OSDN Git Service
Lucas De Marchi [Fri, 18 May 2012 03:23:34 +0000 (00:23 -0300)]
gdbus: add Deprecated annotation in introspection
Lucas De Marchi [Fri, 18 May 2012 03:23:33 +0000 (00:23 -0300)]
gdbus: remove signature and reply from tables
Lucas De Marchi [Fri, 18 May 2012 03:23:32 +0000 (00:23 -0300)]
Do not set signature and reply in GDBus tables
Use GDBUS_* macros, so signature and reply fields are not set in each
method/signal.
Lucas De Marchi [Fri, 18 May 2012 03:23:31 +0000 (00:23 -0300)]
gdbus: loop over args to check message signature
Lucas De Marchi [Fri, 18 May 2012 03:23:30 +0000 (00:23 -0300)]
gdbus: use GDBusArgInfo to generate introspection
By using GDBusArgInfo in methods and signals, the introspection
generation is much simpler and we can add each argument name.
Lucas De Marchi [Fri, 18 May 2012 03:23:29 +0000 (00:23 -0300)]
Convert GDBus methods to use macro helpers
With these macro helpers we can separate in/out arguments and use their
own vector.
Lucas De Marchi [Fri, 18 May 2012 03:23:28 +0000 (00:23 -0300)]
gdbus: add and use helpers for table declarations
Lucas De Marchi [Fri, 18 May 2012 03:23:27 +0000 (00:23 -0300)]
gdbus: add argument info to methods and signals
Lucas De Marchi [Fri, 18 May 2012 03:23:26 +0000 (00:23 -0300)]
Constify GDBus signal tables
Constify signal tables with the following command:
find . -name '*.[ch]' -exec \
sed -i 's/\(GDBusSignalTable .* =\)/const \1/g' {} \;
Lucas De Marchi [Fri, 18 May 2012 03:23:25 +0000 (00:23 -0300)]
Constify GDBus method tables
Constify method tables with the following command:
find . -name '*.[ch]' -exec \
sed -i 's/\(GDBusMethodTable .* =\)/const \1/g' {} \;
Marcel Holtmann [Fri, 18 May 2012 02:56:27 +0000 (19:56 -0700)]
gdbus: Constify introspection method table
Lucas De Marchi [Thu, 17 May 2012 18:19:59 +0000 (15:19 -0300)]
gdbus: do not call memset for terminating NUL
Lucas De Marchi [Thu, 17 May 2012 18:19:58 +0000 (15:19 -0300)]
gdbus: return if method signature is malformed
Alex Elsayed [Thu, 17 May 2012 17:00:27 +0000 (10:00 -0700)]
Add dbus service file that references the systemd unit
This allows bluez to be bus-activated.
Johan Hedberg [Wed, 16 May 2012 10:07:09 +0000 (13:07 +0300)]
hciemu: Fix gcc 4.7 compilation error
This patch fixes the following compilation error with gcc 4.7:
CC test/hciemu.o
test/hciemu.c: In function 'getbdaddrbyname':
test/hciemu.c:1115:3: error: dereferencing type-punned pointer will break strict-aliasing rules [-Werror=strict-aliasing]
Lucas De Marchi [Wed, 16 May 2012 12:26:24 +0000 (09:26 -0300)]
doc: Update possible values of Agent capabilities
Anderson Lizardo [Fri, 11 May 2012 13:09:29 +0000 (09:09 -0400)]
adapter: Remove dead code from set_property()
Lucas De Marchi [Wed, 9 May 2012 12:06:17 +0000 (09:06 -0300)]
build: Do not set CFLAGS/LDFLAGS directly
Set a separate variable for adding warning flags, optimization, etc.
Build systems are not supposed to change CFLAGS and LDFLAGS, these are
user variables.
Doing so we guarantee CFLAGS and LDFLAGS from environment is appended
to the flags used during build. One useful use-case is to temporarily
disable -Werror when using --enable-maintainer-mode, without completely
loosing the warning flags and other parameters in CFLAGS (like -fPIC).
