Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net

Pull networking fixes from David Miller:

 1) When we run a tap on netlink sockets, we have to copy mmap'd SKBs
    instead of cloning them.  From Daniel Borkmann.

 2) When converting classical BPF into eBPF, fix the setting of the
    source reg to BPF_REG_X.  From Tycho Andersen.

 3) Fix igmpv3/mldv2 report parsing in the bridge multicast code, from
    Linus Lussing.

 4) Fix dst refcounting for ipv6 tunnels, from Martin KaFai Lau.

 5) Set NLM_F_REPLACE flag properly when replacing ipv6 routes, from
    Roopa Prabhu.

 6) Add some new cxgb4 PCI device IDs, from Hariprasad Shenai.

 7) Fix headroom tests and SKB leaks in ipv6 fragmentation code, from
    Florian Westphal.

 8) Check DMA mapping errors in bna driver, from Ivan Vecera.

 9) Several 8139cp bug fixes (dev_kfree_skb_any in interrupt context,
    misclearing of interrupt status in TX timeout handler, etc.) from
    David Woodhouse.

10) In tipc, reset SKB header pointer after skb_linearize(), from Erik
    Hugne.

11) Fix autobind races et al. in netlink code, from Herbert Xu with
    help from Tejun Heo and others.

12) Missing SET_NETDEV_DEV in sunvnet driver, from Sowmini Varadhan.

13) Fix various races in timewait timer and reqsk_queue_hadh_req, from
    Eric Dumazet.

14) Fix array overruns in mac80211, from Johannes Berg and Dan
    Carpenter.

15) Fix data race in rhashtable_rehash_one(), from Dmitriy Vyukov.

16) Fix race between poll_one_napi and napi_disable, from Neil Horman.

17) Fix byte order in geneve tunnel port config, from John W Linville.

18) Fix handling of ARP replies over lightweight tunnels, from Jiri
    Benc.

19) We can loop when fib rule dumps cross multiple SKBs, fix from Wilson
    Kok and Roopa Prabhu.

20) Several reference count handling bug fixes in the PHY/MDIO layer
    from Russel King.

21) Fix lockdep splat in ppp_dev_uninit(), from Guillaume Nault.

22) Fix crash in icmp_route_lookup(), from David Ahern.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (116 commits)
  net: Fix panic in icmp_route_lookup
  net: update docbook comment for __mdiobus_register()
  ppp: fix lockdep splat in ppp_dev_uninit()
  net: via/Kconfig: GENERIC_PCI_IOMAP required if PCI not selected
  phy: marvell: add link partner advertised modes
  net: fix net_device refcounting
  phy: add phy_device_remove()
  phy: fixed-phy: properly validate phy in fixed_phy_update_state()
  net: fix phy refcounting in a bunch of drivers
  of_mdio: fix MDIO phy device refcounting
  phy: add proper phy struct device refcounting
  phy: fix mdiobus module safety
  net: dsa: fix of_mdio_find_bus() device refcount leak
  phy: fix of_mdio_find_bus() device refcount leak
  ip6_tunnel: Reduce log level in ip6_tnl_err() to debug
  ip6_gre: Reduce log level in ip6gre_err() to debug
  fib_rules: fix fib rule dumps across multiple skbs
  bnx2x: byte swap rss_key to comply to Toeplitz specs
  net: revert "net_sched: move tp->root allocation into fw_init()"
  lwtunnel: remove source and destination UDP port config option
  ...

6 files changed, 201 insertions, 102 deletions

diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 030fefd..9001133 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -5127,13 +5127,12 @@ static void __ipv6_ifa_notify(int event, struct inet6_ifaddr *ifp)
 
 			rt = addrconf_get_prefix_route(&ifp->peer_addr, 128,
 						       ifp->idev->dev, 0, 0);
-			if (rt && ip6_del_rt(rt))
-				dst_free(&rt->dst);
+			if (rt)
+				ip6_del_rt(rt);
 		}
 		dst_hold(&ifp->rt->dst);
 
-		if (ip6_del_rt(ifp->rt))
-			dst_free(&ifp->rt->dst);
+		ip6_del_rt(ifp->rt);
 
 		rt_genid_bump_ipv6(net);
 		break;
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 418d982..7d2e002 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -155,6 +155,11 @@ static void node_free(struct fib6_node *fn)
 	kmem_cache_free(fib6_node_kmem, fn);
 }
 
