AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
214,552 Commits 1 Branch 0 Tags
0cb1fca8fafe43e6ef87f585e2da79936997cb85
Commit Graph

214552 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Natalie Vock
0cb1fca8fa nir: Use sparse bitset for liveness information 
Some shaders, especially RTPSO shaders that have parts of the PSO
inlined, can become absolutely huge. Using a sparse bitset avoids
quadratic complexity in memory consumption for the liveness information.

This reduces peak memory usage in worst-case tests (hammering
compilation of many huge RTPSOs on 32 threads concurrently) by ~60%,
from 43GB to 18GB.

CPU time (seconds) differences for a workload with mostly small shaders:
Difference at 95.0% confidence
        -5.27 +/- 1.08963
        -0.88811% +/- 0.183626%
        (Student's t, pooled s = 0.629735)

Peak resident set usage for the mostly-small workload:
Difference at 95.0% confidence
        30809 +/- 13394.3
        1.59276% +/- 0.69246%
        (Student's t, pooled s = 7741.09)

CPU time for the heavy workload did not show any difference.

Co-authored-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37908>
 2025-11-06 21:34:33 +00:00
Natalie Vock
a8b75dd0f4 util: Add sparse bitset data structure 
Useful for potentially huge bitsets that are expected to be mostly
filled with zeroes, reducing memory consumption by assuming bits being
zero by default (without wasting memory to store zeroes).

Co-authored-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37908>
 2025-11-06 21:34:33 +00:00
Natalie Vock
1920a99115 util/bitset: Wrap __size in braces 
Otherwise funny things can happen with the < operator because of
precedence rules.

Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37908>
 2025-11-06 21:34:33 +00:00
Dylan Baker
12e22d5bc1 anv: prevent potential, but unlikely, overflow 
The code in question multiplies `uint32_t`s together and assigns them to
a `uint64_t`. It seems rather unlikely at there would be an overflow,
but we might as well do the cast.

CID: 1649587
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38289>
 2025-11-06 21:07:11 +00:00
Dylan Baker
d5199b07d2 anv: assert that we don't overflow 
Our exposed limits say we shouldn't be able to, but let's add an assert
in case something changes, and to help Coverity out.

CID: 1662103
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37583>
 2025-11-06 20:45:41 +00:00
Danylo Piliaiev
01cbd0f24a tu: Fix renderpass-level tracepoints not showing up in binning 
They should be cloned not only per-tile but also for binning IB.

Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38240>
 2025-11-06 19:05:49 +00:00
Danylo Piliaiev
c04e375588 tu: Use cmd->rp_trace u_trace for draw calls 
Fixes: 707c97f634 ("tu: Add tracepoints around draws, with shader sha1s.")

Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38240>
 2025-11-06 19:05:49 +00:00
Yiwei Zhang
4ec2a921d3 panvk: fix mem alloc size for VkBuffer backed by imported blob AHB 
For AHB VkBuffer import, the allocationSize comes from the raw external
AHB props query and it can be larger than the underlying buffer memory
requirement. So we must respect the allocationSize for the actual mem
import to support mapping the whole AHB size, and the dedicated buffer
info has to be stripped to obey the spec.

Test: CtsNativeHardwareTestCases no longer crashes on debug build panvk

Fixes: 66bbd9eec8 ("panvk: implement AHB image deferred init and memory alloc")
Tested-by: Valentine Burley <valentine.burley@collabora.com>
Reviewed-by: Lars-Ivar Hesselberg Simonsen <lars-ivar.simonsen@arm.com>
Reviewed-by: Christoph Pillmayer <christoph.pillmayer@arm.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38274>
 2025-11-06 18:26:01 +00:00
Alyssa Rosenzweig
2d98d44e63 brw,elk: drop unused spirv->nir routines 
Unused since switching to vtn_bindgen2.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38246>
 2025-11-06 17:47:41 +00:00
Emma Anholt
31a4a5ee8c docs: Give more reproducible instructions for how to build the docs. 
I routinely don't update the docs because the build (hawkmoth in
particular) is a pain.  The prior instructions almost worked, except that
on Debian you need to use a venv (which is a good idea to do for py3-clang
as well for reproducibility, anyway), and the versions that were listed
didn't actually run any more due to a revoked dependency.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38266>
 2025-11-06 17:42:45 +00:00
Faith Ekstrand
4909af6bae panvk: Advertise VK_KHR_pipeline_binary 
Closes: https://gitlab.freedesktop.org/panfrost/mesa/-/issues/216
Reviewed-by: Lars-Ivar Hesselberg Simonsen <lars-ivar.simonsen@arm.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38284>
 2025-11-06 11:33:55 -05:00
Faith Ekstrand
acd00c07f6 panvk: Initialize the disk cache earlier 
We want to know whether or not we successfully initialized before
filling out physical device properties.

Reviewed-by: Lars-Ivar Hesselberg Simonsen <lars-ivar.simonsen@arm.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38284>
 2025-11-06 11:33:55 -05:00
Aksel Hjerpbakk
bbe6fff677 panvk: include cmd stages for semaphores on submit 
CTS regressed due to c2a6fb6419 not considering cmd stages for
signaling and waiting. Before this patch panvk did consider cmd stages,
but not semaphore stage masks.

