[all-commits] [llvm/llvm-project] b22721: [AMDGPU] GCNDPPCombine: don't shrink V_ADD_CO_U32 ...

[Jay Foad via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: b22721f01a580ceed64923f528e1b7f3d66a12a9
      https://github.com/llvm/llvm-project/commit/b22721f01a580ceed64923f528e1b7f3d66a12a9
  Author: Jay Foad <jay.foad at amd.com>
  Date:   2021-04-20 (Tue, 20 Apr 2021)

  Changed paths:
    M llvm/lib/Target/AMDGPU/GCNDPPCombine.cpp
    M llvm/test/CodeGen/AMDGPU/dpp_combine.mir

  Log Message:
  -----------
  [AMDGPU] GCNDPPCombine: don't shrink V_ADD_CO_U32 if carry out is used

Don't shrink VOP3 instructions if there are any uses of a carry-out
operand, because the shrunken form of the instruction would write the
carry-out to vcc instead of to a virtual register.

Differential Revision: https://reviews.llvm.org/D100760