Without this patch, fiddling with CFLAGS/LDFLAGS after configure may
result in errors like below:
/usr/bin/ld: tools/rfcomm.o: relocation R_X86_64_32 against `.bss' can
not be used when making a shared object; recompile with -fPIC
tools/rfcomm.o: could not read symbols: Bad value
collect2: error: ld returned 1 exit status
make[1]: *** [tools/rfcomm] Error 1
make: *** [all] Error 2
Reference: http://www.gnu.org/software/automake/manual/html_node/Flag-Variables-Ordering.html
Lucas De Marchi [Wed, 9 May 2012 12:06:16 +0000 (09:06 -0300)]
build: Fix typo in AM_CFLAGS
Vishal Agarwal [Thu, 10 May 2012 13:43:47 +0000 (19:13 +0530)]
textfile: fix missing call to munmap in wite_key
In some cases it was possible that munmap is not called corresponding
to mmap, which might result in future read or writes to fail.
Andrei Emeltchenko [Thu, 10 May 2012 08:08:51 +0000 (11:08 +0300)]
lib: Add definitions for AMP Assoc data
Definitions are needed for decoding A2MP Assoc Rsp.
Paulo Alcantara [Fri, 11 May 2012 15:45:35 +0000 (12:45 -0300)]
core: Remove unused variable
"srcaddr" variable is not used for anything else in
btd_event_remote_name() function. So, remove it for now.
Tedd Ho-Jeong An [Tue, 15 May 2012 17:23:29 +0000 (10:23 -0700)]
hciattach: Add support for Intel Bluetooth device
This patch enables the Intel Bluetooth device (UART sku) over the H4
protocol. It is responsible for bring up the device into known state by
configuring the baudrate and applying the patches, if required.
Chen Ganir [Wed, 9 May 2012 09:21:15 +0000 (12:21 +0300)]
eir: Keep parsed eir_data on error
Do not discard already parsed eir_data on eir data length error.
Data corruption may occur, but this does not mean that we need
to ignore already parsed valid eir data.
Gustavo Padovan [Fri, 11 May 2012 16:18:49 +0000 (13:18 -0300)]
monitor: remove extra black line
Anderson Lizardo [Fri, 27 Apr 2012 14:36:52 +0000 (10:36 -0400)]
hidd: Fix indentation
Anderson Lizardo [Fri, 27 Apr 2012 14:36:51 +0000 (10:36 -0400)]
audio: Fix tab misuse
Anderson Lizardo [Fri, 27 Apr 2012 14:36:50 +0000 (10:36 -0400)]
deviceinfo: Fix indentation and tab misuse
Anderson Lizardo [Fri, 27 Apr 2012 14:36:49 +0000 (10:36 -0400)]
monitor: Fix tab misuse
Anderson Lizardo [Fri, 27 Apr 2012 14:36:48 +0000 (10:36 -0400)]
sap: Fix tab misuse
Anderson Lizardo [Fri, 27 Apr 2012 14:36:47 +0000 (10:36 -0400)]
device: Fix tab misuse
Anderson Lizardo [Fri, 27 Apr 2012 14:36:46 +0000 (10:36 -0400)]
event: Fix tab misuse
Anderson Lizardo [Fri, 27 Apr 2012 14:36:45 +0000 (10:36 -0400)]
tools: Fix tab misuse
Anderson Lizardo [Fri, 27 Apr 2012 14:36:44 +0000 (10:36 -0400)]
lib: Fix tab misuse
Johan Hovold [Fri, 27 Apr 2012 13:03:58 +0000 (15:03 +0200)]
csr_bcsp: fix link establishment
The current link-establishment implementation times out if a packet is
not received after the link is active (i.e. reception of CONF_RESP).
Fix by making sure that the link establishment loop in open is
terminated when ubcsp sets the UBCSP_PACKET_SENT flag (on reception of
CONF_RESP) rather than UBCSP_PACKET_RECEIVED (on reception of any later
packages).
Tom Gundersen [Thu, 3 May 2012 22:21:53 +0000 (00:21 +0200)]
udev: remove udev activation of the blutooth daemon
Udev is not meant to spawn long-running processes, and as of the next
udev version such processes will be killed.
If using systemd, device activation can be used to get the same effect.
On Arch we have been disabling this rule for a long time, even though we
don't use systemd, as it was causing more problems than it was worth.
Tom Gundersen [Thu, 3 May 2012 22:21:52 +0000 (00:21 +0200)]
systemd: don't order against syslog.target
This was only needed in old versions of systemd. All messages are logged
by default to the journal now, no ordering required.
