Yes, I am using imx6. Here is the output of "ethtool -i eth0": # ethtool -i eth0 driver: fec version: Revision: 1.0 firmware-version: expansion-rom-version: bus-info: eth0 supports-statistics: yes supports-test: no supports-eeprom-access: no supports-register-dump: yes supports-priv-flags: no
I am using the vendor kernel. Using a mainline kernel would be a major undertaking due to the build system in place, so I'd like to explore other options first. Incidentally, I initially had the boards connected via a switch that was connected to the corporate network and I was not seeing these large master offsets. I then realized that the switch probably didn't have ports that support HW PTP, so that's why I connected the boards directly and that's when I started seeing this problem. Thank you, Irene. -----Original Message----- From: Richard Cochran [mailto:richardcoch...@gmail.com] Sent: Thursday, July 30, 2020 3:21 PM To: Irene Kravets <ire...@sapling-inc.com> Cc: linuxptp-users@lists.sourceforge.net Subject: Re: [Linuxptp-users] Large master offset on the PTP slave On Thu, Jul 30, 2020 at 05:38:30PM +0000, Irene Kravets wrote: > I am seeing a problem that I hope I can get some help with. I am using two > embedded boards, each has a HW capable Ethernet port. I connected the two > ethernet ports to each other and configured one board to be the PTP master > and the other PTP slave. I am seeing very large master offset values on the > slave even after letting the boards run for a while. > > This is what I see on the master from ptp4l: > > Jul 30 16:12:03 imx6sxsabresd I assume you are using an imx6? Maybe the kernel driver is buggy. What does 'ethtool -i eth0' say? Are you using the vendor kernel? If so, I suggest trying a mainline kernel instead. Thanks, Richard _______________________________________________ Linuxptp-users mailing list Linuxptp-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linuxptp-users
