Merge tag 'net-next-6.17' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next

Pull networking updates from Jakub Kicinski:
 "Core & protocols:

   - Wrap datapath globals into net_aligned_data, to avoid false sharing

   - Preserve MSG_ZEROCOPY in forwarding (e.g. out of a container)

   - Add SO_INQ and SCM_INQ support to AF_UNIX

   - Add SIOCINQ support to AF_VSOCK

   - Add TCP_MAXSEG sockopt to MPTCP

   - Add IPv6 force_forwarding sysctl to enable forwarding per interface

   - Make TCP validation of whether packet fully fits in the receive
     window and the rcv_buf more strict. With increased use of HW
     aggregation a single "packet" can be multiple 100s of kB

   - Add MSG_MORE flag to optimize large TCP transmissions via sockmap,
     improves latency up to 33% for sockmap users

   - Convert TCP send queue handling from tasklet to BH workque

   - Improve BPF iteration over TCP sockets to see each socket exactly
     once

   - Remove obsolete and unused TCP RFC3517/RFC6675 loss recovery code

   - Support enabling kernel threads for NAPI processing on per-NAPI
     instance basis rather than a whole device. Fully stop the kernel
     NAPI thread when threaded NAPI gets disabled. Previously thread
     would stick around until ifdown due to tricky synchronization

   - Allow multicast routing to take effect on locally-generated packets

   - Add output interface argument for End.X in segment routing

   - MCTP: add support for gateway routing, improve bind() handling

   - Don't require rtnl_lock when fetching an IPv6 neighbor over Netlink

   - Add a new neighbor flag ("extern_valid"), which cedes refresh
     responsibilities to userspace. This is needed for EVPN multi-homing
     where a neighbor entry for a multi-homed host needs to be synced
     across all the VTEPs among which the host is multi-homed

   - Support NUD_PERMANENT for proxy neighbor entries

   - Add a new queuing discipline for IETF RFC9332 DualQ Coupled AQM

   - Add sequence numbers to netconsole messages. Unregister
     netconsole's console when all net targets are removed. Code
     refactoring. Add a number of selftests

   - Align IPSec inbound SA lookup to RFC 4301. Only SPI and protocol
     should be used for an inbound SA lookup

   - Support inspecting ref_tracker state via DebugFS

   - Don't force bonding advertisement frames tx to ~333 ms boundaries.
     Add broadcast_neighbor option to send ARP/ND on all bonded links

   - Allow providing upcall pid for the 'execute' command in openvswitch

   - Remove DCCP support from Netfilter's conntrack

   - Disallow multiple packet duplications in the queuing layer

   - Prevent use of deprecated iptables code on PREEMPT_RT

  Driver API:

   - Support RSS and hashing configuration over ethtool Netlink

   - Add dedicated ethtool callbacks for getting and setting hashing
     fields

   - Add support for power budget evaluation strategy in PSE /
     Power-over-Ethernet. Generate Netlink events for overcurrent etc

   - Support DPLL phase offset monitoring across all device inputs.
     Support providing clock reference and SYNC over separate DPLL
     inputs

   - Support traffic classes in devlink rate API for bandwidth
     management

   - Remove rtnl_lock dependency from UDP tunnel port configuration

  Device drivers:

   - Add a new Broadcom driver for 800G Ethernet (bnge)

   - Add a standalone driver for Microchip ZL3073x DPLL

   - Remove IBM's NETIUCV device driver

   - Ethernet high-speed NICs:
      - Broadcom (bnxt):
         - support zero-copy Tx of DMABUF memory
         - take page size into account for page pool recycling rings
      - Intel (100G, ice, idpf):
         - idpf: XDP and AF_XDP support preparations
         - idpf: add flow steering
         - add link_down_events statistic
         - clean up the TSPLL code
         - preparations for live VM migration
      - nVidia/Mellanox:
         - support zero-copy Rx/Tx interfaces (DMABUF and io_uring)
         - optimize context memory usage for matchers
         - expose serial numbers in devlink info
         - support PCIe congestion metrics
      - Meta (fbnic):
         - add 25G, 50G, and 100G link modes to phylink
         - support dumping FW logs
      - Marvell/Cavium:
         - support for CN20K generation of the Octeon chips
      - Amazon:
         - add HW clock (without timestamping, just hypervisor time access)

   - Ethernet virtual:
      - VirtIO net:
         - support segmentation of UDP-tunnel-encapsulated packets
      - Google (gve):
         - support packet timestamping and clock synchronization
      - Microsoft vNIC:
         - add handler for device-originated servicing events
         - allow dynamic MSI-X vector allocation
         - support Tx bandwidth clamping

   - Ethernet NICs consumer, and embedded:
      - AMD:
         - amd-xgbe: hardware timestamping and PTP clock support
      - Broadcom integrated MACs (bcmgenet, bcmasp):
         - use napi_complete_done() return value to support NAPI polling
         - add support for re-starting auto-negotiation
      - Broadcom switches (b53):
         - support BCM5325 switches
         - add bcm63xx EPHY power control
      - Synopsys (stmmac):
         - lots of code refactoring and cleanups
      - TI:
         - icssg-prueth: read firmware-names from device tree
         - icssg: PRP offload support
      - Microchip:
         - lan78xx: convert to PHYLINK for improved PHY and MAC management
         - ksz: add KSZ8463 switch support
      - Intel:
         - support similar queue priority scheme in multi-queue and
           time-sensitive networking (taprio)
         - support packet pre-emption in both
      - RealTek (r8169):
         - enable EEE at 5Gbps on RTL8126
      - Airoha:
         - add PPPoE offload support
         - MDIO bus controller for Airoha AN7583

   - Ethernet PHYs:
      - support for the IPQ5018 internal GE PHY
      - micrel KSZ9477 switch-integrated PHYs:
         - add MDI/MDI-X control support
         - add RX error counters
         - add cable test support
         - add Signal Quality Indicator (SQI) reporting
      - dp83tg720: improve reset handling and reduce link recovery time
      - support bcm54811 (and its MII-Lite interface type)
      - air_en8811h: support resume/suspend
      - support PHY counters for QCA807x and QCA808x
      - support WoL for QCA807x

