[LTS 9.2] net RCU: CVE-2025-40170, CVE-2025-40158, CVE-2025-40135, CVE-2025-21765, CVE-2025-21766, CVE-2024-44985, CVE-2024-44986

drivers/net/vrf.c

@@ -664,18 +664,18 @@ static int vrf_finish_output6(struct net *net, struct sock *sk,
 	skb->protocol = htons(ETH_P_IPV6);
 	skb->dev = dev;
 
-	rcu_read_lock_bh();
+	rcu_read_lock();
 	nexthop = rt6_nexthop((struct rt6_info *)dst, &ipv6_hdr(skb)->daddr);
 	neigh = __ipv6_neigh_lookup_noref(dst->dev, nexthop);
 	if (unlikely(!neigh))
 		neigh = __neigh_create(&nd_tbl, nexthop, dst->dev, false);
 	if (!IS_ERR(neigh)) {
 		sock_confirm_neigh(skb, neigh);
 		ret = neigh_output(neigh, skb, false);
-		rcu_read_unlock_bh();
+		rcu_read_unlock();
 		return ret;
 	}
-	rcu_read_unlock_bh();
+	rcu_read_unlock();
 
 	IP6_INC_STATS(dev_net(dst->dev),
 		      ip6_dst_idev(dst), IPSTATS_MIB_OUTNOROUTES);
@@ -889,7 +889,7 @@ static int vrf_finish_output(struct net *net, struct sock *sk, struct sk_buff *s
 		}
 	}
 
-	rcu_read_lock_bh();
+	rcu_read_lock();
 
 	neigh = ip_neigh_for_gw(rt, skb, &is_v6gw);
 	if (!IS_ERR(neigh)) {
@@ -898,11 +898,11 @@ static int vrf_finish_output(struct net *net, struct sock *sk, struct sk_buff *s
 		sock_confirm_neigh(skb, neigh);
 		/* if crossing protocols, can not use the cached header */
 		ret = neigh_output(neigh, skb, is_v6gw);
-		rcu_read_unlock_bh();
+		rcu_read_unlock();
 		return ret;
 	}
 
-	rcu_read_unlock_bh();
+	rcu_read_unlock();
 	vrf_tx_error(skb->dev, skb);
 	return -EINVAL;
 }

include/linux/rtnetlink.h

@@ -60,16 +60,6 @@ static inline bool lockdep_rtnl_is_held(void)
 #define rcu_dereference_rtnl(p)					\
 	rcu_dereference_check(p, lockdep_rtnl_is_held())
 
-/**
- * rcu_dereference_bh_rtnl - rcu_dereference_bh with debug checking
- * @p: The pointer to read, prior to dereference
- *
- * Do an rcu_dereference_bh(p), but check caller either holds rcu_read_lock_bh()
- * or RTNL. Note : Please prefer rtnl_dereference() or rcu_dereference_bh()
- */
-#define rcu_dereference_bh_rtnl(p)				\
-	rcu_dereference_bh_check(p, lockdep_rtnl_is_held())
-
 /**
  * rtnl_dereference - fetch RCU pointer when updates are prevented by RTNL
  * @p: The pointer to read, prior to dereferencing

include/net/arp.h

@@ -38,11 +38,11 @@ static inline struct neighbour *__ipv4_neigh_lookup(struct net_device *dev, u32
 {
 	struct neighbour *n;
 
-	rcu_read_lock_bh();
+	rcu_read_lock();
 	n = __ipv4_neigh_lookup_noref(dev, key);
 	if (n && !refcount_inc_not_zero(&n->refcnt))
 		n = NULL;
-	rcu_read_unlock_bh();
+	rcu_read_unlock();
 
 	return n;
 }
@@ -51,16 +51,10 @@ static inline void __ipv4_confirm_neigh(struct net_device *dev, u32 key)
 {
 	struct neighbour *n;
 
