[all-commits] [llvm/llvm-project] 080cb8: [X86][AVX] Narrow VPBROADCASTQ->VPBROADCASTD if we...

[Simon Pilgrim via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 080cb83e52c3059a62bbb87142cbcf3f68c14ba2
      https://github.com/llvm/llvm-project/commit/080cb83e52c3059a62bbb87142cbcf3f68c14ba2
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2021-03-23 (Tue, 23 Mar 2021)

  Changed paths:
    M llvm/lib/Target/X86/X86ISelLowering.cpp
    M llvm/test/CodeGen/X86/bitcast-int-to-vector-bool-sext.ll
    M llvm/test/CodeGen/X86/bitcast-int-to-vector-bool-zext.ll

  Log Message:
  -----------
  [X86][AVX] Narrow VPBROADCASTQ->VPBROADCASTD if we don't need the upper bits.

Helps fix cases where we've splatted smaller types to a wider vector element type without needing the upper bits.

Avoid this on AVX512 targets as that can affect broadcast folding.