net: clear the dst when changing skb protocol

A not-so-careful NAT46 BPF program can crash the kernel
if it indiscriminately flips ingress packets from v4 to v6:

  BUG: kernel NULL pointer dereference, address: 0000000000000000
    ip6_rcv_core (net/ipv6/ip6_input.c:190:20)
    ipv6_rcv (net/ipv6/ip6_input.c:306:8)
    process_backlog (net/core/dev.c:6186:4)
    napi_poll (net/core/dev.c:6906:9)
    net_rx_action (net/core/dev.c:7028:13)
    do_softirq (kernel/softirq.c:462:3)
    netif_rx (net/core/dev.c:5326:3)
    dev_loopback_xmit (net/core/dev.c:4015:2)
    ip_mc_finish_output (net/ipv4/ip_output.c:363:8)
    NF_HOOK (./include/linux/netfilter.h:314:9)
    ip_mc_output (net/ipv4/ip_output.c:400:5)
    dst_output (./include/net/dst.h:459:9)
    ip_local_out (net/ipv4/ip_output.c:130:9)
    ip_send_skb (net/ipv4/ip_output.c:1496:8)
    udp_send_skb (net/ipv4/udp.c:1040:8)
    udp_sendmsg (net/ipv4/udp.c:1328:10)

The output interface has a 4->6 program attached at ingress.
We try to loop the multicast skb back to the sending socket.
Ingress BPF runs as part of netif_rx(), pushes a valid v6 hdr
and changes skb->protocol to v6. We enter ip6_rcv_core which
tries to use skb_dst(). But the dst is still an IPv4 one left
after IPv4 mcast output.

Clear the dst in all BPF helpers which change the protocol.
Try to preserve metadata dsts, those may carry non-routing
metadata.

Cc: [email protected]
Reviewed-by: Maciej Żenczykowski <[email protected]>
Acked-by: Daniel Borkmann <[email protected]>
Fixes: d219df6 ("bpf: Add ipip6 and ip6ip decap support for bpf_skb_adjust_room()")
Fixes: 1b00e0d ("bpf: update skb->protocol in bpf_skb_net_grow")
Fixes: 6578171 ("bpf: add bpf_skb_change_proto helper")
Reviewed-by: Willem de Bruijn <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>

Author: kuba-moo

net/core/filter.c

@@ -3233,6 +3233,13 @@ static const struct bpf_func_proto bpf_skb_vlan_pop_proto = {
 	.arg1_type      = ARG_PTR_TO_CTX,
 };
 
+static void bpf_skb_change_protocol(struct sk_buff *skb, u16 proto)
+{
+	skb->protocol = htons(proto);
+	if (skb_valid_dst(skb))
+		skb_dst_drop(skb);
+}
+
 static int bpf_skb_generic_push(struct sk_buff *skb, u32 off, u32 len)
 {
 	/* Caller already did skb_cow() with len as headroom,
@@ -3329,7 +3336,7 @@ static int bpf_skb_proto_4_to_6(struct sk_buff *skb)
 		}
 	}
 
-	skb->protocol = htons(ETH_P_IPV6);
+	bpf_skb_change_protocol(skb, ETH_P_IPV6);
 	skb_clear_hash(skb);
 
 	return 0;
@@ -3359,7 +3366,7 @@ static int bpf_skb_proto_6_to_4(struct sk_buff *skb)
 		}
 	}
 
-	skb->protocol = htons(ETH_P_IP);
+	bpf_skb_change_protocol(skb, ETH_P_IP);
 	skb_clear_hash(skb);
 
 	return 0;
@@ -3550,10 +3557,10 @@ static int bpf_skb_net_grow(struct sk_buff *skb, u32 off, u32 len_diff,
 		/* Match skb->protocol to new outer l3 protocol */
 		if (skb->protocol == htons(ETH_P_IP) &&
 		    flags & BPF_F_ADJ_ROOM_ENCAP_L3_IPV6)
-			skb->protocol = htons(ETH_P_IPV6);
+			bpf_skb_change_protocol(skb, ETH_P_IPV6);
 		else if (skb->protocol == htons(ETH_P_IPV6) &&
 			 flags & BPF_F_ADJ_ROOM_ENCAP_L3_IPV4)
-			skb->protocol = htons(ETH_P_IP);
+			bpf_skb_change_protocol(skb, ETH_P_IP);
 	}
 
 	if (skb_is_gso(skb)) {
@@ -3606,10 +3613,10 @@ static int bpf_skb_net_shrink(struct sk_buff *skb, u32 off, u32 len_diff,
 	/* Match skb->protocol to new outer l3 protocol */
 	if (skb->protocol == htons(ETH_P_IP) &&
 	    flags & BPF_F_ADJ_ROOM_DECAP_L3_IPV6)
-		skb->protocol = htons(ETH_P_IPV6);
+		bpf_skb_change_protocol(skb, ETH_P_IPV6);
 	else if (skb->protocol == htons(ETH_P_IPV6) &&
 		 flags & BPF_F_ADJ_ROOM_DECAP_L3_IPV4)
-		skb->protocol = htons(ETH_P_IP);
+		bpf_skb_change_protocol(skb, ETH_P_IP);
 
 	if (skb_is_gso(skb)) {
 		struct skb_shared_info *shinfo = skb_shinfo(skb);