Commit Graph

273 Commits

Author SHA1 Message Date
Liam 152beedc51 shader_recompiler: always declare image format for image buffers 2023-09-02 17:25:00 -04:00
Liam 024fb4fae9 shader_recompiler: fix emulation of 3D textureGrad 2023-09-01 22:41:01 -04:00
Kelebek1 c7430e51e3 Remove memory allocations in some hot paths 2023-06-22 08:05:10 +01:00
bunnei 3571f28cde video_core: Enable support_descriptor_aliasing on Turnip, disable storage atomic otherwise. 2023-06-03 00:05:58 -07:00
Billy Laws cfbe4b09eb Avoid using VectorExtractDynamic for subgroup mask on Adreno GPUs
This crashes their shader compiler for some reason.
2023-06-03 00:05:31 -07:00
Billy Laws 2beb3051c1 Implement scaled vertex buffer format emulation
These formats are unsupported by mobile GPUs so they need to be emulated in shaders instead.
2023-06-03 00:05:31 -07:00
Kelebek1 2e478a66cf Fix Tears of the Kingdom flickering clouds and depths. 2023-05-11 19:25:24 +01:00
Kelebek1 dc9546fe26 Define SampleMask as an array 2023-04-30 18:37:37 +01:00
Wollnashorn d53aa13a33 shader_recompiler: Use vector arithmetic rather than component-wise in ImageGatherSubpixelOffset
Should be more efficient and better readable
2023-04-08 16:13:08 +02:00
Wollnashorn 45fb154f0d shader_recompiler: Add subpixel offset for correct rounding at ImageGather
On AMD a subpixel offset of 1/512 of the texel size is applied to the texture coordinates at a ImageGather call to ensure the rounding at the texel centers is done the same way as in Maxwell or other Nvidia architectures.
See https://www.reedbeta.com/blog/texture-gathers-and-coordinate-precision/ for more details why this might be necessary.

This should fix shadow artifacts at object edges in Zelda: Breath of the Wild (#9957, #6956).
2023-04-08 16:12:30 +02:00
ameerj bbb60f60d2 spirv: Fix TXQ with MSAA textures 2023-01-29 13:47:30 -05:00
liamwhite a5bc2444d2 Merge pull request #9694 from ameerj/txq-mips
shader_recompiler: TXQ: Skip QueryLevels when possible
2023-01-29 12:27:26 -05:00
bunnei da76935b7d Merge pull request #9682 from ameerj/shader-s32
shader_recompiler: Remove S32 IR type
2023-01-28 14:00:11 -08:00
ameerj 2d2558797c shader_recompiler: TXQ: Skip QueryLevels when possible 2023-01-28 16:25:18 -05:00
ameerj 382717743a shader_recompiler: Remove S32 IR type
The frontend IR opcodes do not distinguish between signed and unsigned integer types.

