Skip to contentSkip to conversation

dccp/tcp: Fix bhash2 issues related to WARN_ON() in inet_csk_get_port().

PATCH Superseded Kuniyuki Iwashima 2 participants 12 messages Nov 18, 2022 – Nov 22, 2022

Subsystems (5)

v3 of 4 Superseded Latest is v4 →

lore
Share & tools
lore
#0 lore raw

v3 cover dccp/tcp: Fix bhash2 issues related to WARN_ON() in inet_csk_get_port().

syzkaller was hitting a WARN_ON() in inet_csk_get_port() in the 4th patch, which was because we forgot to fix up bhash2 bucket when connect() for a socket bound to a wildcard address fails in __inet_stream_connect().

There was a similar report [0], but its repro does not fire the WARN_ON() due to inconsistent error handling.

When connect() for a socket bound to a wildcard address fails, saddr may or may not be reset depending on where the failure happens. When we fail in __inet_stream_connect(), sk->sk_prot->disconnect() resets saddr. OTOH, in (dccp|tcp)_v[46]_connect(), if we fail after inet_hash6?_connect(), we forget to reset saddr.

We fix this inconsistent error handling in the 1st patch, and then we'll fix the bhash2 WARN_ON() issue.

Note that there is still an issue in that we reset saddr without checking if there are conflicting sockets in bhash and bhash2, but this should be another series.

See [1][2] for the previous discussion.

[0]: https://lore.kernel.org/netdev/[email protected]/
[1]: https://lore.kernel.org/netdev/[email protected]/
[2]: https://lore.kernel.org/netdev/[email protected]/
[3]: https://lore.kernel.org/netdev/[email protected]/T/#m00aafedb29ff0b55d5e67aef0252ef1baaf4b6ee

Changes:
v3:
* Patch 3
* Update saddr under the bhash's lock
* Correct Fixes tag
* Change #ifdef in inet_update_saddr() along the recent discussion [3]

v2: https://lore.kernel.org/netdev/[email protected]/
* Add patch 2-4

v1: [2]

Kuniyuki Iwashima (4):
dccp/tcp: Reset saddr on failure after inet6?_hash_connect(). dccp/tcp: Remove NULL check for prev_saddr in

    inet_bhash2_update_saddr().

dccp/tcp: Update saddr under bhash's lock. dccp/tcp: Fixup bhash2 bucket when connect() fails.

 include/net/inet_hashtables.h |  3 +-
 net/dccp/ipv4.c               | 23 ++-------
 net/dccp/ipv6.c               | 24 ++-------
 net/dccp/proto.c              |  3 +-
 net/ipv4/af_inet.c            | 11 +----
 net/ipv4/inet_hashtables.c    | 91 +++++++++++++++++++++++++++++------
 net/ipv4/tcp.c                |  3 +-
 net/ipv4/tcp_ipv4.c           | 21 ++------
 net/ipv6/tcp_ipv6.c           | 20 ++------
 9 files changed, 101 insertions(+), 98 deletions(-)
Signature · 1 line
2.30.2
#1 lore raw

PATCH v3 1/4 dccp/tcp: Reset saddr on failure after inet6?_hash_connect().

When connect() is called on a socket bound to the wildcard address, we change the socket's saddr to a local address. If the socket fails to connect() to the destination, we have to reset the saddr.

However, when an error occurs after inet_hash6?_connect() in (dccp|tcp)_v[46]_conect(), we forget to reset saddr and leave the socket bound to the address.

From the user's point of view, whether saddr is reset or not varies with errno. Let's fix this inconsistent behaviour.

Note that after this patch, the repro [0] will trigger the WARN_ON() in inet_csk_get_port() again, but this patch is not buggy and rather fixes a bug papering over the bhash2's bug for which we need another fix.

For the record, the repro causes -EADDRNOTAVAIL in inet_hash6_connect() by this sequence:

s1 = socket()
s1.setsockopt(SOL_SOCKET, SO_REUSEADDR, 1) s1.bind(('127.0.0.1', 10000)) s1.sendto(b'hello', MSG_FASTOPEN, (('127.0.0.1', 10000)))

  # or s1.connect(('127.0.0.1', 10000))

s2 = socket()
s2.setsockopt(SOL_SOCKET, SO_REUSEADDR, 1) s2.bind(('0.0.0.0', 10000)) s2.connect(('127.0.0.1', 10000)) # -EADDRNOTAVAIL

  s2.listen(32)  # WARN_ON(inet_csk(sk)->icsk_bind2_hash != tb2);

[0]: https://syzkaller.appspot.com/bug?extid5d756bbd1f8b5c8f09

Fixes 3df80d9320bc ("[DCCP]: Introduce DCCPv6")
Fixes 7c657876b63c ("[DCCP]: Initial implementation")
Fixes 1da177e4c3f4 ("Linux-2.6.12-rc2")
---
 net/dccp/ipv4.c     | 2 ++
 net/dccp/ipv6.c     | 2 ++
 net/ipv4/tcp_ipv4.c | 2 ++
 net/ipv6/tcp_ipv6.c | 2 ++
 4 files changed, 8 insertions(+)
net/dccp/ipv4.c +2 -0
index 713b7b8dad7e..40640c26680e 100644
@@ -157,6 +157,8 @@ int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 	 * This unhashes the socket and releases the local port, if necessary.
 	 */
 	dccp_set_state(sk, DCCP_CLOSED);
+	if (!(sk->sk_userlocks & SOCK_BINDADDR_LOCK))
+		inet_reset_saddr(sk);
 	ip_rt_put(rt);
 	sk->sk_route_caps = 0;
 	inet->inet_dport = 0;
net/dccp/ipv6.c +2 -0
index e57b43006074..626166cb6d7e 100644
@@ -985,6 +985,8 @@ static int dccp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
 
 late_failure:
 	dccp_set_state(sk, DCCP_CLOSED);
+	if (!(sk->sk_userlocks & SOCK_BINDADDR_LOCK))
+		inet_reset_saddr(sk);
 	__sk_dst_reset(sk);
 failure:
 	inet->inet_dport = 0;
net/ipv4/tcp_ipv4.c +2 -0
index 87d440f47a70..6a3a732b584d 100644
@@ -343,6 +343,8 @@ int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 	 * if necessary.
 	 */
 	tcp_set_state(sk, TCP_CLOSE);
+	if (!(sk->sk_userlocks & SOCK_BINDADDR_LOCK))
+		inet_reset_saddr(sk);
 	ip_rt_put(rt);
 	sk->sk_route_caps = 0;
 	inet->inet_dport = 0;
net/ipv6/tcp_ipv6.c +2 -0
index 2a3f9296df1e..81b396e5cf79 100644
@@ -359,6 +359,8 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
 
 late_failure:
 	tcp_set_state(sk, TCP_CLOSE);
+	if (!(sk->sk_userlocks & SOCK_BINDADDR_LOCK))
+		inet_reset_saddr(sk);
 failure:
 	inet->inet_dport = 0;
 	sk->sk_route_caps = 0;
Signature · 1 line
2.30.2
#2 lore raw

PATCH v3 2/4 dccp/tcp: Remove NULL check for prev_saddr in inet_bhash2_update_saddr().

In reply to Kuniyuki Iwashima

When we call inet_bhash2_update_saddr(), prev_saddr is always non-NULL. Let's remove the unnecessary test.

---
 net/ipv4/inet_hashtables.c | 11 ++++-------
 1 file changed, 4 insertions(+), 7 deletions(-)
net/ipv4/inet_hashtables.c +4 -7
index 033bf3c2538f..d745f962745e 100644
@@ -877,13 +877,10 @@ int inet_bhash2_update_saddr(struct inet_bind_hashbucket *prev_saddr, struct soc
 
 	head2 = inet_bhashfn_portaddr(hinfo, sk, net, port);
 
-	if (prev_saddr) {
-		spin_lock_bh(&prev_saddr->lock);
-		__sk_del_bind2_node(sk);
-		inet_bind2_bucket_destroy(hinfo->bind2_bucket_cachep,
-					  inet_csk(sk)->icsk_bind2_hash);
-		spin_unlock_bh(&prev_saddr->lock);
-	}
+	spin_lock_bh(&prev_saddr->lock);
+	__sk_del_bind2_node(sk);
+	inet_bind2_bucket_destroy(hinfo->bind2_bucket_cachep, inet_csk(sk)->icsk_bind2_hash);
+	spin_unlock_bh(&prev_saddr->lock);
 
 	spin_lock_bh(&head2->lock);
 	tb2 = inet_bind2_bucket_find(head2, net, port, l3mdev, sk);
Signature · 1 line
2.30.2
#3 lore raw

PATCH v3 3/4 dccp/tcp: Update saddr under bhash's lock.

In reply to Kuniyuki Iwashima

When we call connect() for a socket bound to a wildcard address, we update saddr locklessly. However, it could result in a data race; another thread iterating over bhash might see a corrupted address.

Let's update saddr under the bhash bucket's lock.

Fixes 3df80d9320bc ("[DCCP]: Introduce DCCPv6")
Fixes 7c657876b63c ("[DCCP]: Initial implementation")
Fixes 1da177e4c3f4 ("Linux-2.6.12-rc2")
---
 include/net/inet_hashtables.h |  2 +-
 net/dccp/ipv4.c               | 22 +++-----------
 net/dccp/ipv6.c               | 23 +++------------
 net/ipv4/af_inet.c            | 11 +------
 net/ipv4/inet_hashtables.c    | 55 +++++++++++++++++++++++++++++------
 net/ipv4/tcp_ipv4.c           | 20 +++----------
 net/ipv6/tcp_ipv6.c           | 19 ++----------
 7 files changed, 63 insertions(+), 89 deletions(-)
include/net/inet_hashtables.h +1 -1
index 3af1e927247d..ba06e8b52264 100644
@@ -281,7 +281,7 @@ inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in
  * sk_v6_rcv_saddr (ipv6) changes after it has been binded. The socket's
  * rcv_saddr field should already have been updated when this is called.
  */
-int inet_bhash2_update_saddr(struct inet_bind_hashbucket *prev_saddr, struct sock *sk);
+int inet_bhash2_update_saddr(struct sock *sk, void *saddr, int family);
 
 void inet_bind_hash(struct sock *sk, struct inet_bind_bucket *tb,
 		    struct inet_bind2_bucket *tb2, unsigned short port);
net/dccp/ipv4.c +4 -17
index 40640c26680e..95e376e3b911 100644
@@ -45,11 +45,10 @@ static unsigned int dccp_v4_pernet_id __read_mostly;
 int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 {
 	const struct sockaddr_in *usin = (struct sockaddr_in *)uaddr;
-	struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
-	__be32 daddr, nexthop, prev_sk_rcv_saddr;
 	struct inet_sock *inet = inet_sk(sk);
 	struct dccp_sock *dp = dccp_sk(sk);
 	__be16 orig_sport, orig_dport;
+	__be32 daddr, nexthop;
 	struct flowi4 *fl4;
 	struct rtable *rt;
 	int err;
@@ -91,26 +90,13 @@ int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 		daddr = fl4->daddr;
 
 	if (inet->inet_saddr = 0) {
-		if (inet_csk(sk)->icsk_bind2_hash) {
-			prev_addr_hashbucket -				inet_bhashfn_portaddr(&dccp_hashinfo, sk,
-						      sock_net(sk),
-						      inet->inet_num);
-			prev_sk_rcv_saddr = sk->sk_rcv_saddr;
-		}
-		inet->inet_saddr = fl4->saddr;
-	}
-
-	sk_rcv_saddr_set(sk, inet->inet_saddr);
-
-	if (prev_addr_hashbucket) {
-		err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
+		err = inet_bhash2_update_saddr(sk,  &fl4->saddr, AF_INET);
 		if (err) {
-			inet->inet_saddr = 0;
-			sk_rcv_saddr_set(sk, prev_sk_rcv_saddr);
 			ip_rt_put(rt);
 			return err;
 		}
+	} else {
+		sk_rcv_saddr_set(sk, inet->inet_saddr);
 	}
 
 	inet->inet_dport = usin->sin_port;
net/dccp/ipv6.c +4 -19
index 626166cb6d7e..94c101ed57a9 100644
@@ -934,26 +934,11 @@ static int dccp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
 	}
 
 	if (saddr = NULL) {
-		struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
-		struct in6_addr prev_v6_rcv_saddr;
-
-		if (icsk->icsk_bind2_hash) {
-			prev_addr_hashbucket = inet_bhashfn_portaddr(&dccp_hashinfo,
-								     sk, sock_net(sk),
-								     inet->inet_num);
-			prev_v6_rcv_saddr = sk->sk_v6_rcv_saddr;
-		}
-
 		saddr = &fl6.saddr;
-		sk->sk_v6_rcv_saddr = *saddr;
-
-		if (prev_addr_hashbucket) {
-			err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
-			if (err) {
-				sk->sk_v6_rcv_saddr = prev_v6_rcv_saddr;
-				goto failure;
-			}
-		}
+
+		err = inet_bhash2_update_saddr(sk, saddr, AF_INET6);
+		if (err)
+			goto failure;
 	}
 
 	/* set the source address */
net/ipv4/af_inet.c +1 -9
index 4728087c42a5..0da679411330 100644
@@ -1230,7 +1230,6 @@ EXPORT_SYMBOL(inet_unregister_protosw);
 
 static int inet_sk_reselect_saddr(struct sock *sk)
 {
-	struct inet_bind_hashbucket *prev_addr_hashbucket;
 	struct inet_sock *inet = inet_sk(sk);
 	__be32 old_saddr = inet->inet_saddr;
 	__be32 daddr = inet->inet_daddr;
@@ -1260,16 +1259,8 @@ static int inet_sk_reselect_saddr(struct sock *sk)
 		return 0;
 	}
 
-	prev_addr_hashbucket -		inet_bhashfn_portaddr(tcp_or_dccp_get_hashinfo(sk), sk,
-				      sock_net(sk), inet->inet_num);
-
-	inet->inet_saddr = inet->inet_rcv_saddr = new_saddr;
-
-	err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
+	err = inet_bhash2_update_saddr(sk, &new_saddr, AF_INET);
 	if (err) {
-		inet->inet_saddr = old_saddr;
-		inet->inet_rcv_saddr = old_saddr;
 		ip_rt_put(rt);
 		return err;
 	}
net/ipv4/inet_hashtables.c +46 -9

89 lines not on this page · raw message

net/ipv4/tcp_ipv4.c +4 -16
index 6a3a732b584d..23dd7e9df2d5 100644
@@ -199,15 +199,14 @@ static int tcp_v4_pre_connect(struct sock *sk, struct sockaddr *uaddr,
 /* This will initiate an outgoing connection. */
 int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 {
-	struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
 	struct sockaddr_in *usin = (struct sockaddr_in *)uaddr;
 	struct inet_timewait_death_row *tcp_death_row;
-	__be32 daddr, nexthop, prev_sk_rcv_saddr;
 	struct inet_sock *inet = inet_sk(sk);
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct ip_options_rcu *inet_opt;
 	struct net *net = sock_net(sk);
 	__be16 orig_sport, orig_dport;
+	__be32 daddr, nexthop;
 	struct flowi4 *fl4;
 	struct rtable *rt;
 	int err;
@@ -251,24 +250,13 @@ int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 	tcp_death_row = &sock_net(sk)->ipv4.tcp_death_row;
 
 	if (!inet->inet_saddr) {
-		if (inet_csk(sk)->icsk_bind2_hash) {
-			prev_addr_hashbucket = inet_bhashfn_portaddr(tcp_death_row->hashinfo,
-								     sk, net, inet->inet_num);
-			prev_sk_rcv_saddr = sk->sk_rcv_saddr;
-		}
-		inet->inet_saddr = fl4->saddr;
-	}
-
-	sk_rcv_saddr_set(sk, inet->inet_saddr);
-
-	if (prev_addr_hashbucket) {
-		err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
+		err = inet_bhash2_update_saddr(sk,  &fl4->saddr, AF_INET);
 		if (err) {
-			inet->inet_saddr = 0;
-			sk_rcv_saddr_set(sk, prev_sk_rcv_saddr);
 			ip_rt_put(rt);
 			return err;
 		}
+	} else {
+		sk_rcv_saddr_set(sk, inet->inet_saddr);
 	}
 
 	if (tp->rx_opt.ts_recent_stamp && inet->inet_daddr != daddr) {
net/ipv6/tcp_ipv6.c +3 -16
index 81b396e5cf79..2f3ca3190d26 100644
@@ -292,24 +292,11 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
 	tcp_death_row = &sock_net(sk)->ipv4.tcp_death_row;
 
 	if (!saddr) {
-		struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
-		struct in6_addr prev_v6_rcv_saddr;
-
-		if (icsk->icsk_bind2_hash) {
-			prev_addr_hashbucket = inet_bhashfn_portaddr(tcp_death_row->hashinfo,
-								     sk, net, inet->inet_num);
-			prev_v6_rcv_saddr = sk->sk_v6_rcv_saddr;
-		}
 		saddr = &fl6.saddr;
-		sk->sk_v6_rcv_saddr = *saddr;
 
-		if (prev_addr_hashbucket) {
-			err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
-			if (err) {
-				sk->sk_v6_rcv_saddr = prev_v6_rcv_saddr;
-				goto failure;
-			}
-		}
+		err = inet_bhash2_update_saddr(sk, saddr, AF_INET6);
+		if (err)
+			goto failure;
 	}
 
 	/* set the source address */
Signature · 1 line
2.30.2
#4 lore raw

Re: [PATCH v3 net 3/4] dccp/tcp: Update saddr under bhash's lock.

Quoted 6 lines · Kuniyuki Iwashima

When we call connect() for a socket bound to a wildcard address, we update saddr locklessly. However, it could result in a data race; another thread iterating over bhash might see a corrupted address.

Let's update saddr under the bhash bucket's lock.

Thanks for the quick turnaround!

