The init-sequence configured for the TTY that is being used as data
port must not be launched during the port reopen() sequence; instead
we must run it manually after the port flashing has finished.
This patch initializes `match_info' in registration_status_check_ready()
to NULL by default, such that `match_info' is always initialized even if
`self->priv->modem_3gpp_registration_regex' contains no elements.
Though `self->priv->modem_3gpp_registration_regex' always contains some
elements in the current implementation, it's better not to rely on that.
This patch fixes an ineffective `g_assert (equip_id)' in
modem_load_equipment_identifier_finish(). After mm_parse_gsn() succeeds,
`equip_id' is freed but not reset to NULL, so `g_assert (equip_id)' will
never assert even if `imei', `meid', and `esn' are all NULL (though that
shouldn't happen when mm_parse_gsn() succeeds).
This patch fixes some potential use-after-freed issues in
dms_get_ids_ready(). When an invalid ESN / MEID is retrieved,
`ctx->self->priv->esn' / `ctx->self->priv->meid' is freed but not reset
to NULL. If no IMEI is retrieved, `str' can be set to the already freed
`ctx->self->priv->esn' / `ctx->self->priv->meid' and then propagated to
a GSimpleAsyncResult object.
This patch fixes an issue in disconnect_set_ready(). If
mbim_message_connect_response_parse(), `session_id' and `nw_error' are
not set to a valid value, and thus shouldn't be used.
This patch fixes cause_code_to_delivery_state() by adding two missing
break statements for the case ERROR_CLASS_TEMPORARY and
ERROR_CLASS_PERMANENT in the `switch (error_class)` statement. Without
the break statements, the switch always falls through to the default and
returns MM_SMS_DELIVERY_STATE_UNKNOWN for an `error_class' of value
ERROR_CLASS_TEMPORARY or ERROR_CLASS_PERMANENT.
Currently, when SIM hot swap fails in either mm-iface or plugin, the
ModemManager still opens ports context and prints a message saying that
SIM hot swap is supported and that it's waiting for SIM insertion,
instead of clearly saying that SIM hot swap is not working.
This patch:
1. introduces a new property MM_IFACE_MODEM_SIM_HOT_SWAP_CONFIGURED
which is FALSE by default and set to TRUE only when
setup_sim_hot_swap_finish() succeded.
2. subordinates the completion of SIM hot swap setup (in
mm-broadband-modem) and the related messages to the the value of
MM_IFACE_MODEM_SIM_HOT_SWAP_CONFIGURED
Finally, this patch replaces the MBIM's sim_hot_swap_on private property
with the new property MM_IFACE_MODEM_SIM_HOT_SWAP_CONFIGURED, since they have the
same meaning.
crm_range_ready() misses a return statement after it invokes
interface_initialization_step(), which will lead to an invalid access of
the already freed InitAsyncContext.
==28664== 4 bytes in 1 blocks are definitely lost in loss record 33 of 4,317
==28664== at 0x4C2BE1F: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==28664== by 0x4C2E164: realloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==28664== by 0x644E0CF: g_realloc (in /usr/lib/libglib-2.0.so.0.5200.2)
==28664== by 0x6469C06: ??? (in /usr/lib/libglib-2.0.so.0.5200.2)
==28664== by 0x6469C61: g_string_sized_new (in /usr/lib/libglib-2.0.so.0.5200.2)
==28664== by 0x4E8514A: mm_common_build_sms_storages_string (mm-common-helpers.c:116)
==28664== by 0x489035: cpms_query_ready (mm-broadband-modem.c:5428)
==28664== by 0x5E8E335: g_simple_async_result_complete (in /usr/lib/libgio-2.0.so.0.5200.2)
==28664== by 0x44EADF: at_command_ready (mm-base-modem-at.c:503)
==28664== by 0x5E8E335: g_simple_async_result_complete (in /usr/lib/libgio-2.0.so.0.5200.2)
==28664== by 0x4D8122: serial_command_ready (mm-port-serial-at.c:392)
==28664== by 0x5E8E335: g_simple_async_result_complete (in /usr/lib/libgio-2.0.so.0.5200.2)
==28664==
==28664== 4 bytes in 1 blocks are definitely lost in loss record 34 of 4,317
==28664== at 0x4C2BE1F: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==28664== by 0x4C2E164: realloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==28664== by 0x644E0CF: g_realloc (in /usr/lib/libglib-2.0.so.0.5200.2)
==28664== by 0x6469C06: ??? (in /usr/lib/libglib-2.0.so.0.5200.2)
==28664== by 0x6469C61: g_string_sized_new (in /usr/lib/libglib-2.0.so.0.5200.2)
==28664== by 0x4E8514A: mm_common_build_sms_storages_string (mm-common-helpers.c:116)
==28664== by 0x489067: cpms_query_ready (mm-broadband-modem.c:5430)
==28664== by 0x5E8E335: g_simple_async_result_complete (in /usr/lib/libgio-2.0.so.0.5200.2)
==28664== by 0x44EADF: at_command_ready (mm-base-modem-at.c:503)
==28664== by 0x5E8E335: g_simple_async_result_complete (in /usr/lib/libgio-2.0.so.0.5200.2)
==28664== by 0x4D8122: serial_command_ready (mm-port-serial-at.c:392)
==28664== by 0x5E8E335: g_simple_async_result_complete (in /usr/lib/libgio-2.0.so.0.5200.2)
==28664== 216 (120 direct, 96 indirect) bytes in 5 blocks are definitely lost in loss record 4,109 of 4,317
==28664== at 0x4C2BEEF: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==28664== by 0x644E018: g_malloc (in /usr/lib/libglib-2.0.so.0.5200.2)
==28664== by 0x6465B05: g_slice_alloc (in /usr/lib/libglib-2.0.so.0.5200.2)
==28664== by 0x6469C32: g_string_sized_new (in /usr/lib/libglib-2.0.so.0.5200.2)
==28664== by 0x4D7861: parse_response (mm-port-serial-at.c:142)
==28664== by 0x4D3B52: parse_response_buffer (mm-port-serial.c:857)
==28664== by 0x4D40F2: common_input_available (mm-port-serial.c:981)
==28664== by 0x4D41B4: iochannel_input_available (mm-port-serial.c:1004)
==28664== by 0x64488B4: g_main_context_dispatch (in /usr/lib/libglib-2.0.so.0.5200.2)
==28664== by 0x6448C77: ??? (in /usr/lib/libglib-2.0.so.0.5200.2)
==28664== by 0x6448F91: g_main_loop_run (in /usr/lib/libglib-2.0.so.0.5200.2)
==28664== by 0x438035: main (main.c:180)
Aleksander Morgado
Ben Chan
Carlo Lobrano