Track the pid per submit, so we can print the name and cmdline of
the task which submitted the batch that caused the gpu to hang.
Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
We need to pull the drm_sched_job_init much earlier, but that's very
minor surgery.
v2: Actually fix up cleanup paths by calling drm_sched_job_init, which
I wanted to to in the previous round (and did, for all other drivers).
Spotted by Lucas.
v3: Rebase over renamed functions to add dependencies.
v4: Rebase over patches from Christian.
v5: More rebasing over work from Christian.
Acked-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Cc: Lucas Stach <l.stach@pengutronix.de>
Cc: Russell King <linux+etnaviv@armlinux.org.uk>
Cc: Christian Gmeiner <christian.gmeiner@gmail.com>
Cc: Sumit Semwal <sumit.semwal@linaro.org>
Cc: "Christian König" <christian.koenig@amd.com>
Cc: etnaviv@lists.freedesktop.org
Cc: linux-media@vger.kernel.org
Cc: linaro-mm-sig@lists.linaro.org
Link: https://patchwork.freedesktop.org/patch/msgid/20220331204651.2699107-2-daniel.vetter@ffwll.ch
We can get the excl fence together with the shared ones as well.
v2: rename the member to fences as well
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Lucas Stach <l.stach@pengutronix.de>
Cc: Lucas Stach <l.stach@pengutronix.de>
Cc: Russell King <linux+etnaviv@armlinux.org.uk>
Cc: Christian Gmeiner <christian.gmeiner@gmail.com>
Cc: etnaviv@lists.freedesktop.org
Link: https://patchwork.freedesktop.org/patch/msgid/20220321135856.1331-5-christian.koenig@amd.com
Add device pointer so scheduler's printing can use
DRM_DEV_ERROR() instead, which makes life easier under multiple GPU
scenario.
v2: amend all calls of drm_sched_init()
v3: fill dev pointer for all drm_sched_init() calls
Signed-off-by: Jiawei Gu <Jiawei.Gu@amd.com>
Reviewed-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220221095705.5290-1-Jiawei.Gu@amd.com
Some GPU heavy test programs manage to trigger the hangcheck quite often.
If there are no other GPU users in the system and the test program
exhibits a very regular structure in the commandstreams that are being
submitted, we can end up with two distinct submits managing to trigger
the hangcheck with the FE in a very similar address range. This leads
the hangcheck to believe that the GPU is stuck, while in reality the GPU
is already busy working on a different job. To avoid those spurious
GPU resets, also remember and consider the last completed fence seqno
in the hang check.
Reported-by: Joerg Albert <joerg.albert@iav.de>
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Originally a job was only bound to the queue when we pushed this, but
now that's done in drm_sched_job_init, making that parameter entirely
redundant.
Remove it.
The same applies to the context parameter in
lima_sched_context_queue_task, simplify that too.
