[PATCH] D56864: [x86] vectorize cast ops in lowering to avoid register file transfers

Fri Jan 18 06:55:23 PST 2019

spatel added a comment.

In D56864#1362813 <https://reviews.llvm.org/D56864#1362813>, @RKSimon wrote:

> We still seem to be missing many x86_64 cases?

In the knownbits tests, there are other problems. We have some other op(s) sitting between the extract and convert. Some of that would be fixed with a change for PR39975:
https://bugs.llvm.org/show_bug.cgi?id=39975
...but we may need more patches before that to avoid yet more regressions.

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D56864/new/

https://reviews.llvm.org/D56864