Merge branch 'net-Move-fib_nh-and-fib6_nh-to-a-common-struct'

David Ahern says:

====================
net: Move fib_nh and fib6_nh to a common struct

First set of three with the end goal of enabling IPv6 gateways with IPv4
routes.

This set refactors ipv4 and ipv6 code to create init and release
helpers for each protocol and moving common elements to a fib_nh_common
struct.

v3
- split the reject setting into 2 with helper to the checks. This
  avoids changing cfg->fc_flags in fib6_nh_init

v2
- addressed Ido's comments: cleanup on failure path in nh_init helpers,
  ordering in fib6_nh_release, and removal of RTF_GATEWAY from fib6_info
  uses in mlxsw
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c

@@ -122,7 +122,7 @@ static void mlx5_lag_fib_route_event(struct mlx5_lag *ldev,
 	/* Handle add/replace event */
 	if (fi->fib_nhs == 1) {
 		if (__mlx5_lag_is_active(ldev)) {
-			struct net_device *nh_dev = fi->fib_nh[0].nh_dev;
+			struct net_device *nh_dev = fi->fib_nh[0].fib_nh_dev;
 			int i = mlx5_lag_dev_get_netdev_idx(ldev, nh_dev);
 
 			mlx5_lag_set_port_affinity(ldev, ++i);
@@ -134,10 +134,10 @@ static void mlx5_lag_fib_route_event(struct mlx5_lag *ldev,
 		return;
 
 	/* Verify next hops are ports of the same hca */
-	if (!(fi->fib_nh[0].nh_dev == ldev->pf[0].netdev &&
-	      fi->fib_nh[1].nh_dev == ldev->pf[1].netdev) &&
-	    !(fi->fib_nh[0].nh_dev == ldev->pf[1].netdev &&
-	      fi->fib_nh[1].nh_dev == ldev->pf[0].netdev)) {
+	if (!(fi->fib_nh[0].fib_nh_dev == ldev->pf[0].netdev &&
+	      fi->fib_nh[1].fib_nh_dev == ldev->pf[1].netdev) &&
+	    !(fi->fib_nh[0].fib_nh_dev == ldev->pf[1].netdev &&
+	      fi->fib_nh[1].fib_nh_dev == ldev->pf[0].netdev)) {
 		mlx5_core_warn(ldev->pf[0].dev, "Multipath offload require two ports of the same HCA\n");
 		return;
 	}
@@ -167,7 +167,7 @@ static void mlx5_lag_fib_nexthop_event(struct mlx5_lag *ldev,
 
 	/* nh added/removed */
 	if (event == FIB_EVENT_NH_DEL) {
-		int i = mlx5_lag_dev_get_netdev_idx(ldev, fib_nh->nh_dev);
+		int i = mlx5_lag_dev_get_netdev_idx(ldev, fib_nh->fib_nh_dev);
 
 		if (i >= 0) {
 			i = (i + 1) % 2 + 1; /* peer port */

drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c

@@ -2873,12 +2873,13 @@ mlxsw_sp_nexthop6_group_cmp(const struct mlxsw_sp_nexthop_group *nh_grp,
 		return false;
 
 	list_for_each_entry(mlxsw_sp_rt6, &fib6_entry->rt6_list, list) {
+		struct fib6_nh *fib6_nh = &mlxsw_sp_rt6->rt->fib6_nh;
 		struct in6_addr *gw;
 		int ifindex, weight;
 
-		ifindex = mlxsw_sp_rt6->rt->fib6_nh.nh_dev->ifindex;
-		weight = mlxsw_sp_rt6->rt->fib6_nh.nh_weight;
-		gw = &mlxsw_sp_rt6->rt->fib6_nh.nh_gw;
+		ifindex = fib6_nh->fib_nh_dev->ifindex;
+		weight = fib6_nh->fib_nh_weight;
+		gw = &fib6_nh->fib_nh_gw6;
 		if (!mlxsw_sp_nexthop6_group_has_nexthop(nh_grp, gw, ifindex,
 							 weight))
 			return false;
@@ -2944,7 +2945,7 @@ mlxsw_sp_nexthop6_group_hash(struct mlxsw_sp_fib6_entry *fib6_entry, u32 seed)
 	struct net_device *dev;
 
 	list_for_each_entry(mlxsw_sp_rt6, &fib6_entry->rt6_list, list) {
-		dev = mlxsw_sp_rt6->rt->fib6_nh.nh_dev;
+		dev = mlxsw_sp_rt6->rt->fib6_nh.fib_nh_dev;
 		val ^= dev->ifindex;
 	}
 
@@ -3610,7 +3611,7 @@ static bool mlxsw_sp_nexthop4_ipip_type(const struct mlxsw_sp *mlxsw_sp,
 					const struct fib_nh *fib_nh,
 					enum mlxsw_sp_ipip_type *p_ipipt)
 {
-	struct net_device *dev = fib_nh->nh_dev;
+	struct net_device *dev = fib_nh->fib_nh_dev;
 
 	return dev &&
 	       fib_nh->nh_parent->fib_type == RTN_UNICAST &&
@@ -3637,7 +3638,7 @@ static int mlxsw_sp_nexthop4_type_init(struct mlxsw_sp *mlxsw_sp,
 				       struct fib_nh *fib_nh)
 {
 	const struct mlxsw_sp_ipip_ops *ipip_ops;
-	struct net_device *dev = fib_nh->nh_dev;
+	struct net_device *dev = fib_nh->fib_nh_dev;
 	struct mlxsw_sp_ipip_entry *ipip_entry;
 	struct mlxsw_sp_rif *rif;
 	int err;
@@ -3681,18 +3682,18 @@ static int mlxsw_sp_nexthop4_init(struct mlxsw_sp *mlxsw_sp,
 				  struct mlxsw_sp_nexthop *nh,
 				  struct fib_nh *fib_nh)
 {
-	struct net_device *dev = fib_nh->nh_dev;
+	struct net_device *dev = fib_nh->fib_nh_dev;
 	struct in_device *in_dev;
 	int err;
 
