On Wed, Feb 24, 2021 at 02:42:12PM +0000, Coughlin, David S wrote: > We sourced a grandmaster clock board from a vendor that advertised > GPS integration. We are using an ICD-GPS-153 compliant receiver, and > I had to rewicker ts2phc to interact with it. Most of my changes > were built off of ts2phc_nmea_master.c . We were reconciling our > times with external clocks. Our external clocks were behind what we > were getting out of ts2phc by 40+ seconds. I looked through the > code and saw that it was calling lstab_utc2tai() . I pulled the > thread on that, and saw that a good chunk of our delta is due to > adding back leap seconds.
As the name of the function suggests, the purpose of this call is to convert UTC time values into TAI. > My systems engineer thinks we should just straight convert the UTC > time tag to PTP time. Correct, and that is what the function call does. > Can you shed some light on the decision to add the leap seconds in > when using a GPS upstream master? The reason is that, according to the IEEE 1588 standard, the PTP time scale is TAI. HTH, Richard _______________________________________________ Linuxptp-devel mailing list Linuxptp-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linuxptp-devel