v2:
Rebase on top of msm adopting drm/sched
Reviewed-by: Christian König <christian.koenig@amd.com>
Acked-by: Emma Anholt <emma@anholt.net>
Acked-by: Melissa Wen <mwen@igalia.com>
Reviewed-by: Steven Price <steven.price@arm.com> (v1)
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com> (v1)
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Cc: Lucas Stach <l.stach@pengutronix.de>
Cc: Russell King <linux+etnaviv@armlinux.org.uk>
Cc: Christian Gmeiner <christian.gmeiner@gmail.com>
Cc: Qiang Yu <yuq825@gmail.com>
Cc: Rob Herring <robh@kernel.org>
Cc: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Cc: Steven Price <steven.price@arm.com>
Cc: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Cc: Emma Anholt <emma@anholt.net>
Cc: David Airlie <airlied@linux.ie>
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Sumit Semwal <sumit.semwal@linaro.org>
Cc: "Christian König" <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Nirmoy Das <nirmoy.das@amd.com>
Cc: Dave Airlie <airlied@redhat.com>
Cc: Chen Li <chenli@uniontech.com>
Cc: Lee Jones <lee.jones@linaro.org>
Cc: Deepak R Varma <mh12gx2825@gmail.com>
Cc: Kevin Wang <kevin1.wang@amd.com>
Cc: Luben Tuikov <luben.tuikov@amd.com>
Cc: "Marek Olšák" <marek.olsak@amd.com>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Cc: Dennis Li <Dennis.Li@amd.com>
Cc: Boris Brezillon <boris.brezillon@collabora.com>
Cc: etnaviv@lists.freedesktop.org
Cc: lima@lists.freedesktop.org
Cc: linux-media@vger.kernel.org
Cc: linaro-mm-sig@lists.linaro.org
Cc: Rob Clark <robdclark@gmail.com>
Cc: Sean Paul <sean@poorly.run>
Cc: Melissa Wen <mwen@igalia.com>
Cc: linux-arm-msm@vger.kernel.org
Cc: freedreno@lists.freedesktop.org
Link: https://patchwork.freedesktop.org/patch/msgid/20210805104705.862416-6-daniel.vetter@ffwll.ch
This is a very confusingly named function, because not just does it
init an object, it arms it and provides a point of no return for
pushing a job into the scheduler. It would be nice if that's a bit
clearer in the interface.
But the real reason is that I want to push the dependency tracking
helpers into the scheduler code, and that means drm_sched_job_init
must be called a lot earlier, without arming the job.
v2:
- don't change .gitignore (Steven)
- don't forget v3d (Emma)
v3: Emma noticed that I leak the memory allocated in
drm_sched_job_init if we bail out before the point of no return in
subsequent driver patches. To be able to fix this change
drm_sched_job_cleanup() so it can handle being called both before and
after drm_sched_job_arm().
Also improve the kerneldoc for this.
v4:
- Fix the drm_sched_job_cleanup logic, I inverted the booleans, as
usual (Melissa)
- Christian pointed out that drm_sched_entity_select_rq() also needs
to be moved into drm_sched_job_arm, which made me realize that the
job->id definitely needs to be moved too.
Shuffle things to fit between job_init and job_arm.
v5:
Reshuffle the split between init/arm once more, amdgpu abuses
drm_sched.ready to signal gpu reset failures. Also document this
somewhat. (Christian)
v6:
Rebase on top of the msm drm/sched support. Note that the
drm_sched_job_init() call is completely misplaced, and hence also the
split-out drm_sched_entity_push_job(). I've put in a FIXME which the next
patch will address.
v7: Drop the FIXME in msm, after discussions with Rob I agree it shouldn't
be a problem where it is now.
Acked-by: Christian König <christian.koenig@amd.com>
Acked-by: Melissa Wen <mwen@igalia.com>
Cc: Melissa Wen <melissa.srw@gmail.com>
Acked-by: Emma Anholt <emma@anholt.net>
Acked-by: Steven Price <steven.price@arm.com> (v2)
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com> (v5)
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Cc: Lucas Stach <l.stach@pengutronix.de>
Cc: Russell King <linux+etnaviv@armlinux.org.uk>
Cc: Christian Gmeiner <christian.gmeiner@gmail.com>
Cc: Qiang Yu <yuq825@gmail.com>
Cc: Rob Herring <robh@kernel.org>
Cc: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Cc: Steven Price <steven.price@arm.com>
Cc: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Cc: David Airlie <airlied@linux.ie>
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Sumit Semwal <sumit.semwal@linaro.org>
Cc: "Christian König" <christian.koenig@amd.com>
Cc: Masahiro Yamada <masahiroy@kernel.org>
