51a7aebc86
tu: Refactor BO deletion
...
For VM_BIND, BO deletion will have to be implemented differently in
native drm and virtio. We already have a somewhat awkward situation with
native-specific code in the common BO deletion helper, which we only get
away with because it's for kernels without SET_IOVA in which case virtio
isn't supported. Add a few common helpers for some of the guts, and move
the guts into backend-specific functions.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32533 >
2025-08-25 20:11:57 +00:00
e689b342fe
etnaviv: Optimize sampler view iteration with u_foreach_bit(..)
...
Replace loop over all PIPE_MAX_SAMPLERS with u_foreach_bit(..) to iterate
only over active sampler views. This avoids unnecessary iterations.
Improves drawoverhead test 1 performance by ~10%.
Signed-off-by: Christian Gmeiner <cgmeiner@igalia.com >
Reviewed-by: Lucas Stach <l.stach@pengutronix.de >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36831 >
2025-08-25 19:21:09 +00:00
a6e0cbd110
radv: consolidate cooperative matrix array sizes enumeration
...
Share the code between khr and nv paths to not have to recreate it.
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36544 >
2025-08-25 18:55:08 +00:00
a55c036450
radv: add support for coopmat2 flexible dimensions
...
This allows matricies that are multiples of a base size.
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36544 >
2025-08-25 18:55:08 +00:00
7a96a928a2
nir: add coop mat flexible dimensions lowering.
...
This adds a generic lowering pass for coop mat flexible dimensions.
This should be suitable for all drivers that implement coop mat2 flexible dimensions
or even just lowering sw exposed sizes to hw sizes.
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36544 >
2025-08-25 18:55:08 +00:00
39ffc3f629
zink: when sparse unbinding, always wait on main timeline semaphore
...
this ensures synchronization between gfx/compute work and the sparse queue
fixes KHR-GL46.sparse_buffer_tests.BufferStorageTest_cases_*. on turnip
cc: mesa-stable
Closes : #13747
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36954 >
2025-08-25 14:35:25 -04:00
2cf626d17d
zink: pass ctx to sparse bind functions
...
no functional changes
cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36954 >
2025-08-25 14:35:25 -04:00
c9746103f5
zink: break out unflushed batch waiting into separate function/mechanism
...
this is useful on its own
no functional changes
cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36954 >
2025-08-25 14:35:25 -04:00
483d53e0e3
zink: ALWAYS_INLINE resource inlines
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36954 >
2025-08-25 14:35:25 -04:00
f9bf524a6d
zink: inline zink_resource_access_is_write()
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36954 >
2025-08-25 14:35:25 -04:00
bef37336fb
vulkan/queue: Fix VkTimelineSemaphoreSubmitInfo sanitization
...
We're supposed to completely ignore VkTimelineSemaphoreSubmitInfo if
there aren't any timeline semaphores, including the array lengths, which
is made clear by the various VUs already cited by the code. The
vkQueueSubmit() path correctly handled this when asserting but still
dereferenced pWaitSemaphoreValues unconditionally, which could lead to
dereferencing an invalid pointer if waitSemaphoreValueCount is less than
waitSemaphoreCount. The vkQueueSparseBind() path didn't even assert
correctly. Bring vkQueueSparseBind() in line with vkQueueSubmit()
and make both only dereference the wait/signal array once we've
determined it must be present. While we're here, also fix the assert in
vkQueueSubmit() to disallow a waitSemaphoreValueCount of 0 if there are
timeline semaphores present, which conversely is not allowed.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36989 >
2025-08-25 17:57:47 +00:00
2cd564c1de
anv: Add missing L3 flushes
...
We are reading out some of the parameters from IR data structure those
have been written previously, on some platforms L3 is not coherent, so
explicitly add those flushes.
Cc: mesa-stable
Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com >
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36952 >
2025-08-25 17:36:08 +00:00
4473e21e2f
anv: Enable CS stall for ACCELERATION_STRUCTURE_COPY stage
...
Cc: mesa-stable
Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com >
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36952 >
2025-08-25 17:36:08 +00:00
75d770b4f8
anv: Add missing ACCELERATION_STRUCTURE_READ in barrier handling
...
Cc: mesa-stable
Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com >
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36952 >
2025-08-25 17:36:08 +00:00
fee83f4fcc
d3d12: Move logicop descriptor initialization to after all blits
...
The blit code is destructive on the framebuffer state, which means
that it can set new render targets. If you have 2 BGRA surfaces bound
for logic ops, then after setting up the surface for the first one,
the blit for the second will end up destroying + re-creating the
surface for the first one.
Let's be robust to this by putting the blit in a first pass, and
then actually initializing all of the descriptors in a second pass.
