H3cJP/citra
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
26,080 Commits 7 Branches 3 Tags
698c854d5bcd5c021e94dea9bea44e6b07e1c53d
Commit Graph
100 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
Ameer J db1d32485e GLSL: Prefer known used cbuf sizes 2023-11-26 23:25:29 -05:00
Ameer JandGitHub 1d11fe00a3 Merge branch 'master' into ssbo-align 2023-11-26 21:08:53 -05:00
Ameer J a595ed499d gl_buffer_cache: Batch vertex/tfb buffer binding 2023-11-19 17:17:16 -05:00
Ameer JandGitHub feb60de5c3 shader_recompiler: Fix spelling of "derivate" (#12067) 2023-11-18 13:39:47 +01:00
Ameer J c67644f1da gl_graphics_pipeline: GLASM: Fix transform feedback attribs buffer mode 
GL_SEPARATE_ATTRIBS only applies when multiple buffers are being used, else GL_INTERLEAVED_ATTRIBS handles the cases for a single buffer with potentially more than one attribute
 2023-11-18 00:44:05 -05:00
Ameer J 9343b81afd shader_recompiler: add byteswap pattern workaround for Nvidia 2023-11-17 22:29:12 -05:00
Ameer J 75c5be55af shader_recompiler: Align SSBO offsets in GlobalMemory functions 2023-10-31 20:14:18 -04:00
Ameer J 735612c9b3 buffer_cache: Apply storage buffer alignment only to the offset 2023-10-31 20:10:54 -04:00
Ameer JandBilly Laws 7d34800531 shader_recompiler: Align SSBO offsets to meet host requirements 
Co-Authored-By: Billy Laws <blaws05@gmail.com>
 2023-10-31 20:10:54 -04:00
Ameer J 9e4d606c4c nvidia_flags: Enable GL Threaded optimizations 2023-10-28 21:26:22 -04:00
Ameer J cfe73af6f2 emit_glsl_warp: Fix shfl_in_bounds conditional 2023-10-22 00:45:23 -04:00
Ameer J d1c878fb41 gl_texture_cache: Enable async downloads 2023-08-13 23:17:59 -05:00
Ameer J 9b112107e2 gl_buffer_cache: Enable async downloads 2023-08-13 23:17:54 -05:00
Ameer J 48b87d64de gl_staging_buffer_pool: Refactor allocation variables into a struct 2023-08-13 23:17:47 -05:00
Ameer J f9ef721ca6 gl_graphics_pipeline: Fix GLASM storage buffer detection 2023-08-13 17:06:45 -04:00
Ameer J c34ed4bbd8 gl_graphics_pipeline: GLASM: Fix transform feedback with multiple buffers 2023-08-13 16:50:01 -04:00
Ameer J 5c25712af9 flatten color_values 2023-08-09 18:45:52 -04:00
Ameer J 0f7220c9c8 flatten encoding_values 2023-08-09 18:38:37 -04:00
Ameer J 71857e889e flatten result vector 2023-08-09 18:34:57 -04:00
Ameer J 70f8ffb787 GetUnquantizedWeightVector 2023-08-09 17:45:39 -04:00
Ameer J 9058486b9b Revert "HACK: Avoid swizzling and reuploading ASTC image every frame" 
This reverts commit b18c1fb1bb.
 2023-08-06 14:55:05 -04:00
Ameer J b18c1fb1bb HACK: Avoid swizzling and reuploading ASTC image every frame 2023-08-06 14:54:58 -04:00
Ameer J 913803bf65 Compute Replicate 2023-08-06 14:54:58 -04:00
Ameer J 31a0cff036 minor 2023-08-06 14:54:58 -04:00
Ameer J b36e645fee undo uint 2023-08-06 14:54:58 -04:00
Ameer J 8ce158bce6 Revert "vulkan dims specialization" 
This reverts commit e6243058f2269bd79ac8479d58e55feec2611e9d.
 2023-08-06 14:54:58 -04:00
ameerj 5a78b35b1a vulkan dims specialization 2023-08-06 14:54:58 -04:00
Ameer J 7a0d7e7668 small_block opt 2023-08-06 14:54:58 -04:00
Ameer J fd2051b401 remove TexelWeightParams 2023-08-06 14:54:57 -04:00
Ameer J 75ac7845ce error/void extent funcs 2023-08-06 14:54:57 -04:00
Ameer J 441b847107 more packing 2023-08-06 14:54:57 -04:00
Ameer J f2cf81e0b6 Revert "uint result index" 
This reverts commit 0e978786b5a8e7382005d8b1e16cfa12f3eeb775.
 2023-08-06 14:54:57 -04:00
Ameer J f41fb3ec0b Revert "bfe instead of mod" 
This reverts commit 86006a3b09e8a8c17d2ade61be76736a79e3f58a.
 2023-08-06 14:54:57 -04:00
Ameer J 553dd3e120 Revert "global endpoints" 
This reverts commit d8f5bfd1df2b7469ef6abcee182aa110602d1751.
 2023-08-06 14:54:57 -04:00
Ameer J c077e467c4 global endpoints 2023-08-06 14:54:57 -04:00
Ameer J 5c16559694 bfe instead of mod 2023-08-06 14:54:57 -04:00
Ameer J 6b0b584eba uint result index 2023-08-06 14:54:57 -04:00
Ameer J 05ee37a1f0 amd opts 2023-08-06 14:54:57 -04:00
Ameer J 3494fce864 gl 2023-08-06 14:54:57 -04:00
Ameer J 5248fa926d const, pack result_vector and replicate tables, 
undo amd opts
 2023-08-06 14:54:57 -04:00
Ameer J 998246efc2 minor redundancy cleanup 2023-08-06 14:54:57 -04:00
Ameer J d17a51bc59 extractbits robustness 2023-08-06 14:54:57 -04:00
Ameer J 0078e5a338 reuse vectors memory 2023-08-06 14:54:57 -04:00
Ameer J b8ca47e094 EncodingData pack 2023-08-06 14:54:57 -04:00
Ameer J 27c8bb9615 flattening 2023-08-06 14:54:57 -04:00
Ameer J ac09cc3504 weights refactor 2023-08-06 14:54:57 -04:00
Ameer J 6ff65abd62 params.max_weight 2023-08-06 14:54:57 -04:00
Ameer J e0c59c7b0b skip bits 2023-08-06 14:54:57 -04:00
Ameer J 7ef879b296 restrict 2023-08-06 14:54:57 -04:00
Ameer J cd80cbc420 vulkan_device: Fix subgroup_size_control detection on Vulkan 1.3 2023-08-02 20:45:03 -04:00
Ameer J c7c44dc238 vulkan_device: Fix VK_EXT_subgroup_size_control detection 2023-08-02 19:25:14 -04:00
ameerj 4f160633d3 OpenGL: Limit lmem warmup to NVIDIA 
🐸
 2023-06-25 19:06:51 -04:00
ameerj 405eae3734 shaders: Track local memory usage 2023-06-25 18:59:33 -04:00
ameerj b198339580 emit_glasm: Fix lmem size computation 2023-06-25 18:43:52 -04:00
ameerj 82107b33a2 OpenGL: Add Local Memory warmup shader 2023-06-25 18:43:23 -04:00
ameerj 1fc47361a1 texture_cache: Fix incorrect logic for AccelerateDMA 2023-06-02 18:07:52 -04:00
ameerj ea2e155b0b gl_texture_cache: Fix ASTC CPU decoding with compression disabled 
gl_format was incorrectly being overwritten when compression was disabled
 2023-05-28 13:14:51 -04:00
ameerj cb0a410907 gl_staging_buffers: Optimization to reduce fence waiting 2023-05-28 00:38:47 -04:00
ameerj 642c14f0c7 OpenGL: Make use of persistent buffer maps in buffer cache downloads 
Persistent buffer maps were already used by the texture cache, this extends their usage for the buffer cache.

In my testing, using the memory maps for uploads was slower than the existing "ImmediateUpload" path, so the memory map usage is limited to downloads for the time being.
 2023-05-28 00:38:46 -04:00
ameerj fee91096ca microprofile: Avoid crashing due to OOB stack pos 2023-05-27 22:24:22 -04:00
ameerj 03137086db OpenGL: Prefer glClientWaitSync for OGLSync objects 
At least on Nvidia, glClientWaitSync with a timeout of 0 (non-blocking) is faster than glGetSynciv of GL_SYNC_STATUS.
 2023-03-08 20:29:25 -05:00
ameerj 268942c8fe gl_rasterizer: Implement AccelerateDMA DmaBufferImageCopy 2023-03-06 22:57:52 -05:00
ameerj 6b9cc0ed23 Refactor AccelerateDMA code 2023-03-06 22:57:45 -05:00
ameerj 2ce5bb9bd6 buffer_cache: Add logic for non-NVN storage buffer tracking 2023-02-25 16:24:21 -05:00
ameerj b5bcd8c71b configuration: Add async ASTC decode setting 2023-02-22 18:21:09 -05:00
ameerj 090bc588e5 texture_cache: Add async texture decoding 2023-02-22 00:26:07 -05:00
ameerj 93cf2b3ca8 texture_cache: OpenGL: Implement MSAA uploads and copies 2023-02-11 15:43:07 -05:00
ameerj 5e746da981 kernel: Refactor thread_local variable usage 
On MSVC at least, there seems to be a non-trivial overhead to calling GetHostThreadId().

