dev.c

	NAPI_GRO_CB(skb)->data_offset = 0;
	NAPI_GRO_CB(skb)->frag0 = NULL;
	NAPI_GRO_CB(skb)->frag0_len = 0;

	if (skb->mac_header == skb->tail &&
	    pinfo->nr_frags &&
	    !PageHighMem(skb_frag_page(frag0))) {
		NAPI_GRO_CB(skb)->frag0 = skb_frag_address(frag0);
		NAPI_GRO_CB(skb)->frag0_len = skb_frag_size(frag0);
	}
}

gro_result_t napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb)
{
	skb_gro_reset_offset(skb);

	return napi_skb_finish(dev_gro_receive(napi, skb), skb);
}
EXPORT_SYMBOL(napi_gro_receive);

static void napi_reuse_skb(struct napi_struct *napi, struct sk_buff *skb)
{
	__skb_pull(skb, skb_headlen(skb));
	/* restore the reserve we had after netdev_alloc_skb_ip_align() */
	skb_reserve(skb, NET_SKB_PAD + NET_IP_ALIGN - skb_headroom(skb));
	skb->vlan_tci = 0;
	skb->dev = napi->dev;
	skb->skb_iif = 0;

	napi->skb = skb;
}

struct sk_buff *napi_get_frags(struct napi_struct *napi)
{
	struct sk_buff *skb = napi->skb;

	if (!skb) {
		skb = netdev_alloc_skb_ip_align(napi->dev, GRO_MAX_HEAD);
		if (skb)
			napi->skb = skb;
	}
	return skb;
}
EXPORT_SYMBOL(napi_get_frags);

static gro_result_t napi_frags_finish(struct napi_struct *napi, struct sk_buff *skb,
			       gro_result_t ret)
{
	switch (ret) {
	case GRO_NORMAL:
	case GRO_HELD:
		skb->protocol = eth_type_trans(skb, skb->dev);

		if (ret == GRO_HELD)
			skb_gro_pull(skb, -ETH_HLEN);
		else if (netif_receive_skb(skb))
			ret = GRO_DROP;
		break;

	case GRO_DROP:
	case GRO_MERGED_FREE:
		napi_reuse_skb(napi, skb);
		break;

	case GRO_MERGED:
		break;
	}

	return ret;
}

static struct sk_buff *napi_frags_skb(struct napi_struct *napi)
{
	struct sk_buff *skb = napi->skb;
	struct ethhdr *eth;
	unsigned int hlen;
	unsigned int off;

	napi->skb = NULL;

	skb_reset_mac_header(skb);
	skb_gro_reset_offset(skb);

	off = skb_gro_offset(skb);
	hlen = off + sizeof(*eth);
	eth = skb_gro_header_fast(skb, off);
	if (skb_gro_header_hard(skb, hlen)) {
		eth = skb_gro_header_slow(skb, hlen, off);
		if (unlikely(!eth)) {
			napi_reuse_skb(napi, skb);
			skb = NULL;
			goto out;
		}
	}

	skb_gro_pull(skb, sizeof(*eth));

	/*
	 * This works because the only protocols we care about don't require
	 * special handling.  We'll fix it up properly at the end.
	 */
	skb->protocol = eth->h_proto;

out:
	return skb;
}

gro_result_t napi_gro_frags(struct napi_struct *napi)
{
	struct sk_buff *skb = napi_frags_skb(napi);

	if (!skb)
		return GRO_DROP;

	return napi_frags_finish(napi, skb, dev_gro_receive(napi, skb));
}
EXPORT_SYMBOL(napi_gro_frags);

/*
 * net_rps_action sends any pending IPI's for rps.
 * Note: called with local irq disabled, but exits with local irq enabled.
 */
static void net_rps_action_and_irq_enable(struct softnet_data *sd)
{
#ifdef CONFIG_RPS
	struct softnet_data *remsd = sd->rps_ipi_list;

	if (remsd) {
		sd->rps_ipi_list = NULL;

		local_irq_enable();

