This driver adds pinctrl/GPIO support for Intel Emmitsburg PCH. The
GPIO controller is based on the next generation GPIO hardware but still
compatible with the one supported by the Intel core pinctrl/GPIO driver.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Now that we have yaml bindings for the thermal subsystem, get rid of the
old bindings (thermal.txt).
Replace all references to thermal.txt in the Documentation with a link
to the appropriate YAML bindings using the following search and replace
pattern:
- If the reference is specific to the thermal-sensor-cells property,
replace with a pointer to thermal-sensor.yaml
- If the reference is to the cooling-cells property, replace with a
pointer to thermal-cooling-devices.yaml
- If the reference is generic thermal bindings, replace with a
reference to thermal*.yaml.
Signed-off-by: Amit Kucheria <amit.kucheria@linaro.org>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Link: https://lore.kernel.org/r/e9aacd33071a00568b67e110fa3bcc4d86d3e1e4.1595245166.git.amit.kucheria@linaro.org
The generic netlink is initialized at subsys_initcall, so far after
the thermal init routine and the thermal generic netlink family
initialization.
On ŝome platforms, that leads to a memory corruption.
The fix was sent to netdev@ to move the genetlink framework
initialization at core_initcall.
Move the thermal core initialization to postcore level which is very
close to core level.
Reported-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Tested-by: Marek Szyprowski <m.szyprowski@samsung.com>
Reviewed-by: Amit Kucheria <amit.kucheria@linaro.org>
Link: https://lore.kernel.org/r/20200717164217.18819-2-daniel.lezcano@linaro.org
The initcalls like to play joke. In our case, the thermal-netlink
initcall is called after the thermal-core initcall but this one sends
a notification before the former is initialized. No issue was spotted,
but it could lead to a memory corruption, so instead of relying on the
core_initcall for the thermal-netlink, let's initialize directly from
the thermal-core init routine, so we have full control of the init
ordering.
Reported-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Tested-by: Marek Szyprowski <m.szyprowski@samsung.com>
Reviewed-by: Amit Kucheria <amit.kucheria@linaro.org>
Link: https://lore.kernel.org/r/20200717164217.18819-1-daniel.lezcano@linaro.org
The generic netlink is initialized far after the netlink protocol
itself at subsys_initcall. The devlink is initialized at the same
level, but after, as shown by a disassembly of the vmlinux:
[ ... ]
374 ffff8000115f22c0 <__initcall_devlink_init4>:
375 ffff8000115f22c4 <__initcall_genl_init4>:
[ ... ]
The function devlink_init() calls genl_register_family() before the
generic netlink subsystem is initialized.
As the generic netlink initcall level is set since 2005, it seems that
was not a problem, but now we have the thermal framework initialized
at the core_initcall level which creates the generic netlink family
and sends a notification which leads to a subtle memory corruption
only detectable when the CONFIG_INIT_ON_ALLOC_DEFAULT_ON option is set
with the earlycon at init time.
The thermal framework needs to be initialized early in order to begin
the mitigation as soon as possible. Moving it to postcore_initcall is
acceptable.
This patch changes the initialization level for the generic netlink
family to the core_initcall and comes after the netlink protocol
initialization.
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Acked-by: David S. Miller <davem@davemloft.net>
Tested-by: Marek Szyprowski <m.szyprowski@samsung.com>
Reviewed-by: Amit Kucheria <amit.kucheria@linaro.org>
Link: https://lore.kernel.org/r/20200715074120.8768-1-daniel.lezcano@linaro.org
The variable ret is being initialized with a value that is never read
and it is being updated later with a new value. The initialization is
redundant and can be removed.
Addresses-Coverity: ("Unused value")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Zhang Rui <rui.zhang@intel.com>
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Link: https://lore.kernel.org/r/20200706140747.489075-1-colin.king@canonical.com
When the network is not configured, the netlink is disabled on all
the system. The thermal framework assumed the netlink is always
opt-in.
Fix this by adding a Kconfig option for the netlink notification,
defaulting to yes and depending on CONFIG_NET.
