During overcommit, mesh_config_save is called in asynchronous mode to
avoid blocking Send() calls. This means that update of cfg->write_time
is scheduled via l_idle_oneshot, so if the next Send() gets scheduled
first, the code may see elapsed time of zero.
If this happens, then the overcommit logic was already executed and the
overcommit is pending, so we can just return.
The folllowing functionsa are not being used anywhere in the code and
have been removed:
mesh_net_flush()
mesh_net_prov_caps_get()
mesh_net_priv_key_get()
mesh_net_priv_key_set()
mesh_net_prov_rand()
mesh_net_prov_uni()
mesh_net_id_uuid_set()
mesh_net_test_addr()
mesh_net_test_mode()
mesh_net_uni_range_set()
mesh_net_uni_range_get()
mesh_net_set_window_accuracy()
This removes callback for sending every single access layer packet
since the callback does nothing but print debug satement. This
functionality is mature enough not to require such heavy debugging.
This patch fixes a double free condition in load_remote_sep. Value is
freed, then the inner loop is broken, but the rest of the outer loop
will attempt to free value again.
In a2dp_reconfig, setup is ref-counted by cb_data. However, in the fail
label setup is unref-ed but cb_data is not cleared. This may cause
double unref in the future if cb_data gets executed. Instead, we should
do setup_cb_free to clear cb_data and unref setup.
If command MGMT_OP_START_DISCOVERY fails we reply with btd_error_busy
but we don't remove the client from the list which would prevent the
client to start it once again.
Player Application settings is not mandatory for category 1 so instead
of always listing the settings the code now checks if
AVRCP_FEATURE_PLAYER_SETTINGS is enabled.
When we are in non-interactive mode (data.mode == 1), we should not
terminate the mainloop when stdin is disconnected.
For example, in bash, the following command would terminate immediately
without any output.
: | btmgmt info
When an instance wants to force being discoverable the code shall
actually check if the adapter is in general disverable mode already and
if not set BR/EDR as not supported so that devices scanning don't
assume BR/EDR PHY is connectable when in fact it isn't.
The slave will be the first to respond with CIS Estabished but only the
master has the CIG params set so this switch to master so parameters are
proper.
This ensures that every time l_dbus_message_ref() is used to preserve
a message for a pending method reply, there is a matching call to
l_dbus_message_unref().
When a new node is created as a result of successful completion
of either Join() or Create() or Import() methods and has been
confirmed via successful token delivery to the application,
clean up node's D-Bus resources (application path, element paths, etc)
that have been gathered during the initial GetMAnagedObjects() call.
Also, remove the agent instance associaed with the new node.
These resources will be re-populated after the Attach() call
verifies the node's integrity.
LE Setup ISO Data Path can be used even before the ACL handle exists
when master since Set CIG Parameters will respond with ISO handle it
can then be used to setup the data path in advance of ACL connection.
Adding the Tx Power Level is no longer done via IncludeTxPower, but via
the tx-power value in the Includes array. The previous code did not
throw an error, but neither led to the insertion of the value. As a
result of this change, include_tx_power now adds the Tx Power Level
again.
This fixes a double-free error when destroying the NVM storage of a mesh node.
Cleanly handle two distinct scenarios:
1. When the node is being deleted at runtime. This causes release of
both dynamic memory and NVM storage.
2. During shutdown, we release dynamic memory only.
Running bluetoothd from the build directory I noticed that as soon as
a bluetooth audio device was connect, bluetoothd would be disconnected
from dbus and print
Disconnected from D-Bus. Exiting.
Luiz Augusto von Dentz suggested trying with SELinux disabled and that
solved the problem.
This patch just documents how to disable SELinux before running
bluetoothd from the build directory. I would love to say more about
why that is needed, but could not find anything on the audit log.
Set the the correct vendor and product ids for all UHID/HoG
devices when they are unknown at HoG creation time.
Before this change, when connecting a BT device with multiple HoG
services for the first time, only the first HoG instance's vendor,
product and version fields would be set by the DIS callback. This meant
that all HoG instances except the first would be left with unset values
and their UHID devices would then be created with '0000:0000' as their
vendor:product ids.
hid2hci is no longer a common use. This patch changes the default to not
build and install hid2hci and provide a configure flag --enable-hid2hci
to explicitly enable it.
This fixes the crash that occurs when a controller used by bluetooth-meshd
is removed and then added back again.
Also, correctly restart scanning when the controller is re-enabled.
Backtrace:
0x00005618e754d040 in ?? ()
0x00005618e6e12d9a in io_ready_callback () at mesh/mesh.c:174
0x00005618e6e3d2c8 in l_queue_foreach () at ell/queue.c:441
0x00005618e6e37927 in request_complete () at src/shared/mgmt.c:261
This removes a redundant call to save node configuration in
add_local_node(): there is a number of unconditional calls to
mesh_config_write_<some_node_Setting>() within this function
which would result in the node configuration being saved implicitly.
This allows to update settings of the following composition fields
when an existing node (application) attaches to the daemon:
Company ID (CID), Product ID (PID), Versioin ID (VID),
CRPL (replay protection depth)
This modifies miscellaneous utility functions in mesh-config-json.c:
when writing a new value to a node configuration file, delete
the existing field containing an old value first.
