Re: [gpsd-dev] another git head problem

[Eric S. Raymond]

Gary E. Miller <address@hidden>:
> Yo Eric!
> 
> On Thu, 10 Oct 2013 17:22:07 -0400
> "Eric S. Raymond" <address@hidden> wrote:
> 
> > Live-testing with a SiRF puck and uBlox TCX0 on different USB ports
> > looks good here - command line 
> > 
> >       gpsd -D 5 -N /dev/ttyACM0 /dev/ttyUSB0
> 
> I only put one device on the command line, not two.  And yours
> are so similar if a mis-id happens it would probably not
> be noticeable.

Huh?  One is a SirF device, the other a uBlox.  The packet frames are
different - different header bytes.  So are the log messages from the
packet analysis.
 
> > Can you suggest a way to reproduce your problem?
> 
> I suspect is the key is having the Garmin USB in the mix.  Confirmed since
> cgps show gpsd id'ed the GR-601W as Garmin binary.

That, on the other hand, is plausible.  The Garmin packets aren't
checksummed; this makes it easy for the packet getter to get a
false positive match.

> One thing that has needed doing is getting libusb to get us the
> vendorid:productid of the current dev.  That way less confusion if 
> a Garmin USSB is in the mix.

Agreed.  Can you allocate any time for this?
-- 
                <a href="http://www.catb.org/~esr/";>Eric S. Raymond</a>

signature.asc
Description: Digital signature