Re: [PATCH] ipv6: simplify code structure in addrconf.c/manage_tempaddrs

[Heiner Kallweit <heiner.kallweit@web.de>]

Am 31.03.2014 21:21, schrieb Hannes Frederic Sowa:
> On Mon, Mar 31, 2014 at 08:47:42PM +0200, Heiner Kallweit wrote:
>> read_unlock_bh can be moved out of the if clause
>>
>> Signed-off-by: Heiner Kallweit <heiner.kallweit@web.de>
>> ---
>>  net/ipv6/addrconf.c | 4 +---
>>  1 file changed, 1 insertion(+), 3 deletions(-)
>>
>> diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
>> index 6c7fa08..a65812f 100644
>> --- a/net/ipv6/addrconf.c
>> +++ b/net/ipv6/addrconf.c
>> @@ -2122,6 +2122,7 @@ static void manage_tempaddrs(struct inet6_dev *idev,
>>  		if (!(flags&IFA_F_TENTATIVE))
>>  			ipv6_ifa_notify(0, ift);
>>  	}
>> +	read_unlock_bh(&idev->lock);
>>  
>>  	if ((create || list_empty(&idev->tempaddr_list)) &&
>>  	    idev->cnf.use_tempaddr > 0) {
>> @@ -2130,10 +2131,7 @@ static void manage_tempaddrs(struct inet6_dev *idev,
>>  		 * Also create a temporary address if it's enabled but
>>  		 * no temporary address currently exists.
>>  		 */
>> -		read_unlock_bh(&idev->lock);
>>  		ipv6_create_tempaddr(ifp, NULL);
>> -	} else {
>> -		read_unlock_bh(&idev->lock);
> Doesn't the lock also protect the list_empty? I remember there is
> a list_empty_careful to check concurrently if empty during possible
> modification. I actually would leave it as is.
>
> Thanks,
>
>   Hannes
>
You're right, I missed this. I was a little to fast, sorry.