| 80525c13 |
gdbus: add macro for methods marked as NOREPLY |
Lucas De Marchi |
13 years ago |
1 file, +7, -0 |
| 5c9af798 |
simple-agent: Fix input with python version < 3.0
input() in python < 3.0 is the same as eval(raw_input()) which is not
what we want. With python >= 3.0 in turn raw_input doesn't exist. This
patch fixes support for both versions by a simple try-except clause. |
Johan Hedberg |
13 years ago |
1 file, +11, -5 |
| 9d2f3c94 |
AVDTP: Fix rejecting AVDTP Start if starting flag is set
This is now handled by checking if the command collided. |
Luiz Augusto von Dentz |
13 years ago |
1 file, +2, -3 |
| 83630251 |
audio: Fix aborting A2DP setup while AVDTP Start is in progress
Change return of avdtp_start to -EINPROGRESS so the caller can check if
the operation is in progress and don't abort because of that. |
Luiz Augusto von Dentz |
13 years ago |
2 files, +3, -3 |
| a59e8210 |
AVDTP: Fix responding to ABORT with reject
ABORT command cannot be rejected |
Luiz Augusto von Dentz |
13 years ago |
3 files, +7, -13 |
| 2b3bf7b2 |
AVDTP: Do not respond ABORT command with invalid id
AVDTP spec, 8.15.2 Abort Response:
"If an AVDTP_ABORT_CMD contains an invalid SEID, no response shall be
sent." |
Luiz Augusto von Dentz |
13 years ago |
1 file, +2, -4 |
| 34d9bd7d |
audio: Wait remote side to send AVDTP_START when acting as acceptor
Some devices like Sony Ericsson MW600 reject AVDTP_START if it was the
initiator of the connection, apparently it follows recommendation 12 of
simultaneous use of HFP, A2DP and AVRCP profiles white paper which says:
"If the RD has configured and opened a stream it is also responsible to
start the streaming via GAVDP_START."
If the client is fast enough and try to acquire the transport this cause
an error, so instead of sending AVDTP_START the code now checks if it is
the acceptor of the stream and wait the remote side to send the command. |
Luiz Augusto von Dentz |
13 years ago |
1 file, +17, -1 |
| 4503dba4 |
audio: Fix handling of A2DP abort indication
When an abort is received all setup callbacks should return an error. |
Luiz Augusto von Dentz |
13 years ago |
1 file, +9, -0 |
| c1b89146 |
audio: Fix handling of A2DP start indication
Only process callbacks if avdtp_start was sent, otherwise it may cancel
setup callbacks that were registere via g_idle_add. |
Luiz Augusto von Dentz |
13 years ago |
1 file, +12, -4 |
| ef539d84 |
audio: Fix handling of A2DP open indication
When accepting the open indication all config callbacks should be
notified that open completed. |
Luiz Augusto von Dentz |
13 years ago |
1 file, +11, -0 |
| c5bf6d66 |
audio: Fix handling of A2DP suspend indication
When accepting the suspend indication all callbacks should be notified
that suspend completed.
In addition to this fix not using avdtp_start return in indication
callback as well as in the confirmation. |
Luiz Augusto von Dentz |
13 years ago |
1 file, +33, -5 |
| ee89674c |
audio: Add handling of AVDTP command collision
Check collision for AVDTP Open, Close, Start, Suspend and Abort commands
and if they collided remove the pending request if SEP has accepted the
indication. |
Luiz Augusto von Dentz |
13 years ago |
1 file, +80, -0 |
| e395d8dd |
Fix ALSA plugin having full soname
This is a plugin, so spell -avoid-version correctly so
it doesn't have a full soname. |
Bastien Nocera |
13 years ago |
1 file, +1, -1 |
| ee56337e |
Update tests to be compatible with gi and python3
This patch makes the python tests source-compatible with python 3, while
leaving the interpreter at python 2 for now.
