summaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2018-01-18lib/core: report subtests that hit an igt_warning as WARNINGstuffDaniel Vetter1-1/+6
This is another piece of prep work to push the detection of dmesg warnings into igt itself, so that we can correctly report dmesg issue on a per-subtest basis even when running the entire binary. Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2018-01-18igt/core: Initial simple interleaved kmsg filteringDaniel Vetter1-1/+40
Needs to be beefed up so that dmesg warning (and worse) are re-emmitted as IGT_LOG_WARN. But only if they match one of our filters (which we should probably allow to be extended, e.g. depending upon which driver has been openened). This also requires that we at least parse the basic of kmsg lines (adjusting the timestamp to match our own would be real cool). v2: - Seek to the end of the kmsg buffer before starting the capturing. - Increase linebuffer to avoid dmesg drowning out all the tests messages. Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2018-01-18lib/core: Don't hide non-debug message when filtering for a debug log domainDaniel Vetter1-5/+5
I think this is the more sensible semantics, since this allows you to still follow what's going on with the test at a high level, while filtering for a specific (or multiple specific) debug log domains. For non-debug messages log-domains technically exist, but we're not making much use of them really. Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2018-01-18lib/core: make logging pthread vs. fork safeDaniel Vetter1-8/+17
fork() is a pretty thing in a multithreaded program, it's essentially as bad as handling signals: When we fork the memory snapshot we can interrupts all other threads at any place, including while they're holding mutexes and other fun stuff. libc itself has some internal fork handlers to clear caches and make sure locks stay in a safe place (we've had plenty of fun with e.g. the pid/tid caches when a signal happens too fast). I want to put dmesg capture into igt, into a separate thread (so that dmesg capture nicely interleaves what the test is doing, +/- races), and stuff all the dmesg output into the igt logger. Which means we need to make sure that the log_buffer_mutex is in a predictable state. Since we have 2 calls to fork() extract a little helper for this. v2: Stop using fflush(NULL) - somehow this manages to hit a bug in libc when using a FILE in a separate thread (for capturing dmesg). Instead explicitly flush stderr and stdout only. Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
2018-01-18tests/sw_sync: use igt_fork_helperDaniel Vetter1-36/+18
I'll need to wrap a bit of magic around all the fork() calls in our tests. Simplest way to get there is to roll out the existing helpers, which even saves a bit of boilerplate code. Cc: Robert Foss <robert.foss@collabora.com> Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
2018-01-18igt/pm_rps: Increase load for waitboostingChris Wilson1-0/+15
The waitboost subtests encode internal knowledge of the kernel, and should we wish to change how the kernel functions, we also need to update the test (or reconfigure it somehow to express desired latencies for certain workloads). One such change proposed is to remove the waitboost if the target request is also completed, so update the test to not wait on the current batch. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
2018-01-17igt/gem_tiled_fence_blits: Allocate bo arrayChris Wilson1-3/+10
As we allow more buffers to be allocated to fill larger apertures, we may exceed the static allocation of 4096 buffers. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=104669 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
2018-01-17meson: Refactor get_option() calls for directoriesPetri Latvala8-25/+25
Fetch the configuration values in the toplevel meson.build for all subdirs to share. v2: Also remember tests/intel-ci/meson.build Signed-off-by: Petri Latvala <petri.latvala@intel.com> Cc: Daniel Vetter <daniel.vetter@ffwll.ch> Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2018-01-17CONTRIBUTING: Fix spelling mistake and line lengthSean Paul1-4/+4
Noticed while I was reading it. Makes for a good first contribution, I guess. Changes in v2: - None Changes in v3: - None Reviewed-by: Petri Latvala <petri.latvala@intel.com> Signed-off-by: Sean Paul <seanpaul@chromium.org>
2018-01-17tests/kms_plane_scaling: test for multi pipe with scaling, v3.Jyoti Yadav1-1/+92
