https://gcc.gnu.org/bugzilla/show_bug.cgi?id=117385
--- Comment #3 from Andrew Pinski <pinskia at gcc dot gnu.org> --- I think I will do it differently than maybe_fold_comparisons_from_match_pd though. Since most of the time phiopt will use the comparison statement, it won't be on the stack and just a normal gimple which will be added to the sequence seq1 and it would just work. Plus we already mark the lhs of the sequence (which we don't discard) for the simple dce, we don't need to handle it otherwise special.
