AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
205,885 Commits 1 Branch 0 Tags
38caef8fa2033889ff412f530189fb28210a6788
Commit Graph

205885 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Juan A. Suarez Romero
38caef8fa2 broadcom/ci: update expected test results 
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35156>
 2025-05-26 11:42:45 +02:00
Job Noorman
7ebcc8d402 ir3: don't free constant_data after assembling 
When using shader overrides, the assembler will be called a 2nd time
which will try to dereference the freed constant_data. Fix this by not
explicitly freeing constant_data, it's ralloc'd in the context of the
shader variant anyway so will be freed automatically.

Signed-off-by: Job Noorman <jnoorman@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35128>
 2025-05-26 07:43:16 +00:00
Lionel Landwerlin
5daf4608e5 anv: workaround Sky: Children of the Light 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/12966
Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35111>
 2025-05-26 05:52:30 +00:00
Lionel Landwerlin
828173d79a anv: add support for lower_terminate_to_discard workaround 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35111>
 2025-05-26 05:52:30 +00:00
Lionel Landwerlin
87e57a9bb2 radv: rename radv_lower_terminate_to_discard for wider use 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35111>
 2025-05-26 05:52:30 +00:00
Olivia Lee
65406cf500 panvk/csf: fix provoking vertex mode in partial secondary cmdbufs 
For partial secondary cmdbufs, we emit FBDs/TDs in the primary cmdbuf
before calling the secondary. In order to set the provoking vertex mode
correctly here, we need to look at the mode set by pipelines bound in
the secondary cmdbuf.

This leaves one edge case: reemitting FBDs/TDs in a secondary cmdbuf
after a flush. If the secondary cmdbuf only contains vk_meta draws,
without ever binding a pipeline, we won't know which provoking vertex
mode to use here. This is actually okay, because in that case the
provoking vertex mode doesn't matter for any of the draws in the
secondary, and the FBDs/TDs will be reemitted on the primary with the
correct mode.

Fixes: 7a9f14d3c2 ("panvk: advertise VK_EXT_provoking_vertex")
Signed-off-by: Olivia Lee <olivia.lee@collabora.com>
Tested-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Ryan Mckeever <ryan.mckeever@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34974>
 2025-05-26 04:15:22 +00:00
Olivia Lee
885805560f panvk/csf: fix case where vk_meta is used before PROVOKING_VERTEX_MODE_LAST 
In this case, we need to emit the FBDs and TDs for the meta command
before we know what provoking vertex mode the application is going to
use. To handle this, we make a guess for which provoking vertex mode we
need. Then we use cs_maybe to leave space to flip the provoking vertex
bit if the guess was wrong.

This case is still unhandled on JM.

Fixes: 7a9f14d3c2 ("panvk: advertise VK_EXT_provoking_vertex")
Signed-off-by: Olivia Lee <olivia.lee@collabora.com>
Tested-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Ryan Mckeever <ryan.mckeever@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34974>
 2025-05-26 04:15:22 +00:00
Olivia Lee
4d99346477 panvk: fix case where vk_meta is used after PROVOKING_VERTEX_MODE_LAST 
Because we advertise provokingVertexModePerPipeline=false, the provoking
vertex mode must be set the same for all pipelines used in a renderpass.
vk_meta doesn't care about the provoking vertex mode, but the vulkan api
doesn't provide a way to express this, so it always sets
PROVOKING_VERTEX_MODE_FIRST (the vulkan default). This causes an
assertion failure when vk_meta is used in a renderpass where the
application sets PROVOKING_VERTEX_MODE_LAST.

There are a few different cases here, that need different handling. The
simplest is when vk_meta is used after the first application draw, in
which case we can just ignore the state passed by vk_meta and use the
existing state.

Fixes: 7a9f14d3c2 ("panvk: advertise VK_EXT_provoking_vertex")
Signed-off-by: Olivia Lee <olivia.lee@collabora.com>
Tested-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Ryan Mckeever <ryan.mckeever@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34974>
 2025-05-26 04:15:22 +00:00
Olivia Lee
32177b99d5 panvk: track whether we are in a vk_meta command 
This is needed to handle the provoking vertex mode correctly. vk_meta
doesn't care which provoking vertex mode is used, but there is no way to
express this directly in the vulkan api.