+static void rt6_rcu_free(struct rt6_info *rt)
+{
+	call_rcu(&rt->dst.rcu_head, dst_rcu_free);
+}
+
 static void rt6_free_pcpu(struct rt6_info *non_pcpu_rt)
 {
 	int cpu;
@@ -169,7 +174,7 @@ static void rt6_free_pcpu(struct rt6_info *non_pcpu_rt)
 		ppcpu_rt = per_cpu_ptr(non_pcpu_rt->rt6i_pcpu, cpu);
 		pcpu_rt = *ppcpu_rt;
 		if (pcpu_rt) {
-			dst_free(&pcpu_rt->dst);
+			rt6_rcu_free(pcpu_rt);
 			*ppcpu_rt = NULL;
 		}
 	}
@@ -181,7 +186,7 @@ static void rt6_release(struct rt6_info *rt)
 {
 	if (atomic_dec_and_test(&rt->rt6i_ref)) {
 		rt6_free_pcpu(rt);
-		dst_free(&rt->dst);
+		rt6_rcu_free(rt);
 	}
 }
 
@@ -846,7 +851,7 @@ add:
 		*ins = rt;
 		rt->rt6i_node = fn;
 		atomic_inc(&rt->rt6i_ref);
-		inet6_rt_notify(RTM_NEWROUTE, rt, info);
+		inet6_rt_notify(RTM_NEWROUTE, rt, info, 0);
 		info->nl_net->ipv6.rt6_stats->fib_rt_entries++;
 
 		if (!(fn->fn_flags & RTN_RTINFO)) {
@@ -872,7 +877,7 @@ add:
 		rt->rt6i_node = fn;
 		rt->dst.rt6_next = iter->dst.rt6_next;
 		atomic_inc(&rt->rt6i_ref);
-		inet6_rt_notify(RTM_NEWROUTE, rt, info);
+		inet6_rt_notify(RTM_NEWROUTE, rt, info, NLM_F_REPLACE);
 		if (!(fn->fn_flags & RTN_RTINFO)) {
 			info->nl_net->ipv6.rt6_stats->fib_route_nodes++;
 			fn->fn_flags |= RTN_RTINFO;
@@ -933,6 +938,10 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt,
 	int replace_required = 0;
 	int sernum = fib6_new_sernum(info->nl_net);
 
+	if (WARN_ON_ONCE((rt->dst.flags & DST_NOCACHE) &&
+			 !atomic_read(&rt->dst.__refcnt)))
+		return -EINVAL;
+
 	if (info->nlh) {
 		if (!(info->nlh->nlmsg_flags & NLM_F_CREATE))
 			allow_create = 0;
@@ -1025,6 +1034,7 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt,
 		fib6_start_gc(info->nl_net, rt);
 		if (!(rt->rt6i_flags & RTF_CACHE))
 			fib6_prune_clones(info->nl_net, pn);
+		rt->dst.flags &= ~DST_NOCACHE;
 	}
 
 out:
@@ -1049,7 +1059,8 @@ out:
 			atomic_inc(&pn->leaf->rt6i_ref);
 		}
 #endif
-		dst_free(&rt->dst);
+		if (!(rt->dst.flags & DST_NOCACHE))
+			dst_free(&rt->dst);
 	}
 	return err;
 
@@ -1060,7 +1071,8 @@ out:
 st_failure:
 	if (fn && !(fn->fn_flags & (RTN_RTINFO|RTN_ROOT)))
 		fib6_repair_tree(info->nl_net, fn);
-	dst_free(&rt->dst);
+	if (!(rt->dst.flags & DST_NOCACHE))
+		dst_free(&rt->dst);
 	return err;
 #endif
 }
@@ -1410,7 +1422,7 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp,
 
 	fib6_purge_rt(rt, fn, net);
 
