this is a great memory-saving optimization for drivers that support it,
but for drivesr that can't, it ends up taking slower paths instead
of just consuming more memory
but the year is 2022. we have all the memory. so use more memory if signed
offsets aren't supported
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18856>
Previously both destination and source were floats, so no casting was
performed, but with 7664c85b1d source register was reinterpreted as
unsigned integer, so MOV started casting that integer to float.
Fixes: 7664c85b1d ("intel/compiler: Create and use struct for TASK and MESH thread payloads")
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18886>
All virgl-iris-traces tests are passing now after the crosvm revup that
fixed crosvm crash and after updating test checksums. Enable auto-testing
for the virgl-iris-traces pipeline stage.
Reviewed-by: Corentin Noël <corentin.noel@collabora.com>
Signed-off-by: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18693>
Few virgl-iris-traces tests fail because resultant images now have few
pixels that changed. The actual reason of the change is unknown, we
assume it happened due to a new optimizations. Update the virgl-iris-traces
checksums.
Reviewed-by: Corentin Noël <corentin.noel@collabora.com>
Signed-off-by: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18693>
VirglRenderer::export_blob doesn't need to handle opaque fd. That will
let rutabaga's export_blob to fail and fallback directly to the external
mapping path desired.
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18693>
If a library with only the vertex input interface is created, this
would crash.
This fixes segfault with vkoverhead.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18869>
attribute_mask was unused when I first introduced this but now it's
used again.
This fixes a bunch of GPL regressions.
Fixes: 0feab7b9cf ("radv: prepare the VS input state for prologs created with GPL")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18862>
Bind PIPE_CAP_TEXTURE_SHADOW_LOD with its value so that we are not using
non-existing GLSL on host side.
Rename VIRGL_CAP_FAKE_FP64 into VIRGL_CAP_HOST_IS_GLES
Signed-off-by: Corentin Noël <corentin.noel@collabora.com>
Reviewed-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18296>
The driver no longer bind NULL graphics pipelines, so these checks
are useless.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18873>
I missed this when enabling pipeline libraries, and we were also setting
this to the wrong thing. Previously we were using rasterization state
when parsing depth/stencil indirectly via builder->depth_clip_disable,
which is not allowed with pipeline libraries. Fixing this is a bit
painful because now RB_DEPTH_CNTL can depend on state from both the
fragment shader library and the pre-rasterization library, in addition
to being disabled via output interface state.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18861>
When the array length is a calculation, the C version of the calculation
is provided in altlen while len often contains LaTeX that we can't do
anything with. Use altlen when available. Also, while we're here, wrap
array lengths in parentheses in case they contain math.
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18878>
This flag indicates whether or not the legacy scanout flag is supported.
It defaults to true since that has been the default assumption for the
WSI code up until now.
On NVIDIA hardware, we can't render to linear so, if we don't have
modifiers, we want to automatically fall back to the blit path. In
theory, we could do this inside the driver but it's a giant pain and
much harder to ensure that the blit only happens as part of
vkQueuePresent().
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18826>
