net: better IFF_XMIT_DST_RELEASE support

Testing xmit_more support with netperf and connected UDP sockets,
I found strange dst refcount false sharing.

Current handling of IFF_XMIT_DST_RELEASE is not optimal.

Dropping dst in validate_xmit_skb() is certainly too late in case
packet was queued by cpu X but dequeued by cpu Y

The logical point to take care of drop/force is in __dev_queue_xmit()
before even taking qdisc lock.

As Julian Anastasov pointed out, need for skb_dst() might come from some
packet schedulers or classifiers.

This patch adds new helper to cleanly express needs of various drivers
or qdiscs/classifiers.

Drivers that need skb_dst() in their ndo_start_xmit() should call
following helper in their setup instead of the prior :

	dev->priv_flags &= ~IFF_XMIT_DST_RELEASE;
->
	netif_keep_dst(dev);

Instead of using a single bit, we use two bits, one being
eventually rebuilt in bonding/team drivers.

The other one, is permanent and blocks IFF_XMIT_DST_RELEASE being
rebuilt in bonding/team. Eventually, we could add something
smarter later.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Julian Anastasov <ja@ssi.bg>
Signed-off-by: David S. Miller <davem@davemloft.net>

14 files changed, 23 insertions, 23 deletions

diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index 35a6b6b15e8a..0d441ec8763e 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -799,7 +799,8 @@ void vlan_setup(struct net_device *dev)
         ether_setup(dev);
 
         dev->priv_flags         |= IFF_802_1Q_VLAN;
-        dev->priv_flags         &= ~(IFF_XMIT_DST_RELEASE | IFF_TX_SKB_SHARING);
+        dev->priv_flags         &= ~IFF_TX_SKB_SHARING;
+        netif_keep_dst(dev);
         dev->tx_queue_len       = 0;
 
         dev->netdev_ops         = &vlan_netdev_ops;
diff --git a/net/atm/clip.c b/net/atm/clip.c
index 1d9eaa4f041a..17e55dfecbe2 100644
--- a/net/atm/clip.c
+++ b/net/atm/clip.c
@@ -501,7 +501,7 @@ static void clip_setup(struct net_device *dev)
         /* without any more elaborate queuing. 100 is a reasonable */
         /* compromise between decent burst-tolerance and protection */
         /* against memory hogs. */
-        dev->priv_flags &= ~IFF_XMIT_DST_RELEASE;
+        netif_keep_dst(dev);
 }
 
 static int clip_create(int number)
diff --git a/net/core/dev.c b/net/core/dev.c
index a63b8c43c1b6..3c5bdaa44486 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2665,12 +2665,6 @@ static struct sk_buff *validate_xmit_skb(struct sk_buff *skb, struct net_device
         if (skb->next)
                 return skb;
 
-        /* If device doesn't need skb->dst, release it right now while
-         * its hot in this cpu cache
-         */
-        if (dev->priv_flags & IFF_XMIT_DST_RELEASE)
-                skb_dst_drop(skb);
-
         features = netif_skb_features(skb);
         skb = validate_xmit_vlan(skb, features);
         if (unlikely(!skb))
@@ -2811,8 +2805,6 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q,
                  * waiting to be sent out; and the qdisc is not running -
                  * xmit the skb directly.
                  */
-                if (!(dev->priv_flags & IFF_XMIT_DST_RELEASE))
-                        skb_dst_force(skb);
 
                 qdisc_bstats_update(q, skb);
 
@@ -2827,7 +2819,6 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q,
 
                 rc = NET_XMIT_SUCCESS;
         } else {
-                skb_dst_force(skb);
                 rc = q->enqueue(skb, q) & NET_XMIT_MASK;
                 if (qdisc_run_begin(q)) {
                         if (unlikely(contended)) {
@@ -2924,6 +2915,14 @@ static int __dev_queue_xmit(struct sk_buff *skb, void *accel_priv)
 
         skb_update_prio(skb);
 
