SymbX/ModemManager
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
5,095 Commits 2 Branches 0 Tags
738deebfb4586ffff560d538f7809cf2fdee7132
Commit Graph
2778 Commits
Author SHA1 Message Date
Aleksander Morgado 14bb687b94 iface-modem-cdma: fix use of uninitialized variable 2013-03-22 08:17:48 +01:00
Aleksander Morgado 3b66047d94 broadband-modem-qmi: implement automatic CDMA activation 2013-03-22 08:12:59 +01:00
Aleksander Morgado 6252df0bef broadband-modem-qmi: new generic power cycle action 2013-03-22 08:12:59 +01:00
Aleksander Morgado 3e5251cad7 broadband-modem-qmi: implement initial CDMA activation state loading 2013-03-22 08:12:59 +01:00
Aleksander Morgado 5cc98341ad iface-modem-cdma: load initial activation state during init 2013-03-22 08:12:59 +01:00
Aleksander Morgado 240dc12fd2 iface-modem-cdma: allow reporting activation state changes 2013-03-22 08:12:43 +01:00
Aleksander Morgado ad757bd3bc bearer-qmi: don't try to gather connection settings if we didn't connect 2013-03-21 17:31:11 +01:00
Aleksander Morgado aa84ce98f7 blacklist: ignore Arduino devices 
https://bugs.launchpad.net/bugs/910736
https://bugs.launchpad.net/bugs/1153632
 2013-03-20 21:36:34 +01:00
Aleksander Morgado 0b39072b76 error-helpers: avoid crash when error is not recognized 
Just remove the last element with NULL strings in the array of errors, as we're
safe using G_N_ELEMENTS() to iterate it.

Reported by Jose Maria Gonzalez Calabozo <jmgonzalezc@indra.es>
 2013-03-20 18:16:10 +01:00
Aleksander Morgado f27dbda4a7 iface-modem-cdma: fix checking support for manual activation 2013-03-20 09:50:34 +01:00
Aleksander Morgado b64fc4b588 blacklist: ignore West Mountain radio devices 
https://bugs.launchpad.net/bugs/1154654
 2013-03-13 17:06:10 +01:00
Aleksander Morgado 66c5416e3c broadband-modem-qmi: fix logic to include GSM/WCDM acquisition order preference 2013-03-12 18:50:20 +01:00
Aleksander Morgado 5758fafaf5 modem-helpers-qmi: fix allowed mode translation from QMI to MM 
Which actually fixes allowed mode switching in MM...
 2013-03-12 12:22:22 +01:00
Aleksander Morgado 4b2577fe29 broadband-modem-qmi: fix initial operating mode loading 2013-03-12 11:01:27 +01:00
Aleksander Morgado 2bcc647d3e broadband-modem-qmi: handle 'UimUninitialized' error when checking unlock status 
QMI modems without SIM may report 'UimUninitialized' QMI protocol errors, so
catch those and use them as 'SIM failure' so that they get reported to the user.
 2013-03-12 10:59:14 +01:00
Ben Chan 0dea5b5d72 broadband-modem: fix enable flag in UnsolicitedRegistrationEventsContext 2013-03-06 09:37:51 +01:00
Aleksander Morgado 3ebc8413b0 udev: tag all devices from idVendor 0x0711 as manual-scan-only 
Magic Control Technology Corp (0x0711) manufacturers several USB<->RS232
adapters.
 2013-03-05 19:41:08 +01:00
Aleksander Morgado 8450b7b2a3 core: don't automatically probe ports of USB<->serial adapters 
We should not automatically probe ports marked as coming from USB to serial
adapters, as we're not sure that a modem is behind the adapter. Still, let the
user request a manual scan and have these devices probed in that case.