Quoted 166 lines
Fixes 3df80d9320bc ("[DCCP]: Introduce DCCPv6")
Fixes 7c657876b63c ("[DCCP]: Initial implementation")
Fixes 1da177e4c3f4 ("Linux-2.6.12-rc2")
---
 include/net/inet_hashtables.h |  2 +-
 net/dccp/ipv4.c               | 22 +++-----------
 net/dccp/ipv6.c               | 23 +++------------
 net/ipv4/af_inet.c            | 11 +------
 net/ipv4/inet_hashtables.c    | 55 +++++++++++++++++++++++++++++------
 net/ipv4/tcp_ipv4.c           | 20 +++----------
 net/ipv6/tcp_ipv6.c           | 19 ++----------
 7 files changed, 63 insertions(+), 89 deletions(-)
include/net/inet_hashtables.h +1 -1
index 3af1e927247d..ba06e8b52264 100644
@@ -281,7 +281,7 @@ inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in
  * sk_v6_rcv_saddr (ipv6) changes after it has been binded. The socket's
  * rcv_saddr field should already have been updated when this is called.
  */
-int inet_bhash2_update_saddr(struct inet_bind_hashbucket *prev_saddr, struct sock *sk);
+int inet_bhash2_update_saddr(struct sock *sk, void *saddr, int family);
 
 void inet_bind_hash(struct sock *sk, struct inet_bind_bucket *tb,
                    struct inet_bind2_bucket *tb2, unsigned short port);
net/dccp/ipv4.c +4 -17
index 40640c26680e..95e376e3b911 100644
@@ -45,11 +45,10 @@ static unsigned int dccp_v4_pernet_id __read_mostly;
 int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 {
        const struct sockaddr_in *usin = (struct sockaddr_in *)uaddr;
-       struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
-       __be32 daddr, nexthop, prev_sk_rcv_saddr;
        struct inet_sock *inet = inet_sk(sk);
        struct dccp_sock *dp = dccp_sk(sk);
        __be16 orig_sport, orig_dport;
+       __be32 daddr, nexthop;
        struct flowi4 *fl4;
        struct rtable *rt;
        int err;
@@ -91,26 +90,13 @@ int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
                daddr = fl4->daddr;
 
        if (inet->inet_saddr = 0) {
-               if (inet_csk(sk)->icsk_bind2_hash) {
-                       prev_addr_hashbucket > -                               inet_bhashfn_portaddr(&dccp_hashinfo, sk,
-                                                     sock_net(sk),
-                                                     inet->inet_num);
-                       prev_sk_rcv_saddr = sk->sk_rcv_saddr;
-               }
-               inet->inet_saddr = fl4->saddr;
-       }
-
-       sk_rcv_saddr_set(sk, inet->inet_saddr);
-
-       if (prev_addr_hashbucket) {
-               err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
+               err = inet_bhash2_update_saddr(sk,  &fl4->saddr, AF_INET);
                if (err) {
-                       inet->inet_saddr = 0;
-                       sk_rcv_saddr_set(sk, prev_sk_rcv_saddr);
                        ip_rt_put(rt);
                        return err;
                }
+       } else {
+               sk_rcv_saddr_set(sk, inet->inet_saddr);
        }
 
        inet->inet_dport = usin->sin_port;
net/dccp/ipv6.c +4 -19
index 626166cb6d7e..94c101ed57a9 100644
@@ -934,26 +934,11 @@ static int dccp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
        }
 
        if (saddr = NULL) {
-               struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
-               struct in6_addr prev_v6_rcv_saddr;
-
-               if (icsk->icsk_bind2_hash) {
-                       prev_addr_hashbucket = inet_bhashfn_portaddr(&dccp_hashinfo,
-                                                                    sk, sock_net(sk),
-                                                                    inet->inet_num);
-                       prev_v6_rcv_saddr = sk->sk_v6_rcv_saddr;
-               }
-
                saddr = &fl6.saddr;
-               sk->sk_v6_rcv_saddr = *saddr;
-
-               if (prev_addr_hashbucket) {
-                       err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
-                       if (err) {
-                               sk->sk_v6_rcv_saddr = prev_v6_rcv_saddr;
-                               goto failure;
-                       }
-               }
+
+               err = inet_bhash2_update_saddr(sk, saddr, AF_INET6);
+               if (err)
+                       goto failure;
        }
 
        /* set the source address */
net/ipv4/af_inet.c +1 -9
index 4728087c42a5..0da679411330 100644
@@ -1230,7 +1230,6 @@ EXPORT_SYMBOL(inet_unregister_protosw);
 
 static int inet_sk_reselect_saddr(struct sock *sk)
 {
-       struct inet_bind_hashbucket *prev_addr_hashbucket;
        struct inet_sock *inet = inet_sk(sk);
        __be32 old_saddr = inet->inet_saddr;
        __be32 daddr = inet->inet_daddr;
@@ -1260,16 +1259,8 @@ static int inet_sk_reselect_saddr(struct sock *sk)
                return 0;
        }
 
-       prev_addr_hashbucket > -               inet_bhashfn_portaddr(tcp_or_dccp_get_hashinfo(sk), sk,
-                                     sock_net(sk), inet->inet_num);
-
-       inet->inet_saddr = inet->inet_rcv_saddr = new_saddr;
-
-       err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
+       err = inet_bhash2_update_saddr(sk, &new_saddr, AF_INET);
        if (err) {
-               inet->inet_saddr = old_saddr;
-               inet->inet_rcv_saddr = old_saddr;
                ip_rt_put(rt);
                return err;
        }
net/ipv4/inet_hashtables.c +15 -1
index d745f962745e..fce0bd62d6b5 100644
@@ -858,31 +858,65 @@ inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in
        return &hinfo->bhash2[hash & (hinfo->bhash_size - 1)];
 }
 
-int inet_bhash2_update_saddr(struct inet_bind_hashbucket *prev_saddr, struct sock *sk)
+static void inet_update_saddr(struct sock *sk, void *saddr, int family)
+{
+       if (family = AF_INET) {
+               inet_sk(sk)->inet_saddr = *(__be32 *)saddr;
+               sk_rcv_saddr_set(sk, inet_sk(sk)->inet_saddr);
+       }
+#if IS_ENABLED(CONFIG_IPV6)
+       else {
+               sk->sk_v6_rcv_saddr = *(struct in6_addr *)saddr;
+       }
+#endif
+}
+
+int inet_bhash2_update_saddr(struct sock *sk, void *saddr, int family)
 {
        struct inet_hashinfo *hinfo = tcp_or_dccp_get_hashinfo(sk);
+       struct inet_bind_hashbucket *head, *head2;
        struct inet_bind2_bucket *tb2, *new_tb2;
        int l3mdev = inet_sk_bound_l3mdev(sk);

net/ipv4/inet_hashtables.c inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in · line 888

Quoted 10 lines
net/ipv4/inet_hashtables.c +7 -1
-       struct inet_bind_hashbucket *head2;
        int port = inet_sk(sk)->inet_num;
        struct net *net = sock_net(sk);
+       int bhash, err = 0;
+
+       if (!inet_csk(sk)->icsk_bind2_hash) {
+               /* Not bind()ed before. */
+               inet_update_saddr(sk, saddr, family);
+               goto out;
+       }

I think it would be cleaner if this logic were outside bhash2_update_saddr(), since this mutates the sk's address when the socket hasn't been previously bound to bhash2. I think something like this would be clearer:

static int inet_update_saddr(struct sock *sk, void *saddr, int family)

{
    if (!inet_csk(sk)->icsk_bind2_hash) {
      update_sk_saddr(sk, saddr, family)
      return 0;
    }
    return inet_bhash2_update_saddr(sk, saddr, family);
}

and then from dccp/tcp_ipv4/6_connect(), we just call inet_update_saddr(). This also "moves" the lower-level implementation details (eg underlying bind tables) to inet_hashtables.c, instead of it being mentioned in the higher dccp_tcp_ipv4/6 layers.

What are your thoughts?

net/ipv4/inet_hashtables.c inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in · line 896

Quoted 8 lines
net/ipv4/inet_hashtables.c +8 -0
+
+       bhash = inet_bhashfn(net, port, hinfo->bhash_size);
+       head = &hinfo->bhash[bhash];
+
+       /* If we change saddr locklessly, another thread
+        * iterating over bhash might see corrupted address.
+        */
+       spin_lock_bh(&head->lock);

I don't think we should be acquiring the bhash lock here. I think we only need to acquire it right before we mutate the saddr, and we can release it right after.

Quoted 133 lines
net/ipv4/inet_hashtables.c +16 -7
        /* Allocate a bind2 bucket ahead of time to avoid permanently putting
         * the bhash2 table in an inconsistent state if a new tb2 bucket
         * allocation fails.
         */
        new_tb2 = kmem_cache_alloc(hinfo->bind2_bucket_cachep, GFP_ATOMIC);
-       if (!new_tb2)
-               return -ENOMEM;
+       if (!new_tb2) {
+               err = -ENOMEM;
+               goto unlock;
+       }
 
        head2 = inet_bhashfn_portaddr(hinfo, sk, net, port);
 
-       spin_lock_bh(&prev_saddr->lock);
+       spin_lock(&head2->lock);
        __sk_del_bind2_node(sk);
        inet_bind2_bucket_destroy(hinfo->bind2_bucket_cachep, inet_csk(sk)->icsk_bind2_hash);
-       spin_unlock_bh(&prev_saddr->lock);
+       spin_unlock(&head2->lock);
+
+       inet_update_saddr(sk, saddr, family);
 
-       spin_lock_bh(&head2->lock);
+       head2 = inet_bhashfn_portaddr(hinfo, sk, net, port);
+
+       spin_lock(&head2->lock);
        tb2 = inet_bind2_bucket_find(head2, net, port, l3mdev, sk);
        if (!tb2) {
                tb2 = new_tb2;
@@ -890,12 +924,15 @@ int inet_bhash2_update_saddr(struct inet_bind_hashbucket *prev_saddr, struct soc
        }
        sk_add_bind2_node(sk, &tb2->owners);
        inet_csk(sk)->icsk_bind2_hash = tb2;
-       spin_unlock_bh(&head2->lock);
+       spin_unlock(&head2->lock);
 
        if (tb2 != new_tb2)
                kmem_cache_free(hinfo->bind2_bucket_cachep, new_tb2);
 
-       return 0;
+unlock:
+       spin_unlock_bh(&head->lock);
+out:
+       return err;
 }
 EXPORT_SYMBOL_GPL(inet_bhash2_update_saddr);
 
net/ipv4/tcp_ipv4.c +4 -16
index 6a3a732b584d..23dd7e9df2d5 100644
@@ -199,15 +199,14 @@ static int tcp_v4_pre_connect(struct sock *sk, struct sockaddr *uaddr,
 /* This will initiate an outgoing connection. */
 int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 {
-       struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
        struct sockaddr_in *usin = (struct sockaddr_in *)uaddr;
        struct inet_timewait_death_row *tcp_death_row;
-       __be32 daddr, nexthop, prev_sk_rcv_saddr;
        struct inet_sock *inet = inet_sk(sk);
        struct tcp_sock *tp = tcp_sk(sk);
        struct ip_options_rcu *inet_opt;
        struct net *net = sock_net(sk);
        __be16 orig_sport, orig_dport;
+       __be32 daddr, nexthop;
        struct flowi4 *fl4;
        struct rtable *rt;
        int err;
@@ -251,24 +250,13 @@ int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
        tcp_death_row = &sock_net(sk)->ipv4.tcp_death_row;
 
        if (!inet->inet_saddr) {
-               if (inet_csk(sk)->icsk_bind2_hash) {
-                       prev_addr_hashbucket = inet_bhashfn_portaddr(tcp_death_row->hashinfo,
-                                                                    sk, net, inet->inet_num);
-                       prev_sk_rcv_saddr = sk->sk_rcv_saddr;
-               }
-               inet->inet_saddr = fl4->saddr;
-       }
-
-       sk_rcv_saddr_set(sk, inet->inet_saddr);
-
-       if (prev_addr_hashbucket) {
-               err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
+               err = inet_bhash2_update_saddr(sk,  &fl4->saddr, AF_INET);
                if (err) {
-                       inet->inet_saddr = 0;
-                       sk_rcv_saddr_set(sk, prev_sk_rcv_saddr);
                        ip_rt_put(rt);
                        return err;
                }
+       } else {
+               sk_rcv_saddr_set(sk, inet->inet_saddr);
        }
 
        if (tp->rx_opt.ts_recent_stamp && inet->inet_daddr != daddr) {
net/ipv6/tcp_ipv6.c +3 -17
index 81b396e5cf79..2f3ca3190d26 100644
@@ -292,24 +292,11 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
        tcp_death_row = &sock_net(sk)->ipv4.tcp_death_row;
 
        if (!saddr) {
-               struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
-               struct in6_addr prev_v6_rcv_saddr;
-
-               if (icsk->icsk_bind2_hash) {
-                       prev_addr_hashbucket = inet_bhashfn_portaddr(tcp_death_row->hashinfo,
-                                                                    sk, net, inet->inet_num);
-                       prev_v6_rcv_saddr = sk->sk_v6_rcv_saddr;
-               }
                saddr = &fl6.saddr;
-               sk->sk_v6_rcv_saddr = *saddr;
 
-               if (prev_addr_hashbucket) {
-                       err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
-                       if (err) {
-                               sk->sk_v6_rcv_saddr = prev_v6_rcv_saddr;
-                               goto failure;
-                       }
-               }
+               err = inet_bhash2_update_saddr(sk, saddr, AF_INET6);
+               if (err)
+                       goto failure;
        }
 
        /* set the source address */
--

2.30.2

#5 lore raw

Re: [PATCH v3 net 3/4] dccp/tcp: Update saddr under bhash's lock.

From:   Joanne Koong <[email protected]>
Date:   Fri, 18 Nov 2022 15:20:20 -0800
Quoted 201 lines · Kuniyuki Iwashima
Quoted 6 lines

When we call connect() for a socket bound to a wildcard address, we update saddr locklessly. However, it could result in a data race; another thread iterating over bhash might see a corrupted address.

Let's update saddr under the bhash bucket's lock.

Thanks for the quick turnaround!

Quoted 166 lines
Fixes 3df80d9320bc ("[DCCP]: Introduce DCCPv6")
Fixes 7c657876b63c ("[DCCP]: Initial implementation")
Fixes 1da177e4c3f4 ("Linux-2.6.12-rc2")
---
 include/net/inet_hashtables.h |  2 +-
 net/dccp/ipv4.c               | 22 +++-----------
 net/dccp/ipv6.c               | 23 +++------------
 net/ipv4/af_inet.c            | 11 +------
 net/ipv4/inet_hashtables.c    | 55 +++++++++++++++++++++++++++++------
 net/ipv4/tcp_ipv4.c           | 20 +++----------
 net/ipv6/tcp_ipv6.c           | 19 ++----------
 7 files changed, 63 insertions(+), 89 deletions(-)
include/net/inet_hashtables.h +1 -1
index 3af1e927247d..ba06e8b52264 100644
@@ -281,7 +281,7 @@ inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in
  * sk_v6_rcv_saddr (ipv6) changes after it has been binded. The socket's
  * rcv_saddr field should already have been updated when this is called.
  */
-int inet_bhash2_update_saddr(struct inet_bind_hashbucket *prev_saddr, struct sock *sk);
+int inet_bhash2_update_saddr(struct sock *sk, void *saddr, int family);
 
 void inet_bind_hash(struct sock *sk, struct inet_bind_bucket *tb,
                    struct inet_bind2_bucket *tb2, unsigned short port);
net/dccp/ipv4.c +4 -17
index 40640c26680e..95e376e3b911 100644
@@ -45,11 +45,10 @@ static unsigned int dccp_v4_pernet_id __read_mostly;
 int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 {
        const struct sockaddr_in *usin = (struct sockaddr_in *)uaddr;
-       struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
-       __be32 daddr, nexthop, prev_sk_rcv_saddr;
        struct inet_sock *inet = inet_sk(sk);
        struct dccp_sock *dp = dccp_sk(sk);
        __be16 orig_sport, orig_dport;
+       __be32 daddr, nexthop;
        struct flowi4 *fl4;
        struct rtable *rt;
        int err;
@@ -91,26 +90,13 @@ int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
                daddr = fl4->daddr;
 
        if (inet->inet_saddr = 0) {
-               if (inet_csk(sk)->icsk_bind2_hash) {
-                       prev_addr_hashbucket > > -                               inet_bhashfn_portaddr(&dccp_hashinfo, sk,
-                                                     sock_net(sk),
-                                                     inet->inet_num);
-                       prev_sk_rcv_saddr = sk->sk_rcv_saddr;
-               }
-               inet->inet_saddr = fl4->saddr;
-       }
-
-       sk_rcv_saddr_set(sk, inet->inet_saddr);
-
-       if (prev_addr_hashbucket) {
-               err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
+               err = inet_bhash2_update_saddr(sk,  &fl4->saddr, AF_INET);
                if (err) {
-                       inet->inet_saddr = 0;
-                       sk_rcv_saddr_set(sk, prev_sk_rcv_saddr);
                        ip_rt_put(rt);
                        return err;
                }
+       } else {
+               sk_rcv_saddr_set(sk, inet->inet_saddr);
        }
 
        inet->inet_dport = usin->sin_port;
net/dccp/ipv6.c +4 -19
index 626166cb6d7e..94c101ed57a9 100644
@@ -934,26 +934,11 @@ static int dccp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
        }
 
        if (saddr = NULL) {
-               struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
-               struct in6_addr prev_v6_rcv_saddr;
-
-               if (icsk->icsk_bind2_hash) {
-                       prev_addr_hashbucket = inet_bhashfn_portaddr(&dccp_hashinfo,
-                                                                    sk, sock_net(sk),
-                                                                    inet->inet_num);
-                       prev_v6_rcv_saddr = sk->sk_v6_rcv_saddr;
-               }
-
                saddr = &fl6.saddr;
-               sk->sk_v6_rcv_saddr = *saddr;
-
-               if (prev_addr_hashbucket) {
-                       err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
-                       if (err) {
-                               sk->sk_v6_rcv_saddr = prev_v6_rcv_saddr;
-                               goto failure;
-                       }
-               }
+
+               err = inet_bhash2_update_saddr(sk, saddr, AF_INET6);
+               if (err)
+                       goto failure;
        }
 
        /* set the source address */
net/ipv4/af_inet.c +1 -9
index 4728087c42a5..0da679411330 100644
@@ -1230,7 +1230,6 @@ EXPORT_SYMBOL(inet_unregister_protosw);
 
 static int inet_sk_reselect_saddr(struct sock *sk)
 {
-       struct inet_bind_hashbucket *prev_addr_hashbucket;
        struct inet_sock *inet = inet_sk(sk);
        __be32 old_saddr = inet->inet_saddr;
        __be32 daddr = inet->inet_daddr;
@@ -1260,16 +1259,8 @@ static int inet_sk_reselect_saddr(struct sock *sk)
                return 0;
        }
 
