[PATCH 4.19 136/139] ipv6: annotate some data-races around sk->sk_prot

[PATCH 4.19 136/139] ipv6: annotate some data-races around sk->sk_prot

[Greg Kroah-Hartman]

4.19-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Eric Dumazet <edumazet@google.com>

commit 086d49058cd8471046ae9927524708820f5fd1c7 upstream.

Changes from the original is that the applied code to inet6_sendmsg
and inet6_recvmsg is ported to inet_sendmsg and inet_recvmsg because
the same functions are shared between ipv4 and v6 in 4.19 kernel.

The original commit message is as below.

IPv6 has this hack changing sk->sk_prot when an IPv6 socket
is 'converted' to an IPv4 one with IPV6_ADDRFORM option.

This operation is only performed for TCP and UDP, knowing
their 'struct proto' for the two network families are populated
in the same way, and can not disappear while a reader
might use and dereference sk->sk_prot.

If we think about it all reads of sk->sk_prot while
either socket lock or RTNL is not acquired should be using READ_ONCE().

Also note that other layers like MPTCP, XFRM, CHELSIO_TLS also
write over sk->sk_prot.

BUG: KCSAN: data-race in inet6_recvmsg / ipv6_setsockopt

write to 0xffff8881386f7aa8 of 8 bytes by task 26932 on cpu 0:
 do_ipv6_setsockopt net/ipv6/ipv6_sockglue.c:492 [inline]
 ipv6_setsockopt+0x3758/0x3910 net/ipv6/ipv6_sockglue.c:1019
 udpv6_setsockopt+0x85/0x90 net/ipv6/udp.c:1649
 sock_common_setsockopt+0x5d/0x70 net/core/sock.c:3489
 __sys_setsockopt+0x209/0x2a0 net/socket.c:2180
 __do_sys_setsockopt net/socket.c:2191 [inline]
 __se_sys_setsockopt net/socket.c:2188 [inline]
 __x64_sys_setsockopt+0x62/0x70 net/socket.c:2188
 do_syscall_x64 arch/x86/entry/common.c:50 [inline]
 do_syscall_64+0x44/0xd0 arch/x86/entry/common.c:80
 entry_SYSCALL_64_after_hwframe+0x44/0xae

read to 0xffff8881386f7aa8 of 8 bytes by task 26911 on cpu 1:
 inet6_recvmsg+0x7a/0x210 net/ipv6/af_inet6.c:659
 ____sys_recvmsg+0x16c/0x320
 ___sys_recvmsg net/socket.c:2674 [inline]
 do_recvmmsg+0x3f5/0xae0 net/socket.c:2768
 __sys_recvmmsg net/socket.c:2847 [inline]
 __do_sys_recvmmsg net/socket.c:2870 [inline]
 __se_sys_recvmmsg net/socket.c:2863 [inline]
 __x64_sys_recvmmsg+0xde/0x160 net/socket.c:2863
 do_syscall_x64 arch/x86/entry/common.c:50 [inline]
 do_syscall_64+0x44/0xd0 arch/x86/entry/common.c:80
 entry_SYSCALL_64_after_hwframe+0x44/0xae

value changed: 0xffffffff85e0e980 -> 0xffffffff85e01580

Reported by Kernel Concurrency Sanitizer on:
CPU: 1 PID: 26911 Comm: syz-executor.3 Not tainted 5.17.0-rc2-syzkaller-00316-g0457e5153e0e-dirty #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011

Reported-by: syzbot <syzkaller@googlegroups.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Kazunori Kobayashi <kazunori.kobayashi@miraclelinux.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 net/ipv4/af_inet.c       |   15 +++++++++++----
 net/ipv6/af_inet6.c      |   14 ++++++++++----
 net/ipv6/ipv6_sockglue.c |    6 ++++--
 3 files changed, 25 insertions(+), 10 deletions(-)

--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -789,15 +789,19 @@ EXPORT_SYMBOL(inet_getname);
 int inet_sendmsg(struct socket *sock, struct msghdr *msg, size_t size)
 {
 	struct sock *sk = sock->sk;
+	const struct proto *prot;
 
 	sock_rps_record_flow(sk);
 
