Re: [gpsd-dev] autobaud

[Gary E. Miller]

Yo Eric!

On Tue, 29 Oct 2013 14:17:39 -0400
"Eric S. Raymond" <address@hidden> wrote:

> Gary E. Miller <address@hidden>:
> > It worked a few days ago.  No time to bisect, I'm busy just finding
> > your bugs.  You broke it, you fix it.  I have more bugs and fixes
> > in my queue but not nailed down enough to report.
> 
> Autobauding to 38400 is working here with a BU-353 (SiRF II).  In
> gpsmon I can switch from 4800 to 38400 and back with the expected
> results.

I have a BU-350P.  A SiRF III.

> To make double sure, I set the BU-353 to 38400, used stty to set the
> port speed to 9600, and then watched gpsd -D 5 -N -n sync to 38400
> successfully.

That is what I have.  port at 9600, SiRF at 38400, it never gets there.

> I did notice that the hunt sequence took noticeably longer than it had
> last time I tested this - each hunt step was a bit over a second.  I
> think this is because we keep increasing the size of the
> incoming-packet buffer to deal with various strange non-NMEA devices
> - this implicitly increases the amount of data the autobaud logic
> sniffs before stepping to the next speed/parity/stopbits possibility.

So how many steps is that to 38400?

> Maybe you're not waiting long enough?

I waited several minutes several times.

I'm gonna keep looking at chrony today.  Got some changes in testing
and some doc cleanup in the works.  Plus soome nasty intermittent
failures that migh be related to priv dropping.

RGDS
GARY
---------------------------------------------------------------------------
Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97701
        address@hidden  Tel:+1(541)382-8588

signature.asc
Description: PGP signature