[all-commits] [llvm/llvm-project] fb1d61: [X86][AVX] Fold concat(ps*lq(x, 32), ps*lq(y, 32)) ->...

[Simon Pilgrim via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: fb1d61b7257ccd5ba0c96bcea78d6516384ce5b6
      https://github.com/llvm/llvm-project/commit/fb1d61b7257ccd5ba0c96bcea78d6516384ce5b6
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2021-05-12 (Wed, 12 May 2021)

  Changed paths:
    M llvm/lib/Target/X86/X86ISelLowering.cpp
    M llvm/test/CodeGen/X86/vec_int_to_fp.ll
    M llvm/test/CodeGen/X86/vector-shift-lshr-256.ll
    M llvm/test/CodeGen/X86/vector-shift-shl-256.ll

  Log Message:
  -----------
  [X86][AVX] Fold concat(ps*lq(x,32),ps*lq(y,32)) -> shuffle(concat(x,y),zero) (PR46621)

On AVX1 targets we can handle v4i64 logical shifts by 32 bits as a pair of v8f32 shuffles with zero.

I was hoping to put this in LowerScalarImmediateShift, but performing that early causes regressions where other instructions were respliting the subvectors.