1 files changed, 70 insertions, 78 deletions

diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 1d5c7bf2993..fdde51f4271 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -39,7 +39,10 @@
 #define XFRM_QUEUE_TMO_MAX ((unsigned)(60*HZ))
 #define XFRM_MAX_QUEUE_LEN	100
 
-static struct dst_entry *xfrm_policy_sk_bundles;
+struct xfrm_flo {
+	struct dst_entry *dst_orig;
+	u8 flags;
+};
 
 static DEFINE_SPINLOCK(xfrm_policy_afinfo_lock);
 static struct xfrm_policy_afinfo __rcu *xfrm_policy_afinfo[NPROTO]
@@ -391,7 +394,7 @@ redo:
 			if (h != h0)
 				continue;
 			hlist_del(&pol->bydst);
-			hlist_add_after(entry0, &pol->bydst);
+			hlist_add_behind(&pol->bydst, entry0);
 		}
 		entry0 = &pol->bydst;
 	}
@@ -656,12 +659,12 @@ int xfrm_policy_insert(int dir, struct xfrm_policy *policy, int excl)
 			break;
 	}
 	if (newpos)
-		hlist_add_after(newpos, &policy->bydst);
+		hlist_add_behind(&policy->bydst, newpos);
 	else
 		hlist_add_head(&policy->bydst, chain);
 	xfrm_pol_hold(policy);
 	net->xfrm.policy_count[dir]++;
-	atomic_inc(&flow_cache_genid);
+	atomic_inc(&net->xfrm.flow_cache_genid);
 
 	/* After previous checking, family can either be AF_INET or AF_INET6 */
 	if (policy->family == AF_INET)
@@ -771,7 +774,7 @@ EXPORT_SYMBOL(xfrm_policy_byid);
 
 #ifdef CONFIG_SECURITY_NETWORK_XFRM
 static inline int
-xfrm_policy_flush_secctx_check(struct net *net, u8 type, struct xfrm_audit *audit_info)
+xfrm_policy_flush_secctx_check(struct net *net, u8 type, bool task_valid)
 {
 	int dir, err = 0;
 
@@ -785,10 +788,7 @@ xfrm_policy_flush_secctx_check(struct net *net, u8 type, struct xfrm_audit *audi
 				continue;
 			err = security_xfrm_policy_delete(pol->security);
 			if (err) {
-				xfrm_audit_policy_delete(pol, 0,
-							 audit_info->loginuid,
-							 audit_info->sessionid,
-							 audit_info->secid);
+				xfrm_audit_policy_delete(pol, 0, task_valid);
 				return err;
 			}
 		}
@@ -802,9 +802,7 @@ xfrm_policy_flush_secctx_check(struct net *net, u8 type, struct xfrm_audit *audi
 								pol->security);
 				if (err) {
 					xfrm_audit_policy_delete(pol, 0,
-							audit_info->loginuid,
-							audit_info->sessionid,
-							audit_info->secid);
+								 task_valid);
 					return err;
 				}
 			}
@@ -814,19 +812,19 @@ xfrm_policy_flush_secctx_check(struct net *net, u8 type, struct xfrm_audit *audi
 }
 #else
 static inline int
-xfrm_policy_flush_secctx_check(struct net *net, u8 type, struct xfrm_audit *audit_info)
+xfrm_policy_flush_secctx_check(struct net *net, u8 type, bool task_valid)
 {
 	return 0;
 }
 #endif
 
-int xfrm_policy_flush(struct net *net, u8 type, struct xfrm_audit *audit_info)
+int xfrm_policy_flush(struct net *net, u8 type, bool task_valid)
 {
 	int dir, err = 0, cnt = 0;
 
 	write_lock_bh(&net->xfrm.xfrm_policy_lock);
 
-	err = xfrm_policy_flush_secctx_check(net, type, audit_info);
+	err = xfrm_policy_flush_secctx_check(net, type, task_valid);
 	if (err)
 		goto out;
 
@@ -843,9 +841,7 @@ int xfrm_policy_flush(struct net *net, u8 type, struct xfrm_audit *audit_info)
 			write_unlock_bh(&net->xfrm.xfrm_policy_lock);
 			cnt++;
 
-			xfrm_audit_policy_delete(pol, 1, audit_info->loginuid,
-						 audit_info->sessionid,
-						 audit_info->secid);
+			xfrm_audit_policy_delete(pol, 1, task_valid);
 
 			xfrm_policy_kill(pol);
 
