Re: [PATCH 3/4] power_supply/MFD: twl4030_charger: Allow charger tocontrol the regulator that feeds it.

[NeilBrown]

On Sun, 13 May 2012 20:12:08 +0200 Andi Shyti <andi.shyti@xxxxxxxxx> wrote:

> Hi Neil,
> 
> On Wed, May 09, 2012 at 07:40:40AM +1000, NeilBrown wrote:
> > @@ -183,6 +186,12 @@ static int twl4030_charger_enable_usb(struct twl4030_bci *bci, bool enable)
> >  			return -EACCES;
> >  		}
> >  
> > +		/* Need to keep regulator on */
> > +		if (!bci->usb_enabled) {
> > +			regulator_enable(bci->usb_reg);
> 
> Should you consider here a failure case?

Probably.  Something like this?

+		/* Need to keep regulator on */
+		if (!bci->usb_enabled &&
+		    bci->usb_reg &&
+		    regulator_enable(bci->usb_reg) == 0)
+			bci->usb_enabled = 1;
+

> 
> > +			bci->usb_enabled = 1;
> > +		}
> > +
> >  		/* forcing the field BCIAUTOUSB (BOOT_BCI[1]) to 1 */
> >  		ret = twl4030_clear_set_boot_bci(0, TWL4030_BCIAUTOUSB);
> >  		if (ret < 0)
> > @@ -511,6 +524,8 @@ static int __init twl4030_bci_probe(struct platform_device *pdev)
> >  	bci->usb.num_properties = ARRAY_SIZE(twl4030_charger_props);
> >  	bci->usb.get_property = twl4030_bci_get_property;
> >  
> > +	bci->usb_reg = regulator_get(bci->dev, "bci3v1");
> 
> ... and here.

In what circumstances can that fail, I wonder.
Still, maybe this is best:


+	bci->usb_reg = regulator_get(bci->dev, "bci3v1");
+	if (IS_ERR(bci->usb_reg)) {
+		dev_warn(&bci->dev, "regulator get bci3v1 failed\n");
+		bci->usb_reg = NULL;
+	}
+

Thanks,
NeilBrown

Attachment: signature.asc
Description: PGP signature