AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
200,417 Commits 1 Branch 0 Tags
91f19bcbe072e663c1c4e9cd81f12ed0824a30e9
Commit Graph

200417 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Connor Abbott 91f19bcbe0 ir3: Plumb through two-dimensional UAV loads 
There is native support for D3D-style untyped UAVs, which are an unsized
array of "records."

This will be needed for acceleration structures, because normal SSBO
descriptors aren't large enough to cover all the 128-byte instance
descriptors for the maximum number of instances (2**24).

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28447>
 2025-01-20 01:22:23 +00:00
Connor Abbott b6ae20f2d6 ir3: Support assembling/disassembling ray_intersection and resbase 
These instructions are used for ray tracing.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28447>
 2025-01-20 01:22:23 +00:00
Connor Abbott 3a5cb40c1f ir3: Allow collect sources to be undef 
Undef collect sources will never have a move materialized for them.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28447>
 2025-01-20 01:22:22 +00:00
Connor Abbott c2c48db6a4 vk/bvh: Fix clang build error with turnip 
Make sure that struct vk_bvh_geometry_data is defined before
vk_fill_geometry_data(), to fix this error:

In file included from ../src/freedreno/vulkan/tu_acceleration_structure.cc:29:
../src/vulkan/runtime/vk_acceleration_structure.h:138:1: error: 'vk_fill_geometry_data' has C-linkage specified, but returns incomplete type 'struct vk_bvh_geometry_data' which could be incompatible with C [-Werror,-Wreturn-type-c-linkage]

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28447>
 2025-01-20 01:22:22 +00:00
Caio Oliveira 793cba0e6f intel/brw: Apply conventions to lower_src_modifiers helper 
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33110>
 2025-01-19 08:24:09 -08:00
Timothy Arceri 7d41cfa1a9 glsl: enable layout qualifier if OVR_multiview enabled 
OVR_multiview requires 1.30 but makes use of layout qualifier

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Acked-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33074>
 2025-01-19 01:10:54 +00:00
Caio Oliveira d7d210fed4 intel/brw: Move shuffle_from_32bit_read implementation to brw_builder 
Make it a member function for convenience -- since another
member function uses it.

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33108>
 2025-01-18 20:48:57 +00:00
Caio Oliveira b3001e4946 intel/brw: Move a few builder helpers to brw_builder.h/cpp 
Add brw prefix when necessary.

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33108>
 2025-01-18 20:48:57 +00:00
Lionel Landwerlin 10a4dc529f blorp: disable PS shaders with depth/stencil HiZ ops 
Found on simulation, complaining about SIMD32 shaders enabled when
using MSAA 16x.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: mesa-stable
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30753>
 2025-01-18 17:52:19 +00:00
Caio Oliveira 1043187ec6 intel/brw: Stop using namespace for brw_builder 
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33076>
 2025-01-18 16:12:56 +00:00
Caio Oliveira 5ac82efd35 intel/brw: Rename fs_builder to brw_builder 
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33076>
 2025-01-18 16:12:55 +00:00
Caio Oliveira f2d4c9db92 intel/brw: Rename brw_fs_builder.h to brw_builder.h 
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33076>
 2025-01-18 16:12:54 +00:00
Caio Oliveira f0fe0026c0 intel/brw: Remove extra wrapping around fs_visitor in tests 
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33100>
 2025-01-18 07:41:35 -08:00
Konstantin Seurer 70767b9cd4 docs: Add documentation for NIR unit testing 
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32644>
 2025-01-18 11:02:25 +00:00
Konstantin Seurer 01ec2f59a4 nir/print: Do not print trailing spaces after preds/succs 
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32644>
 2025-01-18 11:02:25 +00:00
Konstantin Seurer 6ddd791d6b nir: Add a test runner 
The test runner can automatically update reference shaders when the
result changes.

Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32644>
 2025-01-18 11:02:25 +00:00
Konstantin Seurer eb3ab68e5e nir/tests: Add reference shaders 
Acked-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32644>
 2025-01-18 11:02:25 +00:00
Konstantin Seurer 8838a0c595 nir/tests: Add a helper for comparing a shader against a string 
This allows unit tests to compare against a reference nir shader instead
of implementing checks for interesting instructions/CF nodes.

Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32644>
 2025-01-18 11:02:25 +00:00
Konstantin Seurer 6d1d15183f nir/tests: Improve shader creation 
Sets some fields so they are not printed and allows specifying a stage.
This decreases the size of reference shaders.

Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32644>
 2025-01-18 11:02:25 +00:00
Konstantin Seurer 305be9cf5e nir/print: Print less unused shader info 
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32644>
 2025-01-18 11:02:25 +00:00
Caio Oliveira 94fa449318 intel/brw: Add missing cases to flags_written() 
These virtual opcodes will write the whole flag set, either directly
(via brw_fill_flag()) or indirectly by using LOAD_LIVE_CHANNELS.

Issue was found when analysing a hang that would disappear
if the lowering of those opcodes was pulled all the way up
right before brw_opt_cmod_propagation (which uses the
flags_written).

Fixes: 019770f026 ("intel/brw: Add SHADER_OPCODE_VOTE_*")
Fixes: 2bd7592b0b ("intel/brw: Add SHADER_OPCODE_BALLOT")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/12347
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/12479
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33085>
 2025-01-18 05:30:23 +00:00
Samuel Pitoiset 0a4584a684 radv: bump maxViewportDimensions to 32K on GFX12 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33064>
 2025-01-17 21:10:23 +00:00
Mike Blumenkrantz 3d7d303dc5 zink: enable maintenance8 
this avoids validation errors from 3d<->2d blits

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33102>
 2025-01-17 20:46:43 +00:00
Mike Blumenkrantz 0d29ddb328 lavapipe: maintenance8 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33098>
 2025-01-17 19:41:45 +00:00
Mike Blumenkrantz c285a2a354 lavapipe: support zs<->color copies 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33098>
 2025-01-17 19:41:45 +00:00
Mike Blumenkrantz 00d478d96d lavapipe: abort on unsupported depth copy ops 
better than silently failing

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33098>
 2025-01-17 19:41:45 +00:00
Mike Blumenkrantz 7e8c13ea64 lavapipe: fix 3D->2D blitting 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33098>
 2025-01-17 19:41:45 +00:00
Samuel Pitoiset 2ba91d1deb radv: promote VK_EXT_depth_clamp_zero_one to KHR 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33090>
 2025-01-17 19:02:02 +00:00
Lionel Landwerlin f96e95fcc9 anv: remove print lowering 
This is handled by the back compiler.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33067>
 2025-01-17 18:09:46 +00:00
Lionel Landwerlin e1074f5bd4 anv: update debug printf example code 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33067>
 2025-01-17 18:09:45 +00:00
Lionel Landwerlin 2603dbd796 nir: make lower-level printf helper respect buffer size 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33067>
 2025-01-17 18:09:45 +00:00
Lionel Landwerlin 58a3ef4160 anv: handle printf buffer size relocations 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33067>
 2025-01-17 18:09:45 +00:00
Lionel Landwerlin d63b5fc8c5 brw: handle load_printf_buffer_size intrinsic 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33067>
 2025-01-17 18:09:45 +00:00
Alyssa Rosenzweig c1e685bfe9 util,vulkan,asahi,hk: hash format strings 
flag day change to use the new infra. as-is this is a bit pointless, but it
unblocks the new bindgen work.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33067>
 2025-01-17 18:09:45 +00:00
Alyssa Rosenzweig 43e79b26de nir/lower_printf: drop static buffer addr lowering 
no longer used, replaced by the new pass.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33067>
 2025-01-17 18:09:45 +00:00
Alyssa Rosenzweig c4bd2e691b agx: defer printf address lowering 
this will be needed with vtn_bindgen2

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33067>
 2025-01-17 18:09:45 +00:00
Alyssa Rosenzweig 07ad850787 nir: add nir_lower_printf_buffer pass 
this is a helper for lowering the printf buffer intrinsics to constants for
backend convenience.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33067>
 2025-01-17 18:09:45 +00:00
Alyssa Rosenzweig 7bc9bbcc6e nir/lower_printf: support dynamic buffer size 
this is required for vtn_bindgen2 where we don't know the buffer size until
the driver-specific code paths, but we need to lower printf (to hash format
strings) in common code. so defer the buffer size decision to an intrinsic.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33067>
 2025-01-17 18:09:45 +00:00
Alyssa Rosenzweig 6db9218ec3 nir/lower_printf: add option to hash format strings 
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33067>
 2025-01-17 18:09:45 +00:00
Alyssa Rosenzweig dd66287e63 util/u_printf: allow printing from singleton 
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33067>
 2025-01-17 18:09:45 +00:00
Alyssa Rosenzweig 007f60c8b8 util/u_printf: add singleton implementation 
Currently, nir_lower_printf depends on a per-nir_shader table, writing out
indices into the printf buffer. This works for real OpenCL implementations
(rusticl, microsoft) which can associate the printf buffer with a particular
kernel, I guess. (Actually it's not clear to me that it works well there either
but that's not my problem.)

