RE: [PATCH 1/2] ASoC: rt5682s: Add driver for ALC5682I-VS codec

["Derek [方德義]" <derek.fang@realtek.com>]

> Subject: Re: [PATCH 1/2] ASoC: rt5682s: Add driver for ALC5682I-VS codec
> 
> On 8/26/2021 4:09 PM, derek.fang@realtek.com wrote:
> > +
> > +static int is_sys_clk_from_plla(struct snd_soc_dapm_widget *w,
> > +		struct snd_soc_dapm_widget *sink)
> > +{
> > +	struct snd_soc_component *component =
> snd_soc_dapm_to_component(w->dapm);
> > +	struct rt5682s_priv *rt5682s =
> snd_soc_component_get_drvdata(component);
> > +
> > +	if ((rt5682s->sysclk_src == RT5682S_CLK_SRC_PLL1) ||
> > +	    (rt5682s->sysclk_src == RT5682S_CLK_SRC_PLL2 &&
> > +	     rt5682s->pll_comb == USE_PLLAB))
> 
> 
> What about rt5682s->pll_comb == USE_PLLA?

Thanks for review.

My design thought of this check function is that
pll_comb would be set via snd_soc_dai_set_pll and 
sysclk_src would be set via snd_soc_dai_set_sysclk.

So I want to make sure the two settings are both correct first
and then power PLLA/B.
Otherwise, not power, although the chance of mistake is rare.

   ___   +-->[PLLA]-+-------------------> PLL1 out
  |Clk|  |          |    ___
  |Mux|--|          +-->|Clk|
  |___|  |	        |Mux|-->[PLLB]--> PLL2 out
	 +------------->|___|

> 
> > +		return 1;
> > +
> > +	return 0;
> > +}
> > +
> > +static int is_sys_clk_from_pllb(struct snd_soc_dapm_widget *w,
> > +		struct snd_soc_dapm_widget *sink)
> > +{
> > +	struct snd_soc_component *component =
> snd_soc_dapm_to_component(w->dapm);
> > +	struct rt5682s_priv *rt5682s =
> snd_soc_component_get_drvdata(component);
> > +
> > +	if (rt5682s->sysclk_src == RT5682S_CLK_SRC_PLL2)
> 
> 
> Should we also test rt5682s->pll_comb here?
> 
> 
> > +		return 1;
> > +
> > +	return 0;
> > +}
> >
> ------Please consider the environment before printing this e-mail.