+	/* IPV6_ADDRFORM can change sk->sk_prot under us. */
+	prot = READ_ONCE(sk->sk_prot);
+
 	/* We may need to bind the socket. */
-	if (!inet_sk(sk)->inet_num && !sk->sk_prot->no_autobind &&
+	if (!inet_sk(sk)->inet_num && !prot->no_autobind &&
 	    inet_autobind(sk))
 		return -EAGAIN;
 
-	return sk->sk_prot->sendmsg(sk, msg, size);
+	return prot->sendmsg(sk, msg, size);
 }
 EXPORT_SYMBOL(inet_sendmsg);
 
@@ -823,14 +827,17 @@ int inet_recvmsg(struct socket *sock, st
 		 int flags)
 {
 	struct sock *sk = sock->sk;
+	const struct proto *prot;
 	int addr_len = 0;
 	int err;
 
 	if (likely(!(flags & MSG_ERRQUEUE)))
 		sock_rps_record_flow(sk);
 
-	err = sk->sk_prot->recvmsg(sk, msg, size, flags & MSG_DONTWAIT,
-				   flags & ~MSG_DONTWAIT, &addr_len);
+	/* IPV6_ADDRFORM can change sk->sk_prot under us. */
+	prot = READ_ONCE(sk->sk_prot);
+	err = prot->recvmsg(sk, msg, size, flags & MSG_DONTWAIT,
+			    flags & ~MSG_DONTWAIT, &addr_len);
 	if (err >= 0)
 		msg->msg_namelen = addr_len;
 	return err;
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -443,11 +443,14 @@ out_unlock:
 int inet6_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
 {
 	struct sock *sk = sock->sk;
+	const struct proto *prot;
 	int err = 0;
 
+	/* IPV6_ADDRFORM can change sk->sk_prot under us. */
+	prot = READ_ONCE(sk->sk_prot);
 	/* If the socket has its own bind function then use it. */
-	if (sk->sk_prot->bind)
-		return sk->sk_prot->bind(sk, uaddr, addr_len);
+	if (prot->bind)
+		return prot->bind(sk, uaddr, addr_len);
 
 	if (addr_len < SIN6_LEN_RFC2133)
 		return -EINVAL;
@@ -558,6 +561,7 @@ int inet6_ioctl(struct socket *sock, uns
 {
 	struct sock *sk = sock->sk;
 	struct net *net = sock_net(sk);
+	const struct proto *prot;
 
 	switch (cmd) {
 	case SIOCGSTAMP:
@@ -578,9 +582,11 @@ int inet6_ioctl(struct socket *sock, uns
 	case SIOCSIFDSTADDR:
 		return addrconf_set_dstaddr(net, (void __user *) arg);
 	default:
-		if (!sk->sk_prot->ioctl)
+		/* IPV6_ADDRFORM can change sk->sk_prot under us. */
+		prot = READ_ONCE(sk->sk_prot);
+		if (!prot->ioctl)
 			return -ENOIOCTLCMD;
-		return sk->sk_prot->ioctl(sk, cmd, arg);
+		return prot->ioctl(sk, cmd, arg);
 	}
 	/*NOTREACHED*/
 	return 0;
--- a/net/ipv6/ipv6_sockglue.c
+++ b/net/ipv6/ipv6_sockglue.c
@@ -224,7 +224,8 @@ static int do_ipv6_setsockopt(struct soc
 				sock_prot_inuse_add(net, sk->sk_prot, -1);
 				sock_prot_inuse_add(net, &tcp_prot, 1);
 				local_bh_enable();
-				sk->sk_prot = &tcp_prot;
+				/* Paired with READ_ONCE(sk->sk_prot) in net/ipv6/af_inet6.c */
+				WRITE_ONCE(sk->sk_prot, &tcp_prot);
 				icsk->icsk_af_ops = &ipv4_specific;
 				sk->sk_socket->ops = &inet_stream_ops;
 				sk->sk_family = PF_INET;
@@ -238,7 +239,8 @@ static int do_ipv6_setsockopt(struct soc
 				sock_prot_inuse_add(net, sk->sk_prot, -1);
 				sock_prot_inuse_add(net, prot, 1);
 				local_bh_enable();
-				sk->sk_prot = prot;
+				/* Paired with READ_ONCE(sk->sk_prot) in net/ipv6/af_inet6.c */
+				WRITE_ONCE(sk->sk_prot, prot);
 				sk->sk_socket->ops = &inet_dgram_ops;
 				sk->sk_family = PF_INET;
 			}

[PATCH 4.19 137/139] ipv6: Fix data races around sk->sk_prot.