   - CAN drivers:
      - rcar_canfd: support for Transceiver Delay Compensation
      - kvaser: report FW versions via devlink dev info

   - WiFi:
      - extended regulatory info support (6 GHz)
      - add statistics and beacon monitor for Multi-Link Operation (MLO)
      - support S1G aggregation, improve S1G support
      - add Radio Measurement action fields
      - support per-radio RTS threshold
      - some work around how FIPS affects wifi, which was wrong (RC4 is
        used by TKIP, not only WEP)
      - improvements for unsolicited probe response handling

   - WiFi drivers:
      - RealTek (rtw88):
         - IBSS mode for SDIO devices
      - RealTek (rtw89):
         - BT coexistence for MLO/WiFi7
         - concurrent station + P2P support
         - support for USB devices RTL8851BU/RTL8852BU
      - Intel (iwlwifi):
         - use embedded PNVM in (to be released) FW images to fix
           compatibility issues
         - many cleanups (unused FW APIs, PCIe code, WoWLAN)
         - some FIPS interoperability
      - MediaTek (mt76):
         - firmware recovery improvements
         - more MLO work
      - Qualcomm/Atheros (ath12k):
         - fix scan on multi-radio devices
         - more EHT/Wi-Fi 7 features
         - encapsulation/decapsulation offload
      - Broadcom (brcm80211):
         - support SDIO 43751 device

   - Bluetooth:
      - hci_event: add support for handling LE BIG Sync Lost event
      - ISO: add socket option to report packet seqnum via CMSG
      - ISO: support SCM_TIMESTAMPING for ISO TS

   - Bluetooth drivers:
      - intel_pcie: support Function Level Reset
      - nxpuart: add support for 4M baudrate
      - nxpuart: implement powerup sequence, reset, FW dump, and FW loading"

* tag 'net-next-6.17' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (1742 commits)
  dpll: zl3073x: Fix build failure
  selftests: bpf: fix legacy netfilter options
  ipv6: annotate data-races around rt->fib6_nsiblings
  ipv6: fix possible infinite loop in fib6_info_uses_dev()
  ipv6: prevent infinite loop in rt6_nlmsg_size()
  ipv6: add a retry logic in net6_rt_notify()
  vrf: Drop existing dst reference in vrf_ip6_input_dst
  net/sched: taprio: align entry index attr validation with mqprio
  net: fsl_pq_mdio: use dev_err_probe
  selftests: rtnetlink.sh: remove esp4_offload after test
  vsock: remove unnecessary null check in vsock_getname()
  igb: xsk: solve negative overflow of nb_pkts in zerocopy mode
  stmmac: xsk: fix negative overflow of budget in zerocopy mode
  dt-bindings: ieee802154: Convert at86rf230.txt yaml format
  net: dsa: microchip: Disable PTP function of KSZ8463
  net: dsa: microchip: Setup fiber ports for KSZ8463
  net: dsa: microchip: Write switch MAC address differently for KSZ8463
  net: dsa: microchip: Use different registers for KSZ8463
  net: dsa: microchip: Add KSZ8463 switch support to KSZ DSA driver
  dt-bindings: net: dsa: microchip: Add KSZ8463 switch support
  ...

1 files changed, 450 insertions, 203 deletions

diff --git a/net/mctp/route.c b/net/mctp/route.c
index d9c8e5a5f9ce..2b2b958ef6a3 100644
--- a/net/mctp/route.c
+++ b/net/mctp/route.c
@@ -17,6 +17,8 @@
 #include <linux/rtnetlink.h>
 #include <linux/skbuff.h>
 
+#include <kunit/static_stub.h>
+
 #include <uapi/linux/if_arp.h>
 
 #include <net/mctp.h>
@@ -32,39 +34,42 @@ static const unsigned long mctp_key_lifetime = 6 * CONFIG_HZ;
 static void mctp_flow_prepare_output(struct sk_buff *skb, struct mctp_dev *dev);
 
 /* route output callbacks */
-static int mctp_route_discard(struct mctp_route *route, struct sk_buff *skb)
+static int mctp_dst_discard(struct mctp_dst *dst, struct sk_buff *skb)
 {
 	kfree_skb(skb);
 	return 0;
 }
 
-static struct mctp_sock *mctp_lookup_bind(struct net *net, struct sk_buff *skb)
+static struct mctp_sock *mctp_lookup_bind_details(struct net *net,
+						  struct sk_buff *skb,
+						  u8 type, u8 dest,
+						  u8 src, bool allow_net_any)
 {
 	struct mctp_skb_cb *cb = mctp_cb(skb);
-	struct mctp_hdr *mh;
 	struct sock *sk;
-	u8 type;
+	u8 hash;
 
-	WARN_ON(!rcu_read_lock_held());
+	WARN_ON_ONCE(!rcu_read_lock_held());
 
-	/* TODO: look up in skb->cb? */
-	mh = mctp_hdr(skb);
-
-	if (!skb_headlen(skb))
-		return NULL;
+	hash = mctp_bind_hash(type, dest, src);
 
-	type = (*(u8 *)skb->data) & 0x7f;
-
-	sk_for_each_rcu(sk, &net->mctp.binds) {
+	sk_for_each_rcu(sk, &net->mctp.binds[hash]) {
 		struct mctp_sock *msk = container_of(sk, struct mctp_sock, sk);
 
+		if (!allow_net_any && msk->bind_net == MCTP_NET_ANY)
+			continue;
+
 		if (msk->bind_net != MCTP_NET_ANY && msk->bind_net != cb->net)
 			continue;
 
 		if (msk->bind_type != type)
 			continue;
 
-		if (!mctp_address_matches(msk->bind_addr, mh->dest))
+		if (msk->bind_peer_set &&
+		    !mctp_address_matches(msk->bind_peer_addr, src))
+			continue;
+
+		if (!mctp_address_matches(msk->bind_local_addr, dest))
 			continue;
 
 		return msk;
@@ -73,6 +78,54 @@ static struct mctp_sock *mctp_lookup_bind(struct net *net, struct sk_buff *skb)
 	return NULL;
 }
 
