RKSimon requested changes to this revision. RKSimon added a comment. This revision now requires changes to proceed.
I really don't support this change - -ffast-math should be allowed to perform aggressive folds and it makes no sense to make an exception for common intrinsics like these. We should just ensure the intrinsics are suitably documented to warn that they are affected by fast math flags like any other operation. An alternative, if you have specific (and common) use cases, would be to add an opt-in command line switch that disables fast math flags on intrinsics. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D107004/new/ https://reviews.llvm.org/D107004 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
