[ntp:questions] Stratum 0 refclocks and PPS
mikes at flatsurface.com
Wed May 9 01:39:25 UTC 2012
On 5/8/2012 9:07 PM, Dave Hart wrote:
> I'm not familiar with the elan-mmcr driver. From your description it
> appears you have a single hardware PPS signal which is exposed to
> software via two different means (perhaps because it's connected to
> both a RS-232 input pin and another PPS input?).
The Soekris NET4501 has a high precision (125 ns?) hardware timer. The
elan-mmcr PPS driver was written by Poul Henning-Kemp (I think) to take
advantage of that. PPS starts the timer, then when the interrupt is
handled, does some math to remove the interrupt latency from the PPS
timestamp. The PPS signal also goes to the serial port DCD, which is
what the Oncore works with, so yes, PPS is available in two places.
The one difference I've noticed is that the regular (DCD) PPS does both
asserts and clears. The elan-mmcr only signals asserts. That works fine
with the Atom driver, but it seems the Oncore driver has a problem with
it (just guessing that may be the cause).
Might there be anyway to force NTP to take one PPS source over the
other? The docs say "While the rules do not forbid it, it is usually not
useful to designate more than one source as preferred; however, if more
than one source is so designated, they are used in the order specified
in the configuration file; that is, if the first one becomes
unselectable, the second one is considered and so forth. This order of
priority is also applicable to multiple PPS drivers, multiple modem
drivers and even multiple local drivers, although that would not
normally be useful." But, setting both Atom and Oncore to prefer doesn't
seem to make a difference (Atom appears first). I'm guessing because
Atom can't be used without a prefer peer which provides TOD.
More information about the questions