@@ -864,10 +860,7 @@ int xfrm_policy_flush(struct net *net, u8 type, struct xfrm_audit *audit_info)
 				write_unlock_bh(&net->xfrm.xfrm_policy_lock);
 				cnt++;
 
-				xfrm_audit_policy_delete(pol, 1,
-							 audit_info->loginuid,
-							 audit_info->sessionid,
-							 audit_info->secid);
+				xfrm_audit_policy_delete(pol, 1, task_valid);
 				xfrm_policy_kill(pol);
 
 				write_lock_bh(&net->xfrm.xfrm_policy_lock);
@@ -1844,7 +1837,7 @@ purge_queue:
 	xfrm_pol_put(pol);
 }
 
-static int xdst_queue_output(struct sk_buff *skb)
+static int xdst_queue_output(struct sock *sk, struct sk_buff *skb)
 {
 	unsigned long sched_next;
 	struct dst_entry *dst = skb_dst(skb);
@@ -1889,13 +1882,14 @@ static int xdst_queue_output(struct sk_buff *skb)
 }
 
 static struct xfrm_dst *xfrm_create_dummy_bundle(struct net *net,
-						 struct dst_entry *dst,
+						 struct xfrm_flo *xflo,
 						 const struct flowi *fl,
 						 int num_xfrms,
 						 u16 family)
 {
 	int err;
 	struct net_device *dev;
+	struct dst_entry *dst;
 	struct dst_entry *dst1;
 	struct xfrm_dst *xdst;
 
@@ -1903,9 +1897,12 @@ static struct xfrm_dst *xfrm_create_dummy_bundle(struct net *net,
 	if (IS_ERR(xdst))
 		return xdst;
 
-	if (net->xfrm.sysctl_larval_drop || num_xfrms <= 0)
+	if (!(xflo->flags & XFRM_LOOKUP_QUEUE) ||
+	    net->xfrm.sysctl_larval_drop ||
+	    num_xfrms <= 0)
 		return xdst;
 
+	dst = xflo->dst_orig;
 	dst1 = &xdst->u.dst;
 	dst_hold(dst);
 	xdst->route = dst;
@@ -1947,7 +1944,7 @@ static struct flow_cache_object *
 xfrm_bundle_lookup(struct net *net, const 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_flo *xflo = (struct xfrm_flo *)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;
@@ -1988,7 +1985,8 @@ xfrm_bundle_lookup(struct net *net, const struct flowi *fl, u16 family, u8 dir,
 			goto make_dummy_bundle;
 	}
 
-	new_xdst = xfrm_resolve_and_create_bundle(pols, num_pols, fl, family, dst_orig);
+	new_xdst = xfrm_resolve_and_create_bundle(pols, num_pols, fl, family,
+						  xflo->dst_orig);
 	if (IS_ERR(new_xdst)) {
 		err = PTR_ERR(new_xdst);
 		if (err != -EAGAIN)
@@ -2022,7 +2020,7 @@ 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_create_dummy_bundle(net, dst_orig, fl, num_xfrms, family);
+	xdst = xfrm_create_dummy_bundle(net, xflo, fl, num_xfrms, family);
 	if (IS_ERR(xdst)) {
 		xfrm_pols_put(pols, num_pols);
 		return ERR_CAST(xdst);
@@ -2110,24 +2108,24 @@ struct dst_entry *xfrm_lookup(struct net *net, struct dst_entry *dst_orig,
 			}
 
 			dst_hold(&xdst->u.dst);
-
-			spin_lock_bh(&net->xfrm.xfrm_policy_sk_bundle_lock);
-			xdst->u.dst.next = xfrm_policy_sk_bundles;
-			xfrm_policy_sk_bundles = &xdst->u.dst;
-			spin_unlock_bh(&net->xfrm.xfrm_policy_sk_bundle_lock);
-
+			xdst->u.dst.flags |= DST_NOCACHE;
 			route = xdst->route;
 		}
 	}
 
 	if (xdst == NULL) {
+		struct xfrm_flo xflo;
+
+		xflo.dst_orig = dst_orig;
+		xflo.flags = flags;
+
 		/* To accelerate a bit...  */
 		if ((dst_orig->flags & DST_NOXFRM) ||
 		    !net->xfrm.policy_count[XFRM_POLICY_OUT])
 			goto nopol;
 
 		flo = flow_cache_lookup(net, fl, family, dir,
-					xfrm_bundle_lookup, dst_orig);
+					xfrm_bundle_lookup, &xflo);
 		if (flo == NULL)
 			goto nopol;
 		if (IS_ERR(flo)) {
@@ -2155,7 +2153,7 @@ struct dst_entry *xfrm_lookup(struct net *net, struct dst_entry *dst_orig,
 			xfrm_pols_put(pols, drop_pols);
 			XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTNOSTATES);
 