+static struct mctp_sock *mctp_lookup_bind(struct net *net, struct sk_buff *skb)
+{
+	struct mctp_sock *msk;
+	struct mctp_hdr *mh;
+	u8 type;
+
+	/* TODO: look up in skb->cb? */
+	mh = mctp_hdr(skb);
+
+	if (!skb_headlen(skb))
+		return NULL;
+
+	type = (*(u8 *)skb->data) & 0x7f;
+
+	/* Look for binds in order of widening scope. A given destination or
+	 * source address also implies matching on a particular network.
+	 *
+	 * - Matching destination and source
+	 * - Matching destination
+	 * - Matching source
+	 * - Matching network, any address
+	 * - Any network or address
+	 */
+
+	msk = mctp_lookup_bind_details(net, skb, type, mh->dest, mh->src,
+				       false);
+	if (msk)
+		return msk;
+	msk = mctp_lookup_bind_details(net, skb, type, MCTP_ADDR_ANY, mh->src,
+				       false);
+	if (msk)
+		return msk;
+	msk = mctp_lookup_bind_details(net, skb, type, mh->dest, MCTP_ADDR_ANY,
+				       false);
+	if (msk)
+		return msk;
+	msk = mctp_lookup_bind_details(net, skb, type, MCTP_ADDR_ANY,
+				       MCTP_ADDR_ANY, false);
+	if (msk)
+		return msk;
+	msk = mctp_lookup_bind_details(net, skb, type, MCTP_ADDR_ANY,
+				       MCTP_ADDR_ANY, true);
+	if (msk)
+		return msk;
+
+	return NULL;
+}
+
 /* A note on the key allocations.
  *
  * struct net->mctp.keys contains our set of currently-allocated keys for
@@ -368,7 +421,7 @@ static int mctp_frag_queue(struct mctp_sk_key *key, struct sk_buff *skb)
 	return 0;
 }
 
-static int mctp_route_input(struct mctp_route *route, struct sk_buff *skb)
+static int mctp_dst_input(struct mctp_dst *dst, struct sk_buff *skb)
 {
 	struct mctp_sk_key *key, *any_key = NULL;
 	struct net *net = dev_net(skb->dev);
@@ -392,6 +445,9 @@ static int mctp_route_input(struct mctp_route *route, struct sk_buff *skb)
 	 */
 	skb_orphan(skb);
 
+	if (skb->pkt_type == PACKET_OUTGOING)
+		skb->pkt_type = PACKET_LOOPBACK;
+
 	/* ensure we have enough data for a header and a type */
 	if (skb->len < sizeof(struct mctp_hdr) + 1)
 		goto out;
@@ -556,39 +612,31 @@ out:
 	return rc;
 }
 
-static unsigned int mctp_route_mtu(struct mctp_route *rt)
-{
-	return rt->mtu ?: READ_ONCE(rt->dev->dev->mtu);
-}
-
-static int mctp_route_output(struct mctp_route *route, struct sk_buff *skb)
+static int mctp_dst_output(struct mctp_dst *dst, struct sk_buff *skb)
 {
-	struct mctp_skb_cb *cb = mctp_cb(skb);
-	struct mctp_hdr *hdr = mctp_hdr(skb);
 	char daddr_buf[MAX_ADDR_LEN];
 	char *daddr = NULL;
-	unsigned int mtu;
 	int rc;
 
 	skb->protocol = htons(ETH_P_MCTP);
+	skb->pkt_type = PACKET_OUTGOING;
 
-	mtu = READ_ONCE(skb->dev->mtu);
-	if (skb->len > mtu) {
+	if (skb->len > dst->mtu) {
 		kfree_skb(skb);
 		return -EMSGSIZE;
 	}
 
-	if (cb->ifindex) {
-		/* direct route; use the hwaddr we stashed in sendmsg */
-		if (cb->halen != skb->dev->addr_len) {
+	/* direct route; use the hwaddr we stashed in sendmsg */
+	if (dst->halen) {
+		if (dst->halen != skb->dev->addr_len) {
 			/* sanity check, sendmsg should have already caught this */
 			kfree_skb(skb);
 			return -EMSGSIZE;
 		}
-		daddr = cb->haddr;
+		daddr = dst->haddr;
 	} else {
 		/* If lookup fails let the device handle daddr==NULL */
-		if (mctp_neigh_lookup(route->dev, hdr->dest, daddr_buf) == 0)
+		if (mctp_neigh_lookup(dst->dev, dst->nexthop, daddr_buf) == 0)
 			daddr = daddr_buf;
 	}
 
@@ -599,7 +647,7 @@ static int mctp_route_output(struct mctp_route *route, struct sk_buff *skb)
 		return -EHOSTUNREACH;
 	}
 
-	mctp_flow_prepare_output(skb, route->dev);
+	mctp_flow_prepare_output(skb, dst->dev);
 
 	rc = dev_queue_xmit(skb);
 	if (rc)
@@ -612,7 +660,8 @@ static int mctp_route_output(struct mctp_route *route, struct sk_buff *skb)
 static void mctp_route_release(struct mctp_route *rt)
 {
 	if (refcount_dec_and_test(&rt->refs)) {
-		mctp_dev_put(rt->dev);
+		if (rt->dst_type == MCTP_ROUTE_DIRECT)
+			mctp_dev_put(rt->dev);
 		kfree_rcu(rt, rcu);
 	}
 }
@@ -628,7 +677,7 @@ static struct mctp_route *mctp_route_alloc(void)
 
 	INIT_LIST_HEAD(&rt->list);
 	refcount_set(&rt->refs, 1);
-	rt->output = mctp_route_discard;
+	rt->output = mctp_dst_discard;
 
 	return rt;
 }
@@ -801,10 +850,16 @@ static struct mctp_sk_key *mctp_lookup_prealloc_tag(struct mctp_sock *msk,
 }
 
 /* routing lookups */
+static unsigned int mctp_route_netid(struct mctp_route *rt)
+{
+	return rt->dst_type == MCTP_ROUTE_DIRECT ?
+		READ_ONCE(rt->dev->net) : rt->gateway.net;
+}
+
 static bool mctp_rt_match_eid(struct mctp_route *rt,
 			      unsigned int net, mctp_eid_t eid)
 {
-	return READ_ONCE(rt->dev->net) == net &&
+	return mctp_route_netid(rt) == net &&
 		rt->min <= eid && rt->max >= eid;
 }
 