-       prev_addr_hashbucket > > -               inet_bhashfn_portaddr(tcp_or_dccp_get_hashinfo(sk), sk,
-                                     sock_net(sk), inet->inet_num);
-
-       inet->inet_saddr = inet->inet_rcv_saddr = new_saddr;
-
-       err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
+       err = inet_bhash2_update_saddr(sk, &new_saddr, AF_INET);
        if (err) {
-               inet->inet_saddr = old_saddr;
-               inet->inet_rcv_saddr = old_saddr;
                ip_rt_put(rt);
                return err;
        }
net/ipv4/inet_hashtables.c +15 -1
index d745f962745e..fce0bd62d6b5 100644
@@ -858,31 +858,65 @@ inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in
        return &hinfo->bhash2[hash & (hinfo->bhash_size - 1)];
 }
 
-int inet_bhash2_update_saddr(struct inet_bind_hashbucket *prev_saddr, struct sock *sk)
+static void inet_update_saddr(struct sock *sk, void *saddr, int family)
+{
+       if (family = AF_INET) {
+               inet_sk(sk)->inet_saddr = *(__be32 *)saddr;
+               sk_rcv_saddr_set(sk, inet_sk(sk)->inet_saddr);
+       }
+#if IS_ENABLED(CONFIG_IPV6)
+       else {
+               sk->sk_v6_rcv_saddr = *(struct in6_addr *)saddr;
+       }
+#endif
+}
+
+int inet_bhash2_update_saddr(struct sock *sk, void *saddr, int family)
 {
        struct inet_hashinfo *hinfo = tcp_or_dccp_get_hashinfo(sk);
+       struct inet_bind_hashbucket *head, *head2;
        struct inet_bind2_bucket *tb2, *new_tb2;
        int l3mdev = inet_sk_bound_l3mdev(sk);

net/ipv4/inet_hashtables.c inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in · line 888

Quoted 10 lines
net/ipv4/inet_hashtables.c +7 -1
-       struct inet_bind_hashbucket *head2;
        int port = inet_sk(sk)->inet_num;
        struct net *net = sock_net(sk);
+       int bhash, err = 0;
+
+       if (!inet_csk(sk)->icsk_bind2_hash) {
+               /* Not bind()ed before. */
+               inet_update_saddr(sk, saddr, family);
+               goto out;
+       }

I think it would be cleaner if this logic were outside bhash2_update_saddr(), since this mutates the sk's address when the socket hasn't been previously bound to bhash2. I think something like this would be clearer:

static int inet_update_saddr(struct sock *sk, void *saddr, int family)

{
    if (!inet_csk(sk)->icsk_bind2_hash) {
      update_sk_saddr(sk, saddr, family)
      return 0;
    }
    return inet_bhash2_update_saddr(sk, saddr, family);
}
Quoted 6 lines

and then from dccp/tcp_ipv4/6_connect(), we just call inet_update_saddr(). This also "moves" the lower-level implementation details (eg underlying bind tables) to inet_hashtables.c, instead of it being mentioned in the higher dccp_tcp_ipv4/6 layers.

What are your thoughts?

Sounds good!
I'll change them like above.

Quoted 13 lines

net/ipv4/inet_hashtables.c inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in · line 896

Quoted 9 lines
net/ipv4/inet_hashtables.c +8 -0
+
+       bhash = inet_bhashfn(net, port, hinfo->bhash_size);
+       head = &hinfo->bhash[bhash];
+
+       /* If we change saddr locklessly, another thread
+        * iterating over bhash might see corrupted address.
+        */
+       spin_lock_bh(&head->lock);

I don't think we should be acquiring the bhash lock here. I think we only need to acquire it right before we mutate the saddr, and we can release it right after.

Exactly, will move it down before "__"inet_update_saddr() :)

Thank you!

Quoted 131 lines
Quoted 131 lines
net/ipv4/inet_hashtables.c +16 -7
        /* Allocate a bind2 bucket ahead of time to avoid permanently putting
         * the bhash2 table in an inconsistent state if a new tb2 bucket
         * allocation fails.
         */
        new_tb2 = kmem_cache_alloc(hinfo->bind2_bucket_cachep, GFP_ATOMIC);
-       if (!new_tb2)
-               return -ENOMEM;
+       if (!new_tb2) {
+               err = -ENOMEM;
+               goto unlock;
+       }
 
        head2 = inet_bhashfn_portaddr(hinfo, sk, net, port);
 
-       spin_lock_bh(&prev_saddr->lock);
+       spin_lock(&head2->lock);
        __sk_del_bind2_node(sk);
        inet_bind2_bucket_destroy(hinfo->bind2_bucket_cachep, inet_csk(sk)->icsk_bind2_hash);
-       spin_unlock_bh(&prev_saddr->lock);
+       spin_unlock(&head2->lock);
+
+       inet_update_saddr(sk, saddr, family);
 
-       spin_lock_bh(&head2->lock);
+       head2 = inet_bhashfn_portaddr(hinfo, sk, net, port);
+
+       spin_lock(&head2->lock);
        tb2 = inet_bind2_bucket_find(head2, net, port, l3mdev, sk);
        if (!tb2) {
                tb2 = new_tb2;
@@ -890,12 +924,15 @@ int inet_bhash2_update_saddr(struct inet_bind_hashbucket *prev_saddr, struct soc
        }
        sk_add_bind2_node(sk, &tb2->owners);
        inet_csk(sk)->icsk_bind2_hash = tb2;
-       spin_unlock_bh(&head2->lock);
+       spin_unlock(&head2->lock);
 
        if (tb2 != new_tb2)
                kmem_cache_free(hinfo->bind2_bucket_cachep, new_tb2);
 
-       return 0;
+unlock:
+       spin_unlock_bh(&head->lock);
+out:
+       return err;
 }
 EXPORT_SYMBOL_GPL(inet_bhash2_update_saddr);
 
net/ipv4/tcp_ipv4.c +4 -16
index 6a3a732b584d..23dd7e9df2d5 100644
@@ -199,15 +199,14 @@ static int tcp_v4_pre_connect(struct sock *sk, struct sockaddr *uaddr,
 /* This will initiate an outgoing connection. */
 int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 {
-       struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
        struct sockaddr_in *usin = (struct sockaddr_in *)uaddr;
        struct inet_timewait_death_row *tcp_death_row;
-       __be32 daddr, nexthop, prev_sk_rcv_saddr;
        struct inet_sock *inet = inet_sk(sk);
        struct tcp_sock *tp = tcp_sk(sk);
        struct ip_options_rcu *inet_opt;
        struct net *net = sock_net(sk);
        __be16 orig_sport, orig_dport;
+       __be32 daddr, nexthop;
        struct flowi4 *fl4;
        struct rtable *rt;
        int err;
@@ -251,24 +250,13 @@ int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
        tcp_death_row = &sock_net(sk)->ipv4.tcp_death_row;
 
        if (!inet->inet_saddr) {
-               if (inet_csk(sk)->icsk_bind2_hash) {
-                       prev_addr_hashbucket = inet_bhashfn_portaddr(tcp_death_row->hashinfo,
-                                                                    sk, net, inet->inet_num);
-                       prev_sk_rcv_saddr = sk->sk_rcv_saddr;
-               }
-               inet->inet_saddr = fl4->saddr;
-       }
-
-       sk_rcv_saddr_set(sk, inet->inet_saddr);
-
-       if (prev_addr_hashbucket) {
-               err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
+               err = inet_bhash2_update_saddr(sk,  &fl4->saddr, AF_INET);
                if (err) {
-                       inet->inet_saddr = 0;
-                       sk_rcv_saddr_set(sk, prev_sk_rcv_saddr);
                        ip_rt_put(rt);
                        return err;
                }
+       } else {
+               sk_rcv_saddr_set(sk, inet->inet_saddr);
        }
 
        if (tp->rx_opt.ts_recent_stamp && inet->inet_daddr != daddr) {
net/ipv6/tcp_ipv6.c +3 -17
index 81b396e5cf79..2f3ca3190d26 100644
@@ -292,24 +292,11 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
        tcp_death_row = &sock_net(sk)->ipv4.tcp_death_row;
 
        if (!saddr) {
-               struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
-               struct in6_addr prev_v6_rcv_saddr;
-
-               if (icsk->icsk_bind2_hash) {
-                       prev_addr_hashbucket = inet_bhashfn_portaddr(tcp_death_row->hashinfo,
-                                                                    sk, net, inet->inet_num);
-                       prev_v6_rcv_saddr = sk->sk_v6_rcv_saddr;
-               }
                saddr = &fl6.saddr;
-               sk->sk_v6_rcv_saddr = *saddr;
 
-               if (prev_addr_hashbucket) {
-                       err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
-                       if (err) {
-                               sk->sk_v6_rcv_saddr = prev_v6_rcv_saddr;
-                               goto failure;
-                       }
-               }
+               err = inet_bhash2_update_saddr(sk, saddr, AF_INET6);
+               if (err)
+                       goto failure;
        }
 
        /* set the source address */
--

2.30.2

#6 lore raw

Re: [PATCH v3 net 3/4] dccp/tcp: Update saddr under bhash's lock.

From:   Kuniyuki Iwashima <[email protected]>
Date:   Fri, 18 Nov 2022 15:57:53 -0800
Quoted 228 lines
From:   Joanne Koong <[email protected]>
Date:   Fri, 18 Nov 2022 15:20:20 -0800
Quoted 201 lines · Kuniyuki Iwashima
Quoted 6 lines

When we call connect() for a socket bound to a wildcard address, we update saddr locklessly. However, it could result in a data race; another thread iterating over bhash might see a corrupted address.

Let's update saddr under the bhash bucket's lock.

Thanks for the quick turnaround!

Quoted 166 lines
Fixes 3df80d9320bc ("[DCCP]: Introduce DCCPv6")
Fixes 7c657876b63c ("[DCCP]: Initial implementation")
Fixes 1da177e4c3f4 ("Linux-2.6.12-rc2")
---
 include/net/inet_hashtables.h |  2 +-
 net/dccp/ipv4.c               | 22 +++-----------
 net/dccp/ipv6.c               | 23 +++------------
 net/ipv4/af_inet.c            | 11 +------
 net/ipv4/inet_hashtables.c    | 55 +++++++++++++++++++++++++++++------
 net/ipv4/tcp_ipv4.c           | 20 +++----------
 net/ipv6/tcp_ipv6.c           | 19 ++----------
 7 files changed, 63 insertions(+), 89 deletions(-)
include/net/inet_hashtables.h +1 -1
index 3af1e927247d..ba06e8b52264 100644
@@ -281,7 +281,7 @@ inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in
  * sk_v6_rcv_saddr (ipv6) changes after it has been binded. The socket's
  * rcv_saddr field should already have been updated when this is called.
  */
-int inet_bhash2_update_saddr(struct inet_bind_hashbucket *prev_saddr, struct sock *sk);
+int inet_bhash2_update_saddr(struct sock *sk, void *saddr, int family);
 
 void inet_bind_hash(struct sock *sk, struct inet_bind_bucket *tb,
                    struct inet_bind2_bucket *tb2, unsigned short port);
net/dccp/ipv4.c +4 -17
index 40640c26680e..95e376e3b911 100644
@@ -45,11 +45,10 @@ static unsigned int dccp_v4_pernet_id __read_mostly;
 int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 {
        const struct sockaddr_in *usin = (struct sockaddr_in *)uaddr;
-       struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
-       __be32 daddr, nexthop, prev_sk_rcv_saddr;
        struct inet_sock *inet = inet_sk(sk);
        struct dccp_sock *dp = dccp_sk(sk);
        __be16 orig_sport, orig_dport;
+       __be32 daddr, nexthop;
        struct flowi4 *fl4;
        struct rtable *rt;
        int err;
@@ -91,26 +90,13 @@ int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
                daddr = fl4->daddr;
 
        if (inet->inet_saddr = 0) {
-               if (inet_csk(sk)->icsk_bind2_hash) {
-                       prev_addr_hashbucket > > > -                               inet_bhashfn_portaddr(&dccp_hashinfo, sk,
-                                                     sock_net(sk),
-                                                     inet->inet_num);
-                       prev_sk_rcv_saddr = sk->sk_rcv_saddr;
-               }
-               inet->inet_saddr = fl4->saddr;
-       }
-
-       sk_rcv_saddr_set(sk, inet->inet_saddr);
-
-       if (prev_addr_hashbucket) {
-               err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
+               err = inet_bhash2_update_saddr(sk,  &fl4->saddr, AF_INET);
                if (err) {
-                       inet->inet_saddr = 0;
-                       sk_rcv_saddr_set(sk, prev_sk_rcv_saddr);
                        ip_rt_put(rt);
                        return err;
                }
+       } else {
+               sk_rcv_saddr_set(sk, inet->inet_saddr);
        }
 
        inet->inet_dport = usin->sin_port;
net/dccp/ipv6.c +4 -19
index 626166cb6d7e..94c101ed57a9 100644
@@ -934,26 +934,11 @@ static int dccp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
        }
 
        if (saddr = NULL) {
-               struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
-               struct in6_addr prev_v6_rcv_saddr;
-
-               if (icsk->icsk_bind2_hash) {
-                       prev_addr_hashbucket = inet_bhashfn_portaddr(&dccp_hashinfo,
-                                                                    sk, sock_net(sk),
-                                                                    inet->inet_num);
-                       prev_v6_rcv_saddr = sk->sk_v6_rcv_saddr;
-               }
-
                saddr = &fl6.saddr;
-               sk->sk_v6_rcv_saddr = *saddr;
-
-               if (prev_addr_hashbucket) {
-                       err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
-                       if (err) {
-                               sk->sk_v6_rcv_saddr = prev_v6_rcv_saddr;
-                               goto failure;
-                       }
-               }
+
+               err = inet_bhash2_update_saddr(sk, saddr, AF_INET6);
+               if (err)
+                       goto failure;
        }
 
        /* set the source address */
net/ipv4/af_inet.c +1 -9
index 4728087c42a5..0da679411330 100644
@@ -1230,7 +1230,6 @@ EXPORT_SYMBOL(inet_unregister_protosw);
 
 static int inet_sk_reselect_saddr(struct sock *sk)
 {
-       struct inet_bind_hashbucket *prev_addr_hashbucket;
        struct inet_sock *inet = inet_sk(sk);
        __be32 old_saddr = inet->inet_saddr;
        __be32 daddr = inet->inet_daddr;
@@ -1260,16 +1259,8 @@ static int inet_sk_reselect_saddr(struct sock *sk)
                return 0;
        }
 
-       prev_addr_hashbucket > > > -               inet_bhashfn_portaddr(tcp_or_dccp_get_hashinfo(sk), sk,
-                                     sock_net(sk), inet->inet_num);
-
-       inet->inet_saddr = inet->inet_rcv_saddr = new_saddr;
-
-       err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
+       err = inet_bhash2_update_saddr(sk, &new_saddr, AF_INET);
        if (err) {
-               inet->inet_saddr = old_saddr;
-               inet->inet_rcv_saddr = old_saddr;
                ip_rt_put(rt);
                return err;
        }
net/ipv4/inet_hashtables.c +15 -1
index d745f962745e..fce0bd62d6b5 100644
@@ -858,31 +858,65 @@ inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in
        return &hinfo->bhash2[hash & (hinfo->bhash_size - 1)];
 }
 
-int inet_bhash2_update_saddr(struct inet_bind_hashbucket *prev_saddr, struct sock *sk)
+static void inet_update_saddr(struct sock *sk, void *saddr, int family)
+{
+       if (family = AF_INET) {
+               inet_sk(sk)->inet_saddr = *(__be32 *)saddr;
+               sk_rcv_saddr_set(sk, inet_sk(sk)->inet_saddr);
+       }
+#if IS_ENABLED(CONFIG_IPV6)
+       else {
+               sk->sk_v6_rcv_saddr = *(struct in6_addr *)saddr;
+       }
+#endif
+}
+
+int inet_bhash2_update_saddr(struct sock *sk, void *saddr, int family)
 {
        struct inet_hashinfo *hinfo = tcp_or_dccp_get_hashinfo(sk);
+       struct inet_bind_hashbucket *head, *head2;
        struct inet_bind2_bucket *tb2, *new_tb2;
        int l3mdev = inet_sk_bound_l3mdev(sk);

net/ipv4/inet_hashtables.c inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in · line 888

Quoted 10 lines
net/ipv4/inet_hashtables.c +7 -1
-       struct inet_bind_hashbucket *head2;
        int port = inet_sk(sk)->inet_num;
        struct net *net = sock_net(sk);
+       int bhash, err = 0;
+
+       if (!inet_csk(sk)->icsk_bind2_hash) {
+               /* Not bind()ed before. */
+               inet_update_saddr(sk, saddr, family);
+               goto out;
+       }

I think it would be cleaner if this logic were outside bhash2_update_saddr(), since this mutates the sk's address when the socket hasn't been previously bound to bhash2. I think something like this would be clearer:

static int inet_update_saddr(struct sock *sk, void *saddr, int family)

{
    if (!inet_csk(sk)->icsk_bind2_hash) {
      update_sk_saddr(sk, saddr, family)
      return 0;
    }
    return inet_bhash2_update_saddr(sk, saddr, family);
}
Quoted 6 lines

and then from dccp/tcp_ipv4/6_connect(), we just call inet_update_saddr(). This also "moves" the lower-level implementation details (eg underlying bind tables) to inet_hashtables.c, instead of it being mentioned in the higher dccp_tcp_ipv4/6 layers.

What are your thoughts?

Sounds good!
I'll change them like above.

Quoted 13 lines

net/ipv4/inet_hashtables.c inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in · line 896

Quoted 9 lines
net/ipv4/inet_hashtables.c +8 -0
+
+       bhash = inet_bhashfn(net, port, hinfo->bhash_size);
+       head = &hinfo->bhash[bhash];
+
+       /* If we change saddr locklessly, another thread
+        * iterating over bhash might see corrupted address.
+        */
+       spin_lock_bh(&head->lock);

I don't think we should be acquiring the bhash lock here. I think we only need to acquire it right before we mutate the saddr, and we can release it right after.

