openvswitch: Fix egress tunnel info.

While transitioning to netdev based vport we broke OVS
feature which allows user to retrieve tunnel packet egress
information for lwtunnel devices.  Following patch fixes it
by introducing ndo operation to get the tunnel egress info.
Same ndo operation can be used for lwtunnel devices and compat
ovs-tnl-vport devices. So after adding such device operation
we can remove similar operation from ovs-vport.

Fixes: 614732eaa12d ("openvswitch: Use regular VXLAN net_device device").
Signed-off-by: Pravin B Shelar <pshelar@nicira.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

12 files changed, 79 insertions, 166 deletions

diff --git a/net/core/dev.c b/net/core/dev.c
index 6bb6470f5b7b..c14748d051e7 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -99,6 +99,7 @@
 #include <linux/rtnetlink.h>
 #include <linux/stat.h>
 #include <net/dst.h>
+#include <net/dst_metadata.h>
 #include <net/pkt_sched.h>
 #include <net/checksum.h>
 #include <net/xfrm.h>
@@ -682,6 +683,32 @@ int dev_get_iflink(const struct net_device *dev)
 EXPORT_SYMBOL(dev_get_iflink);
 
 /**
+ *	dev_fill_metadata_dst - Retrieve tunnel egress information.
+ *	@dev: targeted interface
+ *	@skb: The packet.
+ *
+ *	For better visibility of tunnel traffic OVS needs to retrieve
+ *	egress tunnel information for a packet. Following API allows
+ *	user to get this info.
+ */
+int dev_fill_metadata_dst(struct net_device *dev, struct sk_buff *skb)
+{
+	struct ip_tunnel_info *info;
+
+	if (!dev->netdev_ops  || !dev->netdev_ops->ndo_fill_metadata_dst)
+		return -EINVAL;
+
+	info = skb_tunnel_info_unclone(skb);
+	if (!info)
+		return -ENOMEM;
+	if (unlikely(!(info->mode & IP_TUNNEL_INFO_TX)))
+		return -EINVAL;
+
+	return dev->netdev_ops->ndo_fill_metadata_dst(dev, skb);
+}
+EXPORT_SYMBOL_GPL(dev_fill_metadata_dst);
+
+/**
  *	__dev_get_by_name	- find a device by its name
  *	@net: the applicable net namespace
  *	@name: name to find
diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index bd0679d90519..614521437e30 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -498,10 +498,26 @@ static struct sk_buff *gre_handle_offloads(struct sk_buff *skb,
 					csum ? SKB_GSO_GRE_CSUM : SKB_GSO_GRE);
 }
 
+static struct rtable *gre_get_rt(struct sk_buff *skb,
+				 struct net_device *dev,
+				 struct flowi4 *fl,
+				 const struct ip_tunnel_key *key)
+{
+	struct net *net = dev_net(dev);
+
+	memset(fl, 0, sizeof(*fl));
+	fl->daddr = key->u.ipv4.dst;
+	fl->saddr = key->u.ipv4.src;
+	fl->flowi4_tos = RT_TOS(key->tos);
+	fl->flowi4_mark = skb->mark;
+	fl->flowi4_proto = IPPROTO_GRE;
+
+	return ip_route_output_key(net, fl);
+}
+
 static void gre_fb_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct ip_tunnel_info *tun_info;
-	struct net *net = dev_net(dev);
 	const struct ip_tunnel_key *key;
 	struct flowi4 fl;
 	struct rtable *rt;
@@ -516,14 +532,7 @@ static void gre_fb_xmit(struct sk_buff *skb, struct net_device *dev)
 		goto err_free_skb;
 
 	key = &tun_info->key;
-	memset(&fl, 0, sizeof(fl));
-	fl.daddr = key->u.ipv4.dst;
-	fl.saddr = key->u.ipv4.src;
-	fl.flowi4_tos = RT_TOS(key->tos);
-	fl.flowi4_mark = skb->mark;
-	fl.flowi4_proto = IPPROTO_GRE;
-
-	rt = ip_route_output_key(net, &fl);
+	rt = gre_get_rt(skb, dev, &fl, key);
 	if (IS_ERR(rt))
 		goto err_free_skb;
 
