Merge tag 'net-5.16-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net

Pull networking fixes from Jakub Kicinski:
 "Including fixes from netfilter.

  Current release - regressions:

   - revert "tipc: use consistent GFP flags"

  Previous releases - regressions:

   - igb: fix deadlock caused by taking RTNL in runtime resume path

   - accept UFOv6 packages in virtio_net_hdr_to_skb

   - netfilter: fix regression in looped (broad|multi)cast's MAC
     handling

   - bridge: fix ioctl old_deviceless bridge argument

   - ice: xsk: do not clear status_error0 for ntu + nb_buffs descriptor,
     avoid stalls when multiple sockets use an interface

  Previous releases - always broken:

   - inet: fully convert sk->sk_rx_dst to RCU rules

   - veth: ensure skb entering GRO are not cloned

   - sched: fix zone matching for invalid conntrack state

   - bonding: fix ad_actor_system option setting to default

   - nf_tables: fix use-after-free in nft_set_catchall_destroy()

   - lantiq_xrx200: increase buffer reservation to avoid mem corruption

   - ice: xsk: avoid leaking app buffers during clean up

   - tun: avoid double free in tun_free_netdev"

* tag 'net-5.16-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (45 commits)
  net: stmmac: dwmac-visconti: Fix value of ETHER_CLK_SEL_FREQ_SEL_2P5M
  r8152: sync ocp base
  r8152: fix the force speed doesn't work for RTL8156
  net: bridge: fix ioctl old_deviceless bridge argument
  net: stmmac: ptp: fix potentially overflowing expression
  net: dsa: tag_ocelot: use traffic class to map priority on injected header
  veth: ensure skb entering GRO are not cloned.
  asix: fix wrong return value in asix_check_host_enable()
  asix: fix uninit-value in asix_mdio_read()
  sfc: falcon: Check null pointer of rx_queue->page_ring
  sfc: Check null pointer of rx_queue->page_ring
  net: ks8851: Check for error irq
  drivers: net: smc911x: Check for error irq
  fjes: Check for error irq
  bonding: fix ad_actor_system option setting to default
  igb: fix deadlock caused by taking RTNL in RPM resume path
  gve: Correct order of processing device options
  net: skip virtio_net_hdr_set_proto if protocol already set
  net: accept UFOv6 packages in virtio_net_hdr_to_skb
  docs: networking: replace skb_hwtstamp_tx with skb_tstamp_tx
  ...

5 files changed, 13 insertions, 11 deletions

diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 0189e3cd4a7d..6b5956500436 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -154,7 +154,7 @@ void inet_sock_destruct(struct sock *sk)
 
 	kfree(rcu_dereference_protected(inet->inet_opt, 1));
 	dst_release(rcu_dereference_protected(sk->sk_dst_cache, 1));
-	dst_release(sk->sk_rx_dst);
+	dst_release(rcu_dereference_protected(sk->sk_rx_dst, 1));
 	sk_refcnt_debug_dec(sk);
 }
 EXPORT_SYMBOL(inet_sock_destruct);
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index bbb3d39c69af..2bb28bfd83bf 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -3012,8 +3012,7 @@ int tcp_disconnect(struct sock *sk, int flags)
 	icsk->icsk_ack.rcv_mss = TCP_MIN_MSS;
 	memset(&tp->rx_opt, 0, sizeof(tp->rx_opt));
 	__sk_dst_reset(sk);
-	dst_release(sk->sk_rx_dst);
-	sk->sk_rx_dst = NULL;
+	dst_release(xchg((__force struct dst_entry **)&sk->sk_rx_dst, NULL));
 	tcp_saved_syn_free(tp);
 	tp->compressed_ack = 0;
 	tp->segs_in = 0;
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 246ab7b5e857..0ce46849ec3d 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -5787,7 +5787,7 @@ void tcp_rcv_established(struct sock *sk, struct sk_buff *skb)
 	trace_tcp_probe(sk, skb);
 
 	tcp_mstamp_refresh(tp);
-	if (unlikely(!sk->sk_rx_dst))
+	if (unlikely(!rcu_access_pointer(sk->sk_rx_dst)))
 		inet_csk(sk)->icsk_af_ops->sk_rx_dst_set(sk, skb);
 	/*
 	 *	Header prediction.
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 13d868c43284..084df223b5df 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1701,7 +1701,10 @@ int tcp_v4_do_rcv(struct sock *sk, struct sk_buff *skb)
 	struct sock *rsk;
 
 	if (sk->sk_state == TCP_ESTABLISHED) { /* Fast path */
-		struct dst_entry *dst = sk->sk_rx_dst;
+		struct dst_entry *dst;
+
+		dst = rcu_dereference_protected(sk->sk_rx_dst,
+						lockdep_sock_is_held(sk));
 
 		sock_rps_save_rxhash(sk, skb);
 		sk_mark_napi_id(sk, skb);
@@ -1709,8 +1712,8 @@ int tcp_v4_do_rcv(struct sock *sk, struct sk_buff *skb)
 			if (sk->sk_rx_dst_ifindex != skb->skb_iif ||
 			    !INDIRECT_CALL_1(dst->ops->check, ipv4_dst_check,
 					     dst, 0)) {
+				RCU_INIT_POINTER(sk->sk_rx_dst, NULL);
 				dst_release(dst);
-				sk->sk_rx_dst = NULL;
 			}
 		}
 		tcp_rcv_established(sk, skb);
@@ -1786,7 +1789,7 @@ int tcp_v4_early_demux(struct sk_buff *skb)
 		skb->sk = sk;
 		skb->destructor = sock_edemux;
 		if (sk_fullsock(sk)) {
-			struct dst_entry *dst = READ_ONCE(sk->sk_rx_dst);
+			struct dst_entry *dst = rcu_dereference(sk->sk_rx_dst);
 
 			if (dst)
 				dst = dst_check(dst, 0);
@@ -2201,7 +2204,7 @@ void inet_sk_rx_dst_set(struct sock *sk, const struct sk_buff *skb)
 	struct dst_entry *dst = skb_dst(skb);
 
 	if (dst && dst_hold_safe(dst)) {
-		sk->sk_rx_dst = dst;
+		rcu_assign_pointer(sk->sk_rx_dst, dst);
 		sk->sk_rx_dst_ifindex = skb->skb_iif;
 	}
 }
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 23b05e28490b..15c6b450b8db 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -2250,7 +2250,7 @@ bool udp_sk_rx_dst_set(struct sock *sk, struct dst_entry *dst)
 	struct dst_entry *old;
 
 	if (dst_hold_safe(dst)) {
-		old = xchg(&sk->sk_rx_dst, dst);
+		old = xchg((__force struct dst_entry **)&sk->sk_rx_dst, dst);
 		dst_release(old);
 		return old != dst;
 	}
@@ -2440,7 +2440,7 @@ int __udp4_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,
 		struct dst_entry *dst = skb_dst(skb);
 		int ret;
 
-		if (unlikely(sk->sk_rx_dst != dst))
+		if (unlikely(rcu_dereference(sk->sk_rx_dst) != dst))
 			udp_sk_rx_dst_set(sk, dst);
 
 		ret = udp_unicast_rcv_skb(sk, skb, uh);
@@ -2599,7 +2599,7 @@ int udp_v4_early_demux(struct sk_buff *skb)
 
 	skb->sk = sk;
 	skb->destructor = sock_efree;
-	dst = READ_ONCE(sk->sk_rx_dst);
+	dst = rcu_dereference(sk->sk_rx_dst);
 
 	if (dst)
 		dst = dst_check(dst, 0);