xfrm: force a dst reference in __xfrm_route_forward()
[safe/jmp/linux-2.6] / net / xfrm / xfrm_policy.c
index 38822b3..4bf27d9 100644 (file)
 
 #include "xfrm_hash.h"
 
-int sysctl_xfrm_larval_drop __read_mostly = 1;
-
 DEFINE_MUTEX(xfrm_cfg_mutex);
 EXPORT_SYMBOL(xfrm_cfg_mutex);
 
+static DEFINE_SPINLOCK(xfrm_policy_sk_bundle_lock);
+static struct dst_entry *xfrm_policy_sk_bundles;
 static DEFINE_RWLOCK(xfrm_policy_lock);
 
 static DEFINE_RWLOCK(xfrm_policy_afinfo_lock);
@@ -46,12 +46,13 @@ static struct xfrm_policy_afinfo *xfrm_policy_afinfo[NPROTO];
 
 static struct kmem_cache *xfrm_dst_cache __read_mostly;
 
-static HLIST_HEAD(xfrm_policy_gc_list);
-static DEFINE_SPINLOCK(xfrm_policy_gc_lock);
-
 static struct xfrm_policy_afinfo *xfrm_policy_get_afinfo(unsigned short family);
 static void xfrm_policy_put_afinfo(struct xfrm_policy_afinfo *afinfo);
 static void xfrm_init_pmtu(struct dst_entry *dst);
+static int stale_bundle(struct dst_entry *dst);
+
+static struct xfrm_policy *__xfrm_policy_unlink(struct xfrm_policy *pol,
+                                               int dir);
 
 static inline int
 __xfrm4_selector_match(struct xfrm_selector *sel, struct flowi *fl)
@@ -155,7 +156,7 @@ static void xfrm_policy_timer(unsigned long data)
 
        read_lock(&xp->lock);
 
-       if (xp->walk.dead)
+       if (unlikely(xp->walk.dead))
                goto out;
 
        dir = xfrm_policy_id2dir(xp->index);
@@ -215,6 +216,35 @@ expired:
        xfrm_pol_put(xp);
 }
 
+static struct flow_cache_object *xfrm_policy_flo_get(struct flow_cache_object *flo)
+{
+       struct xfrm_policy *pol = container_of(flo, struct xfrm_policy, flo);
+
+       if (unlikely(pol->walk.dead))
+               flo = NULL;
+       else
+               xfrm_pol_hold(pol);
+
+       return flo;
+}
+
+static int xfrm_policy_flo_check(struct flow_cache_object *flo)
+{
+       struct xfrm_policy *pol = container_of(flo, struct xfrm_policy, flo);
+
+       return !pol->walk.dead;
+}
+
+static void xfrm_policy_flo_delete(struct flow_cache_object *flo)
+{
+       xfrm_pol_put(container_of(flo, struct xfrm_policy, flo));
+}
+
+static const struct flow_cache_ops xfrm_policy_fc_ops = {
+       .get = xfrm_policy_flo_get,
+       .check = xfrm_policy_flo_check,
+       .delete = xfrm_policy_flo_delete,
+};
 
 /* Allocate xfrm_policy. Not used here, it is supposed to be used by pfkeyv2
  * SPD calls.
@@ -235,6 +265,7 @@ struct xfrm_policy *xfrm_policy_alloc(struct net *net, gfp_t gfp)
                atomic_set(&policy->refcnt, 1);
                setup_timer(&policy->timer, xfrm_policy_timer,
                                (unsigned long)policy);
+               policy->flo.ops = &xfrm_policy_fc_ops;
        }
        return policy;
 }
@@ -246,8 +277,6 @@ void xfrm_policy_destroy(struct xfrm_policy *policy)
 {
        BUG_ON(!policy->walk.dead);
 
-       BUG_ON(policy->bundles);
-
        if (del_timer(&policy->timer))
                BUG();
 
@@ -256,63 +285,20 @@ void xfrm_policy_destroy(struct xfrm_policy *policy)
 }
 EXPORT_SYMBOL(xfrm_policy_destroy);
 
-static void xfrm_policy_gc_kill(struct xfrm_policy *policy)
-{
-       struct dst_entry *dst;
-
-       while ((dst = policy->bundles) != NULL) {
-               policy->bundles = dst->next;
-               dst_free(dst);
-       }
-
-       if (del_timer(&policy->timer))
-               atomic_dec(&policy->refcnt);
-
-       if (atomic_read(&policy->refcnt) > 1)
-               flow_cache_flush();
-
-       xfrm_pol_put(policy);
-}
-
-static void xfrm_policy_gc_task(struct work_struct *work)
-{
-       struct xfrm_policy *policy;
-       struct hlist_node *entry, *tmp;
-       struct hlist_head gc_list;
-
-       spin_lock_bh(&xfrm_policy_gc_lock);
-       gc_list.first = xfrm_policy_gc_list.first;
-       INIT_HLIST_HEAD(&xfrm_policy_gc_list);
-       spin_unlock_bh(&xfrm_policy_gc_lock);
-
-       hlist_for_each_entry_safe(policy, entry, tmp, &gc_list, bydst)
-               xfrm_policy_gc_kill(policy);
-}
-static DECLARE_WORK(xfrm_policy_gc_work, xfrm_policy_gc_task);
-
 /* Rule must be locked. Release descentant resources, announce
  * entry dead. The rule must be unlinked from lists to the moment.
  */
 
 static void xfrm_policy_kill(struct xfrm_policy *policy)
 {
-       int dead;
-
-       write_lock_bh(&policy->lock);
-       dead = policy->walk.dead;
        policy->walk.dead = 1;
-       write_unlock_bh(&policy->lock);
 
-       if (unlikely(dead)) {
-               WARN_ON(1);
-               return;
-       }
+       atomic_inc(&policy->genid);
 
-       spin_lock_bh(&xfrm_policy_gc_lock);
-       hlist_add_head(&policy->bydst, &xfrm_policy_gc_list);
-       spin_unlock_bh(&xfrm_policy_gc_lock);
+       if (del_timer(&policy->timer))
+               xfrm_pol_put(policy);
 
-       schedule_work(&xfrm_policy_gc_work);
+       xfrm_pol_put(policy);
 }
 
 static unsigned int xfrm_policy_hashmax __read_mostly = 1 * 1024 * 1024;
@@ -468,16 +454,16 @@ static inline int xfrm_byidx_should_resize(struct net *net, int total)
        return 0;
 }
 