		/* Send pending IPI's to kick RPS processing on remote cpus. */
		while (remsd) {
			struct softnet_data *next = remsd->rps_ipi_next;

			if (cpu_online(remsd->cpu))
				__smp_call_function_single(remsd->cpu,
							   &remsd->csd, 0);
			remsd = next;
		}
	} else
#endif
		local_irq_enable();
}

static int process_backlog(struct napi_struct *napi, int quota)
{
	int work = 0;
	struct softnet_data *sd = container_of(napi, struct softnet_data, backlog);

#ifdef CONFIG_RPS
	/* Check if we have pending ipi, its better to send them now,
	 * not waiting net_rx_action() end.
	 */
	if (sd->rps_ipi_list) {
		local_irq_disable();
		net_rps_action_and_irq_enable(sd);
	}
#endif
	napi->weight = weight_p;
	local_irq_disable();
	while (work < quota) {
		struct sk_buff *skb;
		unsigned int qlen;

		while ((skb = __skb_dequeue(&sd->process_queue))) {
			local_irq_enable();
			__netif_receive_skb(skb);
			local_irq_disable();
			input_queue_head_incr(sd);
			if (++work >= quota) {
				local_irq_enable();
				return work;
			}
		}

		rps_lock(sd);
		qlen = skb_queue_len(&sd->input_pkt_queue);
		if (qlen)
			skb_queue_splice_tail_init(&sd->input_pkt_queue,
						   &sd->process_queue);

		if (qlen < quota - work) {
			/*
			 * Inline a custom version of __napi_complete().
			 * only current cpu owns and manipulates this napi,
			 * and NAPI_STATE_SCHED is the only possible flag set on backlog.
			 * we can use a plain write instead of clear_bit(),
			 * and we dont need an smp_mb() memory barrier.
			 */
			list_del(&napi->poll_list);
			napi->state = 0;

			quota = work + qlen;
		}
		rps_unlock(sd);
	}
	local_irq_enable();

	return work;
}

/**
 * __napi_schedule - schedule for receive
 * @n: entry to schedule
 *
 * The entry's receive function will be scheduled to run
 */
void __napi_schedule(struct napi_struct *n)
{
	unsigned long flags;

	local_irq_save(flags);
	____napi_schedule(&__get_cpu_var(softnet_data), n);
	local_irq_restore(flags);
}
EXPORT_SYMBOL(__napi_schedule);

void __napi_complete(struct napi_struct *n)
{
	BUG_ON(!test_bit(NAPI_STATE_SCHED, &n->state));
	BUG_ON(n->gro_list);

	list_del(&n->poll_list);
	smp_mb__before_clear_bit();
	clear_bit(NAPI_STATE_SCHED, &n->state);
}
EXPORT_SYMBOL(__napi_complete);

void napi_complete(struct napi_struct *n)
{
	unsigned long flags;

	/*
	 * don't let napi dequeue from the cpu poll list
	 * just in case its running on a different cpu
	 */
	if (unlikely(test_bit(NAPI_STATE_NPSVC, &n->state)))
		return;

	napi_gro_flush(n, false);
	local_irq_save(flags);
	__napi_complete(n);
	local_irq_restore(flags);
}
EXPORT_SYMBOL(napi_complete);

void netif_napi_add(struct net_device *dev, struct napi_struct *napi,
		    int (*poll)(struct napi_struct *, int), int weight)
{
	INIT_LIST_HEAD(&napi->poll_list);
	napi->gro_count = 0;
	napi->gro_list = NULL;
	napi->skb = NULL;
	napi->poll = poll;
	napi->weight = weight;
	list_add(&napi->dev_list, &dev->napi_list);
	napi->dev = dev;
#ifdef CONFIG_NETPOLL
	spin_lock_init(&napi->poll_lock);
	napi->poll_owner = -1;
#endif
	set_bit(NAPI_STATE_SCHED, &napi->state);
}
EXPORT_SYMBOL(netif_napi_add);

void netif_napi_del(struct napi_struct *napi)
{
	struct sk_buff *skb, *next;

	list_del_init(&napi->dev_list);
	napi_free_frags(napi);

	for (skb = napi->gro_list; skb; skb = next) {
		next = skb->next;
		skb->next = NULL;
		kfree_skb(skb);
	}

	napi->gro_list = NULL;
	napi->gro_count = 0;
}
EXPORT_SYMBOL(netif_napi_del);

static void net_rx_action(struct softirq_action *h)
{
	struct softnet_data *sd = &__get_cpu_var(softnet_data);
	unsigned long time_limit = jiffies + 2;
	int budget = netdev_budget;
	void *have;

	local_irq_disable();

	while (!list_empty(&sd->poll_list)) {
		struct napi_struct *n;
		int work, weight;

