net: Fix data-races around sysctl_[rw]mem(_offset)?.

[ Upstream commit 02739545951ad4c1215160db7fbf9b7a918d3c0b ]

While reading these sysctl variables, they can be changed concurrently.
Thus, we need to add READ_ONCE() to their readers.

  - .sysctl_rmem
  - .sysctl_rwmem
  - .sysctl_rmem_offset
  - .sysctl_wmem_offset
  - sysctl_tcp_rmem[1, 2]
  - sysctl_tcp_wmem[1, 2]
  - sysctl_decnet_rmem[1]
  - sysctl_decnet_wmem[1]
  - sysctl_tipc_rmem[1]

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>

diff --git a/include/net/sock.h b/include/net/sock.h
index 96f51d4..819c539 100644 (file)
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -2765,18 +2765,18 @@ static inline int sk_get_wmem0(const struct sock *sk, const struct proto *proto)
 {
        /* Does this proto have per netns sysctl_wmem ? */
        if (proto->sysctl_wmem_offset)
-               return *(int *)((void *)sock_net(sk) + proto->sysctl_wmem_offset);
+               return READ_ONCE(*(int *)((void *)sock_net(sk) + proto->sysctl_wmem_offset));
 
-       return *proto->sysctl_wmem;
+       return READ_ONCE(*proto->sysctl_wmem);
 }
 
 static inline int sk_get_rmem0(const struct sock *sk, const struct proto *proto)
 {
        /* Does this proto have per netns sysctl_rmem ? */
        if (proto->sysctl_rmem_offset)
-               return *(int *)((void *)sock_net(sk) + proto->sysctl_rmem_offset);
+               return READ_ONCE(*(int *)((void *)sock_net(sk) + proto->sysctl_rmem_offset));
 
-       return *proto->sysctl_rmem;
+       return READ_ONCE(*proto->sysctl_rmem);
 }
 
 /* Default TCP Small queue budget is ~1 ms of data (1sec >> 10)

diff --git a/net/decnet/af_decnet.c b/net/decnet/af_decnet.c
index dc92a67..7d542eb 100644 (file)
--- a/net/decnet/af_decnet.c
+++ b/net/decnet/af_decnet.c
@@ -480,8 +480,8 @@ static struct sock *dn_alloc_sock(struct net *net, struct socket *sock, gfp_t gf
        sk->sk_family      = PF_DECnet;
        sk->sk_protocol    = 0;
        sk->sk_allocation  = gfp;
-       sk->sk_sndbuf      = sysctl_decnet_wmem[1];
-       sk->sk_rcvbuf      = sysctl_decnet_rmem[1];
+       sk->sk_sndbuf      = READ_ONCE(sysctl_decnet_wmem[1]);
+       sk->sk_rcvbuf      = READ_ONCE(sysctl_decnet_rmem[1]);
 
        /* Initialization of DECnet Session Control Port                */
        scp = DN_SK(sk);

diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 7ba9059..2097eea 100644 (file)
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -458,8 +458,8 @@ void tcp_init_sock(struct sock *sk)
 
        icsk->icsk_sync_mss = tcp_sync_mss;
 
-       WRITE_ONCE(sk->sk_sndbuf, sock_net(sk)->ipv4.sysctl_tcp_wmem[1]);
-       WRITE_ONCE(sk->sk_rcvbuf, sock_net(sk)->ipv4.sysctl_tcp_rmem[1]);
+       WRITE_ONCE(sk->sk_sndbuf, READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_wmem[1]));
+       WRITE_ONCE(sk->sk_rcvbuf, READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_rmem[1]));
 
        sk_sockets_allocated_inc(sk);
        sk->sk_route_forced_caps = NETIF_F_GSO;
@@ -1722,7 +1722,7 @@ int tcp_set_rcvlowat(struct sock *sk, int val)
        if (sk->sk_userlocks & SOCK_RCVBUF_LOCK)
                cap = sk->sk_rcvbuf >> 1;
        else
-               cap = sock_net(sk)->ipv4.sysctl_tcp_rmem[2] >> 1;
+               cap = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_rmem[2]) >> 1;
        val = min(val, cap);
        WRITE_ONCE(sk->sk_rcvlowat, val ? : 1);
 

diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 566745f..e007bdc 100644 (file)
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -426,7 +426,7 @@ static void tcp_sndbuf_expand(struct sock *sk)
 
        if (sk->sk_sndbuf < sndmem)
                WRITE_ONCE(sk->sk_sndbuf,
-                          min(sndmem, sock_net(sk)->ipv4.sysctl_tcp_wmem[2]));
+                          min(sndmem, READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_wmem[2])));
 }
 
