[PATCH] D96022: [LoopVectorize] NFC: Split off clamping from computeFeasibleUserVF into its own function.
Sander de Smalen via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Thu Feb 4 05:16:12 PST 2021
sdesmalen created this revision.
Herald added a subscriber: hiraditya.
sdesmalen requested review of this revision.
Herald added a project: LLVM.
This patch is NFC and cleans up `computeFeasibleUserVF()` by moving the clamping
into a separate function, that can be reused in multiple places, including
computeFeasibleMaxVF. This helps prepare the code to operate on scalable
VFs.
This patch is a preparatory patch with the ultimate goal of making
`computeMaxVF()` return both a max fixed VF and a max scalable VF,
so that `selectVectorizationFactor()` can pick the most cost-effective
vectorization factor.
Repository:
rG LLVM Github Monorepo
https://reviews.llvm.org/D96022
Files:
llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
llvm/test/Transforms/LoopVectorize/AArch64/scalable-vf-hint.ll
-------------- next part --------------
A non-text attachment was scrubbed...
Name: D96022.321391.patch
Type: text/x-patch
Size: 11064 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20210204/dd26e93b/attachment.bin>
More information about the llvm-commits
mailing list