0cfabe0613
aco/lower_to_hw_instr: Add scratch size in call lowering
...
We did this in the preserved spiller previously, but let's move it here.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38281 >
2025-12-08 19:12:55 +00:00
7059068b61
aco/spill: Restore registers spilled by call immediately
...
Makes for better latency hiding if we're not short on registers
otherwise.
On top of RT function calls:
Totals from 7 (0.01% of 81072) affected shaders:
Instrs: 9084 -> 8980 (-1.14%)
CodeSize: 52564 -> 51976 (-1.12%)
SpillSGPRs: 244 -> 248 (+1.64%); split: -3.28%, +4.92%
SpillVGPRs: 360 -> 367 (+1.94%)
Latency: 138989 -> 135669 (-2.39%); split: -2.49%, +0.10%
InvThroughput: 35120 -> 35301 (+0.52%); split: -0.06%, +0.57%
VClause: 258 -> 241 (-6.59%)
SClause: 116 -> 117 (+0.86%)
Copies: 1290 -> 1311 (+1.63%)
Branches: 131 -> 119 (-9.16%)
VALU: 6125 -> 6143 (+0.29%); split: -0.20%, +0.49%
SALU: 920 -> 913 (-0.76%); split: -0.98%, +0.22%
VMEM: 1026 -> 989 (-3.61%)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38281 >
2025-12-08 19:12:55 +00:00
6616f25e43
aco/spill: Create linear VGPRs for spilling ABI-preserved SGPRs
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38281 >
2025-12-08 19:12:55 +00:00
6b2e766617
aco/ra: Handle linear VGPRs allocated by p_startpgm
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38281 >
2025-12-08 19:12:54 +00:00
761efe9163
aco/spill: Reset scratch_rsrc on calls
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38281 >
2025-12-08 19:12:54 +00:00
369a3c0dca
aco/spill: Handle calls
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38281 >
2025-12-08 19:12:54 +00:00
ecc548cd37
aco: Record required call spills during live-var analysis
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38281 >
2025-12-08 19:12:53 +00:00
8bc5fdef53
aco: Remove unused p_reload_preserved def
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38281 >
2025-12-08 19:12:52 +00:00
8b7ad622ee
aco/live_var_analysis: Count linear VGPRs as always preserved by calls
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37907 >
2025-12-08 18:50:31 +00:00
f692ac76ca
radv/rt: Use traversal vars for object origin/direction in ahit/isec
...
We already did the work of transforming the ray data, no need to do it
multiple times.
Should theoretically be a lot better. However, none of the fossils
appear to use object-space ray data in anyhit/intersection shaders. :(
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38809 >
2025-12-08 18:29:49 +00:00
a488203e85
radv/rt: Refactor and split radv_nir_rt_shader.c
...
This splits up radv_nir_rt_shader.c into several parts.
The first part is all ray traversal lowering for RT pipelines, located
at radv_nir_rt_traversal_shader.c. It implements building the traversal
loop, including inlined any-hit/intersection shaders (optionally as a
completely separate shader).
The second part is lowering for individual RT stages (right now,
monolithic vs. CPS-style separate compilation). Each lowering technique
lives in its own file (radv_nir_rt_stage_{monolithic,cps}.c).
Code shared between RT lowering techniques (shader inlining helpers and
storage lowering passes) gets moved into radv_nir_rt_stage_common.c.
One header, radv_nir_rt_stage.h, is the public interface for RT pipeline
stage lowering. Functions exposed to users (really just
radv_pipeline_rt.c) go there. The header for internal shared helpers is
radv_nir_rt_stage_common.c.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38809 >
2025-12-08 18:29:49 +00:00
5681fabdc2
panfrost,panvk: Enable shader_realtime_clock on panthor 1.6
...
shader_realtime_clock requires a newer kernel version in order to enable
GLB_COUNTER_EN this change adds a check on this kernel functionality.