		/* If softirq window is exhuasted then punt.
		 * Allow this to run for 2 jiffies since which will allow
		 * an average latency of 1.5/HZ.
		 */
		if (unlikely(budget <= 0 || time_after(jiffies, time_limit)))
			goto softnet_break;

		local_irq_enable();

		/* Even though interrupts have been re-enabled, this
		 * access is safe because interrupts can only add new
		 * entries to the tail of this list, and only ->poll()
		 * calls can remove this head entry from the list.
		 */
		n = list_first_entry(&sd->poll_list, struct napi_struct, poll_list);

		have = netpoll_poll_lock(n);

		weight = n->weight;

		/* This NAPI_STATE_SCHED test is for avoiding a race
		 * with netpoll's poll_napi().  Only the entity which
		 * obtains the lock and sees NAPI_STATE_SCHED set will
		 * actually make the ->poll() call.  Therefore we avoid
		 * accidentally calling ->poll() when NAPI is not scheduled.
		 */
		work = 0;
		if (test_bit(NAPI_STATE_SCHED, &n->state)) {
			work = n->poll(n, weight);
			trace_napi_poll(n);
		}

		WARN_ON_ONCE(work > weight);

		budget -= work;

		local_irq_disable();

		/* Drivers must not modify the NAPI state if they
		 * consume the entire weight.  In such cases this code
		 * still "owns" the NAPI instance and therefore can
		 * move the instance around on the list at-will.
		 */
		if (unlikely(work == weight)) {
			if (unlikely(napi_disable_pending(n))) {
				local_irq_enable();
				napi_complete(n);
				local_irq_disable();
			} else {
				if (n->gro_list) {
					/* flush too old packets
					 * If HZ < 1000, flush all packets.
					 */
					local_irq_enable();
					napi_gro_flush(n, HZ >= 1000);
					local_irq_disable();
				}
				list_move_tail(&n->poll_list, &sd->poll_list);
			}
		}

		netpoll_poll_unlock(have);
	}
out:
	net_rps_action_and_irq_enable(sd);

#ifdef CONFIG_NET_DMA
	/*
	 * There may not be any more sk_buffs coming right now, so push
	 * any pending DMA copies to hardware
	 */
	dma_issue_pending_all();
#endif

	return;

softnet_break:
	sd->time_squeeze++;
	__raise_softirq_irqoff(NET_RX_SOFTIRQ);
	goto out;
}

static gifconf_func_t *gifconf_list[NPROTO];

/**
 *	register_gifconf	-	register a SIOCGIF handler
 *	@family: Address family
 *	@gifconf: Function handler
 *
 *	Register protocol dependent address dumping routines. The handler
 *	that is passed must not be freed or reused until it has been replaced
 *	by another handler.
 */
int register_gifconf(unsigned int family, gifconf_func_t *gifconf)
{
	if (family >= NPROTO)
		return -EINVAL;
	gifconf_list[family] = gifconf;
	return 0;
}
EXPORT_SYMBOL(register_gifconf);


/*
 *	Map an interface index to its name (SIOCGIFNAME)
 */

/*
 *	We need this ioctl for efficient implementation of the
 *	if_indextoname() function required by the IPv6 API.  Without
 *	it, we would have to search all the interfaces to find a
 *	match.  --pb
 */

static int dev_ifname(struct net *net, struct ifreq __user *arg)
{
	struct net_device *dev;
	struct ifreq ifr;
	unsigned seq;

