net: Fix rxq ref counting

[net-next-2.6.git] / net / core / dev.c

diff --git a/net/core/dev.c b/net/core/dev.c
index a313bab1b754a75272c4912d48611edc11e980df..193eafaabd8867197903909e92cd956da95b5bc4 100644 (file)
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1483,8 +1483,9 @@ int dev_forward_skb(struct net_device *dev, struct sk_buff *skb)
        skb_orphan(skb);
        nf_reset(skb);
 
-       if (!(dev->flags & IFF_UP) ||
-           (skb->len > (dev->mtu + dev->hard_header_len))) {
+       if (unlikely(!(dev->flags & IFF_UP) ||
+                    (skb->len > (dev->mtu + dev->hard_header_len)))) {
+               atomic_long_inc(&dev->rx_dropped);
                kfree_skb(skb);
                return NET_RX_DROP;
        }
@@ -1575,8 +1576,8 @@ EXPORT_SYMBOL(netif_set_real_num_tx_queues);
  *
  *     This must be called either with the rtnl_lock held or before
  *     registration of the net device.  Returns 0 on success, or a
- *     negative error code.  If called before registration, it also
- *     sets the maximum number of queues, and always succeeds.
+ *     negative error code.  If called before registration, it always
+ *     succeeds.
  */
 int netif_set_real_num_rx_queues(struct net_device *dev, unsigned int rxq)
 {
@@ -1592,8 +1593,6 @@ int netif_set_real_num_rx_queues(struct net_device *dev, unsigned int rxq)
                                                  rxq);
                if (rc)
                        return rc;
-       } else {
-               dev->num_rx_queues = rxq;
        }
 
        dev->real_num_rx_queues = rxq;
@@ -2548,6 +2547,7 @@ enqueue:
 
        local_irq_restore(flags);
 
+       atomic_long_inc(&skb->dev->rx_dropped);
        kfree_skb(skb);
        return NET_RX_DROP;
 }
@@ -2702,11 +2702,10 @@ EXPORT_SYMBOL_GPL(br_fdb_test_addr_hook);
  * the ingress scheduler, you just cant add policies on ingress.
  *
  */
-static int ing_filter(struct sk_buff *skb)
+static int ing_filter(struct sk_buff *skb, struct netdev_queue *rxq)
 {
        struct net_device *dev = skb->dev;
        u32 ttl = G_TC_RTTL(skb->tc_verd);
-       struct netdev_queue *rxq;
        int result = TC_ACT_OK;
        struct Qdisc *q;
 
@@ -2720,8 +2719,6 @@ static int ing_filter(struct sk_buff *skb)
        skb->tc_verd = SET_TC_RTTL(skb->tc_verd, ttl);
        skb->tc_verd = SET_TC_AT(skb->tc_verd, AT_INGRESS);
 
-       rxq = &dev->ingress_queue;
-
        q = rxq->qdisc;
        if (q != &noop_qdisc) {
                spin_lock(qdisc_lock(q));
@@ -2737,7 +2734,9 @@ static inline struct sk_buff *handle_ing(struct sk_buff *skb,
                                         struct packet_type **pt_prev,
                                         int *ret, struct net_device *orig_dev)
 {
-       if (skb->dev->ingress_queue.qdisc == &noop_qdisc)
+       struct netdev_queue *rxq = rcu_dereference(skb->dev->ingress_queue);
+
+       if (!rxq || rxq->qdisc == &noop_qdisc)
                goto out;
 
        if (*pt_prev) {
@@ -2745,7 +2744,7 @@ static inline struct sk_buff *handle_ing(struct sk_buff *skb,
                *pt_prev = NULL;
        }
 
-       switch (ing_filter(skb)) {
+       switch (ing_filter(skb, rxq)) {
        case TC_ACT_SHOT:
        case TC_ACT_STOLEN:
                kfree_skb(skb);
@@ -2996,6 +2995,7 @@ ncls:
        if (pt_prev) {
                ret = pt_prev->func(skb, skb->dev, pt_prev, orig_dev);
        } else {
+               atomic_long_inc(&skb->dev->rx_dropped);
                kfree_skb(skb);
                /* Jamal, now you will not able to escape explaining
                 * me how you were going to use this. :-)
@@ -4940,7 +4940,6 @@ static void __netdev_init_queue_locks_one(struct net_device *dev,
 static void netdev_init_queue_locks(struct net_device *dev)
 {
        netdev_for_each_tx_queue(dev, __netdev_init_queue_locks_one, NULL);
-       __netdev_init_queue_locks_one(dev, &dev->ingress_queue, NULL);
 }
 
 unsigned long netdev_fix_features(unsigned long features, const char *name)
@@ -5025,7 +5024,6 @@ static int netif_alloc_rx_queues(struct net_device *dev)
                        return -ENOMEM;
                }
                dev->_rx = rx;
-               atomic_set(&rx->count, count);
 
                /*
                 * Set a pointer to first element in the array which holds the
@@ -5431,14 +5429,14 @@ struct rtnl_link_stats64 *dev_get_stats(struct net_device *dev,
 
        if (ops->ndo_get_stats64) {
                memset(storage, 0, sizeof(*storage));
-               return ops->ndo_get_stats64(dev, storage);
-       }
-       if (ops->ndo_get_stats) {
+               ops->ndo_get_stats64(dev, storage);
+       } else if (ops->ndo_get_stats) {
                netdev_stats_to_stats64(storage, ops->ndo_get_stats(dev));
-               return storage;
+       } else {
+               netdev_stats_to_stats64(storage, &dev->stats);
+               dev_txq_stats_fold(dev, storage);
        }
-       netdev_stats_to_stats64(storage, &dev->stats);
-       dev_txq_stats_fold(dev, storage);
+       storage->rx_dropped += atomic_long_read(&dev->rx_dropped);
        return storage;
 }
 EXPORT_SYMBOL(dev_get_stats);
@@ -5452,11 +5450,29 @@ static void netdev_init_one_queue(struct net_device *dev,
 
 static void netdev_init_queues(struct net_device *dev)
 {
-       netdev_init_one_queue(dev, &dev->ingress_queue, NULL);
        netdev_for_each_tx_queue(dev, netdev_init_one_queue, NULL);
        spin_lock_init(&dev->tx_global_lock);
 }
 
+struct netdev_queue *dev_ingress_queue_create(struct net_device *dev)
+{
+       struct netdev_queue *queue = dev_ingress_queue(dev);
+
+#ifdef CONFIG_NET_CLS_ACT
+       if (queue)
+               return queue;
+       queue = kzalloc(sizeof(*queue), GFP_KERNEL);
+       if (!queue)
+               return NULL;
+       netdev_init_one_queue(dev, queue, NULL);
+       __netdev_init_queue_locks_one(dev, queue, NULL);
+       queue->qdisc = &noop_qdisc;
+       queue->qdisc_sleeping = &noop_qdisc;
+       rcu_assign_pointer(dev->ingress_queue, queue);
+#endif
+       return queue;
+}
+
 /**
  *     alloc_netdev_mq - allocate network device
  *     @sizeof_priv:   size of private data to allocate space for
@@ -5559,6 +5575,8 @@ void free_netdev(struct net_device *dev)
 
        kfree(dev->_tx);
 
+       kfree(rcu_dereference_raw(dev->ingress_queue));
+
        /* Flush device addresses */
        dev_addr_flush(dev);