[LinuxPPS] So near, and yet so far...

[David]

I used to have working serial pps, but abandoned the old patch series.

I have applied patches 1 to 9 of 11, all fine

modprobe'd pps_ldisc - fine

ldattach 18 /dev/ttyS1 - ok

I then got one message/second in the log

Sep 18 00:10:15 astro PPS assert at 370434 on source #0
Sep 18 00:10:16 astro PPS event on source 0 at 1253229016.989842753
Sep 18 00:10:16 astro capture assert seq #56 for source 0
Sep 18 00:10:16 astro PPS assert at 370684 on source #0
Sep 18 00:10:17 astro PPS event on source 0 at 1253229017.989844304
Sep 18 00:10:17 astro capture assert seq #57 for source 0

Looking good...

Copied pps.h and timepps.h into user space and rebuild ntp 4.2.4p7 - it
reported timepps to be 'yes'

I use the TSIP parse clock and PPS - in /etc/ntp.conf I have
...
# to break the ambiguity, use server slaved to demon
server 192.168.69.12 iburst

#the gps TSIP feed
server 127.127.8.0 mode 10 prefer burst
fudge 127.127.8.0 refid TSIP time1 0.040

# the gps PPS
server 127.127.22.0 iburst minpoll 4 maxpoll 4
fudge 127.127.22.0 refid PPS0
...

which has worked in the past.

As soon as I restart ntp, the capture messages cease, and I get log messages

Sep 18 00:10:21 astro ntpd[8245]: ntpd 4.2.4p7 at 1.1607-o Thu Sep 17
22:57:48 UTC 2009 (1)
Sep 18 00:10:21 astro ntpd[8246]: precision = 1.000 usec
Sep 18 00:10:21 astro ntpd[8246]: Listening on interface #0 wildcard,
0.0.0.0#123 Disabled
Sep 18 00:10:21 astro ntpd[8246]: Listening on interface #1 lo,
127.0.0.1#123 Enabled
Sep 18 00:10:21 astro ntpd[8246]: Listening on interface #2 eth0,
192.168.69.4#123 Enabled
Sep 18 00:10:21 astro ntpd[8246]: kernel time sync status 0040
Sep 18 00:10:21 astro ntpd[8246]: PARSE receiver #0: new phase
adjustment 0.040000 s
Sep 18 00:10:21 astro PPS_GETPARAMS: source 0
Sep 18 00:10:21 astro PPS_GETCAP: source 0
Sep 18 00:10:21 astro PPS_SETPARAMS: source 0
Sep 18 00:10:21 astro PPS_GETCAP: source 0
Sep 18 00:10:21 astro PPS_SETPARAMS: source 0
Sep 18 00:10:21 astro ntpd[8246]: frequency initialized -129.754 PPM
from /etc/ntp/ntp.drift
Sep 18 00:10:22 astro PPS_FETCH: source 0
Sep 18 00:10:22 astro timeout 0.000000000

with the fetch / timeout pair repeating for ever

/dev/pps0 does exist, the serial port interrupts are counting one per
second, and ntpd locks to the TSIP serial data successfully.

Is this a known bug, and is there anything I can do to help fix it?

Many thanks

--David