mbox series

[v1,0/6] iio: Introduce and use device_property_match_property_string()

Message ID 20230808162800.61651-1-andriy.shevchenko@linux.intel.com
Headers show
Series iio: Introduce and use device_property_match_property_string() | expand

Message

Andy Shevchenko Aug. 8, 2023, 4:27 p.m. UTC
At least 4 drivers in the kernel are open coding functionality of
proposed device_property_match_property_string() helper.

There are also more existing users, but I want to limit this to one
(biggest user so far) subsystem. Hence, assuming Greg KH is okay
this can be routed via IIO (and again via Greg :-).

The +28 statistics in particular due to documentation.
With more users converted we may actually get rid of some
lines of code in the future.

Andy Shevchenko (6):
  device property: Use fwnode_property_string_array_count()
  device property: Add fwnode_property_match_property_string()
  iio: frequency: adf4377: Switch to
    device_property_match_property_string()
  iio: frequency: admv1014: Switch to
    device_property_match_property_string()
  iio: magnetometer: tmag5273: Switch to
    device_property_match_property_string()
  iio: proximity: sx9324: Switch to
    device_property_match_property_string()

 drivers/base/property.c             | 37 ++++++++++++++++++++++++++++-
 drivers/iio/frequency/adf4377.c     | 16 +++++--------
 drivers/iio/frequency/admv1014.c    | 31 ++++++++++--------------
 drivers/iio/magnetometer/tmag5273.c | 10 +++-----
 drivers/iio/proximity/sx9324.c      | 24 ++++++++-----------
 include/linux/property.h            | 12 ++++++++++
 6 files changed, 79 insertions(+), 51 deletions(-)

Comments

Jonathan Cameron Aug. 9, 2023, 5:59 p.m. UTC | #1
On Tue,  8 Aug 2023 19:27:56 +0300
Andy Shevchenko <andriy.shevchenko@linux.intel.com> wrote:

> Sometimes the users want to match the single value string property
> against an array of predefined strings. Create a helper for them.
> 
> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> ---
>  drivers/base/property.c  | 35 +++++++++++++++++++++++++++++++++++
>  include/linux/property.h | 12 ++++++++++++
>  2 files changed, 47 insertions(+)
> 
> diff --git a/drivers/base/property.c b/drivers/base/property.c
> index 3bb9505f1631..8f8e2a6816bc 100644
> --- a/drivers/base/property.c
> +++ b/drivers/base/property.c
> @@ -498,6 +498,41 @@ int fwnode_property_match_string(const struct fwnode_handle *fwnode,
>  }
>  EXPORT_SYMBOL_GPL(fwnode_property_match_string);
>  
> +/**
> + * fwnode_property_match_property_string - find a property string value in an array and return index
> + * @fwnode: Firmware node to get the property of
> + * @propname: Name of the property holding the string value
> + * @array: String array to search in
> + * @n: Size of the @array
> + *
> + * Find a property string value in a given @array and if it is found return
> + * the index back.
> + *
> + * Return: index, starting from %0, if the string value was found in the @array (success),
> + *	   %-ENOENT when the string value was not found in the @array,
> + *	   %-EINVAL if given arguments are not valid,
> + *	   %-ENODATA if the property does not have a value,
> + *	   %-EPROTO or %-EILSEQ if the property is not a string,
> + *	   %-ENXIO if no suitable firmware interface is present.
> + */
> +int fwnode_property_match_property_string(const struct fwnode_handle *fwnode,
> +	const char *propname, const char * const *array, size_t n)

Hi Andy,

Whilst I'm not 100% sold on adding ever increasing complexity to what we
match, this one feels like a common enough thing to be worth providing.

Looking at the usecases I wonder if it would be better to pass in
an unsigned int *ret which is only updated on a match?

That way the common properties approach of not checking the return value
if we have an optional property would apply.

e.g. patch 3 would end up with a block that looks like:

	st->input_mode = ADMV1014_IQ_MODE;
	device_property_match_property_string(&spi->dev, "adi,input-mode",
					      input_mode_names,
					      ARRAY_SIZE(input_mode_names),
					      &st->input_mode);

Only neat and tidy if the thing being optionally read into is an unsigned int
though (otherwise you still need a local variable)