-void xfrm_spd_getinfo(struct xfrmk_spdinfo *si)
+void xfrm_spd_getinfo(struct net *net, struct xfrmk_spdinfo *si)
 {
        read_lock_bh(&xfrm_policy_lock);
-       si->incnt = init_net.xfrm.policy_count[XFRM_POLICY_IN];
-       si->outcnt = init_net.xfrm.policy_count[XFRM_POLICY_OUT];
-       si->fwdcnt = init_net.xfrm.policy_count[XFRM_POLICY_FWD];
-       si->inscnt = init_net.xfrm.policy_count[XFRM_POLICY_IN+XFRM_POLICY_MAX];
-       si->outscnt = init_net.xfrm.policy_count[XFRM_POLICY_OUT+XFRM_POLICY_MAX];
-       si->fwdscnt = init_net.xfrm.policy_count[XFRM_POLICY_FWD+XFRM_POLICY_MAX];
-       si->spdhcnt = init_net.xfrm.policy_idx_hmask;
+       si->incnt = net->xfrm.policy_count[XFRM_POLICY_IN];
+       si->outcnt = net->xfrm.policy_count[XFRM_POLICY_OUT];
+       si->fwdcnt = net->xfrm.policy_count[XFRM_POLICY_FWD];
+       si->inscnt = net->xfrm.policy_count[XFRM_POLICY_IN+XFRM_POLICY_MAX];
+       si->outscnt = net->xfrm.policy_count[XFRM_POLICY_OUT+XFRM_POLICY_MAX];
+       si->fwdscnt = net->xfrm.policy_count[XFRM_POLICY_FWD+XFRM_POLICY_MAX];
+       si->spdhcnt = net->xfrm.policy_idx_hmask;
        si->spdhmcnt = xfrm_policy_hashmax;
        read_unlock_bh(&xfrm_policy_lock);
 }
@@ -554,7 +540,7 @@ int xfrm_policy_insert(int dir, struct xfrm_policy *policy, int excl)
        struct xfrm_policy *delpol;
        struct hlist_head *chain;
        struct hlist_node *entry, *newpos;
-       struct dst_entry *gc_list;
+       u32 mark = policy->mark.v & policy->mark.m;
 
        write_lock_bh(&xfrm_policy_lock);
        chain = policy_hash_bysel(net, &policy->selector, policy->family, dir);
@@ -563,6 +549,7 @@ int xfrm_policy_insert(int dir, struct xfrm_policy *policy, int excl)
        hlist_for_each_entry(pol, entry, chain, bydst) {
                if (pol->type == policy->type &&
                    !selector_cmp(&pol->selector, &policy->selector) &&
+                   (mark & pol->mark.m) == pol->mark.v &&
                    xfrm_sec_ctx_match(pol->security, policy->security) &&
                    !WARN_ON(delpol)) {
                        if (excl) {
@@ -586,12 +573,8 @@ int xfrm_policy_insert(int dir, struct xfrm_policy *policy, int excl)
        xfrm_pol_hold(policy);
        net->xfrm.policy_count[dir]++;
        atomic_inc(&flow_cache_genid);
-       if (delpol) {
-               hlist_del(&delpol->bydst);
-               hlist_del(&delpol->byidx);
-               list_del(&delpol->walk.all);
-               net->xfrm.policy_count[dir]--;
-       }
+       if (delpol)
+               __xfrm_policy_unlink(delpol, dir);
        policy->index = delpol ? delpol->index : xfrm_gen_index(net, dir);
        hlist_add_head(&policy->byidx, net->xfrm.policy_byidx+idx_hash(net, policy->index));
        policy->curlft.add_time = get_seconds();
@@ -606,40 +589,12 @@ int xfrm_policy_insert(int dir, struct xfrm_policy *policy, int excl)
        else if (xfrm_bydst_should_resize(net, dir, NULL))
                schedule_work(&net->xfrm.policy_hash_work);
 
-       read_lock_bh(&xfrm_policy_lock);
-       gc_list = NULL;
-       entry = &policy->bydst;
-       hlist_for_each_entry_continue(policy, entry, bydst) {
-               struct dst_entry *dst;
-
-               write_lock(&policy->lock);
-               dst = policy->bundles;
-               if (dst) {
-                       struct dst_entry *tail = dst;
-                       while (tail->next)
-                               tail = tail->next;
-                       tail->next = gc_list;
-                       gc_list = dst;
-
-                       policy->bundles = NULL;
-               }
-               write_unlock(&policy->lock);
-       }
-       read_unlock_bh(&xfrm_policy_lock);
-
-       while (gc_list) {
-               struct dst_entry *dst = gc_list;
-
-               gc_list = dst->next;
-               dst_free(dst);
-       }
-
        return 0;
 }
 EXPORT_SYMBOL(xfrm_policy_insert);
 
-struct xfrm_policy *xfrm_policy_bysel_ctx(struct net *net, u8 type, int dir,
-                                         struct xfrm_selector *sel,
+struct xfrm_policy *xfrm_policy_bysel_ctx(struct net *net, u32 mark, u8 type,
+                                         int dir, struct xfrm_selector *sel,
                                          struct xfrm_sec_ctx *ctx, int delete,
                                          int *err)
 {
@@ -653,6 +608,7 @@ struct xfrm_policy *xfrm_policy_bysel_ctx(struct net *net, u8 type, int dir,
        ret = NULL;
        hlist_for_each_entry(pol, entry, chain, bydst) {
                if (pol->type == type &&
+                   (mark & pol->mark.m) == pol->mark.v &&
                    !selector_cmp(sel, &pol->selector) &&
                    xfrm_sec_ctx_match(ctx, pol->security)) {
                        xfrm_pol_hold(pol);
@@ -663,10 +619,7 @@ struct xfrm_policy *xfrm_policy_bysel_ctx(struct net *net, u8 type, int dir,
                                        write_unlock_bh(&xfrm_policy_lock);
                                        return pol;
                                }
-                               hlist_del(&pol->bydst);
-                               hlist_del(&pol->byidx);
-                               list_del(&pol->walk.all);
-                               net->xfrm.policy_count[dir]--;
+                               __xfrm_policy_unlink(pol, dir);
                        }
                        ret = pol;
                        break;
@@ -674,16 +627,14 @@ struct xfrm_policy *xfrm_policy_bysel_ctx(struct net *net, u8 type, int dir,
        }
        write_unlock_bh(&xfrm_policy_lock);
 
-       if (ret && delete) {
-               atomic_inc(&flow_cache_genid);
+       if (ret && delete)
                xfrm_policy_kill(ret);
-       }
        return ret;
 }
 EXPORT_SYMBOL(xfrm_policy_bysel_ctx);
 
-struct xfrm_policy *xfrm_policy_byid(struct net *net, u8 type, int dir, u32 id,
-                                    int delete, int *err)
+struct xfrm_policy *xfrm_policy_byid(struct net *net, u32 mark, u8 type,
+                                    int dir, u32 id, int delete, int *err)
 {
        struct xfrm_policy *pol, *ret;
        struct hlist_head *chain;
@@ -698,7 +649,8 @@ struct xfrm_policy *xfrm_policy_byid(struct net *net, u8 type, int dir, u32 id,
        chain = net->xfrm.policy_byidx + idx_hash(net, id);
        ret = NULL;
        hlist_for_each_entry(pol, entry, chain, byidx) {
-               if (pol->type == type && pol->index == id) {
+               if (pol->type == type && pol->index == id &&
+                   (mark & pol->mark.m) == pol->mark.v) {
                        xfrm_pol_hold(pol);
                        if (delete) {
                                *err = security_xfrm_policy_delete(
@@ -707,10 +659,7 @@ struct xfrm_policy *xfrm_policy_byid(struct net *net, u8 type, int dir, u32 id,
                                        write_unlock_bh(&xfrm_policy_lock);
                                        return pol;
                                }
-                               hlist_del(&pol->bydst);
-                               hlist_del(&pol->byidx);
-                               list_del(&pol->walk.all);
-                               net->xfrm.policy_count[dir]--;
+                               __xfrm_policy_unlink(pol, dir);
                        }
                        ret = pol;
                        break;
@@ -718,10 +667,8 @@ struct xfrm_policy *xfrm_policy_byid(struct net *net, u8 type, int dir, u32 id,
        }
        write_unlock_bh(&xfrm_policy_lock);
 
