RE: [PATCH V4 2/2] iio: light: isl29018: add optional vdd/vdda regulator operation support

[Anson Huang <anson.huang@nxp.com>]

Hi, Phil

Best Regards!
Anson Huang

> -----Original Message-----
> From: Phil Reid [mailto:preid@electromag.com.au]
> Sent: 2018年12月11日 11:56
> To: Anson Huang <anson.huang@nxp.com>; jic23@kernel.org;
> knaack.h@gmx.de; lars@metafoo.de; pmeerw@pmeerw.net;
> robh+dt@kernel.org; mark.rutland@arm.com; linux-iio@vger.kernel.org;
> devicetree@vger.kernel.org; linux-kernel@vger.kernel.org;
> festevam@gmail.com
> Cc: dl-linux-imx <linux-imx@nxp.com>
> Subject: Re: [PATCH V4 2/2] iio: light: isl29018: add optional vdd/vdda
> regulator operation support
> 
> G'day Anson,
> 
> Just pulled up the datasheet for this chip.
> 
> The absolute max for Vdda is speced as Vddd +/-0.5 With a note that Vdda
> should be externally shorted to Vddd.

The data sheet says vdda should be connected to vdd externally, then I think we should just
use one regulator vdd, then it will be much easier for the driver, what do you think?

Anson.

