Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Merge pull request #12066 from ameerj/nvidia-nsanity | liamwhite | 2023-11-19 | 4 | -0/+82 |
|\ | | | | | shader_recompiler: add byteswap pattern workaround for Nvidia | ||||
| * | shader_recompiler: add byteswap pattern workaround for Nvidia | Ameer J | 2023-11-18 | 4 | -0/+82 |
| | | |||||
* | | shader_recompiler: Fix spelling of "derivate" (#12067) | Ameer J | 2023-11-18 | 9 | -58/+59 |
|/ | |||||
* | Merge pull request #11847 from ameerj/glsl-shfl-fix | liamwhite | 2023-10-23 | 2 | -4/+9 |
|\ | | | | | emit_glsl_warp: Fix shfl_in_bounds conditional | ||||
| * | emit_glsl_warp: Fix shfl_in_bounds conditional | Ameer J | 2023-10-22 | 2 | -4/+9 |
| | | |||||
* | | Manually robust on Maxwell and earlier | Kelebek1 | 2023-10-19 | 3 | -4/+59 |
|/ | |||||
* | ci: fix new codespell errors | Liam | 2023-10-03 | 1 | -1/+1 |
| | |||||
* | emit_spirv: fix incorrect use of descriptor index in image atomics | Liam | 2023-09-22 | 2 | -13/+9 |
| | |||||
* | shader_recompiler: skip sampler for buffer textures (#11435) | liamwhite | 2023-09-14 | 3 | -6/+2 |
| | |||||
* | Merge pull request #11436 from liamwhite/bad-format | liamwhite | 2023-09-10 | 1 | -1/+6 |
|\ | | | | | shader_recompiler: always declare image format for image buffers | ||||
| * | shader_recompiler: always declare image format for image buffers | Liam | 2023-09-02 | 1 | -1/+6 |
| | | |||||
* | | Merge pull request #11434 from danilaml/fix-warnings | liamwhite | 2023-09-06 | 1 | -2/+0 |
|\ \ | |/ |/| | msvc: set warning level to /W4 globally | ||||
| * | msvc: set warning level to /W4 globally | Danila Malyutin | 2023-09-03 | 1 | -2/+0 |
| | | | | | | | | And fix a bunch of warnings | ||||
* | | Merge pull request #11383 from FernandoS27/are-you-a-wabbit | liamwhite | 2023-09-02 | 1 | -1/+2 |
|\ \ | |/ |/| | Fix regressions that damaged compute indirect & use reinterpret for copies with different byteblocksizes | ||||
| * | Shader Recompiler: Auto stub special registers and dump pipelines on exception. | Fernando Sahmkow | 2023-08-27 | 1 | -1/+2 |
| | | |||||
* | | shader_recompiler: fix emulation of 3D textureGrad | Liam | 2023-09-02 | 2 | -3/+34 |
|/ | |||||
* | Shader Recomnpiler: implement textuzreGrad 3D emulation constant propagation | Fernando Sahmkow | 2023-08-19 | 8 | -11/+261 |
| | |||||
* | Fix shader dumps with nvdisasm | Kelebek1 | 2023-08-03 | 1 | -1/+1 |
| | | | | | skip fragment shaders when rasterizer is disabled initialize env_ptrs | ||||
* | ssa_rewrite_pass: use proper maps | Liam | 2023-07-23 | 1 | -6/+5 |
| | |||||
* | shaders: Track local memory usage | ameerj | 2023-06-26 | 2 | -0/+5 |
| | |||||
* | emit_glasm: Fix lmem size computation | ameerj | 2023-06-26 | 1 | -1/+1 |
| | |||||
* | Remove memory allocations in some hot paths | Kelebek1 | 2023-06-22 | 4 | -4/+5 |
| | |||||
* | Merge pull request #10699 from liamwhite/conditional-barrier | Matías Locatti | 2023-06-12 | 5 | -0/+51 |
|\ | | | | | shader_recompiler: remove barriers in conditional control flow when device lacks support | ||||
| * | shader_recompiler: remove barriers in conditional control flow when device lacks support | Liam | 2023-06-10 | 5 | -0/+51 |
| | | |||||
* | | shader_recompiler: translate f64 to f32 when unsupported on host | Liam | 2023-06-10 | 5 | -0/+191 |
|/ | |||||
* | video_core: Enable support_descriptor_aliasing on Turnip, disable storage atomic otherwise. | bunnei | 2023-06-03 | 1 | -0/+10 |
| | |||||
* | Avoid using VectorExtractDynamic for subgroup mask on Adreno GPUs | Billy Laws | 2023-06-03 | 2 | -1/+18 |
| | | | | This crashes their shader compiler for some reason. | ||||
* | Implement scaled vertex buffer format emulation | Billy Laws | 2023-06-03 | 5 | -49/+75 |
| | | | | These formats are unsupported by mobile GPUs so they need to be emulated in shaders instead. | ||||
* | shader_recompiler: fix copy-paste error | Liam | 2023-05-26 | 1 | -1/+1 |
| | |||||
* | Fix Tears of the Kingdom flickering clouds and depths. | Kelebek1 | 2023-05-11 | 2 | -12/+3 |
| | |||||
* | Define SampleMask as an array | Kelebek1 | 2023-04-30 | 2 | -2/+4 |
| | |||||
* | shader_recompiler: Use vector arithmetic rather than component-wise in ImageGatherSubpixelOffset | Wollnashorn | 2023-04-08 | 1 | -18/+9 |
| | | | | Should be more efficient and better readable | ||||
* | video_core: Enable ImageGather with subpixel offset on Intel | Wollnashorn | 2023-04-08 | 1 | -3/+3 |
| | |||||
* | shader_recompiler: Add subpixel offset for correct rounding at `ImageGather` | Wollnashorn | 2023-04-08 | 3 | -0/+72 |
| | | | | | | | 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). | ||||
* | general: fix spelling mistakes | Liam | 2023-03-12 | 2 | -41/+41 |
| | |||||
* | buffer_cache: Add logic for non-NVN storage buffer tracking | ameerj | 2023-02-25 | 1 | -1/+7 |
| | |||||
* | Merge pull request #9588 from liamwhite/bylaws-reverts | liamwhite | 2023-02-19 | 4 | -12/+6 |
|\ | | | | | Revert "shader_recompiler: Align SSBO offsets to meet host requirements" | ||||
| * | Revert "shader_recompiler: Align SSBO offsets to meet host requirements" | Liam | 2023-01-07 | 4 | -12/+6 |
| | | | | | | | | This reverts commit 8804a4eb23e0c4f3e4bab03dee7c204bd38bf21e. | ||||
* | | glsl_emit_context: Remove redeclarations of gl_SampleID and gl_SampleMask | ameerj | 2023-02-09 | 1 | -6/+0 |
| | | | | | | | | These built-ins seem to be available without needing to be declared for fragment shaders, similar i.e. to gl_FragDepth | ||||
* | | shader_recompiler/value.h: Remove lingering references to S32 | ameerj | 2023-02-04 | 1 | -11/+0 |
| | | |||||
* | | Merge pull request #9703 from ameerj/txq-ms | liamwhite | 2023-02-02 | 5 | -18/+51 |
|\ \ | | | | | | | shaders: Fix TXQ with MSAA textures | ||||
| * | | spirv: Fix TXQ with MSAA textures | ameerj | 2023-01-29 | 3 | -8/+19 |
| | | | |||||
| * | | emit_glasm_image: Fix TXQ with MSAA textures | ameerj | 2023-01-29 | 1 | -1/+9 |
| | | | |||||
| * | | emit_glsl_image: Implement TXQ with MSAA textures | ameerj | 2023-01-29 | 1 | -9/+23 |
| | | | | | | | | | | | | Also fixes for texture buffers, which do not have mips eithers. | ||||
* | | | Move to Clang Format 15 | Levi Behunin | 2023-01-30 | 3 | -10/+10 |
|/ / | | | | | | | | | | | Depends on https://github.com/yuzu-emu/build-environments/pull/69 clang-15 primary run | ||||
* | | Merge pull request #9699 from ameerj/texture-pass-desc | liamwhite | 2023-01-29 | 1 | -2/+9 |
|\ \ | | | | | | | texture_pass: Fix texture descriptors comparisons | ||||
| * | | texture_pass: Fix texture descriptors comparisons | ameerj | 2023-01-29 | 1 | -2/+9 |
| | | | |||||
* | | | Merge pull request #9698 from ameerj/texture-pass-handle | liamwhite | 2023-01-29 | 1 | -7/+7 |
|\ \ \ | | | | | | | | | texture_pass: Refactor texture handle retrieval | ||||
| * | | | texture_pass: Refactor texture handle retrieval | ameerj | 2023-01-29 | 1 | -7/+7 |
| | | | | |||||
* | | | | Merge pull request #9694 from ameerj/txq-mips | liamwhite | 2023-01-29 | 11 | -29/+37 |
|\ \ \ \ | |/ / / |/| | | | shader_recompiler: TXQ: Skip QueryLevels when possible | ||||
| * | | | shader_recompiler: TXQ: Skip QueryLevels when possible | ameerj | 2023-01-28 | 11 | -29/+37 |
| |/ / | |||||
* | | | Merge pull request #9687 from ameerj/ogl-shader-ms | bunnei | 2023-01-29 | 4 | -33/+46 |
|\ \ \ | | | | | | | | | glasm, glsl: Implement multisampled Image Fetch | ||||
| * | | | emit_glsl_image: Fix ImageFetch for MSAA textures | ameerj | 2023-01-28 | 1 | -6/+11 |
| | | | | |||||
| * | | | glasm: Add MS sampler types | ameerj | 2023-01-27 | 2 | -5/+8 |
| | | | | |||||
| * | | | glsl: Add MS sampler types | ameerj | 2023-01-27 | 1 | -22/+27 |
| |/ / | |||||
* | | | Merge pull request #9682 from ameerj/shader-s32 | bunnei | 2023-01-28 | 13 | -46/+19 |
|\ \ \ | |/ / |/| | | shader_recompiler: Remove S32 IR type | ||||
| * | | shader_recompiler: Remove S32 IR type | ameerj | 2023-01-26 | 13 | -46/+19 |
| | | | | | | | | | | | | | | | | | | 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 | ||||
* | | | spirv: fix multisampled image fetch | Liam | 2023-01-23 | 4 | -2/+16 |
|/ / | |||||
* / | Avoid OOB array access reading passthrough attr mask | Billy Laws | 2023-01-07 | 1 | -1/+1 |
|/ | | | YFC 1.5 extended the size of the varying mask used to hold passthrough attrs without considering this | ||||
* | Run clang-format | Billy Laws | 2023-01-05 | 5 | -23/+33 |
| | |||||
* | shader_recompiler: Fix shuffle partitioning for >64 invoc-per-subgroup GPUs | Billy Laws | 2023-01-05 | 1 | -30/+28 |
| | | | | The existing implementation only supports 64 invoc-per-subgroup GPUs, and misbehaves on adreno when invocations need to be split into 4 emulated subgroups. | ||||
* | shader_recompiler: Add support for lowering geometry passthrough | Billy Laws | 2023-01-05 | 2 | -40/+67 |
| | | | | Reuses most of the existing code for generating the gl_Layer passthrough. Fixes geometry in Nier: Automata on GPUs without HW passthrough support. | ||||
* | shader_recompiler: Align SSBO offsets to meet host requirements | Billy Laws | 2023-01-05 | 4 | -6/+11 |
| | | | | We can take advantage of SSBO addresses being passed in a constant bufer to account for the extra alignment requirements in the shader itself. | ||||
* | shader_recompiler: SPIRV: Only enable int64 feature when supported | Billy Laws | 2023-01-05 | 1 | -1/+1 |
| | |||||
* | shader_recompiler: Add comparison operators to descriptor types | Billy Laws | 2023-01-05 | 1 | -0/+12 |
| | |||||
* | Vulkan: Add a workaround for input_position on Adreno drivers | Billy Laws | 2023-01-05 | 4 | -11/+41 |
| | | | | Adreno drivers will crash compiling geometry shaders if the input position is not wrapped in a gl_in struct. | ||||
* | Video_core: Address feedback | Fernando Sahmkow | 2023-01-04 | 9 | -0/+39 |
| | |||||
* | ShaderCompiler: Inline driver specific constants. | Fernando Sahmkow | 2023-01-03 | 2 | -1/+34 |
| | |||||
* | MacroHLE: Final cleanup and fixes. | Fernando Sahmkow | 2023-01-01 | 1 | -2/+2 |
| | |||||
* | MacroHLE: Add OpenGL Support | Fernando Sahmkow | 2023-01-01 | 2 | -1/+13 |
| | |||||
* | MacroHLE: Add HLE replacement for base vertex and base instance. | Fernando Sahmkow | 2023-01-01 | 12 | -6/+91 |
| | |||||
* | Merge pull request #7450 from FernandoS27/ndc-vulkan | liamwhite | 2022-12-17 | 3 | -3/+5 |
|\ | | | | | Vulkan: Add support for VK_EXT_depth_clip_control. | ||||
| * | Vulkan: Add support for VK_EXT_depth_clip_control. | FernandoS27 | 2022-12-14 | 3 | -3/+5 |
| | | |||||
* | | spirv_emit_context: declare GroupNonUniform capability for SubgroupLocalInvocationId | Liam | 2022-12-14 | 1 | -0/+2 |
|/ | |||||
* | Merge pull request #9300 from ameerj/pch | liamwhite | 2022-12-03 | 3 | -1/+12 |
|\ | | | | | CMake: Use precompiled headers to improve compile times | ||||
| * | CMake: Consolidate common PCH headers | ameerj | 2022-12-01 | 1 | -7/+1 |
| | | |||||
| * | CMake: Use precompiled headers | ameerj | 2022-11-30 | 2 | -0/+18 |
| | | |||||
| * | value.h: remove recursive include | ameerj | 2022-11-30 | 1 | -1/+0 |
| | | |||||
* | | Merge pull request #9289 from liamwhite/fruit-company | liamwhite | 2022-12-03 | 6 | -3/+9 |
|\ \ | | | | | | | general: fix compile for Apple Clang | ||||
| * | | general: fix compile for Apple Clang | Liam | 2022-11-23 | 6 | -3/+9 |
| |/ | |||||
* | | Merge pull request #9303 from liamwhite/new-vulkan-init | Matías Locatti | 2022-12-02 | 3 | -19/+31 |
|\ \ | | | | | | | Vulkan: update initialization | ||||
| * | | Vulkan: update initialization | Liam | 2022-11-27 | 3 | -19/+31 |
| |/ | | | | | | | Co-authored-by: bylaws <bylaws@users.noreply.github.com> | ||||
* / | shader_recompiler: add gl_Layer translation GS for older hardware | Liam | 2022-12-01 | 7 | -1/+165 |
|/ | |||||
* | spirv_emit_context: add missing flat decoration | Liam | 2022-11-19 | 1 | -0/+1 |
| | |||||
* | Merge pull request #9253 from vonchenplus/attr_layer | liamwhite | 2022-11-19 | 5 | -0/+13 |
|\ | | | | | shader: Implement miss attribute layer | ||||
| * | shader: Implement miss attribute layer | FengChen | 2022-11-17 | 5 | -0/+13 |
| | | |||||
* | | Merge pull request #9167 from vonchenplus/tess | liamwhite | 2022-11-11 | 16 | -5/+60 |
|\ \ | | | | | | | video_core: Fix few issues in Tess stage | ||||
| * | | video_core: Fix few issues in Tess stage | FengChen | 2022-11-07 | 16 | -5/+60 |
| |/ | |||||
* / | ir/texture_pass: Use host_info instead of querying Settings::values (#9176) | Morph | 2022-11-11 | 4 | -8/+13 |
|/ | |||||
* | video_core: Fix SNORM texture buffer emulating error (#9001) | Feng Chen | 2022-11-04 | 15 | -16/+115 |
| | |||||
* | Merge pull request #8858 from vonchenplus/mipmap | bunnei | 2022-11-04 | 20 | -1/+163 |
|\ | | | | | video_core: Generate mipmap texture by drawing | ||||
| * | Merge branch 'master' into mipmap | Feng Chen | 2022-09-20 | 1 | -5/+5 |
| |\ | |||||
| * | | video_core: Generate mipmap texture by drawing | FengChen | 2022-09-20 | 20 | -1/+163 |
| | | | |||||
* | | | Revert "shader_recompiler/dead_code_elimination: Add DeadBranchElimination pass" | Feng Chen | 2022-10-25 | 3 | -98/+9 |
| | | | |||||
* | | | Merge pull request #8873 from vonchenplus/fix_legacy_location_error | bunnei | 2022-10-24 | 3 | -19/+33 |
|\ \ \ | | | | | | | | | video_core: Fix legacy to generic location unpaired | ||||
| * | | | Address feedback | FengChen | 2022-10-17 | 1 | -6/+6 |
| | | | | |||||
| * | | | video_core: Fix legacy to generic location unpaired | FengChen | 2022-09-20 | 3 | -15/+29 |