 	nh->nh_grp = nh_grp;
 	nh->key.fib_nh = fib_nh;
 #ifdef CONFIG_IP_ROUTE_MULTIPATH
-	nh->nh_weight = fib_nh->nh_weight;
+	nh->nh_weight = fib_nh->fib_nh_weight;
 #else
 	nh->nh_weight = 1;
 #endif
-	memcpy(&nh->gw_addr, &fib_nh->nh_gw, sizeof(fib_nh->nh_gw));
+	memcpy(&nh->gw_addr, &fib_nh->fib_nh_gw4, sizeof(fib_nh->fib_nh_gw4));
 	err = mlxsw_sp_nexthop_insert(mlxsw_sp, nh);
 	if (err)
 		return err;
@@ -3705,7 +3706,7 @@ static int mlxsw_sp_nexthop4_init(struct mlxsw_sp *mlxsw_sp,
 
 	in_dev = __in_dev_get_rtnl(dev);
 	if (in_dev && IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN(in_dev) &&
-	    fib_nh->nh_flags & RTNH_F_LINKDOWN)
+	    fib_nh->fib_nh_flags & RTNH_F_LINKDOWN)
 		return 0;
 
 	err = mlxsw_sp_nexthop4_type_init(mlxsw_sp, nh, fib_nh);
@@ -3804,7 +3805,7 @@ static void mlxsw_sp_nexthop_rif_gone_sync(struct mlxsw_sp *mlxsw_sp,
 static bool mlxsw_sp_fi_is_gateway(const struct mlxsw_sp *mlxsw_sp,
 				   const struct fib_info *fi)
 {
-	return fi->fib_nh->nh_scope == RT_SCOPE_LINK ||
+	return fi->fib_nh->fib_nh_scope == RT_SCOPE_LINK ||
 	       mlxsw_sp_nexthop4_ipip_type(mlxsw_sp, fi->fib_nh, NULL);
 }
 
@@ -3946,9 +3947,9 @@ mlxsw_sp_rt6_nexthop(struct mlxsw_sp_nexthop_group *nh_grp,
 		struct mlxsw_sp_nexthop *nh = &nh_grp->nexthops[i];
 		struct fib6_info *rt = mlxsw_sp_rt6->rt;
 
-		if (nh->rif && nh->rif->dev == rt->fib6_nh.nh_dev &&
+		if (nh->rif && nh->rif->dev == rt->fib6_nh.fib_nh_dev &&
 		    ipv6_addr_equal((const struct in6_addr *) &nh->gw_addr,
-				    &rt->fib6_nh.nh_gw))
+				    &rt->fib6_nh.fib_nh_gw6))
 			return nh;
 		continue;
 	}
@@ -3966,7 +3967,7 @@ mlxsw_sp_fib4_entry_offload_set(struct mlxsw_sp_fib_entry *fib_entry)
 	    fib_entry->type == MLXSW_SP_FIB_ENTRY_TYPE_BLACKHOLE ||
 	    fib_entry->type == MLXSW_SP_FIB_ENTRY_TYPE_IPIP_DECAP ||
 	    fib_entry->type == MLXSW_SP_FIB_ENTRY_TYPE_NVE_DECAP) {
-		nh_grp->nexthops->key.fib_nh->nh_flags |= RTNH_F_OFFLOAD;
+		nh_grp->nexthops->key.fib_nh->fib_nh_flags |= RTNH_F_OFFLOAD;
 		return;
 	}
 
@@ -3974,9 +3975,9 @@ mlxsw_sp_fib4_entry_offload_set(struct mlxsw_sp_fib_entry *fib_entry)
 		struct mlxsw_sp_nexthop *nh = &nh_grp->nexthops[i];
 
 		if (nh->offloaded)
-			nh->key.fib_nh->nh_flags |= RTNH_F_OFFLOAD;
+			nh->key.fib_nh->fib_nh_flags |= RTNH_F_OFFLOAD;
 		else
-			nh->key.fib_nh->nh_flags &= ~RTNH_F_OFFLOAD;
+			nh->key.fib_nh->fib_nh_flags &= ~RTNH_F_OFFLOAD;
 	}
 }
 
@@ -3992,7 +3993,7 @@ mlxsw_sp_fib4_entry_offload_unset(struct mlxsw_sp_fib_entry *fib_entry)
 	for (i = 0; i < nh_grp->count; i++) {
 		struct mlxsw_sp_nexthop *nh = &nh_grp->nexthops[i];
 
-		nh->key.fib_nh->nh_flags &= ~RTNH_F_OFFLOAD;
+		nh->key.fib_nh->fib_nh_flags &= ~RTNH_F_OFFLOAD;
 	}
 }
 
@@ -4008,19 +4009,20 @@ mlxsw_sp_fib6_entry_offload_set(struct mlxsw_sp_fib_entry *fib_entry)
 	if (fib_entry->type == MLXSW_SP_FIB_ENTRY_TYPE_LOCAL ||
 	    fib_entry->type == MLXSW_SP_FIB_ENTRY_TYPE_BLACKHOLE) {
 		list_first_entry(&fib6_entry->rt6_list, struct mlxsw_sp_rt6,
-				 list)->rt->fib6_nh.nh_flags |= RTNH_F_OFFLOAD;
+				 list)->rt->fib6_nh.fib_nh_flags |= RTNH_F_OFFLOAD;
 		return;
 	}
 
 	list_for_each_entry(mlxsw_sp_rt6, &fib6_entry->rt6_list, list) {
 		struct mlxsw_sp_nexthop_group *nh_grp = fib_entry->nh_group;
+		struct fib6_nh *fib6_nh = &mlxsw_sp_rt6->rt->fib6_nh;
 		struct mlxsw_sp_nexthop *nh;
 
 		nh = mlxsw_sp_rt6_nexthop(nh_grp, mlxsw_sp_rt6);
 		if (nh && nh->offloaded)
-			mlxsw_sp_rt6->rt->fib6_nh.nh_flags |= RTNH_F_OFFLOAD;
+			fib6_nh->fib_nh_flags |= RTNH_F_OFFLOAD;
 		else
-			mlxsw_sp_rt6->rt->fib6_nh.nh_flags &= ~RTNH_F_OFFLOAD;
+			fib6_nh->fib_nh_flags &= ~RTNH_F_OFFLOAD;
 	}
 }
 
