Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6
[safe/jmp/linux-2.6] / net / core / neighbour.c
index 6ba1cde..d8d0ca9 100644 (file)
  *     Harald Welte            Add neighbour cache statistics like rtstat
  */
 
-#include <linux/config.h>
 #include <linux/types.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/socket.h>
-#include <linux/sched.h>
 #include <linux/netdevice.h>
 #include <linux/proc_fs.h>
 #ifdef CONFIG_SYSCTL
 #include <linux/sysctl.h>
 #endif
 #include <linux/times.h>
+#include <net/net_namespace.h>
 #include <net/neighbour.h>
 #include <net/dst.h>
 #include <net/sock.h>
+#include <net/netevent.h>
+#include <net/netlink.h>
 #include <linux/rtnetlink.h>
 #include <linux/random.h>
 #include <linux/string.h>
+#include <linux/log2.h>
 
 #define NEIGH_DEBUG 1
 
 #define PNEIGH_HASHMASK                0xF
 
 static void neigh_timer_handler(unsigned long arg);
-#ifdef CONFIG_ARPD
-static void neigh_app_notify(struct neighbour *n);
-#endif
+static void __neigh_notify(struct neighbour *n, int type, int flags);
+static void neigh_update_notify(struct neighbour *neigh);
 static int pneigh_ifdown(struct neigh_table *tbl, struct net_device *dev);
-void neigh_changeaddr(struct neigh_table *tbl, struct net_device *dev);
 
 static struct neigh_table *neigh_tables;
 #ifdef CONFIG_PROC_FS
-static struct file_operations neigh_stat_seq_fops;
+static const struct file_operations neigh_stat_seq_fops;
 #endif
 
 /*
@@ -104,6 +104,15 @@ static int neigh_blackhole(struct sk_buff *skb)
        return -ENETDOWN;
 }
 
+static void neigh_cleanup_and_release(struct neighbour *neigh)
+{
+       if (neigh->parms->neigh_cleanup)
+               neigh->parms->neigh_cleanup(neigh);
+
+       __neigh_notify(neigh, RTM_DELNEIGH, 0);
+       neigh_release(neigh);
+}
+
 /*
  * It is random distribution in the interval (1/2)*base...(3/2)*base.
  * It corresponds to default IPv6 settings and is not overridable,
@@ -140,7 +149,7 @@ static int neigh_forced_gc(struct neigh_table *tbl)
                                n->dead = 1;
                                shrunk  = 1;
                                write_unlock(&n->lock);
-                               neigh_release(n);
+                               neigh_cleanup_and_release(n);
                                continue;
                        }
                        write_unlock(&n->lock);
@@ -155,6 +164,16 @@ static int neigh_forced_gc(struct neigh_table *tbl)
        return shrunk;
 }
 
+static void neigh_add_timer(struct neighbour *n, unsigned long when)
+{
+       neigh_hold(n);
+       if (unlikely(mod_timer(&n->timer, when))) {
+               printk("NEIGH: BUG, double timer add, state is %x\n",
+                      n->nud_state);
+               dump_stack();
+       }
+}
+
 static int neigh_del_timer(struct neighbour *n)
 {
        if ((n->nud_state & NUD_IN_TIMER) &&
@@ -211,7 +230,7 @@ static void neigh_flush_dev(struct neigh_table *tbl, struct net_device *dev)
                                NEIGH_PRINTK2("neigh %p is stray.\n", n);
                        }
                        write_unlock(&n->lock);
-                       neigh_release(n);
+                       neigh_cleanup_and_release(n);
                }
        }
 }
@@ -250,21 +269,17 @@ static struct neighbour *neigh_alloc(struct neigh_table *tbl)
                        goto out_entries;
        }
 
-       n = kmem_cache_alloc(tbl->kmem_cachep, SLAB_ATOMIC);
+       n = kmem_cache_zalloc(tbl->kmem_cachep, GFP_ATOMIC);
        if (!n)
                goto out_entries;
 
-       memset(n, 0, tbl->entry_size);
-
        skb_queue_head_init(&n->arp_queue);
        rwlock_init(&n->lock);
        n->updated        = n->used = now;
        n->nud_state      = NUD_NONE;
        n->output         = neigh_blackhole;
        n->parms          = neigh_parms_clone(&tbl->parms);
-       init_timer(&n->timer);
-       n->timer.function = neigh_timer_handler;
-       n->timer.data     = (unsigned long)n;
+       setup_timer(&n->timer, neigh_timer_handler, (unsigned long)n);
 
        NEIGH_CACHE_STAT_INC(tbl, allocs);
        n->tbl            = tbl;
@@ -284,14 +299,11 @@ static struct neighbour **neigh_hash_alloc(unsigned int entries)
        struct neighbour **ret;
 
        if (size <= PAGE_SIZE) {
-               ret = kmalloc(size, GFP_ATOMIC);
+               ret = kzalloc(size, GFP_ATOMIC);
        } else {
                ret = (struct neighbour **)
-                       __get_free_pages(GFP_ATOMIC, get_order(size));
+                     __get_free_pages(GFP_ATOMIC|__GFP_ZERO, get_order(size));
        }
-       if (ret)
-               memset(ret, 0, size);
-
        return ret;
 }
 
@@ -312,7 +324,7 @@ static void neigh_hash_grow(struct neigh_table *tbl, unsigned long new_entries)
 
        NEIGH_CACHE_STAT_INC(tbl, hash_grows);
 
-       BUG_ON(new_entries & (new_entries - 1));
+       BUG_ON(!is_power_of_2(new_entries));
        new_hash = neigh_hash_alloc(new_entries);
        if (!new_hash)
                return;
@@ -346,12 +358,13 @@ struct neighbour *neigh_lookup(struct neigh_table *tbl, const void *pkey,
 {
        struct neighbour *n;
        int key_len = tbl->key_len;
-       u32 hash_val = tbl->hash(pkey, dev) & tbl->hash_mask;
-       
+       u32 hash_val;
+
        NEIGH_CACHE_STAT_INC(tbl, lookups);
 
        read_lock_bh(&tbl->lock);
-       for (n = tbl->hash_buckets[hash_val]; n; n = n->next) {
+       hash_val = tbl->hash(pkey, dev);
+       for (n = tbl->hash_buckets[hash_val & tbl->hash_mask]; n; n = n->next) {
                if (dev == n->dev && !memcmp(n->primary_key, pkey, key_len)) {
                        neigh_hold(n);
                        NEIGH_CACHE_STAT_INC(tbl, hits);
@@ -362,17 +375,20 @@ struct neighbour *neigh_lookup(struct neigh_table *tbl, const void *pkey,
        return n;
 }
 
-struct neighbour *neigh_lookup_nodev(struct neigh_table *tbl, const void *pkey)
+struct neighbour *neigh_lookup_nodev(struct neigh_table *tbl, struct net *net,
+                                    const void *pkey)
 {
        struct neighbour *n;
        int key_len = tbl->key_len;
-       u32 hash_val = tbl->hash(pkey, NULL) & tbl->hash_mask;
+       u32 hash_val;
 
        NEIGH_CACHE_STAT_INC(tbl, lookups);
 
        read_lock_bh(&tbl->lock);
-       for (n = tbl->hash_buckets[hash_val]; n; n = n->next) {
-               if (!memcmp(n->primary_key, pkey, key_len)) {
+       hash_val = tbl->hash(pkey, NULL);
+       for (n = tbl->hash_buckets[hash_val & tbl->hash_mask]; n; n = n->next) {
+               if (!memcmp(n->primary_key, pkey, key_len) &&
+                   net_eq(dev_net(n->dev), net)) {
                        neigh_hold(n);
                        NEIGH_CACHE_STAT_INC(tbl, hits);
                        break;
@@ -450,7 +466,30 @@ out_neigh_release:
        goto out;
 }
 
-struct pneigh_entry * pneigh_lookup(struct neigh_table *tbl, const void *pkey,
+struct pneigh_entry *__pneigh_lookup(struct neigh_table *tbl,
+               struct net *net, const void *pkey, struct net_device *dev)
+{
+       struct pneigh_entry *n;
+       int key_len = tbl->key_len;
+       u32 hash_val = *(u32 *)(pkey + key_len - 4);
+
+       hash_val ^= (hash_val >> 16);
+       hash_val ^= hash_val >> 8;
+       hash_val ^= hash_val >> 4;
+       hash_val &= PNEIGH_HASHMASK;
+
+       for (n = tbl->phash_buckets[hash_val]; n; n = n->next) {
+               if (!memcmp(n->key, pkey, key_len) &&
+                   (pneigh_net(n) == net) &&
+                   (n->dev == dev || !n->dev))
+                       break;
+       }
+
+       return n;
+}
+
+struct pneigh_entry * pneigh_lookup(struct neigh_table *tbl,
+                                   struct net *net, const void *pkey,
                                    struct net_device *dev, int creat)
 {
        struct pneigh_entry *n;
@@ -466,6 +505,7 @@ struct pneigh_entry * pneigh_lookup(struct neigh_table *tbl, const void *pkey,
 
        for (n = tbl->phash_buckets[hash_val]; n; n = n->next) {
                if (!memcmp(n->key, pkey, key_len) &&
+                   net_eq(pneigh_net(n), net) &&
                    (n->dev == dev || !n->dev)) {
                        read_unlock_bh(&tbl->lock);
                        goto out;
@@ -476,10 +516,15 @@ struct pneigh_entry * pneigh_lookup(struct neigh_table *tbl, const void *pkey,
        if (!creat)
                goto out;
 
+       ASSERT_RTNL();
+
        n = kmalloc(sizeof(*n) + key_len, GFP_KERNEL);
        if (!n)
                goto out;
 
+#ifdef CONFIG_NET_NS
+       n->net = hold_net(net);
+#endif
        memcpy(n->key, pkey, key_len);
        n->dev = dev;
        if (dev)
@@ -488,6 +533,7 @@ struct pneigh_entry * pneigh_lookup(struct neigh_table *tbl, const void *pkey,
        if (tbl->pconstructor && tbl->pconstructor(n)) {
                if (dev)
                        dev_put(dev);
+               release_net(net);
                kfree(n);
                n = NULL;
                goto out;
@@ -502,7 +548,7 @@ out:
 }
 
 
-int pneigh_delete(struct neigh_table *tbl, const void *pkey,
+int pneigh_delete(struct neigh_table *tbl, struct net *net, const void *pkey,
                  struct net_device *dev)
 {
        struct pneigh_entry *n, **np;
@@ -517,13 +563,15 @@ int pneigh_delete(struct neigh_table *tbl, const void *pkey,
        write_lock_bh(&tbl->lock);
        for (np = &tbl->phash_buckets[hash_val]; (n = *np) != NULL;
             np = &n->next) {
-               if (!memcmp(n->key, pkey, key_len) && n->dev == dev) {
+               if (!memcmp(n->key, pkey, key_len) && n->dev == dev &&
+                   net_eq(pneigh_net(n), net)) {
                        *np = n->next;
                        write_unlock_bh(&tbl->lock);
                        if (tbl->pdestructor)
                                tbl->pdestructor(n);
                        if (n->dev)
                                dev_put(n->dev);
+                       release_net(pneigh_net(n));
                        kfree(n);
                        return 0;
                }
@@ -546,6 +594,7 @@ static int pneigh_ifdown(struct neigh_table *tbl, struct net_device *dev)
                                        tbl->pdestructor(n);
                                if (n->dev)
                                        dev_put(n->dev);
+                               release_net(pneigh_net(n));
                                kfree(n);
                                continue;
                        }
@@ -555,6 +604,13 @@ static int pneigh_ifdown(struct neigh_table *tbl, struct net_device *dev)
        return -ENOENT;
 }
 
