Hi,
I am aware that I am sending this mail to the developers list. I have sent
this to the users list and have got no response. People on this mailing
list might have more info on this application.

We have a setup of a master PC(Linux OS) and slave embedded HW running
Linux OS connected via media gateway(not time aware) and synchronizing
using IEEE 1588. With no application running on the HW, or dedicating a
specific core to ptp, we are able to synchronize the clocks between the two
systems. But when we have our application that uses ethernet bandwidth both
on the receive and transmit side running on all cores, we see very frequent
timeouts. We found that the timeout is happening when the slave responds to
master and waits for the follow-up message from master within 1ms or 10ms
(depending on the tx_timestamp_timeout parameter), the timeout occurs on
the socket. Can you please recommend any way of prioritizing ptp packets
over other traffic?

On a different note, is there an interface or some sort of notification
that identifies when the timestamp registers are corrected so that the
other applications can use this notification to schedule other tasks?

Thanks,
Mani
_______________________________________________
Linuxptp-devel mailing list
Linuxptp-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linuxptp-devel

Reply via email to