[PATCH] D150851: [LoopVectorize] Vectorize select-cmp reduction pattern for increasing integer induction variable

[Mel Chen via Phabricator via llvm-commits]

Mel-Chen added a comment.

In D150851#4535194 <https://reviews.llvm.org/D150851#4535194>, @artagnon wrote:

> Thanks for the detailed thoughts, Mel! My initial reaction is that (2) is probably unworkable due to the fallout, and that (3) isn't general enough. I'm leaning towards trying out (1) -- I'll see how that works out over the next few days.

Sounds good! Before you start implementing, I suggest observing the benchmarks to gather more cases. We choose option 3 because, based on our available benchmarks, all the opportunities for the FindLastIV pattern arise from induction variables that determine the branch condition at the loop latch. Therefore, we decided to go with option 3, as it is an implementation that is not too challenging and provides effective results


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D150851/new/

https://reviews.llvm.org/D150851