It's nice to see just how many components the crc claims to have
when the count don't match what we expect.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
The extra_long_opts passed to igt_*_parse_opts() isn't modified,
so let's make it const.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
The joy of our hardware; don't let two threads attempt to read the same
register at the same time.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Allow the producers to be set with maximum RT priority to verify that
the waiters are not exhibiting priorty-inversion.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Try a different pattern to cascade the cancellation from producers to
their consumers in order to avoid one potential deadlock.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Do the workload before the nop, so that if combining both, there is a
better chance for the spurious interrupts. Emit just one workload batch
(use the nops to generate spurious interrupts) and apply the factor to
the number of copies to make inside the workload - the intention is that
this gives sufficient time for all producers to run concurrently.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Split the distinct phases (generate interrupts, busywork, measure
latency) into separate batches for finer control.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Knowing how long it takes to execute the workload (and how that scales)
is interesting to put the latency figures into perspective.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Late last night I forgot I had only added the llc CPU mmaping and not
the !llc GTT mapping for byt/bsw.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
The goal is measure how long it takes for clients waiting on results to
wakeup after a buffer completes, and in doing so ensure scalibilty of
the kernel to large number of clients.
We spawn a number of producers. Each producer submits a busyload to the
system and records in the GPU the BCS timestamp of when the batch
completes. Then each producer spawns a number of waiters, who wait upon
the batch completion and measure the current BCS timestamp register and
compare against the recorded value.
By varying the number of producers and consumers, we can study different
aspects of the design, in particular how many wakeups the kernel does
for each interrupt (end of batch). The more wakeups on each batch, the
longer it takes for any one client to finish.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
The upper bound for SLOW_QUICK was added for the benefit of the slow
simulator, not because, as I wrongly thought, of the latency
measurements.
SLOW_QUICK was added in
commit d1e862324b747a0ab5d985eaa6830076817231c5
Author: Damien Lespiau <damien.lespiau@intel.com>
Date: Mon Mar 25 20:06:20 2013 +0000
tests: Instrument tests run in simulation to run quickly
and dropped in
commit 89bcdb9022fb7a1f66635b9f2546356ad0c0761a
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date: Tue Dec 8 13:42:50 2015 +0000
igt/gem_exec_nop: Remove nop latency measurements
Reported-by: Ben Widawsky <benjamin.widawsky@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
So there's 3 competing proposals for what wait_ioctl should do wrt
-EIO:
- return -EIO when the gpu is wedged. Not terribly useful for
userspace since it might race with a hang and then there's no
guarantee that a subsequent execbuf won't end up in an -EIO.
Terminally wedge really can only be reliably signalled at execbuf
time, and userspace needs to cope with that (or decide not to
bother).
- EIO for any obj that suffered from a reset. This means big internal
reorginazation in the kernel since currently we track reset stats
per-ctx and not on the obj. That's also what arb robustness wants.
We could do this, but this feels like new ABI territory with the
usual userspace requirements and high hurdles.
- No -EIO at all. Consistent with set_domain_ioctl and simplest to
implement. Which is what this patch does.
We can always opt to change this later on if there's a real need.
To make the test really exercise this do a full wedged gpu hang, to
make sure -EIO doesn't leak out at all.
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
gem_flink_race and prime_self_import have subtests which read the
number of open gem objects from debugfs to determine if objects have
leaked during the test. However the test can fail sporadically if
the number of gem objects changes due to other process activity.
This patch introduces a change to check the number of gem objects
several times to filter out any fluctuations.
v2: Moved the common code to a library and made the loop android
specific (Daniel Vetter)
v3: Renamed get_stable_obj_count -> igt_get_stable_obj_count
Signed-off-by: Derek Morton <derek.j.morton@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
We get build error as we try to cast from ptr to integer
of different size on 32 bit platforms. Use unsigned long
as the cast, it will work with both 32 and 64 bit
systems.
Cc: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Since we hold an exclusive write lock we expect 2 writes to happen
serially, but we expect 2 reads to happen in parallel. Expand the testing
to demonstrate this effect (i.e. we expect read-read to be roughly 2x
faster than write-write for small copies on big core.)
