https://gcc.gnu.org/bugzilla/show_bug.cgi?id=123113
--- Comment #8 from Drea Pinski <pinskia at gcc dot gnu.org> --- (In reply to Drea Pinski from comment #6) > factor_out_conditional_operation is harder to handle because we can't really > do on demand forwarder block creation. Because the phi for the one which we > were going to use need to be found. Or maybe wait the new forwarder block is > the new merge anyways. I need to double check that. Yes it is. > > So the other patches could be simplified anyways ... I have a simplified patch for the first one in testing (plus a secondary patch which does `cfgchanged = true;` change first since that shows up even without this one).
