tunnel: Clear IPCB(skb)->opt before dst_link_failure called
authorBernie Harris <bernie.harris@alliedtelesis.co.nz>
Sun, 21 Feb 2016 23:58:05 +0000 (12:58 +1300)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Wed, 20 Apr 2016 06:41:56 +0000 (15:41 +0900)
commit207485dc4f22f8da29d2ba5e9e41c4c9e84d4e01
treed9a6066b2fb77bc947ddf4eeb97f1f8a21621451
parentd5322b91e61647630a91cacd7e22736c35dd98bd
tunnel: Clear IPCB(skb)->opt before dst_link_failure called

[ Upstream commit 5146d1f151122e868e594c7b45115d64825aee5f ]

IPCB may contain data from previous layers (in the observed case the
qdisc layer). In the observed scenario, the data was misinterpreted as
ip header options, which later caused the ihl to be set to an invalid
value (<5). This resulted in an infinite loop in the mips implementation
of ip_fast_csum.

This patch clears IPCB(skb)->opt before dst_link_failure can be called for
various types of tunnels. This change only applies to encapsulated ipv4
packets.

The code introduced in 11c21a30 which clears all of IPCB has been removed
to be consistent with these changes, and instead the opt field is cleared
unconditionally in ip_tunnel_xmit. The change in ip_tunnel_xmit applies to
SIT, GRE, and IPIP tunnels.

The relevant vti, l2tp, and pptp functions already contain similar code for
clearing the IPCB.

Signed-off-by: Bernie Harris <bernie.harris@alliedtelesis.co.nz>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
net/ipv4/ip_tunnel.c
net/ipv4/udp_tunnel.c
net/ipv6/ip6_gre.c
net/ipv6/ip6_tunnel.c