On 6/8/2020 10:00 AM, Vicente Ricardo (BT-CO/ETP1) via Linuxptp-users > The values of RMS look like they are overflowing, and the values of the > stressed CPU look like they perform better that an unstressed CPU. > So, it is quite possible a stressed out CPU will have less jitter due to scheduling. For example, if the thread that processes the NIC that is doing the PTP traffic is highly active, it may not go to sleep. This could potentially lead to a more stable Tx and Rx timestamp due to the threads not going to sleep in between processing. As for the RMS values, they appear to stabilize once the ptp4l process synchronizes to the slave state. Thanks, Jake _______________________________________________ Linuxptp-users mailing list Linuxptp-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linuxptp-users
