| Age | Commit message (Collapse) | Author |
|---|
|
This patch fixes the following crash, which happens when
MMBroadbandBearer tries to parse the NULL response of a failed +CGDCONT?
query. It also fixes a leak in parse_pdp_list where it fails to free the
GError object return by mm_3gpp_parse_cgdcont_read_response.
Thread 0 *CRASHED* ( SIGSEGV @ 0x00000000 )
0x7feb15564c11 [ModemManager] - mm-modem-helpers.c:850 mm_3gpp_parse_cgdcont_read_response
0x7feb15518e51 [ModemManager] - mm-broadband-bearer.c:927 parse_pdp_list
0x7feb1551ada2 [ModemManager] - mm-base-modem-at.c:189 at_sequence_parse_response
0x7feb151d4b86 [libgio-2.0.so.0.3400.3] - gsimpleasyncresult.c:775 g_simple_async_result_complete
0x7feb1556e223 [ModemManager] - mm-port-serial-at.c:355 serial_command_ready
0x7feb151d4b86 [libgio-2.0.so.0.3400.3] - gsimpleasyncresult.c:775 g_simple_async_result_complete
0x7feb1556a834 [ModemManager] - mm-port-serial.c:141 command_context_complete_and_free
0x7feb1556bc26 [ModemManager] - mm-port-serial.c:734 port_serial_got_response
0x7feb1556c51a [ModemManager] - mm-port-serial.c:758 port_serial_timed_out
0x7feb1508cc33 [libglib-2.0.so.0.3400.3] - gmain.c:4026 g_timeout_dispatch
0x7feb1508c087 [libglib-2.0.so.0.3400.3] - gmain.c:2715 g_main_context_dispatch
0x7feb1508c437 [libglib-2.0.so.0.3400.3] - gmain.c:3290 g_main_context_iterate
0x7feb1508c891 [libglib-2.0.so.0.3400.3] - gmain.c:3484 g_main_loop_run
0x7feb1550ad16 [ModemManager] - main.c:154 main
0x7feb14a9e9c6 [libc-2.15.so] - libc-start.c:234 __libc_start_main
0x7feb1550a808 [ModemManager + 0x00023808
|
|
|
|
AT+COPS=0 in a FXT009 may end up making the device stuck in a weird state
in which it only replies "+CME ERROR: 515" to every AT command. So try to
avoid this command whenever possible.
|
|
Subscriber information like MSISDNs / MDNs may not be readily available
when MMIfaceModem queries for it during modem initialization.
This patch modifies MMBroadbandModemMbim to subscribe to the
MBIM_CID_SUBSCRIBER_READY_STATUS notification and update the
'OwnNumbers' property of the Modem interface when the information
becomes available.
|
|
libmbim renames 'MBIM_CID_DEVICE_SERVICE_SUBSCRIBER_LIST' to
'MBIM_CID_DEVICE_SERVICE_SUBSCRIBE_LIST' to be consistent with the MBIM
specification. This patch updates MMBroadbandModemMbim accordingly.
|
|
We're going to add support for some features only available in that version.
|
|
hdr_subsys_state_info_ready() did not implement the GAsyncReadyCallback
function signature, which led to a crash. This patch fixes
hdr_subsys_state_info_ready() to properly extract the QCDM response from
the GAsyncResult object.
|
|
This is not a license change; the daemon and command line tool are still GPLv2+,
and the libmm-glib library LGPLv2+.
For reference, see libqmi bug:
https://bugs.freedesktop.org/show_bug.cgi?id=74220
|
|
A MBIM device typically delays its response to an 'Open' command when it
has not completed the initialization process. This patch increases the
timeout of the MBIM device open operation from 10s to 30s to handle
situations where the device takes much longer to initialize.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
So that plugins can use it.
|
|
|
|
|
|
|
|
|
|
|
|
For periodic updates, we may want to request an explicit re-check at some point,
instead of waiting for the next scheduled update.
|
|
|
|
When automatic mode (i.e. 2G+3G+none preferred) is selected, we need to
explicitly request it by giving the AT+COPS write command with values (i.e. not
just empty fields).
|
|
We'll need it afterwards for proper mode switching
|
|
So no need to try to handle error cases or fallbacks to support them.
