AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
213,712 Commits 1 Branch 0 Tags
4711fb711cbf0dab3f34def05ab9a6566acf877f
Commit Graph

213712 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Qiang Yu 4711fb711c gallium/blitter: no need to save TS state 
TS does not affect blitter currently.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37505>
 2025-10-17 03:52:19 +00:00
Qiang Yu 71e0895715 mesa,radeonsi: add comments about vertex and mesh pipeline shader states 
They are exclusive in mesa state tracker currently, so add some comments
and assertions for developers.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37505>
 2025-10-17 03:52:19 +00:00
Qiang Yu dcf2399e6f radeonsi: save mesh shader when blit 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37505>
 2025-10-17 03:52:19 +00:00
Qiang Yu ffc3d430db radeonsi: simplify si_update_rasterized_prim while handle mesh shader 
Otherwise mesh shader ends in the "else" section.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37505>
 2025-10-17 03:52:19 +00:00
Qiang Yu 56a437183a radeonsi: si_get_vs support mesh shader 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37505>
 2025-10-17 03:52:19 +00:00
Qiang Yu 7e83962e85 radeonsi: update scratch va for mesh shader 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37505>
 2025-10-17 03:52:19 +00:00
Qiang Yu de4fb088d3 radeonsi: share some vertex pipe function with mesh pipe 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37505>
 2025-10-17 03:52:18 +00:00
Qiang Yu e6e21dfbf2 radeonsi: kill outputs for mesh shader 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37505>
 2025-10-17 03:52:18 +00:00
Qiang Yu 4c315bdbfa radeonsi: lower task/mesh shader io to mem 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37505>
 2025-10-17 03:52:18 +00:00
Qiang Yu 5931dbf7ac radeonsi: add task info to screen 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37505>
 2025-10-17 03:52:18 +00:00
Qiang Yu 73aebeec42 radeonsi: no ngg culling for mesh shader 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37505>
 2025-10-17 03:52:17 +00:00
Qiang Yu 74894150f1 radeonsi: init pm4 state for mesh shader 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37505>
 2025-10-17 03:52:17 +00:00
Qiang Yu ce6a1e7563 radeonsi: init mesh shader args 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37505>
 2025-10-17 03:52:16 +00:00
Qiang Yu 2038134efc radeonsi: calc workgroup size for mesh shader 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37505>
 2025-10-17 03:52:15 +00:00
Qiang Yu 977a3f45bf radeonsi: add task/mesh shader info to si_shader_info 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37505>
 2025-10-17 03:52:15 +00:00
Qiang Yu 8659666089 radeonsi: add si_mesh_resources_add_all_to_bo_list 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37505>
 2025-10-17 03:52:14 +00:00
Qiang Yu b533d39b95 radeonsi: inline uniform support mesh shader 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37505>
 2025-10-17 03:52:14 +00:00
Qiang Yu 8a3ef188c2 radeonsi: add context shader state for mesh shader 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37505>
 2025-10-17 03:52:13 +00:00
Qiang Yu 24d7c9a2a8 radeonsi: handle mesh shader when si_create_shader 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37505>
 2025-10-17 03:52:12 +00:00
Qiang Yu f06a1b0d07 radeonsi: enlarge SI_NUM_SHADERS for mesh shader 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37505>
 2025-10-17 03:52:12 +00:00
Mike Blumenkrantz f74cf45078 zink: consistently set/unset msrtss in begin_rendering 
this has to always be set or unset, never persistent from previous renderpass

Fixes: 5080f2b6f5 ("zink: disable msrtss handling when blitting")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37923>
 2025-10-16 22:22:34 -04:00
Marek Olšák 733ba77bfe r300: fix DXTC blits 
Fixes: 9d359c6d10 - gallium: delete pipe_surface::width and pipe_surface::height
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37912>
 2025-10-16 22:33:50 +00:00
Gert Wollny ba35ac29b6 r600/sfn: drop range pinning for registers after RA 
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37847>
 2025-10-16 20:57:18 +00:00
Gert Wollny 5962add398 r600/sfn: correct register interference range 
If a life range of one register starts in the same instruction where the
life range of another register ends, then
the two ranges don't overlap.

v2: Fix test

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37847>
 2025-10-16 20:57:18 +00:00
Dylan Baker a1b6dbcd67 docs: update calendar for 25.3.0-rc1 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37921>
 2025-10-16 20:52:31 +00:00
José Roberto de Souza ad86a666ae anv: Add support for low latency hint on Xe KMD 
This hint tells KMD and firmware to turn into low latency but high
power usage mode.
i915 already had it now it was implemented in Xe KMD.