 /* 2. Tuning advertised window (window_clamp, rcv_ssthresh)
@@ -461,7 +461,7 @@ static int __tcp_grow_window(const struct sock *sk, const struct sk_buff *skb,
        struct tcp_sock *tp = tcp_sk(sk);
        /* Optimize this! */
        int truesize = tcp_win_from_space(sk, skbtruesize) >> 1;
-       int window = tcp_win_from_space(sk, sock_net(sk)->ipv4.sysctl_tcp_rmem[2]) >> 1;
+       int window = tcp_win_from_space(sk, READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_rmem[2])) >> 1;
 
        while (tp->rcv_ssthresh <= window) {
                if (truesize <= skb->len)
@@ -566,16 +566,17 @@ static void tcp_clamp_window(struct sock *sk)
        struct tcp_sock *tp = tcp_sk(sk);
        struct inet_connection_sock *icsk = inet_csk(sk);
        struct net *net = sock_net(sk);
+       int rmem2;
 
        icsk->icsk_ack.quick = 0;
+       rmem2 = READ_ONCE(net->ipv4.sysctl_tcp_rmem[2]);
 
-       if (sk->sk_rcvbuf < net->ipv4.sysctl_tcp_rmem[2] &&
+       if (sk->sk_rcvbuf < rmem2 &&
            !(sk->sk_userlocks & SOCK_RCVBUF_LOCK) &&
            !tcp_under_memory_pressure(sk) &&
            sk_memory_allocated(sk) < sk_prot_mem_limits(sk, 0)) {
                WRITE_ONCE(sk->sk_rcvbuf,
-                          min(atomic_read(&sk->sk_rmem_alloc),
-                              net->ipv4.sysctl_tcp_rmem[2]));
+                          min(atomic_read(&sk->sk_rmem_alloc), rmem2));
        }
        if (atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf)
                tp->rcv_ssthresh = min(tp->window_clamp, 2U * tp->advmss);
@@ -737,7 +738,7 @@ void tcp_rcv_space_adjust(struct sock *sk)
 
                do_div(rcvwin, tp->advmss);
                rcvbuf = min_t(u64, rcvwin * rcvmem,
-                              sock_net(sk)->ipv4.sysctl_tcp_rmem[2]);
+                              READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_rmem[2]));
                if (rcvbuf > sk->sk_rcvbuf) {
                        WRITE_ONCE(sk->sk_rcvbuf, rcvbuf);
 

diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 51f3131..9c9a0f7 100644 (file)
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -238,7 +238,7 @@ void tcp_select_initial_window(const struct sock *sk, int __space, __u32 mss,
        *rcv_wscale = 0;
        if (wscale_ok) {
                /* Set window scaling on max possible window */
-               space = max_t(u32, space, sock_net(sk)->ipv4.sysctl_tcp_rmem[2]);
+               space = max_t(u32, space, READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_rmem[2]));
                space = max_t(u32, space, sysctl_rmem_max);
                space = min_t(u32, space, *window_clamp);
                *rcv_wscale = clamp_t(int, ilog2(space) - 15,

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 01ede89..7f96e0c 100644 (file)
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -1899,7 +1899,7 @@ static void mptcp_rcv_space_adjust(struct mptcp_sock *msk, int copied)
 
                do_div(rcvwin, advmss);
                rcvbuf = min_t(u64, rcvwin * rcvmem,
-                              sock_net(sk)->ipv4.sysctl_tcp_rmem[2]);
+                              READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_rmem[2]));
 
                if (rcvbuf > sk->sk_rcvbuf) {
                        u32 window_clamp;
@@ -2532,8 +2532,8 @@ static int mptcp_init_sock(struct sock *sk)
        icsk->icsk_ca_ops = NULL;
 
        sk_sockets_allocated_inc(sk);
-       sk->sk_rcvbuf = sock_net(sk)->ipv4.sysctl_tcp_rmem[1];
-       sk->sk_sndbuf = sock_net(sk)->ipv4.sysctl_tcp_wmem[1];
+       sk->sk_rcvbuf = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_rmem[1]);
+       sk->sk_sndbuf = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_wmem[1]);
 
        return 0;
 }

diff --git a/net/tipc/socket.c b/net/tipc/socket.c
index 43509c7..f1c3b8e 100644 (file)
--- a/net/tipc/socket.c
+++ b/net/tipc/socket.c
@@ -517,7 +517,7 @@ static int tipc_sk_create(struct net *net, struct socket *sock,
        timer_setup(&sk->sk_timer, tipc_sk_timeout, 0);
        sk->sk_shutdown = 0;
        sk->sk_backlog_rcv = tipc_sk_backlog_rcv;
-       sk->sk_rcvbuf = sysctl_tipc_rmem[1];
+       sk->sk_rcvbuf = READ_ONCE(sysctl_tipc_rmem[1]);
        sk->sk_data_ready = tipc_data_ready;
        sk->sk_write_space = tipc_write_space;
        sk->sk_destruct = tipc_sock_destruct;