On Fri, Apr 27, 2007 at 02:56:16PM +0200, Johan Borkhuis wrote: > As I understood from several emails the arch/ppc branch is not supported > by the RT-patch, and also will not because of missing functionality. My > BSP (MVME3100, PPC 8540) is based on the ppc architecture however, but I > do need hard realtime performance. > > As far as I know there are a couple of choices I have: > - Port the BSP to the powerpc architecture branch Which is the future proof variant, because then every kernel release will work for your board - minus bugs :) > Before I start on the port of the BSP I would like to know what the > status of the RT-patch on PowerPC is. How good and stable does the patch > run on PowerPC, and how good is the real time performance on the PowerPC? We have it running on MPC5200B, and the 'cyclictest -n -p 90 -t 4' worst case numbers are between 50 and 100 us. For more reliable numbers wait a few days, we'll re-test with the recently released 2.6.21 soon and I can post the results here. > Also I like to know how the RT-patch and Xenomai compare. I did see some > measurements and discussions from a couple of years back between Ingo > and Kristian Benoit. The RT patch and Xenomai had comparable results, > but how is the current situation? Are the current results substantial > better for one of these or for both, or are they still on the same level? I don't think there is significant performace difference, although I cannot provide numbers. Nevertheless, don't only take latencies into account but also the fact that development goes on quickly and one has to have a maintainable system for the future. So IMHO mainline is a better way, at least on a long term perspective. Robert -- Dipl.-Ing. Robert Schwebel | http://www.pengutronix.de Pengutronix - Linux Solutions for Science and Industry Handelsregister: Amtsgericht Hildesheim, HRA 2686 Hannoversche Str. 2, 31134 Hildesheim, Germany Phone: +49-5121-206917-0 | Fax: +49-5121-206917-9 - To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