-	rcu_read_lock_bh();
+	rcu_read_lock();
 	n = __ipv4_neigh_lookup_noref(dev, key);
-	if (n) {
-		unsigned long now = jiffies;
-
-		/* avoid dirtying neighbour */
-		if (READ_ONCE(n->confirmed) != now)
-			WRITE_ONCE(n->confirmed, now);
-	}
-	rcu_read_unlock_bh();
+	neigh_confirm(n);
+	rcu_read_unlock();
 }
 
 void arp_init(void);

include/net/dst.h

@@ -23,7 +23,12 @@
 struct sk_buff;
 
 struct dst_entry {
-	struct net_device       *dev;
+	RH_KABI_REPLACE(
+		struct net_device       *dev,
+		union {
+			struct net_device       *dev;
+			struct net_device __rcu *dev_rcu;
+		})
 	struct  dst_ops	        *ops;
 	unsigned long		_metrics;
 	unsigned long           expires;
@@ -553,6 +558,41 @@ static inline void skb_dst_update_pmtu_no_confirm(struct sk_buff *skb, u32 mtu)
 		dst->ops->update_pmtu(dst, NULL, skb, mtu, false);
 }
 
+static inline struct net_device *dst_dev(const struct dst_entry *dst)
+{
+	return READ_ONCE(dst->dev);
+}
+
+static inline struct net_device *dst_dev_rcu(const struct dst_entry *dst)
+{
+	return rcu_dereference(dst->dev_rcu);
+}
+
+static inline struct net *dst_dev_net_rcu(const struct dst_entry *dst)
+{
+	return dev_net_rcu(dst_dev_rcu(dst));
+}
+
+static inline struct net_device *skb_dst_dev(const struct sk_buff *skb)
+{
+	return dst_dev(skb_dst(skb));
+}
+
+static inline struct net_device *skb_dst_dev_rcu(const struct sk_buff *skb)
+{
+	return dst_dev_rcu(skb_dst(skb));
+}
+
+static inline struct net *skb_dst_dev_net(const struct sk_buff *skb)
+{
+	return dev_net(skb_dst_dev(skb));
+}
+
+static inline struct net *skb_dst_dev_net_rcu(const struct sk_buff *skb)
+{
+	return dev_net_rcu(skb_dst_dev_rcu(skb));
+}
+
 struct dst_entry *dst_blackhole_check(struct dst_entry *dst, u32 cookie);
 void dst_blackhole_update_pmtu(struct dst_entry *dst, struct sock *sk,
 			       struct sk_buff *skb, u32 mtu, bool confirm_neigh);

include/net/ip.h

@@ -440,20 +440,43 @@ static inline bool ip_sk_ignore_df(const struct sock *sk)
 static inline unsigned int ip_dst_mtu_maybe_forward(const struct dst_entry *dst,
 						    bool forwarding)
 {
-	struct net *net = dev_net(dst->dev);
-	unsigned int mtu;
+	const struct rtable *rt = container_of(dst, struct rtable, dst);
+	const struct net_device *dev;
+	unsigned int mtu, res;
+	struct net *net;
+
+	rcu_read_lock();
 
+	dev = dst_dev_rcu(dst);
+	net = dev_net_rcu(dev);
 	if (READ_ONCE(net->ipv4.sysctl_ip_fwd_use_pmtu) ||
 	    ip_mtu_locked(dst) ||
-	    !forwarding)
-		return dst_mtu(dst);
+	    !forwarding) {
+		mtu = rt->rt_pmtu;
+		if (mtu && time_before(jiffies, rt->dst.expires))
+			goto out;
+	}
 
 	/* 'forwarding = true' case should always honour route mtu */
 	mtu = dst_metric_raw(dst, RTAX_MTU);
-	if (!mtu)
-		mtu = min(READ_ONCE(dst->dev->mtu), IP_MAX_MTU);
+	if (mtu)
+		goto out;
+
+	mtu = READ_ONCE(dev->mtu);
+
+	if (unlikely(ip_mtu_locked(dst))) {
+		if (rt->rt_uses_gateway && mtu > 576)
+			mtu = 576;
+	}
+
+out:
+	mtu = min_t(unsigned int, mtu, IP_MAX_MTU);
+
+	res = mtu - lwtunnel_headroom(dst->lwtstate, mtu);
+
+	rcu_read_unlock();
 