+static void neigh_parms_destroy(struct neigh_parms *parms);
+
+static inline void neigh_parms_put(struct neigh_parms *parms)
+{
+       if (atomic_dec_and_test(&parms->refcnt))
+               neigh_parms_destroy(parms);
+}
 
 /*
  *     neighbour must already be out of the table;
@@ -579,16 +635,14 @@ void neigh_destroy(struct neighbour *neigh)
        while ((hh = neigh->hh) != NULL) {
                neigh->hh = hh->hh_next;
                hh->hh_next = NULL;
-               write_lock_bh(&hh->hh_lock);
+
+               write_seqlock_bh(&hh->hh_lock);
                hh->hh_output = neigh_blackhole;
-               write_unlock_bh(&hh->hh_lock);
+               write_sequnlock_bh(&hh->hh_lock);
                if (atomic_dec_and_test(&hh->hh_refcnt))
                        kfree(hh);
        }
 
-       if (neigh->ops && neigh->ops->destructor)
-               (neigh->ops->destructor)(neigh);
-
        skb_queue_purge(&neigh->arp_queue);
 
        dev_put(neigh->dev);
@@ -679,7 +733,7 @@ static void neigh_periodic_timer(unsigned long arg)
                        *np = n->next;
                        n->dead = 1;
                        write_unlock(&n->lock);
-                       neigh_release(n);
+                       neigh_cleanup_and_release(n);
                        continue;
                }
                write_unlock(&n->lock);
@@ -688,16 +742,19 @@ next_elt:
                np = &n->next;
        }
 
-       /* Cycle through all hash buckets every base_reachable_time/2 ticks.
-        * ARP entry timeouts range from 1/2 base_reachable_time to 3/2
-        * base_reachable_time.
+       /* Cycle through all hash buckets every base_reachable_time/2 ticks.
+        * ARP entry timeouts range from 1/2 base_reachable_time to 3/2
+        * base_reachable_time.
         */
        expire = tbl->parms.base_reachable_time >> 1;
        expire /= (tbl->hash_mask + 1);
        if (!expire)
                expire = 1;
 
-       mod_timer(&tbl->gc_timer, now + expire);
+       if (expire>HZ)
+               mod_timer(&tbl->gc_timer, round_jiffies(now + expire));
+       else
+               mod_timer(&tbl->gc_timer, now + expire);
 
        write_unlock(&tbl->lock);
 }
@@ -710,15 +767,6 @@ static __inline__ int neigh_max_probes(struct neighbour *n)
                p->ucast_probes + p->app_probes + p->mcast_probes);
 }
 
-static inline void neigh_add_timer(struct neighbour *n, unsigned long when)
-{
-       if (unlikely(mod_timer(&n->timer, when))) {
-               printk("NEIGH: BUG, double timer add, state is %x\n",
-                      n->nud_state);
-               dump_stack();
-       }
-}
-
 /* Called when a timer expires for a neighbour entry. */
 
 static void neigh_timer_handler(unsigned long arg)
@@ -742,7 +790,7 @@ static void neigh_timer_handler(unsigned long arg)
        }
 
        if (state & NUD_REACHABLE) {
-               if (time_before_eq(now, 
+               if (time_before_eq(now,
                                   neigh->confirmed + neigh->parms->reachable_time)) {
                        NEIGH_PRINTK2("neigh %p is still alive.\n", neigh);
                        next = neigh->confirmed + neigh->parms->reachable_time;
@@ -758,14 +806,16 @@ static void neigh_timer_handler(unsigned long arg)
                        neigh->nud_state = NUD_STALE;
                        neigh->updated = jiffies;
                        neigh_suspect(neigh);
+                       notify = 1;
                }
        } else if (state & NUD_DELAY) {
-               if (time_before_eq(now, 
+               if (time_before_eq(now,
                                   neigh->confirmed + neigh->parms->delay_probe_time)) {
                        NEIGH_PRINTK2("neigh %p is now reachable.\n", neigh);
                        neigh->nud_state = NUD_REACHABLE;
                        neigh->updated = jiffies;
                        neigh_connect(neigh);
+                       notify = 1;
                        next = neigh->confirmed + neigh->parms->reachable_time;
                } else {
                        NEIGH_PRINTK2("neigh %p is probed.\n", neigh);
@@ -813,7 +863,7 @@ static void neigh_timer_handler(unsigned long arg)
                struct sk_buff *skb = skb_peek(&neigh->arp_queue);
                /* keep skb alive even if arp_queue overflows */
                if (skb)
-                       skb_get(skb);
+                       skb = skb_copy(skb, GFP_ATOMIC);
                write_unlock(&neigh->lock);
                neigh->ops->solicit(neigh, skb);
                atomic_inc(&neigh->probes);
@@ -824,10 +874,9 @@ out:
                write_unlock(&neigh->lock);
        }
 
-#ifdef CONFIG_ARPD
-       if (notify && neigh->parms->app_probes)
-               neigh_app_notify(neigh);
-#endif
+       if (notify)
+               neigh_update_notify(neigh);
+
        neigh_release(neigh);
 }
 
@@ -843,13 +892,12 @@ int __neigh_event_send(struct neighbour *neigh, struct sk_buff *skb)
                goto out_unlock_bh;
 
        now = jiffies;
-       
+
        if (!(neigh->nud_state & (NUD_STALE | NUD_INCOMPLETE))) {
                if (neigh->parms->mcast_probes + neigh->parms->app_probes) {
                        atomic_set(&neigh->probes, neigh->parms->ucast_probes);
                        neigh->nud_state     = NUD_INCOMPLETE;
                        neigh->updated = jiffies;
-                       neigh_hold(neigh);
                        neigh_add_timer(neigh, now + 1);
                } else {
                        neigh->nud_state = NUD_FAILED;
@@ -862,7 +910,6 @@ int __neigh_event_send(struct neighbour *neigh, struct sk_buff *skb)
                }
        } else if (neigh->nud_state & NUD_STALE) {
                NEIGH_PRINTK2("neigh %p is delayed.\n", neigh);
-               neigh_hold(neigh);
                neigh->nud_state = NUD_DELAY;
                neigh->updated = jiffies;
                neigh_add_timer(neigh,
@@ -887,17 +934,17 @@ out_unlock_bh:
        return rc;
 }
 
-static __inline__ void neigh_update_hhs(struct neighbour *neigh)
+static void neigh_update_hhs(struct neighbour *neigh)
 {
        struct hh_cache *hh;
-       void (*update)(struct hh_cache*, struct net_device*, unsigned char *) =
-               neigh->dev->header_cache_update;
+       void (*update)(struct hh_cache*, const struct net_device*, const unsigned char *)
+               = neigh->dev->header_ops->cache_update;
 
        if (update) {
                for (hh = neigh->hh; hh; hh = hh->hh_next) {
-                       write_lock_bh(&hh->hh_lock);
+                       write_seqlock_bh(&hh->hh_lock);
                        update(hh, neigh->dev, neigh->ha);
-                       write_unlock_bh(&hh->hh_lock);
+                       write_sequnlock_bh(&hh->hh_lock);
                }
        }
 }
@@ -911,13 +958,13 @@ static __inline__ void neigh_update_hhs(struct neighbour *neigh)
        NEIGH_UPDATE_F_OVERRIDE allows to override existing lladdr,
                                if it is different.
        NEIGH_UPDATE_F_WEAK_OVERRIDE will suspect existing "connected"
-                               lladdr instead of overriding it 
+                               lladdr instead of overriding it
                                if it is different.
                                It also allows to retain current state
                                if lladdr is unchanged.
        NEIGH_UPDATE_F_ADMIN    means that the change is administrative.
 
-       NEIGH_UPDATE_F_OVERRIDE_ISROUTER allows to override existing 
+       NEIGH_UPDATE_F_OVERRIDE_ISROUTER allows to override existing
                                NTF_ROUTER flag.
        NEIGH_UPDATE_F_ISROUTER indicates if the neighbour is known as
                                a router.
@@ -930,9 +977,7 @@ int neigh_update(struct neighbour *neigh, const u8 *lladdr, u8 new,
 {
        u8 old;
        int err;
-#ifdef CONFIG_ARPD
        int notify = 0;
-#endif
        struct net_device *dev;
        int update_isrouter = 0;
 
@@ -942,7 +987,7 @@ int neigh_update(struct neighbour *neigh, const u8 *lladdr, u8 new,
        old    = neigh->nud_state;
        err    = -EPERM;
 
-       if (!(flags & NEIGH_UPDATE_F_ADMIN) && 
+       if (!(flags & NEIGH_UPDATE_F_ADMIN) &&
            (old & (NUD_NOARP | NUD_PERMANENT)))
                goto out;
 
@@ -952,9 +997,7 @@ int neigh_update(struct neighbour *neigh, const u8 *lladdr, u8 new,
                        neigh_suspect(neigh);
                neigh->nud_state = new;
                err = 0;
-#ifdef CONFIG_ARPD
                notify = old & NUD_VALID;
-#endif
                goto out;
        }
 
@@ -968,7 +1011,7 @@ int neigh_update(struct neighbour *neigh, const u8 *lladdr, u8 new,
                   - compare new & old
                   - if they are different, check override flag
                 */
-               if ((old & NUD_VALID) && 
+               if ((old & NUD_VALID) &&
                    !memcmp(lladdr, neigh->ha, dev->addr_len))
                        lladdr = neigh->ha;
        } else {
@@ -1010,13 +1053,11 @@ int neigh_update(struct neighbour *neigh, const u8 *lladdr, u8 new,
 
        if (new != old) {
                neigh_del_timer(neigh);
-               if (new & NUD_IN_TIMER) {
-                       neigh_hold(neigh);
-                       neigh_add_timer(neigh, (jiffies + 
-                                               ((new & NUD_REACHABLE) ? 
+               if (new & NUD_IN_TIMER)
+                       neigh_add_timer(neigh, (jiffies +
+                                               ((new & NUD_REACHABLE) ?
                                                 neigh->parms->reachable_time :
                                                 0)));
-               }
                neigh->nud_state = new;
        }
 
@@ -1026,9 +1067,7 @@ int neigh_update(struct neighbour *neigh, const u8 *lladdr, u8 new,
                if (!(new & NUD_CONNECTED))
                        neigh->confirmed = jiffies -
                                      (neigh->parms->base_reachable_time << 1);
-#ifdef CONFIG_ARPD
                notify = 1;
-#endif
        }
        if (new == old)
                goto out;
@@ -1060,10 +1099,10 @@ out:
                        (neigh->flags & ~NTF_ROUTER);
        }
        write_unlock_bh(&neigh->lock);
-#ifdef CONFIG_ARPD
-       if (notify && neigh->parms->app_probes)
-               neigh_app_notify(neigh);
-#endif
+
+       if (notify)
+               neigh_update_notify(neigh);
+
        return err;
 }
 
@@ -1074,13 +1113,13 @@ struct neighbour *neigh_event_ns(struct neigh_table *tbl,
        struct neighbour *neigh = __neigh_lookup(tbl, saddr, dev,
                                                 lladdr || !dev->addr_len);
        if (neigh)
-               neigh_update(neigh, lladdr, NUD_STALE, 
+               neigh_update(neigh, lladdr, NUD_STALE,
                             NEIGH_UPDATE_F_OVERRIDE);
        return neigh;
 }
 
 static void neigh_hh_init(struct neighbour *n, struct dst_entry *dst,
-                         u16 protocol)
+                         __be16 protocol)
 {
        struct hh_cache *hh;
        struct net_device *dev = dst->dev;
@@ -1089,13 +1128,13 @@ static void neigh_hh_init(struct neighbour *n, struct dst_entry *dst,
                if (hh->hh_type == protocol)
                        break;
 
