Adds, removes and modifies includes in all LE audio
files.
Fixes any found spelling mistakes as well.
Fixes a few places where incorrect types were used.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
The LE Audio implementations do not really support bonding yet,
and removing subs on disconnect is the most effective (and correct)
way of ensuring that we do not subscribe more than once when we
re-discover after reconnection.
The broadcast assistant and the media control client does not
support multiple connections as of this commit, so they needed
special treatment. In the case that we do discovery on multiple
ACL connections, it is important that the existing subscriptions
are removed correctly by calling bt_gatt_unsubscribe.
In order to implement this change properly on some of the clients,
thet had no proper connection references or support
for clearing the data on disconnects, they had to be updated
as well.
The csip_notify.sh test has been disabled, as that expected a
notification in the client, but since this commit removes that
(until bonding is properly supported in the clients), then the
test will fail.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Several places the LE Audio clients called bt_gatt_subscribe without
checking the return value, which could cause some issues in the worst
case, and in the best case, cause some unexpected behavior.
Some implementations had a bit more updating to handle the new
behavior.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
This change makes the optional procedures in the Media Control Client
optional and configurable through Kconfig.
Signed-off-by: Ping Wang <pinw@demant.com>
This implements the deferring of characteristic value notifications to
be sent from system work queue thread context. Notifications issued from
Bluetooth Rx thread might not be sent if there are no L2CAP Tx buffers
available, as the operation is non-blocking to prevent deadlock while
waiting for free buffers.
The same operation issued from other thread context is blocking,
meaning that the thread waits until L2CAP Tx buffers become available.
Thus it's guaranteed the notifications will be sent.
With this patch, the control point operations become blocking until the
control point response is sent. Meaning there might be only one
pending operation waiting for completition. This might be further
improved by queuing the operations if needed.
Fixes: #57444
Signed-off-by: Mariusz Skamra <mariusz.skamra@codecoup.pl>
The discovery parameters may contain values from previous
discovery calls. The simply and future proof solution to
this is to simply memset it before using it.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
If the subscribed callback is called with value_handle = 0,
then it should not be treated as a subscription complete.
This will happen when bt_mcc_discover_mcs is called, as
reset_mcs_inst unsubscribed to all characteristics before
discovering again.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Add macro BT_MCS_VALID_OBJ_ID to verify the OTS Object
ID before attempting any OTS operations.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
With BT_GATT_SUBSCRIBE_FLAG_NO_RESUB the MCC subscriptions would
be resent to the server on reconnection. Since this may trigger an
error, the error will trigger a call to the MCC discover complete
callback.
For MCP We can assume that the server stores the subscriptions if
we are bonded, and thus we can use BT_GATT_SUBSCRIBE_FLAG_NO_RESUB.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
mcc.c used 0x%016x, but for some platforms we need to
be more explicit and use 0x%016llx to avoid compile
warnings.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
When MCC disconnected, it would memset the entire instance.
This causes issues, because the subscription parameters
would also be memset, causing the `notify` callback to
be set to NULL. If this was done as part of a disconnect,
the GATT layer would attempt to call the notify callback
as part of cleaning up the ACL, but if the callback was
set to NULL (from the memset) this caused a fatal error.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
The `bluetooth/common/log.h` and `bluetooth/common/log.c` files have been
removed. Files that were using them have been updated to use
`zephyr/logging/log.h` instead.
Those replacement have been done consequently:
- `/BT_DBG/LOG_DBG/`
- `/BT_ERR/LOG_ERR/`
- `/BT_WARN/LOG_WRN/`
- `/BT_INFO/LOG_INF/`
- `/BT_HEXDUMP_DBG/LOG_HEXDUMP_DBG/`
- `/BT_DBG_OBJ_ID/LOG_DBG_OBJ_ID/`
Also, some files were relying on the `common/log.h` include to include
`zephyr/bluetooth/hci.h`, in those cases the include of `hci.h` has
been added.
For files that were including `common/log.h` but not using any logs,
the include has been removed and not replaced.
Signed-off-by: Théo Battrel <theo.battrel@nordicsemi.no>
mcc.c does not include `conn_internal.h` so it does not
know the `bt_conn` struct, causing logging issues when
trying to log a `bt_conn` pointer.
Fixed by casting the `conn` pointers to `(void *)`.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
One of the notification callback functions did not properly
get the `read_params` pointer, causing an invalid result of
CONTAINER_OF as it would use a NULL pointer.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Removes the global cur_mcs_inst and replace it
with lookup functionality.
This fixes an issue with bt_mcc_discover_mcs not being able
to be called more than once, as well
as streamline the implementation significantly.
This is also the first step towards not only supporting
multipe MCS instances for a single device, but also a
step towards handling multiple devices.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
It was possible that on re-connect, the MCC implementation
would receive notifications when cur_mcs_inst was NULL.
This is partially caused by a bug in gatt.c, but can
and should also be handled in MCC.
This commit ensures that we do not compare handles
on notifications with a NULL instance.
This commit also ensures that subscribe_mcs_char_func
is not called outside the context of the discovery
procedure by setting the subscribe callback
to NULL.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Functions related to string manipulation that were defined in
`common/log.h` has been moved to the `common/bt_str.h` file and their
implementation in `common/bt_str.c`.