-       if (ret && delete) {
-               atomic_inc(&flow_cache_genid);
+       if (ret && delete)
                xfrm_policy_kill(ret);
-       }
        return ret;
 }
 EXPORT_SYMBOL(xfrm_policy_byid);
@@ -780,7 +727,7 @@ xfrm_policy_flush_secctx_check(struct net *net, u8 type, struct xfrm_audit *audi
 
 int xfrm_policy_flush(struct net *net, u8 type, struct xfrm_audit *audit_info)
 {
-       int dir, err = 0;
+       int dir, err = 0, cnt = 0;
 
        write_lock_bh(&xfrm_policy_lock);
 
@@ -791,24 +738,22 @@ int xfrm_policy_flush(struct net *net, u8 type, struct xfrm_audit *audit_info)
        for (dir = 0; dir < XFRM_POLICY_MAX; dir++) {
                struct xfrm_policy *pol;
                struct hlist_node *entry;
-               int i, killed;
+               int i;
 
-               killed = 0;
        again1:
                hlist_for_each_entry(pol, entry,
                                     &net->xfrm.policy_inexact[dir], bydst) {
                        if (pol->type != type)
                                continue;
-                       hlist_del(&pol->bydst);
-                       hlist_del(&pol->byidx);
+                       __xfrm_policy_unlink(pol, dir);
                        write_unlock_bh(&xfrm_policy_lock);
+                       cnt++;
 
                        xfrm_audit_policy_delete(pol, 1, audit_info->loginuid,
                                                 audit_info->sessionid,
                                                 audit_info->secid);
 
                        xfrm_policy_kill(pol);
-                       killed++;
 
                        write_lock_bh(&xfrm_policy_lock);
                        goto again1;
@@ -821,26 +766,24 @@ int xfrm_policy_flush(struct net *net, u8 type, struct xfrm_audit *audit_info)
                                             bydst) {
                                if (pol->type != type)
                                        continue;
-                               hlist_del(&pol->bydst);
-                               hlist_del(&pol->byidx);
-                               list_del(&pol->walk.all);
+                               __xfrm_policy_unlink(pol, dir);
                                write_unlock_bh(&xfrm_policy_lock);
+                               cnt++;
 
                                xfrm_audit_policy_delete(pol, 1,
                                                         audit_info->loginuid,
                                                         audit_info->sessionid,
                                                         audit_info->secid);
                                xfrm_policy_kill(pol);
-                               killed++;
 
                                write_lock_bh(&xfrm_policy_lock);
                                goto again2;
                        }
                }
 
-               net->xfrm.policy_count[dir] -= killed;
        }
-       atomic_inc(&flow_cache_genid);
+       if (!cnt)
+               err = -ESRCH;
 out:
        write_unlock_bh(&xfrm_policy_lock);
        return err;
@@ -925,6 +868,7 @@ static int xfrm_policy_match(struct xfrm_policy *pol, struct flowi *fl,
        int match, ret = -ESRCH;
 
        if (pol->family != family ||
+           (fl->mark & pol->mark.m) != pol->mark.v ||
            pol->type != type)
                return ret;
 
@@ -993,32 +937,37 @@ fail:
        return ret;
 }
 
-static int xfrm_policy_lookup(struct net *net, struct flowi *fl, u16 family,
-                             u8 dir, void **objp, atomic_t **obj_refp)
+static struct xfrm_policy *
+__xfrm_policy_lookup(struct net *net, struct flowi *fl, u16 family, u8 dir)
 {
+#ifdef CONFIG_XFRM_SUB_POLICY
        struct xfrm_policy *pol;
-       int err = 0;
 
-#ifdef CONFIG_XFRM_SUB_POLICY
        pol = xfrm_policy_lookup_bytype(net, XFRM_POLICY_TYPE_SUB, fl, family, dir);
-       if (IS_ERR(pol)) {
-               err = PTR_ERR(pol);
-               pol = NULL;
-       }
-       if (pol || err)
-               goto end;
-#endif
-       pol = xfrm_policy_lookup_bytype(net, XFRM_POLICY_TYPE_MAIN, fl, family, dir);
-       if (IS_ERR(pol)) {
-               err = PTR_ERR(pol);
-               pol = NULL;
-       }
-#ifdef CONFIG_XFRM_SUB_POLICY
-end:
+       if (pol != NULL)
+               return pol;
 #endif
-       if ((*objp = (void *) pol) != NULL)
-               *obj_refp = &pol->refcnt;
-       return err;
+       return xfrm_policy_lookup_bytype(net, XFRM_POLICY_TYPE_MAIN, fl, family, dir);
+}
+
+static struct flow_cache_object *
+xfrm_policy_lookup(struct net *net, struct flowi *fl, u16 family,
+                  u8 dir, struct flow_cache_object *old_obj, void *ctx)
+{
+       struct xfrm_policy *pol;
+
+       if (old_obj)
+               xfrm_pol_put(container_of(old_obj, struct xfrm_policy, flo));
+
+       pol = __xfrm_policy_lookup(net, fl, family, dir);
+       if (IS_ERR_OR_NULL(pol))
+               return ERR_CAST(pol);
+
+       /* Resolver returns two references:
+        * one for cache and one for caller of flow_cache_lookup() */
+       xfrm_pol_hold(pol);
+
+       return &pol->flo;
 }
 
 static inline int policy_to_flow_dir(int dir)
@@ -1049,6 +998,10 @@ static struct xfrm_policy *xfrm_sk_policy_lookup(struct sock *sk, int dir, struc
                int err = 0;
 
                if (match) {
+                       if ((sk->sk_mark & pol->mark.m) != pol->mark.v) {
+                               pol = NULL;
+                               goto out;
+                       }
                        err = security_xfrm_policy_lookup(pol->security,
                                                      fl->secid,
                                                      policy_to_flow_dir(dir));
@@ -1061,6 +1014,7 @@ static struct xfrm_policy *xfrm_sk_policy_lookup(struct sock *sk, int dir, struc
                } else
                        pol = NULL;
        }
+out:
        read_unlock_bh(&xfrm_policy_lock);
        return pol;
 }
@@ -1103,8 +1057,6 @@ int xfrm_policy_delete(struct xfrm_policy *pol, int dir)
        pol = __xfrm_policy_unlink(pol, dir);
        write_unlock_bh(&xfrm_policy_lock);
        if (pol) {
-               if (dir < XFRM_POLICY_MAX)
-                       atomic_inc(&flow_cache_genid);
                xfrm_policy_kill(pol);
                return 0;
        }
@@ -1131,6 +1083,9 @@ int xfrm_sk_policy_insert(struct sock *sk, int dir, struct xfrm_policy *pol)
                __xfrm_policy_link(pol, XFRM_POLICY_MAX+dir);
        }
        if (old_pol)
+               /* Unlinking succeeds always. This is the only function
+                * allowed to delete or replace socket policy.
+                */
                __xfrm_policy_unlink(old_pol, XFRM_POLICY_MAX+dir);
        write_unlock_bh(&xfrm_policy_lock);
 
@@ -1153,6 +1108,7 @@ static struct xfrm_policy *clone_policy(struct xfrm_policy *old, int dir)
                }
                newp->lft = old->lft;
                newp->curlft = old->curlft;
+               newp->mark = old->mark;
                newp->action = old->action;
                newp->flags = old->flags;
                newp->xfrm_nr = old->xfrm_nr;
@@ -1298,18 +1254,6 @@ xfrm_tmpl_resolve(struct xfrm_policy **pols, int npols, struct flowi *fl,
  * still valid.
  */
 
