[LinuxPPS] Re: LinuxPPS - odd behaviour. (Please Read)

[Rodolfo Giometti]

On Wed, Nov 07, 2007 at 12:04:57PM -0700, clemens at dwf.com wrote:
> I just read the comments in one of the nmea patch  postings, 
> and saw something that I hadn't realized before.
> 
> Namely, that the "serial" device (I dont know how to describe it in
> any other way,- possibly "parent") for a given pps<x> device
> must be open for the pps<x> device to function.
> 
> I checked it and that is true.
> 
> This is a truly odd Unix/Linux behaviour.  Isnt there any way
> around this 'problem' ?

No.

> It would mean, that IF one had the serial data on
>     /dev/ttyS0
> and the PPS data coming in on /dev/lp
> that one would have to open
>     /dev/ttyS0
>     /dev/lp
> and /dev/pps<x>
> 
> If nothing else, that's ugly.

The ugly part of this story is that the PPS devices are connected
through other devices (serial ports, parallel ports, and so on).

If your PPS devices was connected to a _dedicated_ device the
"problem" just vanished.

> To do it auto-magically is going to take some ugly code that
> is Linux specific in NTP and we've tried to avoid OS specific code
> whenever possible.

I think that the only solution is to propose a "main" and an
"alternate" device for all refclocks driver.

Ciao,

Rodolfo

-- 

GNU/Linux Solutions                  e-mail:    giometti at enneenne.com
Linux Device Driver                             giometti at gnudd.com
Embedded Systems                     		giometti at linux.it
UNIX programming                     phone:     +39 349 2432127