-       if (!hh && (hh = kmalloc(sizeof(*hh), GFP_ATOMIC)) != NULL) {
-               memset(hh, 0, sizeof(struct hh_cache));
-               rwlock_init(&hh->hh_lock);
+       if (!hh && (hh = kzalloc(sizeof(*hh), GFP_ATOMIC)) != NULL) {
+               seqlock_init(&hh->hh_lock);
                hh->hh_type = protocol;
                atomic_set(&hh->hh_refcnt, 0);
                hh->hh_next = NULL;
-               if (dev->hard_header_cache(n, hh)) {
+
+               if (dev->header_ops->cache(n, hh)) {
                        kfree(hh);
                        hh = NULL;
                } else {
@@ -1123,12 +1162,11 @@ int neigh_compat_output(struct sk_buff *skb)
 {
        struct net_device *dev = skb->dev;
 
-       __skb_pull(skb, skb->nh.raw - skb->data);
+       __skb_pull(skb, skb_network_offset(skb));
 
-       if (dev->hard_header &&
-           dev->hard_header(skb, dev, ntohs(skb->protocol), NULL, NULL,
-                            skb->len) < 0 &&
-           dev->rebuild_header(skb))
+       if (dev_hard_header(skb, dev, ntohs(skb->protocol), NULL, NULL,
+                           skb->len) < 0 &&
+           dev->header_ops->rebuild(skb))
                return 0;
 
        return dev_queue_xmit(skb);
@@ -1145,22 +1183,22 @@ int neigh_resolve_output(struct sk_buff *skb)
        if (!dst || !(neigh = dst->neighbour))
                goto discard;
 
-       __skb_pull(skb, skb->nh.raw - skb->data);
+       __skb_pull(skb, skb_network_offset(skb));
 
        if (!neigh_event_send(neigh, skb)) {
                int err;
                struct net_device *dev = neigh->dev;
-               if (dev->hard_header_cache && !dst->hh) {
+               if (dev->header_ops->cache && !dst->hh) {
                        write_lock_bh(&neigh->lock);
                        if (!dst->hh)
                                neigh_hh_init(neigh, dst, dst->ops->protocol);
-                       err = dev->hard_header(skb, dev, ntohs(skb->protocol),
-                                              neigh->ha, NULL, skb->len);
+                       err = dev_hard_header(skb, dev, ntohs(skb->protocol),
+                                             neigh->ha, NULL, skb->len);
                        write_unlock_bh(&neigh->lock);
                } else {
                        read_lock_bh(&neigh->lock);
-                       err = dev->hard_header(skb, dev, ntohs(skb->protocol),
-                                              neigh->ha, NULL, skb->len);
+                       err = dev_hard_header(skb, dev, ntohs(skb->protocol),
+                                             neigh->ha, NULL, skb->len);
                        read_unlock_bh(&neigh->lock);
                }
                if (err >= 0)
@@ -1188,11 +1226,11 @@ int neigh_connected_output(struct sk_buff *skb)
        struct neighbour *neigh = dst->neighbour;
        struct net_device *dev = neigh->dev;
 
-       __skb_pull(skb, skb->nh.raw - skb->data);
+       __skb_pull(skb, skb_network_offset(skb));
 
        read_lock_bh(&neigh->lock);
-       err = dev->hard_header(skb, dev, ntohs(skb->protocol),
-                              neigh->ha, NULL, skb->len);
+       err = dev_hard_header(skb, dev, ntohs(skb->protocol),
+                             neigh->ha, NULL, skb->len);
        read_unlock_bh(&neigh->lock);
        if (err >= 0)
                err = neigh->ops->queue_xmit(skb);
@@ -1264,28 +1302,49 @@ void pneigh_enqueue(struct neigh_table *tbl, struct neigh_parms *p,
        spin_unlock(&tbl->proxy_queue.lock);
 }
 
+static inline struct neigh_parms *lookup_neigh_params(struct neigh_table *tbl,
+                                                     struct net *net, int ifindex)
+{
+       struct neigh_parms *p;
+
+       for (p = &tbl->parms; p; p = p->next) {
+               if ((p->dev && p->dev->ifindex == ifindex && net_eq(neigh_parms_net(p), net)) ||
+                   (!p->dev && !ifindex))
+                       return p;
+       }
+
+       return NULL;
+}
 
 struct neigh_parms *neigh_parms_alloc(struct net_device *dev,
                                      struct neigh_table *tbl)
 {
-       struct neigh_parms *p = kmalloc(sizeof(*p), GFP_KERNEL);
+       struct neigh_parms *p, *ref;
+       struct net *net;
 
+       net = dev_net(dev);
+       ref = lookup_neigh_params(tbl, net, 0);
+       if (!ref)
+               return NULL;
+
+       p = kmemdup(ref, sizeof(*p), GFP_KERNEL);
        if (p) {
-               memcpy(p, &tbl->parms, sizeof(*p));
                p->tbl            = tbl;
                atomic_set(&p->refcnt, 1);
                INIT_RCU_HEAD(&p->rcu_head);
                p->reachable_time =
                                neigh_rand_reach_time(p->base_reachable_time);
-               if (dev) {
-                       if (dev->neigh_setup && dev->neigh_setup(dev, p)) {
-                               kfree(p);
-                               return NULL;
-                       }
 
-                       dev_hold(dev);
-                       p->dev = dev;
+               if (dev->neigh_setup && dev->neigh_setup(dev, p)) {
+                       kfree(p);
+                       return NULL;
                }
+
+               dev_hold(dev);
+               p->dev = dev;
+#ifdef CONFIG_NET_NS
+               p->net = hold_net(net);
+#endif
                p->sysctl_table = NULL;
                write_lock_bh(&tbl->lock);
                p->next         = tbl->parms.next;
@@ -1325,40 +1384,41 @@ void neigh_parms_release(struct neigh_table *tbl, struct neigh_parms *parms)
        NEIGH_PRINTK1("neigh_parms_release: not found\n");
 }
 
-void neigh_parms_destroy(struct neigh_parms *parms)
+static void neigh_parms_destroy(struct neigh_parms *parms)
 {
+       release_net(neigh_parms_net(parms));
        kfree(parms);
 }
 
+static struct lock_class_key neigh_table_proxy_queue_class;
 
-void neigh_table_init(struct neigh_table *tbl)
+void neigh_table_init_no_netlink(struct neigh_table *tbl)
 {
        unsigned long now = jiffies;
        unsigned long phsize;
 
+#ifdef CONFIG_NET_NS
+       tbl->parms.net = &init_net;
+#endif
        atomic_set(&tbl->parms.refcnt, 1);
        INIT_RCU_HEAD(&tbl->parms.rcu_head);
        tbl->parms.reachable_time =
                          neigh_rand_reach_time(tbl->parms.base_reachable_time);
 
        if (!tbl->kmem_cachep)
-               tbl->kmem_cachep = kmem_cache_create(tbl->id,
-                                                    tbl->entry_size,
-                                                    0, SLAB_HWCACHE_ALIGN,
-                                                    NULL, NULL);
-
-       if (!tbl->kmem_cachep)
-               panic("cannot create neighbour cache");
-
+               tbl->kmem_cachep =
+                       kmem_cache_create(tbl->id, tbl->entry_size, 0,
+                                         SLAB_HWCACHE_ALIGN|SLAB_PANIC,
+                                         NULL);
        tbl->stats = alloc_percpu(struct neigh_statistics);
        if (!tbl->stats)
                panic("cannot create neighbour cache statistics");
-       
+
 #ifdef CONFIG_PROC_FS
-       tbl->pde = create_proc_entry(tbl->id, 0, proc_net_stat);
-       if (!tbl->pde) 
+       tbl->pde = proc_create(tbl->id, 0, init_net.proc_net_stat,
+                              &neigh_stat_seq_fops);
+       if (!tbl->pde)
                panic("cannot create neighbour proc dir entry");
-       tbl->pde->proc_fops = &neigh_stat_seq_fops;
        tbl->pde->data = tbl;
 #endif
 
@@ -1366,33 +1426,45 @@ void neigh_table_init(struct neigh_table *tbl)
        tbl->hash_buckets = neigh_hash_alloc(tbl->hash_mask + 1);
 
        phsize = (PNEIGH_HASHMASK + 1) * sizeof(struct pneigh_entry *);
-       tbl->phash_buckets = kmalloc(phsize, GFP_KERNEL);
+       tbl->phash_buckets = kzalloc(phsize, GFP_KERNEL);
 
        if (!tbl->hash_buckets || !tbl->phash_buckets)
                panic("cannot allocate neighbour cache hashes");
 
-       memset(tbl->phash_buckets, 0, phsize);
-
        get_random_bytes(&tbl->hash_rnd, sizeof(tbl->hash_rnd));
 
        rwlock_init(&tbl->lock);
-       init_timer(&tbl->gc_timer);
-       tbl->gc_timer.data     = (unsigned long)tbl;
-       tbl->gc_timer.function = neigh_periodic_timer;
+       setup_timer(&tbl->gc_timer, neigh_periodic_timer, (unsigned long)tbl);
        tbl->gc_timer.expires  = now + 1;
        add_timer(&tbl->gc_timer);
 
-       init_timer(&tbl->proxy_timer);
-       tbl->proxy_timer.data     = (unsigned long)tbl;
-       tbl->proxy_timer.function = neigh_proxy_process;
-       skb_queue_head_init(&tbl->proxy_queue);
+       setup_timer(&tbl->proxy_timer, neigh_proxy_process, (unsigned long)tbl);
+       skb_queue_head_init_class(&tbl->proxy_queue,
+                       &neigh_table_proxy_queue_class);
 
        tbl->last_flush = now;
        tbl->last_rand  = now + tbl->parms.reachable_time * 20;
+}
+
+void neigh_table_init(struct neigh_table *tbl)
+{
+       struct neigh_table *tmp;
+
+       neigh_table_init_no_netlink(tbl);
        write_lock(&neigh_tbl_lock);
+       for (tmp = neigh_tables; tmp; tmp = tmp->next) {
+               if (tmp->family == tbl->family)
+                       break;
+       }
        tbl->next       = neigh_tables;
        neigh_tables    = tbl;
        write_unlock(&neigh_tbl_lock);
+
+       if (unlikely(tmp)) {
+               printk(KERN_ERR "NEIGH: Registering multiple tables for "
+                      "family %d\n", tbl->family);
+               dump_stack();
+       }
 }
 
 int neigh_table_clear(struct neigh_table *tbl)
@@ -1421,53 +1493,76 @@ int neigh_table_clear(struct neigh_table *tbl)
        kfree(tbl->phash_buckets);
        tbl->phash_buckets = NULL;
 
+       remove_proc_entry(tbl->id, init_net.proc_net_stat);
+
+       free_percpu(tbl->stats);
+       tbl->stats = NULL;
+
+       kmem_cache_destroy(tbl->kmem_cachep);
+       tbl->kmem_cachep = NULL;
+
        return 0;
 }
 
-int neigh_delete(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
+static int neigh_delete(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 {
-       struct ndmsg *ndm = NLMSG_DATA(nlh);
-       struct rtattr **nda = arg;
+       struct net *net = sock_net(skb->sk);
+       struct ndmsg *ndm;
+       struct nlattr *dst_attr;
        struct neigh_table *tbl;
        struct net_device *dev = NULL;
-       int err = -ENODEV;
+       int err = -EINVAL;
+
+       if (nlmsg_len(nlh) < sizeof(*ndm))
+               goto out;
 
-       if (ndm->ndm_ifindex &&
-           (dev = dev_get_by_index(ndm->ndm_ifindex)) == NULL)
+       dst_attr = nlmsg_find_attr(nlh, sizeof(*ndm), NDA_DST);
+       if (dst_attr == NULL)
                goto out;
 
