Merge v6.17.3linux-rolling-stable

Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

8 files changed, 45 insertions, 25 deletions

diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index c48c572f024d..1be0d91620a3 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -318,17 +318,17 @@ static bool icmpv4_xrlim_allow(struct net *net, struct rtable *rt,
 		return true;
 
 	/* No rate limit on loopback */
-	dev = dst_dev(dst);
+	rcu_read_lock();
+	dev = dst_dev_rcu(dst);
 	if (dev && (dev->flags & IFF_LOOPBACK))
 		goto out;
 
-	rcu_read_lock();
 	peer = inet_getpeer_v4(net->ipv4.peers, fl4->daddr,
 			       l3mdev_master_ifindex_rcu(dev));
 	rc = inet_peer_xrlim_allow(peer,
 				   READ_ONCE(net->ipv4.sysctl_icmp_ratelimit));
-	rcu_read_unlock();
 out:
+	rcu_read_unlock();
 	if (!rc)
 		__ICMP_INC_STATS(net, ICMP_MIB_RATELIMITHOST);
 	else
diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index b2584cce90ae..f7012479713b 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -476,14 +476,16 @@ out_fail:
 /* Process an incoming IP datagram fragment. */
 int ip_defrag(struct net *net, struct sk_buff *skb, u32 user)
 {
-	struct net_device *dev = skb->dev ? : skb_dst_dev(skb);
-	int vif = l3mdev_master_ifindex_rcu(dev);
+	struct net_device *dev;
 	struct ipq *qp;
+	int vif;
 
 	__IP_INC_STATS(net, IPSTATS_MIB_REASMREQDS);
 
 	/* Lookup (or create) queue header */
 	rcu_read_lock();
+	dev = skb->dev ? : skb_dst_dev_rcu(skb);
+	vif = l3mdev_master_ifindex_rcu(dev);
 	qp = ip_find(net, ip_hdr(skb), user, vif);
 	if (qp) {
 		int ret, refs = 0;
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index e86a8a862c41..8c568fbddb5f 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -1904,7 +1904,7 @@ static int ipmr_prepare_xmit(struct net *net, struct mr_table *mrt,
 		return -1;
 	}
 
-	encap += LL_RESERVED_SPACE(rt->dst.dev) + rt->dst.header_len;
+	encap += LL_RESERVED_SPACE(dst_dev_rcu(&rt->dst)) + rt->dst.header_len;
 
 	if (skb_cow(skb, encap)) {
 		ip_rt_put(rt);
@@ -1957,7 +1957,7 @@ static void ipmr_queue_fwd_xmit(struct net *net, struct mr_table *mrt,
 	 * result in receiving multiple packets.
 	 */
 	NF_HOOK(NFPROTO_IPV4, NF_INET_FORWARD,
-		net, NULL, skb, skb->dev, rt->dst.dev,
+		net, NULL, skb, skb->dev, dst_dev_rcu(&rt->dst),
 		ipmr_forward_finish);
 	return;
 
@@ -2301,7 +2301,7 @@ int ip_mr_output(struct net *net, struct sock *sk, struct sk_buff *skb)
 
 	guard(rcu)();
 
-	dev = rt->dst.dev;
+	dev = dst_dev_rcu(&rt->dst);
 
 	if (IPCB(skb)->flags & IPSKB_FORWARDED)
 		goto mc_output;
diff --git a/net/ipv4/ping.c b/net/ipv4/ping.c
index 031df4c19fcc..d2c3480df8f7 100644
--- a/net/ipv4/ping.c
+++ b/net/ipv4/ping.c
@@ -77,6 +77,7 @@ static inline struct hlist_head *ping_hashslot(struct ping_table *table,
 
 int ping_get_port(struct sock *sk, unsigned short ident)
 {
+	struct net *net = sock_net(sk);
 	struct inet_sock *isk, *isk2;
 	struct hlist_head *hlist;
 	struct sock *sk2 = NULL;
@@ -90,9 +91,10 @@ int ping_get_port(struct sock *sk, unsigned short ident)
 		for (i = 0; i < (1L << 16); i++, result++) {
 			if (!result)
 				result++; /* avoid zero */
-			hlist = ping_hashslot(&ping_table, sock_net(sk),
-					    result);
+			hlist = ping_hashslot(&ping_table, net, result);
 			sk_for_each(sk2, hlist) {
+				if (!net_eq(sock_net(sk2), net))
+					continue;
 				isk2 = inet_sk(sk2);
 
 				if (isk2->inet_num == result)
@@ -108,8 +110,10 @@ next_port:
 		if (i >= (1L << 16))
 			goto fail;
 	} else {
-		hlist = ping_hashslot(&ping_table, sock_net(sk), ident);
+		hlist = ping_hashslot(&ping_table, net, ident);
 		sk_for_each(sk2, hlist) {
+			if (!net_eq(sock_net(sk2), net))
+				continue;
 			isk2 = inet_sk(sk2);
 
