[PATCH] D40984: [InstCombine] canonicalize shifty abs(): ashr+add+xor --> cmp+neg+sel

[Hal Finkel via Phabricator via llvm-commits]

hfinkel accepted this revision.
hfinkel added a comment.
This revision is now accepted and ready to land.

LGTM (please take a quick look at CodeGen for other non-X86 targets (e.g., PowerPC, AArch64) and make sure that looks reasonable too).


https://reviews.llvm.org/D40984