@@ -4035,7 +4037,7 @@ mlxsw_sp_fib6_entry_offload_unset(struct mlxsw_sp_fib_entry *fib_entry)
 	list_for_each_entry(mlxsw_sp_rt6, &fib6_entry->rt6_list, list) {
 		struct fib6_info *rt = mlxsw_sp_rt6->rt;
 
-		rt->fib6_nh.nh_flags &= ~RTNH_F_OFFLOAD;
+		rt->fib6_nh.fib_nh_flags &= ~RTNH_F_OFFLOAD;
 	}
 }
 
@@ -4913,7 +4915,7 @@ static void mlxsw_sp_rt6_destroy(struct mlxsw_sp_rt6 *mlxsw_sp_rt6)
 static bool mlxsw_sp_fib6_rt_can_mp(const struct fib6_info *rt)
 {
 	/* RTF_CACHE routes are ignored */
-	return (rt->fib6_flags & (RTF_GATEWAY | RTF_ADDRCONF)) == RTF_GATEWAY;
+	return !(rt->fib6_flags & RTF_ADDRCONF) && rt->fib6_nh.fib_nh_has_gw;
 }
 
 static struct fib6_info *
@@ -4972,8 +4974,8 @@ static bool mlxsw_sp_nexthop6_ipip_type(const struct mlxsw_sp *mlxsw_sp,
 					const struct fib6_info *rt,
 					enum mlxsw_sp_ipip_type *ret)
 {
-	return rt->fib6_nh.nh_dev &&
-	       mlxsw_sp_netdev_ipip_type(mlxsw_sp, rt->fib6_nh.nh_dev, ret);
+	return rt->fib6_nh.fib_nh_dev &&
+	       mlxsw_sp_netdev_ipip_type(mlxsw_sp, rt->fib6_nh.fib_nh_dev, ret);
 }
 
 static int mlxsw_sp_nexthop6_type_init(struct mlxsw_sp *mlxsw_sp,
@@ -4983,7 +4985,7 @@ static int mlxsw_sp_nexthop6_type_init(struct mlxsw_sp *mlxsw_sp,
 {
 	const struct mlxsw_sp_ipip_ops *ipip_ops;
 	struct mlxsw_sp_ipip_entry *ipip_entry;
-	struct net_device *dev = rt->fib6_nh.nh_dev;
+	struct net_device *dev = rt->fib6_nh.fib_nh_dev;
 	struct mlxsw_sp_rif *rif;
 	int err;
 
@@ -5026,11 +5028,11 @@ static int mlxsw_sp_nexthop6_init(struct mlxsw_sp *mlxsw_sp,
 				  struct mlxsw_sp_nexthop *nh,
 				  const struct fib6_info *rt)
 {
-	struct net_device *dev = rt->fib6_nh.nh_dev;
+	struct net_device *dev = rt->fib6_nh.fib_nh_dev;
 
 	nh->nh_grp = nh_grp;
-	nh->nh_weight = rt->fib6_nh.nh_weight;
-	memcpy(&nh->gw_addr, &rt->fib6_nh.nh_gw, sizeof(nh->gw_addr));
+	nh->nh_weight = rt->fib6_nh.fib_nh_weight;
+	memcpy(&nh->gw_addr, &rt->fib6_nh.fib_nh_gw6, sizeof(nh->gw_addr));
 	mlxsw_sp_nexthop_counter_alloc(mlxsw_sp, nh);
 
 	list_add_tail(&nh->router_list_node, &mlxsw_sp->router->nexthop_list);
@@ -5053,7 +5055,7 @@ static void mlxsw_sp_nexthop6_fini(struct mlxsw_sp *mlxsw_sp,
 static bool mlxsw_sp_rt6_is_gateway(const struct mlxsw_sp *mlxsw_sp,
 				    const struct fib6_info *rt)
 {
-	return rt->fib6_flags & RTF_GATEWAY ||
+	return rt->fib6_nh.fib_nh_has_gw ||
 	       mlxsw_sp_nexthop6_ipip_type(mlxsw_sp, rt, NULL);
 }
 

drivers/net/ethernet/rocker/rocker_ofdpa.c

@@ -2288,11 +2288,11 @@ static int ofdpa_port_fib_ipv4(struct ofdpa_port *ofdpa_port,  __be32 dst,
 
 	nh = fi->fib_nh;
 	nh_on_port = (fi->fib_dev == ofdpa_port->dev);
-	has_gw = !!nh->nh_gw;
+	has_gw = !!nh->fib_nh_gw4;
 
 	if (has_gw && nh_on_port) {
 		err = ofdpa_port_ipv4_nh(ofdpa_port, flags,
-					 nh->nh_gw, &index);
+					 nh->fib_nh_gw4, &index);
 		if (err)
 			return err;
 
@@ -2749,7 +2749,7 @@ static int ofdpa_fib4_add(struct rocker *rocker,
 				  fen_info->tb_id, 0);
 	if (err)
 		return err;
-	fen_info->fi->fib_nh->nh_flags |= RTNH_F_OFFLOAD;
+	fen_info->fi->fib_nh->fib_nh_flags |= RTNH_F_OFFLOAD;
 	return 0;
 }
 
