Kai Moritz wrote: > Hi folks, > > like Henrique I'm also trying to get orte working together with rtnet. > But I hav some more problems than he: when I try to insert the > ortemanager Module my system freez. The last message on the screen is > something about a rtskb, that cannot be obtained (unfortunatly I've > forgotten to put down the message, though I do not remember it correctly).
Oh, yes, there is something one have to know when trying to run ORTE over RTnet (which is, AFAIK, undocumented yet): due to the old-style kernel design of the ORTE port over RTAI/RTnet (mea culpa), initialisation of sockets takes place in real-time context. This means that the required buffers of that sockets need to be obtained from the global real-time rtskb cache - which is empty by default. Increase it by passing e.g. "rtskb_cache_size=64" during insmod rtnet.ko (or patch the rtnet start script). See also README.pools. Anyway, this should only cause the system to lock-up (run into an endless real-time loop), if some error handling in ORTE is missing. I'm not sure if this is the case with your ORTE version. Petr...? > > I'm using RTAI 3.1 with rtnet 0.8.3 and ORTE 0.3. > > Hey Henrique, > perhaps you could tell me, with wich versions of RTAI, rtnet and ORTE > you are working and send me your configuration-Files (.rtai.config and > so on), so that I can solve my problem! That way we would become two > guys debugging at the same problem ;) > > By the way, I had some serious problems compiling orte. ortespy and > orteping were not build at all and "make install" did only work after > fixing the LIBTOOL-variable in the generated Makefiles. How did you cope > with that problems?! > > Greetings kai > Besides these issues with setting up and running ORTE 0.3 over latest RTnet (which is likely easy to fix), the time has definitely come to move the RTnet support of ORTE out of the dark (kernel space) into the light (userspace) over fusion's POSIX skin. Some month ago, I tried to hack an experimental port which proved that this can work but was far from being usable for ORTE. We would first need some preparation of the build system of ORTE, then fixing remaining compile and maybe design issue (setup/cleanup) will be much easier. That work need not necessarily be contributed by the ORTE developers themselves (although they likely know best what to change)... Jan
signature.asc
Description: OpenPGP digital signature