This is still woefully inefficient but at least it's correct.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36945 >
2025-08-25 17:21:37 +00:00
cd3885c592
d3d12: Move logicop emulation resource from surface to resource
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36945 >
2025-08-25 17:21:37 +00:00
784f93653a
d3d12: Support more logic op formats
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36945 >
2025-08-25 17:21:36 +00:00
91c063493d
d3d12: Stop using util_framebuffer_init
...
For https://gitlab.freedesktop.org/mesa/mesa/-/issues/13262
This just inlines the existing logic. A surface cache needs to be
introduced next.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36945 >
2025-08-25 17:21:36 +00:00
635ac758c9
aco/optimizer: don't create undef copies from p_create_vector
...
p_create_vector allows undef operands, p_parallelcopy doesn't.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/13765
Fixes: 01d20680e2pendingchaos02@gmail.com >
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36963 >
2025-08-25 16:47:38 +00:00
8903bb4618
aco/optimizer: don't apply packed clamp to v_fma_mix
...
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/13758
Fixes: 345bf8a2f2pendingchaos02@gmail.com >
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36963 >
2025-08-25 16:47:38 +00:00
cde83062ad
ci/android: enable cross-building nvk and panvk for Android
...
Enable cross-building nvk and panvk for Android to make it easier to
spot build regressions.
NOTE: nvk is built only for x86_64, and panvk is built only for aarch64.
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Reviewed-by: Valentine Burley <valentine.burley@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36710 >
2025-08-25 15:55:55 +00:00
289913547c
ci: bump DEBIAN_BUILD_TAG to include all the android/rust changes for nvk and panvk
...
Reviewed-by: Valentine Burley <valentine.burley@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36710 >
2025-08-25 15:55:55 +00:00
ca0f0d7b95
meson: handle dep_libdrm before the driver specific libdrm modules
...
Handle dep_libdrm before the driver specific libdrm modules, this is for
when libdrm is built as a subproject, for example libdrm_amdgpu is only
available after dep_libdrm has been added.
Acked-by: Valentine Burley <valentine.burley@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36710 >
2025-08-25 15:55:54 +00:00
9d26eb9a2f
subprojects: fix ignore exception for files under packagefiles/
...
Fix the rule about not ignoring files under subprojects/packagefiles/
which is not covering files in subdirectories of
subprojects/packagefiles
Acked-by: Valentine Burley <valentine.burley@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36710 >
2025-08-25 15:55:53 +00:00
1dd1995388
nvk: silence error when cross-building for Android
...
When cross-building NVK for Android the compiler emits this error:
-----------------------------------------------------------------------
../src/nouveau/vulkan/nvk_image.c:1084:38: error: unused variable 'pdev' [-Werror,-Wunused-variable]
1084 | const struct nvk_physical_device *pdev = nvk_device_physical(dev);
| ^~~~
1 error generated.
-----------------------------------------------------------------------
Mark the variable definition as UNUSED to silence the error.
Acked-by: Valentine Burley <valentine.burley@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36710 >
2025-08-25 15:55:53 +00:00
5ae3343590
ci/android: add comment about updating tags to create-android-cross-file.sh
...
Add a comment about updating tags in .gitlab-ci/image-tags.yml to the
.gitlab-ci/container/create-android-cross-file.sh as it is not immediate
that it gets called from .gitlab-ci/container/debian/android_build.sh
which also requires to update DEBIAN_BUILD_TAG.
Reviewed-by: Valentine Burley <valentine.burley@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36710 >
2025-08-25 15:55:53 +00:00
3cd4cc7d07
ci/android: add rust compiler to create-android-cross-file.sh
...
Add rust compiler definitions to the create-android-cross-file.sh file
to be able to cross-build drivers that use Rust code.
This is inspired by the changes in commit 2f5db959a7faith.ekstrand@collabora.com >
Reviewed-by: Valentine Burley <valentine.burley@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36710 >
2025-08-25 15:55:53 +00:00
fbbb5d15f0
ci: Add missing rust subprojects to meson/build.sh
...
Reviewed-by: Valentine Burley <valentine.burley@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36710 >
2025-08-25 15:55:52 +00:00
dfab0e952b
zink: sprinkle in a bunch of MESA_TRACE_FUNC
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36984 >
2025-08-25 15:34:20 +00:00
9d0203bace
ci/piglit: automatically use baremetal proxy
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36955 >
2025-08-25 14:52:38 +00:00
fa74e939bf
ci/piglit: automatically use LAVA proxy
...
This avoids having to hardcode the proxy in the traces `download-url` or
jobs setting `PIGLIT_REPLAY_EXTRA_ARGS` and accidentally overriding the
default args when the author meant to append.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36955 >
2025-08-25 14:52:38 +00:00
b4e49664f4
vmware/ci: fix caching proxy url
...
FDO_HTTP_CACHE_URI is not defined, but LAVA_HTTP_CACHE_URI is and is the
right URL for this.
This job is currently disabled, but fix it in preparation for when
someone eventually brings it back.
