Kernel PLL, from IRC

Gary E. Miller gem at
Wed Sep 28 22:44:53 UTC 2016

Yo Hal!

On Wed, 28 Sep 2016 15:38:58 -0700
Hal Murray <hmurray at> wrote:

> gem at said:
> >>  New samples arrive every second.  
> > No.  Twice a second.  But now you are confusing the PPS with the
> > PLL.   
> PPS is pulse per second.  There is only 1 pulse each second.  Where
> is the other data coming from?

Two edges per pulse, one pulse per second.  And if the system clock is 
running a bit slow it will miss many PPS.  chronyd slews the clock
up to 8.5, so you thn miss a lot of pulses.

> Are you thinking of the other edge?  ntpd disables collecting data on
> the unused edge.

I hope not.  gpsd sure uses the information from both edges, if it
can get it.  But the bigger problem is the missed PPS.

> > And a LOT more jitter.  
> I think you have jumped from my theoretical description of 2 options
> to an experiment with real data.  What change did you make?

Already discussed at great length on gpsd, and easily demonstrated with
ntpshmmon.  When you start missing edges your time quality degrades.

Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97703
	gem at  Tel:+1 541 382 8588
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 473 bytes
Desc: OpenPGP digital signature
URL: <>

More information about the devel mailing list