Signed-off-by: Olivia Lee <olivia.lee@collabora.com>
Tested-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Ryan Mckeever <ryan.mckeever@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34974>
 2025-05-26 04:15:21 +00:00
Olivia Lee
d60c688317 panvk/csf: set up shared register dump regions for cs functions 
The tiler OOM exception handler allocated a region of memory to dump
save/restored registers. For defining more functions in the future, we
allocate a register dump region for each subqueue, that can hold the
largest number of registers needed by any functions executed on that
subqueue.

This does mean that we cannot have function calls more than one deep. If
we ever need nested function calls, we will have to consider a real
stack.

Signed-off-by: Olivia Lee <olivia.lee@collabora.com>
Tested-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Ryan Mckeever <ryan.mckeever@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34974>
 2025-05-26 04:15:21 +00:00
Olivia Lee
61e7d47270 pan/csf: rename cs_exception_handler to cs_function 
The register save/restore machinery is useful for more general callable
functions, not just exception handlers.

Signed-off-by: Olivia Lee <olivia.lee@collabora.com>
Tested-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Ryan Mckeever <ryan.mckeever@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34974>
 2025-05-26 04:15:21 +00:00
Olivia Lee
83bb97796b pan/csf: add cs_maybe mechanism to retroactively patch cs contents 
We have an edge case with VK_EXT_provoking_vertex where we may need to
emit FBDs and TDs before we know what provoking vertex mode the
application is using for the renderpass. To handle this, we want to
retroactively patch the provoking vertex bit. This commit introduces an
abstraction to do that.

Signed-off-by: Olivia Lee <olivia.lee@collabora.com>
Tested-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Ryan Mckeever <ryan.mckeever@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34974>
 2025-05-26 04:15:21 +00:00
Olivia Lee
952703eefe pan/csf: add cs_builder unit test infrastructure 
For now, just comparing the raw contents of the output buffer. Possibly
in the future we could hook this up to the disassembly from decode_csf.c
to make it a easier to edit.

Signed-off-by: Olivia Lee <olivia.lee@collabora.com>
Tested-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Ryan Mckeever <ryan.mckeever@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34974>
 2025-05-26 04:15:21 +00:00
Olivia Lee
79a1d98e1e pan/csf: make cs_builder.h usable from c++ 
We need to do this in order to test it with gtest. Most of the changes
are just fixing integer truncation warnings.

Signed-off-by: Olivia Lee <olivia.lee@collabora.com>
Tested-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Ryan Mckeever <ryan.mckeever@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34974>
 2025-05-26 04:15:21 +00:00
Qiang Yu
6f2a1e19da nir/opt_varyings: fix mesh shader miss promote varying to flat 
We still allow mesh shader promote constant output to flat, but
mesh shader like geometry shader may store multi vertices'
varying in a single thread. So mesh shader may store different
constant values to different vertices in a single thread, we
should not promote this case to flat.

I'm not using shader_info.mesh.ms_cross_invocation_output_access
because OpenGL does not require IO to have explicit location, so
when nir_shader_gather_info is called in OpenGL GLSL compiler to
compute ms_cross_invocation_output_access, some implicit output
has -1 location which causes ms_cross_invocation_output_access
unset for it.

Cc: mesa-stable
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/13134
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35081>
 2025-05-26 02:07:50 +00:00
Timothy Arceri
bf24d56862 util: add workaround for the game Foundation 
Cc: mesa-stable
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/12882
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35107>
 2025-05-26 00:10:40 +00:00
Timothy Arceri
27945bbd8a mesa: extend linear_as_nearest work around 
Here we allow packed stencils to skip the completeness check also.
Will be used in the following patch for a bug in the game Foundation.

