Re: [gpsd-users] GPS fix last to long

[Krzysztof Drobiński]

On 21.08.2012 11:01, Krzysztof Drobiński wrote:
> Hi Gary,
> thanks for replay.
>
> On 21.08.2012 00:44, Gary E. Miller wrote:
> > Yo Krzysztof!
> >
> > On Sat, 18 Aug 2012 14:29:18 +0200
> > Krzysztof Drobiński <address@hidden> wrote:
> >
> > > Hi All,
> > > I'm using ISM420 receiver. GPSD version is 2.95. I was wondering why
> > > first location reporting by GPSD lasts so long.
> > That can take up to 30 mins from a cold start.
> >
> > > Analysis of output from serial interface shows that time to first fix
> > > is more / less 45s but gpsd reports location after 180s.
> > Serial or USB connected GPS?  If serial, try settng the port speed
> > before starting gpsd.  That saves the auto-baud time.
> Do you mean by compile time --enable-fixed-port-speed option? I 
> already set it. It's fixed to receiver's default baud (4800)
> > > My question is how gpsd decides when position is ready to report?
> > When it gets a position, it reports it.  Which is  AFTER it has 
> > determined the
> > proper port speed and GPS type.
> I think now I understand what causes the delay. gpspipe reports NMEA 
> sentences after 180s as well (after I restarted gpsd).
>
> # gpspipe -r
>
> netlib_connectsock() returns socket on fd 3
>
> {"class":"VERSION","release":"2.95","rev":"2012-08-21T06:02:44","proto_major":3,"proto_minor":3} 
>
>
> {"class":"DEVICES","devices":[{"class":"DEVICE","path":"/dev/ttyS1","activated":1345532244.64,"native":0,"bps":0,"parity":"N","stopbits":1,"cycle":1.00}]} 
>
>
> {"class":"WATCH","enable":true,"json":false,"nmea":true,"raw":0,"scaled":false,"timing":false} 
>
>
> {"class":"DEVICE","path":"/dev/ttyS1","activated":1345532393.81,"driver":"Generic 
> NMEA","native":0,"bps":4800,"parity":"N","stopbits":2,"cycle":1.00}
>
>
> So the issue could be probing the serial port / receiver settings. 
> Last sentence
>
> {"class":"DEVICE","path":"/dev/ttyS1","activated":1345532393.81,"driver":"Generic 
> NMEA","native":0,"bps":4800,"parity":"N","stopbits":2,"cycle":1.00}
>
> occur after about 3 minutes, then NMEA sentences are reported.
> It is weird because baudrate is already set during compilation and log 
> above
> shows that gpsd is still trying to probe terminal parameters..
> > RGDS
> > GARY
> > --------------------------------------------------------------------------- 
> >
> > Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97701
> > address@hidden Tel:+1(541)382-8588
Here is the solution. Attached patch explains wrong behaviour in case of 
fixed baudrate - maybe someone will find it useful.
The problem was that gpsd is using '0' speed for getting default linux 
terminal settings even if  FIXED_PORT_SPEED is defined (It took about 3 
minutes for hunting first valid packet in my case). Additionally there 
is no possibility to set stopbits value (I found it is implemented in 
latest gpsd but there is no such feature in 2.95).

Regards,
Krzysztof

gpsd-2.95-serial.patch
Description: Text Data