Re: [PATCH] scsi: csiostor: replace snprintf with sysfs_emit

[Damien Le Moal]

On 2022/02/09 12:16, Damien Le Moal wrote:
> On 2022/02/09 12:12, Joe Perches wrote:
>> On Wed, 2022-02-09 at 11:36 +0900, Damien Le Moal wrote:
>>> On 2/9/22 09:40, davidcomponentone@xxxxxxxxx wrote:
>>>> From: Yang Guang <yang.guang5@xxxxxxxxxx>
>>>>
>>>> coccinelle report:
>>>> ./drivers/scsi/csiostor/csio_scsi.c:1433:8-16:
>>>> WARNING: use scnprintf or sprintf
>>>> ./drivers/scsi/csiostor/csio_scsi.c:1369:9-17:
>>>> WARNING: use scnprintf or sprintf
>>>> ./drivers/scsi/csiostor/csio_scsi.c:1479:8-16:
>>>> WARNING: use scnprintf or sprintf
>>>>
>>>> Use sysfs_emit instead of scnprintf or sprintf makes more sense.
>> []
>>>> diff --git a/drivers/scsi/csiostor/csio_scsi.c b/drivers/scsi/csiostor/csio_scsi.c
>> []
>>>> @@ -1366,9 +1366,9 @@ csio_show_hw_state(struct device *dev,
>>>>  	struct csio_hw *hw = csio_lnode_to_hw(ln);
>>>>  
>>>>  	if (csio_is_hw_ready(hw))
>>>> -		return snprintf(buf, PAGE_SIZE, "ready\n");
>>>> +		return sysfs_emit(buf, "ready\n");
>>>>  	else
>>>> -		return snprintf(buf, PAGE_SIZE, "not ready\n");
>>>> +		return sysfs_emit(buf, "not ready\n");
>>>
>>> While at it, you could remove the useless "else" above.
>>
>> Or not.  It's fine as is.  It's just a style preference.
> 
> It is. I dislike the useless line of code in this case :)
> 
>>
>> Another style option would be to use a ?: like any of
>>
>> 	return sysfs_emit(buf, "%sready\n", csio_is_hw_ready(hw) ? "" : "not ");
>> or
>> 	return sysfs_emit(buf, "%s\n", csio_is_hw_ready(hw) ? "ready" : "not ready");
>> or
>> 	return sysfs_emit(buf, csio_is_hw_ready(hw) ? "ready\n" : "not ready\n");
> 
> That is nice and can make that
> 	
> 	return sysfs_emit(buf, "%sready\n", csio_is_hw_ready(hw) ? "" : "not ");

Oops. You did have that one listed... Read too quickly...

> 
> too :)
> 


-- 
Damien Le Moal
Western Digital Research