https://github.com/aaronpuchert commented:
I can only tangentially comment on the parser bits, because I'm more familiar with the actual analysis. I like what the change is doing, and it's clear that this should be allowed. But looking at the implementation it seems as if we're struggling with treating this as declaration attribute instead of type attribute. We have to introduce the function parameters in the context of the function pointer variable declaration, instead of in the context of the function prototype itself. As I wrote on other changes, this is all Ok for now. Talk is cheap, and I don't have the time myself to fix this. It's just a bit of technical debt that we're accumulating every time we add functionality that emulates the behavior of type attributes with declaration attributes. https://github.com/llvm/llvm-project/pull/212615 _______________________________________________ cfe-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