-	inet6_rt_notify(RTM_DELROUTE, rt, info);
+	inet6_rt_notify(RTM_DELROUTE, rt, info, 0);
 	rt6_release(rt);
 }
 
diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c
index 4038c69..3c7b931 100644
--- a/net/ipv6/ip6_gre.c
+++ b/net/ipv6/ip6_gre.c
@@ -404,13 +404,13 @@ static void ip6gre_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 		struct ipv6_tlv_tnl_enc_lim *tel;
 		__u32 mtu;
 	case ICMPV6_DEST_UNREACH:
-		net_warn_ratelimited("%s: Path to destination invalid or inactive!\n",
-				     t->parms.name);
+		net_dbg_ratelimited("%s: Path to destination invalid or inactive!\n",
+				    t->parms.name);
 		break;
 	case ICMPV6_TIME_EXCEED:
 		if (code == ICMPV6_EXC_HOPLIMIT) {
-			net_warn_ratelimited("%s: Too small hop limit or routing loop in tunnel!\n",
-					     t->parms.name);
+			net_dbg_ratelimited("%s: Too small hop limit or routing loop in tunnel!\n",
+					    t->parms.name);
 		}
 		break;
 	case ICMPV6_PARAMPROB:
@@ -421,12 +421,12 @@ static void ip6gre_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 		if (teli && teli == be32_to_cpu(info) - 2) {
 			tel = (struct ipv6_tlv_tnl_enc_lim *) &skb->data[teli];
 			if (tel->encap_limit == 0) {
-				net_warn_ratelimited("%s: Too small encapsulation limit or routing loop in tunnel!\n",
-						     t->parms.name);
+				net_dbg_ratelimited("%s: Too small encapsulation limit or routing loop in tunnel!\n",
+						    t->parms.name);
 			}
 		} else {
-			net_warn_ratelimited("%s: Recipient unable to parse tunneled packet!\n",
-					     t->parms.name);
+			net_dbg_ratelimited("%s: Recipient unable to parse tunneled packet!\n",
+					    t->parms.name);
 		}
 		break;
 	case ICMPV6_PKT_TOOBIG:
@@ -634,20 +634,20 @@ static netdev_tx_t ip6gre_xmit2(struct sk_buff *skb,
 	}
 
 	if (!fl6->flowi6_mark)
-		dst = ip6_tnl_dst_check(tunnel);
+		dst = ip6_tnl_dst_get(tunnel);
 
 	if (!dst) {
-		ndst = ip6_route_output(net, NULL, fl6);
+		dst = ip6_route_output(net, NULL, fl6);
 
-		if (ndst->error)
+		if (dst->error)
 			goto tx_err_link_failure;
-		ndst = xfrm_lookup(net, ndst, flowi6_to_flowi(fl6), NULL, 0);
-		if (IS_ERR(ndst)) {
-			err = PTR_ERR(ndst);
-			ndst = NULL;
+		dst = xfrm_lookup(net, dst, flowi6_to_flowi(fl6), NULL, 0);
+		if (IS_ERR(dst)) {
+			err = PTR_ERR(dst);
+			dst = NULL;
 			goto tx_err_link_failure;
 		}
-		dst = ndst;
+		ndst = dst;
 	}
 
 	tdev = dst->dev;
@@ -702,12 +702,9 @@ static netdev_tx_t ip6gre_xmit2(struct sk_buff *skb,
 		skb = new_skb;
 	}
 
-	if (fl6->flowi6_mark) {
-		skb_dst_set(skb, dst);
-		ndst = NULL;
-	} else {
-		skb_dst_set_noref(skb, dst);
-	}
+	if (!fl6->flowi6_mark && ndst)
+		ip6_tnl_dst_set(tunnel, ndst);
+	skb_dst_set(skb, dst);
 
 	proto = NEXTHDR_GRE;
 	if (encap_limit >= 0) {
@@ -762,14 +759,12 @@ static netdev_tx_t ip6gre_xmit2(struct sk_buff *skb,
 	skb_set_inner_protocol(skb, protocol);
 
 	ip6tunnel_xmit(NULL, skb, dev);
-	if (ndst)
-		ip6_tnl_dst_store(tunnel, ndst);
 	return 0;
 tx_err_link_failure:
 	stats->tx_carrier_errors++;
 	dst_link_failure(skb);
 tx_err_dst_release:
-	dst_release(ndst);
+	dst_release(dst);
 	return err;
 }
 
@@ -1223,6 +1218,9 @@ static const struct net_device_ops ip6gre_netdev_ops = {
 
 static void ip6gre_dev_free(struct net_device *dev)
 {
+	struct ip6_tnl *t = netdev_priv(dev);
+
+	ip6_tnl_dst_destroy(t);
 	free_percpu(dev->tstats);
 	free_netdev(dev);
 }
@@ -1245,9 +1243,10 @@ static void ip6gre_tunnel_setup(struct net_device *dev)
 	netif_keep_dst(dev);
 }
 