The tradeoff is that this source is no longer compatible with python
versions < 2.6, and requires gobject-introspection for the glib-based
tests. |
Steve Langasek |
13 years ago |
24 files, +342, -294 |
| ebb2896b |
Remove test/apitest
This script is back from BlueZ 3.x times and unusable now |
Johan Hedberg |
13 years ago |
2 files, +1, -449 |
| a59814ad |
Add org.bluez.service to the tarball
Seeing as we want to install it. |
Bastien Nocera |
13 years ago |
1 file, +1, -1 |
| 094acfdd |
Release 4.100 |
Marcel Holtmann |
13 years ago |
2 files, +27, -1 |
| 933d78a9 |
Update library version |
Marcel Holtmann |
13 years ago |
1 file, +1, -1 |
| cf75b3f8 |
mgmtops: Clear pending_cod_change when powering off |
Johan Hedberg |
13 years ago |
1 file, +1, -0 |
| 00372cde |
mgmtops: Move error message after command status handlers calls
Don't print error message for actions that are expected or recoverable
i.e. add_uuid returning EBUSY status. |
Szymon Janc |
13 years ago |
1 file, +7, -5 |
| 4da87e3e |
mgmtops: Handle EBUSY status from MGMT_OP_ADD_UUID commands
MGMT_OP_ADD_UUID may fail with EBUSY due to ongoing CoD update. In case
of EBUSY error wait for Class Of Device changed event before adding
more UUIDs. |
Szymon Janc |
13 years ago |
1 file, +50, -9 |
| 07b6738f |
audio: Separate profile specific fields from media_transport |
Luiz Augusto von Dentz |
13 years ago |
1 file, +82, -34 |
| 18ffa5c1 |
gattrib: Make event callback identifiers globally unique
The attrib server code relies on these id's to be unique globally and
not just per GAttrib instance. As an easy fix make them global by adding
a static guint to g_attrib_register. |
Johan Hedberg |
13 years ago |
1 file, +2, -2 |
| 1da4a660 |
btmgmt: Fix use after free in mgmt_check_pending
Break loop when pending command was found and callback called. |
Szymon Janc |
13 years ago |
1 file, +1, -0 |
| 283b4b7d |
gattrib: Fix protection of GAttrib while calling callbacks
The callbacks could result with the reference count dropping to 0 and
the object being freed. This patch fixes the issue by adding one extra
reference for the duration of the timeout function. |
Johan Hedberg |
13 years ago |
1 file, +4, -0 |
| c03184ef |
mgmtops: Fix handling of cmd_status(0) |
Johan Hedberg |
13 years ago |
1 file, +6, -0 |
| 7dc92fb8 |
Remove workaround in gatt_connect
This workaround is not necessary anymore since setsockopt is now
checking for minimum MTU. |
Andre Guedes |
13 years ago |
1 file, +0, -13 |
| 71254107 |
Fix gatt_connect for BR/EDR
Use BT_IO_OPT_IMTU instead of BT_IO_OPT_OMTU in bt_io_connect.
We cannot control omtu value since it is negotiated during L2CAP
configuration phase. |
Andre Guedes |
13 years ago |
1 file, +1, -1 |
| ece3234d |
Remove omtu parameter from LE bt_io_connect calls
There is no need to set the omtu of L2CAP ATT fixed channel. We
use the default value. |
Andre Guedes |
13 years ago |
1 file, +0, -1 |
| 5d1f47f1 |
attrib-server: Update GAttrib buffer after Exchange MTU
We should update the GAttrib buffer length after exchanging ATT_MTU. |
Andre Guedes |
13 years ago |
1 file, +1, -0 |
| 5e446024 |
attrib-server: Fix mtu_exchange
If the client requests an ATT_MTU less than the minimum ATT_MTU,
the server should send an Error Response message with Request Not
Supported code.
According to GATT spec, the server shall respond to Exchange MTU
Requests messages with an Exchange MTU Response with the Server
Rx MTU parameter set to the maximum MTU that this server can
receive. Thus, we should get L2CAP imtu value in order to properly
send the Exchange MTU Response message.
Additionally, we should not change the L2CAP ATT fixed channel MTU.
bt_io_set call will always fail since we are not supposed to change
L2CAP MTU after connection is established. |
Andre Guedes |
13 years ago |
1 file, +16, -7 |
| e3598c49 |
attrib-server: Fix gatt_channel MTU value
In attrib_channel_attach, channel->mtu should be initialized according
to ATT_MTU value.
Over BR/EDR, ATT_MTU should be set to the L2CAP imtu negotiated during
L2CAP configuration phase. Over LE, ATT_MTU should be 23 octets. |
Andre Guedes |
13 years ago |
1 file, +7, -7 |
| ea271806 |
gattrib: Fix g_attrib_set_mtu
23 octets is the default (and minimum) ATT_MTU value. If someone tries
to set ATT_MTU less than 23 octets g_attrib_set_mtu should fail (return
FALSE). Additionally, there is no constraint regarding the maximum value
of ATT_MTU, so we should not check for it.
