[PATCH] D123453: [InstCombine] Fold mul nuw+lshr to a single multiplication when the latter is a factor
Sanjay Patel via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Wed Apr 20 08:46:10 PDT 2022
spatel accepted this revision.
spatel added a comment.
This revision is now accepted and ready to land.
LGTM - see inline remark for minor edit to a code comment.
================
Comment at: llvm/lib/Transforms/InstCombine/InstCombineShifts.cpp:1177
+
+ // Check one-use here to make sure we can save the shift instruction,
+ // avoid replace shift to mul as shift is cheaper than mul in general.
----------------
I would phrase this differently:
// The one-use check is not strictly necessary, but codegen may not be able
// to invert the transform and perf may suffer with an extra mul instruction.
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D123453/new/
https://reviews.llvm.org/D123453
More information about the llvm-commits
mailing list