* [PATCH 6.12.y] net: annotate data-races around sk->sk_{data_ready,write_space}
@ 2026-04-13 9:40 Leon Chen
0 siblings, 0 replies; only message in thread
From: Leon Chen @ 2026-04-13 9:40 UTC (permalink / raw)
To: stable, edumazet, daniel, john.fastabend, jakub,
willemdebruijn.kernel, kuniyu, kuba
From: Eric Dumazet <edumazet@google.com>
[ Upstream commit 2ef2b20cf4e04ac8a6ba68493f8780776ff84300 ]
skmsg (and probably other layers) are changing these pointers
while other cpus might read them concurrently.
Add corresponding READ_ONCE()/WRITE_ONCE() annotations
for UDP, TCP and AF_UNIX.
Fixes: 604326b41a6f ("bpf, sockmap: convert to generic sk_msg interface")
Reported-by: syzbot+87f770387a9e5dc6b79b@syzkaller.appspotmail.com
Closes: https://lore.kernel.org/netdev/699ee9fc.050a0220.1cd54b.0009.GAE@google.com/
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: John Fastabend <john.fastabend@gmail.com>
Cc: Jakub Sitnicki <jakub@cloudflare.com>
Cc: Willem de Bruijn <willemdebruijn.kernel@gmail.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@google.com>
Link: https://patch.msgid.link/20260225131547.1085509-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Leon Chen <leonchen.oss@139.com>
---
net/core/skmsg.c | 14 +++++++-------
net/ipv4/tcp.c | 4 ++--
net/ipv4/tcp_bpf.c | 2 +-
net/ipv4/tcp_input.c | 14 ++++++++------
net/ipv4/tcp_minisocks.c | 2 +-
net/ipv4/udp.c | 3 ++-
net/ipv4/udp_bpf.c | 2 +-
net/unix/af_unix.c | 8 ++++----
8 files changed, 26 insertions(+), 23 deletions(-)
diff --git a/net/core/skmsg.c b/net/core/skmsg.c
index 2e89087b95c2..e1e0283e53c1 100644
--- a/net/core/skmsg.c
+++ b/net/core/skmsg.c
@@ -1204,8 +1204,8 @@ void sk_psock_start_strp(struct sock *sk, struct sk_psock *psock)
return;
psock->saved_data_ready = sk->sk_data_ready;
- sk->sk_data_ready = sk_psock_strp_data_ready;
- sk->sk_write_space = sk_psock_write_space;
+ WRITE_ONCE(sk->sk_data_ready, sk_psock_strp_data_ready);
+ WRITE_ONCE(sk->sk_write_space, sk_psock_write_space);
}
void sk_psock_stop_strp(struct sock *sk, struct sk_psock *psock)
@@ -1215,8 +1215,8 @@ void sk_psock_stop_strp(struct sock *sk, struct sk_psock *psock)
if (!psock->saved_data_ready)
return;
- sk->sk_data_ready = psock->saved_data_ready;
- psock->saved_data_ready = NULL;
+ WRITE_ONCE(sk->sk_data_ready, psock->saved_data_ready);
+ WRITE_ONCE(psock->saved_data_ready, NULL);
strp_stop(&psock->strp);
}
@@ -1298,8 +1298,8 @@ void sk_psock_start_verdict(struct sock *sk, struct sk_psock *psock)
return;
psock->saved_data_ready = sk->sk_data_ready;
- sk->sk_data_ready = sk_psock_verdict_data_ready;
- sk->sk_write_space = sk_psock_write_space;
+ WRITE_ONCE(sk->sk_data_ready, sk_psock_verdict_data_ready);
+ WRITE_ONCE(sk->sk_write_space, sk_psock_write_space);
}
void sk_psock_stop_verdict(struct sock *sk, struct sk_psock *psock)
@@ -1310,6 +1310,6 @@ void sk_psock_stop_verdict(struct sock *sk, struct sk_psock *psock)
if (!psock->saved_data_ready)
return;
- sk->sk_data_ready = psock->saved_data_ready;
+ WRITE_ONCE(sk->sk_data_ready, psock->saved_data_ready);
psock->saved_data_ready = NULL;
}
