Fix path MTU discovery for TCP LRO/TSO when forwarding.

When doing LRO (Large Receive Offload), the drivers, currently ix(4)
and lo(4) only, record an upper bound of the size of the original
packets in ph_mss.  When sending, either stack or hardware must
chop the packets with TSO (TCP Segmentation Offload) to that size.
That means we have to call tcp_if_output_tso() before ifp->if_output().
Put that logic into if_output_tso() to avoid code duplication.  As
TCP packets on the wire do not get larger that way, path MTU discovery
should still work.

tested by and OK jan@

1 files changed, 2 insertions, 14 deletions

diff --git a/sys/netinet6/ip6_forward.c b/sys/netinet6/ip6_forward.c
index db68b32eca8..7de1e0ec701 100644
--- a/sys/netinet6/ip6_forward.c
+++ b/sys/netinet6/ip6_forward.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: ip6_forward.c,v 1.111 2023/06/16 19:18:56 bluhm Exp $	*/
+/*	$OpenBSD: ip6_forward.c,v 1.112 2023/07/07 08:05:02 bluhm Exp $	*/
 /*	$KAME: ip6_forward.c,v 1.75 2001/06/29 12:42:13 jinmei Exp $	*/
 
 /*
@@ -319,8 +319,7 @@ reroute:
 	}
 #endif
 
-	error = tcp_if_output_tso(ifp, &m, sin6tosa(sin6), rt, IFCAP_TSOv6,
-	    ifp->if_mtu);
+	error = if_output_tso(ifp, &m, sin6tosa(sin6), rt, ifp->if_mtu);
 	if (error)
 		ip6stat_inc(ip6s_cantforward);
 	else if (m == NULL)
@@ -328,17 +327,6 @@ reroute:
 	if (error || m == NULL)
 		goto senderr;
 
-	/* Check the size after pf_test to give pf a chance to refragment. */
-	if (m->m_pkthdr.len <= ifp->if_mtu) {
-		in6_proto_cksum_out(m, ifp);
-		error = ifp->if_output(ifp, m, sin6tosa(sin6), rt);
-		if (error)
-			ip6stat_inc(ip6s_cantforward);
-		else
-			ip6stat_inc(ip6s_forward);
-		goto senderr;
-	}
-
 	if (mcopy != NULL)
 		icmp6_error(mcopy, ICMP6_PACKET_TOO_BIG, 0, ifp->if_mtu);
 	m_freem(m);