| | |/ | |/| | |||||
* | | | CMakeLists: Disable C4100 and C4324 | Morph | 2022-10-22 | 2 | -8/+0 |
| | | | | | | | | | | | | Disabling C4100 is similar to -Wno-unused-parameter | ||||
* | | | CMakeLists: Remove redundant warnings | Morph | 2022-10-22 | 1 | -2/+0 |
| | | | | | | | | | | | | These warnings are already included in /W3. | ||||
* | | | CMakeLists: Treat MSVC warnings as errors | Morph | 2022-10-22 | 1 | -1/+0 |
| | | | |||||
* | | | general: Enforce C4800 everywhere except in video_core | Morph | 2022-10-22 | 1 | -1/+0 |
| | | | |||||
* | | | CMakeLists: Remove all redundant warnings | Morph | 2022-10-22 | 1 | -8/+2 |
| | | | | | | | | | | | | These are already explicitly or implicitly set in src/CMakeLists.txt | ||||
* | | | General: Fix compilation for GCC | Liam White | 2022-10-06 | 1 | -1/+1 |
| | | | |||||
* | | | Shader Decompiler: implement better tracking for Vulkan samplers. | Fernando Sahmkow | 2022-10-06 | 1 | -9/+59 |
| | | | |||||
* | | | Shader Decompiler: Check for shift when deriving composite samplers. | Fernando Sahmkow | 2022-10-06 | 2 | -3/+35 |
| | | | |||||
* | | | Shader Decompiler: Fix dangerous behavior of invalid iterator insertion. | Fernando Sahmkow | 2022-10-06 | 1 | -3/+3 |
| | | | |||||
* | | | shader_recompiler: add extended LDC to GLASM backend | Liam | 2022-10-02 | 1 | -4/+21 |
| | | | |||||
* | | | chore: fix some typos | Andrea Pappacoda | 2022-09-23 | 2 | -2/+2 |
|/ / | | | | | | | Fix some typos reported by Lintian | ||||
* / | style: General style changes to match with the rest of the codebase | Morph | 2022-08-31 | 1 | -5/+5 |
|/ | |||||
* | video_code: support rectangle texture | FengChen | 2022-08-25 | 10 | -2/+44 |
| | |||||
* | Add missed shader defines. Fixes Xenoblade Chronicles 3 booting with Vulkan. | Kelebek1 | 2022-07-29 | 1 | -2/+3 |
| | |||||
* | chore: make yuzu REUSE compliant | Andrea Pappacoda | 2022-07-27 | 2 | -4/+5 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | [REUSE] is a specification that aims at making file copyright information consistent, so that it can be both human and machine readable. It basically requires that all files have a header containing copyright and licensing information. When this isn't possible, like when dealing with binary assets, generated files or embedded third-party dependencies, it is permitted to insert copyright information in the `.reuse/dep5` file. Oh, and it also requires that all the licenses used in the project are present in the `LICENSES` folder, that's why the diff is so huge. This can be done automatically with `reuse download --all`. The `reuse` tool also contains a handy subcommand that analyzes the project and tells whether or not the project is (still) compliant, `reuse lint`. Following REUSE has a few advantages over the current approach: - Copyright information is easy to access for users / downstream - Files like `dist/license.md` do not need to exist anymore, as `.reuse/dep5` is used instead - `reuse lint` makes it easy to ensure that copyright information of files like binary assets / images is always accurate and up to date To add copyright information of files that didn't have it I looked up who committed what and when, for each file. As yuzu contributors do not have to sign a CLA or similar I couldn't assume that copyright ownership was of the "yuzu Emulator Project", so I used the name and/or email of the commit author instead. [REUSE]: https://reuse.software Follow-up to 01cf05bc75b1e47beb08937439f3ed9339e7b254 | ||||
* | Merge pull request #8383 from Morph1984/shadow-of-the-past | Mai | 2022-06-15 | 1 | -3/+0 |
|\ | | | | | yuzu: Make variable shadowing a compile-time error | ||||
| * | CMakeLists: Make variable shadowing a compile-time error | Morph | 2022-06-14 | 1 | -3/+0 |
| | | | | | | | | Now that the entire project is free of variable shadowing, we can enforce this as a compile time error to prevent any further introduction of this logic bug. | ||||
* | | general: fix compilation on GCC 12 | Liam | 2022-06-14 | 1 | -1/+1 |
| | | |||||
* | | structured_control_flow: Remove constexpr Flow::Block | lat9nq | 2022-06-14 | 1 | -6/+0 |
|/ | | | | | | This seems to be unsupported in newer libstdc++ versions due to Flow::Block's base class being a non-literal type. It's not clear to me why this was permitted in earlier versions. | ||||
* | general: Avoid ambiguous format_to compilation errors | Lioncash | 2022-05-14 | 2 | -2/+2 |
| | | | | | | | Ensures that we're using the fmt version of format_to. These are also the only three outliers. All of the other formatters we have are properly qualified. | ||||
* | GCC 12 fixes | Liam | 2022-04-28 | 1 | -1/+1 |
| | |||||
* | general: Convert source file copyright comments over to SPDX | Morph | 2022-04-23 | 233 | -699/+466 |
| | | | | | 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. | ||||
* | Merge pull request #8133 from liamwhite/gl-spv-cbuf | Fernando S | 2022-04-07 | 6 | -25/+51 |
|\ | | | | | shader_recompiler: support const buffer indirect addressing on OpenGL | ||||
| * | shader_recompiler: Decrease indirect cbuf limit to match hardware | Liam | 2022-04-04 | 1 | -1/+1 |
| | | |||||
| * | shader_compiler: support const buffer indirect addressing in GLSL | Liam | 2022-04-01 | 4 | -9/+38 |
| | | |||||
| * | shader_recompiler: support const buffer indirect addressing on OpenGL SPIR-V | Liam | 2022-04-01 | 3 | -17/+14 |
| | | |||||
* | | fix: typos | Andrea Pappacoda | 2022-04-02 | 2 | -2/+2 |
|/ | |||||
* | Merge pull request #8095 from bylaws/master | Mai M | 2022-03-27 | 3 | -0/+4 |
|\ | | | | | shader_recompiler: Include <bit> header when std::count{r,l}_zero is used | ||||
| * | Include <bit> header when std::count{r,l}_zero is used | Billy Laws | 2022-03-22 | 3 | -0/+4 |
| | | | | | | | | Needed for compilation with older libc++ releases | ||||
* | | dead_code_elimination_pass: Remove unreachable Phi arguments | ameerj | 2022-03-23 | 3 | -0/+36 |
| | | |||||
* | | shader_recompiler/dead_code_elimination: Add DeadBranchElimination pass | ameerj | 2022-03-22 | 1 | -9/+62 |
|/ | | | | This adds a pass to eliminate if(false) branches within the shader code | ||||
* | Merge pull request #8038 from liamwhite/exit-register-detection | Ameer J | 2022-03-22 | 2 | -0/+9 |
|\ | | | | | shader_recompiler/EXIT: increment output register on failed enable test | ||||
| * | Address review comments | Liam | 2022-03-18 | 1 | -1/+1 |
| | | |||||
| * | shader_recompiler/EXIT: skip render targets with no outputs | Liam | 2022-03-18 | 2 | -0/+8 |
| | | |||||
| * | shader_recompiler/EXIT: increment output register on failed enable test | Liam | 2022-03-18 | 1 | -0/+1 |
| | | |||||
* | | general: Fix clang/gcc build errors | ameerj | 2022-03-20 | 1 | -0/+1 |
| | | |||||
* | | shader_recompiler: Reduce unused includes | ameerj | 2022-03-20 | 69 | -106/+7 |
|/ | |||||
* | Address review comments | Liam | 2022-03-17 | 4 | -52/+36 |
| | |||||
* | shader_recompiler: Use functions for indirect const buffer accesses | Liam | 2022-03-17 | 5 | -39/+94 |
| | |||||
* | Address review comments | Liam | 2022-03-17 | 1 | -16/+15 |
| | |||||
* | shader_recompiler: Implement LDC.IS address mode | Liam | 2022-03-16 | 1 | -2/+12 |
| | |||||
* | shader: add support for const buffer indirect addressing | Liam | 2022-03-15 | 2 | -18/+68 |
| | |||||
* | Merge pull request #8008 from ameerj/rescale-offsets-array | Fernando S | 2022-03-15 | 1 | -2/+27 |
|\ | | | | | rescaling_pass: Fix rescaling Color2DArray ImageFetch offsets | ||||
| * | rescaling_pass: Fix rescaling Color2DArray ImageFetch offsets | ameerj | 2022-03-12 | 1 | -2/+27 |
| | | | | | | | | | | | | ImageFetch offsets for 2D array coordinates have a different composite size than the coordinates. The rescaling pass was not taking this into account. Fixes broken shaders when scaling is enabled in Astral Chain, and likely other titles. | ||||
* | | Shader decompiler: do constant propgation before texture pass. | Fernando Sahmkow | 2022-03-13 | 1 | -2/+2 |
| | | |||||
* | | Shader decompiler: Fix storage tracking in deko3d. | Fernando Sahmkow | 2022-03-13 | 1 | -1/+2 |
| | | |||||
* | | emit_spirv, vk_compute_pass: Resolve VS2022 compiler errors | ameerj | 2022-03-12 | 1 | -1/+1 |
|/ | |||||
* | shader_recompiler/LOP3: Use brute force python results within switch/case. | Markus Wick | 2022-03-08 | 2 | -52/+620 |
| | | | | | | | | | | | | | | | | | | | | | Thanks to @asLody for optimizing this function. This raised the focus that this function should be optimized more. The current table assumes that the host GPU is able to invert for free, so only AND,OR,XOR are accumulated in the performance metrik. Performance results: Instructions 0: 8 1: 30 2: 114 3: 80 4: 24 Latency 0: 8 1: 30 2: 194 3: 24 | ||||
* | emit_glsl_atomic: Implement 32x2 fallback atomic ops | ameerj | 2022-01-30 | 1 | -9/+55 |
| | |||||
* | lower_int64_to_int32: Add 64-bit atomic fallbacks | ameerj | 2022-01-30 | 3 | -11/+76 |
| | |||||
* | shaders: Add U64->U32x2 Atomic fallback functions | ameerj | 2022-01-30 | 9 | -1/+469 |
| | |||||
* | spirv_atomic: Define U32x2 storage buffers for 64-bit storage atomics | ameerj | 2022-01-29 | 2 | -3/+3 |
| | | | | | | 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 | ||||
* | Merge pull request #7786 from ameerj/vmnmx-sel | Morph | 2022-01-29 | 1 | -12/+6 |
|\ | | | | | video_minimum_maximum: Implement src operand selectors | ||||
| * | video_minimum_maximum: Implement src operand selectors | ameerj | 2022-01-27 | 1 | -12/+6 |
| | | | | | | | | Used by Pokemon Legends: Arceus | ||||
* | | emit_spirv: Add Xfb execution mode when transform feedback is used | ameerj | 2022-01-28 | 1 | -3/+9 |
|/ | | | | Fixes Transform Feedback on Vulkan AMD drivers. | ||||
* | shader_recompiler: Remove unnecessary [[nodiscard]] | Lioncash | 2022-01-25 | 1 | -2/+1 |
| | | | | | Since ConvertLegacyToGeneric has a void return value, there's nothing that is actually returned by the function. | ||||
* | shader_recompiler: fix potential OOB access | v1993 | 2022-01-17 | 2 | -6/+8 |
| | | | | Found by static analysis with PVS-Studio. Original check wasn't actually checking for OOB and would segfault in case of it. | ||||
* | logging/log.h: move enum class formatter to a separate file ... | liushuyu | 2022-01-10 | 3 | -7/+7 |
| | | | | ... to common/logging/formatter.h | ||||
* | logging: adapt to changes in fmt 8.1 | liushuyu | 2022-01-08 | 2 | -6/+6 |
| | |||||
* | glsl: Remove unreachable return | Narr the Reg | 2022-01-05 | 1 | -1/+0 |
| | |||||
* | ShaderDecompiler: Add a debug option to dump the game's shaders. | Fernando Sahmkow | 2022-01-04 | 1 | -0/+2 |
| | |||||
* | Merge pull request #7629 from ameerj/nv-driver-fixes | Fernando S | 2022-01-03 | 15 | -23/+125 |
|\ | | | | | shaders: Add fixes for NVIDIA drivers 495+ | ||||
| * | glsl: Add boolean reference workaround | ameerj | 2021-12-30 | 3 | -2/+8 |
| | | |||||
| * | glsl_context_get_set: Add alternative cbuf type for broken drivers | ameerj | 2021-12-30 | 3 | -17/+27 |
| | | | | | | | | some drivers have a bug bitwise converting floating point cbuf values to uint variables. This adds a workaround for these drivers to make all cbufs uint and convert to floating point as needed. | ||||
| * | emit_glsl_integer: Use negation work around | ameerj | 2021-12-30 | 1 | -2/+2 |
| | | |||||
| * | shader: Add integer attribute get optimization pass | ameerj | 2021-12-30 | 9 | -0/+86 |
| | | | | | | | | Works around an nvidia driver bug, where casting the integer attributes to float and back to an integer always returned 0. | ||||
| * | emit_glsl_floating_point: Fix FPNeg on newer Nvidia drivers | ameerj | 2021-12-25 | 1 | -2/+2 |
| | | |||||
* | | Merge pull request #7618 from goldenx86/patch-4 | bunnei | 2021-12-29 | 1 | -0/+9 |
|\ \ | | | | | | | Increase boost requirement to 1.78.0 | ||||
| * | | Empty spaces | Matías Locatti | 2021-12-28 | 1 | -1/+1 |
| | | | |||||
| * | | Changes to avoid warnings in SSE4.2 optimized SPIR-V | Matías Locatti | 2021-12-28 | 1 | -0/+9 |
| |/ | |||||
* / | emit_glasm_context_get_set: Fix GetAttribute return value type. | ameerj | 2021-12-25 | 1 | -4/+4 |
|/ | | | | GetAttribute expects an F32 result type at the IR level, this fixes the return value of attributes which were not returning an F32 | ||||
* | Address format clang | vonchenplus | 2021-12-18 | 1 | -36/+36 |
| | |||||
* | Remove spirv handle legacy related code | vonchenplus | 2021-12-18 | 4 | -190/+1 |
| | |||||
* | Remove glsl handle legacy related code | vonchenplus | 2021-12-18 | 3 | -103/+1 |
| | |||||
* | Merge branch 'yuzu-emu:master' into convert_legacy | Feng Chen | 2021-12-18 | 66 | -214/+286 |
|\ | |||||
| * | Merge pull request #7522 from ameerj/shader-recompiler-filenames | Mai M | 2021-12-08 | 65 | -214/+282 |
| |\ | | | | | | | shader_recompiler/backend: Minor organization and refactoring to reduce compile time overhead | ||||
| | * | emit_spirv: Reduce emit_spirv.h include overhead | ameerj | 2021-12-06 | 20 | -3/+20 |
| | | | | | | | | | | | | emit_spirv.h is included in video_core, which was propagating further includes that video_core did not depend on. | ||||
| | * | glasm: Move implemented instructions from not_implemented.cpp | ameerj | 2021-12-06 | 7 | -169/+220 |
| | | | |||||
| | * | shader_recompiler: Adjust emit_context includes | ameerj | 2021-12-06 | 37 | -37/+37 |
| | | | |||||
| | * | shader_recompiler: Rename backend emit_context files | ameerj | 2021-12-05 | 7 | -6/+6 |
| | | | |||||