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Many warnings of the form
gem_pread.c: In function ‘main’:
gem_pread.c:128:8: warning: assignment discards ‘const’ qualifier from
pointer target type [-Wdiscarded-qualifiers]
bps = bytes_per_sec(buf, object_size/usecs*1e6);
Regression from
commit 48c945322b4c5f6443758143cccb9c4c04da4aaa
Author: Ankitprasad Sharma <ankitprasad.r.sharma@intel.com>
Date: Wed Dec 2 14:54:51 2015 +0530
igt/gem_pread: Support to verify pread/pwrite for non-shmem backed obj
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
This reverts commit 4f5efc5c844f6fe69209982463f9220f8f3951ed.
There was a bit a misunderstanding on IRC between Chris&me. We want
basic tests as sanity test to be run in the BAT CI. It's just unfortunate
that right now we have fairly limited ability to absorb new ones, both
because of a pile of existing bugs in the kernel and because the CI
infrastructure is still being scaled out.
The idea was just to remove the BAT tests added yesterday, not all of
the ones we've had for a while longer.
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Grumpily-acked-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
In times past, I added "basic" variants of tests just to ensure that the
general principle of operation was sound before proceeding on to the
main test (which typically looked at thrashing, i.e. were long and
tedious and pointless if the test didn't even work in the normal
situation). Since "basic" now collides with BAT, rename my trivial tests
to "sanitycheck".
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
As the concurrency tests are a good source of stress for
i915_wait_request() (the tests are primarily designed to ensure that GPU
activity of one form or another is completed before access by third
parties), one of the common form of errors we can detect are the
"missing interrupts" (i.e. where the waits do not terminate because of a
race between the interrupt and the seqno write). Add an explicit check
for this error and flag it as a definite fail - which also helps narrow
it down to certain subtests when run as a batch.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
When changing the sysfs GT min frequency, the kernel won't
automatcilly drop the GT frequency to idle unless the GPU
transitions from busy to idle.
Load the GPU after increasing the GT min frequency to force
a busy to idle transition. This matches the behavior when
decreasing the GT min frequency.
Signed-off-by: Bob Paauwe <bob.j.paauwe@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Using same timeout value as kms_fronbuffer_tracking and for
same reasons exposed at 'commit 83582f9b ("kms_frontbuffer_tracking:
Increase the time we wait for PSR.")'
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
These tests exercise the userptr ioctl to create shared buffers
between CPU and GPU. They contain error and normal usage scenarios.
They also contain a couple of stress tests which copy buffers between
CPU and GPU. These tests rely on the softpin patch in order to pin buffers
to a certain VA.
Caveat: These tests were designed to run on 64-bit system. Future work
includes adding logic to ensure these tests can run on 32-bit systems with
PPGTT support. Some tests are currently disabled for 32-bit systems for that
reason.
v2: Added cc and signed-off-by fields
v3: Fixed review comments, added helper functions. Removed userptr error
scenarios covered by existing userptr tests. Modified stress test to have
100K buffers, it now runs for ~30 mins, checks every element has been written
to correctly, and pins buffers at different VMAs.
v4: Changed name to gem_softpin
v5: More fixes. Removed the file based tests, will move them to userptr tests.
Added a function that validates appropriate PPGTT support before running tests.
Optimized stack space and memory footprint in stress test. Removed the eviction
test, will add it back after verifying proper functionality.
v6: Split basic test into userptr and bo
Fixed some coding style issues.
v7: Enhanced invalid vma pinning test to verify 32-bit PPGTT functionality.
Enabled the test for 32-bit PPGTT systems, and verify pinning fails above
32-bit addresses. Enhanced the high adress pinning test to ensure pinning
fails when EXEC_OBJECT_PINNED flag is not used. Some more cosmetic fixes to
close buffer handles. Changed userptr function to used synchronized operations.
v8: Minor change to high address pinning test as per comment.
v9: Skip the tests if softpin support is not present.
v10: Removed trailing white spaces.
v11: Keep alphabetical order in Makefile and gitignore; update error code
returned while trying to pin above the max vm size (EINVAL); test attempt
to pin above 4GB without the support 48b flag.
