This patch simplifies the handling of platform/pci/pnp/sdio device in
find_physical_gudevdevice(). When the code finds the first parent device
under the subsystem 'platform', 'pci', 'pnp', or 'sdio', it stops
traversing up the device tree, so there is no need to keep track of
whether a platform / pci / pnp / sdio device has previously been visited
via those is_platform/is_pci/is_pnp/is_sdio Boolean variables.
A few crashes have been observed in the field with the following signature:
Thread 0 CRASHED [SIGSEGV @ 0x00000000 ] MAGIC SIGNATURE THREAD
0xf53ff5e8 (libglib-2.0.so.0.3600.4 -ghash.c:1732 ) g_str_hash
0xf53fe8c7 (libglib-2.0.so.0.3600.4 -ghash.c:365 ) g_hash_table_lookup
0xb953c3bd (ModemManager -mm-base-manager.c:130 ) device_removed
0xb953cc9f (ModemManager -mm-base-manager.c:408 ) handle_uevent
0xf54c535f (libgobject-2.0.so.0.3600.4 -gclosure.c:777 ) g_closure_invoke
0xf54d6fc7 (libgobject-2.0.so.0.3600.4 -gsignal.c:3584 ) signal_emit_unlocked_R
0xf54d7baf (libgobject-2.0.so.0.3600.4 -gsignal.c:3328 ) g_signal_emit_valist
0xf54d7fa5 (libgobject-2.0.so.0.3600.4 -gsignal.c:3384 ) g_signal_emit
0xf5623819 (libgudev-1.0.so.0.2.0 -gudevclient.c:104 ) monitor_event
0xf540cc51 (libglib-2.0.so.0.3600.4 -gmain.c:3054 ) g_main_context_dispatch
0xf540cfa5 (libglib-2.0.so.0.3600.4 -gmain.c:3701 ) g_main_context_iterate
0xf540d2c5 (libglib-2.0.so.0.3600.4 -gmain.c:3895 ) g_main_loop_run
0xb9539dad (ModemManager -main.c:180 ) main
0xf52d687d (libc-2.23.so -libc-start.c:289 ) __libc_start_main
0xb9539bf3 (ModemManager + 0x0001cbf3 ) _start
0xb95b0fbf (ModemManager -elf-init.c:87 ) __libc_csu_init
0xf56dbe43 (ld-2.23.so + 0x0000be43 ) _dl_sort_fini
0xb9539bbf (ModemManager + 0x0001cbbf ) _init
The crash happens when mm-kernel-device-udev.c:find_physical_gudevdevice()
fails to find the physical device, which eventually leads to a NULL
`physdev_uid' being passed to g_hash_table_lookup() in
mm-base-manager.cc:find_device_by_physdev_uid().
There isn't much information about the actual device that triggered the
udev event, but it's suspected to be a tty or net device on a SDIO bus
and thus associated with a physical device on the 'sdio' subsystem. This
patch updates find_physical_gudevdevice() in mm-kernel-device-udev.c to
handle this particular scenario.
Two issues here:
1) PCH_SLEEP wasn't considered a "wcdma_open" state, but it should be
since the modem is still registered and can be attached in this state
2) If the system mode is WCDMA (eg, not GSM and not GSM/WCDMA) then
MM shouldn't assume GPRS if WCDMA isn't "open", since GPRS isn't
enabled in WCDMA-only mode
Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=100000
The enum was wrong. There isn't actually an L1M_INIT state; the
enum should start with L1M_IDLE. There should also be a
L1M_PCH_SLEEP state between DEACTIVATE and DEEP_SLEEP.
Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=100000
* Allow whitespaces prefixing the values row.
* Allow more than one \r\n between the title and the table header.
Reported-by: Colin Helliwell <colin.helliwell@ln-systems.com>
Don't rely on the automatic disconnection of the signal as the last
reference of the modem object may outlive the device object.
Also, setup a common clear_modem() function to dispose the internal
reference of the modem object, which will take care of the signal
disconnection when required.
https://lists.freedesktop.org/archives/modemmanager-devel/2017-February/003907.html
So that we don't leak the port names allocated within each
MMModemPortInfo.
==261== 672 bytes in 84 blocks are definitely lost in loss record 7,314 of 7,383
==261== at 0x402C51E: malloc (vg_replace_malloc.c:299)
==261== by 0x4484878: g_malloc (gmem.c:94)
==261== by 0x449D51D: g_strdup (gstrfuncs.c:363)
==261== by 0x44B5B73: g_variant_dup_string (gvariant.c:1529)
==261== by 0x44B945E: g_variant_valist_get_nnp (gvariant.c:4775)
==261== by 0x44B945E: g_variant_valist_get_leaf (gvariant.c:4945)
==261== by 0x44B945E: g_variant_valist_get (gvariant.c:5126)
==261== by 0x44B922C: g_variant_valist_get (gvariant.c:5161)
==261== by 0x44B9FC9: g_variant_get_va (gvariant.c:5388)
==261== by 0x44BA3C5: g_variant_get_child (gvariant.c:5486)
==261== by 0x6613FA8: mm_common_ports_variant_to_garray (mm-common-helpers.c:238)
==261== by 0x6601AB9: ensure_internal_ports (mm-modem.c:766)
==261== by 0x6603E42: mm_modem_peek_ports (mm-modem.c:825)
==261== by 0x65CD94D: owns_port (nm-modem-broadband.c:196)
Reported-by: Piotr Figiel <p.figiel@camlintechnologies.com>
==261== 482 bytes in 12 blocks are definitely lost in loss record 7,290 of 7,383
==261== at 0x402C51E: malloc (vg_replace_malloc.c:299)
==261== by 0x4484878: g_malloc (gmem.c:94)
==261== by 0x449D51D: g_strdup (gstrfuncs.c:363)
==261== by 0x44B5B73: g_variant_dup_string (gvariant.c:1529)
==261== by 0x44B945E: g_variant_valist_get_nnp (gvariant.c:4775)
==261== by 0x44B945E: g_variant_valist_get_leaf (gvariant.c:4945)
==261== by 0x44B945E: g_variant_valist_get (gvariant.c:5126)
==261== by 0x44B922C: g_variant_valist_get (gvariant.c:5161)
==261== by 0x44B9FC9: g_variant_get_va (gvariant.c:5388)
==261== by 0x44BA1DB: g_variant_get (gvariant.c:5335)
==261== by 0x664E2EF: mm_gdbus_modem_simple_call_connect_finish (mm-gdbus-modem.c:22451)
==261== by 0x6608A08: simple_connect_ready (mm-modem-simple.c:154)
==261== by 0x429136F: g_task_return_now (gtask.c:1107)
==261== by 0x4291A69: g_task_return (gtask.c:1165)
g_type_init() has been deprecated (and also marked with the attribute
'deprecated') since glib 2.36 as the type system is automatically
initialized. Since the minimum version of glib required by ModemManager
is 2.36, calling g_type_init() isn't necessarily in the ModemManager
code.
