| Age | Commit message (Collapse) | Author |
|---|
|
It was left out.
Signed-off-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
Reviewed-by: Antonio Argenziano <antonio.argenziano@intel.com>
|
|
Actually check the error state exists (!"No error state captured") and
that it contains the expected engine dump.
v2: Throw in some debug clues.
v3: Fail if the file doesn't exist, or empty.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Antonio Argenziano <antonio.argenziano@intel.com>
|
|
We introduced a subslice mask storage per slice in newer kernels
(because of the possibility of asymmetry). As a result the debugfs
output has changed a bit.
v2: Add a comment on why we have a special case on subslice per slice
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
It appears that waiting for a 100us period whereby we are unable to
submit another batch and proclaim the ring full, may have the false
positive where the scheduler intervenes and we are signalled twice
before having slept on ring space. Increasing the interval reduces the
likelihood of the scheduler stealing the cpu from us, but does not
eliminate it. Fortuitously it appears to be a rare false positive.
For the library routine, we can fork a RT process but that seems a bit
overkill!
References: https://bugs.freedesktop.org/show_bug.cgi?id=105343
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Antonio Argenziano <antonio.argenziano@intel.com>
Reviewed-by: Antonio Argenziano <antonio.argenziano@intel.com>
|
|
Having demonstrated that FORCEWAKE_MT is suspect to the same old
concurrent mmio access bug that stalks gen7, we have shown that it is
not viable for userspace to poke around inside FORCEWAKE_MT directly. As
it can not work correctly, remove the test.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
|
|
Prevent the compiler from caching reads/writes to the hw register as we
do want to perform mmio.
Whilst fixing up the mmio access, also ensure that we do not leave the
test with any other bits still set in the forcewake register to prevent
affecting other tests, as spotted by Tvrtko.
v2: Use intel_mmio_use_pci_bar() rather open code the ioremap
v3: Flip igt_wait() checking as it returns true on success, not errno.
v4: Hohum, the mmio bug affects FORCEWAKE_MT as well.
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>
|
|
If we do a global wait while trying to execute spinners in parallel,
it ends badly with a GPU hang.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=104352
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
CPU hotplug, especially CPU0, can be flaky on commodity hardware.
To improve test reliability and reponse times when testing larger runs we
need to handle those cases better.
Handle failures to off-line a CPU by immediately skipping the test, and
failures to on-line a CPU by immediately rebooting the machine.
This patch includes igt_sysrq_reboot implementation from Chris Wilson.
v2: Halt by default, reboot if env variable IGT_REBOOT_ON_FATAL_ERROR is
set. (Petri Latvala)
v3: Add missign docs and update stale comment. (Petri Latvala)
v4: Use pause instead of sleep. (Chris Wilson)
v5: Newlines! (Chris Wilson)
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Petri Latvala <petri.latvala@intel.com>
Cc: Tomi Sarvela <tomi.p.sarvela@intel.com>
Reviewed-by: Petri Latvala <petri.latvala@intel.com>
|
|
Verify that the reported busyness is in line with what would we expect
from a batch which causes a hang and gets kicked out from the engine.
v2: Change to explicit igt_force_gpu_reset instead of guessing when a spin
batch will hang. (Chris Wilson)
v3: Assert and comment test expectations. (Chris Wilson)
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
Just a small variant to apply a continuous context-switch load to all
engines.
v2: Adapt to for_each_physical_engine() and sane gem_context_create()
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Antonio Argenziano <antonio.argenziano@intel.com>
Reviewed-by: Antonio Argenziano <antonio.argenziano@intel.com>
|
|
testdisplay.h includes the glib.h header file but the Makefile does not
explicitly pass a -I option with the path containing that header, hence
causing the build to fail. Note that this doesn't seem to happen with a
recent enough version of cairo, which implicitly provides the correct
-I option.
Signed-off-by: Thierry Reding <treding@nvidia.com>
Reviewed-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
|
|
EXEC_OBJECT_CAPTURE extends the type of buffers we may read during error
capture. Previously we knew that we would only see batch buffers (which
limited the objects to being from gem_create()), but now we need to
check that any buffer the user can create can be read. The first
alternate buffer type is a userptr.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Michał Winiarski <michal.winiarski@intel.com>
|
|
Much like the enable-disable subtest, we're printing a bunch of values
that were meant to try to figure out the issue of the OA unit not
producing reports. After fixing the i915 driver with :
https://patchwork.freedesktop.org/series/39112/
We don't need those values anymore.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
|
|
We're printing a bunch of values that were meant to try to figure out
the issue of the OA unit not producing reports. After fixing the i915
driver with :
https://patchwork.freedesktop.org/series/39112/
We don't need those values anymore. It turns out the issue was simply
a race condition in the driver.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
|
|
Reading from WC is awfully slow as each access is uncached and so
performed synchronously, stalling for the memory load. x86 did introduce
some new instructions in SSE 4.1 to provide a small internal buffer to
accelerate reading back a cacheline at a time from uncached memory, for
this purpose.
