summaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2020-12-18intel-gpu-top: Support for client statscontext-statsTvrtko Ursulin1-11/+523
Adds support for per-client engine busyness stats i915 exports in sysfs and produces output like the below: ========================================================================== intel-gpu-top - 935/ 935 MHz; 0% RC6; 14.73 Watts; 1097 irqs/s IMC reads: 1401 MiB/s IMC writes: 4 MiB/s ENGINE BUSY MI_SEMA MI_WAIT Render/3D/0 63.73% |███████████████████ | 3% 0% Blitter/0 9.53% |██▊ | 6% 0% Video/0 39.32% |███████████▊ | 16% 0% Video/1 15.62% |████▋ | 0% 0% VideoEnhance/0 0.00% | | 0% 0% PID NAME RCS BCS VCS VECS 4084 gem_wsim |█████▌ ||█ || || | 4086 gem_wsim |█▌ || ||███ || | ========================================================================== Apart from the existing physical engine utilization it now also shows utilization per client and per engine class. v2: * Version to match removal of global enable_stats toggle. * Plus various fixes. v3: * Support brief backward jumps in client stats. v4: * Support device selection. v5: * Rebase for class aggregation. Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
2020-12-18intel_gpu_top: Fix interactive mode on serial consoleTvrtko Ursulin1-0/+5
TIOCGWINSZ returns zero columns and rows on serial so lets assume 80x24. Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
2020-12-18intel_gpu_top: Aggregate engine busyness per classTvrtko Ursulin2-13/+197
Similarly to how top(1) handles SMP, we can default to showing engines of a same class as a single bar graph entry. To achieve this a little bit of hackery is employed. PMU sampling is left as is and only at the presentation layer we create a fake set of engines, one for each class, summing and normalizing the load respectively. v2: * Fix building the aggregated engines. * Tidy static variable handling. Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
2020-12-18intel_gpu_top: Support exiting the tool by pressing 'q'Tvrtko Ursulin2-13/+73
Analoguous to top(1) we can enable the user to exit from the tool by pressing 'q' on the console. v2: * Fix sleep period with closed stdin. (Chris) Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
2020-12-18lib/igt_perf: Request CLOCK_MONOTONIC when opening eventsTvrtko Ursulin1-3/+6
Ask for CLOCK_MONOTONIC which is more stable than the default perf clock. (Ability to select a clock has been available since kernel version 4.1.) The change should not have any significant impact on the IGT as whole apart from maybe improving the occasional jitter in tests/tools which use nanosleep(2) and use time slept together with perf reported time delta either in direct or indirect calculations. Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
2020-12-18lib: Fix double lock in igt_free_spins()Chris Wilson1-17/+23
igt_free_spins() took the lock to iterate the list, igt_spin_free() took the lock to remove the list element. We only want one. Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/2823 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Acked-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
2020-12-17i915/perf_pmu: Exercise I915_PMU_SOFTWARE_GT_AWAKE_TIMEChris Wilson4-1/+56
Measure the sample gt-awake time while each engine and every engine is busy. They should all report the same duration, the elapsed runtime of the batch. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
2020-12-17i915/gem_softpin: Check the last 32b page is excludedChris Wilson1-0/+37
In order to prevent issues with 32b stateless address, the last page under 4G is excluded for non-48b objects. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: CQ Tang <cq.tang@intel.com> Reviewed-by: Matthew Auld <matthew.auld@intel.com>
2020-12-17i915/perf_pmu: Replace init/read-other with a pleaChris Wilson1-43/+15
We cannot assume we know how many PMU there are exactly, so pick -1ULL to represent all invalid metrics. Similarly, we have to rely on explicit testing for each PMU to prove their existence and correct functioning. 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>
2020-12-17i915/api_intel_bb: Only assert objects are unmoved for full-ppgttChris Wilson1-19/+22
If we let an object idle in a shared GTT, it may be evicted by the kernel in favour of another client. Thus, we have to be very careful when asserting that two different executions of the same object will be at the same address. If there's an idle point between the two asserts, it will only be guaranteed to hold for full-ppgtt. Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/2810 References: c20db20a7cd7 ("i915/api_intel_bb: Only assert objects are unmoved for full-ppgtt") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com> Acked-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
2020-12-17tests/kms_color_chamelium: fix update successKunal Joshi1-3/+9
Made a silly mistake and didn't update bool success to true, fixing that with this patch. Signed-off-by: Kunal Joshi <kunal1.joshi@intel.com> Reviewed-by: Karthik B S <karthik.b.s@intel.com>
2020-12-16kms_atomic_transition:Convert nx-modeset-transitions subtest to dynamicNidhi Gupta1-10/+27
