[PATCH] D85929: [InstSimplify] Fold min/max intrinsic based on icmp of operands
Sanjay Patel via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Fri Aug 14 11:50:17 PDT 2020
spatel added a comment.
I'm iffy about this. I was hoping that we could either remove the explicit folds or show more wins by using the general icmp simplification.
Based on the test diffs, we're only getting assume improvements, and that doesn't seem worth the effort of full icmp simplify.
It would be nice to show a more meaningful motivating case, but I doubt we will get that until we start canonicalizing to the intrinsics.
Thoughts?
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D85929/new/
https://reviews.llvm.org/D85929
More information about the llvm-commits
mailing list