+        /* If device/qdisc don't need skb->dst, release it right now while
+         * its hot in this cpu cache.
+         */
+        if (dev->priv_flags & IFF_XMIT_DST_RELEASE)
+                skb_dst_drop(skb);
+        else
+                skb_dst_force(skb);
+
         txq = netdev_pick_tx(dev, skb, accel_priv);
         q = rcu_dereference_bh(txq->qdisc);
 
@@ -6674,7 +6673,7 @@ struct net_device *alloc_netdev_mqs(int sizeof_priv, const char *name,
         INIT_LIST_HEAD(&dev->adj_list.lower);
         INIT_LIST_HEAD(&dev->all_adj_list.upper);
         INIT_LIST_HEAD(&dev->all_adj_list.lower);
-        dev->priv_flags = IFF_XMIT_DST_RELEASE;
+        dev->priv_flags = IFF_XMIT_DST_RELEASE | IFF_XMIT_DST_RELEASE_PERM;
         setup(dev);
 
         dev->num_tx_queues = txqs;
diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index 0485ef18d254..12055fdbe716 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -510,7 +510,7 @@ static int ipgre_tunnel_init(struct net_device *dev)
         memcpy(dev->broadcast, &iph->daddr, 4);
 
         dev->flags              = IFF_NOARP;
-        dev->priv_flags         &= ~IFF_XMIT_DST_RELEASE;
+        netif_keep_dst(dev);
         dev->addr_len           = 4;
 
         if (iph->daddr) {
diff --git a/net/ipv4/ip_vti.c b/net/ipv4/ip_vti.c
index e453cb724a95..3e861011e4a3 100644
--- a/net/ipv4/ip_vti.c
+++ b/net/ipv4/ip_vti.c
@@ -364,7 +364,7 @@ static int vti_tunnel_init(struct net_device *dev)
         dev->iflink             = 0;
         dev->addr_len           = 4;
         dev->features           |= NETIF_F_LLTX;
-        dev->priv_flags         &= ~IFF_XMIT_DST_RELEASE;
+        netif_keep_dst(dev);
 
         return ip_tunnel_init(dev);
 }
diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c
index ea88ab3102a8..37096d64730e 100644
--- a/net/ipv4/ipip.c
+++ b/net/ipv4/ipip.c
@@ -289,7 +289,7 @@ static void ipip_tunnel_setup(struct net_device *dev)
         dev->iflink             = 0;
         dev->addr_len           = 4;
         dev->features           |= NETIF_F_LLTX;
-        dev->priv_flags         &= ~IFF_XMIT_DST_RELEASE;
+        netif_keep_dst(dev);
 
         dev->features           |= IPIP_FEATURES;
         dev->hw_features        |= IPIP_FEATURES;
diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c
index 74b677916a70..de3b1c86b8d3 100644
--- a/net/ipv6/ip6_gre.c
+++ b/net/ipv6/ip6_gre.c
@@ -1242,7 +1242,7 @@ static void ip6gre_tunnel_setup(struct net_device *dev)
         dev->flags |= IFF_NOARP;
         dev->iflink = 0;
         dev->addr_len = sizeof(struct in6_addr);
-        dev->priv_flags &= ~IFF_XMIT_DST_RELEASE;
+        netif_keep_dst(dev);
 }
 
 static int ip6gre_tunnel_init(struct net_device *dev)
diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c
index d3e8888ad611..9409887fb664 100644
--- a/net/ipv6/ip6_tunnel.c
+++ b/net/ipv6/ip6_tunnel.c
@@ -1493,7 +1493,7 @@ static void ip6_tnl_dev_setup(struct net_device *dev)
                 dev->mtu -= 8;
         dev->flags |= IFF_NOARP;
         dev->addr_len = sizeof(struct in6_addr);
-        dev->priv_flags &= ~IFF_XMIT_DST_RELEASE;
+        netif_keep_dst(dev);
         /* This perm addr will be used as interface identifier by IPv6 */
         dev->addr_assign_type = NET_ADDR_RANDOM;
         eth_random_addr(dev->perm_addr);
diff --git a/net/ipv6/ip6_vti.c b/net/ipv6/ip6_vti.c
index 5833a2244467..d440bb585524 100644
--- a/net/ipv6/ip6_vti.c
+++ b/net/ipv6/ip6_vti.c
@@ -807,7 +807,7 @@ static void vti6_dev_setup(struct net_device *dev)
         dev->mtu = ETH_DATA_LEN;
         dev->flags |= IFF_NOARP;
         dev->addr_len = sizeof(struct in6_addr);
-        dev->priv_flags &= ~IFF_XMIT_DST_RELEASE;
+        netif_keep_dst(dev);
 }
 
