The Bluetooth Low Energy MIDI code added a few legacy function declarations
that fail when building with -Werror=strict-prototypes. The fix is same as
before: add a void to the empty function argument list.
Signed-off-by: Niklāvs Koļesņikovs <89q1r14hd@relay.firefox.com>
Use stdalign.h instead of union.
Fix some refcounting and return values.
Fail early in add_filters.
Minor style cleanups.
Less magical spa_dbus_async_call.
As server, put node nick in GATT Characteristic User Description, so
that clients have some information which endpoint is which in case we
have multiple. This is not in BLE MIDI standard, but cannot hurt.
As client, make use of that information, if present.
Implement BLE MIDI node, providing an input and output port. MIDI data
received is normalized and produced on output port. Data from input
port is converted to BLE MIDI and sent to the device.
Event timestamps come from device clock, whose offset to wall clock we
determine by correlating packet reception times with their timestamps.
The jitter in packet reception is likely much larger than device clock
error, so we average over that.
Make a real debug context with a log function and move it to a new file.
This way we don't need to redefine a macro.
Make a new context for debugging to a log file. Make new functions to
debug to a log file.
Move the stringbuffer to string utils.
Integrate file/line/func and topics into the debug log.
We can remove some more things from the pipewire log_object function and
also add support for topics.
Add new spa_debugc_ funnctions that take a context. The user should also
redefine the spa_debugc macro to handle the context.
Use this to let some plugins log the pod and format to the log without
using the global logger.
Also use this to remove our custom pod logger function by reusing the
spa one with a custom context.
BlueZ may send multiple PACs in the capabilities delimited by zero LTV.
Handle this case by selecting the "best" one.
The configuration size may also for BAP generally be different from PAC
size.
When reading the timerfd gives an error, we should return right away
because the timeout did not happen.
If we change the timerfd timeout before reading it, we can get -EAGAIN.
Don't log an error in that case but wait for the new timeout.
The free buffer check must happen before writing to check for leftover data in buffers. In case data is left over from previous submission, bitpool mustn't be increased.
Also improved logging by adding bitpool to the log message.
AVDTP in principle allows 62 endpoints, but in practice it appears some
devices (Samsung Galaxy Buds2 Pro, Redmi Buds 3 Lite, probably others)
fail to connect A2DP when the number is somewhere above 24. A2DP
connection works when initiated from the Central, but not when the
device itself does it, so these devices are not fully broken. We should
reduce the number of registered A2DP endpoints to avoid running into
problems with such broken devices.
Some of our source codecs are the same actual codec with the same
configuration, and don't need separate source endpoints.
Allow codecs to not have a registered endpoint (fill_caps == NULL), and
tolerate codecs with the same endpoint name. In codec switch, keep
track separately which of the codecs with the same endpoint name the
local endpoint is currently associated with.
The graph cycle goes: driver timeout -> process output nodes -> process
driver node. Hence, driver should produce buffers in the timeout,
otherwise there's one quantum extra latency.
Make the bluez5 media/sco sources as drivers put a buffer to io before
indicating ready, and as follower do it in process. Also make checks if
io == NULL, and don't set io->status to HAVE_DATA unless there really is
a buffer ready.
Introduce `unregister_media_endpoint()` to unregister the specified
media endpoint object from dbus. Moreover, move the logic that decides
whether or not a particular codec should be registered for a given
direction into `endpoint_should_be_registered()` and use that in
both `(un)register_media_endpoint()`.