[Greg Kroah-Hartman]

4.19-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Kuniyuki Iwashima <kuniyu@amazon.com>

commit 364f997b5cfe1db0d63a390fe7c801fa2b3115f6 upstream.

Commit 086d49058cd8 ("ipv6: annotate some data-races around sk->sk_prot")
fixed some data-races around sk->sk_prot but it was not enough.

Some functions in inet6_(stream|dgram)_ops still access sk->sk_prot
without lock_sock() or rtnl_lock(), so they need READ_ONCE() to avoid
load tearing.

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Kazunori Kobayashi <kazunori.kobayashi@miraclelinux.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 net/core/sock.c          |    6 ++++--
 net/ipv4/af_inet.c       |   23 ++++++++++++++++-------
 net/ipv6/ipv6_sockglue.c |    4 ++--
 3 files changed, 22 insertions(+), 11 deletions(-)

--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -3064,7 +3064,8 @@ int sock_common_getsockopt(struct socket
 {
 	struct sock *sk = sock->sk;
 
-	return sk->sk_prot->getsockopt(sk, level, optname, optval, optlen);
+	/* IPV6_ADDRFORM can change sk->sk_prot under us. */
+	return READ_ONCE(sk->sk_prot)->getsockopt(sk, level, optname, optval, optlen);
 }
 EXPORT_SYMBOL(sock_common_getsockopt);
 
@@ -3105,7 +3106,8 @@ int sock_common_setsockopt(struct socket
 {
 	struct sock *sk = sock->sk;
 
-	return sk->sk_prot->setsockopt(sk, level, optname, optval, optlen);
+	/* IPV6_ADDRFORM can change sk->sk_prot under us. */
+	return READ_ONCE(sk->sk_prot)->setsockopt(sk, level, optname, optval, optlen);
 }
 EXPORT_SYMBOL(sock_common_setsockopt);
 
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -553,22 +553,27 @@ int inet_dgram_connect(struct socket *so
 		       int addr_len, int flags)
 {
 	struct sock *sk = sock->sk;
+	const struct proto *prot;
 	int err;
 
 	if (addr_len < sizeof(uaddr->sa_family))
 		return -EINVAL;
+
+	/* IPV6_ADDRFORM can change sk->sk_prot under us. */
+	prot = READ_ONCE(sk->sk_prot);
+
 	if (uaddr->sa_family == AF_UNSPEC)
-		return sk->sk_prot->disconnect(sk, flags);
+		return prot->disconnect(sk, flags);
 
 	if (BPF_CGROUP_PRE_CONNECT_ENABLED(sk)) {
-		err = sk->sk_prot->pre_connect(sk, uaddr, addr_len);
+		err = prot->pre_connect(sk, uaddr, addr_len);
 		if (err)
 			return err;
 	}
 
 	if (!inet_sk(sk)->inet_num && inet_autobind(sk))
 		return -EAGAIN;
-	return sk->sk_prot->connect(sk, uaddr, addr_len);
+	return prot->connect(sk, uaddr, addr_len);
 }
 EXPORT_SYMBOL(inet_dgram_connect);
 
@@ -731,10 +736,11 @@ EXPORT_SYMBOL(inet_stream_connect);
 int inet_accept(struct socket *sock, struct socket *newsock, int flags,
 		bool kern)
 {
-	struct sock *sk1 = sock->sk;
+	struct sock *sk1 = sock->sk, *sk2;
 	int err = -EINVAL;
-	struct sock *sk2 = sk1->sk_prot->accept(sk1, flags, &err, kern);
 
+	/* IPV6_ADDRFORM can change sk->sk_prot under us. */
+	sk2 = READ_ONCE(sk1->sk_prot)->accept(sk1, flags, &err, kern);
 	if (!sk2)
 		goto do_err;
 
@@ -809,6 +815,7 @@ ssize_t inet_sendpage(struct socket *soc
 		      size_t size, int flags)
 {
 	struct sock *sk = sock->sk;
+	const struct proto *prot;
 
 	sock_rps_record_flow(sk);
 