Quoted 1 line

Exactly, will move it down before "__"inet_update_saddr() :)

And I'll move this down after the first inet_bhashfn_portaddr() to align the same lock semantics with these 3 functions which aquire the bhash's lock before updating a bhash2 bucket.

* inet_csk_get_port()
* __inet_hash_connect()
* inet_put_port()

Quoted 135 lines

Thank you!

Quoted 131 lines
Quoted 131 lines
net/ipv4/inet_hashtables.c +16 -7
        /* Allocate a bind2 bucket ahead of time to avoid permanently putting
         * the bhash2 table in an inconsistent state if a new tb2 bucket
         * allocation fails.
         */
        new_tb2 = kmem_cache_alloc(hinfo->bind2_bucket_cachep, GFP_ATOMIC);
-       if (!new_tb2)
-               return -ENOMEM;
+       if (!new_tb2) {
+               err = -ENOMEM;
+               goto unlock;
+       }
 
        head2 = inet_bhashfn_portaddr(hinfo, sk, net, port);
 
-       spin_lock_bh(&prev_saddr->lock);
+       spin_lock(&head2->lock);
        __sk_del_bind2_node(sk);
        inet_bind2_bucket_destroy(hinfo->bind2_bucket_cachep, inet_csk(sk)->icsk_bind2_hash);
-       spin_unlock_bh(&prev_saddr->lock);
+       spin_unlock(&head2->lock);
+
+       inet_update_saddr(sk, saddr, family);
 
-       spin_lock_bh(&head2->lock);
+       head2 = inet_bhashfn_portaddr(hinfo, sk, net, port);
+
+       spin_lock(&head2->lock);
        tb2 = inet_bind2_bucket_find(head2, net, port, l3mdev, sk);
        if (!tb2) {
                tb2 = new_tb2;
@@ -890,12 +924,15 @@ int inet_bhash2_update_saddr(struct inet_bind_hashbucket *prev_saddr, struct soc
        }
        sk_add_bind2_node(sk, &tb2->owners);
        inet_csk(sk)->icsk_bind2_hash = tb2;
-       spin_unlock_bh(&head2->lock);
+       spin_unlock(&head2->lock);
 
        if (tb2 != new_tb2)
                kmem_cache_free(hinfo->bind2_bucket_cachep, new_tb2);
 
-       return 0;
+unlock:
+       spin_unlock_bh(&head->lock);
+out:
+       return err;
 }
 EXPORT_SYMBOL_GPL(inet_bhash2_update_saddr);
 
net/ipv4/tcp_ipv4.c +4 -16
index 6a3a732b584d..23dd7e9df2d5 100644
@@ -199,15 +199,14 @@ static int tcp_v4_pre_connect(struct sock *sk, struct sockaddr *uaddr,
 /* This will initiate an outgoing connection. */
 int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 {
-       struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
        struct sockaddr_in *usin = (struct sockaddr_in *)uaddr;
        struct inet_timewait_death_row *tcp_death_row;
-       __be32 daddr, nexthop, prev_sk_rcv_saddr;
        struct inet_sock *inet = inet_sk(sk);
        struct tcp_sock *tp = tcp_sk(sk);
        struct ip_options_rcu *inet_opt;
        struct net *net = sock_net(sk);
        __be16 orig_sport, orig_dport;
+       __be32 daddr, nexthop;
        struct flowi4 *fl4;
        struct rtable *rt;
        int err;
@@ -251,24 +250,13 @@ int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
        tcp_death_row = &sock_net(sk)->ipv4.tcp_death_row;
 
        if (!inet->inet_saddr) {
-               if (inet_csk(sk)->icsk_bind2_hash) {
-                       prev_addr_hashbucket = inet_bhashfn_portaddr(tcp_death_row->hashinfo,
-                                                                    sk, net, inet->inet_num);
-                       prev_sk_rcv_saddr = sk->sk_rcv_saddr;
-               }
-               inet->inet_saddr = fl4->saddr;
-       }
-
-       sk_rcv_saddr_set(sk, inet->inet_saddr);
-
-       if (prev_addr_hashbucket) {
-               err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
+               err = inet_bhash2_update_saddr(sk,  &fl4->saddr, AF_INET);
                if (err) {
-                       inet->inet_saddr = 0;
-                       sk_rcv_saddr_set(sk, prev_sk_rcv_saddr);
                        ip_rt_put(rt);
                        return err;
                }
+       } else {
+               sk_rcv_saddr_set(sk, inet->inet_saddr);
        }
 
        if (tp->rx_opt.ts_recent_stamp && inet->inet_daddr != daddr) {
net/ipv6/tcp_ipv6.c +3 -17
index 81b396e5cf79..2f3ca3190d26 100644
@@ -292,24 +292,11 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
        tcp_death_row = &sock_net(sk)->ipv4.tcp_death_row;
 
        if (!saddr) {
-               struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
-               struct in6_addr prev_v6_rcv_saddr;
-
-               if (icsk->icsk_bind2_hash) {
-                       prev_addr_hashbucket = inet_bhashfn_portaddr(tcp_death_row->hashinfo,
-                                                                    sk, net, inet->inet_num);
-                       prev_v6_rcv_saddr = sk->sk_v6_rcv_saddr;
-               }
                saddr = &fl6.saddr;
-               sk->sk_v6_rcv_saddr = *saddr;
 
-               if (prev_addr_hashbucket) {
-                       err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
-                       if (err) {
-                               sk->sk_v6_rcv_saddr = prev_v6_rcv_saddr;
-                               goto failure;
-                       }
-               }
+               err = inet_bhash2_update_saddr(sk, saddr, AF_INET6);
+               if (err)
+                       goto failure;
        }
 
        /* set the source address */
--

2.30.2

#7 lore raw

Re: [PATCH v3 net 3/4] dccp/tcp: Update saddr under bhash's lock.

In reply to Joanne Koong
From:   Joanne Koong <[email protected]>
Date:   Fri, 18 Nov 2022 15:20:20 -0800
Quoted 201 lines · Kuniyuki Iwashima
Quoted 6 lines

When we call connect() for a socket bound to a wildcard address, we update saddr locklessly. However, it could result in a data race; another thread iterating over bhash might see a corrupted address.

Let's update saddr under the bhash bucket's lock.

Thanks for the quick turnaround!

Quoted 166 lines
Fixes 3df80d9320bc ("[DCCP]: Introduce DCCPv6")
Fixes 7c657876b63c ("[DCCP]: Initial implementation")
Fixes 1da177e4c3f4 ("Linux-2.6.12-rc2")
---
 include/net/inet_hashtables.h |  2 +-
 net/dccp/ipv4.c               | 22 +++-----------
 net/dccp/ipv6.c               | 23 +++------------
 net/ipv4/af_inet.c            | 11 +------
 net/ipv4/inet_hashtables.c    | 55 +++++++++++++++++++++++++++++------
 net/ipv4/tcp_ipv4.c           | 20 +++----------
 net/ipv6/tcp_ipv6.c           | 19 ++----------
 7 files changed, 63 insertions(+), 89 deletions(-)
include/net/inet_hashtables.h +1 -1
index 3af1e927247d..ba06e8b52264 100644
@@ -281,7 +281,7 @@ inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in
  * sk_v6_rcv_saddr (ipv6) changes after it has been binded. The socket's
  * rcv_saddr field should already have been updated when this is called.
  */
-int inet_bhash2_update_saddr(struct inet_bind_hashbucket *prev_saddr, struct sock *sk);
+int inet_bhash2_update_saddr(struct sock *sk, void *saddr, int family);
 
 void inet_bind_hash(struct sock *sk, struct inet_bind_bucket *tb,
                    struct inet_bind2_bucket *tb2, unsigned short port);
net/dccp/ipv4.c +4 -17
index 40640c26680e..95e376e3b911 100644
@@ -45,11 +45,10 @@ static unsigned int dccp_v4_pernet_id __read_mostly;
 int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 {
        const struct sockaddr_in *usin = (struct sockaddr_in *)uaddr;
-       struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
-       __be32 daddr, nexthop, prev_sk_rcv_saddr;
        struct inet_sock *inet = inet_sk(sk);
        struct dccp_sock *dp = dccp_sk(sk);
        __be16 orig_sport, orig_dport;
+       __be32 daddr, nexthop;
        struct flowi4 *fl4;
        struct rtable *rt;
        int err;
@@ -91,26 +90,13 @@ int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
                daddr = fl4->daddr;
 
        if (inet->inet_saddr = 0) {
-               if (inet_csk(sk)->icsk_bind2_hash) {
-                       prev_addr_hashbucket > > -                               inet_bhashfn_portaddr(&dccp_hashinfo, sk,
-                                                     sock_net(sk),
-                                                     inet->inet_num);
-                       prev_sk_rcv_saddr = sk->sk_rcv_saddr;
-               }
-               inet->inet_saddr = fl4->saddr;
-       }
-
-       sk_rcv_saddr_set(sk, inet->inet_saddr);
-
-       if (prev_addr_hashbucket) {
-               err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
+               err = inet_bhash2_update_saddr(sk,  &fl4->saddr, AF_INET);
                if (err) {
-                       inet->inet_saddr = 0;
-                       sk_rcv_saddr_set(sk, prev_sk_rcv_saddr);
                        ip_rt_put(rt);
                        return err;
                }
+       } else {
+               sk_rcv_saddr_set(sk, inet->inet_saddr);
        }
 
        inet->inet_dport = usin->sin_port;
net/dccp/ipv6.c +4 -19
index 626166cb6d7e..94c101ed57a9 100644
@@ -934,26 +934,11 @@ static int dccp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
        }
 
        if (saddr = NULL) {
-               struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
-               struct in6_addr prev_v6_rcv_saddr;
-
-               if (icsk->icsk_bind2_hash) {
-                       prev_addr_hashbucket = inet_bhashfn_portaddr(&dccp_hashinfo,
-                                                                    sk, sock_net(sk),
-                                                                    inet->inet_num);
-                       prev_v6_rcv_saddr = sk->sk_v6_rcv_saddr;
-               }
-
                saddr = &fl6.saddr;
-               sk->sk_v6_rcv_saddr = *saddr;
-
-               if (prev_addr_hashbucket) {
-                       err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
-                       if (err) {
-                               sk->sk_v6_rcv_saddr = prev_v6_rcv_saddr;
-                               goto failure;
-                       }
-               }
+
+               err = inet_bhash2_update_saddr(sk, saddr, AF_INET6);
+               if (err)
+                       goto failure;
        }
 
        /* set the source address */
net/ipv4/af_inet.c +1 -9
index 4728087c42a5..0da679411330 100644
@@ -1230,7 +1230,6 @@ EXPORT_SYMBOL(inet_unregister_protosw);
 
 static int inet_sk_reselect_saddr(struct sock *sk)
 {
-       struct inet_bind_hashbucket *prev_addr_hashbucket;
        struct inet_sock *inet = inet_sk(sk);
        __be32 old_saddr = inet->inet_saddr;
        __be32 daddr = inet->inet_daddr;
@@ -1260,16 +1259,8 @@ static int inet_sk_reselect_saddr(struct sock *sk)
                return 0;
        }
 
-       prev_addr_hashbucket > > -               inet_bhashfn_portaddr(tcp_or_dccp_get_hashinfo(sk), sk,
-                                     sock_net(sk), inet->inet_num);
-
-       inet->inet_saddr = inet->inet_rcv_saddr = new_saddr;
-
-       err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
+       err = inet_bhash2_update_saddr(sk, &new_saddr, AF_INET);
        if (err) {
-               inet->inet_saddr = old_saddr;
-               inet->inet_rcv_saddr = old_saddr;
                ip_rt_put(rt);
                return err;
        }
net/ipv4/inet_hashtables.c +15 -1
index d745f962745e..fce0bd62d6b5 100644
@@ -858,31 +858,65 @@ inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in
        return &hinfo->bhash2[hash & (hinfo->bhash_size - 1)];
 }
 
-int inet_bhash2_update_saddr(struct inet_bind_hashbucket *prev_saddr, struct sock *sk)
+static void inet_update_saddr(struct sock *sk, void *saddr, int family)
+{
+       if (family = AF_INET) {
+               inet_sk(sk)->inet_saddr = *(__be32 *)saddr;
+               sk_rcv_saddr_set(sk, inet_sk(sk)->inet_saddr);
+       }
+#if IS_ENABLED(CONFIG_IPV6)
+       else {
+               sk->sk_v6_rcv_saddr = *(struct in6_addr *)saddr;
+       }
+#endif
+}
+
+int inet_bhash2_update_saddr(struct sock *sk, void *saddr, int family)
 {
        struct inet_hashinfo *hinfo = tcp_or_dccp_get_hashinfo(sk);
+       struct inet_bind_hashbucket *head, *head2;
        struct inet_bind2_bucket *tb2, *new_tb2;
        int l3mdev = inet_sk_bound_l3mdev(sk);

net/ipv4/inet_hashtables.c inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in · line 888

Quoted 10 lines
net/ipv4/inet_hashtables.c +7 -1
-       struct inet_bind_hashbucket *head2;
        int port = inet_sk(sk)->inet_num;
        struct net *net = sock_net(sk);
+       int bhash, err = 0;
+
+       if (!inet_csk(sk)->icsk_bind2_hash) {
+               /* Not bind()ed before. */
+               inet_update_saddr(sk, saddr, family);
+               goto out;
+       }

I think it would be cleaner if this logic were outside bhash2_update_saddr(), since this mutates the sk's address when the socket hasn't been previously bound to bhash2. I think something like this would be clearer:

static int inet_update_saddr(struct sock *sk, void *saddr, int family)

{
    if (!inet_csk(sk)->icsk_bind2_hash) {
      update_sk_saddr(sk, saddr, family)
      return 0;
    }
    return inet_bhash2_update_saddr(sk, saddr, family);
}
Quoted 6 lines

and then from dccp/tcp_ipv4/6_connect(), we just call inet_update_saddr(). This also "moves" the lower-level implementation details (eg underlying bind tables) to inet_hashtables.c, instead of it being mentioned in the higher dccp_tcp_ipv4/6 layers.

What are your thoughts?

Hmm.. There are other users of inet_reset_saddr(). If we hide the details of reset & bhash2 logic too, we need a cleanup patch below.

So, I'll keep this part as is in the next spin. If needed, I can post a followup for net-next.

And I noticed UDP has the same problem, this will be another series.

Thank you!

---8<---

include/net/inet_hashtables.h +1 -0
index 95a18c029d5b..6ac7d5b8cbed 100644
@@ -277,6 +277,7 @@ inet_bhashfn_portaddr(const struct inet_hashinfo *hinfo, const struct sock *sk,
 struct inet_bind_hashbucket *
 inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, int port);
 
+void inet_reset_saddr(struct sock *sk);
 int inet_update_saddr(struct sock *sk, void *saddr, int family);
 
 void inet_bind_hash(struct sock *sk, struct inet_bind_bucket *tb,
include/net/ip.h +1 -1
index 144bdfbb25af..43939a235398 100644
@@ -632,7 +632,7 @@ static inline void ip_ipgre_mc_map(__be32 naddr, const unsigned char *broadcast,
 #include <linux/ipv6.h>
 #endif
 
-static __inline__ void inet_reset_saddr(struct sock *sk)
+static __inline__ void __inet_reset_saddr(struct sock *sk)
 {
 	inet_sk(sk)->inet_rcv_saddr = inet_sk(sk)->inet_saddr = 0;
 #if IS_ENABLED(CONFIG_IPV6)
net/ipv4/inet_hashtables.c +6 -0
index b0ababa74c02..4171077b127c 100644
@@ -941,6 +941,12 @@ int inet_update_saddr(struct sock *sk, void *saddr, int family)
 }
 EXPORT_SYMBOL_GPL(inet_update_saddr);
 
+void inet_reset_saddr(struct sock *sk)
+{
+	__inet_reset_saddr(sk);
+}
+EXPORT_SYMBOL_GPL(inet_reset_saddr);
+
 /* RFC 6056 3.3.4.  Algorithm 4: Double-Hash Port Selection Algorithm
  * Note that we use 32bit integers (vs RFC 'short integers')
  * because 2^16 is not a multiple of num_ephemeral and this
net/ipv4/udp.c +1 -1
index 6a320a614e54..a3c64866207d 100644
@@ -1971,7 +1971,7 @@ int __udp_disconnect(struct sock *sk, int flags)
 	sock_rps_reset_rxhash(sk);
 	sk->sk_bound_dev_if = 0;
 	if (!(sk->sk_userlocks & SOCK_BINDADDR_LOCK)) {
-		inet_reset_saddr(sk);
+		__inet_reset_saddr(sk);
 		if (sk->sk_prot->rehash &&
 		    (sk->sk_userlocks & SOCK_BINDPORT_LOCK))
 			sk->sk_prot->rehash(sk);
net/ipv6/af_inet6.c +2 -2
index 024191004982..fea2de97ba54 100644
@@ -411,7 +411,7 @@ static int __inet6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len,
 		      (flags & BIND_FORCE_ADDRESS_NO_PORT))) {
 		if (sk->sk_prot->get_port(sk, snum)) {
 			sk->sk_ipv6only = saved_ipv6only;
-			inet_reset_saddr(sk);
+			__inet_reset_saddr(sk);
 			err = -EADDRINUSE;
 			goto out;
 		}
@@ -419,7 +419,7 @@ static int __inet6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len,
 			err = BPF_CGROUP_RUN_PROG_INET6_POST_BIND(sk);
 			if (err) {
 				sk->sk_ipv6only = saved_ipv6only;
-				inet_reset_saddr(sk);
+				__inet_reset_saddr(sk);
 				if (sk->sk_prot->put_port)
 					sk->sk_prot->put_port(sk);
 				goto out;

---8<---

Quoted 146 lines

net/ipv4/inet_hashtables.c inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in · line 896

Quoted 9 lines
net/ipv4/inet_hashtables.c +8 -0
+
+       bhash = inet_bhashfn(net, port, hinfo->bhash_size);
+       head = &hinfo->bhash[bhash];
+
+       /* If we change saddr locklessly, another thread
+        * iterating over bhash might see corrupted address.
+        */
+       spin_lock_bh(&head->lock);

I don't think we should be acquiring the bhash lock here. I think we only need to acquire it right before we mutate the saddr, and we can release it right after.

