Re: [gpsd-dev] Port to QNX: PPS without SHM?

[Hal Murray]

You work mostly on gpsd.  It's your hammer.  I work mostly on ntpd.  It's my 
hammer.

address@hidden said:
> Not much point, gpsd json is new and problematic, keep the NTPSHM for now.
> You will find it a lot easier, less confusing, and less buggy than the gpsd
> json driver. 

What sort of problems are you seeing with JSON?  It works fine for me.

One of the reasons we (both gpsd and ntpd) put the JSON stuff together is to 
handle cases where shared memory didn't work.  If it doesn't work got this 
case we should fix it.   Maybe the documentation needs work.


> Yes, gpsd needs to be the ne getting the PPS signal.  Do not pass it
> directly to ntpd. 

Why not?  I often run gpsd without a PPS.  It does what I expect.

I don't see any reason why ntpd wouldn't work correctly if it gets the NMEA 
timing info via gpsd and the PPS directly.  That would let gpsd see the 
position info and pass it on to other clients.


> That would be a bad thing.  You have now bypassed all that gpsd does to
> clean up the PPS signal for you. 

ntpd knows how to process PPS signals.

----------

I'm not trying to say "don't feed the PPS to gpsd", but rather if there was a 
problem doing that, it should be possible to feed it directly to ntpd.  That 
might be a useful way to learn more and/or avoid needing to write some code.



-- 
These are my opinions.  I hate spam.