From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Alexey Andriyanov" Subject: [net-next 1/2] ip6_tunnel: put ip6tnl0 FB device into 'any' mode Date: Tue, 21 Oct 2014 12:19:55 +0400 Message-ID: <1413879596-13726-1-git-send-email-alan@al-an.info> Cc: Alexey Andriyanov , David Miller , Eric Dumazet To: Roman Gushchin , netdev@vger.kernel.org Return-path: Received: from forward6l.mail.yandex.net ([84.201.143.139]:44335 "EHLO forward6l.mail.yandex.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750750AbaJUIUB (ORCPT ); Tue, 21 Oct 2014 04:20:01 -0400 Sender: netdev-owner@vger.kernel.org List-ID: The fallback device is in ipv6 mode by default. The mode can not be changed in runtime, so there is no way to decapsulate ip4in6 packets coming from various sources without creating the specific tunnel ifaces for each peer. Cc: David Miller Cc: Eric Dumazet Signed-off-by: Alexey Andriyanov --- net/ipv6/ip6_tunnel.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index 9409887..a48f212 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -1551,7 +1551,8 @@ static int __net_init ip6_fb_tnl_dev_init(struct net_device *dev) if (err) return err; - t->parms.proto = IPPROTO_IPV6; + /* allow any registered unrelying proto for the FB device */ + t->parms.proto = 0; dev_hold(dev); ip6_tnl_link_config(t); -- 1.9.1