Date: Sun, 5 Jul 1998 16:34:27 +0300 (IDT)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
To: Dan Goodman <dog AT fcbob DOT demon DOT co DOT uk>
cc: djgpp AT delorie DOT com
Subject: Re: SIGFPE
In-Reply-To: <899425828.6166.0.nnrp-03.9e989aee@news.demon.co.uk>
Message-ID: <Pine.SUN.3.91.980705163314.23984a-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Precedence: bulk


On Fri, 3 Jul 1998, Dan Goodman wrote:

> A program I have written (ray tracing sort of thing) is suddenly getting
> SIGFPEs, which doesn't stop it from running, but it runs half as slow as it
> used to and crashes RHIDE after it finishes. The error occurs when you
> change one value from a value lower than 0.9 to one greater than or equal to
> 0.9. I am now pretty sure it is not a divide by zero, sqrt(-x) or overflow
> problem, what else might cause a sigfpe and how can I trace the problem?

The DJGPP FAQ list has some advice on how to begin debugging your 
program, in section 9.2.  If that doesn't help, post here the shortest 
program that reproduces the bug.