ogpsd questions

[Daniel Willmann]

On Tue, 30 Sep 2008 00:42:21 +0200
Luca Capello <luca at pca.it> wrote:

> Hi there!
> 
> On Tue, 30 Sep 2008 00:10:10 +0200, Daniel Willmann wrote:
> > On Thu, 4 Sep 2008 00:14:38 +0200
> > Sascha Wessel <wessel at nefkom.net> wrote:
> >> * TODO: power off the device when frameworkd is shutting down
> >>   (`/etc/init.d/fso-frameworkd restart` while the device is
> >> powered on confuses frameworkd)
> >
> > This shouldn't be a problem any more as we force GPS off at ogpsd
> > startup. Of course turning off GPS on regular shutdown would still
> > be nice.
> 
> Can I disagree?  Why can't frameworkd cope with the GPS status?  I
> mean, I don't understand why the GPS cannot be on *before* starting
> frameworkd.  What's the advantage of frameworkd "predating" the GPS?

If GPS was on before frameworkd started it might also be configured in
a way that is bad for us. For example maybe UBX as input is disabled,
how are we going to configure the GPS now?

> The situation should be similar for GSM, bluetooth and WiFi, right?

Well, for GSM we were running into problems already. If GSM was on and
the SIM ready messages were already sent (phonebook, messagebook access
ready) there is *no* way whatsoever to query this status. There's more
stuff like this so basically you want to reset the modem to a known
state. And manage from there.

The bottom line is that one daemon will have to be in charge of the
resource and that daemon will want to make sure it knows exactly which
state it's device is in. If you want to play with GPS without ogpsd
intervening, then simply disable that subsystem.
Or better, help me make the information that you want to access
available through ogpsd.

> >> * Debian bug #494927
> >>   - sometimes disabling of nmea in ubx.py doesn't work
> >>   - zhone shows 90.0/0.0 as location
> >
> > Okay, these are two problems.
> 
> I'm sorry, but wasn't that Debian bug already fixed [1]?

Yeah, I replied too soon. I was even the one who applied the patch. :-)

> > The Discarded data not UBX is pretty much unavoidable since when we
> > power the GPS it will start to output NMEA data. ogpsd configures
> > in- and output to only UBX, but before this command goes through
> > the GPS chip already printed its welcome message.  I could lower
> > the log level of that message to DEBUG, though, so it wouldn't
> > confuse people.
> 
> Yes, please.

Okay, will do as soon as I'm sure that the GPS is put into UBX mode
every time. There are some problems with it at the moment.

Regards,
Daniel Willmann
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 197 bytes
Desc: not available
URL: <http://lists.linuxtogo.org/pipermail/smartphones-standards/attachments/20080930/058ade00/attachment.pgp>