Merge branch 'mpls-next'

Eric W. Biederman says:

====================
mpls: Minor fixes and cleanups

This is a bunch of small changes that have come out of the discussions
of the mpls code and the automated tests that people run against things.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

include/net/neighbour.h

@@ -226,6 +226,7 @@ enum {
 	NEIGH_ND_TABLE = 1,
 	NEIGH_DN_TABLE = 2,
 	NEIGH_NR_TABLES,
+	NEIGH_LINK_TABLE = NEIGH_NR_TABLES /* Pseudo table for neigh_xmit */
 };
 
 static inline int neigh_parms_family(struct neigh_parms *p)

net/core/neighbour.c

@@ -2391,22 +2391,15 @@ void __neigh_for_each_release(struct neigh_table *tbl,
 }
 EXPORT_SYMBOL(__neigh_for_each_release);
 
-int neigh_xmit(int family, struct net_device *dev,
+int neigh_xmit(int index, struct net_device *dev,
 	       const void *addr, struct sk_buff *skb)
 {
-	int err;
-	if (family == AF_PACKET) {
-		err = dev_hard_header(skb, dev, ntohs(skb->protocol),
-				      addr, NULL, skb->len);
-		if (err < 0)
-			goto out_kfree_skb;
-		err = dev_queue_xmit(skb);
-	} else {
+	int err = -EAFNOSUPPORT;
+	if (likely(index < NEIGH_NR_TABLES)) {
 		struct neigh_table *tbl;
 		struct neighbour *neigh;
 
-		err = -ENETDOWN;
-		tbl = neigh_find_table(family);
+		tbl = neigh_tables[index];
 		if (!tbl)
 			goto out;
 		neigh = __neigh_lookup_noref(tbl, addr, dev);
@@ -2417,6 +2410,13 @@ int neigh_xmit(int family, struct net_device *dev,
 			goto out_kfree_skb;
 		err = neigh->output(neigh, skb);
 	}
+	else if (index == NEIGH_LINK_TABLE) {
+		err = dev_hard_header(skb, dev, ntohs(skb->protocol),
+				      addr, NULL, skb->len);
+		if (err < 0)
+			goto out_kfree_skb;
+		err = dev_queue_xmit(skb);
+	}
 out:
 	return err;
 out_kfree_skb:

net/mpls/af_mpls.c

@@ -24,13 +24,13 @@
 #define MAX_VIA_ALEN (ALIGN(MAX_ADDR_LEN, sizeof(unsigned long)))
 
 struct mpls_route { /* next hop label forwarding entry */
-	struct net_device 	*rt_dev;
+	struct net_device __rcu *rt_dev;
 	struct rcu_head		rt_rcu;
 	u32			rt_label[MAX_NEW_LABELS];
 	u8			rt_protocol; /* routing protocol that set this entry */
-	u8			rt_labels:2,
-				rt_via_alen:6;
-	unsigned short		rt_via_family;
+	u8			rt_labels;
+	u8			rt_via_alen;
+	u8			rt_via_table;
 	u8			rt_via[0];
 };
 
@@ -152,7 +152,7 @@ static int mpls_forward(struct sk_buff *skb, struct net_device *dev,
 		goto drop;
 
 	/* Find the output device */
-	out_dev = rt->rt_dev;
+	out_dev = rcu_dereference(rt->rt_dev);
 	if (!mpls_output_possible(out_dev))
 		goto drop;
 
@@ -162,7 +162,7 @@ static int mpls_forward(struct sk_buff *skb, struct net_device *dev,
 	skb_forward_csum(skb);
 
 	/* Verify ttl is valid */
-	if (dec.ttl <= 2)
+	if (dec.ttl <= 1)
 		goto drop;
 	dec.ttl -= 1;
 
@@ -201,7 +201,7 @@ static int mpls_forward(struct sk_buff *skb, struct net_device *dev,
 		}
 	}
 
-	err = neigh_xmit(rt->rt_via_family, out_dev, rt->rt_via, skb);
+	err = neigh_xmit(rt->rt_via_table, out_dev, rt->rt_via, skb);
 	if (err)
 		net_dbg_ratelimited("%s: packet transmission failed: %d\n",
 				    __func__, err);
@@ -225,7 +225,7 @@ static const struct nla_policy rtm_mpls_policy[RTA_MAX+1] = {
 struct mpls_route_config {
 	u32		rc_protocol;
 	u32		rc_ifindex;
-	u16		rc_via_family;
+	u16		rc_via_table;
 	u16		rc_via_alen;
 	u8		rc_via[MAX_VIA_ALEN];
 	u32		rc_label;
@@ -239,7 +239,7 @@ static struct mpls_route *mpls_rt_alloc(size_t alen)
 {
 	struct mpls_route *rt;
 
