Pierre-Eric Pelloux-Prayer
84dcc8b940
ac/info: add ac_fill_bug_info
...
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40656 >
2026-04-14 12:06:17 +00:00
Pierre-Eric Pelloux-Prayer
a26ba344a3
ac/info: remove has_bo_metadata
...
It's unused.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40656 >
2026-04-14 12:06:16 +00:00
Pierre-Eric Pelloux-Prayer
44334d6de0
ac/info: move more memory properties to ac_fill_memory_info
...
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40656 >
2026-04-14 12:06:15 +00:00
Pierre-Eric Pelloux-Prayer
5abf02362f
ac/info: add ac_identify_chip
...
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40656 >
2026-04-14 12:06:14 +00:00
Pierre-Eric Pelloux-Prayer
360e9503d8
ac/info: add ac_fill_hw_ip_info
...
num_instances is initialized to 1 and overriden in ac_query_gpu_info.
has_graphics needs to be set early as other info fields are determined
based on its value.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40656 >
2026-04-14 12:06:14 +00:00
Pierre-Eric Pelloux-Prayer
2bc0e5437d
ac/info: add ac_fill_memory_info
...
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40656 >
2026-04-14 12:06:13 +00:00
Pierre-Eric Pelloux-Prayer
f9c60f3699
ac/info: add ac_fill_tiling_info
...
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40656 >
2026-04-14 12:06:12 +00:00
Pierre-Eric Pelloux-Prayer
79ebbf587e
ac: remove ac_null_device
...
Prefer AMDGPU shim.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40656 >
2026-04-14 12:06:12 +00:00
Pierre-Eric Pelloux-Prayer
42d4ec080c
winsys/amdgpu: remove assert
...
This was useful to validate native context fence code, but
now it has less value and triggers when drm-shim is used.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40656 >
2026-04-14 12:06:10 +00:00
Pierre-Eric Pelloux-Prayer
d97688b6b3
drm-shim: fix shim on GLX
...
xcb_dri3_open_reply_fds will dup the fd received from the server,
so we need to detect this case.
Reviewed-by: Emma Anholt <emma@anholt.net>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40656 >
2026-04-14 12:06:10 +00:00
Aitor Camacho
2ec0925bc1
kk: Fix pre-gfx encoder dependency with gfx encoder
...
macOS-CI / macOS-CI (dri) (push) Waiting to run
macOS-CI / macOS-CI (xlib) (push) Waiting to run
Signed-off-by: Aitor Camacho <aitor@lunarg.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40914 >
2026-04-14 10:48:29 +00:00
Samuel Pitoiset
d39d846cfa
radv/ci: set RADV_EXPERIMENTAL=heap
...
Current VKCTS main doesn't have any tests for that, but next uprev
should contain a bunch and using that envvar will allow us to catch
them easily.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39483 >
2026-04-14 10:10:29 +00:00
Samuel Pitoiset
f4f2a966e3
radv: advertise VK_EXT_descriptor_heap with RADV_EXPERIMENTAL=heap
...
It's not yet enabled by default because it's quite a big extension
and I expect bugs because test coverage isn't very good. It will be
enabled by default in one or two Mesa releases when it's more stable.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39483 >
2026-04-14 10:10:29 +00:00
Samuel Pitoiset
67ae52cdc2
radv: add support for DGC with descriptor heap
...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39483 >
2026-04-14 10:10:28 +00:00
Samuel Pitoiset
17ed105a9b
radv: flush caches with descriptor heap access flags
...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39483 >
2026-04-14 10:10:28 +00:00
Samuel Pitoiset
ef03ca8dab
radv: add support for inherited descriptor heap for secondaries
...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39483 >
2026-04-14 10:10:28 +00:00
Samuel Pitoiset
5df943629d
radv: add support for capture&replay with descriptor heap
...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39483 >
2026-04-14 10:10:28 +00:00
Samuel Pitoiset
7ebf99362d
radv: implement vkGetPhysicalDeviceDescriptorSizeEXT()
...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39483 >
2026-04-14 10:10:28 +00:00
Samuel Pitoiset
d557f21bb6
radv: implement vkCmdPushDataEXT()
...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39483 >
2026-04-14 10:10:28 +00:00
Samuel Pitoiset
5704ad905d
radv: add support for emitting descriptor heaps
...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39483 >
2026-04-14 10:10:27 +00:00
Samuel Pitoiset
86e8d171fc
radv: implement vkCmdBind{Resource,Sampler}HeapEXT()
...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39483 >
2026-04-14 10:10:26 +00:00
Samuel Pitoiset
7e63a2050b
radv: implement vkWrite{Resource,Sampler}DescriptorsEXT()
...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39483 >
2026-04-14 10:10:26 +00:00
Samuel Pitoiset
7da95418a5
radv: copy mapping info for graphics pipeline libraries
...