@@ -813,54 +868,150 @@ static bool mctp_rt_compare_exact(struct mctp_route *rt1,
 				  struct mctp_route *rt2)
 {
 	ASSERT_RTNL();
-	return rt1->dev->net == rt2->dev->net &&
+	return mctp_route_netid(rt1) == mctp_route_netid(rt2) &&
 		rt1->min == rt2->min &&
 		rt1->max == rt2->max;
 }
 
-struct mctp_route *mctp_route_lookup(struct net *net, unsigned int dnet,
-				     mctp_eid_t daddr)
+/* must only be called on a direct route, as the final output hop */
+static void mctp_dst_from_route(struct mctp_dst *dst, mctp_eid_t eid,
+				unsigned int mtu, struct mctp_route *route)
+{
+	mctp_dev_hold(route->dev);
+	dst->nexthop = eid;
+	dst->dev = route->dev;
+	dst->mtu = READ_ONCE(dst->dev->dev->mtu);
+	if (mtu)
+		dst->mtu = min(dst->mtu, mtu);
+	dst->halen = 0;
+	dst->output = route->output;
+}
+
+int mctp_dst_from_extaddr(struct mctp_dst *dst, struct net *net, int ifindex,
+			  unsigned char halen, const unsigned char *haddr)
 {
-	struct mctp_route *tmp, *rt = NULL;
+	struct net_device *netdev;
+	struct mctp_dev *dev;
+	int rc = -ENOENT;
+
+	if (halen > sizeof(dst->haddr))
+		return -EINVAL;
 
 	rcu_read_lock();
 
-	list_for_each_entry_rcu(tmp, &net->mctp.routes, list) {
-		/* TODO: add metrics */
-		if (mctp_rt_match_eid(tmp, dnet, daddr)) {
-			if (refcount_inc_not_zero(&tmp->refs)) {
-				rt = tmp;
-				break;
-			}
-		}
+	netdev = dev_get_by_index_rcu(net, ifindex);
+	if (!netdev)
+		goto out_unlock;
+
+	if (netdev->addr_len != halen) {
+		rc = -EINVAL;
+		goto out_unlock;
 	}
 
+	dev = __mctp_dev_get(netdev);
+	if (!dev)
+		goto out_unlock;
+
+	dst->dev = dev;
+	dst->mtu = READ_ONCE(netdev->mtu);
+	dst->halen = halen;
+	dst->output = mctp_dst_output;
+	dst->nexthop = 0;
+	memcpy(dst->haddr, haddr, halen);
+
+	rc = 0;
+
+out_unlock:
 	rcu_read_unlock();
+	return rc;
+}
 
-	return rt;
+void mctp_dst_release(struct mctp_dst *dst)
+{
+	mctp_dev_put(dst->dev);
+}
+
+static struct mctp_route *mctp_route_lookup_single(struct net *net,
+						   unsigned int dnet,
+						   mctp_eid_t daddr)
+{
+	struct mctp_route *rt;
+
+	list_for_each_entry_rcu(rt, &net->mctp.routes, list) {
+		if (mctp_rt_match_eid(rt, dnet, daddr))
+			return rt;
+	}
+
+	return NULL;
 }
 
-static struct mctp_route *mctp_route_lookup_null(struct net *net,
-						 struct net_device *dev)
+/* populates *dst on successful lookup, if set */
+int mctp_route_lookup(struct net *net, unsigned int dnet,
+		      mctp_eid_t daddr, struct mctp_dst *dst)
 {
-	struct mctp_route *tmp, *rt = NULL;
+	const unsigned int max_depth = 32;
+	unsigned int depth, mtu = 0;
+	int rc = -EHOSTUNREACH;
 
 	rcu_read_lock();
 
-	list_for_each_entry_rcu(tmp, &net->mctp.routes, list) {
-		if (tmp->dev->dev == dev && tmp->type == RTN_LOCAL &&
-		    refcount_inc_not_zero(&tmp->refs)) {
-			rt = tmp;
+	for (depth = 0; depth < max_depth; depth++) {
+		struct mctp_route *rt;
+
+		rt = mctp_route_lookup_single(net, dnet, daddr);
+		if (!rt)
+			break;
+
+		/* clamp mtu to the smallest in the path, allowing 0
+		 * to specify no restrictions
+		 */
+		if (mtu && rt->mtu)
+			mtu = min(mtu, rt->mtu);
+		else
+			mtu = mtu ?: rt->mtu;
+
+		if (rt->dst_type == MCTP_ROUTE_DIRECT) {
+			if (dst)
+				mctp_dst_from_route(dst, daddr, mtu, rt);
+			rc = 0;
 			break;
+
+		} else if (rt->dst_type == MCTP_ROUTE_GATEWAY) {
+			daddr = rt->gateway.eid;
 		}
 	}
 
 	rcu_read_unlock();
 
-	return rt;
+	return rc;
+}
+
+static int mctp_route_lookup_null(struct net *net, struct net_device *dev,
+				  struct mctp_dst *dst)
+{
+	int rc = -EHOSTUNREACH;
+	struct mctp_route *rt;
+
+	rcu_read_lock();
+
+	list_for_each_entry_rcu(rt, &net->mctp.routes, list) {
+		if (rt->dst_type != MCTP_ROUTE_DIRECT || rt->type != RTN_LOCAL)
+			continue;
+
+		if (rt->dev->dev != dev)
+			continue;
+
+		mctp_dst_from_route(dst, 0, 0, rt);
+		rc = 0;
+		break;
+	}
+
+	rcu_read_unlock();
+
+	return rc;
 }
 
-static int mctp_do_fragment_route(struct mctp_route *rt, struct sk_buff *skb,
+static int mctp_do_fragment_route(struct mctp_dst *dst, struct sk_buff *skb,
 				  unsigned int mtu, u8 tag)
 {
 	const unsigned int hlen = sizeof(struct mctp_hdr);
@@ -933,7 +1084,7 @@ static int mctp_do_fragment_route(struct mctp_route *rt, struct sk_buff *skb,
 		skb_ext_copy(skb2, skb);
 
 		/* do route */
-		rc = rt->output(rt, skb2);
+		rc = dst->output(dst, skb2);
 		if (rc)
 			break;
 
