Re: [PATCH 2/2] usb: r8a66597-udc: add support for SUDMAC

[Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>]

Hi,

2011/09/30 18:55, Felipe Balbi wrote:
> Hi,
> 
> On Fri, Sep 30, 2011 at 06:49:15PM +0900, Yoshihiro Shimoda wrote:
>> Hi,
>> 2011/09/30 17:17, Felipe Balbi wrote:
>>> Hi,
>>>
>> < snip >
>>> On Mon, Sep 26, 2011 at 05:43:02PM +0900, Yoshihiro Shimoda wrote:
>>>> @@ -186,6 +187,51 @@ static inline void control_reg_sqclr(struct r8a66597 *r8a66597, u16 pipenum)
>>>>  		printk(KERN_ERR "unexpect pipe num(%d)\n", pipenum);
>>>>  }
>>>>
>>>> +static void control_reg_sqset(struct r8a66597 *r8a66597, u16 pipenum)
>>>> +{
>>>> +	unsigned long offset;
>>>> +
>>>> +	pipe_stop(r8a66597, pipenum);
>>>> +
>>>> +	if (pipenum = 0)
>>>> +		r8a66597_bset(r8a66597, SQSET, DCPCTR);
>>>> +	else if (pipenum < R8A66597_MAX_NUM_PIPE) {
>>>> +		offset = get_pipectr_addr(pipenum);
>>>> +		r8a66597_bset(r8a66597, SQSET, offset);
>>>> +	} else
>>>> +		printk(KERN_ERR "unexpect pipe num(%d)\n", pipenum);
>>>
>>> according to CodingStyle, if one branch as braces, all of them should
>>> have. Also, try to avoid printk(), I guess you have a *dev here to use
>>> dev_err(). Besides this, I guess it looks fine.
>>>
>>
>> I knew the rule, but I mistook the code...
>> I will fix it. And I also use dev_err() instead of printk().
> 
> it might also be wise to add a patch, before this one, converting
> printk() to dev_* macros, I guess ;-)
> 

I think so. :)
I wrote such patch and I will submit it soon.

Best regards,
Yoshihiro Shimoda