@@ -566,6 +575,24 @@ err_free_skb:
 	dev->stats.tx_dropped++;
 }
 
+static int gre_fill_metadata_dst(struct net_device *dev, struct sk_buff *skb)
+{
+	struct ip_tunnel_info *info = skb_tunnel_info(skb);
+	struct rtable *rt;
+	struct flowi4 fl4;
+
+	if (ip_tunnel_info_af(info) != AF_INET)
+		return -EINVAL;
+
+	rt = gre_get_rt(skb, dev, &fl4, &info->key);
+	if (IS_ERR(rt))
+		return PTR_ERR(rt);
+
+	ip_rt_put(rt);
+	info->key.u.ipv4.src = fl4.saddr;
+	return 0;
+}
+
 static netdev_tx_t ipgre_xmit(struct sk_buff *skb,
 			      struct net_device *dev)
 {
@@ -1023,6 +1050,7 @@ static const struct net_device_ops gre_tap_netdev_ops = {
 	.ndo_change_mtu		= ip_tunnel_change_mtu,
 	.ndo_get_stats64	= ip_tunnel_get_stats64,
 	.ndo_get_iflink		= ip_tunnel_get_iflink,
+	.ndo_fill_metadata_dst	= gre_fill_metadata_dst,
 };
 
 static void ipgre_tap_setup(struct net_device *dev)
diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c
index c6a39bf2c3b9..0bf0f406de52 100644
--- a/net/openvswitch/actions.c
+++ b/net/openvswitch/actions.c
@@ -768,7 +768,6 @@ static int output_userspace(struct datapath *dp, struct sk_buff *skb,
 			    struct sw_flow_key *key, const struct nlattr *attr,
 			    const struct nlattr *actions, int actions_len)
 {
-	struct ip_tunnel_info info;
 	struct dp_upcall_info upcall;
 	const struct nlattr *a;
 	int rem;
@@ -796,11 +795,9 @@ static int output_userspace(struct datapath *dp, struct sk_buff *skb,
 			if (vport) {
 				int err;
 
-				upcall.egress_tun_info = &info;
-				err = ovs_vport_get_egress_tun_info(vport, skb,
-								    &upcall);
-				if (err)
-					upcall.egress_tun_info = NULL;
+				err = dev_fill_metadata_dst(vport->dev, skb);
+				if (!err)
+					upcall.egress_tun_info = skb_tunnel_info(skb);
 			}
 
 			break;
diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c
index b816ff871528..c5d08ee37730 100644
--- a/net/openvswitch/datapath.c
+++ b/net/openvswitch/datapath.c
@@ -490,9 +490,8 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb,
 
 	if (upcall_info->egress_tun_info) {
 		nla = nla_nest_start(user_skb, OVS_PACKET_ATTR_EGRESS_TUN_KEY);
-		err = ovs_nla_put_egress_tunnel_key(user_skb,
-						    upcall_info->egress_tun_info,
-						    upcall_info->egress_tun_opts);
+		err = ovs_nla_put_tunnel_info(user_skb,
+					      upcall_info->egress_tun_info);
 		BUG_ON(err);
 		nla_nest_end(user_skb, nla);
 	}
diff --git a/net/openvswitch/datapath.h b/net/openvswitch/datapath.h
index f88038a99f44..67bdecd9fdc1 100644
--- a/net/openvswitch/datapath.h
+++ b/net/openvswitch/datapath.h
@@ -117,7 +117,6 @@ struct ovs_skb_cb {
  */
 struct dp_upcall_info {
 	struct ip_tunnel_info *egress_tun_info;
-	const void *egress_tun_opts;
 	const struct nlattr *userdata;
 	const struct nlattr *actions;
 	int actions_len;
diff --git a/net/openvswitch/flow_netlink.c b/net/openvswitch/flow_netlink.c
index bd710bc37469..38536c137c54 100644
--- a/net/openvswitch/flow_netlink.c
+++ b/net/openvswitch/flow_netlink.c
@@ -717,7 +717,7 @@ static int __ipv4_tun_to_nlattr(struct sk_buff *skb,
 	if ((output->tun_flags & TUNNEL_OAM) &&
 	    nla_put_flag(skb, OVS_TUNNEL_KEY_ATTR_OAM))
 		return -EMSGSIZE;
-	if (tun_opts) {
+	if (swkey_tun_opts_len) {
 		if (output->tun_flags & TUNNEL_GENEVE_OPT &&
 		    nla_put(skb, OVS_TUNNEL_KEY_ATTR_GENEVE_OPTS,
 			    swkey_tun_opts_len, tun_opts))
@@ -749,13 +749,12 @@ static int ipv4_tun_to_nlattr(struct sk_buff *skb,
 	return 0;
 }
 