As the change implies multiple stubs and in order to not pollute the
internal thermal header, the thermal_nelink.h has been added and
included in the thermal_core.h, so this one regain some kind of
clarity.
Reported-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Reviewed-by: Amit Kucheria <amit.kucheria@linaro.org>
Link: https://lore.kernel.org/r/20200707090159.1018-1-daniel.lezcano@linaro.org
Syscon nodes needs at least 2 compatibles to be compliant why yaml documentation.
Signed-off-by: Benjamin Gaignard <benjamin.gaignard@st.com>
Signed-off-by: Alexandre Torgue <alexandre.torgue@st.com>
Syscon nodes needs at least 2 compatibles to be compliant why yaml documentation.
Signed-off-by: Benjamin Gaignard <benjamin.gaignard@st.com>
Signed-off-by: Alexandre Torgue <alexandre.torgue@st.com>
Syscon nodes needs at least 2 compatibles to be compliant why yaml documentation.
Signed-off-by: Benjamin Gaignard <benjamin.gaignard@st.com>
Signed-off-by: Alexandre Torgue <alexandre.torgue@st.com>
Syscon nodes needs at least 2 compatibles to be compliant why yaml documentation.
Signed-off-by: Benjamin Gaignard <benjamin.gaignard@st.com>
Signed-off-by: Alexandre Torgue <alexandre.torgue@st.com>
Since commit ad440432d1 ("dt-bindings: mfd: Ensure 'syscon' has a more specific compatible")
it is required to provide at least 2 compatibles string for syscon node.
This patch document the missing compatibles for stm32 SoCs.
Signed-off-by: Benjamin Gaignard <benjamin.gaignard@st.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Alexandre Torgue <alexandre.torgue@st.com>
Move spi4_pins_a nodes from pinctrl_z to pinctrl as the associated pins
are not in BANK Z.
Fixes: 498a701498 ("ARM: dts: stm32: Add missing pinctrl entries for STM32MP15")
Signed-off-by: Patrick Delaunay <patrick.delaunay@st.com>
Reviewed-by: Manivannan Sadhasivam <mani@kernel.org>
Signed-off-by: Alexandre Torgue <alexandre.torgue@st.com>
Add Thomas and myself as maintainers of the MIPS CPU and GIC IRQchip, MIPS
GIC timer and MIPS CPS CPUidle drivers.
Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Acked-by: Marc Zyngier <maz@kernel.org>
Acked-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
CDMM may be available not only on MIPS R2 architectures, but also on
newer MIPS R5 chips. For instance our P5600 chip has one. Let's mark
the CDMM bus being supported for that MIPS arch too.
Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Reviewed-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Since having and mapping the CDMM block is platform specific, then
instead of just returning a zero-address, lets make the default CDMM
base address search method (mips_cdmm_phys_base()) to do something
useful. For instance to find the address in a dedicated dtb-node in
order to support of-based platforms by default.
Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
It's a Common Device Memory Map controller embedded into the MIPS IP
cores, which dts node is supposed to have compatible and reg properties.
Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Modern device tree bindings are supposed to be created as YAML-files
in accordance with DT schema. This commit replaces MIPS GIC legacy bare
text binding with YAML file. As before the binding file states that the
corresponding dts node is supposed to be compatible with MIPS Global
Interrupt Controller indicated by the "mti,gic" compatible string and
to provide a mandatory interrupt-controller and '#interrupt-cells'
properties. There might be optional registers memory range,
"mti,reserved-cpu-vectors" and "mti,reserved-ipi-vectors" properties
specified.
MIPS GIC also includes a free-running global timer, per-CPU count/compare
timers, and a watchdog. Since currently the GIC Timer is only supported the
DT schema expects an IRQ and clock-phandler charged timer sub-node with
"mti,mips-gic-timer" compatible string.
Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
It's a Cluster Power Controller embedded into the MIPS IP cores.
Currently the corresponding dts node is supposed to have compatible
and reg properties.
Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Reviewed-by: Rob Herring <robh@kernel.org>
Cc: Alexey Malahov <Alexey.Malahov@baikalelectronics.ru>
Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Configure I2C5 on stm32mp15 DK boards. It's available and can be used on:
- Arduino connector
- GPIO expansion connector
Keep it disabled by default, so the pins are kept in their initial state to
lower power consumption. This way they can also be used as GPIO.
