datapath: backport: ovs: propagate per dp max headroom to all vports

Upstream commit:
    commit 3a927bc7cf9d0fbe8f4a8189dd5f8440228f64e7
    Author: Paolo Abeni <[email protected]>

    ovs: propagate per dp max headroom to all vports

    This patch implements bookkeeping support to compute the maximum
    headroom for all the devices in each datapath. When said value
    changes, the underlying devs are notified via the
    ndo_set_rx_headroom method.

    This also increases the internal vports xmit performance.

    Signed-off-by: Paolo Abeni <[email protected]>
    Signed-off-by: David S. Miller <[email protected]>

Signed-off-by: Pravin B Shelar <[email protected]>
Acked-by: Jesse Gross <[email protected]>

acinclude.m4

@@ -469,6 +469,7 @@ AC_DEFUN([OVS_CHECK_LINUX_COMPAT], [
   OVS_GREP_IFELSE([$KSRC/include/linux/netdevice.h], [udp_offload.*uoff],
                   [OVS_DEFINE([HAVE_UDP_OFFLOAD_ARG_UOFF])])
   OVS_GREP_IFELSE([$KSRC/include/linux/netdevice.h], [gro_remcsum])
+  OVS_GREP_IFELSE([$KSRC/include/linux/netdevice.h], [IFF_PHONY_HEADROOM])
 
   OVS_GREP_IFELSE([$KSRC/include/linux/netfilter.h], [nf_hook_state])
   OVS_GREP_IFELSE([$KSRC/include/linux/netfilter.h], [nf_register_net_hook])

datapath/datapath.c

@@ -1953,6 +1953,29 @@ 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;
@@ -2019,6 +2042,12 @@ static int ovs_vport_cmd_new(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_NEW);
 	BUG_ON(err < 0);
+
+	if (netdev_get_fwd_headroom(vport->dev) > dp->max_headroom)
+		update_headroom(dp);
+	else
+		netdev_set_rx_headroom(vport->dev, dp->max_headroom);
+
 	ovs_unlock();
 
 	ovs_notify(&dp_vport_genl_family, &ovs_dp_vport_multicast_group, reply, info);
@@ -2083,8 +2112,10 @@ static int ovs_vport_cmd_set(struct sk_buff *skb, struct genl_info *info)
 
 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;
 
@@ -2106,7 +2137,17 @@ 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, &ovs_dp_vport_multicast_group, reply, info);

datapath/datapath.h

@@ -69,6 +69,8 @@ struct dp_stats_percpu {
  * ovs_mutex and RCU.
  * @stats_percpu: Per-CPU datapath statistics.
  * @net: Reference to net namespace.
+ * @max_headroom: the maximum headroom of all vports in this datapath; it will
+ * be used by all the internal vports in this dp.
  *
  * Context: See the comment on locking at the top of datapath.c for additional
  * locking information.
@@ -90,6 +92,8 @@ struct datapath {
 	possible_net_t net;
 
 	u32 user_features;
+
+	u32 max_headroom;
 };
 
 /**

datapath/linux/compat/include/linux/netdevice.h

@@ -262,4 +262,23 @@ int ovs_dev_fill_metadata_dst(struct net_device *dev, struct sk_buff *skb);
 #define NETDEV_OFFLOAD_PUSH_GENEVE      0x001D
 #endif
 
+#ifndef HAVE_IFF_PHONY_HEADROOM
+
+#define IFF_PHONY_HEADROOM 0
+static inline unsigned netdev_get_fwd_headroom(struct net_device *dev)
+{
+	return 0;
+}
+
+static inline void netdev_set_rx_headroom(struct net_device *dev, int new_hr)
+{
+}
+
+/* set the device rx headroom to the dev's default */
+static inline void netdev_reset_rx_headroom(struct net_device *dev)
+{
+}
+
+#endif
+
 #endif /* __LINUX_NETDEVICE_WRAPPER_H */

datapath/vport-internal_dev.c

@@ -138,13 +138,23 @@ internal_get_stats(struct net_device *dev, struct rtnl_link_stats64 *stats)
 	return stats;
 }
 
+#ifdef HAVE_IFF_PHONY_HEADROOM
+static void internal_set_rx_headroom(struct net_device *dev, int new_hr)
+{
+	dev->needed_headroom = new_hr;
+}
+#endif
+
 static const struct net_device_ops internal_dev_netdev_ops = {
 	.ndo_open = internal_dev_open,
 	.ndo_stop = internal_dev_stop,
 	.ndo_start_xmit = internal_dev_xmit,
 	.ndo_set_mac_address = eth_mac_addr,
 	.ndo_change_mtu = internal_dev_change_mtu,
 	.ndo_get_stats64 = internal_get_stats,
+#ifdef HAVE_IFF_PHONY_HEADROOM
+	.ndo_set_rx_headroom = internal_set_rx_headroom,
+#endif
 };
 
 static struct rtnl_link_ops internal_dev_link_ops __read_mostly = {
@@ -158,7 +168,8 @@ static void do_setup(struct net_device *netdev)
 	netdev->netdev_ops = &internal_dev_netdev_ops;
 
 	netdev->priv_flags &= ~IFF_TX_SKB_SHARING;
-	netdev->priv_flags |= IFF_LIVE_ADDR_CHANGE | IFF_OPENVSWITCH;
+	netdev->priv_flags |= IFF_LIVE_ADDR_CHANGE | IFF_OPENVSWITCH |
+			      IFF_PHONY_HEADROOM;
 	netdev->destructor = internal_dev_destructor;
 	netdev->ethtool_ops = &internal_dev_ethtool_ops;
 	netdev->rtnl_link_ops = &internal_dev_link_ops;
@@ -200,6 +211,9 @@ static struct vport *internal_dev_create(const struct vport_parms *parms)
 		goto error_free_netdev;
 	}
 
+#ifdef HAVE_IFF_PHONY_HEADROOM
+	vport->dev->needed_headroom = vport->dp->max_headroom;
+#endif
 	dev_net_set(vport->dev, ovs_dp_get_net(vport->dp));
 	internal_dev = internal_dev_priv(vport->dev);
 	internal_dev->vport = vport;