Gustavo Padovan [Fri, 4 May 2012 02:01:19 +0000 (23:01 -0300)]
input: increase security level of the connection
Andrzej Kaczmarek [Mon, 7 May 2012 13:09:02 +0000 (15:09 +0200)]
gatttool: Add option to specify LE address type
This patch makes possible to specify LE address type.
After advertising cache was removed from kernel we should always specify
address type for LE link when calling bt_io_connect() as otherwise random
will always be used.
LE address type can be specified either by 'addr-type' or 't' command line
parameter or as additional parameter to 'connect' command in interactive mode.
Possible values are 'public' (default) and 'random'.
Mikel Astiz [Fri, 4 May 2012 13:14:40 +0000 (15:14 +0200)]
media: Enable parallel requests to endpoint
If endpoint is handling several devices, it should be able to handle
multiple requests, one per each.
Mikel Astiz [Fri, 4 May 2012 13:14:39 +0000 (15:14 +0200)]
media: Create multiple transports if needed
During endpoint registration one than one device might be connected.
Thus all matching devices should have one transport each.
Mikel Astiz [Fri, 4 May 2012 13:14:38 +0000 (15:14 +0200)]
media: Split media_endpoint_create
This function is starting to be too long and needs to be split.
After this patch, the resulting code should be exactly equivalent as the
previous implementation.
Mikel Astiz [Fri, 4 May 2012 13:14:37 +0000 (15:14 +0200)]
media: Support multiple transports per endpoint
Several transports may exist for each endpoint, for example if several
HFGW are connected. This should be exposed to the endpoint as one
transport each.
Mikel Astiz [Fri, 4 May 2012 13:14:36 +0000 (15:14 +0200)]
audio: Add multiple device search to manager
This method is useful to search for more than one device fulfulling
certain criteria.
Mikel Astiz [Fri, 4 May 2012 13:14:35 +0000 (15:14 +0200)]
audio: Fix gateway state check
Gateway should be considered active also if connecting or playing.
This could for example lead to manager_find_device() not returning a
device that is connecting, and thus the corresponding endpoint would
never be created in the Media API.
Luiz Augusto von Dentz [Thu, 3 May 2012 14:41:23 +0000 (17:41 +0300)]
SAP: Fix not unregistering dummy interface on exit
Dummy interface is registered on plugin init and is never unregistered
Anderson Lizardo [Fri, 27 Apr 2012 14:57:02 +0000 (10:57 -0400)]
mgmtops: Fix wrong parameter order for btd_event_ltk_notify()
The "master" parameter was in the wrong place.
Jaganath Kanakkassery [Wed, 2 May 2012 11:36:30 +0000 (17:06 +0530)]
audio: Reset hfp and hs handle before connection
When bluez initiates headset connection hfp and hs handle is saved
each time when initiating connection. So at some point if hf service
is removed in remote headset then bluez connects to hs service. But
because of previously stored hfp handle bluez thinks that connection
made to hs service is hfp connection and waits for at commands.
Eventually bluez connection state will be incorrectly set.
Bastien Nocera [Sat, 28 Apr 2012 15:16:56 +0000 (16:16 +0100)]
Fix adaptername shutdown
An inotify watch descriptor isn't a file descriptor, so don't
treat it as such and remove it properly.
Gustavo Padovan [Fri, 27 Apr 2012 02:00:37 +0000 (23:00 -0300)]
monitor: add version information
Gustavo Padovan [Fri, 27 Apr 2012 02:00:36 +0000 (23:00 -0300)]
monitor: add help information
Andrzej Kaczmarek [Wed, 25 Apr 2012 15:25:19 +0000 (17:25 +0200)]
gatttool: Fix handle parsing in cmd_char_write()
errno is not set to 0 before call to strtoll() so cmd_char_write() can fail
even with proper data. Use strtohandle() helper function instead.
Johan Hedberg [Thu, 26 Apr 2012 06:40:31 +0000 (09:40 +0300)]
hciops: Fix event mask for Inquiry Result with RSSI
The right value is 0x02 and not 0x04 (which is "Read Remote Extended
Features Complete").
Andre Guedes [Wed, 25 Apr 2012 01:42:31 +0000 (22:42 -0300)]
l2test: Add option to inform the address type
This patch adds 'V' option to inform the address type. Possible values
are: "bredr", "le_public", and "le_random".
Claudio Takahasi [Wed, 25 Apr 2012 01:42:30 +0000 (22:42 -0300)]
Remove MGMT address type definition
This patch replaces the address type defined in management interface
header by the new address type constants defined in the bluetooth.h.
