http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56725
--- Comment #9 from Paolo Carlini <paolo.carlini at oracle dot com> 2013-03-27 19:52:57 UTC --- Ok, thanks! Thus, if I remember correctly that comment (true meaning warning), looks like we want everywhere if (!permerror...) inform(... right? Otherwise if I misremembering, we should anyway fix the existing permerror/inform sequences, right?
