1 files changed, 96 insertions, 78 deletions

diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index 95e62235b5..74e6f9746f 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -483,24 +483,15 @@ EXPORT_SYMBOL_GPL(__rtnl_link_unregister);
  */
 static void rtnl_lock_unregistering_all(void)
 {
-	struct net *net;
-	bool unregistering;
 	DEFINE_WAIT_FUNC(wait, woken_wake_function);
 
 	add_wait_queue(&netdev_unregistering_wq, &wait);
 	for (;;) {
-		unregistering = false;
 		rtnl_lock();
 		/* We held write locked pernet_ops_rwsem, and parallel
 		 * setup_net() and cleanup_net() are not possible.
 		 */
-		for_each_net(net) {
-			if (atomic_read(&net->dev_unreg_count) > 0) {
-				unregistering = true;
-				break;
-			}
-		}
-		if (!unregistering)
+		if (!atomic_read(&dev_unreg_count))
 			break;
 		__rtnl_unlock();
 
@@ -851,9 +842,22 @@ int rtnl_put_cacheinfo(struct sk_buff *skb, struct dst_entry *dst, u32 id,
 }
 EXPORT_SYMBOL_GPL(rtnl_put_cacheinfo);
 
+void netdev_set_operstate(struct net_device *dev, int newstate)
+{
+	unsigned int old = READ_ONCE(dev->operstate);
+
+	do {
+		if (old == newstate)
+			return;
+	} while (!try_cmpxchg(&dev->operstate, &old, newstate));
+
+	netdev_state_change(dev);
+}
+EXPORT_SYMBOL(netdev_set_operstate);
+
 static void set_operstate(struct net_device *dev, unsigned char transition)
 {
-	unsigned char operstate = dev->operstate;
+	unsigned char operstate = READ_ONCE(dev->operstate);
 
 	switch (transition) {
 	case IF_OPER_UP:
@@ -875,12 +879,7 @@ static void set_operstate(struct net_device *dev, unsigned char transition)
 		break;
 	}
 
-	if (READ_ONCE(dev->operstate) != operstate) {
-		write_lock(&dev_base_lock);
-		WRITE_ONCE(dev->operstate, operstate);
-		write_unlock(&dev_base_lock);
-		netdev_state_change(dev);
-	}
+	netdev_set_operstate(dev, operstate);
 }
 
 static unsigned int rtnl_dev_get_flags(const struct net_device *dev)
@@ -1456,17 +1455,18 @@ static noinline_for_stack int rtnl_fill_vf(struct sk_buff *skb,
 	return 0;
 }
 
-static int rtnl_fill_link_ifmap(struct sk_buff *skb, struct net_device *dev)
+static int rtnl_fill_link_ifmap(struct sk_buff *skb,
+				const struct net_device *dev)
 {
 	struct rtnl_link_ifmap map;
 
 	memset(&map, 0, sizeof(map));
-	map.mem_start   = dev->mem_start;
-	map.mem_end     = dev->mem_end;
-	map.base_addr   = dev->base_addr;
-	map.irq         = dev->irq;
-	map.dma         = dev->dma;
-	map.port        = dev->if_port;
+	map.mem_start = READ_ONCE(dev->mem_start);
+	map.mem_end   = READ_ONCE(dev->mem_end);
+	map.base_addr = READ_ONCE(dev->base_addr);
+	map.irq       = READ_ONCE(dev->irq);
+	map.dma       = READ_ONCE(dev->dma);
+	map.port      = READ_ONCE(dev->if_port);
 
 	if (nla_put_64bit(skb, IFLA_MAP, sizeof(map), &map, IFLA_PAD))
 		return -EMSGSIZE;
@@ -1612,10 +1612,10 @@ static int put_master_ifindex(struct sk_buff *skb, struct net_device *dev)
 static int nla_put_iflink(struct sk_buff *skb, const struct net_device *dev,
 			  bool force)
 {
-	int ifindex = dev_get_iflink(dev);
+	int iflink = dev_get_iflink(dev);
 
-	if (force || dev->ifindex != ifindex)
-		return nla_put_u32(skb, IFLA_LINK, ifindex);
+	if (force || READ_ONCE(dev->ifindex) != iflink)
+		return nla_put_u32(skb, IFLA_LINK, iflink);
 
 	return 0;
 }
@@ -1699,7 +1699,7 @@ static int rtnl_fill_alt_ifnames(struct sk_buff *skb,
 	struct netdev_name_node *name_node;
 	int count = 0;
 