Signed-off-by: Fabrice Gasnier <fabrice.gasnier@st.com>
Signed-off-by: Alain Volmat <alain.volmat@st.com>
Signed-off-by: Alexandre Torgue <alexandre.torgue@st.com>
- Add CAAM clock support for vf610 driver.
- Add MU root clock support for imx8mp driver.
-----BEGIN PGP SIGNATURE-----
iQFIBAABCgAyFiEEFmJXigPl4LoGSz08UFdYWoewfM4FAl8VGBEUHHNoYXduZ3Vv
QGtlcm5lbC5vcmcACgkQUFdYWoewfM407AgAlAlMVIHVNmob88YQc9LKb9eG0os3
R/Qrz2dE4VdZbry2iPY3nfI/v+clUUfKB63I3Uj1knjvqMQDZK2vpzQEo9s31nMY
xpAgPxoSzPYuglLKia982N77vehegZ/slSalDKfgdb0qRZFKTHNq4yByUx8pErVL
mA1RSx5VXuelzwOiRg1RMJpuI/ljYM2f6j7wBiENwNdWbxOZyLtzhwTpFqnP06hh
X/2eYoJiqCpryCCDIK1xf2OHBWbN0tYJPQnJmZl+DmMmYChSs10afGfSiFb5waf+
59pGQsiU4I3Epbe66oLhI7Fk8X5NNlU40Zmo4BaDWQLatPLycswBt2xewg==
=C5MK
-----END PGP SIGNATURE-----
Merge tag 'clk-imx-5.9' of git://git.kernel.org/pub/scm/linux/kernel/git/shawnguo/linux into clk-imx
Pull i.MX clk updates from Shawn Guo:
- Add CAAM clock support for vf610 driver
- Add MU root clock support for imx8mp driver
* tag 'clk-imx-5.9' of git://git.kernel.org/pub/scm/linux/kernel/git/shawnguo/linux:
clk: imx: vf610: add CAAM clock
clk: imx8mp: add mu root clk
It is not enough to select WATCHDOG_CORE, the watchdog part of the driver
now also depends on WATCHDOG. This is currently the best we can do because
alarm support and watchdog support are mutually exclusive.
Reported-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Acked-by: Randy Dunlap <rdunlap@infradead.org> # build-tested
Link: https://lore.kernel.org/r/20200720075250.1019172-1-alexandre.belloni@bootlin.com
- Add support for the new RZ/G2H (R8A774E1) SoC.
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQQ9qaHoIs/1I4cXmEiKwlD9ZEnxcAUCXxFrOQAKCRCKwlD9ZEnx
cEGtAQDI+HcEWlrOPm517lQtsAaBOkrPalNCFGX8Nf+rr3yMWgEAsMffVKeS3Rt4
w7jFWd2NfiV1Qxn+73M6QEdWZeyDJgE=
=OCqr
-----END PGP SIGNATURE-----
Merge tag 'clk-renesas-for-v5.9-tag2' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/renesas-drivers into clk-renesas
Pull Renesas clk driver updates from Geert Uytterhoeven:
- Add support for the new RZ/G2H (R8A774E1) SoC
* tag 'clk-renesas-for-v5.9-tag2' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/renesas-drivers:
clk: renesas: cpg-mssr: Add r8a774e1 support
dt-bindings: clock: renesas,cpg-mssr: Document r8a774e1
clk: renesas: Add r8a774e1 CPG Core Clock Definitions
dt-bindings: power: Add r8a774e1 SYSC power domain definitions
* clk-qcom:
clk: qcom: gcc: Make disp gpll0 branch aon for sc7180/sdm845
ipq806x: gcc: add support for child probe
clk: qcom: msm8996: Make symbol 'cpu_msm8996_clks' static
clk: qcom: ipq8074: Add correct index for PCIe clocks
Adds the usart2 node to stm32mp157c-dk2 board. usart2 pins are connected
to Bluetooth component. usart2 is disabled by default.
