Chris Wilson [Fri, 21 Feb 2014 15:14:08 +0000 (15:14 +0000)]
kms_flip: Skip tests that try to inject GPU hangs if already terminally wedged
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Fri, 21 Feb 2014 15:08:22 +0000 (15:08 +0000)]
kms_flip: Assert that hang_gpu() should only fail is the GPU is already hung
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Fri, 21 Feb 2014 14:42:27 +0000 (14:42 +0000)]
tests: Compile all threaded tests with -pthread when appropriate
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Ville Syrjälä [Fri, 21 Feb 2014 14:08:28 +0000 (16:08 +0200)]
kms_flip: Fail the subtest if page flip hang recovery wasn't actually tested
Context banning can prevent the page flip hang tests from actaully
testing anything, so make the relevant subtests fail in that case.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Ville Syrjälä [Fri, 21 Feb 2014 13:14:33 +0000 (15:14 +0200)]
kms_flip: Restore rings to running state in unhang_gpu()
If things go bad, make sure the rings aren't left in the stopped state.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Paulo Zanoni [Thu, 20 Feb 2014 20:47:53 +0000 (17:47 -0300)]
tests/pm_pc8: try to modprobe msr
Just in case it's compile with M instead of Y. If the module is not
there, the other assertions will catch the problem.
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Paulo Zanoni [Thu, 20 Feb 2014 20:45:53 +0000 (17:45 -0300)]
tests/pm_pc8: try to modprobe i2c-dev
Just in case the module is compiled with M instead of Y. If the module
is not there, the other assertions will catch the problem.
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Paulo Zanoni [Tue, 18 Feb 2014 17:27:19 +0000 (14:27 -0300)]
tests/pm_pc8: add gem-idle subtest
It's triggering WARNs and DRM_ERRORs on current drm-intel-nighly.
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Paulo Zanoni [Mon, 17 Feb 2014 19:29:36 +0000 (16:29 -0300)]
tests/pm_pc8: remove sleep() call when setting up the environment
This sleep was added because sometimes we didn't reach PC8+
residencies, but it was still not enough to prevent the problem every
time, and it is really not needed most of the times. I have
investigated more and it seems that we only have to wait until after
some minutes have past since the machine booted. So just remove the
sleep for now since when you run each subtest in a separate process,
you end up having to sleep at every subtest.
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Paulo Zanoni [Fri, 21 Feb 2014 13:05:54 +0000 (10:05 -0300)]
tests/gen7_forcewake_mt: use -lpthread
Otherwise it fails to compile on my machine.
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Chris Wilson [Fri, 21 Feb 2014 09:38:43 +0000 (09:38 +0000)]
kms_flip: Try to make hang_gpu() robust against hanging the GPU
On a bad day, hanging the GPU may be terminal. Yet even if the GPU is
terminally wedged we expect modesetting (and pageflips) to continue.
That deserves to be a dedicated test, but in the meantime we should
strive to avoid falling over just because the code is not resilient.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Thu, 20 Feb 2014 13:41:41 +0000 (13:41 +0000)]
Add i915_pciids.h
This is a pure copy from the central location at
kernel/include/drm/i915_pciids.h
Chris Wilson [Thu, 20 Feb 2014 13:27:05 +0000 (13:27 +0000)]
tests: Add gen7_forcewake_mt
This is supposed to be snafu...
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Joao Santos [Wed, 19 Feb 2014 14:19:19 +0000 (14:19 +0000)]
build: Skip kms_plane on Android
Skipping this test until we have Cairo in the build.
Signed-off-by: Joao Santos <joao.santos@intel.com>
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Tue, 18 Feb 2014 14:17:10 +0000 (14:17 +0000)]
kms_pipe_crc_basic: Remove spurious inclusion of glib.h
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Tue, 18 Feb 2014 14:16:49 +0000 (14:16 +0000)]
kms_plane: Remove spurious inclusion of glib.h
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Daniel Vetter [Thu, 13 Feb 2014 18:22:53 +0000 (19:22 +0100)]
test/kms_flip: fix up igt logging conversion
Silly me ...
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Thu, 13 Feb 2014 18:22:35 +0000 (19:22 +0100)]
testdisplay: Allow to specify the dpms level
aka the jbarnes-me-harder mode
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Thu, 13 Feb 2014 18:15:38 +0000 (19:15 +0100)]
testdisplay: implement dpms test
For those who wonder about the -j switch ... all the good ones have
been taken, so this is the Jesse-flag.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Thu, 13 Feb 2014 13:43:20 +0000 (14:43 +0100)]
NEWS: Mention new logging support
Daniel Vetter [Thu, 13 Feb 2014 13:25:55 +0000 (14:25 +0100)]
tests/kms_flip: use igt logging
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Thu, 13 Feb 2014 13:12:03 +0000 (14:12 +0100)]
tests/gem_seqno_wrap: use igt logging
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Thu, 13 Feb 2014 13:09:41 +0000 (14:09 +0100)]
test/pm_rps: use igt logging
Cc: Jeff McGee <jeff.mcgee@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Wed, 12 Feb 2014 14:19:15 +0000 (15:19 +0100)]
lib: (somewhat) structured logging support
Apparently there's a bit a need for more verbose output in testcases,
mostly for debugging purposes. At least gem_reset_stats and pm_rps
have a verbose mode.
On top of that we're currently not taking advantage of piglit's "warn"
state all that much. But I think it might be useful for testcases which
are notorious for some kinds of spurious failures, like e.g. the really
nasty timing checks in kms_flip. If we demote some of them to just
warnings we could run the overall tests more often.
Hence this patchs adds a new igt_log function with the three levels DEBUG,
INFO and WARN. Plus a bunch of convenience helpers to keep the test
code tidy.
The level can be set through an enviroment vairable IGT_LOG_LEVEL with
info being the default. Also tests can look at the selected log level in
case they want to run costly debug functions only when needed.
Comments highly welcome, I plan to roll this out over tests which can
use it (not all, imo that's too much churn) once we've settled on the
interfaces/semantics.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Zhao Yakui [Tue, 28 Jan 2014 01:53:50 +0000 (09:53 +0800)]
Assembler/bdw: Remove the unsupported cache agent for WRITE(...)
The Sampler/Constant cache is read-only. And it can't be used as
the target cache agent of WRITE message.
Reviewed-by: Xiang, Haihao <haihao.xiang@intel.com>
Signed-off-by: Zhao Yakui <yakui.zhao@intel.com>
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Thomas Wood [Wed, 12 Feb 2014 15:54:27 +0000 (15:54 +0000)]
tests: fix sub-test enumeration
Ensure sub-tests can be listed correctly by doing any test setup within
an igt_fixture block.
Signed-off-by: Thomas Wood <thomas.wood@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Tue, 11 Feb 2014 23:12:07 +0000 (00:12 +0100)]
NEWS: Mention Damien's new helper library
Daniel Vetter [Tue, 11 Feb 2014 22:54:19 +0000 (23:54 +0100)]
lib: Make igt_skip noreturn
This is a remnant from the subtest code before we had proper
fixture blocks. With those there's no no reason to have igt_skip
calls outside of fixtures or subtests, so we can assert this and
hence always jump out.
Suggested by Thomas Wood since static analyzers got confused about
this.
To check for fallout I've run all tests as non-root so that they'll
all skip. Only very little fallout resulted.
Cc: Thomas Wood <thomas.wood@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Tue, 11 Feb 2014 23:07:11 +0000 (00:07 +0100)]
lib: install exit handler only on success for prefault control
Otherwise we'll hit an igt_skip in the exit handler, which upsets
the new in_fixture||in_subtests checks.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Tue, 11 Feb 2014 23:05:57 +0000 (00:05 +0100)]
test/gem_gtt_hog: bail out earlier
This way we properly skip instead of failing hard. Just result
polish when running on non-intel systems.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Tue, 11 Feb 2014 22:52:43 +0000 (23:52 +0100)]
tests/gem_madvise: use correct main block
igt_simple_main doesn't enumerate subtests ...
