Re: Re: [PATCH] gpiolib: friendly debug information for consumer

["Wang, Dongsheng" <dongsheng.wang@xxxxxxxxxxxxxxxx>]

Hi Linus,

Thanks for your feedback.

On 2018/2/7 21:04:07, "Linus Walleij" <linus.walleij@xxxxxxxxxx> wrote:

>Hi Wang,
>
>thank you for your patch!
>
>On Tue, Jan 23, 2018 at 6:28 AM, Wang Dongsheng
><dongsheng.wang@xxxxxxxxxxxxxxxx> wrote:
>
>>mdiobus always try to get "reset" consumer, but not all platforms
>>have this capability. "failed" makes observer confuse when a
>>consumer can not find.
>>
>>Signed-off-by: Wang Dongsheng <dongsheng.wang@xxxxxxxxxxxxxxxx>
>
>But the message is correct?
>
>>-               dev_dbg(dev, "lookup for GPIO %s failed\n", con_id);
>
>This is happening.
>
>I would be more inclined to agree if it was dev_err() but this is
>not even info, it is debug info.
>
dev_dbg is often used when we develop code. I've been asked recently
about this GPIO "failed", and others may be less familiar with GPIO.
I'm also not familiar with it, just do some quick research here. :)
So them confuse why kernel got this. In fact, there is no such GPIO
feature(external PHY reset GPIO) on our platform. Therefore, I prefer
not to use the word "failed".

May be "No GPIO consumer %s found" is better?


Cheers,
-Dongsheng

>
>>+               dev_dbg(dev, "Can not detect GPIO consumer %s\n", 
>>con_id);
>
>This is not about any "detection", it is about looking something
>up.
>
>
>
>Yours,
>Linus Walleij��.n��������+%������w��{.n�����{��
b���ܨ}���Ơz�j:+v�����w����ޙ��&�)ߡ�a����z�ޗ���ݢj��w�f