Skip to contentSkip to conversation

net: annotate data-races around sock->ops

PATCH Applied Eric Dumazet 4 participants 6 messages Aug 8, 2023 – Aug 28, 2023

Subsystems (8)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 net: annotate data-races around sock->ops

IPV6_ADDRFORM socket option is evil, because it can change sock->ops while other threads might read it. Same issue for sk->sk_family being set to AF_INET.

Adding READ_ONCE() over sock->ops reads is needed for sockets that might be impacted by IPV6_ADDRFORM.

Note that mptcp_is_tcpsk() can also overwrite sock->ops.

Adding annotations for all sk->sk_family reads will require more patches :/

BUG: KCSAN: data-race in ____sys_sendmsg / do_ipv6_setsockopt

write to 0xffff888109f24ca0 of 8 bytes by task 4470 on cpu 0:

do_ipv6_setsockopt+0x2c5e/0x2ce0 net/ipv6/ipv6_sockglue.c:491
ipv6_setsockopt+0x57/0x130 net/ipv6/ipv6_sockglue.c:1012
udpv6_setsockopt+0x95/0xa0 net/ipv6/udp.c:1690
sock_common_setsockopt+0x61/0x70 net/core/sock.c:3663
__sys_setsockopt+0x1c3/0x230 net/socket.c:2273
__do_sys_setsockopt net/socket.c:2284 [inline]
__se_sys_setsockopt net/socket.c:2281 [inline]
__x64_sys_setsockopt+0x66/0x80 net/socket.c:2281
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd

read to 0xffff888109f24ca0 of 8 bytes by task 4469 on cpu 1:

sock_sendmsg_nosec net/socket.c:724 [inline]
sock_sendmsg net/socket.c:747 [inline]
____sys_sendmsg+0x349/0x4c0 net/socket.c:2503
___sys_sendmsg net/socket.c:2557 [inline]
__sys_sendmmsg+0x263/0x500 net/socket.c:2643
__do_sys_sendmmsg net/socket.c:2672 [inline]
__se_sys_sendmmsg net/socket.c:2669 [inline]
__x64_sys_sendmmsg+0x57/0x60 net/socket.c:2669
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd

value changed: 0xffffffff850e32b8 -> 0xffffffff850da890

Reported by Kernel Concurrency Sanitizer on:

CPU: 1 PID: 4469 Comm: syz-executor.1 Not tainted 6.4.0-rc5-syzkaller-00313-g4c605260bc60 #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 05/25/2023
---
 include/linux/net.h      |   2 +-
 net/9p/trans_fd.c        |   4 +-
 net/core/scm.c           |   3 +-
 net/core/skmsg.c         |   8 ++-
 net/core/sock.c          |  24 +++++--
 net/ipv6/ipv6_sockglue.c |   8 +--
 net/mptcp/protocol.c     |   8 +--
 net/socket.c             | 136 +++++++++++++++++++++++----------------
 net/unix/scm.c           |   3 +-
 9 files changed, 118 insertions(+), 78 deletions(-)
include/linux/net.h +1 -1
index 41c608c1b02c26071bffc9a44a075ee65cc56197..c9b4a63791a4594882991a8fd109fe3648bf2997 100644
@@ -123,7 +123,7 @@ struct socket {
 
 	struct file		*file;
 	struct sock		*sk;
-	const struct proto_ops	*ops;
+	const struct proto_ops	*ops; /* Might change with IPV6_ADDRFORM or MPTCP. */
 
 	struct socket_wq	wq;
 };
net/9p/trans_fd.c +2 -2
index 00b684616e8d96b195840dce50ff6e69d5501638..c4015f30f9fa79a4a968f9a0a9aab243f0d460a0 100644
@@ -1019,7 +1019,7 @@ p9_fd_create_tcp(struct p9_client *client, const char *addr, char *args)
 		}
 	}
 
