Michael,

> I think he's suggesting that the code terminated prior to the end of
> the playback, because the sound is being played by interrupt-driven
> code outside your program.

Yes, I understood that he vaguely hinted at (but not said) that.   I guess
it shows that that vagueness irked me.

> That's why the actual playback duration could be longer than the
> execution time of your code.

As described in the test at the end of the post you're responding to, it
isn't.

But, if you doubt that test (or its results) than please do advice another
(better?) one.  I'm rather stuck here, so anything is welcome.

You know whats frustrating - besides obviously noone here owning an RPi
3B(+) board and programming C(++) on it that is ?    Although it was for
another, bit more complex bit of example code my first post in this thread
described that above "does it really gets cut off ?" test in points 1) and
2) .  I think it shows that I was aware of that "could keep playing after
the program ends" possibility.

Regards
Rudy Wieser

P.s.
This one uses a single buffer.   The "cut off" problem is the same.  As
mentioned by the poster.
https://stackoverflow.com/questions/40399046/how-to-play-short-tones-with-alsa

--- SoupGate-Win32 v1.05