On 07/14/2004 02:25 AM Paolo Mantegazza wrote: > Wolfgang Grandegger wrote: >> On 07/13/2004 02:29 PM Philippe Gerum wrote: >> >>>On Wed, 2004-07-07 at 13:37, Wolfgang Grandegger wrote: >>> >>>>Hi, >>>> >>>>I have now commited a first port of RTAI over ADEOS/ppc to the vesuvio >>>>branch. One first observation is, that the latencies and task switch >>>>times are almost doubled on slow PowerPC processors (MPC860 at 50 MHz). >>> >>>>I will give some more information on this port including performance >>>>figures later on. >>>> >>> >>>It would be interesting to know if the IRQ latencies are also observable >>>using the irq_jitter test (or some adaptation of) from the Adeos distro >>>(linux/examples/* IIRC). >> >> >> These figures look resonable but I cannot compare them directly with the >> pre-ADEOS case. From my point of view, the bigger task switch times and >> latencies are simply due to additional code to be processed, which shows >> up on slow processors (with little caches, etc.). Are there any figures >> for x86? I will provide some preliminary ones for PPC today or tomorrow. >> > > The machine I've (PIII 1GHz) shows about .5 us UP switching time for > LXRT, full Linux process, and .35 us for an RTAI proper task (not safe > to be used in production work). The latter figure is practically the > same as for ksched_up. I roughly recall that NEWLXRT had something about > 4 us on a plain Pentium 200 MHz for full Linux processes.
As mentioned in a previous mail to the list, it would be interessting to directly compare RTAI/RTHAL with RTAI-ADEOS on a slow x86 Processor like the 200 MHz Pentium you mentioned. Sometimes ago Philippe presented some figures at FOSDEM (http://download.gna.org/adeos/doc/fosdem-2003.pdf) and they also show some notable difference (for the latencies). Wolfgang.