@@ -945,68 +1096,34 @@ static int mctp_do_fragment_route(struct mctp_route *rt, struct sk_buff *skb,
 	return rc;
 }
 
-int mctp_local_output(struct sock *sk, struct mctp_route *rt,
+int mctp_local_output(struct sock *sk, struct mctp_dst *dst,
 		      struct sk_buff *skb, mctp_eid_t daddr, u8 req_tag)
 {
 	struct mctp_sock *msk = container_of(sk, struct mctp_sock, sk);
-	struct mctp_skb_cb *cb = mctp_cb(skb);
-	struct mctp_route tmp_rt = {0};
 	struct mctp_sk_key *key;
 	struct mctp_hdr *hdr;
 	unsigned long flags;
 	unsigned int netid;
 	unsigned int mtu;
 	mctp_eid_t saddr;
-	bool ext_rt;
 	int rc;
 	u8 tag;
 
-	rc = -ENODEV;
-
-	if (rt) {
-		ext_rt = false;
-		if (WARN_ON(!rt->dev))
-			goto out_release;
-
-	} else if (cb->ifindex) {
-		struct net_device *dev;
-
-		ext_rt = true;
-		rt = &tmp_rt;
-
-		rcu_read_lock();
-		dev = dev_get_by_index_rcu(sock_net(sk), cb->ifindex);
-		if (!dev) {
-			rcu_read_unlock();
-			goto out_free;
-		}
-		rt->dev = __mctp_dev_get(dev);
-		rcu_read_unlock();
-
-		if (!rt->dev)
-			goto out_release;
+	KUNIT_STATIC_STUB_REDIRECT(mctp_local_output, sk, dst, skb, daddr,
+				   req_tag);
 
-		/* establish temporary route - we set up enough to keep
-		 * mctp_route_output happy
-		 */
-		rt->output = mctp_route_output;
-		rt->mtu = 0;
-
-	} else {
-		rc = -EINVAL;
-		goto out_free;
-	}
+	rc = -ENODEV;
 
-	spin_lock_irqsave(&rt->dev->addrs_lock, flags);
-	if (rt->dev->num_addrs == 0) {
+	spin_lock_irqsave(&dst->dev->addrs_lock, flags);
+	if (dst->dev->num_addrs == 0) {
 		rc = -EHOSTUNREACH;
 	} else {
 		/* use the outbound interface's first address as our source */
-		saddr = rt->dev->addrs[0];
+		saddr = dst->dev->addrs[0];
 		rc = 0;
 	}
-	spin_unlock_irqrestore(&rt->dev->addrs_lock, flags);
-	netid = READ_ONCE(rt->dev->net);
+	spin_unlock_irqrestore(&dst->dev->addrs_lock, flags);
+	netid = READ_ONCE(dst->dev->net);
 
 	if (rc)
 		goto out_release;
@@ -1032,15 +1149,13 @@ int mctp_local_output(struct sock *sk, struct mctp_route *rt,
 		tag = req_tag & MCTP_TAG_MASK;
 	}
 
+	skb->pkt_type = PACKET_OUTGOING;
 	skb->protocol = htons(ETH_P_MCTP);
 	skb->priority = 0;
 	skb_reset_transport_header(skb);
 	skb_push(skb, sizeof(struct mctp_hdr));
 	skb_reset_network_header(skb);
-	skb->dev = rt->dev->dev;
-
-	/* cb->net will have been set on initial ingress */
-	cb->src = saddr;
+	skb->dev = dst->dev->dev;
 
 	/* set up common header fields */
 	hdr = mctp_hdr(skb);
@@ -1048,73 +1163,64 @@ int mctp_local_output(struct sock *sk, struct mctp_route *rt,
 	hdr->dest = daddr;
 	hdr->src = saddr;
 
-	mtu = mctp_route_mtu(rt);
+	mtu = dst->mtu;
 
 	if (skb->len + sizeof(struct mctp_hdr) <= mtu) {
 		hdr->flags_seq_tag = MCTP_HDR_FLAG_SOM |
 			MCTP_HDR_FLAG_EOM | tag;
-		rc = rt->output(rt, skb);
+		rc = dst->output(dst, skb);
 	} else {
-		rc = mctp_do_fragment_route(rt, skb, mtu, tag);
+		rc = mctp_do_fragment_route(dst, skb, mtu, tag);
 	}
 
 	/* route output functions consume the skb, even on error */
 	skb = NULL;
 
 out_release:
-	if (!ext_rt)
-		mctp_route_release(rt);
-
-	mctp_dev_put(tmp_rt.dev);
-
-out_free:
 	kfree_skb(skb);
 	return rc;
 }
 
 /* route management */
-static int mctp_route_add(struct mctp_dev *mdev, mctp_eid_t daddr_start,
-			  unsigned int daddr_extent, unsigned int mtu,
-			  unsigned char type)
+
+/* mctp_route_add(): Add the provided route, previously allocated via
+ * mctp_route_alloc(). On success, takes ownership of @rt, which includes a
+ * hold on rt->dev for usage in the route table. On failure a caller will want
+ * to mctp_route_release().
+ *
+ * We expect that the caller has set rt->type, rt->dst_type, rt->min, rt->max,
+ * rt->mtu and either rt->dev (with a reference held appropriately) or
+ * rt->gateway. Other fields will be populated.
+ */
+static int mctp_route_add(struct net *net, struct mctp_route *rt)
 {
-	int (*rtfn)(struct mctp_route *rt, struct sk_buff *skb);
-	struct net *net = dev_net(mdev->dev);
-	struct mctp_route *rt, *ert;
+	struct mctp_route *ert;
 
-	if (!mctp_address_unicast(daddr_start))
+	if (!mctp_address_unicast(rt->min) || !mctp_address_unicast(rt->max))
 		return -EINVAL;
 
-	if (daddr_extent > 0xff || daddr_start + daddr_extent >= 255)
+	if (rt->dst_type == MCTP_ROUTE_DIRECT && !rt->dev)
+		return -EINVAL;
+
+	if (rt->dst_type == MCTP_ROUTE_GATEWAY && !rt->gateway.eid)
 		return -EINVAL;
 
