371468c013
intel/brw: Remove Gfx8- code from lower logical sends
...
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27691 >
2024-02-28 05:45:38 +00:00
1ee29f82d2
intel/brw: Remove Gfx8- code from lower storage image pass
...
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27691 >
2024-02-28 05:45:38 +00:00
6a03280af1
intel/brw: Remove Gfx8- code from NIR conversion
...
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27691 >
2024-02-28 05:45:38 +00:00
2a1c2a1bf1
intel/brw: Remove Gfx8- code from thread payload
...
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27691 >
2024-02-28 05:45:38 +00:00
ed6f0665e0
intel/brw: Remove Gfx8- code from register allocator
...
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27691 >
2024-02-28 05:45:38 +00:00
85eb672325
intel/brw: Remove Gfx8- code from scheduler
...
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27691 >
2024-02-28 05:45:38 +00:00
6e88fa8a77
intel/brw: Remove Gfx8- code from brw_compile_* functions
...
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27691 >
2024-02-28 05:45:38 +00:00
a4bf016582
intel/brw: Remove Gfx8- code from assembler
...
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27691 >
2024-02-28 05:45:38 +00:00
83d7ddebcb
intel/brw: Remove Gfx8- code from disassembler
...
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27691 >
2024-02-28 05:45:38 +00:00
15469b0a63
intel/brw: Remove unused legacy shader stages
...
CLIP, STRIP-FAN and FF Geometry don't apply to Gfx9+.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27691 >
2024-02-28 05:45:38 +00:00
c621f75e7b
intel/brw: Remove now unused vec4-only opcodes
...
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27691 >
2024-02-28 05:45:38 +00:00
a641aa294e
intel/brw: Remove vec4 backend
...
It still exists as part of ELK for older gfx versions.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27691 >
2024-02-28 05:45:37 +00:00
7c23b90537
intel/brw: Always use scalar shaders
...
Remove scalar_stage[] array, since now it is always scalar. This
removes any usage of vec4 shaders in brw.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27691 >
2024-02-28 05:45:37 +00:00
303fd4e935
intel/brw: Move type_size_* functions out of vec4-specific file
...
Will make easier later to delete vec4 files.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27691 >
2024-02-28 05:45:37 +00:00
9bfccc1935
intel/brw: Move brw_compile_* functions out of vec4-specific files
...
These contain code that is both fs and vec4. Will make easier later to
delete vec4 files.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27691 >
2024-02-28 05:45:37 +00:00
c11d7743b3
intel/blorp: Remove Gfx8- references in BRW code
...
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27691 >
2024-02-28 05:45:37 +00:00
6c190bdfe9
intel/clc: Fix file descriptor leak
...
Fix defect reported by Coverity Scan.
Resource leak (RESOURCE_LEAK)
leaked_storage: Variable fp going out of scope leaks the storage it points to.
Fixes: 4fd7495c69vlee@freedesktop.org >
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27778 >
2024-02-28 04:30:33 +00:00
41722c6137
nak: Add support for imad on Volta+ and enable it in simple cases
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27159 >
2024-02-27 21:51:30 -06:00
a747cd1bd5
nak: Move NAK_FS_OUT_COLOR next to the enum
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27159 >
2024-02-27 21:51:30 -06:00
f4fb5277c3
nir: Add an imad opcode
...
Acked-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27159 >
2024-02-27 21:51:30 -06:00
1881d97c27
nak: Implement nir_op_iadd3 on SM70+
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27159 >
2024-02-27 21:51:29 -06:00
0c95d39309
zink: add nvk baseline
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27843 >
2024-02-27 21:50:47 -05:00
9ffb7e0179
zink: update nv blob baseline
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27843 >
2024-02-27 21:50:47 -05:00
efc4ac0d27
nak/sm50: sprinkle OpAnnotate in optimization passes
...
Not only do we want to know where an Op originated from, but also how it got
transformed along the way if possible. Preferably all the way to the final
machine code emitted.
This commit inserts OpAnnotates in some of the optimization passes when
map_instr() or Instr::new_boxed is used.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27158 >
2024-02-28 01:12:03 +00:00
feb2d3e1da
nak/sm50: support annotations through OpAnnotate
...
Add a new op to annotate the IR. This will help debugging and is only
in effect when NAK_DEBUG=annotate is set.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27158 >
2024-02-28 01:12:03 +00:00
a69bd9a70a
nak/sm50: add an annotate debug flag
...
Add a flag so that users can enable debug annotations when printing the IR.
This does nothing for now. A follow-up commit will actually implement
annotations.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27158 >
2024-02-28 01:12:03 +00:00
02774be708
nak/sm50: add a memstream abstraction
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27158 >
2024-02-28 01:12:03 +00:00
e57cf175e2
venus: move feedback on empty last batch to prior batch
...
For submissions with an empty last batch containing no cmd buffers but
with semaphores as zink does, adding feedback to that batch would make
it no longer empty and increase submission overhead on some drivers.
Since feedback order is enforced by barriers, the feedback cmds can
instead be appended to the previous batch (if it exists) so that the
last batch remains empty.
Signed-off-by: Juston Li <justonli@google.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27830 >
2024-02-28 00:56:26 +00:00
0586a3fb22
frontends/va/postproc: do not use efc if image is to be translated
...
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/10658
Signed-off-by: Thong Thai <thong.thai@amd.com >
Tested-by: Andrej Benz <hello@benz.dev >
Reviewed-by: Boyuan Zhang <Boyuan.Zhang@amd.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27802 >
2024-02-27 22:56:04 +00:00
b8c3d18fba
nvk: Advertise VK_EXT_shader_object
...