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index c532803b9957..9c5fc4464783 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -1346,7 +1346,7 @@ int tcp_sendmsg_locked(struct sock *sk, struct msghdr *msg, size_t size)
err = sk_stream_error(sk, flags, err);
/* make sure we wake any epoll edge trigger waiter */
if (unlikely(tcp_rtx_and_write_queues_empty(sk) && err == -EAGAIN)) {
- sk->sk_write_space(sk);
+ READ_ONCE(sk->sk_write_space)(sk);
tcp_chrono_stop(sk, TCP_CHRONO_SNDBUF_LIMITED);
}
return err;
@@ -4023,7 +4023,7 @@ int do_tcp_setsockopt(struct sock *sk, int level, int optname,
break;
case TCP_NOTSENT_LOWAT:
WRITE_ONCE(tp->notsent_lowat, val);
- sk->sk_write_space(sk);
+ READ_ONCE(sk->sk_write_space)(sk);
break;
case TCP_INQ:
if (val > 1 || val < 0)
diff --git a/net/ipv4/tcp_bpf.c b/net/ipv4/tcp_bpf.c
index f5817438f173..b9a58fde9c31 100644
--- a/net/ipv4/tcp_bpf.c
+++ b/net/ipv4/tcp_bpf.c
@@ -725,7 +725,7 @@ int tcp_bpf_update_proto(struct sock *sk, struct sk_psock *psock, bool restore)
WRITE_ONCE(sk->sk_prot->unhash, psock->saved_unhash);
tcp_update_ulp(sk, psock->sk_proto, psock->saved_write_space);
} else {
- sk->sk_write_space = psock->saved_write_space;
+ WRITE_ONCE(sk->sk_write_space, psock->saved_write_space);
/* Pairs with lockless read in sk_clone_lock() */
sock_replace_proto(sk, psock->sk_proto);
}
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 1d9e93a04930..c498588c021d 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -5034,7 +5034,7 @@ static void tcp_data_queue_ofo(struct sock *sk, struct sk_buff *skb)
if (unlikely(tcp_try_rmem_schedule(sk, skb, skb->truesize))) {
NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPOFODROP);
- sk->sk_data_ready(sk);
+ READ_ONCE(sk->sk_data_ready)(sk);
tcp_drop_reason(sk, skb, SKB_DROP_REASON_PROTO_MEM);
return;
}
@@ -5241,7 +5241,7 @@ int tcp_send_rcvq(struct sock *sk, struct msghdr *msg, size_t size)
void tcp_data_ready(struct sock *sk)
{
if (tcp_epollin_ready(sk, sk->sk_rcvlowat) || sock_flag(sk, SOCK_DONE))
- sk->sk_data_ready(sk);
+ READ_ONCE(sk->sk_data_ready)(sk);
}
static void tcp_data_queue(struct sock *sk, struct sk_buff *skb)
@@ -5297,7 +5297,7 @@ static void tcp_data_queue(struct sock *sk, struct sk_buff *skb)
inet_csk(sk)->icsk_ack.pending |=
(ICSK_ACK_NOMEM | ICSK_ACK_NOW);
inet_csk_schedule_ack(sk);
- sk->sk_data_ready(sk);
+ READ_ONCE(sk->sk_data_ready)(sk);
if (skb_queue_len(&sk->sk_receive_queue) && skb->len) {
reason = SKB_DROP_REASON_PROTO_MEM;
@@ -5735,7 +5735,9 @@ static void tcp_new_space(struct sock *sk)
tp->snd_cwnd_stamp = tcp_jiffies32;
}
- INDIRECT_CALL_1(sk->sk_write_space, sk_stream_write_space, sk);
+ INDIRECT_CALL_1(READ_ONCE(sk->sk_write_space),
+ sk_stream_write_space,
+ sk);
}
/* Caller made space either from:
@@ -5941,7 +5943,7 @@ static void tcp_urg(struct sock *sk, struct sk_buff *skb, const struct tcphdr *t
BUG();
WRITE_ONCE(tp->urg_data, TCP_URG_VALID | tmp);
if (!sock_flag(sk, SOCK_DEAD))
- sk->sk_data_ready(sk);
+ READ_ONCE(sk->sk_data_ready)(sk);
}
}
}
@@ -7341,7 +7343,7 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops,
sock_put(fastopen_sk);
goto drop_and_free;
}
- sk->sk_data_ready(sk);
+ READ_ONCE(sk->sk_data_ready)(sk);
bh_unlock_sock(fastopen_sk);