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Tue, 11 Feb 2014 22:50:10 +0000 (23:50 +0100)]
tests/gem_bad_ctx_exec: Properly wrap in igt_fixtures
Caught by the new nasty assert in igt_skip.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Tue, 11 Feb 2014 22:47:50 +0000 (23:47 +0100)]
tests/gem_close_race: Properly wrap init code in fixtures
Blows up otherwise if there's no intel gpu around.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Tue, 11 Feb 2014 22:36:24 +0000 (23:36 +0100)]
lib: fix header include
drm/ path is for the headers from kernel sources, without when using
the pkgconfig libdrm version.
Cc: Damien Lespiau <damien.lespiau@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Damien Lespiau [Tue, 11 Feb 2014 17:53:00 +0000 (17:53 +0000)]
lib/display: Fix typo in igt_plane_get_fd_id() (fb)
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Tue, 11 Feb 2014 17:48:32 +0000 (17:48 +0000)]
kms_pipe_crc_basic: Remove igt_display_get_n_pipes() usage in main
The fixture won't be run when listing the subtests and thus we can't use
igt_display_get_n_pipes() in the list of tests. The alternative here is
to list the subtests with the maximum values and skip when not supported
on the platform.
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Tue, 11 Feb 2014 17:45:48 +0000 (17:45 +0000)]
lib: Fix igt_ouput_get_plane() typo (output)
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Tue, 11 Feb 2014 13:54:10 +0000 (13:54 +0000)]
lib: Don't wait in igt_wait_for_keypress() if stdin is not a tty
When running in a non interactive terminal, there's little point in
trying to wait for some input on stdin.
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Thu, 6 Feb 2014 21:20:35 +0000 (21:20 +0000)]
kms_plane: Start a basic display plane test
We test the sprite plane positionning in there, for now.
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Fri, 7 Feb 2014 14:12:44 +0000 (14:12 +0000)]
lib/display: Fix assertion in set_plane()
When cycling throuth planes, we still want to reach the cursor plane. We
have to special case IGT_PLANE_CURSOR as a shorthand to select the
cursor plane (the last plane on the pipe).
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Thu, 6 Feb 2014 21:18:28 +0000 (21:18 +0000)]
lib/display: Wait for a vblank after SetPlane()
Let's be testing friendly and gently wait for the next vblank before
returning from commit() when needed.
After igt_display_commit() one can safely look at the CRC.
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Thu, 6 Feb 2014 21:06:59 +0000 (21:06 +0000)]
lib/display: Print the fb id, not its pointer in the set_fb() log message
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Thu, 6 Feb 2014 21:06:25 +0000 (21:06 +0000)]
lib/display: Add a way to wait at every commit for inspection
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Thu, 6 Feb 2014 21:06:07 +0000 (21:06 +0000)]
lib: Add a helper to wait for a keypress
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Thu, 6 Feb 2014 19:10:51 +0000 (19:10 +0000)]
kms_pipe_crc_basic: Use kmstest_create_color_fb()
Reducing again the per-test number of lines.
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Thu, 6 Feb 2014 19:09:53 +0000 (19:09 +0000)]
lib: Include drm_fourcc.h from igt_kms.h
This include is needed for the DRM_FORMAT* defines used in the fb
creation helpers.
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Thu, 6 Feb 2014 19:04:58 +0000 (19:04 +0000)]
kms_cursor_crc: Use kmstest_create_color_fb()
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Thu, 6 Feb 2014 19:04:12 +0000 (19:04 +0000)]
lib: Introduce a new helper kmstest_create_color_fb()
We need to create fbs of a single color in a few places. Time to
abstract that out to a helper function.
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Thu, 6 Feb 2014 17:54:33 +0000 (17:54 +0000)]
lib/display: Add support for DRM planes
We can now extend our plane support beyond primary and cursor planes.
