tunnels: Don't apply GRO to multiple layers of encapsulation.
authorJesse Gross <jesse@kernel.org>
Sat, 19 Mar 2016 16:32:01 +0000 (09:32 -0700)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Mon, 31 Oct 2016 10:13:59 +0000 (04:13 -0600)
commit5699b3431e0b14736867484b8669ead2d40f575e
tree88fc51d049d2f4e484b98135418f17a5ae97a2b5
parent419cabdcc4818dd06a1827e5e84df7738900a458
tunnels: Don't apply GRO to multiple layers of encapsulation.

commit fac8e0f579695a3ecbc4d3cac369139d7f819971 upstream.

When drivers express support for TSO of encapsulated packets, they
only mean that they can do it for one layer of encapsulation.
Supporting additional levels would mean updating, at a minimum,
more IP length fields and they are unaware of this.

No encapsulation device expresses support for handling offloaded
encapsulated packets, so we won't generate these types of frames
in the transmit path. However, GRO doesn't have a check for
multiple levels of encapsulation and will attempt to build them.

UDP tunnel GRO actually does prevent this situation but it only
handles multiple UDP tunnels stacked on top of each other. This
generalizes that solution to prevent any kind of tunnel stacking
that would cause problems.

Fixes: bf5a755f ("net-gre-gro: Add GRE support to the GRO stack")
Signed-off-by: Jesse Gross <jesse@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Juerg Haefliger <juerg.haefliger@hpe.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
include/linux/netdevice.h
net/core/dev.c
net/ipv4/af_inet.c
net/ipv4/gre_offload.c
net/ipv4/udp_offload.c
net/ipv6/ip6_offload.c