Date: Tue, 24 Dec 2002 08:07:48 +0200 (IST)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
X-Sender: eliz AT is
To: Richard Dawe <rich AT phekda DOT freeserve DOT co DOT uk>
cc: DJGPP workers <djgpp-workers AT delorie DOT com>
Subject: Re: ld crash when building emu387.dxe
In-Reply-To: <3E075A0F.5C476C7A@phekda.freeserve.co.uk>
Message-ID: <Pine.SUN.3.91.1021224080301.24775C@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Reply-To: djgpp-workers AT delorie DOT com
Errors-To: nobody 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, 23 Dec 2002, Richard Dawe wrote:

> Please note that I also get this crash in another tree, which is built with
> stock versions from Simtel.NET. Both trees have been built with -g with gcc
> 3.2.1 - DWARF-2 debugging information.

I'd suspect ld, not GCC.  So perhaps try with an older version of 
Binutils (I seem to recall other complaints about the stability of the 
latest Binutils release).

> Page fault at eip=000438d0, error=0006

This means the program crashed trying to write (assign a value) to a
bad address.

> Program received signal SIGSEGV, Segmentation fault.
> mark_relocs (finfo=0xfdcb0, input_bfd=0x113870) at ../../bfd/cofflink.c:1349
> 1349              finfo->sym_indices[ irel->r_symndx ] = -1;

Since this is an assignment, I'm guessing that irel->r_symndx has a
garbled value which causes the code to write past the end of the 
sym_indices array.  Alternatively, finfo->sym_indices could itself be a 
bad address.  So please print the values of the variables involved in
this expression and try to figure out which one is garbled and why.