Timur Kristóf
127999989a
aco: Add MESA_SHADER_KERNEL to instruction selection setup.
...
Treat it the same way as MESA_SHADER_COMPUTE.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24112 >
2023-07-12 11:24:44 +00:00
Eric Engestrom
8e383e6d9e
ci: set priority:low tag only on non-Marge pipelines
...
This allows dynamically setting the priority to avoid starving Marge.
Signed-off-by: Eric Engestrom <eric@igalia.com>
Reviewed-by: Martin Roukala (né Peres) <martin.roukala@mupuf.org>
Reviewed-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23976 >
2023-07-11 13:50:38 +00:00
Daniel Schürmann
7e4870e8e5
amd: Do shader binary alignment for prefetch at memory allocation time.
...
This makes it consistent between drivers and compilers.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23799 >
2023-07-11 12:01:45 +00:00
Daniel Schürmann
437bf4fccb
amd: move end-of-code marker padding to ACO.
...
This makes it consistent between drivers and compilers.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23799 >
2023-07-11 12:01:45 +00:00
Daniel Schürmann
e88c0bb945
radv: migrate radv_shader hash to BLAKE3
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23974 >
2023-07-11 09:02:43 +00:00
Daniel Schürmann
8740fb0026
radv/meta: disable disk cache for meta shaders
...
Meta shaders are already stored in a separate cache file,
inserting them into the disk cache is unnecessary.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23909 >
2023-07-11 07:36:53 +00:00
Daniel Schürmann
c778803d67
aco/assembler: change prefetch mode on GFX10.3+ during loops if beneficial
...
Totals from 8864 (6.68% of 132726) affected shaders: GFX11
CodeSize: 90776128 -> 90923760 (+0.16%)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23748 >
2023-07-11 07:15:43 +00:00
Daniel Schürmann
b9c5b273b0
aco/assembler: align loops if it reduces the number of cache lines
...
This is especially beneficial on GFX6-9.
Totals from 11229 (8.46% of 132726) affected shaders: GFX11
CodeSize: 109608640 -> 109840916 (+0.21%)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23748 >
2023-07-11 07:15:43 +00:00
Daniel Schürmann
de8ecc127e
aco/assembler: align resume shaders with cache lines
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23748 >
2023-07-11 07:15:43 +00:00
Yonggang Luo
bd27faf083
treewide: remove unused nir_builder
...
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24038 >
2023-07-10 19:20:17 +00:00
Yonggang Luo
48a25ef700
treewide: Remove all usage of nir_builder_init with nir_builder_create and nir_builder_at
...
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24038 >
2023-07-10 19:20:17 +00:00
Friedrich Vock
e034ba1c44
radv/rt: Miss rays that hit the triangle's v edge
...
The hardware seems to do this as well. Avoids invoking hit shaders twice
at shared edges.
Fixes the fails in watertightness tests on emulated RT.
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24063 >
2023-07-10 18:31:30 +00:00
Friedrich Vock
a096cf784f
radv/rt: Enable exact on software intersection functions
...
These functions need exact output, otherwise watertightness can't be
guaranteed in some cases.
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24063 >
2023-07-10 18:31:30 +00:00
Erik Faye-Lund
6d42711043
radv: do not rely on constant-folding
...
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24004 >
2023-07-10 16:06:40 +02:00
Marek Olšák
146a92dd9f
radeonsi/gfx11: only use SET_*_PAIRS* packets on dGPUs
...
They are not available on APUs.
This adds a new template parameter HAS_PAIRS. into draw functions.
Other places add back the non-pairs code for gfx11.
Fixes: 22f3bcfb - radeonsi/gfx11: use SET_*_REG_PAIRS_PACKED packets for pm4 states
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9259
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24010 >
2023-07-09 04:18:05 -04:00
Friedrich Vock
d92520e19c
Revert "Revert "radv: Enable ray tracing pipelines by default""
...
This reverts commit 49b400423c .
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23978 >
2023-07-08 10:10:15 +00:00
Friedrich Vock
816ed055ac
Revert "radv/rt: Enable RT pipelines on GFX10_3+ excluding vangogh"
...
This reverts commit 7f3874981d .
