[PATCH] D56057: [X86] Individually simplify both operands of PMULDQ/PMULUDQ using the other entry point of SimplifyDemandedBits that allows the one use check of the root node to be suppressed.

[Craig Topper via Phabricator via llvm-commits]

craig.topper added a comment.

Looking at the vector-reduce-mul.ll changes, I don't think this patch is valid. If the first node has multiple uses we can't propagate the input demanded bits to the next operation down.


Repository:
  rL LLVM

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D56057/new/

https://reviews.llvm.org/D56057