Re: [PATCH v4 2/2] iio: adc: Add TI ADS131M0x ADC driver

[Jonathan Cameron <jonathan.cameron@huawei.com>]

On Mon, 8 Dec 2025 00:59:09 +0200
Andy Shevchenko <andriy.shevchenko@intel.com> wrote:

> On Sun, Dec 07, 2025 at 07:33:13PM +0000, Jonathan Cameron wrote:
> > On Tue, 18 Nov 2025 15:18:21 +0100
> > Oleksij Rempel <o.rempel@pengutronix.de> wrote:  
> 
> ...
> 
> > > +	clk = devm_clk_get_enabled(dev, NULL);  
> > 
> > This surprised me, so I went digging.  Anyone know why
> > the stub returns NULL?  Given that the normal function doesn't have
> > that as an allowed return value that seems really odd.
> > 
> > Still, it does, so this code is fine if odd.  
> 
> I believe it has to do something with an optional clocks or so, but OTOH
> I think it's so odd that may be considered as a bug in CCF APIs.

I'd think an optional clock is optional whether or not the generic clock
stuff is built. Hence should use devm_clk_get_optional_enabled()
That one I agree should be stubbed to return NULL (which it is)

Ah well, I don't really care about the answer, just diverted me for
a few mins so I thought I'd comment.

Jonathan

> 
> > > +	if (IS_ERR_OR_NULL(clk)) {
> > > +		if (IS_ERR(clk))
> > > +			ret = PTR_ERR(clk);
> > > +		else
> > > +			ret = -ENODEV;
> > > +
> > > +		return dev_err_probe(dev, ret, "clk get enabled failed\n");
> > > +	}  
>