As tx timestamp timeout is something multiple users encounter, provide information for configuring kthread priorities in manual for ptp4l where tx_timestamp_timeout is already described.
Signed-off-by: Lukasz Plachno <lukasz.plac...@intel.com> --- ptp4l.8 | 36 ++++++++++++++++++++++++++++++++++++ 1 file changed, 36 insertions(+) diff --git a/ptp4l.8 b/ptp4l.8 index 09ff108af102..5e201be3f793 100644 --- a/ptp4l.8 +++ b/ptp4l.8 @@ -1032,6 +1032,42 @@ to maintain the correct offset between UTC and PTP times. See .BR phc2sys (8) manual page for more details. +.SH KTHREAD PRIORITY + +In case of following log: + +.RS +.I timed out while polling for tx timestamp +.I increasing tx_timestamp_timeout or increasing +.I kworker priority may correct this issue, +.I but a driver bug likely causes it +.RE + +one of possible causes is kworker processing timestamps being starved, +user might try manually increasing the priority of the kworker. + +Usually drivers use kworker created by PTP stack in Linux kernel, +those kworkers are named: +.I ptp<decimal number of clock> +, +user can manually bump the priority of the kworker process using chrt. + +Example: +.RS +\f(CWpgrep \-f "ptp[0-9]+" | xargs \-I {} sudo chrt \-f \-\-pid 75 {}\fP +.RE + +Intel ice driver may create multiple kworkers for one physical NIC and +names those processes differently. + +Example for Intel E810 card: +.RS +\f(CWpgrep \-f ice-ptp | xargs \-I {} sudo chrt \-f \-\-pid 75 {}\fP +.RE + +Assigning priority needs careful consideration as assigning too high +priority to any task might make system unstable. + .SH SEE ALSO .BR pmc (8), .BR phc2sys (8) -- 2.34.1 _______________________________________________ Linuxptp-devel mailing list Linuxptp-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linuxptp-devel