[all-commits] [llvm/llvm-project] 1590ca: [AMDGPU] Implement moveToVALU for S_CSELECT_B64 (#...

[Jay Foad via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 1590cac4941e278ab10580c8a19a2720a512054c
      https://github.com/llvm/llvm-project/commit/1590cac4941e278ab10580c8a19a2720a512054c
  Author: Jay Foad <jay.foad at amd.com>
  Date:   2023-11-02 (Thu, 02 Nov 2023)

  Changed paths:
    M llvm/lib/Target/AMDGPU/SIISelLowering.cpp
    M llvm/lib/Target/AMDGPU/SIInstrInfo.cpp
    A llvm/test/CodeGen/AMDGPU/bug-cselect-b64.ll
    M llvm/test/CodeGen/AMDGPU/fix-sgpr-copies.mir

  Log Message:
  -----------
  [AMDGPU] Implement moveToVALU for S_CSELECT_B64 (#70352)

moveToVALU previously only handled S_CSELECT_B64 in the trivial case
where it was semantically equivalent to a copy. Implement the general
case using V_CNDMASK_B64_PSEUDO and implement post-RA expansion of
V_CNDMASK_B64_PSEUDO with immediate as well as register operands.