Jonathan


> +{
> +	const char *string;
> +	int ret;
> +
> +	ret = fwnode_property_read_string(fwnode, propname, &string);
> +	if (ret)
> +		return ret;
> +
> +	ret = match_string(array, n, string);
> +	if (ret < 0)
> +		ret = -ENOENT;
> +
> +	return ret;
> +}
> +EXPORT_SYMBOL_GPL(fwnode_property_match_property_string);
> +
>  /**
>   * fwnode_property_get_reference_args() - Find a reference with arguments
>   * @fwnode:	Firmware node where to look for the reference
> diff --git a/include/linux/property.h b/include/linux/property.h
> index 8c3c6685a2ae..11f3ad6814f2 100644
> --- a/include/linux/property.h
> +++ b/include/linux/property.h
> @@ -97,6 +97,18 @@ static inline bool device_is_compatible(const struct device *dev, const char *co
>  	return fwnode_device_is_compatible(dev_fwnode(dev), compat);
>  }
>  
> +int fwnode_property_match_property_string(const struct fwnode_handle *fwnode,
> +					  const char *propname,
> +					  const char * const *array, size_t n);
> +
> +static inline
> +int device_property_match_property_string(const struct device *dev,
> +					  const char *propname,
> +					  const char * const *array, size_t n)
> +{
> +	return fwnode_property_match_property_string(dev_fwnode(dev), propname, array, n);
> +}
> +
>  int fwnode_property_get_reference_args(const struct fwnode_handle *fwnode,
>  				       const char *prop, const char *nargs_prop,
>  				       unsigned int nargs, unsigned int index,
Andy Shevchenko Aug. 10, 2023, 1:26 p.m. UTC | #2
On Wed, Aug 09, 2023 at 06:59:44PM +0100, Jonathan Cameron wrote:
> On Tue,  8 Aug 2023 19:27:56 +0300
> Andy Shevchenko <andriy.shevchenko@linux.intel.com> wrote:

...

> > +int fwnode_property_match_property_string(const struct fwnode_handle *fwnode,
> > +	const char *propname, const char * const *array, size_t n)
> 
> Hi Andy,
> 
> Whilst I'm not 100% sold on adding ever increasing complexity to what we
> match, this one feels like a common enough thing to be worth providing.

Yep, that's why I considered it's good to add (and because of new comers).

> Looking at the usecases I wonder if it would be better to pass in
> an unsigned int *ret which is only updated on a match?

So the question is here are we going to match (pun intended) the prototype to
the device_property_match*() family of functions or to device_property_read_*()
one. If the latter, this has to be renamed, but then it probably will contradict
the semantics as we are _matching_ against something and not just _reading_
something.

That said, do you agree that current implementation is (slightly) better from
these aspects? Anyway, look at the below.

> That way the common properties approach of not checking the return value
> if we have an optional property would apply.
> 
> e.g. patch 3

Only?

> would end up with a block that looks like:
> 
> 	st->input_mode = ADMV1014_IQ_MODE;
> 	device_property_match_property_string(&spi->dev, "adi,input-mode",
> 					      input_mode_names,
> 					      ARRAY_SIZE(input_mode_names),
> 					      &st->input_mode);
> 
> Only neat and tidy if the thing being optionally read into is an unsigned int
> though (otherwise you still need a local variable)

We also can have a hybrid variant, returning in both sides

  int device_property_match_property_string(..., size_t *index)
  {
	  if (index)
		  *index = ret;
	  return ret;
  }

(also note the correct return type as it has to match to @n).

Would it be still okay or too over engineered?
Rafael J. Wysocki Oct. 17, 2023, 7:19 p.m. UTC | #3
On Mon, Aug 28, 2023 at 8:00 PM Jonathan Cameron <jic23@kernel.org> wrote:
>
> On Thu, 10 Aug 2023 16:26:54 +0300
> Andy Shevchenko <andriy.shevchenko@linux.intel.com> wrote:
>
> > On Wed, Aug 09, 2023 at 06:59:44PM +0100, Jonathan Cameron wrote:
> > > On Tue,  8 Aug 2023 19:27:56 +0300
> > > Andy Shevchenko <andriy.shevchenko@linux.intel.com> wrote:
> >
> > ...
> >
> > > > +int fwnode_property_match_property_string(const struct fwnode_handle *fwnode,
> > > > + const char *propname, const char * const *array, size_t n)
> > >
> > > Hi Andy,
> > >
> > > Whilst I'm not 100% sold on adding ever increasing complexity to what we
> > > match, this one feels like a common enough thing to be worth providing.
> >
> > Yep, that's why I considered it's good to add (and because of new comers).
> >
> > > Looking at the usecases I wonder if it would be better to pass in
> > > an unsigned int *ret which is only updated on a match?
> >
> > So the question is here are we going to match (pun intended) the prototype to
> > the device_property_match*() family of functions or to device_property_read_*()
> > one. If the latter, this has to be renamed, but then it probably will contradict
> > the semantics as we are _matching_ against something and not just _reading_
> > something.
> >
> > That said, do you agree that current implementation is (slightly) better from
> > these aspects? Anyway, look at the below.
> >
> > > That way the common properties approach of not checking the return value
> > > if we have an optional property would apply.
> > >
> > > e.g. patch 3
> >
> > Only?
> I didn't look further :)
>
> >
> > > would end up with a block that looks like:
> > >
> > >     st->input_mode = ADMV1014_IQ_MODE;
> > >     device_property_match_property_string(&spi->dev, "adi,input-mode",
> > >                                           input_mode_names,
> > >                                           ARRAY_SIZE(input_mode_names),
> > >                                           &st->input_mode);
> > >
> > > Only neat and tidy if the thing being optionally read into is an unsigned int
> > > though (otherwise you still need a local variable)
> >
> > We also can have a hybrid variant, returning in both sides
> >
> >   int device_property_match_property_string(..., size_t *index)
> >   {
> >         if (index)
> >                 *index = ret;
> >         return ret;
> >   }
> >
> > (also note the correct return type as it has to match to @n).
> >
> > Would it be still okay or too over engineered?
> >
> Probably over engineered....
>
> Lets stick to what you have.  If various firmware folk are happy with
> the new function that's fine by me.  Rafael?

