[PATCH] D79164: [CostModel] getCFInstrCost
Sam Parker via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Fri Jul 3 00:29:49 PDT 2020
samparker added a comment.
I suspect that the getCFInstr maybe used in an odd way by the vectorizer for these changes to make such a difference. I'm suspect that the vectorizer is multiplying the cost of it, where 0 or 1 would make quite a difference, but where it just shouldn't be... hopefully the cost of a branch and phi doesn't change that much after vectorization / scalarization.
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D79164/new/
https://reviews.llvm.org/D79164
More information about the llvm-commits
mailing list