Adds a high-level documentation page for the Bluetooth Mesh Shell
subsystem, documenting all available commands and their parameters.
Signed-off-by: Trond Einar Snekvik <Trond.Einar.Snekvik@nordicsemi.no>
Resolves a corner case where the segmented sending would be rescheduled
before the segments were done sending. This would happen if the
processing of each segment took longer than the retry delay. The
consequence of this was that some segments were resent multiple times at
the cost of other segments, occasionally causing a loss of transfers.
Rearranges, realigns and documents seg_tx structure bitfields in
transport.c. Removes bitwidth of counter fields, and collecting all flag
fields into one byte.
Signed-off-by: Trond Einar Snekvik <Trond.Einar.Snekvik@nordicsemi.no>
Adds idle wait callback to proxy by counting the number of pending
notifications.
Ensures that all connected nodes have received messages before resetting
the mesh stack.
Signed-off-by: Trond Einar Snekvik <Trond.Einar.Snekvik@nordicsemi.no>
Reverse the order of these checks because invalid PDU type will
never be expected. If PDU type is invalid we should respond with
error 0x0002 - Invalid format.
Signed-off-by: Michał Narajowski <michal.narajowski@codecoup.pl>
Friend Request is also transmitted over local queue and
if both Friend and LPN features are enabled then we would
try to establish friendship with ourselves.
Fixes MESH/NODE/FRND/LPN testcases.
Signed-off-by: Michał Narajowski <michal.narajowski@codecoup.pl>
The hardcoded value was arbitrary and interfered with message cache
mechanism causing invalid behavior and PTS test to fail.
Fixes MESH/NODE/RLY/BV-02-C.
Signed-off-by: Michał Narajowski <michal.narajowski@codecoup.pl>
According to Mesh Profile Test Specification we should still
relay messages with DST address that is RFU.
Fixes MESH/NODE/RLY/BV-01-C.
Signed-off-by: Michał Narajowski <michal.narajowski@codecoup.pl>
Key refresh procedure was ignored on non-primary subnet.
Mesh Profile Specification v1.0.1:
3.10.4 Key Refresh procedure
"This procedure is used when the security of one or more network
keys and/or one or more of the application keys has been compromised
or could be compromised."
"It is possible to update each NetKey independently of all other
NetKeys. A Key Refresh procedure for one NetKey can be in a different
phase to another Key Refresh procedure for other NetKeys."
Signed-off-by: Michał Narajowski <michal.narajowski@codecoup.pl>
Reset provisioning bearers when bt_mesh_reset() is called. Accept
another provisioning attempt after link close.
Signed-off-by: Michał Narajowski <michal.narajowski@codecoup.pl>
This value was used quite often in the file so adding a
descriptive name should improve readability.
Signed-off-by: Michał Narajowski <michal.narajowski@codecoup.pl>
The comparison of a current tx link transaction id caused
us to not send ack and in effect fail provisioning.
Signed-off-by: Michał Narajowski <michal.narajowski@codecoup.pl>
Removes the memset of the prov_bearer_cb in PB-GATT during resets. This
allows the provisioning link to disconnect and reconnect again without
having to call pb_gatt_open.
Fixes#26343.
Signed-off-by: Trond Einar Snekvik <Trond.Einar.Snekvik@nordicsemi.no>
The frnd->last buffer can potentially be NULL if friend_clear is called
after the adv.c buf->busy check, but before the adv_start callback.
The current design is based on the adv.c thread being cooperative, and
therefore not yielding between the busy check and the adv_start
callback, but as the bt_le_adv_start call has to acquire a semaphore,
there's a possibility for friend_clear being called inbetween.
Fixes#26177.
Signed-off-by: Trond Einar Snekvik <Trond.Einar.Snekvik@nordicsemi.no>
Previous mode store function only can store single data,
change this to store as KV model, let's app-layer to manager
model data, other than by stack when node reset.
Signed-off-by: Lingao Meng <mengabc1086@gmail.com>
Adds propagation of error returns from the model init callbacks in
Access, and removing any other checks for successful init in the
foundation models.
Signed-off-by: Trond Einar Snekvik <Trond.Einar.Snekvik@nordicsemi.no>
CC: mesh/access.c
mesh/access.c: In function 'model_has_dst':
mesh/access.c:483:10:
warning: returning 'u16_t *' {aka 'short unsigned int *'} from
a function with return type '_Bool8' {aka 'unsigned char'} makes
integer from pointer without a cast [-Wint-conversion]
483 | return bt_mesh_model_find_group(&mod, dst);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Signed-off-by: chao an <anchao@xiaomi.com>
The transport segmented TX nack and seg_pending fields must be at least
6 bits to avoid overflow for 32 segment messages. This change rearranges
the seg_tx fields to gather all state flag fields in one byte, while
making the counter fields whole bytes.