In kms_atomic_transition subtest,it will test all the pipes defined in IGT, i.e. IGT_MAX_PIPES whether the output is available or not.Later it has to be analysed and discarded as valid skip.To save this time, updated the test to first check the available outputs and then execute the test. V7: -Modified commit message. (Karthik) -Replaced for_each_single_pipe_with_single__output with for_each_connected_output to execute the test only on connected display. (Karthik) v8: -Modified the description subject line. (Petri) -Modified subtests names to remove redundancy. (Petri) -Added extra line before for loop. (Karthik) v9: -Made separate subtests for non-blocking and fencing parameters. (Petri) Signed-off-by: Nidhi Gupta <nidhi1.gupta@intel.com> Reviewed-by: Karthik B S <karthik.b.s@intel.com> Reviewed-by: Petri Latvala <petri.latvala@intel.com>
2020-12-16i915/gem_softpin: Check full placement control under full-ppgttChris Wilson1-0/+63
With full-ppgtt, userspace has complete control over their GTT. Verify that we can place an object at the very beginning and the very end of our GTT. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Matthew Auld <matthew.auld@intel.com>
2020-12-16i915/gem_exec_params: Assert a 4G object does _not_ fit without 48bChris Wilson1-1/+7
Without opting into 48B addressing, objects are strictly limited to being placed only the first (4G - 4K). This is to avoid an issue with stateless 32b addressing being unable to access the last 32b page. Assert that we do indeed fail to fit in a 4G object without setting the EXEC_OBJECT_SUPPORTS_48B_ADDRESS flag. Reported-by: CQ Tang <cq.tang@intel.com> References:: 48ea1e32c39d ("drm/i915/gen9: Set PIN_ZONE_4G end to 4GB - 1 page") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: CQ Tang <cq.tang@intel.com> Reviewed-by: CQ Tang <cq.tang@intel.com>
2020-12-15tests/device_reset: check for i915 in unbind_reset_rebind()Jeremy Cline1-4/+6
Commit 265f1d4a5a14 ("tests/device_reset: Work around for driver unbind issue with audio") introduced an assertion on is_i915_device() by way of intel_get_drm_devid(). Since this is a work-around for an i915-specific issue, guard it with a check on the device type so the test runs on other devices. Signed-off-by: Jeremy Cline <jcline@redhat.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2020-12-15i915/perf_pmu: Verify RC6 measurements before/after suspendChris Wilson1-3/+36
RC6 should work before suspend, and continue to increment while idle after suspend. Should. v2: Include a longer sleep after suspend; it appears we are reticent to idle so soon after waking up. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Acked-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
2020-12-15i915/gem_shrink: Refactor allocation sizing based on available memoryChris Wilson1-5/+6
Refactor the allocation such that we utilise just enough memory pressure to invoke the shrinker, and just enough processes to spread across the CPUs and contend on the shrinker. v2: Reduce over-allocation from mem_size/2 to mem_size/8, and 9 processes per cpu. 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>
2020-12-15i915/gem_exec_schedule: Measure timeslice distribution when oversaturatedChris Wilson1-0/+179
Check that timeslices for an oversaturated system (where there is more work than can be supported by a single engine) are evenly distributed between the clients. 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>
2020-12-15resultgen: avoid null pointer dereference from reallocJeremy Cline1-2/+9
realloc() and friends return NULL if they fail; simplify the new_escaped_json_string() by allocating all the necessary memory up-front and checking for a failed allocation. new_escaped_json_string() can already return NULL since json_oject_new_string_len() returns NULL for various undocumented error paths, and NULL is valid input for json_object_object_add(), which this new_escaped_json_string() is currently exclusively used with. Thus, returning NULL when memory allocation fails should be safe. Signed-off-by: Jeremy Cline <jcline@redhat.com> Reviewed-by: Petri Latvala <petri.latvala@intel.com>
2020-12-15tests/i915/perf: verify reason field in OA reports is never 0Lionel Landwerlin1-0/+90
We're about to remove the filtering in i915 on 0 reason fields because we assume this was a possibility, but it turned out to be a corruption in the tail pointer register that made us read cleared data. This test is here to verify that our assumption hold that the HW never produces such reports. v2: Fix len checking (Umesh) Count report lost events (Umesh) Check report sanity (Umesh) Limit test to 3 times the OA buffer (Umesh) v3: Bump OA sampling frequency to 20us, seeing some lost buffer failures on KBL (Lionel) Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com> Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com> (v2)
2020-12-14i915/gem_exec_balancer: Tighter fairness thresholdChris Wilson1-1/+1
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
2020-12-14tools: Use igt_device_get_pci_device() for known fdChris Wilson2-3/+8
If we know the right device fd, we can find the exact matching pci device. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
2020-12-14lib/i915: Use I915_CONTEXT_PARAM_GTT_SIZEChris Wilson1-1/+1
