dev.c

 *	...
 *	unregister_netdevice(y1);
 *	unregister_netdevice(y2);
 *      ...
 *	rtnl_unlock();
 *	free_netdev(y1);
 *	free_netdev(y2);
 *
 * We are invoked by rtnl_unlock() after it drops the semaphore.
 * This allows us to deal with problems:
 * 1) We can delete sysfs objects which invoke hotplug
 *    without deadlocking with linkwatch via keventd.
 * 2) Since we run with the RTNL semaphore not held, we can sleep
 *    safely in order to wait for the netdev refcnt to drop to zero.
 */
static DEFINE_MUTEX(net_todo_run_mutex);
void netdev_run_todo(void)
{
	struct list_head list;

	/* Need to guard against multiple cpu's getting out of order. */
	mutex_lock(&net_todo_run_mutex);

	/* Not safe to do outside the semaphore.  We must not return
	 * until all unregister events invoked by the local processor
	 * have been completed (either by this todo run, or one on
	 * another cpu).
	 */
	if (list_empty(&net_todo_list))
		goto out;

	/* Snapshot list, allow later requests */
	spin_lock(&net_todo_list_lock);
	list_replace_init(&net_todo_list, &list);
	spin_unlock(&net_todo_list_lock);

	while (!list_empty(&list)) {
		struct net_device *dev
			= list_entry(list.next, struct net_device, todo_list);
		list_del(&dev->todo_list);

		if (unlikely(dev->reg_state != NETREG_UNREGISTERING)) {
			printk(KERN_ERR "network todo '%s' but state %d\n",
			       dev->name, dev->reg_state);
			dump_stack();
			continue;
		}

		dev->reg_state = NETREG_UNREGISTERED;

		netdev_wait_allrefs(dev);

		/* paranoia */
		BUG_ON(atomic_read(&dev->refcnt));
		BUG_TRAP(!dev->ip_ptr);
		BUG_TRAP(!dev->ip6_ptr);
		BUG_TRAP(!dev->dn_ptr);

		if (dev->destructor)
			dev->destructor(dev);

		/* Free network device */
		kobject_put(&dev->dev.kobj);
	}

out:
	mutex_unlock(&net_todo_run_mutex);
}

static struct net_device_stats *internal_stats(struct net_device *dev)
{
	return &dev->stats;
}

/**
 *	alloc_netdev_mq - allocate network device
 *	@sizeof_priv:	size of private data to allocate space for
 *	@name:		device name format string
 *	@setup:		callback to initialize device
 *	@queue_count:	the number of subqueues to allocate
 *
 *	Allocates a struct net_device with private data area for driver use
 *	and performs basic initialization.  Also allocates subquue structs
 *	for each queue on the device at the end of the netdevice.
 */
struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name,
		void (*setup)(struct net_device *), unsigned int queue_count)
{
	void *p;
	struct net_device *dev;
	int alloc_size;

	BUG_ON(strlen(name) >= sizeof(dev->name));

	alloc_size = sizeof(struct net_device) +
		     sizeof(struct net_device_subqueue) * (queue_count - 1);
	if (sizeof_priv) {
		/* ensure 32-byte alignment of private area */
		alloc_size = (alloc_size + NETDEV_ALIGN_CONST) & ~NETDEV_ALIGN_CONST;
		alloc_size += sizeof_priv;
	}
	/* ensure 32-byte alignment of whole construct */
	alloc_size += NETDEV_ALIGN_CONST;

	p = kzalloc(alloc_size, GFP_KERNEL);
	if (!p) {
		printk(KERN_ERR "alloc_netdev: Unable to allocate device.\n");
		return NULL;
	}

	dev = (struct net_device *)
		(((long)p + NETDEV_ALIGN_CONST) & ~NETDEV_ALIGN_CONST);
	dev->padded = (char *)dev - (char *)p;
	dev_net_set(dev, &init_net);

	if (sizeof_priv) {
		dev->priv = ((char *)dev +
			     ((sizeof(struct net_device) +
			       (sizeof(struct net_device_subqueue) *
				(queue_count - 1)) + NETDEV_ALIGN_CONST)
			      & ~NETDEV_ALIGN_CONST));
	}

	dev->egress_subqueue_count = queue_count;
	dev->gso_max_size = GSO_MAX_SIZE;

	dev->get_stats = internal_stats;
	netpoll_netdev_init(dev);
	setup(dev);
	strcpy(dev->name, name);
	return dev;
}
EXPORT_SYMBOL(alloc_netdev_mq);

