7 files changed, 67 insertions, 25 deletions
diff --git a/net/core/dev.c b/net/core/dev.c
index 08ce317fcec8..30a1603a7225 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -397,16 +397,18 @@ static void list_netdevice(struct net_device *dev)
 /* Device list removal
  * caller must respect a RCU grace period before freeing/reusing dev
  */
-static void unlist_netdevice(struct net_device *dev)
+static void unlist_netdevice(struct net_device *dev, bool lock)
 {
 	ASSERT_RTNL();
 
 	/* Unlink dev from the device chain */
-	write_lock(&dev_base_lock);
+	if (lock)
+		write_lock(&dev_base_lock);
 	list_del_rcu(&dev->dev_list);
 	netdev_name_node_del(dev->name_node);
 	hlist_del_rcu(&dev->index_hlist);
-	write_unlock(&dev_base_lock);
+	if (lock)
+		write_unlock(&dev_base_lock);
 
 	dev_base_seq_inc(dev_net(dev));
 }
@@ -4861,7 +4863,10 @@ static u32 netif_receive_generic_xdp(struct sk_buff *skb,
 }
 
 /* When doing generic XDP we have to bypass the qdisc layer and the
- * network taps in order to match in-driver-XDP behavior.
+ * network taps in order to match in-driver-XDP behavior. This also means
+ * that XDP packets are able to starve other packets going through a qdisc,
+ * and DDOS attacks will be more effective. In-driver-XDP use dedicated TX
+ * queues, so they do not have this starvation issue.
  */
 void generic_xdp_tx(struct sk_buff *skb, struct bpf_prog *xdp_prog)
 {
@@ -4873,7 +4878,7 @@ void generic_xdp_tx(struct sk_buff *skb, struct bpf_prog *xdp_prog)
 	txq = netdev_core_pick_tx(dev, skb, NULL);
 	cpu = smp_processor_id();
 	HARD_TX_LOCK(dev, txq, cpu);
-	if (!netif_xmit_stopped(txq)) {
+	if (!netif_xmit_frozen_or_drv_stopped(txq)) {
 		rc = netdev_start_xmit(skb, dev, txq, 0);
 		if (dev_xmit_complete(rc))
 			free_skb = false;
@@ -4881,6 +4886,7 @@ void generic_xdp_tx(struct sk_buff *skb, struct bpf_prog *xdp_prog)
 	HARD_TX_UNLOCK(dev, txq);
 	if (free_skb) {
 		trace_xdp_exception(dev, xdp_prog, XDP_TX);
+		dev_core_stats_tx_dropped_inc(dev);
 		kfree_skb(skb);
 	}
 }
@@ -10043,11 +10049,11 @@ int register_netdevice(struct net_device *dev)
 		goto err_uninit;
 
 	ret = netdev_register_kobject(dev);
-	if (ret) {
-		dev->reg_state = NETREG_UNREGISTERED;
+	write_lock(&dev_base_lock);
+	dev->reg_state = ret ? NETREG_UNREGISTERED : NETREG_REGISTERED;
+	write_unlock(&dev_base_lock);
+	if (ret)
 		goto err_uninit;
-	}
-	dev->reg_state = NETREG_REGISTERED;
 
 	__netdev_update_features(dev);
 
@@ -10329,7 +10335,9 @@ void netdev_run_todo(void)
 			continue;
 		}
 
+		write_lock(&dev_base_lock);
 		dev->reg_state = NETREG_UNREGISTERED;
+		write_unlock(&dev_base_lock);
 		linkwatch_forget_dev(dev);
 	}
 
@@ -10810,9 +10818,10 @@ void unregister_netdevice_many(struct list_head *head)
 
 	list_for_each_entry(dev, head, unreg_list) {
 		/* And unlink it from device chain. */
-		unlist_netdevice(dev);
-
+		write_lock(&dev_base_lock);
+		unlist_netdevice(dev, false);
 		dev->reg_state = NETREG_UNREGISTERING;
+		write_unlock(&dev_base_lock);
 	}
 	flush_all_backlogs();
 
@@ -10959,7 +10968,7 @@ int __dev_change_net_namespace(struct net_device *dev, struct net *net,
 	dev_close(dev);
 
