AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
182,697 Commits 1 Branch 0 Tags
f30fa9dec8612075977df0007eeb8a25af01cc2b
Commit Graph

13699 Commits

Author SHA1 Message Date
Samuel Pitoiset f30fa9dec8 radv: move radv_rt_{common,shader} files to nir/ 
These files contains NIR lowering code for RT.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26851>
 2024-01-03 09:40:29 +00:00
Samuel Pitoiset 02c5365ffa radv: make a couple of NIR RT functions as static 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26851>
 2024-01-03 09:40:29 +00:00
Samuel Pitoiset 685c4b6337 radv: move radv_{emulate,enable}_rt() to radv_physical_device.c 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26851>
 2024-01-03 09:40:29 +00:00
David Heidelberg 9017852de4 ci/radeonsi: disable VA-API testing on raven 
Fails and freezes after Collabora farm outage.

Signed-off-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26820>
 2024-01-02 21:33:51 +00:00
Pierre-Eric Pelloux-Prayer 115b61e51f ac/surface: don't oversize surf_size 
Yet another iteration on the same YUV surfaces.

The change from 87ecfdfbf0 has 2 odd things:
* it's using MAX2(original value, new value) but the point of updating
  surf_slice_size / surf_size is to make it correct relative to the new
  value of surf_pitch
* it's multiplying surf_pitch (= number of elements per row) by height (ok)
  by surf->bpe (= number of bytes per element) by surf->blk_w (= number of
  "horizontal" pixels in an element) so the end unit doesn't make sense.

Fix this by computing a reasonnable value based on unit: the surf_slice_size
is the number of elements per row (surf_pitch) x number of bytes per element
(bpe) x number of rows.

This makes the expected size correct and thus fixes users of eglCreateImageKHR,
like the issue #6131.

I tested a bunch of gst pipelines and ffmpeg scripts on various files I have
and didn't notice any issues (on gfx10.3 and gfx9).

Fixes: 87ecfdfbf0 ("ac/surface: adapt surf_size when modifying surf_pitch")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6131
Acked-by: Marek Olšák <marek.olsak@amd.com>
Tested-by: Nicolas Dufresne <nicolas.dufresne@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26693>
 2024-01-02 14:32:05 +01:00
Georg Lehmann 9ecfd7919b aco: optimize 32bit fsign by using fmulz with Inf 
2 instruction fsign with the power of cursed DX9 floating point rules.

Foz-DB Navi31:
Totals from 3803 (4.86% of 78196) affected shaders:
Instrs: 8436366 -> 8412549 (-0.28%); split: -0.29%, +0.00%
CodeSize: 43174284 -> 43114676 (-0.14%); split: -0.14%, +0.01%
SpillSGPRs: 3241 -> 3247 (+0.19%)
Latency: 66333841 -> 66287361 (-0.07%); split: -0.08%, +0.01%
InvThroughput: 10331902 -> 10316916 (-0.15%); split: -0.15%, +0.01%
VClause: 165455 -> 165472 (+0.01%); split: -0.01%, +0.02%
SClause: 242352 -> 242335 (-0.01%); split: -0.02%, +0.01%
Copies: 604086 -> 605781 (+0.28%); split: -0.04%, +0.32%
Branches: 214017 -> 214013 (-0.00%)
PreSGPRs: 209413 -> 209726 (+0.15%)

Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26765>
 2024-01-02 13:07:30 +01:00
Konstantin Seurer b88ac6b381 nir: Optimize fpow with small constant exponents 
They would be turned into exp(log(a)*b) instead, which is slow.