| * | | general: Add missing copyright notices | ameerj | 2021-12-05 | 1 | -0/+4 |
| |/ | |||||
* / | Implement convert legacy to generic | Feng Chen | 2021-11-19 | 4 | -1/+103 |
|/ | |||||
* | ShaderCache: Better fix for Shuffling gl_FragCoord | Fernando Sahmkow | 2021-11-16 | 1 | -2/+13 |
| | |||||
* | Texture Cahe/Shader decompiler: Resize PointSize on rescaling, refactor and make reaper more agressive on 4Gb GPUs. | FernandoS27 | 2021-11-16 | 1 | -0/+21 |
| | |||||
* | vulkan: Fix rescaling push constant usage | ameerj | 2021-11-16 | 4 | -34/+36 |
| | |||||
* | rescaling_pass: Fix IR errors when unscalable texture types are encountered | ameerj | 2021-11-16 | 1 | -0/+28 |
| | |||||
* | rescaling_pass: Logic simplification and minor style cleanup | ameerj | 2021-11-16 | 2 | -33/+17 |
| | |||||
* | rescaling_pass: Scale ImageFetch offset if it exists | ameerj | 2021-11-16 | 1 | -59/+37 |
| | | | | Plus some code deduplication | ||||
* | rescaling_pass: Enable PatchImageQueryDimensions on fragment stages | ameerj | 2021-11-16 | 1 | -5/+4 |
| | |||||
* | gl_texture_cache/rescaling_pass: minor cleanup | ameerj | 2021-11-16 | 1 | -12/+8 |
| | |||||
* | rescaling_pass: Fix and simplify shuffle/fragcoord pass | ameerj | 2021-11-16 | 1 | -26/+20 |
| | |||||
* | Shader: Don't rescale FragCoord if used by Shuffle | Fernando Sahmkow | 2021-11-16 | 2 | -2/+55 |
| | |||||
* | shader, video_core: Fix GCC build errors | ameerj | 2021-11-16 | 1 | -4/+0 |
| | |||||
* | emit_spirv: Fix RescalingLayout alignment | ameerj | 2021-11-16 | 1 | -0/+1 |
| | |||||
* | RescalingPass: Agregate pixels on texelFetch while on Fragment Shader | Fernando Sahmkow | 2021-11-16 | 1 | -3/+97 |
| | |||||
* | shader: Fix TextureSize check on rescaling. | Fernando Sahmkow | 2021-11-16 | 1 | -27/+21 |
| | |||||
* | emit_spirv: Fix RescalingLayout alignment | ameerj | 2021-11-16 | 1 | -2/+2 |
| | |||||
* | shader: Properly scale image reads and add GL SPIR-V support | ReinUsesLisp | 2021-11-16 | 20 | -51/+171 |
| | | | | Thanks for everything! | ||||
* | shader: Properly blacklist and scale image loads | ReinUsesLisp | 2021-11-16 | 1 | -3/+19 |
| | |||||
* | glsl/glasm: Pass and use scaling parameters in shaders | ReinUsesLisp | 2021-11-16 | 6 | -7/+11 |
| | |||||
* | gl_graphics_pipeline: Add downscale factor to shader uniforms | ameerj | 2021-11-16 | 3 | -4/+5 |
| | |||||
* | spirv: Implement rescaling patching | ReinUsesLisp | 2021-11-16 | 8 | -5/+86 |
| | |||||
* | shader/rescaling_pass: Patch more instructions | ReinUsesLisp | 2021-11-16 | 1 | -4/+101 |
| | |||||
* | shader: Add IsTextureScaled opcode | ReinUsesLisp | 2021-11-16 | 10 | -0/+34 |
| | |||||
* | shader: Add copy constructor to instructions | ReinUsesLisp | 2021-11-16 | 4 | -1/+20 |
| | |||||
* | shader: Add integer division opcodes | ReinUsesLisp | 2021-11-16 | 9 | -0/+37 |
| | |||||
* | shader: Fix rescaling pass | ReinUsesLisp | 2021-11-16 | 1 | -1/+1 |
| | |||||
* | shader: Fix resolution scaling pass | ReinUsesLisp | 2021-11-16 | 5 | -35/+32 |
| | |||||
* | shader: Add resolution down factor opcode | ReinUsesLisp | 2021-11-16 | 9 | -0/+25 |
| | |||||
* | ShaderDecompiler: Add initial support for rescaling. | Fernando Sahmkow | 2021-11-16 | 2 | -0/+73 |
| | |||||
* | Merge pull request #7260 from vonchenplus/spirv_support_legacy_attribute_v2 | bunnei | 2021-11-14 | 3 | -71/+153 |
|\ | | | | | shader: Spirv support legacy attribute v2 | ||||
| * | Simply legacy attribute implement | Feng Chen | 2021-11-04 | 3 | -152/+125 |
| | | |||||
| * | Support gl_FogFragCoord attribute | vonchenplus | 2021-10-31 | 3 | -48/+58 |
| | | |||||
| * | Support gl_BackSecondaryColor attribute | vonchenplus | 2021-10-26 | 3 | -0/+33 |
| | | |||||
| * | Support gl_FrontSecondaryColor attribute | vonchenplus | 2021-10-26 | 3 | -0/+33 |
| | | |||||
| * | Support gl_BackColor attribute | vonchenplus | 2021-10-26 | 3 | -0/+33 |
| | | |||||
* | | Merge pull request #7262 from FernandoS27/Buffalo-buffalo-Buffalo-buffalo-buffalo | bunnei | 2021-11-03 | 7 | -3/+68 |
|\ \ | | | | | | | ShaderCache: Order Phi Arguments from farthest away to nearest. | ||||
| * | | Shader Cahe: Fix Phi Nodes on GLASM. | Fernando Sahmkow | 2021-11-02 | 1 | -1/+1 |
| | | | |||||
| * | | ShaderCache: Fix Phi Nodes Type on OGL. | Fernando Sahmkow | 2021-11-01 | 3 | -2/+30 |
| | | | |||||
| * | | ShaderCache: Order Phi Arguments from farthest away to nearest. | Fernando Sahmkow | 2021-10-31 | 5 | -0/+37 |
| |/ | |||||
* | | Merge pull request #7201 from ameerj/spirv-depth-sampling | Fernando S | 2021-10-30 | 1 | -5/+16 |
|\ \ | |/ |/| | emit_spirv_image: Fix depth image implicit lod sample in non-fragment stages | ||||
| * | emit_spirv_image: Fix depth image implicit lod sample in compute | ameerj | 2021-10-17 | 1 | -5/+16 |
| | | | | | | | | Ensures all drivers behave the same way in this case. | ||||
* | | TexturePass: Fix clamping of images as this allowed negative indices. | Fernando Sahmkow | 2021-10-24 | 1 | -1/+1 |
|/ | |||||
* | Merge pull request #7077 from FernandoS27/face-down | Ameer J | 2021-10-17 | 1 | -1/+2 |
|\ | | | | | A series of fixes to queries and indexed samplers. | ||||
| * | Shader Compiler: avoid overflowed indices on indixed samplers. | Fernando Sahmkow | 2021-10-17 | 1 | -1/+2 |
| | | |||||
* | | style: Remove extra space preceding the :: operator | Morph | 2021-09-29 | 1 | -2/+2 |
| | | |||||
* | | general: Update style to clang-format-12 | ameerj | 2021-09-24 | 1 | -2/+4 |
|/ | |||||
* | Spir-V: Rescale the frag depth to 0,1 mode when -1,1 mode is used in Vulkan. | Fernando Sahmkow | 2021-09-15 | 1 | -1/+7 |
| | |||||
* | Merge pull request #6948 from ameerj/amd-warp-fix | Morph | 2021-09-12 | 2 | -54/+109 |
|\ | | | | | shaders: Fix warp instructions on 64-thread warp devices | ||||
| * | emit_glsl_warp: Fix shuffle ops for 64-thread warp sizes | ameerj | 2021-08-31 | 1 | -24/+36 |
| | | |||||
| * | emit_glsl_warp: Fix ballot related ops for 64-thread warp sizes | ameerj | 2021-08-31 | 1 | -24/+38 |
| | | |||||
| * | emit_spirv_warp: Fix shuffle ops for 64-thread warp sizes | ameerj | 2021-08-31 | 1 | -1/+29 |
| | | |||||
| * | emit_spirv_warp: Fix ballot related ops for 64-thread warp sizes | ameerj | 2021-08-31 | 1 | -10/+11 |
| | | |||||
* | | Merge pull request #6962 from vonchenplus/spirv_support_legacy_attribute | bunnei | 2021-09-08 | 3 | -0/+107 |
|\ \ | | | | | | | renderer_vulkan: Spirv support glsl legacy attribute | ||||
| * | | Detail adjustment | Feng Chen | 2021-09-08 | 1 | -13/+14 |
| | | | |||||
| * | | Detail adjustment | Feng Chen | 2021-09-08 | 2 | -28/+35 |
| | | | |||||
| * | | Re-implement get unused location | Feng Chen | 2021-09-07 | 1 | -30/+30 |
| | | | |||||
| * | | Move attribute related definitions to spirv anonymous namespace | Feng Chen | 2021-09-07 | 4 | -30/+26 |
| | | | |||||
| * | | Dynamic get unused location | Feng Chen | 2021-09-06 | 1 | -27/+49 |
| | | | |||||
| * | | Implement intput and output fixed fnc textures | Feng Chen | 2021-09-06 | 4 | -19/+25 |
| | | | |||||
| * | | Rename parameters | Feng Chen | 2021-09-03 | 5 | -14/+24 |
| | | | |||||
| * | | Fix create GraphicsPipelines crash | Feng Chen | 2021-09-03 | 1 | -5/+5 |
| | | | |||||
| * | | Add input/output location | Feng Chen | 2021-09-02 | 1 | -5/+13 |
| | | | |||||
| * | | Add colorfront and txtcoord support | Feng Chen | 2021-08-31 | 5 | -0/+57 |
| | | | |||||
* | | | Merge pull request #6900 from ameerj/attr-reorder | bunnei | 2021-09-02 | 4 | -10/+133 |
|\ \ \ | |_|/ |/| | | structured_control_flow: Add DemoteCombinationPass | ||||
| * | | structured_control_flow: Skip reordering nested demote branches. | ameerj | 2021-08-30 | 1 | -0/+11 |
| | | | | | | | | | | | | Nested demote branches add complexity with combining the condition if it has not been initialized yet. Skip them for the time being. | ||||
| * | | structured_control_flow: Conditionally invoke demote reorder pass | ameerj | 2021-08-30 | 4 | -10/+16 |
| | | | | | | | | | | | | This is only needed on select drivers when a fragment shader discards/demotes. | ||||
| * | | structured_control_flow: Add DemoteCombinationPass | ameerj | 2021-08-28 | 1 | -1/+107 |
| |/ | | | | | | | | | Some drivers misread data when demotes are interleaved in the program. This moves demote branches to be checked at the end of the program. Fixes "wireframe" issue in Pokemon SwSh on some drivers | ||||
* / | emit_spirv_context_get_set: Fix Get FrontFace return value | ameerj | 2021-08-27 | 1 | -2/+3 |
|/ | | | | The IR expects GetAttribute to return an F32 value. This case was returning a U32 instead. | ||||
* | SPIR-V: Merge two ifs in EmitGetAttribute | Valeri | 2021-08-19 | 1 | -6/+2 |
| | |||||
* | Merge pull request #6767 from ReinUsesLisp/fold-float-pack | Morph | 2021-07-30 | 1 | -0/+4 |
|\ | | | | | shader: Fold UnpackFloat2x16 and PackFloat2x16 | ||||
| * | shader: Fold UnpackFloat2x16 and PackFloat2x16 | ReinUsesLisp | 2021-07-30 | 1 | -0/+4 |
| | | | | | | | | | | Simplifies the code a bit when possible. These instructions should be no-ops codegen wise. | ||||
* | | Merge pull request #6722 from ReinUsesLisp/xmad-opts | bunnei | 2021-07-30 | 2 | -14/+195 |
|\ \ | |/ |/| | shader: Fold integer FMA from Nvidia's pattern | ||||
| * | shader: Fold integer FMA from Nvidia's pattern | ReinUsesLisp | 2021-07-26 | 1 | -0/+175 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fold shaders doing "a * b + c" on integers from the pattern generated by Nvidia's GL compiler. On a somewhat complex compute shader it reduces the code size by 16 instructions from 2 matches on Turing GPUs. On Intel as extracted from KHR_pipeline_executable_properties: Before the optimization: ``` Instruction Count: 2057 Basic Block Count: 45 Scratch Memory Size: 14752 Spill Count: 232 Fill Count: 261 SEND Count: 610 Cycle Count: 11325 ``` After the optimization: ``` Instruction Count: 2046 Basic Block Count: 44 Scratch Memory Size: 13728 Spill Count: 219 Fill Count: 268 SEND Count: 604 Cycle Count: 11367 ``` | ||||
| * | shader: Use TryInstRecursive on XMAD multiply folding | ReinUsesLisp | 2021-07-26 | 1 | -14/+12 |
| | | | | | | | | Simplify a bit the logic. | ||||
| * | shader: Add TryInstRecursive utility to values | ReinUsesLisp | 2021-07-26 | 1 | -0/+8 |
| | | |||||
* | | shader: Mark ConvertF16F32 and ConvertF32F16 as fp16 instructions | ReinUsesLisp | 2021-07-28 | 1 | -0/+2 |
| | | | | | | | | | | | | Fixes instances where fp16 types are not declared on SPIR-V but they are used. This shouldn't happen on master, as it's been uncovered by an additional optimization pass. | ||||
* | | exception: Make constructors explicit | Lioncash | 2021-07-27 | 1 | -4/+4 |
| | | | | | | | | Ensures that exception construction is always explicit. | ||||
* | | exception: Make what() member function nodiscard | Lioncash | 2021-07-27 | 1 | -1/+1 |
| | | |||||
* | | exception: Narrow down specific header | Lioncash | 2021-07-27 | 1 | -1/+1 |
| | | | | | | | | | | We can use the <exception> header instead of pulling in all of the exception-style classes. | ||||
* | | Merge pull request #6724 from lioncash/nodisc-shader | Rodrigo Locatti | 2021-07-26 | 2 | -4/+4 |
|\ \ | | | | | | | shader_recompiler: Remove unnecessary [[nodiscard]] instances | ||||
| * | | shader_recompiler: Remove unnecessary [[nodiscard]] instances | Lioncash | 2021-07-26 | 2 | -4/+4 |
| |/ | | | | | | | | | [[nodiscard]] doesn't do anything on functions with a void return type and causes superfluous warnings. | ||||
* | | Merge pull request #6726 from lioncash/hguard | Rodrigo Locatti | 2021-07-26 | 1 | -0/+2 |
|\ \ | | | | | | | emit_spirv_instructions: Add missing header guard | ||||
| * | | emit_spirv_instructions: Add missing header guard | Lioncash | 2021-07-26 | 1 | -0/+2 |
| |/ | |||||
* | | Merge pull request #6727 from lioncash/topology | Rodrigo Locatti | 2021-07-26 | 1 | -1/+1 |
|\ \ | | | | | | | emit_glasm: Fix LINESS_ADJACENCY typo in InputPrimitive() | ||||
| * | | emit_glasm: Fix LINESS_ADJACENCY typo in InputPrimitive() | Lioncash | 2021-07-26 | 1 | -1/+1 |
| |/ | | | | | | | This should be LINES_ADJACENCY | ||||
* | | Merge pull request #6723 from lioncash/shader | Rodrigo Locatti | 2021-07-26 | 1 | -0/+1 |
|\ \ | | | | | | | object_pool: Add missing return in Chunk move assignment operator | ||||
| * | | object_pool: Add missing return in Chunk move assignment operator | Lioncash | 2021-07-26 | 1 | -0/+1 |
| |/ | | | | | | | Prevents undefined behavior from occurring. | ||||
* / | control_flow: Fix duplicate switch case in OpcodeToken | Lioncash | 2021-07-26 | 1 | -1/+1 |
|/ | | | | This previously duplicated the case of the PBK case above it. | ||||
* | shader: Support out of bound local memory reads and immediate writes | ReinUsesLisp | 2021-07-23 | 1 | -4/+21 |
| | | | | | | | | | Support ignoring immediate out of bound writes. Writing dynamically out of bounds is not yet supported (e.g. R0+0x4). Reading out of bounds yields zero. This is supported checking for the size from the IR; if the input is immediate, the optimization passes will drop it. | ||||
* | shader: Fix disabled attribute default values | ameerj | 2021-07-23 | 1 | -1/+1 |
| | |||||
* | glsl: Simplify FCMP emission | ameerj | 2021-07-23 | 1 | -6/+4 |
| | |||||
* | glsl: Update TessellationControl gl_in | ameerj | 2021-07-23 | 1 | -0/+28 |
| | | | | Adheres to GL_ARB_separate_shader_objects requirements | ||||
* | shader: Implement ISETP.X | ameerj | 2021-07-23 | 4 | -44/+57 |
| | |||||
* | shader: Avoid usage of C++20 ranges to build in clang | ReinUsesLisp | 2021-07-23 | 11 | -39/+47 |