+       ndm = nlmsg_data(nlh);
+       if (ndm->ndm_ifindex) {
+               dev = dev_get_by_index(net, ndm->ndm_ifindex);
+               if (dev == NULL) {
+                       err = -ENODEV;
+                       goto out;
+               }
+       }
+
        read_lock(&neigh_tbl_lock);
        for (tbl = neigh_tables; tbl; tbl = tbl->next) {
-               struct rtattr *dst_attr = nda[NDA_DST - 1];
-               struct neighbour *n;
+               struct neighbour *neigh;
 
                if (tbl->family != ndm->ndm_family)
                        continue;
                read_unlock(&neigh_tbl_lock);
 
-               err = -EINVAL;
-               if (!dst_attr || RTA_PAYLOAD(dst_attr) < tbl->key_len)
+               if (nla_len(dst_attr) < tbl->key_len)
                        goto out_dev_put;
 
                if (ndm->ndm_flags & NTF_PROXY) {
-                       err = pneigh_delete(tbl, RTA_DATA(dst_attr), dev);
+                       err = pneigh_delete(tbl, net, nla_data(dst_attr), dev);
                        goto out_dev_put;
                }
 
-               if (!dev)
-                       goto out;
+               if (dev == NULL)
+                       goto out_dev_put;
 
-               n = neigh_lookup(tbl, RTA_DATA(dst_attr), dev);
-               if (n) {
-                       err = neigh_update(n, NULL, NUD_FAILED, 
-                                          NEIGH_UPDATE_F_OVERRIDE|
-                                          NEIGH_UPDATE_F_ADMIN);
-                       neigh_release(n);
+               neigh = neigh_lookup(tbl, nla_data(dst_attr), dev);
+               if (neigh == NULL) {
+                       err = -ENOENT;
+                       goto out_dev_put;
                }
+
+               err = neigh_update(neigh, NULL, NUD_FAILED,
+                                  NEIGH_UPDATE_F_OVERRIDE |
+                                  NEIGH_UPDATE_F_ADMIN);
+               neigh_release(neigh);
                goto out_dev_put;
        }
        read_unlock(&neigh_tbl_lock);
-       err = -EADDRNOTAVAIL;
+       err = -EAFNOSUPPORT;
+
 out_dev_put:
        if (dev)
                dev_put(dev);
@@ -1475,78 +1570,96 @@ out:
        return err;
 }
 
-int neigh_add(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
+static int neigh_add(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 {
-       struct ndmsg *ndm = NLMSG_DATA(nlh);
-       struct rtattr **nda = arg;
+       struct net *net = sock_net(skb->sk);
+       struct ndmsg *ndm;
+       struct nlattr *tb[NDA_MAX+1];
        struct neigh_table *tbl;
        struct net_device *dev = NULL;
-       int err = -ENODEV;
+       int err;
+
+       err = nlmsg_parse(nlh, sizeof(*ndm), tb, NDA_MAX, NULL);
+       if (err < 0)
+               goto out;
 
-       if (ndm->ndm_ifindex &&
-           (dev = dev_get_by_index(ndm->ndm_ifindex)) == NULL)
+       err = -EINVAL;
+       if (tb[NDA_DST] == NULL)
                goto out;
 
+       ndm = nlmsg_data(nlh);
+       if (ndm->ndm_ifindex) {
+               dev = dev_get_by_index(net, ndm->ndm_ifindex);
+               if (dev == NULL) {
+                       err = -ENODEV;
+                       goto out;
+               }
+
+               if (tb[NDA_LLADDR] && nla_len(tb[NDA_LLADDR]) < dev->addr_len)
+                       goto out_dev_put;
+       }
+
        read_lock(&neigh_tbl_lock);
        for (tbl = neigh_tables; tbl; tbl = tbl->next) {
-               struct rtattr *lladdr_attr = nda[NDA_LLADDR - 1];
-               struct rtattr *dst_attr = nda[NDA_DST - 1];
-               int override = 1;
-               struct neighbour *n;
+               int flags = NEIGH_UPDATE_F_ADMIN | NEIGH_UPDATE_F_OVERRIDE;
+               struct neighbour *neigh;
+               void *dst, *lladdr;
 
                if (tbl->family != ndm->ndm_family)
                        continue;
                read_unlock(&neigh_tbl_lock);
 
-               err = -EINVAL;
-               if (!dst_attr || RTA_PAYLOAD(dst_attr) < tbl->key_len)
+               if (nla_len(tb[NDA_DST]) < tbl->key_len)
                        goto out_dev_put;
+               dst = nla_data(tb[NDA_DST]);
+               lladdr = tb[NDA_LLADDR] ? nla_data(tb[NDA_LLADDR]) : NULL;
 
                if (ndm->ndm_flags & NTF_PROXY) {
+                       struct pneigh_entry *pn;
+
                        err = -ENOBUFS;
-                       if (pneigh_lookup(tbl, RTA_DATA(dst_attr), dev, 1))
+                       pn = pneigh_lookup(tbl, net, dst, dev, 1);
+                       if (pn) {
+                               pn->flags = ndm->ndm_flags;
                                err = 0;
+                       }
                        goto out_dev_put;
                }
 
-               err = -EINVAL;
-               if (!dev)
-                       goto out;
-               if (lladdr_attr && RTA_PAYLOAD(lladdr_attr) < dev->addr_len)
+               if (dev == NULL)
                        goto out_dev_put;
-       
-               n = neigh_lookup(tbl, RTA_DATA(dst_attr), dev);
-               if (n) {
-                       if (nlh->nlmsg_flags & NLM_F_EXCL) {
-                               err = -EEXIST;
-                               neigh_release(n);
+
+               neigh = neigh_lookup(tbl, dst, dev);
+               if (neigh == NULL) {
+                       if (!(nlh->nlmsg_flags & NLM_F_CREATE)) {
+                               err = -ENOENT;
+                               goto out_dev_put;
+                       }
+
+                       neigh = __neigh_lookup_errno(tbl, dst, dev);
+                       if (IS_ERR(neigh)) {
+                               err = PTR_ERR(neigh);
                                goto out_dev_put;
                        }
-                       
-                       override = nlh->nlmsg_flags & NLM_F_REPLACE;
-               } else if (!(nlh->nlmsg_flags & NLM_F_CREATE)) {
-                       err = -ENOENT;
-                       goto out_dev_put;
                } else {
-                       n = __neigh_lookup_errno(tbl, RTA_DATA(dst_attr), dev);
-                       if (IS_ERR(n)) {
-                               err = PTR_ERR(n);
+                       if (nlh->nlmsg_flags & NLM_F_EXCL) {
+                               err = -EEXIST;
+                               neigh_release(neigh);
                                goto out_dev_put;
                        }
-               }
 
-               err = neigh_update(n,
-                                  lladdr_attr ? RTA_DATA(lladdr_attr) : NULL,
-                                  ndm->ndm_state,
-                                  (override ? NEIGH_UPDATE_F_OVERRIDE : 0) |
-                                  NEIGH_UPDATE_F_ADMIN);
+                       if (!(nlh->nlmsg_flags & NLM_F_REPLACE))
+                               flags &= ~NEIGH_UPDATE_F_OVERRIDE;
+               }
 
-               neigh_release(n);
+               err = neigh_update(neigh, lladdr, ndm->ndm_state, flags);
+               neigh_release(neigh);
                goto out_dev_put;
        }
 
        read_unlock(&neigh_tbl_lock);
-       err = -EADDRNOTAVAIL;
+       err = -EAFNOSUPPORT;
+
 out_dev_put:
        if (dev)
                dev_put(dev);
@@ -1556,56 +1669,59 @@ out:
 
 static int neightbl_fill_parms(struct sk_buff *skb, struct neigh_parms *parms)
 {
-       struct rtattr *nest = NULL;
-       
-       nest = RTA_NEST(skb, NDTA_PARMS);
+       struct nlattr *nest;
+
+       nest = nla_nest_start(skb, NDTA_PARMS);
+       if (nest == NULL)
+               return -ENOBUFS;
 
        if (parms->dev)
-               RTA_PUT_U32(skb, NDTPA_IFINDEX, parms->dev->ifindex);
-
-       RTA_PUT_U32(skb, NDTPA_REFCNT, atomic_read(&parms->refcnt));
-       RTA_PUT_U32(skb, NDTPA_QUEUE_LEN, parms->queue_len);
-       RTA_PUT_U32(skb, NDTPA_PROXY_QLEN, parms->proxy_qlen);
-       RTA_PUT_U32(skb, NDTPA_APP_PROBES, parms->app_probes);
-       RTA_PUT_U32(skb, NDTPA_UCAST_PROBES, parms->ucast_probes);
-       RTA_PUT_U32(skb, NDTPA_MCAST_PROBES, parms->mcast_probes);
-       RTA_PUT_MSECS(skb, NDTPA_REACHABLE_TIME, parms->reachable_time);
-       RTA_PUT_MSECS(skb, NDTPA_BASE_REACHABLE_TIME,
+               NLA_PUT_U32(skb, NDTPA_IFINDEX, parms->dev->ifindex);
+
+       NLA_PUT_U32(skb, NDTPA_REFCNT, atomic_read(&parms->refcnt));
+       NLA_PUT_U32(skb, NDTPA_QUEUE_LEN, parms->queue_len);
+       NLA_PUT_U32(skb, NDTPA_PROXY_QLEN, parms->proxy_qlen);
+       NLA_PUT_U32(skb, NDTPA_APP_PROBES, parms->app_probes);
+       NLA_PUT_U32(skb, NDTPA_UCAST_PROBES, parms->ucast_probes);
+       NLA_PUT_U32(skb, NDTPA_MCAST_PROBES, parms->mcast_probes);
+       NLA_PUT_MSECS(skb, NDTPA_REACHABLE_TIME, parms->reachable_time);
+       NLA_PUT_MSECS(skb, NDTPA_BASE_REACHABLE_TIME,
                      parms->base_reachable_time);
-       RTA_PUT_MSECS(skb, NDTPA_GC_STALETIME, parms->gc_staletime);
-       RTA_PUT_MSECS(skb, NDTPA_DELAY_PROBE_TIME, parms->delay_probe_time);
-       RTA_PUT_MSECS(skb, NDTPA_RETRANS_TIME, parms->retrans_time);
-       RTA_PUT_MSECS(skb, NDTPA_ANYCAST_DELAY, parms->anycast_delay);
-       RTA_PUT_MSECS(skb, NDTPA_PROXY_DELAY, parms->proxy_delay);
-       RTA_PUT_MSECS(skb, NDTPA_LOCKTIME, parms->locktime);
+       NLA_PUT_MSECS(skb, NDTPA_GC_STALETIME, parms->gc_staletime);
+       NLA_PUT_MSECS(skb, NDTPA_DELAY_PROBE_TIME, parms->delay_probe_time);
+       NLA_PUT_MSECS(skb, NDTPA_RETRANS_TIME, parms->retrans_time);
+       NLA_PUT_MSECS(skb, NDTPA_ANYCAST_DELAY, parms->anycast_delay);
+       NLA_PUT_MSECS(skb, NDTPA_PROXY_DELAY, parms->proxy_delay);
+       NLA_PUT_MSECS(skb, NDTPA_LOCKTIME, parms->locktime);
 
-       return RTA_NEST_END(skb, nest);
+       return nla_nest_end(skb, nest);
 
-rtattr_failure:
-       return RTA_NEST_CANCEL(skb, nest);
+nla_put_failure:
+       return nla_nest_cancel(skb, nest);
 }
 
-static int neightbl_fill_info(struct neigh_table *tbl, struct sk_buff *skb,
-                             struct netlink_callback *cb)
+static int neightbl_fill_info(struct sk_buff *skb, struct neigh_table *tbl,
+                             u32 pid, u32 seq, int type, int flags)
 {
        struct nlmsghdr *nlh;
        struct ndtmsg *ndtmsg;
 
-       nlh = NLMSG_NEW_ANSWER(skb, cb, RTM_NEWNEIGHTBL, sizeof(struct ndtmsg),
-                              NLM_F_MULTI);
+       nlh = nlmsg_put(skb, pid, seq, type, sizeof(*ndtmsg), flags);
+       if (nlh == NULL)
+               return -EMSGSIZE;
 