	/*
	 *	Fetch the caller's info block.
	 */

	if (copy_from_user(&ifr, arg, sizeof(struct ifreq)))
		return -EFAULT;

retry:
	seq = read_seqcount_begin(&devnet_rename_seq);
	rcu_read_lock();
	dev = dev_get_by_index_rcu(net, ifr.ifr_ifindex);
	if (!dev) {
		rcu_read_unlock();
		return -ENODEV;
	}

	strcpy(ifr.ifr_name, dev->name);
	rcu_read_unlock();
	if (read_seqcount_retry(&devnet_rename_seq, seq))
		goto retry;

	if (copy_to_user(arg, &ifr, sizeof(struct ifreq)))
		return -EFAULT;
	return 0;
}

/*
 *	Perform a SIOCGIFCONF call. This structure will change
 *	size eventually, and there is nothing I can do about it.
 *	Thus we will need a 'compatibility mode'.
 */

static int dev_ifconf(struct net *net, char __user *arg)
{
	struct ifconf ifc;
	struct net_device *dev;
	char __user *pos;
	int len;
	int total;
	int i;

	/*
	 *	Fetch the caller's info block.
	 */

	if (copy_from_user(&ifc, arg, sizeof(struct ifconf)))
		return -EFAULT;

	pos = ifc.ifc_buf;
	len = ifc.ifc_len;

	/*
	 *	Loop over the interfaces, and write an info block for each.
	 */

	total = 0;
	for_each_netdev(net, dev) {
		for (i = 0; i < NPROTO; i++) {
			if (gifconf_list[i]) {
				int done;
				if (!pos)
					done = gifconf_list[i](dev, NULL, 0);
				else
					done = gifconf_list[i](dev, pos + total,
							       len - total);
				if (done < 0)
					return -EFAULT;
				total += done;
			}
		}
	}

	/*
	 *	All done.  Write the updated control block back to the caller.
	 */
	ifc.ifc_len = total;

	/*
	 * 	Both BSD and Solaris return 0 here, so we do too.
	 */
	return copy_to_user(arg, &ifc, sizeof(struct ifconf)) ? -EFAULT : 0;
}

#ifdef CONFIG_PROC_FS

#define BUCKET_SPACE (32 - NETDEV_HASHBITS - 1)

#define get_bucket(x) ((x) >> BUCKET_SPACE)
#define get_offset(x) ((x) & ((1 << BUCKET_SPACE) - 1))
#define set_bucket_offset(b, o) ((b) << BUCKET_SPACE | (o))

static inline struct net_device *dev_from_same_bucket(struct seq_file *seq, loff_t *pos)
{
	struct net *net = seq_file_net(seq);
	struct net_device *dev;
	struct hlist_node *p;
	struct hlist_head *h;
	unsigned int count = 0, offset = get_offset(*pos);

	h = &net->dev_name_head[get_bucket(*pos)];
	hlist_for_each_entry_rcu(dev, p, h, name_hlist) {
		if (++count == offset)
			return dev;
	}

	return NULL;
}

static inline struct net_device *dev_from_bucket(struct seq_file *seq, loff_t *pos)
{
	struct net_device *dev;
	unsigned int bucket;

	do {
		dev = dev_from_same_bucket(seq, pos);
		if (dev)
			return dev;

		bucket = get_bucket(*pos) + 1;
		*pos = set_bucket_offset(bucket, 1);
	} while (bucket < NETDEV_HASHENTRIES);

	return NULL;
}

/*
 *	This is invoked by the /proc filesystem handler to display a device
 *	in detail.
 */
void *dev_seq_start(struct seq_file *seq, loff_t *pos)
	__acquires(RCU)
{
	rcu_read_lock();
	if (!*pos)
		return SEQ_START_TOKEN;

	if (get_bucket(*pos) >= NETDEV_HASHENTRIES)
		return NULL;

	return dev_from_bucket(seq, pos);
}

void *dev_seq_next(struct seq_file *seq, void *v, loff_t *pos)
{
	++*pos;
	return dev_from_bucket(seq, pos);
}

void dev_seq_stop(struct seq_file *seq, void *v)
	__releases(RCU)
{
	rcu_read_unlock();
}

static void dev_seq_printf_stats(struct seq_file *seq, struct net_device *dev)
{
	struct rtnl_link_stats64 temp;
	const struct rtnl_link_stats64 *stats = dev_get_stats(dev, &temp);

