| Message ID | 4b80c3d1dbe3d0ab072f80450c202d9bc88b4b03.1672740602.git.pabeni@redhat.com (mailing list archive) |
|---|---|
| State | Accepted |
| Commit | 2c02d41d71f90a5168391b6a5f2954112ba2307c |
| Delegated to: | Netdev Maintainers |
| Headers | show |
| Series | [net] net/ulp: prevent ULP without clone op from entering the LISTEN status | expand |
Hello: This patch was applied to netdev/net.git (master) by Jakub Kicinski <kuba@kernel.org>: On Tue, 3 Jan 2023 12:19:17 +0100 you wrote: > When an ULP-enabled socket enters the LISTEN status, the listener ULP data > pointer is copied inside the child/accepted sockets by sk_clone_lock(). > > The relevant ULP can take care of de-duplicating the context pointer via > the clone() operation, but only MPTCP and SMC implement such op. > > Other ULPs may end-up with a double-free at socket disposal time. > > [...] Here is the summary with links: - [net] net/ulp: prevent ULP without clone op from entering the LISTEN status https://git.kernel.org/netdev/net/c/2c02d41d71f9 You are awesome, thank you!
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 848ffc3e0239..d1f837579398 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -1200,12 +1200,26 @@ void inet_csk_prepare_forced_close(struct sock *sk) } EXPORT_SYMBOL(inet_csk_prepare_forced_close); +static int inet_ulp_can_listen(const struct sock *sk) +{ + const struct inet_connection_sock *icsk = inet_csk(sk); + + if (icsk->icsk_ulp_ops && !icsk->icsk_ulp_ops->clone) + return -EINVAL; + + return 0; +} + int inet_csk_listen_start(struct sock *sk) { struct inet_connection_sock *icsk = inet_csk(sk); struct inet_sock *inet = inet_sk(sk); int err; + err = inet_ulp_can_listen(sk); + if (unlikely(err)) + return err; + reqsk_queue_alloc(&icsk->icsk_accept_queue); sk->sk_ack_backlog = 0; diff --git a/net/ipv4/tcp_ulp.c b/net/ipv4/tcp_ulp.c index 9ae50b1bd844..05b6077b9f2c 100644 --- a/net/ipv4/tcp_ulp.c +++ b/net/ipv4/tcp_ulp.c @@ -139,6 +139,10 @@ static int __tcp_set_ulp(struct sock *sk, const struct tcp_ulp_ops *ulp_ops) if (sk->sk_socket) clear_bit(SOCK_SUPPORT_ZC, &sk->sk_socket->flags); + err = -EINVAL; + if (!ulp_ops->clone && sk->sk_state == TCP_LISTEN) + goto out_err; + err = ulp_ops->init(sk); if (err) goto out_err;
When an ULP-enabled socket enters the LISTEN status, the listener ULP data pointer is copied inside the child/accepted sockets by sk_clone_lock(). The relevant ULP can take care of de-duplicating the context pointer via the clone() operation, but only MPTCP and SMC implement such op. Other ULPs may end-up with a double-free at socket disposal time. We can't simply clear the ULP data at clone time, as TLS replaces the socket ops with custom ones assuming a valid TLS ULP context is available. Instead completely prevent clone-less ULP sockets from entering the LISTEN status. Fixes: 734942cc4ea6 ("tcp: ULP infrastructure") Reported-by: slipper <slipper.alive@gmail.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com> --- net/ipv4/inet_connection_sock.c | 14 ++++++++++++++ net/ipv4/tcp_ulp.c | 4 ++++ 2 files changed, 18 insertions(+)