https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81245
--- Comment #9 from Andrew Pinski <pinskia at gcc dot gnu.org> --- (In reply to Richard Biener from comment #5) > Actually > > if (fold_stmt (...)) > update_stmt (gsi_stmt (new_gsi)); > > is correct ;) The old update_stmt was redundant (gsi_insert_before updates > the stmt already). fold_stmt may replace the stmt so with your fix you > might just end up updating a stale one... Yes, I agree and in fact my testing showed my patch did not work, I am testing a new patch: Index: tree-if-conv.c =================================================================== --- tree-if-conv.c (revision 249769) +++ tree-if-conv.c (working copy) @@ -1853,7 +1853,8 @@ new_stmt = gimple_build_assign (res, rhs); gsi_insert_before (gsi, new_stmt, GSI_SAME_STMT); gimple_stmt_iterator new_gsi = gsi_for_stmt (new_stmt); - fold_stmt (&new_gsi, ifcvt_follow_ssa_use_edges); + if (fold_stmt (&new_gsi, ifcvt_follow_ssa_use_edges)) + new_stmt = gsi_stmt (new_gsi); update_stmt (new_stmt); if (dump_file && (dump_flags & TDF_DETAILS)) ---- CUT ---- This is so the print below has the correct statement when printing.