net: l2: openthread: Improve Border Router packet forwarding logic

There are some cases when OpenThread opens a sockets and doesn't choose
as default it's internal interface, this leading to usage of
platform UDP module which will then send back the packet to the
OpenThread interface. In this case, the packet should not be treated as
originated from backbone interface.

Backbone router multicast listener callback functionality is improved.
A route with a prefix length of 128 is set
and a multicast address is added for each listener registration.
OpenThread interface joins that multicast address group.

Enabled forwarding capabilities for Backbone interface.
A border router should be able to perform default packet forwarding for
destination addresses with a multicast scope greater than admin-local.
In order to achieve this, multicast routes have been added to those
addreses. [https://datatracker.ietf.org/doc/rfc7346/]

For Border Router application, `ip6_addr_cb` is not installed.
otIp6SubscribeMulticastAddress call would re-register an IPV6 multicast
address which might have been registered by an OpenThread node using
`ipmadd add` command and even if that node performed `ipmaddr del`,
the address was still present in multicast listener table. This also
led to a missing MLDv2 message with that specific multicast IPV6
address.

Signed-off-by: Cristian Bulacu <[email protected]>

Author: Cristib05

subsys/net/l2/openthread/openthread.c

@@ -316,9 +316,12 @@ static int openthread_l2_init(struct net_if *iface)
 	ot_l2_context->iface = iface;
 
 	if (!IS_ENABLED(CONFIG_OPENTHREAD_COPROCESSOR)) {
-		net_mgmt_init_event_callback(&ip6_addr_cb, ipv6_addr_event_handler,
-					     NET_EVENT_IPV6_ADDR_ADD | NET_EVENT_IPV6_MADDR_ADD);
-		net_mgmt_add_event_callback(&ip6_addr_cb);
+		if (!IS_ENABLED(CONFIG_OPENTHREAD_ZEPHYR_BORDER_ROUTER)) {
+			net_mgmt_init_event_callback(&ip6_addr_cb, ipv6_addr_event_handler,
+						     NET_EVENT_IPV6_ADDR_ADD |
+						     NET_EVENT_IPV6_MADDR_ADD);
+			net_mgmt_add_event_callback(&ip6_addr_cb);
+		}
 		net_if_dormant_on(iface);
 
 		openthread_set_receive_cb(ot_receive_handler, (void *)ot_l2_context);

subsys/net/l2/openthread/openthread_border_router.c

@@ -25,6 +25,7 @@
 #include <ipv6.h>
 #include <zephyr/net/net_mgmt.h>
 #include <zephyr/net/openthread.h>
+#include <zephyr/sys/util.h>
 
 #include <inttypes.h>
 #include <stdarg.h>
@@ -38,6 +39,7 @@ static struct net_mgmt_event_callback ail_net_event_ipv4_addr_add_cb;
 #endif /* CONFIG_NET_IPV4 */
 static uint32_t ail_iface_index;
 static struct net_if *ail_iface_ptr;
+static struct net_if *ot_iface_ptr;
 static bool is_border_router_started;
 char otbr_vendor_name[] = OTBR_VENDOR_NAME;
 char otbr_base_service_instance_name[] = OTBR_BASE_SERVICE_INSTANCE_NAME;
@@ -53,6 +55,7 @@ K_MEM_SLAB_DEFINE_STATIC(border_router_messages_slab, sizeof(struct otbr_msg_ctx
 		  CONFIG_OPENTHREAD_ZEPHYR_BORDER_ROUTER_MSG_POOL_NUM, sizeof(void *));
 
 static const char *create_base_name(otInstance *ot_instance, char *base_name);
+static void openthread_border_router_add_route_to_multicast_groups(void);
 
 #if defined(CONFIG_NET_IPV4)
 static void openthread_border_router_check_for_dhcpv4_addr(struct net_if *iface,
@@ -66,8 +69,12 @@ int openthread_start_border_router_services(struct net_if *ot_iface, struct net_
 	otInstance *instance = openthread_get_default_instance();
 	ail_iface_index = (uint32_t)net_if_get_by_iface(ail_iface);
 	ail_iface_ptr = ail_iface;
+	ot_iface_ptr = ot_iface;
 
 	net_if_flag_set(ot_iface, NET_IF_FORWARD_MULTICASTS);
+	net_if_flag_set(ail_iface, NET_IF_FORWARD_MULTICASTS);
+
+	openthread_border_router_add_route_to_multicast_groups();
 
 	openthread_mutex_lock();
 
