unsigned int trace_ctx; - - trace_ctx = tracing_gen_ctx(); - - __trace_stack(tr, trace_ctx, FTRACE_STACK_SKIP); + __trace_stack(tr, tracing_gen_ctx_dec(), FTRACE_STACK_SKIP); } static void From 1d6c39c89f617c9fec6bbae166e25b16a014f7c8 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Tue, 13 May 2025 11:50:32 -0400 Subject: [PATCH 509/555] ring-buffer: Fix persistent buffer when commit page is the reader page The ring buffer is made up of sub buffers (sometimes called pages as they are by default PAGE_SIZE). It has the following "pages": "tail page" - this is the page that the next write will write to "head page" - this is the page that the reader will swap the reader page with. "reader page" - This belongs to the reader, where it will swap the head page from the ring buffer so that the reader does not race with the writer. The writer may end up on the "reader page" if the ring buffer hasn't written more than one page, where the "tail page" and the "head page" are the same. The persistent ring buffer has meta data that points to where these pages exist so on reboot it can re-create the pointers to the cpu_buffer descriptor. But when the commit page is on the reader page, the logic is incorrect. The check to see if the commit page is on the reader page checked if the head page was the reader page, which would never happen, as the head page is always in the ring buffer. The correct check would be to test if the commit page is on the reader page. If that's the case, then it can exit out early as the commit page is only on the reader page when there's only one page of data in the buffer. There's no reason to iterate the ring buffer pages to find the "commit page" as it is already found. To trigger this bug: # echo 1 > /sys/kernel/tracing/instances/boot_mapped/events/syscalls/sys_enter_fchownat/enable # touch /tmp/x # chown sshd /tmp/x # reboot On boot up, the dmesg will have: Ring buffer meta [0] is from previous boot! Ring buffer meta [1] is from previous boot! Ring buffer meta [2] is from previous boot! Ring buffer meta [3] is from previous boot! Ring buffer meta [4] commit page not found Ring buffer meta [5] is from previous boot! Ring buffer meta [6] is from previous boot! Ring buffer meta [7] is from previous boot! Where the buffer on CPU 4 had a "commit page not found" error and that buffer is cleared and reset causing the output to be empty and the data lost. When it works correctly, it has: # cat /sys/kernel/tracing/instances/boot_mapped/trace_pipe <...>-1137 [004] ..... 998.205323: sys_enter_fchownat: __syscall_nr=0x104 (260) dfd=0xffffff9c (4294967196) filename=(0xffffc90000a0002c) user=0x3e8 (1000) group=0xffffffff (4294967295) flag=0x0 (0 Cc: stable@vger.kernel.org Cc: Mathieu Desnoyers Link: https://lore.kernel.org/20250513115032.3e0b97f7@gandalf.local.home Fixes: 5f3b6e839f3ce ("ring-buffer: Validate boot range memory events") Reported-by: Tasos Sahanidis Tested-by: Tasos Sahanidis Reviewed-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/ring_buffer.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index c0f877d39a24..3f9bf562beea 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -1887,10 +1887,12 @@ static void rb_meta_validate_events(struct ring_buffer_per_cpu *cpu_buffer) head_page = cpu_buffer->head_page; - /* If both the head and commit are on the reader_page then we are done. */ - if (head_page == cpu_buffer->reader_page && - head_page == cpu_buffer->commit_page) + /* If the commit_buffer is the reader page, update the commit page */ + if (meta->commit_buffer == (unsigned long)cpu_buffer->reader_page->page) { + cpu_buffer->commit_page = cpu_buffer->reader_page; + /* Nothing more to do, the only page is the reader page */ goto done; + } /* Iterate until finding the commit page */ for (i = 0; i < meta->nr_subbufs + 1; i++, rb_inc_page(&head_page)) { From cd52cc3544e400e53e6d4b7bfc5263e7a867b5ab Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 10 May 2025 11:30:21 -0400 Subject: [PATCH 510/555] bcachefs: Don't strip rebalance_opts from indirect extents Fix bch2_bkey_clear_needs_rebalance(): indirect extents are never supposed to have bch_extent_rebalance stripped off, because that's how we get the IO path options when we don't have the original inode it belonged to. Signed-off-by: Kent Overstreet --- fs/bcachefs/rebalance.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/bcachefs/rebalance.c b/fs/bcachefs/rebalance.c index 4ccdfc1f34aa..623273556aa9 100644 --- a/fs/bcachefs/rebalance.c +++ b/fs/bcachefs/rebalance.c @@ -309,7 +309,7 @@ static int bch2_bkey_clear_needs_rebalance(struct btree_trans *trans, struct btree_iter *iter, struct bkey_s_c k) { - if (!bch2_bkey_rebalance_opts(k)) + if (k.k->type == KEY_TYPE_reflink_v || !bch2_bkey_rebalance_opts(k)) return 0; struct bkey_i *n = bch2_bkey_make_mut(trans, iter, &k, 0); From b1c71cb492bb26dc883b11b5bd085d2467508f84 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 10 May 2025 14:42:37 -0400 Subject: [PATCH 511/555] bcachefs: Fix broken btree_path lock invariants in next_node() This fixes btree locking assert pops users were seeing during evacuate: https://github.com/koverstreet/bcachefs/issues/878 May 09 22:45:02 sharon kernel: bcachefs (68116e25-fa2d-4c6f-86c7-e8b431d792ae): bch2_btree_insert_node(): node not locked at level 1 May 09 22:45:02 sharon kernel: bch2_btree_node_rewrite [bcachefs]: watermark=btree no_check_rw alloc l=0-1 mode=none nodes_written=0 cl.remaining=2 journal_seq=0 May 09 22:45:02 sharon kernel: path: idx 1 ref 1:0 S B btree=alloc level=0 pos 0:3699637:0 0:3698012:1-0:3699637:0 bch2_move_btree.isra.0+0x1db/0x490 [bcachefs] uptodate 0 locks_want 2 May 09 22:45:02 sharon kernel: l=0 locks intent seq 4 node ffff8bd700c93600 May 09 22:45:02 sharon kernel: l=1 locks unlocked seq 1712 node ffff8bd6fd5e7a00 May 09 22:45:02 sharon kernel: l=2 locks unlocked seq 2295 node ffff8bd6cc725400 May 09 22:45:02 sharon kernel: l=3 locks unlocked seq 0 node 0000000000000000 Evacuate walks btree nodes with bch2_btree_iter_next_node() and rewrites them, bch2_btree_update_start() upgrades the path to take intent locks as far as it needs to. But next_node() does low level unlock/relock calls on individual nodes, and didn't handle the case where a path is supposed to be holding multiple intent locks. If a path has locks_want > 1, it needs to be either holding locks on all the btree nodes (at each level) requested, or none of them. Fix this with a bch2_btree_path_downgrade(). Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_iter.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/fs/bcachefs/btree_iter.c b/fs/bcachefs/btree_iter.c index 59fa527ac685..9c9bc7b6c712 100644 --- a/fs/bcachefs/btree_iter.c +++ b/fs/bcachefs/btree_iter.c @@ -1971,6 +1971,12 @@ struct btree *bch2_btree_iter_next_node(struct btree_trans *trans, struct btree_ return NULL; } + /* + * We don't correctly handle nodes with extra intent locks here: + * downgrade so we don't violate locking invariants + */ + bch2_btree_path_downgrade(trans, path); + if (!bch2_btree_node_relock(trans, path, path->level + 1)) { __bch2_btree_path_unlock(trans, path); path->l[path->level].b = ERR_PTR(-BCH_ERR_no_btree_node_relock); From 7b6759b1991d427cf9a562b2891b8c3e87a19c76 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 13 May 2025 12:55:44 -0400 Subject: [PATCH 512/555] bcachefs: Fix livelock in journal_entry_open() When the journal is low on space, we might do discards from journal_res_get() -> journal_entry_open(). Make sure we set j->can_discard correctly, so that if we're low on space but not because discards aren't keeping up we don't livelock. Fixes: 8e4d28036c29 ("bcachefs: Don't aggressively discard the journal") Signed-off-by: Kent Overstreet --- fs/bcachefs/journal_reclaim.c | 19 +++++++++++++------ 1 file changed, 13 insertions(+), 6 deletions(-) diff --git a/fs/bcachefs/journal_reclaim.c b/fs/bcachefs/journal_reclaim.c index 976464d8a695..cc00b0fc40d8 100644 --- a/fs/bcachefs/journal_reclaim.c +++ b/fs/bcachefs/journal_reclaim.c @@ -17,6 +17,8 @@ #include #include +static bool __should_discard_bucket(struct journal *, struct journal_device *); + /* Free space calculations: */ static unsigned journal_space_from(struct journal_device *ja, @@ -203,8 +205,7 @@ void bch2_journal_space_available(struct journal *j) ja->bucket_seq[ja->dirty_idx_ondisk] < j->last_seq_ondisk) ja->dirty_idx_ondisk = (ja->dirty_idx_ondisk + 1) % ja->nr; - if (ja->discard_idx != ja->dirty_idx_ondisk) - can_discard = true; + can_discard |= __should_discard_bucket(j, ja); max_entry_size = min_t(unsigned, max_entry_size, ca->mi.bucket_size); nr_online++; @@ -264,13 +265,19 @@ out: /* Discards - last part of journal reclaim: */ +static bool __should_discard_bucket(struct journal *j, struct journal_device *ja) +{ + unsigned min_free = max(4, ja->nr / 8); + + return bch2_journal_dev_buckets_available(j, ja, journal_space_discarded) < + min_free && + ja->discard_idx != ja->dirty_idx_ondisk; +} + static bool should_discard_bucket(struct journal *j, struct journal_device *ja) { spin_lock(&j->lock); - unsigned min_free = max(4, ja->nr / 8); - - bool ret = bch2_journal_dev_buckets_available(j, ja, journal_space_discarded) < min_free && - ja->discard_idx != ja->dirty_idx_ondisk; + bool ret = __should_discard_bucket(j, ja); spin_unlock(&j->lock); return ret; From 19b22d04cd44ded1ea5af7849aec9cbf4021c852 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 13 May 2025 14:27:01 -0400 Subject: [PATCH 513/555] bcachefs: Don't set btree nodes as accessed on fill Prevent jobs that do lots of scanning (i.e. evacuatee, scrub) from causing OOMs. The shrinker code seems to be having issues when it doesn't do any freeing because it's just flipping off the acccessed bit - and the accessed bit shouldn't be set on first use anyways. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_cache.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/fs/bcachefs/btree_cache.c b/fs/bcachefs/btree_cache.c index 9b80201c7982..899891295797 100644 --- a/fs/bcachefs/btree_cache.c +++ b/fs/bcachefs/btree_cache.c @@ -852,7 +852,6 @@ out: b->sib_u64s[1] = 0; b->whiteout_u64s = 0; bch2_btree_keys_init(b); - set_btree_node_accessed(b); bch2_time_stats_update(&c->times[BCH_TIME_btree_node_mem_alloc], start_time); @@ -1286,6 +1285,10 @@ lock_node: six_unlock_read(&b->c.lock); goto retry; } + + /* avoid atomic set bit if it's not needed: */ + if (!btree_node_accessed(b)) + set_btree_node_accessed(b); } /* XXX: waiting on IO with btree locks held: */ @@ -1301,10 +1304,6 @@ lock_node: prefetch(p + L1_CACHE_BYTES * 2); } - /* avoid atomic set bit if it's not needed: */ - if (!btree_node_accessed(b)) - set_btree_node_accessed(b); - if (unlikely(btree_node_read_error(b))) { six_unlock_read(&b->c.lock); b = ERR_PTR(-BCH_ERR_btree_node_read_err_cached); From 61198e62878477926c9e986eaf6bc40aafb0bf64 Mon Sep 17 00:00:00 2001 From: Alan Huang Date: Tue, 13 May 2025 18:54:26 +0800 Subject: [PATCH 514/555] bcachefs: Fix self deadlock Before invoking bch2_accounting_mem_mod_locked in bch2_gc_accounting_done, we already write locked mark_lock, in bch2_accounting_mem_insert, we lock mark_lock again. Signed-off-by: Alan Huang Signed-off-by: Kent Overstreet --- fs/bcachefs/disk_accounting.c | 17 +++++++++++++++-- fs/bcachefs/disk_accounting.h | 16 +++++++++++----- 2 files changed, 26 insertions(+), 7 deletions(-) diff --git a/fs/bcachefs/disk_accounting.c b/fs/bcachefs/disk_accounting.c index b007319b72e9..1f0422bfae35 100644 --- a/fs/bcachefs/disk_accounting.c +++ b/fs/bcachefs/disk_accounting.c @@ -376,6 +376,19 @@ int bch2_accounting_mem_insert(struct bch_fs *c, struct bkey_s_c_accounting a, return ret; } +int bch2_accounting_mem_insert_locked(struct bch_fs *c, struct bkey_s_c_accounting a, + enum bch_accounting_mode mode) +{ + struct bch_replicas_padded r; + + if (mode != BCH_ACCOUNTING_read && + accounting_to_replicas(&r.e, a.k->p) && + !bch2_replicas_marked_locked(c, &r.e)) + return -BCH_ERR_btree_insert_need_mark_replicas; + + return __bch2_accounting_mem_insert(c, a); +} + static bool accounting_mem_entry_is_zero(struct accounting_mem_entry *e) { for (unsigned i = 0; i < e->nr_counters; i++) @@ -583,7 +596,7 @@ int bch2_gc_accounting_done(struct bch_fs *c) accounting_key_init(&k_i.k, &acc_k, src_v, nr); bch2_accounting_mem_mod_locked(trans, bkey_i_to_s_c_accounting(&k_i.k), - BCH_ACCOUNTING_normal); + BCH_ACCOUNTING_normal, true); preempt_disable(); struct bch_fs_usage_base *dst = this_cpu_ptr(c->usage); @@ -612,7 +625,7 @@ static int accounting_read_key(struct btree_trans *trans, struct bkey_s_c k) percpu_down_read(&c->mark_lock); int ret = bch2_accounting_mem_mod_locked(trans, bkey_s_c_to_accounting(k), - BCH_ACCOUNTING_read); + BCH_ACCOUNTING_read, false); percpu_up_read(&c->mark_lock); return ret; } diff --git a/fs/bcachefs/disk_accounting.h b/fs/bcachefs/disk_accounting.h index abb1f6206fe9..d557b99b3c0a 100644 --- a/fs/bcachefs/disk_accounting.h +++ b/fs/bcachefs/disk_accounting.h @@ -136,6 +136,7 @@ enum bch_accounting_mode { }; int bch2_accounting_mem_insert(struct bch_fs *, struct bkey_s_c_accounting, enum bch_accounting_mode); +int bch2_accounting_mem_insert_locked(struct bch_fs *, struct bkey_s_c_accounting, enum bch_accounting_mode); void bch2_accounting_mem_gc(struct bch_fs *); static inline bool bch2_accounting_is_mem(struct disk_accounting_pos acc) @@ -150,7 +151,8 @@ static inline bool bch2_accounting_is_mem(struct disk_accounting_pos acc) */ static inline int bch2_accounting_mem_mod_locked(struct btree_trans *trans, struct bkey_s_c_accounting a, - enum bch_accounting_mode mode) + enum bch_accounting_mode mode, + bool write_locked) { struct bch_fs *c = trans->c; struct bch_accounting_mem *acc = &c->accounting; @@ -189,7 +191,11 @@ static inline int bch2_accounting_mem_mod_locked(struct btree_trans *trans, while ((idx = eytzinger0_find(acc->k.data, acc->k.nr, sizeof(acc->k.data[0]), accounting_pos_cmp, &a.k->p)) >= acc->k.nr) { - int ret = bch2_accounting_mem_insert(c, a, mode); + int ret = 0; + if (unlikely(write_locked)) + ret = bch2_accounting_mem_insert_locked(c, a, mode); + else + ret = bch2_accounting_mem_insert(c, a, mode); if (ret) return ret; } @@ -206,7 +212,7 @@ static inline int bch2_accounting_mem_mod_locked(struct btree_trans *trans, static inline int bch2_accounting_mem_add(struct btree_trans *trans, struct bkey_s_c_accounting a, bool gc) { percpu_down_read(&trans->c->mark_lock); - int ret = bch2_accounting_mem_mod_locked(trans, a, gc ? BCH_ACCOUNTING_gc : BCH_ACCOUNTING_normal); + int ret = bch2_accounting_mem_mod_locked(trans, a, gc ? BCH_ACCOUNTING_gc : BCH_ACCOUNTING_normal, false); percpu_up_read(&trans->c->mark_lock); return ret; } @@ -259,7 +265,7 @@ static inline int bch2_accounting_trans_commit_hook(struct btree_trans *trans, EBUG_ON(bversion_zero(a->k.bversion)); return likely(!