bpf: Fix bpf_redirect to an ipip/ip6tnl dev

If the bpf program calls bpf_redirect(dev, 0) and dev is
an ipip/ip6tnl, it currently includes the mac header.
e.g. If dev is ipip, the end result is IP-EthHdr-IP instead
of IP-IP.

The fix is to pull the mac header.  At ingress, skb_postpull_rcsum()
is not needed because the ethhdr should have been pulled once already
and then got pushed back just before calling the bpf_prog.
At egress, this patch calls skb_postpull_rcsum().

If bpf_redirect(dev, BPF_F_INGRESS) is called,
it also fails now because it calls dev_forward_skb() which
eventually calls eth_type_trans(skb, dev).  The eth_type_trans()
will set skb->type = PACKET_OTHERHOST because the mac address
does not match the redirecting dev->dev_addr.  The PACKET_OTHERHOST
will eventually cause the ip_rcv() errors out.  To fix this,
____dev_forward_skb() is added.

Joint work with Daniel Borkmann.

Fixes: cfc7381b3002 ("ip_tunnel: add collect_md mode to IPIP tunnel")
Fixes: 8d79266bc48c ("ip6_tunnel: add collect_md mode to IPv6 tunnels")
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@fb.com>
Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
diff --git a/net/core/dev.c b/net/core/dev.c
index eaad4c2..6666b28 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1766,19 +1766,14 @@
 
 int __dev_forward_skb(struct net_device *dev, struct sk_buff *skb)
 {
-	if (skb_orphan_frags(skb, GFP_ATOMIC) ||
-	    unlikely(!is_skb_forwardable(dev, skb))) {
-		atomic_long_inc(&dev->rx_dropped);
-		kfree_skb(skb);
-		return NET_RX_DROP;
+	int ret = ____dev_forward_skb(dev, skb);
+
+	if (likely(!ret)) {
+		skb->protocol = eth_type_trans(skb, dev);
+		skb_postpull_rcsum(skb, eth_hdr(skb), ETH_HLEN);
 	}
 
-	skb_scrub_packet(skb, true);
-	skb->priority = 0;
-	skb->protocol = eth_type_trans(skb, dev);
-	skb_postpull_rcsum(skb, eth_hdr(skb), ETH_HLEN);
-
-	return 0;
+	return ret;
 }
 EXPORT_SYMBOL_GPL(__dev_forward_skb);
 
diff --git a/net/core/filter.c b/net/core/filter.c
index 00351cd..b391209 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -1628,6 +1628,19 @@
 	return dev_forward_skb(dev, skb);
 }
 
+static inline int __bpf_rx_skb_no_mac(struct net_device *dev,
+				      struct sk_buff *skb)
+{
+	int ret = ____dev_forward_skb(dev, skb);
+
+	if (likely(!ret)) {
+		skb->dev = dev;
+		ret = netif_rx(skb);
+	}
+
+	return ret;
+}
+
 static inline int __bpf_tx_skb(struct net_device *dev, struct sk_buff *skb)
 {
 	int ret;
@@ -1647,6 +1660,51 @@
 	return ret;
 }
 
+static int __bpf_redirect_no_mac(struct sk_buff *skb, struct net_device *dev,
+				 u32 flags)
+{
+	/* skb->mac_len is not set on normal egress */
+	unsigned int mlen = skb->network_header - skb->mac_header;
+
+	__skb_pull(skb, mlen);
+
+	/* At ingress, the mac header has already been pulled once.
+	 * At egress, skb_pospull_rcsum has to be done in case that
+	 * the skb is originated from ingress (i.e. a forwarded skb)
+	 * to ensure that rcsum starts at net header.
+	 */
+	if (!skb_at_tc_ingress(skb))
+		skb_postpull_rcsum(skb, skb_mac_header(skb), mlen);
+	skb_pop_mac_header(skb);
+	skb_reset_mac_len(skb);
+	return flags & BPF_F_INGRESS ?
+	       __bpf_rx_skb_no_mac(dev, skb) : __bpf_tx_skb(dev, skb);
+}
+
+static int __bpf_redirect_common(struct sk_buff *skb, struct net_device *dev,
+				 u32 flags)
+{
+	bpf_push_mac_rcsum(skb);
+	return flags & BPF_F_INGRESS ?
+	       __bpf_rx_skb(dev, skb) : __bpf_tx_skb(dev, skb);
+}
+
+static int __bpf_redirect(struct sk_buff *skb, struct net_device *dev,
+			  u32 flags)
+{
+	switch (dev->type) {
+	case ARPHRD_TUNNEL:
+	case ARPHRD_TUNNEL6:
+	case ARPHRD_SIT:
+	case ARPHRD_IPGRE:
+	case ARPHRD_VOID:
+	case ARPHRD_NONE:
+		return __bpf_redirect_no_mac(skb, dev, flags);
+	default:
+		return __bpf_redirect_common(skb, dev, flags);
+	}
+}
+
 BPF_CALL_3(bpf_clone_redirect, struct sk_buff *, skb, u32, ifindex, u64, flags)
 {
 	struct net_device *dev;
@@ -1675,10 +1733,7 @@
 		return -ENOMEM;
 	}
 
-	bpf_push_mac_rcsum(clone);
-
-	return flags & BPF_F_INGRESS ?
-	       __bpf_rx_skb(dev, clone) : __bpf_tx_skb(dev, clone);
+	return __bpf_redirect(clone, dev, flags);
 }
 
 static const struct bpf_func_proto bpf_clone_redirect_proto = {
@@ -1722,10 +1777,7 @@
 		return -EINVAL;
 	}
 
-	bpf_push_mac_rcsum(skb);
-
-	return ri->flags & BPF_F_INGRESS ?
-	       __bpf_rx_skb(dev, skb) : __bpf_tx_skb(dev, skb);
+	return __bpf_redirect(skb, dev, ri->flags);
 }
 
 static const struct bpf_func_proto bpf_redirect_proto = {