Remove GL_EXT_shader_realtime_clock from extensions as this now depends
on kernel version.
Fixes: e9c2c324ashley.smith@collabora.com >
Reviewed-by: Daniel Stone <daniels@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37915 >
2025-12-08 17:58:52 +00:00
7130127e2a
dzn: Suppress new MSVC warning by upconverting to uint64_t
...
Reviewed-by: Marek Olšák <marek.olsak@amd.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38838 >
2025-12-08 17:28:36 +00:00
55e833ae96
mesa: Cast bitmasks to 64-bit before negating
...
Reviewed-by: Marek Olšák <marek.olsak@amd.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38838 >
2025-12-08 17:28:36 +00:00
71dfcd3c96
zink: Fix 64-bit bitmask usage
...
Reviewed-by: Marek Olšák <marek.olsak@amd.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38838 >
2025-12-08 17:28:36 +00:00
dacc56ee53
u_threaded_context: Use 64-bit bitmask utils
...
Reviewed-by: Marek Olšák <marek.olsak@amd.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38838 >
2025-12-08 17:28:36 +00:00
d966fde4f4
ci: Drop hardware-job prerequisite check jobs
...
Stop requiring the `rustfmt` and `yaml-toml-shell-py-test` jobs before
starting hardware jobs.
These prerequisites don't save meaningful resources, as the hardware jobs
are well-behaved nowadays, but they do create unnecessary friction for
users.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/14409
Signed-off-by: Valentine Burley <valentine.burley@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38786 >
2025-12-08 16:59:45 +00:00
f912db3f8d
util/driconf/asahi: Override GL renderer for web browsers
...
Several web sites block clients with "Apple" in the WebGL renderer
string if the reported OS is not one of Apple's.
This check seems to implemented via a 3rd party product which is slowly
rolled out over more web sites. Instead of playing whack-a-mole with
web sites in multiple browsers override the OpenGL renderer in mesa for
known browsers.
Backport-to: 25.3
Signed-off-by: Janne Grunau <j@jannau.net >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38845 >
2025-12-08 16:23:54 +00:00
6246b7be10
r600: improve cayman scissor 1x1 workaround
...
This change improves evergreen_apply_scissor_bug_workaround().
It provides a fully functional workaround for cayman.
Note: this was the last functionality which was working
properly on evergreen but not on cayman.
Here are the tests fixed:
spec/arb_framebuffer_no_attachments/arb_framebuffer_no_attachments-atomic/glscissor: fail pass
spec/arb_framebuffer_no_attachments/arb_framebuffer_no_attachments-query/glscissor: fail pass
deqp-gles31/functional/fbo/no_attachments/interaction/1x1ms0_default_2048x2048ms4: fail pass
deqp-gles31/functional/fbo/no_attachments/npot_size/1x1: fail pass
Fixes: 87a5b07f90patrick9876@free.fr >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38743 >
2025-12-08 15:04:24 +00:00
7f6bd8b003
nir/peephole_select: allow mbcnt_amd
...
It's just alu, so handle it like alu.
Foz-DB Navi21:
Totals from 3 (0.00% of 97591) affected shaders:
Instrs: 433 -> 426 (-1.62%)
CodeSize: 2408 -> 2388 (-0.83%)
Latency: 7520 -> 7925 (+5.39%)
InvThroughput: 857 -> 1009 (+17.74%)
Copies: 55 -> 43 (-21.82%)
Branches: 21 -> 17 (-19.05%)
SALU: 79 -> 76 (-3.80%)
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38828 >
2025-12-08 10:52:48 +00:00
005cc4110c
nir/peephole_select: allow ballot
...
We can allow collapsing control flow around ballot if we update the ballot
condition like we do for discards.
ballot_relaxed needs no condition update, as the result bits are undefined
for inactive invocations.
