Xref: news2.mv.net comp.os.msdos.djgpp:1180
From: Charles Sandmann <sandmann AT clio DOT rice DOT edu>
Newsgroups: comp.os.msdos.djgpp
Subject: Re: Problems with cc1.exe (djgpp v 2.0)
Date: Tue, 13 Feb 1996 21:08:12 CST
Organization: Rice University, Houston, Texas
Lines: 18
Message-ID: <3121521c.sandmann@clio.rice.edu>
References: <4fpm6c$an AT news DOT uni-c DOT dk>
Reply-To: sandmann AT clio DOT rice DOT edu
NNTP-Posting-Host: clio.rice.edu
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp

> I have developed the program at work on a P90 with 16 Mb RAM using
> djgpp v1.12m4 -- where no problems can be observed.

Two differences here - one is the GCC version 2.6.3 vs 2.7.2; the other
is that under DOS you had infinite stack in V1.x if you weren't using
DPMI.

> General Protection Fault at eip=eb9fd; flags=3017
> eax=00000000 ebx=000000ff ecx=00000000 edx=00000002 esi=00000000 edi=000000ef
> ebp=000ecfe4 esp=00003390 cs=e7 ds=ef es=ef fs=cf gs=ff ss=33 error=bbb8

This message comes from CWSDPMI instead of DJGPP (they look much alike), and
the only time you should see this is if you get nasty nested exceptions.
This would probably mean the internal exception handling structures are
scrogged - and the error code indicates a bogus selector was trying to be
loaded.  The EIP and EBP are a bit too close for comfort.  I'll guess
we have a stack overflow.  Try stubediting cc1.exe to have 512K stack
and see if that helps (?)