[PATCH] D130537: AMDGPU: Take care of "tied" operand when removeOperand

Changpeng Fang via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Wed Jul 27 14:57:26 PDT 2022


cfang marked an inline comment as done.
cfang added inline comments.


================
Comment at: llvm/test/CodeGen/AMDGPU/frame-index-elimination-tied-operand.ll:2
+; RUN: llc -march=amdgcn -mcpu=gfx1100 < %s | FileCheck --check-prefix=GCN %s
+
+define amdgpu_kernel void @tied_operand_test() local_unnamed_addr {
----------------
arsenm wrote:
> Can you merge this test with one of the existing frame elimination tests?
There is a difficulty in the merge. This test is only for gfx1100, and we should not check for any other targets. And the existing tests of frame index elimination never check for gfx1100. It is beyond this work to do additional checks for gfx1100 for other works in frame index elimination.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D130537/new/

https://reviews.llvm.org/D130537



More information about the llvm-commits mailing list