[all-commits] [llvm/llvm-project] 57e8f8: [X86][FP16] Fix a bug when Combine the FADD(A, FMA...

[MorningLiu via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 57e8f840b6d33475ca5f023001996ab4bc9035b4
      https://github.com/llvm/llvm-project/commit/57e8f840b6d33475ca5f023001996ab4bc9035b4
  Author: Liu, Chen3 <chen3.liu at intel.com>
  Date:   2021-09-28 (Tue, 28 Sep 2021)

  Changed paths:
    M llvm/lib/Target/X86/X86ISelLowering.cpp
    M llvm/test/CodeGen/X86/avx512fp16-combine-vfmac-fadd.ll
    M llvm/test/CodeGen/X86/avx512fp16-combine-vfmulc-fadd.ll
    M llvm/test/CodeGen/X86/avx512fp16-combine-xor-vfmulc-fadd.ll

  Log Message:
  -----------
  [X86][FP16] Fix a bug when Combine the FADD(A, FMA(B, C, 0)) to FMA(B, C, A).

This bug was introduced by D109953. The operand order of generated FMA
is wrong.

Differential Revision: https://reviews.llvm.org/D110606