It is inherently racy, and we have a better way to ensure that
we won't autostart the service:
dbus_message_set_auto_start()
So use that.
This commit also adds a missing call to `dbus_pending_call_unref()`
and indirectly fixes a type mismatch (`dbus_bool_t` vs. `bool`)
that was present in `is_dbus_service_running()`.
The DBusError passed to `dbus_set_error_from_message()` must
be initialized, otherwise libdbus aborts:
dbus[129473]: arguments to dbus_set_error_from_message() were incorrect,
assertion "(error) == NULL || !dbus_error_is_set ((error))"
failed in file dbus-message.c line 4043.
This is normally a bug in some application using the D-Bus library.
The transport set volume call may take a long time or never complete, so
make them async to not block main loop.
Also reduce log level to info for the failed volume setting, as this is
something the user can do nothing about.
Do BAP audio location selection properly in SelectProperties, now that
BlueZ provides the supported locations there. Remove a previous
workaround.
The audio location in SelectProperties determines the audio channel
allocation, which determines the channel positions.
The duplex polling issue was due to spa_loop_add_source failing
when source and sink were both using the same fd. We now dup, so the
issue no longer exists.
Remove the now unnecessary workaround, and check the return values from
spa_add_source.
Don't emit node error for A2DP duplex sink channel, or when BAP server.
These can occur under normal conditions (remote side suspends
transport), and are not errors.
When a sink contributing to an ISO CIG does not have data, output
silence for it, as long as at least one sink in the CIG is running.
Only if writes to sockets fail, pause all streams to reset
synchronization.
This way we write exactly the same number of packets for each CIS at the
same time, which probably is the best tested configuration in BT
adapters and devices. We also don't then have to pause output if some
sinks are not running or miss their timing, as we generate silence on
the fly.
When using iso-io, have it initialize the codec instance, and have
media-sink uses that instance, so that silence and actual audio are
encoded with the same codec.
Use the "high-reliability" values for QoS parameters instead of the
low-latency ones. Under some condition BlueZ does not pass on the
endpoint QoS values to us, in which case we may end up selecting bad
latency.
Determine correctly if we are resampling, and have the associated delay.
Add off-by-one sample adjustment to the resampling delay, which seems to
correctly align the resampled audio with non-resampled.
Resynchronize ISO streams on playback (re)start, so the stream positions
are aligned immediately. This is better than relying on rate matching
to correct any offsets.
Emit BAP device set nodes, which the session manager can use to combine
the sinks/sources of a device set to a single sink/source.
Emit the actual sinks/sources with media.class=.../Internal to hide them
from pipewire-pulse.
Add separate device set routes to the set leader device. Other routes
of the set members will be marked as unavailable when the set is active.
Accordingly, return failure for attempts to set these unavailable
routes, so that volumes etc. of the "internal" nodes are only controlled
via the device set route.
For ISO server sockets, the QOS struct from getsockopt contains values
with different meaning from ISO client socket. Get the values via DBus
instead, which is right in both cases.
If BAP codec configuration is mono channel with unspecified location,
set the channel position from transport location.
This in principle should be set in SelectProperties, but currently BlueZ
doesn't tell us that yet there, so we hack it up later on.
The number of channels is determined by Audio_Channel_Allocation.
One frame block contains all channels. (BAP v1.0.1 Sec. 4.2)
Fix the handling of frame blocks and counting of numbers of channels. We
support and configure only one frame block per packet.
Consider omitted Audio_Channel_Allocation to indicate MONO stream (see
BAP v1.0.1 Sec 4.3.2).
Use the ISO IO helpers to get synchronized BAP output, and rate match to
the ISO schedule.
The rate matching is necessary, since the driver may be ticking at a
corrected rate, different from the ISO interval rate.
Add factored out helper for ISO socket I/O.
ISO sockets need synchronization of writes and audio position for
different stream fds in the same isochronous group, and it's easier to
separate out the part that coordinates it.
Avoiding unnecessary release + reacquire when nodes restart makes sense
for all transport types. Do timed releases for all transport types, not
only SCO.
Do transport release synchronously for simplicity. BlueZ handles
releasing while acquire is pending, but acquire while release is pending
would fail the acquire.
Otherwise we need to maintain an operation chain to handle trying to
acquire/release while the other operation is pending. This makes things
complex with little gained, as releases generally don't block for a long
time.
Drivers should only read the target_ values in the timeout, update the
timeout with the new duration and then update the position.
For the position we simply need to add the previous duration to the
position and then set the new duration + rate.
Otherwise, everything else should read the duration/rate and not use
the target_ values.
Make BAP nodes align the first sample of their packets at multiples of
the ISO interval, counted in the shared graph sample position. This
skips a few samples (< 10ms) at the start of playback to ensure the
alignment.
Since the sinks align their flush timing to the graph time, this also
results to them sending packets corresponding to the same graph time at
the same real time instants.
Due to packet queues in kernel/controller, the playback may still be off
by multiples of packets. Kernel changes are needed to address that part.
This works towards making BAP left and right channels to be
synchronized in TWS headsets, where the two earpieces currently appear
as different devices.
If transport goes into error state too often, fail instead of trying to
acquire it again.
This avoids getting into a tight acquire->fail->reacquire loop.
