Debugging RTEMS application on Power architecture target

Nick Withers nick.withers at anu.edu.au
Fri Jun 21 06:49:26 UTC 2013


Hi guys,

Newbie RTEMSer here, probably with a newbie question (or two)!

I'm just at the initial stages of a project aiming to replace an ageing
and growing-harder-to-support VME-based data acquisition control system
and am evaluating RTEMS on an MVME3100.

At the moment it's not totally clear to me what (remote, preferably)
debugging options (I'm thinking more than printf() here) I've got for a
Power (MPC8540)-based solution.

>From what I can gather:
  - Low-level RTEMS GDB stubs are not available for the Power
architecture
  - SERDBG isn't supported for the Power architecture
  - rtems_gdb_start() is not available for the Power architecture
  - rgdb...? Dead? Perhaps it's external to the rtems.git sources and I
didn't look hard enough?

...which leaves me looking at the (pretty expensive, IMHO) BDI3000s and
PEEDIs (any other suggestions for alternatives warmly accepted! Anyone
tried a Wind River ICE 2?), over the MVME3100's COP port.


Is my understanding correct? Alternate options I heven't thought of?
Perhaps I could actually run a full-blown GDB on the target? That'd
presumably be a big binary and use a bit of memory (got 256 MiB though),
but still...?

If there aren't any software options, I'm looking primarily at the
BDI3000 as it seems to be the "most supported" for RTEMS use, but could
save over AU/US $1000 going instead with the PEEDI. Anyone got any
recommendations, here? The Wind River ICE 2's $700-odd cheaper than the
PEEDI, if that'd do the job.

Cheers for your time - and for RTEMS!
-- 
Nick Withers

Embedded Systems Programmer
Room 2.26, Building 57
Department of Nuclear Physics
Research School of Physics and Engineering The Australian National
University (CRICOS: 00120C)

eMail: nick.withers at anu.edu.au
Phone: +61 2 6125 2091
Mobile: +61 414 397 446






More information about the users mailing list