922f559c3c
aco: use new disable_wqm for flatlike
...
No Foz-DB changes on GFX1201.
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35970 >
2025-08-15 07:03:46 +00:00
a4c537c5b3
aco: use new disable_wqm for mubuf/mtbuf
...
Foz-DB GFX1201:
Totals from 66 (0.08% of 80251) affected shaders:
Instrs: 45373 -> 45663 (+0.64%); split: -0.01%, +0.65%
CodeSize: 251708 -> 252900 (+0.47%); split: -0.00%, +0.48%
Latency: 278977 -> 278652 (-0.12%); split: -0.14%, +0.02%
InvThroughput: 38259 -> 38245 (-0.04%); split: -0.05%, +0.02%
VClause: 982 -> 962 (-2.04%)
Copies: 2882 -> 2808 (-2.57%)
PreSGPRs: 2564 -> 2599 (+1.37%)
SALU: 4748 -> 5010 (+5.52%)
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35970 >
2025-08-15 07:03:46 +00:00
63af48ae2e
aco/insert_exec: new way to handle instructions that need wqm disabled
...
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35970 >
2025-08-15 07:03:46 +00:00
ca25553b92
aco: add a post-RA pass to disable wqm
...
By disabling WQM post-RA, we don't have RA/spilling mov issues with image_sample
operands that need to be computed in WQM.
We also don't restrict scheduling by inserting exec writes.
The only downside is more scalar ALU usage, but the SALU is almost always underutilized.
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35970 >
2025-08-15 07:03:46 +00:00
34b154866f
aco/insert_exec: remove p_jump_to_epilog from needs exact
...
p_end_wqm will always be emitted before it by isel.
No Foz-DB changes on GFX1201.
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35970 >
2025-08-15 07:03:46 +00:00
5e51c707ed
lavapipe: Revise HAVE_LIBDRM to guard on drm only variables
...
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com >
Reviewed-by: Lucas Fryzek <lfryzek@igalia.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36731 >
2025-08-15 05:58:23 +00:00
da8e990a0a
rusticl: specify FD type when importing fences
...
Acked-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36780 >
2025-08-15 00:13:09 +00:00
f543b36178
rusticl/mesa: port PipeFence to use ThreadSafeCPtr
...
Acked-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36780 >
2025-08-15 00:13:08 +00:00
ec88e83796
rusticl: handle failures when importing fences
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36780 >
2025-08-15 00:13:08 +00:00
c3a331234d
rusticl/icd: sort extension functions by extension name
...
Acked-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36780 >
2025-08-15 00:13:07 +00:00
dda52143d4
rusticl/icd: qualify CLResult inside impl_cl_type_trait_base macro
...
Acked-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36780 >
2025-08-15 00:13:07 +00:00
ad0402fc57
rusticl/util: drop uneccesary Arc in event_list_from_cl
...
Acked-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36780 >
2025-08-15 00:13:06 +00:00
495a46bf49
rusticl/util: use read_and_advance in Properties
...
Acked-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36780 >
2025-08-15 00:13:06 +00:00
4826e37f24
rusticl/util: add read_and_advance methods for pointers
...
Acked-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36780 >
2025-08-15 00:13:06 +00:00
060a084a12
rusticl: fix a bunch of warnings
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36780 >
2025-08-15 00:13:06 +00:00
d6a79f7e9f
ci/tu: generalize the multisample_resolve tess/gs flakes.
...
More variants have shown up, no reason to think formats are special.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36781 >
2025-08-14 19:26:27 +00:00
8fb872159b
ci/tu: Drop highp.scalar xfail.
...
Fixes: 99f09f8047https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36781 >
2025-08-14 19:26:27 +00:00
7090ad706b
ci/tu: Generalize the FDM flakes and link an issue.
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36781 >
2025-08-14 19:26:27 +00:00
a1aed20842
rusticl: implement cl_ext_immutable_memory_objects
...
Reviewed-by: Seán de Búrca <leftmostcat@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36630 >
2025-08-14 18:40:35 +00:00
d483566271
rusticl/mem: turn bool argument into enum in validate_mem_flags
...
Reviewed-by: Seán de Búrca <leftmostcat@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36630 >
2025-08-14 18:40:35 +00:00
bcec4acbd3
rusticl/mem: split out mem_flags validation for creation operations
...
validate_mem_flags is used by clCreate* and clGetSupportedImageFormats
which is a bit akward as those APIs have slightly different semantics on
the image flags and the clCreate* APIs support more flags with more
complex validation rules. And clGetSupportedImageFormats should be very
permissive in what combination of flags to accept.
Reviewed-by: Seán de Búrca <leftmostcat@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36630 >
2025-08-14 18:40:35 +00:00
61ad2bdf7f
rusticl/kernel: create shader constants as immutable
...
Reviewed-by: Seán de Búrca <leftmostcat@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36630 >
2025-08-14 18:40:35 +00:00
c5f57f6594
rusticl/mesa: add ResourceType::Immutable
...
Reviewed-by: Seán de Búrca <leftmostcat@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36630 >
2025-08-14 18:40:34 +00:00
4cb6094f2e
ac,radeonsi/vcn: Use correct swizzle_mode for vcn4
...