-static int ip6gre_tunnel_init(struct net_device *dev)
+static int ip6gre_tunnel_init_common(struct net_device *dev)
 {
 	struct ip6_tnl *tunnel;
+	int ret;
 
 	tunnel = netdev_priv(dev);
 
@@ -1255,16 +1254,37 @@ static int ip6gre_tunnel_init(struct net_device *dev)
 	tunnel->net = dev_net(dev);
 	strcpy(tunnel->parms.name, dev->name);
 
+	dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats);
+	if (!dev->tstats)
+		return -ENOMEM;
+
+	ret = ip6_tnl_dst_init(tunnel);
+	if (ret) {
+		free_percpu(dev->tstats);
+		dev->tstats = NULL;
+		return ret;
+	}
+
+	return 0;
+}
+
+static int ip6gre_tunnel_init(struct net_device *dev)
+{
+	struct ip6_tnl *tunnel;
+	int ret;
+
+	ret = ip6gre_tunnel_init_common(dev);
+	if (ret)
+		return ret;
+
+	tunnel = netdev_priv(dev);
+
 	memcpy(dev->dev_addr, &tunnel->parms.laddr, sizeof(struct in6_addr));
 	memcpy(dev->broadcast, &tunnel->parms.raddr, sizeof(struct in6_addr));
 
 	if (ipv6_addr_any(&tunnel->parms.raddr))
 		dev->header_ops = &ip6gre_header_ops;
 
-	dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats);
-	if (!dev->tstats)
-		return -ENOMEM;
-
 	return 0;
 }
 
@@ -1460,19 +1480,16 @@ static void ip6gre_netlink_parms(struct nlattr *data[],
 static int ip6gre_tap_init(struct net_device *dev)
 {
 	struct ip6_tnl *tunnel;
+	int ret;
 
-	tunnel = netdev_priv(dev);
+	ret = ip6gre_tunnel_init_common(dev);
+	if (ret)
+		return ret;
 
-	tunnel->dev = dev;
-	tunnel->net = dev_net(dev);
-	strcpy(tunnel->parms.name, dev->name);
+	tunnel = netdev_priv(dev);
 
 	ip6gre_tnl_link_config(tunnel, 1);
 
-	dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats);
-	if (!dev->tstats)
-		return -ENOMEM;
-
 	return 0;
 }
 
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 26ea479..92b1aa3 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -586,20 +586,22 @@ int ip6_fragment(struct sock *sk, struct sk_buff *skb,
 	frag_id = ipv6_select_ident(net, &ipv6_hdr(skb)->daddr,
 				    &ipv6_hdr(skb)->saddr);
 
+	hroom = LL_RESERVED_SPACE(rt->dst.dev);
 	if (skb_has_frag_list(skb)) {
 		int first_len = skb_pagelen(skb);
 		struct sk_buff *frag2;
 
 		if (first_len - hlen > mtu ||
 		    ((first_len - hlen) & 7) ||
-		    skb_cloned(skb))
+		    skb_cloned(skb) ||
+		    skb_headroom(skb) < (hroom + sizeof(struct frag_hdr)))
 			goto slow_path;
 
 		skb_walk_frags(skb, frag) {
 			/* Correct geometry. */
 			if (frag->len > mtu ||
 			    ((frag->len & 7) && frag->next) ||
-			    skb_headroom(frag) < hlen)
+			    skb_headroom(frag) < (hlen + hroom + sizeof(struct frag_hdr)))
 				goto slow_path_clean;
 
 			/* Partially cloned skb? */
@@ -616,8 +618,6 @@ int ip6_fragment(struct sock *sk, struct sk_buff *skb,
 
 		err = 0;
 		offset = 0;
-		frag = skb_shinfo(skb)->frag_list;
-		skb_frag_list_init(skb);
 		/* BUILD HEADER */
 
 		*prevhdr = NEXTHDR_FRAGMENT;
@@ -625,8 +625,11 @@ int ip6_fragment(struct sock *sk, struct sk_buff *skb,
 		if (!tmp_hdr) {
 			IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)),
 				      IPSTATS_MIB_FRAGFAILS);
-			return -ENOMEM;
+			err = -ENOMEM;
+			goto fail;
 		}
+		frag = skb_shinfo(skb)->frag_list;
+		skb_frag_list_init(skb);
 
 		__skb_pull(skb, hlen);
 		fh = (struct frag_hdr *)__skb_push(skb, sizeof(struct frag_hdr));
@@ -723,7 +726,6 @@ slow_path:
 	 */
 
