On 3/8/2020 7:42 AM, Richard Cochran wrote: > On Sun, Mar 08, 2020 at 10:25:34AM +0700, Michael Tang wrote: >> HW/SW in slave: >> - CPU: Intel Core i9, runs CentOS 7 with LinuxPTP 2.0 > > Looks like cent os 7 has a really old kernel version, and > >> - NIC: an Intel NIC supports hardware timestamp. I can't remember exactly >> the model name of NIC now. > > IIRC, there are Intel MACs that have bugs when reading the time. > Actually, these bugs do not affect the time stamping, iirc, and so > they do not affect ptp4l but rather phc2sys. You should check your > kernel version and Intel MAC and confirm that you have a working > combination. >
Some of the old Intel MACs to have bugs like this, but I don't recall any on the 10Gb ones, offhand. However, it would be nice to know what board you do have, so we can see if there are any known issues. Since the time jumped by a full second, it could be that something else modified the clock, or that the remote system had a 1second clock jump..? Thanks, Jake > Thanks, > Richard > > > _______________________________________________ > Linuxptp-users mailing list > Linuxptp-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/linuxptp-users > _______________________________________________ Linuxptp-users mailing list Linuxptp-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linuxptp-users