In case only SPIR-V only is imported.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39483 >
2026-04-14 10:10:26 +00:00
Samuel Pitoiset
aeda29c8f7
radv: allow to create pipelines with a NULL pipeline layout
...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39483 >
2026-04-14 10:10:26 +00:00
Samuel Pitoiset
65415b1c89
radv: set descriptor heap sizes/alignments for VTN
...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39483 >
2026-04-14 10:10:25 +00:00
Samuel Pitoiset
652af97f8b
radv/nir: lower descriptor heap in radv_nir_lower_descriptors
...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39483 >
2026-04-14 10:10:25 +00:00
Samuel Pitoiset
8588ddf0df
radv/nir: rename radv_nir_apply_pipeline_layout
...
to radv_nir_lower_descriptors. This is more generic and descriptor
heap support will be added to the same pass.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39483 >
2026-04-14 10:10:25 +00:00
Samuel Pitoiset
0f2f9a9045
radv/nir: adjust lowering of immediate samplers for descriptor heap
...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39483 >
2026-04-14 10:10:25 +00:00
Samuel Pitoiset
a44303ff11
radv/nir: adjust lowering of ycbcr tex instructions for descriptor heap
...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39483 >
2026-04-14 10:10:25 +00:00
Samuel Pitoiset
54e2a1d2e4
radv: call vk_nir_lower_descriptor_heaps()
...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39483 >
2026-04-14 10:10:24 +00:00
Samuel Pitoiset
eac2c1f141
radv: keep track of descriptor heap mapping in the shader layout
...
This will be used by the common lowering.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39483 >
2026-04-14 10:10:24 +00:00
Samuel Pitoiset
cebac5a427
radv/rt: declare shader arguments for resource/sampler heaps
...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39483 >
2026-04-14 10:10:24 +00:00
Samuel Pitoiset
cdfb9a24ba
radv: declare shader arguments for resource/sampler heaps
...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39483 >
2026-04-14 10:10:23 +00:00
Samuel Pitoiset
b29743f91c
radv: add shader info about whether descriptor heap is used
...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39483 >
2026-04-14 10:10:23 +00:00
Samuel Pitoiset
62629d3e3c
radv: use 32-bit memory types for descriptor heap buffers
...
Descriptors must be allocated in the 32-bit addr space.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39483 >
2026-04-14 10:10:23 +00:00
Samuel Pitoiset
e81c375b4a
radv: make radv_make_sampler_descriptor() non-static
...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39483 >
2026-04-14 10:10:23 +00:00
Samuel Pitoiset
1f15a71afe
radv: add support for custom border colors with descriptor heap
...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39483 >
2026-04-14 10:10:23 +00:00
Samuel Pitoiset
8da71ae5a2
radv: add a new helper to make a sampler descriptor
...
For future work.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39483 >
2026-04-14 10:10:22 +00:00
Samuel Pitoiset
afd736bfd7
radv: zero-allocate graphics shader stages
...
This can prevent issues with unitialized memory. RT and compute
stages are already zero-initialized.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39483 >
2026-04-14 10:10:22 +00:00
Samuel Pitoiset
0b016f4bff
nir: add new system values for descriptor heap RT traversal inputs
...
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/39483 >
2026-04-14 10:10:22 +00:00
Erik Faye-Lund
dc4efab205
pan/ci: update expectations
...
These are based on nightlies. It's unclear if one of the t760 flakes
should really be a fail or a flake, but I have another MR that tries to
clean up the flakes pending; I'll address that there, as there's other
related flakes that probably should change as well.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40943 >
2026-04-14 08:51:52 +00:00
Lars-Ivar Hesselberg Simonsen
a23cebcd60
panfrost: Add support for 64 bit gpu_id
...
macOS-CI / macOS-CI (dri) (push) Waiting to run
macOS-CI / macOS-CI (xlib) (push) Waiting to run
Update the gallium driver's use of gpu_id to support 64 bit gpu_ids.
Reviewed-by: Marc Alcala Prieto <marc.alcalaprieto@arm.com>
Reviewed-by: Christoph Pillmayer <christoph.pillmayer@arm.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40921 >
2026-04-14 10:12:25 +02:00
Lars-Ivar Hesselberg Simonsen
fd3aafabe9
pan/model: Expose prod_id and rev functions
...
Some code in gallium was making assumptions of how the gpu_id is laid
out, which will not work for 64 bit gpu_ids.
Expose pan_prod_id and pan_rev from the model to collect this logic in a
single place.
