Pull the individual strands of creating a custom heartbeat requests into
a pair of common functions. This will reduce the number of changes we
will need to make in future.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201224160213.29521-1-chris@chris-wilson.co.uk
Now that the tasklet completely controls scheduling of the requests, and
we postpone scheduling out the old requests, we can keep a hanging
virtual request bound to the engine on which it hung, and remove it from
te queue. On release, it will be returned to the same engine and remain
in its queue until it is scheduled; after which point it will become
eligible for transfer to a sibling. Instead, we could opt to resubmit the
request along the virtual engine on unhold, making it eligible for load
balancing immediately -- but that seems like a pointless optimisation
for a hanging context.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201224135544.1713-8-chris@chris-wilson.co.uk
Having recognised that we do not change the sibling until we schedule
out, we can then defer the decision to resubmit the virtual engine from
the unwind of the active queue to scheduling out of the virtual context.
This improves our resilence in virtual engine scheduling, and should
eliminate the rare cases of gem_exec_balance failing.
By keeping the unwind order intact on the local engine, we can preserve
data dependency ordering while doing a preempt-to-busy pass until we
have determined the new ELSP. This means that if we try to timeslice
between a virtual engine and a data-dependent ordinary request, the pair
will maintain their relative ordering and we will avoid the
resubmission, cancelling the timeslicing until further change.
The dilemma though is that we then may end up in a situation where the
'demotion' of the virtual request to an ordinary request in the engine
queue results in filling the ELSP[] with virtual requests instead of
spreading the load across the engines. To compensate for this, we mark
each virtual request and refuse to resubmit a virtual request in the
secondary ELSP slots, thus forcing subsequent virtual requests to be
scheduled out after timeslicing. By delaying the decision until we
schedule out, we will avoid unnecessary resubmission.
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/2079
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/2098
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201224135544.1713-7-chris@chris-wilson.co.uk
Let's only wait for the list iterator when decoupling the virtual
breadcrumb, as the signaling of all the requests may take a long time,
during which we do not want to keep the tasklet spinning.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201224135544.1713-6-chris@chris-wilson.co.uk
The issue with stale virtual breadcrumbs remain. Now we have the problem
that if the irq-signaler is still referencing the stale breadcrumb as we
transfer it to a new sibling, the list becomes spaghetti. This is a very
small window, but that doesn't stop it being hit infrequently. To
prevent the lists being tangled (the iterator starting on one engine's
b->signalers but walking onto another list), always decouple the virtual
breadcrumb on schedule-out and make sure that the walker has stepped out
of the lists.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201224135544.1713-5-chris@chris-wilson.co.uk
Inside schedule_out, we do extra work upon idling the context, such as
updating the runtime, kicking off retires, kicking virtual engines.
However, if we are in a series of processing single requests per
contexts, we may find ourselves scheduling out the context, only to
immediately schedule it back in during dequeue. This is just extra work
that we can avoid if we keep the context marked as inflight across the
dequeue. This becomes more significant later on for minimising virtual
engine misses.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201224135544.1713-4-chris@chris-wilson.co.uk
Once a virtual engine has been bound to a sibling, it will remain bound
until we finally schedule out the last active request. We can not rebind
the context to a new sibling while it is inflight as the context save
will conflict, hence we wait. As we cannot then use any other sibliing
while the context is inflight, only kick the bound sibling while it
inflight and upon scheduling out the kick the rest (so that we can swap
engines on timeslicing if the previously bound engine becomes
oversubscribed).
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201224135544.1713-3-chris@chris-wilson.co.uk
Rather than going back and forth between the rb_node entry and the
virtual_engine type, store the ve local and reuse it. As the
container_of conversion from rb_node to virtual_engine requires a
variable offset, performing that conversion just once shaves off a bit
of code.
v2: Keep a single virtual engine lookup, for typical use.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201224135544.1713-2-chris@chris-wilson.co.uk
Rather than having special case code for opportunistically calling
process_csb() and performing a direct submit while holding the engine
spinlock for submitting the request, simply call the tasklet directly.
This allows us to retain the direct submission path, including the CS
draining to allow fast/immediate submissions, without requiring any
duplicated code paths, and most importantly greatly simplifying the
control flow by removing reentrancy. This will enable us to close a few
races in the virtual engines in the next few patches.
The trickiest part here is to ensure that paired operations (such as
schedule_in/schedule_out) remain under consistent locking domains,
e.g. when pulled outside of the engine->active.lock
v2: Use bh kicking, see commit 3c53776e29 ("Mark HI and TASKLET
softirq synchronous").
v3: Update engine-reset to be tasklet aware
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201224135544.1713-1-chris@chris-wilson.co.uk
If we want to reuse a fence that is in active use by the GPU, we have to
wait an uncertain amount of time, but if we reuse an inactive fence, we
can change it right away. Loop through the list of available fences
twice, ignoring any active fences on the first pass.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201223122051.4624-1-chris@chris-wilson.co.uk
Pull the GT clock information [used to derive CS timestamps and PM
interval] under the GT so that is it local to the users. In doing so, we
consolidate the two references for the same information, of which the
runtime-info took note of a potential clock source override and scaling
factors.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201223122359.22562-2-chris@chris-wilson.co.uk
We assume that both timestamps are driven off the same clock [reported
to userspace as I915_PARAM_CS_TIMESTAMP_FREQUENCY]. Verify that this is
so by reading the timestamp registers around a busywait (on an otherwise
idle engine so there should be no preemptions).
v2: Icelake (not ehl, nor tgl) seems to be using a fixed 80ns interval
for, and only for, CTX_TIMESTAMP -- or it may be GPU frequency and the
test is always running at maximum frequency?. As far as I can tell, this
isolated change in behaviour is undocumented.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201223122359.22562-1-chris@chris-wilson.co.uk
The caller determines if the failure is an error or not, so avoid
warning when we will try again and succeed. For example,
<7> [111.319321] [drm:intel_guc_fw_upload [i915]] GuC status 0x20
<3> [111.319340] i915 0000:00:02.0: [drm] *ERROR* GuC load failed: status = 0x00000020
<3> [111.319606] i915 0000:00:02.0: [drm] *ERROR* GuC load failed: status: Reset = 0, BootROM = 0x10, UKernel = 0x00, MIA = 0x00, Auth = 0x00
<7> [111.320045] [drm:__uc_init_hw [i915]] GuC fw load failed: -110; will reset and retry 2 more time(s)
<7> [111.322978] [drm:intel_guc_fw_upload [i915]] GuC status 0x8002f0ec
should not have been reported as a _test_ failure, as the GuC was
successfully loaded on the second attempt and the system remained
operational.
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/2797
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201214100949.11387-2-chris@chris-wilson.co.uk
When waiting for the submit, before checking the status of the request,
kick the tasklet to make sure we are processing the submission. This
speeds up submission if we are using any tasklet suppression for
secondary requests.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201222113536.3775-3-chris@chris-wilson.co.uk
Make sure that the request has been submitted to HW before we begin our
wait. This reduces our reliance on the semaphore yield interrupt driving
the preemption request.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201222113536.3775-2-chris@chris-wilson.co.uk
We assume that the contents of the HWSP are lost across suspend, and so
upon resume we must restore critical values such as the timeline seqno.
Keep track of every timeline allocated that uses the HWSP as its storage
and so we can then reset all seqno values by walking that list.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201222104242.10993-1-chris@chris-wilson.co.uk
Primarily used by selftests, but also by runtime debugging of engine
w/a, is a routine to create a temporarily bound buffer for readback.
Almagamate the duplicated routines into one.
Suggested-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201219020343.22681-2-chris@chris-wilson.co.uk
Split the definition, construction and updating of the Logical Ring
Context from the execlist submission interface. The LRC is used by the
HW, irrespective of our different submission backends.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201219020343.22681-1-chris@chris-wilson.co.uk
tasklet_kill() ensures that we _yield_ the processor until a remote
tasklet is completed. However, this leads to a starvation condition as
being at the bottom of the scheduler's runqueue means that anything else
is able to run, including all hogs keeping the tasklet occupied.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201220134858.10510-1-chris@chris-wilson.co.uk
Since we allow removing the timeline map at runtime, there is a risk
that rq->hwsp points into a stale page. To control that risk, we hold
the RCU read lock while reading *rq->hwsp, but we missed a couple of
important barriers. First, the unpinning / removal of the timeline map
must be after all RCU readers into that map are complete, i.e. after an
rcu barrier (in this case courtesy of call_rcu()). Secondly, we must
make sure that the rq->hwsp we are about to dereference under the RCU
lock is valid. In this case, we make the rq->hwsp pointer safe during
i915_request_retire() and so we know that rq->hwsp may become invalid
only after the request has been signaled. Therefore is the request is
not yet signaled when we acquire rq->hwsp under the RCU, we know that
rq->hwsp will remain valid for the duration of the RCU read lock.
This is a very small window that may lead to either considering the
request not completed (causing a delay until the request is checked
again, any wait for the request is not affected) or dereferencing an
invalid pointer.
Fixes: 3adac4689f ("drm/i915: Introduce concept of per-timeline (context) HWSP")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: <stable@vger.kernel.org> # v5.1+
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201218122421.18344-1-chris@chris-wilson.co.uk
Since we wake the GT up before executing a request, and go to sleep as
soon as it is retired, the GT wake time not only represents how long the
device is powered up, but also provides a summary, albeit an overestimate,
of the device runtime (i.e. the rc0 time to compare against rc6 time).
v2: s/busy/awake/
v3: software-gt-awake-time and I915_PMU_SOFTWARE_GT_AWAKE_TIME
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reported-by: kernel test robot <oliver.sang@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201215154456.13954-1-chris@chris-wilson.co.uk
Matthew Brost pointed out that the while-loop on a shared breadcrumb was
inherently fraught with danger as it competed with the other users of
the breadcrumbs. However, in order to completely drain the re-arming irq
worker, the while-loop is a necessity, despite my optimism that we could
force cancellation with a couple of irq_work invocations.
Given that we can't merely drop the while-loop, use an activity counter on
the breadcrumbs to detect when we are parking the breadcrumbs for the
last time.
Based on a patch by Matthew Brost.
Reported-by: Matthew Brost <matthew.brost@intel.com>
Suggested-by: Matthew Brost <matthew.brost@intel.com>
Fixes: 9d5612ca16 ("drm/i915/gt: Defer enabling the breadcrumb interrupt to after submission")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201217091524.10258-1-chris@chris-wilson.co.uk
Now that relay_open() accepts const callbacks, make relay callbacks const.
