Date: Mon, 21 Jun 1999 19:27:28 +0300 (IDT)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
X-Sender: eliz AT is
To: Erik Berglund <erik2 DOT berglund AT telia DOT com>
cc: pavenis AT lanet DOT lv, djgpp-workers AT delorie DOT com
Subject: Re: Re: gcc-crash - and a possible solution
In-Reply-To: <MAPI.Id.0016.00333138303633303030303930303039@MAPI.to.RFC822>
Message-ID: <Pine.SUN.3.91.990621192301.26448C-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Reply-To: djgpp-workers AT delorie DOT com
X-Mailing-List: djgpp-workers AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk


On Mon, 21 Jun 1999, Erik Berglund wrote:

> Because when I changed the stack size with stubedit,
> I was not able to fix the problem, but I could make the traceback
> change, and sometime I got "Abort!" instead of the traceback,
> and in some strange cases, I could make the problem
> disappear by _decreasing_ the stack, but only for a while.

When you change the stack size, the place in memory where the heap lives 
changes as well, because in the DJGPP memory layout, heap comes after 
(i.e. above) the stack.  So, if this bug somehow depends on contents of 
memory used by GCC for its heap, changing the stack size can affect the 
behavior of the bug.

Allocating smaller or larger stack also affects the DPMI server, since it 
has different amounts of free memory at different areas.  This can also 
change the effects.

It might be a good idea to summarize all the different effects you can 
produce by manipulating whatever parameters are important, and post the 
list of results, because looking at all of the effects might bring some 
ideas or ring some bells.