MC7750 libqmi
Tim Harvey
tharvey at gateworks.com
Mon Nov 4 23:37:33 CET 2013
Hi Dan - thanks for the response!
On Mon, Nov 4, 2013 at 12:55 PM, Dan Williams <dcbw at redhat.com> wrote:
> On Mon, 2013-11-04 at 11:19 -0800, Tim Harvey wrote:
>> Heath,
>>
>> I saw your thread here
>> http://lists.freedesktop.org/archives/libqmi-devel/2012-November/000386.html
>> on the libqmi maillist archives and was wondering what the status is?
>>
>> I have a MC7750 that I've established a connection with using libqmi
>> v1.6 but it seems after some time traffic stops being passed and
>> 'qmi-network /dev/cdc-wdm0 status' shows disconnected. If I do a
>> 'qmi-network /dev/cdc-wdm0 start' traffic resumes again. It sounds
>> like this is a different issue than you encountered? I have the same
>> results if I hack qmicli to add QMI_DEVICE_OPEN_FLAGS_NET_802_3 and
>> QMI_DEVICE_OPEN_FLAGS_NET_NO_QOS_HEADER it seems to behave the same.
>> Its possible that I don't have an adequate signal strength as I'm not
>> clear yet how to determine with libqmi what radio band I'm connected
>> with and what an appropriate signal level is.
>
> I've got a 7750 that works fine here; how long is "after some time"?
> I've used mine for at least a couple hours at a time, though only
> stationary. The LTE signal is almost always good (better than -70dbm)
> which might be why it works so well :)
I always seem to drop the connection within 20 minutes. At that point
sometimes I can do a 'qmi-network /dev/cdc-wdm0 start' to restart it,
other times that results in a 'error: cannot re-start network, PDH
already exists' and I have to do a qmi-network stop first.
>
> First, what firmware version?
>
> qmicli -d /dev/cdc-wdm0 --dms-get-revision
Revision: 'SWI9600M_01.00.09.03AP R2492 CARMD-EN-10526 2011/07/01 19:31:09'
I questioned if this was indeed the cmd to get the firmware revision
as the above looks very old. I believe the latest is 3.5.10.9 Sept 11
2013. I'll have to dig into the docs to figure out how to update it
(I believe I have to use a windows host and app)
>
> To determine signal strength and other information, you want:
>
> qmicli -d /dev/cdc-wdm0 --nas-get-signal-strength
> qmicli -d /dev/cdc-wdm0 --nas-get-serving-system
> qmicli -d /dev/cdc-wdm0 --nas-get-signal-info
>
root at OpenWrt:/# qmicli -d /dev/cdc-wdm0 -V
qmicli 1.6.0
Copyright (2012) Aleksander Morgado
License GPLv2+: GNU GPL version 2 or later
<http://gnu.org/licenses/gpl-2.0.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.
root at OpenWrt:/# qmicli -d /dev/cdc-wdm0 --nas-get-signal-strength
[/dev/cdc-wdm0] Successfully got signal strength
Current:
Network 'cdma-1x': '-96 dBm'
Other:
Network 'cdma-1xevdo': '-63 dBm'
RSSI:
Network 'cdma-1x': '-96 dBm'
Network 'cdma-1xevdo': '-63 dBm'
ECIO:
Network 'cdma-1x': '-7.0 dBm'
Network 'cdma-1xevdo': '9.0 dBm'
IO: '-52 dBm'
SINR: (8) '9.0 dB'
root at OpenWrt:/# qmicli -d /dev/cdc-wdm0 --nas-get-serving-system
[/dev/cdc-wdm0] Successfully got serving system:
Registration state: 'registered'
CS: 'attached'
PS: 'attached'
Selected network: '3gpp2'
Radio interfaces: '2'
[0]: 'cdma-1xevdo'
[1]: 'cdma-1x'
Roaming status: '(null)'
Data service capabilities: '2'
[0]: 'cdma'
[1]: 'evdo-rev-a'
CDMA System ID:
SID: '40'
NID: '4'
CDMA Base station info:
Base station ID: '5008'
Latitude: '0.000000'*
Longitude: '0.000000'*
Roaming indicators: '2'
[0]: '(null)' (cdma-1x)
[1]: '(null)' ((null))
Default roaming status: '(null)'
3GPP2 time zone:
Leap seconds: '16' seconds
Local time offset: '-480' minutes
Daylight saving time: 'no'
CDMA P_Rev: '6'
root at OpenWrt:/# qmicli -d /dev/cdc-wdm0 --nas-get-signal-info
error: couldn't get signal info: QMI protocol error (71): 'InvalidQmiCommand'
I believe I recall --nas-get-signal-info working in the past - perhaps
this was changed between libqmi v1.0 and what I'm on now (v1.6)?
Thanks,
Tim
> Most likely you don't care what radio band you're using, you just care
> what access technology (cdma, evdo, umts, lte) you're using and what the
> signal strength is.
>
> Dan
>
More information about the libqmi-devel
mailing list