https://gcc.gnu.org/bugzilla/show_bug.cgi?id=119149
--- Comment #2 from Jonathan Wakely <redi at gcc dot gnu.org> --- Hmm, I don't think it's a dup. In bug 91388 control really does seem to flow off the end (it's just that the function can never be called at all). Here there is clearly a call to a noreturn function so the end of the function is unreachable.