Signed-off-by: Erwan Le Ray <erwan.leray@st.com>
Signed-off-by: Alexandre Torgue <alexandre.torgue@st.com>
Adds uart7 node to stm32mp15xx-dkx and uart7 alias to stm32mp157a-dk1 and
stm32mp157c-dk2 boards. uart7 pins are connected to Arduino connector.
uart7 is disabled by default.
Signed-off-by: Erwan Le Ray <erwan.leray@st.com>
Signed-off-by: Alexandre Torgue <alexandre.torgue@st.com>
Adds the usart3 node to stm32mp157c-ev1 board. usart3 pins are connected to
GPIO Expansion connector. usart3 is disabled by default.
Signed-off-by: Erwan Le Ray <erwan.leray@st.com>
Signed-off-by: Alexandre Torgue <alexandre.torgue@st.com>
Adds usart3 node to stm32mp15xx-dkx and usart3 alias to stm32mp157a-dk1
and stm32mp157c-dk2 boards. usart3 pins are connected to GPIO Expansion
connector. usart3 is disabled by default.
Signed-off-by: Erwan Le Ray <erwan.leray@st.com>
Signed-off-by: Alexandre Torgue <alexandre.torgue@st.com>
Adds usart2_pins_c, usart3_pins_b, usart3_pins_c and uart7_pins_c pins
configurations in stm32mp15-pinctrl.
- usart2_pins_c pins are connected to Bluetooth chip on dk2 board.
- usart3_pins_b pins are connected to GPIO expansion connector on evx board.
- usart3_pins_c pins are connected to GPIO expansion connector on dkx board.
- uart7_pins_c pins are connected to Arduino Uno connector on dkx board.
Signed-off-by: Erwan Le Ray <erwan.leray@st.com>
Signed-off-by: Alexandre Torgue <alexandre.torgue@st.com>
The display gpll0 branch clock inside GCC needs to always be enabled.
Otherwise the AHB clk (disp_cc_mdss_ahb_clk_src) for the display clk
controller (dispcc) will stop clocking while sourcing from gpll0 when
this branch inside GCC is turned off during unused clk disabling. We can
never turn this branch off because the AHB clk for the display subsystem
is needed to read/write any registers inside the display subsystem
including clk related ones. This makes this branch a really easy way to
turn off AHB access to the display subsystem and cause all sorts of
mayhem. Let's just make the clk ops keep the clk enabled forever and
ignore any attempts to disable this clk so that dispcc accesses keep
working.
Signed-off-by: Taniya Das <tdas@codeaurora.org>
Reported-by: Evan Green <evgreen@chromium.org>
Link: https://lore.kernel.org/r/1594796050-14511-1-git-send-email-tdas@codeaurora.org
Fixes: 17269568f7 ("clk: qcom: Add Global Clock controller (GCC) driver for SC7180")
Fixes: 06391eddb6 ("clk: qcom: Add Global Clock controller (GCC) driver for SDM845")
[sboyd@kernel.org: Fill out commit text more]
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
Comes up every few years, gets somewhat tedious to discuss, let's
write this down once and for all.
What I'm not sure about is whether the text should be more explicit in
flat out mandating the amdkfd eviction fences for long running compute
workloads or workloads where userspace fencing is allowed.
v2: Now with dot graph!
v3: Typo (Dave Airlie)
Reviewed-by: Thomas Hellstrom <thomas.hellstrom@intel.com>
Acked-by: Jason Ekstrand <jason@jlekstrand.net>
Acked-by: Christian König <christian.koenig@amd.com>
Acked-by: Daniel Stone <daniels@collabora.com>
Acked-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Jesse Natalie <jenatali@microsoft.com>
Cc: Steve Pronovost <spronovo@microsoft.com>
Cc: Jason Ekstrand <jason@jlekstrand.net>
Cc: Felix Kuehling <Felix.Kuehling@amd.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Cc: Thomas Hellstrom <thomas.hellstrom@intel.com>
Cc: linux-media@vger.kernel.org
Cc: linaro-mm-sig@lists.linaro.org
Cc: linux-rdma@vger.kernel.org
Cc: amd-gfx@lists.freedesktop.org
Cc: intel-gfx@lists.freedesktop.org
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Christian König <christian.koenig@amd.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200709123339.547390-1-daniel.vetter@ffwll.ch
Two in one go:
- it is allowed to call dma_fence_wait() while holding a
dma_resv_lock(). This is fundamental to how eviction works with ttm,
so required.
