Hi, I was wondering whether someone could shed some light on an RTR based approach. If (which I hope not but do fear) the 200 microsecond Rx-ISR engagement time cannot be reduced or overcome, would an RTR approach offer an alternative?
The thought: If I make an RTR call to my node at the beginning of the period and catch the response all within say 250microsec, I have 750 micoseconds for task "work" as long as I can suppress any further disruption from the Rx-ISR (i.e. block the rx interrupt or something during that period) I have never used RTR before but can imagine that it relies on the normal receiving mechanism of the CAN driver to be up which of course would not help. Are there any generic (userspace) C calls with which interrupts can be disabled (e.g. ioctl kind of things) which I could use to keep the Rx-ISR quiet during the 750ms? Regards, Roland. _______________________________________________ Xenomai-help mailing list [email protected] https://mail.gna.org/listinfo/xenomai-help