Link: https://lkml.kernel.org/r/534d089f413db98aa0b94773fa49d5275d0d3c25.1606153547.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Kalle Valo <kvalo@codeaurora.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
core:
- documentation updates
- deprecate DRM_FORMAT_MOD_NONE
- atomic crtc enable/disable rework
- GEM convert drivers to gem object functions
- remove SCATTER_LIST_MAX_SEGMENT
sched:
- avoid infinite waits
ttm:
- remove AGP support
- don't modify caching for swapout
- ttm pinning rework
- major TTM reworks
- new backend allocator
- multihop support
vram-helper:
- top down BO placement fix
- TTM changes
- GEM object support
displayport:
- DP 2.0 DPCD prep work
- DP MST extended DPCD caps
fbdev:
- mark as orphaned
amdgpu:
- Initial Vangogh support
- Green Sardine support
- Dimgrey Cavefish support
- SG display support for renoir
- SMU7 improvements
- gfx9+ modiifier support
- CI BACO fixes
radeon:
- expose voltage via hwmon on SUMO
amdkfd:
- fix unique id handling
i915:
- more DG1 enablement
- bigjoiner support
- integer scaling filter support
- async flip support
- ICL+ DSI command mode
- Improve display shutdown
- Display refactoring
- eLLC machine fbdev loading fix
- dma scatterlist fixes
- TGL hang fixes
- eLLC display buffer caching on SKL+
- MOCS PTE seeting for gen9+
msm:
- Shutdown hook
- GPU cooling device support
- DSI 7nm and 10nm phy/pll updates
- sm8150/sm2850 DPU support
- GEM locking re-work
- LLCC system cache support
aspeed:
- sysfs output config support
ast:
- LUT fix
- new display mode
gma500:
- remove 2d framebuffer accel
panfrost:
- move gpu reset to a worker
exynos:
- new HDMI mode support
mediatek:
- MT8167 support
- yaml bindings
- MIPI DSI phy code moved
etnaviv:
- new perf counter
- more lockdep annotation
hibmc:
- i2c DDC support
ingenic:
- pixel clock reset fix
- reserved memory support
- allow both DMA channels at once
- different pixel format support
- 30/24/8-bit palette modes
tilcdc:
- don't keep vblank irq enabled
vc4:
- new maintainer added
- DSI registration fix
virtio:
- blob resource support
- host visible and cross-device support
- uuid api support
-----BEGIN PGP SIGNATURE-----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=c9rl
-----END PGP SIGNATURE-----
Merge tag 'drm-next-2020-12-11' of git://anongit.freedesktop.org/drm/drm
Pull drm updates from Dave Airlie:
"Not a huge amount of big things here, AMD has support for a few new HW
variants (vangogh, green sardine, dimgrey cavefish), Intel has some
more DG1 enablement. We have a few big reworks of the TTM layers and
interfaces, GEM and atomic internal API reworks cross tree. fbdev is
marked orphaned in here as well to reflect the current reality.
core:
- documentation updates
- deprecate DRM_FORMAT_MOD_NONE
- atomic crtc enable/disable rework
- GEM convert drivers to gem object functions
- remove SCATTER_LIST_MAX_SEGMENT
sched:
- avoid infinite waits
ttm:
- remove AGP support
- don't modify caching for swapout
- ttm pinning rework
- major TTM reworks
- new backend allocator
- multihop support
vram-helper:
- top down BO placement fix
- TTM changes
- GEM object support
displayport:
- DP 2.0 DPCD prep work
- DP MST extended DPCD caps
fbdev:
- mark as orphaned
amdgpu:
- Initial Vangogh support
- Green Sardine support
- Dimgrey Cavefish support
- SG display support for renoir
- SMU7 improvements
- gfx9+ modiifier support
- CI BACO fixes
radeon:
- expose voltage via hwmon on SUMO
amdkfd:
- fix unique id handling
i915:
- more DG1 enablement
- bigjoiner support
- integer scaling filter support
- async flip support
- ICL+ DSI command mode
- Improve display shutdown
- Display refactoring
- eLLC machine fbdev loading fix
- dma scatterlist fixes
- TGL hang fixes
- eLLC display buffer caching on SKL+
- MOCS PTE seeting for gen9+
msm:
- Shutdown hook
- GPU cooling device support
- DSI 7nm and 10nm phy/pll updates
- sm8150/sm2850 DPU support
- GEM locking re-work
- LLCC system cache support
aspeed:
- sysfs output config support
ast:
- LUT fix
- new display mode
gma500:
- remove 2d framebuffer accel
panfrost:
- move gpu reset to a worker
exynos:
- new HDMI mode support
mediatek:
- MT8167 support
- yaml bindings
- MIPI DSI phy code moved
etnaviv:
- new perf counter
- more lockdep annotation
hibmc:
- i2c DDC support
ingenic:
- pixel clock reset fix
- reserved memory support
- allow both DMA channels at once
- different pixel format support
- 30/24/8-bit palette modes
tilcdc:
- don't keep vblank irq enabled
vc4:
- new maintainer added
- DSI registration fix
virtio:
- blob resource support
- host visible and cross-device support
- uuid api support"
* tag 'drm-next-2020-12-11' of git://anongit.freedesktop.org/drm/drm: (1754 commits)
drm/amdgpu: Initialise drm_gem_object_funcs for imported BOs
drm/amdgpu: fix size calculation with stolen vga memory
drm/amdgpu: remove amdgpu_ttm_late_init and amdgpu_bo_late_init
drm/amdgpu: free the pre-OS console framebuffer after the first modeset
drm/amdgpu: enable runtime pm using BACO on CI dGPUs
drm/amdgpu/cik: enable BACO reset on Bonaire
drm/amd/pm: update smu10.h WORKLOAD_PPLIB setting for raven
drm/amd/pm: remove one unsupported smu function for vangogh
drm/amd/display: setup system context for APUs
drm/amd/display: add S/G support for Vangogh
drm/amdkfd: Fix leak in dmabuf import
drm/amdgpu: use AMDGPU_NUM_VMID when possible
drm/amdgpu: fix sdma instance fw version and feature version init
drm/amd/pm: update driver if version for dimgrey_cavefish
drm/amd/display: 3.2.115
drm/amd/display: [FW Promotion] Release 0.0.45
drm/amd/display: Revert DCN2.1 dram_clock_change_latency update
drm/amd/display: Enable gpu_vm_support for dcn3.01
drm/amd/display: Fixed the audio noise during mode switching with HDCP mode on
drm/amd/display: Add wm table for Renoir
...
When we reset the legacy ring context, due to potential corruption over
suspend/resume, remove the valid bit so that we avoid loading garbage.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201210080240.24529-1-chris@chris-wilson.co.uk
The above workaround was added as an engine workaround not a GT
workaround. Moved it to the correct location.
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20201210170615.3107266-1-lucas.demarchi@intel.com
These functions are independent from the backend used and can therefore
be split out of the exelists submission file, so they can be re-used by
the upcoming GuC submission backend.
Based on a patch by Chris Wilson.
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Reviewed-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201209233618.4287-3-chris@chris-wilson.co.uk
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
We want to separate the utility functions for controlling the logical
ring context from the execlists submission mechanism (which is an
overgrown scheduler).
This is similar to Daniele's work to split up the files, but being
selfish I wanted to base it after my own changes to intel_lrc.c petered
out.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201209233618.4287-2-chris@chris-wilson.co.uk
Cleanup intel_lrc.h by moving some of the residual common register
definitions into intel_lrc_reg.h, prior to rebranding and splitting off
the submission backends.
v2: keep the SCHEDULE enum in the old file, since it is specific to the
gvt usage of the execlists submission backend (John)
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> #v2
Cc: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201209233618.4287-1-chris@chris-wilson.co.uk
Tigerlake is plagued by spontaneous DMAR faults [reason 7, next page
table ptr is invalid] which lead to GPU hangs. These faults occur when
an iommu map is immediately reused. Adding further clflushes and
barriers around either the GTT PTE or iommu PTE updates do not prevent
the faults. So far the only effect has been from inducing a delay
between reuse of the iommu on the GPU, and applying the delay at the
iommu map allows for the smallest stable delay.
Note that such a delay is hideous and clearly does not fix the root cause,
and so should only be a bandaid until a complete solution is found. The
delay was determined by running igt/gem_exec_fence/parallel in a loop for
a few hours (unpatched MTBF is about 10s).
We have also seen such DMAR fault [reason 7] errors on other platforms,
notably gen9-gen11, but so far it has only been trivially and
consistently reproduced on Tigerlake.
v2: Leave a tell-tale to know when we apply the vt'd quirk, and as a
reminder to remove it again. Hopefully.
Testcase: igt/gem_exec_fence/parallel
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Acked-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201209164008.5487-2-chris@chris-wilson.co.uk
Document what a masked register is according to bspec so we avoid
developers using the wrong functions to implement WAs.
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20201209045246.2905675-3-lucas.demarchi@intel.com
The use of "masked" in this function is due to its history. Once upon a
time it received a mask and a value as parameter. Since
commit eeec73f8a4 ("drm/i915/gt: Skip rmw for masked registers")
that is not true anymore and now there is a clear and a set parameter.
Depending on the case, that can still be thought as a mask and value,
but there are some subtle differences: what we clear doesn't need to be
the same bits we are setting, particularly when we are using masked
registers.
The fact that we also have "masked registers", i.e. registers whose mask
is stored in the upper 16 bits of the register, makes it even more
confusing, because "masked" in wa_write_masked_or() has little to do
with masked registers, but rather refers to the old mask parameter the
function received (that can also, but not exclusively, be used to write
to masked register).
Avoid the ambiguity and misnomer by renaming it to something else,
hopefully less confusing: wa_write_clr_set(), to designate that we are
doing both clr and set operations in the register.
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20201209045246.2905675-2-lucas.demarchi@intel.com
When using masked registers, there is nothing to clear since a masked
register has the mask in the upper 16b: we can just write to the
location we want and use the mask to control what bits we are writing
to.
However that doesn't mean we don't want to read back the register and
check the value actually matched what we wanted to write, i.e. that
the WA stick. That should be an explicit opt-out for registers that are
either write-only or that are affected by hardware misbehavior.
Moreover both wa_masked_en() and wa_masked_dis() check the WA stick, so
skipping the check just because the field is more than 1 bit is
surprising and error-prone.
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20201209045246.2905675-1-lucas.demarchi@intel.com
We checked the table size against a hardcoded number of entries, and
that number was excluding the special mocs registers at the end.
Fixes: 777a7717d6 ("drm/i915/gt: Program mocs:63 for cache eviction on gen9")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: <stable@vger.kernel.org> # v4.3+
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201127102540.13117-1-chris@chris-wilson.co.uk
(cherry picked from commit 444fbf5d70)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
[backported and updated the Fixes sha]
Currently the check that the unsigned size_t variable i is >= 0
is always true because the unsigned variable will never be negative,
causing the loop to run forever. Fix this by changing the
pre-decrement check to a zero check on i followed by a decrement of i.