-	rt = kzalloc(GFP_KERNEL, sizeof(*rt) + alen);
+	rt = kzalloc(sizeof(*rt) + alen, GFP_KERNEL);
 	if (rt)
 		rt->rt_via_alen = alen;
 	return rt;
@@ -269,13 +269,15 @@ static void mpls_route_update(struct net *net, unsigned index,
 			      struct net_device *dev, struct mpls_route *new,
 			      const struct nl_info *info)
 {
+	struct mpls_route __rcu **platform_label;
 	struct mpls_route *rt, *old = NULL;
 
 	ASSERT_RTNL();
 
-	rt = net->mpls.platform_label[index];
-	if (!dev || (rt && (rt->rt_dev == dev))) {
-		rcu_assign_pointer(net->mpls.platform_label[index], new);
+	platform_label = rtnl_dereference(net->mpls.platform_label);
+	rt = rtnl_dereference(platform_label[index]);
+	if (!dev || (rt && (rtnl_dereference(rt->rt_dev) == dev))) {
+		rcu_assign_pointer(platform_label[index], new);
 		old = rt;
 	}
 
@@ -287,9 +289,14 @@ static void mpls_route_update(struct net *net, unsigned index,
 
 static unsigned find_free_label(struct net *net)
 {
+	struct mpls_route __rcu **platform_label;
+	size_t platform_labels;
 	unsigned index;
-	for (index = 16; index < net->mpls.platform_labels; index++) {
-		if (!net->mpls.platform_label[index])
+
+	platform_label = rtnl_dereference(net->mpls.platform_label);
+	platform_labels = net->mpls.platform_labels;
+	for (index = 16; index < platform_labels; index++) {
+		if (!rtnl_dereference(platform_label[index]))
 			return index;
 	}
 	return LABEL_NOT_SPECIFIED;
@@ -297,6 +304,7 @@ static unsigned find_free_label(struct net *net)
 
 static int mpls_route_add(struct mpls_route_config *cfg)
 {
+	struct mpls_route __rcu **platform_label;
 	struct net *net = cfg->rc_nlinfo.nl_net;
 	struct net_device *dev = NULL;
 	struct mpls_route *rt, *old;
@@ -335,17 +343,18 @@ static int mpls_route_add(struct mpls_route_config *cfg)
 		goto errout;
 
 	err = -EINVAL;
-	if ((cfg->rc_via_family == AF_PACKET) &&
+	if ((cfg->rc_via_table == NEIGH_LINK_TABLE) &&
 	    (dev->addr_len != cfg->rc_via_alen))
 		goto errout;
 
 	/* Append makes no sense with mpls */
-	err = -EINVAL;
+	err = -EOPNOTSUPP;
 	if (cfg->rc_nlflags & NLM_F_APPEND)
 		goto errout;
 
 	err = -EEXIST;
-	old = net->mpls.platform_label[index];
+	platform_label = rtnl_dereference(net->mpls.platform_label);
+	old = rtnl_dereference(platform_label[index]);
 	if ((cfg->rc_nlflags & NLM_F_EXCL) && old)
 		goto errout;
 
@@ -366,8 +375,8 @@ static int mpls_route_add(struct mpls_route_config *cfg)
 	for (i = 0; i < rt->rt_labels; i++)
 		rt->rt_label[i] = cfg->rc_output_label[i];
 	rt->rt_protocol = cfg->rc_protocol;
-	rt->rt_dev = dev;
-	rt->rt_via_family = cfg->rc_via_family;
+	RCU_INIT_POINTER(rt->rt_dev, dev);
+	rt->rt_via_table = cfg->rc_via_table;
 	memcpy(rt->rt_via, cfg->rc_via, cfg->rc_via_alen);
 
 	mpls_route_update(net, index, NULL, rt, &cfg->rc_nlinfo);
@@ -406,14 +415,16 @@ static int mpls_route_del(struct mpls_route_config *cfg)
 
 static void mpls_ifdown(struct net_device *dev)
 {
+	struct mpls_route __rcu **platform_label;
 	struct net *net = dev_net(dev);
 	unsigned index;
 
