[LinuxPPS] performance

James Boddington boddingt at optusnet.com.au
Tue Oct 17 03:45:24 CEST 2006


gnu not unix wrote:
> I have only a few days of running 2.6.18 linux kernel, but
> my experience was that the system would take a long time to
> close the offset, was very touchy, and could get into a state
> where it never seemed to properly stabilize.

Are you getting any sensible results with 2.6.18? It seems that every time I 
reboot my time server with 2.6.18 I get different drift values. So far low 
-30s, -23, -31, -61 and now -50ppm. Except for small fluctuations due to 
temperature change the drift stays relatively constant until the machine is 
rebooted. Had it running for 5 days when the drift was staying around -23ppm. 
The offset generally stayed below 12us. The different drift values after each 
reboot is strange.

2.4.33.3 with ppskit and 2.6.17 with linuxpps I get the same drift values as I 
normally get. About -71ppm for 2.4.33.3 and a bit under -68.5ppm for 2.6.17. 
Both kernels the daily movement is 0.5ppm as the day heats up and cools for the 
night.

Had 2.6.17 running for the last couple of hours and it has settled back down to 
it's usual range of -68.5 to -69ppm.

-- 
    James



More information about the LinuxPPS mailing list