Keep track or the follower state and do the sync from the driver
wakeup.
This improves sync between devices because the sync no longer
depends on when the node was woken up in the graph and scheduled. We
would have been able to handle this difference by using the htimestamp
but that doesn't seen to work reliably on some hardware.
This is also essential for the interrupt based scheduling.
Split the spa_alsa_start() in a prepare and start function. Prepare
will then also reset state and prefill the playback buffer with silence.
This makes it possible to reuse the prepare function in recover().
The threshold depends on the samplerate and even when the graph rate or
quantum didn't change, the driver could be opened in a different
samplerate, which would cause the timeouts to be wrong and cause xruns.
Version 1 does not seem to be packaged in many distros and so they would
need to revert the patch or disable AEC. Enabling both allows for things
to move forwards gracefully.
Make supported codec checks to use profiles, not "is-a-sink" flag, to
determine which codecs can be used.
Fixes bluez5-device checking only source profiles, even when the local
device is only a sink.
port_set_io with SPA_IO_Buffer can be used to enable/disable a port
when the node is running and so the node should make sure the io update
is synchronized with the processing loop.
Use spa_loop_invoke to make sure the mixers handle the port_io updates
correctly.
Setting buffers or a format also needs the port to be disabled so add
some checks for this in the mixers.
Don't just forward the tag and latency events to the follower but let
the audioconvert aggregate and emit the updated tag/latency event
that is then configured on the follower.
When using the DSP mode of the audioconvert, this results in an
accumulated latency/tag from all the DSP ports instead of just
the last DSP port param update.
Put properties with media. prefix in tags in pw-cat.
Instead of watching /dev, use a separate watch for each device.
This is supposed to achieve the same result as the now reverted
88f0dbd6fc ("v4l2: don't set inotify on /dev"):
Doing inotify on /dev is not a good idea because we will be woken up by
a lot of unrelated events.
There is a report of a performance regression on some IO benchmark
because of lock contention within the fsnotify subsystem due to this.
Instead, just watch for attribute changes on the /dev/videoX files
directly. We are only interested in attribute changes, udev should
notify us when the file is added or removed.
Fixes#3439
As part of this, in alsa-udev.c, certain structures and variables referred
to as "device" are renamed to "card". Otherwise, there is ambiguity, since
"device" can mean a udev device, an SPA device, a compress-offload device,
a PCM device etc.
Also, replace "card id" with "card number" to emphasize that these integers
are not actually SPA object IDs.
We're missing the delay in samples plus all of the extra samples of
silence we use to restart the device. This is in the samplerate of
the device.
Convert this to the time domain of the graph before adding it to xrun.
The isinf function returns -1 for negative infinity on glibc, but
the standard guarantees no such behavior (e.g. in C++ it always
returns a bool, on musl libc it's a macro that expands to a bool
expression), saying just that it returns a non-zero value.
This was added in pulseaudio around 15 years ago, and was never
fixed; pipewire then got the code from it. However, we can portably
check against -INFINITY instead (from math.h, already included).
Ref 045c1d6
