Zbigniew Kempczyński authored 2 years ago and unerlige committed 2 years ago

According to documentation and hw team suggestion use source pitch
and aux-ccs in destination object when inplace decompression is
performed (full-resolve operation).

Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
Cc: Karolina Drobnik <karolina.drobnik@intel.com>
Reviewed-by: Karolina Drobnik <karolina.drobnik@intel.com>

Mark Yacoub authored 2 years ago and unerlige committed 2 years ago

[Why]
MST connectors can change connector IDs between plugs. Getting the EDID
prop requires a valid connector ID.

[How]
Get Connector AFTER a plug, just before it's being used.

Fixes: kms_chamelium@dp-edid-read

Signed-off-by: Mark Yacoub <markyacoub@chromium.org>
Reviewed-by: Petri Latvala <petri.latvala@intel.com>

On i915 devices:
kmstest_get_vblank() calls a drm_ioctl(DRM_IOCTL_WAIT_VBLANK), but
vblanks are not updated until a following read() op triggers kernel to
update vblank, ex. "[drm:drm_handle_vblank [drm]] vblank event on xxx".

However, this is not the case for all devices:
kmstest_get_vblank() triggers a drm_ioctl(DRM_IOCTL_WAIT_VBLANK),
and then drm_wait_vblank_ioctl() in drm_vblank.c is called and
returns vblank, ex. "[drm]] crtc 0 returning xxx to client" right away.
In this scenario, the read() op does nothing related to vblank.

Since the vblank can be updated before the read(), the extra vblank
in commit da0b6abeshould also be allowed before the read().

Signed-off-by: Alex Hung <alex.hung@amd.com>
Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>

Decode a bunch of newish child dev fields:
- disable_compression_for_ext_disp
- lttpr_non_transparent
- dp_port_trace_length
- tbt

Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>

Stop decoding most child dev fields if the BDB version
is too old.

Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>

Dump out block 2 contents in the same (big endianish)
order as it's listed in the spec. Also group each bitfield
into its own little block. Makes it a bit easier to cross
check between the two.

Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>

Decode the i2c speed, just for kicks.

Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>

Sync up the the VBT definitions from kernel commit
758b018aa283 ("drm/i915/bios: Add the "Disable compression
for the Display Port/HDMI external display" bit")
and adjust the actual code to match.

Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>

Mark Yacoub authored 2 years ago and Mark Yacoub committed 2 years ago

[Why]
When getting a connector for a port, the connector ID might be invalid
after a replug if it's an MST connector. Getting the connector with just
the connector ID will not always work.

[How]
Check if the connector is MST by checking the port's connector path.
If it is, iterate through all connectors until a connector with a
matching path is found.

TEST=./kms_chamelium --run-subtest dp-hpd [on intel Volteer board with
Chamelium V3 connected]

Signed-off-by: Mark Yacoub <markyacoub@chromium.org>
Reviewed-by: Lyude Paul <lyude@redhat.com>

Riana Tauro authored 2 years ago and Ashutosh Dixit committed 2 years ago

Remove unused rapl functions from igt_power.
No functional changes

Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>

Riana Tauro authored 2 years ago and Ashutosh Dixit committed 2 years ago

igt_power library reads energy using hwmon interface for dgfx and
rapl otherwise. Modify gem_exec_whisper test to use igt_power interface.

Cc: Ashutosh Dixit <ashutosh.dixit@intel.com>
Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>

Riana Tauro authored 2 years ago and Ashutosh Dixit committed 2 years ago

igt_power library reads energy using hwmon interface for dgfx and
rapl otherwise. Modify semaphore-power test to use igt_power interface.

Cc: Ashutosh Dixit <ashutosh.dixit@intel.com>
Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>

Riana Tauro authored 2 years ago and Ashutosh Dixit committed 2 years ago

igt_power library reads energy using hwmon interface for dgfx and
rapl otherwise. Modify rc6-idle and rc6-fence to use igt_power interface.

Cc: Ashutosh Dixit <ashutosh.dixit@intel.com>
Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>

Riana Tauro authored 2 years ago and Ashutosh Dixit committed 2 years ago

Modify igt_power to expose functions to read energy, power
using hwmon and rapl interface

Cc: Ashutosh Dixit <ashutosh.dixit@intel.com>
Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>

Riana Tauro authored 2 years ago and Ashutosh Dixit committed 2 years ago

Rename igt_rapl library to igt_power.
No functional changes

Cc: Ashutosh Dixit <ashutosh.dixit@intel.com>
Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>

Riana Tauro authored 2 years ago and Ashutosh Dixit committed 2 years ago

igt_hwmon exposes methods to open hwmon directories identified by name

v2: Add license(Petri)

Cc: Ashutosh Dixit <ashutosh.dixit@intel.com>
Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>

amdgpu rejects when crtc is on + primary plane is off. As a result,
skipping these scenario when they cannot be turned off together, i.e.
crtc cannot be off when reading crc values.