-			return make_blackhole(net, family, dst_orig);
+			return ERR_PTR(-EREMOTE);
 		}
 
 		err = -EAGAIN;
@@ -2212,6 +2210,23 @@ dropdst:
 }
 EXPORT_SYMBOL(xfrm_lookup);
 
+/* Callers of xfrm_lookup_route() must ensure a call to dst_output().
+ * Otherwise we may send out blackholed packets.
+ */
+struct dst_entry *xfrm_lookup_route(struct net *net, struct dst_entry *dst_orig,
+				    const struct flowi *fl,
+				    struct sock *sk, int flags)
+{
+	struct dst_entry *dst = xfrm_lookup(net, dst_orig, fl, sk,
+					    flags | XFRM_LOOKUP_QUEUE);
+
+	if (IS_ERR(dst) && PTR_ERR(dst) == -EREMOTE)
+		return make_blackhole(net, dst_orig->ops->family, dst_orig);
+
+	return dst;
+}
+EXPORT_SYMBOL(xfrm_lookup_route);
+
 static inline int
 xfrm_secpath_reject(int idx, struct sk_buff *skb, const struct flowi *fl)
 {
@@ -2477,7 +2492,7 @@ int __xfrm_route_forward(struct sk_buff *skb, unsigned short family)
 
 	skb_dst_force(skb);
 
-	dst = xfrm_lookup(net, skb_dst(skb), &fl, NULL, 0);
+	dst = xfrm_lookup(net, skb_dst(skb), &fl, NULL, XFRM_LOOKUP_QUEUE);
 	if (IS_ERR(dst)) {
 		res = 0;
 		dst = NULL;
@@ -2549,33 +2564,15 @@ static struct dst_entry *xfrm_negative_advice(struct dst_entry *dst)
 	return dst;
 }
 
-static void __xfrm_garbage_collect(struct net *net)
-{
-	struct dst_entry *head, *next;
-
-	spin_lock_bh(&net->xfrm.xfrm_policy_sk_bundle_lock);
-	head = xfrm_policy_sk_bundles;
-	xfrm_policy_sk_bundles = NULL;
-	spin_unlock_bh(&net->xfrm.xfrm_policy_sk_bundle_lock);
-
-	while (head) {
-		next = head->next;
-		dst_free(head);
-		head = next;
-	}
-}
-
 void xfrm_garbage_collect(struct net *net)
 {
-	flow_cache_flush();
-	__xfrm_garbage_collect(net);
+	flow_cache_flush(net);
 }
 EXPORT_SYMBOL(xfrm_garbage_collect);
 
 static void xfrm_garbage_collect_deferred(struct net *net)
 {
-	flow_cache_flush_deferred();
-	__xfrm_garbage_collect(net);
+	flow_cache_flush_deferred(net);
 }
 
 static void xfrm_init_pmtu(struct dst_entry *dst)
@@ -2810,21 +2807,19 @@ static struct notifier_block xfrm_dev_notifier = {
 static int __net_init xfrm_statistics_init(struct net *net)
 {
 	int rv;
-
-	if (snmp_mib_init((void __percpu **)net->mib.xfrm_statistics,
-			  sizeof(struct linux_xfrm_mib),
-			  __alignof__(struct linux_xfrm_mib)) < 0)
+	net->mib.xfrm_statistics = alloc_percpu(struct linux_xfrm_mib);
+	if (!net->mib.xfrm_statistics)
 		return -ENOMEM;
 	rv = xfrm_proc_init(net);
 	if (rv < 0)
-		snmp_mib_free((void __percpu **)net->mib.xfrm_statistics);
+		free_percpu(net->mib.xfrm_statistics);
 	return rv;
 }
 
