This improves the type breakdown for some large vectors. For example,
we now get a <4 x s32> and s32 store instead of 5 s32 stores for
<5 x s32>.
Details
Details
Diff Detail
Diff Detail
Paths
| Differential D73118
AMDGPU/GlobalISel: Use more wide vector load/stores ClosedPublic Authored by arsenm on Jan 21 2020, 9:29 AM.
Details
Diff Detail Event Timelinearsenm added a child revision: D73127: AMDGPU/GlobalISel: Widen non-power-of-2 load results.Jan 21 2020, 11:26 AM This revision is now accepted and ready to land.Jan 29 2020, 10:06 AM
Revision Contents
Diff 239348 llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp
llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-load-flat.mir
llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-load-local.mir
llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-load-private.mir
llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-store-global.mir
|