[all-commits] [llvm/llvm-project] 879ac4: [AMDGPU] Fix crash in SIOptimizeExecMaskingPreRA

[Jay Foad via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 879ac4108979a617d502aad520b7282f0f29ce56
      https://github.com/llvm/llvm-project/commit/879ac4108979a617d502aad520b7282f0f29ce56
  Author: Jay Foad <jay.foad at amd.com>
  Date:   2022-04-20 (Wed, 20 Apr 2022)

  Changed paths:
    M llvm/lib/Target/AMDGPU/SIInstrInfo.cpp
    A llvm/test/CodeGen/AMDGPU/atomic-optimizer-strict-wqm.ll

  Log Message:
  -----------
  [AMDGPU] Fix crash in SIOptimizeExecMaskingPreRA

When folding a COPY of exec into another COPY, the call to
TII->isOperandLegal would crash because COPYs don't have defined
register classes for their operands.

Differential Revision: https://reviews.llvm.org/D122737