> > We'll also need to keep the clocks of machine A & B in sync, right? > > Here another type of synchronization could be used, for example > there's a reference implementation that uses TCP packets.
Just to make sure that how this works is clear, you dont need to run something like NTP between the two machines. It would not be precise enough anyway. We synchronize the traces (deduce the offset / drift) a posteriori based on the TCP packets exchanged. Because this synchronization takes into account several packet exchanges over the trace duration, the precision obtained is excellent, much better than the network average latency jitter. We worked on this for several years and now Genevieve implemented it for TMF recently. _______________________________________________ linuxtools-dev mailing list linuxtools-dev@eclipse.org https://dev.eclipse.org/mailman/listinfo/linuxtools-dev