Signed-off-by: Trond Einar Snekvik <Trond.Einar.Snekvik@nordicsemi.no>
Increases the transport segmentented tx seg_o counter to 6 bits to avoid
overflow when sending 32 segments. The check in the send loop would
previously never be false, which causes segments to repeat
unnecessarily.
Signed-off-by: Trond Einar Snekvik <Trond.Einar.Snekvik@nordicsemi.no>
Fixes bug where the transport segmented tx would decrement the attempt
counter every time it ran out of buffers. If transport ran out of
buffers 4 times before the sending could complete, the transfer would
end prematurely.
Moves the attempt decrement to only execute when all segments have been
sent.
Signed-off-by: Trond Einar Snekvik <Trond.Einar.Snekvik@nordicsemi.no>
Ensures that friend messages are enqueued, even if the packet is
received with an appkey is unknown to the friend. Previously, sdu_recv
would return EINVAL if the appkey was unknown, which would prevent the
lower transport layer from adding the packet to the friend queue. This
is irrelevant for the logic in lower transport, and should not be
returned as an error.
Fixes#24014.
Signed-off-by: Trond Einar Snekvik <Trond.Einar.Snekvik@nordicsemi.no>
Convert bluetooth mesh to using k_timeout struct. Many of the mesh
modules uses timeout calculations, so it is most practical to keep
the s32_t type and only initialize a k_timeout_t struct when
calling the kernel.
Signed-off-by: Joakim Andersson <joakim.andersson@nordicsemi.no>
Remove logging of individual threads spread out throughout the
bluetooth subsystem. The stacks can be analysed by enabling the
following options.
CONFIG_THREAD_ANALYZER=y
CONFIG_THREAD_ANALYZER_AUTO=y
CONFIG_THREAD_ANALYZER_RUN_UNLOCKED=y
Optional:
CONFIG_THREAD_NAME=y
Signed-off-by: Joakim Andersson <joakim.andersson@nordicsemi.no>
According Mesh Profile 1.0.1. A application key shall
binding single network key. And Device key shall bind all
network key, and dev key only known by cfg_cli and node self,
only used by cfg_cli & cfg_srv.
Fixes: #21088
Signed-off-by: Lingao Meng <mengabc1086@gmail.com>
Fixes bug where the config client's net_key_status handler would attempt
to pull two key indexes from a message which only holds one.
Fixes#24601.
Signed-off-by: Trond Einar Snekvik <Trond.Einar.Snekvik@nordicsemi.no>
After #22013, bt_le_adv_param got additional fields which were passed to
the bluetooth API uninitialized in the BT Mesh module. This
zero-initializes the entire structure in all usages to avoid passing
uninitialized data now and in the future.
Signed-off-by: Trond Einar Snekvik <Trond.Einar.Snekvik@nordicsemi.no>
Add option for low power node automatically subscribe
all-nodes-address to friend sub list.
Fixes: #24009
Signed-off-by: Lingao Meng <mengabc1086@gmail.com>
Makes a common link_closed function for PB-GATT, getting rid of a bug
where cb_data is reset before the link closed callback. Also ensures
that the link close and reset order is the same in both scenarios.
Signed-off-by: Trond Einar Snekvik <Trond.Einar.Snekvik@nordicsemi.no>
When PB-GATT Procedure timeout, func `bt_mesh_pb_gatt_close`
will also dumplicated with `link_closed()`
Signed-off-by: Lingao Meng <mengabc1086@gmail.com>
Fix seg_tx occupy when no segment not send
and allocate rx_seg when receive such message.
Fixes: #24101
Signed-off-by: Lingao Meng <mengabc1086@gmail.com>
Fixed overflow risk when `poll_timeout` is 1s,
although it is not recommended to use it like this,
when it is used like this, `POLL_TIMEOUT_MAX`
will overflow.
When the poll timeout set like above, the default `6`
req_attemps for first pull request lost,
may cause this procedure bigger than poll_timeout,
Well, stop scanning when lpn terminated friendship,
this will save lots of energy, when lpn mode enable,
the scanning will be start after `FRIEND_REQ_RETRY_TIMEOUT`
Signed-off-by: Lingao Meng <mengabc1086@gmail.com>
Run the int_literal_to_timeout Coccinelle script to fix places where
it is clear that an integer duration is being passed where a timeout
value is required.
