[alsa-devel] [PATCH v2 2/9] ASoC: sun8i-codec: Don't hardcode BCLK / LRCK ratio
Vasily Khoruzhick
anarsoul at gmail.com
Mon Oct 15 17:14:25 CEST 2018
On Monday, October 15, 2018 12:07:14 AM PDT Maxime Ripard wrote:
> On Sat, Oct 13, 2018 at 09:53:22AM -0700, Vasily Khoruzhick wrote:
> > BCLK / LRCK ratio should be sample size * channels, but it was
> > hardcoded to 32 (0x1 is 32 as per A33 and A64 datasheets).
> >
> > Calculate it basing on sample size and number of channels.
> >
> > Signed-off-by: Vasily Khoruzhick <anarsoul at gmail.com>
> > ---
> >
> > sound/soc/sunxi/sun8i-codec.c | 35 ++++++++++++++++++++++++++++++-----
> > 1 file changed, 30 insertions(+), 5 deletions(-)
> >
> > diff --git a/sound/soc/sunxi/sun8i-codec.c b/sound/soc/sunxi/sun8i-codec.c
> > index fb37dd927e33..e681e194ad4c 100644
> > --- a/sound/soc/sunxi/sun8i-codec.c
> > +++ b/sound/soc/sunxi/sun8i-codec.c
> > @@ -52,7 +52,6 @@
> >
> > #define SUN8I_AIF1CLK_CTRL_AIF1_LRCK_INV 13
> > #define SUN8I_AIF1CLK_CTRL_AIF1_BCLK_DIV 9
> > #define SUN8I_AIF1CLK_CTRL_AIF1_LRCK_DIV 6
> >
> > -#define SUN8I_AIF1CLK_CTRL_AIF1_LRCK_DIV_16 (1 << 6)
> >
> > #define SUN8I_AIF1CLK_CTRL_AIF1_WORD_SIZ 4
> > #define SUN8I_AIF1CLK_CTRL_AIF1_WORD_SIZ_16 (1 << 4)
> > #define SUN8I_AIF1CLK_CTRL_AIF1_DATA_FMT 2
> >
> > @@ -257,8 +256,8 @@ static int sun8i_set_fmt(struct snd_soc_dai *dai,
> > unsigned int fmt)>
> > }
> >
> > struct sun8i_codec_clk_div {
> >
> > - u8 div;
> > - u8 val;
> > + unsigned int div;
> > + u8 val;
> >
> > };
> >
> > static const struct sun8i_codec_clk_div sun8i_codec_bclk_div[] = {
> >
> > @@ -300,12 +299,33 @@ static u8 sun8i_codec_get_bclk_div(struct
> > sun8i_codec *scodec,>
> > return best_val;
> >
> > }
> >
> > +static const struct sun8i_codec_clk_div sun8i_codec_lrck_div[] = {
> > + { .div = 16, .val = 0 },
> > + { .div = 32, .val = 1 },
> > + { .div = 64, .val = 2 },
> > + { .div = 128, .val = 3 },
> > + { .div = 256, .val = 4 },
> > +};
>
> That's log2(div) - 4, right? There's no need to store the table if so.
OK, will fix in v3
> Maxime
More information about the Alsa-devel
mailing list