-static struct dst_entry *
-xfrm_find_bundle(struct flowi *fl, struct xfrm_policy *policy, unsigned short family)
-{
-       struct dst_entry *x;
-       struct xfrm_policy_afinfo *afinfo = xfrm_policy_get_afinfo(family);
-       if (unlikely(afinfo == NULL))
-               return ERR_PTR(-EINVAL);
-       x = afinfo->find_bundle(fl, policy);
-       xfrm_policy_put_afinfo(afinfo);
-       return x;
-}
-
 static inline int xfrm_get_tos(struct flowi *fl, int family)
 {
        struct xfrm_policy_afinfo *afinfo = xfrm_policy_get_afinfo(family);
@@ -1325,18 +1269,80 @@ static inline int xfrm_get_tos(struct flowi *fl, int family)
        return tos;
 }
 
-static inline struct xfrm_dst *xfrm_alloc_dst(int family)
+static struct flow_cache_object *xfrm_bundle_flo_get(struct flow_cache_object *flo)
+{
+       struct xfrm_dst *xdst = container_of(flo, struct xfrm_dst, flo);
+       struct dst_entry *dst = &xdst->u.dst;
+
+       if (xdst->route == NULL) {
+               /* Dummy bundle - if it has xfrms we were not
+                * able to build bundle as template resolution failed.
+                * It means we need to try again resolving. */
+               if (xdst->num_xfrms > 0)
+                       return NULL;
+       } else {
+               /* Real bundle */
+               if (stale_bundle(dst))
+                       return NULL;
+       }
+
+       dst_hold(dst);
+       return flo;
+}
+
+static int xfrm_bundle_flo_check(struct flow_cache_object *flo)
+{
+       struct xfrm_dst *xdst = container_of(flo, struct xfrm_dst, flo);
+       struct dst_entry *dst = &xdst->u.dst;
+
+       if (!xdst->route)
+               return 0;
+       if (stale_bundle(dst))
+               return 0;
+
+       return 1;
+}
+
+static void xfrm_bundle_flo_delete(struct flow_cache_object *flo)
+{
+       struct xfrm_dst *xdst = container_of(flo, struct xfrm_dst, flo);
+       struct dst_entry *dst = &xdst->u.dst;
+
+       dst_free(dst);
+}
+
+static const struct flow_cache_ops xfrm_bundle_fc_ops = {
+       .get = xfrm_bundle_flo_get,
+       .check = xfrm_bundle_flo_check,
+       .delete = xfrm_bundle_flo_delete,
+};
+
+static inline struct xfrm_dst *xfrm_alloc_dst(struct net *net, int family)
 {
        struct xfrm_policy_afinfo *afinfo = xfrm_policy_get_afinfo(family);
+       struct dst_ops *dst_ops;
        struct xfrm_dst *xdst;
 
        if (!afinfo)
                return ERR_PTR(-EINVAL);
 
-       xdst = dst_alloc(afinfo->dst_ops) ?: ERR_PTR(-ENOBUFS);
-
+       switch (family) {
+       case AF_INET:
+               dst_ops = &net->xfrm.xfrm4_dst_ops;
+               break;
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+       case AF_INET6:
+               dst_ops = &net->xfrm.xfrm6_dst_ops;
+               break;
+#endif
+       default:
+               BUG();
+       }
+       xdst = dst_alloc(dst_ops) ?: ERR_PTR(-ENOBUFS);
        xfrm_policy_put_afinfo(afinfo);
 
+       xdst->flo.ops = &xfrm_bundle_fc_ops;
+
        return xdst;
 }
 
@@ -1357,7 +1363,8 @@ static inline int xfrm_init_path(struct xfrm_dst *path, struct dst_entry *dst,
        return err;
 }
 
-static inline int xfrm_fill_dst(struct xfrm_dst *xdst, struct net_device *dev)
+static inline int xfrm_fill_dst(struct xfrm_dst *xdst, struct net_device *dev,
+                               struct flowi *fl)
 {
        struct xfrm_policy_afinfo *afinfo =
                xfrm_policy_get_afinfo(xdst->u.dst.ops->family);
@@ -1366,13 +1373,14 @@ static inline int xfrm_fill_dst(struct xfrm_dst *xdst, struct net_device *dev)
        if (!afinfo)
                return -EINVAL;
 
-       err = afinfo->fill_dst(xdst, dev);
+       err = afinfo->fill_dst(xdst, dev, fl);
 
        xfrm_policy_put_afinfo(afinfo);
 
        return err;
 }
 
+
 /* Allocate chain of dst_entry's, attach known xfrm's, calculate
  * all the metrics... Shortly, bundle a bundle.
  */
@@ -1382,6 +1390,7 @@ static struct dst_entry *xfrm_bundle_create(struct xfrm_policy *policy,
                                            struct flowi *fl,
                                            struct dst_entry *dst)
 {
+       struct net *net = xp_net(policy);
        unsigned long now = jiffies;
        struct net_device *dev;
        struct dst_entry *dst_prev = NULL;
@@ -1405,7 +1414,7 @@ static struct dst_entry *xfrm_bundle_create(struct xfrm_policy *policy,
        dst_hold(dst);
 
        for (; i < nx; i++) {
-               struct xfrm_dst *xdst = xfrm_alloc_dst(family);
+               struct xfrm_dst *xdst = xfrm_alloc_dst(net, family);
                struct dst_entry *dst1 = &xdst->u.dst;
 
                err = PTR_ERR(xdst);
@@ -1435,7 +1444,7 @@ static struct dst_entry *xfrm_bundle_create(struct xfrm_policy *policy,
                        dst_hold(dst);
 
                dst1->xfrm = xfrm[i];
-               xdst->genid = xfrm[i]->genid;
+               xdst->xfrm_genid = xfrm[i]->genid;
 
                dst1->obsolete = -1;
                dst1->flags |= DST_HOST;
@@ -1461,7 +1470,7 @@ static struct dst_entry *xfrm_bundle_create(struct xfrm_policy *policy,
        if (!dev)
                goto free_dst;
 
-       /* Copy neighbout for reachability confirmation */
+       /* Copy neighbour for reachability confirmation */
        dst0->neighbour = neigh_clone(dst->neighbour);
 
        xfrm_init_path((struct xfrm_dst *)dst0, dst, nfheader_len);
@@ -1470,7 +1479,7 @@ static struct dst_entry *xfrm_bundle_create(struct xfrm_policy *policy,
        for (dst_prev = dst0; dst_prev != dst; dst_prev = dst_prev->child) {
                struct xfrm_dst *xdst = (struct xfrm_dst *)dst_prev;
 
-               err = xfrm_fill_dst(xdst, dev);
+               err = xfrm_fill_dst(xdst, dev, fl);
                if (err)
                        goto free_dst;
 
@@ -1528,7 +1537,186 @@ xfrm_dst_update_origin(struct dst_entry *dst, struct flowi *fl)
 #endif
 }
 
