From: Eric Dumazet edumazet@google.com
[ Upstream commit 7db48e983930285b765743ebd665aecf9850582b ]
There are few places where we fetch tp->copied_seq while this field can change from IRQ or other cpu.
We need to add READ_ONCE() annotations, and also make sure write sides use corresponding WRITE_ONCE() to avoid store-tearing.
Note that tcp_inq_hint() was already using READ_ONCE(tp->copied_seq)
Signed-off-by: Eric Dumazet edumazet@google.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- net/ipv4/tcp.c | 18 +++++++++--------- net/ipv4/tcp_diag.c | 3 ++- net/ipv4/tcp_input.c | 6 +++--- net/ipv4/tcp_ipv4.c | 2 +- net/ipv4/tcp_minisocks.c | 2 +- net/ipv4/tcp_output.c | 2 +- net/ipv6/tcp_ipv6.c | 2 +- 7 files changed, 18 insertions(+), 17 deletions(-)
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 98e8ee8bb7593..f639c7d608382 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -567,7 +567,7 @@ __poll_t tcp_poll(struct file *file, struct socket *sock, poll_table *wait) (state != TCP_SYN_RECV || tp->fastopen_rsk)) { int target = sock_rcvlowat(sk, 0, INT_MAX);
- if (tp->urg_seq == tp->copied_seq && + if (tp->urg_seq == READ_ONCE(tp->copied_seq) && !sock_flag(sk, SOCK_URGINLINE) && tp->urg_data) target++; @@ -628,7 +628,7 @@ int tcp_ioctl(struct sock *sk, int cmd, unsigned long arg) unlock_sock_fast(sk, slow); break; case SIOCATMARK: - answ = tp->urg_data && tp->urg_seq == tp->copied_seq; + answ = tp->urg_data && tp->urg_seq == READ_ONCE(tp->copied_seq); break; case SIOCOUTQ: if (sk->sk_state == TCP_LISTEN) @@ -1696,9 +1696,9 @@ int tcp_read_sock(struct sock *sk, read_descriptor_t *desc, sk_eat_skb(sk, skb); if (!desc->count) break; - tp->copied_seq = seq; + WRITE_ONCE(tp->copied_seq, seq); } - tp->copied_seq = seq; + WRITE_ONCE(tp->copied_seq, seq);
tcp_rcv_space_adjust(sk);
@@ -1835,7 +1835,7 @@ static int tcp_zerocopy_receive(struct sock *sk, out: up_read(¤t->mm->mmap_sem); if (length) { - tp->copied_seq = seq; + WRITE_ONCE(tp->copied_seq, seq); tcp_rcv_space_adjust(sk);
/* Clean up data we have read: This will do ACK frames. */ @@ -2112,7 +2112,7 @@ int tcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int nonblock, if (urg_offset < used) { if (!urg_offset) { if (!sock_flag(sk, SOCK_URGINLINE)) { - ++*seq; + WRITE_ONCE(*seq, *seq + 1); urg_hole++; offset++; used--; @@ -2134,7 +2134,7 @@ int tcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int nonblock, } }
- *seq += used; + WRITE_ONCE(*seq, *seq + used); copied += used; len -= used;
@@ -2163,7 +2163,7 @@ int tcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int nonblock,
found_fin_ok: /* Process the FIN. */ - ++*seq; + WRITE_ONCE(*seq, *seq + 1); if (!(flags & MSG_PEEK)) sk_eat_skb(sk, skb); break; @@ -2578,7 +2578,7 @@ int tcp_disconnect(struct sock *sk, int flags)
tcp_clear_xmit_timers(sk); __skb_queue_purge(&sk->sk_receive_queue); - tp->copied_seq = tp->rcv_nxt; + WRITE_ONCE(tp->copied_seq, tp->rcv_nxt); tp->urg_data = 0; tcp_write_queue_purge(sk); tcp_fastopen_active_disable_ofo_check(sk); diff --git a/net/ipv4/tcp_diag.c b/net/ipv4/tcp_diag.c index c9e97f304f984..a96b252c742cb 100644 --- a/net/ipv4/tcp_diag.c +++ b/net/ipv4/tcp_diag.c @@ -30,7 +30,8 @@ static void tcp_diag_get_info(struct sock *sk, struct inet_diag_msg *r, } else if (sk->sk_type == SOCK_STREAM) { const struct tcp_sock *tp = tcp_sk(sk);