Add a subtest to display primary and overlay planes on two connected pipes and runs scaling test on both pipes Changes since v1: - Commit first before trying any scaling. (Maarten) - Use the same logic as kms_cursor_legacy to find a pipe and output. (Maarten) - Rework test to work with how cleanup is handled. (Maarten) Changes since v2: - Use get_num_scalers() to only use 2 scalers if available. (Maarten) Signed-off-by: Jyoti Yadav <jyoti.r.yadav@intel.com> Signed-off-by: Mahesh Kumar <mahesh1.kumar@intel.com> Signed-off-by: Vidya Srinivas <vidya.srinivas@intel.com> Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Reviewed-by: Mika Kahola <mika.kahola@intel.com>
2018-01-17tests/kms_plane_scaling: test scaler with clipping clamping, v3.Jyoti Yadav1-0/+43
This patch adds subtest to test scaler clipping and clamping scenario. Changes since v1: - Modify test to work with the changes to kms_plane_scaling. (Maarten) Changes since v2: - Use get_num_scalers() to skip when needed. Signed-off-by: Jyoti Yadav <jyoti.r.yadav@intel.com> Signed-off-by: Mahesh Kumar <mahesh1.kumar@intel.com> Signed-off-by: Vidya Srinivas <vidya.srinivas@intel.com> Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Reviewed-by: Mika Kahola <mika.kahola@intel.com>
2018-01-17tests/kms_plane_scaling: test scaling with tiling rotation and pixel ↵Jyoti Yadav1-1/+129
formats, v3. This patch adds subtest for testing scaling in combination with rotation and pixel formats. Changes since v1: - Rework test to work with the other changes to kms_plane_scaling. (Maarten) - Remove hardcodes for MIN/MAX_SRC_WIDTH, and use the value directly. (Maarten) Changes since v2: - Put rotation and tiling in an array. (Maarten) Signed-off-by: Jyoti Yadav <jyoti.r.yadav@intel.com> Signed-off-by: Mahesh Kumar <mahesh1.kumar@intel.com> Signed-off-by: Vidya Srinivas <vidya.srinivas@intel.com> Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Reviewed-by: Mika Kahola <mika.kahola@intel.com>
2018-01-17tests/kms_plane_scaling: Clean up tests to work better with igt_kms, v2.Maarten Lankhorst1-136/+99
The test only runs on gen9+, so we can safely replace all calls with COMMIT_ATOMIC. Also perform some cleanups by making fb an array, and cleaning up in prepare_crtc. This way failed subtests won't cause failures in other subtests. Changes since v1: - Rebase on top of num_scalers changes. Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Reviewed-by: Mika Kahola <mika.kahola@intel.com>
2018-01-17tests/kms_plane_scaling: Move get_num_scalers to a function, v2.Maarten Lankhorst1-6/+16
The number of scalers can depend on the pipe, so require at least 1 scaler before running any subtests. Changes since v1: - More closely match kernel implementation. (Mika) Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Reviewed-by: Mika Kahola <mika.kahola@intel.com>
2018-01-17tests/kms_plane_scaling: Convert from simple test to full testMaarten Lankhorst1-25/+21
Convert the test to run subtests per pipe, before we start adding more subtests. Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Reviewed-by: Mika Kahola <mika.kahola@intel.com>
2018-01-17tests/kms_plane_scaling: Fix basic scaling test, v3.Mahesh Kumar1-9/+16
PIPEC doesnt have 3rd plane in GEN9. So, we skip the 3rd plane related scaling test where 2nd OVERLAY plane is not available. Restricting downscaling to (9/10)x original size of the image to avoid "Max pixel rate limitation" of the hardware. When testing with a resolution of 1024x768, the test fails in iterate_plane_scaling because height becomes out of bounds. Lastly, instead of a fixed test image, generate a test pattern at runtime to not run into bxt's hardware limits for scaling. Later patches in this series will cover corner cases of scaling. Changes since v1: - Move out the code reshuffle to a separate commit. (Maarten) This makes it more clear what's fixed. Changes since v2: - Fix iterate_plane_scaling on 1024x768. (Maarten) - Instead of upscaling a test image, use the test pattern with a color. (Maarten) Using a 600x600 test pattern doesn't run into BXT's hardware limits for scaling. Signed-off-by: Mahesh Kumar <mahesh1.kumar@intel.com> Signed-off-by: Jyoti Yadav <jyoti.r.yadav@intel.com> Signed-off-by: Vidya Srinivas <vidya.srinivas@intel.com> Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Reviewed-by: Mika Kahola <mika.kahola@intel.com>
2018-01-17tests/kms_plane_scaling: Move the actual test to its own function.Maarten Lankhorst1-109/+117
We will add more subtests in the future, it's more clear if we split out the actual test to its own function first. Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Reviewed-by: Mika Kahola <mika.kahola@intel.com>
