From mboxrd@z Thu Jan 1 00:00:00 1970 From: Alexey Kodanev Subject: [PATCH net-next] ip6_vti: adjust vti mtu according to mtu of output device Date: Wed, 6 Dec 2017 19:38:19 +0300 Message-ID: <1512578299-7573-1-git-send-email-alexey.kodanev@oracle.com> Cc: Steffen Klassert , David Miller , Petr Vorel , Alexey Kodanev To: netdev@vger.kernel.org Return-path: Received: from userp2130.oracle.com ([156.151.31.86]:51971 "EHLO userp2130.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751544AbdLFQaA (ORCPT ); Wed, 6 Dec 2017 11:30:00 -0500 Sender: netdev-owner@vger.kernel.org List-ID: LTP/udp6_ipsec_vti tests fail when sending large UDP datagrams that require fragmentation and underlying device MTU <= 1500. This happens because ip6_vti sets mtu to ETH_DATA_LEN and not updating it depending on a destiantion address. Futhure attempts to send UDP packets may succeed because pmtu get updated on ICMPV6_PKT_TOOBIG in vti6_err(). Here is the example when output device MTU set to 9000: # ip a sh ltp_ns_veth2 ltp_ns_veth2@if7: mtu 9000 ... inet 10.0.0.2/24 scope global ltp_ns_veth2 inet6 fd00::2/64 scope global ... # ip li add vti6 type vti6 local fd00::2 remote fd00::1 # ip li show vti6 vti6@NONE: mtu 1500 ... link/tunnel6 fd00::2 peer fd00::1 After the patch: # ip li add vti6 type vti6 local fd00::2 remote fd00::1 # ip li show vti6 vti6@NONE: mtu 8832 ... link/tunnel6 fd00::2 peer fd00::1 Regarding ip_vti, it already tunes mtu with ip_tunnel_bind_dev(): # ip li add vti4 type vti local 10.0.0.2 remote 10.0.0.1 # ip li sh vti4 vti4@NONE: mtu 8832 ... link/ipip 10.0.0.2 peer 10.0.0.1 Reported-by: Petr Vorel Signed-off-by: Alexey Kodanev --- ip6_vti mtu offset is the same (168) as in ip_vti because ip_vti offset includes two sizes of struct iphdr: in dev->hard_header_len and in t_hlen in ip_tunnel_bind_dev(). I'm not sure if it's correct. net/ipv6/ip6_vti.c | 18 ++++++++++++++++++ 1 files changed, 18 insertions(+), 0 deletions(-) diff --git a/net/ipv6/ip6_vti.c b/net/ipv6/ip6_vti.c index dbb74f3..47e6464 100644 --- a/net/ipv6/ip6_vti.c +++ b/net/ipv6/ip6_vti.c @@ -638,6 +638,24 @@ static void vti6_link_config(struct ip6_tnl *t) dev->flags |= IFF_POINTOPOINT; else dev->flags &= ~IFF_POINTOPOINT; + + if (p->flags & IP6_TNL_F_CAP_XMIT) { + int strict = (ipv6_addr_type(&p->raddr) & + (IPV6_ADDR_MULTICAST | IPV6_ADDR_LINKLOCAL)); + + struct rt6_info *rt = rt6_lookup(t->net, + &p->raddr, &p->laddr, + p->link, strict); + + if (!rt) + return; + + if (rt->dst.dev) { + dev->mtu = max(rt->dst.dev->mtu - dev->hard_header_len, + IPV6_MIN_MTU); + } + ip6_rt_put(rt); + } } /** -- 1.7.1