Re: [PATCH] ASoC: wm8962: add .set_tdm_slot callback function

[Charles Keepax <ckeepax@opensource.cirrus.com>]

On Tue, Jan 27, 2026 at 09:01:02PM +0800, Shengjiu Wang wrote:
> On Tue, Jan 27, 2026 at 7:03 PM Charles Keepax
> <ckeepax@opensource.cirrus.com> wrote:
> >
> > On Tue, Jan 27, 2026 at 06:33:26PM +0800, Shengjiu Wang wrote:
> > > The slot_width can be different with the params_width(), for example,
> > > DSP_A mode, slot_width = 32, but data format is S16_LE, if the word
> > > length is configured to be 16, there is no sound on the right speaker.
> > >
> > > So add .set_tdm_slot() callback function to configure the slot_width and
> > > update the word length according to slot_width in hw_params().
> > >
> > > Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
> > > ---
> > > +static int wm8962_set_tdm_slot(struct snd_soc_dai *dai, unsigned int tx_mask,
> > > +                            unsigned int rx_mask, int slots, int slot_width)
> > > +{
> > > +     struct snd_soc_component *component = dai->component;
> > > +     struct wm8962_priv *wm8962 = snd_soc_component_get_drvdata(component);
> > > +
> > > +     if (slots <= 0 || slot_width <= 0)
> > > +             return 0;
> >
> > You probably shouldn't bail out here, how does one disable TDM if
> > it is no longer required?
> 
> Should we add some checks for the slots and slot_width here?
> or just remove these two lines?
> 

I think its ok to just remove these, if you call set_tdm_slot
with weird numbers that is probably a bug in the calling code,
but not super picky either way.

Thanks,
Charles