-static int stale_bundle(struct dst_entry *dst);
+static int xfrm_expand_policies(struct flowi *fl, u16 family,
+                               struct xfrm_policy **pols,
+                               int *num_pols, int *num_xfrms)
+{
+       int i;
+
+       if (*num_pols == 0 || !pols[0]) {
+               *num_pols = 0;
+               *num_xfrms = 0;
+               return 0;
+       }
+       if (IS_ERR(pols[0]))
+               return PTR_ERR(pols[0]);
+
+       *num_xfrms = pols[0]->xfrm_nr;
+
+#ifdef CONFIG_XFRM_SUB_POLICY
+       if (pols[0] && pols[0]->action == XFRM_POLICY_ALLOW &&
+           pols[0]->type != XFRM_POLICY_TYPE_MAIN) {
+               pols[1] = xfrm_policy_lookup_bytype(xp_net(pols[0]),
+                                                   XFRM_POLICY_TYPE_MAIN,
+                                                   fl, family,
+                                                   XFRM_POLICY_OUT);
+               if (pols[1]) {
+                       if (IS_ERR(pols[1])) {
+                               xfrm_pols_put(pols, *num_pols);
+                               return PTR_ERR(pols[1]);
+                       }
+                       (*num_pols) ++;
+                       (*num_xfrms) += pols[1]->xfrm_nr;
+               }
+       }
+#endif
+       for (i = 0; i < *num_pols; i++) {
+               if (pols[i]->action != XFRM_POLICY_ALLOW) {
+                       *num_xfrms = -1;
+                       break;
+               }
+       }
+
+       return 0;
+
+}
+
+static struct xfrm_dst *
+xfrm_resolve_and_create_bundle(struct xfrm_policy **pols, int num_pols,
+                              struct flowi *fl, u16 family,
+                              struct dst_entry *dst_orig)
+{
+       struct net *net = xp_net(pols[0]);
+       struct xfrm_state *xfrm[XFRM_MAX_DEPTH];
+       struct dst_entry *dst;
+       struct xfrm_dst *xdst;
+       int err;
+
+       /* Try to instantiate a bundle */
+       err = xfrm_tmpl_resolve(pols, num_pols, fl, xfrm, family);
+       if (err < 0) {
+               if (err != -EAGAIN)
+                       XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTPOLERROR);
+               return ERR_PTR(err);
+       }
+
+       dst = xfrm_bundle_create(pols[0], xfrm, err, fl, dst_orig);
+       if (IS_ERR(dst)) {
+               XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTBUNDLEGENERROR);
+               return ERR_CAST(dst);
+       }
+
+       xdst = (struct xfrm_dst *)dst;
+       xdst->num_xfrms = err;
+       if (num_pols > 1)
+               err = xfrm_dst_update_parent(dst, &pols[1]->selector);
+       else
+               err = xfrm_dst_update_origin(dst, fl);
+       if (unlikely(err)) {
+               dst_free(dst);
+               XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTBUNDLECHECKERROR);
+               return ERR_PTR(err);
+       }
+
+       xdst->num_pols = num_pols;
+       memcpy(xdst->pols, pols, sizeof(struct xfrm_policy*) * num_pols);
+       xdst->policy_genid = atomic_read(&pols[0]->genid);
+
+       return xdst;
+}
+
+static struct flow_cache_object *
+xfrm_bundle_lookup(struct net *net, struct flowi *fl, u16 family, u8 dir,
+                  struct flow_cache_object *oldflo, void *ctx)
+{
+       struct dst_entry *dst_orig = (struct dst_entry *)ctx;
+       struct xfrm_policy *pols[XFRM_POLICY_TYPE_MAX];
+       struct xfrm_dst *xdst, *new_xdst;
+       int num_pols = 0, num_xfrms = 0, i, err, pol_dead;
+
+       /* Check if the policies from old bundle are usable */
+       xdst = NULL;
+       if (oldflo) {
+               xdst = container_of(oldflo, struct xfrm_dst, flo);
+               num_pols = xdst->num_pols;
+               num_xfrms = xdst->num_xfrms;
+               pol_dead = 0;
+               for (i = 0; i < num_pols; i++) {
+                       pols[i] = xdst->pols[i];
+                       pol_dead |= pols[i]->walk.dead;
+               }
+               if (pol_dead) {
+                       dst_free(&xdst->u.dst);
+                       xdst = NULL;
+                       num_pols = 0;
+                       num_xfrms = 0;
+                       oldflo = NULL;
+               }
+       }
+
+       /* Resolve policies to use if we couldn't get them from
+        * previous cache entry */
+       if (xdst == NULL) {
+               num_pols = 1;
+               pols[0] = __xfrm_policy_lookup(net, fl, family, dir);
+               err = xfrm_expand_policies(fl, family, pols,
+                                          &num_pols, &num_xfrms);
+               if (err < 0)
+                       goto inc_error;
+               if (num_pols == 0)
+                       return NULL;
+               if (num_xfrms <= 0)
+                       goto make_dummy_bundle;
+       }
+
+       new_xdst = xfrm_resolve_and_create_bundle(pols, num_pols, fl, family, dst_orig);
+       if (IS_ERR(new_xdst)) {
+               err = PTR_ERR(new_xdst);
+               if (err != -EAGAIN)
+                       goto error;
+               if (oldflo == NULL)
+                       goto make_dummy_bundle;
+               dst_hold(&xdst->u.dst);
+               return oldflo;
+       }
+
+       /* Kill the previous bundle */
+       if (xdst) {
+               /* The policies were stolen for newly generated bundle */
+               xdst->num_pols = 0;
+               dst_free(&xdst->u.dst);
+       }
+
+       /* Flow cache does not have reference, it dst_free()'s,
+        * but we do need to return one reference for original caller */
+       dst_hold(&new_xdst->u.dst);
+       return &new_xdst->flo;
+
+make_dummy_bundle:
+       /* We found policies, but there's no bundles to instantiate:
+        * either because the policy blocks, has no transformations or
+        * we could not build template (no xfrm_states).*/
+       xdst = xfrm_alloc_dst(net, family);
+       if (IS_ERR(xdst)) {
+               xfrm_pols_put(pols, num_pols);
+               return ERR_CAST(xdst);
+       }
+       xdst->num_pols = num_pols;
+       xdst->num_xfrms = num_xfrms;
+       memcpy(xdst->pols, pols, sizeof(struct xfrm_policy*) * num_pols);
+
+       dst_hold(&xdst->u.dst);
+       return &xdst->flo;
+
+inc_error:
+       XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTPOLERROR);
+error:
+       if (xdst != NULL)
+               dst_free(&xdst->u.dst);
+       else
+               xfrm_pols_put(pols, num_pols);
+       return ERR_PTR(err);
+}
 
 /* Main function: finds/creates a bundle for given flow.
  *
@@ -1538,245 +1726,152 @@ static int stale_bundle(struct dst_entry *dst);
 int __xfrm_lookup(struct net *net, struct dst_entry **dst_p, struct flowi *fl,
                  struct sock *sk, int flags)
 {
-       struct xfrm_policy *policy;
        struct xfrm_policy *pols[XFRM_POLICY_TYPE_MAX];
-       int npols;
-       int pol_dead;
-       int xfrm_nr;
-       int pi;
-       struct xfrm_state *xfrm[XFRM_MAX_DEPTH];
-       struct dst_entry *dst, *dst_orig = *dst_p;
-       int nx = 0;
-       int err;
-       u32 genid;
-       u16 family;
+       struct flow_cache_object *flo;
+       struct xfrm_dst *xdst;
+       struct dst_entry *dst, *dst_orig = *dst_p, *route;
+       u16 family = dst_orig->ops->family;
        u8 dir = policy_to_flow_dir(XFRM_POLICY_OUT);
+       int i, err, num_pols, num_xfrms = 0, drop_pols = 0;
 
 restart:
-       genid = atomic_read(&flow_cache_genid);
-       policy = NULL;
-       for (pi = 0; pi < ARRAY_SIZE(pols); pi++)
-               pols[pi] = NULL;
-       npols = 0;
-       pol_dead = 0;
-       xfrm_nr = 0;
+       dst = NULL;
+       xdst = NULL;
+       route = NULL;
 
