Setting the extended advertiser as default advertiser
to improve both preformence and reliability.
Signed-off-by: Ingar Kulbrandstad <ingar.kulbrandstad@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>
Convert model_ackd_send and model_send from health_cli.c to a common API
to get rid of code duplication in other client models that implement
synchronous messages' sending.
Signed-off-by: Pavel Vasilyev <pavel.vasilyev@nordicsemi.no>
Change linkage for ctx_shell and default_key in shell.c to external to
share their usage in other places.
Signed-off-by: Pavel Vasilyev <pavel.vasilyev@nordicsemi.no>
This commits renames `health_cli.c` and `cfg_cli.c` under `mesh/shell`
folder to `health.c` and `cfg.c` accordingly. This is to create a
generic pattern for the future, when not only clients, but server models
will also expose some API through shell commands. This means that server
models' shell commands should be located in the same file as client
commands to avoid cluttering `mesh models` subcommand.
Signed-off-by: Pavel Vasilyev <pavel.vasilyev@nordicsemi.no>
This code is a boulerplate that will be needed for many models in mesh
shell module. This commit adds a special macro designed to improve
readability of the code and helps to avoid potential bugs when
copy-pasting identical code.
Signed-off-by: Pavel Vasilyev <pavel.vasilyev@nordicsemi.no>
Not all shell commands may be needed for an end application even if a
certain feature is enabled. In that case they will just occupy flash
space without being used. This commit adds a separate Kconfig for each
feature to make possible to selectively disable some mesh shell commands.
Signed-off-by: Pavel Vasilyev <pavel.vasilyev@nordicsemi.no>
Move mesh shell related options to a new file to avoid cluttering of
the main Bluetooth Mesh Kconfig file when extending it with new options.
Signed-off-by: Pavel Vasilyev <pavel.vasilyev@nordicsemi.no>
Since the net cache is accumulated in positive order, the net
cache uses reverse order traversal, which can filter duplicate
messages more effectively.
Signed-off-by: Lingao Meng <menglingao@xiaomi.com>
Since in bt_mesh_net_recv will call msg_cache_add
and if lpn reject this message, should remove this cache.
But this function sequentially and will not be interrupt
by any thread so, we no need a copy var to save this value.
Signed-off-by: Lingao Meng <menglingao@xiaomi.com>
By design, adv main set can use to send relay mesh messages
but here adv main set is invoked.
Add in the ended, so, adv main set only can be used when all
relay adv sets busy.
and this way does not cause too much delay for adv main set,
due to local message will process firstly.
Signed-off-by: Lingao Meng <menglingao@xiaomi.com>
Adds feature that deletes the shell CDB node after perfoming a config
client reset of another node.
Signed-off-by: Anders Storrø <anders.storro@nordicsemi.no>
Alters several provisioning related features to allow commands to be
functional with application defined Provisioning properties &
capabilities.
Signed-off-by: Anders Storrø <anders.storro@nordicsemi.no>
Some variables shadow global, which is difficult to handle when
porting to other platforms with shadow warnings turned on.
https://github.com/zephyrproject-rtos/zephyr/pull/50581
Signed-off-by: Lingao Meng <menglingao@xiaomi.com>
Adds Kconfig option for the Mesh Shell Health server.
The shell implementation of Health Server test commands is
dependent on the user adding and initializing the shell Health
Server instance in their application to make the commands functional.
To prevent any confussion about how these commands work, a Kconfig
option has been added to allow conditional compilation of these
commands. In this way the user must actively enable this feature, making
the conditional requirements more clear.
Signed-off-by: Anders Storrø <anders.storro@nordicsemi.no>
When the node is part of only one subnet, it starts Network ID
advertisements with infinite timeout. It is not an issue when using
legacy advertising and extended advertising with shared set for local
messages and GATT because any message will stop advertisements. However,
when a separate adv set is used for GATT advertisements,
`struct bt_le_ext_adv_cb.start` callback won't be called, but the
implementation relies on that callback to switch to a next subnet.
Calling `bt_mesh_adv_gatt_update()` stop GATT advertisement so that the
implementation can switch to the next subnet.
Signed-off-by: Pavel Vasilyev <pavel.vasilyev@nordicsemi.no>
Reduce advertising duration to `max_timeout` when advertising node
identity for multiple subnets. This will let the node to interleave
subnets for NODE_ID_TIMEOUT.
Signed-off-by: Pavel Vasilyev <pavel.vasilyev@nordicsemi.no>
`struct bt_le_ext_adv_start_param.timeout` resolution is in 10ms. When
duration is less than 10ms, division will be evaluated to zero, which
will result in indefinite timeout. Set timeout to 10ms to avoid
indefinite timeout. The advertisement will end earlier anyway after the
event is finished.
