[PATCH] D140493: [SROA] Support promotion in presence of variably-indexed loads

[Sergei Barannikov via Phabricator via llvm-commits]

barannikov88 added a comment.

Could ScalarEvolution + SCEVExpander help with computing / materializing the offset?
It is mainly used for loop transformations, but it should be possible to use it for linear code as well.
One caveat: ScalarEvolution does not peek through addrspacecasts (is it safe to ignore them anyway?).


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D140493/new/

https://reviews.llvm.org/D140493