On VCN4 SWIZZLE_MODE_8x8_1D_THIN_12_24BPP use different value
than previous VCN generations
Signed-off-by: Yinjie Yao <yinjie.yao@amd.com >
Reviewed-by: Ruijing Dong <ruijing.dong@amd.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36775 >
2025-08-14 17:24:40 +00:00
0aef7ede1e
zink+radv/ci: give gfx1201 piglit tests a display to use
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36579 >
2025-08-14 18:10:10 +02:00
d0ef3c02c7
zink+radv/ci: give vangogh piglit tests a display to use
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36579 >
2025-08-14 18:10:10 +02:00
ef4b6808b9
zink+radv/ci: give navi31 piglit tests a display to use
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36579 >
2025-08-14 18:10:10 +02:00
c789def7a6
zink+radv/ci: give navi10 piglit tests a display to use
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36579 >
2025-08-14 18:10:10 +02:00
7f0b5a2c92
zink+radv/ci: give polaris10 piglit tests a display to use
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36579 >
2025-08-14 18:10:09 +02:00
003df4bb64
zink+radv/ci: add common fails for the next commits
...
The GPU-specific fails will be in each commit.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36579 >
2025-08-14 18:10:09 +02:00
c038839eb3
zink+radv/ci: fix typo in skips comment
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36648 >
2025-08-14 16:07:04 +00:00
c8c1541cd4
zink+radv/ci: deduplicate zink-radv-*-fails.txt files
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36648 >
2025-08-14 16:07:04 +00:00
2e26e1e86f
zink+radv/ci: deduplicate zink-radv-*-skips.txt lists
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36648 >
2025-08-14 16:07:04 +00:00
90d692bd32
zink+radv/ci: set DRIVER_NAME=zink-radv to allow using common expectation files
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36648 >
2025-08-14 16:07:04 +00:00
3a211679f7
softpipe/ci: drop asan fails that are already tracked as normal fails
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36648 >
2025-08-14 16:07:04 +00:00
6ea6682384
lavapipe/ci: drop asan fails that are already tracked as normal fails
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36648 >
2025-08-14 16:07:04 +00:00
22dcb250cc
ci/deqp-runner: add support for all the prefixes for *-fails.txt files
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36648 >
2025-08-14 16:07:04 +00:00
2440ecc348
ci/deqp-runner: remove duplicate values to avoiding read the same file multiple times
...
Some jobs define multiple of these to the same value; let's simply
ignore those, as they might have a legitimate reason to do it.
This wasn't a problem with skips and flakes as those can contain
duplicates, but fails (next commit) are not allowed to be defined
multiple times.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36648 >
2025-08-14 16:07:04 +00:00
8137eae04d
ci/deqp-runner: add support for all the prefixes for *-flakes.txt files
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36648 >
2025-08-14 16:07:04 +00:00
f8353e7eff
ci/deqp-runner: simplify handling the various *-skips.txt files
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36648 >
2025-08-14 16:07:03 +00:00
44015c5325
ci/deqp-runner: drop implicit skips of GALLIUM_DRIVER or VK_DRIVER
...
The previous commit made these explicit via `DRIVER_NAME`.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36648 >
2025-08-14 16:07:03 +00:00
1b88e4a262
ci: set DRIVER_NAME in jobs that are implicitly inheriting skip lists
...
They were getting these through setting `GALLIUM_DRIVER`, which the next
commit removes.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36648 >
2025-08-14 16:07:03 +00:00
f6f091b9a6
softpipe/ci: document fixed tests
...
Unrelated to this MR, just noticed now.
They are likely missed by the current fraction on that job, which is why
this wasn't noticed.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36648 >
2025-08-14 16:07:03 +00:00
117ac6ac0a
tc: remove user cbuf uploads
...
this is no longer supported (drivers should set prefer_real_buffer_in_constbuf0)
Reviewed-by: Marek Olšák <marek.olsak@amd.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36715 >
2025-08-14 12:47:48 +00:00
f967a85f64
mesa/st: rework thread scheduler handling + add dispatch tracking
...
Reviewed-by: Marek Olšák <marek.olsak@amd.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36715 >
2025-08-14 12:47:47 +00:00
b8cd48e8e8
gallium: always upload cbuf0 when cap is set
...
Reviewed-by: Marek Olšák <marek.olsak@amd.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36715 >
2025-08-14 12:47:47 +00:00
203ac73374
gallium: set prefer_real_buffer_in_constbuf0 for all drivers using tc
...
this isn't really a functional change since tc was always doing
cbuf uploads anyway
Reviewed-by: Marek Olšák <marek.olsak@amd.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36715 >
2025-08-14 12:47:47 +00:00
04c505c200
zink: destroy u_uploaders earlier in context destroy
...
these might get defer-deleted onto the batch states, so ensure they
don't leak in that way
Reviewed-by: Marek Olšák <marek.olsak@amd.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36715 >
2025-08-14 12:47:47 +00:00
cbb6e0277f
r600: stop using util_set_vertex_buffers
...
Reviewed-by: Marek Olšák <marek.olsak@amd.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36715 >
2025-08-14 12:47:47 +00:00
2d1758242f
r300: stop using util_set_vertex_buffers
...
Reviewed-by: Marek Olšák <marek.olsak@amd.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36715 >
2025-08-14 12:47:47 +00:00
Georg Lehmann