summaryrefslogtreecommitdiffstats
path: root/src/device.c
Commit message (Collapse)AuthorAgeFilesLines
* Use MITM for CreatePairedDevice and handle the NoInputNoOuput casesMarcel Holtmann2009-01-161-1/+1
|
* Require MITM for dedicated bonding if the remote side is capable of itJohan Hedberg2009-01-161-0/+5
|
* Use L2CAP_LM for the L2CAP raw socket for nowJohan Hedberg2009-01-161-7/+10
|
* Don't use auth_required. New kernels still need explicit auth requestedJohan Hedberg2009-01-161-6/+1
|
* Don't try to remove bonding->io_id twice (bonding_request_free does it)Johan Hedberg2009-01-161-3/+0
|
* Fix bonding with pre-2.1 devices and newer kernelsJohan Hedberg2009-01-161-0/+5
| | | | | The L2CAP raw connection will return IO_OUT imediately so let the HCI even tracking take care of cleaning up the bonding struct.
* Use dedicated bonding and no MITM for pairingMarcel Holtmann2009-01-151-1/+1
|
* Don't remove linkkeys upon adapter removalJohan Hedberg2009-01-131-2/+3
|
* Add ReverseServiceDiscovery config optionJohan Hedberg2009-01-131-1/+2
| | | | | This option is solely for qualification since the BITE tester in its grand braindeadness doesn't like use doing reverse SDP.
* Add debug statement for reverse service discovery timerMarcel Holtmann2009-01-131-0/+1
|
* Don't do reverse SDP for renewed link keysJohan Hedberg2009-01-131-2/+13
| | | | | | Add a flag for a renewed link key to the device struct so that when we get authentication complete we don't do reverse SDP or PropertyChanged(Paired, true) if a new bonding isn't being created.
* Fix dbus_connection_ref/unref usageJohan Hedberg2009-01-131-1/+3
|
* Move l2raw_connect over to device.c (since it's not needed anywhere else)Johan Hedberg2009-01-121-18/+65
|
* Add support for dedicated bonding using BT_SECURITYJohan Hedberg2009-01-121-5/+22
|
* Handle active connection using device object.Luiz Augusto von Dentz2009-01-111-32/+44
|
* Move bonding creation to device.c.Luiz Augusto von Dentz2009-01-111-23/+669
|
* Update copyright informationMarcel Holtmann2009-01-011-1/+1
|
* Make two more local functions staticMarcel Holtmann2008-12-231-1/+1
|
* Add support for the LegacyPairing propertyJohan Hedberg2008-12-041-0/+7
|
* Once a driver has been removed break out and proceed to the next oneJohan Hedberg2008-11-181-3/+3
|
* Fix parameters to debug()Johan Hedberg2008-11-181-1/+1
|
* Prevent duplicated uuids from being probed.Luiz Augusto von Dentz2008-11-171-0/+5
| | | | | | | Since probe takes the device object and the uuid list it make no sense to have duplicated uuid being probed, since the driver wouldn't detect any difference between them (fetching from the cache would return the very same record).
* Make use of g_timeout_add_seconds whenever possible.Luiz Augusto von Dentz2008-11-171-2/+2
|
* Make use of sdp_copy_record.Luiz Augusto von Dentz2008-11-171-10/+1
|
* Emit DeviceCreated for temporary devices.Luiz Augusto von Dentz2008-11-111-8/+0
| | | | | | DeviceCreated should be emitted for temporary devices since there could be meaningful information for user applications. This also affects DeviceRemoved since it now should be emitted for temporary devices too.
* Code cleanup.Luiz Augusto von Dentz2008-11-101-3/+1
|
* Fix bug which makes DeviceCreated to not being emitted.Luiz Augusto von Dentz2008-11-101-12/+12
| | | | | DeviceCreated was not being emitted when the remote device initiate the connection.
* Remove debug prints that were added by mistake in the previous commitJohan Hedberg2008-11-081-12/+2
|
* Remove duplicate call to btd_device_get_recordJohan Hedberg2008-11-081-2/+12
|
* Simplify record uuid matching logicJohan Hedberg2008-11-081-15/+13
|
* Make device_match_patter() more readable by using more intuitive variable namesJohan Hedberg2008-11-081-10/+9
|
* Fix memory leakJohan Hedberg2008-11-081-4/+4
|
* Fix coding styleJohan Hedberg2008-11-081-5/+3
|
* Add support for device pattern drivers.Luiz Augusto von Dentz2008-11-071-27/+95
| | | | | | The support is necessary for driver like serial which should be able to register RFCOMM UUID but was unable to do so as we were only matching the profile itself.
* Use only L2CAP and PNP UUIDs for initial service discoveryJohan Hedberg2008-11-071-107/+30
| | | | | | | | | Since public browsing is unreliable on so many devices and the fixed list cannot cover all possible profiles (e.g. so far it didnt cover OBEX based profiles) use the L2CAP and PNP UUIDs since in practice they have shown to work with practicly all devices (older BlueZ versions is an exception). If these two UUIDs don't return anything use the public browse group as a last resort.
* Make the variable prefix really globalMarcel Holtmann2008-10-311-3/+2
|
* Use i instead of j in for-loopJohan Hedberg2008-10-291-3/+3
|
* Fix duplicate UUID detectionJohan Hedberg2008-10-291-14/+24
| | | | | | The original code didn't work because the continue statement in the innermost for-loop was supposed to act on the outer for-loop (which it obviously doesn't do).
* Convert Bluetooth UUID-128 to 32 or 16 bit if possible before searchingJohan Hedberg2008-10-291-0/+3
| | | | | When doing SDP queries try to convert Bluetooth UUID-128 to 32 or 16 bit form.
* Use g_dbus_send_message instead of dbus_connection_sendJohan Hedberg2008-10-241-4/+2
|
* Fix fallouts due to GDBusWatchFunction changesMarcel Holtmann2008-10-171-1/+1
|
* Add emit_array_property_changed()Johan Hedberg2008-10-161-2/+2
|
* Rename dbus_connection_emit_property_changed to emit_property_changedJohan Hedberg2008-10-161-16/+12
|
* Add dict_append_array()Johan Hedberg2008-10-161-1/+1
|
* Rename dbus_message_iter_append_dict_entry to dict_append_entryJohan Hedberg2008-10-161-19/+10
|
* Better debug when failing to probe a deviceBastien Nocera2008-10-151-2/+2
| | | | The previous error message was a bit terse.
* Fix uuid list allocation and freeing bug exposed by previous mem leak fixJohan Hedberg2008-10-101-6/+4
|
* Fix memory leakJohan Hedberg2008-10-101-0/+1
|
* Add mechanism for adding new UUIDs from the driver sideJohan Hedberg2008-10-091-5/+28
| | | | | | With this patch drivers can notify the core daemon that a device supports some UUID. This is useful in case we don't have a service record for a profile which the remote device initiates a connection for.
* Make device driver probe calls take UUID's instead of full recordsJohan Hedberg2008-10-091-18/+60
| | | | | This is preparation for allowing the device object to be correctly set up even when we don't have the full service records but only the remote UUID's.
generated by cgit (git 2.34.1) at 2024-06-18 15:55:03 +0000