-	err = csocket->ops->connect(csocket,
+	err = READ_ONCE(csocket->ops)->connect(csocket,
 				    (struct sockaddr *)&sin_server,
 				    sizeof(struct sockaddr_in), 0);
 	if (err < 0) {
@@ -1060,7 +1060,7 @@ p9_fd_create_unix(struct p9_client *client, const char *addr, char *args)
 
 		return err;
 	}
-	err = csocket->ops->connect(csocket, (struct sockaddr *)&sun_server,
+	err = READ_ONCE(csocket->ops)->connect(csocket, (struct sockaddr *)&sun_server,
 			sizeof(struct sockaddr_un) - 1, 0);
 	if (err < 0) {
 		pr_err("%s (%d): problem connecting socket: %s: %d\n",
net/core/scm.c +2 -1
index 3cd7dd377e53fbdf793260fd4e40207a45ba33f0..880027ecf516503c6b98d1190aabca3c3be24e99 100644
@@ -130,6 +130,7 @@ EXPORT_SYMBOL(__scm_destroy);
 
 int __scm_send(struct socket *sock, struct msghdr *msg, struct scm_cookie *p)
 {
+	const struct proto_ops *ops = READ_ONCE(sock->ops);
 	struct cmsghdr *cmsg;
 	int err;
 
@@ -153,7 +154,7 @@ int __scm_send(struct socket *sock, struct msghdr *msg, struct scm_cookie *p)
 		switch (cmsg->cmsg_type)
 		{
 		case SCM_RIGHTS:
-			if (!sock->ops || sock->ops->family != PF_UNIX)
+			if (!ops || ops->family != PF_UNIX)
 				goto error;
 			err=scm_fp_copy(cmsg, &p->fp);
 			if (err<0)
net/core/skmsg.c +6 -2
index a29508e1ff3568583263b9307f7b1a0e814ba76d..e6dfc846018f9217d3fa34c60bd7f6ec0179d487 100644
@@ -1198,13 +1198,17 @@ static int sk_psock_verdict_recv(struct sock *sk, struct sk_buff *skb)
 static void sk_psock_verdict_data_ready(struct sock *sk)
 {
 	struct socket *sock = sk->sk_socket;
+	const struct proto_ops *ops;
 	int copied;
 
 	trace_sk_data_ready(sk);
 
-	if (unlikely(!sock || !sock->ops || !sock->ops->read_skb))
+	if (unlikely(!sock))
 		return;
-	copied = sock->ops->read_skb(sk, sk_psock_verdict_recv);
+	ops = READ_ONCE(sock->ops);
+	if (!ops || !ops->read_skb)
+		return;
+	copied = ops->read_skb(sk, sk_psock_verdict_recv);
 	if (copied >= 0) {
 		struct sk_psock *psock;
 
net/core/sock.c +17 -7
index 49915801d53a3524c756f706672b3f70bbc17cfc..51f7d94eccf7c78d25f8acf9a24ce0828b4f56f4 100644
@@ -1277,14 +1277,19 @@ int sk_setsockopt(struct sock *sk, int level, int optname,
 		break;
 
 	case SO_RCVLOWAT:
+		{
+		int (*set_rcvlowat)(struct sock *sk, int val) = NULL;
+
 		if (val < 0)
 			val = INT_MAX;
-		if (sock && sock->ops->set_rcvlowat)
-			ret = sock->ops->set_rcvlowat(sk, val);
+		if (sock)
+			set_rcvlowat = READ_ONCE(sock->ops)->set_rcvlowat;
+		if (set_rcvlowat)
+			ret = set_rcvlowat(sk, val);
 		else
 			WRITE_ONCE(sk->sk_rcvlowat, val ? : 1);
 		break;
-
+		}
 	case SO_RCVTIMEO_OLD:
 	case SO_RCVTIMEO_NEW:
 		ret = sock_set_timeout(&sk->sk_rcvtimeo, optval,
@@ -1379,11 +1384,16 @@ int sk_setsockopt(struct sock *sk, int level, int optname,
 		break;
 