@@ -2764,7 +2764,7 @@ static int ofdpa_fib4_del(struct rocker *rocker,
 	ofdpa_port = ofdpa_port_dev_lower_find(fen_info->fi->fib_dev, rocker);
 	if (!ofdpa_port)
 		return 0;
-	fen_info->fi->fib_nh->nh_flags &= ~RTNH_F_OFFLOAD;
+	fen_info->fi->fib_nh->fib_nh_flags &= ~RTNH_F_OFFLOAD;
 	return ofdpa_port_fib_ipv4(ofdpa_port, htonl(fen_info->dst),
 				   fen_info->dst_len, fen_info->fi,
 				   fen_info->tb_id, OFDPA_OP_FLAG_REMOVE);
@@ -2791,7 +2791,7 @@ static void ofdpa_fib4_abort(struct rocker *rocker)
 						       rocker);
 		if (!ofdpa_port)
 			continue;
-		flow_entry->fi->fib_nh->nh_flags &= ~RTNH_F_OFFLOAD;
+		flow_entry->fi->fib_nh->fib_nh_flags &= ~RTNH_F_OFFLOAD;
 		ofdpa_flow_tbl_del(ofdpa_port, OFDPA_OP_FLAG_REMOVE,
 				   flow_entry);
 	}

include/linux/inetdevice.h

@@ -237,6 +237,20 @@ static inline struct in_device *__in_dev_get_rtnl(const struct net_device *dev)
 	return rtnl_dereference(dev->ip_ptr);
 }
 
+/* called with rcu_read_lock or rtnl held */
+static inline bool ip_ignore_linkdown(const struct net_device *dev)
+{
+	struct in_device *in_dev;
+	bool rc = false;
+
+	in_dev = rcu_dereference_rtnl(dev->ip_ptr);
+	if (in_dev &&
+	    IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN(in_dev))
+		rc = true;
+
+	return rc;
+}
+
 static inline struct neigh_parms *__in_dev_arp_parms_get_rcu(const struct net_device *dev)
 {
 	struct in_device *in_dev = __in_dev_get_rcu(dev);

include/net/addrconf.h

@@ -425,6 +425,14 @@ static inline void in6_dev_hold(struct inet6_dev *idev)
 	refcount_inc(&idev->refcnt);
 }
 
+/* called with rcu_read_lock held */
+static inline bool ip6_ignore_linkdown(const struct net_device *dev)
+{
+	const struct inet6_dev *idev = __in6_dev_get(dev);
+
+	return !!idev->cnf.ignore_routes_with_linkdown;
+}
+
 void inet6_ifa_finish_destroy(struct inet6_ifaddr *ifp);
 
 static inline void in6_ifa_put(struct inet6_ifaddr *ifp)

include/net/ip6_fib.h

@@ -19,6 +19,7 @@
 #include <linux/notifier.h>
 #include <net/dst.h>
 #include <net/flow.h>
+#include <net/ip_fib.h>
 #include <net/netlink.h>
 #include <net/inetpeer.h>
 #include <net/fib_notifier.h>
@@ -125,13 +126,7 @@ struct rt6_exception {
 #define FIB6_MAX_DEPTH 5
 
 struct fib6_nh {
-	struct in6_addr		nh_gw;
-	struct net_device	*nh_dev;
-	struct lwtunnel_state	*nh_lwtstate;
-
-	unsigned int		nh_flags;
-	atomic_t		nh_upper_bound;
-	int			nh_weight;
+	struct fib_nh_common	nh_common;
 };
 
 struct fib6_info {
@@ -441,13 +436,18 @@ void rt6_get_prefsrc(const struct rt6_info *rt, struct in6_addr *addr)
 
 static inline struct net_device *fib6_info_nh_dev(const struct fib6_info *f6i)
 {
-	return f6i->fib6_nh.nh_dev;
+	return f6i->fib6_nh.fib_nh_dev;
 }
 
+int fib6_nh_init(struct net *net, struct fib6_nh *fib6_nh,
+		 struct fib6_config *cfg, gfp_t gfp_flags,
+		 struct netlink_ext_ack *extack);
+void fib6_nh_release(struct fib6_nh *fib6_nh);
+
 static inline
 struct lwtunnel_state *fib6_info_nh_lwt(const struct fib6_info *f6i)
 {
-	return f6i->fib6_nh.nh_lwtstate;
+	return f6i->fib6_nh.fib_nh_lws;
 }
 
 void inet6_rt_notify(int event, struct fib6_info *rt, struct nl_info *info,

include/net/ip6_route.h

@@ -68,8 +68,8 @@ static inline bool rt6_need_strict(const struct in6_addr *daddr)
 
 static inline bool rt6_qualify_for_ecmp(const struct fib6_info *f6i)
 {
-	return (f6i->fib6_flags & (RTF_GATEWAY|RTF_ADDRCONF|RTF_DYNAMIC)) ==
-	       RTF_GATEWAY;
+	return !(f6i->fib6_flags & (RTF_ADDRCONF|RTF_DYNAMIC)) &&
+		f6i->fib6_nh.fib_nh_has_gw;
 }
 
 void ip6_route_input(struct sk_buff *skb);
@@ -274,9 +274,11 @@ static inline struct in6_addr *rt6_nexthop(struct rt6_info *rt,
 
 static inline bool rt6_duplicate_nexthop(struct fib6_info *a, struct fib6_info *b)
 {
-	return a->fib6_nh.nh_dev == b->fib6_nh.nh_dev &&
-	       ipv6_addr_equal(&a->fib6_nh.nh_gw, &b->fib6_nh.nh_gw) &&
-	       !lwtunnel_cmp_encap(a->fib6_nh.nh_lwtstate, b->fib6_nh.nh_lwtstate);
+	struct fib6_nh *nha = &a->fib6_nh, *nhb = &b->fib6_nh;
+
+	return nha->fib_nh_dev == nhb->fib_nh_dev &&
+	       ipv6_addr_equal(&nha->fib_nh_gw6, &nhb->fib_nh_gw6) &&
+	       !lwtunnel_cmp_encap(nha->fib_nh_lws, nhb->fib_nh_lws);
 }
 
 static inline unsigned int ip6_dst_mtu_forward(const struct dst_entry *dst)

include/net/ip_fib.h

@@ -76,27 +76,48 @@ struct fnhe_hash_bucket {
 #define FNHE_HASH_SIZE		(1 << FNHE_HASH_SHIFT)
 #define FNHE_RECLAIM_DEPTH	5
 
