AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
203,808 Commits 1 Branch 0 Tags
83bf45b0ce94f7d3504b4e2efe0f284e0cae2c09
Commit Graph

203808 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
LingMan 83bf45b0ce meson: Update pest subproject family 
Starting with Rust 1.83 this benign warning is show when compiling the pest dependency:

```
warning: elided lifetime has a name
   --> pest/src/iterators/pairs.rs:330:70
    |
89  | impl<'i, R: RuleType> Pairs<'i, R> {
    |      -- lifetime `'i` declared here
...
330 |     ) -> Filter<FlatPairs<'i, R>, impl FnMut(&Pair<'i, R>) -> bool + '_> {
    |                                                                      ^^ this elided lifetime gets resolved as `'i`
    |
    = note: `#[warn(elided_named_lifetimes)]` on by default
```

Meson, at least as of version 1.7.0, unfortunately does not suppress warnings originating in dependencies.
Upstream has resolved the warning in 2.8.0, so update to that.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34368>
 2025-04-04 21:22:27 +00:00
Caio Oliveira 03aca2d248 brw: Use new bld/exp style in scoreboard tests 
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34354>
 2025-04-04 20:14:53 +00:00
Caio Oliveira 7ee673c195 brw: Add parser of SWSB annotations to use in tests 
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34354>
 2025-04-04 20:14:53 +00:00
Caio Oliveira 81dd3e1527 brw: Return actual progress in brw_lower_scoreboard 
This will be useful later for tests to be used in conjunction with the
EXPECT_PROGRESS / EXPECT_NO_PROGRESS helpers.

Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34354>
 2025-04-04 20:14:53 +00:00
Caio Oliveira 3e727000dd brw: Stop setting SFID in scoreboard tests 
They won't affect the scoreboard, and will get in the
way of a later change.

Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34354>
 2025-04-04 20:14:53 +00:00
Caio Oliveira bcea076aca brw: Use SIMD16 shaders in scoreboard tests for Xe2+ 
Some tests changed to avoid unintended overlap between operands which
would change the SWSB assigned.  In some cases also changed the Gfx12
matching test so they remain equal.

Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34354>
 2025-04-04 20:14:52 +00:00
Caio Oliveira cd486cda48 brw: Use control flow helpers in scoreboard tests 
Also update WHILE to optionally take a predicate (default to NONE).  And
make the predicate in the IF optional (default to NORMAL).

Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34354>
 2025-04-04 20:14:52 +00:00
Eric Engestrom 7c5389695b ci/fluster: don't overwrite FLUSTER_VECTORS_HOST_PATH to a different meaning 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34362>
 2025-04-04 19:36:09 +00:00
Eric Engestrom 8ec11893d9 ci/fluster: use http proxy when checking for the vector files 
It's done properly when downloading it, but not when checking for it.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34362>
 2025-04-04 19:36:09 +00:00
Konstantin e7a44de184 nir/tests: Do not rely on __LINE__ 
__LINE__ can be inconsistent when using different compilers. This patch
changes the test runner to do a simple string find/replace of the test
source file instead of looking for the line where the reference string
starts.

Acked-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33980>
 2025-04-04 19:01:01 +00:00
Danylo Piliaiev c496774c35 freedreno/a6xx: Implement ARB_shader_clock 
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29860>
 2025-04-04 18:22:49 +00:00
Danylo Piliaiev 0e9854a894 tu: Implement VK_KHR_shader_clock 
There is a special address defined in kernel from which ALWAYSON
counter could be read.

Blob uses this sequence to read it:
  getone #l15
  mov.s32s32 r2.y, -4096
  mov.s32s32 r2.z, 131071
  (rpt5)nop
  ldg.u32 r2.w, g[r2.y], 1
  ldg.u32 r2.y, g[r2.y+4], 1
  (sy)(ss)mov.s32s32 r48.x, (last)r2.w
  mov.s32s32 r48.y, (last)r2.y
  l15:

Passes:
 dEQP-VK.glsl.shader_clock.*

Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29860>
 2025-04-04 18:22:49 +00:00
Danylo Piliaiev 4b1b4ee10c freedreno,tu: Read and pass to compiler uche_trap_base 
KGSL always exposed uche_trap_base, and MSM only recently
got support for it.

Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29860>
 2025-04-04 18:22:49 +00:00
Danylo Piliaiev 6a448ca08b freedreno: Bump kernel uapi (linux 6.14) 
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29860>
 2025-04-04 18:22:48 +00:00
Benjamin Otte 0941af995a lavapipe: Don't advertise support for multiplane drm formats 
Fixes: bd4f69a0fe
Signed-off-by: Benjamin Otte <otte@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34190>
 2025-04-04 17:59:43 +00:00
Robert Mader 8b76f521a0 meson: Bump minimum version to 1.3.0 
Builds currently fail on older versions since the commit
below.

Fixes: f35172b6a4 (meson: make CL args common)
Signed-off-by: Robert Mader <robert.mader@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34313>
 2025-04-04 17:15:34 +00:00
Mark Collins e4359cc49c tu/kgsl: Fix KGSL syncobj lifetime in no CB submit 
The temporary syncobj created in the fast path of kgsl_queue_submit was
not being destroyed, and potentially being assigned to multiple syncobjs
without being properly duplicated. This could lead to a use-after-free
or double-free since multiple syncobjs could be assigned the same FD.

Signed-off-by: Mark Collins <mark@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34328>
 2025-04-04 16:54:17 +00:00
Mark Collins cf4bd2e412 tu/kgsl: Revert "Remove zero CB queue submission fast path" 
This reverts commit 0342d34bdb which
introduced a regression in the Turnip's KGSL backend, causing various
sync issues since KGSL doesn't advance the GPU timeline when a submit
without cmdbufs is made. A comment explaining the issue was added to the
code, and the fast path is reintroduced.

Signed-off-by: Mark Collins <mark@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34328>
 2025-04-04 16:54:17 +00:00
Georg Lehmann c70dcd1451 aco/gfx9+: use d16 global/scratch/buffer loads 
Full register loads are not nessecary and prevent packing optimizations.
Global/Scratch is GFX9+ so D16 loads are always supported.
We already used LDS D16 loads.

Foz-DB Navi31(mostly RA noise):
Totals from 716 (0.90% of 79789) affected shaders:
Instrs: 3854176 -> 3854238 (+0.00%); split: -0.00%, +0.00%
CodeSize: 20034440 -> 20035220 (+0.00%); split: -0.00%, +0.00%
Latency: 24410951 -> 24411120 (+0.00%)
InvThroughput: 5181276 -> 5181301 (+0.00%)
Copies: 320258 -> 320317 (+0.02%)
VALU: 2207307 -> 2207366 (+0.00%)

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34346>
 2025-04-04 16:20:39 +00:00
Juan A. Suarez Romero f5e36e382f broadcom/compiler: initialize register 
This fixes issue detected by static analyzer: passed-by-value struct
argument contains uninitialized data (e.g., field: 'file').

Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Reviewed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34050>
 2025-04-04 15:55:13 +00:00
Juan A. Suarez Romero 0e50b09d4a broadcom/compiler: don't use VLA on emit alu 
Using constant-size array instead of variable-length array is preferred
due several issues with the latter.

Particularly, for this case using VLA generates several warnings by
static analyzer: passed-by-value struct argument contains uninitialized
data (e.g., field: 'file').

Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Reviewed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34050>
 2025-04-04 15:55:13 +00:00
Juan A. Suarez Romero 01151f045f broadcom/compiler: use safe iterator to remove instructions 
The current approach has an issue detected by static analyzer: use of
memory after it is freed.

Using a proper iterator makes things safer.

Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Reviewed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34050>
 2025-04-04 15:55:13 +00:00
Juan A. Suarez Romero 0b0d6a36f2 vc4: use safe iterator to remove instructions 
The current approach has an issue detected by static analyzer: use of
memory after it is freed.

Using a proper iterator makes things safer.

Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Reviewed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34050>
 2025-04-04 15:55:13 +00:00
Juan A. Suarez Romero baa4fefe74 broadcom/cle: assert attribute has a value 
This assertions helps to fix several warnings detected by static
analyzer regarding passing null pointers to functions that expects
non-null pointer.

Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Reviewed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34050>
 2025-04-04 15:55:13 +00:00
Juan A. Suarez Romero 47f280778e vc4: assert there are sources when emitting texture 
This assertion fixes an issue detected by static analyzer:
passed-by-value struct argument contains uninitialized data (e.g.,
field: 'file').

Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Reviewed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34050>
 2025-04-04 15:55:13 +00:00
Juan A. Suarez Romero 8f5423820a vc4: add assertion on constant_fold 
This fixes an issue detected by static analyzer: the right operand of
'>>' is a garbage value due to array index out of bounds.

Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Reviewed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34050>
 2025-04-04 15:55:13 +00:00
Juan A. Suarez Romero 54db7c46fa vc4: initialize register 
While this is likely not necessary, it fixes an issue detected by static
analyzer: the left operand of '<=' is a garbage value.

Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Reviewed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34050>
 2025-04-04 15:55:12 +00:00
Juan A. Suarez Romero 6d6a3ab679 v3dv: asserts push constants data is valid 
When pushing constants.

This fixes an issue detected by static analyzer: null pointer passed to
1st parameter expecting 'nonnull'.

Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Reviewed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34050>
 2025-04-04 15:55:12 +00:00
Juan A. Suarez Romero 665df034e7 v3dv: check dynamic offset output 
The output variable must be non NULL when descriptor type is dynamic.

This fixes an issue detected by static analyzer: dereference of null
pointer (loaded from variable 'dynamic_offset').

Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Reviewed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34050>
 2025-04-04 15:55:12 +00:00
Juan A. Suarez Romero ede3feb16d v3dv: asserts struct is always non null 
This fixes an issue detected by static analyzer: access to field 'pNext'
results in a dereference of a null pointer (loaded from variable
'rs_info').

Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Reviewed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34050>
 2025-04-04 15:55:12 +00:00
Juan A. Suarez Romero 2c8411e9d2 vc4: check instruction before setting flags 
This fixes an issue detected by static analyzer: access to field 'sf'
results in a dereference of a null pointer (loaded from variable
'last_inst').

Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Reviewed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34050>
 2025-04-04 15:55:12 +00:00
Lionel Landwerlin 72bc74f0be anv: add shader-hash debug option 
Emits a dummy MI_STORE_DATA_IMM with the shader hash in front of :
   - 3DSTATE_VS
   - 3DSTATE_HS
   - 3DSTATE_DS
   - 3DSTATE_HS
   - 3DSTATE_PS
   - COMPUTE_WALKER / GPGPU_WALKER

Example :

0x00000000:  0x10000002:  MI_STORE_DATA_IMM
0x00000000:  0x10000002 : Dword 0
    DWord Length: 2
    Force Write Completion Check : false
    Store Qword: 0
    Use Global GTT: false
0x00000004:  0xffffe0c0 : Dword 1
    Core Mode Enable: 0
0x00000008:  0x0000effe : Dword 2
    Address: 0xeffeffffe0c0
0x0000000c:  0x126e815a : Dword 3  <------------ shader hash
0x00000010:  0x78100007 : Dword 4
    Immediate Data: 309231962
0x00000000:  0x78100007:  3DSTATE_VS
0x00000000:  0x78100007 : Dword 0
    DWord Length: 7
0x00000004:  0x00000000 : Dword 1
0x00000008:  0x00000000 : Dword 2
    Kernel Start Pointer: 0x00000000
0x0000000c:  0x00040000 : Dword 3
    Software Exception Enable: false
    Accesses UAV: false

