asb added a comment. In D121670#4192158 <https://reviews.llvm.org/D121670#4192158>, @joshua-arch1 wrote:
> It seems that llvm implementation is different from Binutils/GCC. Binutils > didn't regard ntl instructions as aliases of add since encoding space is > reserved for HINT instructions. How does this different implementation approach show up in terms of user-visible behaviour? Looking at the ISA spec, I think the use of the term "reserved" with respect to hint instructions is unfortunate, because I don't think the intended meaning isn't that they are considered a "reserved" encoding as defined in the categories of instructions, as otherwise it would be legal for implementations to trap when encountering a hint, making them impossible to deploy in a backward compatible way. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D121670/new/ https://reviews.llvm.org/D121670 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
