[all-commits] [llvm/llvm-project] 90ec51: [X86] combineConcatVectorOps - fold 512-bit concat...

[Simon Pilgrim via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 90ec51a9ab87c4b0a2a145425be8068f6edebeaf
      https://github.com/llvm/llvm-project/commit/90ec51a9ab87c4b0a2a145425be8068f6edebeaf
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2022-11-01 (Tue, 01 Nov 2022)

  Changed paths:
    M llvm/lib/Target/X86/X86ISelLowering.cpp
    M llvm/test/CodeGen/X86/vector-bitreverse.ll

  Log Message:
  -----------
  [X86] combineConcatVectorOps - fold 512-bit concat(GF2P8AFFINEQB(x,y,c),GF2P8AFFINEQB(z,w,c)) -> GF2P8AFFINEQB(concat(x,z),concat(y,w),c)

Now that D137036 has landed, we just need AVX512F support to generate 512-bit GF2P8AFFINEQB ops