Re: [RFC PATCH v2 2/2] wifi: Add Nordic nRF70 series Wi-Fi driver

[Christophe JAILLET]

Le 01/05/2025 à 12:41, Artur Rojek a écrit :
> Hi Christophe,
>
> thanks for the review!
> Reply inline.
>
> On Fri, Apr 25, 2025 at 9:31 PM Christophe JAILLET
> <christophe.jaillet-39ZsbGIQGT5GWvitb5QawA@xxxxxxxxxxxxxxxx> wrote:
> > Le 22/04/2025 à 19:59, Artur Rojek a écrit :
> > > Introduce support for Nordic Semiconductor nRF70 series wireless
> > > companion IC.
> >
> > Hi,
> > ...
> >
> > > +     /* vpwr is mandatory, but we want to catch the -ENODEV error. */
> > > +     priv->vpwr = devm_regulator_get_optional(dev, "vpwr");
> > > +     if (IS_ERR(priv->vpwr))
> > > +             return dev_err_probe(dev, PTR_ERR(priv->vpwr),
> > > +                                  "Unable to find vpwr-supply property");
> > > +
> > > +     priv->vio = devm_regulator_get_optional(dev, "vio");
> > > +     if (IS_ERR(priv->vio) && PTR_ERR(priv->vio) != -ENODEV) {
> > > +             return dev_err_probe(dev, PTR_ERR(priv->vio),
> > > +                                  "Invalid vio-supply property");
> > > +     }
> >
> > Unneeded extra { }
> >
> > > +
> > > +     irq = of_irq_get_byname(dev->of_node, "host-irq");
> > > +     if (irq <= 0)
> > > +             return dev_err_probe(dev, irq, "Unable to find host-irq\n");
> >
> > If irq is 0, is it expected to return sucess here?
>
> No, return value of 0 is considered IRQ mapping failure, as per
> of_irq_get_byname() documentation. Perhaps it warrants a different error
> message, though.

My point is nor related to the test itself, or the message.

I mean, that, should irq = 0, then it ends to
	return dev_err_probe(dev, 0, "Unable to find host-irq\n");
and the probe returns success.

Maybe something like:
	return dev_err_probe(dev, irq ? irq : -ESOMETHING, "Unable to find 
host-irq\n");

CJ

> Cheers,
> Artur
>
> > > +
> > > +     mutex_init(&priv->write_lock);
> > > +     mutex_init(&priv->read_lock);
> > ...
> >
> > CJ