 	*prevhdr = NEXTHDR_FRAGMENT;
-	hroom = LL_RESERVED_SPACE(rt->dst.dev);
 	troom = rt->dst.dev->needed_tailroom;
 
 	/*
diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c
index b0ab420..eabffbb 100644
--- a/net/ipv6/ip6_tunnel.c
+++ b/net/ipv6/ip6_tunnel.c
@@ -126,36 +126,92 @@ static struct net_device_stats *ip6_get_stats(struct net_device *dev)
  * Locking : hash tables are protected by RCU and RTNL
  */
 
-struct dst_entry *ip6_tnl_dst_check(struct ip6_tnl *t)
+static void ip6_tnl_per_cpu_dst_set(struct ip6_tnl_dst *idst,
+				    struct dst_entry *dst)
 {
-	struct dst_entry *dst = t->dst_cache;
+	write_seqlock_bh(&idst->lock);
+	dst_release(rcu_dereference_protected(
+			    idst->dst,
+			    lockdep_is_held(&idst->lock.lock)));
+	if (dst) {
+		dst_hold(dst);
+		idst->cookie = rt6_get_cookie((struct rt6_info *)dst);
+	} else {
+		idst->cookie = 0;
+	}
+	rcu_assign_pointer(idst->dst, dst);
+	write_sequnlock_bh(&idst->lock);
+}
+
+struct dst_entry *ip6_tnl_dst_get(struct ip6_tnl *t)
+{
+	struct ip6_tnl_dst *idst;
+	struct dst_entry *dst;
+	unsigned int seq;
+	u32 cookie;
 
-	if (dst && dst->obsolete &&
-	    !dst->ops->check(dst, t->dst_cookie)) {
-		t->dst_cache = NULL;
+	idst = raw_cpu_ptr(t->dst_cache);
+
+	rcu_read_lock();
+	do {
+		seq = read_seqbegin(&idst->lock);
+		dst = rcu_dereference(idst->dst);
+		cookie = idst->cookie;
+	} while (read_seqretry(&idst->lock, seq));
+
+	if (dst && !atomic_inc_not_zero(&dst->__refcnt))
+		dst = NULL;
+	rcu_read_unlock();
+
+	if (dst && dst->obsolete && !dst->ops->check(dst, cookie)) {
+		ip6_tnl_per_cpu_dst_set(idst, NULL);
 		dst_release(dst);
-		return NULL;
+		dst = NULL;
 	}
-
 	return dst;
 }
-EXPORT_SYMBOL_GPL(ip6_tnl_dst_check);
+EXPORT_SYMBOL_GPL(ip6_tnl_dst_get);
 
 void ip6_tnl_dst_reset(struct ip6_tnl *t)
 {
-	dst_release(t->dst_cache);
-	t->dst_cache = NULL;
+	int i;
+
+	for_each_possible_cpu(i)
+		ip6_tnl_per_cpu_dst_set(raw_cpu_ptr(t->dst_cache), NULL);
 }
 EXPORT_SYMBOL_GPL(ip6_tnl_dst_reset);
 
-void ip6_tnl_dst_store(struct ip6_tnl *t, struct dst_entry *dst)
+void ip6_tnl_dst_set(struct ip6_tnl *t, struct dst_entry *dst)
+{
+	ip6_tnl_per_cpu_dst_set(raw_cpu_ptr(t->dst_cache), dst);
+
+}
+EXPORT_SYMBOL_GPL(ip6_tnl_dst_set);
+
+void ip6_tnl_dst_destroy(struct ip6_tnl *t)
 {
-	struct rt6_info *rt = (struct rt6_info *) dst;
-	t->dst_cookie = rt6_get_cookie(rt);
-	dst_release(t->dst_cache);
-	t->dst_cache = dst;
+	if (!t->dst_cache)
+		return;
+
+	ip6_tnl_dst_reset(t);
+	free_percpu(t->dst_cache);
 }
-EXPORT_SYMBOL_GPL(ip6_tnl_dst_store);
+EXPORT_SYMBOL_GPL(ip6_tnl_dst_destroy);
+
+int ip6_tnl_dst_init(struct ip6_tnl *t)
+{
+	int i;
+
+	t->dst_cache = alloc_percpu(struct ip6_tnl_dst);
+	if (!t->dst_cache)
+		return -ENOMEM;
+
+	for_each_possible_cpu(i)
+		seqlock_init(&per_cpu_ptr(t->dst_cache, i)->lock);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ip6_tnl_dst_init);
 