 	case SO_PEEK_OFF:
-		if (sock->ops->set_peek_off)
-			ret = sock->ops->set_peek_off(sk, val);
+		{
+		int (*set_peek_off)(struct sock *sk, int val);
+
+		set_peek_off = READ_ONCE(sock->ops)->set_peek_off;
+		if (set_peek_off)
+			ret = set_peek_off(sk, val);
 		else
 			ret = -EOPNOTSUPP;
 		break;
+		}
 
 	case SO_NOFCS:
 		sock_valbool_flag(sk, SOCK_NOFCS, valbool);
@@ -1816,7 +1826,7 @@ int sk_getsockopt(struct sock *sk, int level, int optname,
 	{
 		struct sockaddr_storage address;
 
-		lv = sock->ops->getname(sock, (struct sockaddr *)&address, 2);
+		lv = READ_ONCE(sock->ops)->getname(sock, (struct sockaddr *)&address, 2);
 		if (lv < 0)
 			return -ENOTCONN;
 		if (lv < len)
@@ -1858,7 +1868,7 @@ int sk_getsockopt(struct sock *sk, int level, int optname,
 		break;
 
 	case SO_PEEK_OFF:
-		if (!sock->ops->set_peek_off)
+		if (!READ_ONCE(sock->ops)->set_peek_off)
 			return -EOPNOTSUPP;
 
 		v.val = READ_ONCE(sk->sk_peek_off);
net/ipv6/ipv6_sockglue.c +4 -4
index ae818ff4622486a5966977e970579b193031b305..ca377159967c8aa9c18a80f9b189f4ef41398d01 100644
@@ -474,8 +474,8 @@ int do_ipv6_setsockopt(struct sock *sk, int level, int optname,
 				WRITE_ONCE(sk->sk_prot, &tcp_prot);
 				/* 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;
+				WRITE_ONCE(sk->sk_socket->ops, &inet_stream_ops);
+				WRITE_ONCE(sk->sk_family, PF_INET);
 				tcp_sync_mss(sk, icsk->icsk_pmtu_cookie);
 			} else {
 				struct proto *prot = &udp_prot;
@@ -488,8 +488,8 @@ int do_ipv6_setsockopt(struct sock *sk, int level, int optname,
 
 				/* 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;
+				WRITE_ONCE(sk->sk_socket->ops, &inet_dgram_ops);
+				WRITE_ONCE(sk->sk_family, PF_INET);
 			}
 
 			/* Disable all options not to allocate memory anymore,
net/mptcp/protocol.c +4 -4
index 65ee949a8a445ca29941d836ed65d79ef12a10c2..1c079e83481eaca16cf73c78050a298cf3e556b6 100644
@@ -67,11 +67,11 @@ static bool mptcp_is_tcpsk(struct sock *sk)
 		 * Hand the socket over to tcp so all further socket ops
 		 * bypass mptcp.
 		 */
-		sock->ops = &inet_stream_ops;
+		WRITE_ONCE(sock->ops, &inet_stream_ops);
 		return true;
 #if IS_ENABLED(CONFIG_MPTCP_IPV6)
 	} else if (unlikely(sk->sk_prot == &tcpv6_prot)) {
-		sock->ops = &inet6_stream_ops;
+		WRITE_ONCE(sock->ops, &inet6_stream_ops);
 		return true;
 #endif
 	}
@@ -3683,7 +3683,7 @@ static int mptcp_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
 		goto unlock;
 	}
 
-	err = ssock->ops->bind(ssock, uaddr, addr_len);
+	err = READ_ONCE(ssock->ops)->bind(ssock, uaddr, addr_len);
 	if (!err)
 		mptcp_copy_inaddrs(sock->sk, ssock->sk);
 
@@ -3717,7 +3717,7 @@ static int mptcp_listen(struct socket *sock, int backlog)
 	inet_sk_state_store(sk, TCP_LISTEN);
 	sock_set_flag(sk, SOCK_RCU_FREE);
 
