[PATCH] D41436: [X86][AVX512] Enable variable shuffle combining by default on AVX512 targets
Zvi Rackover via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Thu Dec 21 03:46:18 PST 2017
zvi added a comment.
In https://reviews.llvm.org/D41436#961881, @RKSimon wrote:
> @gadi.haber This change means that KNL will be more aggressive with shuffle combining as well - is that OK?
See my reply here, https://reviews.llvm.org/D41323#961494. I don't know of any reason why applying this for KNL would not be beneficial. @gadi.haber, do you have any means of getting an answer to this?
Repository:
rL LLVM
https://reviews.llvm.org/D41436
More information about the llvm-commits
mailing list