+struct fib_nh_common {
+	struct net_device	*nhc_dev;
+	int			nhc_oif;
+	unsigned int		nhc_flags;
+	struct lwtunnel_state	*nhc_lwtstate;
+	unsigned char		nhc_scope;
+	u8			nhc_family;
+	u8			nhc_has_gw:1,
+				unused:7;
+	union {
+		__be32          ipv4;
+		struct in6_addr ipv6;
+	} nhc_gw;
+
+	int			nhc_weight;
+	atomic_t		nhc_upper_bound;
+};
+
 struct fib_nh {
-	struct net_device	*nh_dev;
+	struct fib_nh_common	nh_common;
 	struct hlist_node	nh_hash;
 	struct fib_info		*nh_parent;
-	unsigned int		nh_flags;
-	unsigned char		nh_scope;
-#ifdef CONFIG_IP_ROUTE_MULTIPATH
-	int			nh_weight;
-	atomic_t		nh_upper_bound;
-#endif
 #ifdef CONFIG_IP_ROUTE_CLASSID
 	__u32			nh_tclassid;
 #endif
-	int			nh_oif;
-	__be32			nh_gw;
 	__be32			nh_saddr;
 	int			nh_saddr_genid;
 	struct rtable __rcu * __percpu *nh_pcpu_rth_output;
 	struct rtable __rcu	*nh_rth_input;
 	struct fnhe_hash_bucket	__rcu *nh_exceptions;
-	struct lwtunnel_state	*nh_lwtstate;
+#define fib_nh_family		nh_common.nhc_family
+#define fib_nh_dev		nh_common.nhc_dev
+#define fib_nh_oif		nh_common.nhc_oif
+#define fib_nh_flags		nh_common.nhc_flags
+#define fib_nh_lws		nh_common.nhc_lwtstate
+#define fib_nh_scope		nh_common.nhc_scope
+#define fib_nh_family		nh_common.nhc_family
+#define fib_nh_has_gw		nh_common.nhc_has_gw
+#define fib_nh_gw4		nh_common.nhc_gw.ipv4
+#define fib_nh_gw6		nh_common.nhc_gw.ipv6
+#define fib_nh_weight		nh_common.nhc_weight
+#define fib_nh_upper_bound	nh_common.nhc_upper_bound
 };
 
 /*
@@ -125,7 +146,7 @@ struct fib_info {
 	int			fib_nhs;
 	struct rcu_head		rcu;
 	struct fib_nh		fib_nh[0];
-#define fib_dev		fib_nh[0].nh_dev
+#define fib_dev		fib_nh[0].fib_nh_dev
 };
 
 
@@ -180,9 +201,9 @@ __be32 fib_info_update_nh_saddr(struct net *net, struct fib_nh *nh);
 	  atomic_read(&(net)->ipv4.dev_addr_genid)) ?	\
 	 FIB_RES_NH(res).nh_saddr :			\
 	 fib_info_update_nh_saddr((net), &FIB_RES_NH(res)))
-#define FIB_RES_GW(res)			(FIB_RES_NH(res).nh_gw)
-#define FIB_RES_DEV(res)		(FIB_RES_NH(res).nh_dev)
-#define FIB_RES_OIF(res)		(FIB_RES_NH(res).nh_oif)
+#define FIB_RES_GW(res)			(FIB_RES_NH(res).fib_nh_gw4)
+#define FIB_RES_DEV(res)		(FIB_RES_NH(res).fib_nh_dev)
+#define FIB_RES_OIF(res)		(FIB_RES_NH(res).fib_nh_oif)
 
 #define FIB_RES_PREFSRC(net, res)	((res).fi->fib_prefsrc ? : \
 					 FIB_RES_SADDR(net, res))
@@ -416,6 +437,15 @@ void fib_select_multipath(struct fib_result *res, int hash);
 void fib_select_path(struct net *net, struct fib_result *res,
 		     struct flowi4 *fl4, const struct sk_buff *skb);
 
+int fib_nh_init(struct net *net, struct fib_nh *fib_nh,
+		struct fib_config *cfg, int nh_weight,
+		struct netlink_ext_ack *extack);
+void fib_nh_release(struct net *net, struct fib_nh *fib_nh);
+int fib_nh_common_init(struct fib_nh_common *nhc, struct nlattr *fc_encap,
+		       u16 fc_encap_type, void *cfg, gfp_t gfp_flags,
+		       struct netlink_ext_ack *extack);
+void fib_nh_common_release(struct fib_nh_common *nhc);
+
 /* Exported by fib_trie.c */
 void fib_trie_init(void);
 struct fib_table *fib_trie_table(u32 id, struct fib_table *alias);

include/trace/events/fib.h

@@ -63,13 +63,16 @@ TRACE_EVENT(fib_table_lookup,
 		}
 
 		if (nh) {
+			struct net_device *dev;
+
 			p32 = (__be32 *) __entry->saddr;
 			*p32 = nh->nh_saddr;
 
 			p32 = (__be32 *) __entry->gw;
-			*p32 = nh->nh_gw;
+			*p32 = nh->fib_nh_gw4;
 
-			__assign_str(name, nh->nh_dev ? nh->nh_dev->name : "-");
+			dev = nh->fib_nh_dev;
+			__assign_str(name, dev ? dev->name : "-");
 		} else {
 			p32 = (__be32 *) __entry->saddr;
 			*p32 = 0;

include/trace/events/fib6.h

@@ -62,8 +62,8 @@ TRACE_EVENT(fib6_table_lookup,
 			__entry->dport = 0;
 		}
 
-		if (f6i->fib6_nh.nh_dev) {
-			__assign_str(name, f6i->fib6_nh.nh_dev);
+		if (f6i->fib6_nh.fib_nh_dev) {
+			__assign_str(name, f6i->fib6_nh.fib_nh_dev);
 		} else {
 			__assign_str(name, "-");
 		}
@@ -75,7 +75,7 @@ TRACE_EVENT(fib6_table_lookup,
 
