OSDN Git Service
Bruna Moreira [Tue, 27 Mar 2012 18:06:47 +0000 (14:06 -0400)]
eir: Read GAP appearance from advertising data
This was introduced on Core Specification Supplement (CSS) v1.
Bruna Moreira [Thu, 29 Mar 2012 14:38:02 +0000 (10:38 -0400)]
device: Appearance characteristic from GAP service
After searching for primary services, the appearance characteristic is
read from GAP service and saved in "appearance" file. Running
GetProperty in device interface, will show the respective icon based in
"appearance" value.
Bruna Moreira [Tue, 27 Mar 2012 13:29:36 +0000 (09:29 -0400)]
storage: Add storage to GAP characteristic
Add functions to read and write remote Appearance characteristic. The values
are saved in "appearance" file.
Bruna Moreira [Tue, 27 Mar 2012 13:29:35 +0000 (09:29 -0400)]
dbus-common: Map GAP Appearance characteristic to icon
Add function appearance_to_icon() to map values from Appearance
Characteristic to icon available.
Note: The Appearance Characteristic is composed of a category (10-bits)
and sub-categories (6-bits). These categories are defined at:
http://developer.bluetooth.org/gatt/characteristics/Pages/
CharacteristicViewer.aspx?u=org.bluetooth.characteristic.gap.appearance.xml
Anderson Lizardo [Tue, 27 Mar 2012 20:43:48 +0000 (16:43 -0400)]
attrib: Attempt to remove attio callback on watcher exit
When registering a watcher, it is attempted to add an attio callback.
Therefore, when the watcher is unregistered or exits, this attio should
be removed if there are no other users.
This is already done when unregistering a watcher, it just missed when
it exits without unregistering.
Claudio Takahasi [Tue, 27 Mar 2012 20:43:47 +0000 (16:43 -0400)]
attrib: Fix missing D-Bus message unref on DiscoverCharacteristics
This patch fixes a missing message unref when DiscoverCharacteristics on
Generic Attribute finishes.
Claudio Takahasi [Tue, 27 Mar 2012 20:43:45 +0000 (16:43 -0400)]
attrib: Register only new characteristics found
This avoids duplicating registration for previously found
characteristics.
Claudio Takahasi [Tue, 27 Mar 2012 20:43:44 +0000 (16:43 -0400)]
attrib: Move Characteristic Discovery reply to a new function
Minor cleanup patch moving the creation of the reply for Discover
All Characteristics call to a new local function in the Generic
Attribute.
Claudio Takahasi [Tue, 27 Mar 2012 20:43:43 +0000 (16:43 -0400)]
attrib: Discover Characteristics if already connected
On Generic Attribute, Discover All Characteristics of a Service can be
started immediately if already connected, without registering a ATTIO
connection callback.
Claudio Takahasi [Tue, 27 Mar 2012 20:43:42 +0000 (16:43 -0400)]
attrib: Centralize ATTIO connection management
On Generic Attribute, register only one ATTIO connection callback to
centralize the connection management.
Also make sure the connection reference is properly dropped only if
there is no pending operation.
Szymon Janc [Thu, 29 Mar 2012 14:12:44 +0000 (16:12 +0200)]
Fix common spelling error in code and comments
Correct spelling is "incoming" not "incomming".
Chen Ganir [Wed, 28 Mar 2012 15:26:44 +0000 (17:26 +0200)]
core: Fix device version in GetProperties
Fix how device version is added to the property dictionary.
Szymon Janc [Wed, 28 Mar 2012 10:03:19 +0000 (12:03 +0200)]
Move common code to sdp to avoid duplication
LanguageBaseAttributeIDList is set from few places and each time static
function was copied and used for that. Move this funtion to sdp code
and use that instead of copying code.
Szymon Janc [Wed, 28 Mar 2012 10:03:18 +0000 (12:03 +0200)]
Bump Device ID service version to 1.3
Device ID information in EIR are specified in DID 1.3 and this is
supported by BlueZ.
Szymon Janc [Wed, 28 Mar 2012 10:03:17 +0000 (12:03 +0200)]
btmgmt: Add support for setting Device ID
Szymon Janc [Wed, 28 Mar 2012 10:03:16 +0000 (12:03 +0200)]
mgmtops: Add support for setting Device ID
Szymon Janc [Wed, 28 Mar 2012 10:03:15 +0000 (12:03 +0200)]
Add support for setting VID source in DeviceID from config file
This allows to set if VID source is Bluetooth SIG or USB. Assigner is
provided as string {bluetooth,usb} and fallback to usb if none is set
in config.