Reviewed-by: Sushma Venkatesh Reddy <sushma.venkatesh.reddy@intel.com>
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33214>
 2025-10-16 20:23:21 +00:00
José Roberto de Souza 0ba6a0a23b intel/dev: Add supports_low_latency_hint to intel_device_info 
Lets query if this feature is supported only once, also in the next
patches support for this feature will be added to Xe KMD.

Reviewed-by: Sushma Venkatesh Reddy <sushma.venkatesh.reddy@intel.com>
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33214>
 2025-10-16 20:23:21 +00:00
Xaver Hugl 892cf427a0 vulkan/wsi: require extended target volume support for scRGB 
It's hardly going to be useful without that

Signed-off-by: Xaver Hugl <xaver.hugl@kde.org>
Fixes: 4b663d56 ("vulkan/wsi: implement support for VK_EXT_hdr_metadata on Wayland")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37920>
 2025-10-16 19:58:02 +00:00
Mary Guillemard e3d9c5da2a mr-label-maker: Remove mapi label 
It doesn't exist anymore.

Signed-off-by: Mary Guillemard <mary.guillemard@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37914>
 2025-10-16 19:25:35 +00:00
Mary Guillemard f65756a31b mr-label-maker: Add poly 
Signed-off-by: Mary Guillemard <mary.guillemard@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37914>
 2025-10-16 19:25:35 +00:00
Mary Guillemard b2accf86d1 poly: Migrate AGX's GS/TESS emulation to common code 
This moves most of the code to a new home: src/poly.
Most precomp kernels logic that could be moved are provided by poly now.

Signed-off-by: Mary Guillemard <mary.guillemard@collabora.com>
Acked-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37914>
 2025-10-16 19:25:35 +00:00
Mary Guillemard 8048004238 asahi/gs: Reuse GS shader compiler options 
Avoid importing internal bits

Signed-off-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37914>
 2025-10-16 19:25:35 +00:00
Mary Guillemard 6f73533094 asahi,nir: Stop relying on zero and scratch page in GS/TESS code 
Introduce new NIR intrinsics to handle getting a "sink" read-only
address and another intrinsic to handle conversion of address to
read-write (allowing implementation to replace the "sink" read-only with
another address like required for Asahi)

Signed-off-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37914>
 2025-10-16 19:25:35 +00:00
Mary Guillemard 60e5abdbaa asahi: Move compiler preprocess out of agx_nir_lower_gs 
We run agx_preprocess_nir as the last step of each new compute shaders
in agx_nir_lower_gs but we could move this out of the pass and makes it
the driver responsability to call it.

Signed-off-by: Mary Guillemard <mary.guillemard@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37914>
 2025-10-16 19:25:35 +00:00
Mary Guillemard a427581c4d asahi/gs: Remove agx_nir_* prefix around static functions 
Only a small detail but git will not go too crazy when I move
everything around at least.

Signed-off-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37914>
 2025-10-16 19:25:35 +00:00
Mary Guillemard 71c4943c37 compiler: rename vs.tes_agx bit to vs.tes_poly 
Preparing to move AGX's GS/TESS lowering code.

Signed-off-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37914>
 2025-10-16 19:25:35 +00:00
Mary Guillemard 1e0c18d6cf nir: Rename stat_query_address_agx to stat_query_address_poly 
This is used by the geometry lowering that we are going to move to
common code.

Signed-off-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37914>
 2025-10-16 19:25:35 +00:00
Mary Guillemard 8a25b88d69 asahi/libagx: Do not expose anything not use externaly 
No need to actually generate NIR bindings for anything we don't need
to.

We are going to copy most of this in Panfrost and that will be required
there...

Signed-off-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37914>
 2025-10-16 19:25:35 +00:00
Mary Guillemard 29f348941d asahi/libagx: Stop exposing fake entrypoint _libagx_prefix_sum 
This was a hack to allow things to build but still could break in the
future, let comply by passing scratch as an argument instead.

Signed-off-by: Mary Guillemard <mary.guillemard@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37914>
 2025-10-16 19:25:35 +00:00
Lionel Landwerlin c20e2733bf Revert "brw: add serialize send stats" 
This reverts commit b8ae4ede60 now that
we have a cycle estimation accounting.

Reviewed-by: Alyssa Anne Rosenzweig <alyssa.rosenzweig@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37816>
 2025-10-16 18:55:06 +00:00
Lionel Landwerlin 14683a045b brw: account for disabled SEND fused message in cycle computation 
This is an alternative Curro proposed to counting the number of
serialized messages.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Alyssa Anne Rosenzweig <alyssa.rosenzweig@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37816>
 2025-10-16 18:55:06 +00:00
Job Noorman 3b2f7ed918 ci,marge_queue: read token from file by default 
This allows token to be stored in ~/.config/gitlab-token instead of
passing them as an argument.

