fdo-mirrors/mesa
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/mesa/mesa.git synced 2026-06-16 00:18:20 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions 1
224064 commits 66 branches 938 tags 993 MiB
Commit graph

224064 commits

This branch
This branch
All branches
Author SHA1 Message Date
Eric Engestrom
127acbb126 ci: bump fedora from 42 to 44 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41620>
 2026-06-10 13:53:26 +00:00
Eric Engestrom
4ebf2e3baa ci: bump bindgen version from 0.71.1 to 0.72.1 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41620>
 2026-06-10 13:53:26 +00:00
Eric Engestrom
dae8bc711d ci: bump rust version from 1.90 to 1.96 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41620>
 2026-06-10 13:53:25 +00:00
Eric Engestrom
47570e74ec meson: exclude known buggy versions of bindgen 
Backport-to: *
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41620>
 2026-06-10 13:53:25 +00:00
Eric Engestrom
09ea05cf23 rusticl: skip bindgen for pipe_shader_state_from_tgsi 
bindgen up to at least 0.72.1 generates invalid code (see below) and
that function is not used, so simply skip it.

    src/gallium/frontends/rusticl/rusticl_mesa_bindings.c:795:81: error: duplicate ‘const’ declaration specifier [-Werror=duplicate-decl-specifier]
      795 | void pipe_shader_state_from_tgsi__extern(struct pipe_shader_state *state, const const struct tgsi_token *tokens) { pipe_shader_state_from_tgsi(state, tokens); }
          |                                                                                 ^~~~~

Backport-to: *
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41620>
 2026-06-10 13:53:25 +00:00
yserrr
38a98a4803 v3d: remove duplicate util_blitter_save_so_targets() call
Some checks are pending
macOS-CI / macOS-CI (dri) (push) Waiting to run
Details
macOS-CI / macOS-CI (xlib) (push) Waiting to run
Details 
v3d_blitter_save() saves the stream output targets twice.

Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/42139>
 2026-06-10 13:36:26 +00:00
Rhys Perry
addc719ec2 radv: workaround has_smem_partial_oob_access_bug 
Just use an existing flag to increase the bo size slightly.

Fixes a ring gfx timeout with
dEQP-VK.spirv_assembly.instruction.compute.opfma.fp32.vec3.undef.denorm_flush.directed
on vega10.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Backport-to: *
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41937>
 2026-06-10 13:01:47 +00:00
Rhys Perry
f7a3884278 ac/gpu_info: add has_smem_partial_oob_access_bug 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Backport-to: *
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41937>
 2026-06-10 13:01:46 +00:00
Rhys Perry
bed7ba2780 aco: schedule split barriers 
Move the s_barrier_signal as earlier and the s_barrier_wait later.

fossil-db (gfx1201):
Totals from 2152 (1.03% of 208640) affected shaders:
Instrs: 1463236 -> 1463248 (+0.00%); split: -0.00%, +0.01%
CodeSize: 7710732 -> 7710720 (-0.00%); split: -0.00%, +0.00%
Latency: 7164883 -> 7159042 (-0.08%); split: -0.10%, +0.01%
InvThroughput: 1593643 -> 1593651 (+0.00%); split: -0.00%, +0.00%
VClause: 30170 -> 30166 (-0.01%)
SClause: 26771 -> 26772 (+0.00%)
Copies: 123002 -> 123004 (+0.00%)
SALU: 221966 -> 221967 (+0.00%)
VOPD: 1680 -> 1681 (+0.06%)

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41364>
 2026-06-10 12:13:19 +00:00
Rhys Perry
26b942c306 aco: use split barrier instructions 
fossil-db (gfx1201):
Totals from 135 (0.06% of 208640) affected shaders:
Instrs: 155940 -> 155932 (-0.01%); split: -0.02%, +0.02%
CodeSize: 905460 -> 905432 (-0.00%); split: -0.02%, +0.01%
Latency: 1910087 -> 1909703 (-0.02%); split: -0.02%, +0.00%
InvThroughput: 886321 -> 886280 (-0.00%)
Copies: 12025 -> 12024 (-0.01%)
VALU: 89681 -> 89679 (-0.00%)
VOPD: 177 -> 178 (+0.56%)

