In 15e8ffec ("tests/perf: respect gen11 restriction in disable
subslices") I fixed the global-sseu-config subtest which was affected
by the same issue. But the global-sseu-config-invalid subtest is also
affected by the same Gen11 restriction.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 779d43cd ("tests/perf: add global sseu parameter tests")
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/1543


Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>

Instantiating the objects can be surprisingly, worryingly slow; even on
big core LLC devices. Rather than worry about that and cause timeouts in
simply setting up the test, apply a limit to the setup time so that we
only use the amount of objects instantiated within the first 20s.

Due to the quadratic behaviour being exposed by the test, this may not
entirely eliminate the timeouts -- but at least the timeouts should then
be for the right reason.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>

In a few cases, we hit a timeout where no process appears to be
deadlocked (i.e. tasks stuck in 'D' with intertwined stacks) but
everything appears to be running happily. Often, they appear to be
fighting over the shrinker, so one naturally presumes we are running low
on memory. But for tests that were designed to run with ample memory to
spare, that is a little disconcerting and I would like to know where the
memory actually went.

sysrq('m'): Will dump current memory info to your console

Sounds like that should do the trick.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Petri Latvala <petri.latvala@intel.com>
Acked-by: Petri Latvala <petri.latvala@intel.com>

To help verify correct deployment, add a --version flag to igt_runner
that just prints the IGT-version text, the same tests would print.

Note that only igt_runner gained the --version flag. igt_resume and
igt_results don't do fancy flag handling, they only accept the
directory to operate as their single arg.

v2: Depend on version.h (CI)

Signed-off-by: Petri Latvala <petri.latvala@intel.com>
Cc: Tomi Sarvela <tomi.p.sarvela@intel.com>
Cc: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
Reviewed-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com>

To help verify correct deployment, add a --version flag that just
prints the IGT-Version text.

Signed-off-by: Petri Latvala <petri.latvala@intel.com>
Cc: Tomi Sarvela <tomi.p.sarvela@intel.com>
Cc: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
Reviewed-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com>

Janusz Krzysztofik authored 6 years ago and Janusz Krzysztofik committed 4 years ago

There is a test which verifies unloading of i915 driver module but no
test exists that checks how a driver behaves when it gets unbound from
a device or when the device gets unplugged.  Implement such test using
sysfs interface.

Two minimalistic subtests - "unbind-rebind" and "unplug-rescan" -
perform the named operations on a DRM device which is believed to be
not in use.  Another pair of subtests named "hotunbind-lateclose" and
hotunplug-lateclose" do the same on a DRM device while keeping its file
descriptor open and close it thereafter.

