tcp: fix fastopen races vs lockless listener
There are multiple races that need fixes :
1) skb_get() + queue skb + kfree_skb() is racy
An accept() can be done on another cpu, data consumed immediately.
tcp_recvmsg() uses __kfree_skb() as it is assumed all skb found in
socket receive queue are private.
Then the kfree_skb() in tcp_rcv_state_process() uses an already freed skb
2) tcp_reqsk_record_syn() needs to be done before tcp_try_fastopen()
for the same reasons.
3) We want to send the SYNACK before queueing child into accept queue,
otherwise we might reintroduce the ooo issue fixed in
commit 7c85af881044 ("tcp: avoid reorders for TFO passive connections")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 2710875..a95c8eb 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -6229,12 +6229,16 @@
tcp_rsk(req)->txhash = net_tx_rndhash();
tcp_openreq_init_rwin(req, sk, dst);
if (!want_cookie) {
- fastopen_sk = tcp_try_fastopen(sk, skb, req, &foc, dst);
tcp_reqsk_record_syn(sk, req, skb);
+ fastopen_sk = tcp_try_fastopen(sk, skb, req, &foc, dst);
}
if (fastopen_sk) {
af_ops->send_synack(fastopen_sk, dst, &fl, req,
skb_get_queue_mapping(skb), &foc, false);
+ /* Add the child socket directly into the accept queue */
+ inet_csk_reqsk_queue_add(sk, req, fastopen_sk);
+ sk->sk_data_ready(sk);
+ bh_unlock_sock(fastopen_sk);
sock_put(fastopen_sk);
} else {
tcp_rsk(req)->tfo_listener = false;