-	list_for_each_entry(name_node, &dev->name_node->list, list) {
+	list_for_each_entry_rcu(name_node, &dev->name_node->list, list) {
 		if (nla_put_string(skb, IFLA_ALT_IFNAME, name_node->name))
 			return -EMSGSIZE;
 		count++;
@@ -1707,6 +1707,7 @@ static int rtnl_fill_alt_ifnames(struct sk_buff *skb,
 	return count;
 }
 
+/* RCU protected. */
 static int rtnl_fill_prop_list(struct sk_buff *skb,
 			       const struct net_device *dev)
 {
@@ -1876,9 +1877,6 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb,
 			goto nla_put_failure;
 	}
 
-	if (rtnl_fill_link_ifmap(skb, dev))
-		goto nla_put_failure;
-
 	if (dev->addr_len) {
 		if (nla_put(skb, IFLA_ADDRESS, dev->addr_len, dev->dev_addr) ||
 		    nla_put(skb, IFLA_BROADCAST, dev->addr_len, dev->broadcast))
@@ -1928,10 +1926,11 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb,
 	rcu_read_lock();
 	if (rtnl_fill_link_af(skb, dev, ext_filter_mask))
 		goto nla_put_failure_rcu;
-	rcu_read_unlock();
-
+	if (rtnl_fill_link_ifmap(skb, dev))
+		goto nla_put_failure_rcu;
 	if (rtnl_fill_prop_list(skb, dev))
-		goto nla_put_failure;
+		goto nla_put_failure_rcu;
+	rcu_read_unlock();
 
 	if (dev->dev.parent &&
 	    nla_put_string(skb, IFLA_PARENT_DEV_NAME,
@@ -2200,25 +2199,22 @@ static int rtnl_valid_dump_ifinfo_req(const struct nlmsghdr *nlh,
 
 static int rtnl_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb)
 {
+	const struct rtnl_link_ops *kind_ops = NULL;
 	struct netlink_ext_ack *extack = cb->extack;
 	const struct nlmsghdr *nlh = cb->nlh;
 	struct net *net = sock_net(skb->sk);
-	struct net *tgt_net = net;
-	int h, s_h;
-	int idx = 0, s_idx;
-	struct net_device *dev;
-	struct hlist_head *head;
+	unsigned int flags = NLM_F_MULTI;
 	struct nlattr *tb[IFLA_MAX+1];
+	struct {
+		unsigned long ifindex;
+	} *ctx = (void *)cb->ctx;
+	struct net *tgt_net = net;
 	u32 ext_filter_mask = 0;
-	const struct rtnl_link_ops *kind_ops = NULL;
-	unsigned int flags = NLM_F_MULTI;
+	struct net_device *dev;
 	int master_idx = 0;
 	int netnsid = -1;
 	int err, i;
 
-	s_h = cb->args[0];
-	s_idx = cb->args[1];
-
 	err = rtnl_valid_dump_ifinfo_req(nlh, cb->strict_check, tb, extack);
 	if (err < 0) {
 		if (cb->strict_check)
@@ -2262,36 +2258,18 @@ static int rtnl_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb)
 		flags |= NLM_F_DUMP_FILTERED;
 
 walk_entries:
-	for (h = s_h; h < NETDEV_HASHENTRIES; h++, s_idx = 0) {
-		idx = 0;
-		head = &tgt_net->dev_index_head[h];
-		hlist_for_each_entry(dev, head, index_hlist) {
-			if (link_dump_filtered(dev, master_idx, kind_ops))
-				goto cont;
-			if (idx < s_idx)
-				goto cont;
-			err = rtnl_fill_ifinfo(skb, dev, net,
-					       RTM_NEWLINK,
-					       NETLINK_CB(cb->skb).portid,
-					       nlh->nlmsg_seq, 0, flags,
-					       ext_filter_mask, 0, NULL, 0,
-					       netnsid, GFP_KERNEL);
-
-			if (err < 0) {
-				if (likely(skb->len))
-					goto out;
-
-				goto out_err;
-			}
-cont:
-			idx++;
-		}
+	err = 0;
+	for_each_netdev_dump(tgt_net, dev, ctx->ifindex) {
+		if (link_dump_filtered(dev, master_idx, kind_ops))
+			continue;
+		err = rtnl_fill_ifinfo(skb, dev, net, RTM_NEWLINK,
+				       NETLINK_CB(cb->skb).portid,
+				       nlh->nlmsg_seq, 0, flags,
+				       ext_filter_mask, 0, NULL, 0,
+				       netnsid, GFP_KERNEL);
+		if (err < 0)
+			break;
 	}
-out:
-	err = skb->len;
-out_err:
-	cb->args[1] = idx;
-	cb->args[0] = h;
 	cb->seq = tgt_net->dev_base_seq;
 	nl_dump_check_consistent(cb, nlmsg_hdr(skb));
 	if (netnsid >= 0)
@@ -2983,11 +2961,9 @@ static int do_setlink(const struct sk_buff *skb,
 	if (tb[IFLA_LINKMODE]) {
 		unsigned char value = nla_get_u8(tb[IFLA_LINKMODE]);
 
