Commit graph

216912 commits

Author SHA1 Message Date
Marek Olšák
ceb2667cf3 gallium/util: print task/mesh statistics in util_end_pipestat_query
Used for debugging. Not used by anything in the tree.

Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39253>
2026-01-12 08:51:15 +00:00
Yiwei Zhang
b41f466045 panvk: fix to defer disk cache init after vk_physical_device_init
Some checks are pending
macOS-CI / macOS-CI (dri) (push) Waiting to run
macOS-CI / macOS-CI (xlib) (push) Waiting to run
Instead of initializing the disk cache earlier, the vk_properties
filling can be deferred after disk cache init is done. Otherwise, the
created disk cache will be zero'ed out in vk_physical_device_init,
ending up with leaked alloc and disabled disk cache (though advertised).

Fixes: acd00c07f6 ("panvk: Initialize the disk cache earlier")
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39248>
2026-01-12 08:34:46 +00:00
Samuel Pitoiset
6f4b3c7c9b ac/perfcounter: re-order GPU perf blocks on GFX11
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39199>
2026-01-12 08:10:33 +00:00
Samuel Pitoiset
653b39989a ac/perfcounter: define more GPU blocks on GFX11
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39199>
2026-01-12 08:10:33 +00:00
Samuel Pitoiset
d391cd0c4d ac/perfcounter: fix number of scoped instances for RMI block
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39199>
2026-01-12 08:10:33 +00:00
Samuel Pitoiset
ea63aa3e8e ac/perfcounter: add missing configuration for GCEA on GFX11
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39199>
2026-01-12 08:10:33 +00:00
Samuel Pitoiset
6722a6332a ac,radv,radeonsi: rename num_spm_counters to num_spm_modules
A module can have different number of counters.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39199>
2026-01-12 08:10:32 +00:00
Samuel Pitoiset
fb43d7bff2 ac/perfcounter: re-order GPU perf blocks on GFX12
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39199>
2026-01-12 08:10:31 +00:00
Samuel Pitoiset
3b6ff80d48 ac/perfcounter: define more GPU blocks on GFX12
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39199>
2026-01-12 08:10:31 +00:00
Samuel Pitoiset
eb37d6ceb7 ac/perfcounter: fix computing number of 16-bit/32-bit SPM counters
Determine them only when both are explicitly 0.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39199>
2026-01-12 08:10:30 +00:00
Samuel Pitoiset
d1efdc7e76 ac/perfcounter: fix number of 32-bit SPM counters
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39199>
2026-01-12 08:10:29 +00:00
Samuel Pitoiset
9fe57d3882 ac/spm: define new per-shader engine blocks
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39199>
2026-01-12 08:10:29 +00:00
Samuel Pitoiset
60fac38491 ac/spm: fix typo in one GPU perf block name
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39199>
2026-01-12 08:10:29 +00:00
Samuel Pitoiset
db02077c8a radv: remove extra instructions after UNREACHABLE
Minor cleanups.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39237>
2026-01-12 07:41:08 +00:00
Samuel Pitoiset
e1e2517664 radv: use UNREACHABLE for illegal texture filter
Found this with a broken CTS test, way easier to crash for isolating
the test case.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39237>
2026-01-12 07:41:08 +00:00
Samuel Pitoiset
91e0f8f1e5 radv/rt: fix a compilation warning about uninitialized fields
Just zero-initialize the layout struct to fix the following warning
because radv_use_bvh8() might return FALSE.

../src/amd/vulkan/radv_acceleration_structure.c: In function ‘radv_update_as_gfx12’:
../src/amd/vulkan/radv_acceleration_structure.c:873:70: warning: ‘layout.bounds_offsets’ may be used uninitialized [-Wmaybe-uninitialized]
  873 |       .bounds = state->build_info->scratchData.deviceAddress + layout.bounds_offsets,
      |                                                                ~~~~~~^~~~~~~~~~~~~~~
../src/amd/vulkan/radv_acceleration_structure.c:866:33: note: ‘layout.bounds_offsets’ was declared here
  866 |    struct update_scratch_layout layout;
      |                                 ^~~~~~

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39228>
2026-01-12 07:18:50 +00:00
Sergi Blanch Torne
548e195eb4 ci: disable Collabora's farm due to maintenance
Planned downtime in the farm:
* Start: 2026-01-12 08:00 UTC
* End: 2026-01-12 16:00 UTC

Signed-off-by: Sergi Blanch Torne <sergi.blanch.torne@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39044>
2026-01-12 07:02:30 +00:00
Dave Airlie
b4023532e3 lavapipe: drop unused defines
Some checks are pending
macOS-CI / macOS-CI (dri) (push) Waiting to run
macOS-CI / macOS-CI (xlib) (push) Waiting to run
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39255>
2026-01-11 20:48:01 +00:00
Dave Airlie
a82c68c439 lavapipe: drop data pointer from lvp_query
There is no need to store this.

Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39255>
2026-01-11 20:48:01 +00:00
Dave Airlie
6d9ad83f8f lavapipe: drop mem pointer and offset from buffer
these aren't used.

Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39255>
2026-01-11 20:48:01 +00:00
Dave Airlie
8b512692f5 lavapipe: repack render attachment.
just a trivial repack

Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39255>
2026-01-11 20:48:01 +00:00
Dave Airlie
e21d1bcea7 lavapipe: drop apiVersion from instance
this isn't used

Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39255>
2026-01-11 20:48:01 +00:00
Mary Guillemard
9a6c57b5c7 hk: Hash the multiview mask for both vertex and fragment stages
Some checks failed
macOS-CI / macOS-CI (dri) (push) Has been cancelled
macOS-CI / macOS-CI (xlib) (push) Has been cancelled
We were only accounting for the multiview mask when the fragment stage
was present causing wrong cached shaders to be possibly used.

This fixes
"dEQP-VK.renderpasses.dynamic_rendering.primary_cmd_buff.*_multiview"
failures.

Signed-off-by: Mary Guillemard <mary@mary.zone>
Cc: mesa-stable
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39238>
2026-01-10 17:42:26 +00:00
Alyssa Ross
f677b3eb48 rocket: fix building for musl
musl follows POSIX and provides ioctl as int ioctl(int, int, ...).

Fixes: 5b829658f7 ("rocket: Initial commit occf a driver for Rockchip's NPU")
Signed-off-by: Alyssa Ross <hi@alyssa.is>
Reviewed-by: Tomeu Vizoso <tomeu@tomeuvizoso.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38561>
2026-01-10 16:01:55 +00:00
Maíra Canal
df85401e8e docs/teflon: Update documentation with more recent output
Some checks are pending
macOS-CI / macOS-CI (dri) (push) Waiting to run
macOS-CI / macOS-CI (xlib) (push) Waiting to run
Signed-off-by: Maíra Canal <mairacanal@riseup.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38942>
2026-01-10 14:05:48 +00:00
Maíra Canal
9bf0ad3b15 docs/envvars: Document Teflon environment variables
Add documentation describing the use of Teflon's environment variables.

Signed-off-by: Maíra Canal <mairacanal@riseup.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38942>
2026-01-10 14:05:48 +00:00
Maíra Canal
7308ebd9cd teflon: List all supported operations on tflite_builtin_op_name()
When running ``Models.Op/yolox_114`` on Etnaviv with TEFLON_DEBUG=verbose,
the delegate currently reports supported operations as unknown, for example:

Teflon delegate: loaded etnaviv driver
idx    type ver support     inputs
=================================================================
  0 unknown v2  supported   in: 0(i8) 1(i32) out: 2(i8)

This happens because not all operations supported by Teflon are mapped
in ``tflite_builtin_op_name()``. Therefore, extend ``tflite_builtin_op_name()``
to include all operations supported, ensuring that the operation type
is reported correctly during debug.

Signed-off-by: Maíra Canal <mairacanal@riseup.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38942>
2026-01-10 14:05:48 +00:00
Maíra Canal
104d1a0d4b teflon: Improve dumped graph formatting
Currently, the graph dumped with ``TEFLON_DEBUG=verbose`` has the
following appearance:

idx type                      inputs                   outputs  operation type-specific
================================================================================================
  0 CONV   88,8,6 7
  1 DWCONV 7,35,34 33
  2 CONV   33,38,36 37
  3 DWCONV 37,41,40 39
  [...]
 20 CONV   9,14,12 13
 21 DWCONV 13,17,16 15
 22 CONV   15,20,18 19
 23 DWCONV 19,23,22 21
 24 CONV   21,26,24 25
 25 DWCONV 25,29,28 27
 26 CONV   27,32,30 31

Due to misaligned fields, the graph output is difficult to read.
Additionally, the "operation type-specific" field is always empty, as
its entries were removed in commit 986f8c7ff2 ("teflon: Support multiple
graph inputs and outputs").

