nikic added a comment. I think our abs intrinsic support is already sufficient to start canonicalizing to the intrinsic variant (the same is not true for min/max intrinsics). We might want to make the InstCombine change before this one, to make sure we don't lose CSE opportunities between the intrinsic and the expanded forms. Though if that's not a concern for these vector intrinsics, then this LG to me.
Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D87101/new/ https://reviews.llvm.org/D87101 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