| | |||||
* | glsl: Clamp shared mem size to GL_MAX_COMPUTE_SHARED_MEMORY_SIZE | ameerj | 2021-07-23 | 2 | -2/+11 |
| | |||||
* | shader_recompiler, video_core: Resolve clang errors | lat9nq | 2021-07-23 | 13 | -41/+39 |
| | | | | | | | | | | Silences the following warnings-turned-errors: -Wsign-conversion -Wunused-private-field -Wbraced-scalar-init -Wunused-variable And some other errors | ||||
* | shader: Manually convert from array<u32> to bitset instead of using bit_cast | ReinUsesLisp | 2021-07-23 | 1 | -2/+3 |
| | |||||
* | glsl: Fix tracking of info.uses_shadow_lod | ameerj | 2021-07-23 | 1 | -4/+4 |
| | |||||
* | shader: Ignore global memory ops on devices lacking int64 support | ameerj | 2021-07-23 | 6 | -30/+77 |
| | |||||
* | dual_vertex_pass: Clang format | ameerj | 2021-07-23 | 1 | -14/+14 |
| | |||||
* | emit_spirv: Workaround VK_KHR_shader_float_controls on fp16 Nvidia | ReinUsesLisp | 2021-07-23 | 2 | -5/+10 |
| | | | | Fix regression on Fire Emblem: Three Houses when using native fp16. | ||||
* | shader: GCC fmt 8.0.0 fixes | lat9nq | 2021-07-23 | 7 | -16/+19 |
| | |||||
* | shader: Account for 33-bit IADD3 scenario | ameerj | 2021-07-23 | 1 | -2/+10 |
| | |||||
* | shader: Only apply shift on register mode for IADD3 | ReinUsesLisp | 2021-07-23 | 1 | -10/+14 |
| | |||||
* | shader: Fix disabled and unwritten attributes and varyings | ReinUsesLisp | 2021-07-23 | 2 | -3/+11 |
| | |||||
* | glsl: Fix shared and local memory declarations | ameerj | 2021-07-23 | 1 | -3/+3 |
| | | | | account for the fact that program.*memory_size is in units of bytes. | ||||
* | opengl: Implement LOP.CC | ameerj | 2021-07-23 | 2 | -6/+38 |
| | | | | Used by MH:Rise | ||||
* | spirv: Fix code emission when descriptor aliasing is unsupported | ReinUsesLisp | 2021-07-23 | 1 | -1/+2 |
| | | | | Fixes OpenGL. | ||||
* | glsl: Declare local memory in main | ameerj | 2021-07-23 | 1 | -3/+3 |
| | |||||
* | glsl: Add passthrough geometry shader support | ameerj | 2021-07-23 | 3 | -7/+27 |
| | |||||
* | shader: Use std::bit_cast instead of Common::BitCast for passthrough | ReinUsesLisp | 2021-07-23 | 1 | -2/+3 |
| | |||||
* | glasm: Add passthrough geometry shader support | ReinUsesLisp | 2021-07-23 | 2 | -7/+26 |
| | |||||
* | shader: Rework varyings and implement passthrough geometry shaders | ReinUsesLisp | 2021-07-23 | 22 | -316/+302 |
| | | | | | | Put all varyings into a single std::bitset with helpers to access it. Implement passthrough geometry shaders using host's. | ||||
* | shader: Only verify shader when graphics debugging is enabled | ReinUsesLisp | 2021-07-23 | 1 | -2/+7 |
| | |||||
* | shader: Unify shader stage types | ReinUsesLisp | 2021-07-23 | 1 | -2/+9 |
| | |||||
* | lower_int64_to_int32: Add missing include | lat9nq | 2021-07-23 | 1 | -0/+1 |
| | |||||
* | shader: Emulate 64-bit integers when not supported | ReinUsesLisp | 2021-07-23 | 1 | -0/+3 |
| | | | | Useful for mobile and Intel Xe devices. | ||||
* | shader: Add int64 to int32 lowering pass | ReinUsesLisp | 2021-07-23 | 3 | -0/+218 |
| | |||||
* | shader: Teach global memory base tracker to follow vectors | ReinUsesLisp | 2021-07-23 | 1 | -15/+14 |
| | |||||
* | shader: Add constant propagation to integer vectors | ReinUsesLisp | 2021-07-23 | 1 | -0/+9 |
| | |||||
* | glsl: Better IAdd Overflow CC fix | ameerj | 2021-07-23 | 2 | -11/+13 |
| | | | | This ensures the original operand values are not overwritten when being used in the overflow detection. | ||||
* | shader: Remove IAbs64 | ReinUsesLisp | 2021-07-23 | 9 | -26/+3 |
| | |||||
* | glsl: Fix IADD CC | ameerj | 2021-07-23 | 2 | -5/+7 |
| | |||||
* | shader_recompiler: Fix IADD3 input partitioning | ameerj | 2021-07-23 | 1 | -14/+13 |
| | |||||
* | shader: Move loop safety tests to code emission | ReinUsesLisp | 2021-07-23 | 16 | -108/+54 |
| | |||||
* | glsl: Remove frag color initialization | ameerj | 2021-07-23 | 1 | -9/+0 |
| | |||||
* | glasm: Implement SetAttribute ViewportMask | ameerj | 2021-07-23 | 2 | -1/+10 |
| | |||||
* | emit_glsl_special: Skip initialization of frag_color0 | ameerj | 2021-07-23 | 1 | -1/+1 |
| | | | | Fixes rendering in Devil May Cry without regressing Ori and the Blind Forest. | ||||
* | shader: Calibrate loop safety threshold | ReinUsesLisp | 2021-07-23 | 1 | -1/+1 |
| | |||||
* | glsl: Add missing ; in EmitSetSampleMask | Morph | 2021-07-23 | 1 | -1/+1 |
| | | | | Fixes shader compilation in Okami HD | ||||
* | glsl: Fix output varying initialization when transform feedback is used | ameerj | 2021-07-23 | 1 | -3/+37 |
| | |||||
* | texture_pass: Fix is_read image qualification | ameerj | 2021-07-23 | 1 | -1/+1 |
| | | | | Atomic operations are considered to have both read and write access. This was not being accounted for. | ||||
* | shader: Align constant buffer sizes to 16 bytes | ReinUsesLisp | 2021-07-23 | 1 | -1/+2 |
| | | | | WAR for AMD reading zeroes on uniform buffers of size 2. | ||||
* | spirv: Properly handle devices without int8 and int16 | ReinUsesLisp | 2021-07-23 | 2 | -39/+67 |
| | |||||
* | spirv: Handle small storage buffer loads on devices with no support | ReinUsesLisp | 2021-07-23 | 2 | -6/+6 |
| | |||||
* | glsl: Fix cbuf component indexing bug falback | ameerj | 2021-07-23 | 1 | -7/+6 |
| | |||||
* | shader: Simplify MergeDualVertexPrograms | ReinUsesLisp | 2021-07-23 | 1 | -6/+4 |
| | |||||
* | shader: Properly manage attributes not written from previous stages | ReinUsesLisp | 2021-07-23 | 10 | -36/+40 |
| | |||||
* | glsl: Only declare fragment outputs on fragment shaders | ReinUsesLisp | 2021-07-23 | 1 | -4/+6 |
| | |||||
* | shader: Split profile and runtime info headers | ReinUsesLisp | 2021-07-23 | 11 | -76/+91 |
| | |||||
* | shader: Add support for native 16-bit floats | ReinUsesLisp | 2021-07-23 | 4 | -4/+26 |
| | |||||
* | shader: Rename maxwell/program.h to translate_program.h | ReinUsesLisp | 2021-07-23 | 3 | -9/+4 |
| | |||||
* | glsl: Obey need_declared_frag_colors to declare and initialize all frag_color | ameerj | 2021-07-23 | 2 | -1/+10 |
| | | | | Fixes Ori and the blind forest title screen | ||||
* | glsl: Address rest of feedback | ameerj | 2021-07-23 | 7 | -21/+64 |
| | |||||
* | glsl: Move gl_Position/generic attribute initialization to EmitProlgue | ameerj | 2021-07-23 | 2 | -14/+12 |
| | |||||
* | glsl: Conditionally use fine/coarse derivatives based on device support | ameerj | 2021-07-23 | 3 | -4/+28 |
| | |||||
* | glsl: Cleanup/Address feedback | ameerj | 2021-07-23 | 9 | -28/+22 |
| | |||||
* | glsl: Add Shader_GLSL logging | ameerj | 2021-07-23 | 3 | -28/+32 |
| | |||||
* | glsl: Add LoopSafety instructions | ameerj | 2021-07-23 | 2 | -0/+10 |
| | |||||
* | glsl: Conditionally add EXT_texture_shadow_lod | ameerj | 2021-07-23 | 3 | -4/+15 |
| | |||||
* | glsl: Add stubs for sparse queries and variable aoffi when not supported | ameerj | 2021-07-23 | 4 | -13/+39 |
| | |||||
* | glsl: Implement legacy varyings | ameerj | 2021-07-23 | 6 | -8/+81 |
| | |||||
* | glsl: Minor cleanup | ameerj | 2021-07-23 | 2 | -19/+15 |
| | |||||
* | glsl: Fix Cbuf getters for F32 type | ameerj | 2021-07-23 | 1 | -12/+15 |
| | |||||
* | glsl: Add immediate index oob checking for Cbuf getters | ameerj | 2021-07-23 | 1 | -0/+16 |
| | |||||
* | glsl: Refactor GetCbuf functions to reduce code duplication | ameerj | 2021-07-23 | 1 | -104/+66 |
| | |||||
* | glsl: Address more feedback. Implement indexed texture reads | ameerj | 2021-07-23 | 5 | -111/+109 |
| | |||||
* | glsl: Remove Signed Integer variables | ameerj | 2021-07-23 | 8 | -43/+13 |
| | |||||
* | glsl: Address Rodrigo's feedback | ameerj | 2021-07-23 | 13 | -75/+87 |
| | |||||
* | glsl: Reorganize backend code, remove unneeded [[maybe_unused]] | ameerj | 2021-07-23 | 12 | -315/+251 |
| | |||||
* | glsl: Implement SampleId and SetSampleMask | ameerj | 2021-07-23 | 3 | -30/+35 |
| | | | | plus some minor refactoring of implementations | ||||
* | glsl: Add gl_PerVertex in for GS | ameerj | 2021-07-23 | 1 | -1/+2 |
| | |||||
* | glsl: Use existing tracking for enabling EXT_shader_image_load_formatted | ameerj | 2021-07-23 | 1 | -15/+1 |
| | |||||
* | glsl: Enable early fragment tests | ameerj | 2021-07-23 | 2 | -4/+7 |
| | |||||
* | glsl: Implement more attribute getters and setters | ameerj | 2021-07-23 | 2 | -12/+60 |
| | |||||
* | glsl: Implement fswzadd | ameerj | 2021-07-23 | 4 | -5/+44 |
| | | | | and wip nv thread shuffle impl | ||||
* | glsl: Implement indexed attribute loads | ameerj | 2021-07-23 | 5 | -29/+64 |
| | |||||
* | glsl: Conditionally add GL_ARB_sparse_texture2 | ameerj | 2021-07-23 | 1 | -2/+3 |
| | |||||
* | glsl: Conditionally use GL_EXT_shader_image_load_formatted | ameerj | 2021-07-23 | 1 | -2/+18 |
| | | | | Fix for SULD.D | ||||
* | glsl: Remove output generic indexing for geometry stage | ameerj | 2021-07-23 | 1 | -5/+3 |
| | |||||
* | glsl: Allow dynamic tracking of variable allocation | ameerj | 2021-07-23 | 3 | -21/+35 |
| | |||||
* | glsl: Implement barriers | ameerj | 2021-07-23 | 3 | -13/+21 |
| | |||||
* | glsl: Implement image atomics and set layer | ameerj | 2021-07-23 | 5 | -153/+202 |
| | | | | along with some more cleanup/oversight fixes | ||||
* | glsl: Fix image gather logic | ameerj | 2021-07-23 | 1 | -0/+4 |
| | |||||
* | glsl: Add cbuf access workaround for devices with component indexing bug | ameerj | 2021-07-23 | 2 | -51/+112 |
| | |||||
* | glsl: Use textureGrad fallback when EXT_texture_shadow_lod is unsupported | ameerj | 2021-07-23 | 3 | -8/+41 |
| | |||||
* | emit_glsl_image: Use immediate offsets when possible | ameerj | 2021-07-23 | 1 | -12/+33 |
| | |||||
* | glsl: Fix <32-bit SSBO writes | ameerj | 2021-07-23 | 4 | -50/+43 |
| | | | | and more cleanup | ||||
* | glsl: Cleanup and address feedback | ameerj | 2021-07-23 | 10 | -86/+69 |
| | |||||
* | glsl: Refactor Global memory functions | ameerj | 2021-07-23 | 2 | -71/+73 |
| | |||||
* | glsl: Increase NUM_VARS that can be allocated | ameerj | 2021-07-23 | 1 | -1/+1 |
| | | | | needed for HW:AoC. | ||||
* | glsl: Implement Load/WriteGlobal | ameerj | 2021-07-23 | 9 | -98/+185 |
| | | | | along with some other misc changes and fixes | ||||
* | glsl: Implement Images | ameerj | 2021-07-23 | 2 | -9/+74 |
| | |||||
* | glsl: skip gl_ViewportIndex write if device does not support it | ameerj | 2021-07-23 | 4 | -8/+17 |
| | |||||
* | glsl: Implement transform feedback | ameerj | 2021-07-23 | 3 | -13/+63 |
| | |||||
* | glsl: Yet another gl_ViewportIndex fix attempt | ameerj | 2021-07-23 | 1 | -3/+19 |
| | |||||
* | glsl: Add gl_ViewportIndex out attribute | ameerj | 2021-07-23 | 1 | -1/+3 |
| | |||||
* | emit_glsl_context_get_set: Remove unused function | lat9nq | 2021-07-23 | 1 | -4/+0 |
| | |||||
* | glsl: Fix precise variable declaration | ameerj | 2021-07-23 | 3 | -24/+25 |
| | | | | and add some more separation in the shader for better debugability when dumped | ||||
* | glsl: Implement tessellation shaders | ameerj | 2021-07-23 | 5 | -27/+146 |
| | |||||
* | glsl: Implement ImageGradient and other texture function variants | ameerj | 2021-07-23 | 2 | -32/+73 |
| | |||||
* | glsl: Fix atomic SSBO offsets | ameerj | 2021-07-23 | 4 | -67/+74 |
| | | | | and implement misc getters | ||||
* | glsl: Implement geometry shaders | ameerj | 2021-07-23 | 4 | -9/+62 |
| | |||||
* | glsl: Use NotImplemented macro with function name output | ameerj | 2021-07-23 | 10 | -104/+103 |
| | |||||
* | glsl: Implement gl_ViewportIndex | ameerj | 2021-07-23 | 3 | -5/+14 |
| | | | | SSBU now working | ||||
* | glsl: SHFL fix and prefer shift operations over divide in glsl shader | ameerj | 2021-07-23 | 5 | -63/+64 |
| | |||||
* | glsl: Implement precise fp variable allocation | ameerj | 2021-07-23 | 4 | -8/+67 |
| | |||||
* | HACK glsl: Write defaults to unused generic attributes | ameerj | 2021-07-23 | 2 | -2/+11 |
| | |||||
* | glsl: Fix ssbo indexing and name shadowing between shader stages | ameerj | 2021-07-23 | 3 | -77/+101 |
| | |||||
* | glsl: implement set clip distance | ameerj | 2021-07-23 | 2 | -0/+15 |
| | | | | and missed a diff in emit_glsl relating to var alloc ref counting | ||||
* | glsl: Rework var alloc to not assign unused results | ameerj | 2021-07-23 | 9 | -49/+91 |
| | |||||
* | glsl: Rework variable allocator to allow for variable reuse | ameerj | 2021-07-23 | 14 | -353/+482 |
| | |||||
* | glsl: Fix ATOM and implement ATOMS | ameerj | 2021-07-23 | 5 | -114/+136 |
| | |||||
* | glsl: Use gl_SubGroupInvocationARB | ameerj | 2021-07-23 | 2 | -8/+7 |
| | |||||
* | glsl: Implement VOTE for subgroup size potentially larger | ameerj | 2021-07-23 | 2 | -19/+36 |
| | |||||
* | glsl: Implement VOTE | ameerj | 2021-07-23 | 4 | -50/+64 |
| | |||||
* | glsl: Implement ST{LS} | ameerj | 2021-07-23 | 6 | -69/+106 |
| | |||||
* | glsl: Implement more instructions used by SMO | ameerj | 2021-07-23 | 1 | -3/+3 |
| | |||||
* | glsl: Implement more instructions used by SMO | ameerj | 2021-07-23 | 5 | -10/+16 |
| | |||||
* | glsl: Fix GetAttribute return values | ameerj | 2021-07-23 | 2 | -7/+9 |
| | | | | fixes font rendering issues as these were used to index into the ssbos | ||||
* | glsl: minor cleanup | ameerj | 2021-07-23 | 4 | -20/+19 |
| | |||||
* | glsl: Fix and implement rest of cbuf access | ameerj | 2021-07-23 | 1 | -7/+43 |
| | |||||
* | glsl: Implement TXQ and other misc changes | ameerj | 2021-07-23 | 5 | -6/+36 |
| | |||||
* | glsl: TLD4 implementation | ameerj | 2021-07-23 | 1 | -2/+89 |
| | |||||
