Why is it ok to use the sync pool?
Because command complete/status is processed in prio: that means on the
same stack as the `bt_recv()` call from the driver.
Why does it fix the issue?
Because the complete/status event goes into a pool that is guaranteed to
have one free buffer any time `bt_recv()` is not executing.
Since the driver is the one calling bt_recv(), it (hopefully) will
finish one `bt_recv()` before starting another one.
Fixes#78223
Co-authored-by: Aleksander Wasaznik <aleksander.wasaznik@nordicsemi.no>
Signed-off-by: Aleksander Wasaznik <aleksander.wasaznik@nordicsemi.no>
Signed-off-by: Jonathan Rico <jonathan.rico@nordicsemi.no>
`struct acl_data` is used even when Host flow control is not enabled.
It is written to through the `acl(buf)` accessor in `conn.c:hci_acl()`.
Hopefully no netbufs were harmed by that :/
Signed-off-by: Jonathan Rico <jonathan.rico@nordicsemi.no>
We can get rid of the view pool for SDU segments :)
We have to make the code slightly more complex :'(
The basic idea is always giving the original SDU buffer to `conn.c` for it
to pull ACL fragments from.
In order to do this, we need to add the PDU headers just-in-time.
`bt_l2cap_send_pdu()` does not add them before putting the PDU on the queue
anymore. They are added by `l2cap_data_pull()` right before the data leaves
`l2cap.c` for `conn.c`.
We also have to inform `conn.c` "out of band" of the real L2CAP PDU size so
it doesn't fragment across segment boundaries. This oob is the new `length`
parameter to the `.pull()` method.
This is the added complexity mentioned above.
Since SDU segmentation concerns only LE-L2CAP, ISO and Classic L2CAP don't
need this extra logic.
Signed-off-by: Jonathan Rico <jonathan.rico@nordicsemi.no>
Instead of allocating segments/fragments and copying data into them, we
allocate segments as "views" (or slices) into the original buffer.
The view also gives access to the headroom of the original buffer, allowing
lower layers to push their headers.
We choose not to allow multiple views into the same buffer as the headroom
of a view would overlap with the data of the previous view.
We mark a buffer as locked (or "in-view") by temporarily setting its
headroom to zero. This effectively stops create_view because the requested
headroom is not available.
Each layer that does some kind of fragmentation and wants to use views for
that needs to maintain a buffer pool (bufsize 0, count = max views) and a
metadata array (size = max views) for the view mechanism to work.
Maximum number of views: number of parallel buffers from the upper layer,
e.g. number of L2CAP channels for L2CAP segmentation or number of ACL
connections for HCI fragmentation.
Reason for the change:
1. prevent deadlocks or (ATT/SMP) requests timing out
2. save time (zero-copy)
3. save memory (gets rid of frag pools)
L2CAP CoC: would either allocate from the `alloc_seg` application callback,
or worse _steal_ from the same pool, or allocate from the global ACL pool.
Conn/HCI: would either allocate from `frag_pool` or the global ACL pool.
Signed-off-by: Jonathan Rico <jonathan.rico@nordicsemi.no>
Co-authored-by: Aleksander Wasaznik <aleksander.wasaznik@nordicsemi.no>
Refactor only. The surrounding ifdefs are intentionally not changed in
this patch. They will be in the near future.
Rename the pool and generalize the documentation to allow using this
pool for other events that fit the same criteria. This pool can be used
for any buffer that is processed synchronously, without negatively
affecting 'num complete' messages. E.g. 'cmd complete/status' can be put
in this pool already.
We will be working towards making the host process all event buffers
synchronously. This is because events have no dedicated flow control,
and discarding events in the driver without informing the host creates
problems. Discarding should instead happen in the host higher layers
when unavoidable.
Signed-off-by: Aleksander Wasaznik <aleksander.wasaznik@nordicsemi.no>
Add 2 new Kconfig promptless options that are shorthand
for whether the ISO configuration can support RX and TX.
This also applies these new options as guards for existing
and missing code pieces.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
`bt_buf_get_cmd_complete` is broken due to
https://github.com/zephyrproject-rtos/zephyr/issues/64158, and fixing it
would require changing its signature and put even more complexity into
the HCI drivers, as it would require the drivers to perform an even
deeper peek into the event in order to observe the opcode.
Instead of the above, this patch removes the use of
`bt_buf_get_cmd_complete` and adds logic to allow the host to accept
command complete events in normal event buffers.
The above means performing a copy into the destination buffer, which is
the original command buffer. This is a small inefficiency for now, but
we should strive to redesign the host into a streaming architecture as
much as possible and handle events immediately instead of retaining
buffers.
This fixes https://github.com/zephyrproject-rtos/zephyr/issues/64158:
Like all command completed events, the completion event for
`BT_HCI_OP_HOST_NUM_COMPLETED_PACKETS` is now placed in normal event
buffers. The the logic where the host discards this event is already
present. Since it's discarded, it will not interfere with the logic
around `bt_dev.cmd_send`.
Signed-off-by: Aleksander Wasaznik <aleksander.wasaznik@nordicsemi.no>
This to ensure we don't fail to send a response and never get an ATT
TIMEOUT due to ACL TX buffer starvation caused by other users of the stack.
Signed-off-by: Jonathan Rico <jonathan.rico@nordicsemi.no>
We should clear the bt_dev.sent_cmd pointer after using it to allocate a
new HCI event buffer in the bt_buf_get_cmd_complete() function.
Otherwise, there is a risk of reusing the same stored net_buf for
multiple consecutive HCI events in case the controller sents duplicate
or invalid event packets.
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
The `bluetooth/common/log.h` and `bluetooth/common/log.c` files have been
removed. Files that were using them have been updated to use
`zephyr/logging/log.h` instead.