(commit_flags & BCH_TRANS_COMMIT_skip_accounting_apply)) - ? bch2_accounting_mem_mod_locked(trans, accounting_i_to_s_c(a), BCH_ACCOUNTING_normal) + ? bch2_accounting_mem_mod_locked(trans, accounting_i_to_s_c(a), BCH_ACCOUNTING_normal, false) : 0; } @@ -271,7 +277,7 @@ static inline void bch2_accounting_trans_commit_revert(struct btree_trans *trans struct bkey_s_accounting a = accounting_i_to_s(a_i); bch2_accounting_neg(a); - bch2_accounting_mem_mod_locked(trans, a.c, BCH_ACCOUNTING_normal); + bch2_accounting_mem_mod_locked(trans, a.c, BCH_ACCOUNTING_normal, false); bch2_accounting_neg(a); } } From 43b9fece2d9687cde58cc7eec4548dd1c35e2198 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Tue, 13 May 2025 21:14:17 -0400 Subject: [PATCH 515/555] bcachefs: Fix set_should_be_locked() call in peek_slot() set_should_be_locked() needs to be called before peek_key_cache(), which traverses other paths and may do a trans unlock/relock. This fixes an assertion pop in path_peek_slot(), when the path we're using is unexpectedly not uptodate. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_iter.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/fs/bcachefs/btree_iter.c b/fs/bcachefs/btree_iter.c index 9c9bc7b6c712..a873ec1baf58 100644 --- a/fs/bcachefs/btree_iter.c +++ b/fs/bcachefs/btree_iter.c @@ -2749,7 +2749,7 @@ struct bkey_s_c bch2_btree_iter_peek_slot(struct btree_trans *trans, struct btre ret = trans_maybe_inject_restart(trans, _RET_IP_); if (unlikely(ret)) { k = bkey_s_c_err(ret); - goto out_no_locked; + goto out; } /* extents can't span inode numbers: */ @@ -2769,13 +2769,15 @@ struct bkey_s_c bch2_btree_iter_peek_slot(struct btree_trans *trans, struct btre ret = bch2_btree_path_traverse(trans, iter->path, iter->flags); if (unlikely(ret)) { k = bkey_s_c_err(ret); - goto out_no_locked; + goto out; } struct btree_path *path = btree_iter_path(trans, iter); if (unlikely(!btree_path_node(path, path->level))) return bkey_s_c_null; + btree_path_set_should_be_locked(trans, path); + if ((iter->flags & BTREE_ITER_cached) || !(iter->flags & (BTREE_ITER_is_extents|BTREE_ITER_filter_snapshots))) { k = bkey_s_c_null; @@ -2796,12 +2798,12 @@ struct bkey_s_c bch2_btree_iter_peek_slot(struct btree_trans *trans, struct btre if (!bkey_err(k)) iter->k = *k.k; /* We're not returning a key from iter->path: */ - goto out_no_locked; + goto out; } - k = bch2_btree_path_peek_slot(trans->paths + iter->path, &iter->k); + k = bch2_btree_path_peek_slot(btree_iter_path(trans, iter), &iter->k); if (unlikely(!k.k)) - goto out_no_locked; + goto out; if (unlikely(k.k->type == KEY_TYPE_whiteout && (iter->flags & BTREE_ITER_filter_snapshots) && @@ -2839,7 +2841,7 @@ struct bkey_s_c bch2_btree_iter_peek_slot(struct btree_trans *trans, struct btre } if (unlikely(bkey_err(k))) - goto out_no_locked; + goto out; next = k.k ? bkey_start_pos(k.k) : POS_MAX; @@ -2861,8 +2863,6 @@ struct bkey_s_c bch2_btree_iter_peek_slot(struct btree_trans *trans, struct btre } } out: - btree_path_set_should_be_locked(trans, btree_iter_path(trans, iter)); -out_no_locked: bch2_btree_iter_verify_entry_exit(iter); bch2_btree_iter_verify(trans, iter); ret = bch2_btree_iter_verify_ret(trans, iter, k); From a12cb6f758177fcf9bdb7d18b4724eaa29023740 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 14 May 2025 13:40:47 -0400 Subject: [PATCH 516/555] bcachefs: Fix accidental O(n^2) in fiemap MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Since bch2_seek_pagecache_data() searches for dirty data, we only want to call it for holes in the extents btree - otherwise we have an accidental O(n^2), as we repeatedly search the same range. Reported-by: Marcin Mirosław Signed-off-by: Kent Overstreet --- fs/bcachefs/fs.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/fs/bcachefs/fs.c b/fs/bcachefs/fs.c index b6801861c66f..4b742e62255b 100644 --- a/fs/bcachefs/fs.c +++ b/fs/bcachefs/fs.c @@ -1429,7 +1429,9 @@ static int bch2_next_fiemap_extent(struct btree_trans *trans, if (ret) goto err; - ret = bch2_next_fiemap_pagecache_extent(trans, inode, start, end, cur); + u64 pagecache_end = k.k ? max(start, bkey_start_offset(k.k)) : end; + + ret = bch2_next_fiemap_pagecache_extent(trans, inode, start, pagecache_end, cur); if (ret) goto err; From d1041d8eab31f9b4c696ee060141d1b0d156f84d Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Fri, 9 May 2025 15:05:19 -0400 Subject: [PATCH 517/555] bcachefs: Fix missing commit in backpointer to missing target Fsck wants to do transaction commits from an outer context; it may have other repair to do (i.e. duplicate backpointers). But when calling backpointer_not_found() from runtime code, i.e. runtime self healing, we should be doing the commit - the outer context expects to just be doing lookups. This fixes bugs where we get stuck spinning, reported as "RCU lock hold time warnings. Signed-off-by: Kent Overstreet --- fs/bcachefs/backpointers.c | 117 +++++++++++++++++++++++++------------ 1 file changed, 79 insertions(+), 38 deletions(-) diff --git a/fs/bcachefs/backpointers.c b/fs/bcachefs/backpointers.c index ff26bb515150..5f195d2280a4 100644 --- a/fs/bcachefs/backpointers.c +++ b/fs/bcachefs/backpointers.c @@ -192,7 +192,8 @@ static inline int bch2_backpointers_maybe_flush(struct btree_trans *trans, static int backpointer_target_not_found(struct btree_trans *trans, struct bkey_s_c_backpointer bp, struct bkey_s_c target_k, - struct bkey_buf *last_flushed) + struct bkey_buf *last_flushed, + bool commit) { struct bch_fs *c = trans->c; struct printbuf buf = PRINTBUF; @@ -228,18 +229,77 @@ static int backpointer_target_not_found(struct btree_trans *trans, } if (fsck_err(trans, backpointer_to_missing_ptr, - "%s", buf.buf)) + "%s", buf.buf)) { ret = bch2_backpointer_del(trans, bp.k->p); + if (ret || !commit) + goto out; + + /* + * Normally, on transaction commit from inside a transaction, + * we'll return -BCH_ERR_transaction_restart_nested, since a + * transaction commit invalidates pointers given out by peek(). + * + * However, since we're updating a write buffer btree, if we + * return a transaction restart and loop we won't see that the + * backpointer has been deleted without an additional write + * buffer flush - and those are expensive. + * + * So we're relying on the caller immediately advancing to the + * next backpointer and starting a new transaction immediately + * after backpointer_get_key() returns NULL: + */ + ret = bch2_trans_commit(trans, NULL, NULL, BCH_TRANS_COMMIT_no_enospc); + } +out: fsck_err: printbuf_exit(&buf); return ret; } -struct bkey_s_c bch2_backpointer_get_key(struct btree_trans *trans, - struct bkey_s_c_backpointer bp, - struct btree_iter *iter, - unsigned iter_flags, - struct bkey_buf *last_flushed) +static struct btree *__bch2_backpointer_get_node(struct btree_trans *trans, + struct bkey_s_c_backpointer bp, + struct btree_iter *iter, + struct bkey_buf *last_flushed, + bool commit) +{ + struct bch_fs *c = trans->c; + + BUG_ON(!bp.v->level); + + bch2_trans_node_iter_init(trans, iter, + bp.v->btree_id, + bp.v->pos, + 0, + bp.v->level - 1, + 0); + struct btree *b = bch2_btree_iter_peek_node(trans, iter); + if (IS_ERR_OR_NULL(b)) + goto err; + + BUG_ON(b->c.level != bp.v->level - 1); + + if (extent_matches_bp(c, bp.v->btree_id, bp.v->level, + bkey_i_to_s_c(&b->key), bp)) + return b; + + if (btree_node_will_make_reachable(b)) { + b = ERR_PTR(-BCH_ERR_backpointer_to_overwritten_btree_node); + } else { + int ret = backpointer_target_not_found(trans, bp, bkey_i_to_s_c(&b->key), + last_flushed, commit); + b = ret ? ERR_PTR(ret) : NULL; + } +err: + bch2_trans_iter_exit(trans, iter); + return b; +} + +static struct bkey_s_c __bch2_backpointer_get_key(struct btree_trans *trans, + struct bkey_s_c_backpointer bp, + struct btree_iter *iter, + unsigned iter_flags, + struct bkey_buf *last_flushed, + bool commit) { struct bch_fs *c = trans->c; @@ -277,10 +337,10 @@ struct bkey_s_c bch2_backpointer_get_key(struct btree_trans *trans, bch2_trans_iter_exit(trans, iter); if (!bp.v->level) { - int ret = backpointer_target_not_found(trans, bp, k, last_flushed); + int ret = backpointer_target_not_found(trans, bp, k, last_flushed, commit); return ret ? bkey_s_c_err(ret) : bkey_s_c_null; } else { - struct btree *b = bch2_backpointer_get_node(trans, bp, iter, last_flushed); + struct btree *b = __bch2_backpointer_get_node(trans, bp, iter, last_flushed, commit); if (b == ERR_PTR(-BCH_ERR_backpointer_to_overwritten_btree_node)) return bkey_s_c_null; if (IS_ERR_OR_NULL(b)) @@ -295,35 +355,16 @@ struct btree *bch2_backpointer_get_node(struct btree_trans *trans, struct btree_iter *iter, struct bkey_buf *last_flushed) { - struct bch_fs *c = trans->c; + return __bch2_backpointer_get_node(trans, bp, iter, last_flushed, true); +} - BUG_ON(!bp.v->level); - - bch2_trans_node_iter_init(trans, iter, - bp.v->btree_id, - bp.v->pos, - 0, - bp.v->level - 1, - 0); - struct btree *b = bch2_btree_iter_peek_node(trans, iter); - if (IS_ERR_OR_NULL(b)) - goto err; - - BUG_ON(b->c.level != bp.v->level - 1); - - if (extent_matches_bp(c, bp.v->btree_id, bp.v->level, - bkey_i_to_s_c(&b->key), bp)) - return b; - - if (btree_node_will_make_reachable(b)) { - b = ERR_PTR(-BCH_ERR_backpointer_to_overwritten_btree_node); - } else { - int ret = backpointer_target_not_found(trans, bp, bkey_i_to_s_c(&b->key), last_flushed); - b = ret ? ERR_PTR(ret) : NULL; - } -err: - bch2_trans_iter_exit(trans, iter); - return b; +struct bkey_s_c bch2_backpointer_get_key(struct btree_trans *trans, + struct bkey_s_c_backpointer bp, + struct btree_iter *iter, + unsigned iter_flags, + struct bkey_buf *last_flushed) +{ + return __bch2_backpointer_get_key(trans, bp, iter, iter_flags, last_flushed, true); } static int bch2_check_backpointer_has_valid_bucket(struct btree_trans *trans, struct bkey_s_c k, @@ -521,7 +562,7 @@ check_existing_bp: struct bkey_s_c_backpointer other_bp = bkey_s_c_to_backpointer(bp_k); struct bkey_s_c other_extent = - bch2_backpointer_get_key(trans, other_bp, &other_extent_iter, 0, NULL); + __bch2_backpointer_get_key(trans, other_bp, &other_extent_iter, 0, NULL, false); ret = bkey_err(other_extent); if (ret == -BCH_ERR_backpointer_to_overwritten_btree_node) ret = 0; From 9c09e59cc55cdf7feb29971fd792fc1947010b79 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Wed, 14 May 2025 18:53:48 -0400 Subject: [PATCH 518/555] bcachefs: fix wrong arg to fsck_err() fsck_err() needs the btree transaction passed to it if there is one - so that it can unlock/relock around prompting userspace for fixing the error. Signed-off-by: Kent Overstreet --- fs/bcachefs/fsck.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/bcachefs/fsck.c b/fs/bcachefs/fsck.c index 7b25cedd3e40..71d428f376a5 100644 --- a/fs/bcachefs/fsck.c +++ b/fs/bcachefs/fsck.c @@ -2446,7 +2446,7 @@ static int check_subvol_path(struct btree_trans *trans, struct btree_iter *iter, u32 parent = le32_to_cpu(s.v->fs_path_parent); if (darray_u32_has(&subvol_path, parent)) { - if (fsck_err(c, subvol_loop, "subvolume loop")) + if (fsck_err(trans, subvol_loop, "subvolume loop")) ret = reattach_subvol(trans, s); break; } From 1fe4a44b7fa3955bcb7b4067c07b778fe90d8ee7 Mon Sep 17 00:00:00 2001 From: Jethro Donaldson Date: Thu, 15 May 2025 01:23:23 +1200 Subject: [PATCH 519/555] smb: client: fix memory leak during error handling for POSIX mkdir The response buffer for the CREATE request handled by smb311_posix_mkdir() is leaked on the error path (goto err_free_rsp_buf) because the structure pointer *rsp passed to free_rsp_buf() is not assigned until *after* the error condition is checked. As *rsp is initialised to NULL, free_rsp_buf() becomes a no-op and the leak is instead reported by __kmem_cache_shutdown() upon subsequent rmmod of cifs.ko if (and only if) the error path has been hit. Pass rsp_iov.iov_base to free_rsp_buf() instead, similar to the code in other functions in smb2pdu.c for which *rsp is assigned late. Cc: stable@vger.kernel.org Signed-off-by: Jethro Donaldson Signed-off-by: Steve French --- fs/smb/client/smb2pdu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/smb/client/smb2pdu.c b/fs/smb/client/smb2pdu.c index 0b35816d551f..4e28632b5fd6 100644 --- a/fs/smb/client/smb2pdu.c +++ b/fs/smb/client/smb2pdu.c @@ -2968,7 +2968,7 @@ replay_again: /* Eventually save off posix specific response info and timestamps */ err_free_rsp_buf: - free_rsp_buf(resp_buftype, rsp); + free_rsp_buf(resp_buftype, rsp_iov.iov_base); kfree(pc_buf); err_free_req: cifs_small_buf_release(req); From 3965c23773e81c476f6de30ccc5d201c59ff9714 Mon Sep 17 00:00:00 2001 From: Paulo Alcantara Date: Mon, 12 May 2025 14:58:36 -0300 Subject: [PATCH 520/555] smb: client: fix zero rsize error messages cifs_prepare_read() might be called with a disconnected channel, where TCP_Server_Info::max_read is set to zero due to reconnect, so calling ->negotiate_rize() will set @rsize to default min IO size (64KiB) and then logging CIFS: VFS: SMB: Zero rsize calculated, using minimum value 65536 If the reconnect happens in cifsd thread, cifs_renegotiate_iosize() will end up being called and then @rsize set to the expected value. Since we can't rely on the value of @server->max_read by the time we call cifs_prepare_read(), try to ->negotiate_rize() only if @cifs_sb->ctx->rsize is zero. Reported-by: Steve French Fixes: c59f7c9661b9 ("smb: client: ensure aligned IO sizes") Signed-off-by: Paulo Alcantara (Red Hat) Signed-off-by: Steve French --- fs/smb/client/file.