-int ovs_nla_put_egress_tunnel_key(struct sk_buff *skb,
-				  const struct ip_tunnel_info *egress_tun_info,
-				  const void *egress_tun_opts)
+int ovs_nla_put_tunnel_info(struct sk_buff *skb,
+			    struct ip_tunnel_info *tun_info)
 {
-	return __ipv4_tun_to_nlattr(skb, &egress_tun_info->key,
-				    egress_tun_opts,
-				    egress_tun_info->options_len);
+	return __ipv4_tun_to_nlattr(skb, &tun_info->key,
+				    ip_tunnel_info_opts(tun_info),
+				    tun_info->options_len);
 }
 
 static int metadata_from_nlattrs(struct net *net, struct sw_flow_match *match,
@@ -2383,10 +2382,7 @@ static int set_action_to_attr(const struct nlattr *a, struct sk_buff *skb)
 		if (!start)
 			return -EMSGSIZE;
 
-		err = ipv4_tun_to_nlattr(skb, &tun_info->key,
-					 tun_info->options_len ?
-					     ip_tunnel_info_opts(tun_info) : NULL,
-					 tun_info->options_len);
+		err = ovs_nla_put_tunnel_info(skb, tun_info);
 		if (err)
 			return err;
 		nla_nest_end(skb, start);
diff --git a/net/openvswitch/flow_netlink.h b/net/openvswitch/flow_netlink.h
index 6ca3f0baf449..47dd142eca1c 100644
--- a/net/openvswitch/flow_netlink.h
+++ b/net/openvswitch/flow_netlink.h
@@ -55,9 +55,9 @@ int ovs_nla_put_mask(const struct sw_flow *flow, struct sk_buff *skb);
 int ovs_nla_get_match(struct net *, struct sw_flow_match *,
 		      const struct nlattr *key, const struct nlattr *mask,
 		      bool log);
-int ovs_nla_put_egress_tunnel_key(struct sk_buff *,
-				  const struct ip_tunnel_info *,
-				  const void *egress_tun_opts);
+
+int ovs_nla_put_tunnel_info(struct sk_buff *skb,
+			    struct ip_tunnel_info *tun_info);
 
 bool ovs_nla_get_ufid(struct sw_flow_id *, const struct nlattr *, bool log);
 int ovs_nla_get_identifier(struct sw_flow_id *sfid, const struct nlattr *ufid,
diff --git a/net/openvswitch/vport-geneve.c b/net/openvswitch/vport-geneve.c
index 2735e9c4a3b8..5f8aaaaa0785 100644
--- a/net/openvswitch/vport-geneve.c
+++ b/net/openvswitch/vport-geneve.c
@@ -52,18 +52,6 @@ static int geneve_get_options(const struct vport *vport,
 	return 0;
 }
 
-static int geneve_get_egress_tun_info(struct vport *vport, struct sk_buff *skb,
-				      struct dp_upcall_info *upcall)
-{
-	struct geneve_port *geneve_port = geneve_vport(vport);
-	struct net *net = ovs_dp_get_net(vport->dp);
-	__be16 dport = htons(geneve_port->port_no);
-	__be16 sport = udp_flow_src_port(net, skb, 1, USHRT_MAX, true);
-
-	return ovs_tunnel_get_egress_info(upcall, ovs_dp_get_net(vport->dp),
-					  skb, IPPROTO_UDP, sport, dport);
-}
-
 static struct vport *geneve_tnl_create(const struct vport_parms *parms)
 {
 	struct net *net = ovs_dp_get_net(parms->dp);
@@ -130,7 +118,6 @@ static struct vport_ops ovs_geneve_vport_ops = {
 	.get_options	= geneve_get_options,
 	.send		= ovs_netdev_send,
 	.owner          = THIS_MODULE,
-	.get_egress_tun_info	= geneve_get_egress_tun_info,
 };
 
 static int __init ovs_geneve_tnl_init(void)
diff --git a/net/openvswitch/vport-gre.c b/net/openvswitch/vport-gre.c
index 4d24481669c9..64225bf5eb40 100644
--- a/net/openvswitch/vport-gre.c
+++ b/net/openvswitch/vport-gre.c
@@ -84,18 +84,10 @@ static struct vport *gre_create(const struct vport_parms *parms)
 	return ovs_netdev_link(vport, parms->name);
 }
 
