This is also encouraged by another wa, Wa_14018813551.
Both workarounds state that StackIDControlOverride_RTGlobals should
always be set to 0 (i.e. 2k).
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30937>
NIR discard_if expects bools, therefore we end with extr CMP in
the end, however we can handle floats just fine (only with the
reversed logic, so it triggers for src < 0).
Shader-db RV530:
total instructions in shared programs: 120787 -> 120687 (-0.08%)
instructions in affected programs: 5559 -> 5459 (-1.80%)
helped: 102
HURT: 8
total temps in shared programs: 16770 -> 16764 (-0.04%)
temps in affected programs: 151 -> 145 (-3.97%)
helped: 8
HURT: 5
total cycles in shared programs: 181520 -> 181444 (-0.04%)
cycles in affected programs: 10932 -> 10856 (-0.70%)
helped: 101
HURT: 12
Signed-off-by: Pavel Ondračka <pavel.ondracka@gmail.com>
Reviewed-by: Filip Gawin <filip@gawin.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30967>
Always use MAX instead of MAD for MOV and we use RM_OMOD_DISABLE
to prevent flushing denormals for MIN/MAX/CMP/CND and prevent
-NaN -> NaN conversion.
Inspired by patch from Filip Gawin.
Signed-off-by: Pavel Ondračka <pavel.ondracka@gmail.com>
Reviewed-by: Filip Gawin <filip@gawin.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30967>
Just emit fake ADD + 0 instead, because we want to emit MAX for MOV in
the next commit and also use RC_OMOD_DISABLE so that we preserve denormals
and correctly handle negative NaNs. Using ADD here will mean we get
RC_OMOD_MUL_1 later which will convert -0 to 0 and therefore KIL will
work correctly.
Signed-off-by: Pavel Ondračka <pavel.ondracka@gmail.com>
Reviewed-by: Filip Gawin <filip@gawin.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30967>
At i945_texture_layout_3d() util_next_power_of_two() is called,
which oversizes the npot-blocks for every level to get power of 2
for width and height. Hardware doesnot expect these oversized
npot-blocks. The call is removed.
Code is added to align allocation of npot-blocks correctly.
Also at i915_texture_layout_3d() the util_next_power_of_two() call
is removed. Besides the computation of block-allocation is changed,
because it still was using classic i915-coding.
Cc: mesa-stable
Signed-off-by: GKraats <vd.kraats@hccnet.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31057>
The OpenCL target defined its own `dep_clang`, which was rather
incomplete, when compared to the version from the main `meson.build`
file.
This commit removes the custom version and relies on the main version of
`dep_clang` in the root level `meson.build`.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/11945
Signed-off-by: Kai Wasserbäch <kai@dev.carbon-project.org>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31364>
Enable build testing for zink and llvmpipe in the debian-no-libdrm job.
EGL is still excluded due to additional dependencies on libdrm.
Signed-off-by: Valentine Burley <valentine.burley@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31430>
Guard the usage of DRM_PRIME_CAP_IMPORT and DRM_PRIME_CAP_EXPORT behind HAVE_LIBDRM.
Fixes: c20d6a9197 ("egl: make dmabuf/modifier display bools a bit more precise/intelligible")
Reviewed-by: Eric Engestrom <eric@igalia.com>
Signed-off-by: Valentine Burley <valentine.burley@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31430>
We are dangerously close to begin with, but having some failures makes
it go over the 20 minutes timeout by ~30s. Let's add 50% to be safe.
Signed-off-by: Martin Roukala (né Peres) <martin.roukala@mupuf.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31415>
X11 starts up if Mesa is compiled with legacy-x11=dri2 and shows
the desktop.
It crashes with segfault as soon as an application is started.
At src/egl/drivers/dri2/platform_x11.c function dri2_from_names
(at src/gallium/frontends/dri/dri2.c) should be called
with offset address instead of with offset value.
Signed-off-by: GKraats <vd.kraats@hccnet.nl>
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31406>
The `.common-rules` need to be executed first, before all the "does this
file exist? is it modified?" logic in the farm rules (and in
api/driver/etc. rules after).
The custom override in windows jobs was bypassing this, resulting in
windows container jobs something being missing, breaking the pipelines.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31390>
Our usage of vk_error to expound on error messages returned by drivers
probably falls more into the warning category.
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31292>
This gives us the possibility of describing other GOB formats in the
future. This also cleans things up a bit as it replaces all of the uses
of GOB_WIDTH/DEPTH with an extent which we can change in the future.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31410>
Take advantage of 3 spare JSL in Collabora lab to load the balance of
those jobs:
job name avg duation (min)
Signed-off-by: Guilherme Gallo <guilherme.gallo@collabora.com>
--- ---
anv-jsl 15
anv-jsl-angle 20
iris-jsl-deqp 18
Signed-off-by: Guilherme Gallo <guilherme.gallo@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31414>
As we don't have capacity for more parallelism atm, increase the
fraction of the CTS for those jobs.
job name average min
Signed-off-by: Guilherme Gallo <guilherme.gallo@collabora.com>
--- ---
radv-stoney-angle 16
radv-stoney-vkcts 15
Signed-off-by: Guilherme Gallo <guilherme.gallo@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31414>
It is taking 19 min on average to run this job.
As we don't have more capacity, introduce a fraction of 2 and create the
`full` version to be run on nightly pipelines.
Signed-off-by: Guilherme Gallo <guilherme.gallo@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31414>
By rebalancing the `a618_vk` job fraction from **2** to **3**, we have
freed up 2 `kingoftown` devices. These devices can now be reallocated to
the `a618_traces` and `a618_skqp` jobs to optimize resource utilization.
Signed-off-by: Guilherme Gallo <guilherme.gallo@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31414>
the original handling here was supposed to be more granular, but I
never got around to making the granularity work
now it should do something like:
* run through all the modifiers and check for one that supports expected usage
* do not ever remove mutable; this should always be supported for SOME modifier
* use LINEAR as a fallback (with mutable)
supporting mutable for srgb is the most important aspect of modifier selection
here since it avoids having to later reallocate the dmabuf
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31413>
