[pulseaudio-discuss] [PATCH 0/2] Suppress the analog-output path when it's redundant

[Alexander E. Patrakov]

2014-08-17 17:52 GMT+06:00 Tanu Kaskinen <tanu.kaskinen at linux.intel.com>:
> The analog-output path should be used only when more specific outputs
> don't exist, so usually it should be suppressed, and currently that
> doesn't happen. One practical problem caused by that is that the
> analog-output port may get chosen during boot if headphones are
> plugged in, because analog-output has higher priority than headphones
> and the speaker port is unavailable[1]. That particular problem could
> maybe be fixed by decreasing the analog-output priority, or by
> tweaking module-switch-on-port-available, but this time I decided to
> fix the problem of analog-output appearing when it's redundant.
>
> [1] https://bugs.freedesktop.org/show_bug.cgi?id=74609
>
> Tanu Kaskinen (2):
>   alsa-mixer: Fix path subset detection
>   alsa-mixer: Ignore some elements in the analog-output path
>
>  src/modules/alsa/alsa-mixer.c                   | 26 +++++++++----------
>  src/modules/alsa/mixer/paths/analog-output.conf | 34 -------------------------
>  2 files changed, 13 insertions(+), 47 deletions(-)

I have looked at and tested both patches on my laptop.

Patch 1/2 gets a full ACK. It makes mixer element handling consistent
with the existing code for jack handling.
Patch 2/2, when applied on top of 1/2, successfully kills the unwanted
Analog output port on my laptop. However, I have not verified that the
set of to-be-removed elements in this patch is indeed minimal and
sufficient. So this patch only gets a "possibly-incomplete ACK".

Maybe it is a good idea to write a test that verifies that the
analog-output path is indeed a subset of all mixer paths that should
kill it?

-- 
Alexander E. Patrakov