[ECOS] Anyone tried uses PMON instead of GDBstubs on VRC4373.

Ling Su lingsu@palmmicro.com
Mon Jan 22 12:26:00 GMT 2001


> Ling> Hi, Dear All, Previously we were working with eCos on
> Ling> DDB-VRC4373 board. But I checked with NEC rep, the VRC4373 is
> Ling> obselete, the new version is VRC4375. These two boards have few
> Ling> differences, most important is the system bus controller
> Ling> (vrc4375 instead of vrc4373). Unfortunately I found the old GDB
> Ling> monitor doesn't work on the new board. But the PMON comes
> Ling> together with the board is working, I wonder if anyone tried use
> Ling> PMON with eCos on NEC DDB-VRC4373 platform. If it works, I guess
> Ling> I can test my old application is working on the new
> Ling> board. Thanks!
>
> Set the CYGSEM_HAL_USE_ROM_MONITOR option to PMON.
>

I changed to Little Endian by enable LSBFIRST macro define, and compile the
code using PMON monitor option. It seems don't work. I found the toolchain
provide by NEC(Actually Cygnus tool) uses a -Tddb.ld option. The ddb.ld
linker script might be different from what we use in eCos (target.ld). Can I
just apply the ddb.ld instead of target.ld?

 I think the most import thing for me now is to get a GDB stub working. I
want to save the time to burn new EPROM, so I follow the document on eCos
website, to run the stub from RAM startup first since I have a working
monitor PMON now. But I am not that clear on how to make a "load and go" GDB
stub working with my PMON. From my current point of view, I don't know if
the linker scripts target.ld is compatible with ddb.ld of PMON. Is this be a
patential problem?

Supposely I change to CYGSEM_HAL_USE_ROM_MONITOR option to PMON, and
CYG_HAL_STARTUP option to RAM, the gdb stub elf format file can work with
PMON, is this understanding correct?

Thanks Jesper for your kindly reply! Thanks Jonathan, too! You always give
me a lot of help!

Regards,
-Ling



More information about the Ecos-discuss mailing list