http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60760
--- Comment #3 from Daniel Krügler <daniel.kruegler at googlemail dot com> --- (In reply to Jonathan Wakely from comment #2) > Yes, we rely on that in initializer_list<>::end(), and clang rejects that. <nod> > I discussed that case with Jason and we think [expr.add] could be clarified > - do you have an existing issue number related to that? I haven't - I myself found the existing wording always clear in that regard.