Replace the open-coded value of 0x3 with the I915_CONTEXT_PARAM_GTT_SIZE from i915_drm.h Suggested-by: Bruce Chang <yu.bruce.chang@intel.com> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Bruce Chang <yu.bruce.chang@intel.com>
2020-12-14i915/gem_exec_balancer: Measure timeslicing fairnessChris Wilson1-0/+167
Oversaturate the virtual engines on the system and check that each workload receives a fair share of the available GPU time. v2: Apply a modicum of statistical integrity. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
2020-12-14core_setmaster: Load modules before accessing device filesPetri Latvala1-0/+9
core_setmaster does special trickery to the dri device files before it calls drm_open_device. Now that drm_load_module() exists for just loading the modules without the other things that drm_open_device() does, use it to ensure the files exist. Signed-off-by: Petri Latvala <petri.latvala@intel.com> Closes: https://gitlab.freedesktop.org/drm/igt-gpu-tools/-/issues/91 Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
2020-12-12lib: Pass device fd to gem_mmappable_aperture_size()Chris Wilson14-145/+138
In order to find the correct aperture size for the test, we want to pass the test's device into the query. Reported-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> Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
2020-12-11intel-ci: Remove blacklisting for perf_pmu/cpu-hotplugChris Wilson1-5/+0
While this regularly breaks upstream, that is also a good reason to keep testing! Let's see if upstream is in a working mood. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Acked-by: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
2020-12-11i915/gem_create: Exercise bo creation while the GPU is busyChris Wilson1-0/+41
We always expect to be able to create new buffer, regardless of the state of the GPU. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Acked-by: Matthew Auld <matthew.auld@intel.com>
2020-12-11i915/gem_exec_fence: Skip timeline tests if not supportedChris Wilson1-1/+1
Report we cannot run the timeline tests (SKIP) if the kernel doesn't support the API. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Acked-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
2020-12-11i915/gem_vm_create: Race vm-destroy against object freeChris Wilson1-0/+67
Matthew postulated that we should be able to hit a race in __i915_vm_close() between the RCU object free and vma unbind viz GEM_BUG_ON(!list_empty(&vm->bound_list)); due to the effect of leaving the vma on the list if we are unable to obtain the kref to the object. Let's try and find that race. In practice, this does not happen because to race the object free vma cleanup against vm close requires a leak of a ppGTT vma. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Matthew Auld <matthew.auld@intel.com> Acked-by: Matthew Auld <matthew.auld@intel.com>
2020-12-11i915/gem_exec_fence: Check a submit chainChris Wilson3-8/+60
Submit a chain of spinners across all the engines, using the submit fence to launch them in parallel. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Acked-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
2020-12-10i915/gem_mmap_gtt: Replace gem_threaded_access_tiledChris Wilson5-134/+15
Concurrent access to a mmap is covered by gem_mmap_gtt/concurrent, if we add tiled access to it, we make gem_threaded_access_tiled entirely redundant. 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>
2020-12-10i915/gem_tiled_blits: Remove libdrm dependencyDominik Grzegorzek1-42/+45
Use intel_bb / intel_buf to remove libdrm dependency. Signed-off-by: Dominik Grzegorzek <dominik.grzegorzek@intel.com> Cc: Zbigniew Kempczyński <zbigniew.kempczynski@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>
2020-12-10i915/gem_unref_active_buffers.c: Remove libdrm dependencyDominik Grzegorzek2-55/+77
Signed-off-by: Dominik Grzegorzek <dominik.grzegorzek@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>
2020-12-10i915/gem_unfence_active_buffers.c: Remove librdm dependencyDominik Grzegorzek1-93/+28
Signed-off-by: Dominik Grzegorzek <dominik.grzegorzek@intel.com> Acked-by: Zbigniew Kempczyński <zbigniew.kempczynski@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>
2020-12-08i915/query: Directly check query results against GETPARAMChris Wilson1-40/+9
Simplify the cross-check by asserting that the existence of an engine in the list matches the existence of the engine as reported by GETPARAM. By using the comparison, we check both directions at once. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Petri Latvala <petri.latvala@intel.com> Reviewed-by: Petri Latvala <petri.latvala@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
2020-12-08i915/query: Cross-check engine list against execbuf interfaceChris Wilson1-1/+26
Check that every engine listed can be used in execbuf. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Andi Shyti <andi.shyti@intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Petri Latvala <petri.latvala@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Andi Shyti <andi.shyti@intel.com>
2020-12-07runner: Don't kill a test on taint if watching timeoutsJanusz Krzysztofik1-6/+20
