[pulseaudio-discuss] [PATCH] switch-on-port-available: check and change default sink or source

[Hui Wang]

On 04/29/2017 09:04 PM, Tanu Kaskinen wrote:
> On Fri, 2017-04-28 at 12:21 +0800, Hui Wang wrote:
>> Hello Tanu,
>>
>> Could you please take a look at this patch, you are the maintainer and
>> recently contributed couple of commits to
>> module-switch-on-port-available.c. :-)
> Well, I'm trying to concentrate on preparing for the release, so I
> don't do patch reviews much at the moment. Now that I read the commit
> message, I have some comments, though.
>
>> On 04/27/2017 11:20 AM, Hui Wang wrote:
>>> Suppose your machine has two sound cards as below:
>>>
>>> Card#0(HDA INTEL HDMI)-> Sink#0(hdmi-stereo)->hdmi-output(priority: 5900)
>>> Card#1(HDA INTEL PCH)->Sink#1(analog-stereo)->headphones(priority: 9000)
>>>
>>> If neither hdmi cable nor headphone plug into the machine, the default
>>> sink will randomly be Sink#0 or Sink#1, let us assume it is Sink#1,
>>> then users plug hdmi cable into the machine, the port hdmi-output will
>>> change to the state PA_AVAILABLE_YES, so the Sink#0 has a port with
>>> state YES, while the Sink#1 still has a port with state NO, in this
>>> situation it is reasonable to change the default_sink to Sink#0, but
>>> current code can't do that.
> This problem should be fixed by these two patches:
> https://patchwork.freedesktop.org/patch/139179/
> https://patchwork.freedesktop.org/patch/139178/
>
> The patches have been reviewed, but I haven't yet pushed them. I'm not
> sure if they'll be in the next release or not (I need to ask Georg and
> Arun if they want to grant a freeze exception for these patches).
>
>>> Let us suppose another situation, both hdmi cable and headphone are
>>> plugged into the machine, and the Sink#0 is the default sink, if users
>>> unplug the hdmi cable, the port hdmi-output is changed to NO while
>>> the port headphone is still kept YES, in this situation it is
>>> reasonable to switch the default_sink to Sink#1, but current code
>>> can't do that.
> This should be fixed too by those two patches mentioned above.
>
Good to know that, look forward to those patches being merged to the 
mainline. :-)

Thank you for your comment and thank you for your patches.

Cheers,
Hui.