Foz-DB Navi21:
Totals from 27 (0.03% of 97591) affected shaders:
Instrs: 2554506 -> 2554469 (-0.00%); split: -0.00%, +0.00%
CodeSize: 13765636 -> 13765684 (+0.00%); split: -0.00%, +0.00%
Latency: 14186667 -> 14186861 (+0.00%); split: -0.00%, +0.00%
InvThroughput: 3542516 -> 3542595 (+0.00%); split: -0.00%, +0.00%
SClause: 52038 -> 52030 (-0.02%)
Copies: 209410 -> 208763 (-0.31%)
Branches: 83716 -> 83399 (-0.38%)
PreSGPRs: 2372 -> 2386 (+0.59%); split: -0.17%, +0.76%
VALU: 1701458 -> 1701482 (+0.00%)
SALU: 369884 -> 370107 (+0.06%); split: -0.00%, +0.07%
SMEM: 67643 -> 67634 (-0.01%)
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38828 >
2025-12-08 10:52:48 +00:00
077b654cc7
nir: don't sink alu that uses ballot(true)
...
Don't sink alu that uses ballot(true), as that can a local system value
and moving the alu then requires a new mov in the old location.
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38829 >
2025-12-08 09:07:54 +00:00
96662cd459
nak/cmat: free the type mapping hash table.
...
Just noticed this on review.
Cc: mesa-stable
Reviewed-by: Karol Herbst <kherbst@redhat.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38843 >
2025-12-08 08:08:56 +10:00
6b67ca2e8f
pan/bi: Keep vectorized phis
...
When the source contains vectorized phis, keep them to avoid
generating unnecessary moves.
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38772 >
2025-12-06 07:22:47 +00:00
d3a890a58e
pan/bi: Handle small vectors in bi_src_index()
...
bit_size <= 32 does not actually guarantee a single component, which
nir_src_as_uint() requires. We could just check num_components == 1 but
it's easy enough to support any vector that fits in 32 bits.
Cc: mesa-stable
Reviewed-by: Romaric Jodin <rjodin@google.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38772 >
2025-12-06 07:22:47 +00:00
a051d4ee6b
nir/lower_io_vars: don't insert output stores for unrelated streams before emits
...
Before every emit_vertex(stream_id = n), we would insert stores for all
outputs, including outputs that are not meant for that stream.
Those stores would end up having no effect while potentially reducing
performance.
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38100 >
2025-12-06 02:27:46 +00:00
0ac9ccb4b4
d3d12: Only try to compute scaled point size for stream 0
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38100 >
2025-12-06 02:27:46 +00:00
0df8aa940c
nir: Use nir_shader_intrinsics_pass in nir_lower_io_to_scalar
...
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38816 >
2025-12-05 22:30:22 +00:00
e9ad86db08
tu: support driconf option force_vk_vendor
...
Some games refuse to start with an unknown (to them) GPU vendor. Enable
the usual way to override the vendor ID in this driver.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38795 >
2025-12-05 22:05:35 +00:00
c0e6b0aca0
zink/ci: update fixed tests
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38587 >
2025-12-05 21:22:41 +00:00
ae36b9e345
radeonsi: enable 64K x 64K textures
...
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com >
Reviewed-by: Qiang Yu <yuq825@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38587 >
2025-12-05 21:22:41 +00:00
6feb3e49f8
st/mesa: remove bogus framebuffer state assertions
...
Reviewed-by: Qiang Yu <yuq825@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38587 >
2025-12-05 21:22:40 +00:00
f46b6042ab
radeonsi: allow 64K viewports
...
max_hw_screen_offset was incorrect
Acked-by: Qiang Yu <yuq825@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38587 >
2025-12-05 21:22:40 +00:00
f6bf120d02
radeonsi: reject textures that don't fit in the CPU address space
...
Reviewed-by: Qiang Yu <yuq825@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38587 >
2025-12-05 21:22:40 +00:00
566c666d57
radeonsi: fix a few non-critical 64-bit integer overflows
...
