[PATCH] D41353: [InstCombine] Adjusting bswap pattern to the new masked shl canonization

[Sanjay Patel via Phabricator via llvm-commits]

spatel added a comment.

If I understand the relationship correctly between the patches, this patch would move an existing failure of bswap matching to a different place. Shouldn't we improve the matching instead? Ie, I was expecting an independent patch from https://reviews.llvm.org/D41233 here.

I added tests at https://reviews.llvm.org/rL322206 to demonstrate.


Repository:
  rL LLVM

https://reviews.llvm.org/D41353