It'll correlate with the value emitted in the pipeline stats from fossil replay :

  $ grep -i 126e815a /tmp/stats.csv
  fossilize.aab93c5c3f965151.1.foz,GRAPHICS,de1b925dec8a8083,507378,498283,303434,vertex,8,50,4,0,1826,0,0,0,8,17,0,0x00000000126e815a,15

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34332>
 2025-04-04 15:18:28 +00:00
Lionel Landwerlin 789f13359a anv: consolidate environment variables 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34332>
 2025-04-04 15:18:28 +00:00
Lionel Landwerlin 713cb0fdc1 anv/hasvk: sort out debug options 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34332>
 2025-04-04 15:18:28 +00:00
Lionel Landwerlin 8a51e097af docs: remove unused env variable 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34332>
 2025-04-04 15:18:28 +00:00
Lionel Landwerlin 43e0f02391 anv/hasvk: consider timeline semaphore support stable 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34332>
 2025-04-04 15:18:28 +00:00
Vlad Zahorodnii c57da522fa vulkan/wsi/wayland: Document why wl_surface_damage() code path ignores provided damage 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34227>
 2025-04-04 14:38:44 +00:00
Vlad Zahorodnii 0c943bbb64 vulkan/wsi/wayland: Damage whole surface using wl_surface_damage_buffer() 
Most compositors work with damage in the buffer local coordinate space.
This change spares the compositors some work converting the provided
INT32_MAX x INT32_MAX damage region to the buffer local coordinate
space. It has no significant performance impact, but it'd still be nice
to use wl_surface_damage_buffer() if possible.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34227>
 2025-04-04 14:38:44 +00:00
Vlad Zahorodnii fd146d04d1 egl/wayland: Damage whole surface using wl_surface_damage_buffer() 
Most compositors work with damage in the buffer local coordinate space.
This change spares the compositors some work converting the provided
INT32_MAX x INT32_MAX damage region to the buffer local coordinate
space. It has no significant performance impact, but it'd still be nice
to use wl_surface_damage_buffer() if possible.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34227>
 2025-04-04 14:38:43 +00:00
Job Noorman 78ef51aa04 ir3/opt_preamble: take alias.rt into account for rewrite cost 
FS outputs can use const registers in alias.rt without a mov so take
this into account when calculating the rewrite cost of instructions.

Totals:
MaxWaves: 2765084 -> 2765130 (+0.00%); split: +0.00%, -0.00%
Instrs: 56289002 -> 56285073 (-0.01%); split: -0.01%, +0.00%
CodeSize: 118071672 -> 118076808 (+0.00%); split: -0.00%, +0.01%
NOPs: 9491112 -> 9492474 (+0.01%); split: -0.00%, +0.02%
MOVs: 1790085 -> 1786768 (-0.19%); split: -0.19%, +0.00%
Full: 2156693 -> 2156607 (-0.00%); split: -0.00%, +0.00%
(ss): 1329812 -> 1329546 (-0.02%); split: -0.03%, +0.01%
(sy): 686396 -> 686386 (-0.00%); split: -0.00%, +0.00%
(ss)-stall: 4995295 -> 4995185 (-0.00%); split: -0.02%, +0.01%
(sy)-stall: 19828966 -> 19828624 (-0.00%); split: -0.01%, +0.01%
Cat0: 10450369 -> 10451731 (+0.01%); split: -0.00%, +0.02%
Cat1: 2787946 -> 2784566 (-0.12%); split: -0.12%, +0.00%
Cat2: 21265787 -> 21264447 (-0.01%)
Cat3: 16207098 -> 16206536 (-0.00%)
Cat7: 1597849 -> 1597840 (-0.00%); split: -0.00%, +0.00%

