In case CIS failed to be established, leave the ASE in enabling state
waiting for peer to retry.
Signed-off-by: Mariusz Skamra <mariusz.skamra@codecoup.pl>
When Receiver Stop Ready operation has completed, the Unicast Server
may request to terminate a CIS and set source ASE state to Qos
configured state immediately. But after CIS disconnected completed,
it will transition source ASE state from Qos Configured to Disabling
state. It's invalid and will cause assert.
we should add Streaming and Enabling state check when CIS disconnected.
Signed-off-by: Hang Fan <fanhang@xiaomi.com>
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>
This fixes invalid error code that was returned in case remote requested
Disable operation with invalid ASE ID.
Fixes: BAP/USR/SPE/BI-01-C
Signed-off-by: Mariusz Skamra <mariusz.skamra@codecoup.pl>
ASE in QoS state already have audio ISO object.
This will just skip the audio ISO allocation and binding in such case
if CIG and CIS parameters did not change.
Signed-off-by: Mariusz Skamra <mariusz.skamra@codecoup.pl>
Unbind ISO when goint to idle state if there's still reference kept.
It may happen that we get CIS disconnection being in non-releasing
state, so we have to handle that case. This fixes regression that was
introduced in 3fa456905d.
Signed-off-by: Mariusz Skamra <mariusz.skamra@codecoup.pl>
This adds common bt_audio_iso pool that will be used across all the
roles/profiles. The pool range is dependent on the CONFIG_BT_ISO_MAX_CHAN
which is the maximum number of ISO connections the host can maintain.
Signed-off-by: Mariusz Skamra <mariusz.skamra@codecoup.pl>
Move the control point work to the EP. The aim is to cleanup the code so
that it there's no EP for ASE that means ASE is in idle state.
Signed-off-by: Mariusz Skamra <mariusz.skamra@codecoup.pl>
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>
If 2 ASEs shared the same CIS and the first ASE had the CIS
connected before the second ASE was QoS configured, then
the CIS was missing either the TX or RX pointer, causing
it to be considered unidirectional by the ISO layer.
This commit fixes this issue by configuring the (static)
pointers at an earlier time, so that the RX and TX QoS
pointers are always valid.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
This changes the API to use proper naming convention, as the code has
been moved out of capabilities.
Signed-off-by: Mariusz Skamra <mariusz.skamra@codecoup.pl>
Macro that was defined in the `common/log.h` has been moved to the newly
created `common/assert.h` file.
Files that were using those macro has been updated consequently. Also, the
include of `common/log.h` has been removed of files that were actually not
using any logging methods.
Some `#include "common/log.h"` have been removed. Those were required
before when including `hal/debug.h`. With this change, `hal/debug.h` no
longer has this requirement, because the replacement, `common/assert.h`
includes what it uses.
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>
The struct bt_ascs holds the conn reference already,
thus it's not needed to store the device address.
Signed-off-by: Mariusz Skamra <mariusz.skamra@codecoup.pl>
There should be functional equivalence between these two forms. And the
'_eq'-form is more readable.
Signed-off-by: Aleksander Wasaznik <aleksander.wasaznik@nordicsemi.no>
Adds check to verify that the audio stream is actually
in the streaming state before handling the
audio data. As per the BAP spec, a stream that
is not in the streaming state shall not send
any data, and this check is to handle any
remote devices that are not working correctly.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
The metadataa_updated callback has existed for a long
time, but was never actually used. The metadata is set
or updated in the enabling and streaming state.
Added checks to see if the state actually changes when going
into these states, and if not, then it means that the
metadata was updated.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
This function exposes list pointer, so that it allows the user to modify
the internal list. This adds bt_audio_foreach_capability iterator finction
that can be used instead.
Signed-off-by: Mariusz Skamra <mariusz.skamra@codecoup.pl>
Use the stream->conn instead of the iso->acl, as the iso->acl
may already have been free'd by the ISO layer when the
disconnected callback is called.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
This patch fixes invalid ASE state transition,
ASE now waits for Receiver Start Ready before
going into Streaming state.
Signed-off-by: Szymon Czapracki <szymon.czapracki@codecoup.pl>
If the CIS was connected before the sink endpoint was
in the enabling state, it never autonousmly went into
the streaming state.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
bt_ascs is per connection, so the control point attribute pointer
should not be part of this structure.
Signed-off-by: Mariusz Skamra <mariusz.skamra@codecoup.pl>
The codec->path_id supplied to the capabilities
were not being used for the endpoint codec.
Important note: This only works if there is a single
capability per codec ID per direction. This limitation
is also an existing limitation in the current
implementation of capabilities.c.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Update the existing ASE EP state checks to be more correct
and aligned with the ASCS specification.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Remove the support for caching ASEs on release.
This is mainly due to the 2-stage design in Zephyr: We
have the bt_audio_stream and the bt_audio_ep, where the
the latter represents an ASE in most cases.