/**
 *	free_netdev - free network device
 *	@dev: device
 *
 *	This function does the last stage of destroying an allocated device
 * 	interface. The reference to the device object is released.
 *	If this is the last reference then it will be freed.
 */
void free_netdev(struct net_device *dev)
{
	release_net(dev_net(dev));

	/*  Compatibility with error handling in drivers */
	if (dev->reg_state == NETREG_UNINITIALIZED) {
		kfree((char *)dev - dev->padded);
		return;
	}

	BUG_ON(dev->reg_state != NETREG_UNREGISTERED);
	dev->reg_state = NETREG_RELEASED;

	/* will free via device release */
	put_device(&dev->dev);
}

/* Synchronize with packet receive processing. */
void synchronize_net(void)
{
	might_sleep();
	synchronize_rcu();
}

/**
 *	unregister_netdevice - remove device from the kernel
 *	@dev: device
 *
 *	This function shuts down a device interface and removes it
 *	from the kernel tables.
 *
 *	Callers must hold the rtnl semaphore.  You may want
 *	unregister_netdev() instead of this.
 */

void unregister_netdevice(struct net_device *dev)
{
	ASSERT_RTNL();

	rollback_registered(dev);
	/* Finish processing unregister after unlock */
	net_set_todo(dev);
}

/**
 *	unregister_netdev - remove device from the kernel
 *	@dev: device
 *
 *	This function shuts down a device interface and removes it
 *	from the kernel tables.
 *
 *	This is just a wrapper for unregister_netdevice that takes
 *	the rtnl semaphore.  In general you want to use this and not
 *	unregister_netdevice.
 */
void unregister_netdev(struct net_device *dev)
{
	rtnl_lock();
	unregister_netdevice(dev);
	rtnl_unlock();
}

EXPORT_SYMBOL(unregister_netdev);

/**
 *	dev_change_net_namespace - move device to different nethost namespace
 *	@dev: device
 *	@net: network namespace
 *	@pat: If not NULL name pattern to try if the current device name
 *	      is already taken in the destination network namespace.
 *
 *	This function shuts down a device interface and moves it
 *	to a new network namespace. On success 0 is returned, on
 *	a failure a netagive errno code is returned.
 *
 *	Callers must hold the rtnl semaphore.
 */

int dev_change_net_namespace(struct net_device *dev, struct net *net, const char *pat)
{
	char buf[IFNAMSIZ];
	const char *destname;
	int err;

	ASSERT_RTNL();

	/* Don't allow namespace local devices to be moved. */
	err = -EINVAL;
	if (dev->features & NETIF_F_NETNS_LOCAL)
		goto out;

	/* Ensure the device has been registrered */
	err = -EINVAL;
	if (dev->reg_state != NETREG_REGISTERED)
		goto out;

	/* Get out if there is nothing todo */
	err = 0;
	if (net_eq(dev_net(dev), net))
		goto out;

	/* Pick the destination device name, and ensure
	 * we can use it in the destination network namespace.
	 */
	err = -EEXIST;
	destname = dev->name;
	if (__dev_get_by_name(net, destname)) {
		/* We get here if we can't use the current device name */
		if (!pat)
			goto out;
		if (!dev_valid_name(pat))
			goto out;
		if (strchr(pat, '%')) {
			if (__dev_alloc_name(net, pat, buf) < 0)
				goto out;
			destname = buf;
		} else
			destname = pat;
		if (__dev_get_by_name(net, destname))
			goto out;
	}

	/*
	 * And now a mini version of register_netdevice unregister_netdevice.
	 */

	/* If device is running close it first. */
	dev_close(dev);

	/* And unlink it from device chain */
	err = -ENODEV;
	unlist_netdevice(dev);

	synchronize_net();

	/* Shutdown queueing discipline. */
	dev_shutdown(dev);

	/* Notify protocols, that we are about to destroy
	   this device. They should clean all the things.
	*/
	call_netdevice_notifiers(NETDEV_UNREGISTER, dev);

	/*
	 *	Flush the unicast and multicast chains
	 */
	dev_addr_discard(dev);

	/* Actually switch the network namespace */
	dev_net_set(dev, net);

