Hey Charles, The Xenomai tests were performed at the userspace level (not even kernelspace or xenomai-kernelspace). We wanted to make a quick test and porting all the drivers to Xenomai seemed like a lot of work.
I agree with you, PREEMPT seems to us specially comfortable and easy to keep up with the changes. 2014-09-18 14:07 GMT-07:00 Charles Steinkuehler <char...@steinkuehler.net>: > On 9/18/2014 3:43 PM, Víctor MV wrote: > > > > According to our benchmarking my recommendation will be to use PREEMPT > > because it's quite easy to activate/mantain. > > > > I'd be happy to attend any questions on this matter. This post > > <http://erlerobot.com/blog/beaglepilot-cyclictests-different-kernels/> > > could be of some interest for some of you. > > How did you manage to get 630 uS worst-case for Xenomai? I've never > seen a worst-case above 100 uS, and IIRC it's more like 75-80 uS > worst-case. > > Does your stress test include heavy IRQ/DMA usage (Ethernet, uSD/eMMC, > Display) or mostly CPU load? > > Regardless, it is much easier to work with PREEMPT or RT_PREEMT If those > meet your needs. > > -- > Charles Steinkuehler > char...@steinkuehler.net > > -- > For more options, visit http://beagleboard.org/discuss > --- > You received this message because you are subscribed to a topic in the > Google Groups "BeagleBoard" group. > To unsubscribe from this topic, visit > https://groups.google.com/d/topic/beagleboard/4eDQvQOkUkc/unsubscribe. > To unsubscribe from this group and all its topics, send an email to > beagleboard+unsubscr...@googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