-       ndtmsg = NLMSG_DATA(nlh);
+       ndtmsg = nlmsg_data(nlh);
 
        read_lock_bh(&tbl->lock);
        ndtmsg->ndtm_family = tbl->family;
        ndtmsg->ndtm_pad1   = 0;
        ndtmsg->ndtm_pad2   = 0;
 
-       RTA_PUT_STRING(skb, NDTA_NAME, tbl->id);
-       RTA_PUT_MSECS(skb, NDTA_GC_INTERVAL, tbl->gc_interval);
-       RTA_PUT_U32(skb, NDTA_THRESH1, tbl->gc_thresh1);
-       RTA_PUT_U32(skb, NDTA_THRESH2, tbl->gc_thresh2);
-       RTA_PUT_U32(skb, NDTA_THRESH3, tbl->gc_thresh3);
+       NLA_PUT_STRING(skb, NDTA_NAME, tbl->id);
+       NLA_PUT_MSECS(skb, NDTA_GC_INTERVAL, tbl->gc_interval);
+       NLA_PUT_U32(skb, NDTA_THRESH1, tbl->gc_thresh1);
+       NLA_PUT_U32(skb, NDTA_THRESH2, tbl->gc_thresh2);
+       NLA_PUT_U32(skb, NDTA_THRESH3, tbl->gc_thresh3);
 
        {
                unsigned long now = jiffies;
@@ -1624,7 +1740,7 @@ static int neightbl_fill_info(struct neigh_table *tbl, struct sk_buff *skb,
                        .ndtc_proxy_qlen        = tbl->proxy_queue.qlen,
                };
 
-               RTA_PUT(skb, NDTA_CONFIG, sizeof(ndc), &ndc);
+               NLA_PUT(skb, NDTA_CONFIG, sizeof(ndc), &ndc);
        }
 
        {
@@ -1633,7 +1749,7 @@ static int neightbl_fill_info(struct neigh_table *tbl, struct sk_buff *skb,
 
                memset(&ndst, 0, sizeof(ndst));
 
-               for_each_cpu(cpu) {
+               for_each_possible_cpu(cpu) {
                        struct neigh_statistics *st;
 
                        st = per_cpu_ptr(tbl->stats, cpu);
@@ -1649,277 +1765,323 @@ static int neightbl_fill_info(struct neigh_table *tbl, struct sk_buff *skb,
                        ndst.ndts_forced_gc_runs        += st->forced_gc_runs;
                }
 
-               RTA_PUT(skb, NDTA_STATS, sizeof(ndst), &ndst);
+               NLA_PUT(skb, NDTA_STATS, sizeof(ndst), &ndst);
        }
 
        BUG_ON(tbl->parms.dev);
        if (neightbl_fill_parms(skb, &tbl->parms) < 0)
-               goto rtattr_failure;
+               goto nla_put_failure;
 
        read_unlock_bh(&tbl->lock);
-       return NLMSG_END(skb, nlh);
+       return nlmsg_end(skb, nlh);
 
-rtattr_failure:
+nla_put_failure:
        read_unlock_bh(&tbl->lock);
-       return NLMSG_CANCEL(skb, nlh);
-nlmsg_failure:
-       return -1;
+       nlmsg_cancel(skb, nlh);
+       return -EMSGSIZE;
 }
 
-static int neightbl_fill_param_info(struct neigh_table *tbl,
+static int neightbl_fill_param_info(struct sk_buff *skb,
+                                   struct neigh_table *tbl,
                                    struct neigh_parms *parms,
-                                   struct sk_buff *skb,
-                                   struct netlink_callback *cb)
+                                   u32 pid, u32 seq, int type,
+                                   unsigned int flags)
 {
        struct ndtmsg *ndtmsg;
        struct nlmsghdr *nlh;
 
-       nlh = NLMSG_NEW_ANSWER(skb, cb, RTM_NEWNEIGHTBL, sizeof(struct ndtmsg),
-                              NLM_F_MULTI);
+       nlh = nlmsg_put(skb, pid, seq, type, sizeof(*ndtmsg), flags);
+       if (nlh == NULL)
+               return -EMSGSIZE;
 
-       ndtmsg = NLMSG_DATA(nlh);
+       ndtmsg = nlmsg_data(nlh);
 
        read_lock_bh(&tbl->lock);
        ndtmsg->ndtm_family = tbl->family;
        ndtmsg->ndtm_pad1   = 0;
        ndtmsg->ndtm_pad2   = 0;
-       RTA_PUT_STRING(skb, NDTA_NAME, tbl->id);
 
-       if (neightbl_fill_parms(skb, parms) < 0)
-               goto rtattr_failure;
+       if (nla_put_string(skb, NDTA_NAME, tbl->id) < 0 ||
+           neightbl_fill_parms(skb, parms) < 0)
+               goto errout;
 
        read_unlock_bh(&tbl->lock);
-       return NLMSG_END(skb, nlh);
-
-rtattr_failure:
+       return nlmsg_end(skb, nlh);
+errout:
        read_unlock_bh(&tbl->lock);
-       return NLMSG_CANCEL(skb, nlh);
-
-nlmsg_failure:
-       return -1;
+       nlmsg_cancel(skb, nlh);
+       return -EMSGSIZE;
 }
-static inline struct neigh_parms *lookup_neigh_params(struct neigh_table *tbl,
-                                                     int ifindex)
-{
-       struct neigh_parms *p;
-       
-       for (p = &tbl->parms; p; p = p->next)
-               if ((p->dev && p->dev->ifindex == ifindex) ||
-                   (!p->dev && !ifindex))
-                       return p;
 
-       return NULL;
-}
+static const struct nla_policy nl_neightbl_policy[NDTA_MAX+1] = {
+       [NDTA_NAME]             = { .type = NLA_STRING },
+       [NDTA_THRESH1]          = { .type = NLA_U32 },
+       [NDTA_THRESH2]          = { .type = NLA_U32 },
+       [NDTA_THRESH3]          = { .type = NLA_U32 },
+       [NDTA_GC_INTERVAL]      = { .type = NLA_U64 },
+       [NDTA_PARMS]            = { .type = NLA_NESTED },
+};
 
-int neightbl_set(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
+static const struct nla_policy nl_ntbl_parm_policy[NDTPA_MAX+1] = {
+       [NDTPA_IFINDEX]                 = { .type = NLA_U32 },
+       [NDTPA_QUEUE_LEN]               = { .type = NLA_U32 },
+       [NDTPA_PROXY_QLEN]              = { .type = NLA_U32 },
+       [NDTPA_APP_PROBES]              = { .type = NLA_U32 },
+       [NDTPA_UCAST_PROBES]            = { .type = NLA_U32 },
+       [NDTPA_MCAST_PROBES]            = { .type = NLA_U32 },
+       [NDTPA_BASE_REACHABLE_TIME]     = { .type = NLA_U64 },
+       [NDTPA_GC_STALETIME]            = { .type = NLA_U64 },
+       [NDTPA_DELAY_PROBE_TIME]        = { .type = NLA_U64 },
+       [NDTPA_RETRANS_TIME]            = { .type = NLA_U64 },
+       [NDTPA_ANYCAST_DELAY]           = { .type = NLA_U64 },
+       [NDTPA_PROXY_DELAY]             = { .type = NLA_U64 },
+       [NDTPA_LOCKTIME]                = { .type = NLA_U64 },
+};
+
+static int neightbl_set(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
 {
+       struct net *net = sock_net(skb->sk);
        struct neigh_table *tbl;
-       struct ndtmsg *ndtmsg = NLMSG_DATA(nlh);
-       struct rtattr **tb = arg;
-       int err = -EINVAL;
+       struct ndtmsg *ndtmsg;
+       struct nlattr *tb[NDTA_MAX+1];
+       int err;
 
-       if (!tb[NDTA_NAME - 1] || !RTA_PAYLOAD(tb[NDTA_NAME - 1]))
-               return -EINVAL;
+       err = nlmsg_parse(nlh, sizeof(*ndtmsg), tb, NDTA_MAX,
+                         nl_neightbl_policy);
+       if (err < 0)
+               goto errout;
 
+       if (tb[NDTA_NAME] == NULL) {
+               err = -EINVAL;
+               goto errout;
+       }
+
+       ndtmsg = nlmsg_data(nlh);
        read_lock(&neigh_tbl_lock);
        for (tbl = neigh_tables; tbl; tbl = tbl->next) {
                if (ndtmsg->ndtm_family && tbl->family != ndtmsg->ndtm_family)
                        continue;
 
-               if (!rtattr_strcmp(tb[NDTA_NAME - 1], tbl->id))
+               if (nla_strcmp(tb[NDTA_NAME], tbl->id) == 0)
                        break;
        }
 
        if (tbl == NULL) {
                err = -ENOENT;
-               goto errout;
+               goto errout_locked;
        }
 
-       /* 
+       /*
         * We acquire tbl->lock to be nice to the periodic timers and
         * make sure they always see a consistent set of values.
         */
        write_lock_bh(&tbl->lock);
 
-       if (tb[NDTA_THRESH1 - 1])
-               tbl->gc_thresh1 = RTA_GET_U32(tb[NDTA_THRESH1 - 1]);
-
-       if (tb[NDTA_THRESH2 - 1])
-               tbl->gc_thresh2 = RTA_GET_U32(tb[NDTA_THRESH2 - 1]);
-
-       if (tb[NDTA_THRESH3 - 1])
-               tbl->gc_thresh3 = RTA_GET_U32(tb[NDTA_THRESH3 - 1]);
-
-       if (tb[NDTA_GC_INTERVAL - 1])
-               tbl->gc_interval = RTA_GET_MSECS(tb[NDTA_GC_INTERVAL - 1]);
-
-       if (tb[NDTA_PARMS - 1]) {
-               struct rtattr *tbp[NDTPA_MAX];
+       if (tb[NDTA_PARMS]) {
+               struct nlattr *tbp[NDTPA_MAX+1];
                struct neigh_parms *p;
-               u32 ifindex = 0;
+               int i, ifindex = 0;
 
-               if (rtattr_parse_nested(tbp, NDTPA_MAX, tb[NDTA_PARMS - 1]) < 0)
-                       goto rtattr_failure;
+               err = nla_parse_nested(tbp, NDTPA_MAX, tb[NDTA_PARMS],
+                                      nl_ntbl_parm_policy);
+               if (err < 0)
+                       goto errout_tbl_lock;
 
-               if (tbp[NDTPA_IFINDEX - 1])
-                       ifindex = RTA_GET_U32(tbp[NDTPA_IFINDEX - 1]);
+               if (tbp[NDTPA_IFINDEX])
+                       ifindex = nla_get_u32(tbp[NDTPA_IFINDEX]);
 
-               p = lookup_neigh_params(tbl, ifindex);
+               p = lookup_neigh_params(tbl, net, ifindex);
                if (p == NULL) {
                        err = -ENOENT;
-                       goto rtattr_failure;
+                       goto errout_tbl_lock;
                }
-       
-               if (tbp[NDTPA_QUEUE_LEN - 1])
-                       p->queue_len = RTA_GET_U32(tbp[NDTPA_QUEUE_LEN - 1]);
-
-               if (tbp[NDTPA_PROXY_QLEN - 1])
-                       p->proxy_qlen = RTA_GET_U32(tbp[NDTPA_PROXY_QLEN - 1]);
-
-               if (tbp[NDTPA_APP_PROBES - 1])
-                       p->app_probes = RTA_GET_U32(tbp[NDTPA_APP_PROBES - 1]);
 
