| .. |
|
ControlFlow
|
fix(shader): move the primitive class to request version 19 and compose the ancillary word with render target index 0
|
2026-10-13 13:18:51 +13:00 |
|
IntermediateRepresentation
|
fix(shader): keep direct VGPR accesses in SSA when the window has only relative reads
|
2026-10-13 13:59:35 +13:00 |
|
Optimization
|
fix(shader): read the ancillary render target index as 0
|
2026-10-13 13:23:56 +13:00 |
|
RdnaDecoder
|
Merge pull request #3827 from CouchRiv/fix-shader-combined-cvt-i16-f16-omod
|
2026-10-12 00:39:20 +02:00 |
|
SpirvBackend
|
feat(shader): give reads the masked-select pass keeps the whole pixel ancillary word
|
2026-10-13 13:06:21 +13:00 |
|
Translation
|
fix(shader): keep direct VGPR accesses in SSA when the window has only relative reads
|
2026-10-13 13:59:35 +13:00 |
|
tests
|
fix(shader): handle cache file allocation failures
|
2026-10-11 19:57:45 +04:00 |
|
BdaAbi.hpp
|
fix(agc): note BDA-written pages at full address width and read back only the claimed ones
|
2026-10-11 12:54:02 +05:30 |
|
CMakeLists.txt
|
perf(agc): build the shader pipeline libraries when graphics stages are resolved at registration
|
2026-10-12 09:50:41 +02:00 |
|
CacheKey.hpp
|
fix(recompiler): run vertex programs as one-lane waves without vertex subgroups
|
2026-10-12 08:50:05 -04:00 |
|
CompiledVariant.hpp
|
refactor(shader): remove resource specialization from artifact key
|
2026-10-08 17:26:38 +03:00 |
|
Deserialize.cpp
|
perf(recompiler): keep descriptor heaps uncompacted and read image modes at run time when modules are built at registration
|
2026-10-12 09:50:32 +02:00 |
|
FragmentOutputSpecialization.hpp
|
feat(agc): implement dual-source blending from the MRT1 export
|
2026-10-10 10:29:00 +02:00 |
|
PipelineSpecialization.hpp
|
feat(shader): give reads the masked-select pass keeps the whole pixel ancillary word
|
2026-10-13 13:06:21 +13:00 |
|
Recompiler.cpp
|
feat(shader): give reads the masked-select pass keeps the whole pixel ancillary word
|
2026-10-13 13:06:21 +13:00 |
|
Recompiler.hpp
|
feat(shader): give reads the masked-select pass keeps the whole pixel ancillary word
|
2026-10-13 13:06:21 +13:00 |
|
RuntimeAbi.hpp
|
perf(recompiler): read vertex attribute component selectors at run time when modules are built at registration
|
2026-10-12 09:50:41 +02:00 |
|
ShaderCacheDirectory.hpp
|
Merge pull request #3785 from MrMha/fix/shader-cache-read-allocation
|
2026-10-12 00:01:43 +02:00 |
|
ShaderCacheVersion.cmake
|
perf(shader): hash only the recompiler's inputs into the shader cache version
|
2026-10-04 03:02:24 +13:00 |
|
ShaderDiskCache.cpp
|
feat(shader): give reads the masked-select pass keeps the whole pixel ancillary word
|
2026-10-13 13:06:21 +13:00 |
|
ShaderDiskCache.hpp
|
feat(shader): give reads the masked-select pass keeps the whole pixel ancillary word
|
2026-10-13 13:06:21 +13:00 |
|
ThreadOwned.hpp
|
fix(agc): free per-thread driver state at thread exit without emutls
|
2026-10-04 19:08:54 -05:00 |
|
VertexInputSpecialization.hpp
|
perf(shader): restore prepared vertex inputs + harden buffer specialization
|
2026-10-08 17:28:55 +03:00 |
c3625c4789