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41364>
 2026-06-10 12:13:19 +00:00
Rhys Perry
a95f841125 aco: add split barrier instructions 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41364>
 2026-06-10 12:13:19 +00:00
Rhys Perry
7c6be36cf4 aco: don't emit waitcnts before subgroup-scope execution barriers 
This delays the waitcnt for has_attr_ring_wait_bug by a few instructions.

fossil-db (gfx1201):
Totals from 9 (0.00% of 208640) affected shaders:
Instrs: 19352 -> 19506 (+0.80%)
CodeSize: 101180 -> 101716 (+0.53%)
Latency: 660221 -> 678782 (+2.81%); split: -0.00%, +2.81%
InvThroughput: 95106 -> 97398 (+2.41%)

fossil-db (navi33):
Totals from 58834 (28.20% of 208626) affected shaders:
Instrs: 22424304 -> 22424571 (+0.00%)
CodeSize: 110198112 -> 110199184 (+0.00%)
Latency: 115894319 -> 126491124 (+9.14%); split: -0.00%, +9.14%
InvThroughput: 19424631 -> 19754358 (+1.70%); split: -0.00%, +1.70%

I don't think the stats are very accurate. This seems to often move the
s_waitcnt down into a divergent branch, but the wait still happens later
if the branch isn't taken, so the wait is counted twice.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41364>
 2026-06-10 12:13:18 +00:00
Rhys Perry
3676c3860e aco: only assume load/store with semantic_atomic is atomic 
ACCESS_ATOMIC was added a while ago.

fossil-db (gfx1201):
Totals from 84 (0.04% of 208640) affected shaders:
Instrs: 74569 -> 74402 (-0.22%)
CodeSize: 379220 -> 378552 (-0.18%)
Latency: 589791 -> 575984 (-2.34%)
InvThroughput: 56042 -> 54921 (-2.00%)

fossil-db (navi33):
Totals from 79 (0.04% of 208626) affected shaders:
Instrs: 69170 -> 69015 (-0.22%)
CodeSize: 349580 -> 348928 (-0.19%)
Latency: 563270 -> 549156 (-2.51%)
InvThroughput: 61245 -> 59887 (-2.22%)

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41364>
 2026-06-10 12:13:18 +00:00
Rhys Perry
a0d5c117fc aco: optimize redundant s_wait_alu vm_vsrc(0) during waitcnt insertion 
fossil-db (gfx1201):
Totals from 143 (0.07% of 208640) affected shaders:
Instrs: 104804 -> 104588 (-0.21%)
CodeSize: 543148 -> 542320 (-0.15%)
Latency: 751702 -> 751446 (-0.03%); split: -0.04%, +0.00%
InvThroughput: 78599 -> 78588 (-0.01%); split: -0.02%, +0.00%

fossil-db (navi33):
Totals from 170 (0.08% of 208626) affected shaders:
Instrs: 107230 -> 106983 (-0.23%)
CodeSize: 554952 -> 553940 (-0.18%)
Latency: 746901 -> 746628 (-0.04%); split: -0.04%, +0.00%
InvThroughput: 102412 -> 102390 (-0.02%)

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41364>
 2026-06-10 12:13:18 +00:00
Rhys Perry
650715b077 aco: fix printing of primitive exports 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41364>
 2026-06-10 12:13:17 +00:00
Rhys Perry
c815c51dcb aco/waitcnt: always use uint32_t for event masks 
This shouldn't fix anything, because event_vmem_bvh was never used here.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41364>
 2026-06-10 12:13:17 +00:00
Rhys Perry
49fb361c0a aco: don't emit workgroup-scope p_barrier for single-wave workgroups 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41364>
 2026-06-10 12:13:17 +00:00
Rhys Perry
e6703f8e68 aco: add cost estimation of s_barrier 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41364>
 2026-06-10 12:13:17 +00:00
Christian Gmeiner
9a8ac361ff panvk: Advertise VK_GOOGLE_display_timing
Some checks are pending
macOS-CI / macOS-CI (dri) (push) Waiting to run
Details
macOS-CI / macOS-CI (xlib) (push) Waiting to run
Details 
Signed-off-by: Christian Gmeiner <cgmeiner@igalia.com>
Reviewed-by: Christoph Pillmayer <christoph.pillmayer@arm.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/42066>
 2026-06-10 11:33:00 +00:00
Christian Gmeiner
9cecd0821b vulkan/wsi: Constify wsi_instance_supports_google_display_timing(..) 
Signed-off-by: Christian Gmeiner <cgmeiner@igalia.com>
Reviewed-by: Christoph Pillmayer <christoph.pillmayer@arm.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/42066>
 2026-06-10 11:33:00 +00:00
Lionel Landwerlin
e8ab81cc3a brw: fix null render target decision 
First, use 64bit values everywhere since shader_info::outputs_written
is a 64bit field.