Claudio Takahasi [Wed, 25 Apr 2012 01:42:29 +0000 (22:42 -0300)]
Remove addr_type_t definition from adapter.h
This patch replaces addr_type_t by the new Bluetooth address type
constants defined in bluetooth.h
Claudio Takahasi [Wed, 25 Apr 2012 01:42:28 +0000 (22:42 -0300)]
Remove leftover field in remote_dev_info struct
Field "bdaddr_type" of the remote_dev_info structure is not being
referenced in the code. Field "type" is already being used to store
the Bluetooth address type.
Claudio Takahasi [Wed, 25 Apr 2012 01:42:27 +0000 (22:42 -0300)]
Add address type for BLE bt_io_connect calls
This patch adds the address type option in bt_io_connect calls for BLE
devices. BR/EDR is the default value, and it is not mandatory to inform
it. For BLE devices, it is necessary to inform if the type is public or
random.
Claudio Takahasi [Wed, 25 Apr 2012 01:42:26 +0000 (22:42 -0300)]
btio: Add address type in bt_io_connect
This patch adds a new BtIO option to allow setting the remote Bluetooth
address type for BLE connections. Allowed values for BT_IO_OPT_DEST_TYPE
option are: BDADDR_BREDR, BDADDR_LE_PUBLIC, and BDADDR_LE_RANDOM.
Claudio Takahasi [Wed, 25 Apr 2012 01:42:25 +0000 (22:42 -0300)]
Add Bluetooth address type in sockaddr_l2
This patch adds the address type information to sockaddr_l2 structure,
allowing the userspace to inform the remote address type required for
LE Create Connection command.
Luiz Augusto von Dentz [Thu, 19 Apr 2012 18:03:35 +0000 (21:03 +0300)]
core: Remove DiscoverSchedulerInterval option from main.conf
This option was designed to be the time where name resolving would happen
in between inquiries, but with the introdution of mgmt interface this is
no longer the case as it treats the discovery as both inquiry/scan + name
resolving and keeping it just to delay discovery rounds is probably
useless.
Gustavo Padovan [Fri, 20 Apr 2012 17:36:15 +0000 (14:36 -0300)]
input: add error message when auth fails
Mikel Astiz [Mon, 23 Apr 2012 08:34:35 +0000 (10:34 +0200)]
core: Fix adapter mode inconsistency
The mode change in btd_adapter_start() doesn't check if the adapter is
discoverable, creating an inconsistency between the plugin and the core.
Using the management interface it is easy to reproduce the problem, by
starting BlueZ while the adapter is discoverable. After this, the state
cannot be changed.
Scott James Remnant [Wed, 18 Apr 2012 22:01:01 +0000 (15:01 -0700)]
hciops: map KeyboardDisplay IO Capability to DisplayYesNo
Since hciops is not used for LE devices, if an agent announces the
KeyboardDisplay IO Capability when creating a bonding, map it to
DisplayYesNo.
This matches the existing change in hciops_set_io_capability which
is only used in the RegisterAgent path.
Johan Hedberg [Wed, 18 Apr 2012 11:15:50 +0000 (14:15 +0300)]
audio: Fix AVDTP Reconfigure Reject message
The specification describes a proper reject response to
AVDTP_Reconfigure, so the previously used unknown command response is
not correct.
Daniel Wagner [Wed, 18 Apr 2012 08:53:37 +0000 (10:53 +0200)]
Makefile: Always add all test scripts to EXTRA_DIST
With commit
e407e9ac6d8bfde31f95ed5d7e7bcecf0cfe48ed
test-attrib, test-proximity and test-thermometer are only packaged
when "./configure --enable-gatt && make dist" is called.
The dist target should be independant on the configure flags.
Syam Sidhardhan [Mon, 16 Apr 2012 13:01:38 +0000 (18:31 +0530)]
eir: Fix incorrect eir_has_data_type() parsing
Updating the "parsed" variable twice inside the for loop, leads to
incorrect parsing.
Syam Sidhardhan [Mon, 16 Apr 2012 13:01:37 +0000 (18:31 +0530)]
eir: Fix incorrect eir_length() parsing
Issue:
The COD value displayed via dbus during inquiry is wrong.
This is because of the incorrect return length of the eir_length(),
which leads to appending the COD at wrong location.
