On Wed, May 09, 2018 at 04:56:29PM +0200, Miroslav Lichvar wrote: > I'm trying to measure the maximum throughput of ptp4l handling unicast > delay requests and I'm seeing some unexpected (at least for me) > behavior. When the rate of requests reaches a threshold, ptp4l > switches to the fault state after missing the TX timestamp of its own > sync message, which is sent just once per second. From tcpdump output > it doesn't look like the sync message was dropped due to a full queue. > > Delay responses don't generate TX timestamps (unlike NSM responses). > How can a high rate of RX timestamps cause a TX timestamp to be > missed? Is this a common HW limitation? I see this happening with > three different drivers (tg3, igb, ixgbe).
What kind of "high" packet rate triggers this problem? Thanks, Richard PS I'm partially offline this week and might not respond right away... ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot _______________________________________________ Linuxptp-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/linuxptp-devel