With this fix we consider both which is what the spec requires

Fixes: c2a6fb6419 ("panvk: cull semaphores in unrelated subqueues")

Reviewed-by: Lars-Ivar Hesselberg Simonsen <lars-ivar.simonsen@arm.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38138>
 2025-11-06 15:57:22 +00:00
Faith Ekstrand
60ad7e8da2 nvk: Advertise VK_KHR_pipeline_binary 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36647>
 2025-11-06 15:27:29 +00:00
Lionel Landwerlin
21aafaea16 anv: enable KHR_pipeline_binary support 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/12802
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36647>
 2025-11-06 15:27:29 +00:00
Faith Ekstrand
cb7df84430 vulkan/runtime: Add an environment variable to validate shader binaries 
Setting MESA_VK_VALIDATE_SHADER_BINARIES will cause the shader code to
round-trip every shader through [de]serialize and only ever use the
deserialized version.  This catches bugs where the driver may drop
things in the [de]serialization process.  It also deserializes the new
shader again and compares it against the original to ensure that
deserialize -> serialize is idempotent.

Acked-by: Eric R. Smith <eric.smith@collabora.com>
Reviewed-by: Lars-Ivar Hesselberg Simonsen <lars-ivar.simonsen@arm.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36647>
 2025-11-06 15:27:29 +00:00
Faith Ekstrand
59a89cd762 vulkan/runtime: Add a vk_compile_shaders() helper 
This is just a wrapper around ops->compile() for now but we'll extend it
in the next commit to add some validation.

Acked-by: Eric R. Smith <eric.smith@collabora.com>
Reviewed-by: Lars-Ivar Hesselberg Simonsen <lars-ivar.simonsen@arm.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36647>
 2025-11-06 15:27:29 +00:00
Lionel Landwerlin
5c47ac640b vulkan/runtime: implement VK_KHR_pipeline_binary 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Co-Authored-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36647>
 2025-11-06 15:27:28 +00:00
Lionel Landwerlin
2e42e03cec vulkan/runtime: track imported stages 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36647>
 2025-11-06 15:27:28 +00:00
Lionel Landwerlin
708cc72b11 vulkan/runtime: switch precomp shaders to blake3 hashes 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36647>
 2025-11-06 15:27:27 +00:00
Lionel Landwerlin
e9c1947ed6 vulkan/runtime: use only blake3_hash to shader key 
To match the VK_MAX_PIPELINE_BINARY_KEY_SIZE_KHR of only 32B.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36647>
 2025-11-06 15:27:26 +00:00
Lionel Landwerlin
e05a9b77b6 vulkan/runtime: split rt shaders hashing from compile 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36647>
 2025-11-06 15:27:26 +00:00
Lionel Landwerlin
08ed1c3da2 vulkan/runtime: split graphics shaders hashing from compile 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36647>
 2025-11-06 15:27:25 +00:00
Lionel Landwerlin
b2d6ead1ee vulkan/runtime: split compute shader hashing from compile 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36647>
 2025-11-06 15:27:24 +00:00
Faith Ekstrand
440e71bdbd vulkan/runtime: Add a get_push_range_for_stage() helper 
This is already duplicated a few times and we're about to duplicate it
more.

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36647>
 2025-11-06 15:27:23 +00:00
Lionel Landwerlin
9a5b0bbba4 vulkan/runtime: use stage flags to track valid stages 
We'll want to have only hashes in vk_pipeline_stage so the
vk_pipeline_stage_is_null() helper won't work.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36647>
 2025-11-06 15:27:23 +00:00
Lionel Landwerlin
8e93938c3f vulkan/runtime: keep the set layouts on the stack until pipeline creation 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36647>
 2025-11-06 15:27:21 +00:00
Lionel Landwerlin
ab0bcefab1 vulkan/runtime: split precomp shader hashing from precomp loading 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36647>
 2025-11-06 15:27:21 +00:00
Lionel Landwerlin
cbc8ec2cc4 vulkan/runtime: drop blake3 hash on precomp shaders 
In order to implement VK_KHR_pipeline_binary, we need to be able to
build hash from pipeline creation structures without looking at the
cache.

The blake3 hash on precomp shaders prevents that as its loading from
cache and potentially apply transformation to NIR.

Let's stick to the hash generated by vk_pipeline_hash_shader_stage(),
it does not look at NIR (except for internal shaders) and already hash
the same information :
  * shader code (SPIR-V, identifier, hash)
  * robustness state
  * specialization constants
  * pipeline flags
  * entry point name
  * subgroup information

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36647>
 2025-11-06 15:27:20 +00:00
Lionel Landwerlin
6279645fed vulkan/runtime: drop some geometry shader hashing 
Following bc64ea2815 ("vulkan: fix shader linking with common
pipelines") we're always linking pre-rasterization shaders together
and the shader hashes are hashed together, so there is no point
hashing :
  - a bitfield of active shaders
  - merged tesselation information

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36647>
 2025-11-06 15:27:19 +00:00
Lionel Landwerlin
fc6d17a290 vulkan/runtime: simplify robustness state hashing 
We're doing the same in vk_pipeline_precomp_shader_create().

