On Thu, Dec 13, 2012 at 1:35 PM, Johannes Berg
<johannes@xxxxxxxxxxxxxxxx> wrote:
> On Thu, 2012-12-13 at 13:27 -0800, Luis R. Rodriguez wrote:
>
>> > diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
>> > index 260b5d8..370cd31 100644
>> > --- a/net/wireless/nl80211.c
>> > +++ b/net/wireless/nl80211.c
>> > @@ -4333,6 +4326,8 @@ static int nl80211_set_reg(struct sk_buff *skb, struct genl_info *info)
>> > }
>> > }
>> >
>> > + mutex_lock(&cfg80211_mutex);
>> > +
>> > r = set_regdom(rd);
>> > rd = NULL;
>>
>> Unless you removed it from previous patches in the series the
>> "bad_reg" goto label has an unlock but above you removed the top
>> locking call so I think you just forgot to remove the unlock on the
>> goto label.
>
> Good catch. Wonder why smatch didn't complain? Fixed.
>> > @@ -1471,6 +1475,7 @@ new_request:
>
> if (last_request != &core_request_world)
> kfree(last_request);
>
>> >
>> > last_request = pending_request;
>> > last_request->intersect = intersect;
>> > + last_request->processed = false;
>>
>> The clearing you are doing is on the last_request but it comes from
>> the pending regulatory request which is always false given that the
>> request is kzalloc'd, not sure if that was needed or how that helped.
>
> Not necessarily, I added a bit more context above -- if it's
> core_request_world the clearing is needed I believe.
Good catch yes, upon reset only if we actually run the core request
upon reset and I do think we do. Yeap:
static void reset_regdomains(bool full_reset)
...
last_request = &core_request_world;
}
With your fix in place:
Acked-by: Luis R. Rodriguez <mcgrof@xxxxxxxxxxxxxxxx>
Luis
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
