https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77308
--- Comment #25 from wilco at gcc dot gnu.org --- (In reply to Bernd Edlinger from comment #24) > (In reply to Bernd Edlinger from comment #23) > > @@ -5020,7 +5020,7 @@ > > (define_insn_and_split "one_cmpldi2" > > [(set (match_operand:DI 0 "s_register_operand" "=w,&r,&r,?w") > > (not:DI (match_operand:DI 1 "s_register_operand" " w, 0, r, w")))] > > BTW: who knows what it is good for to have > =w > w > > on the first alternative, and on the 4th alternative > ?w > w > > "?" does only make the alternative less attractive, but it is otherwise > exactly the same as the first one, right? Alternatives can be disabled, there are flags, eg: (set_attr "arch" "neon_for_64bits,*,*,avoid_neon_for_64bits") I think the NEON alternatives not only do not help at all, they actually generate cause CQ to be far worse even when no NEON instructions are ever generated.
