lntue wrote: > Right, I'm more interested in being able to use on aarch64 the `_Float128` > type as standardized by C23, than the clang specific `__float128` extension. > Does this PR help get us there? > > EDIT: in fact, if we could skip propagating the clang-specific extension > further, and just go straight to C23 `_Float128`, I'd prefer that.
I would second that we should go straight to C23 `_Float128`. The current landscape for float128 supports in C (clang / gcc) and C++ (clang++, g++) : https://github.com/llvm/llvm-project/pull/78017#discussion_r1476974657 So adding `__float128` to aarch64 does not help much to simplify codes that need to work for both clang & gcc. I haven't tested but I think the picture for float128 on riscv would be similar to aarch64. https://github.com/llvm/llvm-project/pull/85070 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