2018-01-16igt/gem_linear_blits: Compute GTT size using 4G limitChris Wilson3-10/+35
Both gem_linear_blits and gem_tiled_blit do not request the full 48b GTT layout for their objects, restricting themselves to 4G. The underlying test that they trigger eviction is unaffected by this restriction, so we can simply reduce their memory requirements to fill the low 4G GTT space and so allow them to run on 48b machines. v2: gem_tiled_fenced_blits as well Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
2018-01-16Update NEWS, bump version to 1.21.Petri Latvala3-5/+58
Signed-off-by: Petri Latvala <petri.latvala@intel.com> Acked-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
2018-01-16meson: Name the project intel-gpu-toolsPetri Latvala1-1/+1
Eventually we're switching the official name to "IGT GPU Tools", but right now there's still a lot of hardcoding to intel-gpu-tools that is to be fixed in the near future. Rename the project in toplevel meson.build to intel-gpu-tools to get meson to generate tarballs roughly the same as autotools in 'dist'. Signed-off-by: Petri Latvala <petri.latvala@intel.com> Reviewed-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
2018-01-16meson: Add quotes in assembler/test/run-test.shPetri Latvala1-6/+6
If the directories contain spaces, run-test.sh fails. Signed-off-by: Petri Latvala <petri.latvala@intel.com> Reviewed-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
2018-01-16meson: Add quotes in man/rst2man.shPetri Latvala1-6/+6
If the directories contain spaces, rst2man.sh fails. Signed-off-by: Petri Latvala <petri.latvala@intel.com> Reviewed-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
2018-01-16meson: Build cnl_compute_wrpllPetri Latvala1-0/+1
Signed-off-by: Petri Latvala <petri.latvala@intel.com> Reviewed-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
2018-01-15tests/kms_plane: Run test for all supported pixel formats, v2.Mahesh Kumar1-0/+103
This patch adds a subtest related to pixel format testing. The test tries to create framebuffer with all supported pixel formats on every plane, and tries to draw them using cairo and commits the same on display. Changes since v1: - Make the test more generic and try on all planes, including legacy cursor. Signed-off-by: Mahesh Kumar <mahesh1.kumar@intel.com> Signed-off-by: Jyoti Yadav <jyoti.r.yadav@intel.com> Signed-off-by: Vidya Srinivas <vidya.srinivas@intel.com> Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Reviewed-by: Mika Kahola <mika.kahola@intel.com>
2018-01-15lib/igt_kms: Add more braces around macrosMaarten Lankhorst1-8/+8
The next patch wants to call for_each_pipe_with_valid_output with *pipe and *output, this fails miserably without these braces. Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Reviewed-by: Mika Kahola <mika.kahola@intel.com>
2018-01-15tests/perf_pmu: Exercise busy stats and lite-restoreTvrtko Ursulin1-0/+72
While developing a fix for an accounting hole in busy stats we realized lite-restore is a potential edge case which would be interesting to check is properly handled. It is unfortnately quite timing sensitive to hit lite-restore in the fashion test needs, so downside of this test is that it sufferes from a high rate of false negatives. v2: * Make the sleep unconditional and use scientific notiation for large constants. (Chris Wilson) * Use gem_quiscent_gpu instead of gem_sync+usleep to ensure context complete was received under execlists. (Chris Wilson) Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
2018-01-15tests/perf_pmu: Verify busyness when PMU is enabled after engine got busyTvrtko Ursulin1-0/+41
Make sure busyness is correctly reported when PMU is enabled after the engine is already busy with a single long batch. v2: * Make the sleep unconditional and use scientific notiation for large constants. (Chris Wilson) Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
2018-01-12tools: Update .gitignorePetri Latvala1-0/+1
Fixes: 834321a5d76a ("tools: Cannonlake port clock programming") Signed-off-by: Petri Latvala <petri.latvala@intel.com> Cc: Mika Kahola <mika.kahola@intel.com> Reviewed-by: Mika Kahola <mika.kahola@intel.com>
2018-01-11Revert "build: make meson more official" damageDaniel Vetter1-3/+3
This reverts the meson.build changes from commit 07c331773dd3bc4dadb164bcd9bc06dbd01de3b6. Reviewed-by: Petri Latvala <petri.latvala@intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2018-01-11run-tests.sh: Allow users to override IGT_TEST_ROOTPetri Latvala1-1/+1
Signed-off-by: Petri Latvala <petri.latvala@intel.com> Cc: Arkadiusz Hiler <arkadiusz.hiler@intel.com> Reviewed-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