Signed-off-by: Pavel Vasilyev <pavel.vasilyev@nordicsemi.no>
Sharing the same pool for both, local and relayed messages has some
disadwantages. In a high-traffic environment the relayed messages may
occupy the whole pool without letting the node to send own messages,
neither to process segmented messages. The vice versa situation is
possible when local messages takes the whole pool so that the node is
not able to relay other messages. Another thing is that it doesn't let
to controll amount of local and relayed messages that the node can send
at once. Controlling the number of messages that the node can relay may
help to reduce the message latency.
This change adds separate net_buf_pool for the relayed messages that
helps to avoid problems described above.
Signed-off-by: Pavel Vasilyev <pavel.vasilyev@nordicsemi.no>
If `struct bt_mesh_prov::unprovisioned_beacon` callback is not set,
don't process the beacon.
Signed-off-by: Pavel Vasilyev <pavel.vasilyev@nordicsemi.no>
When friend entry is cleared by `friend_clear` after an adv buffer has
been sent to that friend in `friend_timeout`, another friend request
coming after this may reuse the same entry. It is possible that
`friend_timeout` will be scheduled twice: once by the friend request
and once by `buf_send_end` of the previously sent buffer.
In that case, the code will assert on line friend.c:1234. It is not
possible to cancel an adv buffer for the friendship that is to be
cleared, therefore, we need to ensure in correctness of the pending_buf
value in start and end callbacks.
Signed-off-by: Pavel Vasilyev <pavel.vasilyev@nordicsemi.no>
When local node enter iv recovery procedure, should clear
replay list at first due to avoid possible wrong replay attack.
At present, `bt_mesh_rpl_clear` schedule a timer to clear rpl,
this would cause local node unable precess mesh message(IV Updated).
Also, in `bt_mesh_rpl_pending_store` also not clear rpl, which check
`BT_MESH_VALID` flag, always valid for iv recovery.
So, Add `rpl_clear_pending` work handler process rpl clear.
Signed-off-by: Lingao Meng <menglingao@xiaomi.com>
The CDB module is a mandatory part of provisioning features in BTM, and
can not be regarded as experimental anymore.
Signed-off-by: Anders Storrø <anders.storro@nordicsemi.no>
Using the passed address is incorrect since the passed address can be
zero to find the lowest available address.
Signed-off-by: Pavel Vasilyev <pavel.vasilyev@nordicsemi.no>
According to MshPRFv1.0.1, section 3.10.7, the Provisioner shall only
use reuse addresses of the deleted nodes after IV Index is updated.
This change prevents CDB reusing addresses of the deleted nodes until IV
Index is updated. Due to a high range of the unicast addresses, CDB only
stores the highest deleted address once a node is deleted. This creates
a limitation where allocating a node with the high primary unicast
address and then deleting it will make CDB skip big range of address
until next IV Index update.
Signed-off-by: Pavel Vasilyev <pavel.vasilyev@nordicsemi.no>
This commit splits out the model related commands from mesh shell to
separate shell modules. The purpose of this alteration is to make model
commands more flexible by letting the application itself define the
model instances.
- Move Health Client commands to separate shell module.
- Move Config Client commands to separate shell module.
- Remove Health Client instance from shell.c.
- Add Health Client instance to mesh_shell test application.
- Add util library for Mesh shell.
- Some refactoring of shell menu layout.
- Update Mesh shell documentation.
Signed-off-by: Anders Storrø <anders.storro@nordicsemi.no>
Moves shell.c to shell subfolder. Done as a separate step to preserve
diff between previous and new revision of shell.c.
Signed-off-by: Anders Storrø <anders.storro@nordicsemi.no>
The current implementation of provisioning features is dependent on CDB
to compile, which is reflected poorly in the Kconfig options for mesh.
This commit alters the Kconfig options for provisioning to be dependent
on the CDB option to compile.
Signed-off-by: Anders Storrø <anders.storro@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>
This PR adds a `*_cli_*` infix to the Config Client API to match
the changes in Health Client. The old API is marked as deprecated.
Signed-off-by: Michal Narajowski <michal.narajowski@codecoup.pl>
Additional uncontrolled delay for transmitting segack affects throughput
of the segmented messages. Though the best throughput can be achieved
with the smallest allowed values, move them to Kconfig option so that
the segack delay can be controlled by a user.
Signed-off-by: Pavel Vasilyev <pavel.vasilyev@nordicsemi.no>
Swaps response scheme for Fault_test command so that it
aligns with description of the documentation.
Signed-off-by: Anders Storrø <anders.storro@nordicsemi.no>
Adds missing feature so that the config client is able to receive
odd number of app index entries.