However, once the endpoint goes into the codec configured
state, then a stream must be bound to it. This would be fine
(although a stream is technically not established at this point),
except that a stream is also, for unicast, specified for a
specific ACL. Once the ACL disconnects, the stream technically
becomes invalid, and if the stream is invalid, so is the endpoint.
Based on that, there's no reason (or even option) to keep the
endpoint in a codec configured state, as it would not
have a valid stream associated with it.
If we want to support an ASE going into the codec
configured state on ASE release (or ACL disconnect),
the relationsship between ASEs and streams needs to
be redefined.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
This fixes crash that happened during source ASE QoS by setting
the tx path that was left uninitialized.
Signed-off-by: Mariusz Skamra <mariusz.skamra@codecoup.pl>
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>
Add support for setting a vendor specific data path using
the BAP API, instead of only support the HCI data path.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Multiple places we assumed that we could simply just do
sink_stream->iso == chan without checking if sink_stream
was NULL.
This commit fixes that by adding proper NULL checks for that,
as well as additional checks before defaulting to the
source_stream.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
This fixes invalid check, which is redundant in fact as the length
is already checked in ascs_parse_metadata function.
The check be test 'greater than' only, as 'equal' condition is valid.
Moreover this fixes return value from ascs_verify_metadata function.
Signed-off-by: Mariusz Skamra <mariusz.skamra@codecoup.pl>
As per ASCS_v1.0
If a client requests a Config QoS operation for an ASE that would
result in more than one (Sink/Source) ASE having identical CIG_ID
and CIS_ID parameter values for that client, the server shall not
accept the Config QoS operation for that ASE.
The server shall send a notification of the ASE Control Point
characteristic to the client, the server shall set the Response_Code
value for that ASE to 0x09 (Invalid Parameter Value), and the server
shall set the Reason value for that ASE to 0x0A
(Invalid_ASE_CIS_Mapping).
Fixes: ASCS/SR/SPE/BI-{15,16}-C
Signed-off-by: Mariusz Skamra <mariusz.skamra@codecoup.pl>
The LE Audio related profiles define specific security requirements that
client have to met to gain access to related characteristics. This
introduces internal wrapper API for GATT characteristics and CCC
descriptors to ensure the security.
Signed-off-by: Mariusz Skamra <mariusz.skamra@codecoup.pl>
This makes the server endpoint to store the ASE pointer instead of
handle. This way, there will be no need to "mock" attribute when
notification is sent.
Signed-off-by: Mariusz Skamra <mariusz.skamra@codecoup.pl>
This fixes crash that happened when client initiated Release operation
on ASE in Idle state. In such case the operation shall be rejected with
Invalid ASE State Machine Response_Code.
Fixes: ASCS/SR/SPE/BI-10-C
Signed-off-by: Mariusz Skamra <mariusz.skamra@codecoup.pl>
This fixes state transition to QoS Configured state triggered by
client-initiated Receiver Stop Ready operation on ASE that has no ISO
attached.
Fixes: ASCS/SR/ACP/BV-12-C
Signed-off-by: Mariusz Skamra <mariusz.skamra@codecoup.pl>
This fixes sending invalid ATT Request Not Supported error code as a
response to Control Point write request.
ATT Request Not Supported error code is used to inform that
"ATT Server does not support the request received from the client",
so it should not be sent in this case.
If the requested opcode is unsupported, ASCS mandates to send ATT
Notification from Control Point with Unsupported opcode Response_Code
only. If the request received from the client is properly formatted,
ATT Write Response can be sent to complete the ATT transaction.
Fixes: ASCS/SR/SPE/BI-01-C
Signed-off-by: Mariusz Skamra <mariusz.skamra@codecoup.pl>
BT_ERR() should really be reserved for local errors only,
thus BT_WARN() is the right choice whenever a remote peer
sends something invalid that we cleanly handle.
Signed-off-by: Mariusz Skamra <mariusz.skamra@codecoup.pl>
As per ASCS_v1.0
"A client-initiated ASE Control operation shall be defined as an
invalid length operation if the Number_of_ASEs parameter value is
less than 1".
Fixes: ASCS/SR/SPE/BI-{02,03,04,05,06}-C
Signed-off-by: Mariusz Skamra <mariusz.skamra@codecoup.pl>
This fixes issue where the ASE state stuck in Releasing state instead of
going to Idle/Codec Configured as test requires.
Looks like there was a race condition possible where the ASE
notification work was run before the stream was detached, so there was
no transition to idle state triggered.
Fixes: ASCS/SR/ACP/BV-24-C
Signed-off-by: Mariusz Skamra <mariusz.skamra@codecoup.pl>
The server should not perform Release operation on ASE in Releasing
state. This fixes crash that might happen in ASCS when ACL gets
disconnected while ASE is in Releasing state so that server tries to
dereference NULL stream.
Signed-off-by: Mariusz Skamra <mariusz.skamra@codecoup.pl>
