[ntp:questions] Re: PPS working or not? FreeBSD 5.4

David L. Mills mills at udel.edu
Tue Jan 31 15:59:40 UTC 2006


With all due respect old chum, where have you been? I have been shrill 
to the point of extreme nastyness on getting rid of per-driver PPS. I'm 
sure I offended everybody with this agenda, especially on getting rid of 
the per-OS PPS interface in favor of the PPSAPI.

The downside of the per-driver PPS is the absence of reliable mitigation 
should something break in the radio or interface. The usual per-driver 
behavior is to take the seconds from the timecode and fraction from the 
PPS, possibly ignoring the timecode fraction, which could be and 
sometimes in my experience is bogus.


Harlan Stenn wrote:
> Dave,
> Just to be Crystal Clear, you are saying it would be best to lose the
> per-driver PPS code Everywhere and let everybody use the ATOM driver,
> correct?
> If so, I'll start working on making that happen.
> H

More information about the questions mailing list