        if (sk && sk->sk_policy[XFRM_POLICY_OUT]) {
-               policy = xfrm_sk_policy_lookup(sk, XFRM_POLICY_OUT, fl);
-               err = PTR_ERR(policy);
-               if (IS_ERR(policy)) {
-                       XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTPOLERROR);
+               num_pols = 1;
+               pols[0] = xfrm_sk_policy_lookup(sk, XFRM_POLICY_OUT, fl);
+               err = xfrm_expand_policies(fl, family, pols,
+                                          &num_pols, &num_xfrms);
+               if (err < 0)
                        goto dropdst;
+
+               if (num_pols) {
+                       if (num_xfrms <= 0) {
+                               drop_pols = num_pols;
+                               goto no_transform;
+                       }
+
+                       xdst = xfrm_resolve_and_create_bundle(
+                                       pols, num_pols, fl,
+                                       family, dst_orig);
+                       if (IS_ERR(xdst)) {
+                               xfrm_pols_put(pols, num_pols);
+                               err = PTR_ERR(xdst);
+                               goto dropdst;
+                       }
+
+                       spin_lock_bh(&xfrm_policy_sk_bundle_lock);
+                       xdst->u.dst.next = xfrm_policy_sk_bundles;
+                       xfrm_policy_sk_bundles = &xdst->u.dst;
+                       spin_unlock_bh(&xfrm_policy_sk_bundle_lock);
+
+                       route = xdst->route;
                }
        }
 
-       if (!policy) {
+       if (xdst == NULL) {
                /* To accelerate a bit...  */
                if ((dst_orig->flags & DST_NOXFRM) ||
                    !net->xfrm.policy_count[XFRM_POLICY_OUT])
                        goto nopol;
 
-               policy = flow_cache_lookup(net, fl, dst_orig->ops->family,
-                                          dir, xfrm_policy_lookup);
-               err = PTR_ERR(policy);
-               if (IS_ERR(policy)) {
-                       XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTPOLERROR);
+               flo = flow_cache_lookup(net, fl, family, dir,
+                                       xfrm_bundle_lookup, dst_orig);
+               if (flo == NULL)
+                       goto nopol;
+               if (IS_ERR(flo)) {
+                       err = PTR_ERR(flo);
                        goto dropdst;
                }
+               xdst = container_of(flo, struct xfrm_dst, flo);
+
+               num_pols = xdst->num_pols;
+               num_xfrms = xdst->num_xfrms;
+               memcpy(pols, xdst->pols, sizeof(struct xfrm_policy*) * num_pols);
+               route = xdst->route;
+       }
+
+       dst = &xdst->u.dst;
+       if (route == NULL && num_xfrms > 0) {
+               /* The only case when xfrm_bundle_lookup() returns a
+                * bundle with null route, is when the template could
+                * not be resolved. It means policies are there, but
+                * bundle could not be created, since we don't yet
+                * have the xfrm_state's. We need to wait for KM to
+                * negotiate new SA's or bail out with error.*/
+               if (net->xfrm.sysctl_larval_drop) {
+                       /* EREMOTE tells the caller to generate
+                        * a one-shot blackhole route. */
+                       dst_release(dst);
+                       xfrm_pols_put(pols, drop_pols);
+                       XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTNOSTATES);
+                       return -EREMOTE;
+               }
+               if (flags & XFRM_LOOKUP_WAIT) {
+                       DECLARE_WAITQUEUE(wait, current);
+
+                       add_wait_queue(&net->xfrm.km_waitq, &wait);
+                       set_current_state(TASK_INTERRUPTIBLE);
+                       schedule();
+                       set_current_state(TASK_RUNNING);
+                       remove_wait_queue(&net->xfrm.km_waitq, &wait);
+
+                       if (!signal_pending(current)) {
+                               dst_release(dst);
+                               goto restart;
+                       }
+
+                       err = -ERESTART;
+               } else
+                       err = -EAGAIN;
+
+               XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTNOSTATES);
+               goto error;
        }
 
-       if (!policy)
+no_transform:
+       if (num_pols == 0)
                goto nopol;
 
-       family = dst_orig->ops->family;
-       pols[0] = policy;
-       npols ++;
-       xfrm_nr += pols[0]->xfrm_nr;
-
-       err = -ENOENT;
-       if ((flags & XFRM_LOOKUP_ICMP) && !(policy->flags & XFRM_POLICY_ICMP))
+       if ((flags & XFRM_LOOKUP_ICMP) &&
+           !(pols[0]->flags & XFRM_POLICY_ICMP)) {
+               err = -ENOENT;
                goto error;
+       }
 
-       policy->curlft.use_time = get_seconds();
+       for (i = 0; i < num_pols; i++)
+               pols[i]->curlft.use_time = get_seconds();
 
