This is the mail archive of the
ecos-discuss@sourceware.org
mailing list for the eCos project.
Re: Porting eCos to the Atmel AT91CAP7
- From: John K <johnk_dev at yahoo dot com>
- To: ecos-discuss at sourceware dot org
- Date: Wed, 19 May 2010 13:40:24 -0700 (PDT)
- Subject: Re: [ECOS] Porting eCos to the Atmel AT91CAP7
- References: <374509.79148.qm@web58608.mail.re3.yahoo.com> <D6050C555CC56940A7AF3265228302761394CA@mail2.STMIRV01.COM>
Christophe,
Basically, a general lack of experience with the ARM architecture,
eCos, GNU tools, and Eclipse is stopping us in the port. It's quite a
steep learning curve trying to fit all of those pieces together and
deal with so many unknowns.
I have been working off the Atmel
AT91SAM7X256 port of eCos, because the processor is quite similar to
the AT91CAP7. Also, I have SAM7X256 sample applications (non-eCos) from
Atmel building and executing unmodified on my CAP7 board. I am also
able to build and debug these sample apps within Eclipse thru a SEGGER
SAM-ICE JTAG debugger.
I have played with the eCos config tool
extensively, built the SAM7X256 target template, and have been trying
to execute the eCos Hello World sample. I was able to run the Hello
World app through my debugger in Eclipse, and it seemed like it was
getting stuck somewhere. When I halted the processor I found it was
stuck in the "wait_pmc_sr_2" loop inside "hal_platform_setup.h", which
is a loop that was waiting for the PLL to lock (I was executing within
Eclipse through my JTAG debugger). Since "hal_platform_setup.h" was
SAM7-specific assembly code, I modified it with the correct
functionality for the CAP7, and it looks like I get past that point
now.
But
now I think I'm lost somewhere in vectors.S - when I execute the app,
it never gets to the breakpoint at main(), my debugger reports "remote
connection closed", and my board and JTAG debugger get into a funky
state where I need to cycle power to the board to get out of it.
That
is where I am stuck now. I have no idea how to figure out what is going
wrong. One thing that might help is if I could set breakpoints within
vectors.S, but I can't seem to link the eCos source to the Eclipse
debug session. I have tried all sorts of different path combinations
within the "Source Lookup Path" of the Eclipse "Debug Configuration" -
the Eclipse "Problems" tab reports "Unresolved Breakpoint" for anything I set within vectors.S.
Any suggestions?
Regards,
John Korsakas
________________________________
From: Christophe Coutand <ccoutand@stmi.com>
To: John K <johnk_dev@yahoo.com>; ecos-discuss@sourceware.org
Sent: Tue, May 18, 2010 5:38:50 PM
Subject: RE: [ECOS] Porting eCos to the Atmel AT91CAP7
RE: [ECOS] Porting eCos to the Atmel AT91CAP7
Hi John,
Just curious, what is actually stopping you in your port?
For consulting, you can check with Zylin.com, I beleive they have an office in US.
Christophe
-----Original Message-----
From: ecos-discuss-owner@ecos.sourceware.org on behalf of John K
Sent: Tue 5/18/2010 2:39 PM
To: ecos-discuss@sourceware.org
Subject: [ECOS] Porting eCos to the Atmel AT91CAP7
Hi everyone,
We have been working on porting eCos to the Atmel CAP7 processor core for a couple weeks now (starting out with the AT91CAP7A-STK target board), and it's going a bit rough. Does anyone else have experience with this processor or perhaps is also working on a port?
Or maybe one of you has some general experience porting eCos (plus building & debugging apps with Eclipse) and can be hired as a consultant to help kick start us? (we've looked at eCosCentric and we were hoping to find someone in the US that would only charge a one-time consulting fee rather than royalties). The resulting port would be fed back to the community.
Regards,
John
johnk_dev@yahoo.com
--
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss
--
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss