You mentioned you were using igb. I believe that driver still relies on a work
queue task to handle the Tx timestamps, as well as overflow check.
If your device needs the overflow check, and it gets skipped that could result
in very wild incorrect timestamping results. However, I do not believe that is
the case with the i210 so I think that can be ruled out.
I am not sure what else could be causing clock instability or variable delay in
the packet transmit and receive handling within the hardware to the point that
it impacts the PTP calculation..
To be exact my network driver is "igb" whose version is 5.4.0. I saw
that there is a new version (5.5.2). Is it convenient to update the
version or I will not notice any change?
Everything is OK when I run ptp4l and phc2sys, under 100ns for both,
that is totally synchronized. But at the moment I run 4 threads of
stress-ng by adding CPU load, suddenly I lose sync. I have to say that
if I force ptp4l and phc2sys to run in one core and the threads in other
cores, everything is OK, but the case is that I want to simulate an
environment where other real-time is sharing cores with ptp4l-phc2sys.
Do you know what I mean?
Thank you very much for your hel, Jacob. I appreciate it. Do not
hesitate to reply again.
Diego G.
_______________________________________________
Linuxptp-users mailing list
Linuxptp-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linuxptp-users