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Thu, 6 Feb 2014 16:26:31 +0000 (16:26 +0000)]
tests/kms_cursor_crc: Use igt_pipe_crc_collect_crc()
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Thu, 6 Feb 2014 16:17:38 +0000 (16:17 +0000)]
lib/crc: Add a helper to read a single CRC value
In this case, we also take care of starting/stopping the CRC collection.
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Thu, 6 Feb 2014 16:17:14 +0000 (16:17 +0000)]
lib/crc: Factor out reading a single CRC value
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Thu, 6 Feb 2014 16:05:19 +0000 (16:05 +0000)]
lib/display: Check if we're trying to use the same pipe on 2 outputs
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Wed, 5 Feb 2014 17:51:50 +0000 (17:51 +0000)]
lib/display: Allow to override the display verbosity with an env variale
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Wed, 5 Feb 2014 17:14:12 +0000 (17:14 +0000)]
kms_cursor_crc: Port the test to the new modeset API
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Thu, 6 Feb 2014 15:45:42 +0000 (15:45 +0000)]
lib/display: Add a way to specify we don't care about the pipe to use
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Wed, 5 Feb 2014 16:36:51 +0000 (16:36 +0000)]
lib/display: Add support for the cursor plane
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Wed, 5 Feb 2014 15:54:13 +0000 (15:54 +0000)]
lib/display: Fix the SetCrtc disabling log message
We were displaying the value of fb_id (0), when the actual interesting
thing about this call it that it'll just down the pipe.
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Wed, 5 Feb 2014 15:32:15 +0000 (15:32 +0000)]
tests/kms_cursor_crc: Move the array of cursor fb_ids to the stack
This array is not used outside ouf create_cursor_fb(). A stack allocated
array is enough.
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Tue, 4 Feb 2014 15:11:29 +0000 (15:11 +0000)]
kms_pipe_crc_basic: Use igt_display_get_n_pipes() instead of hardcoding 3
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Tue, 4 Feb 2014 14:59:39 +0000 (14:59 +0000)]
lib/display: Add an accessor to retrieve the number of pipes
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Tue, 4 Feb 2014 12:07:56 +0000 (12:07 +0000)]
kms_pipe_crc_basic: Use for_each_connected_output()
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Tue, 4 Feb 2014 12:06:57 +0000 (12:06 +0000)]
lib: Introduce a for_each_connected_output() macro
So we can easily cycle through them in tests without knowing too many
internal details about how igt_display_t organize its data.
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Sun, 2 Feb 2014 17:48:24 +0000 (17:48 +0000)]
lib: Introduce symbolic names for display planes
It'd be nice to have symbolic names for planes instead of using an index
in igt_output_get_plane().
We also namespace the enum to not conflict with anyone.
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Mon, 27 Jan 2014 16:32:09 +0000 (16:32 +0000)]
kms_pipe_crc_basic: Port to the new modeset API
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Mon, 27 Jan 2014 16:25:43 +0000 (16:25 +0000)]
lib: Introduce a modeset API
The goals here are:
- Reduce duplicated code in each KMS test
- Provide an API that looks more like what we want for atomic
modesets. The hope is then that it'll be easy to switch, at
run-time, between the "legacy" path and atomic modesets, keeping
the same API for tests.
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Thomas Wood [Fri, 7 Feb 2014 17:03:39 +0000 (17:03 +0000)]
tests/gem_seqno_wrap: fix over allocation of arrays
Signed-off-by: Thomas Wood <thomas.wood@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Thomas Wood [Fri, 7 Feb 2014 17:03:38 +0000 (17:03 +0000)]
assembler: fix condition for printing a warning
Signed-off-by: Thomas Wood <thomas.wood@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Thomas Wood [Fri, 7 Feb 2014 17:03:37 +0000 (17:03 +0000)]
debugger: fix the -p option
Signed-off-by: Thomas Wood <thomas.wood@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Tue, 11 Feb 2014 11:14:09 +0000 (12:14 +0100)]
lib/drmtest: Fix up skip_henceforth checks
We need to first check whether we'll even run the testcase before
deciding to print a premature SKIP/FAIL. Without this and global
skipped resulted in superflous output when running only one testcase.
