Gary E. Miller gem at rellim.com
Thu Sep 29 01:47:01 UTC 2016

Yo All!

I just ran a 6 hour test of --disable-kernel-pll.  See attached.

The left side before 18:45 is yesterday's ntpd.  After that
time is today's ntpd with --disable-kernel-pll.  Nothing else

As the png shows, the frequency jitter declined, but the offset
jitter increased.  Since ntpd cares more about time than frequency,
not a good trade off.

The knerel PLL and the ntpd userspace PLL are supposed to be using the
same algorithm and constants.  Might be worth checking if the constants
are still the same.

Full data for --disable-kernel-pll here: https://pi3.rellim.com/12h/

I'll have 24 hour data Thursday night.

Confounding the case is another RasPi that only changed the ntpd to
today's code.  It also shows increased time offset jitter, but not of
the same magnitude as the userspace PLL test.  It also shows increased
frequency offset jitter, unlike the userpsace PLL test.  So a bit hard
to tease apart the two worsening results.

Full data for the simple ntpd update here: https://pi2.rellim.com/12h/

Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97703
	gem at rellim.com  Tel:+1 541 382 8588
-------------- next part --------------
A non-text attachment was scrubbed...
Name: local-offset.png
Type: image/png
Size: 15725 bytes
Desc: not available
URL: <http://lists.ntpsec.org/pipermail/devel/attachments/20160928/1661cc99/attachment.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 473 bytes
Desc: OpenPGP digital signature
URL: <http://lists.ntpsec.org/pipermail/devel/attachments/20160928/1661cc99/attachment.bin>

More information about the devel mailing list