[NETFILTER]: Don't increase master refcount on expectations

As it's been discussed [1][2]. We shouldn't increase the master conntrack
refcount for non-fulfilled conntracks. During the conntrack destruction,
the expectations are always killed before the conntrack itself, this
guarantees that there won't be any orphan expectation.

[1]https://lists.netfilter.org/pipermail/netfilter-devel/2005-August/020783.html
[2]https://lists.netfilter.org/pipermail/netfilter-devel/2005-August/020904.html

Signed-off-by: Pablo Neira Ayuso <[e-mail]>
Signed-off-by: Patrick McHardy <[e-mail]>
Signed-off-by: David S. Miller <[e-mail]>

net/ipv4/netfilter/ip_conntrack_core.c

···
 	write_unlock_bh(&ip_conntrack_lock);
 }
 
+/* We don't increase the master conntrack refcount for non-fulfilled
+ * conntracks. During the conntrack destruction, the expectations are 
+ * always killed before the conntrack itself */
 struct ip_conntrack_expect *ip_conntrack_expect_alloc(struct ip_conntrack *me)
 {
 	struct ip_conntrack_expect *new;
···
 		return NULL;
 	}
 	new->master = me;
-	atomic_inc(&new->master->ct_general.use);
 	atomic_set(&new->use, 1);
 	return new;
 }
 
 void ip_conntrack_expect_put(struct ip_conntrack_expect *exp)
 {
-	if (atomic_dec_and_test(&exp->use)) {
-		ip_conntrack_put(exp->master);
+	if (atomic_dec_and_test(&exp->use))
 		kmem_cache_free(ip_conntrack_expect_cachep, exp);
-	}
 }
 
 static void ip_conntrack_expect_insert(struct ip_conntrack_expect *exp)