[PATCH] D103814: [LoopVectorize] Don't use strict reductions when reordering is allowed

[Kerry McLaughlin via Phabricator via llvm-commits]

kmclaughlin created this revision.
kmclaughlin added reviewers: sdesmalen, CarolineConcatto, fhahn, spatel.
Herald added a subscriber: hiraditya.
kmclaughlin requested review of this revision.
Herald added a project: LLVM.
Herald added a subscriber: llvm-commits.

If the `-enable-strict-reductions` flag is set to true, then currently we will
always choose to vectorize the loop with strict in-order reductions. This is
not necessary where we allow the reordering of FP operations, such as
when loop hints are passed via metadata.

This patch moves useOrderedReductions so that we can also check whether
loop hints allow reordering, in which case we should use the default
behaviour of vectorizing with unordered reductions.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D103814

Files:
  llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
  llvm/test/Transforms/LoopVectorize/AArch64/scalable-strict-fadd.ll
  llvm/test/Transforms/LoopVectorize/AArch64/strict-fadd.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D103814.350282.patch
Type: text/x-patch
Size: 11307 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20210607/bd1c9306/attachment.bin>