Thread (12 messages) 12 messages, 3 authors, 2019-02-12

Re: [PATCH bpf-next v9 5/7] bpf: add handling of BPF_LWT_REROUTE to lwt_bpf.c

From: David Ahern <hidden>
Date: 2019-02-12 04:40:05

On 2/11/19 5:42 PM, Peter Oskolkov wrote:
quoted hunk ↗ jump to hunk
@@ -88,6 +90,35 @@ static int run_lwt_bpf(struct sk_buff *skb, struct bpf_lwt_prog *lwt,
 	return ret;
 }
 
+static int bpf_lwt_input_reroute(struct sk_buff *skb)
+{
+	int err = -EINVAL;
+
+	if (skb->protocol == htons(ETH_P_IP)) {
+		struct iphdr *iph = ip_hdr(skb);
+
+		err = ip_route_input_noref(skb, iph->daddr, iph->saddr,
+					   iph->tos, skb_dst(skb)->dev);
+	} else if (skb->protocol == htons(ETH_P_IPV6)) {
+#if IS_ENABLED(CONFIG_IPV6)
+		err = ipv6_stub->ipv6_route_input(skb);
+#else
+		pr_warn_once("BPF_LWT_REROUTE input: IPV6 not available\n");
+#endif
The stub defines ipv6_route_input when IPv6 is not enabled.
addrconf_core.o is compiled in if NET is enabled irregardless of
CONFIG_IPV6. ie., you don't need the IS_ENABLED check here.

If a bpf program pushes a v6 header the stub returns -EAFNOSUPPORT based
on patch 4.

+	} else {
+		pr_warn_once("BPF_LWT_REROUTE input: unsupported proto %d\n",
+			     skb->protocol);
You don't need a warn on here; just return -EAFNOSUPPORT.

quoted hunk ↗ jump to hunk
+	}
+
+	if (err)
+		goto err;
+	return dst_input(skb);
+
+err:
+	kfree_skb(skb);
+	return err;
+}
+
 static int bpf_input(struct sk_buff *skb)
 {
 	struct dst_entry *dst = skb_dst(skb);
@@ -99,6 +130,8 @@ static int bpf_input(struct sk_buff *skb)
 		ret = run_lwt_bpf(skb, &bpf->in, dst, NO_REDIRECT);
 		if (ret < 0)
 			return ret;
+		if (ret == BPF_LWT_REROUTE)
+			return bpf_lwt_input_reroute(skb);
 	}
 
 	if (unlikely(!dst->lwtstate->orig_input)) {
@@ -148,6 +181,95 @@ static int xmit_check_hhlen(struct sk_buff *skb)
 	return 0;
 }
 
+static int bpf_lwt_xmit_reroute(struct sk_buff *skb)
+{
+	struct net_device *l3mdev = l3mdev_master_dev_rcu(skb_dst(skb)->dev);
+	int oif = l3mdev ? l3mdev->ifindex : 0;
+	struct dst_entry *dst = NULL;
+	struct sock *sk;
+	struct net *net;
+	bool ipv4;
+	int err;
+
+	if (skb->protocol == htons(ETH_P_IP)) {
+		ipv4 = true;
+	} else if (skb->protocol == htons(ETH_P_IPV6)) {
+		ipv4 = false;
+	} else {
+		pr_warn_once("BPF_LWT_REROUTE xmit: unsupported proto %d\n",
+			     skb->protocol);
same here - no warn on.
+		return -EINVAL;
+	}
+
+	sk = sk_to_full_sk(skb->sk);
+	if (sk) {
+		if (sk->sk_bound_dev_if)
+			oif = sk->sk_bound_dev_if;
+		net = sock_net(sk);
+	} else {
+		net = dev_net(skb_dst(skb)->dev);
+	}
+
+	if (ipv4) {
+		struct iphdr *iph = ip_hdr(skb);
+		struct flowi4 fl4 = {};
+		struct rtable *rt;
+
+		fl4.flowi4_oif = oif;
+		fl4.flowi4_mark = skb->mark;
+		fl4.flowi4_uid = sock_net_uid(net, sk);
+		fl4.flowi4_tos = RT_TOS(iph->tos);
+		fl4.flowi4_flags = FLOWI_FLAG_ANYSRC;
+		fl4.flowi4_proto = iph->protocol;
+		fl4.daddr = iph->daddr;
+		fl4.saddr = iph->saddr;
+
+		rt = ip_route_output_key(net, &fl4);
+		if (IS_ERR(rt) || rt->dst.error)
+			return -EINVAL;
+		dst = &rt->dst;
+	} else {
+#if IS_ENABLED(CONFIG_IPV6)
+		struct ipv6hdr *iph6 = ipv6_hdr(skb);
+		struct flowi6 fl6 = {};
+
+		fl6.flowi6_oif = oif;
+		fl6.flowi6_mark = skb->mark;
+		fl6.flowi6_uid = sock_net_uid(net, sk);
+		fl6.flowlabel = ip6_flowinfo(iph6);
+		fl6.flowi6_proto = iph6->nexthdr;
+		fl6.daddr = iph6->daddr;
+		fl6.saddr = iph6->saddr;
+
+		err = ipv6_stub->ipv6_dst_lookup(net, skb->sk, &dst, &fl6);
+		if (err || IS_ERR(dst) || dst->error)
+			return -EINVAL;
+#else
+		pr_warn_once("BPF_LWT_REROUTE xmit: IPV6 not available\n");
+		return -EINVAL;
+#endif
No #if .. #endif needed. The stub handles it.

Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help