This mechanism is unsuitable for internal driver shaders, where printfs with
unique format strings can come from many different nir_shaders. There are two
current solutions in tree to this for driver CL:

* Honeykrisp: Only use one single nir_shader (libagx). This prevents us from
  using printf in common CL and requires extra driver tracking. It won't work
  with my upcoming vtn_bindgen rework, which is why I'm addressing this now.

* Anv: Offset format-string indices by a dynamic "base identifier" using relocs
  or a push constant, then pool format strings into a table from nir_shader's
  across the device. The problem here is that these indices now depend on the
  order that nir_shaders are seen (which causes a mess for caching if relocs are
  used, or requires extra push constants and extra bookkeping if relocs aren't
  used). And the driver tracking required to do this pooling correctly is even
  more complicated than what Honeykrisp does. I do not want every driver in-tree
  needing to go down this path, and it wouldn't work with my upcoming
  vtn_bindgen.

This MR introduces an alternate approach: rather than writing indices into the
table, we instead hash the format string itself and write the hash. That doesn't
depend on what nir_shader we came from, so we can freely mix & match and get
consistent hashes. That greatly alleviates driver tracking burden. To make that
possible, we need a global hash table mapping hashed format identifiers to the
format strings themselves.

That approach still requires a step to "register" format strings into the table.
That step would not be required if we wrote the actual strings themselves into
the table, but that was ruled out for performance/code size reasons. However, we
do not want drivers to need to explicitly register all the strings they use,
because once we have OpenCL in common code via vtn_bindgen2, drivers won't know
all the strings they use. Fortunately, there's a neat solution for that too.

By making this global table a singleton (with internal locking), vtn_bindgen2
can automatically register format strings via a static constructor. In
conjunction with the infrastructure added here, that eliminates all driver
bookkeeping required for format-strings.

The code itself is inspired by the glsl type singleton. Is it pretty? Not
really, but it gets the job done well.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33067>
 2025-01-17 18:09:45 +00:00
Alyssa Rosenzweig 48dbfba17a util: add u_printf_hash helper 
for hash-based printf. this just shells out to XXH - convenience wrapper.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33067>
 2025-01-17 18:09:45 +00:00
Alyssa Rosenzweig e1368f0a30 nir,util: move printf serializing into util 
there's nothing NIR specific here and these routines will be useful otherwise.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33067>
 2025-01-17 18:09:45 +00:00
Alyssa Rosenzweig 47e16cab5e nir/lower_printf: drop default max buffer size 
no uses and it doesn't make sense.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33067>
 2025-01-17 18:09:45 +00:00
Alyssa Rosenzweig 621ff262bc nir/lower_printf: drop null check 
we derefernce options above.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33067>
 2025-01-17 18:09:45 +00:00
Alyssa Rosenzweig e7a1d704d0 intel: set max_buffer_size to nir_lower_printf 
instead of relying on an implicit value which doesn't make much sense.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33067>
 2025-01-17 18:09:45 +00:00
Gurchetan Singh 61d66a6ea7 gfxstream: fix some integration bugs 
- Fix end2end test target (AOSP-only)
- Fix decoder to use host vk_util.h, not guest
- typo fixes

Reviewed-by: Marcin Radomski <dextero@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33077>
 2025-01-17 17:26:45 +00:00
Samuel Pitoiset 78f5dfca86 vulkan: Update XML and headers to 1.4.305 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33089>
 2025-01-17 16:42:27 +00:00
Mike Blumenkrantz 667745857d lavapipe: handle VK_REMAINING_ARRAY_LAYERS with HIC 
this otherwise goes oob and crashes

cc: mesa-stable

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33071>
 2025-01-17 15:35:09 +00:00
Lu Yao 7679102b45 zink: fix decomposed_attrs val error when zink_vs_key->size is 4 
decomposed_attrs and decomposed_attrs_without_w should get from vs->u32
when zink_vs_key->size is 4.

Fixes: 19fbdb9064 ("zink: move shader keys to be persistent on pipeline state")
Signed-off-by: Lu Yao <yaolu@kylinos.cn>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33079>
 2025-01-17 14:51:49 +00:00