[PATCH] D61250: [PowerPC] Forward the scheduling state for PowerPC target
Qing Shan Zhang via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Mon Jun 17 23:38:33 PDT 2019
steven.zhang added a comment.
In D61250#1545597 <https://reviews.llvm.org/D61250#1545597>, @fhahn wrote:
> Would it be possible to add a test where the forwarding kicks in?
It is not easy. We could easily get some tests that have different code sequence if enable the fwd scheduling, however, it is not easy to explain why, and if that makes sense. Because, we have so many heuristic that would affect the scheduling, and they are working together. Anyway, I will try my best to cook one, Thank you for this suggestion!
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D61250/new/
https://reviews.llvm.org/D61250
More information about the llvm-commits
mailing list