This is the mail archive of the
ecos-discuss@sources.redhat.com
mailing list for the eCos project.
Re: Thread messages problem.
- To: felixwong at i-technologies dot cc
- Subject: Re: [ECOS] Thread messages problem.
- From: Jonathan Larmour <jlarmour at redhat dot com>
- Date: Tue, 30 Oct 2001 03:24:21 +0000
- Cc: Ecos-Discuss <ecos-discuss at sources dot redhat dot com>
- References: <NFBBLLEDKLLJHLKGHKACKEHFCAAA.felixwong@i-technologies.cc>
felixwong@i-technologies.cc wrote:
>
> Hello Jonathan,
>
> I tried to remove the "debugging messages for the kernel device initialisation" part,
> only debug message for my own harddisk device was enabled. Then the $T messages
> gone & the system runs without any error.
>
> However, within my file system where I print out the debug info using "diag_printf" to print
> test values to "ser0" of my EB40, the similar message appears. Have you ever met the
> same problem when you diag_print lots of debug messages using your EB40 board?
The thing is that a $T packet is the debugging stub trying to tell GDB it
hit an exception. Although the addresses are all mucked up as I said. It
may mean stack corruption/overflow, or just your program has done something
silly like access illegal memory, run an illegal instruction, or divide by
zero or something like that.
> Is it related to the serial port buffer (which I set to 1024 byte)? Or anywhere I can set my
> default stack size in eCos configuration tools?
Given this is at initialization time you could probably try to increase the
size of the interrupt stack in the HAL.
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