Re: [PATCH net] tcp: fix various issues for sockets morphing to listen state

[Dmitry Vyukov <dvyukov@google.com>]

On Fri, Mar 3, 2017 at 11:08 PM, Eric Dumazet <eric.dumazet@gmail.com> wrote:
> From: Eric Dumazet <edumazet@google.com>
>
> Dmitry Vyukov reported a divide by 0 triggered by syzkaller, exploiting
> tcp_disconnect() path that was never really considered and/or used
> before syzkaller ;)
>
> I was not able to reproduce the bug, but it seems issues here are the
> three possible actions that assumed they would never trigger on a
> listener.
>
> 1) tcp_write_timer_handler
> 2) tcp_delack_timer_handler
> 3) MTU reduction
>
> Only IPv6 MTU reduction was properly testing TCP_CLOSE and TCP_LISTEN
>  states from tcp_v6_mtu_reduced()

I've run the test as "while ./a.out; do true; done". I've also noticed
that it either reproduces very quickly or does not happen at all. So I
rebooted machine, run the program in loop and then repeated if it does
not happen within seconds.

Retested with your patch, seems to not crash anymore.


> Signed-off-by: Eric Dumazet <edumazet@google.com>
> Reported-by: Dmitry Vyukov <dvyukov@google.com>
> ---
>  net/ipv4/tcp_ipv4.c  |    7 +++++--
>  net/ipv4/tcp_timer.c |    6 ++++--
>  2 files changed, 9 insertions(+), 4 deletions(-)
>
> diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
> index 9a89b8deafae1e9b2e8d1d9bc211c9c30b8dd8ec..8f3ec1365497a58972d31d419f88b34457b5ae39 100644
> --- a/net/ipv4/tcp_ipv4.c
> +++ b/net/ipv4/tcp_ipv4.c
> @@ -279,10 +279,13 @@ EXPORT_SYMBOL(tcp_v4_connect);
>   */
>  void tcp_v4_mtu_reduced(struct sock *sk)
>  {
> -       struct dst_entry *dst;
>         struct inet_sock *inet = inet_sk(sk);
> -       u32 mtu = tcp_sk(sk)->mtu_info;
> +       struct dst_entry *dst;
> +       u32 mtu;
>
> +       if ((1 << sk->sk_state) & (TCPF_LISTEN | TCPF_CLOSE))
> +               return;
> +       mtu = tcp_sk(sk)->mtu_info;
>         dst = inet_csk_update_pmtu(sk, mtu);
>         if (!dst)
>                 return;
> diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
> index 40d893556e6701ace6a02903e53c45822d6fa56d..b2ab411c6d3728fa7dbdebde045532a7317f5166 100644
> --- a/net/ipv4/tcp_timer.c
> +++ b/net/ipv4/tcp_timer.c
> @@ -249,7 +249,8 @@ void tcp_delack_timer_handler(struct sock *sk)
>
>         sk_mem_reclaim_partial(sk);
>
> -       if (sk->sk_state == TCP_CLOSE || !(icsk->icsk_ack.pending & ICSK_ACK_TIMER))
> +       if (((1 << sk->sk_state) & (TCPF_CLOSE | TCPF_LISTEN)) ||
> +           !(icsk->icsk_ack.pending & ICSK_ACK_TIMER))
>                 goto out;
>
>         if (time_after(icsk->icsk_ack.timeout, jiffies)) {
> @@ -552,7 +553,8 @@ void tcp_write_timer_handler(struct sock *sk)
>         struct inet_connection_sock *icsk = inet_csk(sk);
>         int event;
>
> -       if (sk->sk_state == TCP_CLOSE || !icsk->icsk_pending)
> +       if (((1 << sk->sk_state) & (TCPF_CLOSE | TCPF_LISTEN)) ||
> +           !icsk->icsk_pending)
>                 goto out;
>
>         if (time_after(icsk->icsk_timeout, jiffies)) {
>
>