This patch fixes an uninitialized variable issue in
mm_ublox_parse_ugcntrd_response_for_cid(), which uses an uninitialized
`match_info' when `in_cid' is invalid.
This patch fixes a bug in packet_service_status_indication_cb(), which
incorrectly treats the MMBearerStatus enum value returned by
mm_base_bearer_get_status() as a MMBearerConnectionStatus enum value.
MMBearerStatus and MMBearerConnectionStatus can't be used
interchangeably as they have different enum values for the
'disconnected' and 'disconnecting' state.
`g_assert_true' and `g_assert_false' are defined in glib 2.38 or later.
The minimum glib version currently required by ModemMamanger is 2.36.
While `g_assert_true' and `g_assert_false' may be preferred over the
more generic `g_assert', it seems like overkill to bump the minimum glib
version requirement just for that. When more code in ModemManager later
requires newer versions of glib, we can migrate all existing code to use
`g_assert_true' and `g_assert_false' when appropriate.
The suggestion to use specific PDP context CIDs was given by Cinterion
for the special case of the Verizon operator, which 'reserves' specific
CIDs for specific purposes.
We don't want to impose that at the Cinterion plugin level, so remove
the PDP context mapping we had.
Therefore, simplify the connection procedure by just overriding the
'dialing' step of the default 3GPP connection sequence, instead of
overriding the whole connection sequence.
Also, we don't need to override the step to gather IP config because
this is already handled by the generic plugin (for DHCP over a network
interface).
We port to GTask for both 3GPP dial and 3GPP disconnect at the same time.
Better check for ^SWWAN support during the first time a bearer is going
to be created.
The enabling phase isn't the correct one because this logic is only run
whenever a modem is detected but not hotplugged (i.e. this step is to
'reset' the modem to generic runtime settings).
We get as input the ^SWWAN index we're interested in, and we loop
through the list of ^SWWAN lines looking for the one we need.
This updated helper method allows working with multi-line ^SWWAN
responses, e.g. given when more than one PDP context is active.
Group together all connection related logic (e.g. context) and define
the context steps directly within the connection sequence processing.
Also, don't initially run a disconnection before the connection; if that
logic is ever needed we should likely have it in the generic modem, not
done per plugin.
And error out early if not asking for IPv4.
mm_device_create_modem needs a valid object_manager instance to create the new
modem, while the one provided before was cleared out by a call to
mm_device_remove_modem few lines before.
Since the newly created modem refers to the same HW device, we can use
the same object_manager also.
https://bugs.freedesktop.org/show_bug.cgi?id=99160
So that different concurrent runs of this same test don't clash with
each other:
TEST: test-service-generic... (pid=11124)
/MM/Service/Generic/enable-disable: Activating service name='org.freedesktop.ModemManager1'
Successfully activated service 'org.freedesktop.ModemManager1'
** (plugins/.libs/lt-test-service-generic>:11124): ERROR **: Cannot bind socket: Error binding to address: Address already in use
cleaning up pid 11144
FAIL
Looks like gdbus-codegen ends up making paragraphs and that breaks the
XML parsing:
DOC Building HTML
../../../../libmm-glib/generated/mm-gdbus-doc-org.freedesktop.ModemManager1.xml:60: parser error : Opening and ending tag mismatch: variablelist line 59 and para
</para><para> <varlistentry><term><literal>action</literal></term>
^
../../../../libmm-glib/generated/mm-gdbus-doc-org.freedesktop.ModemManager1.xml:110: parser error : Opening and ending tag mismatch: para line 110 and variablelist
</para><para> </variablelist>
^
../xml/ModemManager-dbus-reference.xml:93: element include: XInclude error : could not load ../../../../libmm-glib/generated/mm-gdbus-doc-org.freedesktop.ModemManager1.xml, and no fallback was found
This is so that "make check" doesn't depend on having the rule files
installed in the default udev rules directory, which currently would
break the build:
TEST: test-service-generic... (pid=2601)
/MM/Service/Generic/enable-disable: Activating service name='org.freedesktop.ModemManager1'
Successfully activated service 'org.freedesktop.ModemManager1'
** (/home/aleksander/Development/foss/ModemManager/plugins/.libs/lt-test-service-generic:2601):
ERROR **: Error setting test profile: GDBus.Error:org.freedesktop.ModemManager1.Error.Core.Failed: Failed to find primary AT port
FAIL
GTester: last random seed: R02S3897abaae9df36f8d2eeb679406ab675
make[3]: *** [Makefile:3804: test-nonrecursive] Terminated
