Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <20220927161209.32939-1-kuniyu@amazon.com> <20220927161209.32939-4-kuniyu@amazon.com>
In-Reply-To: <20220927161209.32939-4-kuniyu@amazon.com>
From:   Eric Dumazet <edumazet@google.com>
Date:   Tue, 27 Sep 2022 09:50:04 -0700
Message-ID: <CANn89iKguA1pAc7wUuWVwuSLJ7+dDRLscY0CEJXNPpg8gphJbg@mail.gmail.com>
Subject: Re: [PATCH v1 net 3/5] tcp/udp: Call inet6_destroy_sock() in IPv4 sk_prot->destroy().
To:     Kuniyuki Iwashima <kuniyu@amazon.com>
Cc:     "David S. Miller" <davem@davemloft.net>,
        Jakub Kicinski <kuba@kernel.org>,
        Paolo Abeni <pabeni@redhat.com>,
        David Ahern <dsahern@kernel.org>,
        Kuniyuki Iwashima <kuni1840@gmail.com>,
        netdev <netdev@vger.kernel.org>,
        syzkaller-bugs <syzkaller-bugs@googlegroups.com>,
        LKML <linux-kernel@vger.kernel.org>,
        Vladislav Yasevich <vyasevic@redhat.com>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Tue, Sep 27, 2022 at 9:13 AM Kuniyuki Iwashima <kuniyu@amazon.com> wrote:
>
> Originally, inet6_sk(sk)->XXX were changed under lock_sock(), so we were
> able to clean them up by calling inet6_destroy_sock() during the IPv6 ->
> IPv4 conversion by IPV6_ADDRFORM.  However, commit 03485f2adcde ("udpv6:
> Add lockless sendmsg() support") added a lockless memory allocation path,
> which could cause a memory leak:
>
> setsockopt(IPV6_ADDRFORM)                 sendmsg()
> +-----------------------+                 +-------+
> - do_ipv6_setsockopt(sk, ...)             - udpv6_sendmsg(sk, ...)
>   - lock_sock(sk)                           ^._ called via udpv6_prot
>   - WRITE_ONCE(sk->sk_prot, &tcp_prot)          before WRITE_ONCE()
>   - inet6_destroy_sock()
>   - release_sock(sk)                        - ip6_make_skb(sk, ...)
>                                               ^._ lockless fast path for
>                                                   the non-corking case
>
>                                               - __ip6_append_data(sk, ...)
>                                                 - ipv6_local_rxpmtu(sk, ...)
>                                                   - xchg(&np->rxpmtu, skb)
>                                                     ^._ rxpmtu is never freed.
>
>                                             - lock_sock(sk)
>
> For now, rxpmtu is only the case, but let's call inet6_destroy_sock()
> in both TCP/UDP v4 destroy functions not to miss the future change.
>
> We can consolidate TCP/UDP v4/v6 destroy functions, but such changes
> are too invasive to backport to stable.  So, they can be posted as a
> follow-up later for net-next.
>
> Fixes: 03485f2adcde ("udpv6: Add lockless sendmsg() support")
> Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
> ---
> Cc: Vladislav Yasevich <vyasevic@redhat.com>
> ---
>  net/ipv4/tcp_ipv4.c | 5 +++++
>  net/ipv4/udp.c      | 6 ++++++
>  net/ipv6/tcp_ipv6.c | 1 -
>  3 files changed, 11 insertions(+), 1 deletion(-)
>
> diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
> index 5b019ba2b9d2..035b6c52a243 100644
> --- a/net/ipv4/tcp_ipv4.c
> +++ b/net/ipv4/tcp_ipv4.c
> @@ -2263,6 +2263,11 @@ void tcp_v4_destroy_sock(struct sock *sk)
>         tcp_saved_syn_free(tp);
>
>         sk_sockets_allocated_dec(sk);
> +
> +#if IS_ENABLED(CONFIG_IPV6)
> +       if (sk->sk_prot_creator == &tcpv6_prot)
> +               inet6_destroy_sock(sk);
> +#endif
>  }

This is ugly, and will not compile with CONFIG_IPV6=m, right ?


>  EXPORT_SYMBOL(tcp_v4_destroy_sock);
>
> diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
> index 560d9eadeaa5..cdf131c0a819 100644
> --- a/net/ipv4/udp.c
> +++ b/net/ipv4/udp.c
> @@ -115,6 +115,7 @@
>  #include <net/udp_tunnel.h>
>  #if IS_ENABLED(CONFIG_IPV6)
>  #include <net/ipv6_stubs.h>
> +#include <net/transp_v6.h>
>  #endif
>
>  struct udp_table udp_table __read_mostly;
> @@ -2666,6 +2667,11 @@ void udp_destroy_sock(struct sock *sk)
>                 if (up->encap_enabled)
>                         static_branch_dec(&udp_encap_needed_key);
>         }
> +
> +#if IS_ENABLED(CONFIG_IPV6)
> +       if (sk->sk_prot_creator == &udpv6_prot)
> +               inet6_destroy_sock(sk);
> +#endif
>  }
>
>  /*
> diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
> index e54eee80ce5f..1ff6a92f7774 100644
> --- a/net/ipv6/tcp_ipv6.c
> +++ b/net/ipv6/tcp_ipv6.c
> @@ -1945,7 +1945,6 @@ static int tcp_v6_init_sock(struct sock *sk)
>  static void tcp_v6_destroy_sock(struct sock *sk)
>  {
>         tcp_v4_destroy_sock(sk);
> -       inet6_destroy_sock(sk);
>  }
>
>  #ifdef CONFIG_PROC_FS
> --
> 2.30.2
>