http://gcc.gnu.org/bugzilla/show_bug.cgi?id=39213
--- Comment #14 from Jakub Jelinek <jakub at gcc dot gnu.org> 2010-12-17 10:46:09 UTC --- Created attachment 22795 --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=22795 gcc46-pr39213.patch I wonder if this wouldn't fix it (at least, that's similar to how lex.c guards calling of _cpp_process_line_notes). I can't reproduce it myself, so have to guess...