Second, alpha to coverage should only be considered for draw buffer 0
as stated in the GL spec (quoting Version 4.6 (Core Profile), 17.3.1
Alpha To Coverage) :

   "All alpha values in this section refer only to the alpha component
    of the fragment shader output linked to color number zero, index
    zero (see section 15.2.3)."

Third, the write message setup in brw_compile_fs.cpp was not taking
into account alpha-to-coverage being disabled anymore.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 294644643e ("brw: avoid requiring a valid render target for empty fragment shaders")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/work_items/15625
Tested-by: Christoph Neuhauser <christoph.neuhauser@intel.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/42115>
 2026-06-10 11:08:47 +00:00
Lionel Landwerlin
d698cd7485 iris: only call brw_nir_fs_needs_null_rt() with no render targets 
This helper was only meant to be called once the driver knows it
doesn't have any render target setup, to figure out whether an empty
one needs to be created.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: mesa-stable
Tested-by: Christoph Neuhauser <christoph.neuhauser@intel.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/42115>
 2026-06-10 11:08:47 +00:00
David Rosca
1825627cf8 va: Set contiguous_planes for DMA-BUF imported surfaces 
Set contiguous_planes if all planes share the same fd.
This makes vaDeriveImage work with DMA-BUF imported surfaces.

Reviewed-by: Boyuan Zhang <Boyuan.Zhang@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/42086>
 2026-06-10 10:46:59 +00:00
Georg Lehmann
db382b3cae radeonsi: use exec_size from the aco prolog/epilog callback
Some checks are pending
macOS-CI / macOS-CI (dri) (push) Waiting to run
Details
macOS-CI / macOS-CI (xlib) (push) Waiting to run
Details 
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/42028>
 2026-06-10 08:23:01 +00:00
Georg Lehmann
3bb9ad4d47 radeonsi: use ac_get_instr_prefetch_size 
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/42028>
 2026-06-10 08:23:01 +00:00
Georg Lehmann
4a3e38fd7c radv/gfx11+: program INST_PREF_SIZE for NGG and HS 
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/42028>
 2026-06-10 08:23:01 +00:00
Georg Lehmann
a23fc3f066 radv/gfx12: program SPI_SHADER_PGM_RSRC4_GS for seperately compiled gs 
Cc: mesa-stable

