[LinuxPPS] Any Updates on that issue [PATCHv3 12/16] ntp: add hardpps implementation

[Felix Joussein]

is there any update?

I have made a curious observation:

My server uses the hardpps patch, and behaves like it, but shows some 
odd ntpq and ntptime output:

ntp_gettime() returns code 5 (ERROR)
   time d0595f32.fae7e5d0  Fri, Oct  8 2010  9:13:22.980, (.980101081),
   maximum error 1734 us, estimated error 0 us
ntp_adjtime() returns code 5 (ERROR)
   modes 0x0 (),
   offset 0.413 us, frequency 96.455 ppm, interval 4 s,
   maximum error 1734 us, estimated error 0 us,
   status 0x2007 (PLL,PPSFREQ,PPSTIME,NANO),
   time constant 4, precision 0.001 us, tolerance 500 ppm,
   pps frequency 0.000 ppm, stability 0.000 ppm, jitter 0.000 us,
   intervals 0, jitter exceeded 0, stability exceeded 0, errors 0.

Why do I get an error (return code 5 (ERROR)), if on the other hand 
side, the frequency is other then ZERO and the PPS Signal status is well 
synchronised?

      remote           refid      st t when poll reach   delay   offset 
  jitter
==============================================================================
+ptbtime1.ptb.de .PTB.            1 u    3   16  377   41.513   -0.070 
  0.254
+ptbtime2.ptb.de .PTB.            1 u    2   16  377   41.543   -0.030 
  0.360
oPPS(0)          .ATOM.           0 l    1   16  377    0.000    0.001 
  0.001
associd=0 status=01fd leap_none, sync_pps, 15 events, kern,
version="ntpd 4.2.6p2 at 1.2194-o Sat Oct  2 07:11:48 UTC 2010 (1)",
processor="i686", system="Linux/2.6.35.7", leap=00, stratum=1,
precision=-20, rootdelay=0.000, rootdisp=0.234, refid=ATOM,
reftime=d0595f9f.d9e15abb  Fri, Oct  8 2010  9:15:11.851,
clock=d0595fa0.8c5f9b88  Fri, Oct  8 2010  9:15:12.548, peer=24821, tc=4,
mintc=3, offset=0.001, frequency=96.457, sys_jitter=0.001,
clk_jitter=0.000, clk_wander=0.000, host="timemachine", flags=0x80001,
digest="md5", signature="md5WithRSAEncryption", update=201010080854,
cert="timemachine timemachine 0x0", until=201110060630


Out from the irc #ntp I was told, that the PPSSIGNAL flag is missing, 
and that it should be cleared when there was no pulse in last 10 seconds.

Any ideas?

regards
Felix