-	switch (type) {
+	switch (rt->type) {
 	case RTN_LOCAL:
-		rtfn = mctp_route_input;
+		rt->output = mctp_dst_input;
 		break;
 	case RTN_UNICAST:
-		rtfn = mctp_route_output;
+		rt->output = mctp_dst_output;
 		break;
 	default:
 		return -EINVAL;
 	}
 
-	rt = mctp_route_alloc();
-	if (!rt)
-		return -ENOMEM;
-
-	rt->min = daddr_start;
-	rt->max = daddr_start + daddr_extent;
-	rt->mtu = mtu;
-	rt->dev = mdev;
-	mctp_dev_hold(rt->dev);
-	rt->type = type;
-	rt->output = rtfn;
-
 	ASSERT_RTNL();
+
 	/* Prevent duplicate identical routes. */
 	list_for_each_entry(ert, &net->mctp.routes, list) {
 		if (mctp_rt_compare_exact(rt, ert)) {
-			mctp_route_release(rt);
 			return -EEXIST;
 		}
 	}
@@ -1124,10 +1230,10 @@ static int mctp_route_add(struct mctp_dev *mdev, mctp_eid_t daddr_start,
 	return 0;
 }
 
-static int mctp_route_remove(struct mctp_dev *mdev, mctp_eid_t daddr_start,
-			     unsigned int daddr_extent, unsigned char type)
+static int mctp_route_remove(struct net *net, unsigned int netid,
+			     mctp_eid_t daddr_start, unsigned int daddr_extent,
+			     unsigned char type)
 {
-	struct net *net = dev_net(mdev->dev);
 	struct mctp_route *rt, *tmp;
 	mctp_eid_t daddr_end;
 	bool dropped;
@@ -1141,7 +1247,7 @@ static int mctp_route_remove(struct mctp_dev *mdev, mctp_eid_t daddr_start,
 	ASSERT_RTNL();
 
 	list_for_each_entry_safe(rt, tmp, &net->mctp.routes, list) {
-		if (rt->dev == mdev &&
+		if (mctp_route_netid(rt) == netid &&
 		    rt->min == daddr_start && rt->max == daddr_end &&
 		    rt->type == type) {
 			list_del_rcu(&rt->list);
@@ -1156,12 +1262,32 @@ static int mctp_route_remove(struct mctp_dev *mdev, mctp_eid_t daddr_start,
 
 int mctp_route_add_local(struct mctp_dev *mdev, mctp_eid_t addr)
 {
-	return mctp_route_add(mdev, addr, 0, 0, RTN_LOCAL);
+	struct mctp_route *rt;
+	int rc;
+
+	rt = mctp_route_alloc();
+	if (!rt)
+		return -ENOMEM;
+
+	rt->min = addr;
+	rt->max = addr;
+	rt->dst_type = MCTP_ROUTE_DIRECT;
+	rt->dev = mdev;
+	rt->type = RTN_LOCAL;
+
+	mctp_dev_hold(rt->dev);
+
+	rc = mctp_route_add(dev_net(mdev->dev), rt);
+	if (rc)
+		mctp_route_release(rt);
+
+	return rc;
 }
 
 int mctp_route_remove_local(struct mctp_dev *mdev, mctp_eid_t addr)
 {
-	return mctp_route_remove(mdev, addr, 0, RTN_LOCAL);
+	return mctp_route_remove(dev_net(mdev->dev), mdev->net,
+				 addr, 0, RTN_LOCAL);
 }
 
 /* removes all entries for a given device */
@@ -1172,7 +1298,7 @@ void mctp_route_remove_dev(struct mctp_dev *mdev)
 
 	ASSERT_RTNL();
 	list_for_each_entry_safe(rt, tmp, &net->mctp.routes, list) {
-		if (rt->dev == mdev) {
+		if (rt->dst_type == MCTP_ROUTE_DIRECT && rt->dev == mdev) {
 			list_del_rcu(&rt->list);
 			/* TODO: immediate RTM_DELROUTE */
 			mctp_route_release(rt);
@@ -1189,8 +1315,9 @@ static int mctp_pkttype_receive(struct sk_buff *skb, struct net_device *dev,
 	struct net *net = dev_net(dev);
 	struct mctp_dev *mdev;
 	struct mctp_skb_cb *cb;
-	struct mctp_route *rt;
+	struct mctp_dst dst;
 	struct mctp_hdr *mh;
+	int rc;
 
 	rcu_read_lock();
 	mdev = __mctp_dev_get(dev);
@@ -1232,17 +1359,17 @@ static int mctp_pkttype_receive(struct sk_buff *skb, struct net_device *dev,
 	cb->net = READ_ONCE(mdev->net);
 	cb->ifindex = dev->ifindex;
 
-	rt = mctp_route_lookup(net, cb->net, mh->dest);
+	rc = mctp_route_lookup(net, cb->net, mh->dest, &dst);
 
 	/* NULL EID, but addressed to our physical address */
-	if (!rt && mh->dest == MCTP_ADDR_NULL && skb->pkt_type == PACKET_HOST)
-		rt = mctp_route_lookup_null(net, dev);
+	if (rc && mh->dest == MCTP_ADDR_NULL && skb->pkt_type == PACKET_HOST)
+		rc = mctp_route_lookup_null(net, dev, &dst);
 
-	if (!rt)
+	if (rc)
 		goto err_drop;
 
-	rt->output(rt, skb);
-	mctp_route_release(rt);
+	dst.output(&dst, skb);
+	mctp_dst_release(&dst);
 	mctp_dev_put(mdev);
 
 	return NET_RX_SUCCESS;
@@ -1264,19 +1391,28 @@ static const struct nla_policy rta_mctp_policy[RTA_MAX + 1] = {
 	[RTA_DST]		= { .type = NLA_U8 },
 	[RTA_METRICS]		= { .type = NLA_NESTED },
 	[RTA_OIF]		= { .type = NLA_U32 },
+	[RTA_GATEWAY]		= NLA_POLICY_EXACT_LEN(sizeof(struct mctp_fq_addr)),
+};
+
+static const struct nla_policy rta_metrics_policy[RTAX_MAX + 1] = {
+	[RTAX_MTU]		= { .type = NLA_U32 },
 };
 
