Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!uakari.primate.wisc.edu!ames!sgi!shinobu!odin!moose.sgi.com!jwag
From: jwag@moose.sgi.com (Chris Wagner)
Newsgroups: comp.sys.sgi
Subject: Re: How to force floating point exceptions to dump core
Keywords: core, signal, sigset, NaN, IEEE floating point
Message-ID: <7188@odin.corp.sgi.com>
Date: 3 May 90 15:52:07 GMT
References: <1521@merlin.bhpmrl.oz>
Sender: news@odin.corp.sgi.com
Reply-To: jwag@moose.sgi.com (Chris Wagner)
Organization: Silicon Graphics, Research & Development
Lines: 116

In article <1521@merlin.bhpmrl.oz>, tim@merlin.bhpmrl.oz (Tim Monks) writes:
> How do you force floating point exceptions to dump core in C on an SG ? 
> 
> The reason I want to do this is I've got a longish program which
> generates NaN's somewhere. I thought the easiest way to track
> down the problem was to set up an exception handler which would
> capture SIGFPE signals and take appropriate action. I didn't manage
> to get even that far, I couldn't even make the program baulk at NaN's.
> 
> 
> I've tried some variations on the following with no success:
> 
> --8<----------------------------------------------------------------------
> #include <stdio.h>
> #include <math.h>
> #include <signal.h>
> #include <fp_class.h>
> 
> main(argc, argv)
> 
> int     argc;
> char    **argv;
> 
> {
> 
>     void	Report_class(double);
>     double	x,zero;
> 
> 
>     /* set up a signal trap */
>     signal(SIGFPE, SIG_DFL);
> 
> 
>     /* try to trap Infinity */
>     zero = 0.0;
>     x = 1.0/zero;
> 
>     /* 
>     I wanted the program to dump core on the previous line, 
>     but it carried on, so we have a look at what we got.
>     */
>     Report_class(x);
> 
> 
>     /* Now try to trap NaN */
>     x = zero/zero;
>     Report_class(x);
> }
> 
> 
> 
> void Report_class(double x)
> {
>     int		type;
> 
>     type = fp_class_d(x);
>     if (type == FP_SNAN)
> 	fprintf(stderr,"X = %lf is a signalling NaN\n",x);
>     else if (type == FP_QNAN)
> 	fprintf(stderr,"X = %lf is a quiet NaN\n",x);
>     else if (type == FP_POS_INF)
> 	fprintf(stderr,"X = %lf is positive infinity\n",x);
>     else if (type == FP_NEG_INF)
> 	fprintf(stderr,"X = %lf is negative infinity\n",x);
>     else
> 	fprintf(stderr,"X = %lf is something quite different !!!\n",x);
> }
> --8<----------------------------------------------------------------------
> 
> And the program gives the following output: (240GTX, Irix 3.2.2)
> 	    X = Infinity is positive infinity
> 	    X = NaN is a quiet NaN
> but definitely no core!  
> 
> How do you make quiet NaN's noisy ? (a signaling NaN)
> 
> I've RTFM, but haven't found any enlightenment on this. Can someone
> point me at the relevant FM to read, or provide a solution.
> 
> On this topic, how does SG propagate NaN's and Infinity through
> arithmetic and math library operations ?
> 
> 
> I'll post a summary if there's any demand for one.
> 
> Thanks in advance.
> -- 
> Dr. Tim Monks                                
> 
> Image Processing & Data Analysis Group   |   (direct) (+61-3)566-7448
> BHP Melbourne Research Laboratories      |   (switch) (+61-3)560-7066
> 245 Wellington Rd, Mulgrave, 3170,       |   (fax)    (+61-3)561-6709
> AUSTRALIA                                |   (EMAIL)  tim@merlin.bhpmrl.oz.au


Try including and calling the following function:
#include <asm.h>
#include <regdef.h>
#include <sys/fpu.h>

	.globl	fpe_enable
	.ent 	fpe_enable
fpe_enable:
	cfc1	t0, fpc_csr	# grab fpu status register 
	li	t1, 0xF10
	/*li	t1, 0x3f000*/
	or	t0, t0, t1 	# or in enable bits
	ctc1	t0, fpc_csr	# replace fpu status
	j	ra		# return
	.end	fpe_enable

By default the C runtime startup does not enable floating point exceptions

In our next release a complete fp exception package has been added

Chris Wagner