https://bugzilla.gnome.org/show_bug.cgi?id=647556
https://bugzilla.gnome.org/show_bug.cgi?id=691076
 2013-03-05 19:41:08 +01:00
Ben Chan 3678ae5378 iface-modem-3gpp: handle non-deferrable registration state updates 
This patch changes MMIfaceModem3gpp to differentiate between deferrable
and non-deferrable 3GPP registration state updates. Periodic or
unsolicited registration state updates are deferrable, while internal
updates, e.g. due to modem being disabled, are non-deferrable.
 2013-03-05 19:31:17 +01:00
Ben Chan 1e67312aaf broadband-modem: fix disabling of unsolicited registration events 2013-03-05 19:16:05 +01:00
Aleksander Morgado 9ee9845b1e iface-modem-3gpp: avoid re-setting deferred registration update while disabling 
Don't clear the current deferred registration update until having disabled
and cleaned up unsolicited registrations state messages, or we may end up
re-setting the deferred registration update again meanwhile
 2013-03-01 18:21:19 +01:00
Ben Chan f93bba25f2 iface-modem-3gpp: clear deferred registration state update when disabling 2013-03-01 08:44:43 +01:00
Aleksander Morgado 89461cf7b3 iface-modem: allow falling back to 'failed' state 
Allow mm_iface_modem_update_state() receive 'MM_MODEM_STATE_FAILED', and treat
it as any other change to failed state, but with
'MM_MODEM_STATE_FAILED_REASON_UNKNOWN'.
 2013-02-26 21:17:43 +01:00
Aleksander Morgado 8d74b11bd3 plugin-manager: cancelled tasks are no longer deferred-until-suggested 
The deferred-until-suggested tasks specify tasks that are open/ongoing;
cancelled tasks shouldn't have this flag set.

https://bugzilla.gnome.org/show_bug.cgi?id=694603
 2013-02-25 15:28:47 +01:00
Ben Chan 8488782e14 iface-modem-3gpp: defer registration state update when registration is lost 
This patch defers the update of 3GPP registration state by 15 seconds
when the registration state changes from 'registered' (home / roaming)
to 'searching'. This allows a temporary loss of 3GPP registration to
recover itself when relying on ModemManager to explicitly disconnect and
reconnect to the network.
 2013-02-25 09:10:56 +01:00
Aleksander Morgado 3f8b11e3f0 broadband-modem: fix CUSD response parsing 
When reading the string reply in a +CUSD indication, don't blindly split
using whitespace or comma as field separator, as the string may be a text string
with both whitespaces and commas, e.g.:

+CUSD: 0,"hey hey, something here<LF>***<LF>and something more here"

Also, skip reading the encoding field for now, as we don't use it yet.
 2013-02-22 21:54:57 +01:00
Aleksander Morgado c19ee43cbe broadband-modem: avoid reading already freed memory in USSD operations 2013-02-22 21:49:19 +01:00
Aleksander Morgado 9976625f94 broadband-modem: USSD response may come even before finishing our request 
MBM devices seem to include the '+CUSD: 0' indication before even returning OK to our '+CUSD=1'

