ipv6: Fix ip6_xmit to send fragments if ipfragok is true
authorWei Yongjun <yjwei@cn.fujitsu.com>
Fri, 1 Aug 2008 03:46:47 +0000 (20:46 -0700)
committerDavid S. Miller <davem@davemloft.net>
Fri, 1 Aug 2008 03:46:47 +0000 (20:46 -0700)
SCTP used ip6_xmit() to send fragments after received ICMP packet too
big message. But while send packet used ip6_xmit, the skb->local_df is
not initialized. So when skb if enter ip6_fragment(), the following
code will discard the skb.

ip6_fragment(...)
{
    if (!skb->local_df) {
        ...
        return -EMSGSIZE;
    }
    ...
}

SCTP do the following step:
1. send packet ip6_xmit(skb, ipfragok=0)
2. received ICMP packet too big message
3. if PMTUD_ENABLE: ip6_xmit(skb, ipfragok=1)

This patch fixed the problem by set local_df if ipfragok is true.

Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
Acked-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
net/ipv6/ip6_output.c

index 6811901..a027003 100644 (file)
@@ -236,6 +236,10 @@ int ip6_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl,
        skb_reset_network_header(skb);
        hdr = ipv6_hdr(skb);
 
+       /* Allow local fragmentation. */
+       if (ipfragok)
+               skb->local_df = 1;
+
        /*
         *      Fill in the IPv6 header
         */