[LinuxPPS] LinuxPPS + kernel 2.6.34 = ok

[Udo van den Heuvel]

On 2010-05-24 14:49, Miroslav Lichvar wrote:
>> What oh what could be wrong?
> 
> I'm still not sure why do you need clear events when ntpd is syncing to
> assert.

I have no idea.
It is just different.
And I didn't change the config.

>>From one of the ntpq outputs you sent it seems it's working, just not
> very well. 

Much worse.

Offset for pps should be at least a factor 10 smaller, if not 100 or
1000. Same for jitter: at least a factor 10 if not 100 smaller with good
gps reception.

> Do you have the /dev/gpspp0 link?

I didn't change udev, see the wiki for the setup I use.
# ls -l /dev/*ps*
lrwxrwxrwx 1 root root      5 2010-05-24 14:20 /dev/gps0 -> ttyS0
lrwxrwxrwx 1 root root      4 2010-05-24 14:22 /dev/gpspps0 -> pps0
crw-rw---- 1 root tty  254, 0 2010-05-24 14:22 /dev/pps0
# ls -l /dev/ttyS0
crw-rw---- 1 root dialout 4, 64 2010-05-24 15:21 /dev/ttyS0

# setserial /dev/ttyS0
/dev/ttyS0, UART: 16550A, Port: 0x03f8, IRQ: 4, Flags: low_latency

# stty -a -F /dev/ttyS0
speed 4800 baud; rows 0; columns 0; line = 18;
intr = <undef>; quit = <undef>; erase = <undef>; kill = <undef>; eof =
<undef>; eol = <undef>; eol2 = <undef>; swtch = <undef>; start =
<undef>; stop = <undef>; susp = <undef>; rprnt = <undef>;
werase = <undef>; lnext = <undef>; flush = <undef>; min = 0; time = 0;
-parenb -parodd cs8 -hupcl -cstopb cread clocal -crtscts
ignbrk -brkint ignpar -parmrk -inpck -istrip -inlcr -igncr icrnl -ixon
-ixoff -iuclc -ixany -imaxbel -iutf8
-opost -olcuc -ocrnl -onlcr -onocr -onlret -ofill -ofdel nl0 cr0 tab0
bs0 vt0 ff0
-isig icanon -iexten -echo -echoe -echok -echonl -noflsh -xcase -tostop
-echoprt -echoctl -echoke

> Maybe the NMEA and PPS drivers are
> in conflict as they both try to use the same device. 

This setup has been running like this ever since LinuxPPS was started.

> You could try
> removing the symlink or using only the NMEA driver, or using only the
> PPS driver and let it sync to one of the NTP server instead.
> 
> You could also try to compensate the NMEA offset by setting time2 to
> 0.17 or -0.17.

I'll have a look into this.

Udo