Files
clang-p2996/llvm/lib/Target/AMDGPU/AMDGPUSubtarget.cpp
Jay Foad 4db7422771 [AMDGPU] Improve zeroesHigh16BitsOfDest for GFX9 legacy opcodes
Pseudos like V_MAD_U16 and V_FMA_F16 map down to what GFX9 calls
v_mad_legacy_u16 and v_fma_legacy_f16, which are documented to have the
same zeroing behaviour as on GFX8.

Differential Revision: https://reviews.llvm.org/D115729
2021-12-15 13:14:48 +00:00

38 KiB