Re: [PATCH] ASoC: rt5640: change widget sequence for depop

[Stephen Warren <swarren@wwwdotorg.org>]

On 08/19/2013 08:36 PM, Bard Liao wrote:
>> -----Original Message-----
>> From: Stephen Warren [mailto:swarren@wwwdotorg.org]
>> Sent: Tuesday, August 20, 2013 1:30 AM
>> To: Bard Liao
>> Cc: broonie@kernel.org; lgirdwood@gmail.com; alsa-devel@alsa-project.org;
>> Flove; Oder Chiou; swarren@nvidia.com
>> Subject: Re: [PATCH] ASoC: rt5640: change widget sequence for depop
>>
>> On 08/19/2013 06:18 AM, bardliao@realtek.com wrote:
>>> From: Bard Liao <bardliao@realtek.com>
>>>
>>> Add mute/unmute control in widget event and modify the power on/off
>> sequence to avoid pop noise.
>>
>> This version breaks both headphone and speaker output. I see the following
>> issues:
>>
>> 1)
>>
>> After a reboot, if all of "{HP,Speaker} [LR] Playback Switch" are on/unmuted,
>> and I start playback, then something sets all those controls to off/muted. I
>> suspect this is related to function rt5640_hp_event() twiddling the same
>> RT5640_[LR]_MUTE bits that the hp_[lr]_enable_control controls twiddle.
>>
>> 2)
>>
>> If playback is running, and (1) has caused all output to be muted, and I then
>> run alsamixer and enable/unmute "HP L Playback Switch", I hear sound on
>> both the L and R headphone channels for a very brief time, then output
>> switches to the L channel only. Similar happens if I unmute "HP R Playback
>> Switch" first. Unmuting both switches does produce stereo sound as expected.
>>
>> I don't know if this also happens for the speaker output, since the board is too
>> far away from me to easily tell which speakers are playing.
> 
> I think you didn't apply Lars-Peter's patch as below.
> --- a/sound/soc/soc-dapm.c
> +++ b/sound/soc/soc-dapm.c
> @@ -229,6 +229,8 @@ static int dapm_kcontrol_data_alloc(struct
>   			template.id = snd_soc_dapm_kcontrol;
>   			template.name = kcontrol->id.name;
> 
> +			data->value = template.on_val;
> +
>   			data->widget = snd_soc_dapm_new_control(widget->dapm,
>   				&template);
>   			if (!data->widget) {
> 
> It should be able to solve this issue.
> If I remove this patch I can reproduce the issue.
> Sorry for not informing you about that.

As far as I can tell, Lars-Peter hasn't actually sent such a patch to
the list; he simply mentioned this change inline in the middle of a
thread following an earlier version of this patch.

Lars-Peter, are you planning on sending the change above as a patch for
Mark to apply?

With that patch also applied, symptom (1) I mentioned above is indeed
fixed. However, symptom is still observed.