Signed-off-by: Alex Hung <alex.hung@amd.com>
Reviewed-by: Mark Yacoub <markyacoub@chromium.org>

GPU drivers can reject when crtc is on and primary plane is off,
so turn crtc off with primary plane and turn it on afterwards.

Signed-off-by: Alex Hung <alex.hung@amd.com>
Reviewed-by: Mark Yacoub <markyacoub@chromium.org>

on creating reference images for crcs re-render only parts
which have changed.

Signed-off-by: Juha-Pekka Heikkila <juhapekka.heikkila@gmail.com>
Reviewed-by: Vinod Govindapillai <vinod.govindapillai@intel.com>

on async flip tests there's need to create large solid color
framebuffer. Here is optimized creating this framebuffer
with help of rendercopy.

Signed-off-by: Juha-Pekka Heikkila <juhapekka.heikkila@gmail.com>
Reviewed-by: Vinod Govindapillai <vinod.govindapillai@intel.com>

Skip the tests for cursor sizes that are not supported by the hardware

Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Reviewed-by: Juha-Pekka Heikkila <juhapekka.heikkila@gmail.com>

amdgpu rejects when crtc is on + all planes are off, and it
is necessary to turn off crtc when all planes are off in the
subtest "tiling-none".

This is revised from https://patchwork.freedesktop.org/series/80904/



Signed-off-by: Alex Hung <alex.hung@amd.com>
Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>

legacy format addfb test try to fuzz kernel with random parameters,
these random parameters cause logging into dmesg. Limit fuzzing
rounds to maximum of 10k (or 1 second) to limit size of created log.

Fixes: https://gitlab.freedesktop.org/drm/intel/-/issues/7017


Signed-off-by: Juha-Pekka Heikkila <juhapekka.heikkila@gmail.com>
Reviewed-by: Vinod Govindapillai <vinod.govindapillai@intel.com>

Jake Freeland authored 2 years ago and Kamil Konieczny committed 2 years ago

The `#define sigev_notify_thread_id _sigev_un._tid` macro differs on
FreeBSD. This patch uses ifdefs to stop the compiler from overriding
FreeBSD's existing definition.

Signed-off-by: Jake Freeland <jfree@freebsd.org>
Reviewed-by: Kamil Konieczny <kamil.konieczny@linux.intel.com>

Jake Freeland authored 2 years ago and Kamil Konieczny committed 2 years ago

In FreeBSD, the ptrace() function cannot take a variable
length list of arguments. This patch adds NULL in the @data
parameter to allow for compilation on FreeBSD systems.

Signed-off-by: Jake Freeland <jfree@freebsd.org>
Reviewed-by: Kamil Konieczny <kamil.konieczny@linux.intel.com>

Riana Tauro authored 2 years ago and Kamil Konieczny committed 2 years ago

igt_devices_scan returns with empty list if it is called after
anyone calls igt_devices_free. Fixed this with clearing
igt_devs.devs_scanned inside igt_devices_free. Also while at
this, fix memory leak in it.

Remove redudant code from igt_devices_scan and replace it with
call to igt_devices_free

v2: optimize code in igt_devices_scan (Zbigniew)
v3: update commit message (Janusz)
v4: [Kamil] re-worded commit message and commit description

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

As we are intended to test few cases on internal panels separately,
fix the logic to execute such tests on internal panels only.

'TRANSITION_MODESET_FAST' -> should be only for internal panels
'TRANSITION_MODESET' -> should be only for external panels

Fixes: 2527ce52 (tests/kms_atomic_transition: Convert tests to dynamic)
Cc: Karthik B S <karthik.b.s@intel.com>
Signed-off-by: Bhanuprakash Modem <bhanuprakash.modem@intel.com>
Reviewed-by: Karthik B S <karthik.b.s@intel.com>

DG1, as expected only has a fixed sized bar, so attempting to force the
BAR size should fail. Update the test to account for such devices, and
perform a skip, instead of failing.

Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/6458


Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Nirmoy Das <nirmoy.das@intel.com>
Tested-by: Nirmoy Das <nirmoy.das@intel.com>
Reviewed-by: Nirmoy Das <nirmoy.das@intel.com>

Jake Freeland authored 2 years ago and Petri Latvala committed 2 years ago

Allow benchmark compilation on a FreeBSD system using
the drop-in igt_freebsd header.