-               if (tbp[NDTPA_UCAST_PROBES - 1])
-                       p->ucast_probes =
-                          RTA_GET_U32(tbp[NDTPA_UCAST_PROBES - 1]);
-
-               if (tbp[NDTPA_MCAST_PROBES - 1])
-                       p->mcast_probes =
-                          RTA_GET_U32(tbp[NDTPA_MCAST_PROBES - 1]);
-
-               if (tbp[NDTPA_BASE_REACHABLE_TIME - 1])
-                       p->base_reachable_time =
-                          RTA_GET_MSECS(tbp[NDTPA_BASE_REACHABLE_TIME - 1]);
-
-               if (tbp[NDTPA_GC_STALETIME - 1])
-                       p->gc_staletime =
-                          RTA_GET_MSECS(tbp[NDTPA_GC_STALETIME - 1]);
+               for (i = 1; i <= NDTPA_MAX; i++) {
+                       if (tbp[i] == NULL)
+                               continue;
 
-               if (tbp[NDTPA_DELAY_PROBE_TIME - 1])
-                       p->delay_probe_time =
-                          RTA_GET_MSECS(tbp[NDTPA_DELAY_PROBE_TIME - 1]);
+                       switch (i) {
+                       case NDTPA_QUEUE_LEN:
+                               p->queue_len = nla_get_u32(tbp[i]);
+                               break;
+                       case NDTPA_PROXY_QLEN:
+                               p->proxy_qlen = nla_get_u32(tbp[i]);
+                               break;
+                       case NDTPA_APP_PROBES:
+                               p->app_probes = nla_get_u32(tbp[i]);
+                               break;
+                       case NDTPA_UCAST_PROBES:
+                               p->ucast_probes = nla_get_u32(tbp[i]);
+                               break;
+                       case NDTPA_MCAST_PROBES:
+                               p->mcast_probes = nla_get_u32(tbp[i]);
+                               break;
+                       case NDTPA_BASE_REACHABLE_TIME:
+                               p->base_reachable_time = nla_get_msecs(tbp[i]);
+                               break;
+                       case NDTPA_GC_STALETIME:
+                               p->gc_staletime = nla_get_msecs(tbp[i]);
+                               break;
+                       case NDTPA_DELAY_PROBE_TIME:
+                               p->delay_probe_time = nla_get_msecs(tbp[i]);
+                               break;
+                       case NDTPA_RETRANS_TIME:
+                               p->retrans_time = nla_get_msecs(tbp[i]);
+                               break;
+                       case NDTPA_ANYCAST_DELAY:
+                               p->anycast_delay = nla_get_msecs(tbp[i]);
+                               break;
+                       case NDTPA_PROXY_DELAY:
+                               p->proxy_delay = nla_get_msecs(tbp[i]);
+                               break;
+                       case NDTPA_LOCKTIME:
+                               p->locktime = nla_get_msecs(tbp[i]);
+                               break;
+                       }
+               }
+       }
 
-               if (tbp[NDTPA_RETRANS_TIME - 1])
-                       p->retrans_time =
-                          RTA_GET_MSECS(tbp[NDTPA_RETRANS_TIME - 1]);
+       if (tb[NDTA_THRESH1])
+               tbl->gc_thresh1 = nla_get_u32(tb[NDTA_THRESH1]);
 
-               if (tbp[NDTPA_ANYCAST_DELAY - 1])
-                       p->anycast_delay =
-                          RTA_GET_MSECS(tbp[NDTPA_ANYCAST_DELAY - 1]);
+       if (tb[NDTA_THRESH2])
+               tbl->gc_thresh2 = nla_get_u32(tb[NDTA_THRESH2]);
 
-               if (tbp[NDTPA_PROXY_DELAY - 1])
-                       p->proxy_delay =
-                          RTA_GET_MSECS(tbp[NDTPA_PROXY_DELAY - 1]);
+       if (tb[NDTA_THRESH3])
+               tbl->gc_thresh3 = nla_get_u32(tb[NDTA_THRESH3]);
 
-               if (tbp[NDTPA_LOCKTIME - 1])
-                       p->locktime = RTA_GET_MSECS(tbp[NDTPA_LOCKTIME - 1]);
-       }
+       if (tb[NDTA_GC_INTERVAL])
+               tbl->gc_interval = nla_get_msecs(tb[NDTA_GC_INTERVAL]);
 
        err = 0;
 
-rtattr_failure:
+errout_tbl_lock:
        write_unlock_bh(&tbl->lock);
-errout:
+errout_locked:
        read_unlock(&neigh_tbl_lock);
+errout:
        return err;
 }
 
-int neightbl_dump_info(struct sk_buff *skb, struct netlink_callback *cb)
+static int neightbl_dump_info(struct sk_buff *skb, struct netlink_callback *cb)
 {
-       int idx, family;
-       int s_idx = cb->args[0];
+       struct net *net = sock_net(skb->sk);
+       int family, tidx, nidx = 0;
+       int tbl_skip = cb->args[0];
+       int neigh_skip = cb->args[1];
        struct neigh_table *tbl;
 
-       family = ((struct rtgenmsg *)NLMSG_DATA(cb->nlh))->rtgen_family;
+       family = ((struct rtgenmsg *) nlmsg_data(cb->nlh))->rtgen_family;
 
        read_lock(&neigh_tbl_lock);
-       for (tbl = neigh_tables, idx = 0; tbl; tbl = tbl->next) {
+       for (tbl = neigh_tables, tidx = 0; tbl; tbl = tbl->next, tidx++) {
                struct neigh_parms *p;
 
-               if (idx < s_idx || (family && tbl->family != family))
+               if (tidx < tbl_skip || (family && tbl->family != family))
                        continue;
 
-               if (neightbl_fill_info(tbl, skb, cb) <= 0)
+               if (neightbl_fill_info(skb, tbl, NETLINK_CB(cb->skb).pid,
+                                      cb->nlh->nlmsg_seq, RTM_NEWNEIGHTBL,
+                                      NLM_F_MULTI) <= 0)
                        break;
 
-               for (++idx, p = tbl->parms.next; p; p = p->next, idx++) {
-                       if (idx < s_idx)
+               for (nidx = 0, p = tbl->parms.next; p; p = p->next) {
+                       if (!net_eq(neigh_parms_net(p), net))
+                               continue;
+
+                       if (nidx++ < neigh_skip)
                                continue;
 
-                       if (neightbl_fill_param_info(tbl, p, skb, cb) <= 0)
+                       if (neightbl_fill_param_info(skb, tbl, p,
+                                                    NETLINK_CB(cb->skb).pid,
+                                                    cb->nlh->nlmsg_seq,
+                                                    RTM_NEWNEIGHTBL,
+                                                    NLM_F_MULTI) <= 0)
                                goto out;
                }
 
+               neigh_skip = 0;
        }
 out:
        read_unlock(&neigh_tbl_lock);
-       cb->args[0] = idx;
+       cb->args[0] = tidx;
+       cb->args[1] = nidx;
 
        return skb->len;
 }
 
-static int neigh_fill_info(struct sk_buff *skb, struct neighbour *n,
-                          u32 pid, u32 seq, int event, unsigned int flags)
+static int neigh_fill_info(struct sk_buff *skb, struct neighbour *neigh,
+                          u32 pid, u32 seq, int type, unsigned int flags)
 {
        unsigned long now = jiffies;
-       unsigned char *b = skb->tail;
        struct nda_cacheinfo ci;
-       int locked = 0;
-       u32 probes;
-       struct nlmsghdr *nlh = NLMSG_NEW(skb, pid, seq, event,
-                                        sizeof(struct ndmsg), flags);
-       struct ndmsg *ndm = NLMSG_DATA(nlh);
+       struct nlmsghdr *nlh;
+       struct ndmsg *ndm;
+
+       nlh = nlmsg_put(skb, pid, seq, type, sizeof(*ndm), flags);
+       if (nlh == NULL)
+               return -EMSGSIZE;
 
-       ndm->ndm_family  = n->ops->family;
+       ndm = nlmsg_data(nlh);
+       ndm->ndm_family  = neigh->ops->family;
        ndm->ndm_pad1    = 0;
        ndm->ndm_pad2    = 0;
-       ndm->ndm_flags   = n->flags;
-       ndm->ndm_type    = n->type;
-       ndm->ndm_ifindex = n->dev->ifindex;
-       RTA_PUT(skb, NDA_DST, n->tbl->key_len, n->primary_key);
-       read_lock_bh(&n->lock);
-       locked           = 1;
-       ndm->ndm_state   = n->nud_state;
-       if (n->nud_state & NUD_VALID)
-               RTA_PUT(skb, NDA_LLADDR, n->dev->addr_len, n->ha);
-       ci.ndm_used      = now - n->used;
-       ci.ndm_confirmed = now - n->confirmed;
-       ci.ndm_updated   = now - n->updated;
-       ci.ndm_refcnt    = atomic_read(&n->refcnt) - 1;
-       probes = atomic_read(&n->probes);
-       read_unlock_bh(&n->lock);
-       locked           = 0;
-       RTA_PUT(skb, NDA_CACHEINFO, sizeof(ci), &ci);
-       RTA_PUT(skb, NDA_PROBES, sizeof(probes), &probes);
-       nlh->nlmsg_len   = skb->tail - b;
-       return skb->len;
+       ndm->ndm_flags   = neigh->flags;
+       ndm->ndm_type    = neigh->type;
+       ndm->ndm_ifindex = neigh->dev->ifindex;
 
-nlmsg_failure:
-rtattr_failure:
-       if (locked)
-               read_unlock_bh(&n->lock);
-       skb_trim(skb, b - skb->data);
-       return -1;
+       NLA_PUT(skb, NDA_DST, neigh->tbl->key_len, neigh->primary_key);
+
+       read_lock_bh(&neigh->lock);
+       ndm->ndm_state   = neigh->nud_state;
+       if ((neigh->nud_state & NUD_VALID) &&
+           nla_put(skb, NDA_LLADDR, neigh->dev->addr_len, neigh->ha) < 0) {
+               read_unlock_bh(&neigh->lock);
+               goto nla_put_failure;
+       }
+
+       ci.ndm_used      = now - neigh->used;
+       ci.ndm_confirmed = now - neigh->confirmed;
+       ci.ndm_updated   = now - neigh->updated;
+       ci.ndm_refcnt    = atomic_read(&neigh->refcnt) - 1;
+       read_unlock_bh(&neigh->lock);
+
+       NLA_PUT_U32(skb, NDA_PROBES, atomic_read(&neigh->probes));
+       NLA_PUT(skb, NDA_CACHEINFO, sizeof(ci), &ci);
+
+       return nlmsg_end(skb, nlh);
+
+nla_put_failure:
+       nlmsg_cancel(skb, nlh);
+       return -EMSGSIZE;
 }
 
+static void neigh_update_notify(struct neighbour *neigh)
+{
+       call_netevent_notifiers(NETEVENT_NEIGH_UPDATE, neigh);
+       __neigh_notify(neigh, RTM_NEWNEIGH, 0);
+}
 
 static int neigh_dump_table(struct neigh_table *tbl, struct sk_buff *skb,
                            struct netlink_callback *cb)
 {
+       struct net * net = sock_net(skb->sk);
        struct neighbour *n;
        int rc, h, s_h = cb->args[1];
        int idx, s_idx = idx = cb->args[2];
 
+       read_lock_bh(&tbl->lock);
        for (h = 0; h <= tbl->hash_mask; h++) {
                if (h < s_h)
                        continue;
                if (h > s_h)
                        s_idx = 0;
-               read_lock_bh(&tbl->lock);
-               for (n = tbl->hash_buckets[h], idx = 0; n; n = n->next, idx++) {
-                       if (idx < s_idx)
+               for (n = tbl->hash_buckets[h], idx = 0; n; n = n->next) {
+                       int lidx;
+                       if (dev_net(n->dev) != net)
+                               continue;
+                       lidx = idx++;
+                       if (lidx < s_idx)
                                continue;
                        if (neigh_fill_info(skb, n, NETLINK_CB(cb->skb).pid,
                                            cb->nlh->nlmsg_seq,
@@ -1930,8 +2092,8 @@ static int neigh_dump_table(struct neigh_table *tbl, struct sk_buff *skb,
                                goto out;
                        }
                }
-               read_unlock_bh(&tbl->lock);
        }
+       read_unlock_bh(&tbl->lock);
        rc = skb->len;
 out:
        cb->args[1] = h;
@@ -1939,13 +2101,13 @@ out:
        return rc;
 }
 