(ttyACM0): --> 'AT+CUSD=1,"*111#",15<CR>'
(ttyACM0): <-- '<CR><LF>+CUSD: 0,"reply here"<CR><LF>'
(ttyACM0): <-- '<CR><LF>OK<CR><LF>'
 2013-02-22 21:49:19 +01:00
Aleksander Morgado b1bb8e30b4 broadband-modem: don't assume the returned string is always hex-encoded 2013-02-22 19:43:49 +01:00
Aleksander Morgado b22bba509a broadband-modem: increase default AT command timeout for USSD operations 
May really take more than 3s.
 2013-02-22 17:47:25 +01:00
Aleksander Morgado 75d20c15d4 api: notify in the interface about the reason why the modem is FAILED 
We currently implement 'SIM missing' and 'SIM error', which are probably the
most common ones.
 2013-02-22 16:35:56 +01:00
Aleksander Morgado a18140e8ed plugin: avoid using uninitialized variable 
Reported by: Jose Maria Gonzalez <jmgonzalezc@indra.es>
 2013-02-21 11:15:37 +01:00
Aleksander Morgado 0377ccaa59 manager: during initial scan, add new ports in idles 
Treat each port independently in its own idle, as if we're receiving independent
udev events, otherwise, GSources may not be properly scheduled in the main loop.
 2013-02-20 19:16:05 +01:00
Aleksander Morgado 5b072b9198 plugin: plug memleak 
A new reference to probe was acquired through mm_device_get_probe(), just unref it
when no longer needed. Note that mm_port_probe_run() will take its own reference
for as long as required to complete the operation.
 2013-02-20 17:00:14 +01:00
Aleksander Morgado 688bb2cb3a plugin: plug memleaks 
This is really just to have a nice valgrind/memcheck output report, no big deal
if they were never freed.
 2013-02-20 16:48:41 +01:00
Aleksander Morgado 000bb64225 serial-port: allow specifying some wait time between closing and reopening 
We will now by default wait some time (1s) between port getting fully closed and
the port being reopened again. This logic seems to work for multiple modems
where there is a single port for both AT and data, like my Nokia C7 and Iridium
modems. If this wait time is not applied, the port ends up returning EAGAIN for
every write that we try to do afterwards.
 2013-02-20 15:54:54 +01:00
Aleksander Morgado 9a6fce0a08 serial-port: remove unneeded property name 
Was incorrectly included in a previous commit
 2013-02-20 11:41:53 +01:00
Aleksander Morgado 2267fb0480 plugin: plug memleaks 2013-02-20 09:59:20 +01:00
Aleksander Morgado f27737e031 bearer: plug memleak 2013-02-20 09:53:14 +01:00
Aleksander Morgado 061760a4f8 broadband-modem: plug memleak 2013-02-20 09:50:04 +01:00
Ben Chan bb73ce0aaf iface-modem: fix modem state consolidation upon bearer disconnection 
Patch "iface-modem: fix invalid modem state consolidation" (commit
69aff6183a) incorrectly consolidates the
modem state upon the disconnection of a bearer. The modem state remains
'connected' after the last bearer is disconnected. This patch fixes
that.
 2013-02-20 09:17:08 +01:00
Aleksander Morgado 5fe69d0ebb at-serial-port: plug memleak 2013-02-19 13:17:57 +01:00
Aleksander Morgado f9583cb530 core: update logging levels of several messages 
'info' log level, the default if none specified, included too many logs which
aren't that useful, to try to minimize the noise we produce by default.
 2013-02-19 13:13:57 +01:00
Aleksander Morgado 37ddfe202e plugin-manager: log time required for device probing 2013-02-19 12:52:01 +01:00
Aleksander Morgado ed2b32c682 plugin-manager: if minimum time consumed and all deferred tasks, abort probing 
For the case where we just get a device with all net ports (i.e. all deferred
until result suggested), just abort the probing if the minimum probing time is
consumed.
 2013-02-19 12:45:10 +01:00
Aleksander Morgado 1a11af0f86 broadband-bearer: fallback to primary AT port if no data AT port available 
Spotted by Dan Williams <dcbw@redhat.com> with his ADU960S.
 2013-02-19 09:50:52 +01:00
Aleksander Morgado d593116d68 base-modem: don't run port init sequence if we're just sending a command 
This may happen when sending commands to the modem while in non-enabled state,
like when sending the PIN. In this case, just send the command, don't fully
initialize the port with the initialization sequence.
 2013-02-18 16:06:18 +01:00
Aleksander Morgado 7a58647af7 broadband-bearer: fully reopen data port before flashing 
When disconnecting the data port, first fully close and reopen the port before
flashing. This helps with some modems where the flashing mechanism doesn't
fully work, like this ZTE modem:

