- Fix MIPI vtotal programming off by one on Broxton
- Fix PMU code for GCOV and AutoFDO enabled build
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://lore.kernel.org/r/aFJfykDpUwtmpilE@jlahtine-mobl
Fixes for v6.16-rc3
Display:
- Fixed DP output on SDM845
- Fixed 10nm DSI PLL init
GPU:
- SUBMIT ioctl error path leak fixes
- drm half of stall-on-fault fixes. Note there is a soft dependency,
to get correct mmu fault devcoredumps, on arm-smmu changes which
are not in this branch, but have already been merged by Linus. So
by the time Linus merges this, everything should be peachy.
- a7xx: Missing CP_RESET_CONTEXT_STATE
- Skip GPU component bind if GPU is not in the device table.
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Rob Clark <rob.clark@oss.qualcomm.com>
Link: https://lore.kernel.org/r/CACSVV03=OH74ip8O1xqb8RJWGyM4HFuUnWuR=p3zJR+-ko_AJA@mail.gmail.com
i915_pmu.c may fail to build with GCOV and AutoFDO enabled.
../drivers/gpu/drm/i915/i915_pmu.c:116:3: error: call to '__compiletime_assert_487' declared with 'error' attribute: BUILD_BUG_ON failed: bit > BITS_PER_TYPE(typeof_member(struct i915_pmu, enable)) - 1
116 | BUILD_BUG_ON(bit >
| ^
Here is a way to reproduce the issue:
$ git checkout v6.15
$ mkdir build
$ ./scripts/kconfig/merge_config.sh -O build -n -m <(cat <<EOF
CONFIG_DRM=y
CONFIG_PCI=y
CONFIG_DRM_I915=y
CONFIG_PERF_EVENTS=y
CONFIG_DEBUG_FS=y
CONFIG_GCOV_KERNEL=y
CONFIG_GCOV_PROFILE_ALL=y
CONFIG_AUTOFDO_CLANG=y
EOF
)
$ PATH=${PATH}:${HOME}/llvm-20.1.5-x86_64/bin make LLVM=1 O=build \
olddefconfig
$ PATH=${PATH}:${HOME}/llvm-20.1.5-x86_64/bin make LLVM=1 O=build \
CLANG_AUTOFDO_PROFILE=...PATH_TO_SOME_AFDO_PROFILE... \
drivers/gpu/drm/i915/i915_pmu.o
Although not super sure what happened, by reviewing the code, it should
depend on `__builtin_constant_p(bit)` directly instead of assuming
`__builtin_constant_p(config)` makes `bit` a builtin constant.
Also fix a nit, to reuse the `bit` local variable.
Fixes: a644fde77f ("drm/i915/pmu: Change bitmask of enabled events to u32")
Reviewed-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Tzung-Bi Shih <tzungbi@kernel.org>
Signed-off-by: Tvrtko Ursulin <tursulin@ursulin.net>
Link: https://lore.kernel.org/r/20250612083023.562585-1-tzungbi@kernel.org
(cherry picked from commit 686d773186)
Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
We want to WARN_ON() if info is NULL.
Suggested-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Fixes: 0838fc3e67 ("drm/msm/adreno: Check for recognized GPU before bind")
Tested-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Reported-by: Alexey Klimov <alexey.klimov@linaro.org>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/658631/
In case the BO is in iomem, we can't simply take the vaddr and write to
it. Instead, prepare a separate buffer that is later copied into io
memory. Right now it's just a few words that could be using
xe_map_write32(), but the intention is to grow the WA BB for other
uses.
Fixes: 617d824c53 ("drm/xe: Add WA BB to capture active context utilization")
Cc: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Link: https://lore.kernel.org/r/20250604-wa-bb-fix-v1-1-0dfc5dafcef0@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
(cherry picked from commit ef48715b2d)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
This driver requires of_get_display_timing() from
CONFIG_VIDEOMODE_HELPERS but does not select it. If no other driver
selects it, there will be a failure from the linker if the driver is
built in or modpost if it is a module.