-int neigh_dump_info(struct sk_buff *skb, struct netlink_callback *cb)
+static int neigh_dump_info(struct sk_buff *skb, struct netlink_callback *cb)
 {
        struct neigh_table *tbl;
        int t, family, s_t;
 
        read_lock(&neigh_tbl_lock);
-       family = ((struct rtgenmsg *)NLMSG_DATA(cb->nlh))->rtgen_family;
+       family = ((struct rtgenmsg *) nlmsg_data(cb->nlh))->rtgen_family;
        s_t = cb->args[0];
 
        for (tbl = neigh_tables, t = 0; tbl; tbl = tbl->next, t++) {
@@ -2000,7 +2162,7 @@ void __neigh_for_each_release(struct neigh_table *tbl,
                                np = &n->next;
                        write_unlock(&n->lock);
                        if (release)
-                               neigh_release(n);
+                               neigh_cleanup_and_release(n);
                }
        }
 }
@@ -2011,6 +2173,7 @@ EXPORT_SYMBOL(__neigh_for_each_release);
 static struct neighbour *neigh_get_first(struct seq_file *seq)
 {
        struct neigh_seq_state *state = seq->private;
+       struct net *net = seq_file_net(seq);
        struct neigh_table *tbl = state->tbl;
        struct neighbour *n = NULL;
        int bucket = state->bucket;
@@ -2020,6 +2183,8 @@ static struct neighbour *neigh_get_first(struct seq_file *seq)
                n = tbl->hash_buckets[bucket];
 
                while (n) {
+                       if (!net_eq(dev_net(n->dev), net))
+                               goto next;
                        if (state->neigh_sub_iter) {
                                loff_t fakep = 0;
                                void *v;
@@ -2049,6 +2214,7 @@ static struct neighbour *neigh_get_next(struct seq_file *seq,
                                        loff_t *pos)
 {
        struct neigh_seq_state *state = seq->private;
+       struct net *net = seq_file_net(seq);
        struct neigh_table *tbl = state->tbl;
 
        if (state->neigh_sub_iter) {
@@ -2060,6 +2226,8 @@ static struct neighbour *neigh_get_next(struct seq_file *seq,
 
        while (1) {
                while (n) {
+                       if (!net_eq(dev_net(n->dev), net))
+                               goto next;
                        if (state->neigh_sub_iter) {
                                void *v = state->neigh_sub_iter(state, n, pos);
                                if (v)
@@ -2106,6 +2274,7 @@ static struct neighbour *neigh_get_idx(struct seq_file *seq, loff_t *pos)
 static struct pneigh_entry *pneigh_get_first(struct seq_file *seq)
 {
        struct neigh_seq_state *state = seq->private;
+       struct net *net = seq_file_net(seq);
        struct neigh_table *tbl = state->tbl;
        struct pneigh_entry *pn = NULL;
        int bucket = state->bucket;
@@ -2113,6 +2282,8 @@ static struct pneigh_entry *pneigh_get_first(struct seq_file *seq)
        state->flags |= NEIGH_SEQ_IS_PNEIGH;
        for (bucket = 0; bucket <= PNEIGH_HASHMASK; bucket++) {
                pn = tbl->phash_buckets[bucket];
+               while (pn && !net_eq(pneigh_net(pn), net))
+                       pn = pn->next;
                if (pn)
                        break;
        }
@@ -2126,6 +2297,7 @@ static struct pneigh_entry *pneigh_get_next(struct seq_file *seq,
                                            loff_t *pos)
 {
        struct neigh_seq_state *state = seq->private;
+       struct net *net = seq_file_net(seq);
        struct neigh_table *tbl = state->tbl;
 
        pn = pn->next;
@@ -2133,6 +2305,8 @@ static struct pneigh_entry *pneigh_get_next(struct seq_file *seq,
                if (++state->bucket > PNEIGH_HASHMASK)
                        break;
                pn = tbl->phash_buckets[state->bucket];
+               while (pn && !net_eq(pneigh_net(pn), net))
+                       pn = pn->next;
                if (pn)
                        break;
        }
@@ -2170,6 +2344,7 @@ static void *neigh_get_idx_any(struct seq_file *seq, loff_t *pos)
 }
 
 void *neigh_seq_start(struct seq_file *seq, loff_t *pos, struct neigh_table *tbl, unsigned int neigh_seq_flags)
+       __acquires(tbl->lock)
 {
        struct neigh_seq_state *state = seq->private;
        loff_t pos_minus_one;
@@ -2213,6 +2388,7 @@ out:
 EXPORT_SYMBOL(neigh_seq_next);
 
 void neigh_seq_stop(struct seq_file *seq, void *v)
+       __releases(tbl->lock)
 {
        struct neigh_seq_state *state = seq->private;
        struct neigh_table *tbl = state->tbl;
@@ -2231,7 +2407,7 @@ static void *neigh_stat_seq_start(struct seq_file *seq, loff_t *pos)
 
        if (*pos == 0)
                return SEQ_START_TOKEN;
-       
+
        for (cpu = *pos-1; cpu < NR_CPUS; ++cpu) {
                if (!cpu_possible(cpu))
                        continue;
@@ -2295,7 +2471,7 @@ static int neigh_stat_seq_show(struct seq_file *seq, void *v)
        return 0;
 }
 
-static struct seq_operations neigh_stat_seq_ops = {
+static const struct seq_operations neigh_stat_seq_ops = {
        .start  = neigh_stat_seq_start,
        .next   = neigh_stat_seq_next,
        .stop   = neigh_stat_seq_stop,
@@ -2313,7 +2489,7 @@ static int neigh_stat_seq_open(struct inode *inode, struct file *file)
        return ret;
 };
 
-static struct file_operations neigh_stat_seq_fops = {
+static const struct file_operations neigh_stat_seq_fops = {
        .owner   = THIS_MODULE,
        .open    = neigh_stat_seq_open,
        .read    = seq_read,
@@ -2323,56 +2499,52 @@ static struct file_operations neigh_stat_seq_fops = {
 
 #endif /* CONFIG_PROC_FS */
 
-#ifdef CONFIG_ARPD
-void neigh_app_ns(struct neighbour *n)
+static inline size_t neigh_nlmsg_size(void)
 {
-       struct nlmsghdr  *nlh;
-       int size = NLMSG_SPACE(sizeof(struct ndmsg) + 256);
-       struct sk_buff *skb = alloc_skb(size, GFP_ATOMIC);
-
-       if (!skb)
-               return;
-
-       if (neigh_fill_info(skb, n, 0, 0, RTM_GETNEIGH, 0) < 0) {
-               kfree_skb(skb);
-               return;
-       }
-       nlh                        = (struct nlmsghdr *)skb->data;
-       nlh->nlmsg_flags           = NLM_F_REQUEST;
-       NETLINK_CB(skb).dst_group  = RTNLGRP_NEIGH;
-       netlink_broadcast(rtnl, skb, 0, RTNLGRP_NEIGH, GFP_ATOMIC);
+       return NLMSG_ALIGN(sizeof(struct ndmsg))
+              + nla_total_size(MAX_ADDR_LEN) /* NDA_DST */
+              + nla_total_size(MAX_ADDR_LEN) /* NDA_LLADDR */
+              + nla_total_size(sizeof(struct nda_cacheinfo))
+              + nla_total_size(4); /* NDA_PROBES */
 }
 
-static void neigh_app_notify(struct neighbour *n)
+static void __neigh_notify(struct neighbour *n, int type, int flags)
 {
-       struct nlmsghdr *nlh;
-       int size = NLMSG_SPACE(sizeof(struct ndmsg) + 256);
-       struct sk_buff *skb = alloc_skb(size, GFP_ATOMIC);
+       struct net *net = dev_net(n->dev);
+       struct sk_buff *skb;
+       int err = -ENOBUFS;
 
-       if (!skb)
-               return;
+       skb = nlmsg_new(neigh_nlmsg_size(), GFP_ATOMIC);
+       if (skb == NULL)
+               goto errout;
 
-       if (neigh_fill_info(skb, n, 0, 0, RTM_NEWNEIGH, 0) < 0) {
+       err = neigh_fill_info(skb, n, 0, 0, type, flags);
+       if (err < 0) {
+               /* -EMSGSIZE implies BUG in neigh_nlmsg_size() */
+               WARN_ON(err == -EMSGSIZE);
                kfree_skb(skb);
-               return;
+               goto errout;
        }
-       nlh                        = (struct nlmsghdr *)skb->data;
-       NETLINK_CB(skb).dst_group  = RTNLGRP_NEIGH;
-       netlink_broadcast(rtnl, skb, 0, RTNLGRP_NEIGH, GFP_ATOMIC);
+       err = rtnl_notify(skb, net, 0, RTNLGRP_NEIGH, NULL, GFP_ATOMIC);
+errout:
+       if (err < 0)
+               rtnl_set_sk_err(net, RTNLGRP_NEIGH, err);
 }
 
+#ifdef CONFIG_ARPD
+void neigh_app_ns(struct neighbour *n)
+{
+       __neigh_notify(n, RTM_GETNEIGH, NLM_F_REQUEST);
+}
 #endif /* CONFIG_ARPD */
 
