Possible regression with implicit function redefinition:

Hi all-
I discovered this example (https://godbolt.org/g/ZhViUu) in our internal test suite and discovered that Clang Trunk does not diagnose the "redeclaration as a different kind of symbol" anymore. It seems to me that it SHOULD (since it would if it wasn't an implicit declaration), so I suspect that this is an issue, right?


Yes, that seems like a bug.