References: https://bugs.freedesktop.org/show_bug.cgi?id=74100
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Tue, 11 Feb 2014 11:04:57 +0000 (12:04 +0100)]
tests/gem_reset_stats: Correctly wrap global test code
Tests _must_ be able to run as non-root and on non-intel platforms,
otherwise the enumeration is busted (since QA enumerates testcases on
a gpu-less xeon box).
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Fri, 7 Feb 2014 09:02:41 +0000 (10:02 +0100)]
tests/pm_rps: remove setfreq
It's unused.
Cc: Jeff McGee <jeff.mcgee@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Fri, 7 Feb 2014 08:59:42 +0000 (09:59 +0100)]
tests/pm_rps: Round requested freq correctly
The kernel will round it, so if we don't we'll have a spurious
mismatch. Happens on my machine here with 650-1300MHz range, where the
midpoint is 975.
Cc: Jeff McGee <jeff.mcgee@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Thomas Wood [Thu, 6 Feb 2014 16:31:54 +0000 (16:31 +0000)]
lib: fix signed/unsigned comparison issues
Store the result of set_vt_mode as a signed value so that errors can be
caught correctly.
Signed-off-by: Thomas Wood <thomas.wood@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Thomas Wood [Thu, 6 Feb 2014 16:23:42 +0000 (16:23 +0000)]
assembler: define YY_NO_INPUT to prevent unused symbol warnings
Signed-off-by: Thomas Wood <thomas.wood@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Thu, 6 Feb 2014 06:49:33 +0000 (07:49 +0100)]
tests/pm_rps: Fix compilation on Linux
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Jeff McGee [Fri, 31 Jan 2014 16:13:53 +0000 (10:13 -0600)]
pm_rps: New subtest for gpu reset
Test that RPS functions as expected after a gpu reset.
Signed-off-by: Jeff McGee <jeff.mcgee@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Jeff McGee [Fri, 31 Jan 2014 16:13:52 +0000 (10:13 -0600)]
pm_rps: Add variable load support to load helper
The load helper can be set to HIGH or LOW. HIGH is the original mode
of sending continuous dword store commands. LOW adds a pause between
each command to reduce throughput.
Signed-off-by: Jeff McGee <jeff.mcgee@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Jeff McGee [Fri, 31 Jan 2014 16:13:51 +0000 (10:13 -0600)]
pm_rps: Load helper should stall for last write
This ensures that gpu is quiescent when load helper exits.
Signed-off-by: Jeff McGee <jeff.mcgee@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Jeff McGee [Fri, 31 Jan 2014 16:13:50 +0000 (10:13 -0600)]
pm_rps: Add stop rings injection utility
Signed-off-by: Jeff McGee <jeff.mcgee@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Jeff McGee [Wed, 29 Jan 2014 00:25:06 +0000 (18:25 -0600)]
pm_rps: New subtest min-max-config-loaded
Like subtest basic-api, but additionally requires that current
frequency is increasing to the configured maximum within reasonable
time since we are loaded.
Signed-off-by: Jeff McGee <jeff.mcgee@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Jeff McGee [Wed, 29 Jan 2014 00:25:05 +0000 (18:25 -0600)]
pm_rps: Add load helper for loading gpu
The load helper submits repeated dword store commands to keep the
gpu loaded while subtests running in the parent process check for
expected rps response.
Signed-off-by: Jeff McGee <jeff.mcgee@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Jeff McGee [Wed, 29 Jan 2014 00:25:04 +0000 (18:25 -0600)]
lib/drmtest: Add igt_wait_helper
igt_wait_helper compliments igt_stop_helper and is used when helper
processes are expected to exit naturally.