-	return mtu - lwtunnel_headroom(dst->lwtstate, mtu);
+	return res;
 }
 
 static inline unsigned int ip_skb_dst_mtu(struct sock *sk,

include/net/ip6_route.h

@@ -316,20 +316,21 @@ static inline bool rt6_duplicate_nexthop(struct fib6_info *a, struct fib6_info *
 	       !lwtunnel_cmp_encap(nha->fib_nh_lws, nhb->fib_nh_lws);
 }
 
-static inline unsigned int ip6_dst_mtu_forward(const struct dst_entry *dst)
+static inline unsigned int ip6_dst_mtu_maybe_forward(const struct dst_entry *dst,
+						     bool forwarding)
 {
 	struct inet6_dev *idev;
 	unsigned int mtu;
 
-	if (dst_metric_locked(dst, RTAX_MTU)) {
+	if (!forwarding || dst_metric_locked(dst, RTAX_MTU)) {
 		mtu = dst_metric_raw(dst, RTAX_MTU);
 		if (mtu)
 			goto out;
 	}
 
 	mtu = IPV6_MIN_MTU;
 	rcu_read_lock();
-	idev = __in6_dev_get(dst->dev);
+	idev = __in6_dev_get(dst_dev_rcu(dst));
 	if (idev)
 		mtu = idev->cnf.mtu6;
 	rcu_read_unlock();

include/net/ndisc.h

@@ -395,11 +395,11 @@ static inline struct neighbour *__ipv6_neigh_lookup(struct net_device *dev, cons
 {
 	struct neighbour *n;
 
-	rcu_read_lock_bh();
+	rcu_read_lock();
 	n = __ipv6_neigh_lookup_noref(dev, pkey);
 	if (n && !refcount_inc_not_zero(&n->refcnt))
 		n = NULL;
-	rcu_read_unlock_bh();
+	rcu_read_unlock();
 
 	return n;
 }
@@ -409,33 +409,21 @@ static inline void __ipv6_confirm_neigh(struct net_device *dev,
 {
 	struct neighbour *n;
 
-	rcu_read_lock_bh();
+	rcu_read_lock();
 	n = __ipv6_neigh_lookup_noref(dev, pkey);
-	if (n) {
-		unsigned long now = jiffies;
-
-		/* avoid dirtying neighbour */
-		if (READ_ONCE(n->confirmed) != now)
-			WRITE_ONCE(n->confirmed, now);
-	}
-	rcu_read_unlock_bh();
+	neigh_confirm(n);
+	rcu_read_unlock();
 }
 
 static inline void __ipv6_confirm_neigh_stub(struct net_device *dev,
 					     const void *pkey)
 {
 	struct neighbour *n;
 
-	rcu_read_lock_bh();
+	rcu_read_lock();
 	n = __ipv6_neigh_lookup_noref_stub(dev, pkey);
-	if (n) {
-		unsigned long now = jiffies;
-
-		/* avoid dirtying neighbour */
-		if (READ_ONCE(n->confirmed) != now)
-			WRITE_ONCE(n->confirmed, now);
-	}
-	rcu_read_unlock_bh();
+	neigh_confirm(n);
+	rcu_read_unlock();
 }
 
 /* uses ipv6_stub and is meant for use outside of IPv6 core */

include/net/neighbour.h

@@ -290,14 +290,14 @@ static inline struct neighbour *___neigh_lookup_noref(
 	const void *pkey,
 	struct net_device *dev)
 {
-	struct neigh_hash_table *nht = rcu_dereference_bh(tbl->nht);
+	struct neigh_hash_table *nht = rcu_dereference(tbl->nht);
 	struct neighbour *n;
 	u32 hash_val;
 
 	hash_val = hash(pkey, dev, nht->hash_rnd) >> (32 - nht->hash_shift);
-	for (n = rcu_dereference_bh(nht->hash_buckets[hash_val]);
+	for (n = rcu_dereference(nht->hash_buckets[hash_val]);
 	     n != NULL;
-	     n = rcu_dereference_bh(n->next)) {
+	     n = rcu_dereference(n->next)) {
 		if (n->dev == dev && key_eq(n, pkey))
 			return n;
 	}
@@ -312,6 +312,17 @@ static inline struct neighbour *__neigh_lookup_noref(struct neigh_table *tbl,
 	return ___neigh_lookup_noref(tbl, tbl->key_eq, tbl->hash, pkey, dev);
 }
 