 /**
  * ip6_tnl_lookup - fetch tunnel matching the end-point addresses
@@ -271,6 +327,9 @@ ip6_tnl_unlink(struct ip6_tnl_net *ip6n, struct ip6_tnl *t)
 
 static void ip6_dev_free(struct net_device *dev)
 {
+	struct ip6_tnl *t = netdev_priv(dev);
+
+	ip6_tnl_dst_destroy(t);
 	free_percpu(dev->tstats);
 	free_netdev(dev);
 }
@@ -510,14 +569,14 @@ ip6_tnl_err(struct sk_buff *skb, __u8 ipproto, struct inet6_skb_parm *opt,
 		struct ipv6_tlv_tnl_enc_lim *tel;
 		__u32 mtu;
 	case ICMPV6_DEST_UNREACH:
-		net_warn_ratelimited("%s: Path to destination invalid or inactive!\n",
-				     t->parms.name);
+		net_dbg_ratelimited("%s: Path to destination invalid or inactive!\n",
+				    t->parms.name);
 		rel_msg = 1;
 		break;
 	case ICMPV6_TIME_EXCEED:
 		if ((*code) == ICMPV6_EXC_HOPLIMIT) {
-			net_warn_ratelimited("%s: Too small hop limit or routing loop in tunnel!\n",
-					     t->parms.name);
+			net_dbg_ratelimited("%s: Too small hop limit or routing loop in tunnel!\n",
+					    t->parms.name);
 			rel_msg = 1;
 		}
 		break;
@@ -529,13 +588,13 @@ ip6_tnl_err(struct sk_buff *skb, __u8 ipproto, struct inet6_skb_parm *opt,
 		if (teli && teli == *info - 2) {
 			tel = (struct ipv6_tlv_tnl_enc_lim *) &skb->data[teli];
 			if (tel->encap_limit == 0) {
-				net_warn_ratelimited("%s: Too small encapsulation limit or routing loop in tunnel!\n",
-						     t->parms.name);
+				net_dbg_ratelimited("%s: Too small encapsulation limit or routing loop in tunnel!\n",
+						    t->parms.name);
 				rel_msg = 1;
 			}
 		} else {
-			net_warn_ratelimited("%s: Recipient unable to parse tunneled packet!\n",
-					     t->parms.name);
+			net_dbg_ratelimited("%s: Recipient unable to parse tunneled packet!\n",
+					    t->parms.name);
 		}
 		break;
 	case ICMPV6_PKT_TOOBIG:
@@ -1010,23 +1069,23 @@ static int ip6_tnl_xmit2(struct sk_buff *skb,
 		memcpy(&fl6->daddr, addr6, sizeof(fl6->daddr));
 		neigh_release(neigh);
 	} else if (!fl6->flowi6_mark)
-		dst = ip6_tnl_dst_check(t);
+		dst = ip6_tnl_dst_get(t);
 
 	if (!ip6_tnl_xmit_ctl(t, &fl6->saddr, &fl6->daddr))
 		goto tx_err_link_failure;
 
 	if (!dst) {
-		ndst = ip6_route_output(net, NULL, fl6);
+		dst = ip6_route_output(net, NULL, fl6);
 
-		if (ndst->error)
+		if (dst->error)
 			goto tx_err_link_failure;
-		ndst = xfrm_lookup(net, ndst, flowi6_to_flowi(fl6), NULL, 0);
-		if (IS_ERR(ndst)) {
-			err = PTR_ERR(ndst);
-			ndst = NULL;
+		dst = xfrm_lookup(net, dst, flowi6_to_flowi(fl6), NULL, 0);
+		if (IS_ERR(dst)) {
+			err = PTR_ERR(dst);
+			dst = NULL;
 			goto tx_err_link_failure;
 		}
-		dst = ndst;
+		ndst = dst;
 	}
 
 	tdev = dst->dev;
@@ -1072,12 +1131,11 @@ static int ip6_tnl_xmit2(struct sk_buff *skb,
 		consume_skb(skb);
 		skb = new_skb;
 	}
-	if (fl6->flowi6_mark) {
-		skb_dst_set(skb, dst);
-		ndst = NULL;
-	} else {
-		skb_dst_set_noref(skb, dst);
-	}
+
+	if (!fl6->flowi6_mark && ndst)
+		ip6_tnl_dst_set(t, ndst);
+	skb_dst_set(skb, dst);
+
 	skb->transport_header = skb->network_header;
 