	/* Assign the new device name */
	if (destname != dev->name)
		strcpy(dev->name, destname);

	/* If there is an ifindex conflict assign a new one */
	if (__dev_get_by_index(net, dev->ifindex)) {
		int iflink = (dev->iflink == dev->ifindex);
		dev->ifindex = dev_new_index(net);
		if (iflink)
			dev->iflink = dev->ifindex;
	}

	/* Fixup kobjects */
	netdev_unregister_kobject(dev);
	err = netdev_register_kobject(dev);
	WARN_ON(err);

	/* Add the device back in the hashes */
	list_netdevice(dev);

	/* Notify protocols, that a new device appeared. */
	call_netdevice_notifiers(NETDEV_REGISTER, dev);

	synchronize_net();
	err = 0;
out:
	return err;
}

static int dev_cpu_callback(struct notifier_block *nfb,
			    unsigned long action,
			    void *ocpu)
{
	struct sk_buff **list_skb;
	struct net_device **list_net;
	struct sk_buff *skb;
	unsigned int cpu, oldcpu = (unsigned long)ocpu;
	struct softnet_data *sd, *oldsd;

	if (action != CPU_DEAD && action != CPU_DEAD_FROZEN)
		return NOTIFY_OK;

	local_irq_disable();
	cpu = smp_processor_id();
	sd = &per_cpu(softnet_data, cpu);
	oldsd = &per_cpu(softnet_data, oldcpu);

	/* Find end of our completion_queue. */
	list_skb = &sd->completion_queue;
	while (*list_skb)
		list_skb = &(*list_skb)->next;
	/* Append completion queue from offline CPU. */
	*list_skb = oldsd->completion_queue;
	oldsd->completion_queue = NULL;

	/* Find end of our output_queue. */
	list_net = &sd->output_queue;
	while (*list_net)
		list_net = &(*list_net)->next_sched;
	/* Append output queue from offline CPU. */
	*list_net = oldsd->output_queue;
	oldsd->output_queue = NULL;

	raise_softirq_irqoff(NET_TX_SOFTIRQ);
	local_irq_enable();

	/* Process offline CPU's input_pkt_queue */
	while ((skb = __skb_dequeue(&oldsd->input_pkt_queue)))
		netif_rx(skb);

	return NOTIFY_OK;
}

#ifdef CONFIG_NET_DMA
/**
 * net_dma_rebalance - try to maintain one DMA channel per CPU
 * @net_dma: DMA client and associated data (lock, channels, channel_mask)
 *
 * This is called when the number of channels allocated to the net_dma client
 * changes.  The net_dma client tries to have one DMA channel per CPU.
 */

static void net_dma_rebalance(struct net_dma *net_dma)
{
	unsigned int cpu, i, n, chan_idx;
	struct dma_chan *chan;

	if (cpus_empty(net_dma->channel_mask)) {
		for_each_online_cpu(cpu)
			rcu_assign_pointer(per_cpu(softnet_data, cpu).net_dma, NULL);
		return;
	}

	i = 0;
	cpu = first_cpu(cpu_online_map);

	for_each_cpu_mask(chan_idx, net_dma->channel_mask) {
		chan = net_dma->channels[chan_idx];

		n = ((num_online_cpus() / cpus_weight(net_dma->channel_mask))
		   + (i < (num_online_cpus() %
			cpus_weight(net_dma->channel_mask)) ? 1 : 0));

		while(n) {
			per_cpu(softnet_data, cpu).net_dma = chan;
			cpu = next_cpu(cpu, cpu_online_map);
			n--;
		}
		i++;
	}
}

/**
 * netdev_dma_event - event callback for the net_dma_client
 * @client: should always be net_dma_client
 * @chan: DMA channel for the event
 * @state: DMA state to be handled
 */
static enum dma_state_client
netdev_dma_event(struct dma_client *client, struct dma_chan *chan,
	enum dma_state state)
{
	int i, found = 0, pos = -1;
	struct net_dma *net_dma =
		container_of(client, struct net_dma, client);
	enum dma_state_client ack = DMA_DUP; /* default: take no action */

	spin_lock(&net_dma->lock);
	switch (state) {
	case DMA_RESOURCE_AVAILABLE:
		for (i = 0; i < nr_cpu_ids; i++)
			if (net_dma->channels[i] == chan) {
				found = 1;
				break;
			} else if (net_dma->channels[i] == NULL && pos < 0)
				pos = i;