v2: Run a subprocess with dummy_load instead of external command
    (Antonio).
v3: Run dummy_load from the test process directly (Antonio).
v4: Run dummy_load from inside subtests (Antonio).
v5: Try to restore the device to a working state after each subtest
    (Petri, Daniel).
v6: Run workload inside an igt helper subprocess so resources consumed
    by the workload are cleaned up automatically on workload subprocess
    crash, without affecting test results,
  - move the igt helper with workload back from subtests to initial
    fixture so workload crash also does not affect test results,
  - other cleanups suggested by Katarzyna and Chris.
v7: No changes.
v8: Move workload functions back from fixture to subtests,
  - register different actions and different workloads in respective
    tables and iterate over those tables while enumerating subtests,
  - introduce new subtest flavors by simply omitting module unload step,
  - instead of simply requesting bus rescan or not, introduce action
    specific device recovery helpers, required specifically with those
    new subtests not touching the module,
  - split workload functions in two parts, one spawning the workload,
    the other waiting for its completion,
  - for the new subtests not requiring module unload, run workload
    functions directly from the test process and use new workload
    completion wait functions in place of subprocess completion wait,
  - take more control over logging, longjumps and exit codes in
    workload subprocesses,
  - add some debug messages for easy progress watching,
  - move function API descriptions on top of respective typedefs.
v9: All changes after Daniel's comments - thanks!
  - flatten the code, don't try to create a midlayer (Daniel),
  - provide minimal subtests that even don't keep device open (Daniel),
  - don't use driver unbind in more advanced subtests (Daniel),
  - provide subtests with different level of resources allocated
    during device unplug (Daniel),
  - provide subtests which check driver behavior after device hot
    unplug (Daniel).
v10 Rename variables and function arguments to something that
    indicates they're file descriptors (Daniel),
  - introduce a data structure that contains various file descriptors
    and a helper function to set them all (Daniel),
  - fix strange indentation (Daniel),
  - limit scope to first three subtests as the initial set of tests to
    merge (Daniel).
v11 Fix typos in some comments,
  - use SPDX license identifier,
  - include a per-patch changelog in the commit message (Daniel).
v12 We don't use SPDX license identifiers nor GPL-2.0 in IGT (Petri),
  - avoid chipset, make sure we reopen the same device (Chris),
  - rename subtest "drm_open-hotunplug" to "hotunplug-lateclose",
  - add subtest "hotunbind-lateclose" (less affected by IOMMU issues),
  - move some redundant code to helpers,
  - reorder some helpers,
  - reword some messages and comments,
  - clean up headers.
v13 Add test / subtest descriptions (patchwork).
v14 Extract redundant device rescan and reopen code to a 'healthcheck'
    helper,
  - call igt_abort_on_f() on device reopen failure (Petri),
  - if any timeout set with igt_set_timeout() inside a subtest expires,
    call igt_abort_on_f() from a follow-up igt_fixture (Petri),
  - when running on a i915 device, require GEM and call
    igt_abort_on_f() if no usable GEM is detected on device reopen.
v15 Add the test to CI blacklist (Martin).
v16 Separate blacklist entry with a descriptive comment (Petri).

Signed-off-by: Janusz Krzysztofik <janusz.krzysztofik@intel.com>
Cc: Antonio Argenziano <antonio.argenziano@intel.com>
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Katarzyna Dec <katarzyna.dec@intel.com>
Cc: Martin Peres <martin.peres@linux.intel.com>
Acked-by: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Petri Latvala <petri.latvala@intel.com>

Adding eDP support for Link Layer Compliance tool
of DP-PHY test automation. Current Implementation
of this compliance tool can be used for dp and eDP
both. Hence extending to eDP too.

Signed-off-by: Nidhi Gupta <nidhi1.gupta@intel.com>
Signed-off-by: Ramalingam C <ramalingam.c@intel.com>
Reviewed-by: Animesh Manna <animesh.manna@intel.com>
Link: https://patchwork.freedesktop.org/series/72783/#rev2

Bob Paauwe authored 4 years ago and Jani Nikula committed 4 years ago

Add XYUV8888 to the list of DRM Formats to test.

Also fix the byte order for the format.

Signed-off-by: Bob Paauwe <bob.j.paauwe@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>

It was missing a ")" to be properly parsed.

Reviewed-by: Petri Latvala <petri.latvala@intel.com>
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>

It's takes a long time to setup the test in CI, but how long?
That is the question.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>

Our goal with this gem_*_blits variant is to exercise fence reuse, and
since the fenced blits [of yore] can only fit into the fenceable region
of the mappable aperture, we can limit our working set to cause
thrashing of that region. This dramatically reduces the workload for big
machines that were trying to trash everything.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>

The poll() is proving unreliable, where our tests timeout without the
spinner being terminated. Let's try a blocking read instead!

Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/1676


Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: "Dixit, Ashutosh" <ashutosh.dixit@intel.com>
Reviewed-by: "Dixit, Ashutosh" <ashutosh.dixit@intel.com>

Use igt_dynamic_f for the dynamic discovered CRTC sets for each test.

v2:
The subsubtest names are slightly more compact and forced to conform to
igt_runner/piglit rules:

	A-eDP1
	AB-eDP1-eDP2

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Petri Latvala <petri.latvala@intel.com>
Cc: Martin Peres <martin.peres@linux.intel.com>
Reviewed-by: Petri Latvala <petri.latvala@intel.com>

Ashutosh Dixit authored 4 years ago and Chris Wilson committed 4 years ago

Can't touch the memory if mmap returns MAP_FAILED.

v2: s/(char *)-1)/MAP_FAILED/ (Mike Ruhl)

Fixes: 9cff3be1("igt/gem_shrink: Reduce oom spam")
Signed-off-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Reviewed-by: Michael J. Ruhl <michael.j.ruhl@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

This test has been spamming logs like crazy since 5.7-rc1, filling
every disk in sight and causing CI processing to grind to a halt.

Signed-off-by: Petri Latvala <petri.latvala@intel.com>
Cc: Tomi Sarvela <tomi.p.sarvela@intel.com>
Cc: Martin Peres <martin.peres@linux.intel.com>
Cc: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>

The default scenario is now performing all hotplugs with modes disabled
on all connectors. This is the quickest of the tests and represents
userspace not caring about the new display (e.g. explicitly disabled).

*-hpd-enable-disable-mode covers the most common userspace behavior
where each hotplug event is accompanied by a corresponding enabling /
disabling commit.

*-hpd-with-enabled-mode explicitly targets the scenario where we have
mode enabled and never disable it as we do hotplugs to reproduce the
issue we see with TypeC connectors for ICL and TGL.

v2:
 - refresh igt_display output state after reprobing
 - get mode and set pipe only after we have connector plugged in

v3: fix VGA subtest names (Kunal)

Cc: Kunal Joshi <kunal1.joshi@intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Issue: https://gitlab.freedesktop.org/drm/intel/issues/323


Signed-off-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
Reviewed-by: Kunal Joshi <kunal1.joshi@intel.com>

Each tests resets the state to be sure it starts with a clean slate.
Let's add disabling modeset to the pile, to be sure that we get the
hotplug events and the connector state is correct.

Cc: Kunal Joshi <kunal1.joshi@intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Issue: https://gitlab.freedesktop.org/drm/intel/issues/323


Signed-off-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
Reviewed-by: Kunal Joshi <kunal1.joshi@intel.com>

Doing disabling modeset from igt_display_require() is quite tricky, as
there are some tests that depend on the leftover mode set by fbcon in
order to work + PSR debugfs checks seem to behave a bit weird.

While those issues are being worked on let's do a disabling modeset
between igt_display_require() and chamelium_init() so port
auto-discovery works reliably on all configurations.

Issue: https://gitlab.freedesktop.org/drm/intel/issues/323


Cc: Kunal Joshi <kunal1.joshi@intel.com>
Signed-off-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
Reviewed-by: Kunal Joshi <kunal1.joshi@intel.com>

Different ppgtt should be independent under full-ppgtt.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>

Hang tests are slow since we must wait for the kernel to detect the GPU
hang before it forcibly completes the execution and signals the fence.
The flip is not allowed before the fence is signaled and so must wait.
Our test is that the flip does eventually occur, we only need to do that
once, and let the power of repeated runs across many machines weed out
the corner cases.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Andi Shyti <andi.shyti@intel.com>

Extend initial check for support of MMAP_GTT mapping to userptr with
equivalent checks for each MMAP_OFFSET mapping type supported by i915
driver.  Based on that, extend coverage of process-exit-gtt* subtests
over non-GTT mapping types.  In case of dmabuf-* subtests, use first
supported mapping type if there are any.

v2: Clear 'map' before reuse (Zbigniew).
v3: Kill out-of-context errno check (Chris).

Signed-off-by: Janusz Krzysztofik <janusz.krzysztofik@linux.intel.com>
Reviewed-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>

Upgrade the subtest to use MMAP_GTT API v4 (aka MMAP_OFFSET),
dynamically examine each mapping type supported by i915 driver.

