AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
155,079 Commits 1 Branch 0 Tags
f227fb6da2f4428551d7145c19ac495dc5382d06
Commit Graph

155079 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Alyssa Rosenzweig f227fb6da2 panfrost: Use push constants for indirect draws 
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16916>
 2022-06-08 13:42:42 +00:00
Alyssa Rosenzweig 43884a9b09 panfrost: Use push constants for indirect dispatch 
Much simpler than creating a UBO and relying on it getting optimized to a push
constant, with possible reordering.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16916>
 2022-06-08 13:42:42 +00:00
Alyssa Rosenzweig 90beea75f6 pan/bi: Don't reorder push with no_ubo_to_push 
Otherwise, load_push_constant won't work properly. This could probably be made
to work if we tried hard enough, but we still don't want reordering for internal
(meta) shaders which are layed out deliberately.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16916>
 2022-06-08 13:42:42 +00:00
Alyssa Rosenzweig 17ea1642e2 pan/bi: Implement load_push_constant 
Bifrost supports "fast access uniforms" loaded from a single contiguous buffer.
This maps directly to Vulkan push constants, with some caveats:

* No indirect access. Indirects need to be lowered to a UBO pull.
* Strict alignment requirements. These will be met in practice.

Implement the NIR intrinsic and map it to the native hardware construct.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16916>
 2022-06-08 13:42:42 +00:00
Alyssa Rosenzweig 3c5f1595b8 panfrost: Fix inverted indirect dispatch condition 
Inverted condition -- indirect dispatch gets disabled when WLS is in use, not
the other way around. Not sure how this worked before...

Fixes: fd7b44882c ("panfrost: Use direct dispatch with shared memory")
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16916>
 2022-06-08 13:42:42 +00:00
Chia-I Wu 5c17a04282 turnip: consider render pass costs in autotune 
To be able to sum drawcall cost and render pass cost, the units of costs
are changed to bytes.  With that, tu_autotune_use_bypass can make
decisions by comparing the costs of sysmem rendering and gmem rendering.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16733>
 2022-06-08 12:48:08 +00:00
Chia-I Wu ce118a7002 turnip: estimate render pass costs 
They will be used by autotuner.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16733>
 2022-06-08 12:48:08 +00:00
Chia-I Wu fe9a2374e6 turnip: include stencil test for drawcall costs 
Stencil test reads from and writes to the stencil buffer.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16733>
 2022-06-08 12:48:08 +00:00
Chia-I Wu 2dc10165a1 turnip: base pipeline draw costs on colorWriteMask 
drawcall_base_cost estimates the memory bandwidth per sample.
colorWriteMask should be more accurate than fs outputs.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16733>
 2022-06-08 12:48:08 +00:00
Chia-I Wu 6fe7b92114 turnip: if-checks autotune debug macros 
This avoids bitrot while the compiler can easily optimize away those
checks.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16733>
 2022-06-08 12:48:08 +00:00
Marek Olšák b7cb4d4f6f radeonsi: set the max UBO size same as the max SSBO size 
Reviewed-by: Mihai Preda <mhpreda@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16901>
 2022-06-08 10:23:20 +00:00
Marek Olšák b750844319 radeonsi: compute PIPE_CAP_MAX_TEXEL_BUFFER_ELEMENTS_UINT correctly 
Reviewed-by: Mihai Preda <mhpreda@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16901>
 2022-06-08 10:23:20 +00:00
Marek Olšák aee8ee17a5 radeonsi: change max TBO/SSBO sizes again and rework max alloc size 
Allow 1/4 of the max heap size, but maximum of 512 MB on 32-bit
architectures.

Reviewed-by: Mihai Preda <mhpreda@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16901>
 2022-06-08 10:23:20 +00:00
Marek Olšák c1adb33a93 radeonsi: clamp against MAX_TEXEL_BUFFER_ELEMENTS correctly 
Reviewed-by: Mihai Preda <mhpreda@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16901>
 2022-06-08 10:23:20 +00:00
Marek Olšák 91e533c6aa radeonsi: report correct maximum compute grid sizes 
Reviewed-by: Mihai Preda <mhpreda@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16901>
 2022-06-08 10:23:20 +00:00
Marek Olšák ecda7be628 radeonsi: increase the max compute LDS size to 64KB for gfx7+ 
Reviewed-by: Mihai Preda <mhpreda@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16901>
 2022-06-08 10:23:19 +00:00
Matt Coster 0fbcaa4b3a pvr: debug: Implement PVR_DEBUG variable 
Signed-off-by: Matt Coster <matt.coster@imgtec.com>
Reviewed-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16907>
 2022-06-08 10:51:37 +01:00
Matt Coster 7c615b4103 pvr: csbgen: Add *_unpack() functions for all generated struct types 
Signed-off-by: Matt Coster <matt.coster@imgtec.com>
Reviewed-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16884>
 2022-06-08 09:13:22 +00:00
Matt Coster 4c0941f0d4 pvr: csbgen: Fix "local variable may be undefined" lints 
Signed-off-by: Matt Coster <matt.coster@imgtec.com>
Reviewed-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16884>
 2022-06-08 09:13:22 +00:00
Matt Coster 9f8556a44a pvr: csbgen: Replace map() and filter() with list comprehension 
Guido originally wanted to drop these (along with reduce()) in 3.0, but
compromised to just recommending list comprehension instead [1].