Addresses-Coverity: ("Unsigned compared against 0")
Fixes: bfed6708d6 ("drm/i915: use vmap in shmem_pin_map")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20201002170354.94627-1-colin.king@canonical.com
(cherry picked from commit e70956a249)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
We currently presume that the engine reset is successful, cancelling the
expired preemption timer in the process. However, engine resets can
fail, leaving the timeout still pending and we will then respond to the
timeout again next time the tasklet fires. What we want is for the
failed engine reset to be promoted to a full device reset, which is
kicked by the heartbeat once the engine stops processing events.
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/1168
Fixes: 3a7a92aba8 ("drm/i915/execlists: Force preemption")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: <stable@vger.kernel.org> # v5.5+
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201204151234.19729-2-chris@chris-wilson.co.uk
(cherry picked from commit d997e240ce)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Before reseting the engine, we suspend the execution of the guilty
request, so that we can continue execution with a new context while we
slowly compress the captured error state for the guilty context. However,
if the reset fails, we will promptly attempt to reset the same request
again, and discover the ongoing capture. Ignore the second attempt to
suspend and capture the same request.
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/1168
Fixes: 32ff621fd7 ("drm/i915/gt: Allow temporary suspension of inflight requests")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: <stable@vger.kernel.org> # v5.7+
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201204151234.19729-1-chris@chris-wilson.co.uk
(cherry picked from commit b969540500)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Currently the check that the unsigned size_t variable i is >= 0
is always true because the unsigned variable will never be negative,
causing the loop to run forever. Fix this by changing the
pre-decrement check to a zero check on i followed by a decrement of i.
Addresses-Coverity: ("Unsigned compared against 0")
Fixes: bfed6708d6 ("drm/i915: use vmap in shmem_pin_map")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20201002170354.94627-1-colin.king@canonical.com
Remove the last macro and implement it as a function like the rest of
the operations that don't assume there is a `wal` list, but rather
receive it as argument.
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20201205092542.2325477-4-lucas.demarchi@intel.com
Just ommitting the list it's operating on doesn't save much typing
and adds another way to do the same thing. Just replace it with
wa_masked_dis().
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20201205092542.2325477-3-lucas.demarchi@intel.com
Just ommitting the list it's operating on doesn't save much typing and
adds another way to do the same thing. Just replace it with
wa_masked_en().
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20201205092542.2325477-2-lucas.demarchi@intel.com
Set GS Timer to 224 to prevent a HS/DS hang.
Bspec: 53508
v2: reword commit message and add comment explaining why read
verification is ignored (Chris)
Cc: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Swathi Dhanavanthri <swathi.dhanavanthri@intel.com>
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20201205092542.2325477-1-lucas.demarchi@intel.com
Across a reset, we stop the engine but not the timers. This leaves a
window where the timers have inconsistent state with the engine, but
should only result in a spurious timeout. As we cancel the outstanding
events, also cancel their timers.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201204151234.19729-4-chris@chris-wilson.co.uk
We currently presume that the engine reset is successful, cancelling the
expired preemption timer in the process. However, engine resets can
fail, leaving the timeout still pending and we will then respond to the
timeout again next time the tasklet fires. What we want is for the
failed engine reset to be promoted to a full device reset, which is
kicked by the heartbeat once the engine stops processing events.
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/1168
Fixes: 3a7a92aba8 ("drm/i915/execlists: Force preemption")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: <stable@vger.kernel.org> # v5.5+
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201204151234.19729-2-chris@chris-wilson.co.uk
Before reseting the engine, we suspend the execution of the guilty
request, so that we can continue execution with a new context while we
slowly compress the captured error state for the guilty context. However,
if the reset fails, we will promptly attempt to reset the same request
again, and discover the ongoing capture. Ignore the second attempt to
suspend and capture the same request.
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/1168
Fixes: 32ff621fd7 ("drm/i915/gt: Allow temporary suspension of inflight requests")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: <stable@vger.kernel.org> # v5.7+
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201204151234.19729-1-chris@chris-wilson.co.uk
drm/i915 features for v5.11:
Highlights:
- Enable big joiner to join two pipes to one port to overcome pipe restrictions
(Manasi, Ville, Maarten)
Display:
- More DG1 enabling (Lucas, Aditya)
- Fixes to cases without display (Lucas, José, Jani)
- Initial PSR state improvements (José)
- JSL eDP vswing updates (Tejas)
- Handle EDID declared max 16 bpc (Ville)
- Display refactoring (Ville)
Other:
- GVT features
- Backmerge
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/87czzzkk1s.fsf@intel.com
We treat idling the GT (intel_rps_park) as a downclock event, and reduce
the frequency we intend to restart the GT with. Since the two workloads
are likely related (e.g. a compositor rendering every 16ms), we want to
carry the frequency and load information from across the idling.
However, we do also need to update the frequencies so that workloads
that run for less than 1ms are autotuned by RPS (otherwise we leave
compositors running at max clocks, draining excess power). Conversely,
if we try to run too slowly, the next workload has to run longer. Since
there is a hysteresis in the power graph, below a certain frequency
running a short workload for longer consumes more energy than running it
slightly higher for less time. The exact balance point is unknown
beforehand, but measurements with 30fps media playback indicate that RPe
is a better choice.
Reported-by: Edward Baker <edward.baker@intel.com>
Tested-by: Edward Baker <edward.baker@intel.com>
Fixes: 043cd2d14e ("drm/i915/gt: Leave rps->cur_freq on unpark")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Edward Baker <edward.baker@intel.com>
Cc: Andi Shyti <andi.shyti@intel.com>
Cc: Lyude Paul <lyude@redhat.com>
Cc: <stable@vger.kernel.org> # v5.8+
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201124183521.28623-1-chris@chris-wilson.co.uk
(cherry picked from commit f7ed83cc19)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
As we use a shmemfs file to hold the context state, when not in use it
may be swapped out, such as across suspend. Since we wrote into the
shmemfs without marking the pages as dirty, the contents may be dropped
instead of being written back to swap. On re-using the shmemfs file,
such as creating a new context after resume, the contents of that file
were likely garbage and so the new context could then hang the GPU.
Simply mark the page as being written when copying into the shmemfs
file, and it the new contents will be retained across swapout.
Fixes: be1cb55a07 ("drm/i915/gt: Keep a no-frills swappable copy of the default context state")
Cc: Sudeep Dutt <sudeep.dutt@intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Ramalingam C <ramalingam.c@intel.com>
Signed-off-by: CQ Tang <cq.tang@intel.com>
Signed-off-by: Venkata Ramana Nayana <venkata.ramana.nayana@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: <stable@vger.kernel.org> # v5.8+
Link: https://patchwork.freedesktop.org/patch/msgid/20201127120718.454037-161-matthew.auld@intel.com
(cherry picked from commit a9d71f76cc)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Allow a brief period for continued access to a dead intel_context by
deferring the release of the struct until after an RCU grace period.
As we are using a dedicated slab cache for the contexts, we can defer
the release of the slab pages via RCU, with the caveat that individual
structs may be reused from the freelist within an RCU grace period. To
handle that, we have to avoid clearing members of the zombie struct.
This is required for a later patch to handle locking around virtual
requests in the signaler, as those requests may want to move between
engines and be destroyed while we are holding b->irq_lock on a physical
engine.
v2: Drop mutex_reinit(), if we never mark the mutex as destroyed we
don't need to reset the debug code, at the loss of having the mutex
debug code spot us attempting to destroy a locked mutex.
v3: As the intended use will remain strongly referenced counted, with
very little inflight access across reuse, drop the ctor.
v4: Drop the unrequired change to remove the temporary reference around
dropping the active context, and add back some more missing ctor
operations.
v5: The ctor is back. Tvrtko spotted that ce->signal_lock [introduced
later] maybe accessed under RCU and so needs special care not to be
reinitialised.
v6: Don't mix SLAB_TYPESAFE_BY_RCU and RCU list iteration.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201126140407.31952-3-chris@chris-wilson.co.uk
(cherry picked from commit 14d1eaf088)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Ville noticed that the last mocs entry is used unconditionally by the HW
when it performs cache evictions, and noted that while the value is not
meant to be writable by the driver, we should program it to a reasonable
value nevertheless.
As it turns out, we can change the value of mocs:63 and the value we
were programming into it would cause hard hangs in conjunction with
atomic operations.
v2: Add details from bspec about how it is used by HW
Suggested-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/2707
Fixes: 3bbaba0cea ("drm/i915: Added Programming of the MOCS")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Jason Ekstrand <jason@jlekstrand.net>
Cc: <stable@vger.kernel.org> # v4.3+
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201126140841.1982-1-chris@chris-wilson.co.uk
(cherry picked from commit 977933b5da)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
We treat idling the GT (intel_rps_park) as a downclock event, and reduce
the frequency we intend to restart the GT with. Since the two workloads
are likely related (e.g. a compositor rendering every 16ms), we want to
carry the frequency and load information from across the idling.
However, we do also need to update the frequencies so that workloads
that run for less than 1ms are autotuned by RPS (otherwise we leave
compositors running at max clocks, draining excess power). Conversely,
if we try to run too slowly, the next workload has to run longer. Since
there is a hysteresis in the power graph, below a certain frequency
running a short workload for longer consumes more energy than running it
slightly higher for less time. The exact balance point is unknown
beforehand, but measurements with 30fps media playback indicate that RPe
is a better choice.
Reported-by: Edward Baker <edward.baker@intel.com>
Tested-by: Edward Baker <edward.baker@intel.com>
Fixes: 043cd2d14e ("drm/i915/gt: Leave rps->cur_freq on unpark")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Edward Baker <edward.baker@intel.com>
Cc: Andi Shyti <andi.shyti@intel.com>
Cc: Lyude Paul <lyude@redhat.com>
Cc: <stable@vger.kernel.org> # v5.8+
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201124183521.28623-1-chris@chris-wilson.co.uk
As we use a shmemfs file to hold the context state, when not in use it
may be swapped out, such as across suspend. Since we wrote into the
shmemfs without marking the pages as dirty, the contents may be dropped
instead of being written back to swap. On re-using the shmemfs file,
such as creating a new context after resume, the contents of that file
were likely garbage and so the new context could then hang the GPU.
Simply mark the page as being written when copying into the shmemfs
file, and it the new contents will be retained across swapout.
Fixes: be1cb55a07 ("drm/i915/gt: Keep a no-frills swappable copy of the default context state")
Cc: Sudeep Dutt <sudeep.dutt@intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Ramalingam C <ramalingam.c@intel.com>
Signed-off-by: CQ Tang <cq.tang@intel.com>
Signed-off-by: Venkata Ramana Nayana <venkata.ramana.nayana@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: <stable@vger.kernel.org> # v5.8+
Link: https://patchwork.freedesktop.org/patch/msgid/20201127120718.454037-161-matthew.auld@intel.com
We checked the table size against a hardcoded number of entries, and
that number was excluding the special mocs registers at the end.
