From: "John M. Aldrich" <fighteer AT cs DOT com>
Newsgroups: comp.os.msdos.djgpp
Subject: Re: Failure during timing-sensitive application...
Date: Sun, 22 Sep 1996 23:37:49 -0700
Organization: Three pounds of chaos and a pinch of salt
Lines: 46
Message-ID: <3246303D.4413@cs.com>
References: <Pine DOT SUN DOT 3 DOT 95 DOT 960922195102 DOT 857B-100000 AT conan DOT ids DOT net>
Reply-To: fighteer AT cs DOT com
NNTP-Posting-Host: ppp228.cs.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
To: Tom Sgouros <tomss AT ids DOT net>
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp

Tom Sgouros wrote:
> The problem is this: One time in three (give or take), the program hangs
> mysteriously in the first minute of execution. An unexplainable disk
> access (i.e. there is no program file access) coincides with the hang. If
> it doesn't hang within a minute, it seems to run indefinitely.
> 
> I assume that there is some disk swap or another at work here, but I don't
> know why the program doesn't just start where it left off. I further
> assume that it is because of some error I've made, but I can't find it
> yet.

Here's a possibility:  Your program is just big enough that sometimes it
needs to swap out to virtual memory.  When it does this, one or more of
the interrupt routines you installed get swapped out to disk and the
program immediately crashes.  If this is the case, there are two ways
around it:

1) Reduce the amount of memory used by other applications running on the
computer. (TSR's, drivers, etc.)  Either that or get more RAM.

2) Try locking all code and data used by the interrupt handlers.  The
DJGPP Frequently Asked Questions list (v2/faq201b.zip) covers almost
every aspect of HW interrupts in section 18 (specifically 18.9 thru
18.11).

I am assuming here that you are using some sort of interrupt handlers
because of the precise nature of the timing that seems to be involved.
If you aren't, or the above doesn't solve your problem, try to isolate
the code fragment that causes the crash by placing diagnostic output
(fprintf( stderr, ...) ) in critical places and narrow it down until
you get it.  Then post the critical code fragment or fragments here on
the newsgroup and we'll see what we can do.  :)

Hope this helps!

-- 
John M. Aldrich <fighteer AT cs DOT com>                      

* Anything that happens, happens.
* Anything that, in happening, causes something else to happen,
  causes something else to happen.
* Anything that, in happening, causes itself to happen again, happens
  again.
* It doesn't necessarily do it in chronological order, though.
 
                                       --- Douglas Adams