Szymon Janc [Wed, 28 Mar 2012 10:03:14 +0000 (12:03 +0200)]
Set DeviceID when registering adapter not when sdp server is starting
Adapter ops are registered after sdp server is started. Thus no
adapters were registered yet when setting DID on sdp startup and DID
was never set.
Szymon Janc [Wed, 28 Mar 2012 10:03:13 +0000 (12:03 +0200)]
Parse config DeviceID string on bluetoothd startup
Instead of parsing config DeviceID string on sdp server startup, parse
it in main while reading config file. This allow to store logical DID
values in main_opts (instead of raw config string) and make use of them
in bluetoothd code.
João Paulo Rechi Vita [Tue, 27 Mar 2012 21:14:32 +0000 (18:14 -0300)]
input: fix coding style
Fix line going over 80 columns.
Jefferson Delfes [Tue, 27 Mar 2012 13:21:55 +0000 (09:21 -0400)]
gatttool: Update interactive prompt if connection is lost
In interactive mode, when connection is lost, the prompt used to remain
in "connected" state. This patch fixes that case, by always showing the
actual connection state.
Anderson Lizardo [Tue, 27 Mar 2012 13:06:14 +0000 (09:06 -0400)]
event: Store Class of Device after connection
The Class of Device is usually sent on the EIR returned by the Device
Connected mgmt event. For hciops, the Class of Device is always NULL on
the connection complete event.
The stored Class of Device is used later for Class/Icon device
properties.
Elvis Pfützenreuter [Tue, 27 Mar 2012 13:03:25 +0000 (10:03 -0300)]
HDP: Omit MainChannel when first reliable channel is nil
Do not add MainChannel to property list when HDP first reliable channel
does not exist. Upon channel destruction, ChannelDeleted signal is already
emitted.
This also fixes a crash where MainChannel was filled with invalid path ""
when first reliable channel did not exist.
Michael Brudevold [Fri, 23 Mar 2012 15:03:47 +0000 (10:03 -0500)]
avdtp: fix audio source init check
Frédéric Dalleau [Mon, 19 Mar 2012 14:04:07 +0000 (15:04 +0100)]
sdp: Fix sdp_get_profile_descs for Mac Os X Lion
Mac OS X 10.7.3 publishes the following SDP record for HFP Gateway.
record #2
aid 0x0000 (SrvRecHndl)
uint 0x10003
aid 0x0001 (SrvClassIDList)
< uuid-16 0x111f (Handsfree AG) uuid-16 0x1203 (Audio) >
aid 0x0004 (ProtocolDescList)
< < uuid-16 0x0100 (L2CAP) > <
uuid-16 0x0003 (RFCOMM) uint 0x2 > >
aid 0x0005 (BrwGrpList)
< uuid-16 0x1002 (PubBrwsGrp) >
aid 0x0006 (LangBaseAttrIDList)
< uint 0x656e uint 0x6a uint 0x100 >
aid 0x0009 (BTProfileDescList)
< uuid-16 0x111e (Handsfree) uint 0x105 >
aid 0x0100 (SrvName)
str "Hands Free Audio Gateway"
aid 0x0301 (SuppDataStoresList)
uint 0x0
aid 0x0311 (SuppFeatures)
uint 0x0
BlueZ do not read correctly the version number of the profile.
The patch fixes this.
From my understanding, the SDP profile descriptor is a list. The for
loop in sdp_get_profile_desc iterates that list. Until now, bluez
expected a collection of either uuid or lists with a uuid and a
version number. This can be represented as follow:
list=(uuid, uuid, (uuid, version))
The patch fixes the situation where the profile descriptor list
contains a uuid AND a version number both on the root level and not
enclosed in a sublist. If the next item after uuid is uint16, then
this is the version number. This can be represented as follow:
list=(uuid, uuid, version, uuid, (uuid, version))
Note that core v4.0 spec. states that the Profile Descriptor
list is a data sequence of profile descriptors. Each profile descriptor
is a data sequence whose first element is a UUID and second element is
a version number. This can be represented as follow:
list=((uuid, version), (uuid, version))
Chen Ganir [Mon, 26 Mar 2012 07:35:59 +0000 (09:35 +0200)]
GATT: Replace start/end with att_range
Use att_range struct instead of individual start/end handles in
gatt_primary structure
Chen Ganir [Mon, 26 Mar 2012 07:35:58 +0000 (09:35 +0200)]
GATT: Rename service and char structs
Rename the att_primary to gatt_primary and att_char to gatt_char.