Totals from 2146 (2.52% of 85071) affected shaders:
MaxWaves: 35769 -> 35779 (+0.03%); split: +0.03%, -0.01%
Instrs: 6476835 -> 6465494 (-0.18%); split: -0.18%, +0.00%
CodeSize: 35382288 -> 35347092 (-0.10%); split: -0.10%, +0.00%
SpillSGPRs: 1055 -> 1017 (-3.60%)
Latency: 75211743 -> 75063623 (-0.20%); split: -0.20%, +0.00%
InvThroughput: 17525115 -> 17501745 (-0.13%); split: -0.14%, +0.00%
VClause: 200089 -> 200077 (-0.01%); split: -0.01%, +0.01%
SClause: 293566 -> 293480 (-0.03%); split: -0.03%, +0.00%
Copies: 649631 -> 640516 (-1.40%); split: -1.44%, +0.03%
Branches: 268441 -> 268325 (-0.04%)
PreSGPRs: 146868 -> 146045 (-0.56%)
PreVGPRs: 134125 -> 134128 (+0.00%); split: -0.00%, +0.01%

Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26727>
 2024-01-02 11:16:14 +01:00
Eric Engestrom 7e8db6aedf meson: always define {,DRAW_}LLVM_AVAILABLE one way or the other 
With the usual benefits of `#if` instead of `#ifdef` (mostly the fact
that typos can be build failures instead of silently being interpreted
as if 0).

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3863>
 2023-12-24 10:01:39 +00:00
Samuel Pitoiset 551924aa87 ci: apply two bugfixes for VKCTS 
These are needed for RADV.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26798>
 2023-12-22 20:12:39 +00:00
Eric Engestrom 91e4428dc4 radv/ci: add flake on raven 
Failed in https://gitlab.freedesktop.org/mesa/mesa/-/jobs/53017453

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26788>
 2023-12-21 13:26:21 +00:00
Friedrich Vock b2067001d4 radv/rt: bsearch inlined shaders 
When there are lots of inlined shaders, going over each one and checking
if the call index matches becomes expensive. Instead, use a
binary-search-like selection to skip most of the checks.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26380>
 2023-12-21 12:39:06 +00:00
Timur Kristóf 4d93aac74d radv: Use correct plane and binding index with SDMA. 
This lets us support multi-planar images properly on transfer queues.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Tatsuyuki Ishi <ishitatsuyuki@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26353>
 2023-12-20 13:22:26 +00:00
Timur Kristóf ab4720691c radv: Clean up SDMA chunked copy info struct. 
Remove redundant fields.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Tatsuyuki Ishi <ishitatsuyuki@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26353>
 2023-12-20 13:22:26 +00:00
Timur Kristóf 7fe899a3b6 radv: Use SDMA surface structs for determining unaligned buffer copies. 
This removes a bunch of redundant calculations, thus making the code
less error-prone.

Also move the row pitch calculation to a separate function.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Tatsuyuki Ishi <ishitatsuyuki@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26353>
 2023-12-20 13:22:26 +00:00
Timur Kristóf dab4863396 radv: Pass radv_sdma_surf from copy functions to SDMA. 
This makes it possible to gather the surface information
only once and pass it to the various SDMA functions, therefore
the code can be less error-prone.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Tatsuyuki Ishi <ishitatsuyuki@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26353>
 2023-12-20 13:22:26 +00:00
Timur Kristóf 85fa749c63 radv: Refactor and simplify SDMA surface info functions. 
This makes it possible to call the function just once instead
of calling different functions repeatedly.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Tatsuyuki Ishi <ishitatsuyuki@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26353>
 2023-12-20 13:22:26 +00:00
Timur Kristóf a21cba6799 radv: Unify SDMA surface struct for linear and tiled images. 
Using just one struct for both types of images (and buffers)
will simplify a lot of code.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Tatsuyuki Ishi <ishitatsuyuki@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26353>
 2023-12-20 13:22:26 +00:00
Timur Kristóf 65dfdd3fff radv: Move SDMA function and struct declarations to a new header. 
Very few parts of RADV actually need the SDMA functions, so
moving them to a separate header makes the driver cleaner and
also improves compilation time when SDMA functions change.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Tatsuyuki Ishi <ishitatsuyuki@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26353>
 2023-12-20 13:22:26 +00:00
Samuel Pitoiset 2ce0ea8e7c radv/ci: update CI lists for NAVI10,NAVI31 and RENOIR 
These dynamic rendering failures/flakes are a known issue that will
be resolved soon as part of VKCTS. Until that, make sure CI is green.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26734>
 2023-12-20 13:06:41 +00:00
Bas Nieuwenhuizen 07ad6fd34a radv: Use correct writemask for cooperative matrix ordering. 
Not expecting this to actually fix anything externally visible,
but reduces some invalid usage when the resulting vector is
not 16 elements long (e.g. the C/result matrix).

