On Tue, Nov 28, 2023 at 03:34:54PM -0800, Richard Cochran wrote: > On Tue, Nov 28, 2023 at 12:53:38PM +0100, Miroslav Lichvar wrote: > > However, timestamping only sync messages has an advantage with very > > large number of clients as they don't have timestamp each other's > > delay requets and timestamping of sync messages is more reliable. > > Actually, in hardware is it much simpler and more efficient just to > time stamp every frame. (The reporting can be selectable if the > application doesn't care about some frame types)
I think it depends on the hardware design, which impacts its cost. If all hardware was able to accurately and reliably timestamp all packets, we wouldn't need PTP. PTP was specifically designed to make the hardware support as simple/cheap as possible. It separates event messages from the rest and uses multicast messaging in order to reduce the required timestamping rate. Timestamping only sync messages on clients is just another trick to reduce the required timestamping rate. The hardware can keep just one timestamp at a time and it doesn't matter that if it takes 20 milliseconds to read it over MDIO. It's good enough for clients, even if there is a very large number of them on the same communication path. -- Miroslav Lichvar _______________________________________________ Linuxptp-devel mailing list Linuxptp-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linuxptp-devel