- it is allowed to call dma_fence_wait() from memory reclaim contexts,
specifically from shrinker callbacks (which i915 does), and from mmu
notifier callbacks (which amdgpu does, and which i915 sometimes also
does, and probably always should, but that's kinda a debate). Also
for stuff like HMM we really need to be able to do this, or things
get real dicey.
Consequence is that any critical path necessary to get to a
dma_fence_signal for a fence must never a) call dma_resv_lock nor b)
allocate memory with GFP_KERNEL. Also by implication of
dma_resv_lock(), no userspace faulting allowed. That's some supremely
obnoxious limitations, which is why we need to sprinkle the right
annotations to all relevant paths.
The one big locking context we're leaving out here is mmu notifiers,
added in
commit 23b68395c7
Author: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Mon Aug 26 22:14:21 2019 +0200
mm/mmu_notifiers: add a lockdep map for invalidate_range_start/end
that one covers a lot of other callsites, and it's also allowed to
wait on dma-fences from mmu notifiers. But there's no ready-made
functions exposed to prime this, so I've left it out for now.
v2: Also track against mmu notifier context.
v3: kerneldoc to spec the cross-driver contract. Note that currently
i915 throws in a hard-coded 10s timeout on foreign fences (not sure
why that was done, but it's there), which is why that rule is worded
with SHOULD instead of MUST.
Also some of the mmu_notifier/shrinker rules might surprise SoC
drivers, I haven't fully audited them all. Which is infeasible anyway,
we'll need to run them with lockdep and dma-fence annotations and see
what goes boom.
v4: A spelling fix from Mika
v5: #ifdef for CONFIG_MMU_NOTIFIER. Reported by 0day. Unfortunately
this means lockdep enforcement is slightly inconsistent, it won't spot
GFP_NOIO and GFP_NOFS allocations in the wrong spot if
CONFIG_MMU_NOTIFIER is disabled in the kernel config. Oh well.
v5: Note that only drivers/gpu has a reasonable (or at least
historical) excuse to use dma_fence_wait() from shrinker and mmu
notifier callbacks. Everyone else should either have a better memory
manager model, or better hardware. This reflects discussions with
Jason Gunthorpe.
Cc: Jason Gunthorpe <jgg@mellanox.com>
Cc: Felix Kuehling <Felix.Kuehling@amd.com>
Cc: kernel test robot <lkp@intel.com>
Acked-by: Christian König <christian.koenig@amd.com>
Acked-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@intel.com> (v4)
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Cc: Thomas Hellstrom <thomas.hellstrom@intel.com>
Cc: linux-media@vger.kernel.org
Cc: linaro-mm-sig@lists.linaro.org
Cc: linux-rdma@vger.kernel.org
Cc: amd-gfx@lists.freedesktop.org
Cc: intel-gfx@lists.freedesktop.org
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Christian König <christian.koenig@amd.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200707201229.472834-3-daniel.vetter@ffwll.ch
Design is similar to the lockdep annotations for workers, but with
some twists:
- We use a read-lock for the execution/worker/completion side, so that
this explicit annotation can be more liberally sprinkled around.
With read locks lockdep isn't going to complain if the read-side
isn't nested the same way under all circumstances, so ABBA deadlocks
are ok. Which they are, since this is an annotation only.
- We're using non-recursive lockdep read lock mode, since in recursive
read lock mode lockdep does not catch read side hazards. And we
_very_ much want read side hazards to be caught. For full details of
this limitation see
commit e914985897
Author: Peter Zijlstra <peterz@infradead.org>
Date: Wed Aug 23 13:13:11 2017 +0200
locking/lockdep/selftests: Add mixed read-write ABBA tests
- To allow nesting of the read-side explicit annotations we explicitly
keep track of the nesting. lock_is_held() allows us to do that.
