Refactor out drift compensation implementation so as to
reuse it for Periodic Sync feature.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Refactor ull_conn_tx_ack function as it no longer needs to
return the connection context back to caller.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Detect and handle AD data set race condition between thread
and ISR context.
Fixes#27637.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
It fixes:
../../../../include/toolchain/gcc.h:169:30: warning: statement will
never be executed [-Wswitch-unreachable]
169 | #define __fallthrough __attribute__((fallthrough))
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Fix conditional compilations that fail when combinations
of Broadcaster, Observer, Peripheral and/or Central are
selected to build an application.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
-Wimplicit-fallthrough=2 requires a fallthrough comment or a compiler
to tells gcc that this happens intentionally.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Implementation of Advertising Set Terminated Event on
duration sans use of additional variable in advertising
set context structure.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Fix flash operation timeout due to incorrect use of
secondary ticker to abort any radio in use. Ticker id 0
is reserved for split controller's pipeline preempt timeout.
Using the same ticker id caused the secondary ticker to
not be started if controller is using the same ticker id
for pipeline preempt timeout.
Fixes#26333.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Review rework, added comments, TODOs, FIXMEs and converted
magic number use in advertising set state flags to defines.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Refactor pdu.h defined in anticipation of reused in
Advertising Extensions implementation. To reuse in
calculating ticks_slots etc.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Change the implementation of extended scan PDU rx list to use
the extra pointer as the next field in the linked list node.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Remove us_radio_rdy field in the footer structure as its always
a constant value which can be obtained using a functional
interface to HAL.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
With BT_CTLR_USER_TICKER_ID_RANGE it is possible for vendors to add a
number of ticker nodes for proprietary purposes. The feature depends on
BT_CTLR_USER_EXT.
Signed-off-by: Morten Priess <mtpr@oticon.com>
Split BT_CTLR_CONN_RSSI option into two, the base option enables the
Read RSSI command, while the new BT_CTLR_CONN_RSSI_EVENT enables the
connection RSSI events. There is no handling of RSSI events, only a
BT_INFO log.
Signed-off-by: Joakim Andersson <joakim.andersson@nordicsemi.no>
SMI TX is different than other controller features in that it does not
necessarily imply any software changes; whether SMI TX is supported
may be simply a matter of hardware calibration. This change supports
using the same software on chips that do or do not support SMI TX
depending on calibration.
Signed-off-by: Wolfgang Puffitsch <wopu@demant.com>
Under race conditions it is possible that there is no call
to k_sem_give to the waiting k_sem_take in the ull_disable
function.
ull_disable function checks for reference count before
using a mayfly to schedule lll_disable, which in turn
would close requested currently active role event leading
to done event being propogated to ULL. Done event would
then call the set disabled_cb callback when the reference
count is zero, giving the semaphore to the waiting
k_sem_give in the ull_disable.
Under race conditions if the reference count reached zero
after the reference count check and before the disabled_cb
was assigned in the ull_disable function, then there are
chances that a k_sem_give is not called while ull_disable
proceeds to waiting using k_sem_take.
Fixes#21586.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Pass the scanner LLL context in the generated connection
complete event with unknown connecion id for HCI Create
Connection Cancel Command Response.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Directed advertising timeout released PDU Rx quota which it
should not be.
Relates to assert in #21006.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Add support for Zero Latency IRQs, which avoids any Zephyr
OS or application influenced ISR latencies on the
controller's ISRs.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Use semaphore to synchronize lll_reset completion with HCI thread for
returning (command complete) only when all is done.
Signed-off-by: Morten Priess <mtpr@oticon.com>
Rx nodes are reserved during control procedures for
generation of HCI event on completion. Fix the allocation
and reservation in the form of a linked list per connection
context. Worst case, this list will hold one rx node for
overlapping non-instant control procedure (Data Length
Update) plus two rx node for control procedure with instant
(PHY update with Data Length Update support).
Fixes#19198.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Call LLL reset functions when calling ll_reset to avoid carrying LLL
state across HCI resets. Respective functions already exist in LLL but
had not been called from anywhere.
Signed-off-by: Wolfgang Puffitsch <wopu@demant.com>
Fix missing generation of data length update HCI event when
effective tx and rx timings change due to PHY update
procedure.
