[PATCH] D145301: Add more efficient vector bitcast for AArch64

[Markus Everling via Phabricator via llvm-commits]

Sp00ph added a comment.

Does "vector comparison followed by bitcast" mean that something like this would not get optimized?

  define i16 @cast(<16 x i1> %vec) {
      %ret = bitcast <16 x i1> %vec to i16
      ret i16 %ret
  }

Forgive me if I'm wrong, but I don't see why the preceding comparison is necessary. Couldn't you just do the bit-and trick on any mask vector, regardless of how it was created?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D145301/new/

https://reviews.llvm.org/D145301