Fixes: 9df4703fbb ("radv: Add cooperative matrix lowering.")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26768>
 2023-12-20 11:02:30 +00:00
David Heidelberg 16af090908 ci/lava: separate HW definitions from SW 
Reviewed-by: Guilherme Gallo <guilherme.gallo@collabora.com>
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26722>
 2023-12-20 10:15:44 +00:00
Bas Nieuwenhuizen eaf61adea5 radv: Add option to clear LDS at the end of a shader. 
Only shaders which explicitly allow shared memory are included for
now. The pass is very late to avoid optimizations removing the stores
and to ensure the clear gets added after MS outputs get loaded from LDS.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26679>
 2023-12-20 09:15:45 +00:00
Samuel Pitoiset b5153693cc radv: move image view related code to radv_image_view.c 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26713>
 2023-12-19 09:48:35 +00:00
Samuel Pitoiset 1db86d93f2 radv: move buffer view related code to radv_buffer_view.c 
More code isolation.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26713>
 2023-12-19 09:48:35 +00:00
Samuel Pitoiset 02282fe482 radv: move SI_GS_PER_ES to radv_constants.h 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26713>
 2023-12-19 09:48:35 +00:00
Samuel Pitoiset d057cfc8f6 radv: move more shader related declarations to radv_shader.h 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26713>
 2023-12-19 09:48:35 +00:00
Samuel Pitoiset 0481723cf0 radv: move radv_depth_clamp_mode to radv_cmd_buffer.c 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26713>
 2023-12-19 09:48:35 +00:00
Samuel Pitoiset c0d7ef1e58 radv: move more descriptor related declarations to radv_descriptor_set.h 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26713>
 2023-12-19 09:48:34 +00:00
Samuel Pitoiset a080390c3c radv: move radv_image_is_renderable() to radv_image.c 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26713>
 2023-12-19 09:48:34 +00:00
Samuel Pitoiset 8a5106f91a radv: move RADV_HASH_SHADER_xxx flags to radv_pipeline.c 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26713>
 2023-12-19 09:48:34 +00:00
Samuel Pitoiset 9d15390635 radv: move meta declarations to radv_meta.h 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26713>
 2023-12-19 09:48:34 +00:00
Samuel Pitoiset 939936a328 radv: remove radv_get_tess_output_topology() declaration 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26713>
 2023-12-19 09:48:34 +00:00
Konstantin Seurer 662f86c533 radv/bvh: Stop emitting leaf nodes inside the encoder 
Avoids unnecessary copies.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26725>
 2023-12-18 19:17:21 +00:00
Konstantin Seurer 8fca54a754 radv: Add more offsets acceleration_structure_layout 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26725>
 2023-12-18 19:17:21 +00:00
Tatsuyuki Ishi 533ec9843e radv: Precompute shader max_waves. 
Doing it at bind-time causes a 1.4% overhead (among all driver calls) in
Overwatch 2. !24502 mentions that it can be precomputed in case overhead
is a concern, so do it here.

