On Fri, Feb 04, 2022 at 11:26:30AM +0100, Jakub Jelinek via Gcc-patches wrote:
> On Fri, Feb 04, 2022 at 11:14:05AM +0100, Eric Botcazou wrote:
> > > > --- a/gcc/match.pd
> > > > +++ b/gcc/match.pd
> > > > @@ -401,27 +401,35 @@ DEFINE_INT_AND_FLOAT_ROUND_FN (RINT)
> > > >
> > > > /* X / bool_range_Y is X. */
> > > > (simplify
> > > >
> > > > (div @0 SSA_NAME@1)
> > > >
> > > > - (if (INTEGRAL_TYPE_P (type) && ssa_name_has_boolean_range (@1))
> > > > + (if (INTEGRAL_TYPE_P (type)
> > > > + && ssa_name_has_boolean_range (@1)
> > > > + && !flag_non_call_exceptions)
> > >
> > > ssa_name_has_boolean_range call is certainly more expensive than
> > > !flag_non_call_exceptions check, can you swap those two?
> >
> > But !flag_non_call_exceptions is (almost) always true for the C family of
> > languages, so you're going to penalize them by doing this.
>
> True, but much less so than the other order penalizing Ada/Go.
>
> > > And similarly, TYPE_PRECISION (type) > 1 check is very cheap, can
> > > it be done before the && !integer_zerop (@1) line?
> >
> > Yes, it clearly belongs there.
Anyway, not a big deal.
Jakub