v2: Don't be lazy and handle misalignment.
v3: Switch out of sse41 before emitting the generic memcpy routine
v4: Replace opencoded memcpy_from_wc
v5: Always flush the internal buffer before use (Eric)
v6: Assume bulk moves, so check for dst alignment.
v7: Use _mm_fence for _buitlin_ia32_mfence for consistency, remove
superfluous defines (Ville)
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Eric Anholt <eric@anholt.net>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
|
|
__igt_spin_batch_new() may be used inside a background helper which is
competing against the GPU being reset. As such, we cannot even assert
that the spin->handle is busy immediately after submission as it may
have already been reset by another client writing to i915_wedged.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Abdiel Janulgue <abdiel.janulgue@linux.intel.com>
|
|
Move variable declaration to top of scope to avoid C90 build warning.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
|
|
Mark drrs_set as static to avoid a build warning.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
Since more essential components use libudev, make its dependency
mandatory.
Signed-off-by: Antonio Argenziano <antonio.argenziano@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Petri Latvala <petri.latvala@intel.com>
Cc: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Petri Latvala <petri.latvala@intel.com>
|
|
When tracking down the cause of a particular kernel warning, knowing
which file it is associated with can be a big clue. So write the
filename into the kernel message log prior to opening it.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
|
|
If the specified object can not fit into the GTT due to overlap with a
neighbouring pinned object (not part of the execobjects[]), we expect to
fail with ENOSPC (as we cannot evict, rather than EINVAL for the user
error in a badly constructed execobjects[]). To prevent the tests
causing overlap with other external objects expand the test hole by a
page on either side.
(Setting up the system to deliberately hit ENOSPC is trickier as for
example it requires pinned an object into the scanout with enough free
space on either side to test.)
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
|
|
When searching for a VGA connector to use to test overriding the
connector status, we require the system to have a disconnected VGA
connector, but if a previous test left an override inplace, that may not
exist. Before we check whether the connector is attached to real HW,
first reset the connector status override so that we always get the
actual HW result.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Marta Lofstedt <marta.lofstedt@intel.com>
Reviewed-by: Marta Lofstedt <marta.lofstedt@intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
|
|
In CI runs we every now and then fail to read correct CRC yielding an error
when comparing reference and grabbed CRC's. Let's first fix the test so that
we drain the pipe first and then read the correct CRC.
References: https://bugs.freedesktop.org/show_bug.cgi?id=103166
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Mika Kahola <mika.kahola@intel.com>
|
|
Although we want to specify exactly which physical engine to run on, the
busy ioctl can only return the I915_EXEC_RING identifier, i.e. the
aliased I915_EXEC_BSD for vcs0/vcs1. Horrors.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105248
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
|
|
We current have a single for_each_engine() iterator which we use to
generate both a set of uABI engines and a set of physical engines.
Determining what uABI ring-id corresponds to an actual HW engine is
tricky, so pull that out to a library function and introduce
for_each_physical_engine() for cases where we want to issue requests
once on each HW ring (avoiding aliasing issues).
v2: Remember can_store_dword for gem_sync
v3: Find more open-coded for_each_physical
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>
|
|
Replace custom execbuf ioctl wrapper with the ones in lib.
v2:
- Lib execbuf wrapper is not signal handling friendly. (Chris)
v3:
- EXECBUFFER2_WR != EXECBUFFER2. (Chris)
v4: Drop gem_exec_fence.c changes
Signed-off-by: Antonio Argenziano <antonio.argenziano@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Petri Latvala <petri.latvala@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
With intel_measure_ring_size added as common function we can use it
instead of the local copy
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Antonio Argenziano <antonio.argenziano@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
With intel_measure_ring_size added as common function we can use it
instead of the local copy
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
With intel_measure_ring_size added as common function we can use it
instead of the local copy
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
With igt_cork added as common utility we can use it instead of the
local copy
Signed-off-by: Antonio Argenziano <antonio.argenziano@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
With igt_cork added as common utility we can use it instead of the
local copy
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Antonio Argenziano <antonio.argenziano@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
With intel_measure_ring_size and igt_cork added as common utilities we
can use them instead of the local copy of those utilities
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
With intel_measure_ring_size and igt_cork added as common utilities we
can use them instead of the local copy of those utilities
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
With intel_measure_ring_size and igt_cork added as common utilities we
can use them instead of the local copy of those utilities
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
v3: Make sure we receive a SIGBUS signal when accessing memory of a
purged BO
Add <signal.h> include after rebase (by anholt)
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Eric Anholt <eric@anholt.net>
|
|
Missed the new method for igt_require(gem_has_contexts()) in the rebase.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
|
|
The previous patch said :
"verify that the time is always longer or equal to the period we've
asked for"
This is an obvious error, it only worked on my machine and the CI
because only one longer period was observed. But another CI run caught
the issue :
https://intel-gfx-ci.01.org/tree/drm-tip/IGT_4280/shard-glkb6/igt@perf@oa-exponents.html
Fixes: c3d11ca104fa ("tests/perf: make oa-exponents subtest more reliable")
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
|
|
for chipset support
If the machine doesn't support PSR, it will return -ENODEV from
i915_edp_psr_status, which we want to interpret as unsupported.
