Re: [bpf PATCH v2 2/3] bpf: sockmap, fix transition through disconnect without close

[John Fastabend <john.fastabend@gmail.com>]

On 09/17/2018 10:59 AM, John Fastabend wrote:
> It is possible (via shutdown()) for TCP socks to go trough TCP_CLOSE
> state via tcp_disconnect() without actually calling tcp_close which
> would then call our bpf_tcp_close() callback. Because of this a user
> could disconnect a socket then put it in a LISTEN state which would
> break our assumptions about sockets always being ESTABLISHED state.
> 
> To resolve this rely on the unhash hook, which is called in the
> disconnect case, to remove the sock from the sockmap.
> 

Sorry for the noise will need a v3 actually.

> Reported-by: Eric Dumazet <edumazet@google.com>
> Fixes: 1aa12bdf1bfb ("bpf: sockmap, add sock close() hook to remove socks")
> Signed-off-by: John Fastabend <john.fastabend@gmail.com>
> ---
>  kernel/bpf/sockmap.c |   71 +++++++++++++++++++++++++++++++++++++-------------
>  1 file changed, 52 insertions(+), 19 deletions(-)

[...]


> +}
> +
> +static void bpf_tcp_unhash(struct sock *sk)
> +{
> +	void (*unhash_fun)(struct sock *sk);
> +	struct smap_psock *psock;
> +
> +	rcu_read_lock();
> +	psock = smap_psock_sk(sk);
> +	if (unlikely(!psock)) {
> +		rcu_read_unlock();
> +		release_sock(sk);
                 ^^^^^^^^^^^^^^^^^
> +		return sk->sk_prot->unhash(sk);

		if (sk->sk_prot->unhash) ...
		else return;

Thanks,
John