Cc: Kees Cook <keescook@chromium.org>
Cc: Adam Borowski <kilobyte@angband.pl>
Cc: Nick Terrell <terrelln@fb.com>
Cc: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Cc: Paul Menzel <pmenzel@molgen.mpg.de>
Cc: Sami Tolvanen <samitolvanen@google.com>
Cc: Viresh Kumar <viresh.kumar@linaro.org>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Dave Airlie <airlied@redhat.com>
Cc: Nirmoy Das <nirmoy.das@amd.com>
Cc: Deepak R Varma <mh12gx2825@gmail.com>
Cc: Lee Jones <lee.jones@linaro.org>
Cc: Kevin Wang <kevin1.wang@amd.com>
Cc: Chen Li <chenli@uniontech.com>
Cc: Luben Tuikov <luben.tuikov@amd.com>
Cc: "Marek Olšák" <marek.olsak@amd.com>
Cc: Dennis Li <Dennis.Li@amd.com>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Cc: Sonny Jiang <sonny.jiang@amd.com>
Cc: Boris Brezillon <boris.brezillon@collabora.com>
Cc: Tian Tao <tiantao6@hisilicon.com>
Cc: etnaviv@lists.freedesktop.org
Cc: lima@lists.freedesktop.org
Cc: linux-media@vger.kernel.org
Cc: linaro-mm-sig@lists.linaro.org
Cc: Emma Anholt <emma@anholt.net>
Cc: Rob Clark <robdclark@gmail.com>
Cc: Sean Paul <sean@poorly.run>
Cc: linux-arm-msm@vger.kernel.org
Cc: freedreno@lists.freedesktop.org
Link: https://patchwork.freedesktop.org/patch/msgid/20210817084917.3555822-1-daniel.vetter@ffwll.ch
Mali Midgard/Bifrost GPUs have 3 hardware queues but only a global GPU
reset. This leads to extra complexity when we need to synchronize timeout
works with the reset work. One solution to address that is to have an
ordered workqueue at the driver level that will be used by the different
schedulers to queue their timeout work. Thanks to the serialization
provided by the ordered workqueue we are guaranteed that timeout
handlers are executed sequentially, and can thus easily reset the GPU
from the timeout handler without extra synchronization.
v5:
* Add a new paragraph to the timedout_job() method
v3:
* New patch
v4:
* Actually use the timeout_wq to queue the timeout work
Suggested-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Steven Price <steven.price@arm.com>
Reviewed-by: Lucas Stach <l.stach@pengutronix.de>
Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Acked-by: Christian König <christian.koenig@amd.com>
Cc: Qiang Yu <yuq825@gmail.com>
Cc: Emma Anholt <emma@anholt.net>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: "Christian König" <christian.koenig@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210630062751.2832545-3-boris.brezillon@collabora.com
This patch does not change current behaviour.
The driver's job timeout handler now returns
status indicating back to the DRM layer whether
the device (GPU) is no longer available, such as
after it's been unplugged, or whether all is
normal, i.e. current behaviour.
All drivers which make use of the
drm_sched_backend_ops' .timedout_job() callback
have been accordingly renamed and return the
would've-been default value of
DRM_GPU_SCHED_STAT_NOMINAL to restart the task's
timeout timer--this is the old behaviour, and is
preserved by this patch.
v2: Use enum as the status of a driver's job
timeout callback method.
v3: Return scheduler/device information, rather
than task information.
Cc: Alexander Deucher <Alexander.Deucher@amd.com>
Cc: Andrey Grodzovsky <Andrey.Grodzovsky@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Lucas Stach <l.stach@pengutronix.de>
Cc: Russell King <linux+etnaviv@armlinux.org.uk>
Cc: Christian Gmeiner <christian.gmeiner@gmail.com>
Cc: Qiang Yu <yuq825@gmail.com>
Cc: Rob Herring <robh@kernel.org>
Cc: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Cc: Steven Price <steven.price@arm.com>
Cc: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Cc: Eric Anholt <eric@anholt.net>
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Luben Tuikov <luben.tuikov@amd.com>
Acked-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Acked-by: Christian König <christian.koenig@amd.com>
Acked-by: Steven Price <steven.price@arm.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Link: https://patchwork.freedesktop.org/patch/415095/
The drm scheduler currently expects that the stop/start sequence is always
executed in the timeout handling, as the job at the head of the hardware
execution list is always removed from the ring mirror before the driver
function is called and only inserted back into the list when starting the
scheduler.