ERROR: modpost: "of_get_display_timing" [drivers/gpu/drm/sitronix/st7571-i2c.ko] undefined!
Select CONFIG_VIDEOMODE_HELPERS to resolve the build failure.
Fixes: 4b35f0f41e ("drm/st7571-i2c: add support for Sitronix ST7571 LCD controller")
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Link: https://lore.kernel.org/r/20250610-drm-st7571-i2c-select-videomode-helpers-v1-1-d30b50ff6e64@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
Commit 1017560164 ("drm/meson: use unsigned long long / Hz for
frequency types") attempts to resolve video playback using 59.94Hz.
using YUV420 by changing the clock calculation to use
Hz instead of kHz (thus yielding more precision).
The basic calculation itself is correct, however the comparisions in
meson_vclk_vic_supported_freq() and meson_vclk_setup() don't work
anymore for 59.94Hz modes (using the freq * 1000 / 1001 logic). For
example, drm/edid specifies a 593407kHz clock for 3840x2160@59.94Hz.
With the mentioend commit we convert this to Hz. Then meson_vclk
tries to find a matchig "params" entry (as the clock setup code
currently only supports specific frequencies) by taking the venc_freq
from the params and calculating the "alt frequency" (used for the
59.94Hz modes) from it, which is:
(594000000Hz * 1000) / 1001 = 593406593Hz
Similar calculation is applied to the phy_freq (TMDS clock), which is 10
times the pixel clock.
Implement a new meson_vclk_freqs_are_matching_param() function whose
purpose is to compare if the requested and calculated frequencies. They
may not match exactly (for the reasons mentioned above). Allow the
clocks to deviate slightly to make the 59.94Hz modes again.
Fixes: 1017560164 ("drm/meson: use unsigned long long / Hz for frequency types")
Reported-by: Christian Hewitt <christianshewitt@gmail.com>
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20250609202751.962208-1-martin.blumenstingl@googlemail.com
meson_vclk_vic_supported_freq() has a debug print which includes the
pixel freq. However, within the whole function the pixel freq is
irrelevant, other than checking the end of the params array. Switch to
printing the vclk_freq which is being compared / matched against the
inputs to the function to avoid confusion when analyzing error reports
from users.
Fixes: e5fab2ec9c ("drm/meson: vclk: add support for YUV420 setup")
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20250606221031.3419353-1-martin.blumenstingl@googlemail.com
The "phy" and "vclk" frequency labels were swapped, making it more
difficult to debug driver errors. Swap the label order to make them
match with the actual frequencies printed to correct this.
Fixes: e5fab2ec9c ("drm/meson: vclk: add support for YUV420 setup")
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20250606203729.3311592-1-martin.blumenstingl@googlemail.com
If we have a newer dtb than kernel, we could end up in a situation where
the GPU device is present in the dtb, but not in the drivers device
table. We don't want this to prevent the display from probing. So
check that we recognize the GPU before adding the GPU component.
v2: use %pOF
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/657701/
We are going to want to re-use this before the component is bound, when
we don't yet have the device pointer (but we do have the of node).
v2: use %pOF
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/657705/
To better match add_gpu_components().
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/657700/
The files generated by gen_header.py capture the source path to the
input files and the date. While that can be informative, it varies
based on where and when the kernel was built as the full path is
captured.
Since all of the files that this tool is run on is under the drivers
directory, this modifies the application to strip all of the path before
drivers. Additionally it prints <stripped> instead of the date.
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
Signed-off-by: Bruce Ashfield <bruce.ashfield@gmail.com>
Signed-off-by: Viswanath Kraleti <viswanath.kraleti@oss.qualcomm.com>
Acked-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/655599/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Calling this packet is necessary when we switch contexts because there
are various pieces of state used by userspace to synchronize between BR
and BV that are persistent across submits and we need to make sure that
they are in a "safe" state when switching contexts. Otherwise a
userspace submission in one context could cause another context to
function incorrectly and hang, effectively a denial of service (although
without leaking data). This was missed during initial a7xx bringup.