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c index 851b74f557c1..950aa4f912f5 100644 --- a/fs/smb/client/file.c +++ b/fs/smb/client/file.c @@ -160,8 +160,10 @@ static int cifs_prepare_read(struct netfs_io_subrequest *subreq) server = cifs_pick_channel(tlink_tcon(req->cfile->tlink)->ses); rdata->server = server; - cifs_negotiate_rsize(server, cifs_sb->ctx, - tlink_tcon(req->cfile->tlink)); + if (cifs_sb->ctx->rsize == 0) { + cifs_negotiate_rsize(server, cifs_sb->ctx, + tlink_tcon(req->cfile->tlink)); + } rc = server->ops->wait_mtu_credits(server, cifs_sb->ctx->rsize, &size, &rdata->credits); From 539fbab37881e32ba6a708a100de6db19e1e7e7d Mon Sep 17 00:00:00 2001 From: Jarkko Sakkinen Date: Mon, 7 Apr 2025 15:28:05 +0300 Subject: [PATCH 521/555] tpm: Mask TPM RC in tpm2_start_auth_session() tpm2_start_auth_session() does not mask TPM RC correctly from the callers: [ 28.766528] tpm tpm0: A TPM error (2307) occurred start auth session Process TPM RCs inside tpm2_start_auth_session(), and map them to POSIX error codes. Cc: stable@vger.kernel.org # v6.10+ Fixes: 699e3efd6c64 ("tpm: Add HMAC session start and end functions") Reported-by: Herbert Xu Closes: https://lore.kernel.org/linux-integrity/Z_NgdRHuTKP6JK--@gondor.apana.org.au/ Reviewed-by: Stefano Garzarella Signed-off-by: Jarkko Sakkinen --- drivers/char/tpm/tpm2-sessions.c | 20 ++++++-------------- include/linux/tpm.h | 19 +++++++++++++++++++ 2 files changed, 25 insertions(+), 14 deletions(-) diff --git a/drivers/char/tpm/tpm2-sessions.c b/drivers/char/tpm/tpm2-sessions.c index 3f89635ba5e8..7b5049b3d476 100644 --- a/drivers/char/tpm/tpm2-sessions.c +++ b/drivers/char/tpm/tpm2-sessions.c @@ -40,11 +40,6 @@ * * These are the usage functions: * - * tpm2_start_auth_session() which allocates the opaque auth structure - * and gets a session from the TPM. This must be called before - * any of the following functions. The session is protected by a - * session_key which is derived from a random salt value - * encrypted to the NULL seed. * tpm2_end_auth_session() kills the session and frees the resources. * Under normal operation this function is done by * tpm_buf_check_hmac_response(), so this is only to be used on @@ -963,16 +958,13 @@ err: } /** - * tpm2_start_auth_session() - create a HMAC authentication session with the TPM - * @chip: the TPM chip structure to create the session with + * tpm2_start_auth_session() - Create an a HMAC authentication session + * @chip: A TPM chip * - * This function loads the NULL seed from its saved context and starts - * an authentication session on the null seed, fills in the - * @chip->auth structure to contain all the session details necessary - * for performing the HMAC, encrypt and decrypt operations and - * returns. The NULL seed is flushed before this function returns. + * Loads the ephemeral key (null seed), and starts an HMAC authenticated + * session. The null seed is flushed before the return. * - * Return: zero on success or actual error encountered. + * Returns zero on success, or a POSIX error code. */ int tpm2_start_auth_session(struct tpm_chip *chip) { @@ -1024,7 +1016,7 @@ int tpm2_start_auth_session(struct tpm_chip *chip) /* hash algorithm for session */ tpm_buf_append_u16(&buf, TPM_ALG_SHA256); - rc = tpm_transmit_cmd(chip, &buf, 0, "start auth session"); + rc = tpm_ret_to_err(tpm_transmit_cmd(chip, &buf, 0, "StartAuthSession")); tpm2_flush_context(chip, null_key); if (rc == TPM2_RC_SUCCESS) diff --git a/include/linux/tpm.h b/include/linux/tpm.h index 6c3125300c00..9ac9768cc8f7 100644 --- a/include/linux/tpm.h +++ b/include/linux/tpm.h @@ -257,6 +257,7 @@ enum tpm2_return_codes { TPM2_RC_TESTING = 0x090A, /* RC_WARN */ TPM2_RC_REFERENCE_H0 = 0x0910, TPM2_RC_RETRY = 0x0922, + TPM2_RC_SESSION_MEMORY = 0x0903, }; enum tpm2_command_codes { @@ -437,6 +438,24 @@ static inline u32 tpm2_rc_value(u32 rc) return (rc & BIT(7)) ? rc & 0xbf : rc; } +/* + * Convert a return value from tpm_transmit_cmd() to POSIX error code. + */ +static inline ssize_t tpm_ret_to_err(ssize_t ret) +{ + if (ret < 0) + return ret; + + switch (tpm2_rc_value(ret)) { + case TPM2_RC_SUCCESS: + return 0; + case TPM2_RC_SESSION_MEMORY: + return -ENOMEM; + default: + return -EFAULT; + } +} + #if defined(CONFIG_TCG_TPM) || defined(CONFIG_TCG_TPM_MODULE) extern int tpm_is_tpm2(struct tpm_chip *chip); From 32d495b384a2db7d23c2295e03e6b6edb1c0db8d Mon Sep 17 00:00:00 2001 From: Purva Yeshi Date: Thu, 10 Apr 2025 16:04:42 +0530 Subject: [PATCH 522/555] char: tpm: tpm-buf: Add sanity check fallback in read helpers Fix Smatch-detected issue: drivers/char/tpm/tpm-buf.c:208 tpm_buf_read_u8() error: uninitialized symbol 'value'. drivers/char/tpm/tpm-buf.c:225 tpm_buf_read_u16() error: uninitialized symbol 'value'. drivers/char/tpm/tpm-buf.c:242 tpm_buf_read_u32() error: uninitialized symbol 'value'. Zero-initialize the return values in tpm_buf_read_u8(), tpm_buf_read_u16(), and tpm_buf_read_u32() to guard against uninitialized data in case of a boundary overflow. Add defensive initialization ensures the return values are always defined, preventing undefined behavior if the unexpected happens. Signed-off-by: Purva Yeshi Reviewed-by: Stefano Garzarella Reviewed-by: Jarkko Sakkinen Signed-off-by: Jarkko Sakkinen --- drivers/char/tpm/tpm-buf.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/char/tpm/tpm-buf.c b/drivers/char/tpm/tpm-buf.c index e49a19fea3bd..dc882fc9fa9e 100644 --- a/drivers/char/tpm/tpm-buf.c +++ b/drivers/char/tpm/tpm-buf.c @@ -201,7 +201,7 @@ static void tpm_buf_read(struct tpm_buf *buf, off_t *offset, size_t count, void */ u8 tpm_buf_read_u8(struct tpm_buf *buf, off_t *offset) { - u8 value; + u8 value = 0; tpm_buf_read(buf, offset, sizeof(value), &value); @@ -218,7 +218,7 @@ EXPORT_SYMBOL_GPL(tpm_buf_read_u8); */ u16 tpm_buf_read_u16(struct tpm_buf *buf, off_t *offset) { - u16 value; + u16 value = 0; tpm_buf_read(buf, offset, sizeof(value), &value); @@ -235,7 +235,7 @@ EXPORT_SYMBOL_GPL(tpm_buf_read_u16); */ u32 tpm_buf_read_u32(struct tpm_buf *buf, off_t *offset) { - u32 value; + u32 value = 0; tpm_buf_read(buf, offset, sizeof(value), &value); From 2f661f71fda1fc0c42b7746ca5b7da529eb6b5be Mon Sep 17 00:00:00 2001 From: Michal Suchanek Date: Fri, 4 Apr 2025 10:23:14 +0200 Subject: [PATCH 523/555] tpm: tis: Double the timeout B to 4s With some Infineon chips the timeouts in tpm_tis_send_data (both B and C) can reach up to about 2250 ms. Timeout C is retried since commit de9e33df7762 ("tpm, tpm_tis: Workaround failed command reception on Infineon devices") Timeout B still needs to be extended. The problem is most commonly encountered with context related operation such as load context/save context. These are issued directly by the kernel, and there is no retry logic for them. When a filesystem is set up to use the TPM for unlocking the boot fails, and restarting the userspace service is ineffective. This is likely because ignoring a load context/save context result puts the real TPM state and the TPM state expected by the kernel out of sync. Chips known to be affected: tpm_tis IFX1522:00: 2.0 TPM (device-id 0x1D, rev-id 54) Description: SLB9672 Firmware Revision: 15.22 tpm_tis MSFT0101:00: 2.0 TPM (device-id 0x1B, rev-id 22) Firmware Revision: 7.83 tpm_tis MSFT0101:00: 2.0 TPM (device-id 0x1A, rev-id 16) Firmware Revision: 5.63 Link: https://lore.kernel.org/linux-integrity/Z5pI07m0Muapyu9w@kitsune.suse.cz/ Signed-off-by: Michal Suchanek Reviewed-by: Jarkko Sakkinen Signed-off-by: Jarkko Sakkinen --- drivers/char/tpm/tpm_tis_core.h | 2 +- include/linux/tpm.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/char/tpm/tpm_tis_core.h b/drivers/char/tpm/tpm_tis_core.h index 970d02c337c7..6c3aa480396b 100644 --- a/drivers/char/tpm/tpm_tis_core.h +++ b/drivers/char/tpm/tpm_tis_core.h @@ -54,7 +54,7 @@ enum tis_int_flags { enum tis_defaults { TIS_MEM_LEN = 0x5000, TIS_SHORT_TIMEOUT = 750, /* ms */ - TIS_LONG_TIMEOUT = 2000, /* 2 sec */ + TIS_LONG_TIMEOUT = 4000, /* 4 secs */ TIS_TIMEOUT_MIN_ATML = 14700, /* usecs */ TIS_TIMEOUT_MAX_ATML = 15000, /* usecs */ }; diff --git a/include/linux/tpm.h b/include/linux/tpm.h index 9ac9768cc8f7..a3d8305e88a5 100644 --- a/include/linux/tpm.h +++ b/include/linux/tpm.h @@ -224,7 +224,7 @@ enum tpm2_const { enum tpm2_timeouts { TPM2_TIMEOUT_A = 750, - TPM2_TIMEOUT_B = 2000, + TPM2_TIMEOUT_B = 4000, TPM2_TIMEOUT_C = 200, TPM2_TIMEOUT_D = 30, TPM2_DURATION_SHORT = 20, From a9fb87b8b86918e34ef6bf3316311f41bc1a5b1f Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 13 May 2025 15:13:16 -0700 Subject: [PATCH 524/555] netlink: specs: tc: fix a couple of attribute names Fix up spelling of two attribute names. These are clearly typoes and will prevent C codegen from working. Let's treat this as a fix to get the correction into users' hands ASAP, and prevent anyone depending on the wrong names. Fixes: a1bcfde83669 ("doc/netlink/specs: Add a spec for tc") Link: https://patch.msgid.link/20250513221316.841700-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- Documentation/netlink/specs/tc.yaml | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/Documentation/netlink/specs/tc.yaml b/Documentation/netlink/specs/tc.yaml index aacccea5dfe4..5e1ff04f51f2 100644 --- a/Documentation/netlink/specs/tc.yaml +++ b/Documentation/netlink/specs/tc.yaml @@ -2745,7 +2745,7 @@ attribute-sets: type: u16 byte-order: big-endian - - name: key-l2-tpv3-sid + name: key-l2tpv3-sid type: u32 byte-order: big-endian - @@ -3504,7 +3504,7 @@ attribute-sets: name: rate64 type: u64 - - name: prate4 + name: prate64 type: u64 - name: burst From f3dd5fb2fa494dcbdb10f8d27f2deac8ef61a2fc Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 13 May 2025 15:16:38 -0700 Subject: [PATCH 525/555] netlink: specs: tc: all actions are indexed arrays Some TC filters have actions listed as indexed arrays of nests and some as just nests. They are all indexed arrays, the handling is common across filters. Fixes: 2267672a6190 ("doc/netlink/specs: Update the tc spec") Link: https://patch.msgid.link/20250513221638.842532-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- Documentation/netlink/specs/tc.yaml | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/Documentation/netlink/specs/tc.yaml b/Documentation/netlink/specs/tc.yaml index 5e1ff04f51f2..953aa837958b 100644 --- a/Documentation/netlink/specs/tc.yaml +++ b/Documentation/netlink/specs/tc.yaml @@ -2017,7 +2017,8 @@ attribute-sets: attributes: - name: act - type: nest + type: indexed-array + sub-type: nest nested-attributes: tc-act-attrs - name: police @@ -2250,7 +2251,8 @@ attribute-sets: attributes: - name: act - type: nest + type: indexed-array + sub-type: nest nested-attributes: tc-act-attrs - name: police From 865ab2461375e3a5a2526f91f9a9f17b8931bc9e Mon Sep 17 00:00:00 2001 From: Subbaraya Sundeep Date: Mon, 12 May 2025 18:12:36 +0530 Subject: [PATCH 526/555] octeontx2-pf: macsec: Fix incorrect max transmit size in TX secy MASCEC hardware block has a field called maximum transmit size for TX secy. Max packet size going out of MCS block has be programmed taking into account full packet size which has L2 header,SecTag and ICV. MACSEC offload driver is configuring max transmit size as macsec interface MTU which is incorrect. Say with 1500 MTU of real device, macsec interface created on top of real device will have MTU of 1468(1500 - (SecTag + ICV)). This is causing packets from macsec interface of size greater than or equal to 1468 are not getting transmitted out because driver programmed max transmit size as 1468 instead of 1514(1500 + ETH_HDR_LEN). Fixes: c54ffc73601c ("octeontx2-pf: mcs: Introduce MACSEC hardware offloading") Signed-off-by: Subbaraya Sundeep Reviewed-by: Simon Horman Link: https://patch.msgid.link/1747053756-4529-1-git-send-email-sbhatta@marvell.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/marvell/octeontx2/nic/cn10k_macsec.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/cn10k_macsec.c b/drivers/net/ethernet/marvell/octeontx2/nic/cn10k_macsec.c index f3b9daffaec3..4c7e0f345cb5 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/cn10k_macsec.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/cn10k_macsec.c @@ -531,7 +531,8 @@ static int cn10k_mcs_write_tx_secy(struct otx2_nic *pfvf, if (sw_tx_sc->encrypt) sectag_tci |= (MCS_TCI_E | MCS_TCI_C); - policy = FIELD_PREP(MCS_TX_SECY_PLCY_MTU, secy->netdev->mtu); + policy = FIELD_PREP(MCS_TX_SECY_PLCY_MTU, + pfvf->netdev->mtu + OTX2_ETH_HLEN); /* Write SecTag excluding AN bits(1..0) */ policy |= FIELD_PREP(MCS_TX_SECY_PLCY_ST_TCI, sectag_tci >> 2); policy |= FIELD_PREP(MCS_TX_SECY_PLCY_ST_OFFSET, tag_offset); From 141a8dec88ba257429965f163dd16e005f1a9718 Mon Sep 17 00:00:00 2001 From: Jiawen Wu Date: Tue, 13 May 2025 10:10:07 +0800 Subject: [PATCH 527/555] net: txgbe: Fix to calculate EEPROM checksum for AML devices In the new firmware version, the shadow ram reserves some space to store I2C information, so the checksum calculation needs to skip this section. Otherwise, the driver will fail to probe because the invalid EEPROM checksum. Fixes: 2e5af6b2ae85 ("net: txgbe: Add basic support for new AML devices") Signed-off-by: Jiawen Wu Reviewed-by: Simon Horman Link: https://patch.msgid.link/1C6BF7A937237F5A+20250513021009.145708-2-jiawenwu@trustnetic.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/wangxun/txgbe/txgbe_hw.c | 8 +++++++- drivers/net/ethernet/wangxun/txgbe/txgbe_type.h | 2 ++ 2 files changed, 9 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_hw.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_hw.c index 4b9921b7bb11..a054b259d435 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_hw.c +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_hw.c @@ -99,9 +99,15 @@ static int txgbe_calc_eeprom_checksum(struct wx *wx, u16 *checksum) } local_buffer = eeprom_ptrs; - for (i = 0; i < TXGBE_EEPROM_LAST_WORD; i++) + for (i = 0; i < TXGBE_EEPROM_LAST_WORD; i++) { + if (wx->mac.type == wx_mac_aml) { + if (i >= TXGBE_EEPROM_I2C_SRART_PTR && + i < TXGBE_EEPROM_I2C_END_PTR) + local_buffer[i] = 0xffff; + } if (i != wx->eeprom.sw_region_offset + TXGBE_EEPROM_CHECKSUM) *checksum += local_buffer[i]; + } kvfree(eeprom_ptrs); diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h b/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h index 9c1c26234cad..f423012dec22 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h @@ -158,6 +158,8 @@ #define TXGBE_EEPROM_VERSION_L 0x1D #define TXGBE_EEPROM_VERSION_H 0x1E #define TXGBE_ISCSI_BOOT_CONFIG 0x07 +#define TXGBE_EEPROM_I2C_SRART_PTR 0x580 +#define TXGBE_EEPROM_I2C_END_PTR 0x800 #define TXGBE_MAX_MSIX_VECTORS 64 #define TXGBE_MAX_FDIR_INDICES 63 From 42efa358f033492097b214309eda464387d15318 Mon Sep 17 00:00:00 2001 From: Jiawen Wu Date: Tue, 13 May 2025 10:10:08 +0800 Subject: [PATCH 528/555] net: libwx: Fix FW mailbox reply timeout For the new SW-FW interaction, the timeout waiting for the firmware to return is too short. So that some mailbox commands cannot be completed. Use the 'timeout' parameter instead of fixed timeout value for flexible configuration. Fixes: 2e5af6b2ae85 ("net: txgbe: Add basic support for new AML devices") Signed-off-by: Jiawen Wu Reviewed-by: Simon Horman Link: https://patch.msgid.link/5D5BDE3EA501BDB8+20250513021009.145708-3-jiawenwu@trustnetic.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/wangxun/libwx/wx_hw.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/wangxun/libwx/wx_hw.c b/drivers/net/ethernet/wangxun/libwx/wx_hw.c index aed45abafb1b..ccdc57e9b0d5 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_hw.c +++ b/drivers/net/ethernet/wangxun/libwx/wx_hw.c @@ -434,8 +434,8 @@ static int wx_host_interface_command_r(struct wx *wx, u32 *buffer, wr32m(wx, WX_SW2FW_MBOX_CMD, WX_SW2FW_MBOX_CMD_VLD, WX_SW2FW_MBOX_CMD_VLD); /* polling reply from FW */ - err = read_poll_timeout(wx_poll_fw_reply, reply, reply, 1000, 50000, - true, wx, buffer, send_cmd); + err = read_poll_timeout(wx_poll_fw_reply, reply, reply, 2000, + timeout * 1000, true, wx, buffer, send_cmd); if (err) { wx_err(wx, "Polling from FW messages timeout, cmd: 0x%x, index: %d\n", send_cmd, wx->swfw_index); From 09e76365baa1d3fb7617d48f00662d32dd9d4828 Mon Sep 17 00:00:00 2001 From: Jiawen Wu Date: Tue, 13 May 2025 10:10:09 +0800 Subject: [PATCH 529/555] net: libwx: Fix FW mailbox unknown command For the new SW-FW interaction, missing the error return if there is an unknown command. It causes the driver to mistakenly believe that the interaction is complete. This problem occurs when new driver is paired with old firmware, which does not support the new mailbox commands. Fixes: 2e5af6b2ae85 ("net: txgbe: Add basic support for new AML devices") Signed-off-by: Jiawen Wu Reviewed-by: Simon Horman Link: https://patch.msgid.link/64DBB705D35A0016+20250513021009.145708-4-jiawenwu@trustnetic.