On Thu, Jan 3, 2013 at 1:37 PM, Aleksander Morgado <[email protected]>wrote:

>
> >     > Also, based on my experiments, the modem sometimes seems to hang
> when
> >     > running AT+COPS=?. This patch avoids AT+COPS=? from being issued
> when
> >     > the modem is in LTE mdoe.
> >
> >     From what I can see, it also avoids the Scan when the access tech is
> >     unknown:
> >
> >     +    if ((access_tech == MM_MODEM_ACCESS_TECHNOLOGY_UNKNOWN) ||
> >     +        (access_tech & MM_MODEM_ACCESS_TECHNOLOGY_LTE)) {
> >
> >     So you're really limiting the scanning for networks only to when the
> >     modem is already connected to a given network which is not LTE. I
> don't
> >     think we shouldn't ban from scanning when access tech is UNKNOWN.
> >
> >
> > I include UNKNOWN just to be safe as there may be a period of time when
> > the access technology is not yet determined.
> >
>
> That's what I mean; it should be perfectly valid to be able to scan for
> networks when the modem is not yet registered anywhere, i.e. when the
> access tech is unknown.
>
>
Let me remove the check for unknown access technology.


>  >     Also, the reply to AT+COPS=? may really take a loong time; you sure
> it
> >     gets stuck?
> >
> >
> > From what I've seen before, the scan operation timed out and the modem
> > didn't respond to AT commands anymore. However, that's not frequently
> > reproducible.
> >
>
> So this patch is just avoiding the need to parse an empty list as the
> AT+COPS=? command replies just OK by default when in LTE mode. Is it
> really worth the complication of subclassing the scan method just to
> avoid that? :-/
>
>
The only purpose of this patch is to skip "AT+COPS=?" in LTE mode as that
causes the modem to sometimes behave weirdly :-/



> >
> >     I wonder how the scan works when the Novatel LTE modem is handled by
> the
> >     QMI driver. IIRC Dan said we wouldn't need this specific plugin any
> more
> >     once the modem is handled by the QMI implementation.
> >
> >
> > I haven't tried the QMI driver for this modem, but wonder if it behaves
> > in a similar way as the limitation is most likely a firmware issue.
> >
>
> It would be great if you could check the overall behaviour of the
> Novatel LTE modem with the QMI driver, to see if there is any reason not
> to switch to it...
>
>
Will do


>  --
> Aleksander
>
_______________________________________________
networkmanager-list mailing list
[email protected]
https://mail.gnome.org/mailman/listinfo/networkmanager-list

Reply via email to