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com >
Reviewed-by: Qiang Yu <yuq825@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38587 >
2025-12-05 21:22:40 +00:00
07d2ab29b6
st/mesa: disallow the PBO upload fragment shader
...
It passes image_size to the shader as int32_t.
Reviewed-by: Qiang Yu <yuq825@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38587 >
2025-12-05 21:22:40 +00:00
bd839d4bfc
st/mesa: don't use the PBO GetTexImage compute shader for 64K textures
...
I uses uint16_t for width/height.
Reviewed-by: Qiang Yu <yuq825@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38587 >
2025-12-05 21:22:40 +00:00
72a14518ae
mesa: raise MAX_TEXTURE_LEVELS to 17 to allow 64K mipmap textures
...
Reviewed-by: Qiang Yu <yuq825@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38587 >
2025-12-05 21:22:40 +00:00
610bdf5e9e
mesa: bump MAX_TEXTURE_RECT_SIZE, MAX_RENDERBUFFER_SIZE
...
Reviewed-by: Qiang Yu <yuq825@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38587 >
2025-12-05 21:22:40 +00:00
24ba57259f
mesa: remove MaxTextureMbytes, use the cap instead
...
Reviewed-by: Qiang Yu <yuq825@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38587 >
2025-12-05 21:22:39 +00:00
00f5f0980a
mesa: use size_t for image address computations
...
For 64K x 64K textures.
The regular expressions used to find potentially overflowing multiplications:
(width.*\*.*height|height.*\*.*width|stride.*\*.*height|height.*\*.*stride|row.*\*.*height|height.*\*.*row)
(height.*\*.*depth|depth.*\*.*height|size.*\*.*depth|depth.*\*.*size)
A few things were authored by Pierre-Eric using static analysis to
detect potential overflows.
Reviewed-by: Qiang Yu <yuq825@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38587 >
2025-12-05 21:22:39 +00:00
3dc2582172
mesa: merge mostly duplicated mesa_format_image_size & mesa_format_image_size64
...
One of them returned 32 bits, the other one returned 64 bits.
Reviewed-by: Qiang Yu <yuq825@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38587 >
2025-12-05 21:22:38 +00:00
6507e8ec65
mesa: remove unused make_null_texture
...
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com >
Reviewed-by: Qiang Yu <yuq825@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38587 >
2025-12-05 21:22:37 +00:00
4fba07f74e
mesa: remove unused mesa_store_cleartexsubimage, _mesa_store_compressed_teximage
...
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com >
Reviewed-by: Qiang Yu <yuq825@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38587 >
2025-12-05 21:22:37 +00:00
30b5d5f9b3
mesa: remove unused _mesa_total_texture_memory
...
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com >
Reviewed-by: Qiang Yu <yuq825@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38587 >
2025-12-05 21:22:36 +00:00
88b6e46ee3
mesa: remove unused image debug code
...
Reviewed-by: Qiang Yu <yuq825@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38587 >
2025-12-05 21:22:36 +00:00
4529693e72
gallium/u_blitter: change width/height parameters to 32-bit integer
...
Reviewed-by: Qiang Yu <yuq825@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38587 >
2025-12-05 21:22:36 +00:00
7621b2d56d
gallium: declare pipe_resource::height0 as 32-bit integer for 64K textures
...
Reviewed-by: Qiang Yu <yuq825@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38587 >
2025-12-05 21:22:36 +00:00
3c6477812e
gallium: change pipe_framebuffer_state width/height to 32-bit integer
...
Reviewed-by: Qiang Yu <yuq825@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38587 >
2025-12-05 21:22:36 +00:00
bc1a6b0a41
gallium: change pipe_scissor_state to 32 bit integer
...
Reviewed-by: Qiang Yu <yuq825@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38587 >
2025-12-05 21:22:35 +00:00
Natalie Vock