This adds some unnecessary overhead if the timeout handler determines
that the GPU is still executing jobs normally and just wished to extend
the timeout, but a better solution requires a major rearchitecture of the
scheduler, which is not applicable as a fix.
Fixes: 135517d356 ("drm/scheduler: Avoid accessing freed bad job.")
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Tested-by: Russell King <rmk+kernel@armlinux.org.uk>
Due to the tracking provided by the scheduler we know exactly which
submit is failing. Only dump this single submit and the required
auxiliary information. This cuts down the size of the devcoredumps
by only including relevant information.
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Philipp Zabel <p.zabel@pengutronix.de>
Drop unused includes, move more includes from the generic etnaviv_drv.h to
the units where they are actually used, sort includes.
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
We now destroy finished jobs from the worker thread to make sure that
we never destroy a job currently in timeout processing.
By this we avoid holding lock around ring mirror list in drm_sched_stop
which should solve a deadlock reported by a user.
v2: Remove unused variable.
v4: Move guilty job free into sched code.
v5:
Move sched->hw_rq_count to drm_sched_start to account for counter
decrement in drm_sched_stop even when we don't call resubmit jobs
if guily job did signal.
v6: remove unused variable
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=109692
Acked-by: Chunming Zhou <david1.zhou@amd.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/1555599624-12285-3-git-send-email-andrey.grodzovsky@amd.com
Decauple sched threads stop and start and ring mirror
list handling from the policy of what to do about the
guilty jobs.
When stoppping the sched thread and detaching sched fences
from non signaled HW fenes wait for all signaled HW fences
to complete before rerunning the jobs.
v2: Fix resubmission of guilty job into HW after refactoring.
v4:
Full restart for all the jobs, not only from guilty ring.
Extract karma increase into standalone function.
v5:
Rework waiting for signaled jobs without relying on the job
struct itself as those might already be freed for non 'guilty'
job's schedulers.
Expose karma increase to drivers.
v6:
Use list_for_each_entry_safe_continue and drm_sched_process_job
in case fence already signaled.
Call drm_sched_increase_karma only once for amdgpu and add documentation.
v7:
Wait only for the latest job's fence.
Suggested-by: Christian Koenig <Christian.Koenig@amd.com>
Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The context isn't really related to the cmdbuf, but is a property of
the job. This has been missed when moving to a properly refcounted
etnaviv_gem_submit.
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
New features for 4.21:
amdgpu:
- Support for SDMA paging queue on vega
- Put compute EOP buffers into vram for better performance
- Share more code with amdkfd
- Support for scanout with DCC on gfx9
- Initial kerneldoc for DC
- Updated SMU firmware support for gfx8 chips
- Rework CSA handling for eventual support for preemption
- XGMI PSP support
- Clean up RLC handling
- Enable GPU reset by default on VI, SOC15 dGPUs
- Ring and IB test cleanups
amdkfd:
- Share more code with amdgpu
ttm:
- Move global init out of the drivers
scheduler:
- Track if schedulers are ready for work
- Timeout/fault handling changes to facilitate GPU recovery
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Alex Deucher <alexdeucher@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20181114165113.3751-1-alexander.deucher@amd.com
This patch adds a new API to clean up the scheduler job resources. This
is primarliy needed in cases the job was created but was not queued to
the scheduler queue. Additionally with this change, the layer which
creates the scheduler job also gets to free up the job's resources and
this entails moving the dma_fence_put(finished_fence) to the drivers
ops free handler routines.
Signed-off-by: Sharat Masetty <smasetty@codeaurora.org>
Reviewed-by: Christian König <christian.koenig@amd.com>
Acked-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Make sure we always restart the timer after a timeout and remove the
device specific workarounds.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The GPU hardware fences and the job out-fences are on different timelines
so it's wrong to compare them. Fix this by only looking at the out-fence.