Fixes: af66706acc ("drm/msm/a6xx: Add skeleton A7xx support")
Signed-off-by: Connor Abbott <cwabbott0@gmail.com>
Patchwork: https://patchwork.freedesktop.org/patch/654924/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Based on kgsl.
Fixes: af66706acc ("drm/msm/a6xx: Add skeleton A7xx support")
Signed-off-by: Connor Abbott <cwabbott0@gmail.com>
Patchwork: https://patchwork.freedesktop.org/patch/654922/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
When things go wrong, the GPU is capable of quickly generating millions
of faulting translation requests per second. When that happens, in the
stall-on-fault model each access will stall until it wins the race to
signal the fault and then the RESUME register is written. This slows
processing page faults to a crawl as the GPU can generate faults much
faster than the CPU can acknowledge them. It also means that all
available resources in the SMMU are saturated waiting for the stalled
transactions, so that other transactions such as transactions generated
by the GMU, which shares translation resources with the GPU, cannot
proceed. This causes a GMU watchdog timeout, which leads to a failed
reset because GX cannot collapse when there is a transaction pending and
a permanently hung GPU.
On older platforms with qcom,smmu-v2, it seems that when one transaction
is stalled subsequent faulting transactions are terminated, which avoids
this problem, but the MMU-500 follows the spec here.
To work around these problems, disable stall-on-fault as soon as we get a
page fault until a cooldown period after pagefaults stop. This allows
the GMU some guaranteed time to continue working. We only use
stall-on-fault to halt the GPU while we collect a devcoredump and we
always terminate the transaction afterward, so it's fine to miss some
subsequent page faults. We also keep it disabled so long as the current
devcoredump hasn't been deleted, because in that case we likely won't
capture another one if there's a fault.
After this commit HFI messages still occasionally time out, because the
crashdump handler doesn't run fast enough to let the GMU resume, but the
driver seems to recover from it. This will probably go away after the
HFI timeout is increased.
Signed-off-by: Connor Abbott <cwabbott0@gmail.com>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Patchwork: https://patchwork.freedesktop.org/patch/654891/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Unused since the previous commit.
Signed-off-by: Connor Abbott <cwabbott0@gmail.com>
Patchwork: https://patchwork.freedesktop.org/patch/654890/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Now that we use a threaded IRQ, it should be safe to do this in the
fault handler.
We can also remove fault_info from struct msm_gpu and just pass it
directly.
Signed-off-by: Connor Abbott <cwabbott0@gmail.com>
Patchwork: https://patchwork.freedesktop.org/patch/654889/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
put_unused_fd() doesn't free the installed file, if we've already done
fd_install(). So we need to also free the sync_file.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Patchwork: https://patchwork.freedesktop.org/patch/653583/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
In error paths, we could unref the submit without calling
drm_sched_entity_push_job(), so msm_job_free() will never get
called. Since drm_sched_job_cleanup() will NULL out the
s_fence, we can use that to detect this case.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Patchwork: https://patchwork.freedesktop.org/patch/653584/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
When changing the condition from >= SZ_64K, it was changed to <= SZ_64K.
This disallows migration of 64K, which is the exact minimum allowed.
Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/5057
Fixes: 794f5493f5 ("drm/xe: Strict migration policy for atomic SVM faults")
Cc: stable@vger.kernel.org
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Reviewed-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Signed-off-by: Maarten Lankhorst <dev@lankhorst.se>
Link: https://lore.kernel.org/r/20250521090102.2965100-1-dev@lankhorst.se
(cherry picked from commit 531bef26d1)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Driver unconditionally saves current state on first init in
dsi_pll_10nm_init(), but does not save the VCO rate, only some of the
divider registers. The state is then restored during probe/enable via
msm_dsi_phy_enable() -> msm_dsi_phy_pll_restore_state() ->
dsi_10nm_pll_restore_state().