Quoted 132 lines
net/ipv4/inet_hashtables.c +16 -7
        /* Allocate a bind2 bucket ahead of time to avoid permanently putting
         * the bhash2 table in an inconsistent state if a new tb2 bucket
         * allocation fails.
         */
        new_tb2 = kmem_cache_alloc(hinfo->bind2_bucket_cachep, GFP_ATOMIC);
-       if (!new_tb2)
-               return -ENOMEM;
+       if (!new_tb2) {
+               err = -ENOMEM;
+               goto unlock;
+       }
 
        head2 = inet_bhashfn_portaddr(hinfo, sk, net, port);
 
-       spin_lock_bh(&prev_saddr->lock);
+       spin_lock(&head2->lock);
        __sk_del_bind2_node(sk);
        inet_bind2_bucket_destroy(hinfo->bind2_bucket_cachep, inet_csk(sk)->icsk_bind2_hash);
-       spin_unlock_bh(&prev_saddr->lock);
+       spin_unlock(&head2->lock);
+
+       inet_update_saddr(sk, saddr, family);
 
-       spin_lock_bh(&head2->lock);
+       head2 = inet_bhashfn_portaddr(hinfo, sk, net, port);
+
+       spin_lock(&head2->lock);
        tb2 = inet_bind2_bucket_find(head2, net, port, l3mdev, sk);
        if (!tb2) {
                tb2 = new_tb2;
@@ -890,12 +924,15 @@ int inet_bhash2_update_saddr(struct inet_bind_hashbucket *prev_saddr, struct soc
        }
        sk_add_bind2_node(sk, &tb2->owners);
        inet_csk(sk)->icsk_bind2_hash = tb2;
-       spin_unlock_bh(&head2->lock);
+       spin_unlock(&head2->lock);
 
        if (tb2 != new_tb2)
                kmem_cache_free(hinfo->bind2_bucket_cachep, new_tb2);
 
-       return 0;
+unlock:
+       spin_unlock_bh(&head->lock);
+out:
+       return err;
 }
 EXPORT_SYMBOL_GPL(inet_bhash2_update_saddr);
 
net/ipv4/tcp_ipv4.c +4 -16
index 6a3a732b584d..23dd7e9df2d5 100644
@@ -199,15 +199,14 @@ static int tcp_v4_pre_connect(struct sock *sk, struct sockaddr *uaddr,
 /* This will initiate an outgoing connection. */
 int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 {
-       struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
        struct sockaddr_in *usin = (struct sockaddr_in *)uaddr;
        struct inet_timewait_death_row *tcp_death_row;
-       __be32 daddr, nexthop, prev_sk_rcv_saddr;
        struct inet_sock *inet = inet_sk(sk);
        struct tcp_sock *tp = tcp_sk(sk);
        struct ip_options_rcu *inet_opt;
        struct net *net = sock_net(sk);
        __be16 orig_sport, orig_dport;
+       __be32 daddr, nexthop;
        struct flowi4 *fl4;
        struct rtable *rt;
        int err;
@@ -251,24 +250,13 @@ int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
        tcp_death_row = &sock_net(sk)->ipv4.tcp_death_row;
 
        if (!inet->inet_saddr) {
-               if (inet_csk(sk)->icsk_bind2_hash) {
-                       prev_addr_hashbucket = inet_bhashfn_portaddr(tcp_death_row->hashinfo,
-                                                                    sk, net, inet->inet_num);
-                       prev_sk_rcv_saddr = sk->sk_rcv_saddr;
-               }
-               inet->inet_saddr = fl4->saddr;
-       }
-
-       sk_rcv_saddr_set(sk, inet->inet_saddr);
-
-       if (prev_addr_hashbucket) {
-               err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
+               err = inet_bhash2_update_saddr(sk,  &fl4->saddr, AF_INET);
                if (err) {
-                       inet->inet_saddr = 0;
-                       sk_rcv_saddr_set(sk, prev_sk_rcv_saddr);
                        ip_rt_put(rt);
                        return err;
                }
+       } else {
+               sk_rcv_saddr_set(sk, inet->inet_saddr);
        }
 
        if (tp->rx_opt.ts_recent_stamp && inet->inet_daddr != daddr) {
net/ipv6/tcp_ipv6.c +3 -17
index 81b396e5cf79..2f3ca3190d26 100644
@@ -292,24 +292,11 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
        tcp_death_row = &sock_net(sk)->ipv4.tcp_death_row;
 
        if (!saddr) {
-               struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
-               struct in6_addr prev_v6_rcv_saddr;
-
-               if (icsk->icsk_bind2_hash) {
-                       prev_addr_hashbucket = inet_bhashfn_portaddr(tcp_death_row->hashinfo,
-                                                                    sk, net, inet->inet_num);
-                       prev_v6_rcv_saddr = sk->sk_v6_rcv_saddr;
-               }
                saddr = &fl6.saddr;
-               sk->sk_v6_rcv_saddr = *saddr;
 
-               if (prev_addr_hashbucket) {
-                       err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
-                       if (err) {
-                               sk->sk_v6_rcv_saddr = prev_v6_rcv_saddr;
-                               goto failure;
-                       }
-               }
+               err = inet_bhash2_update_saddr(sk, saddr, AF_INET6);
+               if (err)
+                       goto failure;
        }
 
        /* set the source address */
--

2.30.2

#8 lore raw

Re: [PATCH v3 net 3/4] dccp/tcp: Update saddr under bhash's lock.

Quoted 209 lines · Kuniyuki Iwashima
From:   Joanne Koong <[email protected]>
Date:   Fri, 18 Nov 2022 15:20:20 -0800
Quoted 206 lines · Kuniyuki Iwashima
Quoted 6 lines

When we call connect() for a socket bound to a wildcard address, we update saddr locklessly. However, it could result in a data race; another thread iterating over bhash might see a corrupted address.

Let's update saddr under the bhash bucket's lock.

Thanks for the quick turnaround!

Quoted 166 lines
Fixes 3df80d9320bc ("[DCCP]: Introduce DCCPv6")
Fixes 7c657876b63c ("[DCCP]: Initial implementation")
Fixes 1da177e4c3f4 ("Linux-2.6.12-rc2")
---
 include/net/inet_hashtables.h |  2 +-
 net/dccp/ipv4.c               | 22 +++-----------
 net/dccp/ipv6.c               | 23 +++------------
 net/ipv4/af_inet.c            | 11 +------
 net/ipv4/inet_hashtables.c    | 55 +++++++++++++++++++++++++++++------
 net/ipv4/tcp_ipv4.c           | 20 +++----------
 net/ipv6/tcp_ipv6.c           | 19 ++----------
 7 files changed, 63 insertions(+), 89 deletions(-)
include/net/inet_hashtables.h +1 -1
index 3af1e927247d..ba06e8b52264 100644
@@ -281,7 +281,7 @@ inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in
  * sk_v6_rcv_saddr (ipv6) changes after it has been binded. The socket's
  * rcv_saddr field should already have been updated when this is called.
  */
-int inet_bhash2_update_saddr(struct inet_bind_hashbucket *prev_saddr, struct sock *sk);
+int inet_bhash2_update_saddr(struct sock *sk, void *saddr, int family);
 
 void inet_bind_hash(struct sock *sk, struct inet_bind_bucket *tb,
                    struct inet_bind2_bucket *tb2, unsigned short port);
net/dccp/ipv4.c +4 -17
index 40640c26680e..95e376e3b911 100644
@@ -45,11 +45,10 @@ static unsigned int dccp_v4_pernet_id __read_mostly;
 int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 {
        const struct sockaddr_in *usin = (struct sockaddr_in *)uaddr;
-       struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
-       __be32 daddr, nexthop, prev_sk_rcv_saddr;
        struct inet_sock *inet = inet_sk(sk);
        struct dccp_sock *dp = dccp_sk(sk);
        __be16 orig_sport, orig_dport;
+       __be32 daddr, nexthop;
        struct flowi4 *fl4;
        struct rtable *rt;
        int err;
@@ -91,26 +90,13 @@ int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
                daddr = fl4->daddr;
 
        if (inet->inet_saddr = 0) {
-               if (inet_csk(sk)->icsk_bind2_hash) {
-                       prev_addr_hashbucket > > > -                               inet_bhashfn_portaddr(&dccp_hashinfo, sk,
-                                                     sock_net(sk),
-                                                     inet->inet_num);
-                       prev_sk_rcv_saddr = sk->sk_rcv_saddr;
-               }
-               inet->inet_saddr = fl4->saddr;
-       }
-
-       sk_rcv_saddr_set(sk, inet->inet_saddr);
-
-       if (prev_addr_hashbucket) {
-               err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
+               err = inet_bhash2_update_saddr(sk,  &fl4->saddr, AF_INET);
                if (err) {
-                       inet->inet_saddr = 0;
-                       sk_rcv_saddr_set(sk, prev_sk_rcv_saddr);
                        ip_rt_put(rt);
                        return err;
                }
+       } else {
+               sk_rcv_saddr_set(sk, inet->inet_saddr);
        }
 
        inet->inet_dport = usin->sin_port;
net/dccp/ipv6.c +4 -19
index 626166cb6d7e..94c101ed57a9 100644
@@ -934,26 +934,11 @@ static int dccp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
        }
 
        if (saddr = NULL) {
-               struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
-               struct in6_addr prev_v6_rcv_saddr;
-
-               if (icsk->icsk_bind2_hash) {
-                       prev_addr_hashbucket = inet_bhashfn_portaddr(&dccp_hashinfo,
-                                                                    sk, sock_net(sk),
-                                                                    inet->inet_num);
-                       prev_v6_rcv_saddr = sk->sk_v6_rcv_saddr;
-               }
-
                saddr = &fl6.saddr;
-               sk->sk_v6_rcv_saddr = *saddr;
-
-               if (prev_addr_hashbucket) {
-                       err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
-                       if (err) {
-                               sk->sk_v6_rcv_saddr = prev_v6_rcv_saddr;
-                               goto failure;
-                       }
-               }
+
+               err = inet_bhash2_update_saddr(sk, saddr, AF_INET6);
+               if (err)
+                       goto failure;
        }
 
        /* set the source address */
net/ipv4/af_inet.c +1 -9
index 4728087c42a5..0da679411330 100644
@@ -1230,7 +1230,6 @@ EXPORT_SYMBOL(inet_unregister_protosw);
 
 static int inet_sk_reselect_saddr(struct sock *sk)
 {
-       struct inet_bind_hashbucket *prev_addr_hashbucket;
        struct inet_sock *inet = inet_sk(sk);
        __be32 old_saddr = inet->inet_saddr;
        __be32 daddr = inet->inet_daddr;
@@ -1260,16 +1259,8 @@ static int inet_sk_reselect_saddr(struct sock *sk)
                return 0;
        }
 
-       prev_addr_hashbucket > > > -               inet_bhashfn_portaddr(tcp_or_dccp_get_hashinfo(sk), sk,
-                                     sock_net(sk), inet->inet_num);
-
-       inet->inet_saddr = inet->inet_rcv_saddr = new_saddr;
-
-       err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
+       err = inet_bhash2_update_saddr(sk, &new_saddr, AF_INET);
        if (err) {
-               inet->inet_saddr = old_saddr;
-               inet->inet_rcv_saddr = old_saddr;
                ip_rt_put(rt);
                return err;
        }
net/ipv4/inet_hashtables.c +15 -1
index d745f962745e..fce0bd62d6b5 100644
@@ -858,31 +858,65 @@ inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in
        return &hinfo->bhash2[hash & (hinfo->bhash_size - 1)];
 }
 
-int inet_bhash2_update_saddr(struct inet_bind_hashbucket *prev_saddr, struct sock *sk)
+static void inet_update_saddr(struct sock *sk, void *saddr, int family)
+{
+       if (family = AF_INET) {
+               inet_sk(sk)->inet_saddr = *(__be32 *)saddr;
+               sk_rcv_saddr_set(sk, inet_sk(sk)->inet_saddr);
+       }
+#if IS_ENABLED(CONFIG_IPV6)
+       else {
+               sk->sk_v6_rcv_saddr = *(struct in6_addr *)saddr;
+       }
+#endif
+}
+
+int inet_bhash2_update_saddr(struct sock *sk, void *saddr, int family)
 {
        struct inet_hashinfo *hinfo = tcp_or_dccp_get_hashinfo(sk);
+       struct inet_bind_hashbucket *head, *head2;
        struct inet_bind2_bucket *tb2, *new_tb2;
        int l3mdev = inet_sk_bound_l3mdev(sk);

net/ipv4/inet_hashtables.c inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in · line 888

Quoted 10 lines
net/ipv4/inet_hashtables.c +7 -1
-       struct inet_bind_hashbucket *head2;
        int port = inet_sk(sk)->inet_num;
        struct net *net = sock_net(sk);
+       int bhash, err = 0;
+
+       if (!inet_csk(sk)->icsk_bind2_hash) {
+               /* Not bind()ed before. */
+               inet_update_saddr(sk, saddr, family);
+               goto out;
+       }

I think it would be cleaner if this logic were outside bhash2_update_saddr(), since this mutates the sk's address when the socket hasn't been previously bound to bhash2. I think something like this would be clearer:

static int inet_update_saddr(struct sock *sk, void *saddr, int family)

{
    if (!inet_csk(sk)->icsk_bind2_hash) {
      update_sk_saddr(sk, saddr, family)
      return 0;
    }
    return inet_bhash2_update_saddr(sk, saddr, family);
}

and then from dccp/tcp_ipv4/6_connect(), we just call inet_update_saddr(). This also "moves" the lower-level implementation details (eg underlying bind tables) to inet_hashtables.c, instead of it being mentioned in the higher dccp_tcp_ipv4/6 layers.

Quoted 4 lines
Quoted 1 line

What are your thoughts?

Hmm.. There are other users of inet_reset_saddr(). If we hide the details of reset & bhash2 logic too, we need a cleanup patch below.

I don't think reset needs to be hidden / abstracted away, that sounds generic. But bhash2 sounds like an implementation detail, which is why it seems cleaner to me if we didn't have bhash2 mentioned in the dccp/tcp_ipv4/6 layers.

Quoted 5 lines

So, I'll keep this part as is in the next spin. If needed, I can post a followup for net-next.

And I noticed UDP has the same problem, this will be another series.

Sounds good, looking forward to your UDP patchset.

Quoted 231 lines

Thank you!

---8<---

include/net/inet_hashtables.h +1 -0
index 95a18c029d5b..6ac7d5b8cbed 100644
@@ -277,6 +277,7 @@ inet_bhashfn_portaddr(const struct inet_hashinfo *hinfo, const struct sock *sk,
 struct inet_bind_hashbucket *
 inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, int port);
 
+void inet_reset_saddr(struct sock *sk);
 int inet_update_saddr(struct sock *sk, void *saddr, int family);
 
 void inet_bind_hash(struct sock *sk, struct inet_bind_bucket *tb,
include/net/ip.h +1 -1
index 144bdfbb25af..43939a235398 100644
@@ -632,7 +632,7 @@ static inline void ip_ipgre_mc_map(__be32 naddr, const unsigned char *broadcast,
 #include <linux/ipv6.h>
 #endif
 
-static __inline__ void inet_reset_saddr(struct sock *sk)
+static __inline__ void __inet_reset_saddr(struct sock *sk)
 {
        inet_sk(sk)->inet_rcv_saddr = inet_sk(sk)->inet_saddr = 0;
 #if IS_ENABLED(CONFIG_IPV6)
net/ipv4/inet_hashtables.c +6 -0
index b0ababa74c02..4171077b127c 100644
@@ -941,6 +941,12 @@ int inet_update_saddr(struct sock *sk, void *saddr, int family)
 }
 EXPORT_SYMBOL_GPL(inet_update_saddr);
 
+void inet_reset_saddr(struct sock *sk)
+{
+       __inet_reset_saddr(sk);
+}
+EXPORT_SYMBOL_GPL(inet_reset_saddr);
+
 /* RFC 6056 3.3.4.  Algorithm 4: Double-Hash Port Selection Algorithm
  * Note that we use 32bit integers (vs RFC 'short integers')
  * because 2^16 is not a multiple of num_ephemeral and this
net/ipv4/udp.c +1 -1
index 6a320a614e54..a3c64866207d 100644
@@ -1971,7 +1971,7 @@ int __udp_disconnect(struct sock *sk, int flags)
        sock_rps_reset_rxhash(sk);
        sk->sk_bound_dev_if = 0;
        if (!(sk->sk_userlocks & SOCK_BINDADDR_LOCK)) {
-               inet_reset_saddr(sk);
+               __inet_reset_saddr(sk);
                if (sk->sk_prot->rehash &&
                    (sk->sk_userlocks & SOCK_BINDPORT_LOCK))
                        sk->sk_prot->rehash(sk);
net/ipv6/af_inet6.c +2 -3
index 024191004982..fea2de97ba54 100644
@@ -411,7 +411,7 @@ static int __inet6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len,
                      (flags & BIND_FORCE_ADDRESS_NO_PORT))) {
                if (sk->sk_prot->get_port(sk, snum)) {
                        sk->sk_ipv6only = saved_ipv6only;
-                       inet_reset_saddr(sk);
+                       __inet_reset_saddr(sk);
                        err = -EADDRINUSE;
                        goto out;
                }
@@ -419,7 +419,7 @@ static int __inet6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len,
                        err = BPF_CGROUP_RUN_PROG_INET6_POST_BIND(sk);
                        if (err) {
                                sk->sk_ipv6only = saved_ipv6only;
-                               inet_reset_saddr(sk);
+                               __inet_reset_saddr(sk);
                                if (sk->sk_prot->put_port)
                                        sk->sk_prot->put_port(sk);
                                goto out;
 
---8<---
Quoted 148 lines

net/ipv4/inet_hashtables.c inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in · line 896

Quoted 11 lines
net/ipv4/inet_hashtables.c +8 -0
+
+       bhash = inet_bhashfn(net, port, hinfo->bhash_size);
+       head = &hinfo->bhash[bhash];
+
+       /* If we change saddr locklessly, another thread
+        * iterating over bhash might see corrupted address.
+        */
+       spin_lock_bh(&head->lock);

I don't think we should be acquiring the bhash lock here. I think we only need to acquire it right before we mutate the saddr, and we can release it right after.