Signed-off-by: Anders Storrø <anders.storro@nordicsemi.no>
If LPN loses Subscription List Add Confirmation it repeats
Subscription List Add frame. Friend doesn't check both
transaction number and presence addresses in the list.
That causes situation when Friend fills the same address
full list.
Signed-off-by: Aleksandr Khromykh <aleksandr.khromykh@nordicsemi.no>
With extended advertiser a precision of transmission becomes high.
The first frame is emitted quite quickly. The scanner latency
config option to enable scanner a bit in advance is not sufficient.
Sometimes replays on Friend Poll are lost.
Signed-off-by: Aleksandr Khromykh <aleksandr.khromykh@nordicsemi.no>
Swaps response scheme for Fault_get & Fault_clear command so that it
aligns with description of the documentation.
Signed-off-by: Anders Storrø <anders.storro@nordicsemi.no>
The def command Indicates that a comment block contains documentation
for a #define macro. This is useful if the comment block documents a
macro not adjacent to it, e.g.
```c
/**
* @def MAX(x,y)
* @brief Computes the maximum of @a x and @a y.
*/
#ifdef XXX
#define MAX(x,y) ...
#endif
```
However, it is not necessary if the comment is adjacent to the
definition, e.g.
```c
/**
* @brief Computes the maximum of @a x and @a y.
*/
#define MAX(x,y) ...
```
This patch removes all unnecessary def entries in-tree.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
If LPN lost Friend node (all Poll attempts didn't succeed)
no reason to send Friend Clear to the Friend node.
This makes LPN powersupply inefficient usage.
During Friend Clear sending LPN rejects any frame from
primary subnetwork.
Specificastion doesn't mandate that LPN would send Friend Clear
if polling didn't succeed.
Signed-off-by: Aleksandr Khromykh <aleksandr.khromykh@nordicsemi.no>
Scanner was disabled after Friend Clear command. LNP didn't have
chance to receive Friend Clear Confirmation command back.
If option BT_MESH_LPN_ESTABLISHMENT is off then scanner is left
disabled and next Friend Offer after Friend Request
cannot be received.
PR fixes this.
Signed-off-by: Aleksandr Khromykh <aleksandr.khromykh@nordicsemi.no>
Mesh Profile Specification errata 11310
Section 4.4.4.1
"This model shall support model publication, as defined in Section
4.2.2 of the Mesh Profile specification and model subscription,
as defined in section 4.2.3 of the Mesh Profile specification."
Signed-off-by: Michal Narajowski <michal.narajowski@codecoup.pl>
According Mesh Spec 1.0.1:
Upon receiving and successfully authenticating to
Secure Network beacon for a primary subnet whose
IV Index is 1 or more higher than the current known IV
Index, the node shall set its current IV Index and its
current IV Update procedure state from the values in
this Secure Network beacon.
Look like test_iv_index.c:81
``` C
(bt_mesh_net_iv_update(TEST_IV_IDX + 1, BCN_IV_IN_IDLE));
```
This test case already exists, but the wrong testcase
to test wrong code.
Move `is_iv_recovery` into `bt_mesh_net_iv_update`.
First check whether it is IV recovery, and then carry out
the subsequent IV normal update procedure.
Signed-off-by: Lingao Meng <menglingao@xiaomi.com>
This fixes an issue where IV index stage counting timer is not
started after the node is provisioned. This would have prevented node
from performing IV index updates.
Signed-off-by: Omkar Kulkarni <omkar.kulkarni@nordicsemi.no>
Struct bt_mesh_send_cb.end is called from the mesh advertiser thread.
bt_mesh_reset now requires more stack size, thus can't be called from
the advertiser thread. k_work instance requires less memory than needed
for bt_mesh_reset.
Signed-off-by: Pavel Vasilyev <pavel.vasilyev@nordicsemi.no>
Upon bt_mesh_reset call, all mesh modules has to clear the privisioning
and configuration data. None of the modules store or erase in settings
subsystem immediately, but put this on the mesh settings work. After the
settings work is scheduled, all stored data will be removed and the
device will eventually be unprovisioned. Until then, the device is not
completely unprovisioned, thus calling bt_mesh_prov_enable,
bt_mesh_provision (and bt_mesh_cdb_create) should not be allowed.
Struct bt_mesh_prov has a reset callback stating that after this
callback is called, the device has been reset and can be re-provisioned
again. Also, bt_mesh_reset API description states that after calling
bt_mesh_reset API, the device needs to reenable the provisioning layer
to be provisioned again. But this is not correct since the settings has
to be cleared before the device can be reprovisionined.
This commit makes bt_mesh_reset flush pending settings so that the
device can be reprovisioned immediately and the API will behave as
written in the description.
The same applies to bt_mesh_cdb_clear.
Signed-off-by: Pavel Vasilyev <pavel.vasilyev@nordicsemi.no>
