This is the mail archive of the ecos-discuss@sources.redhat.com mailing list for the eCos project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

RE: Multi-ICE support Status


I tried the new version of code (v2?) ;
here are the results on e7t board:

1. There is a problem in vector.s (or in my setup):

#ifdef CYGHWR_HAL_ARM_ICE_THREAD_SUPPORT
        ldr     r2,=ice_thread_vector
//BLD!        sub     r2,r2,r1        // compute fixed (low memory) address
        ldr     r3,=0x4D494345  // 'MICE'
        str     r3,[r2],#4

in the commented instruction r2 is the ice_tread_vector which resides in
section ".fixed_vectors"
which in my case has the absolute address 0x20 (the vector itself has the
absolute address 0x150)
r1 is the __exception_handlers which has address 0x10000
This way the sub result got negative.

I commented out the sub instruction, now the sting MICE got written to the
address 0x150.
The comment in vectors.s says that it should be address 0x50. Difference in
0x100 is because of the
hal_virtual_vector_table which I can't remove.
So my MultiIce server keeps reporting "Not a good vector or eCos not
running!"

Also when I enabled CYGDBG_HAL_DEBUG_GDB_INCLUDE_STUBS, I start getting GDB
output to my debug console
like : $T02thread:00000006;0f:08c40100;0d:fc4f0400;#24
Because I don't have GDB on the serial port, and GDB waits for input, code
stack.

Looks like I still miss something in flags.

-Benny

> -----Original Message-----
> From: jlarmour@localhost.localdomain
> [mailto:jlarmour@localhost.localdomain]On Behalf Of Jonathan Larmour
> Sent: Thursday, May 16, 2002 6:52 PM
> To: benny@voicepump.com
> Cc: ecos-discuss@sources.redhat.com
> Subject: Re: [ECOS] Multi-ICE support Status
>
>
> benny wrote:
> >
> > My question is now about MultiICE eCOS awareness.
> > First attempt was not to use stubs at all. My application
> worked fine then
> > (breakpoints, memory read/write etc) but I don't have any OS awareness.
> > Then I tried to follow the instructions and rebuild the eCOS library.
> >
> > Two flags:
> > CYGDBG_HAL_DEBUG_GDB_THREAD_SUPPORT
> > CYGDBG_KERNEL_DEBUG_GDB_THREAD_SUPPORT
> > were enabled before (that is how I got eCOS awareness through the serial
> > port)
> >
> > But then I can't find CYGHWR_HAL_ARM_ICE_THREAD_SUPPORT option at all.
> > I use eCOS from the CVS tip - was it checked in? Or I can
> > find it somewhere else?
> > The flag was mentioned once in some document came with eCOS, that's it.
>
> That flag will appear in eCos v2. Which should be out in anoncvs form in
> the next few days. Please don't ask about it yet - we're quite busy on it!
>
> Jifl
> --
> Red Hat, Rustat House, Clifton Road, Cambridge, UK. Tel: +44 (1223) 271062
> --[ "You can complain because roses have thorns, or you ]--
> --[  can rejoice because thorns have roses." -Lincoln   ]--
> Opinions==mine


-- 
Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos
and search the list archive: http://sources.redhat.com/ml/ecos-discuss


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]