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/wangxun/libwx/wx_hw.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/net/ethernet/wangxun/libwx/wx_hw.c b/drivers/net/ethernet/wangxun/libwx/wx_hw.c index ccdc57e9b0d5..490d34233d38 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_hw.c +++ b/drivers/net/ethernet/wangxun/libwx/wx_hw.c @@ -442,6 +442,12 @@ static int wx_host_interface_command_r(struct wx *wx, u32 *buffer, goto rel_out; } + if (hdr->cmd_or_resp.ret_status == 0x80) { + wx_err(wx, "Unknown FW command: 0x%x\n", send_cmd); + err = -EINVAL; + goto rel_out; + } + /* expect no reply from FW then return */ if (!return_data) goto rel_out; From 1bdea6fad6fb985ff13828373c48e337c4e939f9 Mon Sep 17 00:00:00 2001 From: Bo-Cun Chen Date: Tue, 13 May 2025 05:27:30 +0100 Subject: [PATCH 530/555] net: ethernet: mtk_eth_soc: fix typo for declaration MT7988 ESW capability Since MTK_ESW_BIT is a bit number rather than a bitmap, it causes MTK_HAS_CAPS to produce incorrect results. This leads to the ETH driver not declaring MAC capabilities correctly for the MT7988 ESW. Fixes: 445eb6448ed3 ("net: ethernet: mtk_eth_soc: add basic support for MT7988 SoC") Signed-off-by: Bo-Cun Chen Signed-off-by: Daniel Golle Reviewed-by: Michal Swiatkowski Link: https://patch.msgid.link/b8b37f409d1280fad9c4d32521e6207f63cd3213.1747110258.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mediatek/mtk_eth_soc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index 22a532695fb0..6c92072b4c28 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -4748,7 +4748,7 @@ static int mtk_add_mac(struct mtk_eth *eth, struct device_node *np) } if (mtk_is_netsys_v3_or_greater(mac->hw) && - MTK_HAS_CAPS(mac->hw->soc->caps, MTK_ESW_BIT) && + MTK_HAS_CAPS(mac->hw->soc->caps, MTK_ESW) && id == MTK_GMAC1_ID) { mac->phylink_config.mac_capabilities = MAC_ASYM_PAUSE | MAC_SYM_PAUSE | From bf449f35e77fd44017abf991fac1f9ab7705bbe0 Mon Sep 17 00:00:00 2001 From: Hariprasad Kelam Date: Tue, 13 May 2025 12:45:54 +0530 Subject: [PATCH 531/555] octeontx2-af: Fix CGX Receive counters Each CGX block supports 4 logical MACs (LMACS). Receive counters CGX_CMR_RX_STAT0-8 are per LMAC and CGX_CMR_RX_STAT9-12 are per CGX. Due a bug in previous patch, stale Per CGX counters values observed. Fixes: 66208910e57a ("octeontx2-af: Support to retrieve CGX LMAC stats") Signed-off-by: Hariprasad Kelam Link: https://patch.msgid.link/20250513071554.728922-1-hkelam@marvell.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/marvell/octeontx2/af/cgx.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/net/ethernet/marvell/octeontx2/af/cgx.c b/drivers/net/ethernet/marvell/octeontx2/af/cgx.c index 0b27a695008b..971993586fb4 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/cgx.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/cgx.c @@ -717,6 +717,11 @@ int cgx_get_rx_stats(void *cgxd, int lmac_id, int idx, u64 *rx_stat) if (!is_lmac_valid(cgx, lmac_id)) return -ENODEV; + + /* pass lmac as 0 for CGX_CMR_RX_STAT9-12 */ + if (idx >= CGX_RX_STAT_GLOBAL_INDEX) + lmac_id = 0; + *rx_stat = cgx_read(cgx, lmac_id, CGXX_CMRX_RX_STAT0 + (idx * 8)); return 0; } From 380b75d3078626aadd0817de61f3143f5db6e393 Mon Sep 17 00:00:00 2001 From: Michael Kelley Date: Mon, 12 May 2025 17:06:00 -0700 Subject: [PATCH 532/555] Drivers: hv: Allow vmbus_sendpacket_mpb_desc() to create multiple ranges vmbus_sendpacket_mpb_desc() is currently used only by the storvsc driver and is hardcoded to create a single GPA range. To allow it to also be used by the netvsc driver to create multiple GPA ranges, no longer hardcode as having a single GPA range. Allow the calling driver to specify the rangecount in the supplied descriptor. Update the storvsc driver to reflect this new approach. Cc: # 6.1.x Signed-off-by: Michael Kelley Link: https://patch.msgid.link/20250513000604.1396-2-mhklinux@outlook.com Signed-off-by: Jakub Kicinski --- drivers/hv/channel.c | 6 +++--- drivers/scsi/storvsc_drv.c | 1 + 2 files changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/hv/channel.c b/drivers/hv/channel.c index fb8cd8469328..4ffd5eaa7817 100644 --- a/drivers/hv/channel.c +++ b/drivers/hv/channel.c @@ -1136,9 +1136,10 @@ int vmbus_sendpacket_pagebuffer(struct vmbus_channel *channel, EXPORT_SYMBOL_GPL(vmbus_sendpacket_pagebuffer); /* - * vmbus_sendpacket_multipagebuffer - Send a multi-page buffer packet + * vmbus_sendpacket_mpb_desc - Send one or more multi-page buffer packets * using a GPADL Direct packet type. - * The buffer includes the vmbus descriptor. + * The desc argument must include space for the VMBus descriptor. The + * rangecount field must already be set. */ int vmbus_sendpacket_mpb_desc(struct vmbus_channel *channel, struct vmbus_packet_mpb_array *desc, @@ -1160,7 +1161,6 @@ int vmbus_sendpacket_mpb_desc(struct vmbus_channel *channel, desc->length8 = (u16)(packetlen_aligned >> 3); desc->transactionid = VMBUS_RQST_ERROR; /* will be updated in hv_ringbuffer_write() */ desc->reserved = 0; - desc->rangecount = 1; bufferlist[0].iov_base = desc; bufferlist[0].iov_len = desc_size; diff --git a/drivers/scsi/storvsc_drv.c b/drivers/scsi/storvsc_drv.c index 35db061ae3ec..2e6b2412d2c9 100644 --- a/drivers/scsi/storvsc_drv.c +++ b/drivers/scsi/storvsc_drv.c @@ -1819,6 +1819,7 @@ static int storvsc_queuecommand(struct Scsi_Host *host, struct scsi_cmnd *scmnd) return SCSI_MLQUEUE_DEVICE_BUSY; } + payload->rangecount = 1; payload->range.len = length; payload->range.offset = offset_in_hvpg; From 4f98616b855cb0e3b5917918bb07b44728eb96ea Mon Sep 17 00:00:00 2001 From: Michael Kelley Date: Mon, 12 May 2025 17:06:01 -0700 Subject: [PATCH 533/555] hv_netvsc: Use vmbus_sendpacket_mpb_desc() to send VMBus messages netvsc currently uses vmbus_sendpacket_pagebuffer() to send VMBus messages. This function creates a series of GPA ranges, each of which contains a single PFN. However, if the rndis header in the VMBus message crosses a page boundary, the netvsc protocol with the host requires that both PFNs for the rndis header must be in a single "GPA range" data structure, which isn't possible with vmbus_sendpacket_pagebuffer(). As the first step in fixing this, add a new function netvsc_build_mpb_array() to build a VMBus message with multiple GPA ranges, each of which may contain multiple PFNs. Use vmbus_sendpacket_mpb_desc() to send this VMBus message to the host. There's no functional change since higher levels of netvsc don't maintain or propagate knowledge of contiguous PFNs. Based on its input, netvsc_build_mpb_array() still produces a separate GPA range for each PFN and the behavior is the same as with vmbus_sendpacket_pagebuffer(). But the groundwork is laid for a subsequent patch to provide the necessary grouping. Cc: # 6.1.x Signed-off-by: Michael Kelley Link: https://patch.msgid.link/20250513000604.1396-3-mhklinux@outlook.com Signed-off-by: Jakub Kicinski --- drivers/net/hyperv/netvsc.c | 50 +++++++++++++++++++++++++++++++++---- 1 file changed, 45 insertions(+), 5 deletions(-) diff --git a/drivers/net/hyperv/netvsc.c b/drivers/net/hyperv/netvsc.c index d6f5b9ea3109..5882d6670200 100644 --- a/drivers/net/hyperv/netvsc.c +++ b/drivers/net/hyperv/netvsc.c @@ -1055,6 +1055,42 @@ static int netvsc_dma_map(struct hv_device *hv_dev, return 0; } +/* Build an "array" of mpb entries describing the data to be transferred + * over VMBus. After the desc header fields, each "array" entry is variable + * size, and each entry starts after the end of the previous entry. The + * "offset" and "len" fields for each entry imply the size of the entry. + * + * The pfns are in HV_HYP_PAGE_SIZE, because all communication with Hyper-V + * uses that granularity, even if the system page size of the guest is larger. + * Each entry in the input "pb" array must describe a contiguous range of + * guest physical memory so that the pfns are sequential if the range crosses + * a page boundary. The offset field must be < HV_HYP_PAGE_SIZE. + */ +static inline void netvsc_build_mpb_array(struct hv_page_buffer *pb, + u32 page_buffer_count, + struct vmbus_packet_mpb_array *desc, + u32 *desc_size) +{ + struct hv_mpb_array *mpb_entry = &desc->range; + int i, j; + + for (i = 0; i < page_buffer_count; i++) { + u32 offset = pb[i].offset; + u32 len = pb[i].len; + + mpb_entry->offset = offset; + mpb_entry->len = len; + + for (j = 0; j < HVPFN_UP(offset + len); j++) + mpb_entry->pfn_array[j] = pb[i].pfn + j; + + mpb_entry = (struct hv_mpb_array *)&mpb_entry->pfn_array[j]; + } + + desc->rangecount = page_buffer_count; + *desc_size = (char *)mpb_entry - (char *)desc; +} + static inline int netvsc_send_pkt( struct hv_device *device, struct hv_netvsc_packet *packet, @@ -1097,6 +1133,9 @@ static inline int netvsc_send_pkt( packet->dma_range = NULL; if (packet->page_buf_cnt) { + struct vmbus_channel_packet_page_buffer desc; + u32 desc_size; + if (packet->cp_partial) pb += packet->rmsg_pgcnt; @@ -1106,11 +1145,12 @@ static inline int netvsc_send_pkt( goto exit; } - ret = vmbus_sendpacket_pagebuffer(out_channel, - pb, packet->page_buf_cnt, - &nvmsg, sizeof(nvmsg), - req_id); - + netvsc_build_mpb_array(pb, packet->page_buf_cnt, + (struct vmbus_packet_mpb_array *)&desc, + &desc_size); + ret = vmbus_sendpacket_mpb_desc(out_channel, + (struct vmbus_packet_mpb_array *)&desc, + desc_size, &nvmsg, sizeof(nvmsg), req_id); if (ret) netvsc_dma_unmap(ndev_ctx->device_ctx, packet); } else { From 41a6328b2c55276f89ea3812069fd7521e348bbf Mon Sep 17 00:00:00 2001 From: Michael Kelley Date: Mon, 12 May 2025 17:06:02 -0700 Subject: [PATCH 534/555] hv_netvsc: Preserve contiguous PFN grouping in the page buffer array Starting with commit dca5161f9bd0 ("hv_netvsc: Check status in SEND_RNDIS_PKT completion message") in the 6.3 kernel, the Linux driver for Hyper-V synthetic networking (netvsc) occasionally reports "nvsp_rndis_pkt_complete error status: 2".[1] This error indicates that Hyper-V has rejected a network packet transmit request from the guest, and the outgoing network packet is dropped. Higher level network protocols presumably recover and resend the packet so there is no functional error, but performance is slightly impacted. Commit dca5161f9bd0 is not the cause of the error -- it only added reporting of an error that was already happening without any notice. The error has presumably been present since the netvsc driver was originally introduced into Linux. The root cause of the problem is that the netvsc driver in Linux may send an incorrectly formatted VMBus message to Hyper-V when transmitting the network packet. The incorrect formatting occurs when the rndis header of the VMBus message crosses a page boundary due to how the Linux skb head memory is aligned. In such a case, two PFNs are required to describe the location of the rndis header, even though they are contiguous in guest physical address (GPA) space. Hyper-V requires that two rndis header PFNs be in a single "GPA range" data struture, but current netvsc code puts each PFN in its own GPA range, which Hyper-V rejects as an error. The incorrect formatting occurs only for larger packets that netvsc must transmit via a VMBus "GPA Direct" message. There's no problem when netvsc transmits a smaller packet by copying it into a pre- allocated send buffer slot because the pre-allocated slots don't have page crossing issues. After commit 14ad6ed30a10 ("net: allow small head cache usage with large MAX_SKB_FRAGS values") in the 6.14-rc4 kernel, the error occurs much more frequently in VMs with 16 or more vCPUs. It may occur every few seconds, or even more frequently, in an ssh session that outputs a lot of text. Commit 14ad6ed30a10 subtly changes how skb head memory is allocated, making it much more likely that the rndis header will cross a page boundary when the vCPU count is 16 or more. The changes in commit 14ad6ed30a10 are perfectly valid -- they just had the side effect of making the netvsc bug more prominent. Current code in init_page_array() creates a separate page buffer array entry for each PFN required to identify the data to be transmitted. Contiguous PFNs get separate entries in the page buffer array, and any information about contiguity is lost. Fix the core issue by having init_page_array() construct the page buffer array to represent contiguous ranges rather than individual pages. When these ranges are subsequently passed to netvsc_build_mpb_array(), it can build GPA ranges that contain multiple PFNs, as required to avoid the error "nvsp_rndis_pkt_complete error status: 2". If instead the network packet is sent by copying into a pre-allocated send buffer slot, the copy proceeds using the contiguous ranges rather than individual pages, but the result of the copying is the same. Also fix rndis_filter_send_request() to construct a contiguous range, since it has its own page buffer array. This change has a side benefit in CoCo VMs in that netvsc_dma_map() calls dma_map_single() on each contiguous range instead of on each page. This results in fewer calls to dma_map_single() but on larger chunks of memory, which should reduce contention on the swiotlb. Since the page buffer array now contains one entry for each contiguous range instead of for each individual page, the number of entries in the array can be reduced, saving 208 bytes of stack space in netvsc_xmit() when MAX_SKG_FRAGS has the default value of 17. [1] https://bugzilla.kernel.org/show_bug.cgi?id=217503 Closes: https://bugzilla.kernel.org/show_bug.cgi?id=217503 Cc: # 6.1.x Signed-off-by: Michael Kelley Link: https://patch.msgid.link/20250513000604.1396-4-mhklinux@outlook.com Signed-off-by: Jakub Kicinski --- drivers/net/hyperv/hyperv_net.h | 12 ++++++ drivers/net/hyperv/netvsc_drv.c | 63 ++++++++----------------------- drivers/net/hyperv/rndis_filter.c | 24 +++--------- 3 files changed, 32 insertions(+), 67 deletions(-) diff --git a/drivers/net/hyperv/hyperv_net.h b/drivers/net/hyperv/hyperv_net.h index 70f7cb383228..76725f25abd5 100644 --- a/drivers/net/hyperv/hyperv_net.h +++ b/drivers/net/hyperv/hyperv_net.h @@ -893,6 +893,18 @@ struct nvsp_message { sizeof(struct nvsp_message)) #define NETVSC_MIN_IN_MSG_SIZE sizeof(struct vmpacket_descriptor) +/* Maximum # of contiguous data ranges that can make up a trasmitted packet. + * Typically it's the max SKB fragments plus 2 for the rndis packet and the + * linear portion of the SKB. But if MAX_SKB_FRAGS is large, the value may + * need to be limited to MAX_PAGE_BUFFER_COUNT, which