Quoted 132 lines
net/ipv4/inet_hashtables.c +16 -7
        /* Allocate a bind2 bucket ahead of time to avoid permanently putting
         * the bhash2 table in an inconsistent state if a new tb2 bucket
         * allocation fails.
         */
        new_tb2 = kmem_cache_alloc(hinfo->bind2_bucket_cachep, GFP_ATOMIC);
-       if (!new_tb2)
-               return -ENOMEM;
+       if (!new_tb2) {
+               err = -ENOMEM;
+               goto unlock;
+       }
 
        head2 = inet_bhashfn_portaddr(hinfo, sk, net, port);
 
-       spin_lock_bh(&prev_saddr->lock);
+       spin_lock(&head2->lock);
        __sk_del_bind2_node(sk);
        inet_bind2_bucket_destroy(hinfo->bind2_bucket_cachep, inet_csk(sk)->icsk_bind2_hash);
-       spin_unlock_bh(&prev_saddr->lock);
+       spin_unlock(&head2->lock);
+
+       inet_update_saddr(sk, saddr, family);
 
-       spin_lock_bh(&head2->lock);
+       head2 = inet_bhashfn_portaddr(hinfo, sk, net, port);
+
+       spin_lock(&head2->lock);
        tb2 = inet_bind2_bucket_find(head2, net, port, l3mdev, sk);
        if (!tb2) {
                tb2 = new_tb2;
@@ -890,12 +924,15 @@ int inet_bhash2_update_saddr(struct inet_bind_hashbucket *prev_saddr, struct soc
        }
        sk_add_bind2_node(sk, &tb2->owners);
        inet_csk(sk)->icsk_bind2_hash = tb2;
-       spin_unlock_bh(&head2->lock);
+       spin_unlock(&head2->lock);
 
        if (tb2 != new_tb2)
                kmem_cache_free(hinfo->bind2_bucket_cachep, new_tb2);
 
-       return 0;
+unlock:
+       spin_unlock_bh(&head->lock);
+out:
+       return err;
 }
 EXPORT_SYMBOL_GPL(inet_bhash2_update_saddr);
 
net/ipv4/tcp_ipv4.c +4 -16
index 6a3a732b584d..23dd7e9df2d5 100644
@@ -199,15 +199,14 @@ static int tcp_v4_pre_connect(struct sock *sk, struct sockaddr *uaddr,
 /* This will initiate an outgoing connection. */
 int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 {
-       struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
        struct sockaddr_in *usin = (struct sockaddr_in *)uaddr;
        struct inet_timewait_death_row *tcp_death_row;
-       __be32 daddr, nexthop, prev_sk_rcv_saddr;
        struct inet_sock *inet = inet_sk(sk);
        struct tcp_sock *tp = tcp_sk(sk);
        struct ip_options_rcu *inet_opt;
        struct net *net = sock_net(sk);
        __be16 orig_sport, orig_dport;
+       __be32 daddr, nexthop;
        struct flowi4 *fl4;
        struct rtable *rt;
        int err;
@@ -251,24 +250,13 @@ int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
        tcp_death_row = &sock_net(sk)->ipv4.tcp_death_row;
 
        if (!inet->inet_saddr) {
-               if (inet_csk(sk)->icsk_bind2_hash) {
-                       prev_addr_hashbucket = inet_bhashfn_portaddr(tcp_death_row->hashinfo,
-                                                                    sk, net, inet->inet_num);
-                       prev_sk_rcv_saddr = sk->sk_rcv_saddr;
-               }
-               inet->inet_saddr = fl4->saddr;
-       }
-
-       sk_rcv_saddr_set(sk, inet->inet_saddr);
-
-       if (prev_addr_hashbucket) {
-               err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
+               err = inet_bhash2_update_saddr(sk,  &fl4->saddr, AF_INET);
                if (err) {
-                       inet->inet_saddr = 0;
-                       sk_rcv_saddr_set(sk, prev_sk_rcv_saddr);
                        ip_rt_put(rt);
                        return err;
                }
+       } else {
+               sk_rcv_saddr_set(sk, inet->inet_saddr);
        }
 
        if (tp->rx_opt.ts_recent_stamp && inet->inet_daddr != daddr) {
net/ipv6/tcp_ipv6.c +3 -17
index 81b396e5cf79..2f3ca3190d26 100644
@@ -292,24 +292,11 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
        tcp_death_row = &sock_net(sk)->ipv4.tcp_death_row;
 
        if (!saddr) {
-               struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
-               struct in6_addr prev_v6_rcv_saddr;
-
-               if (icsk->icsk_bind2_hash) {
-                       prev_addr_hashbucket = inet_bhashfn_portaddr(tcp_death_row->hashinfo,
-                                                                    sk, net, inet->inet_num);
-                       prev_v6_rcv_saddr = sk->sk_v6_rcv_saddr;
-               }
                saddr = &fl6.saddr;
-               sk->sk_v6_rcv_saddr = *saddr;
 
-               if (prev_addr_hashbucket) {
-                       err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
-                       if (err) {
-                               sk->sk_v6_rcv_saddr = prev_v6_rcv_saddr;
-                               goto failure;
-                       }
-               }
+               err = inet_bhash2_update_saddr(sk, saddr, AF_INET6);
+               if (err)
+                       goto failure;
        }
 
        /* set the source address */
--

2.30.2

#9 lore raw

Re: [PATCH v3 net 3/4] dccp/tcp: Update saddr under bhash's lock.

From:   Joanne Koong <[email protected]>
Date:   Mon, 21 Nov 2022 14:52:56 -0800
Quoted 212 lines · Kuniyuki Iwashima
Quoted 211 lines
From:   Joanne Koong <[email protected]>
Date:   Fri, 18 Nov 2022 15:20:20 -0800
Quoted 207 lines · Kuniyuki Iwashima
Quoted 6 lines

When we call connect() for a socket bound to a wildcard address, we update saddr locklessly. However, it could result in a data race; another thread iterating over bhash might see a corrupted address.

Let's update saddr under the bhash bucket's lock.

Thanks for the quick turnaround!

Quoted 166 lines
Fixes 3df80d9320bc ("[DCCP]: Introduce DCCPv6")
Fixes 7c657876b63c ("[DCCP]: Initial implementation")
Fixes 1da177e4c3f4 ("Linux-2.6.12-rc2")
---
 include/net/inet_hashtables.h |  2 +-
 net/dccp/ipv4.c               | 22 +++-----------
 net/dccp/ipv6.c               | 23 +++------------
 net/ipv4/af_inet.c            | 11 +------
 net/ipv4/inet_hashtables.c    | 55 +++++++++++++++++++++++++++++------
 net/ipv4/tcp_ipv4.c           | 20 +++----------
 net/ipv6/tcp_ipv6.c           | 19 ++----------
 7 files changed, 63 insertions(+), 89 deletions(-)
include/net/inet_hashtables.h +1 -1
index 3af1e927247d..ba06e8b52264 100644
@@ -281,7 +281,7 @@ inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in
  * sk_v6_rcv_saddr (ipv6) changes after it has been binded. The socket's
  * rcv_saddr field should already have been updated when this is called.
  */
-int inet_bhash2_update_saddr(struct inet_bind_hashbucket *prev_saddr, struct sock *sk);
+int inet_bhash2_update_saddr(struct sock *sk, void *saddr, int family);
 
 void inet_bind_hash(struct sock *sk, struct inet_bind_bucket *tb,
                    struct inet_bind2_bucket *tb2, unsigned short port);
net/dccp/ipv4.c +4 -17
index 40640c26680e..95e376e3b911 100644
@@ -45,11 +45,10 @@ static unsigned int dccp_v4_pernet_id __read_mostly;
 int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 {
        const struct sockaddr_in *usin = (struct sockaddr_in *)uaddr;
-       struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
-       __be32 daddr, nexthop, prev_sk_rcv_saddr;
        struct inet_sock *inet = inet_sk(sk);
        struct dccp_sock *dp = dccp_sk(sk);
        __be16 orig_sport, orig_dport;
+       __be32 daddr, nexthop;
        struct flowi4 *fl4;
        struct rtable *rt;
        int err;
@@ -91,26 +90,13 @@ int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
                daddr = fl4->daddr;
 
        if (inet->inet_saddr = 0) {
-               if (inet_csk(sk)->icsk_bind2_hash) {
-                       prev_addr_hashbucket > > > > -                               inet_bhashfn_portaddr(&dccp_hashinfo, sk,
-                                                     sock_net(sk),
-                                                     inet->inet_num);
-                       prev_sk_rcv_saddr = sk->sk_rcv_saddr;
-               }
-               inet->inet_saddr = fl4->saddr;
-       }
-
-       sk_rcv_saddr_set(sk, inet->inet_saddr);
-
-       if (prev_addr_hashbucket) {
-               err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
+               err = inet_bhash2_update_saddr(sk,  &fl4->saddr, AF_INET);
                if (err) {
-                       inet->inet_saddr = 0;
-                       sk_rcv_saddr_set(sk, prev_sk_rcv_saddr);
                        ip_rt_put(rt);
                        return err;
                }
+       } else {
+               sk_rcv_saddr_set(sk, inet->inet_saddr);
        }
 
        inet->inet_dport = usin->sin_port;
net/dccp/ipv6.c +4 -19
index 626166cb6d7e..94c101ed57a9 100644
@@ -934,26 +934,11 @@ static int dccp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
        }
 
        if (saddr = NULL) {
-               struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
-               struct in6_addr prev_v6_rcv_saddr;
-
-               if (icsk->icsk_bind2_hash) {
-                       prev_addr_hashbucket = inet_bhashfn_portaddr(&dccp_hashinfo,
-                                                                    sk, sock_net(sk),
-                                                                    inet->inet_num);
-                       prev_v6_rcv_saddr = sk->sk_v6_rcv_saddr;
-               }
-
                saddr = &fl6.saddr;
-               sk->sk_v6_rcv_saddr = *saddr;
-
-               if (prev_addr_hashbucket) {
-                       err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
-                       if (err) {
-                               sk->sk_v6_rcv_saddr = prev_v6_rcv_saddr;
-                               goto failure;
-                       }
-               }
+
+               err = inet_bhash2_update_saddr(sk, saddr, AF_INET6);
+               if (err)
+                       goto failure;
        }
 
        /* set the source address */
net/ipv4/af_inet.c +1 -9
index 4728087c42a5..0da679411330 100644
@@ -1230,7 +1230,6 @@ EXPORT_SYMBOL(inet_unregister_protosw);
 
 static int inet_sk_reselect_saddr(struct sock *sk)
 {
-       struct inet_bind_hashbucket *prev_addr_hashbucket;
        struct inet_sock *inet = inet_sk(sk);
        __be32 old_saddr = inet->inet_saddr;
        __be32 daddr = inet->inet_daddr;
@@ -1260,16 +1259,8 @@ static int inet_sk_reselect_saddr(struct sock *sk)
                return 0;
        }
 
-       prev_addr_hashbucket > > > > -               inet_bhashfn_portaddr(tcp_or_dccp_get_hashinfo(sk), sk,
-                                     sock_net(sk), inet->inet_num);
-
-       inet->inet_saddr = inet->inet_rcv_saddr = new_saddr;
-
-       err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
+       err = inet_bhash2_update_saddr(sk, &new_saddr, AF_INET);
        if (err) {
-               inet->inet_saddr = old_saddr;
-               inet->inet_rcv_saddr = old_saddr;
                ip_rt_put(rt);
                return err;
        }
net/ipv4/inet_hashtables.c +15 -1
index d745f962745e..fce0bd62d6b5 100644
@@ -858,31 +858,65 @@ inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in
        return &hinfo->bhash2[hash & (hinfo->bhash_size - 1)];
 }
 
-int inet_bhash2_update_saddr(struct inet_bind_hashbucket *prev_saddr, struct sock *sk)
+static void inet_update_saddr(struct sock *sk, void *saddr, int family)
+{
+       if (family = AF_INET) {
+               inet_sk(sk)->inet_saddr = *(__be32 *)saddr;
+               sk_rcv_saddr_set(sk, inet_sk(sk)->inet_saddr);
+       }
+#if IS_ENABLED(CONFIG_IPV6)
+       else {
+               sk->sk_v6_rcv_saddr = *(struct in6_addr *)saddr;
+       }
+#endif
+}
+
+int inet_bhash2_update_saddr(struct sock *sk, void *saddr, int family)
 {
        struct inet_hashinfo *hinfo = tcp_or_dccp_get_hashinfo(sk);
+       struct inet_bind_hashbucket *head, *head2;
        struct inet_bind2_bucket *tb2, *new_tb2;
        int l3mdev = inet_sk_bound_l3mdev(sk);

net/ipv4/inet_hashtables.c inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in · line 888

Quoted 10 lines
net/ipv4/inet_hashtables.c +7 -1
-       struct inet_bind_hashbucket *head2;
        int port = inet_sk(sk)->inet_num;
        struct net *net = sock_net(sk);
+       int bhash, err = 0;
+
+       if (!inet_csk(sk)->icsk_bind2_hash) {
+               /* Not bind()ed before. */
+               inet_update_saddr(sk, saddr, family);
+               goto out;
+       }

I think it would be cleaner if this logic were outside bhash2_update_saddr(), since this mutates the sk's address when the socket hasn't been previously bound to bhash2. I think something like this would be clearer:

static int inet_update_saddr(struct sock *sk, void *saddr, int family)

{
    if (!inet_csk(sk)->icsk_bind2_hash) {
      update_sk_saddr(sk, saddr, family)
      return 0;
    }
    return inet_bhash2_update_saddr(sk, saddr, family);
}

and then from dccp/tcp_ipv4/6_connect(), we just call inet_update_saddr(). This also "moves" the lower-level implementation details (eg underlying bind tables) to inet_hashtables.c, instead of it being mentioned in the higher dccp_tcp_ipv4/6 layers.

What are your thoughts?

Quoted 7 lines
Quoted 2 lines

Hmm.. There are other users of inet_reset_saddr(). If we hide the details of reset & bhash2 logic too, we need a cleanup patch below.

I don't think reset needs to be hidden / abstracted away, that sounds generic. But bhash2 sounds like an implementation detail, which is why it seems cleaner to me if we didn't have bhash2 mentioned in the dccp/tcp_ipv4/6 layers.

Sorry, my description was not clear.

Now we have inet_bhash2_update_saddr() and inet_bhash2_reset_saddr().

If we hide bhash2 logic as inet_update_saddr(),

- inet_update_saddr
- if (...)
- __inet_update_saddr
- inet_bhash2_update_saddr
- __inet_update_saddr

then we will change inet_bhash2_reset_saddr() logic like

- inet_reset_saddr
- if (...)
- __inet_reset_saddr
- inet_bhash2_reset_saddr
- __inet_reset_saddr

inet_update_saddr() is what we newly added, but inet_reset_saddr() is not. There are existing users of inet_reset_saddr(), e.g. UDP, and we need a patch below.

If we don't clean up, other protocol not using bhash2 also go through this condition via inet_reset_saddr(), which is originally unnecessary.

if (!inet_csk(sk)->icsk_bind2_hash)

What do you think ?

Quoted 238 lines
Quoted 4 lines

So, I'll keep this part as is in the next spin. If needed, I can post a followup for net-next.

And I noticed UDP has the same problem, this will be another series.

Sounds good, looking forward to your UDP patchset.

Quoted 231 lines

Thank you!

---8<---

include/net/inet_hashtables.h +1 -0
index 95a18c029d5b..6ac7d5b8cbed 100644
@@ -277,6 +277,7 @@ inet_bhashfn_portaddr(const struct inet_hashinfo *hinfo, const struct sock *sk,
 struct inet_bind_hashbucket *
 inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, int port);
 
+void inet_reset_saddr(struct sock *sk);
 int inet_update_saddr(struct sock *sk, void *saddr, int family);
 
 void inet_bind_hash(struct sock *sk, struct inet_bind_bucket *tb,
include/net/ip.h +1 -1
index 144bdfbb25af..43939a235398 100644
@@ -632,7 +632,7 @@ static inline void ip_ipgre_mc_map(__be32 naddr, const unsigned char *broadcast,
 #include <linux/ipv6.h>
 #endif
 
-static __inline__ void inet_reset_saddr(struct sock *sk)
+static __inline__ void __inet_reset_saddr(struct sock *sk)
 {
        inet_sk(sk)->inet_rcv_saddr = inet_sk(sk)->inet_saddr = 0;
 #if IS_ENABLED(CONFIG_IPV6)
net/ipv4/inet_hashtables.c +6 -0
index b0ababa74c02..4171077b127c 100644
@@ -941,6 +941,12 @@ int inet_update_saddr(struct sock *sk, void *saddr, int family)
 }
 EXPORT_SYMBOL_GPL(inet_update_saddr);
 
+void inet_reset_saddr(struct sock *sk)
+{
+       __inet_reset_saddr(sk);
+}
+EXPORT_SYMBOL_GPL(inet_reset_saddr);
+
 /* RFC 6056 3.3.4.  Algorithm 4: Double-Hash Port Selection Algorithm
  * Note that we use 32bit integers (vs RFC 'short integers')
  * because 2^16 is not a multiple of num_ephemeral and this
net/ipv4/udp.c +1 -1
index 6a320a614e54..a3c64866207d 100644
@@ -1971,7 +1971,7 @@ int __udp_disconnect(struct sock *sk, int flags)
        sock_rps_reset_rxhash(sk);
        sk->sk_bound_dev_if = 0;
        if (!(sk->sk_userlocks & SOCK_BINDADDR_LOCK)) {
-               inet_reset_saddr(sk);
+               __inet_reset_saddr(sk);
                if (sk->sk_prot->rehash &&
                    (sk->sk_userlocks & SOCK_BINDPORT_LOCK))
                        sk->sk_prot->rehash(sk);
net/ipv6/af_inet6.c +2 -3
index 024191004982..fea2de97ba54 100644
@@ -411,7 +411,7 @@ static int __inet6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len,
                      (flags & BIND_FORCE_ADDRESS_NO_PORT))) {
                if (sk->sk_prot->get_port(sk, snum)) {
                        sk->sk_ipv6only = saved_ipv6only;
-                       inet_reset_saddr(sk);
+                       __inet_reset_saddr(sk);
                        err = -EADDRINUSE;
                        goto out;
                }
@@ -419,7 +419,7 @@ static int __inet6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len,
                        err = BPF_CGROUP_RUN_PROG_INET6_POST_BIND(sk);
                        if (err) {
                                sk->sk_ipv6only = saved_ipv6only;
-                               inet_reset_saddr(sk);
+                               __inet_reset_saddr(sk);
                                if (sk->sk_prot->put_port)
                                        sk->sk_prot->put_port(sk);
                                goto out;
 
---8<---
Quoted 148 lines

net/ipv4/inet_hashtables.c inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in · line 896

Quoted 11 lines
net/ipv4/inet_hashtables.c +8 -0
+
+       bhash = inet_bhashfn(net, port, hinfo->bhash_size);
+       head = &hinfo->bhash[bhash];
+
+       /* If we change saddr locklessly, another thread
+        * iterating over bhash might see corrupted address.
+        */
+       spin_lock_bh(&head->lock);

I don't think we should be acquiring the bhash lock here. I think we only need to acquire it right before we mutate the saddr, and we can release it right after.