Characteristic and Service do not exist in the ATT spec, only in
GATT.
Chen Ganir [Mon, 19 Mar 2012 14:57:53 +0000 (16:57 +0200)]
GATT: Profile support for EnableGatt
Add support for the EnableGatt for all GATT profiles.
Chen Ganir [Mon, 19 Mar 2012 14:57:52 +0000 (16:57 +0200)]
GATT: Rename AttributeServer switch
Rename the AttributeServer main.conf option to EnableGatt and
change its purpose to enable/disable all GATT related activity.
Chen Ganir [Mon, 19 Mar 2012 14:57:51 +0000 (16:57 +0200)]
GATT: Remove individual config switches
Remove individual GATT plugin configuration switches and add a
new master gatt switch called --enable-gatt to enable/disable all
GATT related plugins at once.
Vinicius Costa Gomes [Tue, 20 Mar 2012 17:49:33 +0000 (14:49 -0300)]
Fix not marking a reverse SDP discovery as "reverse"
Some devices "hide" some of their records while connected to other
devices. Setting a service discovery as reverse makes BlueZ to not
remove the records that are "hidden".
This patch fixes a mistake that reverse parameter was changed to false
when it should be true.
Chan-yeol Park [Wed, 22 Feb 2012 14:15:26 +0000 (23:15 +0900)]
Remove leftover code of main_opts operation
David Herrmann [Wed, 22 Feb 2012 20:29:51 +0000 (21:29 +0100)]
wiimote: Detect Wii Remotes by device name
We currently detect Wii Remotes by DID information. However, not all Wii
Remotes (especially 3rd party Wii Remotes) provide DID information.
Moreover, they provide no way of identifying them aside from their
device name. Hence, this adds an experimental name-detection for Wii
Remotes.
Claudio Takahasi [Mon, 12 Mar 2012 20:05:27 +0000 (17:05 -0300)]
input: Remove unneeded alias
Claudio Takahasi [Mon, 12 Mar 2012 20:05:26 +0000 (17:05 -0300)]
input: Remove unneeded Bluetooth Address arguments
Vinicius Costa Gomes [Wed, 14 Mar 2012 22:07:40 +0000 (19:07 -0300)]
attrib-server: Fix sending "not supported" for indications/notifications
Now that our Attribute Client is able to handle notifications and
indications we don't need to respond that they are not supported.
Arnaud Mouiche [Thu, 25 Aug 2011 10:09:05 +0000 (12:09 +0200)]
HFP gateway: fix failure on very first GSM connection
This patch fix the very first incoming connection from a GSM device
(playing the gateway role), when 'device->gateway' is NULL (when we
didn't perform a SDP browse request yet)
we add the service with 'btd_device_add_uuid(device->btd_dev,
remote_uuid)' but we provide HFP_HS_UUID as remote_uuid. Consequently,
the HFP headset service is activated instead the gateway service.
Arik Nemtsov [Thu, 8 Mar 2012 13:44:39 +0000 (15:44 +0200)]
Revert "core: Add provision for terminating an ATT connection"
This reverts commit
f89a77478af78d41c80ab7605662382b9e4e1c36.
This is not needed and actually introduces a bug. When the "Disconnect"
API of device is called device->attrib is unref-ed via a watch set on
G_IO_HUP. The channel is shutdown when the last reference is removed.
The code introduced here shuts down the channel and prevents the watch
from getting called. This means we leak a reference to device->attrib.
This can cause a number of bad things. For example, if the device is
temporary, it will never be freed, and we won't be able to pair to it
again.
Arik Nemtsov [Thu, 8 Mar 2012 13:44:38 +0000 (15:44 +0200)]
device: don't auto-connect on disc-cb attio callback registration
If a device is already connected, don't auto-connect if we register
a disconnect-only attio callback. This will obviously fail.