Those replacement have been done consequently:
- `/BT_DBG/LOG_DBG/`
- `/BT_ERR/LOG_ERR/`
- `/BT_WARN/LOG_WRN/`
- `/BT_INFO/LOG_INF/`
- `/BT_HEXDUMP_DBG/LOG_HEXDUMP_DBG/`
- `/BT_DBG_OBJ_ID/LOG_DBG_OBJ_ID/`
Also, some files were relying on the `common/log.h` include to include
`zephyr/bluetooth/hci.h`, in those cases the include of `hci.h` has
been added.
For files that were including `common/log.h` but not using any logs,
the include has been removed and not replaced.
Signed-off-by: Théo Battrel <theo.battrel@nordicsemi.no>
CONFIG_BT_BUF_EVT_DISCARDABLE_COUNT is defined by default and the
value ranges from 1 to 255.
So, checking if this parameter is defined seems to be redundant
Signed-off-by: Ahmed Moheb <ahmed.moheb@nordicsemi.no>
In order to bring consistency in-tree, migrate all subsystems code to
the new prefix <zephyr/...>. Note that the conversion has been scripted,
refer to zephyrproject-rtos#45388 for more details.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
The BT_BUF_RX_SIZE did not take the CONFIG_BT_ISO_RX_MTU
into account. Add BT_BUF_ISO_RX_SIZE which
depend on CONFIG_BT_ISO and use that for the
BT_BUF_RX_SIZE macro.
Furthermore, move the BT_BUF_RX_COUNT macro definitions
into buf.h and update that to account for ISO RX as well.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Update the macro prototype to explicitly require the length of the
desired user data. Update all in-tree usage of this macro.
Signed-off-by: Jordan Yates <jordan.yates@data61.csiro.au>
Splitt he ISO broadcast config into broadcaster and
sync receiver. This will allow a device that only
wants to one of the roles to have a much more optimized
configuration.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Change so that num_completed_packets event handling is also
enabled for broadcast ISO only builds. This is because sending
data on a broadcast ISO still generates this event.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Refactor and simplify the bluetooth buffer configurations to improve the
easy of configurations and eliminate invalid ones.
By moving configurations out of host and controller specific
configurations and into a common one it becomes easier to configure
the host and controller separately as the same configurations can be
used as would be for a combined build.
All HCI configurations are now given exluding the matching HCI header,
which eases the configuration as the application don't have to know the
different header sizes.
The BT_RX_BUF_LEN is split into ACL and Event, as well as the suprising
use of Command size.
BT_L2CAP_RX_MTU is removed as the stack does not support reassembling of
HCI ACL data to larger L2CAP PDUs. The application will have to set
ACL RX size and account for the L2CAP PDU header itself.
BT_EATT_RX_MTU was removed as it is only used for setting a different
default value for another option which leads to the stuck kconfig symbol
problem.
The configurations can be updated according to the table below:
** New configuration | ** Old configuration
All configurations
BT_BUF_ACL_RX_SIZE | BT_L2CAP_RX_MTU + 4
BT_BUF_ACL_RX_SIZE | BT_RX_BUF_LEN - 4
BT_BUF_EVT_RX_SIZE | BT_RX_BUF_LEN - 2
BT_BUF_CMD_TX_SIZE | BT_RX_BUF_LEN - 3
BT_BUF_CMD_TX_COUNT | BT_HCI_CMD_COUNT
BT_BUF_EVT_RX_COUNT | BT_RX_BUF_COUNT
BT_BUF_ACL_RX_COUNT | BT_RX_BUF_COUNT
BT_BUF_ACL_RX_COUNT | BT_ACL_RX_COUNT
BT_BUF_EVT_DISCARDABLE_SIZE | BT_DISCARDABLE_BUF_SIZE - 2
BT_BUF_EVT_DISCARDABLE_COUNT | BT_DISCARDABLE_BUF_COUNT
Controller-build
BT_BUF_ACL_TX_SIZE | BT_CTLR_TX_BUFFERS_SIZE
BT_BUF_ACL_TX_COUNT | BT_CTLR_TX_BUFFER
HCI-bridge
BT_BUF_ACL_TX_SIZE | BT_HCI_ACL_DATA_SIZE
BT_BUF_ACL_TX_COUNT | 6
Fixed invalid configurations setting either BT_L2CAP_RX_MTU or
BT_CTLR_DATA_LENGTH_MAX larger than BT_RX_BUF_LEN could lead to buffer
overruns.
Fix advertising report max data length calculation.
This always used the BT_DISCARDABLE_BUF_SIZE macro but this feature
can be turned off and advertising reports will be allocated from the RX
buffer in that case. Also controller-build does not have this buffer
(in hci_raw.c). Also the wrong HCI header was used in the calculation,
HCI event header should have been used instead of HCI ACL header.
Signed-off-by: Joakim Andersson <joakim.andersson@nordicsemi.no>
Add macros for calculating buffer sizes, accounting for the various
L2CAP and HCI headers needed as well as the reserved bytes needed.
Signed-off-by: Joakim Andersson <joakim.andersson@nordicsemi.no>
ISO is a building block for BT_AUDIO but it is not only
useful for AUDIO, and as such should be possible to
enable without enabling BT_AUDIO.
This commit moves iso.c and iso_internal.h to the
host directory (from host/audio) and removes
the CMakeLists.txt.
The /audio directory is left intact for the Kconfig options
it provides, and as a directory for future BLE Audio
content.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Refactor out implementation of functions defined in buf.h to its own
source file buf.c
Signed-off-by: Joakim Andersson <joakim.andersson@nordicsemi.no>
