cherry-pick Bluetooth Host commits #2625
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
theob-pro
force-pushed
the
bring-bt-host-changes
branch
from
ba3abbc to
e098314
Compare
PavelVPV
approved these changes
Mar 18, 2025
theob-pro
force-pushed
the
bring-bt-host-changes
branch
from
e098314 to
c18a0e4
Compare
…eration fails When `bt_le_oob_get_local` or `bt_le_ext_adv_oob_get_local` is called and SMP is enabled, `bt_smp_le_oob_generate_sc_data` is called to generate a Random Number and a Confirmation Value needed for OOB data. These values are based on the device's public key. The public key is generated only once when `bt_smp_init` is called. If public key generation fails, the callback passed to `bt_pub_key_get` is called with `pkey` set to NULL. The `bt_smp_pkey_ready` callback gets called, but it doesn't release the `sc_local_pkey_ready` semaphore thus leaving `bt_smp_le_oob_generate_sc_data` wait for semaphore with `K_FOREVER`. This commit replaces the semaphore with a conditional variable and requests a public key again if the key is NULL thus solving 2 issues: - handling the case where the callback was triggered notifying about the completion of the public key request, but the key was not generated, - handling the case where multiple threads trying to acquire the same sempahore. The timeout is used instead of K_FOREVER to avoid cases when callback has never been triggered. Signed-off-by: Pavel Vasilyev <[email protected]> (cherry picked from commit 9757ffa)
…_WHITELIST macro Macro has been deprecated since 2.7 LTS(!) and is using non-inclusive terminology. Drop it. Signed-off-by: Benjamin Cabé <[email protected]> (cherry picked from commit 110c49f)
When scanning is started on bt_bap_broadcast_assistant_scan_start then BAP_BA_FLAG_SCANNING flag for the broadcast assistant instance should not be cleared. Fixes #85937 Signed-off-by: Jens Rehhoff Thomsen <[email protected]> (cherry picked from commit c443343)
…nd buffer Remove nested allocation of HCI command buffer so that one less HCI command buffer is needed starting extended advertising with privacy enabled. Signed-off-by: Vinayak Kariappa Chettimada <[email protected]> (cherry picked from commit cc1b534)
The gatt docs are very sparce, this commit adds imrpovements for the GATT server API part. Others will follow. Signed-off-by: Kyra Lengfeld <[email protected]> (cherry picked from commit 775cc35)
… BT RX thread This commit alignes the timeout value for allocating buffers within att on the BT RX thread, making it consistent within att.c, see bt_att_req_alloc. We are inferring in many bt_gatt_* functions that if called from a BT RX thread (which is inherently the case if called from a callback when running a Bluetooth application), we don't block and instead return -ENOMEM when the ATT request queue is full, avoiding a deadlock. This promise is fulfilled within bt_att_req_alloc, where the timeout for allocation of the request slab is set to K_NO_WAIT if we are on the BT RX thread. Unfortunately, we break this promise in bt_att_chan_create_pdu, where the timeout for allocation of the att pool is still K_FOREVER and deadlocks can (and do) occur when too many requests are sent yet the pool is depleted. Note: Both req_slab and att_pool sizes are defined by CONFIG_BT_ATT_TX_COUNT. If applications start getting -ENOMEM with this change, they were at risk of such a deadlock, and may increase CONFIG_BT_ATT_TX_COUNT to allocate the att pool for their requests. Note: This possible deadlock has been flying under the radar, as att_pools are freed when the HCI driver has sent it to the controller (instead of when receiving the response, as it happens with req_slabs) and due to the att_pool and the req_slab being both sized by CONFIG_BT_ATT_TX_COUNT, and req_slab being allocated before and returning -ENOMEM already if there is no space, it takes a more specific situation to deplete the att_pool but not the req_slab pool at this point. Note: Ideally, we don't want functions to behave differently depending on which thread they are running, and while this commit makes it more consistent, it should be considered a workaround solution. Signed-off-by: Kyra Lengfeld <[email protected]> (cherry picked from commit 6464ffa)
