vlan: use struct netdev_queue counters instead of dev->stats

We can update netdev_queue tx_bytes/tx_packets/tx_dropped counters instead
of dev->stats ones, to reduce number of cache lines dirtied in xmit path.

This fixes a performance problem on SMP when many different cpus take
vlan tx path.

Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

2 files changed, 9 insertions, 8 deletions

diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index 25ba41e35338..8faacee68633 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -290,7 +290,7 @@ static int vlan_dev_hard_header(struct sk_buff *skb, struct net_device *dev,
 
 static int vlan_dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
-        struct net_device_stats *stats = &dev->stats;
+        struct netdev_queue *txq = netdev_get_tx_queue(dev, 0);
         struct vlan_ethhdr *veth = (struct vlan_ethhdr *)(skb->data);
 
         /* Handle non-VLAN frames if they are sent to us, for example by DHCP.
@@ -309,7 +309,7 @@ static int vlan_dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev)
                 vlan_tci |= vlan_dev_get_egress_qos_mask(dev, skb);
                 skb = __vlan_put_tag(skb, vlan_tci);
                 if (!skb) {
-                        stats->tx_dropped++;
+                        txq->tx_dropped++;
                         return NETDEV_TX_OK;
                 }
 
@@ -317,8 +317,8 @@ static int vlan_dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev)
                         vlan_dev_info(dev)->cnt_inc_headroom_on_tx++;
         }
 
-        stats->tx_packets++;
-        stats->tx_bytes += skb->len;
+        txq->tx_packets++;
+        txq->tx_bytes += skb->len;
 
         skb->dev = vlan_dev_info(dev)->real_dev;
         dev_queue_xmit(skb);
@@ -328,15 +328,15 @@ static int vlan_dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev)
 static int vlan_dev_hwaccel_hard_start_xmit(struct sk_buff *skb,
                                             struct net_device *dev)
 {
-        struct net_device_stats *stats = &dev->stats;
+        struct netdev_queue *txq = netdev_get_tx_queue(dev, 0);
         u16 vlan_tci;
 
         vlan_tci = vlan_dev_info(dev)->vlan_id;
         vlan_tci |= vlan_dev_get_egress_qos_mask(dev, skb);
         skb = __vlan_hwaccel_put_tag(skb, vlan_tci);
 
-        stats->tx_packets++;
-        stats->tx_bytes += skb->len;
+        txq->tx_packets++;
+        txq->tx_bytes += skb->len;
 
         skb->dev = vlan_dev_info(dev)->real_dev;
         dev_queue_xmit(skb);
diff --git a/net/8021q/vlanproc.c b/net/8021q/vlanproc.c
index 3628e0a81b40..b55a091a33df 100644
--- a/net/8021q/vlanproc.c
+++ b/net/8021q/vlanproc.c
@@ -279,13 +279,14 @@ static int vlandev_seq_show(struct seq_file *seq, void *offset)
 {
         struct net_device *vlandev = (struct net_device *) seq->private;
         const struct vlan_dev_info *dev_info = vlan_dev_info(vlandev);
-        struct net_device_stats *stats = &vlandev->stats;
+        const struct net_device_stats *stats;
         static const char fmt[] = "%30s %12lu\n";
         int i;
 
         if (!is_vlan_dev(vlandev))
                 return 0;
 
+        stats = dev_get_stats(vlandev);
         seq_printf(seq,
                    "%s  VID: %d  REORDER_HDR: %i  dev->priv_flags: %hx\n",
                    vlandev->name, dev_info->vlan_id,