Also, we should not change the L2CAP ATT fixed channel MTU. bt_io_set
call will always fail since we are not supposed to change L2CAP MTU
after connection is established. |
Andre Guedes |
13 years ago |
1 file, +0, -8 |
| 185c7ce8 |
gattrib: Fix GAttrib buffer allocation
GAttrib buffer should be allocated according to ATT_MTU value. Over
BR/EDR, ATT_MTU should be set to the L2CAP imtu negotiated during
L2CAP configuration phase. Over LE, ATT_MTU should be 23 octets. |
Andre Guedes |
13 years ago |
1 file, +20, -12 |
| 33523ab3 |
gattrib: Fix a premature response timer removal
A timer is set when a response is expected. The timer is removed when
data is received, regardless of whether or not the data is a response.
As a result, the timer may be cleared even though a response was not
received and there would be no way to detect a command timeout.
Fix this by clearing the timer only after verifying a response was
received. |
Ido Yariv |
13 years ago |
1 file, +5, -5 |
| fec848db |
att: Start application error codes from the beginning of the range
Application error codes start off with 0x80 onwards. In CSSv2 the range
0xE0-0xFF is reserved for "Common Profile Error Codes" so we should
ensure not to mix that with our own definitions. |
Johan Hedberg |
13 years ago |
1 file, +3, -3 |
| 5232f762 |
attrib: Fix always requiring high security
If we get an insufficient encryption error for ATT it doesn't
necessarily mean that we need to have high secirity. If we have an
unauthenticated key and are unencrypted it could also mean that we just
need to encrypt the link (medium security). Always requiring high
security would make the connection fail for unauthenticated keys. |
Johan Hedberg |
13 years ago |
1 file, +9, -1 |
| 3b3e8867 |
gattrib: Fix command timeout handling
This patch fixes command timeout handling. Previously attrib_destroy was
explicitly called which ignored any reference holders. This patch fixes
the issue by first passing errors to command callbacks and after that
marking the GAttrib object as stale so no further operations can be
done. |
Johan Hedberg |
13 years ago |
3 files, +34, -1 |
| 6c5efac6 |
simple-agent: Use a higher (60s) timeout for pairing |
Johan Hedberg |
13 years ago |
1 file, +1, -0 |
| 8446f70d |
Makefile.tools: Add missing test_mpris_player_SOURCES |
Johan Hedberg |
13 years ago |
1 file, +1, -0 |
| 1801a4d4 |
mgmtops: Fix endianness conversion
Flags field has 32-bit size. |
Jefferson Delfes |
13 years ago |
1 file, +1, -1 |
| 29405cc5 |
core: Remove pending connection attempt
Fix connection attempt being triggered after device is removed. Pending
ATT re-connection attempt should be cancelled when all ATTIO callbacks
are unregistered. |
Claudio Takahasi |
13 years ago |
1 file, +5, -0 |
| 6f338a8f |
core: Fix connected callback not being called
ATTIO callbacks are not being added in the list when there is a
connection attempt pending. |
Claudio Takahasi |
13 years ago |
1 file, +9, -12 |
| cc4838e6 |
core: Fix repeated connection attempt
This patch fixes possible repeated connections attempt to the same
remote device when there is an attempt pending, since device drivers
can request independently ATT connections. |
Claudio Takahasi |
13 years ago |
1 file, +1, -1 |
| 9bfc21c2 |
core: Move UUID string definition to lib |
Daniel Wagner |
13 years ago |
4 files, +4, -3 |
| 5c3c3fef |
pnat: Move UUID string definition to lib |
Daniel Wagner |
13 years ago |
2 files, +4, -2 |
| 3af0412b |
input: Move UUID string definition to lib |
Daniel Wagner |
13 years ago |
5 files, +6, -3 |
| cce53f0c |
health: Move UUID string definition to lib |
Daniel Wagner |
13 years ago |
4 files, +6, -4 |
| 9e80406e |
serial: Move UUID string definition to lib |
Daniel Wagner |
13 years ago |
2 files, +3, -2 |
| 0a87ad90 |
thermometer: Move UUID string definition to lib |
Daniel Wagner |
13 years ago |
3 files, +6, -7 |