http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55710
--- Comment #4 from Paolo Carlini <paolo.carlini at oracle dot com> 2012-12-15 19:03:10 UTC --- Oops, 4.6.x says that NSDMIs are needed for line 10 and rejects it, that misled me. Then, what can I say, probably the issue isn't a regression and it's only superficially similar to PR 55015 (which is already fixed)