Hi All, I'm completing a short piece of work examining the effects of running
software and hardware based PTP implementations in different virtualised
environments/under different conditions. The important thing to test is the
error between the synced virtual clock and UTC. Ideally I would be able to
record the time at the virtual clock at every pulse of a trusted clock, this is
a line of research I will follow but I am not sure if it'll be feasible. An
alternative option we are looking at is to use the difference between the
master-slave delay and the slave to master delay as an indicator of the likely
error, similar to this paper on NTP. I have a few questions I am hoping you can
help with: 1. Does this seem like a sensible measure? Intuitively it makes
sense to me but I am not familiar enough with PTP to be certain I am not
missing something. 2. Does ptp4l provide some means of extracting this
information? PMC seems to provide one half of this: the meanPathDelay being the
average master-slave delay but not the other And on another topic 3. Is there
anyway for a slave to request a re-sync? If for example it knows it is out of
sync. My understanding is that the sync's are always started by the
grandmaster. Thanks in advance for your help, All the best,
_______________________________________________
Linuxptp-users mailing list
Linuxptp-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linuxptp-users
