http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40778
--- Comment #11 from Frank Ch. Eigler <fche at redhat dot com> 2012-01-17 16:23:15 UTC --- Jakub's patch makes sense to me in the sense that it's the least modification over what's there. Unfortunately, I do not recall what other problems there were with instrumenting general DECL_ARTIFICIAL !mf_marked_p functions. Perhaps at the time, there was a problem propagating the mf_marked_p flag.