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23978 >
2023-07-08 10:10:15 +00:00
Friedrich Vock
0804e150e4
radv/ci: Skip ray tracing tests on vangogh
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23978 >
2023-07-08 10:10:15 +00:00
David Heidelberg
66a01b9377
ci/amd: switch all possible jobs from X11 to Wayland
...
Saved time is around 3 - 4 minutes per job,
decreasing runtime from ~ 19 minutes to ~ 15 minutes.
Acked-by: Daniel Stone <daniels@collabora.com>
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23964 >
2023-07-08 00:16:33 +02:00
Jesse Natalie
511116d630
radv: Fix label name
...
Reviewed-by: Eric Engestrom <eric@igalia.com>
Acked-by: Daniel Stone <daniels@collabora.com>
Acked-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24042 >
2023-07-07 17:40:38 +00:00
Emma Anholt
ee7e440743
ci/radv: Add testing of the GLES CTS using ANGLE on stoney.
...
We're interested in a Vulkan-only stack in Chrome OS, where Android's GLES
would be provided by ANGLE-over-Venus-over-RADV. Let's get some testing
covering ANGLE-on-RADV first.
This is structured as a single partial job pre-merge to catch most
regressions, and 2 longer manual jobs to do full coverage for when you
need to update the xfails list.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20163 >
2023-07-06 23:12:30 +00:00
Emma Anholt
0811a51e9a
ci/amd: Report flakes to #amd-ci on OFTC.
...
Until we get better tooling for crawling job logs, this is our best way of
tracking and updating flaky tests.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20163 >
2023-07-06 23:12:30 +00:00
Friedrich Vock
4880c827d6
radv: Re-enable RT pipeline capture/replay handles
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23516 >
2023-07-05 15:58:25 +00:00
Friedrich Vock
169583a4ad
radv/rt: Rework radv_GetRayTracingCaptureReplayShaderGroupHandlesKHR
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23516 >
2023-07-05 15:58:25 +00:00
Friedrich Vock
fccf6fbeec
radv/rt: Replay shader allocations according to capture/replay handle
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23516 >
2023-07-05 15:58:25 +00:00
Friedrich Vock
3e9bd821f1
radv/rt: Associate capture/replay handles with stages
...
For stages where the capture/replay handle is only known after compiling
and uploading the shader, the shader needs to be relocated to the VA
corresponding to the capture/replay address.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23516 >
2023-07-05 15:58:25 +00:00
Friedrich Vock
eee0068943
radv/rt: Only compare the non-recursive capture/replay handle
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23516 >
2023-07-05 15:58:25 +00:00
Friedrich Vock
59d269c48e
radv: Add radv_rt_capture_replay_handle
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23516 >
2023-07-05 15:58:25 +00:00
Friedrich Vock
e3bd54d2a8
radv: Add support for creating capture/replay shaders
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23516 >
2023-07-05 15:58:25 +00:00
Friedrich Vock
4f192b9af4
radv: Split up implementation of radv_shader_create
...
This will make it easy to re-use the split-up parts for creating
replayed shaders.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23516 >
2023-07-05 15:58:25 +00:00
Friedrich Vock
51f2fa1a5e
radv: Break up radv_shader_nir_to_asm
...
radv_shader_nir_to_asm actually had 3 functions: compiling the NIR to
asm, uploading the shaders and generating debug info for them.
This reduces the functionality of radv_shader_nir_to_asm to only compile
NIR to asm. Uploading the shader and generating debug info is split into
separate functions.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23516 >
2023-07-05 15:58:25 +00:00
Friedrich Vock
878a731c77
radv: Add radv_shader_reupload
...
Used for relocating RT shaders with capture/replay.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23516 >
2023-07-05 15:58:24 +00:00
Friedrich Vock
744357477e
radv: Add utilities to serialize and deserialize shader allocation info
...
Can be used to capture/replay an arbitrary sequence of shader
allocations while preserving VAs.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23516 >
2023-07-05 15:58:24 +00:00
Friedrich Vock
d23e41de6c
radv: Add option to allocate shaders in replayable VA range
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23516 >
2023-07-05 15:58:24 +00:00
Friedrich Vock
ec9f5b7777
radv: Move shader arena allocation to a separate function
...