Totals from 730 (0.36% of 200220) affected shaders:
MaxWaves: 6308 -> 6354 (+0.73%); split: +0.79%, -0.06%
Instrs: 258235 -> 254306 (-1.52%); split: -1.59%, +0.07%
CodeSize: 698806 -> 703942 (+0.73%); split: -0.28%, +1.02%
NOPs: 21040 -> 22402 (+6.47%); split: -1.85%, +8.33%
MOVs: 9426 -> 6109 (-35.19%); split: -35.52%, +0.33%
Full: 8914 -> 8828 (-0.96%); split: -1.03%, +0.07%
(ss): 5118 -> 4852 (-5.20%); split: -6.58%, +1.39%
(sy): 2118 -> 2108 (-0.47%); split: -1.18%, +0.71%
(ss)-stall: 17360 -> 17250 (-0.63%); split: -4.57%, +3.94%
(sy)-stall: 34921 -> 34579 (-0.98%); split: -5.90%, +4.92%
Cat0: 24734 -> 26096 (+5.51%); split: -1.58%, +7.09%
Cat1: 12311 -> 8931 (-27.46%); split: -27.70%, +0.24%
Cat2: 106329 -> 104989 (-1.26%)
Cat3: 100547 -> 99985 (-0.56%)
Cat7: 3646 -> 3637 (-0.25%); split: -0.91%, +0.66%

Signed-off-by: Job Noorman <jnoorman@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34279>
 2025-04-04 14:17:10 +00:00
Vignesh Raman 7959250d1e s3_upload: improve url validation and error message 
Ensure s3_upload correctly validates the S3 folder url by requiring it
to end with /. This prevents wrong uploads to invalid paths, such as
file urls. Also improve the error message.

Signed-off-by: Vignesh Raman <vignesh.raman@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34255>
 2025-04-04 13:32:45 +00:00
Zan Dobersek 248edb43c3 tu: allow D3D-compatible texture coordinate rounding 
When running under DXVK or vkd3d, the texture coordinate rounding behavior
should match D3D expectations. On Adreno, this behavior can be toggled
through the SP_TP_MODE_CNTL register.

A driconf-based option is introduced to help set the relevant register flag
that enables this behavior.

This fixes the cause of test_sampler_rounding test case failure in vkd3d on
Turnip's side, but a small change in vkd3d is also required, so the test
failure expectation isn't removed yet.

Signed-off-by: Zan Dobersek <zdobersek@igalia.com>
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33987>
 2025-04-04 10:09:47 +00:00
Zan Dobersek 3b1ca55b40 freedreno/registers: add useful A6XX_SP_TP_MODE_CNTL bitfields 
Add additional bitfields for the A6XX_SP_TP_MODE_CNTL registers, ones that
we already use and the texcoord rounding mode bitfield that we'll need for
D3D-over-Vulkan implementations.

Signed-off-by: Zan Dobersek <zdobersek@igalia.com>
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33987>
 2025-04-04 10:09:47 +00:00
Benjamin Lee e183650aa4 panvk/csf: fix uninitialized read in utrace_clone_init_builder 
Previous code assumed that the caller of utrace_clone_init_builder would
fill some parameters of the builder config, but we were not. Instead,
initialize these from the csif props the same as all the other builder
instances.

Signed-off-by: Benjamin Lee <benjamin.lee@collabora.com>
Fixes: 3096cf2a5d ("panvk/csf: flush and process trace events for all cmdbufs")
Reviewed-by: Eric R. Smith <eric.smith@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34270>
 2025-04-04 09:43:02 +00:00
Hyunjun Ko 2b0df6c564 anv: Use vk_video_derive_h265_scaling_list 
Signed-off-by: Hyunjun Ko <zzoon@igalia.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34096>
 2025-04-04 07:23:48 +00:00
MaciejDziuban f31a33905a radv: Use vk_video_derive_h265_scaling_list 
This commit makes radv use vk_video_derive_h265_scaling_list, which properly
applies default scaling lists whenever they're needed. It also simplifies
update_h265_scaling function into a simple memcpy. The firmware interface
struct and Vulkan's StdVideoH265ScalingLists struct both have identical memory
layouts, so it's not neccessary divide it into multiple copies with offsets.

