net: set SOCK_RCU_FREE before inserting socket into hashtable

[ Upstream commit 871019b ]

We've started to see the following kernel traces:

 WARNING: CPU: 83 PID: 0 at net/core/filter.c:6641 sk_lookup+0x1bd/0x1d0

 Call Trace:
  <IRQ>
  __bpf_skc_lookup+0x10d/0x120
  bpf_sk_lookup+0x48/0xd0
  bpf_sk_lookup_tcp+0x19/0x20
  bpf_prog_<redacted>+0x37c/0x16a3
  cls_bpf_classify+0x205/0x2e0
  tcf_classify+0x92/0x160
  __netif_receive_skb_core+0xe52/0xf10
  __netif_receive_skb_list_core+0x96/0x2b0
  napi_complete_done+0x7b5/0xb70
  <redacted>_poll+0x94/0xb0
  net_rx_action+0x163/0x1d70
  __do_softirq+0xdc/0x32e
  asm_call_irq_on_stack+0x12/0x20
  </IRQ>
  do_softirq_own_stack+0x36/0x50
  do_softirq+0x44/0x70

__inet_hash can race with lockless (rcu) readers on the other cpus:

  __inet_hash
    __sk_nulls_add_node_rcu
    <- (bpf triggers here)
    sock_set_flag(SOCK_RCU_FREE)

Let's move the SOCK_RCU_FREE part up a bit, before we are inserting
the socket into hashtables. Note, that the race is really harmless;
the bpf callers are handling this situation (where listener socket
doesn't have SOCK_RCU_FREE set) correctly, so the only
annoyance is a WARN_ONCE.

More details from Eric regarding SOCK_RCU_FREE timeline:

Commit 3b24d85 ("tcp/dccp: do not touch listener sk_refcnt under
synflood") added SOCK_RCU_FREE. At that time, the precise location of
sock_set_flag(sk, SOCK_RCU_FREE) did not matter, because the thread calling
__inet_hash() owns a reference on sk. SOCK_RCU_FREE was only tested
at dismantle time.

Commit 6acc9b4 ("bpf: Add helper to retrieve socket in BPF")
started checking SOCK_RCU_FREE _after_ the lookup to infer whether
the refcount has been taken care of.

Fixes: 6acc9b4 ("bpf: Add helper to retrieve socket in BPF")
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>

net/ipv4/inet_hashtables.c

@@ -731,12 +731,12 @@ int __inet_hash(struct sock *sk, struct sock *osk)
 		if (err)
 			goto unlock;
 	}
+	sock_set_flag(sk, SOCK_RCU_FREE);
 	if (IS_ENABLED(CONFIG_IPV6) && sk->sk_reuseport &&
 		sk->sk_family == AF_INET6)
 		__sk_nulls_add_node_tail_rcu(sk, &ilb2->nulls_head);
 	else
 		__sk_nulls_add_node_rcu(sk, &ilb2->nulls_head);
-	sock_set_flag(sk, SOCK_RCU_FREE);
 	sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1);
 unlock:
 	spin_unlock(&ilb2->lock);