From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id S1757154AbaISHE1 (ORCPT <rfc822;w@1wt.eu>);
	Fri, 19 Sep 2014 03:04:27 -0400
Received: from mail-pd0-f181.google.com ([209.85.192.181]:39256 "EHLO
	mail-pd0-f181.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1755428AbaISHE0 (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Fri, 19 Sep 2014 03:04:26 -0400
Message-ID: <1411110263.7106.301.camel@edumazet-glaptop2.roam.corp.google.com>
Subject: Re: [PATCH net-next] net: keep original skb which only needs header
 checking during software GSO
From: Eric Dumazet <eric.dumazet@gmail.com>
To: Jason Wang <jasowang@redhat.com>
Cc: davem@davemloft.net, netdev@vger.kernel.org, linux-kernel@vger.kernel.org
Date: Fri, 19 Sep 2014 00:04:23 -0700
In-Reply-To: <1411108686-20602-1-git-send-email-jasowang@redhat.com>
References: <1411108686-20602-1-git-send-email-jasowang@redhat.com>
Content-Type: text/plain; charset="UTF-8"
X-Mailer: Evolution 3.2.3-0ubuntu6 
Content-Transfer-Encoding: 7bit
Mime-Version: 1.0
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On Fri, 2014-09-19 at 14:38 +0800, Jason Wang wrote:
> Commit ce93718fb7cdbc064c3000ff59e4d3200bdfa744 ("net: Don't keep
> around original SKB when we software segment GSO frames") frees the
> original skb after software GSO even for dodgy gso skbs. This breaks
> the stream throughput from untrusted sources, since only header
> checking was done during software GSO instead of a true
> segmentation. This patch fixes this by freeing the original gso skb
> only when it was really segmented by software.
> 
> Fixes ce93718fb7cdbc064c3000ff59e4d3200bdfa744 ("net: Don't keep
> around original SKB when we software segment GSO frames.")
> 
> CC: David S. Miller <davem@davemloft.net>
> Signed-off-by: Jason Wang <jasowang@redhat.com>
> ---
>  net/core/dev.c | 6 ++++--
>  1 file changed, 4 insertions(+), 2 deletions(-)
> 
> diff --git a/net/core/dev.c b/net/core/dev.c
> index e916ba8..b7a0e1d 100644
> --- a/net/core/dev.c
> +++ b/net/core/dev.c
> @@ -2694,10 +2694,12 @@ struct sk_buff *validate_xmit_skb(struct sk_buff *skb, struct net_device *dev)
>  		struct sk_buff *segs;
>  
>  		segs = skb_gso_segment(skb, features);
> -		kfree_skb(skb);
>  		if (IS_ERR(segs))
>  			
> -		skb = segs;
> +		else if (segs) {
> +			kfree_skb(skb);
> +			skb = segs;
> +		}
>  	} else {
>  		if (skb_needs_linearize(skb, features) &&
>  		    __skb_linearize(skb))

Good catch !

While we are at it, could you use consume_skb() instead of kfree_skb(),
and add missing {} (CodingStyle) ?

if (IS_ERR(segs)) {
	segs = NULL;
} else if (segs) {
	consume_skb(skb);
	skb = segs;
}


Thanks !