[Chicken-users] Re: [q-lang-users] Re: IEEE float arithmetic

[Abdulaziz Ghuloum]

John Cowan wrote:

> Albert Graef scripsit:
>
>
> > But you want the code at least not to do any harm on systems with _no_ 
> > IEEE floats. I'm not sure, but couldn't dividing by 0.0 on such systems 
> > kill your program with SIGFPE?
>
>
> I don't know.  I've never done serious floating-point work that
> had to run on a non-IEEE system (everything I did on a PDP-8, PDP-11,
> or pre-IEEE VAX was pretty much either integer-only or floats-as-integers,
> as in most Basics).

Are there any processors nowadays that only support non-IEEE flonums?

> > Another point is that division is a much more costly operation than 
> > adding and subtracting.
>
>
> True.

But you don't need division to perform that test.  You can do the
following to be sure:

static double pos_inf = +1.0/0.0;
static double neg_inf = -1.0/0.0;
int isinf(double f){
  return f==pos_inf || f==neg_inf;
}

Now I have to say that gcc is brain damaged sometimes since it
constant-folds the 1.0/0.0 and -1.0/0.0 on PPC but performs the
actual division at runtime on Intel!  The above code should be
cheap on either although I'm not really impressed by the output
of gcc.

Aziz,,,