-static int gre_get_egress_tun_info(struct vport *vport, struct sk_buff *skb,
-				   struct dp_upcall_info *upcall)
-{
-	return ovs_tunnel_get_egress_info(upcall, ovs_dp_get_net(vport->dp),
-					  skb, IPPROTO_GRE, 0, 0);
-}
-
 static struct vport_ops ovs_gre_vport_ops = {
 	.type		= OVS_VPORT_TYPE_GRE,
 	.create		= gre_create,
 	.send		= ovs_netdev_send,
-	.get_egress_tun_info	= gre_get_egress_tun_info,
 	.destroy	= ovs_netdev_tunnel_destroy,
 	.owner		= THIS_MODULE,
 };
diff --git a/net/openvswitch/vport-vxlan.c b/net/openvswitch/vport-vxlan.c
index c11413d5075f..e1c9c0888037 100644
--- a/net/openvswitch/vport-vxlan.c
+++ b/net/openvswitch/vport-vxlan.c
@@ -146,31 +146,12 @@ static struct vport *vxlan_create(const struct vport_parms *parms)
 	return ovs_netdev_link(vport, parms->name);
 }
 
-static int vxlan_get_egress_tun_info(struct vport *vport, struct sk_buff *skb,
-				     struct dp_upcall_info *upcall)
-{
-	struct vxlan_dev *vxlan = netdev_priv(vport->dev);
-	struct net *net = ovs_dp_get_net(vport->dp);
-	__be16 dst_port = vxlan_dev_dst_port(vxlan);
-	__be16 src_port;
-	int port_min;
-	int port_max;
-
-	inet_get_local_port_range(net, &port_min, &port_max);
-	src_port = udp_flow_src_port(net, skb, 0, 0, true);
-
-	return ovs_tunnel_get_egress_info(upcall, net,
-					  skb, IPPROTO_UDP,
-					  src_port, dst_port);
-}
-
 static struct vport_ops ovs_vxlan_netdev_vport_ops = {
 	.type			= OVS_VPORT_TYPE_VXLAN,
 	.create			= vxlan_create,
 	.destroy		= ovs_netdev_tunnel_destroy,
 	.get_options		= vxlan_get_options,
 	.send			= ovs_netdev_send,
-	.get_egress_tun_info	= vxlan_get_egress_tun_info,
 };
 
