Re: [PATCH v6 5/6] ASoC: codecs: wcd938x: add mux control support for hp audio mux

[Johan Hovold <johan@kernel.org>]

On Thu, Apr 03, 2025 at 01:19:57PM +0100, Srinivas Kandagatla wrote:
> On 02/04/2025 10:12, Johan Hovold wrote:
> > On Thu, Mar 27, 2025 at 10:06:32AM +0000, Srinivas Kandagatla wrote:
> >> From: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
> >>
> >> On some platforms to minimise pop and click during switching between
> >> CTIA and OMTP headset an additional HiFi mux is used. Most common
> >> case is that this switch is switched on by default, but on some
> >> platforms this needs a regulator enable.
> >>
> >> move to using mux control to enable both regulator and handle gpios,
> >> deprecate the usage of gpio.
> >>
> >> Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
> >> Tested-by: Christopher Obbard <christopher.obbard@linaro.org>
> > 
> >> @@ -3261,11 +3276,26 @@ static int wcd938x_populate_dt_data(struct wcd938x_priv *wcd938x, struct device
> >>   		return dev_err_probe(dev, wcd938x->reset_gpio,
> >>   				     "Failed to get reset gpio\n");
> >>   
> >> -	wcd938x->us_euro_gpio = devm_gpiod_get_optional(dev, "us-euro",
> >> -						GPIOD_OUT_LOW);
> >> -	if (IS_ERR(wcd938x->us_euro_gpio))
> >> -		return dev_err_probe(dev, PTR_ERR(wcd938x->us_euro_gpio),
> >> -				     "us-euro swap Control GPIO not found\n");
> >> +	wcd938x->us_euro_mux = devm_mux_control_get(dev, NULL);

> > Running with this patch on the CRD I noticed that this now prints an
> > error as there is no optional mux (or gpio) defined:
> > 
> > 	wcd938x_codec audio-codec: /audio-codec: failed to get mux-control (0)
> 
> This is not from codec driver, mux control is throwing up this.

Yeah, the error is printed by the mux code, but it's this codec change
that starts triggering it.

> > You need to suppress that error in mux_get() to allow for optional muxes
> > to be looked up like this.

> I have a plan for this,
> 
> I proposed some changes to mux api for exclusive apis at 
> https://lkml.org/lkml/2025/3/26/955
> 
> This should also allow us to easily add an optional api, which I plan to 
> do once i get some feedback on this patch.

It may just be as simple as demoting that dev_err() in mux_get() to
dev_dbg() and possibly make sure any callers log errors themselves.

A quick grep shows there are only 12 users or so and most of them
already seem to be logging mux lookup errors.

Johan