This slightly reworks the host_thread_id variable to reduce some of the complexity around its usage, along with some small refactors around current_thread and dummy thread
 2023-02-10 20:43:06 -05:00
ameerj eb9f16dce4 buffer_base: Partially revert changes from #9559 
This fixes a regression where Yoshi's Crafted World (and potentially other titles) would enter an infinite loop when GPU Accuracy was set to "Normal"
 2023-02-08 19:37:23 -05:00
ameerj 04139cb3ed glsl_emit_context: Remove redeclarations of gl_SampleID and gl_SampleMask 
These built-ins seem to be available without needing to be declared for fragment shaders, similar i.e. to gl_FragDepth
 2023-02-08 19:34:39 -05:00
ameerj 4678f53463 shader_recompiler/value.h: Remove lingering references to S32 2023-02-04 00:13:47 -05:00
ameerj 979e4d9950 fsp_srv: Copy HLE Read Buffer for OutputAccessLogToSdCard 2023-02-03 00:12:16 -05:00
ameerj 2d2522693e Revert "Merge pull request #9718 from yuzu-emu/revert-9508-hle-ipc-buffer-span" 
This reverts commit 25fc5c0e11, reversing
changes made to af20e25081.
 2023-02-03 00:08:45 -05:00
ameerj 720ff38097 gl_compute_pipeline: Force context flush when loading shader cache 2023-01-29 21:32:12 -05:00
ameerj 01eeda74a6 gl_graphics_pipeline: Force context flush when loading shader cache 2023-01-29 20:26:49 -05:00
ameerj a63e17566a spirv: Fix TXQ with MSAA textures 2023-01-29 13:47:30 -05:00
ameerj a1d8306bfd emit_glasm_image: Fix TXQ with MSAA textures 2023-01-29 13:42:34 -05:00
ameerj 08feba2b56 emit_glsl_image: Implement TXQ with MSAA textures 
Also fixes for texture buffers, which do not have mips eithers.
 2023-01-29 13:31:47 -05:00
ameerj 0d820f2dab texture_pass: Fix texture descriptors comparisons 2023-01-28 21:23:52 -05:00
ameerj 9bb429999e texture_pass: Refactor texture handle retrieval 2023-01-28 21:11:38 -05:00
ameerj 2c2e019a44 shader_recompiler: TXQ: Skip QueryLevels when possible 2023-01-28 16:25:18 -05:00
ameerj c0cedbae94 emit_glsl_image: Fix ImageFetch for MSAA textures 2023-01-28 14:39:27 -05:00
ameerj c2fb7b64ce texture_cache: Adjust image view sizes by MSAA samples 2023-01-28 00:15:29 -05:00
ameerj 0f795603fc glasm: Add MS sampler types 2023-01-27 02:09:18 -05:00
ameerj 5710e90150 glsl: Add MS sampler types 2023-01-27 02:09:17 -05:00
ameerj 93cc6e4d99 shader_recompiler: Remove S32 IR type 
The frontend IR opcodes do not distinguish between signed and unsigned integer types.

Fixes broken shaders when IR validation/graphics debugging is enabled for shaders that used BitCastS32F32
 2023-01-25 22:03:15 -05:00
ameerj b0722591c9 hle_ipc: Use thread_local ReadBuffer 2022-12-28 18:46:54 -05:00
ameerj a1490d77ac hle_ipc: Rename ReadBufferSpan to ReadBuffer 2022-12-28 18:46:54 -05:00
ameerj 59c0f85407 hle_ipc: Rename ReadBuffer to ReadBufferCopy 
Indicates explicitly that a copy is occurring
 2022-12-28 18:46:54 -05:00
ameerj 7c1bf68bb4 bsd: Use std::span for read payloads 
Allows the use of HLERequestContext::ReadBufferSpan
 2022-12-28 18:46:54 -05:00
ameerj f517f82416 nvdrv: Use std::span for inputs 
Allows the use of HLERequestContext::ReadBufferSpan
 2022-12-28 18:46:53 -05:00
ameerj e7032d9e64 hidbus: Use ReadBufferSpan 2022-12-28 18:46:53 -05:00
ameerj 32d01a39b0 nvflinger: Split Parcel class into InputParcel and OutputParcel 
The usages of the Parcel class were already unique to either Read or Write operations.
Avoids needing a vector of the input payload for the InputParcel use-case, instead it can remain as a span.
 2022-12-25 17:19:10 -05:00
ameerj 7ffd624248 service: Use ReadBufferSpan where it is trivial to do so 2022-12-25 17:04:02 -05:00
ameerj 7584d36922 texture_cache: Use Common::ScratchBuffer for swizzle buffers 2022-12-25 15:47:41 -05:00
ameerj 1209d428f1 texture_cache: Use pre-allocated buffer for texture downloads 2022-12-25 15:38:36 -05:00
ameerj c448b3af2f texture_cache: Use pre-allocated buffer for texture uploads 2022-12-25 15:38:36 -05:00
ameerj fbc375f0de fsp_srv: Use ReadBufferSpan 2022-12-25 13:21:15 -05:00
ameerj 9349f06963 hle_ipc: Add ReadBufferSpan function 
Returns a std::span to the buffer address, rather than create a copy of the memory into a std::vector
 2022-12-25 13:20:59 -05:00
ameerj c6590ad07b scratch_buffer: Explicitly defing resize and resize_destructive functions 
resize keeps previous data intact when the buffer grows
resize_destructive destroys the previous data when the buffer grows
 2022-12-19 22:40:50 -05:00