 		} else if (f6i) {
 			in6 = (struct in6_addr *)__entry->gw;
-			*in6 = f6i->fib6_nh.nh_gw;
+			*in6 = f6i->fib6_nh.fib_nh_gw6;
 		}
 	),
 

net/core/filter.c

@@ -4634,12 +4634,12 @@ static int bpf_ipv4_fib_lookup(struct net *net, struct bpf_fib_lookup *params,
 	nh = &res.fi->fib_nh[res.nh_sel];
 
 	/* do not handle lwt encaps right now */
-	if (nh->nh_lwtstate)
+	if (nh->fib_nh_lws)
 		return BPF_FIB_LKUP_RET_UNSUPP_LWT;
 
-	dev = nh->nh_dev;
-	if (nh->nh_gw)
-		params->ipv4_dst = nh->nh_gw;
+	dev = nh->fib_nh_dev;
+	if (nh->fib_nh_gw4)
+		params->ipv4_dst = nh->fib_nh_gw4;
 
 	params->rt_metric = res.fi->fib_priority;
 
@@ -4748,13 +4748,13 @@ static int bpf_ipv6_fib_lookup(struct net *net, struct bpf_fib_lookup *params,
 			return BPF_FIB_LKUP_RET_FRAG_NEEDED;
 	}
 
-	if (f6i->fib6_nh.nh_lwtstate)
+	if (f6i->fib6_nh.fib_nh_lws)
 		return BPF_FIB_LKUP_RET_UNSUPP_LWT;
 
-	if (f6i->fib6_flags & RTF_GATEWAY)
-		*dst = f6i->fib6_nh.nh_gw;
+	if (f6i->fib6_nh.fib_nh_has_gw)
+		*dst = f6i->fib6_nh.fib_nh_gw6;
 
-	dev = f6i->fib6_nh.nh_dev;
+	dev = f6i->fib6_nh.fib_nh_dev;
 	params->rt_metric = f6i->fib6_metric;
 
 	/* xdp and cls_bpf programs are run in RCU-bh so rcu_read_lock_bh is

net/ipv4/fib_frontend.c

@@ -324,16 +324,16 @@ bool fib_info_nh_uses_dev(struct fib_info *fi, const struct net_device *dev)
 	for (ret = 0; ret < fi->fib_nhs; ret++) {
 		struct fib_nh *nh = &fi->fib_nh[ret];
 
-		if (nh->nh_dev == dev) {
+		if (nh->fib_nh_dev == dev) {
 			dev_match = true;
 			break;
-		} else if (l3mdev_master_ifindex_rcu(nh->nh_dev) == dev->ifindex) {
+		} else if (l3mdev_master_ifindex_rcu(nh->fib_nh_dev) == dev->ifindex) {
 			dev_match = true;
 			break;
 		}
 	}
 #else
-	if (fi->fib_nh[0].nh_dev == dev)
+	if (fi->fib_nh[0].fib_nh_dev == dev)
 		dev_match = true;
 #endif
 
@@ -390,7 +390,7 @@ static int __fib_validate_source(struct sk_buff *skb, __be32 src, __be32 dst,
 
 	dev_match = fib_info_nh_uses_dev(res.fi, dev);
 	if (dev_match) {
-		ret = FIB_RES_NH(res).nh_scope >= RT_SCOPE_HOST;
+		ret = FIB_RES_NH(res).fib_nh_scope >= RT_SCOPE_HOST;
 		return ret;
 	}
 	if (no_addr)
@@ -402,7 +402,7 @@ static int __fib_validate_source(struct sk_buff *skb, __be32 src, __be32 dst,
 	ret = 0;
 	if (fib_lookup(net, &fl4, &res, FIB_LOOKUP_IGNORE_LINKSTATE) == 0) {
 		if (res.type == RTN_UNICAST)
-			ret = FIB_RES_NH(res).nh_scope >= RT_SCOPE_HOST;
+			ret = FIB_RES_NH(res).fib_nh_scope >= RT_SCOPE_HOST;
 	}
 	return ret;
 

net/ipv4/fib_trie.c

@@ -1471,18 +1471,16 @@ int fib_table_lookup(struct fib_table *tb, const struct flowi4 *flp,
 			continue;
 		for (nhsel = 0; nhsel < fi->fib_nhs; nhsel++) {
 			const struct fib_nh *nh = &fi->fib_nh[nhsel];
-			struct in_device *in_dev = __in_dev_get_rcu(nh->nh_dev);
 
-			if (nh->nh_flags & RTNH_F_DEAD)
+			if (nh->fib_nh_flags & RTNH_F_DEAD)
 				continue;
-			if (in_dev &&
-			    IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN(in_dev) &&
-			    nh->nh_flags & RTNH_F_LINKDOWN &&
+			if (ip_ignore_linkdown(nh->fib_nh_dev) &&
+			    nh->fib_nh_flags & RTNH_F_LINKDOWN &&
 			    !(fib_flags & FIB_LOOKUP_IGNORE_LINKSTATE))
 				continue;
 			if (!(flp->flowi4_flags & FLOWI_FLAG_SKIP_NH_OIF)) {
 				if (flp->flowi4_oif &&
-				    flp->flowi4_oif != nh->nh_oif)
+				    flp->flowi4_oif != nh->fib_nh_oif)
 					continue;
 			}
 