* | glsl: Implement TLD instruction | ameerj | 2021-07-23 | 1 | -1/+55 |
| | |||||
* | glsl: Implement TEXS | ameerj | 2021-07-23 | 1 | -1/+29 |
| | |||||
* | glsl: Cleanup texture functions | ameerj | 2021-07-23 | 1 | -13/+11 |
| | |||||
* | shader_recompiler: GCC fixes | lat9nq | 2021-07-23 | 14 | -3/+13 |
| | |||||
* | glsl: Implement TEX depth functions | ameerj | 2021-07-23 | 2 | -4/+46 |
| | |||||
* | glsl: Implement TEX ImageSample functions | ameerj | 2021-07-23 | 3 | -11/+71 |
| | |||||
* | glsl: Rework Shuffle emit instructions to align with SPIR-V | ameerj | 2021-07-23 | 1 | -19/+40 |
| | |||||
* | glsl: Better Storage access and wip warps | ameerj | 2021-07-23 | 8 | -62/+133 |
| | |||||
* | glsl: Fix integer conversions, implement clamp CC | ameerj | 2021-07-23 | 2 | -27/+36 |
| | |||||
* | glsl: Implement IADD CC | ameerj | 2021-07-23 | 2 | -2/+17 |
| | |||||
* | glsl: SSBO access fixes and wip SampleExplicitLod implementation. | ameerj | 2021-07-23 | 2 | -4/+19 |
| | |||||
* | glsl: WIP var forward declaration | ameerj | 2021-07-23 | 6 | -49/+60 |
| | | | | to fix Loop control flow. | ||||
* | glsl: Fix bindings, add some CC ops | ameerj | 2021-07-23 | 8 | -57/+91 |
| | |||||
* | glsl: remove unused headers | ameerj | 2021-07-23 | 14 | -34/+10 |
| | |||||
* | glsl: Implement derivatives and YDirection | ameerj | 2021-07-23 | 8 | -81/+87 |
| | | | | plus some other misc additions/changed | ||||
* | glsl: Fix non-immediate buffer access | ameerj | 2021-07-23 | 12 | -72/+133 |
| | | | | and many other misc implementations | ||||
* | glsl: textures wip | ameerj | 2021-07-23 | 9 | -75/+139 |
| | |||||
* | glsl: Implement some attribute getters and setters | ameerj | 2021-07-23 | 9 | -191/+337 |
| | |||||
* | glsl: Track S32 atomics | ameerj | 2021-07-23 | 3 | -6/+16 |
| | |||||
* | glsl: Update phi node management | ameerj | 2021-07-23 | 4 | -21/+53 |
| | |||||
* | glsl: Fix floating point compare ops | ameerj | 2021-07-23 | 1 | -28/+28 |
| | | | | Logic for ordered/unordered ops was wrong. | ||||
* | glsl: Query GL Device for FP16 extension support | ameerj | 2021-07-23 | 2 | -2/+9 |
| | |||||
* | glsl: Simply FP storage atomics | ameerj | 2021-07-23 | 2 | -48/+28 |
| | |||||
* | glsl: F16x2 storage atomics | ameerj | 2021-07-23 | 7 | -58/+64 |
| | |||||
* | glsl: Revert ssbo aliasing. Storage Atomics impl | ameerj | 2021-07-23 | 5 | -75/+134 |
| | |||||
* | glsl: implement phi nodes | ameerj | 2021-07-23 | 4 | -20/+54 |
| | |||||
* | glsl: Wip storage atomic ops | ameerj | 2021-07-23 | 10 | -327/+414 |
| | |||||
* | glsl: Implement FCMP | ameerj | 2021-07-23 | 3 | -242/+185 |
| | |||||
* | glsl: Add a more robust fp formatter | ameerj | 2021-07-23 | 4 | -9/+14 |
| | |||||
* | glsl: More FP fixes | ameerj | 2021-07-23 | 2 | -9/+16 |
| | |||||
* | glsl: FP function fixes | ameerj | 2021-07-23 | 7 | -17/+25 |
| | |||||
* | glsl: More FP instructions/fixes | ameerj | 2021-07-23 | 5 | -28/+41 |
| | |||||
* | glsl: Add many FP32/64 instructions | ameerj | 2021-07-23 | 12 | -765/+1011 |
| | |||||
* | glsl: Implement more Integer ops | ameerj | 2021-07-23 | 3 | -119/+72 |
| | |||||
* | glsl: Implement BF* | ameerj | 2021-07-23 | 3 | -9/+10 |
| | |||||
* | glsl: Implement a few Integer instructions | ameerj | 2021-07-23 | 10 | -260/+398 |
| | |||||
* | glsl: Use std::string_view for Emit function args. | ameerj | 2021-07-23 | 6 | -760/+838 |
| | |||||
* | glsl: Pass IR::Inst& to Emit functions | ameerj | 2021-07-23 | 6 | -171/+169 |
| | |||||
* | glsl: INeg and IAdd negate tests | ameerj | 2021-07-23 | 3 | -94/+106 |
| | |||||
* | glsl: Reusable typed variables. IADD32 | ameerj | 2021-07-23 | 6 | -203/+311 |
| | |||||
* | glsl: Fix program linking and cbuf | ameerj | 2021-07-23 | 2 | -3/+5 |
| | |||||
* | glsl: Fix "reg" allocing | ameerj | 2021-07-23 | 10 | -898/+938 |
| | | | | based on glasm with some tweaks | ||||
* | glsl: Initial backend | ameerj | 2021-07-23 | 27 | -0/+3292 |
| | |||||
* | spirv: Reduce log severity of mismatching denorm rules | ReinUsesLisp | 2021-07-23 | 1 | -2/+2 |
| | |||||
* | shader: Fix loop safety to SSA pass | ReinUsesLisp | 2021-07-23 | 2 | -2/+4 |
| | |||||
* | shader: Add logging | ReinUsesLisp | 2021-07-23 | 13 | -28/+30 |
| | |||||
* | shader: Add shader loop safety check settings | lat9nq | 2021-07-23 | 9 | -33/+130 |
| | | | | Also add a setting for enable Nsight Aftermath. | ||||
* | shader: Comment why the array component is not read in TMML | ReinUsesLisp | 2021-07-23 | 1 | -0/+2 |
| | |||||
* | tmml: Remove index component from coords vec | ameerj | 2021-07-23 | 1 | -4/+3 |
| | | | | The lod query functions exposed by the rendering API's do not make use of the texturearray layer indexing. | ||||
* | spirv/convert: Catch more signed operations oversights | ameerj | 2021-07-23 | 1 | -5/+5 |
| | | | | The sign bit on integers of size < 32 was not properly preserved in casts | ||||
* | spirv/convert: Catch more broken signed operations on Nvidia OpenGL | ReinUsesLisp | 2021-07-23 | 1 | -0/+6 |
| | | | | | BitCast U32 to S32 before converting to float on drivers with broken signed operations. | ||||
* | shader_environment: Add shader_local_memory_crs_size to local memory size | ameerj | 2021-07-23 | 1 | -2/+2 |
| | | | | Fixes DOOM 2016 missing local memory | ||||
* | shader: Fix VertexA Shaders. | FernandoS27 | 2021-07-23 | 3 | -14/+30 |
| | |||||
* | shader: Add 2D and 3D variants to SUATOM and SURED | ReinUsesLisp | 2021-07-23 | 1 | -0/+4 |
| | | | | Used by Claybook. | ||||
* | shader: Avoid CPU side undefined behavior on I2F | ReinUsesLisp | 2021-07-23 | 1 | -0/+2 |
| | |||||
* | glasm: Use ARB_derivative_control conditionally | ReinUsesLisp | 2021-07-23 | 3 | -7/+30 |
| | |||||
* | buffer_cache: Reduce uniform buffer size from shader usage | ReinUsesLisp | 2021-07-23 | 2 | -3/+17 |
| | | | | Increases performance significantly on certain titles. | ||||
* | emit_glasm_context_get_set: Remove unused variable | lat9nq | 2021-07-23 | 1 | -1/+0 |
| | |||||
* | shader,glasm: Implement legacy texcoord loads | ReinUsesLisp | 2021-07-23 | 3 | -54/+29 |
| | |||||
* | glasm: Implement legacy varyings | ReinUsesLisp | 2021-07-23 | 1 | -17/+56 |
| | |||||
* | shader: Track legacy varyings | ReinUsesLisp | 2021-07-23 | 2 | -17/+105 |
| | |||||
* | shader: Add support for "negative" and unaligned offsets | ReinUsesLisp | 2021-07-23 | 3 | -8/+13 |
| | | | | | | | | | "Negative" offsets don't exist. They are shown as such due to a bug in nvdisasm. Unaligned offsets have been proved to read the aligned offset. For example, when reading an U32, if the offset is 6, the offset read will be 4. | ||||
* | shader: Implement ISCADD32I | ReinUsesLisp | 2021-07-23 | 1 | -17/+31 |
| | |||||
* | spirv: Fix output generics with components | ReinUsesLisp | 2021-07-23 | 1 | -1/+1 |
| | |||||
* | opengl: Declare fragment outputs even if they are not used | ReinUsesLisp | 2021-07-23 | 4 | -10/+9 |
| | | | | | | Fixes Ori and the Blind Forest's menu on GLASM. For some reason (probably high level optimizations) it is not sanitized on SPIR-V for OpenGL. Vulkan is unaffected by this change. | ||||
* | shader: Always initialize up reference in structure control flow | ReinUsesLisp | 2021-07-23 | 1 | -31/+36 |
| | | | | Fixes ubsan issue. | ||||
* | shader: Fix ImageWrite indexing | ReinUsesLisp | 2021-07-23 | 1 | -1/+1 |
| | |||||
* | spirv: Fix image and image buffer descriptor index usage | ReinUsesLisp | 2021-07-23 | 1 | -5/+7 |
| | |||||
* | glasm: Fix immediate texture coordinate | ReinUsesLisp | 2021-07-23 | 1 | -0/+1 |
| | |||||
* | shader: Clang-format secondary textures | ReinUsesLisp | 2021-07-23 | 1 | -2/+2 |
| | |||||
* | shader: Fix secondary textures | ReinUsesLisp | 2021-07-23 | 1 | -2/+2 |
| | |||||
* | shader: Fix TMML queries | ReinUsesLisp | 2021-07-23 | 1 | -5/+9 |
| | |||||
* | shader: Fix FSwizzleAdd folding when going through phi nodes | ReinUsesLisp | 2021-07-23 | 1 | -2/+2 |
| | |||||
* | shader/exception: Fix compilation errors on gcc | ReinUsesLisp | 2021-07-23 | 1 | -6/+6 |
| | |||||
* | glasm: Reduce reg allocation leaks from an exception to a log | ReinUsesLisp | 2021-07-23 | 1 | -1/+1 |
| | |||||
* | shader: Handle host exceptions | ReinUsesLisp | 2021-07-23 | 4 | -13/+43 |
| | |||||
* | glasm: Use integer lod for TXQ | ReinUsesLisp | 2021-07-23 | 2 | -2/+2 |
| | |||||
* | glasm: Fix global memory fallbacks | ReinUsesLisp | 2021-07-23 | 1 | -9/+10 |
| | |||||
* | Revert "glasm: Skip phi moves on undefined instructions" | ReinUsesLisp | 2021-07-23 | 2 | -16/+1 |
| | | | | Causes regressions on Bowser's Fury. | ||||
* | glasm: Remove unintentional '\n' on Undef32 | ReinUsesLisp | 2021-07-23 | 1 | -1/+1 |
| | |||||
* | glasm: Use storage buffers instead of global memory when possible | ReinUsesLisp | 2021-07-23 | 6 | -370/+383 |
| | |||||
* | glasm: Implement Y direction | ReinUsesLisp | 2021-07-23 | 4 | -3/+9 |
| | |||||
* | glasm: Skip phi moves on undefined instructions | ReinUsesLisp | 2021-07-23 | 2 | -1/+16 |
| | |||||
* | glasm: Implement undef instructions | ReinUsesLisp | 2021-07-23 | 2 | -15/+15 |
| | |||||
* | glasm: Fix global memory callbacks | ReinUsesLisp | 2021-07-23 | 1 | -5/+6 |
| | |||||
* | video_core,shader: Clang-format fixes | ReinUsesLisp | 2021-07-23 | 2 | -2/+2 |
| | |||||
* | glasm: Release phi node registers after they are no longer needed | ReinUsesLisp | 2021-07-23 | 2 | -38/+54 |
| | |||||
* | glasm: Remove unintentionally committed fmt::prints | ReinUsesLisp | 2021-07-23 | 1 | -2/+0 |
| | |||||
* | glasm: Fix INeg32 on negative immediates | ReinUsesLisp | 2021-07-23 | 1 | -1/+5 |
| | |||||
* | glasm: Remove unnecessary value types | ReinUsesLisp | 2021-07-23 | 3 | -47/+6 |
| | |||||
* | glasm: Throw when there are register leaks | ReinUsesLisp | 2021-07-23 | 2 | -0/+7 |
| | |||||
* | glasm: Catch more register leaks | ReinUsesLisp | 2021-07-23 | 8 | -41/+114 |
| | | | | | | | | | | | | | Add support for null registers. These are used when an instruction has no usages. This comes handy when an instruction is only used for its CC value, with the caveat of having to invalidate all pseudo-instructions before defining the instruction itself in the register allocator. This commits changes this. Workaround a bug on Nvidia's condition codes conditional execution using branches. | ||||
* | glasm: Fix usage counting on phi nodes | ReinUsesLisp | 2021-07-23 | 3 | -8/+22 |
| | |||||
* | glasm: Implement global memory fallbacks | ReinUsesLisp | 2021-07-23 | 2 | -50/+89 |
| | |||||
* | glasm: Implement int64 add and subtract | ReinUsesLisp | 2021-07-23 | 2 | -8/+6 |
| | |||||
* | emit_glasm_context_get_set: Remove unused variable | lat9nq | 2021-07-23 | 1 | -1/+0 |
| | |||||
* | glasm: Implement indirect attribute loads | ReinUsesLisp | 2021-07-23 | 4 | -6/+65 |
| | |||||
* | glasm: Implement image atomics | ReinUsesLisp | 2021-07-23 | 3 | -166/+153 |
| | |||||
* | glasm: Reorder unreachable image atomic insts | ReinUsesLisp | 2021-07-23 | 1 | -66/+66 |
| | | | | Reorder them to the bottom of the file for readability. | ||||
* | glasm: Implement gl_Layer stores | ReinUsesLisp | 2021-07-23 | 1 | -0/+7 |
| | |||||
* | glasm: Implement SampleId | ReinUsesLisp | 2021-07-23 | 2 | -3/+3 |
| | |||||
* | glasm: Implement IsHelperInvocation | ReinUsesLisp | 2021-07-23 | 2 | -3/+3 |
| | |||||
* | glasm: Fix EmitVertex's optimization | ReinUsesLisp | 2021-07-23 | 1 | -1/+1 |
| | |||||
* | gl_shader_cache,glasm: Conditionally use typeless image reads extension | ReinUsesLisp | 2021-07-23 | 1 | -2/+4 |
| | |||||
* | glasm: Implement forced early Z | ReinUsesLisp | 2021-07-23 | 1 | -2/+6 |
| | |||||
* | glasm: Simplify patch reads | ReinUsesLisp | 2021-07-23 | 1 | -5/+2 |
| | |||||
* | glasm: Fix output patch reads | ReinUsesLisp | 2021-07-23 | 2 | -13/+22 |
| | | | | With this, Luigi's Mansion's sand renders properly. | ||||
* | shader: Split profile and runtime information in separate structs | ReinUsesLisp | 2021-07-23 | 11 | -71/+88 |
| | |||||
* | emit_glasm_context_get_and_set.cpp: Add missing semicolons | ameerj | 2021-07-23 | 1 | -2/+2 |
| | |||||
* | glasm: Fix patch attribute declarations | ReinUsesLisp | 2021-07-23 | 1 | -1/+1 |
| | |||||
* | glasm: Implement FSWZADD | ameerj | 2021-07-23 | 3 | -4/+28 |
| | |||||
* | glasm: Implement PrimitiveId attribute read | ReinUsesLisp | 2021-07-23 | 1 | -0/+3 |
| | |||||
* | glasm: Implement clip distance stores | ReinUsesLisp | 2021-07-23 | 2 | -0/+15 |
| | |||||
* | glasm: Fix tessellation input attributes | ReinUsesLisp | 2021-07-23 | 1 | -2/+5 |
| | |||||
* | glasm: Add missing semicolon on tesscoord reading | ReinUsesLisp | 2021-07-23 | 1 | -1/+1 |
| | |||||
* | glasm: Fix tessellation headers | ReinUsesLisp | 2021-07-23 | 1 | -2/+2 |
| | |||||
* | glasm: Add tessellation shader declarations | ReinUsesLisp | 2021-07-23 | 1 | -0/+35 |
| | |||||
* | glasm: Implement TessellationEvaluationPoint | ReinUsesLisp | 2021-07-23 | 1 | -0/+4 |
| | |||||
* | glasm: Implement patch memory | ReinUsesLisp | 2021-07-23 | 3 | -6/+51 |
| | |||||
* | glasm: Fix InvocationId declaration | ReinUsesLisp | 2021-07-23 | 1 | -1/+1 |
| | |||||
* | glasm: Implement InvocationId | ReinUsesLisp | 2021-07-23 | 2 | -2/+5 |
| | |||||
* | glasm: Optimize EmitVertex into EMIT | ReinUsesLisp | 2021-07-23 | 1 | -1/+5 |
| | |||||
* | glasm: Implement geometry shader attribute reads | ReinUsesLisp | 2021-07-23 | 2 | -4/+18 |
| | |||||
* | glasm: Properly declare attributes on geometry programs | ReinUsesLisp | 2021-07-23 | 3 | -6/+14 |
| | |||||
