Re: [Linuxptp-users] time stamping accuracy of Freescale P2020
PTP IEEE 1588 stack for Linux
Brought to you by:
rcochran
From: Richard C. <ric...@gm...> - 2013-04-22 13:19:31
|
On Mon, Apr 22, 2013 at 12:26:08PM +0200, Mozhdeh Kamel wrote: > I think I found the answer: > > There is no hardware timestamping in VSC7385 L2 switch. Time stamping in > P2020 has done in the MAC before the FIFO, so it's still in hardware. Yes, the P2020 does MAC time stamping on all three ports. The switch in front of the one port is not a PTP switch, and so it will degrade the PTP performance slightly. > So, is this accuracy enough? or maybe it is enough for testing? I don't have a P2020, but I used to work with one in a PTP test lab. There we measured the master/slave offset using a PPS signal, and IIRC, the P2020 as slave synchonized to within 100 nanoseconds of the master. HTH, Richard |