05-21-2020, 03:33 PM
Just finished the real git-bisect process.
To cut to the chase, here's the output (where, in our context, "git bisect good" means the last of the succession of bisect builds led to a working mpd/bluez-alsa setup and "first bad commit" identifies the first commit which leads to a non-working mpd/bluez-alsa setup):
I'll set up a clean-room Raspbian system to see what I can see.
Regards,
Kent
To cut to the chase, here's the output (where, in our context, "git bisect good" means the last of the succession of bisect builds led to a working mpd/bluez-alsa setup and "first bad commit" identifies the first commit which leads to a non-working mpd/bluez-alsa setup):
Code:
pi@moode3b:~/MPD $ git bisect good
c926021599c77b3011beec856cad535a5bedb1af is the first bad commit
commit c926021599c77b3011beec856cad535a5bedb1af
Author: Max Kellermann <max@musicpd.org>
Date: Fri Jun 28 08:27:48 2019 +0200
output/alsa: always redo DrainInternal() after writing
Draining isn't finished just because the period_buffer has run empty.
It is only finished after snd_pcm_drain() has succeeded.
:100644 100644 1d066d2901e48a7ca1b565de944f3b3849164a0d 542cd5d048f0609224e9de363bcabd3d7d0c2c97 M NEWS
:040000 040000 987fdf549bf480b76e3e0078313222569bb0e559 5ea2319f77527a0ffebb364c83f3e521994d1672 M src
I'll set up a clean-room Raspbian system to see what I can see.
Regards,
Kent