Rather than the rings, which weren't shared between userspace and kernel
space in Zephyr like they are in Linux with io_uring, use atomic mpsc
queues for submission and completion queues.
Most importantly this removes a potential head of line blocker in the
submission queue as the sqe would be held until a task is completed.
As additional bonuses this avoids some additional locks and restrictions
about what can be submitted and where. It also removes the need for
two executors as all chains/transactions are done concurrently.
Lastly this opens up the possibility for a common pool of sqe's to
allocate from potentially saving lots of memory.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
Fix use of uninitialized ticks_slot value in calculation of
cis_offset_min when creating CIS. Calculate the ticks_slot
value earlier when committing the CIG parameters.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Do not enqueue node rx buffers for generating invalid ISO
data from Tx ISR while supplying the node rx buffer to
Radio. This causes node rx pool corruption and also MIC
failures in the central ISO LLL.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Fix peripheral ISO data ack mechanism to correctly update
the SN when PDUs fail to be received.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
When there is an active CIG for a new CIS create, then use
strict cis_offset_min/cis_offset_max so that the new CIS
is scheduled inside the active CIG event.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
'ps_wakeup_mode' is used to set the wake up interval
to either 'dtim or 'listen_interval'. By default the
sta wakes up every dtim interval.
With 'ps_wakeup_mode' set to 'listen_interval' sta
will wake up for every listen interval period configured.
Signed-off-by: Ajay Parida <ajay.parida@nordicsemi.no>
listen interval is the time periods the STAs may be in idle
without listening beacons.
By default STA wakes up for every DTIM period.
If listen interval based power save is used
STA uses configured listen interval period(default 10
beacon intervals) to wake up.
Signed-off-by: Ajay Parida <ajay.parida@nordicsemi.no>
Fix ztress_execute() out-of-bounds in case it is called with timer
and for maximum number of threads allowed (CONFIG_ZTRESS_MAX_THREADS).
Signed-off-by: Dmitrii Golovanov <dmitrii.golovanov@intel.com>
Add shell based on sip_svc subsystem to access the SDM mailbox in
EL3 layer using ASYNC sip_svc request.
Signed-off-by: Mahesh Rao <mahesh.rao@intel.com>
Introduce a new SiP SVC subsystem to provide ARM Silicon Provider based
supervisory call services. SiP SVC service provides the capability to
send SMC/HVC call from kernel running at EL1 to hypervisor/
secure monitor firmware running at EL2/EL3 and also added SiP SVC shell
commands to exercise the service.
Signed-off-by: Mahesh Rao <mahesh.rao@intel.com>
Prevents verifying the image has been uploaded successfully using
the hash provided at the start of the upload if the hash is not
the correct length of an SHA256 hash.
Signed-off-by: Jamie McCrae <jamie.mccrae@nordicsemi.no>
Cast `dhkey` to `void*` to avoid a warning from the logging subsystem:
```
<wrn> cbprintf_package: (unsigned) char * used for %p argument.
It's recommended to cast it to void * because it may cause misbehavior
in certain configurations
```
Signed-off-by: Herman Berget <herman.berget@nordicsemi.no>
This allows to move flash operations from the context that triggered
IVI update to the mesh settings work item that runs on the system
workqueue. After this change, the outdated RPL entries will be removed
in the setting work item triggered by store_pending function in
mesh/settings.c. This is required to for a case where the mesh settings
work item is running on a separate thread instead of the system
workqueue to unblock the system workqueue eventually.
Signed-off-by: Pavel Vasilyev <pavel.vasilyev@nordicsemi.no>
Bitsets across byte boundaries does not work properly for big-endian;
Converted to use two uint8's instead
Signed-off-by: Troels Nilsson <trnn@demant.com>
As we are using a generic validation function for limits, due to data
type mismatch the check for TWT interval overflow and fails.
Fix the data type and accordingly the maximum value in the help.
Signed-off-by: Krishna T <krishna.t@nordicsemi.no>
Add command to add a source by the broadcast ID. This will
initiate scanning for the broadcast ID, making it unnecessary
for the user to keep track of both the broadcast ID, the SID
and the address, and just focus on the (properly) known
broadcast ID.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
The cmd_bap_broadcast_assistant_broadcast_code used decimal, instead
of hex, for the broadcast code, making it a lot harder to use.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Always print/log the 24-bits (as hex) of the broadcast ID to better
use it with e.g. the shell.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
If starting the discovery procedure multiple times, the
subscribe function returns -EALREADY which needs to be handled.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Moved the implementation of the pa_decode_base to a public
function, and modified it to return better error codes.