* | glasm: Declare geometry program headers | ReinUsesLisp | 2021-07-23 | 1 | -0/+35 |
| | |||||
* | glasm: Fix potential aliasing bug on cube array samples | ReinUsesLisp | 2021-07-23 | 2 | -35/+44 |
| | |||||
* | glasm: Implement ImageWrite | ReinUsesLisp | 2021-07-23 | 1 | -4/+7 |
| | |||||
* | glasm: Implement ImageRead | ReinUsesLisp | 2021-07-23 | 4 | -4/+56 |
| | |||||
* | glasm: Implement EmitVertex and EndPrimitive | ReinUsesLisp | 2021-07-23 | 2 | -4/+8 |
| | |||||
* | glasm: Implement ImageGradient | ReinUsesLisp | 2021-07-23 | 2 | -7/+65 |
| | |||||
* | glasm: Implement 64-bit shifts | ReinUsesLisp | 2021-07-23 | 2 | -12/+14 |
| | |||||
* | glasm: Implement barriers | ReinUsesLisp | 2021-07-23 | 1 | -3/+3 |
| | |||||
* | glasm: Fix compute stage name | ReinUsesLisp | 2021-07-23 | 1 | -1/+1 |
| | |||||
* | glasm: Fix phi instruction types | ReinUsesLisp | 2021-07-23 | 1 | -1/+1 |
| | |||||
* | glasm: Implement PREC on relevant instructions | ReinUsesLisp | 2021-07-23 | 1 | -6/+12 |
| | |||||
* | glasm: Implement stores to gl_ViewportIndex | ReinUsesLisp | 2021-07-23 | 4 | -7/+29 |
| | |||||
* | glasm: Implement gl_PointSize stores | ReinUsesLisp | 2021-07-23 | 1 | -0/+3 |
| | |||||
* | glasm: Implement gl_PointCoord | ReinUsesLisp | 2021-07-23 | 1 | -0/+4 |
| | |||||
* | glasm: Implement ImageQueryLod | ReinUsesLisp | 2021-07-23 | 1 | -3/+5 |
| | |||||
* | glasm: Implement ImageFetch | ReinUsesLisp | 2021-07-23 | 4 | -13/+38 |
| | |||||
* | glasm: Implement IADD.CC | ameerj | 2021-07-23 | 1 | -1/+26 |
| | |||||
* | glasm: Implement BFE.CC | ReinUsesLisp | 2021-07-23 | 1 | -0/+8 |
| | |||||
* | glasm: Implement SelectU1 | ReinUsesLisp | 2021-07-23 | 2 | -4/+5 |
| | |||||
* | glasm: Implement gl_WorkGroupID | ReinUsesLisp | 2021-07-23 | 2 | -3/+3 |
| | |||||
* | glasm: Implement TXQ and improve texture info reads | ReinUsesLisp | 2021-07-23 | 2 | -50/+51 |
| | |||||
* | glasm: Implement gl_FrongFacing attribute | ReinUsesLisp | 2021-07-23 | 1 | -0/+3 |
| | |||||
* | glasm: Support textures used in more than one stage | ReinUsesLisp | 2021-07-23 | 3 | -4/+24 |
| | |||||
* | glasm: Implement textureGather instructions | ReinUsesLisp | 2021-07-23 | 2 | -15/+97 |
| | |||||
* | glasm: Implement gl_FragDepth and gl_SampleMask stores | ReinUsesLisp | 2021-07-23 | 2 | -5/+5 |
| | |||||
* | glasm: Do not alias ConditionRef for now | ReinUsesLisp | 2021-07-23 | 2 | -3/+2 |
| | | | | | Immediate condition refs where not handled correctly. Just move the value for now. | ||||
* | shader: Read branch conditions from an instruction | ReinUsesLisp | 2021-07-23 | 12 | -16/+36 |
| | | | | Fixes the identity removal pass. | ||||
* | glasm: Implement InstanceId and VertexId | ReinUsesLisp | 2021-07-23 | 1 | -0/+6 |
| | |||||
* | glasm: Add missing return value on move assignment | ReinUsesLisp | 2021-07-23 | 1 | -0/+1 |
| | |||||
* | glasm: Fix aliased bitcasts ref counting | ReinUsesLisp | 2021-07-23 | 3 | -13/+42 |
| | |||||
* | glasm: Remove unintentional comma on vector insert | ReinUsesLisp | 2021-07-23 | 1 | -1/+1 |
| | |||||
* | glasm: Implement TEX and TEXS instructions | ReinUsesLisp | 2021-07-23 | 10 | -69/+275 |
| | | | | | Remove lod clamp from texture instructions with lod, as this is not needed (nor supported). | ||||
* | glasm: Add support for non-2D texture samples | ReinUsesLisp | 2021-07-23 | 1 | -4/+26 |
| | |||||
* | glasm: Reorder unreachable image instructions to the bottom | ReinUsesLisp | 2021-07-23 | 1 | -97/+97 |
| | |||||
* | glasm: Add support for texture offsets | ReinUsesLisp | 2021-07-23 | 1 | -11/+15 |
| | |||||
* | glasm: Improve texture sampling instructions | ReinUsesLisp | 2021-07-23 | 2 | -50/+70 |
| | |||||
* | emit_glasm: Enable ARB_draw_buffers when needed | ReinUsesLisp | 2021-07-23 | 2 | -1/+5 |
| | |||||
* | emit_glasm: Add support for reading position attributes | ReinUsesLisp | 2021-07-23 | 1 | -3/+13 |
| | |||||
* | shader_recompiler: GCC fixes | lat9nq | 2021-07-23 | 7 | -58/+55 |
| | | | | | Fixes members of unnamed union not being accessible, and one function without a declaration. | ||||
* | glasm: Implement rest of shared mem | ameerj | 2021-07-23 | 2 | -35/+29 |
| | |||||
* | shader: Use a non-trivial dummy to construct ASL node union | ReinUsesLisp | 2021-07-23 | 1 | -1/+6 |
| | |||||
* | emit_spirv: Jump to loop body with local variable | ReinUsesLisp | 2021-07-23 | 1 | -1/+1 |
| | | | | Silence unused variable warning | ||||
* | glasm: Implement derivative instructions on GLASM | ReinUsesLisp | 2021-07-23 | 2 | -12/+12 |
| | |||||
* | glasm: Initial (broken) implementation of TEX on GLASM | ReinUsesLisp | 2021-07-23 | 3 | -299/+386 |
| | |||||
* | glasm: Implement some graphics instructions on GLASM | ReinUsesLisp | 2021-07-23 | 2 | -6/+5 |
| | |||||
* | glasm: Add Void type to GLASM values | ReinUsesLisp | 2021-07-23 | 3 | -0/+15 |
| | |||||
* | glasm: Add graphics specific shader declarations to GLASM | ReinUsesLisp | 2021-07-23 | 2 | -6/+63 |
| | |||||
* | glasm: Implement local memory for glasm | ameerj | 2021-07-23 | 4 | -9/+12 |
| | |||||
* | emit_spirv: Add missing block in case | ReinUsesLisp | 2021-07-23 | 1 | -1/+2 |
| | |||||
* | glasm: Initial implementation of phi nodes on GLASM | ReinUsesLisp | 2021-07-23 | 12 | -25/+117 |
| | |||||
* | glasm: Write result to scalar on integer comparison instructions | ReinUsesLisp | 2021-07-23 | 1 | -10/+10 |
| | |||||
* | glasm: Declare NV_shader_thread_group when needed | ReinUsesLisp | 2021-07-23 | 1 | -3/+4 |
| | |||||
* | glasm: Rework control flow introducing a syntax list | ReinUsesLisp | 2021-07-23 | 33 | -505/+437 |
| | | | | | This commit regresses VertexA shaders, their transformation pass has to be adapted to the new control flow. | ||||
* | glasm: Implement Storage atomics | ameerj | 2021-07-23 | 5 | -109/+156 |
| | | | | | StorageAtomicExchangeU64 is failing test seemingly due to failure storing 64-bit result into the register | ||||
* | glasm: Ensure reg alloc order across compilers on GLASM | ReinUsesLisp | 2021-07-23 | 1 | -11/+14 |
| | | | | | | | | | | | | | | | | | | Use a struct constructor to serialize register allocation arguments to ensure registers are allocated in the same order regardless of the compiler used. The A and B functions can be called in any order when passed as arguments to "foo": foo(A(), B()) But the order is guaranteed for curly-braced constructor calls in classes: Foo{A(), B()} Use this to get consistent behavior. | ||||
* | glasm: Enable unintentionally disabled register aliasing on GLASM | ReinUsesLisp | 2021-07-23 | 1 | -16/+11 |
| | |||||
* | glasm: Review all GLASM insts to be aware of register aliasing | ReinUsesLisp | 2021-07-23 | 4 | -20/+51 |
| | |||||
* | glasm: Implement shuffle and vote instructions on GLASM | ReinUsesLisp | 2021-07-23 | 10 | -100/+166 |
| | |||||
* | glasm: Add MUFU instructions to GLASM | ReinUsesLisp | 2021-07-23 | 2 | -21/+22 |
| | |||||
* | glasm: Implement IAbs64 and INeg64 on GLASM | ReinUsesLisp | 2021-07-23 | 2 | -6/+6 |
| | |||||
* | shader: Add floating-point rounding to I2F | ReinUsesLisp | 2021-07-23 | 3 | -35/+42 |
| | |||||
* | glasm: Properly clamp Fp64 on GLASM | ReinUsesLisp | 2021-07-23 | 1 | -6/+6 |
| | |||||
* | glasm: Fix register allocation when moving immediate on GLASM | ReinUsesLisp | 2021-07-23 | 3 | -42/+89 |
| | |||||
* | glasm: Implement SelectU64 on GLASM | ReinUsesLisp | 2021-07-23 | 2 | -4/+20 |
| | |||||
* | glasm: Fix clamps so the min value has priority on NAN on GLASM | ReinUsesLisp | 2021-07-23 | 1 | -12/+15 |
| | |||||
* | glasm: Fix moving U64 immediates to registers in GLASM | ReinUsesLisp | 2021-07-23 | 2 | -3/+4 |
| | |||||
* | glasm: Implement storage atomic ops | ameerj | 2021-07-23 | 4 | -305/+358 |
| | |||||
* | glasm: Add conversion instructions to GLASM | ReinUsesLisp | 2021-07-23 | 9 | -282/+351 |
| | |||||
* | glasm: Add fp min/max insts and fix store for fp64 on GLASM | ReinUsesLisp | 2021-07-23 | 2 | -10/+8 |
| | |||||
* | glasm: Add logical instructions on GLASM | ReinUsesLisp | 2021-07-23 | 2 | -12/+12 |
| | |||||
* | glasm: Remove duplicated Fp64 pack instructions on GLASM | ReinUsesLisp | 2021-07-23 | 1 | -8/+0 |
| | |||||
* | glasm: Remove unnecesary new white space on Clamp GLASM | ReinUsesLisp | 2021-07-23 | 1 | -4/+4 |
| | |||||
* | glasm: Add floating-point comparisons on GLASM | ReinUsesLisp | 2021-07-23 | 3 | -120/+116 |
| | |||||
* | emit_glasm: Implement more integer alu ops | ameerj | 2021-07-23 | 2 | -47/+41 |
| | |||||
* | glasm: Reimplement bitwise ops and BFI/BFE | ameerj | 2021-07-23 | 4 | -88/+108 |
| | |||||
* | glasm: Initial GLASM fp64 support | ReinUsesLisp | 2021-07-23 | 9 | -55/+152 |
| | |||||
* | glasm: Implement GLASM fp16 packing and move bitwise insns | ReinUsesLisp | 2021-07-23 | 4 | -66/+77 |
| | |||||
* | glasm: Remove unused functions left from rebase | ReinUsesLisp | 2021-07-23 | 1 | -12/+0 |
| | |||||
* | glasm: Specify namespace when using FormatTo | ReinUsesLisp | 2021-07-23 | 1 | -6/+6 |
| | |||||
* | glasm: Implement more GLASM composite instructions | ReinUsesLisp | 2021-07-23 | 2 | -54/+63 |
| | |||||
* | glasm: Make GLASM aware of types | ReinUsesLisp | 2021-07-23 | 12 | -1244/+1380 |
| | |||||
* | glasm: Use CMP.S for Select32 | ameerj | 2021-07-23 | 3 | -12/+8 |
| | | | | also fixes ADD and SUB to use U modifier | ||||
* | glasm: Implement more logical ops | ameerj | 2021-07-23 | 2 | -5/+5 |
| | |||||
* | glasm: Implement BFI, BFE | ameerj | 2021-07-23 | 4 | -138/+164 |
| | | | | Along with implementations of common instructions along the way | ||||
* | glasm: Use BitField instead of C bitfields | ReinUsesLisp | 2021-07-23 | 2 | -8/+12 |
| | |||||
* | glasm: Remove unused argument in identity instructions on GLASM | ReinUsesLisp | 2021-07-23 | 1 | -7/+7 |
| | |||||
* | glasm: Implement basic GLASM instructions | ReinUsesLisp | 2021-07-23 | 10 | -840/+1173 |
| | |||||
* | glasm: Changes to GLASM register allocator and emit context | ReinUsesLisp | 2021-07-23 | 4 | -26/+64 |
| | |||||
* | glasm: Add GLASM backend infrastructure | ReinUsesLisp | 2021-07-23 | 28 | -4/+3115 |
| | |||||
* | shader: ISET.X implementation | ameerj | 2021-07-23 | 1 | -8/+58 |
| | |||||
* | shader: Fixup SPIR-V emit header namespaces | ReinUsesLisp | 2021-07-23 | 1 | -2/+2 |
| | |||||
* | Move SPIR-V emission functions to their own header | ReinUsesLisp | 2021-07-23 | 24 | -572/+631 |
| | |||||
* | shader: Optimize NVN Fallthrough | FernandoS27 | 2021-07-23 | 4 | -9/+83 |
| | |||||
* | shader: Stub SR_AFFINITY | FernandoS27 | 2021-07-23 | 1 | -0/+3 |
| | |||||
* | shader: Implement Int32 SUATOM/SURED | ameerj | 2021-07-23 | 17 | -6/+733 |
| | |||||
* | shader: Initial OpenGL implementation | ReinUsesLisp | 2021-07-23 | 3 | -0/+12 |
| | |||||
* | spirv: Be aware of NAN unaware drivers | ReinUsesLisp | 2021-07-23 | 1 | -18/+40 |
| | |||||
* | spirv: Add SSBO read fallbacks when no aliasing is available | ReinUsesLisp | 2021-07-23 | 1 | -37/+99 |
| | |||||
* | spirv: Add OpKill fallback to demote | ReinUsesLisp | 2021-07-23 | 1 | -2/+6 |
| | |||||
* | spirv: Do not enable ShaderLayer | ReinUsesLisp | 2021-07-23 | 1 | -3/+0 |
| | | | | This is enabled by an extension instead of the capability. | ||||
* | spirv: Enable DemoteToHelperInvocationEXT only when supported | ReinUsesLisp | 2021-07-23 | 1 | -1/+1 |
| | |||||
* | spirv: Use OriginLowerLeft when requested | ReinUsesLisp | 2021-07-23 | 1 | -1/+5 |
| | |||||
* | spirv: Only add image operands mask when needed | ReinUsesLisp | 2021-07-23 | 1 | -5/+9 |
| | |||||
* | spirv: Workaround image unsigned offset bug | ReinUsesLisp | 2021-07-23 | 2 | -9/+26 |
| | | | | | Workaround bug on Nvidia's OpenGL SPIR-V compiler when using unsigned texture offsets. | ||||
* | spirv: Add int8 and int16 capabilities only when supported | ReinUsesLisp | 2021-07-23 | 1 | -2/+2 |
| | |||||
* | spirv: Add integer clamping workarounds | ReinUsesLisp | 2021-07-23 | 1 | -4/+34 |
| | | | | Workaround more bugs on Nvidia's OpenGL SPIR-V compiler. | ||||
* | spirv: Implement int8 and int16 conversion fallbacks | ReinUsesLisp | 2021-07-23 | 1 | -19/+80 |
| | |||||
* | spirv: Support OpenGL uniform buffers and change bindings | ReinUsesLisp | 2021-07-23 | 5 | -56/+163 |
| | |||||
* | spirv: Desambiguate descriptor names | ReinUsesLisp | 2021-07-23 | 1 | -9/+37 |
| | | | | | Worksaround a bug on Nvidia's OpenGL SPIR-V compiler where names are used for name matching. | ||||
* | shader: Add OpenGL shader profile options | ReinUsesLisp | 2021-07-23 | 1 | -0/+11 |
| | |||||
* | shader: Remove shader util | ReinUsesLisp | 2021-07-23 | 4 | -176/+0 |
| | |||||
* | shader: Address feedback | FernandoS27 | 2021-07-23 | 4 | -35/+33 |
| | |||||
* | shader: Implement VertexA stage | FernandoS27 | 2021-07-23 | 11 | -0/+166 |
| | |||||
* | shader: Implement delegation of Exit to dispatcher on CFG | FernandoS27 | 2021-07-23 | 2 | -3/+47 |
| | |||||
* | shader: Fix IADD3.CC | ameerj | 2021-07-23 | 1 | -12/+5 |
| | |||||
* | shader: Fix BFE s32 undefined check | ameerj | 2021-07-23 | 1 | -1/+1 |
| | | | | Our unit tests were hitting this exception. | ||||
* | shader: Fix error checking in bitfieldExtract and implement bitfieldInsert folding | ReinUsesLisp | 2021-07-23 | 1 | -5/+14 |
| | |||||
* | shader: Fix storage type when reading patches on tess control | ReinUsesLisp | 2021-07-23 | 1 | -1/+2 |
| | |||||
* | shader: Fix VMNMX selector B | ReinUsesLisp | 2021-07-23 | 1 | -1/+2 |
| | |||||
* | shader: Increase the maximum number of storage buffers | ReinUsesLisp | 2021-07-23 | 1 | -1/+1 |
