Log - f8b51c5f90c60aceedef9ed185eeac58f501075a - toolchain/llvm-project

c40d9f2 AMDGPU/GCN: Bring processors in sync with AMDGPUUsage by Konstantin Zhuravlyov · 8 years ago
5f7f32c [AMDGPU] SDWA: add support for PRESERVE into SDWA peephole. by Sam Kolton · 8 years ago
70b9282 AMDGPU: Fix -enable-var-scope violations by Matt Arsenault · 8 years ago
b81495d AMDGPU: Match load d16 hi instructions by Matt Arsenault · 8 years ago
8728c5f AMDGPU: Cleanup subtarget features by Matt Arsenault · 8 years ago
4e309b0 AMDGPU: Start selecting global instructions by Matt Arsenault · 8 years ago
6c29c5a AMDGPU: Allow SIShrinkInstructions to work in non-SSA by Matt Arsenault · 8 years ago
982aee6 [AMDGPU] Switch scalarize global loads ON by default by Alexander Timofeev · 8 years ago
e4a7413 Revert r307026, "[AMDGPU] Switch scalarize global loads ON by default" by NAKAMURA Takumi · 8 years ago
ea7f08b [AMDGPU] Switch scalarize global loads ON by default by Alexander Timofeev · 8 years ago
0330660 [AMDGPU] Untangle SDWA pass from SIShrinkInstructions by Stanislav Mekhanoshin · 9 years ago
9fa1696 [AMDGPU] Resubmit SDWA peephole: enable by default by Sam Kolton · 9 years ago
d4f70c7 Revert r299536. [AMDGPU] SDWA peephole: enable by default. by Ivan Krasin · 9 years ago
34e2978 [AMDGPU] SDWA peephole: enable by default by Sam Kolton · 9 years ago
8edfaee AMDGPU: Remove unnecessary ands when f16 is legal by Matt Arsenault · 9 years ago
3dbeefa AMDGPU: Mark all unspecified CC functions in tests as amdgpu_kernel by Matt Arsenault · 9 years ago
86e02ce AMDGPU: Fix unnecessary ands when packing f16 vectors by Matt Arsenault · 9 years ago
eb522e6 AMDGPU: Support v2i16/v2f16 packed operations by Matt Arsenault · 9 years ago
9dba9bd AMDGPU: Use source modifiers with f16->f32 conversions by Matt Arsenault · 9 years ago
c79dc70 AMDGPU: Fix f16 fabs/fneg by Matt Arsenault · 9 years ago