[LinuxPPS] another refclock_nmea patch ([Bug 610] Generic NMEA GPS Receiver driver w/ PPS expects PPS after NMEA data)

[Udo van den Heuvel]

Rodolfo Giometti wrote:
> On Tue, Nov 06, 2007 at 05:58:45PM +0100, Udo van den Heuvel wrote:
>> Regardless: why does atom on it's own not work?
>> Why is `cat` needed? Or NMEA?
> 
> Because your PPS line is connetced with the serial line and it must be
> opened in order to report PPS events.

So this new LinuxPPS situation of always splitting data and PPS is not
handy here. It requires extra work where it is not necessarily needed.

Only in the non-standard cases of GPIO, etc extra effort is required and
only then changes to the standard habits should be needed.

> This because you use a serial port as PPS source input device, if you
> use a dedicated device, a CPU's GPIO for instance, you don't need this
> dirty trick.

So if LinuxPPS /dev/pps0 is for /dev/ttyS0 and we use atom, LinuxPPS
does not (under water) open the serial port to make sure

> 
> This can be avoided if you add to the ATOM refclock the possibility to
> define a main and an alternate device as is in NMEA.
> 
> Ciao,
> 
> Rodolfo
>