@@ -2653,7 +2651,7 @@ static unsigned int fib_flag_trans(int type, __be32 mask, const struct fib_info
 
 	if (type == RTN_UNREACHABLE || type == RTN_PROHIBIT)
 		flags = RTF_REJECT;
-	if (fi && fi->fib_nh->nh_gw)
+	if (fi && fi->fib_nh->fib_nh_gw4)
 		flags |= RTF_GATEWAY;
 	if (mask == htonl(0xFFFFFFFF))
 		flags |= RTF_HOST;
@@ -2704,7 +2702,7 @@ static int fib_route_seq_show(struct seq_file *seq, void *v)
 				   "%d\t%08X\t%d\t%u\t%u",
 				   fi->fib_dev ? fi->fib_dev->name : "*",
 				   prefix,
-				   fi->fib_nh->nh_gw, flags, 0, 0,
+				   fi->fib_nh->fib_nh_gw4, flags, 0, 0,
 				   fi->fib_priority,
 				   mask,
 				   (fi->fib_advmss ?

net/ipv4/route.c

@@ -644,7 +644,7 @@ static void update_or_create_fnhe(struct fib_nh *nh, __be32 daddr, __be32 gw,
 	unsigned int i;
 	int depth;
 
-	genid = fnhe_genid(dev_net(nh->nh_dev));
+	genid = fnhe_genid(dev_net(nh->fib_nh_dev));
 	hval = fnhe_hashfun(daddr);
 
 	spin_lock_bh(&fnhe_lock);
@@ -1356,7 +1356,7 @@ u32 ip_mtu_from_fib_result(struct fib_result *res, __be32 daddr)
 {
 	struct fib_info *fi = res->fi;
 	struct fib_nh *nh = &fi->fib_nh[res->nh_sel];
-	struct net_device *dev = nh->nh_dev;
+	struct net_device *dev = nh->fib_nh_dev;
 	u32 mtu = 0;
 
 	if (dev_net(dev)->ipv4.sysctl_ip_fwd_use_pmtu ||
@@ -1374,7 +1374,7 @@ u32 ip_mtu_from_fib_result(struct fib_result *res, __be32 daddr)
 	if (likely(!mtu))
 		mtu = min(READ_ONCE(dev->mtu), IP_MAX_MTU);
 
-	return mtu - lwtunnel_headroom(nh->nh_lwtstate, mtu);
+	return mtu - lwtunnel_headroom(nh->fib_nh_lws, mtu);
 }
 
 static bool rt_bind_exception(struct rtable *rt, struct fib_nh_exception *fnhe,
@@ -1531,8 +1531,8 @@ static void rt_set_nexthop(struct rtable *rt, __be32 daddr,
 	if (fi) {
 		struct fib_nh *nh = &FIB_RES_NH(*res);
 
-		if (nh->nh_gw && nh->nh_scope == RT_SCOPE_LINK) {
-			rt->rt_gateway = nh->nh_gw;
+		if (nh->fib_nh_gw4 && nh->fib_nh_scope == RT_SCOPE_LINK) {
+			rt->rt_gateway = nh->fib_nh_gw4;
 			rt->rt_uses_gateway = 1;
 		}
 		ip_dst_init_metrics(&rt->dst, fi->fib_metrics);
@@ -1540,7 +1540,7 @@ static void rt_set_nexthop(struct rtable *rt, __be32 daddr,
 #ifdef CONFIG_IP_ROUTE_CLASSID
 		rt->dst.tclassid = nh->nh_tclassid;
 #endif
-		rt->dst.lwtstate = lwtstate_get(nh->nh_lwtstate);
+		rt->dst.lwtstate = lwtstate_get(nh->fib_nh_lws);
 		if (unlikely(fnhe))
 			cached = rt_bind_exception(rt, fnhe, daddr, do_cache);
 		else if (do_cache)
@@ -2075,7 +2075,7 @@ out:	return err;
 	if (do_cache) {
 		struct fib_nh *nh = &FIB_RES_NH(*res);
 
-		rth->dst.lwtstate = lwtstate_get(nh->nh_lwtstate);
+		rth->dst.lwtstate = lwtstate_get(nh->fib_nh_lws);
 		if (lwtunnel_input_redirect(rth->dst.lwtstate)) {
 			WARN_ON(rth->dst.input == lwtunnel_input);
 			rth->dst.lwtstate->orig_input = rth->dst.input;
@@ -2264,8 +2264,8 @@ static struct rtable *__mkroute_output(const struct fib_result *res,
 		} else {
 			if (unlikely(fl4->flowi4_flags &
 				     FLOWI_FLAG_KNOWN_NH &&
-				     !(nh->nh_gw &&
-				       nh->nh_scope == RT_SCOPE_LINK))) {
+				     !(nh->fib_nh_gw4 &&
+				       nh->fib_nh_scope == RT_SCOPE_LINK))) {
 				do_cache = false;
 				goto add;
 			}

net/ipv6/addrconf.c

@@ -173,7 +173,8 @@ static int addrconf_ifdown(struct net_device *dev, int how);
 static struct fib6_info *addrconf_get_prefix_route(const struct in6_addr *pfx,
 						  int plen,
 						  const struct net_device *dev,
-						  u32 flags, u32 noflags);
+						  u32 flags, u32 noflags,
+						  bool no_gw);
 
 static void addrconf_dad_start(struct inet6_ifaddr *ifp);
 static void addrconf_dad_work(struct work_struct *w);
@@ -1230,10 +1231,8 @@ cleanup_prefix_route(struct inet6_ifaddr *ifp, unsigned long expires, bool del_r
 {
 	struct fib6_info *f6i;
 
