Merge tag 'net-5.18-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net

Pull networking fixes from Paolo Abeni:
 "Including fixes from bpf and netfilter.

  Current release - new code bugs:

   - mctp: correct mctp_i2c_header_create result

   - eth: fungible: fix reference to __udivdi3 on 32b builds

   - eth: micrel: remove latencies support lan8814

  Previous releases - regressions:

   - bpf: resolve to prog->aux->dst_prog->type only for BPF_PROG_TYPE_EXT

   - vrf: fix packet sniffing for traffic originating from ip tunnels

   - rxrpc: fix a race in rxrpc_exit_net()

   - dsa: revert "net: dsa: stop updating master MTU from master.c"

   - eth: ice: fix MAC address setting

  Previous releases - always broken:

   - tls: fix slab-out-of-bounds bug in decrypt_internal

   - bpf: support dual-stack sockets in bpf_tcp_check_syncookie

   - xdp: fix coalescing for page_pool fragment recycling

   - ovs: fix leak of nested actions

   - eth: sfc:
      - add missing xdp queue reinitialization
      - fix using uninitialized xdp tx_queue

   - eth: ice:
      - clear default forwarding VSI during VSI release
      - fix broken IFF_ALLMULTI handling
      - synchronize_rcu() when terminating rings

   - eth: qede: confirm skb is allocated before using

   - eth: aqc111: fix out-of-bounds accesses in RX fixup

   - eth: slip: fix NPD bug in sl_tx_timeout()"

* tag 'net-5.18-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (61 commits)
  drivers: net: slip: fix NPD bug in sl_tx_timeout()
  bpf: Adjust bpf_tcp_check_syncookie selftest to test dual-stack sockets
  bpf: Support dual-stack sockets in bpf_tcp_check_syncookie
  myri10ge: fix an incorrect free for skb in myri10ge_sw_tso
  net: usb: aqc111: Fix out-of-bounds accesses in RX fixup
  qede: confirm skb is allocated before using
  net: ipv6mr: fix unused variable warning with CONFIG_IPV6_PIMSM_V2=n
  net: phy: mscc-miim: reject clause 45 register accesses
  net: axiemac: use a phandle to reference pcs_phy
  dt-bindings: net: add pcs-handle attribute
  net: axienet: factor out phy_node in struct axienet_local
  net: axienet: setup mdio unconditionally
  net: sfc: fix using uninitialized xdp tx_queue
  rxrpc: fix a race in rxrpc_exit_net()
  net: openvswitch: fix leak of nested actions
  net: ethernet: mv643xx: Fix over zealous checking of_get_mac_address()
  net: openvswitch: don't send internal clone attribute to the userspace.
  net: micrel: Fix KS8851 Kconfig
  ice: clear cmd_type_offset_bsz for TX rings
  ice: xsk: fix VSI state check in ice_xsk_wakeup()
  ...

2 files changed, 45 insertions, 17 deletions

diff --git a/net/mctp/af_mctp.c b/net/mctp/af_mctp.c
index f0702d920d8d..e22b0cbb2f35 100644
--- a/net/mctp/af_mctp.c
+++ b/net/mctp/af_mctp.c
@@ -93,13 +93,13 @@ out_release:
 static int mctp_sendmsg(struct socket *sock, struct msghdr *msg, size_t len)
 {
 	DECLARE_SOCKADDR(struct sockaddr_mctp *, addr, msg->msg_name);
-	const int hlen = MCTP_HEADER_MAXLEN + sizeof(struct mctp_hdr);
 	int rc, addrlen = msg->msg_namelen;
 	struct sock *sk = sock->sk;
 	struct mctp_sock *msk = container_of(sk, struct mctp_sock, sk);
 	struct mctp_skb_cb *cb;
 	struct mctp_route *rt;
-	struct sk_buff *skb;
+	struct sk_buff *skb = NULL;
+	int hlen;
 
 	if (addr) {
 		const u8 tagbits = MCTP_TAG_MASK | MCTP_TAG_OWNER |
@@ -129,6 +129,34 @@ static int mctp_sendmsg(struct socket *sock, struct msghdr *msg, size_t len)
 	if (addr->smctp_network == MCTP_NET_ANY)
 		addr->smctp_network = mctp_default_net(sock_net(sk));
 