+	platform_label = rtnl_dereference(net->mpls.platform_label);
 	for (index = 0; index < net->mpls.platform_labels; index++) {
-		struct mpls_route *rt = net->mpls.platform_label[index];
+		struct mpls_route *rt = rtnl_dereference(platform_label[index]);
 		if (!rt)
 			continue;
-		if (rt->rt_dev != dev)
+		if (rtnl_dereference(rt->rt_dev) != dev)
 			continue;
 		rt->rt_dev = NULL;
 	}
@@ -437,15 +448,22 @@ static struct notifier_block mpls_dev_notifier = {
 };
 
 static int nla_put_via(struct sk_buff *skb,
-		       u16 family, const void *addr, int alen)
+		       u8 table, const void *addr, int alen)
 {
+	static const int table_to_family[NEIGH_NR_TABLES + 1] = {
+		AF_INET, AF_INET6, AF_DECnet, AF_PACKET,
+	};
 	struct nlattr *nla;
 	struct rtvia *via;
+	int family = AF_UNSPEC;
 
 	nla = nla_reserve(skb, RTA_VIA, alen + 2);
 	if (!nla)
 		return -EMSGSIZE;
 
+	if (table <= NEIGH_NR_TABLES)
+		family = table_to_family[table];
+
 	via = nla_data(nla);
 	via->rtvia_family = family;
 	memcpy(via->rtvia_addr, addr, alen);
@@ -588,21 +606,23 @@ static int rtm_to_route_config(struct sk_buff *skb,  struct nlmsghdr *nlh,
 			struct rtvia *via = nla_data(nla);
 			if (nla_len(nla) < offsetof(struct rtvia, rtvia_addr))
 				goto errout;
-			cfg->rc_via_family = via->rtvia_family;
 			cfg->rc_via_alen   = nla_len(nla) -
 				offsetof(struct rtvia, rtvia_addr);
 			if (cfg->rc_via_alen > MAX_VIA_ALEN)
 				goto errout;
 
 			/* Validate the address family */
-			switch(cfg->rc_via_family) {
+			switch(via->rtvia_family) {
 			case AF_PACKET:
+				cfg->rc_via_table = NEIGH_LINK_TABLE;
 				break;
 			case AF_INET:
+				cfg->rc_via_table = NEIGH_ARP_TABLE;
 				if (cfg->rc_via_alen != 4)
 					goto errout;
 				break;
 			case AF_INET6:
+				cfg->rc_via_table = NEIGH_ND_TABLE;
 				if (cfg->rc_via_alen != 16)
 					goto errout;
 				break;
@@ -653,6 +673,7 @@ static int mpls_rtm_newroute(struct sk_buff *skb, struct nlmsghdr *nlh)
 static int mpls_dump_route(struct sk_buff *skb, u32 portid, u32 seq, int event,
 			   u32 label, struct mpls_route *rt, int flags)
 {
+	struct net_device *dev;
 	struct nlmsghdr *nlh;
 	struct rtmsg *rtm;
 
@@ -674,9 +695,10 @@ static int mpls_dump_route(struct sk_buff *skb, u32 portid, u32 seq, int event,
 	if (rt->rt_labels &&
 	    nla_put_labels(skb, RTA_NEWDST, rt->rt_labels, rt->rt_label))
 		goto nla_put_failure;
-	if (nla_put_via(skb, rt->rt_via_family, rt->rt_via, rt->rt_via_alen))
+	if (nla_put_via(skb, rt->rt_via_table, rt->rt_via, rt->rt_via_alen))
 		goto nla_put_failure;
-	if (rt->rt_dev && nla_put_u32(skb, RTA_OIF, rt->rt_dev->ifindex))
+	dev = rtnl_dereference(rt->rt_dev);
+	if (dev && nla_put_u32(skb, RTA_OIF, dev->ifindex))
 		goto nla_put_failure;
 	if (nla_put_labels(skb, RTA_DST, 1, &label))
 		goto nla_put_failure;
@@ -692,6 +714,8 @@ static int mpls_dump_route(struct sk_buff *skb, u32 portid, u32 seq, int event,
 static int mpls_dump_routes(struct sk_buff *skb, struct netlink_callback *cb)
 {
 	struct net *net = sock_net(skb->sk);
+	struct mpls_route __rcu **platform_label;
+	size_t platform_labels;
 	unsigned int index;
 
 	ASSERT_RTNL();
@@ -700,9 +724,11 @@ static int mpls_dump_routes(struct sk_buff *skb, struct netlink_callback *cb)
 	if (index < 16)
 		index = 16;
 
-	for (; index < net->mpls.platform_labels; index++) {
+	platform_label = rtnl_dereference(net->mpls.platform_label);
+	platform_labels = net->mpls.platform_labels;
+	for (; index < platform_labels; index++) {
 		struct mpls_route *rt;
-		rt = net->mpls.platform_label[index];
+		rt = rtnl_dereference(platform_label[index]);
 		if (!rt)
 			continue;
 