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34096>
 2025-04-04 07:23:48 +00:00
MaciejDziuban 4072286f07 vulkan: Add default scaling lists for H265 
H265 specification defines default scaling lists to use whenever scaling lists
are not specified in neither sps nor pps. Currently drivers ignore this
requirement and set the lists to zero. This commits adds a helper function
vk_video_derive_h265_scaling_list (similar to its h264 counterpart) that
selects either sps or pps lists and falls back to default values if neither
were specified. The default values were taken from ITU-T H265 specification
(revision 8), section 7.4.5.

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34096>
 2025-04-04 07:23:48 +00:00
MaciejDziuban a1bf7192e5 vulkan: handle use_default_scaling_matrix_mask in h264 decoder 
H264 specification defines this field to force usage of the default
scaling lists even if they are specified in ScalingList4x4 and
ScalingList8x8.

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34096>
 2025-04-04 07:23:47 +00:00
Ian Romanick 20cce95ce5 brw/opt: Don't call brw_opt_copy_propagation before brw_lower_load_reg 
On a 36c/72t Xeon system, performance of replaying
hogwarts_legacy.dx12vk-ultra.foz was improved 1.3% +/- 0.77% (n=10).

I picked MTL for the fossil-db results because it was the most negative.

shader-db:

All Intel platforms had fairly similar results. (Lunar Lake)
total instructions in shared programs: 16964217 -> 16964216 (<.01%)
instructions in affected programs: 51777 -> 51776 (<.01%)
helped: 20 / HURT: 27

total cycles in shared programs: 892934916 -> 893041912 (0.01%)
cycles in affected programs: 51245298 -> 51352294 (0.21%)
helped: 96 /HURT: 78

fossil-db:

All Intel platforms had similar results. (Meteor Lake shown)
Totals:
Instrs: 233678547 -> 233678944 (+0.00%); split: -0.00%, +0.00%
Cycle count: 24398049850 -> 24400490877 (+0.01%); split: -0.01%, +0.02%
Max live registers: 42145052 -> 42145038 (-0.00%); split: -0.00%, +0.00%

Totals from 1141 (0.14% of 805934) affected shaders:
Instrs: 1546001 -> 1546398 (+0.03%); split: -0.01%, +0.03%
Cycle count: 1201746062 -> 1204187089 (+0.20%); split: -0.14%, +0.34%
Max live registers: 84247 -> 84233 (-0.02%); split: -0.03%, +0.01%

Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31497>
 2025-04-04 06:45:02 +00:00
Ian Romanick 991a2f510b brw/sat: Eliminate non-defs saturate propagation 
The intervening_saturating_copy test is removed. The defs version of the
pass does not handle this case. It should not occur often in practice
anyway. Copy propagation and brw_nir_opt_fsat should prevent this
scenario from happening.

No shader-db changes on any Intel platform.

fossil-db:

All Intel platforms had similar results. (Lunar Lake shown)
Totals:
Instrs: 212677275 -> 212677278 (+0.00%)
Cycle count: 30466062848 -> 30466056040 (-0.00%)

Totals from 1 (0.00% of 706300) affected shaders:
Instrs: 1343 -> 1346 (+0.22%)
Cycle count: 411664 -> 404856 (-1.65%)

v2: Stop counting ip. The non-defs part of the pass was the only thing
that used it.

v3: Also delete "if (block != def->block) continue;" code. I noticed
this while working on some other changes to this function. It's the last
thing in the loop, so it's totally useless. Delete some other spurious
continues too.

Reviewed-by: Caio Oliveira <caio.oliveira@intel.com> [v2]
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31497>
 2025-04-04 06:45:02 +00:00