https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82177
--- Comment #2 from Nuno Lopes <nunoplopes at sapo dot pt> --- (In reply to Andrew Pinski from comment #1) > I don't see how this can ever be defined code. > > By definition yi will never be equal to pi unless by accident. Sure, that's ok, but then pi = &x, and so the store '*(int*)pi = 15' should write to x. Right now it's printing x=0, which cannot be correct.