Restoring calls dsi_pll_10nm_vco_set_rate() with
pll_10nm->vco_current_rate=0, which basically overwrites existing rate of
VCO and messes with clock hierarchy, by setting frequency to 0 to clock
tree. This makes anyway little sense - VCO rate was not saved, so
should not be restored.
If PLL was not configured configure it to minimum rate to avoid glitches
and configuring entire in clock hierarchy to 0 Hz.
Suggested-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://lore.kernel.org/r/sz4kbwy5nwsebgf64ia7uq4ee7wbsa5uy3xmlqwcstsbntzcov@ew3dcyjdzmi2/
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Fixes: a4ccc37693 ("drm/msm/dsi_pll_10nm: restore VCO rate during
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/654796/
Link: https://lore.kernel.org/r/20250520111325.92352-2-krzysztof.kozlowski@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
When widebus was enabled for DisplayPort in commit c7c4122026
("drm/msm/dp: enable widebus on all relevant chipsets") it was clarified
that it is only supported on DPU 5.0.0 onwards which includes SC7180 on
DPU revision 6.2. However, this patch missed that the description
structure for SC7180 is also reused for SDM845 (because of identical
io_start address) which is only DPU 4.0.0, leading to a wrongly enbled
widebus feature and corruption on that platform.
Create a separate msm_dp_desc_sdm845 structure for this SoC compatible,
with the wide_bus_supported flag turned off.
Fixes: c7c4122026 ("drm/msm/dp: enable widebus on all relevant chipsets")
Signed-off-by: James A. MacInnes <james.a.macinnes@gmail.com>
[DB: reworded commit text following Marijn's suggestion]
Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/636944/
Link: https://lore.kernel.org/r/20250212-sdm845_dp-v2-1-4954e51458f4@gmail.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Move this API to the canonical timer_*() namespace.
[ tglx: Redone against pre rc1 ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/all/aB2X0jCKQO56WdMt@gmail.com
Replace comma between expressions with semicolons.
Using a ',' in place of a ';' can have unintended side effects.
Although that is not the case here, it is seems best to use ';'
unless ',' is intended.
Found by inspection.
No functional change intended.
Compile tested only.
Signed-off-by: Chen Ni <nichen@iscas.ac.cn>
Fixes: cd3c62282b ("drm/nouveau/gsp: add usermode class id to gpu hal")
Signed-off-by: Dave Airlie <airlied@redhat.com>
Link: https://lore.kernel.org/r/20250603061027.1310267-1-nichen@iscas.ac.cn
- A couple of vm init fixes (Matt Auld)
- Hwmon fixes (Karthik)
- Drop reduntant conversion to bool (Raag)
- Fix CONFIG_INTEL_VSEC dependency (Arnd)
- Rework eviction rejection of bound external bos (Thomas)
- Stop re-submitting signalled jobs (Matt Auld)
- A couple of pxp fixes (Daniele)
- Add back a fix that got lost in a merge (Matt Auld)
- Create LRC bo without VM (Niranjana)
- Fix for the above fix (Maciej)
-----BEGIN PGP SIGNATURE-----
iHUEABYKAB0WIQRskUM7w1oG5rx2IZO4FpNVCsYGvwUCaEHpLAAKCRC4FpNVCsYG
v94QAQCECvLESR1cSyrP/TdIf67x9l2IpNgBKIjrgmF3RAV5agD/VALTsfkTVu7m
ldfuG7whfePPP9Hkkw+HjsDGJMuC1AQ=
=cOlw
-----END PGP SIGNATURE-----
Merge tag 'drm-xe-next-fixes-2025-06-05' of https://gitlab.freedesktop.org/drm/xe/kernel into drm-next
Driver Changes:
- A couple of vm init fixes (Matt Auld)
- Hwmon fixes (Karthik)
- Drop reduntant conversion to bool (Raag)
- Fix CONFIG_INTEL_VSEC dependency (Arnd)
- Rework eviction rejection of bound external bos (Thomas)
- Stop re-submitting signalled jobs (Matt Auld)
- A couple of pxp fixes (Daniele)
- Add back a fix that got lost in a merge (Matt Auld)
- Create LRC bo without VM (Niranjana)
- Fix for the above fix (Maciej)
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Thomas Hellstrom <thomas.hellstrom@linux.intel.com>
Link: https://lore.kernel.org/r/aEHq44uIAZwfK-mG@fedora
There is unmatched xe_vm_unlock() in the __xe_exec_queue_init().