-/* Common part for RTM_NEWROUTE and RTM_DELROUTE parsing.
- * tb must hold RTA_MAX+1 elements.
+/* base parsing; common to both _lookup and _populate variants.
+ *
+ * For gateway routes (which have a RTA_GATEWAY, and no RTA_OIF), we populate
+ * *gatweayp. for direct routes (RTA_OIF, no RTA_GATEWAY), we populate *mdev.
  */
-static int mctp_route_nlparse(struct sk_buff *skb, struct nlmsghdr *nlh,
-			      struct netlink_ext_ack *extack,
-			      struct nlattr **tb, struct rtmsg **rtm,
-			      struct mctp_dev **mdev, mctp_eid_t *daddr_start)
+static int mctp_route_nlparse_common(struct net *net, struct nlmsghdr *nlh,
+				     struct netlink_ext_ack *extack,
+				     struct nlattr **tb, struct rtmsg **rtm,
+				     struct mctp_dev **mdev,
+				     struct mctp_fq_addr *gatewayp,
+				     mctp_eid_t *daddr_start)
 {
-	struct net *net = sock_net(skb->sk);
+	struct mctp_fq_addr *gateway = NULL;
+	unsigned int ifindex = 0;
 	struct net_device *dev;
-	unsigned int ifindex;
 	int rc;
 
 	rc = nlmsg_parse(nlh, sizeof(struct rtmsg), tb, RTA_MAX,
@@ -1292,11 +1428,44 @@ static int mctp_route_nlparse(struct sk_buff *skb, struct nlmsghdr *nlh,
 	}
 	*daddr_start = nla_get_u8(tb[RTA_DST]);
 
-	if (!tb[RTA_OIF]) {
-		NL_SET_ERR_MSG(extack, "ifindex missing");
+	if (tb[RTA_OIF])
+		ifindex = nla_get_u32(tb[RTA_OIF]);
+
+	if (tb[RTA_GATEWAY])
+		gateway = nla_data(tb[RTA_GATEWAY]);
+
+	if (ifindex && gateway) {
+		NL_SET_ERR_MSG(extack,
+			       "cannot specify both ifindex and gateway");
+		return -EINVAL;
+
+	} else if (ifindex) {
+		dev = __dev_get_by_index(net, ifindex);
+		if (!dev) {
+			NL_SET_ERR_MSG(extack, "bad ifindex");
+			return -ENODEV;
+		}
+		*mdev = mctp_dev_get_rtnl(dev);
+		if (!*mdev)
+			return -ENODEV;
+		gatewayp->eid = 0;
+
+	} else if (gateway) {
+		if (!mctp_address_unicast(gateway->eid)) {
+			NL_SET_ERR_MSG(extack, "bad gateway");
+			return -EINVAL;
+		}
+
+		gatewayp->eid = gateway->eid;
+		gatewayp->net = gateway->net != MCTP_NET_ANY ?
+			gateway->net :
+			READ_ONCE(net->mctp.default_net);
+		*mdev = NULL;
+
+	} else {
+		NL_SET_ERR_MSG(extack, "no route output provided");
 		return -EINVAL;
 	}
-	ifindex = nla_get_u32(tb[RTA_OIF]);
 
 	*rtm = nlmsg_data(nlh);
 	if ((*rtm)->rtm_family != AF_MCTP) {
@@ -1304,82 +1473,157 @@ static int mctp_route_nlparse(struct sk_buff *skb, struct nlmsghdr *nlh,
 		return -EINVAL;
 	}
 
-	dev = __dev_get_by_index(net, ifindex);
-	if (!dev) {
-		NL_SET_ERR_MSG(extack, "bad ifindex");
-		return -ENODEV;
-	}
-	*mdev = mctp_dev_get_rtnl(dev);
-	if (!*mdev)
-		return -ENODEV;
-
-	if (dev->flags & IFF_LOOPBACK) {
-		NL_SET_ERR_MSG(extack, "no routes to loopback");
+	if ((*rtm)->rtm_type != RTN_UNICAST) {
+		NL_SET_ERR_MSG(extack, "rtm_type must be RTN_UNICAST");
 		return -EINVAL;
 	}
 
 	return 0;
 }
 
-static const struct nla_policy rta_metrics_policy[RTAX_MAX + 1] = {
-	[RTAX_MTU]		= { .type = NLA_U32 },
-};
-
-static int mctp_newroute(struct sk_buff *skb, struct nlmsghdr *nlh,
-			 struct netlink_ext_ack *extack)
+/* Route parsing for lookup operations; we only need the "route target"
+ * components (ie., network and dest-EID range).
+ */
+static int mctp_route_nlparse_lookup(struct net *net, struct nlmsghdr *nlh,
+				     struct netlink_ext_ack *extack,
+				     unsigned char *type, unsigned int *netid,
+				     mctp_eid_t *daddr_start,
+				     unsigned int *daddr_extent)
 {
 	struct nlattr *tb[RTA_MAX + 1];
+	struct mctp_fq_addr gw;
+	struct mctp_dev *mdev;
+	struct rtmsg *rtm;
+	int rc;
+
+	rc = mctp_route_nlparse_common(net, nlh, extack, tb, &rtm,
+				       &mdev, &gw, daddr_start);
+	if (rc)
+		return rc;
+
+	if (mdev) {
+		*netid = mdev->net;
+	} else if (gw.eid) {
+		*netid = gw.net;
+	} else {
+		/* bug: _nlparse_common should not allow this */
+		return -1;
+	}
+
+	*type = rtm->rtm_type;
+	*daddr_extent = rtm->rtm_dst_len;
+
+	return 0;
+}
+
+/* Full route parse for RTM_NEWROUTE: populate @rt. On success,
+ * MCTP_ROUTE_DIRECT routes (ie, those with a direct dev) will hold a reference
+ * to that dev.
+ */
+static int mctp_route_nlparse_populate(struct net *net, struct nlmsghdr *nlh,
+				       struct netlink_ext_ack *extack,
+				       struct mctp_route *rt)
+{
 	struct nlattr *tbx[RTAX_MAX + 1];
+	struct nlattr *tb[RTA_MAX + 1];
+	unsigned int daddr_extent;
+	struct mctp_fq_addr gw;
 	mctp_eid_t daddr_start;
-	struct mctp_dev *mdev;
+	struct mctp_dev *dev;
 	struct rtmsg *rtm;
-	unsigned int mtu;
+	u32 mtu = 0;
 	int rc;
 