Analysis:
After appending the COD at the end of the eir data, we can see
there are some '00' present in the eir field length in the eir file.
XX:XX:XX:XX:XX:XX 07095359414D5043020A040B0312111F110C110E110311
0000000000000000000000040D000142
Fix:
Corrected the length calculation in eir_length(), which is determining,
which position the COD should append
Syam Sidhardhan [Mon, 16 Apr 2012 13:01:36 +0000 (18:31 +0530)]
sdp: Fix compilation warning due to data type mismatch
In certain cross compiling environment, if we build the code with
enabling the maintainer mode, the following error happens.
cc1: warnings being treated as errors
lib/sdp.c: In function 'sdp_process':
lib/sdp.c:4111:6: error: comparison between signed and unsigned integer
expressions
make[2]: *** [lib/sdp.lo] Error 1
make[1]: *** [all] Error 2
Chan-yeol Park [Mon, 16 Apr 2012 11:56:34 +0000 (20:56 +0900)]
audio: Add missing debug message in authorization
Chan-yeol Park [Mon, 16 Apr 2012 11:56:33 +0000 (20:56 +0900)]
adapter: Add warning if there is no authorization agent
Sometimes BlueZ Agent is disappear suddenly or not registered
because of agent problem. but Bluez just prints "Operation Not
permiited"
This will help debugging.
Chan-yeol Park [Mon, 16 Apr 2012 11:56:32 +0000 (20:56 +0900)]
src: Add function for printing warnings
Marcel Holtmann [Mon, 16 Apr 2012 16:22:24 +0000 (18:22 +0200)]
build: Remove glib-compat.h support
Marcel Holtmann [Mon, 16 Apr 2012 16:17:30 +0000 (18:17 +0200)]
build: Require GLib 2.28 or later
Marcel Holtmann [Mon, 16 Apr 2012 15:57:08 +0000 (17:57 +0200)]
gdbus: Remove unneeded NEED_DBUS_WATCH_GET_UNIX_FD check
Marcel Holtmann [Mon, 16 Apr 2012 16:15:25 +0000 (18:15 +0200)]
build: Require D-Bus 1.4 or later
Chan-yeol Park [Mon, 16 Apr 2012 02:59:59 +0000 (11:59 +0900)]
audio: Add SBC NULL check for GST AVDTP signaling
Due to RF condition or headset malfunctioning, gst audio plug-in could
not get SBC (mandatory codec) information.
Chan-yeol Park [Sat, 14 Apr 2012 05:43:24 +0000 (14:43 +0900)]
audio: Fix default disabled values in audio.conf
Disabled features in audio.conf should be the same with source code
"audio/manager.c:struct enabled_interfaces enabled".
Szymon Janc [Fri, 13 Apr 2012 10:11:43 +0000 (12:11 +0200)]
input: Disconnect SDP before connecting HID if HIDSDPDisable is set
If remote HID device has HIDSDPDisable attribute present and set to
true host should disconnect SDP channel before connecting HID control
channel. Such devices multiplex resources between SDP and HID L2CAP
channels and will reject connection to HID control channel if SDP
channel is already connected.
Szymon Janc [Fri, 13 Apr 2012 10:11:42 +0000 (12:11 +0200)]
sdp-client: Add ability to clear cached session
This will be usefull for disconnecting SDP L2CAP channel before cached
session timeout fires.
Anderson Lizardo [Thu, 12 Apr 2012 16:04:48 +0000 (12:04 -0400)]
Makefile.am: Fix indentation
Only these two lines were using spaces instead of tabs.
