Merge branch 'ndo_set_rx_headroom'

Paolo Abeni says:

====================
bridge/ovs: avoid skb head copy on frame forwarding

Currently, while when an OVS or Linux bridge is used to forward frames towards
some tunnel device, a skb_head_copy() may occur if the ingress device do not
provide enough headroom for the tx encapsulation.

This patch series tries to address the issue implementing a new ndo operation to
allow the master device to control the headroom used when allocating the skb on
frame reception.

Said operation is used by the Linux bridge to notify the bridged ports of
needed_headroom changes, and similar bookkeeping and behaviour is also added to
openvswitch, on a per datapath basis.

Finally, the operation is implemented for veth and tun device, which give
performance improvement in the 6-12% range when forwarding frames from said
devices towards a vxlan tunnel.

v2:
- fix netdev_get_fwd_headroom() behaviour
- remove some code duplication with the netdev_set_rx_headroom() and
   netdev_reset_rx_headroom() helpers
- handle headroom reset on [v]port removal/deletion
- initialize tun align to the old default value

v3:
- fix a comment typo
====================

Acked-by: Pravin B Shelar <pshelar@ovn.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

1 files changed, 40 insertions, 0 deletions

diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c
index c4e8455d5d56..e6a7d494df24 100644
--- a/net/openvswitch/datapath.c
+++ b/net/openvswitch/datapath.c
@@ -1908,6 +1908,29 @@ static struct vport *lookup_vport(struct net *net,
 		return ERR_PTR(-EINVAL);
 }
 
+/* Called with ovs_mutex */
+static void update_headroom(struct datapath *dp)
+{
+	unsigned dev_headroom, max_headroom = 0;
+	struct net_device *dev;
+	struct vport *vport;
+	int i;
+
+	for (i = 0; i < DP_VPORT_HASH_BUCKETS; i++) {
+		hlist_for_each_entry_rcu(vport, &dp->ports[i], dp_hash_node) {
+			dev = vport->dev;
+			dev_headroom = netdev_get_fwd_headroom(dev);
+			if (dev_headroom > max_headroom)
+				max_headroom = dev_headroom;
+		}
+	}
+
+	dp->max_headroom = max_headroom;
+	for (i = 0; i < DP_VPORT_HASH_BUCKETS; i++)
+		hlist_for_each_entry_rcu(vport, &dp->ports[i], dp_hash_node)
+			netdev_set_rx_headroom(vport->dev, max_headroom);
+}
+
 static int ovs_vport_cmd_new(struct sk_buff *skb, struct genl_info *info)
 {
 	struct nlattr **a = info->attrs;
@@ -1973,6 +1996,12 @@ restart:
 
 	err = ovs_vport_cmd_fill_info(vport, reply, info->snd_portid,
 				      info->snd_seq, 0, OVS_VPORT_CMD_NEW);
+
+	if (netdev_get_fwd_headroom(vport->dev) > dp->max_headroom)
+		update_headroom(dp);
+	else
+		netdev_set_rx_headroom(vport->dev, dp->max_headroom);
+
 	BUG_ON(err < 0);
 	ovs_unlock();
 
@@ -2039,8 +2068,10 @@ exit_unlock_free:
 
 static int ovs_vport_cmd_del(struct sk_buff *skb, struct genl_info *info)
 {
+	bool must_update_headroom = false;
 	struct nlattr **a = info->attrs;
 	struct sk_buff *reply;
+	struct datapath *dp;
 	struct vport *vport;
 	int err;
 
@@ -2062,7 +2093,16 @@ static int ovs_vport_cmd_del(struct sk_buff *skb, struct genl_info *info)
 	err = ovs_vport_cmd_fill_info(vport, reply, info->snd_portid,
 				      info->snd_seq, 0, OVS_VPORT_CMD_DEL);
 	BUG_ON(err < 0);
+
+	/* the vport deletion may trigger dp headroom update */
+	dp = vport->dp;
+	if (netdev_get_fwd_headroom(vport->dev) == dp->max_headroom)
+		must_update_headroom = true;
+	netdev_reset_rx_headroom(vport->dev);
 	ovs_dp_detach_port(vport);
+
+	if (must_update_headroom)
+		update_headroom(dp);
 	ovs_unlock();
 
 	ovs_notify(&dp_vport_genl_family, reply, info);