Leftover from commit fbeaad071a ("drm/xe: Create LRC BO without VM")
Fixes: 2b0a0ce0c2 ("drm/xe: Create LRC BO without VM")
Signed-off-by: Maciej Patelczyk <maciej.patelczyk@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Niranjana Vishwanathapura <niranjana.vishwanathapura@intel.com>
Link: https://lore.kernel.org/r/20250530135627.2821612-1-maciej.patelczyk@intel.com
(cherry picked from commit 28b996ce73)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Specifying VM during lrc->bo creation requires VM's reference
to be held for the lifetime of lrc->bo as it will use VM's dma
reservation object. Using VM's dma reservation object for
lrc->bo doesn't provide any advantage. Hence do not pass VM
while creating lrc->bo.
v2: Use xe_bo_unpin_map_no_vm (Matthew Brost)
Fixes: 264eecdba2 ("drm/xe: Decouple xe_exec_queue and xe_lrc")
Signed-off-by: Niranjana Vishwanathapura <niranjana.vishwanathapura@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250529052031.2429120-2-niranjana.vishwanathapura@intel.com
(cherry picked from commit fbeaad071a)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Daniele noticed that the fix in commit 2d2be279f1 ("drm/xe: fix UAF
around queue destruction") looks to have been unintentionally removed as
part of handling a conflict in some past merge commit. Add it back.
Fixes: ac44ff7cec ("Merge tag 'drm-xe-fixes-2024-10-10' of https://gitlab.freedesktop.org/drm/xe/kernel into drm-fixes")
Reported-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: <stable@vger.kernel.org> # v6.12+
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250603174213.1543579-2-matthew.auld@intel.com
(cherry picked from commit 9d9fca62dc)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
The expected flow of operations when using PXP is to query the PXP
status and wait for it to transition to "ready" before attempting to
create an exec_queue. This flow is followed by the Mesa driver, but
there is no guarantee that an incorrectly coded (or malicious) app
will not attempt to create the queue first without querying the status.
Therefore, we need to clarify what the expected behavior of the queue
creation ioctl is in this scenario.
Currently, the ioctl always fails with an -EBUSY code no matter the
error, but for consistency it is better to distinguish between "failed
to init" (-EIO) and "not ready" (-EBUSY), the same way the query ioctl
does. Note that, while this is a change in the return code of an ioctl,
the behavior of the ioctl in this particular corner case was not clearly
spec'd, so no one should have been relying on it (and we know that Mesa,
which is the only known userspace for this, didn't).
v2: Minor rework of the doc (Rodrigo)
Fixes: 72d479601d ("drm/xe/pxp/uapi: Add userspace and LRC support for PXP-using queues")
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: John Harrison <John.C.Harrison@Intel.com>
Cc: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: John Harrison <John.C.Harrison@Intel.com>
Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250522225401.3953243-7-daniele.ceraolospurio@intel.com
(cherry picked from commit 21784ca960)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
The define of the extension type was accidentally used instead of the
one of the property itself. They're both zero, so no functional issue,
but we should use the correct define for code correctness.
