[LinuxPPS] ntp update

[Paul]

I don't know how this 'bug' ever got given any serious consideration. If
the pps pulse comes before or after depends on the hardware, and
adjusting either way is trivial. But - much more serious is that some
hardware 'jumps' by whole seconds (easy to spot - suddenly you are a
whole second off the rest of the internet), you adjust the delay and
after a day or so it jumps again. I use the Jupiter module and it
displays this behaviour, so my module is only wired for pps with
disambiguating from network servers (which is pretty infrequent as I
have a network of peers).

On Thu, 2009-10-15 at 19:27 +0200, Udo van den Heuvel wrote:
> On 2009-10-15 19:19, Udo van den Heuvel wrote:
> > On 2009-10-15 19:15, Reg Clemens wrote:
> >>> https://support.ntp.org/bugs/show_bug.cgi?id=610
> >>>
> >>> So please try the mentioned ntp-dev version(s) if you use the NMEA driver.
> >>>
> >>
> >> Im not too sure how much good this is going to do.
> >> Someone else (not me) reported problems with the KERNEL patch, in that they
> >> were not getting 'ticks'.
> >>
> >> I have reported that time_pps_create fails with the current patch.
> > 
> > This is about the patch to ntpd, the refclock NMEA driver.
> 
> I was asked to add to the test request:
> 
> Minimally they should
> be warned to test this with an OLDER version of the KERNEL, and not
> the current one.
> 
> So please do not use 2.6.31 or newer for now to test this ntp-dev
> version for NMEA refclock.
> 
> Udo
> 
> _______________________________________________
> LinuxPPS mailing list
> LinuxPPS at ml.enneenne.com
> http://ml.enneenne.com/cgi-bin/mailman/listinfo/linuxpps
> Wiki: http://wiki.enneenne.com/index.php/LinuxPPS_support