…_uuid bt_gatt_find_by_uuid used bt_gatt_attr_value_handle but that function only works to get the value handle of a characteristic declaration, i.e. if the UUID is not BT_UUID_GATT_CHRC then it would always return handle = 0. This meant that bt_gatt_find_by_uuid would always use handle = 0 as the starting handle for non-BT_UUID_GATT_CHRC attributes, instead of the handle of the provided attr. This was not an issue for any UUIDs that may only exist once on a GATT server, which is most UUIDs, but for UUIDs like the BT_UUID_TBS_* UUIDs that may be multiple instances of, it would always return the first attribute rather than the one starting from the provided start attr. This commit also ensures that we do not overflow the `end_handle` when adding 2 uint16_t values. Signed-off-by: Emil Gydesen <[email protected]> (cherry picked from commit e4c5bb9)
Adds improvement to the GATT API documentation by providing examples of where (and how) certain structs are used, and adds specification references to the fields of bt_gatt_cpf to clarify their purpose. Signed-off-by: Håvard Reierstad <[email protected]> (cherry picked from commit c7a19da)
This commit adds improvements for the GATT client API part. Signed-off-by: Kyra Lengfeld <[email protected]> (cherry picked from commit b136f6d)
…tures The function supports reading multiple controller-based values. It is effectively a copy of struct bt_dev_le but in a more application-oriented definition. It was chosen to keep the features as an array rather than a 64-bit value, as the comparison macros work on arrays and that there already exists new bits > 64 in the core spec which is not yet supported by Zephyr. It is being smoke tested in a generic GATT client test, as the individual values may depend on several Kconfig options. Signed-off-by: Emil Gydesen <[email protected]> (cherry picked from commit 44c5c1d)
When allocation fails in `hci_rx_cb`, we should return immediately instead of continuing the execution of the function. Also, we need to free the allocated buffer from the heap when sending to the message queue fails to avoid memory leak. Signed-off-by: Axel Le Bourhis <[email protected]> (cherry picked from commit 0034e12)
…mplete event As the previous version of the MCXW71 BLE Controller wasn't sending a command complete event after setting the BD address, we used a workaround to directly send the HCI command without using zephyr's API. Now, on the latest version of MCXW72 and MCXW71, this issue is fixed, so we need to use `bt_hci_cmd_send_sync` to properly expect the command complete event. Signed-off-by: Axel Le Bourhis <[email protected]> (cherry picked from commit e5a1c0b)
…when bt disable remove the power off BLE controller function since the hci_reset cmd is enough, also for wifi-ble coex scenario, need to maintain BLE controller for some info in SMU region Signed-off-by: Ying Zhang <[email protected]> (cherry picked from commit 6b16d10)
- added HCI command to send hci-reset in bt_disable() func, and remove its call from the driver close() call. - remove rsp buf pass to 'hci_reset_complete()' func as status is already checked under 'bt_hci_cmd_send_sync()'. Signed-off-by: Nirav Agrawal <[email protected]> (cherry picked from commit 438701f)
Clarifying `bt_gatt_attr` docs. Signed-off-by: Kyra Lengfeld <[email protected]> (cherry picked from commit c31b962)
This commit adds more detailed description to the "Bluetooth APIs" and "GAP" group within the bluetooth header. Signed-off-by: Kyra Lengfeld <[email protected]> (cherry picked from commit 3570783)
This commit changes log level for already registered callback from warning to debug as this is cause unnecessary noise and doesn't indicate actual issue. Caller gets a certain error code for this case which can be handled properly. Signed-off-by: Pavel Vasilyev <[email protected]> (cherry picked from commit 886faba)
The api has not changed since its introduction some time ago, mark it as stable. Signed-off-by: Herman Berget <[email protected]> (cherry picked from commit 8b909b4)
…ection Subrating The API has not changed since it was introduced so should no longer be considered experimental. Signed-off-by: Timothy Keys <[email protected]> (cherry picked from commit 7ef8116)
Fix the following warning
att.c:734:3: warning: label followed by a declaration is a C23 extension
[-Wc23-extensions]
734 | k_tid_t current_thread = k_current_get();
| ^
By wrapping that code as a compound statement
Signed-off-by: Alberto Escolar Piedras <[email protected]>
(cherry picked from commit 1dab77b)
theob-pro
force-pushed
the
bring-bt-host-changes
branch
from
c18a0e4 to
f0f398b
Compare
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
16 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Cherry-pick Bluetooth Host commits