http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53363
--- Comment #14 from Jason Merrill <jason at gcc dot gnu.org> 2013-03-05 15:41:15 UTC --- (In reply to comment #13) > What is supposed to be tested? Should the whole test skipped on darwin or only > the dg-final? The whole test; the test is verifying that the x86 PIC thunk doesn't have debug info, but Darwin doesn't emit that thunk.