> > The number of switches is related to the number of tasks running in this > test, nofpu reduces this number. So that is ok.
I understand this, but both tests X2.4.9+PPC2.4.25 and X2.5.6+PPC-2.6.34 I run without(!) the option --nofpu. And in both cases there were 22 tasks. With X2.4.9+2.4.25 I got about 1170 CTX-switches and with 2.5.6+2.6.34 about 4300 CTW-switches. Only one test I run with --nofpu option (X2.5.6+PPC2.6.34) (mail form 13:25 o´clock)! and got about 2200 CTX-switches and there the difference is due to reduced number of taks. >The problem with this > test is that switches/sec values are sampled by a regular linux thread > which nanosleeps, so at least over 2.4, the delay is not accurate. So > the number of switches observed can't be either. > > I still have to check over 2.6 + hires if we can still explain this 2% > offset the same way. But in (2.6.34)+2.5.6 the display task is still an ordinary linux task (__real_pthread_create, as far as I can see). Why should it be more accurate? Sorry for steeling your time and asking stupid questions. Roderik > > > > > >Spare time has become a luxury over the last months. I'll try to find a > > >time slot next week to have a look at this again. > > > > Tanks a lot! > > I don´t dare to say, but I am on vacation for the rest of this and the > > next week, so > I only can do testing as of 4. of July again. > > > > Ok. > > > Roderik > > > > ---------------------------------------------------------------------------- > > Xenomai 2.4.9 on PPC-2.4.25: > > RW24:/fat/sbin # switchtest > > == Testing FPU check routines... > > r0: 1 != 2 > > r1: 1 != 2 > > r2: 1 != 2 > > r3: 1 != 2 > > r4: 1 != 2 > > r5: 1 != 2 > > r6: 1 != 2 > > r7: 1 != 2 > > r8: 1 != 2 > > r9: 1 != 2 > > r10: 1 != 2 > > r11: 1 != 2 > > r12: 1 != 2 > > r13: 1 != 2 > > r14: 1 != 2 > > r15: 1 != 2 > > r16: 1 != 2 > > r17: 1 != 2 > > r18: 1 != 2 > > r19: 1 != 2 > > r20: 1 != 2 > > r21: 1 != 2 > > r22: 1 != 2 > > r23: 1 != 2 > > r24: 1 != 2 > > r25: 1 != 2 > > r26: 1 != 2 > > r27: 1 != 2 > > r28: 1 != 2 > > r29: 1 != 2 > > r30: 1 != 2 > > r31: 1 != 2 > > == FPU check routines: OK. > > == Threads: sleeper_ufps-0 rtk-1 rtk-2 rtk_fp-3 rtk_fp-4 rtk_fp_ufpp-5 > rtk_fp_ufpp-6 rtup-7 rtup-8 rtup_ufpp-9 rtup_ufpp-10 rtus-11 rtus-12 > rtus_ufps-13 > rtus_ufps-14 rtuo-15 rtuo-16 rtuo_ufpp-17 rtuo_ufpp-18 rtuo_ufps-19 rtuo_ufps- > 20 rtuo_ufpp_ufps-21 rtuo_ufpp_ufps-22 > > RTT| 00:00:01 > > RTH|ctx switches|-------total > > RTD| 1150| 1150 > > RTD| 1150| 2300 > > RTD| 1173| 3473 > > RTD| 1173| 4646 > > RTD| 1150| 5796 > > RTD| 1173| 6969 > > RTD| 1150| 8119 > > RTD| 1173| 9292 > > RTD| 1150| 10442 > > RTD| 1173| 11615 > > RTD| 1173| 12788 > > RTD| 1150| 13938 > > RTD| 1173| 15111 > > RTD| 1173| 16284 > > RTD| 1150| 17434 > > RTD| 1150| 18584 > > RTD| 1150| 19734 > > RTD| 1173| 20907 > > RTD| 1150| 22057 > > RTD| 1150| 23207 > > RTD| 1150| 24357 > > RTT| 00:00:22 > > RTH|ctx switches|-------total > > RTD| 1173| 25530 > > RTD| 1173| 26703 > > RTD| 1173| 27876 > > RTD| 