 static int __init ovs_vxlan_tnl_init(void)
diff --git a/net/openvswitch/vport.c b/net/openvswitch/vport.c
index 12a36ac21eda..320c765ce44a 100644
--- a/net/openvswitch/vport.c
+++ b/net/openvswitch/vport.c
@@ -479,61 +479,3 @@ void ovs_vport_deferred_free(struct vport *vport)
 	call_rcu(&vport->rcu, free_vport_rcu);
 }
 EXPORT_SYMBOL_GPL(ovs_vport_deferred_free);
-
-int ovs_tunnel_get_egress_info(struct dp_upcall_info *upcall,
-			       struct net *net,
-			       struct sk_buff *skb,
-			       u8 ipproto,
-			       __be16 tp_src,
-			       __be16 tp_dst)
-{
-	struct ip_tunnel_info *egress_tun_info = upcall->egress_tun_info;
-	const struct ip_tunnel_info *tun_info = skb_tunnel_info(skb);
-	const struct ip_tunnel_key *tun_key;
-	u32 skb_mark = skb->mark;
-	struct rtable *rt;
-	struct flowi4 fl;
-
-	if (unlikely(!tun_info))
-		return -EINVAL;
-	if (ip_tunnel_info_af(tun_info) != AF_INET)
-		return -EINVAL;
-
-	tun_key = &tun_info->key;
-
-	/* Route lookup to get srouce IP address.
-	 * The process may need to be changed if the corresponding process
-	 * in vports ops changed.
-	 */
-	rt = ovs_tunnel_route_lookup(net, tun_key, skb_mark, &fl, ipproto);
-	if (IS_ERR(rt))
-		return PTR_ERR(rt);
-
-	ip_rt_put(rt);
-
-	/* Generate egress_tun_info based on tun_info,
-	 * saddr, tp_src and tp_dst
-	 */
-	ip_tunnel_key_init(&egress_tun_info->key,
-			   fl.saddr, tun_key->u.ipv4.dst,
-			   tun_key->tos,
-			   tun_key->ttl,
-			   tp_src, tp_dst,
-			   tun_key->tun_id,
-			   tun_key->tun_flags);
-	egress_tun_info->options_len = tun_info->options_len;
-	egress_tun_info->mode = tun_info->mode;
-	upcall->egress_tun_opts = ip_tunnel_info_opts(egress_tun_info);
-	return 0;
-}
-EXPORT_SYMBOL_GPL(ovs_tunnel_get_egress_info);
-
-int ovs_vport_get_egress_tun_info(struct vport *vport, struct sk_buff *skb,
-				  struct dp_upcall_info *upcall)
-{
-	/* get_egress_tun_info() is only implemented on tunnel ports. */
-	if (unlikely(!vport->ops->get_egress_tun_info))
-		return -EINVAL;
-
-	return vport->ops->get_egress_tun_info(vport, skb, upcall);
-}
diff --git a/net/openvswitch/vport.h b/net/openvswitch/vport.h
index a413f3ae6a7b..d341ad6f3afe 100644
--- a/net/openvswitch/vport.h
+++ b/net/openvswitch/vport.h
@@ -27,7 +27,6 @@
 #include <linux/skbuff.h>
 #include <linux/spinlock.h>
 #include <linux/u64_stats_sync.h>
-#include <net/route.h>
 
 #include "datapath.h"
 
@@ -53,16 +52,6 @@ int ovs_vport_set_upcall_portids(struct vport *, const struct nlattr *pids);
 int ovs_vport_get_upcall_portids(const struct vport *, struct sk_buff *);
 u32 ovs_vport_find_upcall_portid(const struct vport *, struct sk_buff *);
 
-int ovs_tunnel_get_egress_info(struct dp_upcall_info *upcall,
-			       struct net *net,
-			       struct sk_buff *,
-			       u8 ipproto,
-			       __be16 tp_src,
-			       __be16 tp_dst);
-
-int ovs_vport_get_egress_tun_info(struct vport *vport, struct sk_buff *skb,
-				  struct dp_upcall_info *upcall);
-
 /**
  * struct vport_portids - array of netlink portids of a vport.
  *                        must be protected by rcu.
@@ -140,8 +129,6 @@ struct vport_parms {
  * have any configuration.
  * @send: Send a packet on the device.
  * zero for dropped packets or negative for error.
- * @get_egress_tun_info: Get the egress tunnel 5-tuple and other info for
- * a packet.
  */
 struct vport_ops {
 	enum ovs_vport_type type;
@@ -154,9 +141,6 @@ struct vport_ops {
 	int (*get_options)(const struct vport *, struct sk_buff *);
 
 	void (*send)(struct vport *, struct sk_buff *);
-	int (*get_egress_tun_info)(struct vport *, struct sk_buff *,
-				   struct dp_upcall_info *upcall);
-
 	struct module *owner;
 	struct list_head list;
 };
@@ -215,25 +199,6 @@ static inline const char *ovs_vport_name(struct vport *vport)
 int ovs_vport_ops_register(struct vport_ops *ops);
 void ovs_vport_ops_unregister(struct vport_ops *ops);
 
-static inline struct rtable *ovs_tunnel_route_lookup(struct net *net,
-						     const struct ip_tunnel_key *key,
-						     u32 mark,
-						     struct flowi4 *fl,
-						     u8 protocol)
-{
-	struct rtable *rt;
-
-	memset(fl, 0, sizeof(*fl));
-	fl->daddr = key->u.ipv4.dst;
-	fl->saddr = key->u.ipv4.src;
-	fl->flowi4_tos = RT_TOS(key->tos);
-	fl->flowi4_mark = mark;
-	fl->flowi4_proto = protocol;
-
-	rt = ip_route_output_key(net, fl);
-	return rt;
-}
-
 static inline void ovs_vport_send(struct vport *vport, struct sk_buff *skb)
 {
 	vport->ops->send(vport, skb);