neighbour: Split pneigh_lookup().

pneigh_lookup() has ASSERT_RTNL() in the middle of the function, which
is confusing.

When called with the last argument, creat, 0, pneigh_lookup() literally
looks up a proxy neighbour entry.  This is the case of the reader path
as the fast path and RTM_GETNEIGH.

pneigh_lookup(), however, creates a pneigh_entry when called with creat 1
from RTM_NEWNEIGH and SIOCSARP, which require RTNL.

Let's split pneigh_lookup() into two functions.

We will convert all the reader paths to RCU, and read_lock_bh(&tbl->lock)
in the new pneigh_lookup() will be dropped.

Signed-off-by: Kuniyuki Iwashima <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>

Author: q2ven

include/net/neighbour.h

@@ -376,10 +376,11 @@ unsigned long neigh_rand_reach_time(unsigned long base);
 void pneigh_enqueue(struct neigh_table *tbl, struct neigh_parms *p,
 		    struct sk_buff *skb);
 struct pneigh_entry *pneigh_lookup(struct neigh_table *tbl, struct net *net,
-				   const void *key, struct net_device *dev,
-				   int creat);
+				   const void *key, struct net_device *dev);
 struct pneigh_entry *__pneigh_lookup(struct neigh_table *tbl, struct net *net,
 				     const void *key, struct net_device *dev);
+struct pneigh_entry *pneigh_create(struct neigh_table *tbl, struct net *net,
+				   const void *key, struct net_device *dev);
 int pneigh_delete(struct neigh_table *tbl, struct net *net, const void *key,
 		  struct net_device *dev);
 

net/core/neighbour.c

@@ -28,6 +28,7 @@
 #include <net/neighbour.h>
 #include <net/arp.h>
 #include <net/dst.h>
+#include <net/ip.h>
 #include <net/sock.h>
 #include <net/netevent.h>
 #include <net/netlink.h>
@@ -746,24 +747,44 @@ struct pneigh_entry *__pneigh_lookup(struct neigh_table *tbl,
 }
 EXPORT_SYMBOL_GPL(__pneigh_lookup);
 
-struct pneigh_entry * pneigh_lookup(struct neigh_table *tbl,
-				    struct net *net, const void *pkey,
-				    struct net_device *dev, int creat)
+struct pneigh_entry *pneigh_lookup(struct neigh_table *tbl,
+				   struct net *net, const void *pkey,
+				   struct net_device *dev)
+{
+	struct pneigh_entry *n;
+	unsigned int key_len;
+	u32 hash_val;
+
+	key_len = tbl->key_len;
+	hash_val = pneigh_hash(pkey, key_len);
+
+	read_lock_bh(&tbl->lock);
+	n = __pneigh_lookup_1(tbl->phash_buckets[hash_val],
+			      net, pkey, key_len, dev);
+	read_unlock_bh(&tbl->lock);
+
+	return n;
+}
+EXPORT_IPV6_MOD(pneigh_lookup);
+
+struct pneigh_entry *pneigh_create(struct neigh_table *tbl,
+				   struct net *net, const void *pkey,
+				   struct net_device *dev)
 {
 	struct pneigh_entry *n;
 	unsigned int key_len = tbl->key_len;
 	u32 hash_val = pneigh_hash(pkey, key_len);
 
+	ASSERT_RTNL();
+
 	read_lock_bh(&tbl->lock);
 	n = __pneigh_lookup_1(tbl->phash_buckets[hash_val],
 			      net, pkey, key_len, dev);
 	read_unlock_bh(&tbl->lock);
 
-	if (n || !creat)
+	if (n)
 		goto out;
 
-	ASSERT_RTNL();
-
 	n = kzalloc(sizeof(*n) + key_len, GFP_KERNEL);
 	if (!n)
 		goto out;
@@ -787,8 +808,6 @@ struct pneigh_entry * pneigh_lookup(struct neigh_table *tbl,
 out:
 	return n;
 }
-EXPORT_SYMBOL(pneigh_lookup);
-
 
 int pneigh_delete(struct neigh_table *tbl, struct net *net, const void *pkey,
 		  struct net_device *dev)
@@ -2007,7 +2026,7 @@ static int neigh_add(struct sk_buff *skb, struct nlmsghdr *nlh,
 		}
 
 		err = -ENOBUFS;
-		pn = pneigh_lookup(tbl, net, dst, dev, 1);
+		pn = pneigh_create(tbl, net, dst, dev);
 		if (pn) {
 			pn->flags = ndm_flags;
 			pn->permanent = !!(ndm->ndm_state & NUD_PERMANENT);
@@ -3036,7 +3055,7 @@ static int neigh_get(struct sk_buff *in_skb, struct nlmsghdr *nlh,
 	if (ndm->ndm_flags & NTF_PROXY) {
 		struct pneigh_entry *pn;
 
-		pn = pneigh_lookup(tbl, net, dst, dev, 0);
+		pn = pneigh_lookup(tbl, net, dst, dev);
 		if (!pn) {
 			NL_SET_ERR_MSG(extack, "Proxy neighbour entry not found");
 			err = -ENOENT;

net/ipv4/arp.c

@@ -864,7 +864,7 @@ static int arp_process(struct net *net, struct sock *sk, struct sk_buff *skb)
 			    (arp_fwd_proxy(in_dev, dev, rt) ||
 			     arp_fwd_pvlan(in_dev, dev, rt, sip, tip) ||
 			     (rt->dst.dev != dev &&
-			      pneigh_lookup(&arp_tbl, net, &tip, dev, 0)))) {
+			      pneigh_lookup(&arp_tbl, net, &tip, dev)))) {
 				n = neigh_event_ns(&arp_tbl, sha, &sip, dev);
 				if (n)
 					neigh_release(n);
@@ -1089,7 +1089,7 @@ static int arp_req_set_public(struct net *net, struct arpreq *r,
 	if (mask) {
 		__be32 ip = ((struct sockaddr_in *)&r->arp_pa)->sin_addr.s_addr;
 
-		if (!pneigh_lookup(&arp_tbl, net, &ip, dev, 1))
+		if (!pneigh_create(&arp_tbl, net, &ip, dev))
 			return -ENOBUFS;
 		return 0;
 	}

net/ipv6/ip6_output.c

@@ -563,7 +563,7 @@ int ip6_forward(struct sk_buff *skb)
 
 	/* XXX: idev->cnf.proxy_ndp? */
 	if (READ_ONCE(net->ipv6.devconf_all->proxy_ndp) &&
-	    pneigh_lookup(&nd_tbl, net, &hdr->daddr, skb->dev, 0)) {
+	    pneigh_lookup(&nd_tbl, net, &hdr->daddr, skb->dev)) {
 		int proxied = ip6_forward_proxy_check(skb);
 		if (proxied > 0) {
 			/* It's tempting to decrease the hop limit

net/ipv6/ndisc.c

@@ -1100,7 +1100,7 @@ static enum skb_drop_reason ndisc_recv_na(struct sk_buff *skb)
 		if (lladdr && !memcmp(lladdr, dev->dev_addr, dev->addr_len) &&
 		    READ_ONCE(net->ipv6.devconf_all->forwarding) &&
 		    READ_ONCE(net->ipv6.devconf_all->proxy_ndp) &&
-		    pneigh_lookup(&nd_tbl, net, &msg->target, dev, 0)) {
+		    pneigh_lookup(&nd_tbl, net, &msg->target, dev)) {
 			/* XXX: idev->cnf.proxy_ndp */
 			goto out;
 		}