From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: [PATCH] net/ipv6/sit.c: return unhandled skb to tunnel4_rcv Date: Tue, 30 Nov 2010 07:12:44 +0100 Message-ID: <1291097564.2725.15.camel@edumazet-laptop> References: <20101130053234.GA28609@mcafee.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: netdev@vger.kernel.org To: David McCullough Return-path: Received: from mail-ww0-f44.google.com ([74.125.82.44]:36810 "EHLO mail-ww0-f44.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751730Ab0K3GMu (ORCPT ); Tue, 30 Nov 2010 01:12:50 -0500 Received: by wwa36 with SMTP id 36so5671986wwa.1 for ; Mon, 29 Nov 2010 22:12:49 -0800 (PST) In-Reply-To: <20101130053234.GA28609@mcafee.com> Sender: netdev-owner@vger.kernel.org List-ID: Le mardi 30 novembre 2010 =C3=A0 15:32 +1000, David McCullough a =C3=A9= crit : > Hi all, >=20 > I found a problem using an IPv6 over IPv4 tunnel. When CONFIG_IPV6_S= IT > was enabled, the packets would be rejected as net/ipv6/sit.c was catc= hing > all IPPROTO_IPV6 packets and returning an ICMP port unreachable error= =2E >=20 > I think this patch fixes the problem cleanly. I believe the code in > net/ipv4/tunnel4.c:tunnel4_rcv takes care of it properly if none of t= he > handlers claim the skb. >=20 > Patch is against 2.6.35, bit still applies to current. >=20 > Cheers, > Davidm >=20 > Signed-off-by: David McCullough >=20 > diff -u -r1.1.1.45 sit.c > --- a/net/ipv6/sit.c 12 Aug 2010 00:16:26 -0000 1.1.1.45 > +++ b/net/ipv6/sit.c 30 Nov 2010 05:01:19 -0000 > @@ -575,8 +575,9 @@ > return 0; > } > =20 > - icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0); > + /* no tunnel matched, let upstream know, ipsec may handle it */ > rcu_read_unlock(); > + return 1; > out: > kfree_skb(skb); > return 0; >=20 Good catch ! Acked-by: Eric Dumazet