Syam Sidhardhan [Thu, 12 Apr 2012 15:03:10 +0000 (20:33 +0530)]
Add Syam's email address
Syam Sidhardhan [Thu, 12 Apr 2012 15:03:09 +0000 (20:33 +0530)]
doc: Fix typo
Syam Sidhardhan [Thu, 12 Apr 2012 15:03:08 +0000 (20:33 +0530)]
src: Fix newline before EOF
Syam Sidhardhan [Thu, 12 Apr 2012 15:03:07 +0000 (20:33 +0530)]
input: Fix newline before EOF
Syam Sidhardhan [Thu, 12 Apr 2012 15:03:06 +0000 (20:33 +0530)]
audio: Fix newline before EOF
Syam Sidhardhan [Thu, 12 Apr 2012 15:03:05 +0000 (20:33 +0530)]
network: Remove header declared but not defined
Syam Sidhardhan [Thu, 12 Apr 2012 15:03:04 +0000 (20:33 +0530)]
headset: Remove header declared but not defined
Syam Sidhardhan [Thu, 12 Apr 2012 15:03:03 +0000 (20:33 +0530)]
adapter: Remove headers declared but not defined
Syam Sidhardhan [Thu, 12 Apr 2012 15:03:02 +0000 (20:33 +0530)]
sap-u8500: Fix empty parameter list in function declaration
Syam Sidhardhan [Thu, 12 Apr 2012 15:03:01 +0000 (20:33 +0530)]
sdpd-request: Fix passing zero instead of NULL
Syam Sidhardhan [Thu, 12 Apr 2012 15:03:00 +0000 (20:33 +0530)]
agent: Fix double D-Bus message unref
Inside the g_dbus_send_message() we are already doing the
D-BUS message unref.
Syam Sidhardhan [Thu, 12 Apr 2012 15:02:57 +0000 (20:32 +0530)]
gateway: Fix D-Bus reply memory leak
Re-arrange the code to avoid the D-Bus memory leak. Earlier if rfcomm
is not connected, then there was a reply memory leak.
Syam Sidhardhan [Thu, 12 Apr 2012 15:02:56 +0000 (20:32 +0530)]
gateway: Fix D-Bus memory leak in audio
Syam Sidhardhan [Thu, 12 Apr 2012 15:02:55 +0000 (20:32 +0530)]
gstavdtpsink: Fix D-Bus memory leak in audio
Syam Sidhardhan [Thu, 12 Apr 2012 15:02:54 +0000 (20:32 +0530)]
agent: Free the agent request properly
In the case of error, Unref the req->msg which is allocated inside the
pincode_request_new()
Syam Sidhardhan [Thu, 12 Apr 2012 15:02:53 +0000 (20:32 +0530)]
avrcp: Fix the response ctype of SetPlayerApplicationSettingValue
Request ctype of "SetPlayerApplicationSettingValue" is CONTROL,
so response should be ACCEPTED instead of STABLE.
Syam Sidhardhan [Thu, 12 Apr 2012 15:02:52 +0000 (20:32 +0530)]
health: Fix memory leak
Michal Labedzki [Wed, 11 Apr 2012 11:05:18 +0000 (13:05 +0200)]
AVRCP: Add missing capability for TrackReachedEnd
Add missing capability for TrackReachedEnd.
Anderson Lizardo [Tue, 10 Apr 2012 20:22:48 +0000 (16:22 -0400)]
audio: Fix memory leak when headset server probe() fails
If RFCOMM is disabled on the kernel, headset_server_probe() fails.
Relevant log messages:
audio/manager.c:headset_server_probe() path /org/bluez/499/hci0
src/adapter.c:btd_adapter_ref() 0x4bb4f78: ref=6
audio/manager.c:audio_adapter_ref() 0x4ca3010: ref=1
socket(STREAM, RFCOMM): Protocol not supported (93)
audio/manager.c:audio_adapter_unref() 0x4ca3010: ref=0
src/adapter.c:btd_adapter_unref() 0x4bb4f78: ref=5
audio-headset: Operation not permitted (1)
The powered callback should only be registered if adapter driver probe
was successful. The callback unregister was moved to the beginning of
headset_server_remove() for consistency.
This fixes this memory leak:
==499== 8 bytes in 1 blocks are definitely lost in loss record 44 of 182
==499== at 0x4826444: malloc (vg_replace_malloc.c:263)
==499== by 0x4877243: g_malloc (gmem.c:132)
==499== by 0x488D088: g_slice_alloc (gslice.c:836)
==499== by 0x488E8A5: g_slist_append (gslist.c:230)
==499== by 0x18AEEE: btd_adapter_register_powered_callback
(adapter.c:3416)
==499== by 0x11AF61: headset_server_probe (manager.c:919)
==499== by 0x18B67B: probe_driver (adapter.c:2033)
==499== by 0x1908F5: adapter_init (adapter.c:2048)
==499== by 0x189D20: btd_manager_register_adapter (manager.c:397)
==499== by 0x1649AF: mgmt_cmd_complete (mgmtops.c:1075)
==499== by 0x16665E: mgmt_event (mgmtops.c:1780)
==499== by 0x48B2EFA: g_io_unix_dispatch (giounix.c:162)