 #ifdef CONFIG_SYSCTL
 
 static struct neigh_sysctl_table {
        struct ctl_table_header *sysctl_header;
-       ctl_table               neigh_vars[__NET_NEIGH_MAX];
-       ctl_table               neigh_dev[2];
-       ctl_table               neigh_neigh_dir[2];
-       ctl_table               neigh_proto_dir[2];
-       ctl_table               neigh_root_dir[2];
-} neigh_sysctl_template = {
+       struct ctl_table neigh_vars[__NET_NEIGH_MAX];
+       char *dev_name;
+} neigh_sysctl_template __read_mostly = {
        .neigh_vars = {
                {
                        .ctl_name       = NET_NEIGH_MCAST_SOLICIT,
@@ -2396,7 +2568,6 @@ static struct neigh_sysctl_table {
                        .proc_handler   = &proc_dointvec,
                },
                {
-                       .ctl_name       = NET_NEIGH_RETRANS_TIME,
                        .procname       = "retrans_time",
                        .maxlen         = sizeof(int),
                        .mode           = 0644,
@@ -2441,27 +2612,40 @@ static struct neigh_sysctl_table {
                        .proc_handler   = &proc_dointvec,
                },
                {
-                       .ctl_name       = NET_NEIGH_ANYCAST_DELAY,
                        .procname       = "anycast_delay",
                        .maxlen         = sizeof(int),
                        .mode           = 0644,
                        .proc_handler   = &proc_dointvec_userhz_jiffies,
                },
                {
-                       .ctl_name       = NET_NEIGH_PROXY_DELAY,
                        .procname       = "proxy_delay",
                        .maxlen         = sizeof(int),
                        .mode           = 0644,
                        .proc_handler   = &proc_dointvec_userhz_jiffies,
                },
                {
-                       .ctl_name       = NET_NEIGH_LOCKTIME,
                        .procname       = "locktime",
                        .maxlen         = sizeof(int),
                        .mode           = 0644,
                        .proc_handler   = &proc_dointvec_userhz_jiffies,
                },
                {
+                       .ctl_name       = NET_NEIGH_RETRANS_TIME_MS,
+                       .procname       = "retrans_time_ms",
+                       .maxlen         = sizeof(int),
+                       .mode           = 0644,
+                       .proc_handler   = &proc_dointvec_ms_jiffies,
+                       .strategy       = &sysctl_ms_jiffies,
+               },
+               {
+                       .ctl_name       = NET_NEIGH_REACHABLE_TIME_MS,
+                       .procname       = "base_reachable_time_ms",
+                       .maxlen         = sizeof(int),
+                       .mode           = 0644,
+                       .proc_handler   = &proc_dointvec_ms_jiffies,
+                       .strategy       = &sysctl_ms_jiffies,
+               },
+               {
                        .ctl_name       = NET_NEIGH_GC_INTERVAL,
                        .procname       = "gc_interval",
                        .maxlen         = sizeof(int),
@@ -2490,62 +2674,34 @@ static struct neigh_sysctl_table {
                        .mode           = 0644,
                        .proc_handler   = &proc_dointvec,
                },
-               {
-                       .ctl_name       = NET_NEIGH_RETRANS_TIME_MS,
-                       .procname       = "retrans_time_ms",
-                       .maxlen         = sizeof(int),
-                       .mode           = 0644,
-                       .proc_handler   = &proc_dointvec_ms_jiffies,
-                       .strategy       = &sysctl_ms_jiffies,
-               },
-               {
-                       .ctl_name       = NET_NEIGH_REACHABLE_TIME_MS,
-                       .procname       = "base_reachable_time_ms",
-                       .maxlen         = sizeof(int),
-                       .mode           = 0644,
-                       .proc_handler   = &proc_dointvec_ms_jiffies,
-                       .strategy       = &sysctl_ms_jiffies,
-               },
-       },
-       .neigh_dev = {
-               {
-                       .ctl_name       = NET_PROTO_CONF_DEFAULT,
-                       .procname       = "default",
-                       .mode           = 0555,
-               },
-       },
-       .neigh_neigh_dir = {
-               {
-                       .procname       = "neigh",
-                       .mode           = 0555,
-               },
-       },
-       .neigh_proto_dir = {
-               {
-                       .mode           = 0555,
-               },
-       },
-       .neigh_root_dir = {
-               {
-                       .ctl_name       = CTL_NET,
-                       .procname       = "net",
-                       .mode           = 0555,
-               },
+               {},
        },
 };
 
 int neigh_sysctl_register(struct net_device *dev, struct neigh_parms *p,
-                         int p_id, int pdev_id, char *p_name, 
+                         int p_id, int pdev_id, char *p_name,
                          proc_handler *handler, ctl_handler *strategy)
 {
-       struct neigh_sysctl_table *t = kmalloc(sizeof(*t), GFP_KERNEL);
+       struct neigh_sysctl_table *t;
        const char *dev_name_source = NULL;
-       char *dev_name = NULL;
-       int err = 0;
 
+#define NEIGH_CTL_PATH_ROOT    0
+#define NEIGH_CTL_PATH_PROTO   1
+#define NEIGH_CTL_PATH_NEIGH   2
+#define NEIGH_CTL_PATH_DEV     3
+
+       struct ctl_path neigh_path[] = {
+               { .procname = "net",     .ctl_name = CTL_NET, },
+               { .procname = "proto",   .ctl_name = 0, },
+               { .procname = "neigh",   .ctl_name = 0, },
+               { .procname = "default", .ctl_name = NET_PROTO_CONF_DEFAULT, },
+               { },
+       };
+
+       t = kmemdup(&neigh_sysctl_template, sizeof(*t), GFP_KERNEL);
        if (!t)
-               return -ENOBUFS;
-       memcpy(t, &neigh_sysctl_template, sizeof(*t));
+               goto err;
+
        t->neigh_vars[0].data  = &p->mcast_probes;
        t->neigh_vars[1].data  = &p->ucast_probes;
        t->neigh_vars[2].data  = &p->app_probes;
@@ -2558,77 +2714,73 @@ int neigh_sysctl_register(struct net_device *dev, struct neigh_parms *p,
        t->neigh_vars[9].data  = &p->anycast_delay;
        t->neigh_vars[10].data = &p->proxy_delay;
        t->neigh_vars[11].data = &p->locktime;
+       t->neigh_vars[12].data  = &p->retrans_time;
+       t->neigh_vars[13].data  = &p->base_reachable_time;
 
        if (dev) {
                dev_name_source = dev->name;
-               t->neigh_dev[0].ctl_name = dev->ifindex;
-               t->neigh_vars[12].procname = NULL;
-               t->neigh_vars[13].procname = NULL;
-               t->neigh_vars[14].procname = NULL;
-               t->neigh_vars[15].procname = NULL;
+               neigh_path[NEIGH_CTL_PATH_DEV].ctl_name = dev->ifindex;
+               /* Terminate the table early */
+               memset(&t->neigh_vars[14], 0, sizeof(t->neigh_vars[14]));
        } else {
-               dev_name_source = t->neigh_dev[0].procname;
-               t->neigh_vars[12].data = (int *)(p + 1);
-               t->neigh_vars[13].data = (int *)(p + 1) + 1;
-               t->neigh_vars[14].data = (int *)(p + 1) + 2;
-               t->neigh_vars[15].data = (int *)(p + 1) + 3;
+               dev_name_source = neigh_path[NEIGH_CTL_PATH_DEV].procname;
+               t->neigh_vars[14].data = (int *)(p + 1);
+               t->neigh_vars[15].data = (int *)(p + 1) + 1;
+               t->neigh_vars[16].data = (int *)(p + 1) + 2;
+               t->neigh_vars[17].data = (int *)(p + 1) + 3;
        }
 
-       t->neigh_vars[16].data  = &p->retrans_time;
-       t->neigh_vars[17].data  = &p->base_reachable_time;
 
        if (handler || strategy) {
                /* RetransTime */
                t->neigh_vars[3].proc_handler = handler;
                t->neigh_vars[3].strategy = strategy;
                t->neigh_vars[3].extra1 = dev;
+               if (!strategy)
+                       t->neigh_vars[3].ctl_name = CTL_UNNUMBERED;
                /* ReachableTime */
                t->neigh_vars[4].proc_handler = handler;
                t->neigh_vars[4].strategy = strategy;
                t->neigh_vars[4].extra1 = dev;
+               if (!strategy)
+                       t->neigh_vars[4].ctl_name = CTL_UNNUMBERED;
                /* RetransTime (in milliseconds)*/
-               t->neigh_vars[16].proc_handler = handler;
-               t->neigh_vars[16].strategy = strategy;
-               t->neigh_vars[16].extra1 = dev;
+               t->neigh_vars[12].proc_handler = handler;
+               t->neigh_vars[12].strategy = strategy;
+               t->neigh_vars[12].extra1 = dev;
+               if (!strategy)
+                       t->neigh_vars[12].ctl_name = CTL_UNNUMBERED;
                /* ReachableTime (in milliseconds) */
-               t->neigh_vars[17].proc_handler = handler;
-               t->neigh_vars[17].strategy = strategy;
-               t->neigh_vars[17].extra1 = dev;
+               t->neigh_vars[13].proc_handler = handler;
+               t->neigh_vars[13].strategy = strategy;
+               t->neigh_vars[13].extra1 = dev;
+               if (!strategy)
+                       t->neigh_vars[13].ctl_name = CTL_UNNUMBERED;
        }
 
-       dev_name = kstrdup(dev_name_source, GFP_KERNEL);
-       if (!dev_name) {
-               err = -ENOBUFS;
+       t->dev_name = kstrdup(dev_name_source, GFP_KERNEL);
+       if (!t->dev_name)
                goto free;
-       }
 
-       t->neigh_dev[0].procname = dev_name;
+       neigh_path[NEIGH_CTL_PATH_DEV].procname = t->dev_name;
+       neigh_path[NEIGH_CTL_PATH_NEIGH].ctl_name = pdev_id;
+       neigh_path[NEIGH_CTL_PATH_PROTO].procname = p_name;
+       neigh_path[NEIGH_CTL_PATH_PROTO].ctl_name = p_id;
 
-       t->neigh_neigh_dir[0].ctl_name = pdev_id;
-
-       t->neigh_proto_dir[0].procname = p_name;
-       t->neigh_proto_dir[0].ctl_name = p_id;
-
-       t->neigh_dev[0].child          = t->neigh_vars;
-       t->neigh_neigh_dir[0].child    = t->neigh_dev;
-       t->neigh_proto_dir[0].child    = t->neigh_neigh_dir;
-       t->neigh_root_dir[0].child     = t->neigh_proto_dir;
-
-       t->sysctl_header = register_sysctl_table(t->neigh_root_dir, 0);
-       if (!t->sysctl_header) {
-               err = -ENOBUFS;
+       t->sysctl_header =
+               register_net_sysctl_table(neigh_parms_net(p), neigh_path, t->neigh_vars);
+       if (!t->sysctl_header)
                goto free_procname;
-       }
+
        p->sysctl_table = t;
        return 0;
 
-       /* error path */
- free_procname:
-       kfree(dev_name);
- free:
+free_procname:
+       kfree(t->dev_name);
+free:
        kfree(t);
-
-       return err;
+err:
+       return -ENOBUFS;
 }
 
 void neigh_sysctl_unregister(struct neigh_parms *p)
@@ -2637,22 +2789,33 @@ void neigh_sysctl_unregister(struct neigh_parms *p)
                struct neigh_sysctl_table *t = p->sysctl_table;
                p->sysctl_table = NULL;
                unregister_sysctl_table(t->sysctl_header);
-               kfree(t->neigh_dev[0].procname);
+               kfree(t->dev_name);
                kfree(t);
        }
 }
 
 #endif /* CONFIG_SYSCTL */
 
+static int __init neigh_init(void)
+{
+       rtnl_register(PF_UNSPEC, RTM_NEWNEIGH, neigh_add, NULL);
+       rtnl_register(PF_UNSPEC, RTM_DELNEIGH, neigh_delete, NULL);
+       rtnl_register(PF_UNSPEC, RTM_GETNEIGH, NULL, neigh_dump_info);
+
+       rtnl_register(PF_UNSPEC, RTM_GETNEIGHTBL, NULL, neightbl_dump_info);
+       rtnl_register(PF_UNSPEC, RTM_SETNEIGHTBL, neightbl_set, NULL);
+
+       return 0;
+}
+
+subsys_initcall(neigh_init);
+
 EXPORT_SYMBOL(__neigh_event_send);
-EXPORT_SYMBOL(neigh_add);
 EXPORT_SYMBOL(neigh_changeaddr);
 EXPORT_SYMBOL(neigh_compat_output);
 EXPORT_SYMBOL(neigh_connected_output);
 EXPORT_SYMBOL(neigh_create);
-EXPORT_SYMBOL(neigh_delete);
 EXPORT_SYMBOL(neigh_destroy);
-EXPORT_SYMBOL(neigh_dump_info);
 EXPORT_SYMBOL(neigh_event_ns);
 EXPORT_SYMBOL(neigh_ifdown);
 EXPORT_SYMBOL(neigh_lookup);
@@ -2663,12 +2826,11 @@ EXPORT_SYMBOL(neigh_rand_reach_time);
 EXPORT_SYMBOL(neigh_resolve_output);
 EXPORT_SYMBOL(neigh_table_clear);
 EXPORT_SYMBOL(neigh_table_init);
+EXPORT_SYMBOL(neigh_table_init_no_netlink);
 EXPORT_SYMBOL(neigh_update);
-EXPORT_SYMBOL(neigh_update_hhs);
 EXPORT_SYMBOL(pneigh_enqueue);
 EXPORT_SYMBOL(pneigh_lookup);
-EXPORT_SYMBOL(neightbl_dump_info);
-EXPORT_SYMBOL(neightbl_set);
+EXPORT_SYMBOL_GPL(__pneigh_lookup);
 
 #ifdef CONFIG_ARPD
 EXPORT_SYMBOL(neigh_app_ns);