On Mon, Jan 02, 2017 at 12:44:33PM +0530, jeeja.kp@xxxxxxxxx wrote:
> +static int ssp_set_clk(struct snd_soc_dapm_widget *w,
> + struct snd_kcontrol *k, int event)
> +{
> + struct snd_soc_dai *cpu_dai = NULL;
> +
> + cpu_dai = skl_get_dai_widget(w);
> + if (!cpu_dai)
> + return -EIO;
> +
> + /* Enable/Disable the SSP clk */
> + if (SND_SOC_DAPM_EVENT_ON(event))
> + return snd_soc_dai_set_tristate(cpu_dai, 0);
> + else
> + return snd_soc_dai_set_tristate(cpu_dai, 1);
> +}
Erk, no - you definitely want to just move the whole thing to DAPM I
think. We shoudn't be manually fiddling with the tristating from driver
code when the core will also try to manage tristating, at some point
this widget and the core will inevitably end up fighting with each
other.
Attachment:
signature.asc
Description: PGP signature
_______________________________________________ Alsa-devel mailing list Alsa-devel@xxxxxxxxxxxxxxxx http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