Frédéric Dalleau [Mon, 20 Feb 2012 20:57:57 +0000 (21:57 +0100)]
sdp: Check type of sdp data before dereferencing
Mikel Astiz [Tue, 13 Mar 2012 13:19:56 +0000 (14:19 +0100)]
audio: fix io channel shutdown on error
In the unlikely case of service record allocation error, the io channel
should be properly shut down.
Mikel Astiz [Tue, 13 Mar 2012 13:19:55 +0000 (14:19 +0100)]
audio: fix missing io channel shutdown
Channel must be explicitly shut down because otherwise the reference
counter never reaches zero, due to the server installed by bt_io_listen.
Mikel Astiz [Tue, 13 Mar 2012 13:19:54 +0000 (14:19 +0100)]
audio: fix missing unref in case of error
audio_adapter_get() increases the reference counter of the adapter, so
it's necessary to decrement it in case of error.
Michal Labedzki [Wed, 7 Mar 2012 16:23:14 +0000 (17:23 +0100)]
avrcp: Add error code for rejected vendor command
Fix response for any vendor commands when there are no players
registered. According to the specification responses should be
REJECTED with error code.
"For error response PDU the response parameter is always the
error code independent of the response format defined for
ACCEPTED PDU response for the corresponding PDU command."
(source: section 4.7.9 of AVRCP 1.3 specification)
Michal Labedzki [Thu, 1 Mar 2012 14:45:33 +0000 (15:45 +0100)]
avrcp: Fix response ctype of AbortContinuingResponse
Request ctype of "AbortContinuingResponse" is CONTROL, so response
should be ACCEPTED instead of STABLE. This affect PTS Test Case for
TP/RCR/BV-04-C.
Johan Hedberg [Mon, 12 Mar 2012 12:03:57 +0000 (14:03 +0200)]
test-discovery: Fix printing non-ASCII characters
Frédéric Dalleau [Tue, 21 Feb 2012 16:46:50 +0000 (17:46 +0100)]
media: register disconnect watch at transport add
disconnect watch is not removed when the media owner is freed if
an error occurs while resuming in acquire (id == 0).
Frédéric Dalleau [Tue, 21 Feb 2012 16:46:49 +0000 (17:46 +0100)]
gateway: Fix crash if SCO connection fails
In some situations, a connect callback is created, but
this callback is not added to media_owner. Thus when the owner
is destroyed and at rfcomm disconnect, the callback is executed
with an invalid pointer.
Anderson Lizardo [Tue, 6 Mar 2012 20:46:54 +0000 (16:46 -0400)]
Add emulator/btvirt to .gitignore
Marcel Holtmann [Tue, 6 Mar 2012 17:04:22 +0000 (09:04 -0800)]
Release 4.99
Marcel Holtmann [Tue, 6 Mar 2012 17:02:51 +0000 (09:02 -0800)]
Update library version
Marcel Holtmann [Tue, 6 Mar 2012 16:59:58 +0000 (08:59 -0800)]
lib: Update company identifiers
Johan Hedberg [Tue, 6 Mar 2012 00:36:22 +0000 (16:36 -0800)]
lib: Add a2mp.h to lib_headers
Peter Krystad [Tue, 6 Mar 2012 00:14:48 +0000 (16:14 -0800)]
lib: Add A2MP definitions
Szymon Janc [Mon, 5 Mar 2012 19:05:25 +0000 (20:05 +0100)]
mgmtops: Use DBG not error for printing name in read_info_complete
This is debug not an error message.
Szymon Janc [Mon, 5 Mar 2012 19:05:24 +0000 (20:05 +0100)]
mgmtops: Add debug print to update_settings
Print settings for easier tracking what has changed.
Szymon Janc [Mon, 5 Mar 2012 19:05:23 +0000 (20:05 +0100)]
mgmtops: When settings discoverable also set connectable if not set yet
Otherwise command would be rejected e.g. when setting discoverable at
bluetoothd startup.
Szymon Janc [Mon, 5 Mar 2012 18:58:39 +0000 (19:58 +0100)]
Remove not needed total variable in get_ltk_info
This variable is not really needed and was mostly dead assigned.
Jeff Hansen [Mon, 5 Mar 2012 03:44:19 +0000 (20:44 -0700)]
input: Set up uinput device again, if it was previously closed.
If you connect a PS3 controller to bluetoothd as an input device, then
take the batteries out of a PS3 controller, then put them back in and
push a few buttons, it will eventually cause an error which causes the
uinput socket to be closed. It will then re-connect to bluetoothd, but
the uinput socket fd will be -1, so it needs to be set up again after
the re-connect.
