Ian Thompson
2017-01-09 20:53:51 UTC
We are running some embedded boards based on the Altera SOC with its stmicro/stmmac.
The v1.7 code runs well.
Jan 9 19:24:29 localhost user.info ptp4l: [265569.544] selected /dev/ptp0 as PTP clock
Jan 9 19:24:29 localhost user.notice ptp4l: [265569.545] port 1: INITIALIZING to LISTENING on INITIALIZE
Jan 9 19:24:29 localhost user.notice ptp4l: [265569.545] port 0: INITIALIZING to LISTENING on INITIALIZE
Jan 9 19:24:29 localhost user.notice ptp4l: [265569.668] port 1: new foreign master 000cec.fffe.0a085d-1
Jan 9 19:24:31 localhost user.notice ptp4l: [265571.668] selected best master clock 000cec.fffe.0a085d
Jan 9 19:24:31 localhost user.notice ptp4l: [265571.668] port 1: LISTENING to UNCALIBRATED on RS_SLAVE
I've just compiled v1.8 and replaced the v1.7 binary, no other changes, and it fails ....
Jan 9 19:19:07 localhost user.info ptp4l_18: [265247.971] selected /dev/ptp0 as PTP clock
Jan 9 19:19:07 localhost user.info ptp4l_18: [265247.973] driver rejected most general HWTSTAMP filter
Jan 9 19:19:07 localhost user.err ptp4l_18: [265247.973] ioctl SIOCSHWTSTAMP failed: Numerical result out of range
Jan 9 19:19:07 localhost user.notice ptp4l_18: [265247.973] port 1: INITIALIZING to FAULTY on INITIALIZE
Jan 9 19:19:07 localhost user.notice ptp4l_18: [265247.974] port 0: INITIALIZING to LISTENING on INITIALIZE
Jan 9 19:19:07 localhost user.notice ptp4l_18: [265247.974] port 1: link up
Jan 9 19:19:07 localhost user.info ptp4l_18: [265247.975] driver rejected most general HWTSTAMP filter
Jan 9 19:19:07 localhost user.err ptp4l_18: [265247.975] ioctl SIOCSHWTSTAMP failed: Numerical result out of range
Jan 9 19:19:07 localhost user.notice ptp4l_18: [265247.975] port 1: FAULTY to FAULTY on FAULT_CLEARED
The function that generates the failed ioctl error hasn't changed, so is this a problem with the added RTNL link code?
Any help is appreciated.
Regards
Ian T.
The v1.7 code runs well.
Jan 9 19:24:29 localhost user.info ptp4l: [265569.544] selected /dev/ptp0 as PTP clock
Jan 9 19:24:29 localhost user.notice ptp4l: [265569.545] port 1: INITIALIZING to LISTENING on INITIALIZE
Jan 9 19:24:29 localhost user.notice ptp4l: [265569.545] port 0: INITIALIZING to LISTENING on INITIALIZE
Jan 9 19:24:29 localhost user.notice ptp4l: [265569.668] port 1: new foreign master 000cec.fffe.0a085d-1
Jan 9 19:24:31 localhost user.notice ptp4l: [265571.668] selected best master clock 000cec.fffe.0a085d
Jan 9 19:24:31 localhost user.notice ptp4l: [265571.668] port 1: LISTENING to UNCALIBRATED on RS_SLAVE
I've just compiled v1.8 and replaced the v1.7 binary, no other changes, and it fails ....
Jan 9 19:19:07 localhost user.info ptp4l_18: [265247.971] selected /dev/ptp0 as PTP clock
Jan 9 19:19:07 localhost user.info ptp4l_18: [265247.973] driver rejected most general HWTSTAMP filter
Jan 9 19:19:07 localhost user.err ptp4l_18: [265247.973] ioctl SIOCSHWTSTAMP failed: Numerical result out of range
Jan 9 19:19:07 localhost user.notice ptp4l_18: [265247.973] port 1: INITIALIZING to FAULTY on INITIALIZE
Jan 9 19:19:07 localhost user.notice ptp4l_18: [265247.974] port 0: INITIALIZING to LISTENING on INITIALIZE
Jan 9 19:19:07 localhost user.notice ptp4l_18: [265247.974] port 1: link up
Jan 9 19:19:07 localhost user.info ptp4l_18: [265247.975] driver rejected most general HWTSTAMP filter
Jan 9 19:19:07 localhost user.err ptp4l_18: [265247.975] ioctl SIOCSHWTSTAMP failed: Numerical result out of range
Jan 9 19:19:07 localhost user.notice ptp4l_18: [265247.975] port 1: FAULTY to FAULTY on FAULT_CLEARED
The function that generates the failed ioctl error hasn't changed, so is this a problem with the added RTNL link code?
Any help is appreciated.
Regards
Ian T.