[ag-automation] configuration concept for fieldbus stack

[Robert Schwebel]

On Tue, Jul 10, 2007 at 05:22:05PM +0200, Dieter Hess wrote:
> the xml-file is just an external representation of the parameters in
> the fddi-interface.

I've added the PTX config format, adapted to this modbus example, here:
https://www.osadl.org/svn/osadl/software/fieldbus-framework/fddi-20070618-1-trunk/busconfig/Pengutronix

It follows a stricter scheme than yours (the idea is to be able to check
much of the content in the xml layer, not in the software), but anyway,
it's just for reference. Since last friday, I have a Wago bus coupler
running with the current trunk; it is far from being complete, but
anyway, it's progress.

The configuration concept makes it possible to have different
configuration formats per IO device; in the end, each driver vendor
could be able to provide his very own config format for his card, and
the framework would be able to handle it down to the driver, who in turn
is able to understand that special format.

Although the design is so flexible that this is possible, I'd feel much
better if the OSADL Fieldbus Framework would propose it's own XML
configuration format. In the end, it would be really great to have a
unified configuration tool which is able to configure _all_ cards and
interfaces. Nevertheless, the flexibility is necessary, because I
suppose that vendors have their own ideas how things should look like.
In this context it would be very interesting to here about oppinions
from other PLC companies than 3S - in the end, OSADL is about
collaboration - so please help the fieldbus stack project on it's way
towards world domination :-)

One question: the sniplet Mr. Hess has sent is for one IO device only;
would we want to have

1) one configuration file for the whole system, plus a pointer to which
   entry is meant

2) one configuration file per IO device, plus a second configuration
   which defines which IO devices do exist in the system?

What are peoples' strategies here? Please send your suggestions.

My current plan is to finish the modbus example first; when it is
running in a less hardcoded way than it is now, I'll have a look at what
the Homag people sent wrt. the async interface.

Regards,
Robert Schwebel
-- 
 Dipl.-Ing. Robert Schwebel | http://www.pengutronix.de
 Pengutronix - Linux Solutions for Science and Industry
   Handelsregister:  Amtsgericht Hildesheim, HRA 2686
     Hannoversche Str. 2, 31134 Hildesheim, Germany
   Phone: +49-5121-206917-0 |  Fax: +49-5121-206917-9