- r->idiag_rqueue = max_t(int, READ_ONCE(tp->rcv_nxt) - tp->copied_seq, 0); + r->idiag_rqueue = max_t(int, READ_ONCE(tp->rcv_nxt) - + READ_ONCE(tp->copied_seq), 0); r->idiag_wqueue = tp->write_seq - tp->snd_una; } if (info) diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index c7c3df30d8bc1..effba01eed84e 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -5897,7 +5897,7 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb, /* Remember, tcp_poll() does not lock socket! * Change state from SYN-SENT only after copied_seq * is initialized. */ - tp->copied_seq = tp->rcv_nxt; + WRITE_ONCE(tp->copied_seq, tp->rcv_nxt);
smc_check_reset_syn(tp);
@@ -5972,7 +5972,7 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb, }
WRITE_ONCE(tp->rcv_nxt, TCP_SKB_CB(skb)->seq + 1); - tp->copied_seq = tp->rcv_nxt; + WRITE_ONCE(tp->copied_seq, tp->rcv_nxt); tp->rcv_wup = TCP_SKB_CB(skb)->seq + 1;
/* RFC1323: The window in SYN & SYN/ACK segments is @@ -6134,7 +6134,7 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb) tcp_rearm_rto(sk); } else { tcp_init_transfer(sk, BPF_SOCK_OPS_PASSIVE_ESTABLISHED_CB); - tp->copied_seq = tp->rcv_nxt; + WRITE_ONCE(tp->copied_seq, tp->rcv_nxt); } smp_mb(); tcp_set_state(sk, TCP_ESTABLISHED); diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 229ae272b2400..f96720e31ef63 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2431,7 +2431,7 @@ static void get_tcp4_sock(struct sock *sk, struct seq_file *f, int i) * we might find a transient negative value. */ rx_queue = max_t(int, READ_ONCE(tp->rcv_nxt) - - tp->copied_seq, 0); + READ_ONCE(tp->copied_seq), 0);
seq_printf(f, "%4d: %08X:%04X %08X:%04X %02X %08X:%08X %02X:%08lX " "%08X %5u %8d %lu %d %pK %lu %lu %u %u %d", diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 7ba8a90772b0a..0b1a04fa54392 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -470,7 +470,7 @@ struct sock *tcp_create_openreq_child(const struct sock *sk,
seq = treq->rcv_isn + 1; newtp->rcv_wup = seq; - newtp->copied_seq = seq; + WRITE_ONCE(newtp->copied_seq, seq); WRITE_ONCE(newtp->rcv_nxt, seq); newtp->segs_in = 1;
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 3cfefec819758..662aa48173b81 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -3381,7 +3381,7 @@ static void tcp_connect_init(struct sock *sk) else tp->rcv_tstamp = tcp_jiffies32; tp->rcv_wup = tp->rcv_nxt; - tp->copied_seq = tp->rcv_nxt; + WRITE_ONCE(tp->copied_seq, tp->rcv_nxt);
inet_csk(sk)->icsk_rto = tcp_timeout_init(sk); inet_csk(sk)->icsk_retransmits = 0; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 2e76ebfdc907d..de9b9c0bf18f7 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1839,7 +1839,7 @@ static void get_tcp6_sock(struct seq_file *seq, struct sock *sp, int i) * we might find a transient negative value. */ rx_queue = max_t(int, READ_ONCE(tp->rcv_nxt) - - tp->copied_seq, 0); + READ_ONCE(tp->copied_seq), 0);
seq_printf(seq, "%4d: %08X%08X%08X%08X:%04X %08X%08X%08X%08X:%04X "