@@ -780,9 +806,9 @@ static int resize_platform_label_table(struct net *net, size_t limit)
 		rt0 = mpls_rt_alloc(lo->addr_len);
 		if (!rt0)
 			goto nort0;
-		rt0->rt_dev = lo;
+		RCU_INIT_POINTER(rt0->rt_dev, lo);
 		rt0->rt_protocol = RTPROT_KERNEL;
-		rt0->rt_via_family = AF_PACKET;
+		rt0->rt_via_table = NEIGH_LINK_TABLE;
 		memcpy(rt0->rt_via, lo->dev_addr, lo->addr_len);
 	}
 	if (limit > LABEL_IPV6_EXPLICIT_NULL) {
@@ -790,15 +816,15 @@ static int resize_platform_label_table(struct net *net, size_t limit)
 		rt2 = mpls_rt_alloc(lo->addr_len);
 		if (!rt2)
 			goto nort2;
-		rt2->rt_dev = lo;
+		RCU_INIT_POINTER(rt2->rt_dev, lo);
 		rt2->rt_protocol = RTPROT_KERNEL;
-		rt2->rt_via_family = AF_PACKET;
+		rt2->rt_via_table = NEIGH_LINK_TABLE;
 		memcpy(rt2->rt_via, lo->dev_addr, lo->addr_len);
 	}
 
 	rtnl_lock();
 	/* Remember the original table */
-	old = net->mpls.platform_label;
+	old = rtnl_dereference(net->mpls.platform_label);
 	old_limit = net->mpls.platform_labels;
 
 	/* Free any labels beyond the new table */
@@ -815,19 +841,19 @@ static int resize_platform_label_table(struct net *net, size_t limit)
 	/* If needed set the predefined labels */
 	if ((old_limit <= LABEL_IPV6_EXPLICIT_NULL) &&
 	    (limit > LABEL_IPV6_EXPLICIT_NULL)) {
-		labels[LABEL_IPV6_EXPLICIT_NULL] = rt2;
+		RCU_INIT_POINTER(labels[LABEL_IPV6_EXPLICIT_NULL], rt2);
 		rt2 = NULL;
 	}
 
 	if ((old_limit <= LABEL_IPV4_EXPLICIT_NULL) &&
 	    (limit > LABEL_IPV4_EXPLICIT_NULL)) {
-		labels[LABEL_IPV4_EXPLICIT_NULL] = rt0;
+		RCU_INIT_POINTER(labels[LABEL_IPV4_EXPLICIT_NULL], rt0);
 		rt0 = NULL;
 	}
 
 	/* Update the global pointers */
 	net->mpls.platform_labels = limit;
-	net->mpls.platform_label = labels;
+	rcu_assign_pointer(net->mpls.platform_label, labels);
 
 	rtnl_unlock();
 
@@ -903,6 +929,8 @@ static int mpls_net_init(struct net *net)
 
 static void mpls_net_exit(struct net *net)
 {
+	struct mpls_route __rcu **platform_label;
+	size_t platform_labels;
 	struct ctl_table *table;
 	unsigned int index;
 
@@ -910,8 +938,8 @@ static void mpls_net_exit(struct net *net)
 	unregister_net_sysctl_table(net->mpls.ctl);
 	kfree(table);
 
-	/* An rcu grace period haselapsed since there was a device in
-	 * the network namespace (and thus the last in fqlight packet)
+	/* An rcu grace period has passed since there was a device in
+	 * the network namespace (and thus the last in flight packet)
 	 * left this network namespace.  This is because
 	 * unregister_netdevice_many and netdev_run_todo has completed
 	 * for each network device that was in this network namespace.
@@ -920,14 +948,16 @@ static void mpls_net_exit(struct net *net)
 	 * freeing the platform_label table.
 	 */
 	rtnl_lock();
-	for (index = 0; index < net->mpls.platform_labels; index++) {
-		struct mpls_route *rt = net->mpls.platform_label[index];
-		rcu_assign_pointer(net->mpls.platform_label[index], NULL);
+	platform_label = rtnl_dereference(net->mpls.platform_label);
+	platform_labels = net->mpls.platform_labels;
+	for (index = 0; index < platform_labels; index++) {
+		struct mpls_route *rt = rtnl_dereference(platform_label[index]);
+		RCU_INIT_POINTER(platform_label[index], NULL);
 		mpls_rt_free(rt);
 	}
 	rtnl_unlock();
 
-	kvfree(net->mpls.platform_label);
+	kvfree(platform_label);
 }
 
 static struct pernet_operations mpls_net_ops = {

net/openvswitch/Kconfig

@@ -6,6 +6,7 @@ config OPENVSWITCH
 	tristate "Open vSwitch"
 	depends on INET
 	select LIBCRC32C
+	select MPLS
 	select NET_MPLS_GSO
 	---help---
 	  Open vSwitch is a multilayer Ethernet switch targeted at virtualized