is the max # of entries + * in a GPA direct packet sent to netvsp over VMBus. + */ +#if MAX_SKB_FRAGS + 2 < MAX_PAGE_BUFFER_COUNT +#define MAX_DATA_RANGES (MAX_SKB_FRAGS + 2) +#else +#define MAX_DATA_RANGES MAX_PAGE_BUFFER_COUNT +#endif + /* Estimated requestor size: * out_ring_size/min_out_msg_size + in_ring_size/min_in_msg_size */ diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c index c51b318b8a72..929f6b3de768 100644 --- a/drivers/net/hyperv/netvsc_drv.c +++ b/drivers/net/hyperv/netvsc_drv.c @@ -326,43 +326,10 @@ static u16 netvsc_select_queue(struct net_device *ndev, struct sk_buff *skb, return txq; } -static u32 fill_pg_buf(unsigned long hvpfn, u32 offset, u32 len, - struct hv_page_buffer *pb) -{ - int j = 0; - - hvpfn += offset >> HV_HYP_PAGE_SHIFT; - offset = offset & ~HV_HYP_PAGE_MASK; - - while (len > 0) { - unsigned long bytes; - - bytes = HV_HYP_PAGE_SIZE - offset; - if (bytes > len) - bytes = len; - pb[j].pfn = hvpfn; - pb[j].offset = offset; - pb[j].len = bytes; - - offset += bytes; - len -= bytes; - - if (offset == HV_HYP_PAGE_SIZE && len) { - hvpfn++; - offset = 0; - j++; - } - } - - return j + 1; -} - static u32 init_page_array(void *hdr, u32 len, struct sk_buff *skb, struct hv_netvsc_packet *packet, struct hv_page_buffer *pb) { - u32 slots_used = 0; - char *data = skb->data; int frags = skb_shinfo(skb)->nr_frags; int i; @@ -371,28 +338,28 @@ static u32 init_page_array(void *hdr, u32 len, struct sk_buff *skb, * 2. skb linear data * 3. skb fragment data */ - slots_used += fill_pg_buf(virt_to_hvpfn(hdr), - offset_in_hvpage(hdr), - len, - &pb[slots_used]); + pb[0].offset = offset_in_hvpage(hdr); + pb[0].len = len; + pb[0].pfn = virt_to_hvpfn(hdr); packet->rmsg_size = len; - packet->rmsg_pgcnt = slots_used; + packet->rmsg_pgcnt = 1; - slots_used += fill_pg_buf(virt_to_hvpfn(data), - offset_in_hvpage(data), - skb_headlen(skb), - &pb[slots_used]); + pb[1].offset = offset_in_hvpage(skb->data); + pb[1].len = skb_headlen(skb); + pb[1].pfn = virt_to_hvpfn(skb->data); for (i = 0; i < frags; i++) { skb_frag_t *frag = skb_shinfo(skb)->frags + i; + struct hv_page_buffer *cur_pb = &pb[i + 2]; + u64 pfn = page_to_hvpfn(skb_frag_page(frag)); + u32 offset = skb_frag_off(frag); - slots_used += fill_pg_buf(page_to_hvpfn(skb_frag_page(frag)), - skb_frag_off(frag), - skb_frag_size(frag), - &pb[slots_used]); + cur_pb->offset = offset_in_hvpage(offset); + cur_pb->len = skb_frag_size(frag); + cur_pb->pfn = pfn + (offset >> HV_HYP_PAGE_SHIFT); } - return slots_used; + return frags + 2; } static int count_skb_frag_slots(struct sk_buff *skb) @@ -483,7 +450,7 @@ static int netvsc_xmit(struct sk_buff *skb, struct net_device *net, bool xdp_tx) struct net_device *vf_netdev; u32 rndis_msg_size; u32 hash; - struct hv_page_buffer pb[MAX_PAGE_BUFFER_COUNT]; + struct hv_page_buffer pb[MAX_DATA_RANGES]; /* If VF is present and up then redirect packets to it. * Skip the VF if it is marked down or has no carrier. diff --git a/drivers/net/hyperv/rndis_filter.c b/drivers/net/hyperv/rndis_filter.c index 82747dfacd70..9e73959e61ee 100644 --- a/drivers/net/hyperv/rndis_filter.c +++ b/drivers/net/hyperv/rndis_filter.c @@ -225,8 +225,7 @@ static int rndis_filter_send_request(struct rndis_device *dev, struct rndis_request *req) { struct hv_netvsc_packet *packet; - struct hv_page_buffer page_buf[2]; - struct hv_page_buffer *pb = page_buf; + struct hv_page_buffer pb; int ret; /* Setup the packet to send it */ @@ -235,27 +234,14 @@ static int rndis_filter_send_request(struct rndis_device *dev, packet->total_data_buflen = req->request_msg.msg_len; packet->page_buf_cnt = 1; - pb[0].pfn = virt_to_phys(&req->request_msg) >> - HV_HYP_PAGE_SHIFT; - pb[0].len = req->request_msg.msg_len; - pb[0].offset = offset_in_hvpage(&req->request_msg); - - /* Add one page_buf when request_msg crossing page boundary */ - if (pb[0].offset + pb[0].len > HV_HYP_PAGE_SIZE) { - packet->page_buf_cnt++; - pb[0].len = HV_HYP_PAGE_SIZE - - pb[0].offset; - pb[1].pfn = virt_to_phys((void *)&req->request_msg - + pb[0].len) >> HV_HYP_PAGE_SHIFT; - pb[1].offset = 0; - pb[1].len = req->request_msg.msg_len - - pb[0].len; - } + pb.pfn = virt_to_phys(&req->request_msg) >> HV_HYP_PAGE_SHIFT; + pb.len = req->request_msg.msg_len; + pb.offset = offset_in_hvpage(&req->request_msg); trace_rndis_send(dev->ndev, 0, &req->request_msg); rcu_read_lock_bh(); - ret = netvsc_send(dev->ndev, packet, NULL, pb, NULL, false); + ret = netvsc_send(dev->ndev, packet, NULL, &pb, NULL, false); rcu_read_unlock_bh(); return ret; From 5bbc644bbf4e97a05bc0cb052189004588ff8a09 Mon Sep 17 00:00:00 2001 From: Michael Kelley Date: Mon, 12 May 2025 17:06:03 -0700 Subject: [PATCH 535/555] hv_netvsc: Remove rmsg_pgcnt init_page_array() now always creates a single page buffer array entry for the rndis message, even if the rndis message crosses a page boundary. As such, the number of page buffer array entries used for the rndis message must no longer be tracked -- it is always just 1. Remove the rmsg_pgcnt field and use "1" where the value is needed. Cc: # 6.1.x Signed-off-by: Michael Kelley Link: https://patch.msgid.link/20250513000604.1396-5-mhklinux@outlook.com Signed-off-by: Jakub Kicinski --- drivers/net/hyperv/hyperv_net.h | 1 - drivers/net/hyperv/netvsc.c | 7 +++---- drivers/net/hyperv/netvsc_drv.c | 1 - 3 files changed, 3 insertions(+), 6 deletions(-) diff --git a/drivers/net/hyperv/hyperv_net.h b/drivers/net/hyperv/hyperv_net.h index 76725f25abd5..cb6f5482d203 100644 --- a/drivers/net/hyperv/hyperv_net.h +++ b/drivers/net/hyperv/hyperv_net.h @@ -158,7 +158,6 @@ struct hv_netvsc_packet { u8 cp_partial; /* partial copy into send buffer */ u8 rmsg_size; /* RNDIS header and PPI size */ - u8 rmsg_pgcnt; /* page count of RNDIS header and PPI */ u8 page_buf_cnt; u16 q_idx; diff --git a/drivers/net/hyperv/netvsc.c b/drivers/net/hyperv/netvsc.c index 5882d6670200..720104661d7f 100644 --- a/drivers/net/hyperv/netvsc.c +++ b/drivers/net/hyperv/netvsc.c @@ -953,8 +953,7 @@ static void netvsc_copy_to_send_buf(struct netvsc_device *net_device, + pend_size; int i; u32 padding = 0; - u32 page_count = packet->cp_partial ? packet->rmsg_pgcnt : - packet->page_buf_cnt; + u32 page_count = packet->cp_partial ? 1 : packet->page_buf_cnt; u32 remain; /* Add padding */ @@ -1137,7 +1136,7 @@ static inline int netvsc_send_pkt( u32 desc_size; if (packet->cp_partial) - pb += packet->rmsg_pgcnt; + pb++; ret = netvsc_dma_map(ndev_ctx->device_ctx, packet, pb); if (ret) { @@ -1299,7 +1298,7 @@ int netvsc_send(struct net_device *ndev, packet->send_buf_index = section_index; if (packet->cp_partial) { - packet->page_buf_cnt -= packet->rmsg_pgcnt; + packet->page_buf_cnt--; packet->total_data_buflen = msd_len + packet->rmsg_size; } else { packet->page_buf_cnt = 0; diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c index 929f6b3de768..d8b169ac0343 100644 --- a/drivers/net/hyperv/netvsc_drv.c +++ b/drivers/net/hyperv/netvsc_drv.c @@ -343,7 +343,6 @@ static u32 init_page_array(void *hdr, u32 len, struct sk_buff *skb, pb[0].len = len; pb[0].pfn = virt_to_hvpfn(hdr); packet->rmsg_size = len; - packet->rmsg_pgcnt = 1; pb[1].offset = offset_in_hvpage(skb->data); pb[1].len = skb_headlen(skb); From 45a442fe369e6c4e0b4aa9f63b31c3f2f9e2090e Mon Sep 17 00:00:00 2001 From: Michael Kelley Date: Mon, 12 May 2025 17:06:04 -0700 Subject: [PATCH 536/555] Drivers: hv: vmbus: Remove vmbus_sendpacket_pagebuffer() With the netvsc driver changed to use vmbus_sendpacket_mpb_desc() instead of vmbus_sendpacket_pagebuffer(), the latter has no remaining callers. Remove it. Cc: # 6.1.x Signed-off-by: Michael Kelley Link: https://patch.msgid.link/20250513000604.1396-6-mhklinux@outlook.com Signed-off-by: Jakub Kicinski --- drivers/hv/channel.c | 59 ------------------------------------------ include/linux/hyperv.h | 7 ----- 2 files changed, 66 deletions(-) diff --git a/drivers/hv/channel.c b/drivers/hv/channel.c index 4ffd5eaa7817..35f26fa1ffe7 100644 --- a/drivers/hv/channel.c +++ b/drivers/hv/channel.c @@ -1076,65 +1076,6 @@ int vmbus_sendpacket(struct vmbus_channel *channel, void *buffer, } EXPORT_SYMBOL(vmbus_sendpacket); -/* - * vmbus_sendpacket_pagebuffer - Send a range of single-page buffer - * packets using a GPADL Direct packet type. This interface allows you - * to control notifying the host. This will be useful for sending - * batched data. Also the sender can control the send flags - * explicitly. - */ -int vmbus_sendpacket_pagebuffer(struct vmbus_channel *channel, - struct hv_page_buffer pagebuffers[], - u32 pagecount, void *buffer, u32 bufferlen, - u64 requestid) -{ - int i; - struct vmbus_channel_packet_page_buffer desc; - u32 descsize; - u32 packetlen; - u32 packetlen_aligned; - struct kvec bufferlist[3]; - u64 aligned_data = 0; - - if (pagecount > MAX_PAGE_BUFFER_COUNT) - return -EINVAL; - - /* - * Adjust the size down since vmbus_channel_packet_page_buffer is the - * largest size we support - */ - descsize = sizeof(struct vmbus_channel_packet_page_buffer) - - ((MAX_PAGE_BUFFER_COUNT - pagecount) * - sizeof(struct hv_page_buffer)); - packetlen = descsize + bufferlen; - packetlen_aligned = ALIGN(packetlen, sizeof(u64)); - - /* Setup the descriptor */ - desc.type = VM_PKT_DATA_USING_GPA_DIRECT; - desc.flags = VMBUS_DATA_PACKET_FLAG_COMPLETION_REQUESTED; - desc.dataoffset8 = descsize >> 3; /* in 8-bytes granularity */ - desc.length8 = (u16)(packetlen_aligned >> 3); - desc.transactionid = VMBUS_RQST_ERROR; /* will be updated in hv_ringbuffer_write() */ - desc.reserved = 0; - desc.rangecount = pagecount; - - for (i = 0; i < pagecount; i++) { - desc.range[i].len = pagebuffers[i].len; - desc.range[i].offset = pagebuffers[i].offset; - desc.range[i].pfn = pagebuffers[i].pfn; - } - - bufferlist[0].iov_base = &desc; - bufferlist[0].iov_len = descsize; - bufferlist[1].iov_base = buffer; - bufferlist[1].iov_len = bufferlen; - bufferlist[2].iov_base = &aligned_data; - bufferlist[2].iov_len = (packetlen_aligned - packetlen); - - return hv_ringbuffer_write(channel, bufferlist, 3, requestid, NULL); -} -EXPORT_SYMBOL_GPL(vmbus_sendpacket_pagebuffer); - /* * vmbus_sendpacket_mpb_desc - Send one or more multi-page buffer packets * using a GPADL Direct packet type. diff --git a/include/linux/hyperv.h b/include/linux/hyperv.h index 675959fb97ba..ea806cef4a0d 100644 --- a/include/linux/hyperv.h +++ b/include/linux/hyperv.h @@ -1161,13 +1161,6 @@ extern int vmbus_sendpacket(struct vmbus_channel *channel, enum vmbus_packet_type type, u32 flags); -extern int vmbus_sendpacket_pagebuffer(struct vmbus_channel *channel, - struct hv_page_buffer pagebuffers[], - u32 pagecount, - void *buffer, - u32 bufferlen, - u64 requestid); - extern int vmbus_sendpacket_mpb_desc(struct vmbus_channel *channel, struct vmbus_packet_mpb_array *mpb, u32 desc_size, From 78ab4be549533432d97ea8989d2f00b508fa68d8 Mon Sep 17 00:00:00 2001 From: Fedor Pchelkin Date: Tue, 6 May 2025 14:55:39 +0300 Subject: [PATCH 537/555] wifi: mt76: disable napi on driver removal A warning on driver removal started occurring after commit 9dd05df8403b ("net: warn if NAPI instance wasn't shut down"). Disable tx napi before deleting it in mt76_dma_cleanup(). WARNING: CPU: 4 PID: 18828 at net/core/dev.c:7288 __netif_napi_del_locked+0xf0/0x100 CPU: 4 UID: 0 PID: 18828 Comm: modprobe Not tainted 6.15.0-rc4 #4 PREEMPT(lazy) Hardware name: ASUS System Product Name/PRIME X670E-PRO WIFI, BIOS 3035 09/05/2024 RIP: 0010:__netif_napi_del_locked+0xf0/0x100 Call Trace: mt76_dma_cleanup+0x54/0x2f0 [mt76] mt7921_pci_remove+0xd5/0x190 [mt7921e] pci_device_remove+0x47/0xc0 device_release_driver_internal+0x19e/0x200 driver_detach+0x48/0x90 bus_remove_driver+0x6d/0xf0 pci_unregister_driver+0x2e/0xb0 __do_sys_delete_module.isra.0+0x197/0x2e0 do_syscall_64+0x7b/0x160 entry_SYSCALL_64_after_hwframe+0x76/0x7e Tested with mt7921e but the same pattern can be actually applied to other mt76 drivers calling mt76_dma_cleanup() during removal. Tx napi is enabled in their *_dma_init() functions and only toggled off and on again inside their suspend/resume/reset paths. So it should be okay to disable tx napi in such a generic way. Found by Linux Verification Center (linuxtesting.org). Fixes: 2ac515a5d74f ("mt76: mt76x02: use napi polling for tx cleanup") Cc: stable@vger.kernel.org Signed-off-by: Fedor Pchelkin Tested-by: Ming Yen Hsieh Link: https://patch.msgid.link/20250506115540.19045-1-pchelkin@ispras.ru Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/dma.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/wireless/mediatek/mt76/dma.c b/drivers/net/wireless/mediatek/mt76/dma.c index 844af16ee551..35b4ec91979e 100644 --- a/drivers/net/wireless/mediatek/mt76/dma.c +++ b/drivers/net/wireless/mediatek/mt76/dma.c @@ -1011,6 +1011,7 @@ void mt76_dma_cleanup(struct mt76_dev *dev) int i; mt76_worker_disable(&dev->tx_worker); + napi_disable(&dev->tx_napi); netif_napi_del(&dev->tx_napi); for (i = 0; i < ARRAY_SIZE(dev->phys); i++) { From 0aa8496adda570c2005410a30df963a16643a3dc Mon Sep 17 00:00:00 2001 From: Ming Yen Hsieh Date: Fri, 9 May 2025 09:04:20 +0800 Subject: [PATCH 538/555] wifi: mt76: mt7925: fix missing hdr_trans_tlv command for broadcast wtbl Ensure that the hdr_trans_tlv command is included in the broadcast wtbl to prevent the IPv6 and multicast packet from being dropped by the chip. Cc: stable@vger.kernel.org Fixes: cb1353ef3473 ("wifi: mt76: mt7925: integrate *mlo_sta_cmd and *sta_cmd") Reported-by: Benjamin Xiao Tested-by: Niklas Schnelle Signed-off-by: Ming Yen Hsieh Link: https://lore.kernel.org/lkml/EmWnO5b-acRH1TXbGnkx41eJw654vmCR-8_xMBaPMwexCnfkvKCdlU5u19CGbaapJ3KRu-l3B-tSUhf8CCQwL0odjo6Cd5YG5lvNeB-vfdg=@pm.me/ Link: https://patch.msgid.link/20250509010421.403022-1-mingyen.hsieh@mediatek.com Signed-off-by: Felix Fietkau --- drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c index e61da76b2097..14b1f603fb62 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c @@ -1924,14 +1924,14 @@ mt7925_mcu_sta_cmd(struct mt76_phy *phy, mt7925_mcu_sta_mld_tlv(skb, info->vif, info->link_sta->sta); mt7925_mcu_sta_eht_mld_tlv(skb, info->vif, info->link_sta->sta); } - - mt7925_mcu_sta_hdr_trans_tlv(skb, info->vif, info->link_sta); } if (!info->enable) { mt7925_mcu_sta_remove_tlv(skb); mt76_connac_mcu_add_tlv(skb, STA_REC_MLD_OFF, sizeof(struct tlv)); + } else { + mt7925_mcu_sta_hdr_trans_tlv(skb, info->vif, info->link_sta); } return mt76_mcu_skb_send_msg(dev, skb, info->cmd, true); From dcb479fde00be9a151c047d0a7c0626b64eb0019 Mon Sep 17 00:00:00 2001 From: Subbaraya Sundeep Date: Mon, 12 May 2025 18:22:37 +0530 Subject: [PATCH 539/555] octeontx2-pf: Do not reallocate all ntuple filters If ntuple filters count is modified followed by unicast filters count using devlink then the ntuple count set by user is ignored and all the ntuple filters are being reallocated. Fix this by storing the ntuple count set by user. Without this patch, say if user tries to modify ntuple count as 8 followed by ucast filter count as 4 using devlink commands then ntuple count is being reverted to default value 16 i.e, not retaining user set value 8. Fixes: 39c469188b6d ("octeontx2-pf: Add ucast filter count configurability via devlink.") Signed-off-by: Subbaraya Sundeep Reviewed-by: Simon Horman Link: https://patch.msgid.link/1747054357-5850-1-git-send-email-sbhatta@marvell.