Properly align the input and output fields and remove the
"operation type-specific" column. Also, widen the "type" field to
accommodate the longest currently supported operation name. The resulting
output looks like this:

idx type            inputs               outputs
==========================================================================
  0 CONV            88,8,6               7
  1 DWCONV          7,35,34              33
  2 CONV            33,38,36             37
  3 DWCONV          37,41,40             39
  [...]
 20 CONV            9,14,12              13
 21 DWCONV          13,17,16             15
 22 CONV            15,20,18             19
 23 DWCONV          19,23,22             21
 24 CONV            21,26,24             25
 25 DWCONV          25,29,28             27
 26 CONV            27,32,30             31

Signed-off-by: Maíra Canal <mairacanal@riseup.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38942>
2026-01-10 14:05:48 +00:00
Konstantin Seurer
077292f65b radv/bvh: Use box16 nodes when bvh8 is not used
Some checks are pending
macOS-CI / macOS-CI (dri) (push) Waiting to run
macOS-CI / macOS-CI (xlib) (push) Waiting to run
Using box16 nodes trades bvh quality for memory bandwidth which seems to
be roughly equal in performance.

Stats assuming box16 nodes are as expensive as box32 nodes:
Totals from 7668 (79.68% of 9624) affected BVHs:
compacted_size: 951666944 -> 742347648 (-22.00%)
max_depth: 57606 -> 57615 (+0.02%)
sah: 129114796242 -> 129998517775 (+0.68%); split: -0.00%, +0.68%
scene_sah: 188564162 -> 192063633 (+1.86%); split: -0.02%, +1.88%
box16_node_count: 0 -> 3270600 (+inf%)
box32_node_count: 3365707 -> 95100 (-97.17%)

Reviewed-by: Natalie Vock <natalie.vock@gmx.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37883>
2026-01-10 11:36:28 +01:00
Konstantin Seurer
543a88af99 radv/bvh: Add radv_aabb16 and use it for box16 nodes
Reviewed-by: Natalie Vock <natalie.vock@gmx.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37883>
2026-01-10 11:36:19 +01:00
Konstantin Seurer
fefdad9249 radv/rra: Count box16 nodes properly
Otherwise rra won't allocate memory when loading the capture.

Reviewed-by: Natalie Vock <natalie.vock@gmx.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37883>
2026-01-10 11:34:18 +01:00
Konstantin Seurer
39d58a55a7 aco: Add support to f2f16 with rtpi/rtni
Those rounding modes are needed when computing 16-bit bounding boxes
since the bounding box must not get smaller.

Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37883>
2026-01-10 11:34:12 +01:00
Konstantin Seurer
f156743b0f spirv: Add internal f2f16 opcodes
The OpFConvert+FPRoundingModeRTP/FPRoundingModeRTN cannot be used
because GL_EXT_spirv_intrinsics does not allow decorations. Instead,
we need opcodes that encode the rounding mode so that they can be used
in glsl code.

Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37883>
2026-01-10 11:34:07 +01:00
Konstantin Seurer
6d9cd36db6 nir: Add f2f16_ru/rd opcodes
Those are variants of f2f16 that always round up/down. Constant folding
requires nextafter that supports half floats (util_nextafter).

Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37883>
2026-01-10 11:33:23 +01:00
Mel Henning
78c8b8f131 nak/instr_sched_prepass: Fix RegOut special case
Some checks are pending
macOS-CI / macOS-CI (dri) (push) Waiting to run
macOS-CI / macOS-CI (xlib) (push) Waiting to run
This fixes the register count accounting to match the equivalent
special case in liveness.rs

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/14614
Fixes: b55b8da012 ("nak: Add a prepass instruction scheduler")
Acked-by: Mary Guillemard <mary@mary.zone>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39220>
2026-01-09 23:31:45 +00:00
Mel Henning
9ff8ee49f4 nak: Copy-prop bindless cbuf handles
Totals:
CodeSize: 4503463376 -> 4502059168 (-0.03%); split: -0.03%, +0.00%
Number of GPRs: 11377380 -> 11377668 (+0.00%); split: -0.00%, +0.00%
Static cycle count: 2871017064 -> 2870884663 (-0.00%); split: -0.01%, +0.00%
Spills to reg: 210798 -> 210728 (-0.03%)
Fills from reg: 200514 -> 200489 (-0.01%)
Max warps/SM: 7361864 -> 7361904 (+0.00%)