Fixes: 977933b5da ("drm/i915/gt: Program mocs:63 for cache eviction on gen9")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: <stable@vger.kernel.org> # v4.3+
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201127102540.13117-1-chris@chris-wilson.co.uk
If while we are cancelling the breadcrumb signaling, we find that the
request is already completed, move it to the irq signaler and let it be
signaled.
v2: Tweak reference counting so that we only acquire a new reference on
adding to a signal list, as opposed to a hidden i915_request_put of the
caller's reference.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201126140407.31952-5-chris@chris-wilson.co.uk
Allow a brief period for continued access to a dead intel_context by
deferring the release of the struct until after an RCU grace period.
As we are using a dedicated slab cache for the contexts, we can defer
the release of the slab pages via RCU, with the caveat that individual
structs may be reused from the freelist within an RCU grace period. To
handle that, we have to avoid clearing members of the zombie struct.
This is required for a later patch to handle locking around virtual
requests in the signaler, as those requests may want to move between
engines and be destroyed while we are holding b->irq_lock on a physical
engine.
v2: Drop mutex_reinit(), if we never mark the mutex as destroyed we
don't need to reset the debug code, at the loss of having the mutex
debug code spot us attempting to destroy a locked mutex.
v3: As the intended use will remain strongly referenced counted, with
very little inflight access across reuse, drop the ctor.
v4: Drop the unrequired change to remove the temporary reference around
dropping the active context, and add back some more missing ctor
operations.
v5: The ctor is back. Tvrtko spotted that ce->signal_lock [introduced
later] maybe accessed under RCU and so needs special care not to be
reinitialised.
v6: Don't mix SLAB_TYPESAFE_BY_RCU and RCU list iteration.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201126140407.31952-3-chris@chris-wilson.co.uk
Pull the repeated check for the last active request being completed to a
single spot, when deciding whether or not execlist preemption is
required.
In doing so, we remove the tasklet kick, introduced with the completion
checks in commit 35f3fd8182 ("drm/i915/execlists: Workaround switching
back to a completed context"), if we find the request was completed but
have not yet seen the corresponding CS event. This was devolving into a
busy spin of the tasklet while we waited for the event as the delivery
was not as instantaneous as expected. Under load this is sufficient to
exhaust the tasklet softirq timeslice, and force ksoftirqd. Quite
noticeable overhead for no apparent improvement in latency.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201126140407.31952-2-chris@chris-wilson.co.uk
Since the introduction of preempt-to-busy, requests can complete in the
background, even while they are not on the engine->active.requests list.
As such, the engine->active.request list itself is not in strict
retirement order, and we have to scan the entire list while unwinding to
not miss any. However, if the request is completed we currently leave it
on the list [until retirement], but we could just as simply remove it
and stop treating it as active. We would only have to then traverse it
once while unwinding in quick succession.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201126140407.31952-1-chris@chris-wilson.co.uk
Ville noticed that the last mocs entry is used unconditionally by the HW
when it performs cache evictions, and noted that while the value is not
meant to be writable by the driver, we should program it to a reasonable
value nevertheless.
As it turns out, we can change the value of mocs:63 and the value we
were programming into it would cause hard hangs in conjunction with
atomic operations.
v2: Add details from bspec about how it is used by HW
Suggested-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/2707
Fixes: 3bbaba0cea ("drm/i915: Added Programming of the MOCS")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Jason Ekstrand <jason@jlekstrand.net>
Cc: <stable@vger.kernel.org> # v4.3+
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201126140841.1982-1-chris@chris-wilson.co.uk
Since preempt-to-busy, we may unsubmit a request while it is still on
the HW and completes asynchronously. That means it may be retired and in
the process destroy the virtual engine (as the user has closed their
context), but that engine may still be holding onto the unsubmitted
compelted request. Therefore we need to potentially cleanup the old
request on destroying the virtual engine. We also have to keep the
virtual_engine alive until after the sibling's execlists_dequeue() have
finished peeking into the virtual engines, for which we serialise with
RCU.
v2: Be paranoid and flush the tasklet as well.
v3: And flush the tasklet before the engines, as the tasklet may
re-attach an rb_node after our removal from the siblings.
Fixes: 6d06779e86 ("drm/i915: Load balancing across a virtual engine")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201123113717.20500-4-chris@chris-wilson.co.uk
(cherry picked from commit 46eecfccb4)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
We currently want to keep the interrupt enabled until the interrupt after
which we have no more work to do. This heuristic was broken by us
kicking the irq-work on adding a completed request without attaching a
signaler -- hence it appearing to the irq-worker that an interrupt had
fired when we were idle.
Fixes: 2854d86632 ("drm/i915/gt: Replace intel_engine_transfer_stale_breadcrumbs")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201123113717.20500-3-chris@chris-wilson.co.uk
(cherry picked from commit 3aef910d26)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Move the register slow register write and readback from out of the
critical path for execlists submission and delay it until the following
worker, shaving off around 200us. Note that the same signal_irq_work() is
allowed to run concurrently on each CPU (but it will only be queued once,
once running though it can be requeued and reexecuted) so we have to
remember to lock the global interactions as we cannot rely on the
signal_irq_work() itself providing the serialisation (in constrast to a
tasklet).
By pushing the arm/disarm into the central signaling worker we can close
the race for disarming the interrupt (and dropping its associated
GT wakeref) on parking the engine. If we loose the race, that GT wakeref
may be held indefinitely, preventing the machine from sleeping while
the GPU is ostensibly idle.
v2: Move the self-arming parking of the signal_irq_work to a flush of
the irq-work from intel_breadcrumbs_park().
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/2271
Fixes: e23005604b ("drm/i915/gt: Hold context/request reference while breadcrumbs are active")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201123113717.20500-1-chris@chris-wilson.co.uk
(cherry picked from commit 9d5612ca16)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
CT event handler is called under the gt->irq_lock from the interrupt
handling paths so make it the same from the init path. I don't think this
mismatch caused any functional issue but we need to wean the code of the
global i915->irq_lock.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201120095636.1987395-2-tvrtko.ursulin@linux.intel.com
Guc->mmio_msg is set under the guc->irq_lock in guc_get_mmio_msg so it
should be consumed under the same lock from guc_handle_mmio_msg.
I am not sure if the overall flow here makes complete sense but at least
the correct lock is now used.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20201120095636.1987395-1-tvrtko.ursulin@linux.intel.com
Since preempt-to-busy, we may unsubmit a request while it is still on
the HW and completes asynchronously. That means it may be retired and in
the process destroy the virtual engine (as the user has closed their
context), but that engine may still be holding onto the unsubmitted
compelted request. Therefore we need to potentially cleanup the old
request on destroying the virtual engine. We also have to keep the
virtual_engine alive until after the sibling's execlists_dequeue() have
finished peeking into the virtual engines, for which we serialise with
RCU.
v2: Be paranoid and flush the tasklet as well.
v3: And flush the tasklet before the engines, as the tasklet may
re-attach an rb_node after our removal from the siblings.
Fixes: 6d06779e86 ("drm/i915: Load balancing across a virtual engine")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201123113717.20500-4-chris@chris-wilson.co.uk
We currently want to keep the interrupt enabled until the interrupt after
which we have no more work to do. This heuristic was broken by us
kicking the irq-work on adding a completed request without attaching a
signaler -- hence it appearing to the irq-worker that an interrupt had
fired when we were idle.
Fixes: 2854d86632 ("drm/i915/gt: Replace intel_engine_transfer_stale_breadcrumbs")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201123113717.20500-3-chris@chris-wilson.co.uk
Move the register slow register write and readback from out of the
critical path for execlists submission and delay it until the following
worker, shaving off around 200us. Note that the same signal_irq_work() is
allowed to run concurrently on each CPU (but it will only be queued once,
once running though it can be requeued and reexecuted) so we have to
remember to lock the global interactions as we cannot rely on the
signal_irq_work() itself providing the serialisation (in constrast to a
tasklet).
By pushing the arm/disarm into the central signaling worker we can close
the race for disarming the interrupt (and dropping its associated
GT wakeref) on parking the engine. If we loose the race, that GT wakeref
may be held indefinitely, preventing the machine from sleeping while
the GPU is ostensibly idle.
v2: Move the self-arming parking of the signal_irq_work to a flush of
the irq-work from intel_breadcrumbs_park().
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/2271
Fixes: e23005604b ("drm/i915/gt: Hold context/request reference while breadcrumbs are active")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201123113717.20500-1-chris@chris-wilson.co.uk
The old IPS interface did not match the RPS interface that we tried to
plug it into (bool vs int return). Once repaired, our minimal
selftesting is finally happy!
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201121190352.15996-1-chris@chris-wilson.co.uk
If we run out of ring space, or exceed the desired runtime, we wish to
stop the subtest. Put these checks together, so that we always keep the
requests flushed on completion.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201120140314.24749-3-chris@chris-wilson.co.uk
Include the active timelines for debugfs/i915_engine_info, so that we
can see which have unready requests inflight which are not shown
otherwise.
Suggested-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201119165616.10834-4-chris@chris-wilson.co.uk
We plan to expand upon the number of available statuses for when we
pretty-print the requests along the timelines, and so need a new set of
flags. We have settled upon:
Unready [U]
- initial status after being submitted, the request is not
ready for execution as it is waiting for external fences
Ready [R]
- all fences the request was waiting on have been signaled,
and the request is now ready for execution and will be
in a backend queue
- a ready request may still need to wait on semaphores
[internal fences]
Ready/virtual [V]
- same as ready, but queued over multiple backends
Executing [E]
- the request has been transferred from the backend queue and
submitted for execution on HW
- a completed request may still be regarded as executing, its
status may not be updated until it is retired and removed
from the lists
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201119165616.10834-3-chris@chris-wilson.co.uk
Extract i915_request_show for reuse in other request chain pretty
printers.
For a bonus point, quietly change the seqno format from %llx to %lld to
match everywhere else.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201119165616.10834-2-chris@chris-wilson.co.uk
Forcing mocs:1 [used for our winsys follows-pte mode] to be cached
caused display glitches. Though it is documented as deprecated (and so
likely behaves as uncached) use the follow-pte bit and force it out of
L3 cache.
Testcase: igt/kms_frontbuffer_tracking
Testcase: igt/kms_big_fb
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Ayaz A Siddiqui <ayaz.siddiqui@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201015122138.30161-4-chris@chris-wilson.co.uk
(cherry picked from commit a04ac82736)
Fixes: 849c0fe9e8 ("drm/i915/gt: Initialize reserved and unspecified MOCS indices")
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
[Rodrigo: Updated Fixes tag]
Since we allocate some breadcrumbs for the virtual engine, and the
virtual engine has a custom destructor, we also need to free the
breadcrumbs after use.