2018-01-11build: make meson more officialDaniel Vetter5-9/+24
I also noticed that meson.sh doesn't set the prefix for patch submission. Fix that (even thought hopefully real soon igt will move to its own list). v2: Review from Petri. Cc: Petri Latvala <petri.latvala@intel.com> Cc: Arkadiusz Hiler <arkadiusz.hiler@intel.com> Reviewed-by: Petri Latvala <petri.latvala@intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
2018-01-11overlay: Update .gitignoreRhys Kidd1-0/+1
Fixes: 865a47ca ("overlay: parse tracepoints from sysfs to figure out fields' location") Signed-off-by: Rhys Kidd <rhyskidd@gmail.com> Reviewed-by: Petri Latvala <petri.latvala@intel.com>
2018-01-11include inttypes.h for PRI definesMike Frysinger9-0/+9
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=96620 Signed-off-by: Mike Frysinger <vapier@gentoo.org> Reviewed-by: Petri Latvala <petri.latvala@intel.com>
2018-01-10lib/igt_kms.c: Unconditionally include poll.hPetri Latvala1-1/+1
Commit 98c64b33a793 ("lib/igt_kms: Drop all stale events on first commit.") added a use of poll() to igt_kms.c, but that file only includes poll.h when HAVE_UDEV is defined. Move the include outside the UDEV conditional. Fixes: 98c64b33a793 ("lib/igt_kms: Drop all stale events on first commit.") Signed-off-by: Petri Latvala <petri.latvala@intel.com> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Rhys Kidd <rhyskidd@gmail.com>
2018-01-08tools: Cannonlake port clock programmingMika Kahola2-0/+527
Cannonlake port clock programming tests and verifies DPLL legal dividers P, Q, and K. This tests adds two reference clocks 19.2MHz and 24MHz to test algorithm's capability to find P, Q, and K dividers as well as DCO frequency for different symbol clock rates. The test compares two algorithms, the reference with double precision and i915 implementation with fixed point precision. In case of a difference in computation the difference on dividers is printed out to the screen. Signed-off-by: Mika Kahola <mika.kahola@intel.com> Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
2018-01-08scripts/trace.pl: Optimize event parsing and processingTvrtko Ursulin1-65/+52
A couple of small optimizations which altogether bring around 30% improvement in my testing. 1. Do less string processing on tracepoints names and push more of the check into the if-ladder. 2. Pull out common db key and ctx processing and cache common values in local vars. 3. Key value pair parsing is faster with a regexp. 4. Avoid sorting the db hash multiple times if possible. v2: * Use faster key-value splitting method. (John Harrison) v3: * Fix floating-point to int time conversion. v4: * Fix refactoring and logic fails. (John Harrison) Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: John Harrison <John.C.Harrison@intel.com> Reviewed-by: John Harrison <John.C.Harrison@Intel.com>
2018-01-08test/kms_psr_sink_crc - subtests psr_basic and psr_drrs need test cleanupMarta Lofstedt1-0/+2
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=104260 Signed-off-by: Marta Lofstedt <marta.lofstedt@intel.com> Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
2018-01-08meson: use message() rather than warning()Lucas De Marchi1-1/+1
warning() was only added to the meson interpreter in 0.44 which is currently the last version. Let's use message() as we are currently requiring meson > 0.40. Otherwise we get the following error: Meson encountered an error in file overlay/meson.build, line 62, column 1: Unknown function "warning". Fixes: 865a47ca ("overlay: parse tracepoints from sysfs to figure out fields' location") Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Cc: Rhys Kidd <rhyskidd@gmail.com> Cc: Daniel Vetter <daniel.vetter@ffwll.ch> Cc: Petri Latvala <petri.latvala@intel.com> Reviewed-by: Rhys Kidd <rhyskidd@gmail.com>
2018-01-04lib/gem: Reset the global seqno at the start of each testChris Wilson1-1/+15
When we require GEM, reset the global seqno. This gives each test a clean slate to work with, and avoids left-over state from previous tests impacting on the next. In particular, somes tests may be setting up long sequence of stalling batches not expecting to hit a seqno wraparound (leftover from, for example, gem_exec_whisper), causing long GPU hangs and incompletes in CI if they do. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
2018-01-04igt/perf_pmu: Skip GEM checks for repeated spin_batch allocationsChris Wilson1-1/+2
Each call to igt_spin_batch_new_fence will do a stalling check to verify that GEM is functional before submitting the spinning batch. In a loop, this means that we may end up waiting for our earlier spinning batches... Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