Cc: Michel Thierry <michel.thierry@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Signed-off-by: Michel Thierry <michel.thierry@intel.com> (v11)
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Since commit c8beadb811bdc2b9c21f95144852f3e55867b546
Author: Derek Morton <derek.j.morton@intel.com>
Date: Wed Nov 11 14:46:58 2015 +0000
tests/gem_exec_nop: Improved test run time
the stablity of the measurement regressed and it is once again no longer
a reliable metric for detecting regressions. Give up.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Derek Morton <derek.j.morton@intel.com>
Add a new subtest that makes sure old stale modes get pruned from the
connector's mode list when the EDID changes.
v2: s/drmModeGetConnector/drmModeGetConnectorCurrent/ since
kmstest_force_edid() already takes care of doing the heavier
call for us (Daniel)
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Although kms_frontbuffer_tracking already has psr-suspend testcase
this one here can complement it by testing different combination
and mainly covering 2 different cases individually:
1. wait-for-psr, suspend-resume tehn run 1 operation.
2. suspend-resume, wait-for-psr then run 1 operation.
v2: Remove no-suspend option since this should be done with piglit
if necessary for now.
v3: argh! remove remaining no-suspend checks...
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
commit 75b286e821 ("tests/kms_psr_sink_crc: test even
if PSR is disabled by default")' force PSR enabling without
respecting the no-psr (running-with-psr-disabled) option.
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Even with all sink crc re-works we still have platforms
where after 6 vblanks it is unable to calculate the
sink crc. But if we don't get the sink crc it isn't true
that test failed, but that we have no ways to say test
passed or failed.
So let's print a message and move forward in case sink crc
cannot help us to know if the screen has been updated.
v2: Also include a message on setup_sink_crc and also
only skip when it is mandatory, i.e. when running for PSR.
Acked-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Unfortunately Sink CRC is not 100% reliable for all platforms.
So we cannot block FBC tests nor skip them when we are getting
unreliable Sink CRC results, or not getting them at all.
Acked-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
With commit (drm/i915: Delay first PSR activation.) in kernel
PSR might take a bit longer to really activate after the modeset.
The first PSR activation after modeset is taking 5 times the panel
power cycle delay time, which is 600ms for our machines here.
So timeout here needs to be a minimum of 3s. However let's use
5s as the safe value in case we find machines with higher power
cycle delay.
Since we do a lot of assert(psr_disabled), this commit is increasing
the time it takes to run the whole set of PSR tests by a few minutes,
which had been reduced by commit f4db3b18841
("kms_frontbuffer_tracking: reduce the PSR wait timeout to 2s").
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Since commit
commit aed242ff7ebb697e4dff912bd4dc7ec7192f7581
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date: Wed Mar 18 09:48:21 2015 +0000
drm/i915: Relax RPS contraints to allows setting minfreq on idle
it is now possible that the current frequency will drop be the user
specified minimum frequency to the "idle" or RPn frequency. Update the
pm_rps tests to reflect that droping below the user specified minimum
is no longer considered a failure.
v2: Add check RPn <= current freq. (Me)
v3: Use RPn instead of MIN frequency in idle check (Imre)
Signed-off-by: Bob Paauwe <bob.j.paauwe@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
This way we correctly auto-skip instead of falling over the
lack of i915 debugfs files first and fail the testcase due to
that.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Instead of failing. We might want to move this into i915 tests
eventually, but this is good for now.
v2: Use the correct exit code (Derek) and use the new symbolic values
Thomas added.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
wait-ioctl skips a couple of side-effects of retiring, so provoke them
using set-domain before we sleep.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Forcing connector state is a basic piece of our test infrastructure
that we use in all the kms_ tests. It allows us to run tests even if
no outputs are connected.
They're also really fast, so perfect candidates for inclusion into the
BAT set.
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