[1]: https://www.artima.com/weblogs/viewpost.jsp?thread=98196

Signed-off-by: Matt Coster <matt.coster@imgtec.com>
Reviewed-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16884>
 2022-06-08 09:13:22 +00:00
Matt Coster 2285ddc720 pvr: csbgen: Isolate "main" function with __name__ check 
Signed-off-by: Matt Coster <matt.coster@imgtec.com>
Reviewed-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16884>
 2022-06-08 09:13:22 +00:00
Matt Coster a8b80c610b pvr: csbgen: Add typing information 
This is generally PEP-484 stuff, but there is one functional change.
The base class Node needed to have an add() method to allow typed
dynamic dispatch. This could have been decorated @abstractmethod, but
that would require an error-raising implementation on all leaf-type
nodes. Instead, I added a base implementation that just errors out with
information from the subclass instance.

As a simple optimization, subclass implementations of add() (instead of
raising the same (or similar) error) now call super().add() in the
case of invalid child nodes.

Signed-off-by: Matt Coster <matt.coster@imgtec.com>
Reviewed-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16884>
 2022-06-08 09:13:22 +00:00
Matt Coster f2de3716d8 pvr: csbgen: Remove unused function parameters 
Signed-off-by: Matt Coster <matt.coster@imgtec.com>
Reviewed-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16884>
 2022-06-08 09:13:22 +00:00
Matt Coster 14ca712362 pvr: csbgen: Remove "dim" functionality from Group 
This allowed for nested groups rendered as arrays. Support for this had
mostly been removed already; this removes the additional value to make
typing easier.

Signed-off-by: Matt Coster <matt.coster@imgtec.com>
Reviewed-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16884>
 2022-06-08 09:13:22 +00:00
Matt Coster 6961c31938 pvr: csbgen: Cleanup imports 
Signed-off-by: Matt Coster <matt.coster@imgtec.com>
Reviewed-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16884>
 2022-06-08 09:13:22 +00:00
Matt Coster 34a76ec4e6 pvr: csbgen: Make some loops more pythonic 
Signed-off-by: Matt Coster <matt.coster@imgtec.com>
Reviewed-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16884>
 2022-06-08 09:13:22 +00:00
Matt Coster b813ca5d3b pvr: csbgen: Add a missing @staticmethod decorator 
Signed-off-by: Matt Coster <matt.coster@imgtec.com>
Reviewed-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16884>
 2022-06-08 09:13:22 +00:00
Matt Coster 89d6a1cfe4 pvr: csbgen: Formatting pass (PEP-8 plus other minor changes) 
Signed-off-by: Matt Coster <matt.coster@imgtec.com>
Reviewed-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16884>
 2022-06-08 09:13:22 +00:00
Timur Kristóf 0280b526d5 radv: Implement mesh shader scratch ring. 
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16737>
 2022-06-08 08:43:51 +00:00
Timur Kristóf 6056583ae1 ac/nir/ngg: Use variables for outputs without cross-invocation access. 
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16737>
 2022-06-08 08:43:51 +00:00
Timur Kristóf b664279755 ac/nir/ngg: Use mesh shader scratch ring when outputs don't fit LDS. 
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16737>
 2022-06-08 08:43:51 +00:00
Timur Kristóf 304a0e948b ac/nir/ngg: Clean up mesh shader output LDS layout. 
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16737>
 2022-06-08 08:43:51 +00:00
Timur Kristóf 02c87e66e9 nir: Introduce new intrinsics for AMD specific mesh shader task ring. 
The mesh shader task ring is a buffer in VRAM which we will use to
store some mesh shader outputs that don't fit into LDS.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16737>
 2022-06-08 08:43:51 +00:00
Timur Kristóf b730f91247 radv: Implement task shader draw and payload rings. 
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16737>
 2022-06-08 08:43:51 +00:00
Timur Kristóf ac5ab8d227 ac: Add task shader ring information. 
Similarly to tessellation rings information, move the task
rings info to ac_gpu_info.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16737>
 2022-06-08 08:43:51 +00:00
Timur Kristóf 086e499b47 ac: Add RDNA2 task+mesh shader draw packet opcodes. 
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16737>
 2022-06-08 08:43:51 +00:00
Karmjit Mahil 721fb18a38 pvr: Fix incorrect samples to cr_isp_aa_mode conversion. 
This commit fixes the use of 'case 3' where 'case 4' should have
been.