-       switch (policy->action) {
-       default:
-       case XFRM_POLICY_BLOCK:
+       if (num_xfrms < 0) {
                /* Prohibit the flow */
                XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTPOLBLOCK);
                err = -EPERM;
                goto error;
-
-       case XFRM_POLICY_ALLOW:
-#ifndef CONFIG_XFRM_SUB_POLICY
-               if (policy->xfrm_nr == 0) {
-                       /* Flow passes not transformed. */
-                       xfrm_pol_put(policy);
-                       return 0;
-               }
-#endif
-
-               /* Try to find matching bundle.
-                *
-                * LATER: help from flow cache. It is optional, this
-                * is required only for output policy.
-                */
-               dst = xfrm_find_bundle(fl, policy, family);
-               if (IS_ERR(dst)) {
-                       XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTBUNDLECHECKERROR);
-                       err = PTR_ERR(dst);
-                       goto error;
-               }
-
-               if (dst)
-                       break;
-
-#ifdef CONFIG_XFRM_SUB_POLICY
-               if (pols[0]->type != XFRM_POLICY_TYPE_MAIN) {
-                       pols[1] = xfrm_policy_lookup_bytype(net,
-                                                           XFRM_POLICY_TYPE_MAIN,
-                                                           fl, family,
-                                                           XFRM_POLICY_OUT);
-                       if (pols[1]) {
-                               if (IS_ERR(pols[1])) {
-                                       XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTPOLERROR);
-                                       err = PTR_ERR(pols[1]);
-                                       goto error;
-                               }
-                               if (pols[1]->action == XFRM_POLICY_BLOCK) {
-                                       XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTPOLBLOCK);
-                                       err = -EPERM;
-                                       goto error;
-                               }
-                               npols ++;
-                               xfrm_nr += pols[1]->xfrm_nr;
-                       }
-               }
-
-               /*
-                * Because neither flowi nor bundle information knows about
-                * transformation template size. On more than one policy usage
-                * we can realize whether all of them is bypass or not after
-                * they are searched. See above not-transformed bypass
-                * is surrounded by non-sub policy configuration, too.
-                */
-               if (xfrm_nr == 0) {
-                       /* Flow passes not transformed. */
-                       xfrm_pols_put(pols, npols);
-                       return 0;
-               }
-
-#endif
-               nx = xfrm_tmpl_resolve(pols, npols, fl, xfrm, family);
-
-               if (unlikely(nx<0)) {
-                       err = nx;
-                       if (err == -EAGAIN && sysctl_xfrm_larval_drop) {
-                               /* EREMOTE tells the caller to generate
-                                * a one-shot blackhole route.
-                                */
-                               XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTNOSTATES);
-                               xfrm_pol_put(policy);
-                               return -EREMOTE;
-                       }
-                       if (err == -EAGAIN && (flags & XFRM_LOOKUP_WAIT)) {
-                               DECLARE_WAITQUEUE(wait, current);
-
-                               add_wait_queue(&net->xfrm.km_waitq, &wait);
-                               set_current_state(TASK_INTERRUPTIBLE);
-                               schedule();
-                               set_current_state(TASK_RUNNING);
-                               remove_wait_queue(&net->xfrm.km_waitq, &wait);
-
-                               nx = xfrm_tmpl_resolve(pols, npols, fl, xfrm, family);
-
-                               if (nx == -EAGAIN && signal_pending(current)) {
-                                       XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTNOSTATES);
-                                       err = -ERESTART;
-                                       goto error;
-                               }
-                               if (nx == -EAGAIN ||
-                                   genid != atomic_read(&flow_cache_genid)) {
-                                       xfrm_pols_put(pols, npols);
-                                       goto restart;
-                               }
-                               err = nx;
-                       }
-                       if (err < 0) {
-                               XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTNOSTATES);
-                               goto error;
-                       }
-               }
-               if (nx == 0) {
-                       /* Flow passes not transformed. */
-                       xfrm_pols_put(pols, npols);
-                       return 0;
-               }
-
-               dst = xfrm_bundle_create(policy, xfrm, nx, fl, dst_orig);
-               err = PTR_ERR(dst);
-               if (IS_ERR(dst)) {
-                       XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTBUNDLEGENERROR);
-                       goto error;
-               }
-
-               for (pi = 0; pi < npols; pi++) {
-                       read_lock_bh(&pols[pi]->lock);
-                       pol_dead |= pols[pi]->walk.dead;
-                       read_unlock_bh(&pols[pi]->lock);
-               }
-
-               write_lock_bh(&policy->lock);
-               if (unlikely(pol_dead || stale_bundle(dst))) {
-                       /* Wow! While we worked on resolving, this
-                        * policy has gone. Retry. It is not paranoia,
-                        * we just cannot enlist new bundle to dead object.
-                        * We can't enlist stable bundles either.
-                        */
-                       write_unlock_bh(&policy->lock);
-                       dst_free(dst);
-
-                       if (pol_dead)
-                               XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTPOLDEAD);
-                       else
-                               XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTBUNDLECHECKERROR);
-                       err = -EHOSTUNREACH;
-                       goto error;
-               }
-
-               if (npols > 1)
-                       err = xfrm_dst_update_parent(dst, &pols[1]->selector);
-               else
-                       err = xfrm_dst_update_origin(dst, fl);
-               if (unlikely(err)) {
-                       write_unlock_bh(&policy->lock);
-                       dst_free(dst);
-                       XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTBUNDLECHECKERROR);
-                       goto error;
-               }
-
-               dst->next = policy->bundles;
-               policy->bundles = dst;
-               dst_hold(dst);
-               write_unlock_bh(&policy->lock);
+       } else if (num_xfrms > 0) {
+               /* Flow transformed */
+               *dst_p = dst;
+               dst_release(dst_orig);
+       } else {
+               /* Flow passes untransformed */
+               dst_release(dst);
        }
-       *dst_p = dst;
-       dst_release(dst_orig);
-       xfrm_pols_put(pols, npols);
+ok:
+       xfrm_pols_put(pols, drop_pols);
        return 0;
 
+nopol:
+       if (!(flags & XFRM_LOOKUP_ICMP))
+               goto ok;
+       err = -ENOENT;
 error:
-       xfrm_pols_put(pols, npols);
+       dst_release(dst);
 dropdst:
        dst_release(dst_orig);
        *dst_p = NULL;
+       xfrm_pols_put(pols, drop_pols);
        return err;
-
-nopol:
-       err = -ENOENT;
-       if (flags & XFRM_LOOKUP_ICMP)
-               goto dropdst;
-       return 0;
 }
 EXPORT_SYMBOL(__xfrm_lookup);
 
@@ -1935,9 +2030,16 @@ int __xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb,
                }
        }
 
-       if (!pol)
-               pol = flow_cache_lookup(net, &fl, family, fl_dir,
-                                       xfrm_policy_lookup);
+       if (!pol) {
+               struct flow_cache_object *flo;
+
+               flo = flow_cache_lookup(net, &fl, family, fl_dir,
+                                       xfrm_policy_lookup, NULL);
+               if (IS_ERR_OR_NULL(flo))
+                       pol = ERR_CAST(flo);
+               else
+                       pol = container_of(flo, struct xfrm_policy, flo);
+       }
 
        if (IS_ERR(pol)) {
                XFRM_INC_STATS(net, LINUX_MIB_XFRMINPOLERROR);
@@ -2043,14 +2145,20 @@ int __xfrm_route_forward(struct sk_buff *skb, unsigned short family)
 {
        struct net *net = dev_net(skb->dev);
        struct flowi fl;
+       struct dst_entry *dst;
+       int res;
 
        if (xfrm_decode_session(skb, &fl, family) < 0) {
-               /* XXX: we should have something like FWDHDRERROR here. */
-               XFRM_INC_STATS(net, LINUX_MIB_XFRMINHDRERROR);
+               XFRM_INC_STATS(net, LINUX_MIB_XFRMFWDHDRERROR);
                return 0;
        }
 
-       return xfrm_lookup(net, &skb->dst, &fl, NULL, 0) == 0;
+       skb_dst_force(skb);
+       dst = skb_dst(skb);
+
+       res = xfrm_lookup(net, &dst, &fl, NULL, 0) == 0;
+       skb_dst_set(skb, dst);
+       return res;
 }
 EXPORT_SYMBOL(__xfrm_route_forward);
 
@@ -2102,7 +2210,6 @@ EXPORT_SYMBOL(xfrm_dst_ifdown);
 static void xfrm_link_failure(struct sk_buff *skb)
 {
        /* Impossible. Such dst must be popped before reaches point of failure. */
-       return;
 }
 
 static struct dst_entry *xfrm_negative_advice(struct dst_entry *dst)
@@ -2116,69 +2223,22 @@ static struct dst_entry *xfrm_negative_advice(struct dst_entry *dst)
        return dst;
 }
 
-static void prune_one_bundle(struct xfrm_policy *pol, int (*func)(struct dst_entry *), struct dst_entry **gc_list_p)
+static void __xfrm_garbage_collect(struct net *net)
 {
-       struct dst_entry *dst, **dstp;
+       struct dst_entry *head, *next;
 
-       write_lock(&pol->lock);
-       dstp = &pol->bundles;
-       while ((dst=*dstp) != NULL) {
-               if (func(dst)) {
-                       *dstp = dst->next;
-                       dst->next = *gc_list_p;
-                       *gc_list_p = dst;
-               } else {
-                       dstp = &dst->next;
-               }
-       }
-       write_unlock(&pol->lock);
-}
+       flow_cache_flush();
 
-static void xfrm_prune_bundles(struct net *net, int (*func)(struct dst_entry *))
-{
-       struct dst_entry *gc_list = NULL;
-       int dir;
+       spin_lock_bh(&xfrm_policy_sk_bundle_lock);
+       head = xfrm_policy_sk_bundles;
+       xfrm_policy_sk_bundles = NULL;
+       spin_unlock_bh(&xfrm_policy_sk_bundle_lock);
 