Quoted 132 lines
net/ipv4/inet_hashtables.c +16 -7
        /* Allocate a bind2 bucket ahead of time to avoid permanently putting
         * the bhash2 table in an inconsistent state if a new tb2 bucket
         * allocation fails.
         */
        new_tb2 = kmem_cache_alloc(hinfo->bind2_bucket_cachep, GFP_ATOMIC);
-       if (!new_tb2)
-               return -ENOMEM;
+       if (!new_tb2) {
+               err = -ENOMEM;
+               goto unlock;
+       }
 
        head2 = inet_bhashfn_portaddr(hinfo, sk, net, port);
 
-       spin_lock_bh(&prev_saddr->lock);
+       spin_lock(&head2->lock);
        __sk_del_bind2_node(sk);
        inet_bind2_bucket_destroy(hinfo->bind2_bucket_cachep, inet_csk(sk)->icsk_bind2_hash);
-       spin_unlock_bh(&prev_saddr->lock);
+       spin_unlock(&head2->lock);
+
+       inet_update_saddr(sk, saddr, family);
 
-       spin_lock_bh(&head2->lock);
+       head2 = inet_bhashfn_portaddr(hinfo, sk, net, port);
+
+       spin_lock(&head2->lock);
        tb2 = inet_bind2_bucket_find(head2, net, port, l3mdev, sk);
        if (!tb2) {
                tb2 = new_tb2;
@@ -890,12 +924,15 @@ int inet_bhash2_update_saddr(struct inet_bind_hashbucket *prev_saddr, struct soc
        }
        sk_add_bind2_node(sk, &tb2->owners);
        inet_csk(sk)->icsk_bind2_hash = tb2;
-       spin_unlock_bh(&head2->lock);
+       spin_unlock(&head2->lock);
 
        if (tb2 != new_tb2)
                kmem_cache_free(hinfo->bind2_bucket_cachep, new_tb2);
 
-       return 0;
+unlock:
+       spin_unlock_bh(&head->lock);
+out:
+       return err;
 }
 EXPORT_SYMBOL_GPL(inet_bhash2_update_saddr);
 
net/ipv4/tcp_ipv4.c +4 -16
index 6a3a732b584d..23dd7e9df2d5 100644
@@ -199,15 +199,14 @@ static int tcp_v4_pre_connect(struct sock *sk, struct sockaddr *uaddr,
 /* This will initiate an outgoing connection. */
 int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 {
-       struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
        struct sockaddr_in *usin = (struct sockaddr_in *)uaddr;
        struct inet_timewait_death_row *tcp_death_row;
-       __be32 daddr, nexthop, prev_sk_rcv_saddr;
        struct inet_sock *inet = inet_sk(sk);
        struct tcp_sock *tp = tcp_sk(sk);
        struct ip_options_rcu *inet_opt;
        struct net *net = sock_net(sk);
        __be16 orig_sport, orig_dport;
+       __be32 daddr, nexthop;
        struct flowi4 *fl4;
        struct rtable *rt;
        int err;
@@ -251,24 +250,13 @@ int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
        tcp_death_row = &sock_net(sk)->ipv4.tcp_death_row;
 
        if (!inet->inet_saddr) {
-               if (inet_csk(sk)->icsk_bind2_hash) {
-                       prev_addr_hashbucket = inet_bhashfn_portaddr(tcp_death_row->hashinfo,
-                                                                    sk, net, inet->inet_num);
-                       prev_sk_rcv_saddr = sk->sk_rcv_saddr;
-               }
-               inet->inet_saddr = fl4->saddr;
-       }
-
-       sk_rcv_saddr_set(sk, inet->inet_saddr);
-
-       if (prev_addr_hashbucket) {
-               err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
+               err = inet_bhash2_update_saddr(sk,  &fl4->saddr, AF_INET);
                if (err) {
-                       inet->inet_saddr = 0;
-                       sk_rcv_saddr_set(sk, prev_sk_rcv_saddr);
                        ip_rt_put(rt);
                        return err;
                }
+       } else {
+               sk_rcv_saddr_set(sk, inet->inet_saddr);
        }
 
        if (tp->rx_opt.ts_recent_stamp && inet->inet_daddr != daddr) {
net/ipv6/tcp_ipv6.c +3 -17
index 81b396e5cf79..2f3ca3190d26 100644
@@ -292,24 +292,11 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
        tcp_death_row = &sock_net(sk)->ipv4.tcp_death_row;
 
        if (!saddr) {
-               struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
-               struct in6_addr prev_v6_rcv_saddr;
-
-               if (icsk->icsk_bind2_hash) {
-                       prev_addr_hashbucket = inet_bhashfn_portaddr(tcp_death_row->hashinfo,
-                                                                    sk, net, inet->inet_num);
-                       prev_v6_rcv_saddr = sk->sk_v6_rcv_saddr;
-               }
                saddr = &fl6.saddr;
-               sk->sk_v6_rcv_saddr = *saddr;
 
-               if (prev_addr_hashbucket) {
-                       err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
-                       if (err) {
-                               sk->sk_v6_rcv_saddr = prev_v6_rcv_saddr;
-                               goto failure;
-                       }
-               }
+               err = inet_bhash2_update_saddr(sk, saddr, AF_INET6);
+               if (err)
+                       goto failure;
        }
 
        /* set the source address */
--

2.30.2

#10 lore raw

Re: [PATCH v3 net 3/4] dccp/tcp: Update saddr under bhash's lock.

Quoted 247 lines · Kuniyuki Iwashima
From:   Joanne Koong <[email protected]>
Date:   Mon, 21 Nov 2022 14:52:56 -0800
Quoted 212 lines · Kuniyuki Iwashima
Quoted 211 lines
From:   Joanne Koong <[email protected]>
Date:   Fri, 18 Nov 2022 15:20:20 -0800
Quoted 207 lines · Kuniyuki Iwashima
Quoted 6 lines

When we call connect() for a socket bound to a wildcard address, we update saddr locklessly. However, it could result in a data race; another thread iterating over bhash might see a corrupted address.

Let's update saddr under the bhash bucket's lock.

Thanks for the quick turnaround!

Quoted 166 lines
Fixes 3df80d9320bc ("[DCCP]: Introduce DCCPv6")
Fixes 7c657876b63c ("[DCCP]: Initial implementation")
Fixes 1da177e4c3f4 ("Linux-2.6.12-rc2")
---
 include/net/inet_hashtables.h |  2 +-
 net/dccp/ipv4.c               | 22 +++-----------
 net/dccp/ipv6.c               | 23 +++------------
 net/ipv4/af_inet.c            | 11 +------
 net/ipv4/inet_hashtables.c    | 55 +++++++++++++++++++++++++++++------
 net/ipv4/tcp_ipv4.c           | 20 +++----------
 net/ipv6/tcp_ipv6.c           | 19 ++----------
 7 files changed, 63 insertions(+), 89 deletions(-)
include/net/inet_hashtables.h +1 -1
index 3af1e927247d..ba06e8b52264 100644
@@ -281,7 +281,7 @@ inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in
  * sk_v6_rcv_saddr (ipv6) changes after it has been binded. The socket's
  * rcv_saddr field should already have been updated when this is called.
  */
-int inet_bhash2_update_saddr(struct inet_bind_hashbucket *prev_saddr, struct sock *sk);
+int inet_bhash2_update_saddr(struct sock *sk, void *saddr, int family);
 
 void inet_bind_hash(struct sock *sk, struct inet_bind_bucket *tb,
                    struct inet_bind2_bucket *tb2, unsigned short port);
net/dccp/ipv4.c +4 -17
index 40640c26680e..95e376e3b911 100644
@@ -45,11 +45,10 @@ static unsigned int dccp_v4_pernet_id __read_mostly;
 int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 {
        const struct sockaddr_in *usin = (struct sockaddr_in *)uaddr;
-       struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
-       __be32 daddr, nexthop, prev_sk_rcv_saddr;
        struct inet_sock *inet = inet_sk(sk);
        struct dccp_sock *dp = dccp_sk(sk);
        __be16 orig_sport, orig_dport;
+       __be32 daddr, nexthop;
        struct flowi4 *fl4;
        struct rtable *rt;
        int err;
@@ -91,26 +90,13 @@ int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
                daddr = fl4->daddr;
 
        if (inet->inet_saddr = 0) {
-               if (inet_csk(sk)->icsk_bind2_hash) {
-                       prev_addr_hashbucket > > > > > -                               inet_bhashfn_portaddr(&dccp_hashinfo, sk,
-                                                     sock_net(sk),
-                                                     inet->inet_num);
-                       prev_sk_rcv_saddr = sk->sk_rcv_saddr;
-               }
-               inet->inet_saddr = fl4->saddr;
-       }
-
-       sk_rcv_saddr_set(sk, inet->inet_saddr);
-
-       if (prev_addr_hashbucket) {
-               err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
+               err = inet_bhash2_update_saddr(sk,  &fl4->saddr, AF_INET);
                if (err) {
-                       inet->inet_saddr = 0;
-                       sk_rcv_saddr_set(sk, prev_sk_rcv_saddr);
                        ip_rt_put(rt);
                        return err;
                }
+       } else {
+               sk_rcv_saddr_set(sk, inet->inet_saddr);
        }
 
        inet->inet_dport = usin->sin_port;
net/dccp/ipv6.c +4 -19
index 626166cb6d7e..94c101ed57a9 100644
@@ -934,26 +934,11 @@ static int dccp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
        }
 
        if (saddr = NULL) {
-               struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
-               struct in6_addr prev_v6_rcv_saddr;
-
-               if (icsk->icsk_bind2_hash) {
-                       prev_addr_hashbucket = inet_bhashfn_portaddr(&dccp_hashinfo,
-                                                                    sk, sock_net(sk),
-                                                                    inet->inet_num);
-                       prev_v6_rcv_saddr = sk->sk_v6_rcv_saddr;
-               }
-
                saddr = &fl6.saddr;
-               sk->sk_v6_rcv_saddr = *saddr;
-
-               if (prev_addr_hashbucket) {
-                       err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
-                       if (err) {
-                               sk->sk_v6_rcv_saddr = prev_v6_rcv_saddr;
-                               goto failure;
-                       }
-               }
+
+               err = inet_bhash2_update_saddr(sk, saddr, AF_INET6);
+               if (err)
+                       goto failure;
        }
 
        /* set the source address */
net/ipv4/af_inet.c +1 -9
index 4728087c42a5..0da679411330 100644
@@ -1230,7 +1230,6 @@ EXPORT_SYMBOL(inet_unregister_protosw);
 
 static int inet_sk_reselect_saddr(struct sock *sk)
 {
-       struct inet_bind_hashbucket *prev_addr_hashbucket;
        struct inet_sock *inet = inet_sk(sk);
        __be32 old_saddr = inet->inet_saddr;
        __be32 daddr = inet->inet_daddr;
@@ -1260,16 +1259,8 @@ static int inet_sk_reselect_saddr(struct sock *sk)
                return 0;
        }
 
-       prev_addr_hashbucket > > > > > -               inet_bhashfn_portaddr(tcp_or_dccp_get_hashinfo(sk), sk,
-                                     sock_net(sk), inet->inet_num);
-
-       inet->inet_saddr = inet->inet_rcv_saddr = new_saddr;
-
-       err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
+       err = inet_bhash2_update_saddr(sk, &new_saddr, AF_INET);
        if (err) {
-               inet->inet_saddr = old_saddr;
-               inet->inet_rcv_saddr = old_saddr;
                ip_rt_put(rt);
                return err;
        }
net/ipv4/inet_hashtables.c +15 -1
index d745f962745e..fce0bd62d6b5 100644
@@ -858,31 +858,65 @@ inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in
        return &hinfo->bhash2[hash & (hinfo->bhash_size - 1)];
 }
 
-int inet_bhash2_update_saddr(struct inet_bind_hashbucket *prev_saddr, struct sock *sk)
+static void inet_update_saddr(struct sock *sk, void *saddr, int family)
+{
+       if (family = AF_INET) {
+               inet_sk(sk)->inet_saddr = *(__be32 *)saddr;
+               sk_rcv_saddr_set(sk, inet_sk(sk)->inet_saddr);
+       }
+#if IS_ENABLED(CONFIG_IPV6)
+       else {
+               sk->sk_v6_rcv_saddr = *(struct in6_addr *)saddr;
+       }
+#endif
+}
+
+int inet_bhash2_update_saddr(struct sock *sk, void *saddr, int family)
 {
        struct inet_hashinfo *hinfo = tcp_or_dccp_get_hashinfo(sk);
+       struct inet_bind_hashbucket *head, *head2;
        struct inet_bind2_bucket *tb2, *new_tb2;
        int l3mdev = inet_sk_bound_l3mdev(sk);

net/ipv4/inet_hashtables.c inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in · line 888

Quoted 10 lines
net/ipv4/inet_hashtables.c +7 -1
-       struct inet_bind_hashbucket *head2;
        int port = inet_sk(sk)->inet_num;
        struct net *net = sock_net(sk);
+       int bhash, err = 0;
+
+       if (!inet_csk(sk)->icsk_bind2_hash) {
+               /* Not bind()ed before. */
+               inet_update_saddr(sk, saddr, family);
+               goto out;
+       }

I think it would be cleaner if this logic were outside bhash2_update_saddr(), since this mutates the sk's address when the socket hasn't been previously bound to bhash2. I think something like this would be clearer:

static int inet_update_saddr(struct sock *sk, void *saddr, int family)

{
    if (!inet_csk(sk)->icsk_bind2_hash) {
      update_sk_saddr(sk, saddr, family)
      return 0;
    }
    return inet_bhash2_update_saddr(sk, saddr, family);
}

and then from dccp/tcp_ipv4/6_connect(), we just call inet_update_saddr(). This also "moves" the lower-level implementation details (eg underlying bind tables) to inet_hashtables.c, instead of it being mentioned in the higher dccp_tcp_ipv4/6 layers.

What are your thoughts?

Quoted 7 lines
Quoted 2 lines

Hmm.. There are other users of inet_reset_saddr(). If we hide the details of reset & bhash2 logic too, we need a cleanup patch below.

I don't think reset needs to be hidden / abstracted away, that sounds generic. But bhash2 sounds like an implementation detail, which is why it seems cleaner to me if we didn't have bhash2 mentioned in the dccp/tcp_ipv4/6 layers.

Sorry, my description was not clear.

Now we have inet_bhash2_update_saddr() and inet_bhash2_reset_saddr().

If we hide bhash2 logic as inet_update_saddr(),

- inet_update_saddr
- if (...)
- __inet_update_saddr
- inet_bhash2_update_saddr
- __inet_update_saddr

then we will change inet_bhash2_reset_saddr() logic like

- inet_reset_saddr
- if (...)
- __inet_reset_saddr
- inet_bhash2_reset_saddr
- __inet_reset_saddr

inet_update_saddr() is what we newly added, but inet_reset_saddr() is not. There are existing users of inet_reset_saddr(), e.g. UDP, and we need a patch below.

Quoted 7 lines

If we don't clean up, other protocol not using bhash2 also go through this condition via inet_reset_saddr(), which is originally unnecessary.

if (!inet_csk(sk)->icsk_bind2_hash)

What do you think ?

Thanks for clarifying! I think this is a minor detail we can revisit in the future if need be, and that we should just go with what you have in this patch.

Quoted 239 lines
Quoted 239 lines
Quoted 5 lines

So, I'll keep this part as is in the next spin. If needed, I can post a followup for net-next.

And I noticed UDP has the same problem, this will be another series.

Sounds good, looking forward to your UDP patchset.

Quoted 231 lines

Thank you!

