When a device is removed, close() will error. Warn about this but
continue closing our other things including marking the device
as closed so that we don't try to close it again later.
Fixes#413
Don't change the resampler delay value, we need it to make sure
we keep samples around for the next round. With small period sizes,
we set the delay to 0 and mess up the resampler and cause dropouts
and clicking.
Fixes#287
Log device name and stream direction
Don't log xrun errors in alsa, we signal the xrun signal, which should
take care of logging if any.
Log quantum in xrun.
ratelimit xrun messages.
We only set the OUT flag when we put the buffer on an io area and
might need to recycle later.
Before placing the output buffer into the io area, recycle any
buffer that might be left in there.
Only emit the ready callback when we have some data queued.
If we get an error after receovering, ignore it and assume we filled
the buffer completely. We need to do this or otherwise we stop the
timer and audio stops.
This allows the upstream node to put buffers back to its pool in case
they were left around in the ready list locally when the alsa-pcm-sink
was last paused.
Fixes#203
Add some padding, tweak some padding
Remove count in the clock, it's useless
For video frames we will want to use metadata to place
this on individual buffers.
Add a clock name to the clock, remove the old api/clock_id. This makes
it easier to add descriptive names
Place the alsa card number in the clock name.
Check the clock name of the master clock and if it matches our own
clock, disable rate matching.
Add flags to the rate match io area
Add flag to activate/deactivate rate match
Set active flag in rate match when slaved
Update rate before starting resample
