[Bug c++/69843] C++: Inconsistent treatment of may_alias attribute and forward declarations

[fw at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=69843

--- Comment #2 from Florian Weimer <fw at gcc dot gnu.org> ---
(In reply to Jakub Jelinek from comment #1)
> Is the forward declaration here in glibc headers?
> If yes, using __attribute__((may_alias)) there too wouldn't hurt.
> Or do you mean users forward declare standard types before including glibc
> headers?

The latter.  I don't see a forward declaration inside glibc itself.