| Age | Commit message (Collapse) | Author |
|---|
|
Make the property validation more thorough:
- validate property flags
- make sure there's an expected number of values/enums
- make sure the possible values make sense
- make sure the current value makes sense
- actually iterate through all planes/crtc/connectors to
check their properties
- make sure encoders don't expose properties while at it
- check that atomic props aren't exposed to non-atomic clients
Still passes on my ivb. Not tested anything else so far.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com>
|
|
Batch functions were copy/pasted across several libs.
With moving it into intel_batchbuffer lib test can now be
easly maintained without worrying that we forgot to modify
older version of lib.
v2: Added documentation into lib and rebased patch
v3: Fixed typos and rebased patch
v4: Fixed documentation issues
v5: Rename, clean up of leftovers from previous version
and documentation polishing
v6: Fixing assert
Signed-off-by: Lukasz Kalamarz <lukasz.kalamarz@intel.com>
Cc: Katarzyna Dec <katarzyna.dec@intel.com>
Cc: Radoslaw Szwichtenberg <radoslaw.szwichtenberg@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Katarzyna Dec <katarzyna.dec@intel.com>
|
|
structure
With a removal of all references to state variable there is
no longer any need to keep it in structure.
v5: Rebase
Signed-off-by: Lukasz Kalamarz <lukasz.kalamarz@intel.com>
Cc: Katarzyna Dec <katarzyna.dec@intel.com>
Cc: Radoslaw Szwichtenberg <radoslaw.szwichtenberg@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Katarzyna Dec <katarzyna.dec@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
|
|
This patch apply coding standard into rendercopy_gen7 library.
No functional changes were made.
v5: Fixed wrong indent
Signed-off-by: Lukasz Kalamarz <lukasz.kalamarz@intel.com>
Cc: Katarzyna Dec <katarzyna.dec@intel.com>
Cc: Radoslaw Szwichtenberg <radoslaw.szwichtenberg@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Katarzyna Dec <katarzyna.dec@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
|
|
This lib was written in a different manner than all other libs,
which was causing some issues during refactoring. Previous
implementation was using two pointers (ptr and state) to
access two parts of batchbuffer in the same time. Current
implementation is reusing ptr with a downside of needing
to ensure the state/command ordering is proper.
v3: Modified commit message and added batch boundary check
v5: Rebase
Signed-off-by: Lukasz Kalamarz <lukasz.kalamarz@intel.com>
Cc: Katarzyna Dec <katarzyna.dec@intel.com>
Cc: Radoslaw Szwichtenberg <radoslaw.szwichtenberg@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Katarzyna Dec <katarzyna.dec@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
|
|
This patch applies coding standard into gen6_render library.
No functional changes were made.
Signed-off-by: Lukasz Kalamarz <lukasz.kalamarz@intel.com>
Cc: Katarzyna Dec <katarzyna.dec@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Katarzyna Dec <katarzyna.dec@intel.com>
|
|
This patch is reordering registers and related shifts/fields
to be in one place and also is sorting registers definitions
in an ascending order by it memory address.
v4: Fixed missing values and corrected commit message
Signed-off-by: Lukasz Kalamarz <lukasz.kalamarz@intel.com>
Cc: Katarzyna Dec <katarzyna.dec@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Katarzyna Dec <katarzyna.dec@intel.com>
|
|
Remove duplicated registers, shifts and values from lib.
Signed-off-by: Lukasz Kalamarz <lukasz.kalamarz@intel.com>
Cc: Katarzyna Dec <katarzyna.dec@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Katarzyna Dec <katarzyna.dec@intel.com>
|
|
Media fill function can be reused on Gen10. Code has not
changed. Let's reuse gen9.
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Katarzyna Dec <katarzyna.dec@intel.com>
Reviewed-by: Radoslaw Szwichtenberg <radoslaw.szwichtenberg@intel.com>
Reviewed-by: Lukasz Kalamarz <lukasz.kalamarz@intel.com>
|
|
While I am making changes in gpgpu and media fill area let's
adjust code to our coding style.
v2: rebased on series new version (patch is now last from
series so change seems larger)
v3: rebased
Signed-off-by: Katarzyna Dec <katarzyna.dec@intel.com>
Cc: Lukasz Kalamarz <lukasz.kalamarz@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Radoslaw Szwichtenberg <radoslaw.szwichtenberg@intel.com>
|
|
There are missing parameters for Gen8 configuration of gpgpu_fill
that are causing GPU hangs on newer hardware. We need to set the
number of threads in TG in gen8_fill_interface_descriptor. This
field was omitted (apparently without any side effects), but
according to bspec from BDW this field cannot be set to 0. We also
need to use pipeline selection mask to gen9_gpgpu_fillfunc, which
is necessary from SKL.