 /**
diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c
index 0d4e27466f82..6eab37cf5345 100644
--- a/net/ipv6/sit.c
+++ b/net/ipv6/sit.c
@@ -1364,7 +1364,7 @@ static void ipip6_tunnel_setup(struct net_device *dev)
         dev->hard_header_len    = LL_MAX_HEADER + t_hlen;
         dev->mtu                = ETH_DATA_LEN - t_hlen;
         dev->flags              = IFF_NOARP;
-        dev->priv_flags        &= ~IFF_XMIT_DST_RELEASE;
+        netif_keep_dst(dev);
         dev->iflink             = 0;
         dev->addr_len           = 4;
         dev->features           |= NETIF_F_LLTX;
diff --git a/net/sched/cls_flow.c b/net/sched/cls_flow.c
index a5d2b20db560..4ac515f2a6ce 100644
--- a/net/sched/cls_flow.c
+++ b/net/sched/cls_flow.c
@@ -493,6 +493,8 @@ static int flow_change(struct net *net, struct sk_buff *in_skb,
         tcf_exts_change(tp, &fnew->exts, &e);
         tcf_em_tree_change(tp, &fnew->ematches, &t);
 
+        netif_keep_dst(qdisc_dev(tp->q));
+
         if (tb[TCA_FLOW_KEYS]) {
                 fnew->keymask = keymask;
                 fnew->nkeys   = nkeys;
diff --git a/net/sched/cls_route.c b/net/sched/cls_route.c
index 6f22baae0afa..109a329b7198 100644
--- a/net/sched/cls_route.c
+++ b/net/sched/cls_route.c
@@ -524,6 +524,7 @@ static int route4_change(struct net *net, struct sk_buff *in_skb,
                 if (f->handle < f1->handle)
                         break;
 
+        netif_keep_dst(qdisc_dev(tp->q));
         rcu_assign_pointer(f->next, f1);
         rcu_assign_pointer(*fp, f);
 
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 2b349a4de3c8..38d58e6cef07 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -47,7 +47,6 @@ EXPORT_SYMBOL(default_qdisc_ops);
 
 static inline int dev_requeue_skb(struct sk_buff *skb, struct Qdisc *q)
 {
-        skb_dst_force(skb);
         q->gso_skb = skb;
         q->qstats.requeues++;
         q->q.qlen++;    /* it's still part of the queue */
@@ -218,8 +217,6 @@ static inline int qdisc_restart(struct Qdisc *q)
         if (unlikely(!skb))
                 return 0;
 
-        WARN_ON_ONCE(skb_dst_is_noref(skb));
-
         root_lock = qdisc_lock(q);
         dev = qdisc_dev(q);
         txq = skb_get_tx_queue(dev, skb);
diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c
index 5cd291bd00e4..6ada42396a24 100644
--- a/net/sched/sch_teql.c
+++ b/net/sched/sch_teql.c
@@ -470,7 +470,7 @@ static __init void teql_master_setup(struct net_device *dev)
         dev->tx_queue_len       = 100;
         dev->flags              = IFF_NOARP;
         dev->hard_header_len    = LL_MAX_HEADER;
-        dev->priv_flags         &= ~IFF_XMIT_DST_RELEASE;
+        netif_keep_dst(dev);
 }
 
 static LIST_HEAD(master_dev_list);