negative latency

[Philippe Gerum]

On Wed, 2004-06-02 at 10:48, b3nuts wrote:
> Hi all,
> 
> Thank you for your kind help. I understand getting
> negative latencies is quite normal thanks to your
> explanations. By taking a closer look at the original
> code I saw it could have been expected. 
> 
> Now, I performed a test on an Intel Celeron 1.2 gHz
> with kernel 2.4.22, adeos rtai3.0r4, 1 proc, UP
> scheduler.
> 
> Here is what I did for the calibration step :
> 
> root at _:/usr/realtime/calibration# ./calibrate -r -k -c
> 
> *** NUMBER OF REQUESTS: 3. ***
> 
> *** '#define SETUP_TIME_8254 2949', IN USE 2011 ***
> 
> 
> *** KERNEL SPACE LATENCY CALIBRATION, wait 5 seconds
> for it ... ***
> 
> *** '#define LATENCY_8254 2948' (IN USE 4700)
> 
> 
> ->>> FREQ CALIB (PRINTING EVERY 5 SECONDS, press enter
> to end calibrating) <<<-
> 
> 
> ->>> MEASURED CPU_FREQ: 1199145665 (Hz) [5 (s)], IN
> USE 1199172000 (Hz) <<<-
> 
> ->>> MEASURED CPU_FREQ: 1199145894 (Hz) [10 (s)], IN
> USE 1199172000 (Hz) <<<-
> 
> ->>> MEASURED CPU_FREQ: 1199145858 (Hz) [15 (s)], IN
> USE 1199172000 (Hz) <<<-
> 
> 
> *** '#define CPU_FREQ 1199145858', IN USE 1199172000
> ***
> 
> 
> 
> 
> And then, here are the latency results in kernel space
> :
> 
> 
> root at _:/usr/realtime/testsuite/kern/latency# insmod
> rt_process.o
> root at _:/usr/realtime/testsuite/kern/latency# ./run
> *
> *
> * Type ^C to stop this application.
> *
> *
> 
> ## RTAI latency calibration tool ##
> # period = 100000 (ns)
> # avrgtime = 1 (s)
> # check overall worst case
> # do not use the FPU
> # start the timer
> # timer_mode is oneshot
> 
> 2004/06/2 10:27:51   min:    -2073    max:    49022   
> average:    -1928
> 2004/06/2 10:27:51   min:    -2075    max:    49038   
> average:    -1921
> 2004/06/2 10:27:51   min:    -2075    max:    52678   
> average:    -1921
> 2004/06/2 10:27:51   min:    -2075    max:    52678   
> average:    -1668
> 2004/06/2 10:27:51   min:    -2075    max:    52678   
> average:    -1864
> 2004/06/2 10:27:51   min:    -2075    max:    52678   
> average:    -1640
> 2004/06/2 10:27:51   min:    -2078    max:    52678   
> average:    -1591
> 2004/06/2 10:27:52   min:    -2078    max:    52678   
> average:    -1933
> 2004/06/2 10:27:53   min:    -2078    max:    52678   
> average:    -1909
> 

Are you running the test in a X-window environment, or on a plain
console/VT?

> 
> Compared to what I've been reading so far, I would
> rank these results to... awful  :-)
> 
> Did I miss something else ?
> The USB support is enabled but I'm not sure disabling
> it would cover up the whole problem. 

It's worth trying anyway, because USB is a known source of high jittery.

> 
> Any further help would be greatly appreciated. Thank
> you in advance.
> Kind regards,
> 
> Marc.
> 
> 
> 
> 
> --- Steven Krekels <steven.krekels at alcatel.be> a
> écrit : > Have a look at:
> > 
> >
> https://mail.rtai.org/pipermail/rtai/2003-September/005021.html
> > 
> > Cheers,
> > 
> > 	Steven.
> 
> 
> 	
> 
> 	
> 		
> Yahoo! Mail : votre e-mail personnel et gratuit qui vous suit partout ! 
> Créez votre Yahoo! Mail sur http://fr.benefits.yahoo.com/
> 
> Dialoguez en direct avec vos amis grâce à Yahoo! Messenger !Téléchargez Yahoo! Messenger sur http://fr.messenger.yahoo.com
> 
> _______________________________________________
> RTAI mailing list
> RTAI at rtai.org
> https://mail.rtai.org/cgi-bin/mailman/listinfo/rtai
-- 

Philippe.