Signed-off-by: Peter Bigot <peter.bigot@nordicsemi.no>
The publication context is checked for NULL in bt_mesh_model_publish()
however it was dereferenced before that. Move the assignment to
ctx.send_rel to the same place where other ctx members are set.
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Deprecate BT_LE_ADV defines in hci.h that are expected to be used by the
application in the scan received callback to identify the advertising
PDU type. These defines are mixing HCI input parameters and advertising
PDU types. Internally it is acceptable to mix these, but at the API we
should to mix in them.
Signed-off-by: Joakim Andersson <joakim.andersson@nordicsemi.no>
Rename filter_dup parameters used for scanning filter options to the
more generic name options, and make scan filter options follow same
naming patters as advertising and initiator scan options.
Signed-off-by: Joakim Andersson <joakim.andersson@nordicsemi.no>
Splits PB-ADV and PB-GATT into separate modules with a common interface
to modularize prov.c.
Additional trivial fixes from testing:
- Reduces warnings for normal occurances like repeated packets.
- Makes link ack a non-reliable packet to prevent it from being repeated
until prov invite.
- Provisioner does not send link fail, but closes the link (as per spec
section 5.4.4). This prevents lingering zombie links on both sides.
Signed-off-by: Trond Einar Snekvik <Trond.Einar.Snekvik@nordicsemi.no>
Allocates segmented message buffers as slabs in a common pool for RX and
TX. This reduces memory requirements for both TX and RX, as TX messages
can be stored without the network and advertising buffer overhead, and
RX can use only the slabs it needs, instead of allocating a full size
segmented message. This approach also removes the need for decrypting
the segments for each retransmission, reducing overall processing load.
Slab based segmentation for tx also introduces queuing of segmented
messages, which allows the application layer to send multiple messages
to the same destination without violating Bluetooth Mesh specification
v1.0.1, section 3.6.4.1. This mechanism is provided through a flag that
blocks segmented messages to a destination which a message is already
being sent to until the previous message finishes.
This changes the SDU size configuration to a symmetrical
RX_SEG_MAX/TX_SEG_MAX pair of configurations, plus a new segment pool
side configuration. It also removes the binding between the TX_SEG_MAX
config and the advertising buffers, reducing the minimum advertising
buffer count from 6 to 3.
Signed-off-by: Trond Einar Snekvik <Trond.Einar.Snekvik@nordicsemi.no>
The Friend queue uses the message SeqAuth to determine whether the
message is already in the queue. To facilitate this, the SeqAuth is
passed around as a pointer throughout the transport modules. In the
bt_mesh_ctl_send functions, this parameter is also exposed in the API,
but the internal usage is inconsistent and buggy. Also, no one actually
uses this parameter.
- Removes seq_auth param from bt_mesh_ctl_send, instead passing NULL
directly to the friend module, to enforce its addition to the queue.
- Makes the seq_auth pointer const throughout the friend module.
Signed-off-by: Trond Einar Snekvik <Trond.Einar.Snekvik@nordicsemi.no>
The commit e85dd8af5d changed the way the BT_ECC Kconfig option
is enabled, however it got the dependency wrong. The dependency should
only look at BT_SMP_OOB_LEGACY_PAIR_ONLY if BT_SMP was also enable.
This broke e.g. the build of the mesh_demo app for the BBC
micro:bit since the memory consumption jumped up by roughly 2k.
This patch fixes the issue, and in the same go makes the Mesh handling
consistent by also using a conditional default rather than select.
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Changes the behavior of the message cache to optimize for cache
capacity. Previously, the message cache's primary function was to avoid
decrypting messages multiple times, although the cache's main function
in the spec is to avoid message rebroadcasting. Optimizing for minimal
decryption causes us to fill the network cache faster, which in turn
causes more cache misses, potentially outweighing the advantage.
Now stores src + seq in message cache instead of field hash value. This
cuts cache size in two, while including more of the sequence number than
before.
Adds messages to the cache only after the packet is successfully
decrypted. This reduces noise in the cache, and ensures that no
invalid deobfuscations are added.
Additionally, this fixes a bug where multiple calls to net_decrypt with
the same packet failed, as the message cache found its own entry from
the previous call.
Signed-off-by: Trond Einar Snekvik <Trond.Einar.Snekvik@nordicsemi.no>