Totals from 28087 (14.27% of 196789) affected shaders:
CodeSize: 870417856 -> 869013648 (-0.16%); split: -0.17%, +0.01%
Number of GPRs: 2273399 -> 2273687 (+0.01%); split: -0.00%, +0.02%
Static cycle count: 780733864 -> 780601463 (-0.02%); split: -0.04%, +0.02%
Spills to reg: 135380 -> 135310 (-0.05%)
Fills from reg: 131832 -> 131807 (-0.02%)
Max warps/SM: 891108 -> 891148 (+0.00%)

Reviewed-by: Mary Guillemard <mary@mary.zone>
Reviewed-by: Lorenzo Rossi <git@rossilorenzo.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39122>
2026-01-09 23:08:29 +00:00
Mel Henning
e92ab5db56 nak/copy_prop: Split out prop_to_ssa_values helper
Reviewed-by: Mary Guillemard <mary@mary.zone>
Reviewed-by: Lorenzo Rossi <git@rossilorenzo.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39122>
2026-01-09 23:08:29 +00:00
Aaron Ruby
ab01793281 gfxstream: Partial revert of "gfxstream: revert "gfxstream: Add Vulkan func/structs for passing debugging data to host""
i.e. A revert^2 of 8f8bc7ca3b, to undo the
revert in cerealgenerator.py.

It needs to track latest file-naming in upstream host gfxstream:
https://github.com/google/gfxstream/blob/main/host/common/include/gfxstream/host/process_resources.h

Reviewed-by: Gurchetan Singh <gurchetansingh@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39241>
2026-01-09 22:50:18 +00:00
Karol Herbst
777f971d68 docs: reorder and add zink to CL subgroup entries
Some checks are pending
macOS-CI / macOS-CI (dri) (push) Waiting to run
macOS-CI / macOS-CI (xlib) (push) Waiting to run
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38015>
2026-01-09 21:53:28 +00:00
Karol Herbst
2466c5a94e rusticl: properly check for subgroup support
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38015>
2026-01-09 21:53:28 +00:00
Karol Herbst
2e2b86c64f clc: handle all optional subgroup extensions
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38015>
2026-01-09 21:53:28 +00:00
Karol Herbst
1c87c79893 llvmpipe: advertise support for subgroups in all stages
Works on CL and also in use by lavapipe. However it needs the
shader_subgroup_size cap to be set to be actually advertized in OpenGL.

Also support subgroup rotate while at it.

Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38015>
2026-01-09 21:53:28 +00:00
Karol Herbst
2a5227981e gallium: add SUBGROUP_FEATURE bits for rotate and rotate_clustered
The bit values are taken from Vulkan to make it easy for Zink. Those new
subgroup features will be used by rusticl for cl_khr_subgroup_rotate.

Reviewed-by: Adam Jackson <ajax@redhat.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38015>
2026-01-09 21:53:28 +00:00
Alyssa Rosenzweig
235e868ef7 ac/nir: use nir_is_shared_access
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39219>
2026-01-09 20:51:13 +00:00
Alyssa Rosenzweig
00e705add1 pan/mdg: use nir_is_shared_access
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39219>
2026-01-09 20:51:13 +00:00
Alyssa Rosenzweig
b26f9f42c4 agx: use nir_is_shared_access
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39219>
2026-01-09 20:51:12 +00:00
Alyssa Rosenzweig
43efc1cc7e brw: use nir_is_shared_access
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39219>
2026-01-09 20:51:12 +00:00
Alyssa Rosenzweig
4e59199cbb nir: add nir_is_shared_access helper
This is helpful to identify shared mem access for writing more generic code
operating on nir intrinsics.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39219>
2026-01-09 20:51:12 +00:00
Caio Oliveira
d160b7726a brw/scoreboard: Disable nomask workaround for Xe2+
Some checks are pending
macOS-CI / macOS-CI (dri) (push) Waiting to run
macOS-CI / macOS-CI (xlib) (push) Waiting to run
The issue was caused by fused EU feature that is not used in Xe2+
anymore.

Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36659>
2026-01-09 17:25:00 +00:00
Caio Oliveira
47a6ef3fef brw/scoreboard: Use a predicate helper for the nomask workaround
If it wasn't for the workaround, it wouldn't be necessary to track the
whether instructions are exec_all or not.  The workaround affects
results when mixing a dep and inst with different exec_all.

Add the predicate so that, when the workaround is disabled, none of
the effects of having different exec_all will kick in, all them will
be considered `exec_all = true`.

This patch don't change any behavior, just adds the predicate.

Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36659>
2026-01-09 17:25:00 +00:00