This modifies the GATT client characteristic WriteValue D-Bus method to
not check that the characteristic supports the requested type of
write when the "type" option is set to "command".
Before this change, if the "type" option was used and it was set to
"reliable" or "request", then BlueZ would attempt the write even if the
characteristic does not support that write type. On the other hand, if
"type" was set to "command" or was not specified, the method would
return a org.bluez.Error.NotSupported error without attempting to write.
After this change, the WriteValue method will consistently always
ignore the characteristic properties and attempt to write when the
"type" option is used instead of having a different behavior for the
"command" type.
This patch proposes an Advertisement Monitoring API for an application
to register a job of monitoring ADV reports with content filter and
RSSI thresholds.
This change adds a new property to indicate the support for concurrent
roles which means that the controller has reported the appropriate
LE_Supported_States (hdev->le_states) and that the controller's driver
has reported correctly handling the various reported states.
Without this patch ConnectProfile DBus method does not return failure if
profile connection failed and some other profile was already connected.
This is not correct behavior as ConnectProfile DBus method should always
return error when specified profile failed to connect. This patch fixes
this it.
Some bluetooth headsets do not support connecting more then one bluetooth
profile (e.g. in parallel A2DP and HSP, or HSP and HFP) and issuing
connect() syscall for second profile returns just ECONNREFUSED.
Prior this patch bluetooth daemon for such situation reported following
message to log:
Unable to get connect data for Headset Voice gateway: getpeername: Transport endpoint is not connected (107)
Message is incorrect as connect() syscall failed, not getpeername(). This
patch fixes this problem and logs correct error message:
Headset Voice gateway failed connect to XX:XX:XX:XX:XX:XX: Connection refused (111)
Main problem was in ext_connect() function which called bt_io_get() for
retrieving remote address (BT_IO_OPT_DEST) and if it failed then original
error from connect() syscall was masked. Because it is not possible to
retrieve BT_IO_OPT_DEST for unconnected socket, original destination
address for error message is propagated via connect_add() function in btio.
This change removes the Provisioner Initiator from the python test
script, along with the menu commands for create, scan and add. This
makes it a cleaner Python3 example of a simple On/Off node, and a
fuller featured Provisioner/Config Client still exists in the tools
subdirectory.
To avoid a double hog free, need to add a ref
when adding the hog to the slist.
This bug has been reproduced with gamepad-8718
which was connecting/disconnecting frantically.
Fix also a typo in the method hog_attach_instance
Class is using only 3 bytes so make sure GCC is aware of that.
src/device.c:397:3: note: ‘sprintf’ output between 9 and 11 bytes into a destination of size 9
397 | sprintf(class, "0x%6.6x", device->class);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
cc1: all warnings being treated as errors
This one is a false positive but since we never use more than
UINPUT_MAX_NAME_SIZE bytes of name we can silence GCC by reducing
size of source string.
CC profiles/audio/bluetoothd-avctp.o
In function ‘uinput_create’,
inlined from ‘init_uinput’ at profiles/audio/avctp.c:1259:20:
profiles/audio/avctp.c:1188:3: error: ‘strncpy’ output may be truncated copying 79 bytes from a string of length 248 [-Werror=stringop-truncation]
1188 | strncpy(dev.name, name, UINPUT_MAX_NAME_SIZE);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
cc1: all warnings being treated as errors
status and state are used to hold various enum types depending on test
and callback passed. Define them as int to avoid warnings about enum
assignment from invalid type.
JoinComplete() dbus method calls are the only time that node tokens are
delivered to client Applications, so if the call fails for any reason
(including time-outs) the daemon has a way to clean-up the stale unused
node data.
This patch changes Import and CreateNetwork API to deliver tokens via
the JoinComplete method call. When application doesn't raise any error
during handling JoinComplete then it is assumed that the token has been
saved, otherwise when application replies with an error message then the
node is removed.
If Application is not be able to reliably store the token, the daemon
will end up with a uncontrollable node in its database.
Let's fix the issue by always delivering tokens using JoinComplete call,
and expecting a reply - if the application return an error, daemon will
get rid of the node.
Remote Audio Volume Control property in SDP record for HSP HS role
indicates if device supports volume control.
It is required for D-Bus agents which implements audio part of HSP
profile to know if remote device supports volume control or not.
With this change bluez exports status of this SDP property via firt bit
in Features entry in NewConnection() DBus callback method, like for HFP
profile.
This patch fixes invalid app_path on 'Join' method call - the daemon
tried to use the value of app_root API argument, while it should use
path discovered by scanning result of GetManagedObjects() call.
If adapter is discoverable (or LE-only) advertising was failing due
to too big adv data being set. Make sure there is enough space left
for flags if needed.
When daemon receives beacon with IV=n+1, IVU=False it will
start sending messages with new IV and set sequence to 0.
However if daemon receives another beacon with IV=n+1, IVU=True it
will go back to sending messages with old IV=n (IVU set to True).
Because sequence number has been reset those messages will be dropped
by replay protection and node will lose communication.
Once IV is updated daemon should not go back to using the old value.
This patch adds beacon rejection if IV has already been updated.
Unit test would signal failures if run on commandline, but not if
run as part of a unit test sweep. This change forces an exit(1) on
any detected fail, and is caught by test suite runner.
- Remove application_encrypt/application_decrypt
- Make mesh_crypto_privacy_counter private, add mesh_crypto_pecb
- Make *_nonce functions private and align their implementation to be
more consistent
- Refactor network_encrypt/network_decrypt to use *_nonce functions and
rename them to packet_encrypt/packet_decrypt
- Refactor packet_encode/packet_decode