Files that were using those functions has been updated consequently.
Signed-off-by: Théo Battrel <theo.battrel@nordicsemi.no>
Replace unnecessary `BT_HEXDUMP_DBG` with `LOG_HEXDUMP_DBG`.
`BT_HEXDUMP_DBG` did an extra cast to `const uint8_t *` on its first
argument. This was probably the reason it was introduced. But the
current definition of `LOG_HEXDUMP_DBG` does this cast for us.
Signed-off-by: Théo Battrel <theo.battrel@nordicsemi.no>
As of today <zephyr/zephyr.h> is 100% equivalent to <zephyr/kernel.h>.
This patch proposes to then include <zephyr/kernel.h> instead of
<zephyr/zephyr.h> since it is more clear that you are including the
Kernel APIs and (probably) nothing else. <zephyr/zephyr.h> sounds like a
catch-all header that may be confusing. Most applications need to
include a bunch of other things to compile, e.g. driver headers or
subsystem headers like BT, logging, etc.
The idea of a catch-all header in Zephyr is probably not feasible
anyway. Reason is that Zephyr is not a library, like it could be for
example `libpython`. Zephyr provides many utilities nowadays: a kernel,
drivers, subsystems, etc and things will likely grow. A catch-all header
would be massive, difficult to keep up-to-date. It is also likely that
an application will only build a small subset. Note that subsystem-level
headers may use a catch-all approach to make things easier, though.
NOTE: This patch is **NOT** removing the header, just removing its usage
in-tree. I'd advocate for its deprecation (add a #warning on it), but I
understand many people will have concerns.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
MCC need to check characteristic properties before subscribe as not all
MCS characteristics support notify.
Signed-off-by: Hang Fan <fanhang@xiaomi.com>
Use the same function at discovery complete, whether discovery failed
or succeeded.
Rename the discovery_failure() to discovery_complete(), and use that
also when discovery completes successfully.
Signed-off-by: Asbjørn Sæbø <asbjorn.sabo@nordicsemi.no>
The media control client used to subscribe to GMCS characteristics as
the characteristics were discovered, in effect queuing ATT requests.
Since then, the ATT MTU exchange has become automatic, leading to
larger ATT MTUs and more discovered characteristics per ATT response,
which again lead to the client enqueuing more subscriptions and
running out of buffers, with discovery failing as a result.
This commit moves the subscription to GMCS characteristics to a
separate sequence, so that it happens serially after discovery, rather
than in parallel with the discovery.
- Remove separate discovery parameters for the characteristics - they
are no longer needed.
- Add a subscription callback and a couple of subscription functions
The subscription for characteristics for the included OTS has not been
modified. Only two of these are subscribed to, and doing that is
currently not a problem. And I need to better understand the
interaction with OTS and how these subscriptipons are used.
Signed-off-by: Asbjørn Sæbø <asbjorn.sabo@nordicsemi.no>
Refactor subscription to get chaining into callback
Move the start of discovery of included services to a separate
function, in preparation for rewriting discovery.
Signed-off-by: Asbjørn Sæbø <asbjorn.sabo@nordicsemi.no>
Logging v1 has been removed and log_strdup wrapper function is no
longer needed. Removing the function and its use in the tree.
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
In order to bring consistency in-tree, migrate all subsystems code to
the new prefix <zephyr/...>. Note that the conversion has been scripted,
refer to zephyrproject-rtos#45388 for more details.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
Changes the media control client API to pass structs by reference
instead of by value.
Also change internals of test to pass struct by reference.
Signed-off-by: Asbjørn Sæbø <asbjorn.sabo@nordicsemi.no>
Rename the data and metadata callbacks to _read(), for consistency
with the corresponding function calls.
Signed-off-by: Asbjørn Sæbø <asbjorn.sabo@nordicsemi.no>
Move the instance as the first parameter of the client function calls
and callbacks, for consistency with the server implementation.
Signed-off-by: Asbjørn Sæbø <asbjorn.sabo@nordicsemi.no>
Merge the metadata types for client and server.
This is part of merging the ots_client and ots header files.
Signed-off-by: Asbjørn Sæbø <asbjorn.sabo@nordicsemi.no>
Move the metadata request bitfields to the common header files.
This is part of merging the ots_client and ots header files.
Signed-off-by: Asbjørn Sæbø <asbjorn.sabo@nordicsemi.no>
This commit adds the Media Control Client from the topic-le-audio
branch. This is a part of the upmerge of the le-audio media control
files.
This client has been developed and maintained over a couple of years
now, and is mature. During the development it has passed both
IOP-testing and PTS qualification testing.
The commit is a pure copy of the files and content in the
topic-le-audio branch, with the following exceptions:
- files are in bluetooth/audio instead of bluetooth/host/audio, with
some include paths updated as a consequence
- as a consequence, CMake files and Kconfig files updates are done in
other locations
- a macro for debug output of Object ID values has been (temporarily)
added to mcc.h, to avoid a dependency
- a blank line added after a declaration and an overlong line split,
to pass check_compliance
- copyrights have been updated
Signed-off-by: Asbjørn Sæbø <asbjorn.sabo@nordicsemi.no>
