[all-commits] [llvm/llvm-project] 2b731b: AMDGPU: Take care of "tied" operand when removeOpe...

[Changpeng Fang via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 2b731b30a7e7948ba37649ab1af4be542578104a
      https://github.com/llvm/llvm-project/commit/2b731b30a7e7948ba37649ab1af4be542578104a
  Author: Changpeng Fang <Changpeng.Fang at amd.com>
  Date:   2022-07-28 (Thu, 28 Jul 2022)

  Changed paths:
    M llvm/lib/Target/AMDGPU/SIRegisterInfo.cpp
    A llvm/test/CodeGen/AMDGPU/frame-index-elimination-tied-operand.mir
    M llvm/test/CodeGen/AMDGPU/frame-index-elimination.ll

  Log Message:
  -----------
  AMDGPU: Take care of "tied" operand when removeOperand

Summary:
  Flat scratch load of D16 type by default has tied vdst_in operand (with vdst). This should be taken
care of at the time of "removeOperand" in eliminateFrameIndex. Otherwise we will hit an assert saying
"Cannot move tied operands". This patch unties vdst_in before the move, and retie it with vdst afterwards.

Reviewers:
  arsenm, foad

Differential Revision: https://reviews.llvm.org/D130537