Note that this line also has another bug that will be addressed by the
next commit.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36955 >
2025-08-25 14:52:37 +00:00
2851a305fd
virgl/ci: drop invalid but overridden empty caching proxy
...
The variable is undefined, leading to an empty proxy being set, but
luckily it's actually overridden back to the default args, so this was
just dead code.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36955 >
2025-08-25 14:52:37 +00:00
af203146f2
panvk: properly handle errors from utrace_context_init
...
Without this, I get failures in the following CTS test:
dEQP-VK.api.device_init.create_instance_device_intentional_alloc_fail.basic
Fixes: 05006c21ddchristoph.pillmayer@arm.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36980 >
2025-08-25 14:15:57 +00:00
e1fcc403d4
radeonsi/vcn: Fix HEVC encode transform_skip with old FW
...
Cc: mesa-stable
Reviewed-by: Ruijing Dong <ruijing.dong@amd.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36845 >
2025-08-25 13:55:40 +00:00
26fc4798c1
radeonsi/vcn: Fix HEVC encode cu_qp_delta with old FW
...
VCN2 needs interface version bump to enable it.
Cc: mesa-stable
Reviewed-by: Ruijing Dong <ruijing.dong@amd.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36845 >
2025-08-25 13:55:39 +00:00
e945bb3649
radeonsi/vcn: Fix compatibility with old FW for encode
...
The interface version can't be higher than the FW version, otherwise
FW will reject all encode jobs.
Cc: mesa-stable
Reviewed-by: Ruijing Dong <ruijing.dong@amd.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36845 >
2025-08-25 13:55:39 +00:00
51d5c0d537
radeonsi/vcn: Disable H264/5 constrained intra pred with rate control
...
There is a FW issue when using constrained intra prediction with rate
control enabled, causing unexpected quality degradation.
Disable it until FW fix is available.
Cc: mesa-stable
Reviewed-by: Ruijing Dong <ruijing.dong@amd.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36837 >
2025-08-25 13:34:00 +00:00
5cde06b7b1
ci-tron: drop meaningless timestamp in initial section message
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36903 >
2025-08-25 12:46:54 +00:00
78de1e93ee
ci/ci-tron: uprev the job submission template
...
This brings some nice log readability improvements which were asked by
various developers.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36903 >
2025-08-25 12:46:54 +00:00
102892a458
radv: clear RADV_CMD_DIRTY_xxx bits outside of the caller in most cases
...
Prevent risks if an early return is added at some point.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36912 >
2025-08-25 12:17:39 +00:00
3ecf9a17f8
radv: stop abusing dirty_dynamic when binding a NULL fragment shader
...
RADV_DYNAMIC_RASTERIZATION_SAMPLES triggers the following states:
- FS (not needed when FS is NULL)
- MSAA (already triggered)
- BINNING (use radv_get_ps_iter_samples() on GFX9)
- OCCLUSION_QUERY (doesn't use the PS)
- DB_SHADER_CONTROL (already triggered)
- RAST_SAMPLES (use radv_get_ps_iter_samples())
- NGGC (doesn't use the PS)
So this can be simplified to BINNING (gfx9) | RAST_SAMPLES.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36912 >
2025-08-25 12:17:38 +00:00
19c712c8ef
radv: rename rast_prim to vgt_outprim_type everywhere
...
To avoid confusion between the primitive topology and the output
rasterized primitive.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36912 >
2025-08-25 12:17:38 +00:00
5dec043cee
radv: dirty the rast samples state when VRS is forced to 1x1
...
Only these two states use force_sample_iter_shading_rate.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36912 >
2025-08-25 12:17:37 +00:00
8120d14ee0
radv: dirty RADV_CMD_DIRTY_xx states when binding sample shading state
...
This reduces the number of states that are re-emitted but the logic
is mostly duplicated because sample shading can be set from the
fragment shader or the graphics pipeline. Could be refactored
eventually.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36912 >
2025-08-25 12:17:34 +00:00
4afff72e91
radv: dirty the rast sample states for VRS att/OOO rast
...
These fields are only used by radv_emit_rast_samples_state().
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36912 >
2025-08-25 12:17:32 +00:00
4dd05c9a91
radv: remove dead code when setting dynamic primitive topology
...
This is already handled slightly above in the same function. Also
state->dirty isn't for RADV_DYNAMIC_xxx and there is no corresponding
RADV_CMD_DIRTY_xxx either.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36912 >
2025-08-25 12:17:31 +00:00
dbcc9aa942
radv: remove radv_cmd_buffer_flush_dynamic_state()
...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36912 >
2025-08-25 12:17:31 +00:00
12f8508363
radv: add a new dirty bit for the VGT prim state
...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36912 >
2025-08-25 12:17:31 +00:00
4096009ea1
radv: add a new dirty bit for the patch control points state
...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36912 >
2025-08-25 12:17:30 +00:00
Connor Abbott