Patrick Ohly [Mon, 16 Jan 2012 10:11:55 +0000 (11:11 +0100)]
lib: Fix compile issue when using in C++
The compiler error is:
/usr/include/bluetooth/bluetooth.h::131:9: error: invalid conversion from 'void*' to 'bt_get_le64(void*)::<anonymous struct>*'
...
The reason is that C++, in contrast to C, does not allow conversion of
void * to anything, and this code gets compiled as C++ when the app is
written in C++. The macro with the assignment itself is older, but only
recent Bluez starts to use it in inline functions, thus triggering the
problem.
This patch keeps the "struct __attribute__((packed))" magic and merely
changes the typecast so that it works in C and C++. Like the existing
macro this patch relies on support for typeof.
Johan Hedberg [Sat, 3 Mar 2012 03:12:38 +0000 (19:12 -0800)]
core: Fix calling stop_discovery through btd_adapter_stop
Luiz Augusto von Dentz [Thu, 1 Mar 2012 09:15:04 +0000 (11:15 +0200)]
core: remove set_limited_discoverable from adapter_ops driver
Core should not longer need to take care of details of discoverable bits
Luiz Augusto von Dentz [Thu, 1 Mar 2012 09:15:03 +0000 (11:15 +0200)]
core: Make adapter_ops->set_discoverable to take discoverable timeout
This enables the driver to implements its own handling of the timeout
Luiz Augusto von Dentz [Fri, 2 Mar 2012 08:51:18 +0000 (10:51 +0200)]
btiotest: Add option to update security level while connecting
Johan Hedberg [Fri, 2 Mar 2012 23:34:34 +0000 (15:34 -0800)]
mgmtops: Use proper boolean value for mgmt_set_powered call
Szymon Janc [Fri, 2 Mar 2012 14:58:36 +0000 (15:58 +0100)]
SAP: Make PADDING4 macro a bit more robust
Bitwise operations are done before arithmetic operations and this might
cause undesired results if macro is called with parameter of form x+y.
Johan Hedberg [Fri, 2 Mar 2012 22:29:08 +0000 (00:29 +0200)]
core: Remove name and class setting from btd_adapter_start()
Especially for mgmtops it's important that these values are pushed down
in the stack *before* powering on, so btd_adapter_start() is the wrong
place. Instead, adapter functions are added to hciops/mgmtops can fetch
these values at the right point during adapter init and push them down
in the stack (to the kernel).
Johan Hedberg [Fri, 2 Mar 2012 01:24:13 +0000 (03:24 +0200)]
mgmtops: Use error() instead of DBG() for command failures
Johan Hedberg [Fri, 2 Mar 2012 01:17:28 +0000 (03:17 +0200)]
mgmt: Add missing error code definitions
Johan Hedberg [Fri, 2 Mar 2012 00:52:39 +0000 (02:52 +0200)]
mgmt: Move string helpers to lib/mgmt.c
Marcel Holtmann [Fri, 2 Mar 2012 00:56:59 +0000 (16:56 -0800)]
monitor: Add support for storing btsnoop formatted logs
Johan Hedberg [Fri, 2 Mar 2012 00:13:42 +0000 (02:13 +0200)]
mgmtops: Fix clearing of pending_uuids after g_slist_free_full()
Johan Hedberg [Thu, 1 Mar 2012 23:38:42 +0000 (01:38 +0200)]
audio: Remove redundant state_changed() call in headset_server_probe()
Johan Hedberg [Thu, 1 Mar 2012 23:38:02 +0000 (01:38 +0200)]
core: Remove bogus adapter->up check from probe_driver()
Jefferson Delfes [Wed, 29 Feb 2012 20:17:31 +0000 (16:17 -0400)]
gatttool: Add command completion for interactive mode
Anderson Lizardo [Wed, 29 Feb 2012 19:57:15 +0000 (15:57 -0400)]
mgmtops: Add CANCEL_PAIR_DEVICE command complete handling
This avoids the "Unknown command complete for opcode 26" error.