Reported-by: Matt Coster <matt.coster@imgtec.com>
Signed-off-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com>
Reviewed-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16883>
 2022-06-08 08:35:44 +00:00
Karmjit Mahil 61db3154b4 pvr: Write immutable descriptor words in vkAllocateDescriptorSets(). 
Signed-off-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com>
Reviewed-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16883>
 2022-06-08 08:35:44 +00:00
Karmjit Mahil 81077a4c7d pvr: Add csb enum helper funciton header. 
A new header is introduced which groups all vk, uint, rogue types
to csb enum type conversions.

Signed-off-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com>
Reviewed-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16883>
 2022-06-08 08:35:44 +00:00
Georg Lehmann d8493e5310 radv, aco: Lower txf offset in NIR. 
Foz-DB Navi21:
Totals from 384 (0.28% of 134913) affected shaders:
VGPRs: 29736 -> 29536 (-0.67%)
CodeSize: 2455796 -> 2452652 (-0.13%); split: -0.13%, +0.01%
MaxWaves: 6350 -> 6358 (+0.13%)
Instrs: 457743 -> 456273 (-0.32%); split: -0.33%, +0.01%
Latency: 6680266 -> 6730612 (+0.75%); split: -0.03%, +0.78%
InvThroughput: 1562936 -> 1599375 (+2.33%); split: -0.05%, +2.38%
VClause: 9258 -> 9291 (+0.36%); split: -0.14%, +0.50%
SClause: 15713 -> 15707 (-0.04%); split: -0.08%, +0.04%
Copies: 26878 -> 27021 (+0.53%); split: -0.03%, +0.56%
PreVGPRs: 27259 -> 27230 (-0.11%); split: -0.11%, +0.01%

Signed-off-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16869>
 2022-06-08 08:13:01 +00:00
Pierre-Eric Pelloux-Prayer b81f05e94d radeonsi: set size in si_texture_get_handle 
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6507
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6491
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16813>
 2022-06-08 09:42:47 +02:00
Pierre-Eric Pelloux-Prayer bce227611d va/surface: set the correct size in vaExportSurfaceHandle 
The size must be the size of the total object, not the size
of the resource.
For instance, when using a single object for a multi-plane
format, the size of each plane should be equal to the size
of the underlying object to match libva's documentation:

        /** Total size of this object (may include regions which are
         *  not part of the surface). */
        uint32_t size;

Fixes: 13b79266e4 ("frontend/va: Setting the size of VADRMPRIMESurfaceDescriptor")
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16813>
 2022-06-08 09:42:44 +02:00
Pierre-Eric Pelloux-Prayer 8c07d7cbdd gallium: add size attribute to winsys_handle 
This is needed at least for the VA-API frontend where the exporter
needs to know the total size of the object.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16813>
 2022-06-08 09:42:42 +02:00
Pierre-Eric Pelloux-Prayer baf0e66dae gallium: update winsys_handle documentation 
The function names are resource_***_handle, not textures_***_handles.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16813>
 2022-06-08 09:42:26 +02:00
Mike Blumenkrantz cb5973a3dd zink: only update layout when doing mixed zs attachment renderpass check 
avoid doing any sort of extra work here and avoid multi-context weirdness

cc: mesa-stable

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16900>
 2022-06-08 02:55:05 +00:00
Mike Blumenkrantz 8f9c21b15e zink: break out of zs mixed layout update loop when work is done 
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16900>
 2022-06-08 02:55:05 +00:00
Jordan Justen ffb0c97caf intel: Build mi_builder_test whenever build-tests is set 
Previously `install-intel-gpu-tests` controlled this, but now
`install-intel-gpu-tests` will only be used to decide if it should be
installed.

Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16902>
 2022-06-07 18:26:02 +00:00
Jason Ekstrand 04fae05f7d lavapipe: Use vk_image_buffer_copy_layout 
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16873>
 2022-06-07 17:57:42 +00:00
Jason Ekstrand 420717b2ce panvk: Use vk_image_buffer_copy_layout 
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16873>
 2022-06-07 17:57:42 +00:00
Jason Ekstrand 8d6b7eb5ce radv: Use vk_image_buffer_copy_layout 
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16873>
 2022-06-07 17:57:42 +00:00