2018-01-04igt/gem_busy: Remove repeated use of igt_spin_batch_newChris Wilson1-5/+5
igt_spin_batch_new() includes a throttling check that GEM works, which breaks trying to create multiple spin batches, use __igt_spin_batch_new() instead, after verifying GEM works. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
2018-01-04igt/kms_flip: Do igt_require_gem() just onceChris Wilson1-4/+7
Since igt_spin_batch_new() will do a stalling GEM check, it is not advisable to use it within loops. Perform the igt_require_gem() upfront and then use __igt_spin_batch_new() inside the test loop. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
2018-01-03kms_atomic_transition: Remove skip_on_unsupported_nonblocking_modeset.Maarten Lankhorst1-47/+0
This weas a workaround to gracefully skip on i915 before nonblocking modeset support was added, but this is no longer needed. Remove the code, since it's safe to always assume such support is enabled. Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=104471 Reported-by: Marta Lofstedt <marta.lofstedt@intel.com> Reviewed-by: Marta Lofstedt <marta.lofstedt@intel.com>
2018-01-02tests/kms_cursor_legacy: Rework the 2x-*-vs-cursor-* tests.Maarten Lankhorst1-72/+156
Using the fancy new DRM_CAP_CRTC_IN_VBLANK_EVENT cap I can finally make this test the work I originally intended to. For the !modeset case that means performing a pageflip on both crtc's, then requeueing as soon as the event is delivered and then check the vblank counter against the original value, it should be advanced by 1. The modeset case is slightly more complicated, ideally it's handled the same, but if we can't perform a modeset and pageflip at the same time, fall back to queueing both in a single commit, in which case we can say nothing about the vblank counter. There is a small race between flip_done and hw_done, so make flip_nonblocking retry for a second when encountering -EBUSY. Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=101634 Reviewed-by: Mika Kahola <mika.kahola@intel.com>
2018-01-02tests/kms_cursor_legacy: Perform lazy cleanup between testsMaarten Lankhorst1-76/+12
Instead of assuming each subtest cleans up after itself, assume it fails and doesn't. Now that igt_kms can clean up stale events, we can just force each subtest to only clean up its framebuffers, which isn't harmful if it failed. The nonblocking modeset test is removed, this was a workaround only needed for intel because it supported nonblocking commit only for the !modeset case, this has been fixed. Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Reviewed-by: Mika Kahola <mika.kahola@intel.com>
2018-01-02lib/igt_kms: Drop all stale events on first commit.Maarten Lankhorst2-1/+46
I've been trying to make kms_cursor_legacy work when subtests fail. Other subtests will start failing too because of expired events or stale pipe crc. The latter can be resolved in the test, but the former could affect other tests Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> [mlankhorst: Change return status to int, so callers can see how many events are swallowed.] Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
2017-12-27lib: Convert sw_sync to use sync_file uapi imported from the kernelChris Wilson3-46/+112
Similar to how we are now importing the drm uapi directly into igt, we also would like to have a copy of auxiliary uAPI such as sync_file. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
2017-12-22igt/gem_exec_await: Flush the WCB before attempting to queue more workChris Wilson1-0/+1
Ensure that the terminating write into WC-memory is flushed before we might trigger a wait for ring space. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
2017-12-22tests/perf_pmu: Simplify interrupt testingTvrtko Ursulin1-72/+69
Rather than calibrate and emit nop batches, use a manually signalled chain of spinners to generate the desired interrupts. v2: Two flavours of interrupt generation. (Chris Wilson) Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
2017-12-22lib/dummyload: Support returning output fenceTvrtko Ursulin2-8/+82
Support creating spin batches which return an output fence using new __igt_spin_batch_new_fence / igt_spin_batch_new_fence API. This will be used fromthe perf_pmu@interrupts test to ensure user interrupt generation from a batch with controlled duration. v2: Support out fence with multiple engines as well. (Chris Wilson) Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> (v1) Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-03 17:38:15 +0000