 	proto = fl6->flowi6_proto;
@@ -1101,14 +1159,12 @@ static int ip6_tnl_xmit2(struct sk_buff *skb,
 	ipv6h->saddr = fl6->saddr;
 	ipv6h->daddr = fl6->daddr;
 	ip6tunnel_xmit(NULL, skb, dev);
-	if (ndst)
-		ip6_tnl_dst_store(t, ndst);
 	return 0;
 tx_err_link_failure:
 	stats->tx_carrier_errors++;
 	dst_link_failure(skb);
 tx_err_dst_release:
-	dst_release(ndst);
+	dst_release(dst);
 	return err;
 }
 
@@ -1573,12 +1629,21 @@ static inline int
 ip6_tnl_dev_init_gen(struct net_device *dev)
 {
 	struct ip6_tnl *t = netdev_priv(dev);
+	int ret;
 
 	t->dev = dev;
 	t->net = dev_net(dev);
 	dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats);
 	if (!dev->tstats)
 		return -ENOMEM;
+
+	ret = ip6_tnl_dst_init(t);
+	if (ret) {
+		free_percpu(dev->tstats);
+		dev->tstats = NULL;
+		return ret;
+	}
+
 	return 0;
 }
 
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 53617d7..f204089 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -1322,8 +1322,7 @@ static void ip6_link_failure(struct sk_buff *skb)
 	if (rt) {
 		if (rt->rt6i_flags & RTF_CACHE) {
 			dst_hold(&rt->dst);
-			if (ip6_del_rt(rt))
-				dst_free(&rt->dst);
+			ip6_del_rt(rt);
 		} else if (rt->rt6i_node && (rt->rt6i_flags & RTF_DEFAULT)) {
 			rt->rt6i_node->fn_sernum = -1;
 		}
@@ -1886,9 +1885,11 @@ int ip6_route_info_create(struct fib6_config *cfg, struct rt6_info **rt_ret)
 			rt->dst.input = ip6_pkt_prohibit;
 			break;
 		case RTN_THROW:
+		case RTN_UNREACHABLE:
 		default:
 			rt->dst.error = (cfg->fc_type == RTN_THROW) ? -EAGAIN
-					: -ENETUNREACH;
+					: (cfg->fc_type == RTN_UNREACHABLE)
+					? -EHOSTUNREACH : -ENETUNREACH;
 			rt->dst.output = ip6_pkt_discard_out;
 			rt->dst.input = ip6_pkt_discard;
 			break;
@@ -2028,7 +2029,8 @@ static int __ip6_del_rt(struct rt6_info *rt, struct nl_info *info)
 	struct fib6_table *table;
 	struct net *net = dev_net(rt->dst.dev);
 
-	if (rt == net->ipv6.ip6_null_entry) {
+	if (rt == net->ipv6.ip6_null_entry ||
+	    rt->dst.flags & DST_NOCACHE) {
 		err = -ENOENT;
 		goto out;
 	}
@@ -2515,6 +2517,7 @@ struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev,
 	rt->rt6i_dst.addr = *addr;
 	rt->rt6i_dst.plen = 128;
 	rt->rt6i_table = fib6_get_table(net, RT6_TABLE_LOCAL);
+	rt->dst.flags |= DST_NOCACHE;
 
 	atomic_set(&rt->dst.__refcnt, 1);
 
@@ -3303,7 +3306,8 @@ errout:
 	return err;
 }
 
-void inet6_rt_notify(int event, struct rt6_info *rt, struct nl_info *info)
+void inet6_rt_notify(int event, struct rt6_info *rt, struct nl_info *info,
+		     unsigned int nlm_flags)
 {
 	struct sk_buff *skb;
 	struct net *net = info->nl_net;
@@ -3318,7 +3322,7 @@ void inet6_rt_notify(int event, struct rt6_info *rt, struct nl_info *info)
 		goto errout;
 
 	err = rt6_fill_node(net, skb, rt, NULL, NULL, 0,
-				event, info->portid, seq, 0, 0, 0);
+				event, info->portid, seq, 0, 0, nlm_flags);
 	if (err < 0) {
 		/* -EMSGSIZE implies BUG in rt6_nlmsg_size() */
 		WARN_ON(err == -EMSGSIZE);