Commit c9f21106 authored by David S. Miller's avatar David S. Miller
Browse files

Merge branch 'net-ipv4-sysctl-races-part-3' 



Kuniyuki Iwashima says:

====================
sysctl: Fix data-races around ipv4_net_table (Round 3).

This series fixes data-races around 21 knobs after
igmp_link_local_mcast_reports in ipv4_net_table.

These 4 knobs are skipped because they are safe.

  - tcp_congestion_control: Safe with RCU and xchg().
  - tcp_available_congestion_control: Read only.
  - tcp_allowed_congestion_control: Safe with RCU and spinlock().
  - tcp_fastopen_key: Safe with RCU and xchg()

So, round 4 will start with fib_multipath_use_neigh.
====================

Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
parents 1e20904e 021266ec

drivers/net/amt.c

+2 −2
Original line number Diff line number Diff line
@@ -563,7 +563,7 @@ static struct sk_buff *amt_build_igmp_gq(struct amt_dev *amt) 
	ihv3->nsrcs	= 0;

	ihv3->resv	= 0;

	ihv3->suppress	= false;

	ihv3->qrv	= amt->net->ipv4.sysctl_igmp_qrv;

	ihv3->qrv	= READ_ONCE(amt->net->ipv4.sysctl_igmp_qrv);

	ihv3->csum	= 0;

	csum		= &ihv3->csum;

	csum_start	= (void *)ihv3;
@@ -3095,7 +3095,7 @@ static int amt_newlink(struct net *net, struct net_device *dev, 
		goto err;

	}

	if (amt->mode == AMT_MODE_RELAY) {

		amt->qrv = amt->net->ipv4.sysctl_igmp_qrv;

		amt->qrv = READ_ONCE(amt->net->ipv4.sysctl_igmp_qrv);

		amt->qri = 10;

		dev->needed_headroom = amt->stream_dev->needed_headroom +

				       AMT_RELAY_HLEN;

include/net/tcp.h

+9 −5
Original line number Diff line number Diff line
@@ -1493,21 +1493,24 @@ static inline int keepalive_intvl_when(const struct tcp_sock *tp) 
{

	struct net *net = sock_net((struct sock *)tp);



	return tp->keepalive_intvl ? : net->ipv4.sysctl_tcp_keepalive_intvl;

	return tp->keepalive_intvl ? :

		READ_ONCE(net->ipv4.sysctl_tcp_keepalive_intvl);

}



static inline int keepalive_time_when(const struct tcp_sock *tp)

{

	struct net *net = sock_net((struct sock *)tp);



	return tp->keepalive_time ? : net->ipv4.sysctl_tcp_keepalive_time;

	return tp->keepalive_time ? :

		READ_ONCE(net->ipv4.sysctl_tcp_keepalive_time);

}



static inline int keepalive_probes(const struct tcp_sock *tp)

{

	struct net *net = sock_net((struct sock *)tp);



	return tp->keepalive_probes ? : net->ipv4.sysctl_tcp_keepalive_probes;

	return tp->keepalive_probes ? :

		READ_ONCE(net->ipv4.sysctl_tcp_keepalive_probes);

}



static inline u32 keepalive_time_elapsed(const struct tcp_sock *tp)
@@ -1520,7 +1523,8 @@ static inline u32 keepalive_time_elapsed(const struct tcp_sock *tp) 


static inline int tcp_fin_time(const struct sock *sk)

{

	int fin_timeout = tcp_sk(sk)->linger2 ? : sock_net(sk)->ipv4.sysctl_tcp_fin_timeout;

	int fin_timeout = tcp_sk(sk)->linger2 ? :

		READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_fin_timeout);

	const int rto = inet_csk(sk)->icsk_rto;



	if (fin_timeout < (rto << 2) - (rto >> 1))
@@ -2023,7 +2027,7 @@ void __tcp_v4_send_check(struct sk_buff *skb, __be32 saddr, __be32 daddr); 
static inline u32 tcp_notsent_lowat(const struct tcp_sock *tp)

{

	struct net *net = sock_net((struct sock *)tp);

	return tp->notsent_lowat ?: net->ipv4.sysctl_tcp_notsent_lowat;

	return tp->notsent_lowat ?: READ_ONCE(net->ipv4.sysctl_tcp_notsent_lowat);

}



bool tcp_stream_memory_free(const struct sock *sk, int wake);

net/core/filter.c

+2 −2
Original line number Diff line number Diff line
@@ -7041,7 +7041,7 @@ BPF_CALL_5(bpf_tcp_check_syncookie, struct sock *, sk, void *, iph, u32, iph_len 
	if (sk->sk_protocol != IPPROTO_TCP || sk->sk_state != TCP_LISTEN)

		return -EINVAL;



	if (!sock_net(sk)->ipv4.sysctl_tcp_syncookies)

	if (!READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_syncookies))

		return -EINVAL;



	if (!th->ack || th->rst || th->syn)
@@ -7116,7 +7116,7 @@ BPF_CALL_5(bpf_tcp_gen_syncookie, struct sock *, sk, void *, iph, u32, iph_len, 
	if (sk->sk_protocol != IPPROTO_TCP || sk->sk_state != TCP_LISTEN)

		return -EINVAL;



	if (!sock_net(sk)->ipv4.sysctl_tcp_syncookies)

	if (!READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_syncookies))

		return -ENOENT;



	if (!th->syn || th->ack || th->fin || th->rst)

net/core/sock_reuseport.c

+2 −2
Original line number Diff line number Diff line
@@ -387,7 +387,7 @@ void reuseport_stop_listen_sock(struct sock *sk) 
		prog = rcu_dereference_protected(reuse->prog,

						 lockdep_is_held(&reuseport_lock));



		if (sock_net(sk)->ipv4.sysctl_tcp_migrate_req ||

		if (READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_migrate_req) ||

		    (prog && prog->expected_attach_type == BPF_SK_REUSEPORT_SELECT_OR_MIGRATE)) {

			/* Migration capable, move sk from the listening section

			 * to the closed section.
@@ -545,7 +545,7 @@ struct sock *reuseport_migrate_sock(struct sock *sk, 
	hash = migrating_sk->sk_hash;

	prog = rcu_dereference(reuse->prog);

	if (!prog || prog->expected_attach_type != BPF_SK_REUSEPORT_SELECT_OR_MIGRATE) {

		if (sock_net(sk)->ipv4.sysctl_tcp_migrate_req)

		if (READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_migrate_req))

			goto select_by_hash;

		goto failure;

	}

net/ipv4/af_inet.c

+1 −1
Original line number Diff line number Diff line
@@ -217,7 +217,7 @@ int inet_listen(struct socket *sock, int backlog) 
		 * because the socket was in TCP_LISTEN state previously but

		 * was shutdown() rather than close().

		 */

		tcp_fastopen = sock_net(sk)->ipv4.sysctl_tcp_fastopen;

		tcp_fastopen = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_fastopen);

		if ((tcp_fastopen & TFO_SERVER_WO_SOCKOPT1) &&

		    (tcp_fastopen & TFO_SERVER_ENABLE) &&

		    !inet_csk(sk)->icsk_accept_queue.fastopenq.max_qlen) {