v2: rebased on refactored library
v3: Removed replacing gen7_emit_interface_descriptor_load with gen8
version in gen9_gpgpgu_fillfunc, because during refactoring gen8
function was removed.
v4: rebase on series new version
Signed-off-by: Katarzyna Dec <katarzyna.dec@intel.com>
Cc: Lukasz Kalamarz <lukasz.kalamarz@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Radoslaw Szwichtenberg <radoslaw.szwichtenberg@intel.com>
|
|
After moving all functions needed for gpgpu and media fill testing
there is a lot of duplications which can be removed:
Library media_fill_gen8 and media_fill_gen8lp for CHT was removed,
media state flush for !CHT was added to gen7_emit_media_objects.
Many gen8 functions were replaced with gen7 version with devid
parameter (gen7_fill_curbe_load, gen7_emit_interface_descriptor,
gen7_fill_binding_table, gen7_emit_media_objects). Unified fill kernel
function so it is applicable to all gens and both media and gpgpu
(merged gen7_fill_media_kernel and gen8_fill_media_kernel).
Duplicated constants like GEN8_MEDIA_VFE_STATE, GEN8_MEDIA_CURBE_LOAD,
GEN8_MEDIA_INTERFACE_DESCRIPTOR_LOAD, GEN8_MEDIA_OBJECT were
replaced by GEN7 version. However this constants were not removed
from gen8_media.h library, because they are used by other tests
for Gen8+. More refactoring in this gen*_media.h libraries is needed.
It seems that further unification of *_fillfunc functions will
introduce more confusion in understanding what the tests are doing
and what were changes between Gens.
v2: Moved some reduntant changes from Move gpgpu/media fill to gpu_fill...
to this patch. Applied comments from review.
v3: rebase
Signed-off-by: Katarzyna Dec <katarzyna.dec@intel.com>
Cc: Lukasz Kalamarz <lukasz.kalamarz@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Radoslaw Szwichtenberg <radoslaw.szwichtenberg@intel.com>
|
|
Gpgpu_fill and media_fill libraries are very similar and many
functions can be shared. I have created library gpu_fill with
all functions needed for implementing gpgpu_fill and media_fill
tests for all Gens. For reviewing and debugging purposes this patch
should be only moving functions from few libraries to one removing
functions identical for both media and gpgpu.
Places in the code that required more changes:
Removing gen7_fill_gpgpu_kernel function that is identical to
gen7_fill_media_kernel and introduces conflict with moving
genX_fill_interface_descriptor, which are the same for media and gpgpu.
Function gen8_fill_media_kernel is not removed in this patch
(although it is identical with gen7 version), because this patch
should be as much as possible functions movement.
gen8_fill_interface_descriptor was unified for media and gpgpu
by adding kernel and its size as a parameter (this parameters
were missing in media gen8, gen8lp and gen9 functions)
gen8_emit_state_base_address was unified, the one for gpgpu was
configured like it would be using indirect state (while we are
using CURBE). I have checked that media fill version
(OUT_BATCH(0 | BASE_ADDRESS_MODIFY)) works fine on gpgpu gen8 and newer.
v2: Changed code layout. GenX_fill_media_kernel was identical to
genX_fill_gpgpu_kernel so this function was unified to
gen7_fill_kernel. There were 2 very similar functions
gen8_emit_state_base_address for media and gpgpu, where the one
for gpgpu was configured like it would be using indirect state
(while we are using CURBE). I have checked if media fill version
works fine in gpgpu test on Gen8 and unified them.
v3: Made patch easier for reviewing moving changes unifying code for
various gens (that were included v1) to other patch, leaving only
the most critical code changes.
v5: Added copyrights and #ifndef to gpu_fill.h
v5: Added copyrights to gpu_fill.c
Signed-off-by: Katarzyna Dec <katarzyna.dec@intel.com>
Cc: Lukasz Kalamarz <lukasz.kalamarz@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Reviewed-by: Radoslaw Szwichtenberg <radoslaw.szwichtenberg@intel.com>
|
|
This will fix the PSR tests to fail slightly faster, since they wait
indefinitely for a CRC that never comes during open.
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
[mlankhorst: Increase timeout to 10 seconds for HSW CRC w/a.]
Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
|
|
When we set the test to its first state, we will clear FIFO underrun
status through the debugfs, so that any underruns caused by the tests
will not affect other tests. In particular this also resets the FBC
FIFO underrun status, so previous failures won't affect other tests
that depend on FBC.
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105681
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
|
|
Pretend that outputs without modes are disabled, and try reprobing
if no modes are initially detected, just in case that the reprobe
will find modes.
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105798
Reviewed-by: Marta Lofstedt <marta.lofstedt@intel.com>
|
|
All too frequently, we fail our memory checks to a leak in the driver.
While we give every opportunity for the driver to release the memory
before we fail, if we do dump the meminfo and slabinfo, if available, so
we can assign blame^W^W resolve the problem quickly.
References: https://bugs.freedesktop.org/show_bug.cgi?id=105967
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Martin Peres <martin.peres@linux.intel.com>
Cc: Tomi Sarvela <tomi.p.sarvela@intel.com>
Reviewed-by: Antonio Argenziano <antonio.argenziano@intel.com>
|
|
gen4/5 require a DRM_MASTER to use MI_STORE_DW, make it so.
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>
|
|
Contexts executing when reset triggers are potentialy corrupt so trying to
use them from a subsequent test (like the default context) can hang the
GPU or even the driver.
Workaround that by always creating a dedicated context which will be
running when GPU reset happens.
v2:
* Export and use gem_reopen_device so the test works on old gens as well.
(Chris Wilson)
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
Wrong file descriptor was passed to the iterator. This had currently no
effect, since it wasn't used in the macro, but needs to be fixed.
At the same time make the macro consistent by checking for engine presence
like the other iterators do.
Added __for_each_engine_class_instance which does not check for engine
presence and so is useful for enumerating all possible engines - like for
instance for subtest enumeration.
And another 'wrong fd used' fixlet in the render node subtests.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reported-by: Michel Thierry <michel.thierry@intel.com>
Cc: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Michel Thierry <michel.thierry@intel.com>
|
|
Only do a full probe when the initial connector status is unknown, and
otherwise skip it. This should speed up kms tests ever so slightly.
Also set pipe to PIPE_NONE for igt_output_refresh(), to make the
connector info slightly more accurate.
Changes since v1:
- Instead of a force_reprobe tristate, only force reprobe from
igt_display_init() when an unknown connection is found.
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
Reviewed-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
|
|
Pollable spin batch exports a spin->running pointer which can be checked
by dereferencing it to see if the spinner is actually executing on the
GPU.
This is useful for tests which want to make sure they do not proceed with
their next step whilst the spinner is potentially only being processed by
the driver and not actually executing.
Pollable spinner can be created with igt_spin_batch_new_poll or
__igt_spin_batch_new_poll, after which igt_spin_busywait_until_running can
be used to busy wait until it is executing.
v2:
* Move READ_ONCE to igt_core.
* Add igt_spin_busywait_until_running. (Chris Wilson)
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
Some tests measure the render's ring size but are actually meant to
measure the smallest across all engines. This patch adds measuring the
smallest size in gem_measure_ring_size_inflight() given the appropriate
parameter.
v2:
- Only expose high level API. (Chris)
v3:
- Use ALL_ENGINES macro.
Suggested-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Antonio Argenziano <antonio.argenziano@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
Ignore the shennigans of the test surrounding the library call to
gem_test_engine() and focus on answering the query of whether the engine
exists and is operational.
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>
Tested-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
|
|
Decouple the prop name arrays from the actual enum values
by using named initializers. Should make it less likely that we
end up with the wrong property on account of getting the order
wrong.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Daniel Vetter <daniel@ffwll.ch> #irc
|
|
Make a bunch of string arrays const.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Daniel Vetter <daniel@ffwll.ch> #irc
|
|
igt_display_init() now unconditionally tries to get the plane IN_FORMATS
blob, which causes a hard assert failure if we don't have the property
present.
Check if it's there and return early if not.
Signed-off-by: Daniel Stone <daniels@collabora.com>
Fixes: 98f7614bd725 ("lib: Parse plane IN_FORMATS blobifiers into a nicer form")
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
|
|
In interfaces where a parameter allow to select an engine, we usually
use '-1' or '~0u' to select all engines. This patch replaces magic
numbers with a named constant.
Suggested-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Antonio Argenziano <antonio.argenziano@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
igt_get_single_output_for_pipe() will give a valid output for a pipe,
for_each_pipe_with_single_output will iterate over all pipes, and
will be called for each pipe with an output once.