Fixes: b3786b2937 ("drm/i915/gt: Distinguish the virtual breadcrumbs from the irq breadcrumbs")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201118133839.1783-1-chris@chris-wilson.co.uk
(cherry picked from commit 45e50f48b7)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Since we allocate some breadcrumbs for the virtual engine, and the
virtual engine has a custom destructor, we also need to free the
breadcrumbs after use.
Fixes: b3786b2937 ("drm/i915/gt: Distinguish the virtual breadcrumbs from the irq breadcrumbs")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201118133839.1783-1-chris@chris-wilson.co.uk
The presumption was that some time would always elapse between recording
the start and the finish of a context switch. This turns out to be a
regular occurrence and emitting a debug statement superfluous.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201117113103.21480-4-chris@chris-wilson.co.uk
Just like for rkl and tgl, this should be permanent as well for dg1
instead just for A0. The commit making it permanent for those platforms
ended up "racing" with the commit adding the DG1 WAs, so now fix that up.
v2: Add "tgl,dg1" to WA comment (Matt)
Cc: Swathi Dhanavanthri <swathi.dhanavanthri@intel.com>
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201027043228.696518-3-lucas.demarchi@intel.com
I forgot to free the old list when growing past 16 entries.
Luckily, as much as I checked, none of the current platforms has more than
16 workarounds on a single list.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Fixes: 452420d22d ("drm/i915: Fuse per-context workaround handling with the common framework")
Reported-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20201113132510.2298483-1-tvrtko.ursulin@linux.intel.com
(cherry picked from commit 77c296966e)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Some media power gates are disabled by default. commit 5d86923060
("drm/i915/tgl: Enable VD HCP/MFX sub-pipe power gating")
tried to enable it, but it duplicated an existent register.
So, the main PG setup sequences ended up overwriting it.
So, let's now merge this to the main PG setup sequence.
v2: (Chris): s/BIT/REG_BIT, remove useless comment,
remove useless =0, use the right gt,
remove rc6 sequence doubt from commit message.
Fixes: 5d86923060 ("drm/i915/tgl: Enable VD HCP/MFX sub-pipe power gating")
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: stable@vger.kernel.org#v5.5+
Cc: Dale B Stimson <dale.b.stimson@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20201111072859.1186070-1-rodrigo.vivi@intel.com
(cherry picked from commit 695dc55b57)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
I forgot to free the old list when growing past 16 entries.
Luckily, as much as I checked, none of the current platforms has more than
16 workarounds on a single list.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Fixes: 452420d22d ("drm/i915: Fuse per-context workaround handling with the common framework")
Reported-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20201113132510.2298483-1-tvrtko.ursulin@linux.intel.com
Cross-subsystem Changes:
- DMA mapped scatterlist fixes in i915 to unblock merging of
https://lkml.org/lkml/2020/9/27/70 (Tvrtko, Tom)
Driver Changes:
- Fix for user reported issue #2381 (Graphical output stops with "switching to inteldrmfb from simple"):
Mark ininitial fb obj as WT on eLLC machines to avoid rcu lockup during fbdev init (Ville, Chris)
- Fix for Tigerlake (and earlier) to avoid spurious empty CSB events leading to hang (Chris, Bruce)
- Delay execlist processing for Tigerlake to avoid hang (Chris)
- Fix for Tigerlake RCS engine health check through heartbeat (Chris)
- Fix for Tigerlake reserved MOCS entries (Ayaz, Chris)
- Fix Media power gate sequence on Tigerlake (Rodrigo)
- Enable eLLC caching of display buffers for SKL+ (Ville)
- Support parsing of oversize batches on Gen9 (Matt, Chris)
- Exclude low pages (128KiB) of stolen from use to avoid thrashing during reset (Chris)
- Flush engines before Tigerlake breadcrumbs (Chris)
- Use the local HWSP offset during submission (Chris)
- Flush coherency domains on first set-domain-ioctl (Chris, Zbigniew)
- Use the active reference on the vma while capturing to avoid use-after-free (Chris)
- Fix MOCS PTE setting for gen9+ (Ville)
- Avoid NULL dereference on IPS driver callback while unbinding i915 (Chris)
- Avoid NULL dereference from PT/PD stash allocation error (Matt)
- Hold request reference for canceling an active context (Chris)
- Avoid infinite loop on x86-32 when mapping a lot of objects (Chris)
- Disallow WC mappings when processor doesn't support them (Chris)
- Return correct error in i915_gem_object_copy_blt() error path (Dan)
- Return correct error in intel_context_create_request() error path (Maarten)
- Tune down GuC communication enabled/disabled messages to debug (Jani)
- Fix rebased commit "Remove i915_request.lock requirement for execution callbacks" (Chris)
- Cancel outstanding work after disabling heartbeats on an engine (Chris)
- Signal cancelled requests (Chris)
- Retire cancelled requests on unload (Chris)
- Scrub HW state on driver remove (Chris)
- Undo forced context restores after trivial preemptions (Chris)
- Handle PCI unbind in PMU code (Tvrtko)
- Fix CPU hotplug with multiple GPUs in PMU code (Trtkko)
- Correctly set SFC capability for video engines (Venkata)
- Update GuC code to use firmware v49.0.1 (John, Matthew B., Daniele, Oscar, Michel, Rodrigo, Michal)
- Improve GuC warnings on loading failure (John)
- Avoid ownership race in buffer pool by clearing age (Chris)
- Use MMIO to read CSB in case of failure (Chris, Mika)
- Show engine properties in engine state dump to indicate changes (Chris, Joonas)
- Break up error capture compression loops with cond_resched() (Chris)
- Reduce GPU error capture mutex hold time to avoid khungtaskd (Chris)
- Serialise debugfs i915_gem_objects with ctx->mutex (Chris)
- Always test execution status on closing the context and close if not persistent (Chris)
- Avoid mixing integer types during batch copies (Chris, Jared)
- Skip over MI_NOOP when parsing to avoid overhead (Chris)
- Hold onto an explicit ref to i915_vma_work.pinned (Chris)
- Perform all asynchronous waits prior to marking payload start (Chris)
- Pull phys pread/pwrite implementations to the backend (Matt)
- Improve record of hung engines in error state (Tvrtko)
- Allow backends to override pread implementation (Matt)
- Reinforce LRC poisoning checks to confirm context survives execution (Chris)
- Fix memory region max size calculation (Matt)
- Fix order when adding blocks to memory region (Matt)
- Eliminate unused intel_virtual_engine_get_sibling func (Chris)
- Cleanup kasan warning for on-stack (unsigned long) casting (Chris)
- Onion unwind for scratch page allocation failure (Chris)
- Poison stolen pages before use (Chris)
- Selftest improvements (Chris)
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201112163407.GA20320@jlahtine-mobl.ger.corp.intel.com
Some media power gates are disabled by default. commit 5d86923060
("drm/i915/tgl: Enable VD HCP/MFX sub-pipe power gating")
tried to enable it, but it duplicated an existent register.
So, the main PG setup sequences ended up overwriting it.
So, let's now merge this to the main PG setup sequence.
v2: (Chris): s/BIT/REG_BIT, remove useless comment,
remove useless =0, use the right gt,
remove rc6 sequence doubt from commit message.
Fixes: 5d86923060 ("drm/i915/tgl: Enable VD HCP/MFX sub-pipe power gating")
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: stable@vger.kernel.org#v5.5+
Cc: Dale B Stimson <dale.b.stimson@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20201111072859.1186070-1-rodrigo.vivi@intel.com
-----BEGIN PGP SIGNATURE-----
iQFSBAABCAA8FiEEq68RxlopcLEwq+PEeb4+QwBBGIYFAl+oiOgeHHRvcnZhbGRz
QGxpbnV4LWZvdW5kYXRpb24ub3JnAAoJEHm+PkMAQRiGKBQIAJw6oad/FA7j9OO2
dMoaXb8UaBehGWgW2rdfWrFPV5v0DBnp/GkdRpLoZIjV3W4mBfnog7bIa4Eswlxo
Y8sZxo5/3JlgJQUkHvzR1TYk5z61lHkUw9Kj/cCyx6YdbjSl19AfFsnhQVVMuyp9
TXL2c7hxkHlw8eBGrymVu0Ip7Zq0x8O2g+8nQpmRcvaR6SBuSHdikDF/iWCtU1YW
wpk5eWEVaAO67keZOz6b+aCFHqjFX+1dUBBuPnslucYLR73Qi16hfaU9pebe97Gb
lX/MJ1bR9BeRp314cF0PYbm4WhKjRLudHOFJH8x3dj/BiYNrFK3SJGLiiTwsrAZ8
kytU0Xs=
=Ke/D
-----END PGP SIGNATURE-----
Merge v5.10-rc3 into drm-next
We need commit f8f6ae5d07 ("mm: always have io_remap_pfn_range() set
pgprot_decrypted()") to be able to merge Jason's cleanup patch.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Between events which trigger engine and GPU resets and capturing the error
state we lose information on which engine triggered the reset. Improve
this by passing in the hung engine mask down to error capture.
Result is that the list of engines in user visible "GPU HANG: ecode
<gen>:<engines>:<ecode>, <process>" is now a list of hanging and not just
active engines. Most importantly the displayed process is now the one
which was actually hung.
v2:
* Stub prototype. (Chris)
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20201104134743.916027-1-tvrtko.ursulin@linux.intel.com
SFC capability of video engines is not set correctly because i915
is testing for incorrect bits.
Fixes: c5d3e39caa ("drm/i915: Engine discovery query")
Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Venkata Sandeep Dhanalakota <venkata.s.dhanalakota@intel.com>
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: <stable@vger.kernel.org> # v5.3+
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20201106011842.36203-1-daniele.ceraolospurio@intel.com
In a simple test case that writes to scratch and then busy-waits for the
batch to be signaled, we observe that the signal is before the write is
posted. That is bad news.
Splitting the flush + write_dword into two separate flush_dw prevents
the issue from being reproduced, we can presume the post-sync op is not
so post-sync.
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/216
Testcase: igt/gem_exec_fence/parallel
Testcase: igt/i915_selftest/live/gt_timelines
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: stable@vger.kernel.org
Acked-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201102221057.29626-2-chris@chris-wilson.co.uk
(cherry picked from commit 09212e81e5)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Add another lower level to emit_ggtt_write so that the GGTT nature of
the write is not hardcoded into the emitter.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201102221057.29626-1-chris@chris-wilson.co.uk
(cherry picked from commit 2739d8cfc5)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
We wrap the timeline on construction of the next request, but there may
still be requests in flight that have not yet finalized the breadcrumb.
(The breadcrumb is delayed as we need engine-local offsets, and for the
virtual engine that is not known until execution.) As such, by the time
we write to the timeline's HWSP offset it may have changed, and we
should use the value we preserved in the request instead.