 	/* And unlink it from device chain */
-	unlist_netdevice(dev);
+	unlist_netdevice(dev, true);
 
 	synchronize_net();
 
diff --git a/net/core/filter.c b/net/core/filter.c
index 5af58eb48587..7950f7520765 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -6158,7 +6158,6 @@ static int bpf_push_seg6_encap(struct sk_buff *skb, u32 type, void *hdr, u32 len
 	if (err)
 		return err;
 
-	ipv6_hdr(skb)->payload_len = htons(skb->len - sizeof(struct ipv6hdr));
 	skb_set_transport_header(skb, sizeof(struct ipv6hdr));
 
 	return seg6_lookup_nexthop(skb, NULL, 0);
@@ -6516,10 +6515,21 @@ __bpf_sk_lookup(struct sk_buff *skb, struct bpf_sock_tuple *tuple, u32 len,
 					   ifindex, proto, netns_id, flags);
 
 	if (sk) {
-		sk = sk_to_full_sk(sk);
-		if (!sk_fullsock(sk)) {
+		struct sock *sk2 = sk_to_full_sk(sk);
+
+		/* sk_to_full_sk() may return (sk)->rsk_listener, so make sure the original sk
+		 * sock refcnt is decremented to prevent a request_sock leak.
+		 */
+		if (!sk_fullsock(sk2))
+			sk2 = NULL;
+		if (sk2 != sk) {
 			sock_gen_put(sk);
-			return NULL;
+			/* Ensure there is no need to bump sk2 refcnt */
+			if (unlikely(sk2 && !sock_flag(sk2, SOCK_RCU_FREE))) {
+				WARN_ONCE(1, "Found non-RCU, unreferenced socket!");
+				return NULL;
+			}
+			sk = sk2;
 		}
 	}
 
@@ -6553,10 +6563,21 @@ bpf_sk_lookup(struct sk_buff *skb, struct bpf_sock_tuple *tuple, u32 len,
 					 flags);
 
 	if (sk) {
-		sk = sk_to_full_sk(sk);
-		if (!sk_fullsock(sk)) {
+		struct sock *sk2 = sk_to_full_sk(sk);
+
+		/* sk_to_full_sk() may return (sk)->rsk_listener, so make sure the original sk
+		 * sock refcnt is decremented to prevent a request_sock leak.
+		 */
+		if (!sk_fullsock(sk2))
+			sk2 = NULL;
+		if (sk2 != sk) {
 			sock_gen_put(sk);
-			return NULL;
+			/* Ensure there is no need to bump sk2 refcnt */
+			if (unlikely(sk2 && !sock_flag(sk2, SOCK_RCU_FREE))) {
+				WARN_ONCE(1, "Found non-RCU, unreferenced socket!");
+				return NULL;
+			}
+			sk = sk2;
 		}
 	}
 
@@ -7020,7 +7041,7 @@ BPF_CALL_5(bpf_tcp_check_syncookie, struct sock *, sk, void *, iph, u32, iph_len
 	if (sk->sk_protocol != IPPROTO_TCP || sk->sk_state != TCP_LISTEN)
 		return -EINVAL;
 
-	if (!sock_net(sk)->ipv4.sysctl_tcp_syncookies)
+	if (!READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_syncookies))
 		return -EINVAL;
 
 	if (!th->ack || th->rst || th->syn)
@@ -7095,7 +7116,7 @@ BPF_CALL_5(bpf_tcp_gen_syncookie, struct sock *, sk, void *, iph, u32, iph_len,
 	if (sk->sk_protocol != IPPROTO_TCP || sk->sk_state != TCP_LISTEN)
 		return -EINVAL;
 