This is in line with what Chris just fixed for FBC.
V2: Copy-pasted to the correct place
Signed-off-by: Marta Lofstedt <marta.lofstedt@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
chipset support
If the machine doesn't support PSR, it will return -ENODEV from
i915_edp_psr_status, which we want to interpret as unsupported.
This is in line with what Chris just fixed for FBC.
V2: Copy-pasted to the correct position.
Signed-off-by: Marta Lofstedt <marta.lofstedt@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
My bad,
perf_pmu.c: In function ‘accuracy’:
perf_pmu.c:1533:4: warning: format ‘%lu’ expects argument of type ‘long unsigned int’, but argument 5 has type ‘uint64_t’ [-Wformat]
perf_pmu.c:1533:4: warning: format ‘%lu’ expects argument of type ‘long unsigned int’, but argument 6 has type ‘uint64_t’ [-Wformat]
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
|
|
Apollolake machine in the shards cannot bring the CPU0 back online so
skip the test on all Broxtons for now.
v2: Fix inverted check.
v3: igt_skip_on. (Chris Wilson)
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Acked-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
Convert the busy pwm from using a single calibration pass with a fixed
target into a self-correcting pwm that tries to adjust how long to sleep
on each pwm in order to converge at the target busy %%.
Being self-correcting, it should fare better against the more variable
systems CI presents.
v2: Be fair and equally strict for low/high busy %%
v3: target_idle_us and calculate expected from timing of each individual pass
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105157
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>
|
|
If the machine doesn't support FBC, it will return -ENODEV from
i915_fbc_info, which we want to interpret as unsupported.
Reported-by: Marta Lofstedt<marta.lofstedt@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Marta Lofstedt<marta.lofstedt@intel.com>
|
|
A new context assumes that all of its registers are in the default state
when it is created. What may happen is that a register written by one
context may leak into the second, causing mass confusion.
v2: Extend back to Sandybridge (etc)
v3: Check context preserves registers across suspend/hibernate and resets.
v4: Complete the remapping onto the new class:instance
v5: Not like that, like this, try again to use class:instance
v6: Prepare for retrospective gen4 contexts!
v7: Repaint register set name to nonpriv, as this is what bspec calls the
registers that are writable by userspace.
v8: Fix a typo for LRM on gen8
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
|
|
We know the OA unit might skip some reports from time to time (reasons
include pressure on memory controller, power management, ...). So
rather than checking that the time between periodic reports is about
the period we asked for, let's verify that the time is always longer
or equal to the period we've asked for.
We still have to leave some room for errors. Here is dump of an error
in this updated test :
(perf:405) DEBUG: report0019 ts=e217de20 hw_id=0x00000014 delta=64
(perf:405) DEBUG: report0020 ts=e217de60 hw_id=0x00000014 delta=64
(perf:405) DEBUG: report0021 ts=e217dea0 hw_id=0x00000014 delta=64
(perf:405) DEBUG: report0022 ts=e217df66 hw_id=0x00000014 delta=198 ******
(perf:405) DEBUG: report0023 ts=e217dfa0 hw_id=0x00000014 delta=58 ******
(perf:405) DEBUG: report0024 ts=e217dfe0 hw_id=0x00000014 delta=64
(perf:405) DEBUG: report0025 ts=e217e020 hw_id=0x00000014 delta=64
(perf:405) DEBUG: report0026 ts=e217e060 hw_id=0x00000014 delta=64
As you can see there is a discrepency in the periodic reports. I have
no explanation for it. This isn't a programming error since the same
context has correct periods before and after, so it must be some kind
of hardware glitch/corner-case that hasn't be been documented.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
|
|
Ensure that we always use every context at least once before we start
running the stress-test.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
|
|
One weird issue we see in bug 104676 is that the hangs are too fast on
HSW! So force the use of the slow spinners that do not try to trigger
a hang by injecting random bytes into the batch.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=104676
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
|
|
s/16/MAX_ELSP_QLEN/ as appropriate
v2: Use ARRAY_SIZE for loop bounds over fixed size arrays
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
|
|
It is useful to dump the request layout between engines help debug
ordering issues and stuck preemption, so add it to preempt_other().
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
|
|
icl offers a much reduced context space, and in its simplest setup we
cannot allocate one context per priority level, so trim the number and
reuse the same context for multiple priority requests.
v2: Bump the MAX to 1024 (still lower than the ~4096 previously in use)
v3: Also limit NCTX to MAX_CONTEXTS for wide-*
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Antonio Argenziano <antonio.argenziano@intel.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
|
