These should really use v32f32, but were defined as v32i32
due to the lack of the v32f32 type.
Details
Details
Diff Detail
Diff Detail
Paths
| Differential D64667
[AMDGPU] use v32f32 for 3 mfma intrinsics ClosedPublic Authored by rampitec on Jul 12 2019, 2:43 PM.
Details Summary These should really use v32f32, but were defined as v32i32
Diff Detail Event TimelineHerald added subscribers: t-tye, tpr, dstuttard and 6 others. · View Herald TranscriptJul 12 2019, 2:43 PM This revision is now accepted and ready to land.Jul 12 2019, 2:52 PM Closed by commit rL365972: [AMDGPU] use v32f32 for 3 mfma intrinsics (authored by rampitec). · Explain WhyJul 12 2019, 3:41 PM This revision was automatically updated to reflect the committed changes.
Revision Contents
Diff 209599 include/llvm/IR/Intrinsics.td
include/llvm/IR/IntrinsicsAMDGPU.td
lib/Target/AMDGPU/AMDGPUISelLowering.cpp
lib/Target/AMDGPU/SIISelLowering.cpp
lib/Target/AMDGPU/SIInstrInfo.td
lib/Target/AMDGPU/SIInstructions.td
lib/Target/AMDGPU/SIRegisterInfo.td
test/CodeGen/AMDGPU/agpr-register-count.ll
test/CodeGen/AMDGPU/llvm.amdgcn.mfma.ll
test/CodeGen/AMDGPU/spill-agpr.ll
test/CodeGen/AMDGPU/spill-vgpr-to-agpr.ll
|