-	if (!sock_net(sk)->ipv4.sysctl_tcp_syncookies)
+	if (!READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_syncookies))
 		return -ENOENT;
 
 	if (!th->syn || th->ack || th->fin || th->rst)
diff --git a/net/core/flow_offload.c b/net/core/flow_offload.c
index 73f68d4625f3..929f6379a279 100644
--- a/net/core/flow_offload.c
+++ b/net/core/flow_offload.c
@@ -595,3 +595,9 @@ int flow_indr_dev_setup_offload(struct net_device *dev,	struct Qdisc *sch,
 	return (bo && list_empty(&bo->cb_list)) ? -EOPNOTSUPP : count;
 }
 EXPORT_SYMBOL(flow_indr_dev_setup_offload);
+
+bool flow_indr_dev_exists(void)
+{
+	return !list_empty(&flow_block_indr_dev_list);
+}
+EXPORT_SYMBOL(flow_indr_dev_exists);
diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c
index e319e242dddf..a3642569fe53 100644
--- a/net/core/net-sysfs.c
+++ b/net/core/net-sysfs.c
@@ -33,6 +33,7 @@ static const char fmt_dec[] = "%d\n";
 static const char fmt_ulong[] = "%lu\n";
 static const char fmt_u64[] = "%llu\n";
 
+/* Caller holds RTNL or dev_base_lock */
 static inline int dev_isalive(const struct net_device *dev)
 {
 	return dev->reg_state <= NETREG_REGISTERED;
diff --git a/net/core/secure_seq.c b/net/core/secure_seq.c
index 5f85e01d4093..b0ff6153be62 100644
--- a/net/core/secure_seq.c
+++ b/net/core/secure_seq.c
@@ -64,7 +64,7 @@ u32 secure_tcpv6_ts_off(const struct net *net,
 		.daddr = *(struct in6_addr *)daddr,
 	};
 
-	if (net->ipv4.sysctl_tcp_timestamps != 1)
+	if (READ_ONCE(net->ipv4.sysctl_tcp_timestamps) != 1)
 		return 0;
 
 	ts_secret_init();
@@ -120,7 +120,7 @@ EXPORT_SYMBOL(secure_ipv6_port_ephemeral);
 #ifdef CONFIG_INET
 u32 secure_tcp_ts_off(const struct net *net, __be32 saddr, __be32 daddr)
 {
-	if (net->ipv4.sysctl_tcp_timestamps != 1)
+	if (READ_ONCE(net->ipv4.sysctl_tcp_timestamps) != 1)
 		return 0;
 
 	ts_secret_init();
diff --git a/net/core/skmsg.c b/net/core/skmsg.c
index 22b983ade0e7..b0fcd0200e84 100644
--- a/net/core/skmsg.c
+++ b/net/core/skmsg.c
@@ -699,6 +699,11 @@ struct sk_psock *sk_psock_init(struct sock *sk, int node)
 
 	write_lock_bh(&sk->sk_callback_lock);
 
+	if (sk_is_inet(sk) && inet_csk_has_ulp(sk)) {
+		psock = ERR_PTR(-EINVAL);
+		goto out;
+	}
+
 	if (sk->sk_user_data) {
 		psock = ERR_PTR(-EBUSY);
 		goto out;
diff --git a/net/core/sock_reuseport.c b/net/core/sock_reuseport.c
index 3f00a28fe762..5daa1fa54249 100644
--- a/net/core/sock_reuseport.c
+++ b/net/core/sock_reuseport.c
@@ -387,7 +387,7 @@ void reuseport_stop_listen_sock(struct sock *sk)
 		prog = rcu_dereference_protected(reuse->prog,
 						 lockdep_is_held(&reuseport_lock));
 
-		if (sock_net(sk)->ipv4.sysctl_tcp_migrate_req ||
+		if (READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_migrate_req) ||
 		    (prog && prog->expected_attach_type == BPF_SK_REUSEPORT_SELECT_OR_MIGRATE)) {
 			/* Migration capable, move sk from the listening section
 			 * to the closed section.
@@ -545,7 +545,7 @@ struct sock *reuseport_migrate_sock(struct sock *sk,
 	hash = migrating_sk->sk_hash;
 	prog = rcu_dereference(reuse->prog);
 	if (!prog || prog->expected_attach_type != BPF_SK_REUSEPORT_SELECT_OR_MIGRATE) {
-		if (sock_net(sk)->ipv4.sysctl_tcp_migrate_req)
+		if (READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_migrate_req))
 			goto select_by_hash;
 		goto failure;
 	}