On 2 Feb 2019 at 1:54, Artem Panfilov wrote: > Hi Frantisek, > Thank you for your work! I have to mention you on the mailing list. > > I found your paper "Hardware Assisted IEEE 1588 Clock Synchronization > Under Linux", and there is similar work "Precision Time Protocol for > Spectroscope Synchronization" from ZHAW university. > > I'm currently working on stratum-1 time server with GPSDO on board. > The GPSDO driver exposes PHC interface, where timescale is TAI. We > have two Synopsys NICs on board that have external timestamping > support which are disciplined with GPSDO PHC clock. The results are > accurate enough. I could share measurement results taken from Calnex > Sentinel. > > I don't know how external timestamping support fits linuxptp project. > It's very rare use-case, but I would like to have it in upstream. > Thanks
Dear Artem, apologies for splashing water into your fireplace. Apart from me being irrelevant to linuxptp maintenance, personally I have nothing against having your hooks accepted into upstream code. (Per analogiam, thinking about how many odd things exist in the Linux kernel for instance...) The motivation for having things upstream is clear. Please note that I am not the author of "Hardware Assisted IEEE 1588 Clock Synchronization Under Linux" - that was written by Ferencz Balint, who's a proper academic researcher. I'm just a private tinkerer. His work predates my own attempts by several years. I have learned the crucial bits from his work. He's the pioneer, I am his humble apprentice. We are not the same person and we have never met. Frank _______________________________________________ Linuxptp-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/linuxptp-devel