Signed-off-by: Jeff McGee <jeff.mcgee@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Jeff McGee [Wed, 29 Jan 2014 00:25:03 +0000 (18:25 -0600)]
pm_rps: New subtest min-max-config-idle
Like subtest basic-api, but additionally requires that current
frequency is settling to the configured minimum within reasonable
time since we are idle.
Signed-off-by: Jeff McGee <jeff.mcgee@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Imre Deak [Wed, 5 Feb 2014 11:48:26 +0000 (13:48 +0200)]
lib/drmtest: fix hang in quiescent_gpu_at_exit due to signal-unsafe asprintf
I got kms_flip in hung state with the backtrace below, while the parent process
waiting for the signal helper to exit. It was quite easy to reproduce the bug
by running
kms_flip --run-subtest=flip-vs-dpms-off-vs-modeset
With the change I couldn't reproduce it.
TODO: audit/fix other signal handlers with signal-unsafe functions
0 0x00007f9a1362018b in ?? () from /lib/x86_64-linux-gnu/libc.so.6
1 0x00007f9a1359df81 in ?? () from /lib/x86_64-linux-gnu/libc.so.6
2 0x00007f9a1359b6cf in ?? () from /lib/x86_64-linux-gnu/libc.so.6
3 0x00007f9a13628eb6 in __vasprintf_chk () from /lib/x86_64-linux-gnu/libc.so.6
4 0x00007f9a13628e72 in __asprintf_chk () from /lib/x86_64-linux-gnu/libc.so.6
5 0x000000000040a4a2 in asprintf (__fmt=0x417441 "/dev/dri/card%u", __ptr=0x7fff1a972c08)
at /usr/include/x86_64-linux-gnu/bits/stdio2.h:178
6 drm_get_card () at drmtest.c:190
7 0x000000000040a54a in __drm_open_any () at drmtest.c:229
8 0x000000000040a846 in quiescent_gpu_at_exit (sig=<optimized out>) at drmtest.c:281
9 0x0000000000408759 in call_exit_handlers (sig=3) at drmtest.c:1519
10 fatal_sig_handler (sig=3) at drmtest.c:1543
11 <signal handler called>
12 0x00007f9a13596770 in ?? () from /lib/x86_64-linux-gnu/libc.so.6
13 0x00007f9a135d8f3f in fork () from /lib/x86_64-linux-gnu/libc.so.6
14 0x000000000040b3af in __igt_fork_helper (proc=0x61d8cc <signal_helper>) at drmtest.c:1199
15 0x000000000040b4ce in igt_fork_signal_helper () at drmtest.c:751
16 0x0000000000404167 in main (argc=<optimized out>, argv=<optimized out>) at kms_flip.c:1533
v2:
- can't use snprintf as that's also signal-unsafe, so just use a dup'ed fd (Chris,Daniel)
Signed-off-by: Imre Deak <imre.deak@intel.com>
Chris Wilson [Tue, 4 Feb 2014 14:14:31 +0000 (14:14 +0000)]
tests: Add gem_madvise
Exercise that calling madvise produces expected results
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Tue, 4 Feb 2014 12:10:37 +0000 (13:10 +0100)]
fix make distcheck
Bunch of explicit include paths needed adjustments and
eviction_common.c needs to be added to the dist files.
This has been broken in the following three commits:
commit
42bcd05eb3f1545fbf9c397c3f37c3f6a27c5da4
Author: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Date: Mon Feb 3 10:59:41 2014 +0000
tests/eviction_common: Avoid submitting duplicate objects
commit
b92b397d5e4ffe10ed92fb2fc6711758ef5bcf89
Author: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Date: Tue Dec 3 15:09:57 2013 +0000
build: Make sure asserts are enabled for tests
commit
8c33e9ad111d27b84db4daac43809a807f262632
Author: Ben Widawsky <benjamin.widawsky@intel.com>
Date: Fri Dec 6 20:35:29 2013 -0800
intel-gpu-tools: Version informatio
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Ben Widawsky <benjamin.widawsky@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Tvrtko Ursulin [Mon, 3 Feb 2014 10:59:41 +0000 (10:59 +0000)]
tests/eviction_common: Avoid submitting duplicate objects
Make sure selection loop does not generate duplicates
when it picks a subset of objects for a single exec buffer.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Tvrtko Ursulin [Mon, 3 Feb 2014 10:59:40 +0000 (10:59 +0000)]
tests/gem_evict_everything: Factor out eviction logic
In preparation for userptr test we move the eviction logic
into a common file so it can be used from both test cases.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Damien Lespiau [Wed, 29 Jan 2014 15:53:33 +0000 (15:53 +0000)]
gem_ring_sync_copy: Add a ring to ring synchronization test
The goal of this test is to ensure that we respect inter ring
dependencies. A more detailed description of what it tests is in a
comment.