@@ -817,8 +824,10 @@ ssize_t inet_sendpage(struct socket *soc
 	    inet_autobind(sk))
 		return -EAGAIN;
 
-	if (sk->sk_prot->sendpage)
-		return sk->sk_prot->sendpage(sk, page, offset, size, flags);
+	/* IPV6_ADDRFORM can change sk->sk_prot under us. */
+	prot = READ_ONCE(sk->sk_prot);
+	if (prot->sendpage)
+		return prot->sendpage(sk, page, offset, size, flags);
 	return sock_no_sendpage(sock, page, offset, size, flags);
 }
 EXPORT_SYMBOL(inet_sendpage);
--- a/net/ipv6/ipv6_sockglue.c
+++ b/net/ipv6/ipv6_sockglue.c
@@ -224,7 +224,7 @@ static int do_ipv6_setsockopt(struct soc
 				sock_prot_inuse_add(net, sk->sk_prot, -1);
 				sock_prot_inuse_add(net, &tcp_prot, 1);
 				local_bh_enable();
-				/* Paired with READ_ONCE(sk->sk_prot) in net/ipv6/af_inet6.c */
+				/* Paired with READ_ONCE(sk->sk_prot) in inet6_stream_ops */
 				WRITE_ONCE(sk->sk_prot, &tcp_prot);
 				icsk->icsk_af_ops = &ipv4_specific;
 				sk->sk_socket->ops = &inet_stream_ops;
@@ -239,7 +239,7 @@ static int do_ipv6_setsockopt(struct soc
 				sock_prot_inuse_add(net, sk->sk_prot, -1);
 				sock_prot_inuse_add(net, prot, 1);
 				local_bh_enable();
-				/* Paired with READ_ONCE(sk->sk_prot) in net/ipv6/af_inet6.c */
+				/* Paired with READ_ONCE(sk->sk_prot) in inet6_dgram_ops */
 				WRITE_ONCE(sk->sk_prot, prot);
 				sk->sk_socket->ops = &inet_dgram_ops;
 				sk->sk_family = PF_INET;

[PATCH 4.19 138/139] tcp: Fix data races around icsk->icsk_af_ops.

[Greg Kroah-Hartman]

4.19-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Kuniyuki Iwashima <kuniyu@amazon.com>

commit f49cd2f4d6170d27a2c61f1fecb03d8a70c91f57 upstream.

setsockopt(IPV6_ADDRFORM) and tcp_v6_connect() change icsk->icsk_af_ops
under lock_sock(), but tcp_(get|set)sockopt() read it locklessly.  To
avoid load/store tearing, we need to add READ_ONCE() and WRITE_ONCE()
for the reads and writes.

Thanks to Eric Dumazet for providing the syzbot report:

BUG: KCSAN: data-race in tcp_setsockopt / tcp_v6_connect

write to 0xffff88813c624518 of 8 bytes by task 23936 on cpu 0:
tcp_v6_connect+0x5b3/0xce0 net/ipv6/tcp_ipv6.c:240
__inet_stream_connect+0x159/0x6d0 net/ipv4/af_inet.c:660
inet_stream_connect+0x44/0x70 net/ipv4/af_inet.c:724
__sys_connect_file net/socket.c:1976 [inline]
__sys_connect+0x197/0x1b0 net/socket.c:1993
__do_sys_connect net/socket.c:2003 [inline]
__se_sys_connect net/socket.c:2000 [inline]
__x64_sys_connect+0x3d/0x50 net/socket.c:2000
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd

read to 0xffff88813c624518 of 8 bytes by task 23937 on cpu 1:
tcp_setsockopt+0x147/0x1c80 net/ipv4/tcp.c:3789
sock_common_setsockopt+0x5d/0x70 net/core/sock.c:3585
__sys_setsockopt+0x212/0x2b0 net/socket.c:2252
__do_sys_setsockopt net/socket.c:2263 [inline]
__se_sys_setsockopt net/socket.c:2260 [inline]
__x64_sys_setsockopt+0x62/0x70 net/socket.c:2260
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd

value changed: 0xffffffff8539af68 -> 0xffffffff8539aff8

Reported by Kernel Concurrency Sanitizer on:
CPU: 1 PID: 23937 Comm: syz-executor.5 Not tainted
6.0.0-rc4-syzkaller-00331-g4ed9c1e971b1-dirty #0

