Re: [MM qmi-support] "Set Operating Mode" on shutdown and more..

>>> I actually got the connection up now using the method above (mmcli +
>>> ifup wwan0) so it seems to be close to functional now.
>> Regarding connectivity; QMI-powered MM is fully functional. Missing
>> things include messaging and location, which are scheduled for next
>> weeks.
> Okay. What about selection of network technology (2/3/4G/LTE) and
> preference between them?
> I am not sure if this command is the one to use, but it does not work:
> $ mmcli -m 1 --set-preferred-mode lte
> error: setting preferred mode requires list of allowed modes

You need to set a list of allowed modes and only then you can select one
of them to be preferred. The modem may not support the combination you
provide, though. You set allowed modes with "--set-allowed-modes"; e.g.:

 --set-allowed-modes=2g|3g --set-preferred-mode=3g

> I see:
> ModemManager[27013]: <warn>  couldn't load current allowed/preferred
> modes: 'Loading allowed modes is not supported by this device'

Seems your device doesn't have a recent enough NAS service version in
order to be able to select allowed modes.

>>> How is the status of a NM that can use this new MM? Which parts/tasks
>>> are missing there?
>> Like, everything is missing. The idea is to modify NM to talk to MM
>> through the new libmm-glib. My personal list of things TODO includes now
>> 2 big items only: multipart support merge in MM git master and then the
>> NM integration. Any help with the second one is very appreciated, now
>> that all MM plugins are ported.
>>> Btw: seems like the latest MM branch code did not build.
>> Make sure you have the latest libqmi (from git master) before trying to
>> build the qmi-support branch in ModemManager :-)
> I did, but maybe hit a bad timing. Anyway, master version now builds and
> connection still works with my ZTE modem, but I got a crash with the
> Sierra modem (which is not Icera):
> ERROR:mm-port-probe.c:531:serial_probe_at_icera_result_processor:
> assertion failed: (g_variant_is_of_type (result, G_VARIANT_TYPE_BOOLEAN))
> which means that it is unusable with this modem now (at
> 72602a395105006736ecf5829ba33ffcefce3692).

Interesting; can you get me debug logs of the issue? Or better, a
stacktrace after running ModemManager with G_DEBUG=fatal_warnings; e.g.:
sudo G_DEBUG=fatal_warnings gdb --args /usr/sbin/ModemManager --debug



[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]