Signed-off-by: Job Noorman <jnoorman@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37916>
 2025-10-16 17:40:51 +00:00
Lionel Landwerlin 77fb8fb062 anv: fix image-to-image copies of TileW images 
The intermediate buffer between the 2 images is linear, its stride
should be a function of the tile's logical width.

Normally this should map to the values reported by ISL except for
TileW where for some reason it was decided to report 128 for TileW
instead of the actual 64 size (see isl_tiling_get_info() ISL_TILING_W
case)

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: mesa-stable
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37902>
 2025-10-16 17:07:32 +00:00
Job Noorman 75604ff945 ir3: allow (neg) on sel.b on a6xx gen4+ 
Setting the (neg) flag on a sel.b source behaves as fneg on a6xx gen4+.

Totals from 22733 (13.80% of 164705) affected shaders:
MaxWaves: 277060 -> 277292 (+0.08%); split: +0.11%, -0.03%
Instrs: 17676148 -> 17634471 (-0.24%); split: -0.47%, +0.24%
CodeSize: 34824114 -> 34693740 (-0.37%); split: -0.44%, +0.06%
NOPs: 3466984 -> 3487984 (+0.61%); split: -1.17%, +1.78%
MOVs: 521091 -> 522791 (+0.33%); split: -1.87%, +2.20%
Full: 315929 -> 315699 (-0.07%); split: -0.09%, +0.02%
(ss): 473545 -> 472947 (-0.13%); split: -1.36%, +1.23%
(sy): 195612 -> 195743 (+0.07%); split: -1.00%, +1.07%
(ss)-stall: 1928887 -> 1922757 (-0.32%); split: -1.96%, +1.64%
(sy)-stall: 4965071 -> 4972119 (+0.14%); split: -1.43%, +1.57%
STPs: 777 -> 762 (-1.93%)
LDPs: 2168 -> 2117 (-2.35%)
Preamble Instrs: 3465691 -> 3462635 (-0.09%); split: -0.09%, +0.00%
Last helper: 4666320 -> 4593331 (-1.56%); split: -2.81%, +1.24%
Last baryf: 235724 -> 230049 (-2.41%); split: -4.56%, +2.15%
Subgroup size: 2021248 -> 2021952 (+0.03%); split: +0.07%, -0.03%
Cat0: 3792738 -> 3814197 (+0.57%); split: -1.08%, +1.64%
Cat1: 757480 -> 759260 (+0.23%); split: -1.28%, +1.52%
Cat2: 6960677 -> 6897218 (-0.91%)
Cat6: 78290 -> 78224 (-0.08%)
Cat7: 422101 -> 420710 (-0.33%); split: -1.41%, +1.09%

Signed-off-by: Job Noorman <jnoorman@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37906>
 2025-10-16 16:46:18 +00:00
Job Noorman eaef6d048b ir3: add has_sel_b_fneg compiler flag 
a6xx+ support (neg) on sel.b which behaves as fneg. Add a compiler flag
for this.

Signed-off-by: Job Noorman <jnoorman@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37906>
 2025-10-16 16:46:18 +00:00
Job Noorman c44f8fe7f7 ir3: move ir3_catN_absneg to ir3.c 
The following commit needs to use ir3_compiler whose definition isn't
available in ir3.h.

Signed-off-by: Job Noorman <jnoorman@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37906>
 2025-10-16 16:46:18 +00:00
David Rosca 09ff0fa005 frontends/va: Move remainig processing functions to postproc.c 
Reviewed-by: Leo Liu <leo.liu@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37545>
 2025-10-16 16:33:15 +00:00
David Rosca 7a5270d4df frontends/va: Move decode functions to separate file 
Reviewed-by: Leo Liu <leo.liu@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37545>
 2025-10-16 16:33:15 +00:00
David Rosca ba0a059129 frontends/va: Move encode functions to separate file 
Reviewed-by: Leo Liu <leo.liu@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37545>
 2025-10-16 16:33:15 +00:00
Utku Iseri d9d0001b7d panvk: fix for clearing render targets with 8+ layers 
It's valid for the tiler desc to be 0 when the tiler isn't being
used. Currently, we set the descriptor based on an offset over the
pointer in the gfx state, and if this is 0, we end up setting it to
just the offset when there are more than 8 layers on a target.

Reviewed-by: Lars-Ivar Hesselberg Simonsen <lars-ivar.simonsen@arm.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37837>
 2025-10-16 16:13:33 +00:00