Re: Use assert()s for optimization?

[Alex Rosenberg <alexr at spies dot com>]

On Nov 7, 2003, at 1:00 AM, Richard Guenther wrote:

> On Thu, 6 Nov 2003, Richard Guenther wrote:
>
> > On Thu, 6 Nov 2003, Falk Hueffner wrote:
> >
> > > Richard Guenther <rguenth@tat.physik.uni-tuebingen.de> writes:
> > >
> > > > Is there a way to turn an assert() into optimization hints for gcc?
> > > > Like in
> > > >
> > > > double mysqrt(double x)
> > > > {
> > > >   assert(x >= 0.0);
> > > >   return sqrt(x);
> > > > }
...
> double mysqrt(double x)
> {
>   if (!(x >= 0.0))
>     abort();
>   return sqrt(x);
> }
... should be transformed into ....
>   if (!(x >= 0)) {
>      ??
>   } else {
>     return sqrt(x);
>   }

Unfortunately, assert(exp) is guaranteed not to evaluate "exp" when 
NDEBUG is set.

In this case, evaluating "x >= 0.0" may result in a change to the FP 
exception state.

This transformation is interesting only in the case where the 
expression has no side-effects.

+------------------------------------------------------------+
| Alexander M. Rosenberg           <mailto:alexr@_spies.com> |
| Nobody cares what I say. Remove the underscore to mail me. |