Re: [PATCH 1/2] extcon: Use the unique id for external connector instead of string

[Chanwoo Choi <cw00.choi@samsung.com>]

On 05/20/2015 12:24 PM, Krzysztof Kozlowski wrote:
> On 19.05.2015 21:28, Chanwoo Choi wrote:
>> Hi Krzysztof,
>>
>>>>> +
>>>>> +       if (i == edev->max_supported)
>>>>> +               return -EINVAL;
>>>>>
>>>>>         if (index < 0 || (edev->max_supported && edev->max_supported <= index))
>>>>>                 return -EINVAL;
>>>>> @@ -605,7 +644,7 @@ static void dummy_sysfs_dev_release(struct device *dev)
>>>>>   *
>>>>>   * Return the pointer of extcon device if success or ERR_PTR(err) if fail
>>>>>   */
>>>>> -struct extcon_dev *extcon_dev_allocate(const char **supported_cable)
>>>>> +struct extcon_dev *extcon_dev_allocate(const enum extcon *supported_cable)
>>>>
>>>> I think you also have to update the documentation. At least for
>>>> {devm}_extcon_dev_allocate but maybe in other places too. Previously
>>>> the documentation states that supported_cable is an array of strings.
>>>> Additionally AFAIU now it must end with EXTCON_NONE. This
>>>> sentinel-like info must be clearly documented.
>>
>> As you commented, extcon must need to document the update and usage method of extcon core.
>> The mainline kernel includes only the 'Documentation/extcon/porting-android-switch-class'.
>> Namely, the mainline kernel have not included the document showing the usage method of extcon core.
> 
> I mean documentation for the function. The kerneldoc above functions says:
> 
> * @supported_cable:    Array of supported cable names ending with NULL.
> *                      If supported_cable is NULL, cable name related APIs
> *                      are disabled.

Ah. I misunderstand on previous mail. I'll fix upper description.

> 
> AFAIU now the supported cable must end with EXTCON_NONE. Is it correct?

Right. 

Thanks,
Chanwoo Choi