Fixes: 41a97c4a12 ("drm/xe/pxp/uapi: Add API to mark a BO as using PXP")
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://lore.kernel.org/r/20250522225401.3953243-6-daniele.ceraolospurio@intel.com
(cherry picked from commit 1d891ee820)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Customer is reporting a really subtle issue where we get random DMAR
faults, hangs and other nasties for kernel migration jobs when stressing
stuff like s2idle/s3/s4. The explosions seems to happen somewhere
after resuming the system with splats looking something like:
PM: suspend exit
rfkill: input handler disabled
xe 0000:00:02.0: [drm] GT0: Engine reset: engine_class=bcs, logical_mask: 0x2, guc_id=0
xe 0000:00:02.0: [drm] GT0: Timedout job: seqno=24496, lrc_seqno=24496, guc_id=0, flags=0x13 in no process [-1]
xe 0000:00:02.0: [drm] GT0: Kernel-submitted job timed out
The likely cause appears to be a race between suspend cancelling the
worker that processes the free_job()'s, such that we still have pending
jobs to be freed after the cancel. Following from this, on resume the
pending_list will now contain at least one already complete job, but it
looks like we call drm_sched_resubmit_jobs(), which will then call
run_job() on everything still on the pending_list. But if the job was
already complete, then all the resources tied to the job, like the bb
itself, any memory that is being accessed, the iommu mappings etc. might
be long gone since those are usually tied to the fence signalling.
This scenario can be seen in ftrace when running a slightly modified
xe_pm IGT (kernel was only modified to inject artificial latency into
free_job to make the race easier to hit):
xe_sched_job_run: dev=0000:00:02.0, fence=0xffff888276cc8540, seqno=0, lrc_seqno=0, gt=0, guc_id=0, batch_addr=0x000000146910 ...
xe_exec_queue_stop: dev=0000:00:02.0, 3:0x2, gt=0, width=1, guc_id=0, guc_state=0x0, flags=0x13
xe_exec_queue_stop: dev=0000:00:02.0, 3:0x2, gt=0, width=1, guc_id=1, guc_state=0x0, flags=0x4
xe_exec_queue_stop: dev=0000:00:02.0, 4:0x1, gt=1, width=1, guc_id=0, guc_state=0x0, flags=0x3
xe_exec_queue_stop: dev=0000:00:02.0, 1:0x1, gt=1, width=1, guc_id=1, guc_state=0x0, flags=0x3
xe_exec_queue_stop: dev=0000:00:02.0, 4:0x1, gt=1, width=1, guc_id=2, guc_state=0x0, flags=0x3
xe_exec_queue_resubmit: dev=0000:00:02.0, 3:0x2, gt=0, width=1, guc_id=0, guc_state=0x0, flags=0x13
xe_sched_job_run: dev=0000:00:02.0, fence=0xffff888276cc8540, seqno=0, lrc_seqno=0, gt=0, guc_id=0, batch_addr=0x000000146910 ...
.....
xe_exec_queue_memory_cat_error: dev=0000:00:02.0, 3:0x2, gt=0, width=1, guc_id=0, guc_state=0x3, flags=0x13
So the job_run() is clearly triggered twice for the same job, even
though the first must have already signalled to completion during
suspend. We can also see a CAT error after the re-submit.
To prevent this only resubmit jobs on the pending_list that have not yet
signalled.
v2:
- Make sure to re-arm the fence callbacks with sched_start().
v3 (Matt B):
- Stop using drm_sched_resubmit_jobs(), which appears to be deprecated
and just open-code a simple loop such that we skip calling run_job()
on anything already signalled.
Link: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/4856
Fixes: dd08ebf6c3 ("drm/xe: Introduce a new DRM driver for Intel GPUs")
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: William Tseng <william.tseng@intel.com>
Cc: <stable@vger.kernel.org> # v6.8+
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Link: https://lore.kernel.org/r/20250528113328.289392-2-matthew.auld@intel.com
(cherry picked from commit 38fafa9f39)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
For preempt_fence mode VM's we're rejecting eviction of
shared bos during VM_BIND. However, since we do this in the
move() callback, we're getting an eviction failure warning from
TTM. The TTM callback intended for these things is
eviction_valuable().
However, the latter doesn't pass in the struct ttm_operation_ctx
needed to determine whether the caller needs this.