- The wait-side annotation is a write lock, and entirely done within
dma_fence_wait() for everyone by default.
- To be able to freely annotate helper functions I want to make it ok
to call dma_fence_begin/end_signalling from soft/hardirq context.
First attempt was using the hardirq locking context for the write
side in lockdep, but this forces all normal spinlocks nested within
dma_fence_begin/end_signalling to be spinlocks. That bollocks.
The approach now is to simple check in_atomic(), and for these cases
entirely rely on the might_sleep() check in dma_fence_wait(). That
will catch any wrong nesting against spinlocks from soft/hardirq
contexts.
The idea here is that every code path that's critical for eventually
signalling a dma_fence should be annotated with
dma_fence_begin/end_signalling. The annotation ideally starts right
after a dma_fence is published (added to a dma_resv, exposed as a
sync_file fd, attached to a drm_syncobj fd, or anything else that
makes the dma_fence visible to other kernel threads), up to and
including the dma_fence_wait(). Examples are irq handlers, the
scheduler rt threads, the tail of execbuf (after the corresponding
fences are visible), any workers that end up signalling dma_fences and
really anything else. Not annotated should be code paths that only
complete fences opportunistically as the gpu progresses, like e.g.
shrinker/eviction code.
The main class of deadlocks this is supposed to catch are:
Thread A:
mutex_lock(A);
mutex_unlock(A);
dma_fence_signal();
Thread B:
mutex_lock(A);
dma_fence_wait();
mutex_unlock(A);
Thread B is blocked on A signalling the fence, but A never gets around
to that because it cannot acquire the lock A.
Note that dma_fence_wait() is allowed to be nested within
dma_fence_begin/end_signalling sections. To allow this to happen the
read lock needs to be upgraded to a write lock, which means that any
other lock is acquired between the dma_fence_begin_signalling() call and
the call to dma_fence_wait(), and still held, this will result in an
immediate lockdep complaint. The only other option would be to not
annotate such calls, defeating the point. Therefore these annotations
cannot be sprinkled over the code entirely mindless to avoid false
positives.
Originally I hope that the cross-release lockdep extensions would
alleviate the need for explicit annotations:
https://lwn.net/Articles/709849/
But there's a few reasons why that's not an option:
- It's not happening in upstream, since it got reverted due to too
many false positives:
commit e966eaeeb6
Author: Ingo Molnar <mingo@kernel.org>
Date: Tue Dec 12 12:31:16 2017 +0100
locking/lockdep: Remove the cross-release locking checks
This code (CONFIG_LOCKDEP_CROSSRELEASE=y and CONFIG_LOCKDEP_COMPLETIONS=y),
while it found a number of old bugs initially, was also causing too many
false positives that caused people to disable lockdep - which is arguably
a worse overall outcome.
- cross-release uses the complete() call to annotate the end of
critical sections, for dma_fence that would be dma_fence_signal().
But we do not want all dma_fence_signal() calls to be treated as
critical, since many are opportunistic cleanup of gpu requests. If
these get stuck there's still the main completion interrupt and
workers who can unblock everyone. Automatically annotating all
dma_fence_signal() calls would hence cause false positives.
- cross-release had some educated guesses for when a critical section
starts, like fresh syscall or fresh work callback. This would again
cause false positives without explicit annotations, since for
dma_fence the critical sections only starts when we publish a fence.
- Furthermore there can be cases where a thread never does a
dma_fence_signal, but is still critical for reaching completion of
fences. One example would be a scheduler kthread which picks up jobs
and pushes them into hardware, where the interrupt handler or
another completion thread calls dma_fence_signal(). But if the
scheduler thread hangs, then all the fences hang, hence we need to
manually annotate it. cross-release aimed to solve this by chaining
cross-release dependencies, but the dependency from scheduler thread
to the completion interrupt handler goes through hw where
cross-release code can't observe it.
