This function is sparsely documented and a look at the bluez-alsa
source code shows that implementations make undocumented assumptions
on the `struct pollfd` array parameter which can lead to strange
effects.

This function implements the "group" feature on top of the
Database::Visit() method.

Since we switched from autotools to Meson in commit
94592c14, we don't need to include
`config.h` early to properly enable large file support.  Meson passes
the required macros on the compiler command line instead of defining
them in `config.h`.

This means we can include `config.h` at any time, whenever we want to
check its macros, and there are no ordering constraints.

This compile-time check got lost during the Meson transition.

This version should have been incremented in 0.21.2 because this
version added new search operators.  In 0.21.4, we'll have more
protocol bug fixes which may be important for clients to know.

Works around a problem where MPD goes into a busy loop because
snd_pcm_drain() always returns `-EAGAIN` without making any progress
(fixes #425).

This problem was triggered by snd_pcm_drain() after snd_pcm_cancel()
and snd_pcm_prepare(), but without submitting any data with
snd_pcm_writei().

I believe this is a kernel bug: in non-blocking mode, the kernel's
snd_pcm_drain() function returns early.  In this mode, it only checks
whether snd_pcm_drain_done() has been called already, but
snd_pcm_drain_done() is never called if no data was submitted.

In blocking mode, the following `for` loop detects this condition, so
snd_pcm_drain_done() is not necessary, but without this extra check,
we get `-EAGAIN` forever.

test/run_storage.cxx depends on EventThread/EventLoop from libevent.a.
Depend on it explicitly. This addresses build failure with
-Dtest=true -Dcurl=disabled -Ddbus=disabled

This fixes a problem which caused a failure with snd_pcm_writei()
because snd_pcm_drain() had already been called in the previous
iteration.  This commit makes sure that snd_pcm_drain() is only called
after the final snd_pcm_writei() call.

This fixes discarded samples at the end of playback.

When a playback error has occurred, MPD would never recover until one
restarts MPD.

MPD's default is 100ms, which is too long for the real-time I/O
thread.  The OutputThread has 100us, but the real-time I/O thread
might have tighter deadlines.

This change has currently no effect (I believe), because nobody uses
timers on the RTIO thread.

CancelInternal() doesn't need to be protected because it is called
synchronously from Cancel().

Fixes a theoretical race condition which could occur in Drain() (but
was extremely unlikely).

If our `ring_buffer` is smaller than the ALSA-PCM buffer (if the
latter has more than the 4 periods we allocate), it can happen that
the start threshold is crossed and ALSA switches to
`SND_PCM_STATE_RUNNING`, but the `ring_buffer` is empty.  In this
case, MPDD will generate silence, even though the ALSA-PCM buffer has
enough data.  This causes stuttering (#420).

This commit amends an older workaround for a similar problem (commit
e08598e7) by adding a snd_pcm_avail()
check, and only generate silence if there is less than one period of
data in the ALSA-PCM buffer.

Fixes #420

The method Cancel() assumes that the `period_buffer` must be empty
when `active==false`, but that is not the case when Play() fails.

Of course the assertion in Cancel() is not 100% correct, but I decided
to rather fix this in LockCaughtError() because the `period_buffer`
should only be accessed from within the RTIO thread, and this is the
only code path where `active` can be set to `false` with a non-empty
`period_buffer`.

Fixes #423

This implements real error handling, and avoids calling
CancelInternal() from this code path.

alsa-lib doesn't set errno, it returns errors as negative integers.

This check was added 9 years ago in commit
4dc25d39 to work around a dmix bug
which I assume has been fixed long ago.

Removing this fixes another corner case: if draining is requested
before the start threshold is reached, the PCM is still in
SND_PCM_STATE_PREPARED but not yet SND_PCM_STATE_RUNNING, which means
the submitted data will never be played.  This corner case is
realistic when playing songs shorter than the ALSA buffer (if the
buffer is very large).

This fixes a corner case which has probably never occurred and
probably never will: if Cancel() is called, and then Play() followed
by Drain(), the plugin should really play that data.  However
currently, this never happens, because snd_pcm_prepare() is never
called.