ipv6: Support new BIG TCP without HBH

Currently, BIG TCP IPv6 inserts a hop-by-hop extension header with a
jumbo payload option to reflect the real length of the packet bigger
than 65535 bytes. New kernels will drop this extension header and just
calculate the packet length from skb->len, like it's currently done for
BIG TCP IPv4.

Reflect the future kernel change in tcpdump and support parsing such
packets.

BIG TCP only sets payload_len = 0 when the actual payload length is
65536 or more. Check that explicitly, so that smaller packets with
payload_len = 0 are still considered invalid, like in the test called
icmpv6-length-zero. Do not check the transport protocol, though, because
TCP may be inner in an encapsulated packet, e.g., in UDP-based tunnels,
such as VXLAN and GENEVE.

Signed-off-by: Maxim Mikityanskiy <maxim@isovalent.com>

Author: gentoo-root

print-ip6.c

@@ -466,14 +466,23 @@ ip6_print(netdissect_options *ndo, const u_char *bp, u_int length)
 
 					/*
 					 * OK, we didn't see any extension
-					 * header, but that means we have
-					 * no payload, so set the length
-					 * to the IPv6 header length,
+					 * header - presume BIG TCP without
+					 * a Jumbo Payload option, but only
+					 * if it's actually >64k. Otherwise,
+					 * we have no payload, so set the
+					 * length to the IPv6 header length,
 					 * and change the snapshot length
 					 * accordingly.
 					 */
-					len = sizeof(struct ip6_hdr);
-					nd_change_snaplen(ndo, bp, len);
+					if (length > 65535) {
+						len = length;
+						if (ndo->ndo_vflag)
+							ND_PRINT("[real length %u, presumed BIG TCP] ",
+								 len);
+					} else {
+						len = sizeof(struct ip6_hdr);
+						nd_change_snaplen(ndo, bp, len);
+					}
 
 					/*
 					 * Now subtract the length of