-	rc = mctp_route_nlparse(skb, nlh, extack, tb,
-				&rtm, &mdev, &daddr_start);
-	if (rc < 0)
+	rc = mctp_route_nlparse_common(net, nlh, extack, tb, &rtm,
+				       &dev, &gw, &daddr_start);
+	if (rc)
 		return rc;
 
-	if (rtm->rtm_type != RTN_UNICAST) {
-		NL_SET_ERR_MSG(extack, "rtm_type must be RTN_UNICAST");
+	daddr_extent = rtm->rtm_dst_len;
+
+	if (daddr_extent > 0xff || daddr_extent + daddr_start >= 255) {
+		NL_SET_ERR_MSG(extack, "invalid eid range");
 		return -EINVAL;
 	}
 
-	mtu = 0;
 	if (tb[RTA_METRICS]) {
 		rc = nla_parse_nested(tbx, RTAX_MAX, tb[RTA_METRICS],
 				      rta_metrics_policy, NULL);
-		if (rc < 0)
+		if (rc < 0) {
+			NL_SET_ERR_MSG(extack, "incorrect RTA_METRICS format");
 			return rc;
+		}
 		if (tbx[RTAX_MTU])
 			mtu = nla_get_u32(tbx[RTAX_MTU]);
 	}
 
-	rc = mctp_route_add(mdev, daddr_start, rtm->rtm_dst_len, mtu,
-			    rtm->rtm_type);
+	rt->type = rtm->rtm_type;
+	rt->min = daddr_start;
+	rt->max = daddr_start + daddr_extent;
+	rt->mtu = mtu;
+	if (gw.eid) {
+		rt->dst_type = MCTP_ROUTE_GATEWAY;
+		rt->gateway.eid = gw.eid;
+		rt->gateway.net = gw.net;
+	} else {
+		rt->dst_type = MCTP_ROUTE_DIRECT;
+		rt->dev = dev;
+		mctp_dev_hold(rt->dev);
+	}
+
+	return 0;
+}
+
+static int mctp_newroute(struct sk_buff *skb, struct nlmsghdr *nlh,
+			 struct netlink_ext_ack *extack)
+{
+	struct net *net = sock_net(skb->sk);
+	struct mctp_route *rt;
+	int rc;
+
+	rt = mctp_route_alloc();
+	if (!rt)
+		return -ENOMEM;
+
+	rc = mctp_route_nlparse_populate(net, nlh, extack, rt);
+	if (rc < 0)
+		goto err_free;
+
+	if (rt->dst_type == MCTP_ROUTE_DIRECT &&
+	    rt->dev->dev->flags & IFF_LOOPBACK) {
+		NL_SET_ERR_MSG(extack, "no routes to loopback");
+		rc = -EINVAL;
+		goto err_free;
+	}
+
+	rc = mctp_route_add(net, rt);
+	if (!rc)
+		return 0;
+
+err_free:
+	mctp_route_release(rt);
 	return rc;
 }
 
 static int mctp_delroute(struct sk_buff *skb, struct nlmsghdr *nlh,
 			 struct netlink_ext_ack *extack)
 {
-	struct nlattr *tb[RTA_MAX + 1];
+	struct net *net = sock_net(skb->sk);
+	unsigned int netid, daddr_extent;
+	unsigned char type = RTN_UNSPEC;
 	mctp_eid_t daddr_start;
-	struct mctp_dev *mdev;
-	struct rtmsg *rtm;
 	int rc;
 
-	rc = mctp_route_nlparse(skb, nlh, extack, tb,
-				&rtm, &mdev, &daddr_start);
+	rc = mctp_route_nlparse_lookup(net, nlh, extack, &type, &netid,
+				       &daddr_start, &daddr_extent);
 	if (rc < 0)
 		return rc;
 
 	/* we only have unicast routes */
-	if (rtm->rtm_type != RTN_UNICAST)
+	if (type != RTN_UNICAST)
 		return -EINVAL;
 
-	rc = mctp_route_remove(mdev, daddr_start, rtm->rtm_dst_len, RTN_UNICAST);
+	rc = mctp_route_remove(net, netid, daddr_start, daddr_extent, type);
 	return rc;
 }
 
@@ -1405,7 +1649,6 @@ static int mctp_fill_rtinfo(struct sk_buff *skb, struct mctp_route *rt,
 	hdr->rtm_tos = 0;
 	hdr->rtm_table = RT_TABLE_DEFAULT;
 	hdr->rtm_protocol = RTPROT_STATIC; /* everything is user-defined */
-	hdr->rtm_scope = RT_SCOPE_LINK; /* TODO: scope in mctp_route? */
 	hdr->rtm_type = rt->type;
 
 	if (nla_put_u8(skb, RTA_DST, rt->min))
@@ -1422,13 +1665,17 @@ static int mctp_fill_rtinfo(struct sk_buff *skb, struct mctp_route *rt,
 
 	nla_nest_end(skb, metrics);
 
-	if (rt->dev) {
+	if (rt->dst_type == MCTP_ROUTE_DIRECT) {
+		hdr->rtm_scope = RT_SCOPE_LINK;
 		if (nla_put_u32(skb, RTA_OIF, rt->dev->dev->ifindex))
 			goto cancel;
+	} else if (rt->dst_type == MCTP_ROUTE_GATEWAY) {
+		hdr->rtm_scope = RT_SCOPE_UNIVERSE;
+		if (nla_put(skb, RTA_GATEWAY,
+			    sizeof(rt->gateway), &rt->gateway))
+			goto cancel;
 	}
 
-	/* TODO: conditional neighbour physaddr? */
-
 	nlmsg_end(skb, nlh);
 
 	return 0;
@@ -1475,7 +1722,7 @@ static int __net_init mctp_routes_net_init(struct net *net)
 	struct netns_mctp *ns = &net->mctp;
 
 	INIT_LIST_HEAD(&ns->routes);
-	INIT_HLIST_HEAD(&ns->binds);
+	hash_init(ns->binds);
 	mutex_init(&ns->bind_lock);
 	INIT_HLIST_HEAD(&ns->keys);
 	spin_lock_init(&ns->keys_lock);