[all-commits] [llvm/llvm-project] c872fa: [AMDGPU] Do not generate S_CMP_LG_U64 on gfx7

Piotr Sobczak via All-commits all-commits at lists.llvm.org
Mon Oct 19 05:45:23 PDT 2020


  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: c872faf6e0913d05ead41975d72eeb9fc085ce2b
      https://github.com/llvm/llvm-project/commit/c872faf6e0913d05ead41975d72eeb9fc085ce2b
  Author: Piotr Sobczak <Piotr.Sobczak at amd.com>
  Date:   2020-10-19 (Mon, 19 Oct 2020)

  Changed paths:
    M llvm/lib/Target/AMDGPU/SIISelLowering.cpp
    M llvm/test/CodeGen/AMDGPU/expand-scalar-carry-out-select-user.ll

  Log Message:
  -----------
  [AMDGPU] Do not generate S_CMP_LG_U64 on gfx7

S_CMP_LG_U64 was added in gfx8 and is guarded by hasScalarCompareEq64().

Rewrite S_CMP_LG_U64 to S_OR_B32 + S_CMP_LG_U32 for targets that
do not support 64-bit scalar compare.

Differential Revision: https://reviews.llvm.org/D89536




More information about the All-commits mailing list