Re: [chrony-users] PPS Synchronization Verification

[ Thread Index | Date Index | More chrony.tuxfamily.org/chrony-users Archives ]


On Mon, Nov 02, 2020 at 08:02:12AM -0800, Bill Unruh wrote:
> 
> On Mon, 2 Nov 2020, Miroslav Lichvar wrote:
> 
> > On Mon, Nov 02, 2020 at 12:24:32PM +0000, Chang, Benjamin wrote:
> > > The PPS signal is coming from the NTP server (it's a syncserver S650) and is connected via a 232 line. I essentially did this https://www.crc.id.au/2016/09/24/adding-a-pps-source-to-ntpd/ but switched the ntpd kernel with the chrony kernel. Should I set an offset of .00002 for the PPS?
> 
> 20 microseconds for an interrupt? That seems pretty high. One tests I ran
> about 10 years ago, (sending a timed signal over on the parallel port lines to
> the serial port interrupt line) the latency was closer to or less than 1 microsecond.

Modern hardware seems to be worse. There are other delays than the CPU
handling an interrupt, e.g. due to various bridges between the CPU and
UART. On one machine I have measured an offset of about 27
microseconds between PPS on serial port and hardware-timestamped NTP
(accurate to 1us).

-- 
Miroslav Lichvar


-- 
To unsubscribe email chrony-users-request@xxxxxxxxxxxxxxxxxxxx 
with "unsubscribe" in the subject.
For help email chrony-users-request@xxxxxxxxxxxxxxxxxxxx 
with "help" in the subject.
Trouble?  Email listmaster@xxxxxxxxxxxxxxxxxxxx.


Mail converted by MHonArc 2.6.19+ http://listengine.tuxfamily.org/