[chrony-users] asymmetric phy delay?

[Nuno Gonçalves <nunojpg@xxxxxxxxx>]

I have 2 distinct SOCs, both running as close as possible
kernel and same chrony.

Both are PPS synchronized. And tests have been done with and without
nohz=off on the cmdline.

I have multiple boards of each SOC on a local network checking each
other with noselect.

My results are very surprising.

SOC A to A shows a offset of about 50us+-25us
SOC A to B shows a offset of about 0us+25us
SOC B to B shows a offset of about 75us+-25us
SOC B to A shows a offset of about 130us+-25us

The two conclusions is that for a pair of boards of the same type, the
delays appear to be identical and not symmetric as expected and that
when using boards of different types the estimated delays are
different in both sides.

I understand that a eventual asymmetric tx/rx delay on the phy could
cause this unbalance, but my understanding of the NTP protocol is that
this would cancel each other on a query/response cycle of the same
board type, which is not happening.

Any idea what is causing this and if I can do anything to calibrate it?

Thanks

Attachment: B monitors B.png
Description: PNG image

Attachment: B monitors A.png
Description: PNG image

Attachment: A monitors A.png
Description: PNG image

Attachment: A monitors B.png
Description: PNG image