The tests relies on having a blit function for the ring, so is currently
only checking synchronization between the render and blitter ring.
v2: Actually create an inter-ring dependency by making the first copy on
ring2 and the second on ring2, not both on ring2.
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Rafael Barbalho [Fri, 31 Jan 2014 14:57:40 +0000 (14:57 +0000)]
android: Change tests & tools directory to use the lib directory
Instead of recompiling the lib directory for every tool or tests we can just
re-use the static library. This also has the nice side effect of fixing the
android "mm" command to allow android users to only rebuild the local changes
in either the test or tools directory.
Signed-off-by: Rafael Barbalho <rafael.barbalho@intel.com>
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Rafael Barbalho [Fri, 31 Jan 2014 14:57:39 +0000 (14:57 +0000)]
android: Add the generation of vesion.h & config.h to lib
The lib directory is used by all the tests & tools so it should really be
the place where the generate files are created.
Signed-off-by: Rafael Barbalho <rafael.barbalho@intel.com>
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Rafael Barbalho [Fri, 31 Jan 2014 14:57:38 +0000 (14:57 +0000)]
android: Add makefile for the lib directory
The lib directory should really be compiled as a static library on its own
and be re-used by any tests or tools that require it.
Signed-off-by: Rafael Barbalho <rafael.barbalho@intel.com>
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Rafael Barbalho [Fri, 31 Jan 2014 14:57:37 +0000 (14:57 +0000)]
android: Clean-up common makefile directives
All the Android.mk files had the same directives to find the path of
libdrm & libpciaccess. These are no longer required as the android system
now allows the libraries being used to export include paths to dependant
modules & programs.
Signed-off-by: Rafael Barbalho <rafael.barbalho@intel.com>
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Rafael Barbalho [Fri, 31 Jan 2014 14:57:36 +0000 (14:57 +0000)]
android: Handle the case when android doesn't have mmap64
Not all versions of Android have the mmap64 call, thus a suitable alternative
must be called.
Signed-off-by: Rafael Barbalho <rafael.barbalho@intel.com>
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Ben Widawsky [Fri, 31 Jan 2014 05:33:56 +0000 (05:33 +0000)]
quick_dump/broadwell: Add gen7 ring mode registers
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Ben Widawsky [Fri, 31 Jan 2014 05:31:07 +0000 (05:31 +0000)]
quick_dump/haswell: Add gen7 stuff + VECS_MODE
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Ben Widawsky [Fri, 31 Jan 2014 05:29:27 +0000 (05:29 +0000)]
quick_dump/gen7: Add other GFX_MODE registers
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Ben Widawsky [Fri, 31 Jan 2014 04:46:06 +0000 (04:46 +0000)]
quick_dump: Add a file argument
-f, or --file will simply dump the file in the specified format. All
other options will be skipped. This is helpful when one wants to create
their own list for temporary debug, or if one requires a subset of
registers to work around things like fused off PCH.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Ben Widawsky [Fri, 31 Jan 2014 04:42:50 +0000 (04:42 +0000)]
quick_dump: Refactor to allow future code
Extract everything possible, and create a main function which is pretty
darn simple. The goal is to make the various paths the program can take
based on command line options (one more is coming) more decipherable.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>