Cc: mesa-stable
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35107>
 2025-05-26 00:10:39 +00:00
Faith Ekstrand
38f8e82dd2 nvk: 64-bit atomics exist on Kepler B+ 
64-bit atomics exist on Kepler B but not Kepler A.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35149>
 2025-05-25 00:39:56 +00:00
Faith Ekstrand
aad4410b51 nak/sm20: Don't allow 64-bit atomics 
The encoding exists and the disassembler will disassemble it but the
hardware throws an ILLEGAL_INSTRUCTION_ENCODING if we ever try to
execute one.  The proprietary driver doesn't expose any 64-bit atomics
features on Kepler A so we assume the hardware just doesn't do them.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35149>
 2025-05-25 00:39:56 +00:00
Faith Ekstrand
4d752ad123 nvk: Disable R64_[US]INT formats when shaderImageInt64Atomics isn't supported 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35149>
 2025-05-25 00:39:56 +00:00
Lorenzo Rossi
00bbb68c08 nak/sm32: Fix various encoding bugs 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35059>
 2025-05-25 00:33:47 +02:00
Lorenzo Rossi
150a61ab2f nak/sm32: Fix shfl.up register alignment 
For some obscure reason, shfl.up on sm32 requires the lane register to
be 128-bit aligned, this can be fixed with minimal code changes by
aligning the register in legalization.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35059>
 2025-05-24 21:20:49 +02:00
Collabora's Gfx CI Team
3b6d7cdd23 Uprev ANGLE to ec4d8f8e4d3f1a9e64430abd8b6b436253732adf 
db71e8fa7c...ec4d8f8e4d

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35040>
 2025-05-24 07:35:01 +00:00
Marek Olšák
32c419744e mesa: remove the old GL name (ID) allocator 
Split from the commit that enabled the new GL name (ID) allocator.

Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31515>
 2025-05-24 05:05:03 +00:00
sarbes
40497ca3a9 lima: genxml-ify PP frame registers 
Most (all?) PP frame registers are now documented in genxml. Most of the magic values written to the registers are gone.

v2:
- moved default values out of genxml
- commented on stencil truncation

v3:
- fixed comment typo

Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34210>
 2025-05-24 00:45:27 +00:00
sarbes
eb61a607d8 lima: add 'unorm8' format to genxml 
This MR allows packing of (normalized) floats into an 8 bit field.

v2:
- removed default attribute

Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34210>
 2025-05-24 00:45:27 +00:00
Christian Gmeiner
41f2da1a6e treewide: Do not use NIR_PASS_V for nir_divergence_analysis(..) 
Signed-off-by: Christian Gmeiner <cgmeiner@igalia.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35131>
 2025-05-23 21:19:25 +00:00
Caleb Callaway
e7454f5318 intel/debug: shader dump filter 
v2: Fixes filtering for various brw shader dump logic

Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35061>
 2025-05-23 19:57:02 +00:00
Caleb Callaway
5a36452158 intel/perf: intel_monitor documentation tweak 
Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35061>
 2025-05-23 19:57:02 +00:00
Mel Henning
59308f92bf nak: Call nir_lower_undef_to_zero 
Totals:
CodeSize: 4571686864 -> 4545122720 (-0.58%); split: -0.59%, +0.01%
Number of GPRs: 10994283 -> 10963889 (-0.28%); split: -0.28%, +0.00%
SLM Size: 1893964 -> 1855380 (-2.04%)
Static cycle count: 1114502308 -> 1104322907 (-0.91%); split: -0.93%, +0.02%
Spills to memory: 573363 -> 480689 (-16.16%)
Fills from memory: 573363 -> 480689 (-16.16%)
Spills to reg: 1625502 -> 458804 (-71.77%); split: -71.78%, +0.00%
Fills from reg: 878478 -> 303068 (-65.50%); split: -65.50%, +0.00%
Max warps/SM: 7235748 -> 7245516 (+0.13%)

Totals from 25050 (12.75% of 196502) affected shaders:
CodeSize: 1870779504 -> 1844215360 (-1.42%); split: -1.45%, +0.03%
Number of GPRs: 2578387 -> 2547993 (-1.18%); split: -1.18%, +0.00%
SLM Size: 890840 -> 852256 (-4.33%)
Static cycle count: 439061010 -> 428881609 (-2.32%); split: -2.37%, +0.05%
Spills to memory: 268807 -> 176133 (-34.48%)
Fills from memory: 268807 -> 176133 (-34.48%)
Spills to reg: 1485538 -> 318840 (-78.54%); split: -78.54%, +0.00%
Fills from reg: 753623 -> 178213 (-76.35%); split: -76.35%, +0.00%
Max warps/SM: 556844 -> 566612 (+1.75%)

Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35100>
 2025-05-23 19:41:31 +00:00
Mel Henning
9d620fabd2 nak: Fix a perf regression in tex lowering 
These lines look like they were mistakenly introduced, and cause a
significant perf hit. Eg. this fix improves the Horizon Zero Dawn
in-game benchamark by ~42% on my ampere machine (5992 pts -> 8517 pts).

