VPP-1392: VXLAN fails with IP fragmentation

Not only is it wasteful to send all fragments back through ip4-lookup, but
it doesn't work with tunnel mechanisms that don't have IP enabled on their
payload side.

Change-Id: Ic92d95982dddaa70969a2a6ea2f98edec7614425
Signed-off-by: Ole Troan <ot@cisco.com>
diff --git a/src/vnet/ip/ip6_forward.c b/src/vnet/ip/ip6_forward.c
index efd5e0d..5abbba5 100644
--- a/src/vnet/ip/ip6_forward.c
+++ b/src/vnet/ip/ip6_forward.c
@@ -1567,7 +1567,7 @@
 	{
 	  /* IP fragmentation */
 	  ip_frag_set_vnet_buffer (b, 0, adj_packet_bytes,
-				   IP6_FRAG_NEXT_IP6_LOOKUP, 0);
+				   IP6_FRAG_NEXT_IP6_REWRITE, 0);
 	  *next = IP6_REWRITE_NEXT_FRAGMENT;
 	}
       else