---8<---

include/net/inet_hashtables.h +1 -0
index 95a18c029d5b..6ac7d5b8cbed 100644
@@ -277,6 +277,7 @@ inet_bhashfn_portaddr(const struct inet_hashinfo *hinfo, const struct sock *sk,
 struct inet_bind_hashbucket *
 inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, int port);
 
+void inet_reset_saddr(struct sock *sk);
 int inet_update_saddr(struct sock *sk, void *saddr, int family);
 
 void inet_bind_hash(struct sock *sk, struct inet_bind_bucket *tb,
include/net/ip.h +1 -1
index 144bdfbb25af..43939a235398 100644
@@ -632,7 +632,7 @@ static inline void ip_ipgre_mc_map(__be32 naddr, const unsigned char *broadcast,
 #include <linux/ipv6.h>
 #endif
 
-static __inline__ void inet_reset_saddr(struct sock *sk)
+static __inline__ void __inet_reset_saddr(struct sock *sk)
 {
        inet_sk(sk)->inet_rcv_saddr = inet_sk(sk)->inet_saddr = 0;
 #if IS_ENABLED(CONFIG_IPV6)
net/ipv4/inet_hashtables.c +6 -0
index b0ababa74c02..4171077b127c 100644
@@ -941,6 +941,12 @@ int inet_update_saddr(struct sock *sk, void *saddr, int family)
 }
 EXPORT_SYMBOL_GPL(inet_update_saddr);
 
+void inet_reset_saddr(struct sock *sk)
+{
+       __inet_reset_saddr(sk);
+}
+EXPORT_SYMBOL_GPL(inet_reset_saddr);
+
 /* RFC 6056 3.3.4.  Algorithm 4: Double-Hash Port Selection Algorithm
  * Note that we use 32bit integers (vs RFC 'short integers')
  * because 2^16 is not a multiple of num_ephemeral and this
net/ipv4/udp.c +1 -1
index 6a320a614e54..a3c64866207d 100644
@@ -1971,7 +1971,7 @@ int __udp_disconnect(struct sock *sk, int flags)
        sock_rps_reset_rxhash(sk);
        sk->sk_bound_dev_if = 0;
        if (!(sk->sk_userlocks & SOCK_BINDADDR_LOCK)) {
-               inet_reset_saddr(sk);
+               __inet_reset_saddr(sk);
                if (sk->sk_prot->rehash &&
                    (sk->sk_userlocks & SOCK_BINDPORT_LOCK))
                        sk->sk_prot->rehash(sk);
net/ipv6/af_inet6.c +2 -3
index 024191004982..fea2de97ba54 100644
@@ -411,7 +411,7 @@ static int __inet6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len,
                      (flags & BIND_FORCE_ADDRESS_NO_PORT))) {
                if (sk->sk_prot->get_port(sk, snum)) {
                        sk->sk_ipv6only = saved_ipv6only;
-                       inet_reset_saddr(sk);
+                       __inet_reset_saddr(sk);
                        err = -EADDRINUSE;
                        goto out;
                }
@@ -419,7 +419,7 @@ static int __inet6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len,
                        err = BPF_CGROUP_RUN_PROG_INET6_POST_BIND(sk);
                        if (err) {
                                sk->sk_ipv6only = saved_ipv6only;
-                               inet_reset_saddr(sk);
+                               __inet_reset_saddr(sk);
                                if (sk->sk_prot->put_port)
                                        sk->sk_prot->put_port(sk);
                                goto out;
 
---8<---
Quoted 148 lines

net/ipv4/inet_hashtables.c inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in · line 896

Quoted 11 lines
net/ipv4/inet_hashtables.c +8 -0
+
+       bhash = inet_bhashfn(net, port, hinfo->bhash_size);
+       head = &hinfo->bhash[bhash];
+
+       /* If we change saddr locklessly, another thread
+        * iterating over bhash might see corrupted address.
+        */
+       spin_lock_bh(&head->lock);

I don't think we should be acquiring the bhash lock here. I think we only need to acquire it right before we mutate the saddr, and we can release it right after.

Quoted 132 lines
net/ipv4/inet_hashtables.c +16 -7
        /* Allocate a bind2 bucket ahead of time to avoid permanently putting
         * the bhash2 table in an inconsistent state if a new tb2 bucket
         * allocation fails.
         */
        new_tb2 = kmem_cache_alloc(hinfo->bind2_bucket_cachep, GFP_ATOMIC);
-       if (!new_tb2)
-               return -ENOMEM;
+       if (!new_tb2) {
+               err = -ENOMEM;
+               goto unlock;
+       }
 
        head2 = inet_bhashfn_portaddr(hinfo, sk, net, port);
 
-       spin_lock_bh(&prev_saddr->lock);
+       spin_lock(&head2->lock);
        __sk_del_bind2_node(sk);
        inet_bind2_bucket_destroy(hinfo->bind2_bucket_cachep, inet_csk(sk)->icsk_bind2_hash);
-       spin_unlock_bh(&prev_saddr->lock);
+       spin_unlock(&head2->lock);
+
+       inet_update_saddr(sk, saddr, family);
 
-       spin_lock_bh(&head2->lock);
+       head2 = inet_bhashfn_portaddr(hinfo, sk, net, port);
+
+       spin_lock(&head2->lock);
        tb2 = inet_bind2_bucket_find(head2, net, port, l3mdev, sk);
        if (!tb2) {
                tb2 = new_tb2;
@@ -890,12 +924,15 @@ int inet_bhash2_update_saddr(struct inet_bind_hashbucket *prev_saddr, struct soc
        }
        sk_add_bind2_node(sk, &tb2->owners);
        inet_csk(sk)->icsk_bind2_hash = tb2;
-       spin_unlock_bh(&head2->lock);
+       spin_unlock(&head2->lock);
 
        if (tb2 != new_tb2)
                kmem_cache_free(hinfo->bind2_bucket_cachep, new_tb2);
 
-       return 0;
+unlock:
+       spin_unlock_bh(&head->lock);
+out:
+       return err;
 }
 EXPORT_SYMBOL_GPL(inet_bhash2_update_saddr);
 
net/ipv4/tcp_ipv4.c +4 -16
index 6a3a732b584d..23dd7e9df2d5 100644
@@ -199,15 +199,14 @@ static int tcp_v4_pre_connect(struct sock *sk, struct sockaddr *uaddr,
 /* This will initiate an outgoing connection. */
 int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 {
-       struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
        struct sockaddr_in *usin = (struct sockaddr_in *)uaddr;
        struct inet_timewait_death_row *tcp_death_row;
-       __be32 daddr, nexthop, prev_sk_rcv_saddr;
        struct inet_sock *inet = inet_sk(sk);
        struct tcp_sock *tp = tcp_sk(sk);
        struct ip_options_rcu *inet_opt;
        struct net *net = sock_net(sk);
        __be16 orig_sport, orig_dport;
+       __be32 daddr, nexthop;
        struct flowi4 *fl4;
        struct rtable *rt;
        int err;
@@ -251,24 +250,13 @@ int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
        tcp_death_row = &sock_net(sk)->ipv4.tcp_death_row;
 
        if (!inet->inet_saddr) {
-               if (inet_csk(sk)->icsk_bind2_hash) {
-                       prev_addr_hashbucket = inet_bhashfn_portaddr(tcp_death_row->hashinfo,
-                                                                    sk, net, inet->inet_num);
-                       prev_sk_rcv_saddr = sk->sk_rcv_saddr;
-               }
-               inet->inet_saddr = fl4->saddr;
-       }
-
-       sk_rcv_saddr_set(sk, inet->inet_saddr);
-
-       if (prev_addr_hashbucket) {
-               err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
+               err = inet_bhash2_update_saddr(sk,  &fl4->saddr, AF_INET);
                if (err) {
-                       inet->inet_saddr = 0;
-                       sk_rcv_saddr_set(sk, prev_sk_rcv_saddr);
                        ip_rt_put(rt);
                        return err;
                }
+       } else {
+               sk_rcv_saddr_set(sk, inet->inet_saddr);
        }
 
        if (tp->rx_opt.ts_recent_stamp && inet->inet_daddr != daddr) {
net/ipv6/tcp_ipv6.c +3 -17
index 81b396e5cf79..2f3ca3190d26 100644
@@ -292,24 +292,11 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
        tcp_death_row = &sock_net(sk)->ipv4.tcp_death_row;
 
        if (!saddr) {
-               struct inet_bind_hashbucket *prev_addr_hashbucket = NULL;
-               struct in6_addr prev_v6_rcv_saddr;
-
-               if (icsk->icsk_bind2_hash) {
-                       prev_addr_hashbucket = inet_bhashfn_portaddr(tcp_death_row->hashinfo,
-                                                                    sk, net, inet->inet_num);
-                       prev_v6_rcv_saddr = sk->sk_v6_rcv_saddr;
-               }
                saddr = &fl6.saddr;
-               sk->sk_v6_rcv_saddr = *saddr;
 
-               if (prev_addr_hashbucket) {
-                       err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk);
-                       if (err) {
-                               sk->sk_v6_rcv_saddr = prev_v6_rcv_saddr;
-                               goto failure;
-                       }
-               }
+               err = inet_bhash2_update_saddr(sk, saddr, AF_INET6);
+               if (err)
+                       goto failure;
        }
 
        /* set the source address */
--

2.30.2

#11 lore raw

PATCH v3 4/4 dccp/tcp: Fixup bhash2 bucket when connect() fails.

In reply to Kuniyuki Iwashima

If a socket bound to a wildcard address fails to connect(), we only reset saddr and keep the port. Then, we have to fix up the bhash2 bucket; otherwise, the bucket has an inconsistent address in the list.

Also, listen() for such a socket will fire the WARN_ON() in inet_csk_get_port(). [0]

Note that when a system runs out of memory, we give up fixing the bucket and unlink sk from bhash and bhash2 by inet_put_port().

[0]:

WARNING: CPU: 0 PID: 207 at net/ipv4/inet_connection_sock.c:548 inet_csk_get_port (net/ipv4/inet_connection_sock.c:548 (discriminator 1))
Modules linked in:
CPU: 0 PID: 207 Comm: bhash2_prev_rep Not tainted 6.1.0-rc3-00799-gc8421681c845 #63
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.0-1.amzn2022.0.1 04/01/2014
RIP: 0010:inet_csk_get_port (net/ipv4/inet_connection_sock.c:548 (discriminator 1))
Code: 74 a7 eb 93 48 8b 54 24 18 0f b7 cb 4c 89 e6 4c 89 ff e8 48 b2 ff ff 49 8b 87 18 04 00 00 e9 32 ff ff ff 0f 0b e9 34 ff ff ff <0f> 0b e9 42 ff ff ff 41 8b 7f 50 41 8b 4f 54 89 fe 81 f6 00 00 ff
RSP: 0018:ffffc900003d7e50 EFLAGS: 00010202
RAX: ffff8881047fb500 RBX: 0000000000004e20 RCX: 0000000000000000
RDX: 000000000000000a RSI: 00000000fffffe00 RDI: 00000000ffffffff
RBP: ffffffff8324dc00 R08: 0000000000000001 R09: 0000000000000001
R10: 0000000000000001 R11: 0000000000000001 R12: 0000000000000000
R13: 0000000000000001 R14: 0000000000004e20 R15: ffff8881054e1280
FS:  00007f8ac04dc740(0000) GS:ffff88842fc00000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000020001540 CR3: 00000001055fa003 CR4: 0000000000770ef0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
PKRU: 55555554
Call Trace:
 <TASK>
 inet_csk_listen_start (net/ipv4/inet_connection_sock.c:1205)
 inet_listen (net/ipv4/af_inet.c:228)
 __sys_listen (net/socket.c:1810)
 __x64_sys_listen (net/socket.c:1819 net/socket.c:1817 net/socket.c:1817)
 do_syscall_64 (arch/x86/entry/common.c:50 arch/x86/entry/common.c:80)
 entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:120)
RIP: 0033:0x7f8ac051de5d
Code: ff c3 66 2e 0f 1f 84 00 00 00 00 00 90 f3 0f 1e fa 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 93 af 1b 00 f7 d8 64 89 01 48
RSP: 002b:00007ffc1c177248 EFLAGS: 00000206 ORIG_RAX: 0000000000000032
RAX: ffffffffffffffda RBX: 0000000020001550 RCX: 00007f8ac051de5d
RDX: ffffffffffffff80 RSI: 0000000000000000 RDI: 0000000000000004
RBP: 00007ffc1c177270 R08: 0000000000000018 R09: 0000000000000007
R10: 0000000020001540 R11: 0000000000000206 R12: 00007ffc1c177388
R13: 0000000000401169 R14: 0000000000403e18 R15: 00007f8ac0723000
 </TASK>
Fixes 28044fc1d495 ("net: Add a bhash2 table hashed by port and address")
---
 include/net/inet_hashtables.h |  1 +
 net/dccp/ipv4.c               |  3 +--
 net/dccp/ipv6.c               |  3 +--
 net/dccp/proto.c              |  3 +--
 net/ipv4/inet_hashtables.c    | 35 ++++++++++++++++++++++++++++++++---
 net/ipv4/tcp.c                |  3 +--
 net/ipv4/tcp_ipv4.c           |  3 +--
 net/ipv6/tcp_ipv6.c           |  3 +--
 8 files changed, 39 insertions(+), 15 deletions(-)
include/net/inet_hashtables.h +1 -0
index ba06e8b52264..69174093078f 100644
@@ -282,6 +282,7 @@ inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in
  * rcv_saddr field should already have been updated when this is called.
  */
 int inet_bhash2_update_saddr(struct sock *sk, void *saddr, int family);
+void inet_bhash2_reset_saddr(struct sock *sk);
 
 void inet_bind_hash(struct sock *sk, struct inet_bind_bucket *tb,
 		    struct inet_bind2_bucket *tb2, unsigned short port);
net/dccp/ipv4.c +1 -2
index 95e376e3b911..b780827f5e0a 100644
@@ -143,8 +143,7 @@ int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 	 * This unhashes the socket and releases the local port, if necessary.
 	 */
 	dccp_set_state(sk, DCCP_CLOSED);
-	if (!(sk->sk_userlocks & SOCK_BINDADDR_LOCK))
-		inet_reset_saddr(sk);
+	inet_bhash2_reset_saddr(sk);
 	ip_rt_put(rt);
 	sk->sk_route_caps = 0;
 	inet->inet_dport = 0;
net/dccp/ipv6.c +1 -2
index 94c101ed57a9..602f3432d80b 100644
@@ -970,8 +970,7 @@ static int dccp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
 
 late_failure:
 	dccp_set_state(sk, DCCP_CLOSED);
-	if (!(sk->sk_userlocks & SOCK_BINDADDR_LOCK))
-		inet_reset_saddr(sk);
+	inet_bhash2_reset_saddr(sk);
 	__sk_dst_reset(sk);
 failure:
 	inet->inet_dport = 0;
net/dccp/proto.c +1 -2
index c548ca3e9b0e..85e35c5e8890 100644
@@ -279,8 +279,7 @@ int dccp_disconnect(struct sock *sk, int flags)
 
 	inet->inet_dport = 0;
 
-	if (!(sk->sk_userlocks & SOCK_BINDADDR_LOCK))
-		inet_reset_saddr(sk);
+	inet_bhash2_reset_saddr(sk);
 
 	sk->sk_shutdown = 0;
 	sock_reset_flag(sk, SOCK_DONE);
net/ipv4/inet_hashtables.c +32 -3
index fce0bd62d6b5..e657da6dced0 100644
@@ -871,7 +871,7 @@ static void inet_update_saddr(struct sock *sk, void *saddr, int family)
 #endif
 }
 
-int inet_bhash2_update_saddr(struct sock *sk, void *saddr, int family)
+static int __inet_bhash2_update_saddr(struct sock *sk, void *saddr, int family, bool reset)
 {
 	struct inet_hashinfo *hinfo = tcp_or_dccp_get_hashinfo(sk);
 	struct inet_bind_hashbucket *head, *head2;
@@ -883,7 +883,11 @@ int inet_bhash2_update_saddr(struct sock *sk, void *saddr, int family)
 
 	if (!inet_csk(sk)->icsk_bind2_hash) {
 		/* Not bind()ed before. */
-		inet_update_saddr(sk, saddr, family);
+		if (reset)
+			inet_reset_saddr(sk);
+		else
+			inet_update_saddr(sk, saddr, family);
+
 		goto out;
 	}
 
@@ -901,6 +905,16 @@ int inet_bhash2_update_saddr(struct sock *sk, void *saddr, int family)
 	 */
 	new_tb2 = kmem_cache_alloc(hinfo->bind2_bucket_cachep, GFP_ATOMIC);
 	if (!new_tb2) {
+		if (reset) {
+			/* The (INADDR_ANY, port) bucket might have already
+			 * been freed, then we cannot fixup icsk_bind2_hash,
+			 * so we give up and unlink sk from bhash/bhash2 not
+			 * to leave inconsistency in bhash2.
+			 */
+			inet_put_port(sk);
+			inet_reset_saddr(sk);
+		}
+
 		err = -ENOMEM;
 		goto unlock;
 	}
@@ -912,7 +926,10 @@ int inet_bhash2_update_saddr(struct sock *sk, void *saddr, int family)
 	inet_bind2_bucket_destroy(hinfo->bind2_bucket_cachep, inet_csk(sk)->icsk_bind2_hash);
 	spin_unlock(&head2->lock);
 
-	inet_update_saddr(sk, saddr, family);
+	if (reset)
+		inet_reset_saddr(sk);
+	else
+		inet_update_saddr(sk, saddr, family);
 
 	head2 = inet_bhashfn_portaddr(hinfo, sk, net, port);
 
@@ -934,8 +951,20 @@ int inet_bhash2_update_saddr(struct sock *sk, void *saddr, int family)
 out:
 	return err;
 }
+
+int inet_bhash2_update_saddr(struct sock *sk, void *saddr, int family)
+{
+	return __inet_bhash2_update_saddr(sk, saddr, family, false);
+}
 EXPORT_SYMBOL_GPL(inet_bhash2_update_saddr);
 
+void inet_bhash2_reset_saddr(struct sock *sk)
+{
+	if (!(sk->sk_userlocks & SOCK_BINDADDR_LOCK))
+		__inet_bhash2_update_saddr(sk, NULL, 0, true);
+}
+EXPORT_SYMBOL_GPL(inet_bhash2_reset_saddr);
+
 /* RFC 6056 3.3.4.  Algorithm 4: Double-Hash Port Selection Algorithm
  * Note that we use 32bit integers (vs RFC 'short integers')
  * because 2^16 is not a multiple of num_ephemeral and this
net/ipv4/tcp.c +1 -2
index 54836a6b81d6..4f2205756cfe 100644
@@ -3114,8 +3114,7 @@ int tcp_disconnect(struct sock *sk, int flags)
 
 	inet->inet_dport = 0;
 
-	if (!(sk->sk_userlocks & SOCK_BINDADDR_LOCK))
-		inet_reset_saddr(sk);
+	inet_bhash2_reset_saddr(sk);
 
 	sk->sk_shutdown = 0;
 	sock_reset_flag(sk, SOCK_DONE);
net/ipv4/tcp_ipv4.c +1 -2
index 23dd7e9df2d5..da46357f501b 100644
@@ -331,8 +331,7 @@ int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 	 * if necessary.
 	 */
 	tcp_set_state(sk, TCP_CLOSE);
-	if (!(sk->sk_userlocks & SOCK_BINDADDR_LOCK))
-		inet_reset_saddr(sk);
+	inet_bhash2_reset_saddr(sk);
 	ip_rt_put(rt);
 	sk->sk_route_caps = 0;
 	inet->inet_dport = 0;
net/ipv6/tcp_ipv6.c +1 -2
index 2f3ca3190d26..f0548dbcabd2 100644
@@ -346,8 +346,7 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
 
 late_failure:
 	tcp_set_state(sk, TCP_CLOSE);
-	if (!(sk->sk_userlocks & SOCK_BINDADDR_LOCK))
-		inet_reset_saddr(sk);
+	inet_bhash2_reset_saddr(sk);
 failure:
 	inet->inet_dport = 0;
 	sk->sk_route_caps = 0;
Signature · 1 line
2.30.2