-		write_lock(&dev_base_lock);
 		if (dev->link_mode ^ value)
 			status |= DO_SETLINK_NOTIFY;
-		dev->link_mode = value;
-		write_unlock(&dev_base_lock);
+		WRITE_ONCE(dev->link_mode, value);
 	}
 
 	if (tb[IFLA_VFINFO_LIST]) {
@@ -6508,6 +6484,46 @@ static int rtnl_mdb_del(struct sk_buff *skb, struct nlmsghdr *nlh,
 
 /* Process one rtnetlink message. */
 
+static int rtnl_dumpit(struct sk_buff *skb, struct netlink_callback *cb)
+{
+	rtnl_dumpit_func dumpit = cb->data;
+	int err;
+
+	/* Previous iteration have already finished, avoid calling->dumpit()
+	 * again, it may not expect to be called after it reached the end.
+	 */
+	if (!dumpit)
+		return 0;
+
+	err = dumpit(skb, cb);
+
+	/* Old dump handlers used to send NLM_DONE as in a separate recvmsg().
+	 * Some applications which parse netlink manually depend on this.
+	 */
+	if (cb->flags & RTNL_FLAG_DUMP_SPLIT_NLM_DONE) {
+		if (err < 0 && err != -EMSGSIZE)
+			return err;
+		if (!err)
+			cb->data = NULL;
+
+		return skb->len;
+	}
+	return err;
+}
+
+static int rtnetlink_dump_start(struct sock *ssk, struct sk_buff *skb,
+				const struct nlmsghdr *nlh,
+				struct netlink_dump_control *control)
+{
+	if (control->flags & RTNL_FLAG_DUMP_SPLIT_NLM_DONE) {
+		WARN_ON(control->data);
+		control->data = control->dump;
+		control->dump = rtnl_dumpit;
+	}
+
+	return netlink_dump_start(ssk, skb, nlh, control);
+}
+
 static int rtnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh,
 			     struct netlink_ext_ack *extack)
 {
@@ -6552,6 +6568,7 @@ static int rtnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh,
 		}
 		owner = link->owner;
 		dumpit = link->dumpit;
+		flags = link->flags;
 
 		if (type == RTM_GETLINK - RTM_BASE)
 			min_dump_alloc = rtnl_calcit(skb, nlh);
@@ -6569,8 +6586,9 @@ static int rtnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh,
 				.dump		= dumpit,
 				.min_dump_alloc	= min_dump_alloc,
 				.module		= owner,
+				.flags		= flags,
 			};
-			err = netlink_dump_start(rtnl, skb, nlh, &c);
+			err = rtnetlink_dump_start(rtnl, skb, nlh, &c);
 			/* netlink_dump_start() will keep a reference on
 			 * module if dump is still in progress.
 			 */
@@ -6716,7 +6734,7 @@ void __init rtnetlink_init(void)
 	register_netdevice_notifier(&rtnetlink_dev_notifier);
 
 	rtnl_register(PF_UNSPEC, RTM_GETLINK, rtnl_getlink,
-		      rtnl_dump_ifinfo, 0);
+		      rtnl_dump_ifinfo, RTNL_FLAG_DUMP_SPLIT_NLM_DONE);
 	rtnl_register(PF_UNSPEC, RTM_SETLINK, rtnl_setlink, NULL, 0);
 	rtnl_register(PF_UNSPEC, RTM_NEWLINK, rtnl_newlink, NULL, 0);
 	rtnl_register(PF_UNSPEC, RTM_DELLINK, rtnl_dellink, NULL, 0);