Re: [PATCH 2/2] HID: i2c-hid: support the regulator

[Brian Norris]

Hi Benjamin,

On Wed, Sep 14, 2016 at 09:36:03AM +0200, Benjamin Tissoires wrote:
> On Sep 05 2016 or thereabouts, Caesar Wang wrote:
> > diff --git a/drivers/hid/i2c-hid/i2c-hid.c b/drivers/hid/i2c-hid/i2c-hid.c
> > index b3ec4f2..07cc7aa 100644
> > --- a/drivers/hid/i2c-hid/i2c-hid.c
> > +++ b/drivers/hid/i2c-hid/i2c-hid.c
> > @@ -38,6 +38,7 @@
> >  #include <linux/acpi.h>
> >  #include <linux/of.h>
> >  #include <linux/gpio/consumer.h>
> > +#include <linux/regulator/consumer.h>
> >  
> >  #include <linux/i2c/i2c-hid.h>
> >  
> > @@ -152,6 +153,7 @@ struct i2c_hid {
> >  
> >  	bool			irq_wake_enabled;
> >  	struct mutex		reset_lock;
> > +	struct regulator	*supply;
> >  };
> >  
> >  static int __i2c_hid_command(struct i2c_client *client,
> > @@ -968,6 +970,21 @@ static int i2c_hid_probe(struct i2c_client *client,
> >  	if (!ihid)
> >  		return -ENOMEM;
> >  
> > +	ihid->supply = devm_regulator_get(&client->dev, "power");
> > +	if (IS_ERR(ihid->supply)) {
> 
> I am not familiar with regulators, but what if (like 99% of the
> available i2c-hid devices) there is no regulator attached to the device?
> 
> Will the pointer be null? Will there be a dummy regulator?
> 
> It seems at first sight that you are adding a requirement on the devices
> which is not part of the spec, and which will break every existing
> systems but yours. Again, I might be wrong, so please provide more
> information.

The default behavior of regulator_get() is to provide a dummy regulator
if none is found. So the pointer is never NULL, and it won't break
devices without a regulator. If you don't want a dummy regulator you
would use regulator_get_optional() instead, and you would then need to
handle ERR_PTR(-ENODEV) specifically.

Brian