Hello,
On Thursday, July 01, 2010 2:25 PM Mark Brown wrote:
> On Thu, Jul 01, 2010 at 08:07:45AM +0200, Marek Szyprowski wrote:
>
> > +static struct regulator_consumer_supply aquila_ldo3_consumers[] = {
> > + { .supply = "VMIPI_1.1V", },
> > +};
>
> > +static struct regulator_consumer_supply aquila_ldo8_consumers[] = {
> > + { .supply = "VADC_3.3V", },
> > +};
>
> Almost all of these regulator supplies should be removed. Except in
> some exceptional cases all supplies should specify a struct device (the
> only one that really exists at the minute is CPU core due to lack of
> devices for CPUfreq). In general if you're defining a supply name that
> is the same as the rail on the board rather than a pin on a chip you're
> not using the API correctly.
Ok. I will remove them now and add later with the appropriate devices.
One more question - should I enable the regulators in the driver itself
or in the platform callback (like a "poweron()" callback)?
If I put regulator enabling into the driver how can one make it working
on a board without regulators (chip powered all the time)? Should the
driver ignore errors from regulator_get()?
What if one regulator needs enabling other to operate properly
(parent-child hierarchy)?
Best regards
--
Marek Szyprowski
Samsung Poland R&D Center
--
To unsubscribe from this list: send the line "unsubscribe linux-samsung-soc" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