|
|
|
|
ensure_internal_ports() didn't deep-copy the ports array, which
meant that anything calling mm_modem_port_info_array_free() was
freeing the internal memory owned by MMModem if it retrieved
the array with mm_modem_get_ports().
https://bugzilla.gnome.org/show_bug.cgi?id=724900
|
|
By setting the 'RemainingAttempts' field of a MBIM_CID_PIN response to
0xffffffff, a MBIM modem indicates that it does not support the
information on the number of remaining attempts for pin-related
operations. This patch modifies MMBroadbandModemMbim to handle this case
properly, instead of reporting a very large unlock retries count.
|
|
The Cinterion PHS8 reports a single AT port to be used as data port, and no
other port as primary. In this case, recover the tty as primary AT port, instead
of failing with an error.
|
|
Powering off Wavecom devices is done via the AT^CPOF=1 command. These devices
seem to not only power-off, but also power-cycle when this command is received,
which means that the user needs to make sure to fully cut the power to the
device when the SetPowerState() method replies.
|
|
Powering off Cinterion modems is done with the AT^SMSO command. Once this
command is sent to the device, ModemManager still needs to wait to get an
unsolicited ^SHUTDOWN message from the modem, which is when the shutdown
is considered complete.
Newer Cinterion devices, though, don't seem to follow this rule and just
reply OK to AT^SMSO (i.e. no ^SHUTDOWN), but instead, they will remove the
tty ports.
|
|
The power off command will transition the device to the OFFLINE state. After
this, the modem will not be able to go online again unless it is power-cycled,
either externally (cut power and re-establish power) or through another
transition to QMI_DMES_OPERATING_MODE_RESET (which ModemManager will not
implement for this case).
|
|
|
|
There are systems in which the modem power source can be controlled externally,
e.g. to switch it on or off. In these cases, it is sometimes advisable to be
able to tell the modem to cleanly shut off completely before the power is cut.
So, allow transitioning to the OFF power state if the modem supports it, even if
afterwards the modem could be completely unreachable.
|
|
Observing registration updates via the MBIM_CID_REGISTER_STATE
notification alone is not always sufficient to detect if the modem is
disconnected from the network. In case of a network-initiated
disconnect, it is possible that the modem is disconnected from the
network but remains registered.
This patch modifies MMBroadbandModemMbim to subscribe to the
MBIM_CID_CONNECT notification. Upon receiving a notification of a
deactivated session ID, the corresponding bearer is reported as
disconnected.
|
|
|
|
mbim_message_device_service_subscribe_list"
This reverts commit 8327db70b9b5b08481ce1666d7a0192f306ff5ee.
libmbim change not committed yet...
|
|
libmbim renames 'MBIM_CID_DEVICE_SERVICE_SUBSCRIBER_LIST' to
'MBIM_CID_DEVICE_SERVICE_SUBSCRIBE_LIST' to be consistent with the MBIM
specification. This patch updates MMBroadbandModemMbim accordingly.
|
|
This patch simply fixes the following debug message:
from:
<debug> (Huawei) couldn't turn off unsolicited messages insecondary ports: 'Unknown error'
to:
<debug> (Huawei) couldn't turn off unsolicited messages in secondary ports: 'Unknown error'
|
|
|
|
|
|
This patch fixes the following type mismatch in MMPortSerial::port_serial_process_command():
mm-port-serial.c:612:21: error: comparison of unsigned expression < 0 is always false [-Werror,-Wtautological-compare]
if (written < 0) {
~~~~~~~ ^ ~
|
|
This patch fixes the out-of-bounds array accesses in test-port-context.c,
which is detected by AddressSanitizer, by checking the index against the
array length before accessing the array.
|
|
This patches fixes the following error when testing ModemManager with
glib < 2.36, which still requires g_type_init() to be called:
TEST: test-service-generic... (pid=20166)
/MM/Service/Generic:
(/work/ModemManager/plugins/.libs/lt-test-service-generic:20166): GLib-GObject-CRITICAL **: gtype.c:2720: You forgot to call g_type_init()
FAIL
|
|
Caused by a merge conflict between 81ee0783 and 529e94b0.
|
|
At some point rfcomm serial ports stopped having parents in sysfs,
so checks to get the physical device fail because the rfcomm port
is /sys/devices/virtual/rfcommX and has no parents. So we have
to fall back on checking the interface name.
|
|
|