-	err = ssock->ops->listen(ssock, backlog);
+	err = READ_ONCE(ssock->ops)->listen(ssock, backlog);
 	inet_sk_state_store(sk, inet_sk_state_load(ssock->sk));
 	if (!err) {
 		sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1);
net/socket.c +80 -56

438 lines not on this page · raw message

net/unix/scm.c +2 -1
index f9152881d77f636f9500d1b57fdda584df845fc2..e9dde7176c8a3a1100aaf659991ceeedf8ef223d 100644
@@ -29,10 +29,11 @@ struct sock *unix_get_socket(struct file *filp)
 	/* Socket ? */
 	if (S_ISSOCK(inode->i_mode) && !(filp->f_mode & FMODE_PATH)) {
 		struct socket *sock = SOCKET_I(inode);
+		const struct proto_ops *ops = READ_ONCE(sock->ops);
 		struct sock *s = sock->sk;
 
 		/* PF_UNIX ? */
-		if (s && sock->ops && sock->ops->family == PF_UNIX)
+		if (s && ops && ops->family == PF_UNIX)
 			u_sock = s;
 	} else {
 		/* Could be an io_uring instance */
Signature · 1 line
2.41.0.640.ga95def55d0-goog
Reviewed-by on v1 1/1
#1 lore raw

From: Eric Dumazet <[email protected]>
Date: Tue, 8 Aug 2023 13:58:09 +0000

Quoted 40 lines

IPV6_ADDRFORM socket option is evil, because it can change sock->ops while other threads might read it. Same issue for sk->sk_family being set to AF_INET.

Adding READ_ONCE() over sock->ops reads is needed for sockets that might be impacted by IPV6_ADDRFORM.

Note that mptcp_is_tcpsk() can also overwrite sock->ops.

Adding annotations for all sk->sk_family reads will require more patches :/

BUG: KCSAN: data-race in ____sys_sendmsg / do_ipv6_setsockopt

write to 0xffff888109f24ca0 of 8 bytes by task 4470 on cpu 0:

do_ipv6_setsockopt+0x2c5e/0x2ce0 net/ipv6/ipv6_sockglue.c:491
ipv6_setsockopt+0x57/0x130 net/ipv6/ipv6_sockglue.c:1012
udpv6_setsockopt+0x95/0xa0 net/ipv6/udp.c:1690
sock_common_setsockopt+0x61/0x70 net/core/sock.c:3663
__sys_setsockopt+0x1c3/0x230 net/socket.c:2273
__do_sys_setsockopt net/socket.c:2284 [inline]
__se_sys_setsockopt net/socket.c:2281 [inline]
__x64_sys_setsockopt+0x66/0x80 net/socket.c:2281
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd

read to 0xffff888109f24ca0 of 8 bytes by task 4469 on cpu 1:

sock_sendmsg_nosec net/socket.c:724 [inline]
sock_sendmsg net/socket.c:747 [inline]
____sys_sendmsg+0x349/0x4c0 net/socket.c:2503
___sys_sendmsg net/socket.c:2557 [inline]
__sys_sendmmsg+0x263/0x500 net/socket.c:2643
__do_sys_sendmmsg net/socket.c:2672 [inline]
__se_sys_sendmmsg net/socket.c:2669 [inline]
__x64_sys_sendmmsg+0x57/0x60 net/socket.c:2669
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd
Quoted 8 lines

value changed: 0xffffffff850e32b8 -> 0xffffffff850da890

Reported by Kernel Concurrency Sanitizer on:

CPU: 1 PID: 4469 Comm: syz-executor.1 Not tainted 6.4.0-rc5-syzkaller-00313-g4c605260bc60 #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 05/25/2023
#2 lore raw
In reply to Eric Dumazet
This patch was applied to netdev/net-next.git (main) by Jakub Kicinski <[email protected]>:

Hello:

This patch was applied to netdev/net-next.git (main) by Jakub Kicinski <[email protected]>:

Quoted 8 lines · On Tue, 8 Aug 2023 13:58:09 +0000 you wrote:

IPV6_ADDRFORM socket option is evil, because it can change sock->ops while other threads might read it. Same issue for sk->sk_family being set to AF_INET.