> 
> 
> On 11/12/2018 11:43 am, Anson Huang wrote:
> > Hi, Phil
> >
> > Best Regards!
> > Anson Huang
> >
> >> -----Original Message-----
> >> From: Phil Reid [mailto:preid@electromag.com.au]
> >> Sent: 2018年12月11日 11:36
> >> To: Anson Huang <anson.huang@nxp.com>; jic23@kernel.org;
> >> knaack.h@gmx.de; lars@metafoo.de; pmeerw@pmeerw.net;
> >> robh+dt@kernel.org; mark.rutland@arm.com; linux-iio@vger.kernel.org;
> >> devicetree@vger.kernel.org; linux-kernel@vger.kernel.org;
> >> festevam@gmail.com
> >> Cc: dl-linux-imx <linux-imx@nxp.com>
> >> Subject: Re: [PATCH V4 2/2] iio: light: isl29018: add optional
> >> vdd/vdda regulator operation support
> >>
> >> On 11/12/2018 11:24 am, Anson Huang wrote:
> >>> The light sensor's power supply could be controlled by regulator on
> >>> some platforms, such as i.MX6Q-SABRESD board, the light sensor
> >>> isl29023's power supply is controlled by a GPIO fixed regulator,
> >>> need to make sure the regulator is enabled before any operation of
> >>> sensor, this patch adds optional vdd/vdda regulator operation support.
> >>>
> >>> Signed-off-by: Anson Huang <Anson.Huang@nxp.com>
> >>> ---
> >>> ChangeLog since V3:
> >>> 	- improve the error handling of devm_regulator_get_optional;
> >>> 	- make sure regulators are disabled in error path.
> >>> ---
> >>>    drivers/iio/light/isl29018.c | 83
> >> ++++++++++++++++++++++++++++++++++++++++++--
> >>>    1 file changed, 80 insertions(+), 3 deletions(-)
> >>>
> >>> diff --git a/drivers/iio/light/isl29018.c
> >>> b/drivers/iio/light/isl29018.c index b45400f..a21652b 100644
> >>> --- a/drivers/iio/light/isl29018.c
> >>> +++ b/drivers/iio/light/isl29018.c
> >>> @@ -23,6 +23,7 @@
> >>>    #include <linux/mutex.h>
> >>>    #include <linux/delay.h>
> >>>    #include <linux/regmap.h>
> >>> +#include <linux/regulator/consumer.h>
> >>>    #include <linux/slab.h>
> >>>    #include <linux/iio/iio.h>
> >>>    #include <linux/iio/sysfs.h>
> >>> @@ -95,6 +96,8 @@ struct isl29018_chip {
> >>>    	struct isl29018_scale	scale;
> >>>    	int			prox_scheme;
> >>>    	bool			suspended;
> >>> +	struct regulator	*vdd_reg;
> >>> +	struct regulator	*vdda_reg;
> >>>    };
> >>>
> >>>    static int isl29018_set_integration_time(struct isl29018_chip
> >>> *chip, @@ -735,6 +738,34 @@ static int isl29018_probe(struct
> >>> i2c_client *client,
> >>>
> >>>    	mutex_init(&chip->lock);
> >>>
> >>> +	chip->vdd_reg = devm_regulator_get_optional(&client->dev, "vdd");
> >>> +	if (!IS_ERR(chip->vdd_reg)) {
> >>> +		err = regulator_enable(chip->vdd_reg);
> >>> +		if (err) {
> >>> +			dev_err(&client->dev, "failed to enable VDD regulator\n");
> >>> +			return err;
> >>> +		}
> >>> +	} else {
> >>> +		err = PTR_ERR(chip->vdd_reg);
> >>> +		if (err != -ENODEV)
> >>> +			return err;
> >>> +	}
> >>> +
> >>> +	chip->vdda_reg = devm_regulator_get_optional(&client->dev,
> "vdda");
> >>> +	if (!IS_ERR(chip->vdda_reg)) {
> >>> +		err = regulator_enable(chip->vdda_reg);
> >>> +		if (err) {
> >>> +			dev_err(&client->dev, "failed to enable VDDA
> regulator\n");
> >>> +			if (!IS_ERR(chip->vdd_reg))
> >>> +				regulator_disable(chip->vdd_reg);
> >>> +			return err;
> Not sure about this case at the call to enable failed so I think you only want
> the first one to be disabled.
> You could add another goto statement thou, see below.
> 
> 
> >>> +		}
> >>> +	} else {
> >>> +		err = PTR_ERR(chip->vdda_reg);
> >>> +		if (err != -ENODEV)
> >>> +			return err;
> >> maybe goto disable_regulators; to disable vdd.
> >
> > Agree, I will use " goto disable_regulators;" in both here and upper regulator
> enable fail case.
> > Please help review V5, thanks.
> 
> Here its safe to call both as vdda_reg will be an error ptr.
> 
> >
> > Anson
> >
> >>
> >>> +	}
> >>> +
> >>>    	chip->type = dev_id;
> >>>    	chip->calibscale = 1;
> >>>    	chip->ucalibscale = 0;
> >>> @@ -747,12 +778,12 @@ static int isl29018_probe(struct i2c_client
> *client,
> >>>    	if (IS_ERR(chip->regmap)) {
> >>>    		err = PTR_ERR(chip->regmap);
> >>>    		dev_err(&client->dev, "regmap initialization fails: %d\n", err);
> >>> -		return err;
> >>> +		goto disable_regulators;
> >>>    	}
> >>>
> >>>    	err = isl29018_chip_init(chip);
> >>>    	if (err)
> >>> -		return err;
> >>> +		goto disable_regulators;
> >>>
> >>>    	indio_dev->info = isl29018_chip_info_tbl[dev_id].indio_info;
> >>>    	indio_dev->channels = isl29018_chip_info_tbl[dev_id].channels;
> >>> @@ -761,13 +792,24 @@ static int isl29018_probe(struct i2c_client
> *client,
> >>>    	indio_dev->dev.parent = &client->dev;
> >>>    	indio_dev->modes = INDIO_DIRECT_MODE;
> >>>
> >>> -	return devm_iio_device_register(&client->dev, indio_dev);
> >>> +	err = devm_iio_device_register(&client->dev, indio_dev);
> >>> +	if (!err)
> >>> +		return 0;
> >>> +
> >>> +disable_regulators:
> >>> +	if (!IS_ERR(chip->vdd_reg))
> >>> +		regulator_disable(chip->vdd_reg);
> >>> +	if (!IS_ERR(chip->vdda_reg))
> >>> +		regulator_disable(chip->vdda_reg);
> >>> +
> eg: extra label here.
> Order is changed thou, which may be a problem.
> It's in the reverse order to the enable, which is usually what you want but the
> datasheet concerns above may be an issue.
> 
> 
> disable_regulators:
> 	if (!IS_ERR(chip->vdda_reg))
> 		regulator_disable(chip->vdda_reg);
> disable_regulator_vdd:
> 	if (!IS_ERR(chip->vdd_reg))
> 		regulator_disable(chip->vdd_reg);
> 
> >>> +	return err;
> >>>    }
> >>>
> >> [snip]
> >>
> 
> 
> 
> --
> Regards
> Phil