Though the window is small and infrequent (at full flow we can expect a
timeline's seqno to wrap once every 30 minutes), the impact of writing
the old seqno into the new HWSP is severe: the old requests are never
completed, and the new requests are completed before they are even
submitted.
Fixes: ebece75392 ("drm/i915: Keep timeline HWSP allocated until idle across the system")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: <stable@vger.kernel.org> # v5.2+
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201022064127.10159-1-chris@chris-wilson.co.uk
(cherry picked from commit c10f6019d0)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
In a simple test case that writes to scratch and then busy-waits for the
batch to be signaled, we observe that the signal is before the write is
posted. That is bad news.
Splitting the flush + write_dword into two separate flush_dw prevents
the issue from being reproduced, we can presume the post-sync op is not
so post-sync.
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/216
Testcase: igt/gem_exec_fence/parallel
Testcase: igt/i915_selftest/live/gt_timelines
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: stable@vger.kernel.org
Acked-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201102221057.29626-2-chris@chris-wilson.co.uk
Clear out some pointers when objects have been de-allocated. This
makes it much easier to track down use-after-free type issues.
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201028145826.2949180-4-John.C.Harrison@Intel.com
Rather than just saying 'GuC failed to load: -110', actually print out
the GuC status register and break it down into the individual fields.
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201028145826.2949180-3-John.C.Harrison@Intel.com
The latest GuC firmware includes a number of interface changes that
require driver updates to match.
* Starting from Gen11, the ID to be provided to GuC needs to contain
the engine class in bits [0..2] and the instance in bits [3..6].
NOTE: this patch breaks pointer dereferences in some existing GuC
functions that use the guc_id to dereference arrays but these functions
are not used for now as we have GuC submission disabled and we will
update these functions in follow up patch which requires new IDs.
* The new GuC requires the additional data structure (ADS) and associated
'private_data' pointer to be setup. This is basically a scratch area
of memory that the GuC owns. The size is read from the CSS header.
* There is now a physical to logical engine mapping table in the ADS
which needs to be configured in order for the firmware to load. For
now, the table is initialised with a 1 to 1 mapping.
* GUC_CTL_CTXINFO has been removed from the initialization params.
* reg_state_buffer is maintained internally by the GuC as part of
the private data.
* The ADS layout has changed significantly. This patch updates the
shared structure and also adds better documentation of the layout.
* While i915 does not use GuC doorbells, the firmware now requires
that some initialisation is done.
* The number of engine classes and instances supported in the ADS has
been increased.
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Oscar Mateo <oscar.mateo@intel.com>
Signed-off-by: Michel Thierry <michel.thierry@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Michal Winiarski <michal.winiarski@intel.com>
Cc: Tomasz Lis <tomasz.lis@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201028145826.2949180-2-John.C.Harrison@Intel.com
fbcon/fonts:
- Two patches to prevent OOB access
ttm:
- fix for evicition value range check
amdgpu:
- Sienna Cichlid fixes
- MST manager resource leak fix
- GPU reset fix
amdkfd:
- Luxmark fix for Navi1x
i915:
- Tweak initial DPCD backlight.enabled value (Sean)
- Initialize reserved MOCS indices (Ayaz)
- Mark initial fb obj as WT on eLLC machines to avoid rcu lockup (Ville)
- Support parsing of oversize batches (Chris)
- Delay execlists processing for TGL (Chris)
- Use the active reference on the vma during error capture (Chris)
- Widen CSB pointer (Chris)
- Wait for CSB entries on TGL (Chris)
- Fix unwind for scratch page allocation (Chris)
- Exclude low patches of stolen memory (Chris)
- Force VT'd workarounds when running as a guest OS (Chris)
- Drop runtime-pm assert from vpgu io accessors (Chris)
-----BEGIN PGP SIGNATURE-----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=rAUM
-----END PGP SIGNATURE-----
Merge tag 'drm-next-2020-10-23' of git://anongit.freedesktop.org/drm/drm
Pull more drm fixes from Dave Airlie:
"This should be the last round of things for rc1, a bunch of i915
fixes, some amdgpu, more font OOB fixes and one ttm fix just found
reading code:
fbcon/fonts:
- Two patches to prevent OOB access
ttm:
- fix for evicition value range check
amdgpu:
- Sienna Cichlid fixes
- MST manager resource leak fix
- GPU reset fix
amdkfd:
- Luxmark fix for Navi1x
i915:
- Tweak initial DPCD backlight.enabled value (Sean)
- Initialize reserved MOCS indices (Ayaz)
- Mark initial fb obj as WT on eLLC machines to avoid rcu lockup (Ville)
- Support parsing of oversize batches (Chris)
- Delay execlists processing for TGL (Chris)
- Use the active reference on the vma during error capture (Chris)
- Widen CSB pointer (Chris)
- Wait for CSB entries on TGL (Chris)
- Fix unwind for scratch page allocation (Chris)
- Exclude low patches of stolen memory (Chris)
- Force VT'd workarounds when running as a guest OS (Chris)
- Drop runtime-pm assert from vpgu io accessors (Chris)"
* tag 'drm-next-2020-10-23' of git://anongit.freedesktop.org/drm/drm: (31 commits)
drm/amdgpu: correct the cu and rb info for sienna cichlid
drm/amd/pm: remove the average clock value in sysfs
drm/amd/pm: fix pp_dpm_fclk
Revert drm/amdgpu: disable sienna chichlid UMC RAS
drm/amd/pm: fix pcie information for sienna cichlid
drm/amdkfd: Use same SQ prefetch setting as amdgpu
drm/amd/swsmu: correct wrong feature bit mapping
drm/amd/psp: Fix sysfs: cannot create duplicate filename
drm/amd/display: Avoid MST manager resource leak.
drm/amd/display: Revert "drm/amd/display: Fix a list corruption"
drm/amdgpu: update golden setting for sienna_cichlid
drm/amd/swsmu: add missing feature map for sienna_cichlid
drm/amdgpu: correct the gpu reset handling for job != NULL case
drm/amdgpu: add rlc iram and dram firmware support
drm/amdgpu: add function to program pbb mode for sienna cichlid
drm/i915: Drop runtime-pm assert from vgpu io accessors
drm/i915: Force VT'd workarounds when running as a guest OS
drm/i915: Exclude low pages (128KiB) of stolen from use
drm/i915/gt: Onion unwind for scratch page allocation failure
drm/ttm: fix eviction valuable range check.
...
intel_timeline_read_hwsp() is used to support semaphore waits between
engines, that may themselves be deferred for arbitrary periods -- that
is the read of the target request's HWSP is at an indeterminant point in
the future. To support this, we need to prevent overwriting a HWSP that
is being watched across a seqno wrap (otherwise the next request will
write its value into the old HWSP preventing the watcher from making
progress, ad infinitum.) To simulate the observer across a wrap, let's
create a request that reads from the HWSP and dispatch it at different
points around a wrap to see if the value is lost.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201021220411.5777-2-chris@chris-wilson.co.uk
We wrap the timeline on construction of the next request, but there may
still be requests in flight that have not yet finalized the breadcrumb.
(The breadcrumb is delayed as we need engine-local offsets, and for the
virtual engine that is not known until execution.) As such, by the time
we write to the timeline's HWSP offset it may have changed, and we
should use the value we preserved in the request instead.
Though the window is small and infrequent (at full flow we can expect a
timeline's seqno to wrap once every 30 minutes), the impact of writing
the old seqno into the new HWSP is severe: the old requests are never
completed, and the new requests are completed before they are even
submitted.
Fixes: ebece75392 ("drm/i915: Keep timeline HWSP allocated until idle across the system")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: <stable@vger.kernel.org> # v5.2+
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201022064127.10159-1-chris@chris-wilson.co.uk
Since Ironlake uses intel_ips.ko for its dynamic frequency adjustment,
we do not have direct control over the frequency management so such
tests are defunct. Similarly, we can't check the gen6+ RPS registers on
Ironlake.
Hopefully this catches all the invalid tests now that Ironlake has
rejoined the dynamic GPU frequency club. There is an opportunity for the
reader to add tests to exercise MEMINTRSTS and co.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201022210814.23004-1-chris@chris-wilson.co.uk
Let's unmask the PCU event irq _after_ we've set up the
hardware and software to deal with the fallout. We can
also drop the PCU event bit from DEIER except when we
need it for rps.
And on the disable side we replace the hand rolled (and
unlocked) DEIER/IIR/IMR frobbing with ilk_disable_display_irq().
Ocd does require me to reorder it to be symmetric with
the enable path however.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201021131443.25616-5-ville.syrjala@linux.intel.com
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
There is no GEN6_RPSTAT1 on ILK. Instead of reading that let's
try to get the same information from MEMSTAT_ILK. At least it
seems to track MEMSWCTL frequency request perfectly on my ILK.
It needs the same invert trick as the request value.
We don't want to put the invert thing into intel_gpu_freq()
and intel_freq_opcode() because that would incorrectly invert
the min/max/etc frequencies also.
One day someone might want to reverse engineer the formula for
converting these numbers to Hz, but for now we'll just report
them raw.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201021131443.25616-2-ville.syrjala@linux.intel.com
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
In order to test how fast the heartbeat can respond, we measure with the
interval set to its minimum. Before we measure though, we want to be
sure we start with a fresh pulse, and so wait until any old one is
complete. During that wait though, we were continually flushing the
work, and so continually re-evaluating to see if the pulse was complete,
and each attempt would count as an unresponsive system. If the engine
did not complete the request in the couple of busy-spins, it would flag
an error. This is unfortunate, so let's not busy-spin waiting for the
old heartbeat, but terminate it and start afresh.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201019142841.32273-1-chris@chris-wilson.co.uk
The GPU is trashing the low pages of its reserved memory upon reset. If
we are using this memory for ringbuffers, then we will dutiful resubmit
the trashed rings after the reset causing further resets, and worse. We
must exclude this range from our own use. The value of 128KiB was found
by empirical measurement (and verified now with a selftest) on gen9.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: stable@vger.kernel.org
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201019165005.18128-2-chris@chris-wilson.co.uk
(cherry picked from commit d3606757e6)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
In switching to using objects for our ppGTT scratch pages, care was not
taken to avoid trying to unref NULL objects on failure. And for gen6
ppGTT, it appears we forgot entirely to unwind after a partial allocation
failure.
