From: Thomas Demmer <demmer AT LSTM DOT Ruhr-UNI-Bochum DOT De>
Newsgroups: comp.os.msdos.djgpp
Subject: Re: sqrt() in DJGPP
Date: Fri, 20 Feb 1998 10:37:15 +0100
Organization: Lehrstuhl fuer Stroemungsmechanik
Lines: 53
Message-ID: <34ED4ECB.5A5C0CBC@LSTM.Ruhr-UNI-Bochum.De>
References: <01bd3da9$a0052120$0cf234d1 AT mpanian DOT awinc DOT com>
NNTP-Posting-Host: bvb.lstm.ruhr-uni-bochum.de
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Precedence: bulk

Lucas Panian wrote:
> 
> Hi,
> 
> When I try to run my program, it crashes with SIGFPE...I used symify to
> find that it was in the following section of code(all variables are already
> declared)
> 
> A = Y1 * ( Z2 - Z3 ) + Y2 * ( Z3 - Z1 ) + Y3 * ( Z1 - Z2 );
> B = Z1 * ( X2 - X3 ) + Z2 * ( X3 - X1 ) + Z3 * ( X1 - X2 );
> C = X1 * ( Y2 - Y3 ) + X1 * ( Y3 - Y1 ) + X3 * ( Y1 - Y2 );
> 
> Distance = sqrt (A*A + B*B + C*C);
> 
> Normal.X = (A / Distance) + VPoint[0]->Lx;  // It crashes on this line.
> Normal.Y = (B / Distance) + VPoint[0]->Ly;
> Normal.Z = (C / Distance) + VPoint[0]->Lz;
> 
> I'm pretty sure SIGFPE is a 'Not-A-Number' error. I'm pretty sure the bad
> number is 'Distance.'(have done a little bit of detective work...). So,
> they only thing I can think of is an error in the return value of the
> sqrt() function. Am I right in this assumption? Any help would be greatly
> appreciated. Thanks. ;)

I doubt that this is a bug in sqrt(), as it basically boils down to
one or two 87-assembler instructions. More probable, 
you forgot to include <math.h>, so the compiler assumes the
prototype 
int sqrt( _what_ever_A_s__type_is );
or, Distance happens to be zero, which may or may not be
possible following the logic of your code. Recompile the
program with the options
gcc -Wall -pedantic-errors foo.c

to find how to avoid the first possibility, 
re-think your program's logic to avoid the
second.


-- 
Ciao
Tom

*************************************************************
* Thomas Demmer                                             *
* Lehrstuhl fuer Stroemungsmechanik                         *
* Ruhr-Uni-Bochum                                           *
* Universitaetsstr. 150                                     *
* D-44780  Bochum                                           *
* Tel: +49 234 700 6434                                     *
* Fax: +49 234 709 4162                                     *
* http://www.lstm.ruhr-uni-bochum.de/~demmer                *
*************************************************************