erichkeane wrote: > > So the problem with statement attributes is that we don't really have a > > good way to do instantiation of them on a template, which is why we held > > off on this in the first place. The infrastructure for instantiation DOES > > happen on decl attributes automatically anyway, so that made it easier to > > do. > > So please ensure there is some sort of hooks in the statement attributes to > > ensure that they get a chance to instantatiate themselves (and add tests > > with templates in them!). > > This is a fair point... how are standard statement attributes handled inside > of templates? (Assuming clang supports that.)
All our handling of attributes is pretty manual for anything with arguments. We have a bit of an automatic handling for many attributes, but arguments aren't. I think there is a section in `SemaInstantiateDecl.cpp` that handles a lot of it (`InstantiateAttrs`) and you can see there is a LOT of manual handling there https://github.com/llvm/llvm-project/pull/110334 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
