Hello Kim,

Please have a look at *rt*.*wiki*.kernel.org

If you use a preempt-rt Kernel you shoud note following points:
- for realtime processes use SCHED_FIFO or SCHED_RR
- you can raise the priority up to 99 (only root)
- if you want to have a nice realtime performance for your control process, you should have a equal or higher priority than the IRQ-Threads

You need not raise the priority of the master kernel thread. It has no functionality in the realtime context
It will only scan or reconfigure the bus, if you change the bus structure.

Greatings

Andreas


Am 19.01.2011 15:52, schrieb Kim H. Madsen:

Hi

We are using the user space interface with the application running at a real time priority on a PREEMPT_RT system.

Until now we have been running the EtherCAT master on priority 0 (SCHED_OTHER).

We are considering to raise the priority of the ETherCAT to a real time priority (SCHED_FIFO). We consider setting the priority to 50 so that it is locate just a bow the sirg kernel threads. Is this sane? Or is it better place the master priority vice?

Best regards

*Kim Hedegaard Madsen*
/DEIF Wind Power Technology/


_______________________________________________
etherlab-users mailing list
[email protected]
http://lists.etherlab.org/mailman/listinfo/etherlab-users


--

-----------------------------------------------------------------------

Dipl.-Ing. Andreas Stewering-Bone           Amtsgericht Essen HRB 11500
Ingenieurgemeinschaft IgH                   USt-Id.-Nr.: DE 174 626 722
Gesellschaft für Ingenieurleistungen mbH    Geschäftsführung:
Heinz-Bäcker-Str. 34                        Dr.-Ing. S. Rotthäuser
D-45356 Essen                               Dr.-Ing. T. Finke
Tel.: +49 201 / 360-14-15                   Dr.-Ing. W. Hagemeister
Fax.: +49 201 / 360-14-14                   Tel.: +49 201 / 360-14-0
[email protected]        http://www.igh-essen.com

-----------------------------------------------------------------------

_______________________________________________
etherlab-users mailing list
[email protected]
http://lists.etherlab.org/mailman/listinfo/etherlab-users

Reply via email to