1173| 29049 > > RTD| 1150| 30199 > > RTD| 1150| 31349 > > RTD| 1150| 32499 > > RTD| 1173| 33672 > > RTD| 1173| 34845 > > RTD| 1173| 36018 > > RTD| 1150| 37168 > > RTD| 1150| 38318 > > RTD| 1150| 39468 > > RTD| 1173| 40641 > > RTD| 1150| 41791 > > RTD| 1150| 42941 > > RTD| 1173| 44114 > > RTD| 1150| 45264 > > RTD| 1150| 46414 > > RTD| 1150| 47564 > > RTD| 1173| 48737 > > RTT| 00:00:43 > > RW24:/fat/sbin # cat /proc/xenomai/version > > 2.4.9 > > RW24:/fat/sbin # cat /proc/xenomai/sched > > CPU PID PRI PERIOD TIMEOUT TIMEBASE STAT NAME > > 0 0 -1 0 0 master R ROOT > > 0 0 99 200000000 45608637 master D rt-watchdog > > RW24:/fat/sbin # cat /proc/xenomai/stat > > CPU PID MSW CSW PF STAT %CPU NAME > > 0 0 0 38989 0 00500080 98.3 ROOT > > 0 0 0 701 0 00000084 0.0 rt-watchdog > > 0 0 0 14730 0 00000000 0.1 IRQ256: [timer] > > RW24:/fat/sbin # > > > > > > > -----Ursprüngliche Nachricht----- > > > Von: Gilles Chanteperdrix [mailto:gilles.chanteperd...@xenomai.org] > > > Gesendet: Dienstag, 21. Juni 2011 13:33 > > > An: Wildenburg, Roderik RAEK1 MRA > > > Cc: r...@xenomai.org; xenomai-help@gna.org > > > Betreff: Re: AW: AW: [Xenomai-help] Xenomai 2.5.6 with PPC-Kernel 2.4.25 > > > > > > On 06/21/2011 01:22 PM, roderik.wildenb...@manroland.com wrote: > > > >> 2% off seems a lot for a transient load, and this would not happen on a > > > >> periodic basis anyway. This needs to be investigated. Could you run > > > >> switchtest in nofpu mode? > > > > > > > > Still fluctuates about 2% (see !! mark) but now more erratic. The > > > > reduced > > > number of context switches is caused by the reduced number of tasks? > > > > > > switchest works by switching context between several tasks. One of this > > > task does sleep, in order to avoid starving linux, and also prints the > > > numbers approximately every second. > > > > > > When there are less task, we enter the sleeping task more often, so, > > > yes, there are less context switches. > > > > > > But more importantly, since, the time when we print the result is so > > > imprecise, some variations are normal, so, chances are that the 2% > > > variation is normal. > > > > > > -- > > > Gilles. > > > > -------------------------------------------------------- > > manroland AG > > Vorsitzender des Aufsichtsrates: Hanno C. Fiedler > > Vorstand: Gerd Finkbeiner (Vorsitzender), Dr. Ingo Koch, Dr. Markus Rall, > > Paul > Steidle > > Sitz der Gesellschaft: Offenbach am Main, Registergericht: Amtsgericht > Offenbach HRB-Nr. 42592 > > USt-Ident-Nr. DE 250200933 > > > > -- > Philippe. > > -------------------------------------------------------- manroland AG Vorsitzender des Aufsichtsrates: Hanno C. Fiedler Vorstand: Gerd Finkbeiner (Vorsitzender), Dr. Ingo Koch, Dr. Markus Rall, Paul Steidle Sitz der Gesellschaft: Offenbach am Main, Registergericht: Amtsgericht Offenbach HRB-Nr. 42592 USt-Ident-Nr. DE 250200933 _______________________________________________ Xenomai-help mailing list Xenomai-help@gna.org https://mail.gna.org/listinfo/xenomai-help
