[PATCH bpf-next v4 7/9] net: ip: make ip_route_input() return drop reasons

Menglong Dong posted 9 patches 1 month ago
There is a newer version of this series
[PATCH bpf-next v4 7/9] net: ip: make ip_route_input() return drop reasons
Posted by Menglong Dong 1 month ago
In this commit, we make ip_route_input() return skb drop reasons that come
from ip_route_input_noref().

Meanwhile, adjust all the call to it.

Signed-off-by: Menglong Dong <dongml2@chinatelecom.cn>
---
v4:
- replace the variable "err" with "reason" for the return value of
  ip_route_input()
---
 include/net/route.h             |  7 ++++---
 net/bridge/br_netfilter_hooks.c | 11 ++++++-----
 net/ipv4/icmp.c                 |  2 +-
 net/ipv4/ip_options.c           |  2 +-
 net/ipv6/seg6_local.c           | 14 +++++++-------
 5 files changed, 19 insertions(+), 17 deletions(-)

diff --git a/include/net/route.h b/include/net/route.h
index 11674f7c6be6..f4ab5412c9c9 100644
--- a/include/net/route.h
+++ b/include/net/route.h
@@ -210,8 +210,9 @@ int ip_route_use_hint(struct sk_buff *skb, __be32 daddr, __be32 saddr,
 		      dscp_t dscp, struct net_device *dev,
 		      const struct sk_buff *hint);
 
-static inline int ip_route_input(struct sk_buff *skb, __be32 dst, __be32 src,
-				 dscp_t dscp, struct net_device *devin)
+static inline enum skb_drop_reason
+ip_route_input(struct sk_buff *skb, __be32 dst, __be32 src, dscp_t dscp,
+	       struct net_device *devin)
 {
 	enum skb_drop_reason reason;
 
@@ -224,7 +225,7 @@ static inline int ip_route_input(struct sk_buff *skb, __be32 dst, __be32 src,
 	}
 	rcu_read_unlock();
 
-	return reason ? -EINVAL : 0;
+	return reason;
 }
 
 void ipv4_update_pmtu(struct sk_buff *skb, struct net *net, u32 mtu, int oif,
diff --git a/net/bridge/br_netfilter_hooks.c b/net/bridge/br_netfilter_hooks.c
index 17a5f5923d61..110cffc24a1d 100644
--- a/net/bridge/br_netfilter_hooks.c
+++ b/net/bridge/br_netfilter_hooks.c
@@ -373,8 +373,8 @@ static int br_nf_pre_routing_finish(struct net *net, struct sock *sk, struct sk_
 	struct nf_bridge_info *nf_bridge = nf_bridge_info_get(skb);
 	struct net_device *dev = skb->dev, *br_indev;
 	const struct iphdr *iph = ip_hdr(skb);
+	enum skb_drop_reason reason;
 	struct rtable *rt;
-	int err;
 
 	br_indev = nf_bridge_get_physindev(skb, net);
 	if (!br_indev) {
@@ -390,9 +390,9 @@ static int br_nf_pre_routing_finish(struct net *net, struct sock *sk, struct sk_
 	}
 	nf_bridge->in_prerouting = 0;
 	if (br_nf_ipv4_daddr_was_changed(skb, nf_bridge)) {
-		err = ip_route_input(skb, iph->daddr, iph->saddr,
-				     ip4h_dscp(iph), dev);
-		if (err) {
+		reason = ip_route_input(skb, iph->daddr, iph->saddr,
+					ip4h_dscp(iph), dev);
+		if (reason) {
 			struct in_device *in_dev = __in_dev_get_rcu(dev);
 