Hardware name: Google Google Compute Engine/Google Compute Engine,
BIOS Google 08/26/2022

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Reported-by: syzbot <syzkaller@googlegroups.com>
Reported-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Kazunori Kobayashi <kazunori.kobayashi@miraclelinux.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 net/ipv4/tcp.c           |   10 ++++++----
 net/ipv6/ipv6_sockglue.c |    3 ++-
 net/ipv6/tcp_ipv6.c      |    6 ++++--
 3 files changed, 12 insertions(+), 7 deletions(-)

--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -3130,8 +3130,9 @@ int tcp_setsockopt(struct sock *sk, int
 	const struct inet_connection_sock *icsk = inet_csk(sk);
 
 	if (level != SOL_TCP)
-		return icsk->icsk_af_ops->setsockopt(sk, level, optname,
-						     optval, optlen);
+		/* Paired with WRITE_ONCE() in do_ipv6_setsockopt() and tcp_v6_connect() */
+		return READ_ONCE(icsk->icsk_af_ops)->setsockopt(sk, level, optname,
+								optval, optlen);
 	return do_tcp_setsockopt(sk, level, optname, optval, optlen);
 }
 EXPORT_SYMBOL(tcp_setsockopt);
@@ -3653,8 +3654,9 @@ int tcp_getsockopt(struct sock *sk, int
 	struct inet_connection_sock *icsk = inet_csk(sk);
 
 	if (level != SOL_TCP)
-		return icsk->icsk_af_ops->getsockopt(sk, level, optname,
-						     optval, optlen);
+		/* Paired with WRITE_ONCE() in do_ipv6_setsockopt() and tcp_v6_connect() */
+		return READ_ONCE(icsk->icsk_af_ops)->getsockopt(sk, level, optname,
+								optval, optlen);
 	return do_tcp_getsockopt(sk, level, optname, optval, optlen);
 }
 EXPORT_SYMBOL(tcp_getsockopt);
--- a/net/ipv6/ipv6_sockglue.c
+++ b/net/ipv6/ipv6_sockglue.c
@@ -226,7 +226,8 @@ static int do_ipv6_setsockopt(struct soc
 				local_bh_enable();
 				/* Paired with READ_ONCE(sk->sk_prot) in inet6_stream_ops */
 				WRITE_ONCE(sk->sk_prot, &tcp_prot);
-				icsk->icsk_af_ops = &ipv4_specific;
+				/* Paired with READ_ONCE() in tcp_(get|set)sockopt() */
+				WRITE_ONCE(icsk->icsk_af_ops, &ipv4_specific);
 				sk->sk_socket->ops = &inet_stream_ops;
 				sk->sk_family = PF_INET;
 				tcp_sync_mss(sk, icsk->icsk_pmtu_cookie);
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -229,7 +229,8 @@ static int tcp_v6_connect(struct sock *s
 		sin.sin_port = usin->sin6_port;
 		sin.sin_addr.s_addr = usin->sin6_addr.s6_addr32[3];
 
-		icsk->icsk_af_ops = &ipv6_mapped;
+		/* Paired with READ_ONCE() in tcp_(get|set)sockopt() */
+		WRITE_ONCE(icsk->icsk_af_ops, &ipv6_mapped);
 		sk->sk_backlog_rcv = tcp_v4_do_rcv;
 #ifdef CONFIG_TCP_MD5SIG
 		tp->af_specific = &tcp_sock_ipv6_mapped_specific;
@@ -239,7 +240,8 @@ static int tcp_v6_connect(struct sock *s
 
 		if (err) {
 			icsk->icsk_ext_hdr_len = exthdrlen;
-			icsk->icsk_af_ops = &ipv6_specific;
+			/* Paired with READ_ONCE() in tcp_(get|set)sockopt() */
+			WRITE_ONCE(icsk->icsk_af_ops, &ipv6_specific);
 			sk->sk_backlog_rcv = tcp_v6_do_rcv;
 #ifdef CONFIG_TCP_MD5SIG
 			tp->af_specific = &tcp_sock_ipv6_specific;