Peter Krystad [Wed, 15 Feb 2012 19:04:11 +0000 (11:04 -0800)]
lib: Add L2CAP Create/Move Channel definitions
Johan Hedberg [Wed, 29 Feb 2012 18:59:54 +0000 (12:59 -0600)]
mgmt: Add error code definitions
Johan Hedberg [Wed, 29 Feb 2012 18:59:31 +0000 (12:59 -0600)]
mgmt-api: Add error code descriptions
Marcel Holtmann [Wed, 29 Feb 2012 17:05:12 +0000 (09:05 -0800)]
monitor: Add defines for connection packet type changes
Johan Hedberg [Tue, 28 Feb 2012 00:05:42 +0000 (02:05 +0200)]
lib: Add missing host feature defines
This patch add missing SSP and "Simultaneous LE & BR/EDR" feature bit
definitions to hci.h.
Marcel Holtmann [Mon, 27 Feb 2012 18:56:43 +0000 (10:56 -0800)]
emulator: Add initial version of new emulator
Marcel Holtmann [Mon, 27 Feb 2012 18:52:14 +0000 (10:52 -0800)]
monitor: Add Bluetooth HCI declarations
Marcel Holtmann [Mon, 27 Feb 2012 18:35:31 +0000 (10:35 -0800)]
monitor: Use better signal integration
Marcel Holtmann [Mon, 27 Feb 2012 17:19:14 +0000 (09:19 -0800)]
hciemu: Fix wrong error code in host control
Marcel Holtmann [Mon, 27 Feb 2012 08:55:34 +0000 (00:55 -0800)]
monitor: Add support for hcidump fallback
Marcel Holtmann [Mon, 27 Feb 2012 06:42:41 +0000 (22:42 -0800)]
lib: Make unaligned access functions const
Marcel Holtmann [Mon, 27 Feb 2012 06:12:34 +0000 (22:12 -0800)]
lib: Add definition for HCI_CHANNEL_MONITOR
Marcel Holtmann [Mon, 27 Feb 2012 03:11:44 +0000 (19:11 -0800)]
monitor: Fix some minor white space damages
Marcel Holtmann [Mon, 27 Feb 2012 02:39:56 +0000 (18:39 -0800)]
monitor: Add Intel copyright to new monitor tool
Marcel Holtmann [Mon, 27 Feb 2012 02:39:23 +0000 (18:39 -0800)]
hciattach: Remove some unneeded includes
Marcel Holtmann [Mon, 27 Feb 2012 02:34:29 +0000 (18:34 -0800)]
hciattach: Add new HCI UART flags
Andrei Emeltchenko [Fri, 24 Feb 2012 08:44:25 +0000 (10:44 +0200)]
l2test: Add BT Channel Policy option
Make l2test able to set channel policy socket option.
Andrei Emeltchenko [Fri, 24 Feb 2012 08:44:24 +0000 (10:44 +0200)]
l2test: Clean up lookup table code
Make lookup table code available for others and create
common functions.
Andrei Emeltchenko [Fri, 24 Feb 2012 08:44:23 +0000 (10:44 +0200)]
lib: Copy L2CAP chan policy defines from kernel
Those defines will be used by user space.
Luiz Augusto von Dentz [Fri, 24 Feb 2012 09:06:11 +0000 (11:06 +0200)]
mgmt-api: Format fixes
Siarhei Siamashka [Fri, 24 Feb 2012 12:13:16 +0000 (14:13 +0200)]
sbc: fix "always_inline function might not be inlinable" warning
"__attribute__((always_inline))" does not replace "inline" and they
still need to be used together. This fixes "always_inline function
might not be inlinable [-Wattributes]" warning in gcc 4.7
Anderson Lizardo [Fri, 24 Feb 2012 14:42:31 +0000 (10:42 -0400)]
mgmtops: Add missing compatibility header include
glib-compat.h needs to be included when either g_slist_free_full() or
g_list_free_full() is used on a C file. Otherwise, it will fail to build
on older (supported) Glib versions that lack these functions.
Fixes this build error on GLib 2.24.1:
plugins/mgmtops.c: In function ‘remove_controller’:
plugins/mgmtops.c:180: error: implicit declaration of function
‘g_slist_free_full’
Johan Hedberg [Fri, 24 Feb 2012 10:53:22 +0000 (12:53 +0200)]
mgmtops: Fix uninitialized discovery type value
Johan Hedberg [Fri, 24 Feb 2012 09:01:12 +0000 (11:01 +0200)]
mgmtops: Set discovery type based on real capabilities