netdev - Re: Unix Socket buffer attribution

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <1358951180.12374.787.camel@edumazet-glaptop>
Date:	Wed, 23 Jan 2013 06:26:20 -0800
From:	Eric Dumazet <eric.dumazet@...il.com>
To:	Cong Wang <xiyou.wangcong@...il.com>
Cc:	netdev@...r.kernel.org
Subject: Re: Unix Socket buffer attribution

On Wed, 2013-01-23 at 11:42 +0000, Cong Wang wrote:
> On Tue, 22 Jan 2013 at 02:01 GMT, Yannick Koehler <yannick@...hler.name> wrote:
> >
> >   I believe that the problem is that once we move the skb into the
> > client's receive queue we need to decrease the sk_wmem_alloc variable
> > of the server socket since that skb is no more tied to the server.
> > The code should then account for this memory as part of the
> > sk_rmem_alloc variable on the client's socket.  The function
> > "skb_set_owner_r(skb,owner)" would seem to be the function to do that,
> > so it would seem to me.
> 
> Something like below??
> 
> -------->
> 
> diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
> index 0c61236..e273072 100644
> --- a/net/unix/af_unix.c
> +++ b/net/unix/af_unix.c
> @@ -1205,6 +1205,7 @@ restart:
>  
>  	unix_state_unlock(sk);
>  
> +	skb_set_owner_r(skb, other);
>  	/* take ten and and send info to listening sock */
>  	spin_lock(&other->sk_receive_queue.lock);
>  	__skb_queue_tail(&other->sk_receive_queue, skb);
> @@ -1578,6 +1579,7 @@ restart:
>  
>  	if (sock_flag(other, SOCK_RCVTSTAMP))
>  		__net_timestamp(skb);
> +	skb_set_owner_r(skb, other);
>  	maybe_add_creds(skb, sock, other);
>  	skb_queue_tail(&other->sk_receive_queue, skb);
>  	if (max_level > unix_sk(other)->recursion_level)
> @@ -1693,6 +1695,7 @@ static int unix_stream_sendmsg(struct kiocb *kiocb, struct socket *sock,
>  		    (other->sk_shutdown & RCV_SHUTDOWN))
>  			goto pipe_err_free;
>  
> +		skb_set_owner_r(skb, other);
>  		maybe_add_creds(skb, sock, other);
>  		skb_queue_tail(&other->sk_receive_queue, skb);
>  		if (max_level > unix_sk(other)->recursion_level)
> 

So what prevents a malicious program to DOS the machine ?

Current behavior is on purpose. Limited, predictable, but less holes.

Some applications might depend on the current flow control : Limiting
the working set also permits to keep cpu caches hot.

If you want to change it, better do a full analysis, because hackers
will do it.

Its probably doable, but with a "man unix" change.



--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html