Fixes: d16e75e55f ("nak: Lower texture inputs for Kepler B")
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35100>
 2025-05-23 19:41:31 +00:00
Eric Engestrom
162f1f5566 delete xa leftovers 
Fixes: 3be2c47db2 ("delete the XA frontend")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35136>
 2025-05-23 18:54:04 +00:00
Eric Engestrom
e3d8f4c135 lavapipe/ci: skip timing out tests 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35135>
 2025-05-23 18:30:30 +00:00
Eric Engestrom
31a5039b38 lavapipe/ci: document flakes 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35135>
 2025-05-23 18:30:30 +00:00
Eric Engestrom
84bbee22f4 rpi/ci: skip timing out test 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35135>
 2025-05-23 18:30:30 +00:00
Eric Engestrom
30dedadc53 rpi/ci: document flakes 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35135>
 2025-05-23 18:30:30 +00:00
Eric Engestrom
cde3351213 amd/ci: document radv flakes 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35135>
 2025-05-23 18:30:30 +00:00
Eric Engestrom
286204404a amd/ci: document radeonsi flakes 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35135>
 2025-05-23 18:30:30 +00:00
Eric Engestrom
2beeac9484 r300/ci: document fixed test 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35135>
 2025-05-23 18:30:30 +00:00
Eric Engestrom
5969df364a radeonsi/ci: document fixed piglit tests 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35135>
 2025-05-23 18:30:30 +00:00
Eric Engestrom
7707e17cd4 radv/ci: document fixed angle test 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35135>
 2025-05-23 18:30:30 +00:00
Mike Blumenkrantz
00aaef9f12 delete gallium-nine 
farewell, old friend

Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Acked-by: Marek Olšák <marek.olsak@amd.com>
Acked-by: Axel Davy <davyaxel0@gmail.com>
Acked-by: David Heidelberg <david@ixit.cz>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34887>
 2025-05-23 13:43:37 -04:00
Mike Blumenkrantz
3be2c47db2 delete the XA frontend 
this is unmaintained and untested

Acked-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34823>
 2025-05-23 12:46:23 -04:00
Sil Vilerino
c4b6285c81 mr-label-maker.yml: Add mediafoundation tag and associated directories 
Reviewed-by: Pohsiang Hsu <pohhsu@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34843>
 2025-05-23 15:30:04 +00:00
Pohsiang (John) Hsu
d348fd5fb5 mediafoundation: Add mediafoundation frontend 
Reviewed-by: <yuboxie@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34843>
 2025-05-23 15:30:04 +00:00
Rob Clark
45a2f02876 ci: Disable fd-farm 
Take the google farm offline in preparation for shipping.

Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35093>
 2025-05-23 15:27:16 +00:00
Mike Blumenkrantz
42b303c7b0 lavapipe: handle counterOffset in vkCmdDrawIndirectByteCountEXT 
fixes dEQP-VK.transform_feedback.simple.draw_indirect*counter_offset*

cc: mesa-stable

Reviewed-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35076>
 2025-05-23 14:53:02 +00:00
Mike Blumenkrantz
753d3e71d3 llvmpipe: disable conditional rendering mem for blits 
u_blitter doesn't support this, and changing u_blitter to support a niche
lavapipe feature seems like overkill

fixes dEQP-VK.conditional_rendering.conditional_ignore.resolve_image*

cc: mesa-stable

Reviewed-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35076>
 2025-05-23 14:53:02 +00:00
José Roberto de Souza
ddca50584c intel: Return PTL stepping 
Without this no temporary workaround is applied to PTL as by
default INTEL_STEPPING_RELEASE is returned and it is larger than
any stepping.

Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35110>
 2025-05-23 13:52:27 +00:00
Mike Blumenkrantz
c78d6bc4c3 ci: disable panfrost-t720-gles2 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34054>
 2025-05-23 13:21:29 +00:00