| | | | | | Compute shaders spill uniform buffers on storage buffers, increasing the expected number. | ||||
* | shader: Remove identity removal pass for better build times | ReinUsesLisp | 2021-07-23 | 1 | -1/+0 |
| | |||||
* | shader: Add more strict validation the pass | ReinUsesLisp | 2021-07-23 | 1 | -0/+42 |
| | |||||
* | shader: Fix forward referencing identity instructions when inserting phi | ReinUsesLisp | 2021-07-23 | 1 | -11/+13 |
| | |||||
* | shader: Remove invalidated blocks in dead code elimination pass | ReinUsesLisp | 2021-07-23 | 1 | -3/+6 |
| | |||||
* | shader: Add missing UndoUse case for GetSparseFromOp | ReinUsesLisp | 2021-07-23 | 1 | -0/+4 |
| | |||||
* | shader: Simplify code in opcodes.h to fix Intellisense | ReinUsesLisp | 2021-07-23 | 1 | -8/+6 |
| | | | | | | | | Avoid using std::array to fix Intellisense not properly compiling this code and disabling itself on all files that include it. While we are at it, change the code to use u8 instead of size_t for the number of instructions in an opcode. | ||||
* | shader: Implement indexed textures | ReinUsesLisp | 2021-07-23 | 7 | -93/+189 |
| | |||||
* | shader: Refactor atomic_operations_global_memory | ameerj | 2021-07-23 | 1 | -44/+36 |
| | |||||
* | shader: add missing include guard in half_floating_point_helper.h | ameerj | 2021-07-23 | 1 | -0/+2 |
| | |||||
* | shader: Fix gcc warnings | ReinUsesLisp | 2021-07-23 | 2 | -2/+2 |
| | |||||
* | shader: Inline common Value getters | ReinUsesLisp | 2021-07-23 | 2 | -109/+102 |
| | |||||
* | shader: Intrusively store in a block if it's sealed or not | ReinUsesLisp | 2021-07-23 | 2 | -3/+11 |
| | |||||
* | cmake: Link to common in shader_recompiler | ReinUsesLisp | 2021-07-23 | 1 | -1/+1 |
| | |||||
* | shader: Improve goto removal algorithm complexity | ReinUsesLisp | 2021-07-23 | 1 | -49/+28 |
| | | | | | Find sibling node containing a nephew searching from the nephew itself instead of the uncle. | ||||
* | shader: Use memset to reset instruction arguments | ReinUsesLisp | 2021-07-23 | 2 | -4/+7 |
| | |||||
* | shader: Inline common Value functions into the header | ReinUsesLisp | 2021-07-23 | 2 | -19/+23 |
| | |||||
* | shader: Move microinstruction header to the value header | ReinUsesLisp | 2021-07-23 | 19 | -180/+161 |
| | |||||
* | shader: Move siblings check to a separate function and comment them out | ReinUsesLisp | 2021-07-23 | 1 | -16/+21 |
| | |||||
* | shader: Intrusively store register values in block for SSA pass | ReinUsesLisp | 2021-07-23 | 2 | -21/+53 |
| | |||||
* | shader: Inline common Opcode and Inst functions | ReinUsesLisp | 2021-07-23 | 4 | -112/+83 |
| | |||||
* | shader: Inline common IR::Block methods | ReinUsesLisp | 2021-07-23 | 2 | -17/+12 |
| | |||||
* | shader: Use a small_vector for phi blocks | ReinUsesLisp | 2021-07-23 | 1 | -1/+2 |
| | |||||
* | shader: Calculate number of arguments in an opcode at compile time | ReinUsesLisp | 2021-07-23 | 1 | -3/+12 |
| | |||||
* | shader: Implement D3D samplers | ReinUsesLisp | 2021-07-23 | 3 | -12/+76 |
| | |||||
* | shader: Add constant propagation for arithmetic right shifts | ReinUsesLisp | 2021-07-23 | 1 | -0/+3 |
| | |||||
* | shader: Simplify code for local memory | ReinUsesLisp | 2021-07-23 | 1 | -6/+11 |
| | |||||
* | shader: Add NVN storage buffer fallbacks | ReinUsesLisp | 2021-07-23 | 9 | -62/+214 |
| | | | | | | | When we can't track the SSBO origin of a global memory instruction, leave it as a global memory operation and assume these pointers are in the NVN storage buffer slots, then apply a linear search in the shader's runtime. | ||||
* | spirv: Fix ViewportMask | ReinUsesLisp | 2021-07-23 | 1 | -1/+2 |
| | |||||
* | spirv: Replace Constant/ConstantComposite with Const helper | ameerj | 2021-07-23 | 12 | -112/+101 |
| | |||||
* | shader: Address feedback | FernandoS27 | 2021-07-23 | 2 | -7/+10 |
| | |||||
* | shader: Implement F2F (Imm) | FernandoS27 | 2021-07-23 | 1 | -2/+28 |
| | |||||
* | shader: Implement IADD3.CC/.X | FernandoS27 | 2021-07-23 | 1 | -7/+22 |
| | |||||
* | shader: Address feedback | FernandoS27 | 2021-07-23 | 4 | -7/+4 |
| | |||||
* | shader: Add coarse derivatives | FernandoS27 | 2021-07-23 | 7 | -8/+28 |
| | |||||
* | shader: Implement fine derivates constant propagation | FernandoS27 | 2021-07-23 | 9 | -0/+101 |
| | |||||
* | shader: Implement SR_Y_DIRECTION | FernandoS27 | 2021-07-23 | 7 | -0/+18 |
| | |||||
* | shader: Fix Phi node types | ReinUsesLisp | 2021-07-23 | 2 | -4/+4 |
| | |||||
* | shader: Fix memory barriers | ReinUsesLisp | 2021-07-23 | 8 | -62/+30 |
| | |||||
* | spirv: Fix implicit lod type | ReinUsesLisp | 2021-07-23 | 2 | -1/+5 |
| | |||||
* | spirv: Use explicit lods outside of fragment shaders | ReinUsesLisp | 2021-07-23 | 1 | -5/+16 |
| | |||||
* | spirv: Use ConstOffset instead of Offset when possible | ReinUsesLisp | 2021-07-23 | 3 | -21/+67 |
| | |||||
* | shader: Implement BFE and BFI CC | ameerj | 2021-07-23 | 3 | -14/+17 |
| | | | | Fix two bugs in BFI. | ||||
* | shader: Implement SampleMask | ReinUsesLisp | 2021-07-23 | 11 | -2/+22 |
| | |||||
* | shader: Implement PIXLD.MY_INDEX | ReinUsesLisp | 2021-07-23 | 13 | -4/+69 |
| | |||||
* | spirv: Bitcast non-F32 output attributes to their type before store | ReinUsesLisp | 2021-07-23 | 1 | -13/+28 |
| | |||||
* | spirv: Implement ViewportMask with NV_viewport_array2 | ReinUsesLisp | 2021-07-23 | 7 | -0/+20 |
| | |||||
* | spirv: Bitcast non-F32 attributes to F32 | ReinUsesLisp | 2021-07-23 | 1 | -7/+9 |
| | |||||
* | shader: Implement PrimitiveId | ReinUsesLisp | 2021-07-23 | 5 | -0/+10 |
| | |||||
* | shader: Implement tessellation shaders, polygon mode and invocation id | ReinUsesLisp | 2021-07-23 | 22 | -88/+555 |
| | |||||
* | shader: Mark atomic instructions as writes | ReinUsesLisp | 2021-07-23 | 1 | -0/+27 |
| | |||||
* | spirv: Implement image buffers | ReinUsesLisp | 2021-07-23 | 5 | -23/+86 |
| | |||||
* | spirv: Implement Layer stores | ReinUsesLisp | 2021-07-23 | 6 | -9/+30 |
| | |||||
* | spirv: Fix alpha test | FernandoS27 | 2021-07-23 | 1 | -0/+5 |
| | |||||
* | spirv: Fix non-atomic 64-bit store | ameerj | 2021-07-23 | 1 | -1/+1 |
| | |||||
* | spirv: Implement alpha test | ameerj | 2021-07-23 | 2 | -1/+59 |
| | |||||
* | shader: Implement transform feedbacks and define file format | ReinUsesLisp | 2021-07-23 | 8 | -16/+116 |
| | |||||
* | shader: Implement early Z tests | ReinUsesLisp | 2021-07-23 | 2 | -0/+4 |
| | |||||
* | shader: Document and relax cache control on surface instructions | ReinUsesLisp | 2021-07-23 | 1 | -10/+11 |
| | |||||
* | spirv: Rework storage buffers and shader memory | ReinUsesLisp | 2021-07-23 | 8 | -499/+553 |
| | |||||
* | shader: Fix fixed pipeline point size on geometry shaders | ReinUsesLisp | 2021-07-23 | 1 | -10/+18 |
| | |||||
* | shader: Add constant propagation for *&^| binary operations | ReinUsesLisp | 2021-07-23 | 1 | -0/+12 |
| | |||||
* | shader: Implement geometry shaders | ReinUsesLisp | 2021-07-23 | 12 | -84/+221 |
| | |||||
* | shader: Implement OUT | ReinUsesLisp | 2021-07-23 | 10 | -17/+73 |
| | |||||
* | internal_stage_buffer_entry_read: Remove pragma optimize off | lat9nq | 2021-07-23 | 1 | -2/+0 |
| | |||||
* | shader: Stub SR_INVOCATION_INFO | ReinUsesLisp | 2021-07-23 | 1 | -2/+5 |
| | |||||
* | shader: Stub ISBERD | ReinUsesLisp | 2021-07-23 | 3 | -4/+56 |
| | |||||
* | shader: Fix CC in I2I | ReinUsesLisp | 2021-07-23 | 1 | -0/+2 |
| | |||||
* | spirv: Define StorageImageWriteWithoutFormat capability when used | ReinUsesLisp | 2021-07-23 | 3 | -0/+9 |
| | |||||
* | shader: Simplify FLO and throw on CC | ReinUsesLisp | 2021-07-23 | 1 | -12/+13 |
| | |||||
* | shader: Mark blocks with no end branch as unreachable | ReinUsesLisp | 2021-07-23 | 1 | -2/+7 |
| | |||||
* | shader: Implement LOP CC | ReinUsesLisp | 2021-07-23 | 3 | -12/+29 |
| | |||||
* | shader: Implement SR_THREAD_KILL | ReinUsesLisp | 2021-07-23 | 10 | -0/+22 |
| | |||||
* | shader: Apply sign bit in FCMP (imm) | ReinUsesLisp | 2021-07-23 | 1 | -1/+1 |
| | |||||
* | shader: Implement ATOM/S and RED | ameerj | 2021-07-23 | 18 | -19/+1724 |
| | |||||
* | spirv: Move phi node patching to a separate function | ReinUsesLisp | 2021-07-23 | 1 | -13/+16 |
| | |||||
* | spirv: Guard against typeless image reads on unsupported devices | ReinUsesLisp | 2021-07-23 | 5 | -1/+16 |
| | |||||
* | shader: Move LaneId to the warp emission file and fix AMD | ReinUsesLisp | 2021-07-23 | 5 | -7/+11 |
| | |||||
* | spirv: Fix forward declarations on phi nodes | ReinUsesLisp | 2021-07-23 | 1 | -47/+25 |
| | |||||
* | shader: Mark ImageWrite with side effects | ReinUsesLisp | 2021-07-23 | 1 | -0/+3 |
| | |||||
* | shader: Implement CC for ISET, FSET, PSET, CSET, and DSET | FernandoS27 | 2021-07-23 | 18 | -13/+136 |
| | | | | Throw when other instructions are missing CC. | ||||
* | shader: Remove outdated comment in F2I | ReinUsesLisp | 2021-07-23 | 1 | -4/+0 |
| | |||||
* | shader: Implement SULD and SUST | ReinUsesLisp | 2021-07-23 | 23 | -137/+597 |
| | |||||
* | shader: Fix Windows build issues | ReinUsesLisp | 2021-07-23 | 1 | -1/+1 |
| | |||||
* | shader: Address feedback + clang format | lat9nq | 2021-07-23 | 11 | -22/+20 |
| | |||||
* | shader_recompiler,video_core: Cleanup some GCC and Clang errors | lat9nq | 2021-07-23 | 59 | -297/+289 |
| | | | | | | | | | | | | | | | | | Mostly fixing unused *, implicit conversion, braced scalar init, fpermissive, and some others. Some Clang errors likely remain in video_core, and std::ranges is still a pertinent issue in shader_recompiler shader_recompiler: cmake: Force bracket depth to 1024 on Clang Increases the maximum fold expression depth thread_worker: Include condition_variable Don't use list initializers in control flow Co-authored-by: ReinUsesLisp <reinuseslisp@airmail.cc> | ||||
* | shader: Fix FCMP immediate variant | ReinUsesLisp | 2021-07-23 | 1 | -1/+9 |
| | |||||
* | shader: Fix dangling labels | ReinUsesLisp | 2021-07-23 | 1 | -0/+5 |
| | |||||
* | shader: Interact texture buffers with buffer cache | ReinUsesLisp | 2021-07-23 | 3 | -29/+29 |
| | |||||
* | shader: Fix F2I | ReinUsesLisp | 2021-07-23 | 1 | -1/+1 |
| | |||||
* | shader: Fix TextureGrad | ReinUsesLisp | 2021-07-23 | 1 | -1/+1 |
| | |||||
* | shader: Implement texture buffers | ReinUsesLisp | 2021-07-23 | 6 | -23/+125 |
| | |||||
* | shader: Address feedback | FernandoS27 | 2021-07-23 | 5 | -53/+54 |
| | |||||
* | shader: Implement indexed Position and ClipDistances | FernandoS27 | 2021-07-23 | 3 | -11/+100 |
| | |||||
* | shader: Implement indexed attributes | FernandoS27 | 2021-07-23 | 12 | -35/+279 |
| | |||||
* | shader: Implement AL2P | FernandoS27 | 2021-07-23 | 3 | -4/+36 |
| | |||||
* | shader: Fix BRX tracking | FernandoS27 | 2021-07-23 | 2 | -3/+4 |
| | |||||
* | shader: Move recursive SSA rewrite to the heap | ReinUsesLisp | 2021-07-23 | 1 | -29/+89 |
| | |||||
* | shader: Fix ShadowCube declaration type, set number of pipeline threads based on hardware | FernandoS27 | 2021-07-23 | 1 | -1/+1 |
| | |||||
* | shader: Fix splits on blocks using indirect branches | ReinUsesLisp | 2021-07-23 | 3 | -17/+38 |
| | |||||
* | shader: Eliminate orphan blocks more efficiently | ReinUsesLisp | 2021-07-23 | 1 | -7/+8 |
| | |||||
* | shader: Add subgroup masks | ReinUsesLisp | 2021-07-23 | 10 | -45/+169 |
| | |||||
* | shader: Implement BAR and fix memory barriers | ReinUsesLisp | 2021-07-23 | 7 | -5/+79 |
| | |||||
* | shader: Abstract breadth searches and use the abstraction | ReinUsesLisp | 2021-07-23 | 4 | -104/+106 |
| | |||||
* | shader: Reimplement GetCbufU64 as GetCbufU32x2 | ReinUsesLisp | 2021-07-23 | 9 | -22/+21 |
| | | | | It may generate better code on some compilers and it's easier to handle. | ||||
* | shader: Remove unused header in VOTE | ReinUsesLisp | 2021-07-23 | 1 | -2/+0 |
| | |||||
* | shader: Rework global memory tracking to use breadth-first search | ReinUsesLisp | 2021-07-23 | 1 | -69/+80 |
| | |||||
* | shader: Fix fp16 merge when using native fp16 | ReinUsesLisp | 2021-07-23 | 1 | -3/+3 |
| | |||||
* | shader: Fix FADD32I | ReinUsesLisp | 2021-07-23 | 1 | -6/+4 |
| | |||||
* | shader: Fix undetected bug from review | FernandoS27 | 2021-07-23 | 1 | -0/+3 |
| | |||||
* | shader: Address feedback | FernandoS27 | 2021-07-23 | 3 | -13/+16 |
| | |||||
* | shader: "Implement" NOP | FernandoS27 | 2021-07-23 | 1 | -1/+1 |
| | |||||
* | shader: Address Feedback | FernandoS27 | 2021-07-23 | 16 | -211/+60 |
| | |||||
* | shader: Implement SR_LaneId | FernandoS27 | 2021-07-23 | 7 | -0/+15 |
| | |||||
* | shader: Fix shared memory on cool drivers | FernandoS27 | 2021-07-23 | 1 | -0/+1 |
| | |||||
* | shader: Implement MEMBAR | FernandoS27 | 2021-07-23 | 9 | -11/+121 |
| | |||||
* | shader: Improve VOTE.VTG stub | FernandoS27 | 2021-07-23 | 7 | -4/+147 |
| | |||||
* | shader: Mark SSBOs as written when they are | FernandoS27 | 2021-07-23 | 2 | -2/+30 |
| | |||||
* | shader: Implement ViewportIndex | FernandoS27 | 2021-07-23 | 7 | -2/+32 |
| | |||||
* | shader: Stub TLD4's PTP when it isn't constant | FernandoS27 | 2021-07-23 | 1 | -1/+2 |
| | |||||
* | shader: Stub VOTE.VTG | FernandoS27 | 2021-07-23 | 4 | -4/+15 |
| | |||||
* | shader: Fold composite extract | FernandoS27 | 2021-07-23 | 1 | -0/+62 |
| | |||||
* | shader: Fold comparisons and Pack/Unpack16 | FernandoS27 | 2021-07-23 | 1 | -1/+41 |
| | |||||
* | shader: Fix branches to visited virtual blocks | ReinUsesLisp | 2021-07-23 | 2 | -0/+12 |
| | |||||
