netdev - [PATCH net-next 1/6] tcp: set TCP

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20230804144616.3938718-2-edumazet@google.com>
Date: Fri,  4 Aug 2023 14:46:11 +0000
From: Eric Dumazet <edumazet@...gle.com>
To: "David S . Miller" <davem@...emloft.net>, Jakub Kicinski <kuba@...nel.org>, 
	Paolo Abeni <pabeni@...hat.com>
Cc: netdev@...r.kernel.org, eric.dumazet@...il.com, 
	Soheil Hassas Yeganeh <soheil@...gle.com>, Eric Dumazet <edumazet@...gle.com>
Subject: [PATCH net-next 1/6] tcp: set TCP_SYNCNT locklessly

icsk->icsk_syn_retries can safely be set without locking the socket.

We have to add READ_ONCE() annotations in tcp_fastopen_synack_timer()
and tcp_write_timeout().

Signed-off-by: Eric Dumazet <edumazet@...gle.com>
---
 net/ipv4/tcp.c       | 15 ++++++---------
 net/ipv4/tcp_timer.c |  9 ++++++---
 2 files changed, 12 insertions(+), 12 deletions(-)

diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index aca5620cf3ba20be38d81b1b526c22623b145ff7..bcbb33a8c152abe2a060abd644689b54bcca1daa 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -3291,9 +3291,7 @@ int tcp_sock_set_syncnt(struct sock *sk, int val)
 	if (val < 1 || val > MAX_TCP_SYNCNT)
 		return -EINVAL;
 
-	lock_sock(sk);
 	WRITE_ONCE(inet_csk(sk)->icsk_syn_retries, val);
-	release_sock(sk);
 	return 0;
 }
 EXPORT_SYMBOL(tcp_sock_set_syncnt);
@@ -3462,6 +3460,12 @@ int do_tcp_setsockopt(struct sock *sk, int level, int optname,
 	if (copy_from_sockptr(&val, optval, sizeof(val)))
 		return -EFAULT;
 
+	/* Handle options that can be set without locking the socket. */
+	switch (optname) {
+	case TCP_SYNCNT:
+		return tcp_sock_set_syncnt(sk, val);
+	}
+
 	sockopt_lock_sock(sk);
 
 	switch (optname) {
@@ -3569,13 +3573,6 @@ int do_tcp_setsockopt(struct sock *sk, int level, int optname,
 		else
 			WRITE_ONCE(tp->keepalive_probes, val);
 		break;
-	case TCP_SYNCNT:
-		if (val < 1 || val > MAX_TCP_SYNCNT)
-			err = -EINVAL;
-		else
-			WRITE_ONCE(icsk->icsk_syn_retries, val);
-		break;
-
 	case TCP_SAVE_SYN:
 		/* 0: disable, 1: enable, 2: start from ether_header */
 		if (val < 0 || val > 2)
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
index 470f581eedd438b3bbd6ae4973c7a6f01ee1724f..66040ab457d46ffa2fac62f875b636f567157793 100644
--- a/net/ipv4/tcp_timer.c
+++ b/net/ipv4/tcp_timer.c
@@ -239,7 +239,8 @@ static int tcp_write_timeout(struct sock *sk)
 	if ((1 << sk->sk_state) & (TCPF_SYN_SENT | TCPF_SYN_RECV)) {
 		if (icsk->icsk_retransmits)
 			__dst_negative_advice(sk);
-		retry_until = icsk->icsk_syn_retries ? :
+		/* Paired with WRITE_ONCE() in tcp_sock_set_syncnt() */
+		retry_until = READ_ONCE(icsk->icsk_syn_retries) ? :
 			READ_ONCE(net->ipv4.sysctl_tcp_syn_retries);
 
 		max_retransmits = retry_until;
@@ -421,8 +422,10 @@ static void tcp_fastopen_synack_timer(struct sock *sk, struct request_sock *req)
 
 	req->rsk_ops->syn_ack_timeout(req);
 
-	/* add one more retry for fastopen */
-	max_retries = icsk->icsk_syn_retries ? :
+	/* Add one more retry for fastopen.
+	 * Paired with WRITE_ONCE() in tcp_sock_set_syncnt()
+	 */
+	max_retries = READ_ONCE(icsk->icsk_syn_retries) ? :
 		READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_synack_retries) + 1;
 
 	if (req->num_timeout >= max_retries) {
-- 
2.41.0.640.ga95def55d0-goog