Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1403740 > unrolled thread

Re: [PATCH v2 2/2] ASoC: da7219: Convert driver to use generic device/fwnode functions

Started byAndy Shevchenko <andriy.shevchenko@linux.intel.com>
First post2016-05-19 15:30 +0200
Last post2016-05-24 13:00 +0200
Articles 2 — 1 participant

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.


Contents

  Re: [PATCH v2 2/2] ASoC: da7219: Convert driver to use generic  device/fwnode functions Andy Shevchenko <andriy.shevchenko@linux.intel.com> - 2016-05-19 15:30 +0200
    Re: [PATCH v2 2/2] ASoC: da7219: Convert driver to use generic  device/fwnode functions Andy Shevchenko <andriy.shevchenko@linux.intel.com> - 2016-05-24 13:00 +0200

#1403740 — Re: [PATCH v2 2/2] ASoC: da7219: Convert driver to use generic device/fwnode functions

FromAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Date2016-05-19 15:30 +0200
SubjectRe: [PATCH v2 2/2] ASoC: da7219: Convert driver to use generic device/fwnode functions
Message-ID<rAwbw-kQ-33@gated-at.bofh.it>
On Tue, 2016-05-17 at 11:28 +0100, Adam Thomson wrote:
> This change converts the driver from using the of_* functions to using
> the device_* and fwnode_* functions for accssing FW related data.
> 
> Signed-off-by: Adam Thomson <Adam.Thomson.Opensource@diasemi.com>
> Tested-by: Sathyanarayana Nujella <sathyanarayana.nujella@intel.com>

@@ -538,97 +538,96 @@ static enum da7219_aad_adc_1bit_rpt
>  	}
>  }
> 
> -static struct da7219_aad_pdata *da7219_aad_of_to_pdata(struct
> snd_soc_codec *codec)
> +static struct da7219_aad_pdata *da7219_aad_fw_to_pdata(struct
> snd_soc_codec *codec)
>  {
> -	struct device_node *np = codec->dev->of_node;
> -	struct device_node *aad_np = of_find_node_by_name(np,
> "da7219_aad");
> +	struct device *dev = codec->dev;
> +	struct i2c_client *i2c = to_i2c_client(dev);
> +	struct fwnode_handle *aad_np =
> +		device_get_named_child_node(dev, "da7219_aad");

I would suggest to do an assignment below...

>  	struct da7219_aad_pdata *aad_pdata;
> -	const char *of_str;
> -	u32 of_val32;
> +	const char *fw_str;
> +	u32 fw_val32;
> 


...right here.
Same amount of LOC, but less difficult to see from where aad_np comes.

>  	if (!aad_np)
>  		return NULL;
> 

> @@ -769,9 +768,9 @@ int da7219_aad_init(struct snd_soc_codec *codec)
>  	da7219->aad = da7219_aad;
>  	da7219_aad->codec = codec;
> 
> -	/* Handle any DT/platform data */
> -	if ((codec->dev->of_node) && (da7219->pdata))
> -		da7219->pdata->aad_pdata =
> da7219_aad_of_to_pdata(codec);
> +	/* Handle any DT/ACPI/platform data */
> +     if ((da7219->pdata) && (!da7219->pdata->aad_pdata))

Redundant parens, twice.

-- 
Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Intel Finland Oy

[toc] | [next] | [standalone]


#1406056

FromAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Date2016-05-24 13:00 +0200
Message-ID<rCie6-2bW-19@gated-at.bofh.it>
In reply to#1403740
On Mon, 2016-05-23 at 22:53 +0000, Opensource [Adam Thomson] wrote:
> On May 19, 2015 14:28, Andy Shevchenko wrote:
> 
> > > -static struct da7219_aad_pdata *da7219_aad_of_to_pdata(struct
> > > snd_soc_codec *codec)
> > > +static struct da7219_aad_pdata *da7219_aad_fw_to_pdata(struct
> > > snd_soc_codec *codec)
> > >  {
> > > -	struct device_node *np = codec->dev->of_node;
> > > -	struct device_node *aad_np = of_find_node_by_name(np,
> > > "da7219_aad");
> > > +	struct device *dev = codec->dev;
> > > +	struct i2c_client *i2c = to_i2c_client(dev);
> > > +	struct fwnode_handle *aad_np =
> > > +		device_get_named_child_node(dev, "da7219_aad");
> > 
> > I would suggest to do an assignment below...
> > 
> > >  	struct da7219_aad_pdata *aad_pdata;
> > > -	const char *of_str;
> > > -	u32 of_val32;
> > > +	const char *fw_str;
> > > +	u32 fw_val32;
> > > 
> > 
> > 
> > ...right here.
> > Same amount of LOC, but less difficult to see from where aad_np
> > comes.
> > 
> > >  	if (!aad_np)
> > >  		return NULL;
> > > 
> 
> To be fair the allocation of 'aad_np' is only a few lines above so
> this really
> doesn't seem to make much difference in my opinion. It really
> shouldn't be hard
> for someone to spot where it's allocated.

Better to have it exactly before check. Just a readability and future
maintenance. (Someone might insert something in between, and a matter of
fact already did)

Though I agree this is minor.

> 
> > > @@ -769,9 +768,9 @@ int da7219_aad_init(struct snd_soc_codec
> > > *codec)
> > >  	da7219->aad = da7219_aad;
> > >  	da7219_aad->codec = codec;
> > > 
> > > -	/* Handle any DT/platform data */
> > > -	if ((codec->dev->of_node) && (da7219->pdata))
> > > -		da7219->pdata->aad_pdata =
> > > da7219_aad_of_to_pdata(codec);
> > > +	/* Handle any DT/ACPI/platform data */
> > > +     if ((da7219->pdata) && (!da7219->pdata->aad_pdata))
> > 
> > Redundant parens, twice.
> 
> Not essential, but looks cleaner to me. Unless there's a real demand
> to change,
> I'd like to leave this as is.

It's really unusual pattern and doesn't add any value

Compare
 if ((da7219->pdata) && (!da7219->pdata->aad_pdata)) 
to
 if (da7219->pdata && !da7219->pdata->aad_pdata)

Latter looks cleaner.

-- 
Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Intel Finland Oy

[toc] | [prev] | [standalone]


Back to top | Article view | linux.kernel


csiph-web