From: Philippe Guibert <philippe.guibert@6wind.com>
Date: Mon, 13 Feb 2023 11:18:33 +0000 (+0100)
Subject: bgpd: clarify when the vpnv6 nexthop length must be modified
X-Git-Tag: base_9.0~379^2
X-Git-Url: https://git.puffer.fish/?a=commitdiff_plain;h=refs%2Fpull%2F12795%2Fhead;p=mirror%2Ffrr.git

bgpd: clarify when the vpnv6 nexthop length must be modified

Using a route-map to update the local ipv6 address has to be
better clarified. Actually, when a VPN SAFI is used, the nexthop
length must be changed to 48 bytes. Other cases, the length will
be 32 bytes.

Fixes: 9795e9f23465 ("bgpd: fix when route-map changes the link local
nexthop for vpnv6")

Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
---

diff --git a/bgpd/bgp_routemap.c b/bgpd/bgp_routemap.c
index 5aa3ec4fd1..4a2029c026 100644
--- a/bgpd/bgp_routemap.c
+++ b/bgpd/bgp_routemap.c
@@ -3650,17 +3650,29 @@ route_set_ipv6_nexthop_local(void *rule, const struct prefix *p, void *object)
 {
 	struct in6_addr *address;
 	struct bgp_path_info *path;
+	struct bgp_dest *dest;
+	struct bgp_table *table = NULL;
 
 	/* Fetch routemap's rule information. */
 	address = rule;
 	path = object;
+	dest = path->net;
+
+	if (!dest)
+		return RMAP_OKAY;
+
+	table = bgp_dest_table(dest);
+	if (!table)
+		return RMAP_OKAY;
 
 	/* Set next hop value. */
 	path->attr->mp_nexthop_local = *address;
 
 	/* Set nexthop length. */
-	if (path->attr->mp_nexthop_len != BGP_ATTR_NHLEN_IPV6_GLOBAL_AND_LL &&
-	    path->attr->mp_nexthop_len != BGP_ATTR_NHLEN_VPNV6_GLOBAL_AND_LL)
+	if (table->safi == SAFI_MPLS_VPN || table->safi == SAFI_ENCAP ||
+	    table->safi == SAFI_EVPN)
+		path->attr->mp_nexthop_len = BGP_ATTR_NHLEN_VPNV6_GLOBAL_AND_LL;
+	else
 		path->attr->mp_nexthop_len = BGP_ATTR_NHLEN_IPV6_GLOBAL_AND_LL;
 
 	SET_FLAG(path->attr->rmap_change_flags,