[PATCH] D135869: [AMDGPU][DAG] Only apply trunc/shift combine to 16 bit types
Matt Arsenault via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Fri Oct 14 09:27:06 PDT 2022
arsenm added a comment.
> The shift amount check seems to be wrong. I think the correct condition is ShiftAmt <= (32 - VT.getScalarSizeInBits()) https://alive2.llvm.org/ce/z/uYZ9tq
For right shifts. For left shifts it's size.
GlobalISel also seems to have half ported this combine. For some reason it's only handling shl
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D135869/new/
https://reviews.llvm.org/D135869
More information about the llvm-commits
mailing list