PERFORCE change 170842 for review

Nathan Whitehorn nwhitehorn at freebsd.org
Fri Nov 20 16:11:49 UTC 2009


Hans Petter Selasky wrote:
> On Thursday 19 November 2009 23:47:59 Nathan Whitehorn wrote:
>   
>> Hans Petter Selasky wrote:
>>     
>>> http://p4web.freebsd.org/chv.cgi?CH=170842
>>>
>>> Change 170842 by hselasky at hselasky_laptop001 on 2009/11/19 22:34:49
>>>
>>>
>>> 	USB input:
>>> 		 - ATP patch from Rohit Grover:
>>> 		   - fixes some minor issues and
>>> 		     makes the control transfer
>>> 		     fully asynchronous
>>>       
>> [...]
>>
>>     
>>> @@ -1530,7 +1574,7 @@
>>>  		return (ENXIO);
>>>
>>>  	if (usbd_lookup_id_by_uaa(atp_devs, sizeof(atp_devs), uaa) == 0)
>>> -		return BUS_PROBE_SPECIFIC;
>>> +		return 0;
>>>  	else
>>>  		return ENXIO;
>>>  }
>>>       
>> Why are you replacing symbolic constants with less informative numeric
>> ones? -Nathan
>>     
>
> Because returning zero in probe has special meaning and is hardcoded in the 
> subr_bus.c code aswell. The other return values will not be changed.
>   
It's the same thing as far as the code is concerned, of course, my 
complaint was merely a style issue. Using the constant makes the meaning 
of the return value clearer, especially since this driver uses this 
return value to override the BUS_PROBE_GENERIC priority of ums(4). 
Changing it from the constant that was already there seemed like a step 
backward in readability.
-Nathan


More information about the p4-projects mailing list