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9648
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27024 >
2024-02-27 22:17:09 +00:00
fb564040a7
nvk: Advertise VK_KHR_graphics_pipeline_library
...
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9635
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27024 >
2024-02-27 22:17:09 +00:00
813b253939
nvk: Switch to shader objects
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27024 >
2024-02-27 22:17:09 +00:00
4001658c18
nvk: Use vk_render_pass_state::attachments for write masks
...
This pulls everything into nvk_cmd_draw.c where it's a bit easier to
manage. When the time comes for switching to EXT_shader_object, this
will let us handle VK_EXT_dynamic_rendering_unused_attachments via the
common vk_pipeline code.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27024 >
2024-02-27 22:17:09 +00:00
839629634f
nvk: Move nir_lower_patch_vertices to nvk_lower_nir()
...
As long as it happens after we merge tess info between the two stages
(it does) then there's no need to have it in the pipeline code. It's
just an optimization anyway.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27024 >
2024-02-27 22:17:09 +00:00
bd76444257
nvk: Pass an array of descriptor sets to nvk_lower_nir
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27024 >
2024-02-27 22:17:09 +00:00
a4f519d72d
nvk: Move populate_fs_key to nvk_shader.c
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27024 >
2024-02-27 22:17:09 +00:00
045741ac30
nvk/shader: Refactor some helpers
...
This puts them in the form we need for vk_shader.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27024 >
2024-02-27 22:17:09 +00:00
626f38e25e
nvk: Populate vk_descriptor_set_layout::blake3
...
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27024 >
2024-02-27 22:17:09 +00:00
9308e8d90d
vulkan: Add generic graphics and compute VkPipeline implementations
...
These implementations are built on top of vk_shader. For the most part,
the driver shouldn't notice a difference between draws consuming
pipelines vs. draws consuming shaders. The only real difference is
that, when vk_driver_shader_ops::compile() is called for pipelines, a
struct vk_graphics_pipeline_state is provided. For shader objects, the
state object will be NULL indicating that all state is unknown. Besides
that, all the rest of the differences between Vulkan 1.0 pipelines,
VK_EXT_graphics_pipeline_library, and VK_EXT_shader_object are handled
by the Vulkan runtime code.
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27024 >
2024-02-27 22:17:09 +00:00
c488dc9f50
vulkan: Add a BLAKE3 hash to vk_descriptor_set_layout
...
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27024 >
2024-02-27 22:17:09 +00:00
682b99a63f
vulkan: Add push constant ranges to vk_pipeline_layout
...
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27024 >
2024-02-27 22:17:09 +00:00
e2cb395a1f
vulkan: Add a vk_pipeline base struct
...
We need to be able to thunk through a destroy callback if we want to
have different kinds of pipelines implemented in different parts of the
stack.
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27024 >
2024-02-27 22:17:09 +00:00
5e71e6f3f6
vulkan: Add a new dynamic state for render pass attachments
...
This is useful for implementing VK_EXT_dynamic_rendering_unused_attachments
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27024 >
2024-02-27 22:17:09 +00:00
6ec177b116
vulkan: Rework vk_render_pass_state::attachments
...
The new bitfield has a separat flag for each of the color attachments.
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27024 >
2024-02-27 22:17:09 +00:00
c09c086c12
vulkan: Add a vk_render_pass_state_has_attachment_info() helper
...
We already have a helper like this internally. Give it a better name
and expose it.
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27024 >
2024-02-27 22:17:09 +00:00
9f62008bff
vulkan: Add runtime code for VK_EXT_shader_object
...
This adds a new base vk_shader object along with vtables for creating,
binding, and working with shader objects.
Unlike other parts of the runtime, the new shader object code is a bit
more sanitized and opinionated than just handing you the Vulkan
entrypoints. For one thing, the create_shaders() calback takes a NIR
shader, not SPIR-V. Conversion of SPIR-V into NIR, handling of magic
meta NIR shaders, etc. is all done in common code. [De]serialization is
done via `struct blob` and the common code does a checksum of the binary
and handles rejecting invalid binaries based on shaderBinaryUUID and
shaderBinaryVersion. This should make life a bit easier for driver
authors as well as provides a bit nicer interface for building the
common pipeline implementation on top of shader objects.
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27024 >
2024-02-27 22:17:09 +00:00
0d225c9e43
vk/graphics_state: Add stubs required by VK_EXT_shader_objects
...
Because these functions were introduced by VK_EXT_shader_objects, we
technically have to expose them even though they have to do with NV
extensions that no one else supports.
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27024 >
2024-02-27 22:17:09 +00:00
657b8e5264
vk/graphics_state: Remove bogus assert in CmdSetSampleMaskEXT
...
We're supposed to just ignore samples above what we support, and there's
no VU matching this assert. Fixes a crash in
dEQP-VK.pipeline.shader_object_unlinked_spirv.extended_dynamic_state.misc.sample_shading_dynamic_sample_count.
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com >
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27024 >
2024-02-27 22:17:08 +00:00
6ad294202e
vulkan: Move the descriptor set limit to vk_limits.h
...
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27024 >
2024-02-27 22:17:08 +00:00
498d58a5f8
vulkan: Add a vk_get_subgroup_size() helper
...
No reason to duplicate this logic between pipelines and shader objects.
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27024 >
2024-02-27 22:17:08 +00:00
Caio Oliveira