Michael Haberler wrote: > in addition to the rt-preempt measurements reported before, I built a xenomai > 'GPIO wiggler' in C, using the (actually working) xenomi hires timers, and > ran it on an idle rpi, and then a few times under some load (2 compiles). > > I collated everything here: http://linuxcnc.mah.priv.at/rpi/rpi-rtperf.html > > Well, these results are not great, but they aren't horrible, either. A mediocre X-86 CPU can do a whole lot worse. There is still this issue (which may not exist at all but is just a misreading of data) that a long string of tasks in an RT thread may have large delays due to cache purging. That may be the final hurdle to check out on ARM CPUs.
Jon ------------------------------------------------------------------------------ Got visibility? Most devs has no idea what their production app looks like. Find out how fast your code is with AppDynamics Lite. http://ad.doubleclick.net/clk;262219671;13503038;y? http://info.appdynamics.com/FreeJavaPerformanceDownload.html _______________________________________________ Emc-developers mailing list Emc-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-developers
