Re: [PATCH v3 6/6] regulator: bd71837: BD71837 PMIC regulator driver

[Matti Vaittinen]

Hello,

On Tue, May 29, 2018 at 12:47:40PM +0100, Mark Brown wrote:
> On Tue, May 29, 2018 at 01:02:15PM +0300, Matti Vaittinen wrote:
> 
> > @@ -0,0 +1,677 @@
> > +/* SPDX-License-Identifier: GPL-2.0 */
> > +/* Copyright (C) 2018 ROHM Semiconductors */
> > +/*
> > + * bd71837-regulator.c ROHM BD71837MWV regulator driver
> > + */
> 
> The SPDX header (and the rest of the block) need to be C++ comments.

Oh. My bad. I mixed up the C and Cpp style comments. So I will convert
the SPDX and following block to // (Cpp -style) comments.
> 
> > +static int bd71837_regulator_set_regmap(struct regulator_dev *rdev, int set)
> > +{
> > +	int ret = -EINVAL;
> > +	struct bd71837_pmic *pmic = rdev->reg_data;
> > +
> > +	/* According to the data sheet we must not change regulator voltage
> > +	 * when it is enabled. Thus we need to check if regulator is enabled
> > +	 * before changing the voltage. This mutex is used to avoid race where
> > +	 * we might enable regulator after it's state has been checked but
> > +	 * before the voltage is changed
> > +	 */
> > +	mutex_lock(&pmic->mtx);
> > +	if (!set)
> > +		ret = regulator_disable_regmap(rdev);
> > +	else
> > +		ret = regulator_enable_regmap(rdev);
> > +	mutex_unlock(&pmic->mtx);
> > +
> 
> This still has the weird locking/wrapper thing going on.  The regulator
> core will ensure that only one operation is called on a given regulator
> at once.
Ok. I''ll remove the mutex and wrapper since core handles this. Also,
after re-reading the data sheet I see that this restriction to voltage
changes (regulator must be disabled when voltage is changed) is not
mentioned for first 4 bucks. I will confirm if they can be changed when
enabled and also reflect this in next patch set.

> 
> > +static const struct regulator_desc bd71837_regulators[] = {
> > +	{
> > +	 .name = "buck1",
> 
> The indentation style here is weird, please follow CodingStyle.  Looks
> like the second level is just indented by a space for some reason, and
> there's similar problems elsewhere.
I'll change this too.

Br,
	Matti Vaittinen