The arena size is also needed for capture/replay.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23516 >
2023-07-05 15:58:24 +00:00
Friedrich Vock
91241014e8
radv: Add radv_shader_free_list
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23516 >
2023-07-05 15:58:24 +00:00
Joshua Ashton
59cf3702c4
radv: Remove unused pipeline param from radv_generate_pipeline_key
...
Signed-off-by: Joshua Ashton <joshua@froggi.es>
Reviewed-by: Friedrich Vock <friedrich.vock@gmx.de>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23912 >
2023-07-04 21:05:16 +01:00
Joshua Ashton
5ca8390e5f
radv: Advertise VK_EXT_pipeline_robustness
...
Signed-off-by: Joshua Ashton <joshua@froggi.es>
Reviewed-by: Friedrich Vock <friedrich.vock@gmx.de>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23912 >
2023-07-04 21:05:09 +01:00
Joshua Ashton
c9ac960b6a
radv: Implement VK_EXT_pipeline_robustness
...
Funnel the data from the pNext into the new pipeline key members for buffer robustness
Signed-off-by: Joshua Ashton <joshua@froggi.es>
Reviewed-by: Friedrich Vock <friedrich.vock@gmx.de>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23912 >
2023-07-04 21:04:38 +01:00
Joshua Ashton
3e269c7a4e
radv: Rename radv_required_subgroup_info to radv_shader_stage_key
...
Additionally, prefix the members that are subgroup related with subgroup_
We will use this structure to store pipeline robustness information.
Signed-off-by: Joshua Ashton <joshua@froggi.es>
Reviewed-by: Friedrich Vock <friedrich.vock@gmx.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23912 >
2023-07-04 21:04:34 +01:00
Joshua Ashton
2c6125c1c5
radv: Split and move buffer robustness to shader key
...
Will be used by pipeline robustness.
There is also loads of space in the first dword.
Signed-off-by: Joshua Ashton <joshua@froggi.es>
Reviewed-by: Friedrich Vock <friedrich.vock@gmx.de>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23912 >
2023-07-04 21:04:30 +01:00
Joshua Ashton
3e36b47b96
radv: Rename radv_nir_compiler_options::robust_buffer_access to robust_buffer_access_llvm
...
This is only used by LLVM, make that clearer.
Signed-off-by: Joshua Ashton <joshua@froggi.es>
Reviewed-by: Friedrich Vock <friedrich.vock@gmx.de>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23912 >
2023-07-04 21:04:21 +01:00
Joshua Ashton
f52ebc8b40
radv: Refactor buffer robustness to an enum
...
Will be taken advantage of with pipeline robustness.
Signed-off-by: Joshua Ashton <joshua@froggi.es>
Reviewed-by: Friedrich Vock <friedrich.vock@gmx.de>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23912 >
2023-07-04 21:04:10 +01:00
Yonggang Luo
0fbec6dd79
radv: Use nir_remove_non_entrypoints in radv_shader.c
...
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23902 >
2023-07-03 21:45:35 +00:00
Konstantin Seurer
20beebb041
amd: Move ac_hw_stage to its own file
...
Otherwise ACO has to include ac_shader_util.h which also includes NIR.
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Friedrich Vock <friedrich.vock@gmx.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23906 >
2023-07-03 21:12:45 +00:00
Connor Abbott
d322a80170
radv: Fix radv_pipeline_is_blend_enabled
...
This was relying on cb being NULL instead of just gracefully handling
it, and it will stop being NULL once we start tracking attachment count
as state. Moreover is was broken in the case where only the blend enable
is dynamic.
Reviewed-by: Tatsuyuki Ishi <ishitatsuyuki@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22301 >
2023-07-03 19:51:05 +00:00
Tatsuyuki Ishi
94b92718d9
radv: Enable weak reference cache for device->mem_cache.
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23878 >
2023-07-03 17:59:07 +00:00
Daniel Schürmann
3bd72eec1e
radv: reference pipeline cache object in radv_pipeline
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23878 >
2023-07-03 17:59:07 +00:00
Konstantin Seurer
1ea963c00b
radv: Use nir_builder_at
...
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23883 >
2023-07-03 15:21:37 +00:00
Rhys Perry
ae48fae658
aco: remove 64-bit integer conversion opcodes
...
These are handled by nir_lower_int64 now.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23926 >
2023-07-03 10:38:28 +00:00