Fixes: 89351925a4 ("drm/i915/gt: Switch to object allocations for page directories")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201019083444.1286-1-chris@chris-wilson.co.uk
(cherry picked from commit fa812ce96a)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
The GPU is trashing the low pages of its reserved memory upon reset. If
we are using this memory for ringbuffers, then we will dutiful resubmit
the trashed rings after the reset causing further resets, and worse. We
must exclude this range from our own use. The value of 128KiB was found
by empirical measurement (and verified now with a selftest) on gen9.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: stable@vger.kernel.org
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201019165005.18128-2-chris@chris-wilson.co.uk
On Tigerlake, we are seeing a repeat of commit d8f5053117 ("drm/i915/icl:
Forcibly evict stale csb entries") where, presumably, due to a missing
Global Observation Point synchronisation, the write pointer of the CSB
ringbuffer is updated _prior_ to the contents of the ringbuffer. That is
we see the GPU report more context-switch entries for us to parse, but
those entries have not been written, leading us to process stale events,
and eventually report a hung GPU.
However, this effect appears to be much more severe than we previously
saw on Icelake (though it might be best if we try the same approach
there as well and measure), and Bruce suggested the good idea of resetting
the CSB entry after use so that we can detect when it has been updated by
the GPU. By instrumenting how long that may be, we can set a reliable
upper bound for how long we should wait for:
513 late, avg of 61 retries (590 ns), max of 1061 retries (10099 ns)
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/2045
References: d8f5053117 ("drm/i915/icl: Forcibly evict stale csb entries")
References: HSDES#22011327657, HSDES#1508287568
Suggested-by: Bruce Chang <yu.bruce.chang@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Bruce Chang <yu.bruce.chang@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: stable@vger.kernel.org # v5.4
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200915134923.30088-2-chris@chris-wilson.co.uk
(cherry picked from commit 233c1ae3c8)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
A CSB entry is 64b, and it is simpler for us to treat it as an array of
64b entries than as an array of pairs of 32b entries.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200915134923.30088-1-chris@chris-wilson.co.uk
(cherry picked from commit f24a44e52f)
(cherry picked from commit 3d4dbe0e0f0d04ebcea917b7279586817da8cf46)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
We may try to preempt the currently executing request, only to find that
after unravelling all the dependencies that the original executing
context is still the earliest in the topological sort and re-submitted
back to HW (if we do detect some change in the ELSP that requires
re-submission). However, due to the way we check for wrap-around during
the unravelling, we mark any context that has been submitted just once
(i.e. with the rq->wa_tail set, but the ring->tail earlier) as
potentially wrapping and requiring a forced restore on resubmission.
This was expected to be not a problem, as it was anticipated that most
unwinding for preemption would result in a context switch and the few
that did not would be lost in the noise. It did not take long for
someone to find one particular workload where the cost of those extra
context restores was measurable.
However, since we know the wa_tail is of fixed size, and we know that a
request must be larger than the wa_tail itself, we can safely maintain
the check for request wrapping and check against a slightly future point
in the ring that includes an expected wa_tail. (That is if the
ring->tail is already set to rq->wa_tail, including another 8 bytes in
the check does not invalidate the incremental wrap detection.)
Fixes: 8ab3a3812a ("drm/i915/gt: Incrementally check for rewinding")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Bruce Chang <yu.bruce.chang@intel.com>
Cc: Ramalingam C <ramalingam.c@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: <stable@vger.kernel.org> # v5.4+
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201002083425.4605-1-chris@chris-wilson.co.uk
(cherry picked from commit bb65548e3c)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
When running gem_exec_nop, it floods the system with many requests (with
the goal of userspace submitting faster than the HW can process a single
empty batch). This causes the driver to continually resubmit new
requests onto the end of an active context, a flood of lite-restore
preemptions. If we time this just right, Tigerlake hangs.
Inserting a small delay between the processing of CS events and
submitting the next context, prevents the hang. Naturally it does not
occur with debugging enabled. The suspicion then is that this is related
to the issues with the CS event buffer, and inserting an mmio read of
the CS pointer status appears to be very successful in preventing the
hang. Other registers, or uncached reads, or plain mb, do not prevent
the hang, suggesting that register is key -- but that the hang can be
prevented by a simple udelay, suggests it is just a timing issue like
that encountered by commit 233c1ae3c8 ("drm/i915/gt: Wait for CSB
entries on Tigerlake"). Also note that the hang is not prevented by
applying CTX_DESC_FORCE_RESTORE, or by inserting a delay on the GPU
between requests.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Bruce Chang <yu.bruce.chang@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: stable@vger.kernel.org
Acked-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201015195023.32346-1-chris@chris-wilson.co.uk
(cherry picked from commit 6ca7217dff)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
In order to avoid functional breakage of mis-programmed applications that
have grown to depend on unused MOCS entries, we are programming
those entries to be equal to fully cached ("L3 + LLC") entry.
These reserved and unspecified entries should not be used as they may be
changed to less performant variants with better coherency in the future
if more entries are needed.
v2: As suggested by Lucas De Marchi to utilise __init_mocs_table for
programming default value, setting I915_MOCS_PTE index of tgl_mocs_table
with desired value.
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Tomasz Lis <tomasz.lis@intel.com>
Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Francisco Jerez <currojerez@riseup.net>
Cc: Mathew Alwin <alwin.mathew@intel.com>
Cc: Mcguire Russell W <russell.w.mcguire@intel.com>
Cc: Spruit Neil R <neil.r.spruit@intel.com>
Cc: Zhou Cheng <cheng.zhou@intel.com>
Cc: Benemelis Mike G <mike.g.benemelis@intel.com>
Signed-off-by: Ayaz A Siddiqui <ayaz.siddiqui@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Acked-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20200729102539.134731-2-ayaz.siddiqui@intel.com
Cc: stable@vger.kernel.org
(cherry picked from commit 4d8a5cfe3b)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
In switching to using objects for our ppGTT scratch pages, care was not
taken to avoid trying to unref NULL objects on failure. And for gen6
ppGTT, it appears we forgot entirely to unwind after a partial allocation
failure.
Fixes: 89351925a4 ("drm/i915/gt: Switch to object allocations for page directories")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201019083444.1286-1-chris@chris-wilson.co.uk
shmem_pin_map somewhat awkwardly reimplements vmap using alloc_vm_area and
manual pte setup. The only practical difference is that alloc_vm_area
prefeaults the vmalloc area PTEs, which doesn't seem to be required here
(and could be added to vmap using a flag if actually required). Switch to
use vmap, and use vfree to free both the vmalloc mapping and the page
array, as well as dropping the references to each page.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Juergen Gross <jgross@suse.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: "Matthew Wilcox (Oracle)" <willy@infradead.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Nitin Gupta <ngupta@vflare.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Stefano Stabellini <sstabellini@kernel.org>
Cc: Uladzislau Rezki (Sony) <urezki@gmail.com>
Link: https://lkml.kernel.org/r/20201002122204.1534411-7-hch@lst.de
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Repeat our sanitychecks from before execution to after execution. One
expects that if we were to see these, the gpu would already be on fire,
but the timing may be informative.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201015190816.31763-1-chris@chris-wilson.co.uk
We may try to preempt the currently executing request, only to find that
after unravelling all the dependencies that the original executing
context is still the earliest in the topological sort and re-submitted
back to HW (if we do detect some change in the ELSP that requires
re-submission). However, due to the way we check for wrap-around during
the unravelling, we mark any context that has been submitted just once
(i.e. with the rq->wa_tail set, but the ring->tail earlier) as
potentially wrapping and requiring a forced restore on resubmission.
This was expected to be not a problem, as it was anticipated that most
unwinding for preemption would result in a context switch and the few
that did not would be lost in the noise. It did not take long for
someone to find one particular workload where the cost of those extra
context restores was measurable.
However, since we know the wa_tail is of fixed size, and we know that a
request must be larger than the wa_tail itself, we can safely maintain
the check for request wrapping and check against a slightly future point
in the ring that includes an expected wa_tail. (That is if the
ring->tail is already set to rq->wa_tail, including another 8 bytes in
the check does not invalidate the incremental wrap detection.)
Fixes: 8ab3a3812a ("drm/i915/gt: Incrementally check for rewinding")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Bruce Chang <yu.bruce.chang@intel.com>
Cc: Ramalingam C <ramalingam.c@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: <stable@vger.kernel.org> # v5.4+
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201002083425.4605-1-chris@chris-wilson.co.uk
When running gem_exec_nop, it floods the system with many requests (with
the goal of userspace submitting faster than the HW can process a single
empty batch). This causes the driver to continually resubmit new
requests onto the end of an active context, a flood of lite-restore
preemptions. If we time this just right, Tigerlake hangs.
Inserting a small delay between the processing of CS events and
submitting the next context, prevents the hang. Naturally it does not
occur with debugging enabled. The suspicion then is that this is related
to the issues with the CS event buffer, and inserting an mmio read of
the CS pointer status appears to be very successful in preventing the
hang. Other registers, or uncached reads, or plain mb, do not prevent
the hang, suggesting that register is key -- but that the hang can be
prevented by a simple udelay, suggests it is just a timing issue like
that encountered by commit 233c1ae3c8 ("drm/i915/gt: Wait for CSB
entries on Tigerlake"). Also note that the hang is not prevented by
applying CTX_DESC_FORCE_RESTORE, or by inserting a delay on the GPU
between requests.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Bruce Chang <yu.bruce.chang@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: stable@vger.kernel.org
Acked-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201015195023.32346-1-chris@chris-wilson.co.uk
DG1 shares some workarounds with TGL and RKL and also has some
additional workarounds of its own.
v2: Correct location of Wa_1408615072 (JohnH).
v3: Apply WAs 1606700617, 18011464164 and 22010931296 to DG1 (José)
v4 (Anusha)
- Add Wa_22010271021
- s/Wa_14010096844/Wa_1409836686
v5:
- Extend Wa_14010919138 to all revs (Matt Atwood)
- Power gate media is global gen12 design. (Rodrigo)
- Rebase (Lucas)
v6: use REG_BIT() to fix checkpatch warning (Lucas)
BSpec: 53508
Cc: Matt Atwood <matthew.s.atwood@intel.com>
Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Cc: José Roberto de Souza <jose.souza@intel.com>
Signed-off-by: Stuart Summers <stuart.summers@intel.com>
Signed-off-by: Anusha Srivatsa <anusha.srivatsa@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201014191937.1266226-8-lucas.demarchi@intel.com
Forcing mocs:1 [used for our winsys follows-pte mode] to be cached
caused display glitches. Though it is documented as deprecated (and so
likely behaves as uncached) use the follow-pte bit and force it out of
L3 cache.
Fixes: 4d8a5cfe3b ("drm/i915/gt: Initialize reserved and unspecified MOCS indices")
Testcase: igt/kms_frontbuffer_tracking
Testcase: igt/kms_big_fb
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Ayaz A Siddiqui <ayaz.siddiqui@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201015122138.30161-4-chris@chris-wilson.co.uk
Since SKL the eLLC has been sitting on the far side of the system
agent, meaning the display engine can utilize it. Let's enable that.
I chose WB for the caching mode, because my numbers are indicating
that WT might actually be WB and WC might actually be UC. I'm not
100% sure that is indeed the case but at least my simple rendercopy
based benchmark didn't see any difference in performance.