Fixes BT LL TS 5.1.0 test:
LL/CON/MAS/BV-52-C [Master Receiving Data, LE Coded, CI
Change]
Relates to #17097.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Fix ULL implementation that uses conditional compilation by
replacing back to use of #if defined(...) in code accessing
compiled out struct members.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
When more than one simultaneous connections are active,
transmitting data packets to peer, a termination causes
host to use the flushed pending number of completed packets
count for other active connections. This is on reception
of HCI disconnection complete event. But the controller has
not yet released any of the pending enqueued Tx buffers
which was happening after the disconnection event was
dispatched to HCI layer.
The fix here is to dispatch the disconnection complete event
from the LLL context after pending Tx buffers have been
flushed and the buffers get returned to Tx pool in the ULL
context. This way buffers are in the Tx pool before host
get to process the disconnection complete event.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Fix incorrect ticker/mayfly user id used in scheduling the
abort of a radio event. Incorrect use of thread context as
the caller caused the abort function to be scheduled from
thread context while being called from ULL high context
level in reality. This could cause corruption of mayfly
scheduling.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
When multiple simultaneous peripheral connections are
supported, restarting connectable advertising by host on
peripheral connection establishment failed in controller.
This prevented establishing new connections while first
connection was active.
The failure was caused by a bug in the way controller was
using quota for Rx PDU buffers. As the quota count was
release before the connection complete event rx PDU buffer
being released, the Rx PDUs needed to reserve for
connection complete event for new connectable advertising
was not available. This caused the connectable advertising
enable to fail.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Added to allow vendor specific increase of user operation capacity for
ULL_HIGH, to support queuing additional ticker operations.
Signed-off-by: Morten Priess <mtpr@oticon.com>
When both central and peripheral are supported, one each Rx
node will be needed by connectable advertising and the
initiator to generate connection complete, hence
conditionally set the count.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Added ll_rx_link_inc_quota to allow vendor to manipulate
mem_link_rx.quota_pdu when cleaning up nodes via ull_vendor.h.
Signed-off-by: Morten Priess <mtpr@oticon.com>
Post Tx pool corruption fix, clean up code and add comments
explaining the use of Tx node next field used to indicate
the Tx node's allocation from Control or Data pool.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
When compiling a sample like a beacon, none of the `if defined()`
statements in the switch case are actually compiled in, so the compiler
complains of an orphan mem_release();break; sequence.
Add a type that we know will always be compiled in in order to avoid
this issue.
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
Reduce the maximum number of pending event done elements by
decoupling it from the maximum pipeline elements.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Fix the pipeline prepare or resume dequeue loop from
overflow. Fixes premature exit of dequeue of done FIFO
which caused the pipeline to overflow due to previously
placed done is not dequeued correctly and next event not
being put back into the pipeline.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Conditional compile the use of Rx Pool used to dispatch ULL
context generated messages towards LL thread context, which
is presently only used in connections.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
move entropy.h to drivers/entropy.h and
create a shim for backward-compatibility.
No functional changes to the headers.
A warning in the shim can be controlled with CONFIG_COMPAT_INCLUDES.
Related to #16539
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Moved and renamed ull_entropy_get to lll_entropy_get, placed under
vendor specific ll_sw. This is needed for SW implemented entropy,
to allow vendor implementation of faster, less secure random number
generator for randomizing ADV timing.
Signed-off-by: Morten Priess <mtpr@oticon.com>
Fix pending Tx control buffer leak on supervision timeout.
Queued tx buffers in LLL consists of both data and control
PDUs but only data buffers got correctly released.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Add some missing fatal asserts that need to be caught to
avoid unexpected failures in the implementation of the
architecture.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Added an internal ull_update_mark function to detect race
conditions while stopping ticker instances during slave
drift, disconnection and connection update.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Implements hooks to implement user protocols in ull.c
A user defined init function can now be called, this code is gated by
the CONFIG_BT_CTLR_USER_EXT define.
A user defined PDU length can now also be defined using the Kconfig
CONFIG_BT_RX_USER_PDU_LEN
Signed-off-by: Asger Munk Nielsen <asmk@oticon.com>
Fix compilation issues in central_hr or peripheral_hr sample when using
split LL and privacy is enabled.
Signed-off-by: Joakim Andersson <joakim.andersson@nordicsemi.no>
Code refactored to allow calling of a proprietary rx demux function.
This will enable implementation of proprietary protocols and
functionality that is not yet public, while keeping a common zephyr
code base.
