This is the mail archive of the
ecos-discuss@sources.redhat.com
mailing list for the eCos project.
RE: eCos gdb-stub problem (on SA1100)
- To: Andreas Bürgel <ab at genologic dot de>
- Subject: RE: [ECOS] eCos gdb-stub problem (on SA1100)
- From: Gary Thomas <gthomas at cambridge dot redhat dot com>
- Date: Mon, 18 Jun 2001 13:27:35 -0600 (MDT)
- Cc: ecos-discuss at sources dot redhat dot com
- Organization: Red Hat, Inc.
On 18-Jun-2001 Andreas Bürgel wrote:
> While trying to track down my application-problem I got the current
> CVS-source-tree and made a new gdb-stub. Before I used a gdb-stub
> compiled with two(?) month old cvs-tree and a current cvs tree for
> application development.
> I can connect to the stub with gdb, but download fails after some
> hundred bytes. I get the .rom_vectors segment transferred, but after
> transferring approx. 1220 bytes of the .text segment I get packet errors
> and the transfer hangs. The hardware hangs and shows the same symptoms
> as in my other posting today - my debug LED is on and stays on. At this
> time of the download, only the stub code contains LED interaction (some
> blinking).
> I'm quite sure I didn't change very spectacular things in my
> stub/startup code, so why does it work with a two month old eCos-tree
> but not with the current?
It sounds to me like there is a problem with memory layout of the
code you are trying to load and the stubs themselves.
Can you provide a little more information? One thing that could be
useful would be a map of what you are loading. How about the output
from 'arm-elf-objdump -h' for both your GDB stubs and your application.