Also if I configure things to do LLCeLLC+WT I still get cache dirt
on my screen, suggesting that is in fact operating in WB mode
anyway. This is also the reason I had to fix the MOCS target cache
to really say PTE rather than LLC+eLLC.
Since SKL the eLLC has been sitting on the far side of the system agent,
meaning the display engine can utilize it. Let's enable that.
Eero's earlier benchmarks numbers:
"* Results in GfxBench and Unigine (Valley/Heaven) tests were within daily
variation on the tested SKL machines
* SKL GT4e (128MB eLLC) / Wayland / Weston:
+15-20% SynMark TexMem512 (512MB of textures)
+4-6% SynMark TerrainFly*, CSCloth, ShMapVsm
-5-10% SynMark TexMem128 (128MB of textures)
* SKL GT3e (64MB eLLC) / Xorg / Unity:
+4-8% GpuTest Triangle fullscreen (FullHD)
-5-10% GpuTest Triangle windowed (1/2 screen)
* SKL GT2 (no eLLC) / Xorg / Unity:
* Some of the higher FPS SynMark pixel and vertex shader tests
are few percent higher, more than daily variance
=> Do you see any reason why this machine would be impacted
although it doesn't eLLC?"
Caveats:
- Still haven't tested with a prime setup
- Still not entirely sure this a good idea, but I've been
using it on my cfl anyway :)
v2: Split the MOCS PTE change out
Cc: Eero Tamminen <eero.t.tamminen@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20201007120329.17076-3-ville.syrjala@linux.intel.com
Link: https://patchwork.freedesktop.org/patch/msgid/20201015122138.30161-3-chris@chris-wilson.co.uk
Recently we came across requirement to identify EHL and JSL
platform to program them differently. Thus Split the basic
platform definition, macros, and PCI IDs to differentiate
between EHL and JSL platforms. Also, IS_ELKHARTLAKE is replaced
with IS_JSL_EHL everywhere.
Changes since V1 :
- Rebased to avoid merge conflicts
- Added missed check for jasperlake in intel_uc_fw.c
Cc : Matt Roper <matthew.d.roper@intel.com>
Cc : Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Tejas Upadhyay <tejaskumarx.surendrakumar.upadhyay@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201013192948.63470-1-tejaskumarx.surendrakumar.upadhyay@intel.com
In order to avoid functional breakage of mis-programmed applications that
have grown to depend on unused MOCS entries, we are programming
those entries to be equal to fully cached ("L3 + LLC") entry.
These reserved and unspecified entries should not be used as they may be
changed to less performant variants with better coherency in the future
if more entries are needed.
v2: As suggested by Lucas De Marchi to utilise __init_mocs_table for
programming default value, setting I915_MOCS_PTE index of tgl_mocs_table
with desired value.
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Tomasz Lis <tomasz.lis@intel.com>
Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Francisco Jerez <currojerez@riseup.net>
Cc: Mathew Alwin <alwin.mathew@intel.com>
Cc: Mcguire Russell W <russell.w.mcguire@intel.com>
Cc: Spruit Neil R <neil.r.spruit@intel.com>
Cc: Zhou Cheng <cheng.zhou@intel.com>
Cc: Benemelis Mike G <mike.g.benemelis@intel.com>
Signed-off-by: Ayaz A Siddiqui <ayaz.siddiqui@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Acked-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20200729102539.134731-2-ayaz.siddiqui@intel.com
Cc: stable@vger.kernel.org
The updated bspec forcewake table also provides us with new multicast
ranges that should be reflected in our workaround code.
Note that there are different types of multicast registers with
different styles of replication and different steering registers. The
i915 MCR range lists we're updating here are only used to ensure we can
verify workarounds properly (i.e., if we can't steer register reads we
don't want to verify workarounds where an unsteered read might hit a
fused-off instance of the unit). Because of this, we don't need to
include any of the multicast ranges where all instances of the register
will always present and fusing doesn't play a role. Specifically, that
means that we are not including the MCR ranges designated as "SQIDI" in
the bspec.
Bspec: 66696
Cc: Caz Yokoyama <caz.yokoyama@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: José Roberto de Souza <jose.souza@intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201009194442.3668677-4-matthew.d.roper@intel.com
Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
The power well that we've been referring to as the 'blitter' well is
actually more of a general GT power well which contains a lot of things
other than the blitter engine registers. The FORCEWAKE_BLITTER name in
the code was used for historic reasons, but no longer matches how the
bspec describes this power well and just causes confusion for people not
familiar with this area of the code. Let's rename it to FORCEWAKE_GT to
more accurately describe the role of the power well and match how the
modern bspec refers to it.
v2:
- Add a comment noting that the GT power well includes the blitter
engine. (Jose)
Bspec: 66696, 66534, 67609
Cc: José Roberto de Souza <jose.souza@intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201009194442.3668677-2-matthew.d.roper@intel.com
Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
DG1 has a new MOCS table. We still use the old definition of the table,
but as for any dgfx card it doesn't contain the control_value values
(these values don't matter as we won't program them).
Bspec: 45101
v2: Reword the comment to state that the last few entries are reserved
instead of "the last two". DG1 reserves four instead of two from
previous platforms (from Matt Roper)
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201007002210.3678024-3-lucas.demarchi@intel.com
Since we track the idle_pulse for flushing the barriers and avoid
re-emitting the pulse upon idling if no futher action is required, this
also impacts the heartbeat. Before emitting a fresh heartbeat, we look
at the engine idle status and assume that if the pulse was the last
request emitted along the heartbeat, the engine is idling and a
heartbeat pulse not required. This assumption fails, but we can reuse
the idle pulse as the heartbeat if we are yet to emit one, and so track
the status of that pulse for our engine health check.
This impacts tgl/rcs0 as we rely on the heartbeat for our healthcheck for
the normal preemption detection mechanism is disabled by default.
Testcase: igt/gem_exec_schedule/preempt-hang/rcs0 #tgl
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201006094653.7558-1-chris@chris-wilson.co.uk
As the previous patch fixed the places where we walk the whole scatterlist
for DMA addresses, this patch fixes the random lookup functionality.
To achieve this we have to add a second lookup iterator and add a
i915_gem_object_get_sg_dma helper, to be used analoguous to existing
i915_gem_object_get_sg_dma. Therefore two lookup caches are maintained per
object and they are flushed at the same point for simplicity. (Strictly
speaking the DMA cache should be flushed from i915_gem_gtt_finish_pages,
but today this conincides with unsetting of the pages in general.)
Partial VMA view is then fixed to use the new DMA lookup and properly
query sg length.
v2:
* Checkpatch.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Lu Baolu <baolu.lu@linux.intel.com>
Cc: Tom Murphy <murphyt7@tcd.ie>
Cc: Logan Gunthorpe <logang@deltatee.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20201006092508.1064287-2-tvrtko.ursulin@linux.intel.com
When walking DMA mapped scatterlists sg_dma_len has to be used since it
can be different (coalesced) from the backing store entry.
This also means we have to end the walk when encountering a zero length
DMA entry and cannot rely on the normal sg list end marker.
Both issues were there in theory for some time but were hidden by the fact
Intel IOMMU driver was never coalescing entries. As there are ongoing
efforts to change this we need to start handling it.
v2:
* Use unsigned int for local storing sg_dma_len. (Logan)
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
References: 85d1225ec0 ("drm/i915: Introduce & use new lightweight SGL iterators")
References: b31144c0da ("drm/i915: Micro-optimise gen6_ppgtt_insert_entries()")
Reported-by: Tom Murphy <murphyt7@tcd.ie>
Suggested-by: Tom Murphy <murphyt7@tcd.ie> # __sgt_iter
Suggested-by: Logan Gunthorpe <logang@deltatee.com> # __sgt_iter
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201006092508.1064287-1-tvrtko.ursulin@linux.intel.com
Currently we do a final scrub of the HW state upon release. However,
when rebinding the device, this is too late as the device may either
have been partially rebound or the device is no longer accessible. If
the device has been removed before release, the reset goes astray
leaving the device in an inconsistent state, unlikely to work without a
full PCI reset. Furthermore, if the device is partially rebound before
the HW scrubbing, there may be leftover HW state that should have been
scrubbed. Either way, we need to push the scrubbing earlier before the
removal, so into unregister. The danger is that on older machines,
resetting the GPU also impact the display engine and so the reset should
be after modesetting is disabled (and before reuse we need to recover
modesetting).
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/2508
Testcase: igt/core_hotunplug
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200929112639.24223-1-chris@chris-wilson.co.uk
When using fake lmem for tests, we are overriding the setting in
device info for dgfx devices. Current users of IS_DGFX() except one are
correct. However, as we add support for DG1, we are going to use it in
additional places to trigger dgfx-only code path.
In future if we need we can use HAS_LMEM() instead of IS_DGFX() in the
places that make sense to also contemplate fake lmem use.
v2: update gen8_gmch_probe() to use HAS_LMEM(): we need to steal the
mappable aperture later(which is fine since it doesn't exist on "DGFX"),
and use it as a substitute for LMEMBAR. The !mappable aperture property
is also useful since it exercises some other parts of the code too.
(Matthew Auld)
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20201001063917.3133475-1-lucas.demarchi@intel.com
Currently, we check we can send a pulse prior to disabling the
heartbeat to verify that we can change the heartbeat, but since we may
re-evaluate execution upon changing the heartbeat interval we need another
pulse afterwards to refresh execution.
v2: Tvrtko asked if we could reduce the double pulse to a single, which
opened up a discussion of how we should handle the pulse-error after
attempting to change the property, and the desire to serialise
adjustment of the property with its validating pulse, and unwind upon
failure.
Fixes: 9a40bddd47 ("drm/i915/gt: Expose heartbeat interval via sysfs")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: <stable@vger.kernel.org> # v5.7+
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Acked-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200928221510.26044-2-chris@chris-wilson.co.uk
(cherry picked from commit 3dd66a94de)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
We only allow persistent requests to remain on the GPU past the closure
of their containing context (and process) so long as they are continuously
checked for hangs or allow other requests to preempt them, as we need to
ensure forward progress of the system. If we allow persistent contexts
to remain on the system after the the hangcheck mechanism is disabled,
the system may grind to a halt. On disabling the mechanism, we sent a
pulse along the engine to remove all executing contexts from the engine
which would check for hung contexts -- but we did not prevent those
contexts from being resubmitted if they survived the final hangcheck.
Fixes: 9a40bddd47 ("drm/i915/gt: Expose heartbeat interval via sysfs")
Testcase: igt/gem_ctx_persistence/heartbeat-stop
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: <stable@vger.kernel.org> # v5.7+
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Acked-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200928221510.26044-1-chris@chris-wilson.co.uk
(cherry picked from commit 7a991cd3e3)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Chris Wilson