	seq_printf(seq, "%6s: %7llu %7llu %4llu %4llu %4llu %5llu %10llu %9llu "
		   "%8llu %7llu %4llu %4llu %4llu %5llu %7llu %10llu\n",
		   dev->name, stats->rx_bytes, stats->rx_packets,
		   stats->rx_errors,
		   stats->rx_dropped + stats->rx_missed_errors,
		   stats->rx_fifo_errors,
		   stats->rx_length_errors + stats->rx_over_errors +
		    stats->rx_crc_errors + stats->rx_frame_errors,
		   stats->rx_compressed, stats->multicast,
		   stats->tx_bytes, stats->tx_packets,
		   stats->tx_errors, stats->tx_dropped,
		   stats->tx_fifo_errors, stats->collisions,
		   stats->tx_carrier_errors +
		    stats->tx_aborted_errors +
		    stats->tx_window_errors +
		    stats->tx_heartbeat_errors,
		   stats->tx_compressed);
}

/*
 *	Called from the PROCfs module. This now uses the new arbitrary sized
 *	/proc/net interface to create /proc/net/dev
 */
static int dev_seq_show(struct seq_file *seq, void *v)
{
	if (v == SEQ_START_TOKEN)
		seq_puts(seq, "Inter-|   Receive                            "
			      "                    |  Transmit\n"
			      " face |bytes    packets errs drop fifo frame "
			      "compressed multicast|bytes    packets errs "
			      "drop fifo colls carrier compressed\n");
	else
		dev_seq_printf_stats(seq, v);
	return 0;
}

static struct softnet_data *softnet_get_online(loff_t *pos)
{
	struct softnet_data *sd = NULL;

	while (*pos < nr_cpu_ids)
		if (cpu_online(*pos)) {
			sd = &per_cpu(softnet_data, *pos);
			break;
		} else
			++*pos;
	return sd;
}

static void *softnet_seq_start(struct seq_file *seq, loff_t *pos)
{
	return softnet_get_online(pos);
}

static void *softnet_seq_next(struct seq_file *seq, void *v, loff_t *pos)
{
	++*pos;
	return softnet_get_online(pos);
}

static void softnet_seq_stop(struct seq_file *seq, void *v)
{
}

static int softnet_seq_show(struct seq_file *seq, void *v)
{
	struct softnet_data *sd = v;

	seq_printf(seq, "%08x %08x %08x %08x %08x %08x %08x %08x %08x %08x\n",
		   sd->processed, sd->dropped, sd->time_squeeze, 0,
		   0, 0, 0, 0, /* was fastroute */
		   sd->cpu_collision, sd->received_rps);
	return 0;
}

static const struct seq_operations dev_seq_ops = {
	.start = dev_seq_start,
	.next  = dev_seq_next,
	.stop  = dev_seq_stop,
	.show  = dev_seq_show,
};

static int dev_seq_open(struct inode *inode, struct file *file)
{
	return seq_open_net(inode, file, &dev_seq_ops,
			    sizeof(struct seq_net_private));
}

static const struct file_operations dev_seq_fops = {
	.owner	 = THIS_MODULE,
	.open    = dev_seq_open,
	.read    = seq_read,
	.llseek  = seq_lseek,
	.release = seq_release_net,
};

static const struct seq_operations softnet_seq_ops = {
	.start = softnet_seq_start,
	.next  = softnet_seq_next,
	.stop  = softnet_seq_stop,
	.show  = softnet_seq_show,
};

static int softnet_seq_open(struct inode *inode, struct file *file)
{
	return seq_open(file, &softnet_seq_ops);
}

static const struct file_operations softnet_seq_fops = {
	.owner	 = THIS_MODULE,
	.open    = softnet_seq_open,
	.read    = seq_read,
	.llseek  = seq_lseek,
	.release = seq_release,
};

static void *ptype_get_idx(loff_t pos)
{
	struct packet_type *pt = NULL;
	loff_t i = 0;
	int t;

	list_for_each_entry_rcu(pt, &ptype_all, list) {
		if (i == pos)
			return pt;
		++i;
	}

	for (t = 0; t < PTYPE_HASH_SIZE; t++) {
		list_for_each_entry_rcu(pt, &ptype_base[t], list) {
			if (i == pos)
				return pt;
			++i;
		}
	}
	return NULL;
}

static void *ptype_seq_start(struct seq_file *seq, loff_t *pos)
	__acquires(RCU)
{
	rcu_read_lock();
	return *pos ? ptype_get_idx(*pos - 1) : SEQ_START_TOKEN;
}

static void *ptype_seq_next(struct seq_file *seq, void *v, loff_t *pos)
{
	struct packet_type *pt;
	struct list_head *nxt;
	int hash;

