[PATCH v2] net/openvswitch: check Ethernet header length before key_extract()

When a packet arrives on an ARPHRD_NONE device (e.g. TUN),
ovs_flow_key_extract() trusts the user-provided skb->protocol field: if
it is ETH_P_TEB, the packet is classified as MAC_PROTO_ETHERNET and
key_extract() is called without ensuring the skb has ETH_HLEN (14) bytes
of linear data. key_extract() unconditionally pulls 2 * ETH_ALEN bytes
for MAC addresses and parse_ethertype() pulls 2 more, either of which
triggers a kernel BUG in __skb_pull() when the linear area is too small.

  kernel BUG at include/linux/skbuff.h:2848!
  RIP: 0010:key_extract+0xa7e/0xd90 net/openvswitch/flow.c:933
  ovs_flow_key_extract+0x419/0xa70
  ovs_vport_receive+0x222/0x390
  netdev_frame_hook+0x3e0/0x630
  tun_get_user+0x2d0c/0x38e0

Fixed by adding checks in key_extract() before pulling the Ethernet
header.

Fixes: 217ac77a3c25 ("openvswitch: allow L3 netdev ports")
Reported-by: AutonomousCodeSecurity@microsoft.com
Link: https://lore.kernel.org/all/20260721143602.64677-1-blbllhy@gmail.com (local)
Signed-off-by: Cen Zhang (Microsoft) <redacted>
---
v2: Moved the check into key_extract() per Ilya Maximets.

 net/openvswitch/flow.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/net/openvswitch/flow.c b/net/openvswitch/flow.c
index 66366982f604..de7b77d6702a 100644
--- a/net/openvswitch/flow.c
+++ b/net/openvswitch/flow.c

@@ -926,6 +926,9 @@ static int key_extract(struct sk_buff *skb, struct sw_flow_key *key)
 		skb_reset_network_header(skb);
 		key->eth.type = skb->protocol;
 	} else {
+		if (unlikely(!pskb_may_pull(skb, ETH_HLEN)))
+			return -EINVAL;
+
 		eth = eth_hdr(skb);
 		ether_addr_copy(key->eth.src, eth->h_source);
 		ether_addr_copy(key->eth.dst, eth->h_dest);

-- 
2.53.0