Also fixes valgrind warning due to uninitialized fields

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: mesa-stable
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36647>
 2025-11-06 15:27:19 +00:00
Lionel Landwerlin
f56e118ecd vulkan/runtime: split out partitioning logic 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36647>
 2025-11-06 15:27:18 +00:00
Faith Ekstrand
69d7fcd613 pan: Move point size and viewport lowering to postprocess 
Panvk calls pan_preprocess_nir() from its preprocess hook that it hands
off to the Vulkan pipeline code.  That hook gets called before we have
the opportunity to lower geometry shaders.  This means that we get our
viewports lowered for the VS and then the geometry shader is trying to
work on lowered viewports, which is wrong.  Instead, we want to lower
later and only apply the viewport transform in the shader that runs as
the hardware VS.

Reviewed-by: Olivia Lee <olivia.lee@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38265>
 2025-11-06 14:57:32 +00:00
Faith Ekstrand
6c5f981ba8 pan/bi: Move lower_noperspective*() to postprocess() 
Reviewed-by: Olivia Lee <olivia.lee@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38265>
 2025-11-06 14:57:32 +00:00
Faith Ekstrand
0ccadf7a86 nir: Check the deref mode in lower_point_size() 
This is more robust because it ensures that we only ever check the
location on something that we know is an outupt.  Also, if it's an
output then we know (thanks, validation!) that it's a variable.

Reviewed-by: Olivia Lee <olivia.lee@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38265>
 2025-11-06 14:57:31 +00:00
Faith Ekstrand
5ed35866c2 nir: Handle lowered I/O in lower_viewport_transform() 
While we're here, make the variable handling a little more robust by
checking the deref mode before assuming there's a reachable variable.

Reviewed-by: Olivia Lee <olivia.lee@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38265>
 2025-11-06 14:57:31 +00:00
Faith Ekstrand
949a4458ea pan/bi: Call constant folding in postprocess() 
The comment explicitly calls out pan_nir_lower_store_component(), which
is in a different function call so it's a bit weird to have it in the
caller.  Also, we already do this in postprocess() on midgard so it
makes more sense to just move it into bifrost.

Reviewed-by: Olivia Lee <olivia.lee@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38265>
 2025-11-06 14:57:31 +00:00
Faith Ekstrand
0fae56e100 pan: roll lower_texture() into postprocess() 
Every caller of pan_shader_lower_texture() immediatly called
pan_shader_postprocess() and every caller of pan_shader_postprocess()
lowered textures except blend shaders and those don't texture anyway.

Reviewed-by: Olivia Lee <olivia.lee@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38265>
 2025-11-06 14:57:30 +00:00
Samuel Pitoiset
dcb9ca1aa6 radv/ci: remove RADV_PERFTEST=video_encode,video_code for GFX6-7 
It's not supported at all, so no need to force enable all video
extensions.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38280>
 2025-11-06 14:35:19 +00:00
Daniel Schürmann
f37bdd664d amd/common: rename ac_fake_hw_db.h -> ac_surface_test.h 
As the surface tests are the only user now.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37525>
 2025-11-06 13:53:07 +00:00
Daniel Schürmann
7e0ba40264 radeonsi: use ac_null_device_create() when AMD_FORCE_FAMILY is set 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37525>
 2025-11-06 13:53:07 +00:00
Daniel Schürmann
4d68056e83 radv: delete winsys/null/* 
The null device works without winsys, and now can be found in
amd/common/ac_null_device.c.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37525>
 2025-11-06 13:53:07 +00:00
Daniel Schürmann
7d6de8b17e amd, radv: create null device without winsys 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37525>
 2025-11-06 13:53:07 +00:00
Daniel Schürmann
04ebbc3c41 radv: remove radeon_winsys::get_chip_name() and use info->marketing_name directly 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37525>
 2025-11-06 13:53:07 +00:00
Daniel Schürmann
797ddfe4d2 radv/device: return early in radv_CreateDevice() if creating a null device 
Also reorder initialization, so that everything required for compilation
is done first.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37525>
 2025-11-06 13:53:07 +00:00
Daniel Schürmann
23ef756496 radv/null_device: set more options which affect compilation 
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37525>
 2025-11-06 13:53:06 +00:00
Daniel Schürmann
4ad9fb9d2e radv/null_device: don't attempt to upload shaders 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37525>
 2025-11-06 13:53:06 +00:00
Daniel Schürmann
6c2d614412 radv: hash keep_executable_info into shader key rather than device cache key 
for consistency.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37525>
 2025-11-06 13:53:06 +00:00
Daniel Schürmann
5ef8930636 radv: skip shader cache if trap handler is enabled 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37525>
 2025-11-06 13:53:06 +00:00
Jose Maria Casanova Crespo
a58db214bb v3dv: Enable VK_FORMAT_B8G8R8A8_SNORM format 
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38278>
 2025-11-06 13:34:53 +00:00