Signed-off-by: Jake Freeland <jfree@freebsd.org>
Acked-by: Petri Latvala <petri.latvala@intel.com>

Jake Freeland authored 2 years ago and Petri Latvala committed 2 years ago

FreeBSD uses iic instead of i2c for I/O management.
This patch includes the FreeBSD iic header and uses
a macro substitution to convert the i2c addr member
into the iic slave member.

Signed-off-by: Jake Freeland <jfree@freebsd.org>
Acked-by: Petri Latvala <petri.latvala@intel.com>

Jake Freeland authored 2 years ago and Petri Latvala committed 2 years ago

Use sysctlbyname() to gather RAM and swap information for
any FreeBSD system using igt.

Signed-off-by: Jake Freeland <jfree@freebsd.org>
Acked-by: Petri Latvala <petri.latvala@intel.com>

Jake Freeland authored 2 years ago and Petri Latvala committed 2 years ago

Both Linux and FreeBSD have a <sys/syscall.h>, but some
operating systems do not. This patch opens up syscall.h
access to FreeBSD through using #ifdef __FreeBSD__

Signed-off-by: Jake Freeland <jfree@freebsd.org>
Acked-by: Petri Latvala <petri.latvala@intel.com>

Jake Freeland authored 2 years ago and Petri Latvala committed 2 years ago

Use #ifdef __linux__ to limit all Linux-specific headers to
Linux systems

Signed-off-by: Jake Freeland <jfree@freebsd.org>
Acked-by: Petri Latvala <petri.latvala@intel.com>

Jake Freeland authored 2 years ago and Petri Latvala committed 2 years ago

This header attempts to be a simple drop-in to enable FreeBSD
support in a given source file. It includes macro substitutions
that replace Linux-specific phrases with their FreeBSD equivalents.

Signed-off-by: Jake Freeland <jfree@freebsd.org>
Acked-by: Petri Latvala <petri.latvala@intel.com>

MDThasleem authored 2 years ago and Bhanuprakash Modem committed 2 years ago

Sanitize the system state before starting the subtest.

v2: Created a new functions for each subtest instead
    of managing the full logic within the igt_subtest().
v3: Minor changes.
v4: Replaced igt_assert_f with igt_require_f in prune-stale-modes.
v5: Added goto statement if connector_id found true.
v6: Minor changes.
v7: Added func pointer to subtests.

Signed-off-by: Mohammed Thasleem <mohammed.thasleem@intel.com>

Jake Freeland authored 2 years ago and Kamil Konieczny committed 2 years ago

In subtest_leak(), mmap() is called with the flag PROT_WRITE,
but no PROT_READ. Later in the function, the mapped memory is
read using `children[i]`. In FreeBSD, the lack of PROT_READ
causes SIGSEGV. Adding the PROT_READ flag to the mmap() call
fixes this.

Signed-off-by: Jake Freeland <jfree@freebsd.org>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Kamil Konieczny <kamil.konieczny@linux.intel.com>

Nirmoy Das authored 2 years ago and Kamil Konieczny committed 2 years ago

After each subtest gpu should be idle or else busy
engine leaks into next subtest.

References: https://gitlab.freedesktop.org/drm/intel/-/issues/5888


Suggested-by: Chris Wilson <chris.p.wilson@intel.com>
Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
Reviewed-by: Kamil Konieczny <kamil.konieczny@linux.intel.com>

Rahul Janga authored 2 years ago and Kamil Konieczny committed 2 years ago

Add test description to all the available subtests.

Signed-off-by: Janga Rahul Kumar <janga.rahul.kumar@intel.com>
Reviewed-by: Kamil Konieczny <kamil.konieczny@linux.intel.com>

Jeevan B authored 2 years ago and KunalJoshi95 committed 2 years ago

Converting the existing subtests to dynamic subtests.
Adding support for each_pipe, currently PSR2 is supported on
few pipes so breaking the loop for other pipes currently.

v2: Add commit message and change break to continue.
v3: Check for pipe supporting PSR2 or not and then add to dynamic subtest.
v4: Check PSR2 support once in fixture then iterate pipes.
v5: Increase array size for pipes and outputs.

Signed-off-by: Jeevan B <jeevan.b@intel.com>
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>

Jeevan B authored 2 years ago and KunalJoshi95 committed 2 years ago

Converting the existing subtests to dynamic subtests.
Adding support for each_pipe, currently PSR2 is supported on
few pipes so breaking the loop for other pipes currently.

v2: Check for pipe supporting PSR2 or not and then add to dynamic subtest.
v3: Check PSR2 support once in fixture then iterate pipes.
v4: Add output part into dynamic name.
v5: Add the missing code and increase pipes and outputs array size.

Signed-off-by: Jeevan B <jeevan.b@intel.com>
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>