+	/* direct addressing */
+	if (msk->addr_ext && addrlen >= sizeof(struct sockaddr_mctp_ext)) {
+		DECLARE_SOCKADDR(struct sockaddr_mctp_ext *,
+				 extaddr, msg->msg_name);
+		struct net_device *dev;
+
+		rc = -EINVAL;
+		rcu_read_lock();
+		dev = dev_get_by_index_rcu(sock_net(sk), extaddr->smctp_ifindex);
+		/* check for correct halen */
+		if (dev && extaddr->smctp_halen == dev->addr_len) {
+			hlen = LL_RESERVED_SPACE(dev) + sizeof(struct mctp_hdr);
+			rc = 0;
+		}
+		rcu_read_unlock();
+		if (rc)
+			goto err_free;
+		rt = NULL;
+	} else {
+		rt = mctp_route_lookup(sock_net(sk), addr->smctp_network,
+				       addr->smctp_addr.s_addr);
+		if (!rt) {
+			rc = -EHOSTUNREACH;
+			goto err_free;
+		}
+		hlen = LL_RESERVED_SPACE(rt->dev->dev) + sizeof(struct mctp_hdr);
+	}
+
 	skb = sock_alloc_send_skb(sk, hlen + 1 + len,
 				  msg->msg_flags & MSG_DONTWAIT, &rc);
 	if (!skb)
@@ -147,8 +175,8 @@ static int mctp_sendmsg(struct socket *sock, struct msghdr *msg, size_t len)
 	cb = __mctp_cb(skb);
 	cb->net = addr->smctp_network;
 
-	/* direct addressing */
-	if (msk->addr_ext && addrlen >= sizeof(struct sockaddr_mctp_ext)) {
+	if (!rt) {
+		/* fill extended address in cb */
 		DECLARE_SOCKADDR(struct sockaddr_mctp_ext *,
 				 extaddr, msg->msg_name);
 
@@ -159,17 +187,9 @@ static int mctp_sendmsg(struct socket *sock, struct msghdr *msg, size_t len)
 		}
 
 		cb->ifindex = extaddr->smctp_ifindex;
+		/* smctp_halen is checked above */
 		cb->halen = extaddr->smctp_halen;
 		memcpy(cb->haddr, extaddr->smctp_haddr, cb->halen);
-
-		rt = NULL;
-	} else {
-		rt = mctp_route_lookup(sock_net(sk), addr->smctp_network,
-				       addr->smctp_addr.s_addr);
-		if (!rt) {
-			rc = -EHOSTUNREACH;
-			goto err_free;
-		}
 	}
 
 	rc = mctp_local_output(sk, rt, skb, addr->smctp_addr.s_addr,
diff --git a/net/mctp/route.c b/net/mctp/route.c
index d5e7db83fe9d..3b24b8d18b5b 100644
--- a/net/mctp/route.c
+++ b/net/mctp/route.c
@@ -503,6 +503,11 @@ static int mctp_route_output(struct mctp_route *route, struct sk_buff *skb)
 
 	if (cb->ifindex) {
 		/* direct route; use the hwaddr we stashed in sendmsg */
+		if (cb->halen != skb->dev->addr_len) {
+			/* sanity check, sendmsg should have already caught this */
+			kfree_skb(skb);
+			return -EMSGSIZE;
+		}
 		daddr = cb->haddr;
 	} else {
 		/* If lookup fails let the device handle daddr==NULL */
@@ -512,7 +517,7 @@ static int mctp_route_output(struct mctp_route *route, struct sk_buff *skb)
 
 	rc = dev_hard_header(skb, skb->dev, ntohs(skb->protocol),
 			     daddr, skb->dev->dev_addr, skb->len);
-	if (rc) {
+	if (rc < 0) {
 		kfree_skb(skb);
 		return -EHOSTUNREACH;
 	}
@@ -756,7 +761,7 @@ static int mctp_do_fragment_route(struct mctp_route *rt, struct sk_buff *skb,
 {
 	const unsigned int hlen = sizeof(struct mctp_hdr);
 	struct mctp_hdr *hdr, *hdr2;
-	unsigned int pos, size;
+	unsigned int pos, size, headroom;
 	struct sk_buff *skb2;
 	int rc;
 	u8 seq;
@@ -770,6 +775,9 @@ static int mctp_do_fragment_route(struct mctp_route *rt, struct sk_buff *skb,
 		return -EMSGSIZE;
 	}
 
+	/* keep same headroom as the original skb */
+	headroom = skb_headroom(skb);
+
 	/* we've got the header */
 	skb_pull(skb, hlen);
 
@@ -777,7 +785,7 @@ static int mctp_do_fragment_route(struct mctp_route *rt, struct sk_buff *skb,
 		/* size of message payload */
 		size = min(mtu - hlen, skb->len - pos);
 
-		skb2 = alloc_skb(MCTP_HEADER_MAXLEN + hlen + size, GFP_KERNEL);
+		skb2 = alloc_skb(headroom + hlen + size, GFP_KERNEL);
 		if (!skb2) {
 			rc = -ENOMEM;
 			break;
@@ -793,7 +801,7 @@ static int mctp_do_fragment_route(struct mctp_route *rt, struct sk_buff *skb,
 			skb_set_owner_w(skb2, skb->sk);
 
 		/* establish packet */
-		skb_reserve(skb2, MCTP_HEADER_MAXLEN);
+		skb_reserve(skb2, headroom);
 		skb_reset_network_header(skb2);
 		skb_put(skb2, hlen + size);
 		skb2->transport_header = skb2->network_header + hlen;