max_waves is stored directly in the radv_shader struct, because
ac_shader_config conforms to LLVM ABI and we cannot add anything custom,
and radv_shader_info needs to be determined from NIR only.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26692>
 2023-12-18 14:31:25 +00:00
Tatsuyuki Ishi 1161f22c27 radv: Move up radv_get_max_waves, radv_get_max_scratch_waves. 
To avoid forward declaration.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26692>
 2023-12-18 14:31:25 +00:00
Tatsuyuki Ishi e444908d65 radv: Simplify shader config assignment. 
We don't hash this struct so direct assignment here is OK.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26692>
 2023-12-18 14:31:24 +00:00
Samuel Pitoiset 4353b0ad72 radv: move emitting the fb mip tail workaround when rendering begins 
It doesn't have to be emitted in the draw path.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26589>
 2023-12-18 13:06:29 +00:00
Samuel Pitoiset 7dd7e551b1 radv: stop checking FMASK for the fb mip tail workaround 
Vulkan doesn't allow mipmaps with MSAA images, so checking for FMASK
shouldn't have any effect.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26589>
 2023-12-18 13:06:29 +00:00
Samuel Pitoiset 57efe44f43 radv: add missing HTILE support for fb mip tail workaround 
PAL also applies to depth/stencil images with HTILE.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26589>
 2023-12-18 13:06:29 +00:00
Friedrich Vock f9a394b753 radv/rt: Initialize unused children in PLOC early-exit 
Bad things happen when these aren't initialized.

Fixes GPU hangs when loading Avatar: Frontiers of Pandora savegames.

Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26712>
 2023-12-15 17:52:00 +00:00
Ruijing Dong 434a53ebbe radesonsi/vcn add qp_map definition 
This is for enabling ROI (region of interest)
feature in VAAPI interface. It will support
both CQP and rate control mode.

Reviewed-by: Boyuan Zhang <boyuan.zhang@amd.com>
Signed-off-by: Ruijing Dong <ruijing.dong@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26659>
 2023-12-15 14:06:34 +00:00
Samuel Pitoiset 768c737273 radv: remove some declared but unused functions/macros 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26682>
 2023-12-15 08:40:24 +00:00
Samuel Pitoiset 2f79ed1831 radv: stop asserting some image create info fields 
Vulkan Validation Layers already check that and these assertions never
caught anything.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26682>
 2023-12-15 08:40:23 +00:00
Eric Engestrom 871ea3bb88 radv/ci: add flake 
Failed in https://gitlab.freedesktop.org/mesa/mesa/-/jobs/52773139

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26704>
 2023-12-14 22:13:45 +00:00
Daniel Schürmann 42e9ba1c70 aco: remove VCCZ and EXECZ register handling 
We don't use these registers and since RDNA3 removed the explicit usage,
it is unlikely that we will properly support them in the future.
Removing the registers from the ACO IR prevents accidentally using them
without proper support.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26664>
 2023-12-14 20:08:28 +00:00
Samuel Pitoiset 1ef5feac5e radv: fix binding partial depth/stencil views with dynamic rendering 
With dynamic rendering, it's allowed to begin rendering with depth or
stencil only but still with a depth/stencil format. The test below
checks that unbound part of ds isn't modified, if depth is bound and
stencil not and vice versa.

This fixes a recent CTS
dEQP-VK.dynamic_rendering.primary_cmd_buff.basic.partial_binding_depth_stencil.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25350>
 2023-12-14 12:54:23 +00:00
Samuel Pitoiset 98ea540158 radv: add support for MRT compaction with PS epilogs 
Now that PS epilogs are always compiled during cmdbuf recording, we
have all information to enable MRT compaction, for optimal performance.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26667>
 2023-12-14 09:51:26 +01:00
Samuel Pitoiset 8760b7fab7 radv: cleanup radv_pipeline_generate_ps_epilog_key() 
This has no effects because key->spi_shader_col_format isn't used when
the graphics pipeline needs to compile a PS epilog.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26663>
 2023-12-14 08:18:46 +00:00
Samuel Pitoiset 60e6e3f8e6 radv: cleanup gathering PS info with/without PS epilogs 
To make it clear that some fields aren't used with PS epilogs.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26663>
 2023-12-14 08:18:46 +00:00