Adding READ_ONCE() over sock->ops reads is needed for sockets that might be impacted by IPV6_ADDRFORM.

[...]

Here is the summary with links:
- [net-next] net: annotate data-races around sock->ops

    https://git.kernel.org/netdev/net-next/c/1ded5e5a5931

You are awesome, thank you!

Signature · 2 lines
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html
maintainer
#3 lore raw
In reply to Eric Dumazet

Hi,

Quoted 11 lines · Eric Dumazet

IPV6_ADDRFORM socket option is evil, because it can change sock->ops while other threads might read it. Same issue for sk->sk_family being set to AF_INET.

Adding READ_ONCE() over sock->ops reads is needed for sockets that might be impacted by IPV6_ADDRFORM.

Note that mptcp_is_tcpsk() can also overwrite sock->ops.

Adding annotations for all sk->sk_family reads will require more patches :/

I was unable to give the above a proper look before due to OoO on my side.

The mptcp code calls mptcp_is_tcpsk() only before the fd for the newly accepted socket is installed, so we should not have concurrent racing access to sock->ops?!? Do you have any related splat handy?

Thanks,

Paolo

#4 lore raw
Quoted 16 lines · Paolo Abeni

Hi,

Quoted 11 lines · Eric Dumazet

IPV6_ADDRFORM socket option is evil, because it can change sock->ops while other threads might read it. Same issue for sk->sk_family being set to AF_INET.

Adding READ_ONCE() over sock->ops reads is needed for sockets that might be impacted by IPV6_ADDRFORM.

Note that mptcp_is_tcpsk() can also overwrite sock->ops.

Adding annotations for all sk->sk_family reads will require more patches :/

Quoted 7 lines

I was unable to give the above a proper look before due to OoO on my side.

The mptcp code calls mptcp_is_tcpsk() only before the fd for the newly accepted socket is installed, so we should not have concurrent racing access to sock->ops?!? Do you have any related splat handy?

syzbot splat was on another layer. I tried to fix all sites that could trigger a similar issue.

BUG: KCSAN: data-race in ____sys_sendmsg / do_ipv6_setsockopt

write to 0xffff888109f24ca0 of 8 bytes by task 4470 on cpu 0:

do_ipv6_setsockopt+0x2c5e/0x2ce0 net/ipv6/ipv6_sockglue.c:491
ipv6_setsockopt+0x57/0x130 net/ipv6/ipv6_sockglue.c:1012
udpv6_setsockopt+0x95/0xa0 net/ipv6/udp.c:1690
sock_common_setsockopt+0x61/0x70 net/core/sock.c:3663
__sys_setsockopt+0x1c3/0x230 net/socket.c:2273
__do_sys_setsockopt net/socket.c:2284 [inline]
__se_sys_setsockopt net/socket.c:2281 [inline]
__x64_sys_setsockopt+0x66/0x80 net/socket.c:2281
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd

read to 0xffff888109f24ca0 of 8 bytes by task 4469 on cpu 1:

sock_sendmsg_nosec net/socket.c:724 [inline]
sock_sendmsg net/socket.c:747 [inline]
____sys_sendmsg+0x349/0x4c0 net/socket.c:2503
___sys_sendmsg net/socket.c:2557 [inline]
__sys_sendmmsg+0x263/0x500 net/socket.c:2643
__do_sys_sendmmsg net/socket.c:2672 [inline]
__se_sys_sendmmsg net/socket.c:2669 [inline]
__x64_sys_sendmmsg+0x57/0x60 net/socket.c:2669
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd

value changed: 0xffffffff850e32b8 -> 0xffffffff850da890

Reported by Kernel Concurrency Sanitizer on:

CPU: 1 PID: 4469 Comm: syz-executor.1 Not tainted
6.4.0-rc5-syzkaller-00313-g4c605260bc60 #0
Hardware name: Google Google Compute Engine/Google Compute Engine,

BIOS Google 05/25/2023

Maybe MPTCP side (mptcp_is_tcpsk()) was ok,
but the fact that mptcp_is_tcpsk() was able to write over sock->ops was a bit strange to me.

mptcp_is_tcpsk() should answer a question, with a read-only argument.