sock_put(fastopen_sk);
} else {
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index f3e4fc957219..adddfb7d934c 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -928,7 +928,7 @@ enum skb_drop_reason tcp_child_process(struct sock *parent, struct sock *child,
reason = tcp_rcv_state_process(child, skb);
/* Wakeup parent, send SIGIO */
if (state == TCP_SYN_RECV && child->sk_state != state)
- parent->sk_data_ready(parent);
+ READ_ONCE(parent->sk_data_ready)(parent);
} else {
/* Alas, it is possible again, because we do lookup
* in main socket hash table and lock on listening
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index d1090a1f860c..27694334e410 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -1622,7 +1622,8 @@ int __udp_enqueue_schedule_skb(struct sock *sk, struct sk_buff *skb)
spin_unlock(&list->lock);
if (!sock_flag(sk, SOCK_DEAD))
- INDIRECT_CALL_1(sk->sk_data_ready, sock_def_readable, sk);
+ INDIRECT_CALL_1(READ_ONCE(sk->sk_data_ready),
+ sock_def_readable, sk);
busylock_release(busy);
return 0;
diff --git a/net/ipv4/udp_bpf.c b/net/ipv4/udp_bpf.c
index 91233e37cd97..779a3a03762f 100644
--- a/net/ipv4/udp_bpf.c
+++ b/net/ipv4/udp_bpf.c
@@ -158,7 +158,7 @@ int udp_bpf_update_proto(struct sock *sk, struct sk_psock *psock, bool restore)
int family = sk->sk_family == AF_INET ? UDP_BPF_IPV4 : UDP_BPF_IPV6;
if (restore) {
- sk->sk_write_space = psock->saved_write_space;
+ WRITE_ONCE(sk->sk_write_space, psock->saved_write_space);
sock_replace_proto(sk, psock->sk_proto);
return 0;
}
diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index 59911ac719b1..aa6395fd58bb 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -1707,7 +1707,7 @@ static int unix_stream_connect(struct socket *sock, struct sockaddr *uaddr,
__skb_queue_tail(&other->sk_receive_queue, skb);
spin_unlock(&other->sk_receive_queue.lock);
unix_state_unlock(other);
- other->sk_data_ready(other);
+ READ_ONCE(other->sk_data_ready)(other);
sock_put(other);
return 0;
@@ -2175,7 +2175,7 @@ static int unix_dgram_sendmsg(struct socket *sock, struct msghdr *msg,
scm_stat_add(other, skb);
skb_queue_tail(&other->sk_receive_queue, skb);
unix_state_unlock(other);
- other->sk_data_ready(other);
+ READ_ONCE(other->sk_data_ready)(other);
sock_put(other);
scm_destroy(&scm);
return len;
@@ -2243,7 +2243,7 @@ static int queue_oob(struct socket *sock, struct msghdr *msg, struct sock *other
sk_send_sigurg(other);
unix_state_unlock(other);
- other->sk_data_ready(other);
+ READ_ONCE(other->sk_data_ready)(other);
return err;
}
@@ -2354,7 +2354,7 @@ static int unix_stream_sendmsg(struct socket *sock, struct msghdr *msg,
scm_stat_add(other, skb);
skb_queue_tail(&other->sk_receive_queue, skb);
unix_state_unlock(other);
- other->sk_data_ready(other);
+ READ_ONCE(other->sk_data_ready)(other);
sent += size;
}
--
2.35.3
^ permalink raw reply related [flat|nested] only message in thread
only message in thread, other threads:[~2026-04-13 9:40 UTC | newest]
Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-04-13 9:40 [PATCH 6.12.y] net: annotate data-races around sk->sk_{data_ready,write_space} Leon Chen
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox