[PATCH] D148713: [LegalizeTypes][VP] Widen load/store of fixed length vectors to VP ops

[Fraser Cormack via Phabricator via llvm-commits]

frasercrmck added a comment.

Sorry, dropped the ball on this. The code looks good to me.

I wonder if someone with a better sense of architectural performance can say this is likely to be faster? Fetching/writing less memory must be good, right? On top of that we're doing fewer vector extracts so fewer instructions in general.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D148713/new/

https://reviews.llvm.org/D148713