Signed-off-by: Asger Munk Nielsen <asmk@oticon.com>
The old footer was appended after PDU using pointer arithmetic. Now
the footer fields have been moved to the header struct, the
footer fields are now statically located in the data structure,
this is type safe and fields can be referred to by their actual
names rather than indirectly through reference to other members,
thus avoiding pointer arithmetic. Secondly, this change will pave
the way for adding other meta data in the future.
Signed-off-by: Asger Munk Nielsen <asmk@oticon.com>
Fix incorrectly defined Rx and Tx buffer sizes. Wrong
calculation allocated more memory than necessary.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Implement a lower ISR latency ULL processing design. Instead
of looping use ISR/mayfly tail-chaining to process
successive ULL messages.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Implemented ULL to yield from processing in an infinite loop
if current PDU being handled is deferred.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
* Rename the stack local 'done' to 'evdone' so as to disambiguate from
MFIFO_DEQUEUE_PEEK(done) which is actually 'mfifo_done'.
* add comment on ull_slave_done
* add comments to addr_us_get
* add comments to HCI_CLASS
Signed-off-by: Mark Ruvald Pedersen <mped@oticon.com>
In ull.c ll_rx_get, a configuration without CONFIG_BT_CONN would
sometimes return an uninitialized node_rx.
In ull_scan.c, the scanning channel was not initialized to 0 (37). This
would cause new scanning to start at random index (0,1,2).
Signed-off-by: Morten Priess <mtpr@oticon.com>
Ported implementation of Data Length Update Procedure to
ULL/LLL architecture of the Link Layer.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
mem_pdu_rx is not runtime resized meaning PDU_RX_SIZE_MIN is no longer
relevant.
Dynamic resizing was a left-over from the old bt controller.
Its size field was only ever set to PDU_RX_SIZE_MIN.
PDU_RX_SIZE_MIN only accounted for advertise PDUs.
So we remove size field from mem_pdu_rx.
However, the rest of the BT controller expects to be able to pull
RX_CNT-number of valid nodes from the pool at init/reset (rx_alloc)
time. This will fail unless we inflate the capacity by one.
To avoid inflation-by-one, mem_* would have to be rewritten to avoid
always-one-extra pattern.
Signed-off-by: Mark Ruvald Pedersen <mped@oticon.com>
Refactor PDU_RX_POOL_SIZE into its constituents and name each term.
No functional change expected.
Signed-off-by: Mark Ruvald Pedersen <mped@oticon.com>
* Clarify certain functions and data structures in upper link layer.
* Disambiguate node rx event done, add commentary.
MFIFO_DEQUEUE macro does string concatenation.
Thus MFIFO_DEQUEUE(done) does not refer to done (the stack local), but
mfifo_done.
No functional change expected.
Signed-off-by: Mark Ruvald Pedersen <mped@oticon.com>
ull_ref_inc already existed, but not ull_ref_dec.
No functional change expected.
Consistency is preferred due to code navigation.
Signed-off-by: Mark Ruvald Pedersen <mped@oticon.com>
There are issues using lowercase min and max macros when compiling a C++
application with a third-party toolchain such as GNU ARM Embedded when
using some STL headers i.e. <chrono>.
This is because there are actual C++ functions called min and max
defined in some of the STL headers and these macros interfere with them.
By changing the macros to UPPERCASE, which is consistent with almost all
other pre-processor macros this naming conflict is avoided.
All files that use these macros have been updated.
Signed-off-by: Carlos Stuart <carlosstuart1970@gmail.com>
Rename ull_tx_ack_put to ll_tx_ack_put as ack is enqueued
into LL thread context from ULL ISR context.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Remove the redundant proof of concept template temporary
role implementation from the repository.
Relates to #12860.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
This is a squash merge of commits introducing the new split
Upper Link Layer and Lower Link Layer architecture of the
Bluetooth Low Energy controller.
This introduces a new, improved Link Layer based on the
concept of split responsibilities; The Upper Link Layer
(ULL) is in charge of control procedures, inter-event
scheduling and overall role management. The code for the
ULL is shared among all hardware implementations. The
Lower Link Layer (LLL) is responsible for the intra-event
scheduling and vendor specific radio hardware access.
The communication between ULL and LLL is achieved through
a set of FIFOs that contain both control and data packets.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Signed-off-by: Alberto Escolar Piedras <alpi@oticon.com>
Signed-off-by: Wolfgang Puffitsch <wopu@oticon.com>
Signed-off-by: Morten Priess <mtpr@oticon.com>
