netdev - Re: [Patch net] unix: correct sk_rmem

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] [day] [month] [year] [list]

Date:	Wed, 23 Jan 2013 07:03:47 -0800
From:	Eric Dumazet <eric.dumazet@...il.com>
To:	Cong Wang <xiyou.wangcong@...il.com>
Cc:	netdev@...r.kernel.org, Yannick Koehler <yannick@...hler.name>,
	"David S. Miller" <davem@...emloft.net>
Subject: Re: [Patch net] unix: correct sk_rmem_alloc accounting

On Wed, 2013-01-23 at 22:43 +0800, Cong Wang wrote:
> From: Cong Wang <xiyou.wangcong@...il.com>
> 
> Yannick reported [1] we probably forgot to account ->sk_rmem_alloc
> before moving the skb to other->sk_receive_queue. I believe
> he is right. So, just call skb_set_owner_r() before every time
> we queuing skb into other->sk_receive_queue. 
> 
> 1. http://marc.info/?l=linux-netdev&m=135882012924930&w=2
> 
> Reported-by: Yannick Koehler <yannick@...hler.name>
> Cc: Yannick Koehler <yannick@...hler.name>
> Cc: Eric Dumazet <eric.dumazet@...il.com>
> Cc: David S. Miller <davem@...emloft.net>
> Signed-off-by: Cong Wang <xiyou.wangcong@...il.com>
> 
> ---
> diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
> index 5b5c876..7e9dba3 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);
> @@ -1579,6 +1580,7 @@ restart:
>  	if (sock_flag(other, SOCK_RCVTSTAMP))
>  		__net_timestamp(skb);
>  	maybe_add_creds(skb, sock, other);
> +	skb_set_owner_r(skb, other);
>  	skb_queue_tail(&other->sk_receive_queue, skb);
>  	if (max_level > unix_sk(other)->recursion_level)
>  		unix_sk(other)->recursion_level = max_level;
> @@ -1694,6 +1696,7 @@ static int unix_stream_sendmsg(struct kiocb *kiocb, struct socket *sock,
>  			goto pipe_err_free;
>  
>  		maybe_add_creds(skb, sock, other);
> +		skb_set_owner_r(skb, other);
>  		skb_queue_tail(&other->sk_receive_queue, skb);
>  		if (max_level > unix_sk(other)->recursion_level)
>  			unix_sk(other)->recursion_level = max_level;

I Nack this patch, for reasons already given.



--
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