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/42028>
 2026-06-10 08:23:01 +00:00
Georg Lehmann
22a8788dea aco: add exec_size to prolog/epilog callback 
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/42028>
 2026-06-10 08:23:01 +00:00
Georg Lehmann
9b0536da1f radv/gfx11+: program INST_PREF_SIZE for pixel shaders 
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/42028>
 2026-06-10 08:23:00 +00:00
Georg Lehmann
dd3c7756cc radv/gfx11+: program INST_PREF_SIZE for compute 
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/42028>
 2026-06-10 08:22:59 +00:00
Georg Lehmann
e1b649d333 radv: remove gfx6 code from ngg emission 
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/42028>
 2026-06-10 08:22:59 +00:00
Georg Lehmann
ae9eb6298b amd/common: add helper for INST_PREF_SIZE 
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/42028>
 2026-06-10 08:22:59 +00:00
Georg Lehmann
f341fb0742 amd/common: don't pass radeon_info to ac_align_shader_binary_for_prefetch 
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/42028>
 2026-06-10 08:22:59 +00:00
Georg Lehmann
5c3a6c938b amd/gpu_info: precompute instruction prefetch distance 
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/42028>
 2026-06-10 08:22:58 +00:00
Samuel Pitoiset
ec6748468d ci: add a new option called profile in ci_run_n_monitor.py 
It seems very useful, at least to me, to have some predefined profiles
for vkd3d or VKCTS main uprevs.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/42110>
 2026-06-10 07:59:20 +00:00
Samuel Pitoiset
47298397dc util/drirc: remove the driver option in drirc_validate 
Each driver use its own drirc file now, so the option is useless.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41881>
 2026-06-10 07:17:14 +00:00
Samuel Pitoiset
957eb2b5f0 radv,anv: remove useless includes for drirc stuff 
No longer needed.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41881>
 2026-06-10 07:17:14 +00:00
Thomas H.P. Andersen
c0531ece56 nvk: use the new generation script for drirc 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41881>
 2026-06-10 07:17:14 +00:00
Samuel Pitoiset
4c66fc1383 venus: use drirc_gen 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41881>
 2026-06-10 07:17:14 +00:00
Samuel Pitoiset
86406ca87d v3dv: use drirc_gen 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41881>
 2026-06-10 07:17:14 +00:00
Samuel Pitoiset
8b422baac8 hk: use drirc_gen 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41881>
 2026-06-10 07:17:14 +00:00
Samuel Pitoiset
309c2a213a panvk: use drirc_gen 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41881>
 2026-06-10 07:17:14 +00:00
Samuel Pitoiset
281ae60b5e pvr: use drirc_gen 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41881>
 2026-06-10 07:17:14 +00:00
Samuel Pitoiset
dbbd251b6f dzn: use drirc_gen 
WSI options are already NULL, so they are never used.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41881>
 2026-06-10 07:17:14 +00:00
Samuel Pitoiset
0dd9b61c80 util/drirc_gen: allow to override the defaults VK WSI common options 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41881>
 2026-06-10 07:17:14 +00:00
Samuel Pitoiset
482e9f3002 util/drirc_gen: add heap_memory_percent to common VK options 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41881>
 2026-06-10 07:17:14 +00:00
Samuel Pitoiset
34c343aad1 util/drirc_gen: prevent generating empty structs 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41881>
 2026-06-10 07:17:14 +00:00
Samuel Pitoiset
59d6d5e45f util/drirc_gen: fix generating 64-bit driconf options 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41881>
 2026-06-10 07:17:14 +00:00
Samuel Pitoiset
8e008649cb util/drirc_gen: change the driconf DTD to not require one app/engine entry 
To be able to validate empty files for drivers that don't have any
driconf entries yet.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/41881>
 2026-06-10 07:17:14 +00:00
Job Noorman
1a9a0a15f7 ir3: lower undef booleans to zero
Some checks are pending
macOS-CI / macOS-CI (dri) (push) Waiting to run
Details
macOS-CI / macOS-CI (xlib) (push) Waiting to run
Details 
We are generally fine with undefs. However, booleans are only allowed to
contain 0/1 but are stored in 16b registers. Undefs may cause such
registers to be uninitialized and contain values other than 0/1. This
especially happens with undef booleans in phi srcs, which are explicitly
left uninitialized. In general, such non-0/1 values don't cause problems
because we mostly use booleans by comparing them to 0. However, they do
cause problems in special cases like `inot(x)` which we lower to `sub(1,
x)` which only works if true==1.

Fixes misrenderings in "Kingdoms of Amalur: Reckoning".

Totals from 12 (0.01% of 176258) affected shaders:
Instrs: 14590 -> 14615 (+0.17%)
CodeSize: 29796 -> 29808 (+0.04%)
NOPs: 3091 -> 3098 (+0.23%); split: -0.03%, +0.26%
MOVs: 735 -> 748 (+1.77%)
(sy)-stall: 4509 -> 4508 (-0.02%)
Cat0: 3471 -> 3483 (+0.35%); split: -0.03%, +0.37%
Cat1: 1257 -> 1270 (+1.03%)

Signed-off-by: Job Noorman <jnoorman@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/42051>
 2026-06-10 06:05:05 +00:00