 			/* If err equals -EHOSTUNREACH the error is due to a
@@ -402,7 +402,8 @@ static int br_nf_pre_routing_finish(struct net *net, struct sock *sk, struct sk_
 			 * martian destinations: loopback destinations and destination
 			 * 0.0.0.0. In both cases the packet will be dropped because the
 			 * destination is the loopback device and not the bridge. */
-			if (err != -EHOSTUNREACH || !in_dev || IN_DEV_FORWARD(in_dev))
+			if (reason != SKB_DROP_REASON_IP_INADDRERRORS || !in_dev ||
+			    IN_DEV_FORWARD(in_dev))
 				goto free_skb;
 
 			rt = ip_route_output(net, iph->daddr, 0,
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index 23664434922e..69ea33f64a54 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -545,7 +545,7 @@ static struct rtable *icmp_route_lookup(struct net *net, struct flowi4 *fl4,
 		orefdst = skb_in->_skb_refdst; /* save old refdst */
 		skb_dst_set(skb_in, NULL);
 		err = ip_route_input(skb_in, fl4_dec.daddr, fl4_dec.saddr,
-				     dscp, rt2->dst.dev);
+				     dscp, rt2->dst.dev) ? -EINVAL : 0;
 
 		dst_release(&rt2->dst);
 		rt2 = skb_rtable(skb_in);
diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c
index 81e86e5defee..e3321932bec0 100644
--- a/net/ipv4/ip_options.c
+++ b/net/ipv4/ip_options.c
@@ -618,7 +618,7 @@ int ip_options_rcv_srr(struct sk_buff *skb, struct net_device *dev)
 		orefdst = skb->_skb_refdst;
 		skb_dst_set(skb, NULL);
 		err = ip_route_input(skb, nexthop, iph->saddr, ip4h_dscp(iph),
-				     dev);
+				     dev) ? -EINVAL : 0;
 		rt2 = skb_rtable(skb);
 		if (err || (rt2->rt_type != RTN_UNICAST && rt2->rt_type != RTN_LOCAL)) {
 			skb_dst_drop(skb);
diff --git a/net/ipv6/seg6_local.c b/net/ipv6/seg6_local.c
index c74705ead984..ac1dbd492c22 100644
--- a/net/ipv6/seg6_local.c
+++ b/net/ipv6/seg6_local.c
@@ -954,10 +954,10 @@ static int input_action_end_dx4_finish(struct net *net, struct sock *sk,
 				       struct sk_buff *skb)
 {
 	struct dst_entry *orig_dst = skb_dst(skb);
+	enum skb_drop_reason reason;
 	struct seg6_local_lwt *slwt;
 	struct iphdr *iph;
 	__be32 nhaddr;
-	int err;
 
 	slwt = seg6_local_lwtunnel(orig_dst->lwtstate);
 
@@ -967,9 +967,9 @@ static int input_action_end_dx4_finish(struct net *net, struct sock *sk,
 
 	skb_dst_drop(skb);
 
-	err = ip_route_input(skb, nhaddr, iph->saddr, 0, skb->dev);
-	if (err) {
-		kfree_skb(skb);
+	reason = ip_route_input(skb, nhaddr, iph->saddr, 0, skb->dev);
+	if (reason) {
+		kfree_skb_reason(skb, reason);
 		return -EINVAL;
 	}
 
@@ -1174,8 +1174,8 @@ static struct sk_buff *end_dt_vrf_core(struct sk_buff *skb,
 static int input_action_end_dt4(struct sk_buff *skb,
 				struct seg6_local_lwt *slwt)
 {
+	enum skb_drop_reason reason;
 	struct iphdr *iph;
-	int err;
 
 	if (!decap_and_validate(skb, IPPROTO_IPIP))
 		goto drop;
@@ -1193,8 +1193,8 @@ static int input_action_end_dt4(struct sk_buff *skb,
 
 	iph = ip_hdr(skb);
 
-	err = ip_route_input(skb, iph->daddr, iph->saddr, 0, skb->dev);
-	if (unlikely(err))
+	reason = ip_route_input(skb, iph->daddr, iph->saddr, 0, skb->dev);
+	if (unlikely(reason))
 		goto drop;
 
 	return dst_input(skb);
-- 
2.39.5