-	f6i = addrconf_get_prefix_route(&ifp->addr,
-				       ifp->prefix_len,
-				       ifp->idev->dev,
-				       0, RTF_GATEWAY | RTF_DEFAULT);
+	f6i = addrconf_get_prefix_route(&ifp->addr, ifp->prefix_len,
+					ifp->idev->dev, 0, RTF_DEFAULT, true);
 	if (f6i) {
 		if (del_rt)
 			ip6_del_rt(dev_net(ifp->idev->dev), f6i);
@@ -2402,7 +2401,8 @@ addrconf_prefix_route(struct in6_addr *pfx, int plen, u32 metric,
 static struct fib6_info *addrconf_get_prefix_route(const struct in6_addr *pfx,
 						  int plen,
 						  const struct net_device *dev,
-						  u32 flags, u32 noflags)
+						  u32 flags, u32 noflags,
+						  bool no_gw)
 {
 	struct fib6_node *fn;
 	struct fib6_info *rt = NULL;
@@ -2419,7 +2419,9 @@ static struct fib6_info *addrconf_get_prefix_route(const struct in6_addr *pfx,
 		goto out;
 
 	for_each_fib6_node_rt_rcu(fn) {
-		if (rt->fib6_nh.nh_dev->ifindex != dev->ifindex)
+		if (rt->fib6_nh.fib_nh_dev->ifindex != dev->ifindex)
+			continue;
+		if (no_gw && rt->fib6_nh.fib_nh_has_gw)
 			continue;
 		if ((rt->fib6_flags & flags) != flags)
 			continue;
@@ -2717,7 +2719,7 @@ void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len, bool sllao)
 					       pinfo->prefix_len,
 					       dev,
 					       RTF_ADDRCONF | RTF_PREFIX_RT,
-					       RTF_GATEWAY | RTF_DEFAULT);
+					       RTF_DEFAULT, true);
 
 		if (rt) {
 			/* Autoconf prefix route */
@@ -4588,10 +4590,8 @@ static int modify_prefix_route(struct inet6_ifaddr *ifp,
 	struct fib6_info *f6i;
 	u32 prio;
 
-	f6i = addrconf_get_prefix_route(&ifp->addr,
-					ifp->prefix_len,
-					ifp->idev->dev,
-					0, RTF_GATEWAY | RTF_DEFAULT);
+	f6i = addrconf_get_prefix_route(&ifp->addr, ifp->prefix_len,
+					ifp->idev->dev, 0, RTF_DEFAULT, true);
 	if (!f6i)
 		return -ENOENT;
 
@@ -5972,7 +5972,8 @@ static void __ipv6_ifa_notify(int event, struct inet6_ifaddr *ifp)
 			struct fib6_info *rt;
 
 			rt = addrconf_get_prefix_route(&ifp->peer_addr, 128,
-						       ifp->idev->dev, 0, 0);
+						       ifp->idev->dev, 0, 0,
+						       false);
 			if (rt)
 				ip6_del_rt(net, rt);
 		}

net/ipv6/ip6_fib.c

@@ -199,10 +199,7 @@ void fib6_info_destroy_rcu(struct rcu_head *head)
 		free_percpu(f6i->rt6i_pcpu);
 	}
 
-	lwtstate_put(f6i->fib6_nh.nh_lwtstate);
-
-	if (f6i->fib6_nh.nh_dev)
-		dev_put(f6i->fib6_nh.nh_dev);
+	fib6_nh_release(&f6i->fib6_nh);
 
 	ip_fib_metrics_put(f6i->fib6_metrics);
 
@@ -2297,6 +2294,7 @@ static int ipv6_route_seq_show(struct seq_file *seq, void *v)
 {
 	struct fib6_info *rt = v;
 	struct ipv6_route_iter *iter = seq->private;
+	unsigned int flags = rt->fib6_flags;
 	const struct net_device *dev;
 
 	seq_printf(seq, "%pi6 %02x ", &rt->fib6_dst.addr, rt->fib6_dst.plen);
@@ -2306,15 +2304,17 @@ static int ipv6_route_seq_show(struct seq_file *seq, void *v)
 #else
 	seq_puts(seq, "00000000000000000000000000000000 00 ");
 #endif
-	if (rt->fib6_flags & RTF_GATEWAY)
-		seq_printf(seq, "%pi6", &rt->fib6_nh.nh_gw);
-	else
+	if (rt->fib6_nh.fib_nh_has_gw) {
+		flags |= RTF_GATEWAY;
+		seq_printf(seq, "%pi6", &rt->fib6_nh.fib_nh_gw6);
+	} else {
 		seq_puts(seq, "00000000000000000000000000000000");
+	}
 
-	dev = rt->fib6_nh.nh_dev;
+	dev = rt->fib6_nh.fib_nh_dev;
 	seq_printf(seq, " %08x %08x %08x %08x %8s\n",
 		   rt->fib6_metric, atomic_read(&rt->fib6_ref), 0,
-		   rt->fib6_flags, dev ? dev->name : "");
+		   flags, dev ? dev->name : "");
 	iter->w.leaf = NULL;
 	return 0;
 }

net/ipv6/ndisc.c

@@ -1276,8 +1276,8 @@ static void ndisc_router_discovery(struct sk_buff *skb)
 	rt = rt6_get_dflt_router(net, &ipv6_hdr(skb)->saddr, skb->dev);
 
 	if (rt) {
-		neigh = ip6_neigh_lookup(&rt->fib6_nh.nh_gw,
-					 rt->fib6_nh.nh_dev, NULL,
+		neigh = ip6_neigh_lookup(&rt->fib6_nh.fib_nh_gw6,
+					 rt->fib6_nh.fib_nh_dev, NULL,
 					  &ipv6_hdr(skb)->saddr);
 		if (!neigh) {
 			ND_PRINTK(0, err,
@@ -1306,8 +1306,8 @@ static void ndisc_router_discovery(struct sk_buff *skb)
 			return;
 		}
 
-		neigh = ip6_neigh_lookup(&rt->fib6_nh.nh_gw,
-					 rt->fib6_nh.nh_dev, NULL,
+		neigh = ip6_neigh_lookup(&rt->fib6_nh.fib_nh_gw6,
+					 rt->fib6_nh.fib_nh_dev, NULL,
 					  &ipv6_hdr(skb)->saddr);
 		if (!neigh) {
 			ND_PRINTK(0, err,