In short, without manual annotations and careful review of the start
and end of critical sections, cross-relese dependency tracking doesn't
work. We need explicit annotations.
v2: handle soft/hardirq ctx better against write side and dont forget
EXPORT_SYMBOL, drivers can't use this otherwise.
v3: Kerneldoc.
v4: Some spelling fixes from Mika
v5: Amend commit message to explain in detail why cross-release isn't
the solution.
v6: Pull out misplaced .rst hunk.
Acked-by: Christian König <christian.koenig@amd.com>
Acked-by: Dave Airlie <airlied@redhat.com>
Cc: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Cc: Thomas Hellstrom <thomas.hellstrom@intel.com>
Cc: linux-media@vger.kernel.org
Cc: linaro-mm-sig@lists.linaro.org
Cc: linux-rdma@vger.kernel.org
Cc: amd-gfx@lists.freedesktop.org
Cc: intel-gfx@lists.freedesktop.org
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Christian König <christian.koenig@amd.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200707201229.472834-2-daniel.vetter@ffwll.ch
When running qmi-firmware-update on the Sierra Wireless EM7305 in a Toshiba
laptop, it changed product ID to 0x9062 when entering QDL mode:
usb 2-4: new high-speed USB device number 78 using xhci_hcd
usb 2-4: New USB device found, idVendor=1199, idProduct=9062, bcdDevice= 0.00
usb 2-4: New USB device strings: Mfr=1, Product=2, SerialNumber=0
usb 2-4: Product: EM7305
usb 2-4: Manufacturer: Sierra Wireless, Incorporated
The upgrade could complete after running
# echo 1199 9062 > /sys/bus/usb-serial/drivers/qcserial/new_id
qcserial 2-4:1.0: Qualcomm USB modem converter detected
usb 2-4: Qualcomm USB modem converter now attached to ttyUSB0
Signed-off-by: Erik Ekman <erik@kryo.se>
Link: https://lore.kernel.org/r/20200717185118.3640219-1-erik@kryo.se
Cc: stable@vger.kernel.org
Signed-off-by: Johan Hovold <johan@kernel.org>
The set-led command is eight bytes long and starts with a command byte
followed by six bytes of RGB data and ends with a byte encoding a
frequency (see iuu_led() and iuu_rgbf_fill_buffer()).
The led activity helpers had a few long-standing bugs which corrupted
the command packets by inserting a second command byte and thereby
offsetting the RGB data and dropping the frequency in non-xmas mode.
In xmas mode, a related off-by-one error left the frequency field
uninitialised.
Fixes: 60a8fc0171 ("USB: add iuu_phoenix driver")
Reported-by: George Spelvin <lkml@sdf.org>
Link: https://lore.kernel.org/r/20200716085056.31471-1-johan@kernel.org
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>
The helper doesn't expose any not-mapable memory resources.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://patchwork.freedesktop.org/patch/377649/
Use tabs where possible and remove multiple blanks lines.
Signed-off-by: Patrick Delaunay <patrick.delaunay@st.com>
Signed-off-by: Alexandre Torgue <alexandre.torgue@st.com>
Add support for child probing needed for tsens driver that share the
same regs of gcc for this platform.
Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com>
Reviewed-by: Amit Kucheria <amit.kucheria@linaro.org>
Link: https://lore.kernel.org/r/20200716022817.30439-2-ansuelsmth@gmail.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
We forgot to support the xfrm policy hold queue when
VTI was implemented. This patch adds everything we
need so that we can use the policy hold queue together
with VTI interfaces.
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
The commits "xfrm: Move dst->path into struct xfrm_dst"
and "net: Create and use new helper xfrm_dst_child()."
changed xfrm bundle handling under the assumption
that xdst->path and dst->child are not a NULL pointer
only if dst->xfrm is not a NULL pointer. That is true
with one exception. If the xfrm hold queue is used
to wait until a SA is installed by the key manager,
we create a dummy bundle without a valid dst->xfrm
pointer. The current xfrm bundle handling crashes
in that case. Fix this by extending the NULL check
of dst->xfrm with a test of the DST_XFRM_QUEUE flag.
Fixes: 0f6c480f23 ("xfrm: Move dst->path into struct xfrm_dst")
Fixes: b92cf4aab8 ("net: Create and use new helper xfrm_dst_child().")
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Cristian Ciocaltea