Instead, attach the needed information to the vm under the
vm->resv, until we've been able to update TTM to provide the
needed information. And add sufficient lockdep checks to prevent
misuse and races.
v2:
- Fix a copy-paste error in xe_vm_clear_validating()
v3:
- Fix kerneldoc errors.
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Fixes: 0af944f0e3 ("drm/xe: Reject BO eviction if BO is bound to current VM")
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250528164105.234718-1-thomas.hellstrom@linux.intel.com
(cherry picked from commit 9d5558649f)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
The XE driver can be built with or without VSEC support, but fails to link as
built-in if vsec is in a loadable module:
x86_64-linux-ld: vmlinux.o: in function `xe_vsec_init':
(.text+0x1e83e16): undefined reference to `intel_vsec_register'
The normal fix for this is to add a 'depends on INTEL_VSEC || !INTEL_VSEC',
forcing XE to be a loadable module as well, but that causes a circular
dependency:
symbol DRM_XE depends on INTEL_VSEC
symbol INTEL_VSEC depends on X86_PLATFORM_DEVICES
symbol X86_PLATFORM_DEVICES is selected by DRM_XE
The problem here is selecting a symbol from another subsystem, so change
that as well and rephrase the 'select' into the corresponding dependency.
Since X86_PLATFORM_DEVICES is 'default y', there is no change to
defconfig builds here.
Fixes: 0c45e76fcc ("drm/xe/vsec: Support BMG devices")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250529172355.2395634-2-lucas.demarchi@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
(cherry picked from commit e4931f8be3)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Move power2/curr2_crit to channel 1 i.e power1/curr1_crit as this
represents the entire card critical power/current.
v2: Update the date of curr1_crit also in hwmon documentation.
Signed-off-by: Karthik Poosa <karthik.poosa@intel.com>
Fixes: 345dadc4f6 ("drm/xe/hwmon: Add infra to support card power and energy attributes")
Reviewed-by: Badal Nilawar <badal.nilawar@intel.com>
Link: https://lore.kernel.org/r/20250529163458.2354509-3-karthik.poosa@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
(cherry picked from commit 25e963a09e)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Add support to manage power limits using pcode mailbox commands
for supported platforms.
v2:
- Address review comments. (Badal)
- Use mailbox commands instead of registers to manage power limits
for BMG.
- Clamp the maximum power limit to GPU firmware default value.
v3:
- Clamp power limit in write also for platforms with mailbox support.
v4:
- Remove unnecessary debug prints. (Badal)
v5:
- Update description of variable pl1_on_boot to fix kernel-doc error.
v6:
- Improve commit message, refer to BIOS as GPU firmware.
- Change macro READ_PL_FROM_BIOS to READ_PL_FROM_FW.
- Rectify drm_warn to drm_info.
Signed-off-by: Karthik Poosa <karthik.poosa@intel.com>
Fixes: e90f7a58e6 ("drm/xe/hwmon: Add HWMON support for BMG")
Reviewed-by: Badal Nilawar <badal.nilawar@intel.com>
Link: https://lore.kernel.org/r/20250529163458.2354509-2-karthik.poosa@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
(cherry picked from commit 7596d839f6)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
In xe_vm_close_and_put() we need to be able to call xe_svm_fini(),
however during vm creation we can call this on the error path, before
having actually initialised the svm state, leading to various splats
followed by a fatal NPD.
Fixes: 6fd979c2f3 ("drm/xe: Add SVM init / close / fini to faulting VMs")
Link: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/4967
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250514152424.149591-4-matthew.auld@intel.com
(cherry picked from commit 4f296d77cf)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
In xe_vm_close_and_put() we need to be able to call
flush_work(rebind_work), however during vm creation we can call this on
the error path, before having actually set up the worker, leading to a
splat from flush_work().
It looks like we can simply move the worker init step earlier to fix
this.
Fixes: dd08ebf6c3 ("drm/xe: Introduce a new DRM driver for Intel GPUs")
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: <stable@vger.kernel.org> # v6.8+
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250514152424.149591-3-matthew.auld@intel.com
(cherry picked from commit 96af397aa1)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>