 static void xfrm_statistics_fini(struct net *net)
 {
 	xfrm_proc_fini(net);
-	snmp_mib_free((void __percpu **)net->mib.xfrm_statistics);
+	free_percpu(net->mib.xfrm_statistics);
 }
 #else
 static int __net_init xfrm_statistics_init(struct net *net)
@@ -2889,21 +2884,14 @@ out_byidx:
 
 static void xfrm_policy_fini(struct net *net)
 {
-	struct xfrm_audit audit_info;
 	unsigned int sz;
 	int dir;
 
 	flush_work(&net->xfrm.policy_hash_work);
 #ifdef CONFIG_XFRM_SUB_POLICY
-	audit_info.loginuid = INVALID_UID;
-	audit_info.sessionid = (unsigned int)-1;
-	audit_info.secid = 0;
-	xfrm_policy_flush(net, XFRM_POLICY_TYPE_SUB, &audit_info);
+	xfrm_policy_flush(net, XFRM_POLICY_TYPE_SUB, false);
 #endif
-	audit_info.loginuid = INVALID_UID;
-	audit_info.sessionid = (unsigned int)-1;
-	audit_info.secid = 0;
-	xfrm_policy_flush(net, XFRM_POLICY_TYPE_MAIN, &audit_info);
+	xfrm_policy_flush(net, XFRM_POLICY_TYPE_MAIN, false);
 
 	WARN_ON(!list_empty(&net->xfrm.policy_all));
 
@@ -2940,15 +2928,19 @@ static int __net_init xfrm_net_init(struct net *net)
 	rv = xfrm_sysctl_init(net);
 	if (rv < 0)
 		goto out_sysctl;
+	rv = flow_cache_init(net);
+	if (rv < 0)
+		goto out;
 
 	/* Initialize the per-net locks here */
 	spin_lock_init(&net->xfrm.xfrm_state_lock);
 	rwlock_init(&net->xfrm.xfrm_policy_lock);
-	spin_lock_init(&net->xfrm.xfrm_policy_sk_bundle_lock);
 	mutex_init(&net->xfrm.xfrm_cfg_mutex);
 
 	return 0;
 
+out:
+	xfrm_sysctl_fini(net);
 out_sysctl:
 	xfrm_policy_fini(net);
 out_policy:
@@ -2961,6 +2953,7 @@ out_statistics:
 
 static void __net_exit xfrm_net_exit(struct net *net)
 {
+	flow_cache_fini(net);
 	xfrm_sysctl_fini(net);
 	xfrm_policy_fini(net);
 	xfrm_state_fini(net);
@@ -3013,15 +3006,14 @@ static void xfrm_audit_common_policyinfo(struct xfrm_policy *xp,
 	}
 }
 
-void xfrm_audit_policy_add(struct xfrm_policy *xp, int result,
-			   kuid_t auid, unsigned int sessionid, u32 secid)
+void xfrm_audit_policy_add(struct xfrm_policy *xp, int result, bool task_valid)
 {
 	struct audit_buffer *audit_buf;
 
 	audit_buf = xfrm_audit_start("SPD-add");
 	if (audit_buf == NULL)
 		return;
-	xfrm_audit_helper_usrinfo(auid, sessionid, secid, audit_buf);
+	xfrm_audit_helper_usrinfo(task_valid, audit_buf);
 	audit_log_format(audit_buf, " res=%u", result);
 	xfrm_audit_common_policyinfo(xp, audit_buf);
 	audit_log_end(audit_buf);
@@ -3029,14 +3021,14 @@ void xfrm_audit_policy_add(struct xfrm_policy *xp, int result,
 EXPORT_SYMBOL_GPL(xfrm_audit_policy_add);
 
 void xfrm_audit_policy_delete(struct xfrm_policy *xp, int result,
-			      kuid_t auid, unsigned int sessionid, u32 secid)
+			      bool task_valid)
 {
 	struct audit_buffer *audit_buf;
 
 	audit_buf = xfrm_audit_start("SPD-delete");
 	if (audit_buf == NULL)
 		return;
-	xfrm_audit_helper_usrinfo(auid, sessionid, secid, audit_buf);
+	xfrm_audit_helper_usrinfo(task_valid, audit_buf);
 	audit_log_format(audit_buf, " res=%u", result);
 	xfrm_audit_common_policyinfo(xp, audit_buf);
 	audit_log_end(audit_buf);