[alsa-devel] [PATCH 44/54 v2] ASoC: switch over to use snd_soc_register_component() on sh4 siu

[Stephen Warren]

On 03/21/2013 06:21 PM, Kuninori Morimoto wrote:
> 
> Hi Stephen
> 
> Thank you for your checking
> 
> 
>>>  static struct snd_soc_dai_link migor_dai = {
>>>  	.name = "wm8978",
>>>  	.stream_name = "WM8978",
>>> -	.cpu_dai_name = "siu-i2s-dai",
>>> +	.cpu_dai_name = "siu-pcm-audio",
>>
>> I'm not sure what that change is for. The need for it should be
>> described in the commit description.
>>
>>> diff --git a/sound/soc/sh/siu_dai.c b/sound/soc/sh/siu_dai.c
>>
>>> @@ -783,7 +787,8 @@ static int siu_probe(struct platform_device *pdev)
>>>  	dev_set_drvdata(&pdev->dev, info);
>>>  
>>>  	/* register using ARRAY version so we can keep dai name */
>>> -	ret = snd_soc_register_dais(&pdev->dev, &siu_i2s_dai, 1);
>>> +	ret = snd_soc_register_component(&pdev->dev, &siu_i2s_component,
>>> +					 &siu_i2s_dai, 1);
>>
>> That changes from snd_soc_register_dais() to snd_soc_register_dai().
>> Will that be a problem?
> 
> I think .cpu_dai_name will be "siu-i2s-dai" if it used dais(),
> and it will be "siu-pcm-audio" if it used dai()

Ah right, the difference between fmt_single_name() and
fmt_multiple_name() is dev_name() vs. dai_drv->name; I thought it was
just about including the device ID from dev_name(), or chopping it off.
So, I think this is fine.

A description of the change to migor_dai would be useful to explain this
though.