Signed-off-by: Janusz Krzysztofik <janusz.krzysztofik@linux.intel.com>
Reviewed-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>

Include the current spin->timerfd status in the warning message when the
timeout did not fire.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>

Allow for a timerfd of 0, just in case some test closes stdin.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>

Some HW woes are forcing us to store EU activity counters in a
different way.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Acked-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>

Check that our threaded timer to expire the igt_spin_t does fire!

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>

Since we probe the existence of the engines, we expect errno to be
randomly set prior to starting the test. Lets reset it to 0 so that any
error before the igt_dynamic_f test is not mistakenly printed by a
failed test.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>

There was an error in

commit 0ab05a51 ("tests/kms_concurrent: Test for maximum number of planes")

that calculates the maximum number of supported planes. The patch proposes to
test first if an atomic commit is successful with the selected number of planes.
In case of a failure, the number of planes is reduced by one assuming that the
failure was caused by bandwidth limit. The test loops at least as many iterations
as there are planes defined by the platform.

v2: Fix CI build failure (CI)

Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Reviewed-by: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com>

This reverts commit 2b65609b.

Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Acked-by: Petri Latvala <petri.latvala@intel.com>

S4-devices is a debug only path, and while it can prove informative as
to the nature of suspend issues, being a debug only path it is not so
relevant towards the driver health. Relegate it to the shards and idle
runs, so we can shave 20s off BAT.

gem_exec_suspend@basic itself is just a test suspend, a sanitycheck to
ensure the test is functional, for manual debugging of the tests.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Martin Peres <martin.peres@linux.intel.com>
Cc: Petri Latvala <petri.latvala@intel.com>
Acked-by: Martin Peres <martin.peres@linux.intel.com>

Sometimes the bg_load only wakes up once or twice in 3s. That's
just unbelievable, so include some measurements to see how long the
load spends in submission & waiting.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>

Tell the naughty child to use a separate i915 so that there is no
possibility that it should overlap with anything else.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>

Chamelium provides us with hsync_offset value which is read form its
receiver register described as "Hsync start edge to H active start
edge", vsync_offest is analogous.

Let's calculate the correct things on IGT side.

v2: keep the original checks for HDMI (Petri)

Cc: Petri Latvala <petri.latvala@intel.com>
Cc: Nidhi Gupta <nidhi1.gupta@intel.com>
Signed-off-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
Reviewed-by: Petri Latvala <petri.latvala@intel.com>

The goal is to create just enough of a workload that will run long
enough to delay others and be detectable.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>

As now kernel enables FBC on linear surfaces on GEN9+, it will allow
FBC to be enabled on fbcon, so doing the changes he to tests this new
case.

Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>

Every time fbc_wait_until_enabled() is called from
fbc_wait_until_update() it was waiting the whole 5 seconds of timeout
to return false as expected.
To save most 99% of this adding here fbc_wait_until_disabled().

v5:
- Fixed fbc_wait_until_update() test

Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>

The premise of the test is that one client does not block the other, to
which we only need to verify that the ready client proceeds unhindered
by the abusive client. So we only need assert that the ready client
completes within a shorter time than the abusive client.

It would be nice that we could assert that under all conditions it never
takes more than a millisecond to submit a no-op; but that goal is a bit
too far for today.

Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/1654


Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>

Include the reason why we are dumping the task state (test timeout) in
the kmsg log prior to the task state. Hopefully this helps when reading
the dump.

v2: Use asprintf to combine the strings into one to avoid error prone
manual string handling and enjoy one single write() into the kmsg.

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>

i915 is making fences not mandatory to enable FBC in newer platforms.
As BSpec do not have restrictions against tiling formats to enable
FBC it will be possible to enable FBC with linear tiling, so lets test
it.

v2:
- changed from GEN11 to GEN9 following kernel patches

Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Petri Latvala <petri.latvala@intel.com>
Reviewed-by: Juha-Pekka Heikkila <juhapekka.heikkila@gmail.com>
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>