	++*pos;
	if (v == SEQ_START_TOKEN)
		return ptype_get_idx(0);

	pt = v;
	nxt = pt->list.next;
	if (pt->type == htons(ETH_P_ALL)) {
		if (nxt != &ptype_all)
			goto found;
		hash = 0;
		nxt = ptype_base[0].next;
	} else
		hash = ntohs(pt->type) & PTYPE_HASH_MASK;

	while (nxt == &ptype_base[hash]) {
		if (++hash >= PTYPE_HASH_SIZE)
			return NULL;
		nxt = ptype_base[hash].next;
	}
found:
	return list_entry(nxt, struct packet_type, list);
}

static void ptype_seq_stop(struct seq_file *seq, void *v)
	__releases(RCU)
{
	rcu_read_unlock();
}

static int ptype_seq_show(struct seq_file *seq, void *v)
{
	struct packet_type *pt = v;

	if (v == SEQ_START_TOKEN)
		seq_puts(seq, "Type Device      Function\n");
	else if (pt->dev == NULL || dev_net(pt->dev) == seq_file_net(seq)) {
		if (pt->type == htons(ETH_P_ALL))
			seq_puts(seq, "ALL ");
		else
			seq_printf(seq, "%04x", ntohs(pt->type));

		seq_printf(seq, " %-8s %pF\n",
			   pt->dev ? pt->dev->name : "", pt->func);
	}

	return 0;
}

static const struct seq_operations ptype_seq_ops = {
	.start = ptype_seq_start,
	.next  = ptype_seq_next,
	.stop  = ptype_seq_stop,
	.show  = ptype_seq_show,
};

static int ptype_seq_open(struct inode *inode, struct file *file)
{
	return seq_open_net(inode, file, &ptype_seq_ops,
			sizeof(struct seq_net_private));
}

static const struct file_operations ptype_seq_fops = {
	.owner	 = THIS_MODULE,
	.open    = ptype_seq_open,
	.read    = seq_read,
	.llseek  = seq_lseek,
	.release = seq_release_net,
};


static int __net_init dev_proc_net_init(struct net *net)
{
	int rc = -ENOMEM;

	if (!proc_net_fops_create(net, "dev", S_IRUGO, &dev_seq_fops))
		goto out;
	if (!proc_net_fops_create(net, "softnet_stat", S_IRUGO, &softnet_seq_fops))
		goto out_dev;
	if (!proc_net_fops_create(net, "ptype", S_IRUGO, &ptype_seq_fops))
		goto out_softnet;

	if (wext_proc_init(net))
		goto out_ptype;
	rc = 0;
out:
	return rc;
out_ptype:
	proc_net_remove(net, "ptype");
out_softnet:
	proc_net_remove(net, "softnet_stat");
out_dev:
	proc_net_remove(net, "dev");
	goto out;
}

static void __net_exit dev_proc_net_exit(struct net *net)
{
	wext_proc_exit(net);

	proc_net_remove(net, "ptype");
	proc_net_remove(net, "softnet_stat");
	proc_net_remove(net, "dev");
}

static struct pernet_operations __net_initdata dev_proc_ops = {
	.init = dev_proc_net_init,
	.exit = dev_proc_net_exit,
};

static int __init dev_proc_init(void)
{
	return register_pernet_subsys(&dev_proc_ops);
}
#else
#define dev_proc_init() 0
#endif	/* CONFIG_PROC_FS */


struct netdev_upper {
	struct net_device *dev;
	bool master;
	struct list_head list;
	struct rcu_head rcu;
	struct list_head search_list;
};

static void __append_search_uppers(struct list_head *search_list,
				   struct net_device *dev)
{
	struct netdev_upper *upper;

	list_for_each_entry(upper, &dev->upper_dev_list, list) {
		/* check if this upper is not already in search list */
		if (list_empty(&upper->search_list))
			list_add_tail(&upper->search_list, search_list);
	}
}

static bool __netdev_search_upper_dev(struct net_device *dev,
				      struct net_device *upper_dev)
{
	LIST_HEAD(search_list);
	struct netdev_upper *upper;
	struct netdev_upper *tmp;
	bool ret = false;