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h | 1 + drivers/net/ethernet/marvell/octeontx2/nic/otx2_devlink.c | 1 + drivers/net/ethernet/marvell/octeontx2/nic/otx2_flows.c | 3 ++- 3 files changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h index 1e88422825be..d6b4b74e4002 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h @@ -356,6 +356,7 @@ struct otx2_flow_config { struct list_head flow_list_tc; u8 ucast_flt_cnt; bool ntuple; + u16 ntuple_cnt; }; struct dev_hw_ops { diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_devlink.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_devlink.c index 33ec9a7f7c03..e13ae5484c19 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_devlink.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_devlink.c @@ -41,6 +41,7 @@ static int otx2_dl_mcam_count_set(struct devlink *devlink, u32 id, if (!pfvf->flow_cfg) return 0; + pfvf->flow_cfg->ntuple_cnt = ctx->val.vu16; otx2_alloc_mcam_entries(pfvf, ctx->val.vu16); return 0; diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_flows.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_flows.c index 47bfd1fb37d4..64c6d9162ef6 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_flows.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_flows.c @@ -247,7 +247,7 @@ int otx2_mcam_entry_init(struct otx2_nic *pfvf) mutex_unlock(&pfvf->mbox.lock); /* Allocate entries for Ntuple filters */ - count = otx2_alloc_mcam_entries(pfvf, OTX2_DEFAULT_FLOWCOUNT); + count = otx2_alloc_mcam_entries(pfvf, flow_cfg->ntuple_cnt); if (count <= 0) { otx2_clear_ntuple_flow_info(pfvf, flow_cfg); return 0; @@ -307,6 +307,7 @@ int otx2_mcam_flow_init(struct otx2_nic *pf) INIT_LIST_HEAD(&pf->flow_cfg->flow_list_tc); pf->flow_cfg->ucast_flt_cnt = OTX2_DEFAULT_UNICAST_FLOWS; + pf->flow_cfg->ntuple_cnt = OTX2_DEFAULT_FLOWCOUNT; /* Allocate bare minimum number of MCAM entries needed for * unicast and ntuple filters. From 811d6a923b40fc130f91abf49151f57cf9ac2a6f Mon Sep 17 00:00:00 2001 From: Vinod Koul Date: Thu, 15 May 2025 11:42:13 +0100 Subject: [PATCH 540/555] dmaengine: mediatek: drop unused variable Commit 157ae5ffd76a dmaengine: mediatek: Fix a possible deadlock error in mtk_cqdma_tx_status() fixed locks but kept unused varibale leading to warning and build failure (due to warning treated as errors) drivers/dma/mediatek/mtk-cqdma.c: In function 'mtk_cqdma_find_active_desc': drivers/dma/mediatek/mtk-cqdma.c:423:23: error: unused variable 'flags' [-Werror=unused-variable] 423 | unsigned long flags; | ^~~~~ Fix by dropping this unused flag Reported-by: Stephen Rothwell Fixes: 157ae5ffd76a ("dmaengine: mediatek: Fix a possible deadlock error in mtk_cqdma_tx_status()") Signed-off-by: Vinod Koul --- drivers/dma/mediatek/mtk-cqdma.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/dma/mediatek/mtk-cqdma.c b/drivers/dma/mediatek/mtk-cqdma.c index e35271ac1eed..47c8adfdc155 100644 --- a/drivers/dma/mediatek/mtk-cqdma.c +++ b/drivers/dma/mediatek/mtk-cqdma.c @@ -420,7 +420,6 @@ static struct virt_dma_desc *mtk_cqdma_find_active_desc(struct dma_chan *c, { struct mtk_cqdma_vchan *cvc = to_cqdma_vchan(c); struct virt_dma_desc *vd; - unsigned long flags; list_for_each_entry(vd, &cvc->pc->queue, node) if (vd->tx.cookie == cookie) { From 7b9938a14460e8ec7649ca2e80ac0aae9815bf02 Mon Sep 17 00:00:00 2001 From: Nicolas Chauvet Date: Thu, 15 May 2025 12:21:32 +0200 Subject: [PATCH 541/555] ALSA: usb-audio: Add sample rate quirk for Microdia JP001 USB Camera Microdia JP001 does not support reading the sample rate which leads to many lines of "cannot get freq at ep 0x84". This patch adds the USB ID to quirks.c and avoids those error messages. usb 7-4: New USB device found, idVendor=0c45, idProduct=636b, bcdDevice= 1.00 usb 7-4: New USB device strings: Mfr=2, Product=1, SerialNumber=3 usb 7-4: Product: JP001 usb 7-4: Manufacturer: JP001 usb 7-4: SerialNumber: JP001 usb 7-4: 3:1: cannot get freq at ep 0x84 Cc: Signed-off-by: Nicolas Chauvet Link: https://patch.msgid.link/20250515102132.73062-1-kwizart@gmail.com Signed-off-by: Takashi Iwai --- sound/usb/quirks.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/sound/usb/quirks.c b/sound/usb/quirks.c index eb192834db68..dbbc9eb935a4 100644 --- a/sound/usb/quirks.c +++ b/sound/usb/quirks.c @@ -2242,6 +2242,8 @@ static const struct usb_audio_quirk_flags_table quirk_flags_table[] = { QUIRK_FLAG_CTL_MSG_DELAY_1M), DEVICE_FLG(0x0c45, 0x6340, /* Sonix HD USB Camera */ QUIRK_FLAG_GET_SAMPLE_RATE), + DEVICE_FLG(0x0c45, 0x636b, /* Microdia JP001 USB Camera */ + QUIRK_FLAG_GET_SAMPLE_RATE), DEVICE_FLG(0x0d8c, 0x0014, /* USB Audio Device */ QUIRK_FLAG_CTL_MSG_DELAY_1M), DEVICE_FLG(0x0ecb, 0x205c, /* JBL Quantum610 Wireless */ From 82bbe02b2500ef0a62053fe2eb84773fe31c5a0a Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Fri, 9 May 2025 11:46:45 -0700 Subject: [PATCH 542/555] wifi: mac80211: Set n_channels after allocating struct cfg80211_scan_request Make sure that n_channels is set after allocating the struct cfg80211_registered_device::int_scan_req member. Seen with syzkaller: UBSAN: array-index-out-of-bounds in net/mac80211/scan.c:1208:5 index 0 is out of range for type 'struct ieee80211_channel *[] __counted_by(n_channels)' (aka 'struct ieee80211_channel *[]') This was missed in the initial conversions because I failed to locate the allocation likely due to the "sizeof(void *)" not matching the "channels" array type. Reported-by: syzbot+4bcdddd48bb6f0be0da1@syzkaller.appspotmail.com Closes: https://lore.kernel.org/lkml/680fd171.050a0220.2b69d1.045e.GAE@google.com/ Fixes: e3eac9f32ec0 ("wifi: cfg80211: Annotate struct cfg80211_scan_request with __counted_by") Signed-off-by: Kees Cook Reviewed-by: Gustavo A. R. Silva Link: https://patch.msgid.link/20250509184641.work.542-kees@kernel.org Signed-off-by: Johannes Berg --- net/mac80211/main.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/net/mac80211/main.c b/net/mac80211/main.c index 741e6c7edcb7..6b6de43d9420 100644 --- a/net/mac80211/main.c +++ b/net/mac80211/main.c @@ -1354,10 +1354,12 @@ int ieee80211_register_hw(struct ieee80211_hw *hw) hw->wiphy->software_iftypes |= BIT(NL80211_IFTYPE_MONITOR); - local->int_scan_req = kzalloc(sizeof(*local->int_scan_req) + - sizeof(void *) * channels, GFP_KERNEL); + local->int_scan_req = kzalloc(struct_size(local->int_scan_req, + channels, channels), + GFP_KERNEL); if (!local->int_scan_req) return -ENOMEM; + local->int_scan_req->n_channels = channels; eth_broadcast_addr(local->int_scan_req->bssid); From 08fb624802d8786253994d8ebdbbcdaa186f04f5 Mon Sep 17 00:00:00 2001 From: Andrew Bresticker Date: Wed, 14 May 2025 10:13:20 -0700 Subject: [PATCH 543/555] irqchip/riscv-imsic: Start local sync timer on correct CPU When starting the local sync timer to synchronize the state of a remote CPU it should be added on the CPU to be synchronized, not the initiating CPU. This results in interrupt delivery being delayed until the timer eventually runs (due to another mask/unmask/migrate operation) on the target CPU. Fixes: 0f67911e821c ("irqchip/riscv-imsic: Separate next and previous pointers in IMSIC vector") Signed-off-by: Andrew Bresticker Signed-off-by: Thomas Gleixner Reviewed-by: Anup Patel Link: https://lore.kernel.org/all/20250514171320.3494917-1-abrestic@rivosinc.com --- drivers/irqchip/irq-riscv-imsic-state.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/irqchip/irq-riscv-imsic-state.c b/drivers/irqchip/irq-riscv-imsic-state.c index bdf5cd2037f2..62f76950a113 100644 --- a/drivers/irqchip/irq-riscv-imsic-state.c +++ b/drivers/irqchip/irq-riscv-imsic-state.c @@ -208,17 +208,17 @@ skip: } #ifdef CONFIG_SMP -static void __imsic_local_timer_start(struct imsic_local_priv *lpriv) +static void __imsic_local_timer_start(struct imsic_local_priv *lpriv, unsigned int cpu) { lockdep_assert_held(&lpriv->lock); if (!timer_pending(&lpriv->timer)) { lpriv->timer.expires = jiffies + 1; - add_timer_on(&lpriv->timer, smp_processor_id()); + add_timer_on(&lpriv->timer, cpu); } } #else -static inline void __imsic_local_timer_start(struct imsic_local_priv *lpriv) +static inline void __imsic_local_timer_start(struct imsic_local_priv *lpriv, unsigned int cpu) { } #endif @@ -233,7 +233,7 @@ void imsic_local_sync_all(bool force_all) if (force_all) bitmap_fill(lpriv->dirty_bitmap, imsic->global.nr_ids + 1); if (!__imsic_local_sync(lpriv)) - __imsic_local_timer_start(lpriv); + __imsic_local_timer_start(lpriv, smp_processor_id()); raw_spin_unlock_irqrestore(&lpriv->lock, flags); } @@ -278,7 +278,7 @@ static void __imsic_remote_sync(struct imsic_local_priv *lpriv, unsigned int cpu return; } - __imsic_local_timer_start(lpriv); + __imsic_local_timer_start(lpriv, cpu); } } #else From 92ec4855034b2c4d13f117558dc73d20581fa9ff Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Wed, 14 May 2025 14:48:05 +0200 Subject: [PATCH 544/555] mlxsw: spectrum_router: Fix use-after-free when deleting GRE net devices The driver only offloads neighbors that are constructed on top of net devices registered by it or their uppers (which are all Ethernet). The device supports GRE encapsulation and decapsulation of forwarded traffic, but the driver will not offload dummy neighbors constructed on top of GRE net devices as they are not uppers of its net devices: # ip link add name gre1 up type gre tos inherit local 192.0.2.1 remote 198.51.100.1 # ip neigh add 0.0.0.0 lladdr 0.0.0.0 nud noarp dev gre1 $ ip neigh show dev gre1 nud noarp 0.0.0.0 lladdr 0.0.0.0 NOARP (Note that the neighbor is not marked with 'offload') When the driver is reloaded and the existing configuration is replayed, the driver does not perform the same check regarding existing neighbors and offloads the previously added one: # devlink dev reload pci/0000:01:00.0 $ ip neigh show dev gre1 nud noarp 0.0.0.0 lladdr 0.0.0.0 offload NOARP If the neighbor is later deleted, the driver will ignore the notification (given the GRE net device is not its upper) and will therefore keep referencing freed memory, resulting in a use-after-free [1] when the net device is deleted: # ip neigh del 0.0.0.0 lladdr 0.0.0.0 dev gre1 # ip link del dev gre1 Fix by skipping neighbor replay if the net device for which the replay is performed is not our upper. [1] BUG: KASAN: slab-use-after-free in mlxsw_sp_neigh_entry_update+0x1ea/0x200 Read of size 8 at addr ffff888155b0e420 by task ip/2282 [...] Call Trace: dump_stack_lvl+0x6f/0xa0 print_address_description.constprop.0+0x6f/0x350 print_report+0x108/0x205 kasan_report+0xdf/0x110 mlxsw_sp_neigh_entry_update+0x1ea/0x200 mlxsw_sp_router_rif_gone_sync+0x2a8/0x440 mlxsw_sp_rif_destroy+0x1e9/0x750 mlxsw_sp_netdevice_ipip_ol_event+0x3c9/0xdc0 mlxsw_sp_router_netdevice_event+0x3ac/0x15e0 notifier_call_chain+0xca/0x150 call_netdevice_notifiers_info+0x7f/0x100 unregister_netdevice_many_notify+0xc8c/0x1d90 rtnl_dellink+0x34e/0xa50 rtnetlink_rcv_msg+0x6fb/0xb70 netlink_rcv_skb+0x131/0x360 netlink_unicast+0x426/0x710 netlink_sendmsg+0x75a/0xc20 __sock_sendmsg+0xc1/0x150 ____sys_sendmsg+0x5aa/0x7b0 ___sys_sendmsg+0xfc/0x180 __sys_sendmsg+0x121/0x1b0 do_syscall_64+0xbb/0x1d0 entry_SYSCALL_64_after_hwframe+0x4b/0x53 Fixes: 8fdb09a7674c ("mlxsw: spectrum_router: Replay neighbours when RIF is made") Signed-off-by: Ido Schimmel Reviewed-by: Petr Machata Signed-off-by: Petr Machata Link: https://patch.msgid.link/c53c02c904fde32dad484657be3b1477884e9ad6.1747225701.git.petrm@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c index 464821dd492d..a2033837182e 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c @@ -3014,6 +3014,9 @@ static int mlxsw_sp_neigh_rif_made_sync(struct mlxsw_sp *mlxsw_sp, .rif = rif, }; + if (!mlxsw_sp_dev_lower_is_port(mlxsw_sp_rif_dev(rif))) + return 0; + neigh_for_each(&arp_tbl, mlxsw_sp_neigh_rif_made_sync_each, &rms); if (rms.err) goto err_arp; From 325eb217e41fa14f307c7cc702bd18d0bb38fe84 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Tue, 13 May 2025 23:29:08 -0700 Subject: [PATCH 545/555] bnxt_en: bring back rtnl_lock() in the bnxt_open() path Error recovery, PCIe AER, resume, and TX timeout will invoke bnxt_open() with netdev_lock only. This will cause RTNL assert failure in netif_set_real_num_tx_queues(), netif_set_real_num_tx_queues(), and netif_set_real_num_tx_queues(). Example error recovery assert: RTNL: assertion failed at net/core/dev.c (3178) WARNING: CPU: 3 PID: 3392 at net/core/dev.c:3178 netif_set_real_num_tx_queues+0x1fd/0x210 Call Trace: ? __pfx_bnxt_msix+0x10/0x10 [bnxt_en] __bnxt_open_nic+0x1ef/0xb20 [bnxt_en] bnxt_open+0xda/0x130 [bnxt_en] bnxt_fw_reset_task+0x21f/0x780 [bnxt_en] process_scheduled_works+0x9d/0x400 For now, bring back rtnl_lock() in all these code paths that can invoke bnxt_open(). In the bnxt_queue_start() error path, we don't have rtnl_lock held so we just change it to call netif_close() instead of bnxt_reset_task() for simplicity. This error path is unlikely so it should be fine. Fixes: 004b5008016a ("eth: bnxt: remove most dependencies on RTNL") Reviewed-by: Kalesh AP Reviewed-by: Pavan Chebbi Reviewed-by: Andy Gospodarek Signed-off-by: Michael Chan Link: https://patch.msgid.link/20250514062908.2766677-1-michael.chan@broadcom.