[PATCH] D30200: [SLP] Fix for PR31880: shuffle and vectorize repeated scalar ops on extracted elements

[Simon Pilgrim via Phabricator via llvm-commits]

RKSimon added a comment.

@dorit I've implemented most of the x86 single source shuffle costs at https://reviews.llvm.org/rL310632, please can you see if this has improved things for you?


Repository:
  rL LLVM

https://reviews.llvm.org/D30200