The implementation was moved to be publically
accessible, rather than internal in the Broadcast Sink,
as e.g. a Broadcast Assistant need to be able to decode the
BASE too. This wil also allow an application to decode it
itself, if it so wanted to.
This also adds a log module for the audio.c file, as logging
would otherwise not be supported.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Verify that PAST would be possible before providing
BT_BAP_BASS_PA_REQ_SYNC_PAST as the PA sync value to the
Scan Delegator.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Make the `pa_sync` field of the
bt_bap_broadcast_assistant_add_src_param and
bt_bap_broadcast_assistant_mod_src_param to boolean
instead of uint8_t, as it represents a boolean value.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Fix possible NULL dereference at net_pkt_cursor_operate() and
resolve 'Dereference after null check' warnings from the Coverity
static code analysis.
Signed-off-by: Dmitrii Golovanov <dmitrii.golovanov@intel.com>
Added paired_ep to bt_bap_ep_get_info, which would allow to check
whether an endpoint is paired with another endpoint, this would
indicate that its coupled to a bidirectional iso.
Signed-off-by: Fredrik Danebjer <frdn@demant.com>
TFM is not the only provider for PSA API, which means
the PSA config shouldn't depend only on it.
Signed-off-by: Piotr Jasiński <piotr.jasinski@nordicsemi.no>
The receiver_ready field can only be set while in the enabling
state, and is a requirement to go into the streaming state.
If for any reason we go into a state that is not enabling
or streaming, then the receiver_ready field should be
cleared, as the circumstances are changed.
This also fixes an issue where if the CIS disconnects,
then we go from the streaming state to the
qos configured state, without clearing it, causing
incorrect behavior if we attempt to connect the streams
again.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Change fixes check for upgrade-only to ensure that the the new image
version is greater than version of the currently running image.
Signed-off-by: Marek Pieta <Marek.Pieta@nordicsemi.no>
Mesh models may have a data that needs to be stored persistently.
Currently, the models should call bt_mesh_model_data_store and the store
will happen in the calling context. Most likely that it will be called
in BT RX thread as this is the context from which model's opcodes
handlers are called. Thus, the thread will be blocked until the store is
finished.
Another issues is that some models may have states that changes
frequently. Triggering the store on every state change may wear out
flash. Therefore, the models need to implement some postpone mechanism
to reduce the flash wear out.
The mesh stack has already implemented the mechanism of deferred store
with its own settings. The models could use it instead of implementing
their own mechanism.
In combination with the mesh settings workqueue, the models can store
their data without blocking the stack work.
Signed-off-by: Pavel Vasilyev <pavel.vasilyev@nordicsemi.no>
Currently mesh settings are stored in the system workqueue context.
Most of other stack functionality, such that advertisements (incl
relay), loopback, transport sar, beacons transmission, etc. is also
processed in the system workqueue context. When a massive amount of
data needs to be stored and in particularly when page erase needs to
be triggered by GC of NVS subsystem to allocate flash pages, the
execution of the stack (and other functionality that uses the system
workqueue) will be blocked until storing is finished. For example,
right after the provisioning of a erased device, a node may not be
responsive for up to 400ms before it can continue sending messages.
The waiting time may increase if there is a GATT connection in the
mean time.
When write or erase operation is triggered, the flash driver waits for
Bluetooth controller to allocate a time needed to perform the operation.
During the whole operation, the context from which the operation was
triggered is put to sleep. This allows other threads to run until
Bluetooth controller finds the time for the flash driver. In other words,
every settings_save_one or settings_delete should be considered as
rescheduling points.
Considering this, Bluetooth mesh can use another thread to store its
settings, thus releasing the system workqueue for other tasks including
the operation of the stack itself.
The consistency of the data to be stored is guaranteed by the current
implementation where the data is copied to another struct before calling
settings_save_one. The pending flag of a particular module is dropped in
settings.c before starting to store the corresponding data. Thus, if
during the sleep the node receives a message that triggers a change in a
module which data is currently being stored, the pending flag will be
restored and the new change will be stored eventually.
Having this option enabled including with the partial erase, will make
the node more responsive in the described situations.
Signed-off-by: Pavel Vasilyev <pavel.vasilyev@nordicsemi.no>
The generic BT shell header, bt.h, contained quite a few things
related to audio, which should be put in the audio.h shell header
file.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Updated nRF5 Series LLL implementation to support multiple
peripheral CIS establishment from peer central devices.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