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 36 ++++++++++++++++++----- 1 file changed, 29 insertions(+), 7 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 86a5de44b6f3..6afc2ab6fad2 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -14013,13 +14013,28 @@ static void bnxt_unlock_sp(struct bnxt *bp) netdev_unlock(bp->dev); } +/* Same as bnxt_lock_sp() with additional rtnl_lock */ +static void bnxt_rtnl_lock_sp(struct bnxt *bp) +{ + clear_bit(BNXT_STATE_IN_SP_TASK, &bp->state); + rtnl_lock(); + netdev_lock(bp->dev); +} + +static void bnxt_rtnl_unlock_sp(struct bnxt *bp) +{ + set_bit(BNXT_STATE_IN_SP_TASK, &bp->state); + netdev_unlock(bp->dev); + rtnl_unlock(); +} + /* Only called from bnxt_sp_task() */ static void bnxt_reset(struct bnxt *bp, bool silent) { - bnxt_lock_sp(bp); + bnxt_rtnl_lock_sp(bp); if (test_bit(BNXT_STATE_OPEN, &bp->state)) bnxt_reset_task(bp, silent); - bnxt_unlock_sp(bp); + bnxt_rtnl_unlock_sp(bp); } /* Only called from bnxt_sp_task() */ @@ -14027,9 +14042,9 @@ static void bnxt_rx_ring_reset(struct bnxt *bp) { int i; - bnxt_lock_sp(bp); + bnxt_rtnl_lock_sp(bp); if (!test_bit(BNXT_STATE_OPEN, &bp->state)) { - bnxt_unlock_sp(bp); + bnxt_rtnl_unlock_sp(bp); return; } /* Disable and flush TPA before resetting the RX ring */ @@ -14068,7 +14083,7 @@ static void bnxt_rx_ring_reset(struct bnxt *bp) } if (bp->flags & BNXT_FLAG_TPA) bnxt_set_tpa(bp, true); - bnxt_unlock_sp(bp); + bnxt_rtnl_unlock_sp(bp); } static void bnxt_fw_fatal_close(struct bnxt *bp) @@ -14960,15 +14975,17 @@ static void bnxt_fw_reset_task(struct work_struct *work) bp->fw_reset_state = BNXT_FW_RESET_STATE_OPENING; fallthrough; case BNXT_FW_RESET_STATE_OPENING: - while (!netdev_trylock(bp->dev)) { + while (!rtnl_trylock()) { bnxt_queue_fw_reset_work(bp, HZ / 10); return; } + netdev_lock(bp->dev); rc = bnxt_open(bp->dev); if (rc) { netdev_err(bp->dev, "bnxt_open() failed during FW reset\n"); bnxt_fw_reset_abort(bp, rc); netdev_unlock(bp->dev); + rtnl_unlock(); goto ulp_start; } @@ -14988,6 +15005,7 @@ static void bnxt_fw_reset_task(struct work_struct *work) bnxt_dl_health_fw_status_update(bp, true); } netdev_unlock(bp->dev); + rtnl_unlock(); bnxt_ulp_start(bp, 0); bnxt_reenable_sriov(bp); netdev_lock(bp->dev); @@ -15936,7 +15954,7 @@ err_reset: rc); napi_enable_locked(&bnapi->napi); bnxt_db_nq_arm(bp, &cpr->cp_db, cpr->cp_raw_cons); - bnxt_reset_task(bp, true); + netif_close(dev); return rc; } @@ -16752,6 +16770,7 @@ static int bnxt_resume(struct device *device) struct bnxt *bp = netdev_priv(dev); int rc = 0; + rtnl_lock(); netdev_lock(dev); rc = pci_enable_device(bp->pdev); if (rc) { @@ -16796,6 +16815,7 @@ static int bnxt_resume(struct device *device) resume_exit: netdev_unlock(bp->dev); + rtnl_unlock(); bnxt_ulp_start(bp, rc); if (!rc) bnxt_reenable_sriov(bp); @@ -16961,6 +16981,7 @@ static void bnxt_io_resume(struct pci_dev *pdev) int err; netdev_info(bp->dev, "PCI Slot Resume\n"); + rtnl_lock(); netdev_lock(netdev); err = bnxt_hwrm_func_qcaps(bp); @@ -16978,6 +16999,7 @@ static void bnxt_io_resume(struct pci_dev *pdev) netif_device_attach(netdev); netdev_unlock(netdev); + rtnl_unlock(); bnxt_ulp_start(bp, err); if (!err) bnxt_reenable_sriov(bp); From 491deb9b8c4ad12fe51d554a69b8165b9ef9429f Mon Sep 17 00:00:00 2001 From: Pengtao He Date: Wed, 14 May 2025 21:20:13 +0800 Subject: [PATCH 546/555] net/tls: fix kernel panic when alloc_page failed We cannot set frag_list to NULL pointer when alloc_page failed. It will be used in tls_strp_check_queue_ok when the next time tls_strp_read_sock is called. This is because we don't reset full_len in tls_strp_flush_anchor_copy() so the recv path will try to continue handling the partial record on the next call but we dettached the rcvq from the frag list. Alternative fix would be to reset full_len. Unable to handle kernel NULL pointer dereference at virtual address 0000000000000028 Call trace: tls_strp_check_rcv+0x128/0x27c tls_strp_data_ready+0x34/0x44 tls_data_ready+0x3c/0x1f0 tcp_data_ready+0x9c/0xe4 tcp_data_queue+0xf6c/0x12d0 tcp_rcv_established+0x52c/0x798 Fixes: 84c61fe1a75b ("tls: rx: do not use the standard strparser") Signed-off-by: Pengtao He Link: https://patch.msgid.link/20250514132013.17274-1-hept.hept.hept@gmail.com Signed-off-by: Jakub Kicinski --- net/tls/tls_strp.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/tls/tls_strp.c b/net/tls/tls_strp.c index 77e33e1e340e..65b0da6fdf6a 100644 --- a/net/tls/tls_strp.c +++ b/net/tls/tls_strp.c @@ -396,7 +396,6 @@ static int tls_strp_read_copy(struct tls_strparser *strp, bool qshort) return 0; shinfo = skb_shinfo(strp->anchor); - shinfo->frag_list = NULL; /* If we don't know the length go max plus page for cipher overhead */ need_spc = strp->stm.full_len ?: TLS_MAX_PAYLOAD_SIZE + PAGE_SIZE; @@ -412,6 +411,8 @@ static int tls_strp_read_copy(struct tls_strparser *strp, bool qshort) page, 0, 0); } + shinfo->frag_list = NULL; + strp->copy_mode = 1; strp->stm.offset = 0; From b3ca9eef6646576ad506a96d941d87a69f66732a Mon Sep 17 00:00:00 2001 From: Gerhard Engleder Date: Wed, 14 May 2025 21:56:57 +0200 Subject: [PATCH 547/555] tsnep: fix timestamping with a stacked DSA driver This driver is susceptible to a form of the bug explained in commit c26a2c2ddc01 ("gianfar: Fix TX timestamping with a stacked DSA driver") and in Documentation/networking/timestamping.rst section "Other caveats for MAC drivers", specifically it timestamps any skb which has SKBTX_HW_TSTAMP, and does not consider if timestamping has been enabled in adapter->hwtstamp_config.tx_type. Evaluate the proper TX timestamping condition only once on the TX path (in tsnep_xmit_frame_ring()) and store the result in an additional TX entry flag. Evaluate the new TX entry flag in the TX confirmation path (in tsnep_tx_poll()). This way SKBTX_IN_PROGRESS is set by the driver as required, but never evaluated. SKBTX_IN_PROGRESS shall not be evaluated as it can be set by a stacked DSA driver and evaluating it would lead to unwanted timestamps. Fixes: 403f69bbdbad ("tsnep: Add TSN endpoint Ethernet MAC driver") Suggested-by: Vladimir Oltean Signed-off-by: Gerhard Engleder Reviewed-by: Vladimir Oltean Link: https://patch.msgid.link/20250514195657.25874-1-gerhard@engleder-embedded.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/engleder/tsnep_main.c | 30 ++++++++++++++-------- 1 file changed, 19 insertions(+), 11 deletions(-) diff --git a/drivers/net/ethernet/engleder/tsnep_main.c b/drivers/net/ethernet/engleder/tsnep_main.c index 625245b0845c..eba73246f986 100644 --- a/drivers/net/ethernet/engleder/tsnep_main.c +++ b/drivers/net/ethernet/engleder/tsnep_main.c @@ -67,6 +67,8 @@ #define TSNEP_TX_TYPE_XDP_NDO_MAP_PAGE (TSNEP_TX_TYPE_XDP_NDO | TSNEP_TX_TYPE_MAP_PAGE) #define TSNEP_TX_TYPE_XDP (TSNEP_TX_TYPE_XDP_TX | TSNEP_TX_TYPE_XDP_NDO) #define TSNEP_TX_TYPE_XSK BIT(12) +#define TSNEP_TX_TYPE_TSTAMP BIT(13) +#define TSNEP_TX_TYPE_SKB_TSTAMP (TSNEP_TX_TYPE_SKB | TSNEP_TX_TYPE_TSTAMP) #define TSNEP_XDP_TX BIT(0) #define TSNEP_XDP_REDIRECT BIT(1) @@ -386,8 +388,7 @@ static void tsnep_tx_activate(struct tsnep_tx *tx, int index, int length, if (entry->skb) { entry->properties = length & TSNEP_DESC_LENGTH_MASK; entry->properties |= TSNEP_DESC_INTERRUPT_FLAG; - if ((entry->type & TSNEP_TX_TYPE_SKB) && - (skb_shinfo(entry->skb)->tx_flags & SKBTX_IN_PROGRESS)) + if ((entry->type & TSNEP_TX_TYPE_SKB_TSTAMP) == TSNEP_TX_TYPE_SKB_TSTAMP) entry->properties |= TSNEP_DESC_EXTENDED_WRITEBACK_FLAG; /* toggle user flag to prevent false acknowledge @@ -479,7 +480,8 @@ static int tsnep_tx_map_frag(skb_frag_t *frag, struct tsnep_tx_entry *entry, return mapped; } -static int tsnep_tx_map(struct sk_buff *skb, struct tsnep_tx *tx, int count) +static int tsnep_tx_map(struct sk_buff *skb, struct tsnep_tx *tx, int count, + bool do_tstamp) { struct device *dmadev = tx->adapter->dmadev; struct tsnep_tx_entry *entry; @@ -505,6 +507,9 @@ static int tsnep_tx_map(struct sk_buff *skb, struct tsnep_tx *tx, int count) entry->type = TSNEP_TX_TYPE_SKB_INLINE; mapped = 0; } + + if (do_tstamp) + entry->type |= TSNEP_TX_TYPE_TSTAMP; } else { skb_frag_t *frag = &skb_shinfo(skb)->frags[i - 1]; @@ -558,11 +563,12 @@ static int tsnep_tx_unmap(struct tsnep_tx *tx, int index, int count) static netdev_tx_t tsnep_xmit_frame_ring(struct sk_buff *skb, struct tsnep_tx *tx) { - int count = 1; struct tsnep_tx_entry *entry; + bool do_tstamp = false; + int count = 1; int length; - int i; int retval; + int i; if (skb_shinfo(skb)->nr_frags > 0) count += skb_shinfo(skb)->nr_frags; @@ -579,7 +585,13 @@ static netdev_tx_t tsnep_xmit_frame_ring(struct sk_buff *skb, entry = &tx->entry[tx->write]; entry->skb = skb; - retval = tsnep_tx_map(skb, tx, count); + if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) && + tx->adapter->hwtstamp_config.tx_type == HWTSTAMP_TX_ON) { + skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS; + do_tstamp = true; + } + + retval = tsnep_tx_map(skb, tx, count, do_tstamp); if (retval < 0) { tsnep_tx_unmap(tx, tx->write, count); dev_kfree_skb_any(entry->skb); @@ -591,9 +603,6 @@ static netdev_tx_t tsnep_xmit_frame_ring(struct sk_buff *skb, } length = retval; - if (skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) - skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS; - for (i = 0; i < count; i++) tsnep_tx_activate(tx, (tx->write + i) & TSNEP_RING_MASK, length, i == count - 1); @@ -844,8 +853,7 @@ static bool tsnep_tx_poll(struct tsnep_tx *tx, int napi_budget) length = tsnep_tx_unmap(tx, tx->read, count); - if ((entry->type & TSNEP_TX_TYPE_SKB) && - (skb_shinfo(entry->skb)->tx_flags & SKBTX_IN_PROGRESS) && + if (((entry->type & TSNEP_TX_TYPE_SKB_TSTAMP) == TSNEP_TX_TYPE_SKB_TSTAMP) && (__le32_to_cpu(entry->desc_wb->properties) & TSNEP_DESC_EXTENDED_WRITEBACK_FLAG)) { struct skb_shared_hwtstamps hwtstamps; From 0afc44d8cdf6029cce0a92873f0de5ac9416cec8 Mon Sep 17 00:00:00 2001 From: Taehee Yoo Date: Wed, 14 May 2025 15:40:28 +0000 Subject: [PATCH 548/555] net: devmem: fix kernel panic when netlink socket close after module unload Kernel panic occurs when a devmem TCP socket is closed after NIC module is unloaded. This is Devmem TCP unregistration scenarios. number is an order. (a)netlink socket close (b)pp destroy (c)uninstall result 1 2 3 OK 1 3 2 (d)Impossible 2 1 3 OK 3 1 2 (e)Kernel panic 2 3 1 (d)Impossible 3 2 1 (d)Impossible (a) netdev_nl_sock_priv_destroy() is called when devmem TCP socket is closed. (b) page_pool_destroy() is called when the interface is down. (c) mp_ops->uninstall() is called when an interface is unregistered. (d) There is no scenario in mp_ops->uninstall() is called before page_pool_destroy(). Because unregister_netdevice_many_notify() closes interfaces first and then calls mp_ops->uninstall(). (e) netdev_nl_sock_priv_destroy() accesses struct net_device to acquire netdev_lock(). But if the interface module has already been removed, net_device pointer is invalid, so it causes kernel panic. In summary, there are only 3 possible scenarios. A. sk close -> pp destroy -> uninstall. B. pp destroy -> sk close -> uninstall. C. pp destroy -> uninstall -> sk close. Case C is a kernel panic scenario. In order to fix this problem, It makes mp_dmabuf_devmem_uninstall() set binding->dev to NULL. It indicates an bound net_device was unregistered. It makes netdev_nl_sock_priv_destroy() do not acquire netdev_lock() if binding->dev is NULL. A new binding->lock is added to protect a dev of a binding. So, lock ordering is like below. priv->lock netdev_lock(dev) binding->lock Tests: Scenario A: ./ncdevmem -s 192.168.1.4 -c 192.168.1.2 -f $interface -l -p 8000 \ -v 7 -t 1 -q 1 & pid=$! sleep 10 kill $pid ip link set $interface down modprobe -rv $module Scenario B: ./ncdevmem -s 192.168.1.4 -c 192.168.1.2 -f $interface -l -p 8000 \ -v 7 -t 1 -q 1 & pid=$! sleep 10 ip link set $interface down kill $pid modprobe -rv $module Scenario C: ./ncdevmem -s 192.168.1.4 -c 192.168.1.2 -f $interface -l -p 8000 \ -v 7 -t 1 -q 1 & pid=$! sleep 10 modprobe -rv $module sleep 5 kill $pid Splat looks like: Oops: general protection fault, probably for non-canonical address 0xdffffc001fffa9f7: 0000 [#1] SMP DEBUG_PAGEALLOC KASAN NOPTI KASAN: probably user-memory-access in range [0x00000000fffd4fb8-0x00000000fffd4fbf] CPU: 0 UID: 0 PID: 2041 Comm: ncdevmem Tainted: G B W 6.15.0-rc1+ #2 PREEMPT(undef) 0947ec89efa0fd68838b78e36aa1617e97ff5d7f Tainted: [B]=BAD_PAGE, [W]=WARN RIP: 0010:__mutex_lock (./include/linux/sched.h:2244 kernel/locking/mutex.c:400 kernel/locking/mutex.c:443 kernel/locking/mutex.c:605 kernel/locking/mutex.c:746) Code: ea 03 80 3c 02 00 0f 85 4f 13 00 00 49 8b 1e 48 83 e3 f8 74 6a 48 b8 00 00 00 00 00 fc ff df 48 8d 7b 34 48 89 fa 48 c1 ea 03 <0f> b6 f RSP: 0018:ffff88826f7ef730 EFLAGS: 00010203 RAX: dffffc0000000000 RBX: 00000000fffd4f88 RCX: ffffffffaa9bc811 RDX: 000000001fffa9f7 RSI: 0000000000000008 RDI: 00000000fffd4fbc RBP: ffff88826f7ef8b0 R08: 0000000000000000 R09: ffffed103e6aa1a4 R10: 0000000000000007 R11: ffff88826f7ef442 R12: fffffbfff669f65e R13: ffff88812a830040 R14: ffff8881f3550d20 R15: 00000000fffd4f88 FS: 0000000000000000(0000) GS:ffff888866c05000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000563bed0cb288 CR3: 00000001a7c98000 CR4: 00000000007506f0 PKRU: 55555554 Call Trace: ... netdev_nl_sock_priv_destroy (net/core/netdev-genl.c:953 (discriminator 3)) genl_release (net/netlink/genetlink.c:653 net/netlink/genetlink.c:694 net/netlink/genetlink.c:705) ... netlink_release (net/netlink/af_netlink.c:737) ... __sock_release (net/socket.c:647) sock_close (net/socket.c:1393) Fixes: 1d22d3060b9b ("net: drop rtnl_lock for queue_mgmt operations") Signed-off-by: Taehee Yoo Acked-by: Stanislav Fomichev Link: https://patch.msgid.link/20250514154028.1062909-1-ap420073@gmail.com Signed-off-by: Jakub Kicinski --- net/core/devmem.c | 7 +++++++ net/core/devmem.h | 2 ++ net/core/netdev-genl.c | 11 +++++++++++ 3 files changed, 20 insertions(+) diff --git a/net/core/devmem.c b/net/core/devmem.c index 6e27a47d0493..2db428ab6b8b 100644 --- a/net/core/devmem.c +++ b/net/core/devmem.c @@ -200,6 +200,8 @@ net_devmem_bind_dmabuf(struct net_device *dev, unsigned int dmabuf_fd, refcount_set(&binding->ref, 1); + mutex_init(&binding->lock); + binding->dmabuf = dmabuf; binding->attachment = dma_buf_attach(binding->dmabuf, dev->dev.parent); @@ -379,6 +381,11 @@ static void mp_dmabuf_devmem_uninstall(void *mp_priv, xa_for_each(&binding->bound_rxqs, xa_idx, bound_rxq) { if (bound_rxq == rxq) { xa_erase(&binding->bound_rxqs, xa_idx); + if (xa_empty(&binding->bound_rxqs)) { + mutex_lock(&binding->lock); + binding->dev = NULL; + mutex_unlock(&binding->lock); + } break; } } diff --git a/net/core/devmem.h b/net/core/devmem.h index 7fc158d52729..a1aabc9685cc 100644 --- a/net/core/devmem.h +++ b/net/core/devmem.h @@ -20,6 +20,8 @@ struct net_devmem_dmabuf_binding { struct sg_table *sgt; struct net_device *dev; struct gen_pool *chunk_pool; + /* Protect dev */ + struct mutex lock; /* The user holds a ref (via the netlink API) for as long as they want * the binding to remain alive. Each page pool using this binding holds diff --git a/net/core/netdev-genl.c b/net/core/netdev-genl.c index dae9f0d432fb..a877693fecd6 100644 --- a/net/core/netdev-genl.c +++ b/net/core/netdev-genl.c @@ -979,14 +979,25 @@ void netdev_nl_sock_priv_destroy(struct netdev_nl_sock *priv) { struct net_devmem_dmabuf_binding *binding; struct net_devmem_dmabuf_binding *temp; + netdevice_tracker dev_tracker; struct net_device *dev; mutex_lock(&priv->lock); list_for_each_entry_safe(binding, temp, &priv->bindings, list) { + mutex_lock(&binding->lock); dev = binding->dev; + if (!dev) { + mutex_unlock(&binding->lock); + net_devmem_unbind_dmabuf(binding); + continue; + } + netdev_hold(dev, &dev_tracker, GFP_KERNEL); + mutex_unlock(&binding->lock); + netdev_lock(dev); net_devmem_unbind_dmabuf(binding); netdev_unlock(dev); + netdev_put(dev, &dev_tracker); } mutex_unlock(&priv->lock); } From 99bcd91fabada0dbb1d5f0de44532d8008db93c6 Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Thu, 8 May 2025 16:44:52 +0300 Subject: [PATCH 549/555] perf/x86/intel: Fix segfault with PEBS-via-PT with sample_freq Currently, using PEBS-via-PT with a sample frequency instead of a sample period, causes a segfault. For example: BUG: kernel NULL pointer dereference, address: 0000000000000195 ? __die_body.cold+0x19/0x27 ? page_fault_oops+0xca/0x290 ? exc_page_fault+0x7e/0x1b0 ? asm_exc_page_fault+0x26/0x30 ? intel_pmu_pebs_event_update_no_drain+0x40/0x60 ? intel_pmu_pebs_event_update_no_drain+0x32/0x60 intel_pmu_drain_pebs_icl+0x333/0x350 handle_pmi_common+0x272/0x3c0 intel_pmu_handle_irq+0x10a/0x2e0 perf_event_nmi_handler+0x2a/0x50 That happens because intel_pmu_pebs_event_update_no_drain() assumes all the pebs_enabled bits represent counter indexes, which is not always the case. In this particular case, bits 60 and 61 are set for PEBS-via-PT purposes. The behaviour of PEBS-via-PT with sample frequency is questionable because although a PMI is generated (PEBS_PMI_AFTER_EACH_RECORD), the period is not adjusted anyway. Putting that aside, fix intel_pmu_pebs_event_update_no_drain() by passing the mask of counter bits instead of 'size'. Note, prior to the Fixes commit, 'size' would be limited to the maximum counter index, so the issue was not hit. Fixes: 722e42e45c2f1 ("perf/x86: Support counter mask") Signed-off-by: Adrian Hunter Signed-off-by: Ingo Molnar Reviewed-by: Kan Liang Cc: Peter Zijlstra Cc: Ingo Molnar Cc: Alexander Shishkin Cc: Arnaldo Carvalho de Melo Cc: Jiri Olsa Cc: Namhyung Kim Cc: Ian Rogers Cc: linux-perf-users@vger.kernel.org Link: https://lore.kernel.org/r/20250508134452.73960-1-adrian.hunter@intel.com --- arch/x86/events/intel/ds.