Bug#600155: [11/17] r6040: Fix multicast list iteration when hash filter is used

[Jack Stone <jwjstone@fastmail.fm>]

On 22/10/2010 22:18, Ben Hutchings wrote:
> On Fri, 2010-10-22 at 20:29 +0100, Jack Stone wrote:
>> On 22/10/2010 20:23, Greg KH wrote:
>>> On Fri, Oct 22, 2010 at 08:13:27PM +0100, Jack Stone wrote:
>>>> On 22/10/2010 19:39, Greg KH wrote:
>>>>>  drivers/net/r6040.c |    1 +
>>>>>  1 file changed, 1 insertion(+)
>>>>>
>>>>> --- a/drivers/net/r6040.c
>>>>> +++ b/drivers/net/r6040.c
>>>>> @@ -976,6 +976,7 @@ static void r6040_multicast_list(struct
>>>>>  		iowrite16(hash_table[3], ioaddr + MAR3);
>>>>>  	}
>>>>>  	/* Multicast Address 1~4 case */
>>>>> +	dmi = dev->mc_list;
>>>>>  	for (i = 0, dmi; (i < dev->mc_count) && (i < MCAST_MAX); i++) {
>>>> Any reason for the dmi in the above line? As far as I can see it is a
>>>> nop.
>>>
>>> Look closer at the for loop please.
>>
>> Maybe I'm missing something but:
>> for (i = 0, ---->dmi <----; (i < ...
>>
>> The dmi here still doesn't seem to do anything?
> 
> It doesn't, but it doesn't do any harm either.  The loop has been
> rewritten in mainline.

Agreed, it causes no problems, but it seems like it was intended to be
the dmi init, i.e.

for (i = 0, dmi = dev->mc_list; ...

I suppose it doesn't really matter but either removing the dmi in the
for loop or moving the dmi init in there would make more sense to me.

I really should learn to explain myself fully. Sorry for taking up
your time.

Thanks,

Jack