We may still be interested in results of a test even if it has tainted the kernel. On the other hand, we need to kill the test on taint if no other means of killing it on a jam is active. If abort on both kernel taint or a timeout is requested, decrease all potential timeouts significantly while the taint is detected instead of aborting immediately. However, report the taint as the reason of the abort if a timeout decreased by the taint expires. v2: Fix missing show_kernel_task_state() lost on rebase conflict resolution (Chris - thanks!) Signed-off-by: Janusz Krzysztofik <janusz.krzysztofik@linux.intel.com> Reviewed-by: Petri Latvala <petri.latvala@intel.com>
2020-12-07tests/kms_color_chamelium: time optimizationKunal Joshi1-9/+9
Instead of going through all the delta even if we got success with one, now breaking when we pass, thus saving some time and decreasing load on chamelium for capturing the rest of the frames. Signed-off-by: Kunal Joshi <kunal1.joshi@intel.com> Reviewed-by: Uma Shankar <uma.shankar@intel.com>
2020-12-06i915/gem_ctx_exec: Check spinner requirements before forkChris Wilson1-3/+5
Do a check to see if we support a pollable spinner before forking to avoid upsetting libigt. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Acked-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
2020-12-04i915/gem_ctx_exec: Exercise execution along context while closing itChris Wilson1-0/+84
Race the execution and interrupt handlers along a context, while closing it at a random time. v2: Some comments to handwave away the knowledge of internal implementation details. 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>
2020-12-03i915/sysfs_timeslice_duration: Measure error thresholdChris Wilson2-21/+46
Use the elapsed time for the fastest timeslice to define the acceptable error threshold. This gives us a lot more leeway for the slow devices. Maybe too much leeway? Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Acked-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
2020-12-03i915/api_intel_bb: Only assert objects are unmoved for full-ppgttChris Wilson1-4/+15
If we let an object idle in a shared GTT, it may be evicted by the kernel in favour of another client. Thus, we have to be very careful when asserting that two different executions of the same object will be at the same address. If there's an idle point between the two asserts, it will only be guaranteed to hold for full-ppgtt. Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/2754 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com> Reviewed-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
2020-12-01tests/kms_plane: optimize pixel format testsJuha-Pekka Heikkila1-53/+109
On packed formats there's no need to show separate colors on separate fbs. Here augmented path with packed color handling which check colors just on one fb. This cannot be directly applied to planar yuv formats because of scaler use with those formats. On my ICL this drop test execution time from 44.91s to 21.98s v2 (vsyrjala): paint entire screen instead of lines. small refinement. Signed-off-by: Juha-Pekka Heikkila <juhapekka.heikkila@gmail.com> Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Reviewed-by: Patnana Venkata Sai <venkata.sai.patnana@intel.com>
2020-11-30i915/gem_request_retire: Switch from random blitter loads to dummyChris Wilson3-160/+34
Use the spinners to provide exactly the right amount of background busyness. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
2020-11-30test/i915/gem_exec_reloc: Restore interclient testingsChris Wilson1-0/+219
These tests were removed under a blatantly false statement to hide a regression. Denying that userspace can deadlock other clients, does not stop the violations. Introducing such deadlocks deliberately is a cause for concern. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2020-11-30tools/lsgpu: Add -n switch to list devices using vendor:device hex idZbigniew Kempczyński1-1/+7
Default device list prefers vendor and device names. Add -n switch to display vendor/device as hex strings. Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Petri Latvala <petri.latvala@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
2020-11-30tools/intel_gpu_top: Add generation info in headerZbigniew Kempczyński3-3/+10
In multi device world we may want to see generation of device we're tracking counters. Add pretty name of the device to be more verbose. Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Petri Latvala <petri.latvala@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
2020-11-30lib/igt_device_scan: Remember vendor/device for pci devicesZbigniew Kempczyński2-13/+142
If we want to use pci device id for not opened device we need to keep it in card structure. Export igt_device_get_pretty_name() function to the caller to allow return pretty name (for lsgpu, intel_gpu_top and others). Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Petri Latvala <petri.latvala@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
2020-11-28i915/gem_exec_capture: Trim the blocking workloadChris Wilson1-1/+11
While we want the capture to last long enough to delay the concurrent client, we don't want to wait forever for the capture to complete to proceed with the testing. Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/2559 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Acked-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-03 18:12:30 +0000