On Fri, Mar 03, 2017 at 06:51:28PM +0200, Andy Shevchenko wrote:
> On Tue, Feb 28, 2017 at 4:37 AM, Eva Rachel Retuya <eraret...@gmail.com> 
> wrote:
> > Move I2C-specific code into its own file and rely on regmap to access
> > registers. The core code provides access to x, y, z and scale readings.
> 
> Portion of minor comments.
> 
> > +config ADXL345_I2C
> > +       tristate "Analog Devices ADXL345 3-Axis Digital Accelerometer I2C 
> > Driver"
> 
> >         depends on !(INPUT_ADXL34X=y || INPUT_ADXL34X=m)
> 
> I'm wondering if it works in a form of
> 
> depends on INPUT_ADXL34X=n
> 

Yes it works. Will change it into this form.

> > +int adxl345_common_probe(struct device *dev, struct regmap *regmap,
> > +                        const char *name);
> > +int adxl345_common_remove(struct device *dev);
> 
> I think a "common" word is redundant.
> 

OK. I will use "core" instead.

> > - * IIO driver for ADXL345
> > - * 7-bit I2C slave address: 0x1D (ALT ADDRESS pin tied to VDDIO) or
> > - * 0x53 (ALT ADDRESS pin grounded)
> 
> > + * IIO core driver for ADXL345
> 
> Should not it be at the beginning of header comment?
> 

Ack.

> > +static int adxl345_i2c_probe(struct i2c_client *client,
> > +                            const struct i2c_device_id *id)
> > +{
> 
> > +       struct regmap *regmap;
> > +       const char *name = NULL;
> 
> Reverse tree order, please.
> 
> > +
> > +       regmap = devm_regmap_init_i2c(client, &adxl345_i2c_regmap_config);
> > +       if (IS_ERR(regmap)) {
> > +               dev_err(&client->dev, "Error initializing i2c regmap: %d\n",
> > +                       (int)PTR_ERR(regmap));
> > +               return PTR_ERR(regmap);
> > +       }
> > +
> 
> > +       if (id)
> > +               name = id->name;
> > +
> > +       return adxl345_common_probe(&client->dev, regmap, name);
> 
> Do you need temporary variable?
> 
> return adxl345_probe(&client->dev, regmap, id ? id->name : NULL);
> 

Will remove it and use the form you suggest.

Thanks,
Eva

> -- 
> With Best Regards,
> Andy Shevchenko

Reply via email to