 			/* BUG? Why is this reuse and not reuseaddr? ping.c
@@ -129,7 +133,7 @@ next_port:
 		pr_debug("was not hashed\n");
 		sk_add_node_rcu(sk, hlist);
 		sock_set_flag(sk, SOCK_RCU_FREE);
-		sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1);
+		sock_prot_inuse_add(net, sk->sk_prot, 1);
 	}
 	spin_unlock(&ping_table.lock);
 	return 0;
@@ -188,6 +192,8 @@ static struct sock *ping_lookup(struct net *net, struct sk_buff *skb, u16 ident)
 	}
 
 	sk_for_each_rcu(sk, hslot) {
+		if (!net_eq(sock_net(sk), net))
+			continue;
 		isk = inet_sk(sk);
 
 		pr_debug("iterate\n");
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index baa43e5966b1..5582ccd673ee 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -413,11 +413,11 @@ static struct neighbour *ipv4_neigh_lookup(const struct dst_entry *dst,
 					   const void *daddr)
 {
 	const struct rtable *rt = container_of(dst, struct rtable, dst);
-	struct net_device *dev = dst_dev(dst);
+	struct net_device *dev;
 	struct neighbour *n;
 
 	rcu_read_lock();
-
+	dev = dst_dev_rcu(dst);
 	if (likely(rt->rt_gw_family == AF_INET)) {
 		n = ip_neigh_gw4(dev, rt->rt_gw4);
 	} else if (rt->rt_gw_family == AF_INET6) {
@@ -1026,7 +1026,7 @@ static void __ip_rt_update_pmtu(struct rtable *rt, struct flowi4 *fl4, u32 mtu)
 		return;
 
 	rcu_read_lock();
-	net = dev_net_rcu(dst_dev(dst));
+	net = dst_dev_net_rcu(dst);
 	if (mtu < net->ipv4.ip_rt_min_pmtu) {
 		lock = true;
 		mtu = min(old_mtu, net->ipv4.ip_rt_min_pmtu);
@@ -1326,7 +1326,7 @@ static unsigned int ipv4_default_advmss(const struct dst_entry *dst)
 	struct net *net;
 
 	rcu_read_lock();
-	net = dev_net_rcu(dst_dev(dst));
+	net = dst_dev_net_rcu(dst);
 	advmss = max_t(unsigned int, ipv4_mtu(dst) - header_size,
 				   net->ipv4.ip_rt_min_advmss);
 	rcu_read_unlock();
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index ad76556800f2..89040007c7b7 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -3099,8 +3099,8 @@ bool tcp_check_oom(const struct sock *sk, int shift)
 
 void __tcp_close(struct sock *sk, long timeout)
 {
+	bool data_was_unread = false;
 	struct sk_buff *skb;
-	int data_was_unread = 0;
 	int state;
 
 	WRITE_ONCE(sk->sk_shutdown, SHUTDOWN_MASK);
@@ -3119,11 +3119,12 @@ void __tcp_close(struct sock *sk, long timeout)
 	 *  reader process may not have drained the data yet!
 	 */
 	while ((skb = __skb_dequeue(&sk->sk_receive_queue)) != NULL) {
-		u32 len = TCP_SKB_CB(skb)->end_seq - TCP_SKB_CB(skb)->seq;
+		u32 end_seq = TCP_SKB_CB(skb)->end_seq;
 
 		if (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN)
-			len--;
-		data_was_unread += len;
+			end_seq--;
+		if (after(end_seq, tcp_sk(sk)->copied_seq))
+			data_was_unread = true;
 		__kfree_skb(skb);
 	}
 
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 71b76e98371a..64f93668a845 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -4890,12 +4890,23 @@ static int tcp_prune_queue(struct sock *sk, const struct sk_buff *in_skb);
 
 /* Check if this incoming skb can be added to socket receive queues
  * while satisfying sk->sk_rcvbuf limit.
+ *
+ * In theory we should use skb->truesize, but this can cause problems
+ * when applications use too small SO_RCVBUF values.
+ * When LRO / hw gro is used, the socket might have a high tp->scaling_ratio,
+ * allowing RWIN to be close to available space.
+ * Whenever the receive queue gets full, we can receive a small packet
+ * filling RWIN, but with a high skb->truesize, because most NIC use 4K page
+ * plus sk_buff metadata even when receiving less than 1500 bytes of payload.
+ *
+ * Note that we use skb->len to decide to accept or drop this packet,
+ * but sk->sk_rmem_alloc is the sum of all skb->truesize.
  */
 static bool tcp_can_ingest(const struct sock *sk, const struct sk_buff *skb)
 {
-	unsigned int new_mem = atomic_read(&sk->sk_rmem_alloc) + skb->truesize;
+	unsigned int rmem = atomic_read(&sk->sk_rmem_alloc);
 
-	return new_mem <= sk->sk_rcvbuf;
+	return rmem + skb->len <= sk->sk_rcvbuf;
 }
 
 static int tcp_try_rmem_schedule(struct sock *sk, const struct sk_buff *skb,
diff --git a/net/ipv4/tcp_metrics.c b/net/ipv4/tcp_metrics.c
index 03c068ea27b6..10e86f1008e9 100644
--- a/net/ipv4/tcp_metrics.c
+++ b/net/ipv4/tcp_metrics.c
@@ -170,7 +170,7 @@ static struct tcp_metrics_block *tcpm_new(struct dst_entry *dst,
 	struct net *net;
 
 	spin_lock_bh(&tcp_metrics_lock);
-	net = dev_net_rcu(dst_dev(dst));
+	net = dst_dev_net_rcu(dst);
 
 	/* While waiting for the spin-lock the cache might have been populated
 	 * with this entry and so we have to check again.
@@ -273,7 +273,7 @@ static struct tcp_metrics_block *__tcp_get_metrics_req(struct request_sock *req,
 		return NULL;
 	}
 
-	net = dev_net_rcu(dst_dev(dst));
+	net = dst_dev_net_rcu(dst);
 	hash ^= net_hash_mix(net);
 	hash = hash_32(hash, tcp_metrics_hash_log);
 
@@ -318,7 +318,7 @@ static struct tcp_metrics_block *tcp_get_metrics(struct sock *sk,
 	else
 		return NULL;
 
-	net = dev_net_rcu(dst_dev(dst));
+	net = dst_dev_net_rcu(dst);
 	hash ^= net_hash_mix(net);
 	hash = hash_32(hash, tcp_metrics_hash_log);