Front page | perl.perl5.porters |
Postings from December 2000
On Sat, Dec 02, 2000 at 06:34:26PM +0000, Nicholas Clark wrote:
> On Tue, Nov 14, 2000 at 10:30:06AM -0600, Jarkko Hietaniemi wrote:
> > > OK. What does a scalar with IV and NV both OK mean?
> >
> > I *think* it *should* mean that the cached IV and NV are both valid
> > *and* that there is no loss of precision: (IV)nv == iv && (NV)iv == nv.
> > Obviously, this doesn't seem to be the case.
>
> OK. How do I detect that 18446744073709551616.0 isn't 18446744073709551615 ?
>
> Round here:
>
> #include <stdio.h>
> int main (void) {
> unsigned long long max=0xFFFFFFFFFFFFFFFF;
> double imprecision;
>
> imprecision = (double) max;
> printf ("imprecision=%.1f\tmax=%llu\n", imprecision, max);
> if (imprecision == ((double) max))
> puts ("imprecision == (double) max");
> if (((unsigned long long) imprecision) == max)
> puts ("(unsigned long long) imprecision == max");
> return 0;
> }
>
> Gives:
>
> imprecision=18446744073709551616.0 max=18446744073709551615
> imprecision == (double) max
> (unsigned long long) imprecision == max
In alpha where sizeof(long long) == sizeof(long) == sizeof(double) == 8:
imprecision=18446744073709552000.0 max=lu
imprecision == (double) max
You can see how the low order bits fade into garbage...
(btw, fwiw, long long is still thought to be a non std language extension,
so in the actual test we should use UV)
> How do I detect that the UV and the NV aren't equal, when all the casting
> conspires to keep it from me? xor the UV low bits with something and see
> if that also is preserved?
> [I *think* that this only happens on UV_MAX, and that we need can detect
> if the platform needs this double test during Configure]
>
> Nicholas Clark
--
$jhi++; # http://www.iki.fi/jhi/
# There is this special biologist word we use for 'stable'.
# It is 'dead'. -- Jack Cohen
Thread Previous
|
Thread Next