    Before:
    [1360325270.357678] [mm-bearer.c:690] mm_bearer_disconnect(): Disconnecting bearer '/org/freedesktop/ModemManager1/Bearer/0'
    [1360325270.357796] [mm-iface-modem.c:1198] mm_iface_modem_update_state(): Modem /org/freedesktop/ModemManager1/Modem/1: state changed (connected -> disconnecting)
    [1360325270.358503] [mm-broadband-bearer.c:1539] disconnect_3gpp(): Sending PDP context deactivation in secondary port...
    [1360325270.358549] [mm-serial-port.c:927] mm_serial_port_open(): (ttyUSB4) device open count is 2 (open)
    [1360325270.358633] [mm-at-serial-port.c:392] debug_log(): (ttyUSB4): --> 'AT+CGACT=0,2<CR>'
    [1360325270.375131] [mm-at-serial-port.c:392] debug_log(): (ttyUSB4): <-- '<CR><LF>OK<CR><LF>'
    [1360325270.375251] [mm-serial-port.c:967] mm_serial_port_close(): (ttyUSB7) device open count is 1 (close)
    [1360325270.375282] [mm-serial-port.c:927] mm_serial_port_open(): (ttyUSB7) device open count is 2 (open)
    [1360325270.375300] [mm-broadband-bearer.c:1497] cgact_secondary_ready(): Flash primary port...
    [1360325270.375833] [mm-serial-port.c:967] mm_serial_port_close(): (ttyUSB4) device open count is 1 (close)
    [1360325271.377539] [mm-broadband-bearer.c:1455] primary_flash_3gpp_ready(): PDP disconnection already sent in secondary port
    [1360325271.377634] [mm-serial-port.c:967] mm_serial_port_close(): (ttyUSB7) device open count is 1 (close)
    [1360325271.377680] [mm-port.c:149] mm_port_set_connected(): (ttyUSB7): port now disconnected
    [1360325271.377721] [mm-bearer.c:623] disconnect_ready(): Disconnected bearer '/org/freedesktop/ModemManager1/Bearer/0'
    [1360325286.415396] [mm-serial-port.c:927] mm_serial_port_open(): (ttyUSB7) device open count is 2 (open)
    [1360325286.415510] [mm-at-serial-port.c:392] debug_log(): (ttyUSB7): --> 'AT+ZPAS?<CR>'
    [1360325287.400589] [mm-broadband-modem.c:1717] modem_load_signal_quality(): loading signal quality...
    [1360325287.400679] [mm-serial-port.c:927] mm_serial_port_open(): (ttyUSB7) device open count is 3 (open)
    [1360325289.401873] [mm-iface-modem.c:762] access_technologies_check_ready(): Couldn't refresh access technologies: 'Serial command timed out'
    [1360325289.401940] [mm-serial-port.c:967] mm_serial_port_close(): (ttyUSB7) device open count is 2 (close)
    [1360325289.401998] [mm-at-serial-port.c:392] debug_log(): (ttyUSB7): --> 'AT+CIND?<CR>'
    [1360325292.403162] [mm-serial-port.c:927] mm_serial_port_open(): (ttyUSB7) device open count is 3 (open)
    [1360325292.403247] [mm-serial-port.c:967] mm_serial_port_close(): (ttyUSB7) device open count is 2 (close
    ...

    After:
    [1360326617.604527] [mm-bearer.c:690] mm_bearer_disconnect(): Disconnecting bearer '/org/freedesktop/ModemManager1/Bearer/0'
    [1360326617.604622] [mm-iface-modem.c:1198] mm_iface_modem_update_state(): Modem /org/freedesktop/ModemManager1/Modem/0: state changed (connected -> disconnecting)
    [1360326617.605006] [mm-broadband-bearer.c:1535] disconnect_3gpp(): Sending PDP context deactivation in secondary port...
    [1360326617.605034] [mm-serial-port.c:929] mm_serial_port_open(): (ttyUSB4) device open count is 2 (open)
    [1360326617.605088] [mm-at-serial-port.c:392] debug_log(): (ttyUSB4): --> 'AT+CGACT=0,2<CR>'
    [1360326617.620075] [mm-at-serial-port.c:392] debug_log(): (ttyUSB4): <-- '<CR><LF>OK<CR><LF>'
    [1360326617.620206] [mm-broadband-bearer.c:1493] cgact_secondary_ready(): Flash primary port...
    [1360326617.620241] [mm-serial-port.c:1334] mm_serial_port_flash(): (ttyUSB7) reopening before flash (2)
    [1360326617.620265] [mm-serial-port.c:969] mm_serial_port_close(): (ttyUSB7) device open count is 1 (close)
    [1360326617.620286] [mm-serial-port.c:969] mm_serial_port_close(): (ttyUSB7) device open count is 0 (close)
    [1360326617.620320] [mm-serial-port.c:985] mm_serial_port_close(): (ttyUSB7) closing serial port...
    [1360326617.620356] [mm-port.c:149] mm_port_set_connected(): (ttyUSB7): port now disconnected
    [1360326617.621355] [mm-serial-port.c:1017] mm_serial_port_close(): (ttyUSB7) serial port closed
    [1360326617.621403] [mm-serial-port.c:860] mm_serial_port_open(): (ttyUSB7) opening serial port...
    [1360326617.622836] [mm-serial-port.c:417] real_config_fd(): (ttyUSB7): port attributes not fully set
 2013-02-18 15:41:26 +01:00
Aleksander Morgado 553bebe825 serial-port: new method to reopen a port 
We will fully close the port and the re-open it again keeping the same
open_count that we had before.
 2013-02-18 15:41:26 +01:00
Aleksander Morgado f8e3a8adb9 broadband-bearer: try to run CGACT first on the primary port 
It may happen that the primary port is not the data port; if so, try with the
primary port first, and otherwise fallback to the secondary port.
 2013-02-18 15:41:26 +01:00