Re: [PATCH v3] ASoC: Intel: kbl_rt5663_rt5514_max98927: Split be_hw_params_fixup function

Hi,

I am also ok with Harsha patch. I have checked it on my Eve and it looks ok.

Best regards,

Lukasz

śr., 1 lip 2020 o 19:08 Pierre-Louis Bossart pierre-louis.bossart@linux.intel.com napisał(a):

...

...

...

Tested and the following is something we can use without creating a new

dailink.

...

        struct snd_soc_dai *codec_dai = asoc_rtd_to_codec(rtd, 0);
if (!strcmp(codec_dai->name, KBL_REALTEK_DMIC_CODEC_DAI)) {
        if (params_channels(params) == 2 ||

DMIC_CH(dmic_constraints) == 2)

...

                channels->min = channels->max = 2;
        else
                channels->min = channels->max = 4;
} else {
        rate->min = rate->max = 48000;
        channels->min = channels->max = 2;
        snd_mask_none(fmt);
        snd_mask_set_format(fmt, pcm_fmt);
}

Pierre, thoughts?

thanks Harsha, that looks like what I had in mind, but my earlier question was why we deal with the rates and formats only in the last case?

The speaker codec supported only 16 bit. (Vendor mentioned) For playback on this platform, only 48Khz was used.

ok then, as long as Harsha and Lukasz are aligned I'm fine. Thanks!