Hi Pekka,
On Wed, Aug 18, 2010 at 07:34:07PM +0300, Pekka Pessi wrote:
> Hi Samuel,
>
> 2010/8/18 Samuel Ortiz <[email protected]>:
> >> Try to get IMSI immediately after SimManager interface is available.
> >>
> >> Listen to SIM events in Ofono plugin, add device when IMSI
> > I suppose you want to do that in order for ConnMan to be able to control GSM
> > devices through the Online property ? That is to say have it controlling
> > both
> > GPRS enabled and GPRS less modems ? If that's so I'd really appreciate to
> > get
> > this sort of details from either the log message above or from a 0/6 patch
> > explaining in details where you're trying to go.
>
> OK, I'll be more explicit about the nature of the changes.
Thanks in advance.
> > Some additional comments:
> >> - DBG("path %s imsi %s", path, imsi);
> >> + DBG("path %s imsi %s", path, imsi ? imsi : "<missing>");
> > That's not needed, you'll get a "nil" string here.
>
> I do get a crash with libglib 2.24.1 on Ubuntu, that is why the change.
weird, but fair enough.
> >> connman_device_set_string(device, "Path", path);
> >> + connman_device_set_string(device, "IMSI", imsi); /* XXX */
> > What's the XXX for ?
>
> My intention was to use not-yet-exposed connman_device_get_ident()
> instead of separate IMSI string.
Ok, then adding that as a comment would be a good idea.
> >> @@ -1126,7 +1162,7 @@ static void modem_properties_reply(DBusPendingCall
> >> *call, void *user_data)
> >> if (!powered)
> >> modem_change_powered(path, TRUE);
> >>
> >> - if (has_gprs)
> >> + if (has_sim)
> >> get_imsi(path);
> > So here get_imsi will end up creating a device which then will query the
> > GPRS
> > interface and try to build networks out of that. Would it make sense to
> > remove
> > check_networks() from add_device() then ? Networks would only be built once
> > the GPRS interface is up. I guess that would work with the current code and
> > if
> > not we should add the networks once we get the GPRS interface from
> > modem_changed().
>
> I'll do that. It just seemed to work fine, I did not read the
> debugging log close enough.
It probably works fine, it just seems more logical to create network once we
know they're actually existing.
Cheers,
Samuel.
--
Intel Open Source Technology Centre
http://oss.intel.com/
_______________________________________________
connman mailing list
[email protected]
http://lists.connman.net/listinfo/connman