I suggest changing the name of the helper to better reflect what it is doing.

maintainer
#5 lore raw
Quoted 61 lines · Eric Dumazet
Quoted 13 lines · Paolo Abeni
Quoted 11 lines · Eric Dumazet

IPV6_ADDRFORM socket option is evil, because it can change sock->ops while other threads might read it. Same issue for sk->sk_family being set to AF_INET.

Adding READ_ONCE() over sock->ops reads is needed for sockets that might be impacted by IPV6_ADDRFORM.

Note that mptcp_is_tcpsk() can also overwrite sock->ops.

Adding annotations for all sk->sk_family reads will require more patches :/

Quoted 7 lines

I was unable to give the above a proper look before due to OoO on my side.

The mptcp code calls mptcp_is_tcpsk() only before the fd for the newly accepted socket is installed, so we should not have concurrent racing access to sock->ops?!? Do you have any related splat handy?

syzbot splat was on another layer. I tried to fix all sites that could trigger a similar issue.

BUG: KCSAN: data-race in ____sys_sendmsg / do_ipv6_setsockopt

write to 0xffff888109f24ca0 of 8 bytes by task 4470 on cpu 0:

do_ipv6_setsockopt+0x2c5e/0x2ce0 net/ipv6/ipv6_sockglue.c:491
ipv6_setsockopt+0x57/0x130 net/ipv6/ipv6_sockglue.c:1012
udpv6_setsockopt+0x95/0xa0 net/ipv6/udp.c:1690
sock_common_setsockopt+0x61/0x70 net/core/sock.c:3663
__sys_setsockopt+0x1c3/0x230 net/socket.c:2273
__do_sys_setsockopt net/socket.c:2284 [inline]
__se_sys_setsockopt net/socket.c:2281 [inline]
__x64_sys_setsockopt+0x66/0x80 net/socket.c:2281
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd

read to 0xffff888109f24ca0 of 8 bytes by task 4469 on cpu 1:

sock_sendmsg_nosec net/socket.c:724 [inline]
sock_sendmsg net/socket.c:747 [inline]
____sys_sendmsg+0x349/0x4c0 net/socket.c:2503
___sys_sendmsg net/socket.c:2557 [inline]
__sys_sendmmsg+0x263/0x500 net/socket.c:2643
__do_sys_sendmmsg net/socket.c:2672 [inline]
__se_sys_sendmmsg net/socket.c:2669 [inline]
__x64_sys_sendmmsg+0x57/0x60 net/socket.c:2669
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd

value changed: 0xffffffff850e32b8 -> 0xffffffff850da890

Reported by Kernel Concurrency Sanitizer on:

CPU: 1 PID: 4469 Comm: syz-executor.1 Not tainted
6.4.0-rc5-syzkaller-00313-g4c605260bc60 #0
Hardware name: Google Google Compute Engine/Google Compute Engine,

BIOS Google 05/25/2023

Quoted 7 lines

Maybe MPTCP side (mptcp_is_tcpsk()) was ok,
but the fact that mptcp_is_tcpsk() was able to write over sock->ops was a bit strange to me.

mptcp_is_tcpsk() should answer a question, with a read-only argument.

I suggest changing the name of the helper to better reflect what it is doing.

Thanks for double checking. Indeed such helper name is confusing, and should be renamed; I'll try to take care of it after the merge window.

Cheers,

Paolo