This is the mail archive of the
ecos-discuss@sources.redhat.com
mailing list for the eCos project.
Re: MBX redboot
- To: Jesper Skov <jskov at redhat dot com>
- Subject: Re: [ECOS] MBX redboot
- From: Jonathan Larmour <jlarmour at redhat dot com>
- Date: Wed, 23 May 2001 22:11:52 +0100
- Cc: ecos-discuss at sources dot redhat dot com
- Organization: Red Hat UK Ltd.
- References: <DDD04F4480FED411AB5D00508BFD7A441783AF@CTSINTDLSXUA> <3AFAE16E.3788227@redhat.com> <ot4ruc60th.fsf@thinktwice.zoftcorp.adsl.dk>
Jesper Skov wrote:
>
> >>>>> "Jonathan" == Jonathan Larmour <jlarmour@redhat.com> writes:
>
> >> "Natarajan, Mekala (CTS)" wrote:
> >>
> >> How to debug the rom monitor(redboot).
>
> Jonathan> No you can't debug redboot itself very easily because even a
> Jonathan> RAM based RedBoot will steal most of the exception vectors
> Jonathan> etc. for itself (as will any application that has GDB stubs
> Jonathan> included). And quite right too - you wouldn't be able to use
> Jonathan> a RAM based RedBoot to upgrade if it still relied on the
> Jonathan> ROM/Flash based one to run.
>
> Actually, debugging the vector magic can be done simply by changing
> the location of the virtual vector table in the RAM application
> (RedBoot).
>
> Then the ROM monitor can be used for debugging most of the RAM app
> startup initialization, including virtual vector setup. You may have
> to hack a few things to prevent the RAM app from taking over exception
> vectors and/or serial control. But it's definitely doable.
Well, yes, if you want to explicitly override this behaviour in the sources
then fair enough - anything is possible :-). Although I suppose it might be
worthwhile you adding this to your current draft of the porting guide as a
tip?
Jifl
--
Red Hat, Rustat House, Clifton Road, Cambridge, UK. Tel: +44 (1223) 271062
Maybe this world is another planet's Hell -Aldous Huxley || Opinions==mine