	__append_search_uppers(&search_list, dev);
	list_for_each_entry(upper, &search_list, search_list) {
		if (upper->dev == upper_dev) {
			ret = true;
			break;
		}
		__append_search_uppers(&search_list, upper->dev);
	}
	list_for_each_entry_safe(upper, tmp, &search_list, search_list)
		INIT_LIST_HEAD(&upper->search_list);
	return ret;
}

static struct netdev_upper *__netdev_find_upper(struct net_device *dev,
						struct net_device *upper_dev)
{
	struct netdev_upper *upper;

	list_for_each_entry(upper, &dev->upper_dev_list, list) {
		if (upper->dev == upper_dev)
			return upper;
	}
	return NULL;
}

/**
 * netdev_has_upper_dev - Check if device is linked to an upper device
 * @dev: device
 * @upper_dev: upper device to check
 *
 * Find out if a device is linked to specified upper device and return true
 * in case it is. Note that this checks only immediate upper device,
 * not through a complete stack of devices. The caller must hold the RTNL lock.
 */
bool netdev_has_upper_dev(struct net_device *dev,
			  struct net_device *upper_dev)
{
	ASSERT_RTNL();

	return __netdev_find_upper(dev, upper_dev);
}
EXPORT_SYMBOL(netdev_has_upper_dev);

/**
 * netdev_has_any_upper_dev - Check if device is linked to some device
 * @dev: device
 *
 * Find out if a device is linked to an upper device and return true in case
 * it is. The caller must hold the RTNL lock.
 */
bool netdev_has_any_upper_dev(struct net_device *dev)
{
	ASSERT_RTNL();

	return !list_empty(&dev->upper_dev_list);
}
EXPORT_SYMBOL(netdev_has_any_upper_dev);

/**
 * netdev_master_upper_dev_get - Get master upper device
 * @dev: device
 *
 * Find a master upper device and return pointer to it or NULL in case
 * it's not there. The caller must hold the RTNL lock.
 */
struct net_device *netdev_master_upper_dev_get(struct net_device *dev)
{
	struct netdev_upper *upper;

	ASSERT_RTNL();

	if (list_empty(&dev->upper_dev_list))
		return NULL;

	upper = list_first_entry(&dev->upper_dev_list,
				 struct netdev_upper, list);
	if (likely(upper->master))
		return upper->dev;
	return NULL;
}
EXPORT_SYMBOL(netdev_master_upper_dev_get);

/**
 * netdev_master_upper_dev_get_rcu - Get master upper device
 * @dev: device
 *
 * Find a master upper device and return pointer to it or NULL in case
 * it's not there. The caller must hold the RCU read lock.
 */
struct net_device *netdev_master_upper_dev_get_rcu(struct net_device *dev)
{
	struct netdev_upper *upper;

	upper = list_first_or_null_rcu(&dev->upper_dev_list,
				       struct netdev_upper, list);
	if (upper && likely(upper->master))
		return upper->dev;
	return NULL;
}
EXPORT_SYMBOL(netdev_master_upper_dev_get_rcu);

static int __netdev_upper_dev_link(struct net_device *dev,
				   struct net_device *upper_dev, bool master)
{
	struct netdev_upper *upper;

	ASSERT_RTNL();

	if (dev == upper_dev)
		return -EBUSY;

	/* To prevent loops, check if dev is not upper device to upper_dev. */
	if (__netdev_search_upper_dev(upper_dev, dev))
		return -EBUSY;

	if (__netdev_find_upper(dev, upper_dev))
		return -EEXIST;

	if (master && netdev_master_upper_dev_get(dev))
		return -EBUSY;

	upper = kmalloc(sizeof(*upper), GFP_KERNEL);
	if (!upper)
		return -ENOMEM;

	upper->dev = upper_dev;