Re: TCP-MD5 checksum failure on x86_64 SMP

[Stephen Hemminger <shemminger@vyatta.com>]

On Wed, 12 May 2010 05:20:21 +0200
Eric Dumazet <eric.dumazet@gmail.com> wrote:

> Le mardi 11 mai 2010 à 22:50 +0200, Eric Dumazet a écrit :
> > Le mardi 11 mai 2010 à 04:08 +0000, Bijay Singh a écrit :
> > > Hi Eric,
> > > 
> > > I guess that makes me the enviable one. So I am keen to test out this feature completely, as long as I know what to do as a next step, directions, patches.
> > > 
> > > Thanks
> > 
> > 
> > I believe third problem comes from commit 4957faad
> > (TCPCT part 1g: Responder Cookie => Initiator), from William Allen
> > Simpson.
> 
> And a fourth problem might be that tcp_md5_hash_skb_data() is not
> frag_list aware ?
> 
> 
> 
> 
> diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
> index 8ce2974..56ee0f2 100644
> --- a/net/ipv4/tcp.c
> +++ b/net/ipv4/tcp.c
> @@ -2985,6 +2985,7 @@ int tcp_md5_hash_skb_data(struct tcp_md5sig_pool *hp,
>  	const unsigned head_data_len = skb_headlen(skb) > header_len ?
>  				       skb_headlen(skb) - header_len : 0;
>  	const struct skb_shared_info *shi = skb_shinfo(skb);
> +	struct sk_buff *frag_iter;
>  
>  	sg_init_table(&sg, 1);
>  
> @@ -2999,6 +3000,10 @@ int tcp_md5_hash_skb_data(struct tcp_md5sig_pool *hp,
>  			return 1;
>  	}
>  
> +	skb_walk_frags(skb, frag_iter)
> +		if (tcp_md5_hash_skb_data(hp, frag_iter, 0))
> +			return 1;
> +
>  	return 0;
>  }

Yes, that looks like a possible bug, not sure what hardware
generates frag_list.