http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59189
--- Comment #2 from Kazumoto Kojima <kkojima at gcc dot gnu.org> --- It seems that simply there was no user of the valid CFG after m_d_r pass. > Maybe this should be fixed by putting the introduced unconditional > branch instruction into a separate basic block. Sounds plausible.