Cc: <stable@vger.kernel.org>
Fixes: 2c83a726d6 (drm/etnaviv: bring back progress check in job
timeout handler)
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
having a delayed work item per job is redundant as we only need one
per scheduler to track the time out the currently executing job.
v2: the first element of the ring mirror list is the currently
executing job so we don't need a additional variable for it
v3: squash in fixes for v3d and etnaviv
Signed-off-by: Nayan Deshmukh <nayan26deshmukh@gmail.com>
Suggested-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
From: Lucas Stach <l.stach@pengutronix.de>
"not much to de-stage this time. Changes from Philipp and Souptick to
use memset32 more and switch the fault handler to the new vm_fault_t
and two small fixes for issues that can be hit in rare corner cases
from me."
Signed-off-by: Dave Airlie <airlied@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/1533563808.2809.7.camel@pengutronix.de
The documentation of drm_sched_job_init and drm_sched_entity_push_job has
been clarified. Both functions should be called under a shared lock, to
avoid jobs getting pushed into the scheduler queue in a different order
than their sched_fence seqnos, which will confuse checks that are looking
at the seqnos to infer information about completion order.
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
entity has a scheduler field and we don't need the sched argument
in any of the functions where entity is provided.
Signed-off-by: Nayan Deshmukh <nayan26deshmukh@gmail.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
When the hangcheck handler was replaced by the DRM scheduler timeout
handling we dropped the forward progress check, as this might allow
clients to hog the GPU for a long time with a big job.
It turns out that even reasonably well behaved clients like the
Armada Xorg driver occasionally trip over the 500ms timeout. Bring
back the forward progress check to get rid of the userspace regression.
We would still like to fix userspace to submit smaller batches
if possible, but that is for another day.
Cc: <stable@vger.kernel.org>
Fixes: 6d7a20c077 (drm/etnaviv: replace hangcheck with scheduler timeout)
Reported-by: Russell King <linux@armlinux.org.uk>
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Eric Anholt <eric@anholt.net>
This replaces the repetitive GPL-2.0 license text in code and header files
with the SPDX tags. Generated hardware headers aren't changed, as any changes
there need to be done in the upstream rnndb repository.
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
The current limit of 2 leads to some GPU idle times, as the usual
IRQ latency leads to up to 3 jobs getting signaled at once with some
standard workloads.
A larger HW job limit might lead to slightly worse QoS, but we accept
that to not sacrifice GPU throughput in the common case.
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
etnaviv_sched_dependency() and etnaviv_sched_run_job() are only
used in this file, so make them static.
This fixes the following sparse warnings:
drivers/gpu/drm/etnaviv/etnaviv_sched.c:30:18: warning: symbol 'etnaviv_sched_dependency' was not declared. Should it be static?
drivers/gpu/drm/etnaviv/etnaviv_sched.c:81:18: warning: symbol 'etnaviv_sched_run_job' was not declared. Should it be static?
Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
This replaces the etnaviv internal hangcheck logic with the job timeout
handling provided by the DRM scheduler. This simplifies the driver further
and allows to replay jobs after a GPU reset, so only minimal state is lost.
This introduces a user-visible change in that we don't allow jobs to run
indefinitely as long as they make progress anymore, as this introduces
quality of service issues when multiple processes are using the GPU.
Userspace is now responsible to flush jobs in a way that the finish in a
reasonable time, where reasonable is currently defined as less than 500ms.
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Move the fence dependency handling to the scheduler where it belongs.
Jobs with unsignaled dependencies just get to sit in the scheduler queue
without holding any locks.
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
This hooks in the DRM GPU scheduler. No improvement yet, as all the
dependency handling is still done in etnaviv_gem_submit. This just
replaces the actual GPU submit by passing through the scheduler.
Allows to get rid of the retire worker, as this is now driven by the
scheduler.
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Christian Gmeiner