Re: [PATCH 4/4] bonding: fix multiple 3ad mode sysfs race conditions

[Nikolay Aleksandrov <nikolay@redhat.com>]

On 05/15/2013 03:53 PM, Sergei Shtylyov wrote:
> Hello.
> 
> On 15-05-2013 16:32, Nikolay Aleksandrov wrote:
> 
>> When bond_3ad_get_active_agg_info() is used in all show_ad_ functions
>> it is not protected against slave manipulation and since it walks over
>> the slaves and uses them, this can easily result in NULL pointer
>> dereference or use of freed memory.
> 
>> Signed-off-by: Nikolay Aleksandrov <nikolay@redhat.com>
>> ---
>>   drivers/net/bonding/bond_sysfs.c | 21 ++++++++++++++++-----
>>   1 file changed, 16 insertions(+), 5 deletions(-)
> 
>> diff --git a/drivers/net/bonding/bond_sysfs.c b/drivers/net/bonding/bond_sysfs.c
>> index 77ea237..81ef36a 100644
>> --- a/drivers/net/bonding/bond_sysfs.c
>> +++ b/drivers/net/bonding/bond_sysfs.c
> [...]
>> @@ -1333,7 +1344,7 @@ static ssize_t bonding_show_ad_aggregator(struct device *d,
>>       if (bond->params.mode == BOND_MODE_8023AD) {
>>           struct ad_info ad_info;
>>           count = sprintf(buf, "%d\n",
>> -                (bond_3ad_get_active_agg_info(bond, &ad_info))
>> +                (get_active_agg_info(bond, &ad_info))
>>                   ?  0 : ad_info.aggregator_id);
>>       }
>>
>> @@ -1355,7 +1366,7 @@ static ssize_t bonding_show_ad_num_ports(struct device *d,
>>       if (bond->params.mode == BOND_MODE_8023AD) {
>>           struct ad_info ad_info;
>>           count = sprintf(buf, "%d\n",
>> -                (bond_3ad_get_active_agg_info(bond, &ad_info))
>> +                (get_active_agg_info(bond, &ad_info))
>>                   ?  0 : ad_info.ports);
>>       }
>>
>> @@ -1377,7 +1388,7 @@ static ssize_t bonding_show_ad_actor_key(struct device *d,
>>       if (bond->params.mode == BOND_MODE_8023AD) {
>>           struct ad_info ad_info;
>>           count = sprintf(buf, "%d\n",
>> -                (bond_3ad_get_active_agg_info(bond, &ad_info))
>> +                (get_active_agg_info(bond, &ad_info))
>>                   ?  0 : ad_info.actor_key);
>>       }
>>
>> @@ -1399,7 +1410,7 @@ static ssize_t bonding_show_ad_partner_key(struct device
>> *d,
>>       if (bond->params.mode == BOND_MODE_8023AD) {
>>           struct ad_info ad_info;
>>           count = sprintf(buf, "%d\n",
>> -                (bond_3ad_get_active_agg_info(bond, &ad_info))
>> +                (get_active_agg_info(bond, &ad_info))
>>                   ?  0 : ad_info.partner_key);
>>       }
> 
>    Perhaps it's time to get rid of the useless parens around function call in ?:
> operator?
> 
> WBR, Sergei
> 
> 
Perhaps it is :-) but I think to take care of this and other styling problems in
the bonding when I submit the trivial style fix patch which I have in my queue.
I might as well fix this one here if the others require it, should I submit a v2 ?

Cheers,
 Nik