Hi Miroslav Lichvar,

Sorry for long log files. Next time I will upload less sized log.

> The reason for chronyc sources sometimes showing 2 as LastRx, even
> when no samples are lost, seems to be incorrect rounding of the sample
> time. I can fix that.

Will there be a release for this fix.
Current version I am using is 3.5 of chrony for which yocto recipe exists.

Thanks & Regards
Sarveshwar.K

On Thu, Feb 2, 2023 at 3:10 PM Miroslav Lichvar <[email protected]> wrote:

> On Thu, Feb 02, 2023 at 11:52:12AM +0530, sarveshwar k wrote:
> > Hi Miroslav,
> >
> > I have enabled the refclocks logs where I see the PPS signal are
> available.
> > And GPS UTC time comes very fast within a second.
> > Attaches is the refclocks logs. Can you please review once?
>
> Please don't post such large files to the list. Few hundred lines of
> each would be enough, or if needed, upload it on a file hosting
> service and just post the links here.
>
> The logs mostly look good to me. There are some larger offset, but
> it's difficult to tell what is the cause. Could be varying load on the
> CPU, unstable temperature, etc.
>
> The reason for chronyc sources sometimes showing 2 as LastRx, even
> when no samples are lost, seems to be incorrect rounding of the sample
> time. I can fix that.
>
> > CPU is running good and power saving features are not there. Not tried
> with
> > higher PPS rate. Using pps-gpio-poll as kernel module, I see the CPU gets
> > occupied waiting for the edge of clock right.
>
> Right, it's polling for the pin to change. It's more accurate, but
> comes at a cost of the CPU load.
>
> > Through which parameter of chronyc output can we know the synchronization
> > is in micro/nano/seconds?
>
> The RMS offset in tracking report is an indicator of stability. To
> determine accuracy you would need a more accurate time source, e.g. a
> NIC with HW-timestamping PPS input.
>
> > I have made makestep to step if difference is more than 100ns
> continuously,
> > with which I see the system time  is many times 0 seconds slower than NTP
> > time (from the output of tracking). Does this configuration effects
> > anything.
>
> That doesn't do anything useful. It's just stepping the clock to the
> current estimate of true time.
>
> --
> Miroslav Lichvar
>
>
> --
> To unsubscribe email [email protected]
> with "unsubscribe" in the subject.
> For help email [email protected]
> with "help" in the subject.
> Trouble?  Email [email protected].
>
>

Reply via email to