Reviewed-by: Marc Alcala Prieto <marc.alcalaprieto@arm.com>
Reviewed-by: Christoph Pillmayer <christoph.pillmayer@arm.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40921 >
2026-04-14 10:12:00 +02:00
Erik Faye-Lund
739e3d20f0
docs/panfrost: fix heading-levels
...
macOS-CI / macOS-CI (dri) (push) Waiting to run
macOS-CI / macOS-CI (xlib) (push) Waiting to run
These headings shouldn't all be on the same level; the first one is the
heading for the article, but the subsequent ones should be subheadings.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40933 >
2026-04-14 05:39:23 +00:00
Tapani Pälli
745dc66b69
anv: fix Wa_14024015672 interaction in blorp
...
macOS-CI / macOS-CI (dri) (push) Waiting to run
macOS-CI / macOS-CI (xlib) (push) Waiting to run
This was causing issue within hiz clears, apply workaround only
when doing RT writes (not for WM_HZ ops).
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/work_items/14533
Fixes: 7be8af1dad ("anv: deal with Wa 14024015672 on the blorp path")
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40913 >
2026-04-14 03:15:14 +00:00
Karol Herbst
f3ce8fe90b
nak: properly copy prop neg/abs float sources for flushed values
...
macOS-CI / macOS-CI (dri) (push) Waiting to run
macOS-CI / macOS-CI (xlib) (push) Waiting to run
This allows us to copy prop fadd.ftz -rZ, -|ssa| into consuming
instructions giving us nice gains across the board.
Totals from 1033868 (85.24% of 1212873) affected shaders:
CodeSize: 8813536528 -> 8355226128 (-5.20%); split: -5.21%, +0.01%
Number of GPRs: 44954066 -> 44299483 (-1.46%); split: -1.52%, +0.06%
SLM Size: 799688 -> 798544 (-0.14%)
Static cycle count: 4646939330 -> 4485129185 (-3.48%); split: -3.67%, +0.18%
Spills to memory: 35405 -> 33136 (-6.41%); split: -6.41%, +0.01%
Fills from memory: 35405 -> 33136 (-6.41%); split: -6.41%, +0.01%
Spills to reg: 196547 -> 196231 (-0.16%); split: -1.22%, +1.06%
Fills from reg: 201227 -> 200988 (-0.12%); split: -1.00%, +0.88%
Max warps/SM: 44143984 -> 44306960 (+0.37%); split: +0.38%, -0.01%
Reviewed-by: Mary Guillemard <mary@mary.zone>
Reviewed-by: Mel Henning <mhenning@darkrefraction.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40897 >
2026-04-13 23:15:10 +00:00
Karol Herbst
8170f18d9b
nak/copy_prop: allow modified F16v2 and F16 sources
...
Seems to help a couple of shaders using MUFU.F16
Totals from 178 (0.01% of 1212873) affected shaders:
CodeSize: 5929856 -> 5925088 (-0.08%); split: -0.08%, +0.00%
Static cycle count: 8667151 -> 8665940 (-0.01%); split: -0.02%, +0.00%
Reviewed-by: Mel Henning <mhenning@darkrefraction.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40897 >
2026-04-13 23:15:10 +00:00
Rhys Perry
a6b86d43d3
ir3/ra: fix copy-paste error
...
I don't entirely understand what this is all doing, but this looks like a
copy-paste error.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Backport-to: 26.0
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40480 >
2026-04-13 22:28:15 +00:00
Rhys Perry
54af9431d1
ir3/array_to_ssa: initialize ir3_instruction::data
...
This should have already been NULL because this looks like the first time
this field is used, but that's a bit fragile.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40480 >
2026-04-13 22:28:15 +00:00
Rhys Perry
4f0fb5784f
ir3/array_to_ssa: skip remove_trivial_phi for non-array phis
...
remove_trivial_phi() mostly does nothing for non-array phis, but it
rewrites sources if their definining instruction are trivial phis.
In the case of trivial phis in the loop continue block (for loops with
divergent non-trivial continues), we might need to keep those if they
write a shared register, because the source of the trivial phi will not be
reachable from the loop header phi.
In this example, the predecessors of the continue block should be block2,
but the physical predecessors are block2 and block3, requiring a phi in
the continue block which will then be lowered by ir3_lower_shared_phis.
loop {
block1:
a = phi 0, b
if (divergent) {
block2:
b = a + 1
continue;
}
block3:
break;
}
Fixes RA validation error when compiling blackmythwukong/5645a84e669a6179
from radv_fossils.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Backport-to: 26.0
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40480 >
2026-04-13 22:28:15 +00:00