https://github.com/andykaylor commented:
I don't think attaching these operand bundles to the existing set of intrinsics is what we want to do. I'd prefer to deprecate the existing intrinsics and define some new scheme that uses these bundles on intrinsics that may be used in other ways. One of the original reasons that we talked about the idea of using operand bundles this way is that we wanted a way to indicate strict FP constraints on vector predication intrinsics without requiring an entire second set of intrinsics. I'd like to see a set of generic floating point intrinsics (possibly the existing non-constrained intrinsics) that could be modified with operand bundles in the way you've implemented here, but I guess we'd need some mechanism to ensure that the operand bundles were respected. https://github.com/llvm/llvm-project/pull/109798 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