Sorry for the delay, I've lost track of this.

Honestly, I have no strong opinion, but I think that this is going to
reduce some code duplication which is a valid purpose, so please feel
free to add

Acked-by: Rafael J. Wysocki <rafael@kernel.org>

to this patch.

Thanks!
Andy Shevchenko Oct. 17, 2023, 7:43 p.m. UTC | #4
On Tue, Oct 17, 2023 at 09:19:30PM +0200, Rafael J. Wysocki wrote:
> On Mon, Aug 28, 2023 at 8:00 PM Jonathan Cameron <jic23@kernel.org> wrote:

...

> Sorry for the delay, I've lost track of this.

NP!

> Honestly, I have no strong opinion, but I think that this is going to
> reduce some code duplication which is a valid purpose, so please feel
> free to add
> 
> Acked-by: Rafael J. Wysocki <rafael@kernel.org>
> 
> to this patch.

Thank you!

Jonathan, are we all set for applying this series?
Jonathan Cameron Oct. 18, 2023, 7:37 p.m. UTC | #5
On Tue, 17 Oct 2023 22:43:04 +0300
Andy Shevchenko <andriy.shevchenko@linux.intel.com> wrote:

> On Tue, Oct 17, 2023 at 09:19:30PM +0200, Rafael J. Wysocki wrote:
> > On Mon, Aug 28, 2023 at 8:00 PM Jonathan Cameron <jic23@kernel.org> wrote:  
> 
> ...
> 
> > Sorry for the delay, I've lost track of this.  
> 
> NP!
> 
> > Honestly, I have no strong opinion, but I think that this is going to
> > reduce some code duplication which is a valid purpose, so please feel
> > free to add
> > 
> > Acked-by: Rafael J. Wysocki <rafael@kernel.org>
> > 
> > to this patch.  
> 
> Thank you!
> 
> Jonathan, are we all set for applying this series?
> 
Applied, but it might end up as 6.8 material depending on exactly how
timing turns out.  I have one pull request sent and I'm not sure I'll get
another one in this cycle. Given I just applied some big drivers I'd like to, but
not sure yet...


Jonathan
Andy Shevchenko Oct. 19, 2023, 12:05 p.m. UTC | #6
On Wed, Oct 18, 2023 at 08:37:55PM +0100, Jonathan Cameron wrote:
> On Tue, 17 Oct 2023 22:43:04 +0300
> Andy Shevchenko <andriy.shevchenko@linux.intel.com> wrote:
> > On Tue, Oct 17, 2023 at 09:19:30PM +0200, Rafael J. Wysocki wrote:
> > > On Mon, Aug 28, 2023 at 8:00 PM Jonathan Cameron <jic23@kernel.org> wrote:  

...

> > > Sorry for the delay, I've lost track of this.  
> > 
> > NP!
> > 
> > > Honestly, I have no strong opinion, but I think that this is going to
> > > reduce some code duplication which is a valid purpose, so please feel
> > > free to add
> > > 
> > > Acked-by: Rafael J. Wysocki <rafael@kernel.org>
> > > 
> > > to this patch.  
> > 
> > Thank you!
> > 
> > Jonathan, are we all set for applying this series?
> > 
> Applied, but it might end up as 6.8 material depending on exactly how
> timing turns out.  I have one pull request sent and I'm not sure I'll get
> another one in this cycle. Given I just applied some big drivers I'd like to, but
> not sure yet...

It's fine, I'm not in hurry with this and thank you!