Fixes broken shaders when IR validation/graphics debugging is enabled for shaders that used BitCastS32F32
2023-01-25 22:03:15 -05:00
Liam ac94832cb6 spirv: fix multisampled image fetch 2023-01-23 12:03:19 -05:00
Billy Laws 535e297ebd Run clang-format 2023-01-05 22:18:10 +00:00
Billy Laws c7c901dd63 shader_recompiler: Fix shuffle partitioning for >64 invoc-per-subgroup GPUs
The existing implementation only supports 64 invoc-per-subgroup GPUs, and misbehaves on adreno when invocations need to be split into 4 emulated subgroups.
2023-01-05 22:13:07 +00:00
Billy Laws 3772cfc976 shader_recompiler: SPIRV: Only enable int64 feature when supported 2023-01-05 22:13:07 +00:00
Billy Laws 77aea7e2b4 Vulkan: Add a workaround for input_position on Adreno drivers
Adreno drivers will crash compiling geometry shaders if the input position is not wrapped in a gl_in struct.
2023-01-05 22:13:07 +00:00
Fernando Sahmkow 1fdf24a081 Video_core: Address feedback 2023-01-04 14:39:42 -05:00
Fernando Sahmkow 7842543573 MacroHLE: Add HLE replacement for base vertex and base instance. 2023-01-01 16:43:57 -05:00
liamwhite db106151d7 Merge pull request #7450 from FernandoS27/ndc-vulkan
Vulkan: Add support for VK_EXT_depth_clip_control.
2022-12-17 16:08:10 -05:00
FernandoS27 11f29ec173 Vulkan: Add support for VK_EXT_depth_clip_control. 2022-12-13 21:39:18 -05:00
Liam 3b502d3095 spirv_emit_context: declare GroupNonUniform capability for SubgroupLocalInvocationId 2022-12-13 18:25:53 -05:00
Liam 0a077cb467 Vulkan: update initialization
Co-authored-by: bylaws <bylaws@users.noreply.github.com>
2022-11-27 14:58:28 -05:00
Liam 4b905e9680 spirv_emit_context: add missing flat decoration 2022-11-18 22:05:28 -05:00
liamwhite d96606b091 Merge pull request #9253 from vonchenplus/attr_layer
shader: Implement miss attribute layer
2022-11-18 22:04:18 -05:00
FengChen 12f2ab01bb shader: Implement miss attribute layer 2022-11-17 22:45:14 +08:00
FengChen 348c6e5b28 video_core: Fix few issues in Tess stage 2022-11-07 15:42:42 +08:00
Feng Chen bcf6a7c0dd video_core: Fix SNORM texture buffer emulating error (#9001) 2022-11-04 02:39:42 -04:00
FengChen 38e9c78843 video_core: Generate mipmap texture by drawing 2022-09-20 11:55:43 +08:00
FengChen 98656c14a8 video_code: support rectangle texture 2022-08-25 12:45:58 +08:00
Kelebek1 9427a0f535 Add missed shader defines. Fixes Xenoblade Chronicles 3 booting with Vulkan. 2022-07-29 06:12:39 +01:00
Morph 2b87305d31 general: Convert source file copyright comments over to SPDX
This formats all copyright comments according to SPDX formatting guidelines.
Additionally, this resolves the remaining GPLv2 only licensed files by relicensing them to GPLv2.0-or-later.
2022-04-23 05:55:32 -04:00
Liam 528cba63b6 shader_recompiler: support const buffer indirect addressing on OpenGL SPIR-V 2022-04-01 11:17:54 -04:00
Billy Laws ef9b177ecb Include <bit> header when std::count{r,l}_zero is used
Needed for compilation with older libc++ releases
2022-03-22 21:11:24 +00:00
ameerj 0df188787a shader_recompiler: Reduce unused includes 2022-03-20 02:25:08 -04:00
Liam 83b1c9de37 Address review comments 2022-03-17 14:48:18 -04:00
Liam 95b9f62686 shader_recompiler: Use functions for indirect const buffer accesses 2022-03-17 13:30:21 -04:00
Liam f3382e6339 Address review comments 2022-03-17 09:30:41 -04:00
Liam 8515dba789 shader: add support for const buffer indirect addressing 2022-03-14 19:43:32 -04:00
ameerj 048522e98b emit_spirv, vk_compute_pass: Resolve VS2022 compiler errors 2022-03-12 02:54:33 -05:00
ameerj 0c5e641757 shaders: Add U64->U32x2 Atomic fallback functions 2022-01-29 19:55:53 -05:00
ameerj c3d768426c spirv_atomic: Define U32x2 storage buffers for 64-bit storage atomics
Some drivers do not support 64-bit atomics, and fallback to atomically modifying U32x2 vectors. This change ensures that U32x2 storage vectors are defined in the spir-v shader when 64-bit atomics are used.

Fixes a hang on some devices, notably Intel GPUs, when booting Pokemon Legends Arceus
2022-01-28 19:00:04 -05:00
ameerj 712b7883be emit_spirv: Add Xfb execution mode when transform feedback is used
Fixes Transform Feedback on Vulkan AMD drivers.
2022-01-28 16:32:48 -05:00
v1993 970b362ca1 shader_recompiler: fix potential OOB access
Found by static analysis with PVS-Studio. Original check wasn't actually checking for OOB and would segfault in case of it.
2022-01-17 21:50:51 +03:00
Fernando S 8a790b09a7 Merge pull request #7629 from ameerj/nv-driver-fixes
shaders: Add fixes for NVIDIA drivers 495+
2022-01-03 00:39:59 +01:00
ameerj 22a4d26479 shader: Add integer attribute get optimization pass
Works around an nvidia driver bug, where casting the integer attributes to float and back to an integer always returned 0.
2021-12-29 19:03:49 -05:00
Matías Locatti 244dff4e6e Empty spaces 2021-12-28 18:50:51 -03:00