Date: Wed, 10 May 2000 08:16:47 -0400 (EDT)
Message-Id: <200005101216.IAA11783@indy.delorie.com>
From: Eli Zaretskii <eliz AT delorie DOT com>
To: Stefan =?iso-8859-1?Q?Sp=E4nnare?= <stefans AT astro DOT lu DOT se>
CC: djgpp AT delorie DOT com
In-reply-to: <3918626E.B83E232C@astro.lu.se> (message from Stefan
	=?iso-8859-1?Q?Sp=E4nnare?= on Tue, 09 May 2000 21:09:34 +0200)
Subject: Re: Differences between Win 95 and Win 98
References:  <3918626E DOT B83E232C AT astro DOT lu DOT se>
Reply-To: djgpp AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: djgpp AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk

> Date: Tue, 09 May 2000 21:09:34 +0200
> From: Stefan =?iso-8859-1?Q?Sp=E4nnare?= <stefans AT astro DOT lu DOT se>
> 
> I wonder if there is a difference (perhaps memory or stack related)
> between Windows 95 and Windows 98 (perhaps also between
> Windows 98 and Windows 98 second edition) when running
> C-programs compiled with DJGPP ?

I know for a fact that there are differences in how different versions
of Windows 9X handle DPMI programs.  However, the details are not
known, because we can hardly expect Microsoft to tell us what's
changed between these versions.

> I have noticed that my program "sspiel2.c" sometimes craches
> when running Windows 98 second edition but not when running
> Windows 95.

If the program prints the standard DJGPP crash message, it might be a
good idea to post here these crash messages, after you've run SYMIFY
on them.  The information there might hold some crucial clues to
what's going on.

> The problem only arises when calculating e=exp(1)
> using "Recursive Binary Splitting" which is heavily recursive (the
> other parts of the program are not recursive).

Did you look into a possibility of stack overflow (see section 12.2 of
the DJGPP FAQ for an explanation how to identify stack overflows)?
Did you try to enlarge the run-time stack of your program?