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/arch/x86/events/intel/ds.c b/arch/x86/events/intel/ds.c index 9b20acc0e932..8d86e91bd5e5 100644 --- a/arch/x86/events/intel/ds.c +++ b/arch/x86/events/intel/ds.c @@ -2465,8 +2465,9 @@ static void intel_pmu_drain_pebs_core(struct pt_regs *iregs, struct perf_sample_ setup_pebs_fixed_sample_data); } -static void intel_pmu_pebs_event_update_no_drain(struct cpu_hw_events *cpuc, int size) +static void intel_pmu_pebs_event_update_no_drain(struct cpu_hw_events *cpuc, u64 mask) { + u64 pebs_enabled = cpuc->pebs_enabled & mask; struct perf_event *event; int bit; @@ -2477,7 +2478,7 @@ static void intel_pmu_pebs_event_update_no_drain(struct cpu_hw_events *cpuc, int * It needs to call intel_pmu_save_and_restart_reload() to * update the event->count for this case. */ - for_each_set_bit(bit, (unsigned long *)&cpuc->pebs_enabled, size) { + for_each_set_bit(bit, (unsigned long *)&pebs_enabled, X86_PMC_IDX_MAX) { event = cpuc->events[bit]; if (event->hw.flags & PERF_X86_EVENT_AUTO_RELOAD) intel_pmu_save_and_restart_reload(event, 0); @@ -2512,7 +2513,7 @@ static void intel_pmu_drain_pebs_nhm(struct pt_regs *iregs, struct perf_sample_d } if (unlikely(base >= top)) { - intel_pmu_pebs_event_update_no_drain(cpuc, size); + intel_pmu_pebs_event_update_no_drain(cpuc, mask); return; } @@ -2626,7 +2627,7 @@ static void intel_pmu_drain_pebs_icl(struct pt_regs *iregs, struct perf_sample_d (hybrid(cpuc->pmu, fixed_cntr_mask64) << INTEL_PMC_IDX_FIXED); if (unlikely(base >= top)) { - intel_pmu_pebs_event_update_no_drain(cpuc, X86_PMC_IDX_MAX); + intel_pmu_pebs_event_update_no_drain(cpuc, mask); return; } From dd24f87f65c957f30e605e44961d2fd53a44c780 Mon Sep 17 00:00:00 2001 From: Ming Lei Date: Fri, 16 May 2025 00:26:01 +0800 Subject: [PATCH 550/555] ublk: fix dead loop when canceling io command Commit: f40139fde527 ("ublk: fix race between io_uring_cmd_complete_in_task and ublk_cancel_cmd") adds a request state check in ublk_cancel_cmd(), and if the request is started, skips canceling this uring_cmd. However, the current uring_cmd may be in ACTIVE state, without block request coming to the uring command. Meantime, if the cached request in tag_set.tags[tag] has been delivered to ublk server and reycycled, then this uring_cmd can't be canceled. ublk requests are aborted in ublk char device release handler, which depends on canceling all ACTIVE uring_cmd. So it causes a dead loop. Fix this issue by not taking a stale request into account when canceling uring_cmd in ublk_cancel_cmd(). Reported-by: Shinichiro Kawasaki Closes: https://lore.kernel.org/linux-block/mruqwpf4tqenkbtgezv5oxwq7ngyq24jzeyqy4ixzvivatbbxv@4oh2wzz4e6qn/ Fixes: f40139fde527 ("ublk: fix race between io_uring_cmd_complete_in_task and ublk_cancel_cmd") Signed-off-by: Ming Lei Link: https://lore.kernel.org/r/20250515162601.77346-1-ming.lei@redhat.com [axboe: rewording of commit message] Signed-off-by: Jens Axboe --- drivers/block/ublk_drv.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/block/ublk_drv.c b/drivers/block/ublk_drv.c index f9032076bc06..dc104c025cd5 100644 --- a/drivers/block/ublk_drv.c +++ b/drivers/block/ublk_drv.c @@ -1708,7 +1708,7 @@ static void ublk_cancel_cmd(struct ublk_queue *ubq, unsigned tag, * that ublk_dispatch_req() is always called */ req = blk_mq_tag_to_rq(ub->tag_set.tags[ubq->q_id], tag); - if (req && blk_mq_request_started(req)) + if (req && blk_mq_request_started(req) && req->tag == tag) return; spin_lock(&ubq->cancel_lock); From 03680913744de17fa49e62b1d8f71bab42b0b721 Mon Sep 17 00:00:00 2001 From: Lukas Bulwahn Date: Fri, 16 May 2025 11:08:10 +0200 Subject: [PATCH 551/555] x86/mm: Remove duplicated word in warning message Commit bbeb69ce3013 ("x86/mm: Remove CONFIG_HIGHMEM64G support") introduces a new warning message MSG_HIGHMEM_TRIMMED, which accidentally introduces a duplicated 'for for' in the warning message. Remove this duplicated word. This was noticed while reviewing for references to obsolete kernel build config options. Fixes: bbeb69ce3013 ("x86/mm: Remove CONFIG_HIGHMEM64G support") Signed-off-by: Lukas Bulwahn Signed-off-by: Ingo Molnar Cc: Andy Lutomirski Cc: Arnd Bergmann Cc: Dave Hansen Cc: H. Peter Anvin Cc: Peter Zijlstra Cc: kernel-janitors@vger.kernel.org Link: https://lore.kernel.org/r/20250516090810.556623-1-lukas.bulwahn@redhat.com --- arch/x86/mm/init_32.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c index ad662cc4605c..148eba50265a 100644 --- a/arch/x86/mm/init_32.c +++ b/arch/x86/mm/init_32.c @@ -565,7 +565,7 @@ static void __init lowmem_pfn_init(void) "only %luMB highmem pages available, ignoring highmem size of %luMB!\n" #define MSG_HIGHMEM_TRIMMED \ - "Warning: only 4GB will be used. Support for for CONFIG_HIGHMEM64G was removed!\n" + "Warning: only 4GB will be used. Support for CONFIG_HIGHMEM64G was removed!\n" /* * We have more RAM than fits into lowmem - we try to put it into * highmem, also taking the highmem=x boot parameter into account: From 6d6d7f91cc8c111d40416ac9240a3bb9396c5235 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sat, 10 May 2025 10:50:13 -0400 Subject: [PATCH 552/555] NFSv4/pnfs: Reset the layout state after a layoutreturn If there are still layout segments in the layout plh_return_lsegs list after a layout return, we should be resetting the state to ensure they eventually get returned as well. Fixes: 68f744797edd ("pNFS: Do not free layout segments that are marked for return") Signed-off-by: Trond Myklebust --- fs/nfs/pnfs.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index 10fdd065a61c..fc7c5fb10198 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -745,6 +745,14 @@ pnfs_mark_matching_lsegs_invalid(struct pnfs_layout_hdr *lo, return remaining; } +static void pnfs_reset_return_info(struct pnfs_layout_hdr *lo) +{ + struct pnfs_layout_segment *lseg; + + list_for_each_entry(lseg, &lo->plh_return_segs, pls_list) + pnfs_set_plh_return_info(lo, lseg->pls_range.iomode, 0); +} + static void pnfs_free_returned_lsegs(struct pnfs_layout_hdr *lo, struct list_head *free_me, @@ -1292,6 +1300,7 @@ void pnfs_layoutreturn_free_lsegs(struct pnfs_layout_hdr *lo, pnfs_mark_matching_lsegs_invalid(lo, &freeme, range, seq); pnfs_free_returned_lsegs(lo, &freeme, range, seq); pnfs_set_layout_stateid(lo, stateid, NULL, true); + pnfs_reset_return_info(lo); } else pnfs_mark_layout_stateid_invalid(lo, &freeme); out_unlock: From 28511504f3ac73ebf45cbbe0dafeca1026e9a8f3 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sat, 10 May 2025 11:05:36 -0400 Subject: [PATCH 553/555] NFS/pnfs: Fix the error path in pnfs_layoutreturn_retry_later_locked() If there isn't a valid layout, or the layout stateid has changed, the cleanup after a layout return should clear out the old data. Signed-off-by: Trond Myklebust --- fs/nfs/pnfs.c | 30 +++++++++++++----------------- 1 file changed, 13 insertions(+), 17 deletions(-) diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index fc7c5fb10198..3adb7d0dbec7 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -1254,21 +1254,15 @@ static void pnfs_clear_layoutcommit(struct inode *inode, static void pnfs_layoutreturn_retry_later_locked(struct pnfs_layout_hdr *lo, const nfs4_stateid *arg_stateid, - const struct pnfs_layout_range *range) + const struct pnfs_layout_range *range, + struct list_head *freeme) { - const struct pnfs_layout_segment *lseg; - u32 seq = be32_to_cpu(arg_stateid->seqid); - if (pnfs_layout_is_valid(lo) && - nfs4_stateid_match_other(&lo->plh_stateid, arg_stateid)) { - list_for_each_entry(lseg, &lo->plh_return_segs, pls_list) { - if (pnfs_seqid_is_newer(lseg->pls_seq, seq) || - !pnfs_should_free_range(&lseg->pls_range, range)) - continue; - pnfs_set_plh_return_info(lo, range->iomode, seq); - break; - } - } + nfs4_stateid_match_other(&lo->plh_stateid, arg_stateid)) + pnfs_reset_return_info(lo); + else + pnfs_mark_layout_stateid_invalid(lo, freeme); + pnfs_clear_layoutreturn_waitbit(lo); } void pnfs_layoutreturn_retry_later(struct pnfs_layout_hdr *lo, @@ -1276,11 +1270,12 @@ void pnfs_layoutreturn_retry_later(struct pnfs_layout_hdr *lo, const struct pnfs_layout_range *range) { struct inode *inode = lo->plh_inode; + LIST_HEAD(freeme); spin_lock(&inode->i_lock); - pnfs_layoutreturn_retry_later_locked(lo, arg_stateid, range); - pnfs_clear_layoutreturn_waitbit(lo); + pnfs_layoutreturn_retry_later_locked(lo, arg_stateid, range, &freeme); spin_unlock(&inode->i_lock); + pnfs_free_lseg_list(&freeme); } void pnfs_layoutreturn_free_lsegs(struct pnfs_layout_hdr *lo, @@ -1716,6 +1711,7 @@ void pnfs_roc_release(struct nfs4_layoutreturn_args *args, struct inode *inode = args->inode; const nfs4_stateid *res_stateid = NULL; struct nfs4_xdr_opaque_data *ld_private = args->ld_private; + LIST_HEAD(freeme); switch (ret) { case -NFS4ERR_BADSESSION: @@ -1724,9 +1720,9 @@ void pnfs_roc_release(struct nfs4_layoutreturn_args *args, case -NFS4ERR_NOMATCHING_LAYOUT: spin_lock(&inode->i_lock); pnfs_layoutreturn_retry_later_locked(lo, &args->stateid, - &args->range); - pnfs_clear_layoutreturn_waitbit(lo); + &args->range, &freeme); spin_unlock(&inode->i_lock); + pnfs_free_lseg_list(&freeme); break; case 0: if (res->lrs_present) From dcd21b609d4abc7303f8683bce4f35d78d7d6830 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sun, 27 Apr 2025 18:21:06 -0400 Subject: [PATCH 554/555] NFS: Avoid flushing data while holding directory locks in nfs_rename() The Linux client assumes that all filehandles are non-volatile for renames within the same directory (otherwise sillyrename cannot work). However, the existence of the Linux 'subtree_check' export option has meant that nfs_rename() has always assumed it needs to flush writes before attempting to rename. Since NFSv4 does allow the client to query whether or not the server exhibits this behaviour, and since knfsd does actually set the appropriate flag when 'subtree_check' is enabled on an export, it should be OK to optimise away the write flushing behaviour in the cases where it is clearly not needed. Signed-off-by: Trond Myklebust Reviewed-by: Jeff Layton --- fs/nfs/client.c | 2 ++ fs/nfs/dir.c | 15 ++++++++++++++- include/linux/nfs_fs_sb.h | 12 +++++++++--- 3 files changed, 25 insertions(+), 4 deletions(-) diff --git a/fs/nfs/client.c b/fs/nfs/client.c index 2115c1189c2d..6d63b958c4bb 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -1105,6 +1105,8 @@ struct nfs_server *nfs_create_server(struct fs_context *fc) if (server->namelen == 0 || server->namelen > NFS2_MAXNAMLEN) server->namelen = NFS2_MAXNAMLEN; } + /* Linux 'subtree_check' borkenness mandates this setting */ + server->fh_expire_type = NFS_FH_VOL_RENAME; if (!(fattr->valid & NFS_ATTR_FATTR)) { error = ctx->nfs_mod->rpc_ops->getattr(server, ctx->mntfh, diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index bd23fc736b39..d0e0b435a843 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -2676,6 +2676,18 @@ nfs_unblock_rename(struct rpc_task *task, struct nfs_renamedata *data) unblock_revalidate(new_dentry); } +static bool nfs_rename_is_unsafe_cross_dir(struct dentry *old_dentry, + struct dentry *new_dentry) +{ + struct nfs_server *server = NFS_SB(old_dentry->d_sb); + + if (old_dentry->d_parent != new_dentry->d_parent) + return false; + if (server->fh_expire_type & NFS_FH_RENAME_UNSAFE) + return !(server->fh_expire_type & NFS_FH_NOEXPIRE_WITH_OPEN); + return true; +} + /* * RENAME * FIXME: Some nfsds, like the Linux user space nfsd, may generate a @@ -2763,7 +2775,8 @@ int nfs_rename(struct mnt_idmap *idmap, struct inode *old_dir, } - if (S_ISREG(old_inode->i_mode)) + if (S_ISREG(old_inode->i_mode) && + nfs_rename_is_unsafe_cross_dir(old_dentry, new_dentry)) nfs_sync_inode(old_inode); task = nfs_async_rename(old_dir, new_dir, old_dentry, new_dentry, must_unblock ? nfs_unblock_rename : NULL); diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h index 71319637a84e..ee03f3cef30c 100644 --- a/include/linux/nfs_fs_sb.h +++ b/include/linux/nfs_fs_sb.h @@ -213,6 +213,15 @@ struct nfs_server { char *fscache_uniq; /* Uniquifier (or NULL) */ #endif + /* The following #defines numerically match the NFSv4 equivalents */ +#define NFS_FH_NOEXPIRE_WITH_OPEN (0x1) +#define NFS_FH_VOLATILE_ANY (0x2) +#define NFS_FH_VOL_MIGRATION (0x4) +#define NFS_FH_VOL_RENAME (0x8) +#define NFS_FH_RENAME_UNSAFE (NFS_FH_VOLATILE_ANY | NFS_FH_VOL_RENAME) + u32 fh_expire_type; /* V4 bitmask representing file + handle volatility type for + this filesystem */ u32 pnfs_blksize; /* layout_blksize attr */ #if IS_ENABLED(CONFIG_NFS_V4) u32 attr_bitmask[3];/* V4 bitmask representing the set @@ -236,9 +245,6 @@ struct nfs_server { u32 acl_bitmask; /* V4 bitmask representing the ACEs that are supported on this filesystem */ - u32 fh_expire_type; /* V4 bitmask representing file - handle volatility type for - this filesystem */ struct pnfs_layoutdriver_type *pnfs_curr_ld; /* Active layout driver */ struct rpc_wait_queue roc_rpcwaitq; void *pnfs_ld_data; /* per mount point data */ From a5806cd506af5a7c19bcd596e4708b5c464bfd21 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 18 May 2025 13:57:29 -0700 Subject: [PATCH 555/555] Linux 6.15-rc7 --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index 0eb9e6c49b32..a9edd0303653 100644 --- a/Makefile +++ b/Makefile @@ -2,7 +2,7 @@ VERSION = 6 PATCHLEVEL = 15 SUBLEVEL = 0 -EXTRAVERSION = -rc6 +EXTRAVERSION = -rc7 NAME = Baby Opossum Posse # *DOCUMENTATION*