-       read_lock_bh(&xfrm_policy_lock);
-       for (dir = 0; dir < XFRM_POLICY_MAX * 2; dir++) {
-               struct xfrm_policy *pol;
-               struct hlist_node *entry;
-               struct hlist_head *table;
-               int i;
-
-               hlist_for_each_entry(pol, entry,
-                                    &net->xfrm.policy_inexact[dir], bydst)
-                       prune_one_bundle(pol, func, &gc_list);
-
-               table = net->xfrm.policy_bydst[dir].table;
-               for (i = net->xfrm.policy_bydst[dir].hmask; i >= 0; i--) {
-                       hlist_for_each_entry(pol, entry, table + i, bydst)
-                               prune_one_bundle(pol, func, &gc_list);
-               }
+       while (head) {
+               next = head->next;
+               dst_free(head);
+               head = next;
        }
-       read_unlock_bh(&xfrm_policy_lock);
-
-       while (gc_list) {
-               struct dst_entry *dst = gc_list;
-               gc_list = dst->next;
-               dst_free(dst);
-       }
-}
-
-static int unused_bundle(struct dst_entry *dst)
-{
-       return !atomic_read(&dst->__refcnt);
-}
-
-static void __xfrm_garbage_collect(struct net *net)
-{
-       xfrm_prune_bundles(net, unused_bundle);
-}
-
-static int xfrm_flush_bundles(struct net *net)
-{
-       xfrm_prune_bundles(net, stale_bundle);
-       return 0;
 }
 
 static void xfrm_init_pmtu(struct dst_entry *dst)
@@ -2238,7 +2298,9 @@ int xfrm_bundle_ok(struct xfrm_policy *pol, struct xfrm_dst *first,
                        return 0;
                if (dst->xfrm->km.state != XFRM_STATE_VALID)
                        return 0;
-               if (xdst->genid != dst->xfrm->genid)
+               if (xdst->xfrm_genid != dst->xfrm->genid)
+                       return 0;
+               if (xdst->policy_genid != atomic_read(&xdst->pols[0]->genid))
                        return 0;
 
                if (strict && fl &&
@@ -2289,6 +2351,7 @@ EXPORT_SYMBOL(xfrm_bundle_ok);
 
 int xfrm_policy_register_afinfo(struct xfrm_policy_afinfo *afinfo)
 {
+       struct net *net;
        int err = 0;
        if (unlikely(afinfo == NULL))
                return -EINVAL;
@@ -2312,6 +2375,27 @@ int xfrm_policy_register_afinfo(struct xfrm_policy_afinfo *afinfo)
                xfrm_policy_afinfo[afinfo->family] = afinfo;
        }
        write_unlock_bh(&xfrm_policy_afinfo_lock);
+
+       rtnl_lock();
+       for_each_net(net) {
+               struct dst_ops *xfrm_dst_ops;
+
+               switch (afinfo->family) {
+               case AF_INET:
+                       xfrm_dst_ops = &net->xfrm.xfrm4_dst_ops;
+                       break;
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+               case AF_INET6:
+                       xfrm_dst_ops = &net->xfrm.xfrm6_dst_ops;
+                       break;
+#endif
+               default:
+                       BUG();
+               }
+               *xfrm_dst_ops = *afinfo->dst_ops;
+       }
+       rtnl_unlock();
+
        return err;
 }
 EXPORT_SYMBOL(xfrm_policy_register_afinfo);
@@ -2342,6 +2426,22 @@ int xfrm_policy_unregister_afinfo(struct xfrm_policy_afinfo *afinfo)
 }
 EXPORT_SYMBOL(xfrm_policy_unregister_afinfo);
 
+static void __net_init xfrm_dst_ops_init(struct net *net)
+{
+       struct xfrm_policy_afinfo *afinfo;
+
+       read_lock_bh(&xfrm_policy_afinfo_lock);
+       afinfo = xfrm_policy_afinfo[AF_INET];
+       if (afinfo)
+               net->xfrm.xfrm4_dst_ops = *afinfo->dst_ops;
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+       afinfo = xfrm_policy_afinfo[AF_INET6];
+       if (afinfo)
+               net->xfrm.xfrm6_dst_ops = *afinfo->dst_ops;
+#endif
+       read_unlock_bh(&xfrm_policy_afinfo_lock);
+}
+
 static struct xfrm_policy_afinfo *xfrm_policy_get_afinfo(unsigned short family)
 {
        struct xfrm_policy_afinfo *afinfo;
@@ -2365,7 +2465,7 @@ static int xfrm_dev_event(struct notifier_block *this, unsigned long event, void
 
        switch (event) {
        case NETDEV_DOWN:
-               xfrm_flush_bundles(dev_net(dev));
+               __xfrm_garbage_collect(dev_net(dev));
        }
        return NOTIFY_DONE;
 }
@@ -2379,19 +2479,19 @@ static int __net_init xfrm_statistics_init(struct net *net)
 {
        int rv;
 
-       if (snmp_mib_init((void **)net->mib.xfrm_statistics,
+       if (snmp_mib_init((void __percpu **)net->mib.xfrm_statistics,
                          sizeof(struct linux_xfrm_mib)) < 0)
                return -ENOMEM;
        rv = xfrm_proc_init(net);
        if (rv < 0)
-               snmp_mib_free((void **)net->mib.xfrm_statistics);
+               snmp_mib_free((void __percpu **)net->mib.xfrm_statistics);
        return rv;
 }
 
 static void xfrm_statistics_fini(struct net *net)
 {
        xfrm_proc_fini(net);
-       snmp_mib_free((void **)net->mib.xfrm_statistics);
+       snmp_mib_free((void __percpu **)net->mib.xfrm_statistics);
 }
 #else
 static int __net_init xfrm_statistics_init(struct net *net)
@@ -2471,7 +2571,6 @@ static void xfrm_policy_fini(struct net *net)
        audit_info.sessionid = -1;
        audit_info.secid = 0;
        xfrm_policy_flush(net, XFRM_POLICY_TYPE_MAIN, &audit_info);
-       flush_work(&xfrm_policy_gc_work);
 
        WARN_ON(!list_empty(&net->xfrm.policy_all));
 
@@ -2504,8 +2603,14 @@ static int __net_init xfrm_net_init(struct net *net)
        rv = xfrm_policy_init(net);
        if (rv < 0)
                goto out_policy;
+       xfrm_dst_ops_init(net);
+       rv = xfrm_sysctl_init(net);
+       if (rv < 0)
+               goto out_sysctl;
        return 0;
 
+out_sysctl:
+       xfrm_policy_fini(net);
 out_policy:
        xfrm_state_fini(net);
 out_state:
@@ -2516,6 +2621,7 @@ out_statistics:
 
 static void __net_exit xfrm_net_exit(struct net *net)
 {
+       xfrm_sysctl_fini(net);
        xfrm_policy_fini(net);
        xfrm_state_fini(net);
        xfrm_statistics_fini(net);
@@ -2690,7 +2796,6 @@ static int xfrm_policy_migrate(struct xfrm_policy *pol,
                               struct xfrm_migrate *m, int num_migrate)
 {
        struct xfrm_migrate *mp;
-       struct dst_entry *dst;
        int i, j, n = 0;
 
        write_lock_bh(&pol->lock);
@@ -2715,10 +2820,7 @@ static int xfrm_policy_migrate(struct xfrm_policy *pol,
                               sizeof(pol->xfrm_vec[i].saddr));
                        pol->xfrm_vec[i].encap_family = mp->new_family;
                        /* flush bundles */
-                       while ((dst = pol->bundles) != NULL) {
-                               pol->bundles = dst->next;
-                               dst_free(dst);
-                       }
+                       atomic_inc(&pol->genid);
                }
        }