@@ -266,19 +273,43 @@ static void ot_bbr_multicast_listener_handler(void *context,
 					      otBackboneRouterMulticastListenerEvent event,
 					      const otIp6Address *address)
 {
-	struct openthread_context *ot_context = context;
-	struct in6_addr mcast_prefix = {0};
+	struct openthread_context *ot_context = (struct openthread_context *)context;
+	struct in6_addr recv_addr = {0};
+	struct net_if_mcast_addr *mcast_addr = NULL;
+	struct net_route_entry_mcast *entry = NULL;
 
-	memcpy(mcast_prefix.s6_addr, address->mFields.m32, sizeof(otIp6Address));
+	memcpy(recv_addr.s6_addr, address->mFields.m32, sizeof(otIp6Address));
 
 	if (event == OT_BACKBONE_ROUTER_MULTICAST_LISTENER_ADDED) {
-		net_route_mcast_add((struct net_if *)ot_context->iface, &mcast_prefix, 16);
+		entry = net_route_mcast_lookup(&recv_addr);
+		if (entry == NULL) {
+			entry = net_route_mcast_add(ot_context->iface, &recv_addr,
+						    NUM_BITS(struct in6_addr));
+		}
+		if (entry != NULL) {
+			/*
+			 * No need to perform mcast_lookup explicitly as it's already done in
+			 * net_if_ipv6_maddr_add call. If it's found, NULL will be returned
+			 * and maddr_join will not be performed.
+			 */
+			mcast_addr = net_if_ipv6_maddr_add(ot_context->iface,
+							   (const struct in6_addr *)&recv_addr);
+			if (mcast_addr != NULL) {
+				net_if_ipv6_maddr_join(ot_context->iface, mcast_addr);
+			}
+		}
 	} else {
-		struct net_route_entry_mcast *route_to_del = net_route_mcast_lookup(&mcast_prefix);
+		struct net_route_entry_mcast *route_to_del = net_route_mcast_lookup(&recv_addr);
+		struct net_if_mcast_addr *addr_to_del;
 
+		addr_to_del = net_if_ipv6_maddr_lookup(&recv_addr, &(ot_context->iface));
 		if (route_to_del != NULL) {
 			net_route_mcast_del(route_to_del);
 		}
+
+		if (addr_to_del != NULL && net_if_ipv6_maddr_is_joined(addr_to_del)) {
+			net_if_ipv6_maddr_leave(ot_context->iface, addr_to_del);
+		}
 	}
 }
 
@@ -479,6 +510,18 @@ static bool openthread_border_router_check_unicast_packet_forwarding_policy(stru
 		return false;
 	}
 
+	/*
+	 * This is the case when a packet from OpenThread stack is sent via UDP platform.
+	 * Packet will be eventually returned to OpenThread interface, but it won't have
+	 * orig_iface set, an indication that the packet was not forwarded from another interface.
+	 * In this case, this function should not check and let the packet be handled by
+	 * 15.4 layer.
+	 */
+
+	if (net_pkt_orig_iface(pkt) != ail_iface_ptr && net_pkt_iface(pkt) == ot_iface_ptr) {
+		return true;
+	}
+
 	/* An IPv6 packet with a link-local source address or a link-local destination address
 	 * is never forwarded.
 	 */
@@ -536,3 +579,29 @@ bool openthread_border_router_check_packet_forwarding_rules(struct net_pkt *pkt)
 
 	return true;
 }
+
+static void openthread_border_router_add_route_to_multicast_groups(void)
+{
+	static uint8_t mcast_group_idx[] = {
+		0x04, /** Admin-Local scope multicast address */
+		0x05, /** Site-Local scope multicast address */
+		0x08, /** Organization-Local scope multicast address */
+		0x0e, /** Global scope multicast address */
+	};
+	struct in6_addr addr = {0};
+	struct net_if_mcast_addr *mcast_addr = NULL;
+	struct net_route_entry_mcast *entry = NULL;
+
+	ARRAY_FOR_EACH(mcast_group_idx, i) {
+
+		net_ipv6_addr_create(&addr, (0xff << 8) | mcast_group_idx[i], 0, 0, 0, 0, 0, 0, 1);
+		entry = net_route_mcast_add(ail_iface_ptr, &addr, NUM_BITS(struct in6_addr));
+		if (entry != NULL) {
+			mcast_addr = net_if_ipv6_maddr_add(ail_iface_ptr,
+							   (const struct in6_addr *)&addr);
+			if (mcast_addr != NULL) {
+				net_if_ipv6_maddr_join(ail_iface_ptr, mcast_addr);
+			}
+		}
+	}
+}