[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Broken isnan(3) and/or DQNAN on Tru64
From: |
Ludovic Courtès |
Subject: |
Broken isnan(3) and/or DQNAN on Tru64 |
Date: |
Fri, 01 May 2009 15:33:39 +0200 |
User-agent: |
Gnus/5.13 (Gnus v5.13) Emacs/23.0.90 (gnu/linux) |
Hello,
On `alphaev56-dec-osf5.1b', NaN handling seems to be deeply broken.
Consider this piece of code:
--8<---------------cut here---------------start------------->8---
#include <math.h>
extern unsigned int DQNAN[2];
int
main (int argc, char *argv[])
{
return isnan (* (double *) DQNAN);
}
--8<---------------cut here---------------end--------------->8---
When compiled with GCC 4.0.2, it bails out with SIGFPE.
Guile's 1.8 series does not use Gnulib (the next release series will use
Gnulib), so it's been relying on these facilities, which are also used
in Octave. I'm wondering whether it worked with older Tru64 versions.
I know Gnulib's `isnan' module re-implements isnan(3) following the IEEE
specs, but in your experience is there any way to get a working isnan(3)
on Tru64? Or perhaps it's DQNAN that's broken?
Thanks in advance,
Ludo'.
- Broken isnan(3) and/or DQNAN on Tru64,
Ludovic Courtès <=
- Re: Broken isnan(3) and/or DQNAN on Tru64, Bruno Haible, 2009/05/01
- Re: Broken isnan(3) and/or DQNAN on Tru64, Ludovic Courtès, 2009/05/02
- Re: Broken isnan(3) and/or DQNAN on Tru64, Bruno Haible, 2009/05/02
- Re: Broken isnan(3) and/or DQNAN on Tru64, Ludovic Courtès, 2009/05/02
- Re: Broken isnan(3) and/or DQNAN on Tru64, Bruno Haible, 2009/05/02
- Re: Broken isnan(3) and/or DQNAN on Tru64, Ludovic Courtès, 2009/05/03
- Re: Broken isnan(3) and/or DQNAN on Tru64, Bruno Haible, 2009/05/03
- Re: Broken isnan(3) and/or DQNAN on Tru64, Ludovic Courtès, 2009/05/10