Changes since v1:
- pipe = __outputs - __output returned a negative number for pipe,
correctly set it to __output - __outputs.
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
|
|
Maintain straight up arrays of format+modifier tuples for each plane,
and also collect up a similar device wide array. These will make it
easy to confirm whether each plane (or the whole device) supports a
specific format+modifier pair.
v2: s/igt_hweight64/igt_hweight/
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
Add the binary hamming weight helper to igt_aux.h.
v2: Add just the one macro that works for 64 and 32 bits (Chris)
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
Found when compiling igt with clang:
[284/819] Compiling C object 'tests/kms_crtc_background_color@exe/kms_crtc_background_color.c.o'.
../tests/kms_crtc_background_color.c:140:48: warning: implicit conversion from enumeration type
'enum igt_atomic_crtc_properties' to different enumeration type 'enum igt_atomic_connector_properties' [-Wenum-conversion]
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
|
|
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
|
|
Right now we define this only for big core skus and leave the gt field
to 0 to mean unknown.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
Taken from drm-tip :
commit 1e6aa7e55c28ecd842b8b4599e4273c2429ee061
Author: Jani Nikula <jani.nikula@intel.com>
Date: Tue Mar 6 12:41:55 2018 +0200
drm/i915/icl: do not save DDI A/E sharing bit for ICL
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Acked-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
|
|
valgrind complains we feed uninitialised stack into the CONTEXT_SETPARAM
ioctl. It is unused by the kernel, but valgrind doesn't know that and
it's easy enough to clear the struct to prevent the warning.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
|
|
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>
|
|
Rename kick_fbcon() into bind_fbcon() so that it's not so
confusing.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
Looks like unbinding the dummy con doesn't work on all machines. Instead
we have to bind fbcon (which is what we used to do before commit
d18fca7f6cf3 ("lib/sysfs: Fix fbcon rebind")). Since some machines need
the bind and others need the unbind let's do both. The most obvious
difference I observed between two machines that behave differently
was the order in which the console drivers were listed in sysfs.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
Some tests (the ones which call igt_setup_runtime_pm and
igt_pm_enable_audio_runtime_pm) change default system configuration and
never restore it.
The configured runtime suspend is aggressive and may influence behaviour
of subsequent tests, so it is better to restore to previous values on test
exit.
This way system behaviour, with regards to a random sequence of executed
tests, will be more consistent from one run to another.
v2: Read failure means no runtime pm support so don't assert on it.
v3: Install exit handler before the write to close the Ctrl^C race. (Imre)
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> # v2
Acked-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
|
|
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>
|
|
A couple of bugs inside the hang injector, the worst being that the
presumed_offset of the reloc didn't match the batch; so if the reloc was
skipped (as the presumed_offset matched the reloc offset), the batch
wasn't updated and so we may not have generated a hanging batch at all!
Secondly, the MI_BATCH_BUFFER_START was not correct for all gen.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.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>
|
|
gem_handle_to_libdrm_bo creates a new handle through the use of flink,
which makes kms_frontbuffer_tracking through a single run not complete
when it starts leaking 50MB objects. Add the missing bo_unref to kill
the duplicated handle and fix the leaks!
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
Unused since commit 1f53e0dcec7a ("meson: Don't install headers").
Suggested-by: Petri Latvala <petri.latvala@intel.com>
Signed-off-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
Acked-by: Petri Latvala <petri.latvala@intel.com>
|
|
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>
|
|
Some warnings still exist, but they are about missing descriptions for
structs enums and their members.
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
|
|
This gets rid of the following warnings:
lib/igt_kms.c:2688: warning: Symbol name not found at the start of the comment block.
lib/igt_kms.c:2738: warning: Symbol name not found at the start of the comment block.
lib/igt_kms.c:2788: warning: Symbol name not found at the start of the comment block.
lib/igt_kms.c:2812: warning: Parameter description for igt_pipe_obj_replace_prop_blob::pipe is not used from source code comment block.
lib/igt_kms.c:2812: warning: Parameter description for igt_pipe_obj_replace_prop_blob::pipe_obj is missing in source code comment block.
lib/igt_kms.c:3794: warning: Parameter description for igt_cleanup_hotplug::mon is missing in source code comment block.
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
|
|
Export the kmsg() function for use by tests to write into the kernel
message log, useful for tests to inline their progress with kernel error
messages.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com>
|