* | shader: Fix dependency on identity removal pass | ReinUsesLisp | 2021-07-23 | 2 | -3/+8 |
| | |||||
* | shader: Fix constant propagation to use reverse post order | ReinUsesLisp | 2021-07-23 | 1 | -1/+2 |
| | |||||
* | shader: Implement LDG .U.128 as .128 | ReinUsesLisp | 2021-07-23 | 1 | -3/+2 |
| | |||||
* | shader: Unroll "using enum" for opcode declarations | ReinUsesLisp | 2021-07-23 | 1 | -1/+27 |
| | |||||
* | spirv: Remove unnecesary variable for clip distances | ReinUsesLisp | 2021-07-23 | 2 | -6/+2 |
| | |||||
* | shader: Implement ClipDistance | FernandoS27 | 2021-07-23 | 5 | -0/+36 |
| | |||||
* | shader: Fix TXD | FernandoS27 | 2021-07-23 | 2 | -2/+2 |
| | |||||
* | shader: Address feedback | FernandoS27 | 2021-07-23 | 4 | -52/+48 |
| | |||||
* | shader: Always pass a lod for TexelFetch | ReinUsesLisp | 2021-07-23 | 3 | -25/+17 |
| | |||||
* | shader: Implement TXD | FernandoS27 | 2021-07-23 | 4 | -10/+183 |
| | |||||
* | shader: Implement ImageGradient | FernandoS27 | 2021-07-23 | 8 | -2/+84 |
| | |||||
* | shader: Implement TMML partially | FernandoS27 | 2021-07-23 | 6 | -13/+137 |
| | |||||
* | shader,spirv: Implement ImageQueryLod. | FernandoS27 | 2021-07-23 | 9 | -1/+38 |
| | |||||
* | shader: Implement TLDS | FernandoS27 | 2021-07-23 | 3 | -4/+253 |
| | |||||
* | shader: Implement TLD | FernandoS27 | 2021-07-23 | 7 | -14/+173 |
| | |||||
* | spirv: Add fixed pipeline point size | ReinUsesLisp | 2021-07-23 | 3 | -1/+8 |
| | |||||
* | shader: Add PointCoord attribute | FernandoS27 | 2021-07-23 | 5 | -0/+16 |
| | |||||
* | shader: Add PointSize attribute | ameerj | 2021-07-23 | 5 | -0/+13 |
| | |||||
* | shader: Store type of phi nodes in flags | ReinUsesLisp | 2021-07-23 | 3 | -2/+11 |
| | | | | This is needed because pseudo-instructions where invalidated. | ||||
* | shader: Fix indirect branches to scheduler instructions | ReinUsesLisp | 2021-07-23 | 3 | -7/+17 |
| | |||||
* | spirv: Fix default output attribute initialization | ReinUsesLisp | 2021-07-23 | 1 | -3/+3 |
| | |||||
* | shader: Add missing new lines | ReinUsesLisp | 2021-07-23 | 1 | -0/+2 |
| | |||||
* | shader: Implement FSWZADD | ameerj | 2021-07-23 | 14 | -4/+87 |
| | |||||
* | shader: Implement BRX | FernandoS27 | 2021-07-23 | 20 | -47/+388 |
| | |||||
* | shader: Fix alignment checks on RZ | ReinUsesLisp | 2021-07-23 | 1 | -1/+1 |
| | |||||
* | shader: Implement I2I CC | ameerj | 2021-07-23 | 3 | -24/+45 |
| | |||||
* | shader: Implement I2I SAT | ameerj | 2021-07-23 | 6 | -10/+52 |
| | |||||
* | shader: Fix ISCADD logic for PO/CC | ameerj | 2021-07-23 | 1 | -7/+8 |
| | |||||
* | shader: Implement LDS, STS, LDL, and STS and use SPIR-V 1.4 when available | ReinUsesLisp | 2021-07-23 | 17 | -17/+626 |
| | |||||
* | shader: Implement ISCADD CC | ameerj | 2021-07-23 | 1 | -1/+4 |
| | |||||
* | shader: Implement VMAD, VMNMX, VSETP | ameerj | 2021-07-23 | 9 | -23/+319 |
| | |||||
* | shader: Add missing I2I exception when CC is used | ReinUsesLisp | 2021-07-23 | 1 | -0/+4 |
| | |||||
* | shader: Better interpolation and disabled attributes support | ReinUsesLisp | 2021-07-23 | 7 | -23/+96 |
| | |||||
* | spirv: Remove dependencies on Environment when generating SPIR-V | ReinUsesLisp | 2021-07-23 | 4 | -9/+12 |
| | |||||
* | shader: Implement front face | ReinUsesLisp | 2021-07-23 | 5 | -0/+12 |
| | |||||
* | shader: Fix structured control flow on KIL instructions | ReinUsesLisp | 2021-07-23 | 2 | -3/+7 |
| | | | | | This could potentially leave unvisited blocks, leading to illegal phi nodes. | ||||
* | shader: Fix TXQ | FernandoS27 | 2021-07-23 | 1 | -1/+1 |
| | |||||
* | shader: Implement TXQ and fix FragDepth | ReinUsesLisp | 2021-07-23 | 14 | -21/+172 |
| | |||||
* | shader: Refactor PTP and other minor changes | ReinUsesLisp | 2021-07-23 | 14 | -123/+67 |
| | |||||
* | shader: Add IR opcode for ImageFetch | FernandoS27 | 2021-07-23 | 7 | -5/+55 |
| | |||||
* | shader: Implement TLD4.PTP | FernandoS27 | 2021-07-23 | 15 | -28/+111 |
| | |||||
* | shader: Fix Array Indices in TEX/TLD4 | FernandoS27 | 2021-07-23 | 2 | -6/+6 |
| | |||||
* | shader: Implement FragDepth | FernandoS27 | 2021-07-23 | 2 | -1/+7 |
| | |||||
* | shader: Implement TLD4S. | FernandoS27 | 2021-07-23 | 3 | -4/+134 |
| | |||||
* | shader: Implement TLD4 and TLD4_B | FernandoS27 | 2021-07-23 | 13 | -11/+315 |
| | |||||
* | shader: Implement SHFL | ameerj | 2021-07-23 | 16 | -69/+284 |
| | |||||
* | shader: Track first bindless argument instead of the instruction itself | ReinUsesLisp | 2021-07-23 | 1 | -1/+1 |
| | |||||
* | shader: Properly insert Prologue instruction | ReinUsesLisp | 2021-07-23 | 1 | -1/+2 |
| | |||||
* | shader: Minor style nits | ReinUsesLisp | 2021-07-23 | 1 | -2/+4 |
| | |||||
* | shader: Fix F2I | FernandoS27 | 2021-07-23 | 10 | -9/+147 |
| | |||||
* | shader: Implement NDC [-1, 1], attribute types and default varying initialization | ReinUsesLisp | 2021-07-23 | 12 | -40/+149 |
| | |||||
* | shader: Fix use-after-free bug in object_pool | ReinUsesLisp | 2021-07-23 | 1 | -3/+3 |
| | |||||
* | shader: Implement VOTE | ameerj | 2021-07-23 | 14 | -5/+167 |
| | |||||
* | shader: Fix TEX mask | ReinUsesLisp | 2021-07-23 | 1 | -1/+3 |
| | |||||
* | vk_pipeline_cache: Add pipeline cache | ReinUsesLisp | 2021-07-23 | 4 | -8/+15 |
| | |||||
* | shader: Fold interpolation multiplications | ReinUsesLisp | 2021-07-23 | 1 | -0/+34 |
| | |||||
* | shader: Better but still partial interpolation support | ReinUsesLisp | 2021-07-23 | 1 | -5/+7 |
| | |||||
* | shader: Implement DMNMX, DSET, DSETP | ameerj | 2021-07-23 | 15 | -59/+208 |
| | |||||
* | shader: Implement FADD32I | FernandoS27 | 2021-07-23 | 1 | -2/+15 |
| | |||||
* | shader: Implement F2F | FernandoS27 | 2021-07-23 | 6 | -20/+192 |
| | |||||
* | shader: Add missing fp64 usage flags | ReinUsesLisp | 2021-07-23 | 1 | -0/+34 |
| | |||||
* | shader: Implement DMUL and DFMA | ameerj | 2021-07-23 | 8 | -30/+111 |
| | | | | Also add a missing const on DADD | ||||
* | shader: Add FP64 register load/store helpers | ameerj | 2021-07-23 | 3 | -21/+24 |
| | |||||
* | shader: Add support for fp16 comparisons and misc fixes | ReinUsesLisp | 2021-07-23 | 11 | -14/+56 |
| | |||||
* | shader: Fix floating point comparison for FP16 | FernandoS27 | 2021-07-23 | 5 | -32/+56 |
| | |||||
* | shader: Implement HSETP2 | FernandoS27 | 2021-07-23 | 3 | -12/+117 |
| | |||||
* | shader: Implement HSET2 | FernandoS27 | 2021-07-23 | 5 | -14/+119 |
| | |||||
* | shader: Implement HMUL2 | FernandoS27 | 2021-07-23 | 3 | -16/+144 |
| | |||||
* | shader: Implement HFMA2 | FernandoS27 | 2021-07-23 | 5 | -20/+192 |
| | |||||
* | spirv: Implement VertexId and InstanceId, refactor code | ReinUsesLisp | 2021-07-23 | 9 | -144/+243 |
| | |||||
* | shader: Refactor half floating instructions | FernandoS27 | 2021-07-23 | 4 | -58/+84 |
| | |||||
* | shader: Implement I2F | ReinUsesLisp | 2021-07-23 | 16 | -69/+427 |
| | |||||
* | shader: Implement ISCADD (imm) | ReinUsesLisp | 2021-07-23 | 1 | -2/+2 |
| | |||||
* | shader: Implement LOP32I | ReinUsesLisp | 2021-07-23 | 2 | -18/+45 |
| | |||||
* | shader: Add partial rasterizer integration | ReinUsesLisp | 2021-07-23 | 34 | -156/+629 |
| | |||||
* | shader: Implement DADD | ameerj | 2021-07-23 | 8 | -14/+132 |
| | |||||
* | shader: Implement CSET and CSETP | ameerj | 2021-07-23 | 6 | -15/+114 |
| | |||||
* | shader: Reorder phi nodes when redefined as undefined opcodes | ReinUsesLisp | 2021-07-23 | 1 | -1/+9 |
| | |||||
* | shader: Fix instruction transitions in and out of Phi | ReinUsesLisp | 2021-07-23 | 1 | -9/+11 |
| | |||||
* | shader: Implement FSET and FSETP | ameerj | 2021-07-23 | 9 | -94/+204 |
| | | | | Also fix oversight with adding SignedZeroInfNanPreserve execution mode. | ||||
* | shader: Implement TEXS | ReinUsesLisp | 2021-07-23 | 8 | -7/+287 |
| | |||||
* | shader: Implement CAL inlining function calls | ReinUsesLisp | 2021-07-23 | 24 | -330/+286 |
| | |||||
* | spirv: Add SignedZeroInfNanPreserve logic | ameerj | 2021-07-23 | 2 | -0/+8 |
| | |||||
* | shader: Implement FMNMX | ameerj | 2021-07-23 | 8 | -25/+101 |
| | | | | And add a const in FCMP | ||||
* | shader: Fix rebase issue | ReinUsesLisp | 2021-07-23 | 1 | -1/+0 |
| | |||||
* | shader: Implement FCMP | ameerj | 2021-07-23 | 9 | -50/+203 |
| | | | | still need to configure some settings for NV denorm flush and intel NaN | ||||
* | shader: Partial implementation of LDC | ReinUsesLisp | 2021-07-23 | 16 | -50/+405 |
| | |||||
* | shader: Initial support for textures and TEX | ReinUsesLisp | 2021-07-23 | 29 | -341/+1378 |
| | |||||
* | shader: Implement R2P | ameerj | 2021-07-23 | 8 | -15/+88 |
| | |||||
* | shader: Implement SHF | ameerj | 2021-07-23 | 8 | -31/+119 |
| | |||||
* | shader: Implement LEA | ameerj | 2021-07-23 | 9 | -29/+136 |
| | |||||
* | shader: Deduplicate HADD2 code | ReinUsesLisp | 2021-07-23 | 1 | -19/+16 |
| | |||||
* | shader: Implement I2I | ameerj | 2021-07-23 | 3 | -12/+100 |
| | |||||
* | shader: Implement HADD2 | ReinUsesLisp | 2021-07-23 | 12 | -42/+400 |
| | |||||
* | shader: Implement LOP and LOP3 | ameerj | 2021-07-23 | 8 | -31/+227 |
| | |||||
* | shader: Implement IADD3 | ameerj | 2021-07-23 | 3 | -12/+104 |
| | |||||
* | shader: Implement PSETP | ameerj | 2021-07-23 | 4 | -5/+40 |
| | |||||
* | Implement PSET, refactor common comparison funcs | ameerj | 2021-07-23 | 9 | -101/+88 |
| | |||||
* | shader: Implement FLO | ameerj | 2021-07-23 | 8 | -18/+75 |
| | |||||
* | shader: Implement ISET, add common_funcs | ameerj | 2021-07-23 | 8 | -50/+150 |
| | |||||
* | shader: Make IMNMX, SHR, SEL stylistically more consistent | ameerj | 2021-07-23 | 3 | -5/+5 |
| | |||||
* | shader: Implement ICMP | ameerj | 2021-07-23 | 3 | -16/+84 |
| | |||||
* | shader: Implement IMNMX | ameerj | 2021-07-23 | 8 | -12/+105 |
| | |||||
* | shader: Implement BFI | ameerj | 2021-07-23 | 3 | -16/+57 |
| | |||||
* | shader: Implement BFE | ameerj | 2021-07-23 | 3 | -12/+67 |
| | |||||
* | shader: Implement POPC | ameerj | 2021-07-23 | 8 | -12/+59 |
| | |||||
* | shader: Implement SHR | ameerj | 2021-07-23 | 8 | -18/+80 |
| | |||||
* | shader: Implement SEL | ameerj | 2021-07-23 | 4 | -16/+53 |
| | |||||
* | spirv: Move phi arguments emit to a separate function | ReinUsesLisp | 2021-07-23 | 1 | -27/+27 |
| | |||||
* | shader: Avoid infinite recursion when tracking global memory | ReinUsesLisp | 2021-07-23 | 1 | -5/+26 |
| | |||||
* | shader: Fix conditional execution of exit instructions | ReinUsesLisp | 2021-07-23 | 2 | -5/+6 |
| | |||||
* | spirv: Add support for self-referencing phi nodes | ReinUsesLisp | 2021-07-23 | 1 | -3/+10 |
| | |||||
* | shader: Fix control flow | ReinUsesLisp | 2021-07-23 | 8 | -20/+39 |
| | |||||
* | shader: Implement more of XMAD and FFMA32I and fix XMAD.CBCC | ReinUsesLisp | 2021-07-23 | 5 | -28/+76 |
| | |||||
* | shader: FMUL, select, RRO, and MUFU fixes | ReinUsesLisp | 2021-07-23 | 18 | -119/+507 |
| | |||||
* | shader: Fix MOV(reg), add SHL variants and emit neg and abs instructions | ReinUsesLisp | 2021-07-23 | 4 | -11/+11 |
| | |||||
* | spirv: Fixes and Intel specific workarounds | ReinUsesLisp | 2021-07-23 | 10 | -32/+43 |
| | |||||
* | shader: Rename, implement FADD.SAT and P2R (imm) | ReinUsesLisp | 2021-07-23 | 17 | -125/+211 |
| | |||||
* | shader: Add denorm flush support | ReinUsesLisp | 2021-07-23 | 15 | -60/+210 |
| | |||||
* | spirv: Add lower fp16 to fp32 pass | ReinUsesLisp | 2021-07-23 | 28 | -276/+465 |
| | |||||
* | shader: Primitive Vulkan integration | ReinUsesLisp | 2021-07-23 | 28 | -498/+573 |
| | |||||
* | shader: Add XMAD multiplication folding optimization | ReinUsesLisp | 2021-07-23 | 1 | -5/+77 |
| | |||||
* | shader: Simplify ISCADD | ReinUsesLisp | 2021-07-23 | 1 | -6/+1 |
| | |||||
* | shader: Add utility to resolve identities on a value | ReinUsesLisp | 2021-07-23 | 2 | -0/+8 |
| | |||||
* | spirv: Implement EmitIdentity | ReinUsesLisp | 2021-07-23 | 2 | -3/+3 |
| | |||||
* | spirv: Initial bindings support | ReinUsesLisp | 2021-07-23 | 22 | -292/+671 |
| | |||||
* | shader: Improve object pool | ReinUsesLisp | 2021-07-23 | 3 | -50/+66 |
| | |||||
* | shader: Fix tracking | ReinUsesLisp | 2021-07-23 | 1 | -50/+72 |
| | |||||
* | shader: Add support for forward declarations | ReinUsesLisp | 2021-07-23 | 10 | -68/+79 |
| | |||||
* | shader: Support SSA loops on IR | ReinUsesLisp | 2021-07-23 | 12 | -46/+150 |
| | |||||
* | shader: Misc fixes | ReinUsesLisp | 2021-07-23 | 10 | -89/+104 |
| | |||||
* | shader: Initial implementation of an AST | ReinUsesLisp | 2021-07-23 | 32 | -589/+1345 |
| | |||||
* | spirv: Initial SPIR-V support | ReinUsesLisp | 2021-07-23 | 18 | -34/+1400 |
| | |||||
* | shader: Better constant folding | ReinUsesLisp | 2021-07-23 | 2 | -13/+48 |
| | |||||
* | shader: Properly store phi on Inst | ReinUsesLisp | 2021-07-23 | 6 | -75/+132 |
| | |||||
* | shader: Add pools and rename files | ReinUsesLisp | 2021-07-23 | 30 | -108/+255 |
| | |||||
* | shader: Make typed IR | ReinUsesLisp | 2021-07-23 | 19 | -269/+495 |
| | |||||
* | shader: Remove illegal character in SSA pass | ReinUsesLisp | 2021-07-23 | 1 | -1/+1 |
| | |||||
* | shader: Constant propagation and global memory to storage buffer | ReinUsesLisp | 2021-07-23 | 17 | -63/+652 |
| | |||||
* | shader: Initial instruction support | ReinUsesLisp | 2021-07-23 | 28 | -334/+1450 |
| | |||||
* | shader: SSA and dominance | ReinUsesLisp | 2021-07-23 | 24 | -77/+570 |
| | |||||
* | shader: Initial recompiler work | ReinUsesLisp | 2021-07-23 | 56 | -0/+7060 |