+static inline void neigh_confirm(struct neighbour *n)
+{
+	if (n) {
+		unsigned long now = jiffies;
+
+		/* avoid dirtying neighbour */
+		if (READ_ONCE(n->confirmed) != now)
+			WRITE_ONCE(n->confirmed, now);
+	}
+}
+
 void neigh_table_init(int index, struct neigh_table *tbl);
 int neigh_table_clear(int index, struct neigh_table *tbl);
 struct neighbour *neigh_lookup(struct neigh_table *tbl, const void *pkey,

include/net/nexthop.h

@@ -497,29 +497,6 @@ static inline struct fib6_nh *nexthop_fib6_nh(struct nexthop *nh)
 	return NULL;
 }
 
-/* Variant of nexthop_fib6_nh().
- * Caller should either hold rcu_read_lock_bh(), or RTNL.
- */
-static inline struct fib6_nh *nexthop_fib6_nh_bh(struct nexthop *nh)
-{
-	struct nh_info *nhi;
-
-	if (nh->is_group) {
-		struct nh_group *nh_grp;
-
-		nh_grp = rcu_dereference_bh_rtnl(nh->nh_grp);
-		nh = nexthop_mpath_select(nh_grp, 0);
-		if (!nh)
-			return NULL;
-	}
-
-	nhi = rcu_dereference_bh_rtnl(nh->nh_info);
-	if (nhi->family == AF_INET6)
-		return &nhi->fib6_nh;
-
-	return NULL;
-}
-
 static inline struct net_device *fib6_info_nh_dev(struct fib6_info *f6i)
 {
 	struct fib6_nh *fib6_nh;

include/net/route.h

@@ -358,10 +358,15 @@ static inline int inet_iif(const struct sk_buff *skb)
 static inline int ip4_dst_hoplimit(const struct dst_entry *dst)
 {
 	int hoplimit = dst_metric_raw(dst, RTAX_HOPLIMIT);
-	struct net *net = dev_net(dst->dev);
 
-	if (hoplimit == 0)
+	if (hoplimit == 0) {
+		const struct net *net;
+
+		rcu_read_lock();
+		net = dst_dev_net_rcu(dst);
 		hoplimit = READ_ONCE(net->ipv4.sysctl_ip_default_ttl);
+		rcu_read_unlock();
+	}
 	return hoplimit;
 }
 

include/net/sock.h

@@ -2192,13 +2192,10 @@ static inline void sock_confirm_neigh(struct sk_buff *skb, struct neighbour *n)
 {
 	if (skb_get_dst_pending_confirm(skb)) {
 		struct sock *sk = skb->sk;
-		unsigned long now = jiffies;
 
-		/* avoid dirtying neighbour */
-		if (READ_ONCE(n->confirmed) != now)
-			WRITE_ONCE(n->confirmed, now);
 		if (sk && READ_ONCE(sk->sk_dst_pending_confirm))
 			WRITE_ONCE(sk->sk_dst_pending_confirm, 0);
+		neigh_confirm(n);
 	}
 }
 

net/core/dst.c

@@ -154,7 +154,7 @@ void dst_dev_put(struct dst_entry *dst)
 		dst->ops->ifdown(dst, dev, true);
 	dst->input = dst_discard;
 	dst->output = dst_discard_out;
-	dst->dev = blackhole_netdev;
+	rcu_assign_pointer(dst->dev_rcu, blackhole_netdev);
 	dev_replace_track(dev, blackhole_netdev, &dst->dev_tracker,
 			  GFP_ATOMIC);
 }
@@ -265,7 +265,7 @@ unsigned int dst_blackhole_mtu(const struct dst_entry *dst)
 {
 	unsigned int mtu = dst_metric_raw(dst, RTAX_MTU);
 
-	return mtu ? : dst->dev->mtu;
+	return mtu ? : dst_dev(dst)->mtu;
 }
 EXPORT_SYMBOL_GPL(dst_blackhole_mtu);