		if (!found && pos >= 0) {
			ack = DMA_ACK;
			net_dma->channels[pos] = chan;
			cpu_set(pos, net_dma->channel_mask);
			net_dma_rebalance(net_dma);
		}
		break;
	case DMA_RESOURCE_REMOVED:
		for (i = 0; i < nr_cpu_ids; i++)
			if (net_dma->channels[i] == chan) {
				found = 1;
				pos = i;
				break;
			}

		if (found) {
			ack = DMA_ACK;
			cpu_clear(pos, net_dma->channel_mask);
			net_dma->channels[i] = NULL;
			net_dma_rebalance(net_dma);
		}
		break;
	default:
		break;
	}
	spin_unlock(&net_dma->lock);

	return ack;
}

/**
 * netdev_dma_regiser - register the networking subsystem as a DMA client
 */
static int __init netdev_dma_register(void)
{
	net_dma.channels = kzalloc(nr_cpu_ids * sizeof(struct net_dma),
								GFP_KERNEL);
	if (unlikely(!net_dma.channels)) {
		printk(KERN_NOTICE
				"netdev_dma: no memory for net_dma.channels\n");
		return -ENOMEM;
	}
	spin_lock_init(&net_dma.lock);
	dma_cap_set(DMA_MEMCPY, net_dma.client.cap_mask);
	dma_async_client_register(&net_dma.client);
	dma_async_client_chan_request(&net_dma.client);
	return 0;
}

#else
static int __init netdev_dma_register(void) { return -ENODEV; }
#endif /* CONFIG_NET_DMA */

/**
 *	netdev_compute_feature - compute conjunction of two feature sets
 *	@all: first feature set
 *	@one: second feature set
 *
 *	Computes a new feature set after adding a device with feature set
 *	@one to the master device with current feature set @all.  Returns
 *	the new feature set.
 */
int netdev_compute_features(unsigned long all, unsigned long one)
{
	/* if device needs checksumming, downgrade to hw checksumming */
	if (all & NETIF_F_NO_CSUM && !(one & NETIF_F_NO_CSUM))
		all ^= NETIF_F_NO_CSUM | NETIF_F_HW_CSUM;

	/* if device can't do all checksum, downgrade to ipv4/ipv6 */
	if (all & NETIF_F_HW_CSUM && !(one & NETIF_F_HW_CSUM))
		all ^= NETIF_F_HW_CSUM
			| NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM;

	if (one & NETIF_F_GSO)
		one |= NETIF_F_GSO_SOFTWARE;
	one |= NETIF_F_GSO;

	/* If even one device supports robust GSO, enable it for all. */
	if (one & NETIF_F_GSO_ROBUST)
		all |= NETIF_F_GSO_ROBUST;

	all &= one | NETIF_F_LLTX;

	if (!(all & NETIF_F_ALL_CSUM))
		all &= ~NETIF_F_SG;
	if (!(all & NETIF_F_SG))
		all &= ~NETIF_F_GSO_MASK;

	return all;
}
EXPORT_SYMBOL(netdev_compute_features);

static struct hlist_head *netdev_create_hash(void)
{
	int i;
	struct hlist_head *hash;

	hash = kmalloc(sizeof(*hash) * NETDEV_HASHENTRIES, GFP_KERNEL);
	if (hash != NULL)
		for (i = 0; i < NETDEV_HASHENTRIES; i++)
			INIT_HLIST_HEAD(&hash[i]);

	return hash;
}

/* Initialize per network namespace state */
static int __net_init netdev_init(struct net *net)
{
	INIT_LIST_HEAD(&net->dev_base_head);

	net->dev_name_head = netdev_create_hash();
	if (net->dev_name_head == NULL)
		goto err_name;

	net->dev_index_head = netdev_create_hash();
	if (net->dev_index_head == NULL)
		goto err_idx;

	return 0;

err_idx:
	kfree(net->dev_name_head);
err_name:
	return -ENOMEM;
}

static void __net_exit netdev_exit(struct net *net)
{
	kfree(net->dev_name_head);
	kfree(net->dev_index_head);
}

static struct pernet_operations __net_initdata netdev_net_ops = {
	.init = netdev_init,
	.exit = netdev_exit,
};

static void __net_exit default_device_exit(struct net *net)
{
	struct net_device *dev, *next;
	/*
	 * Push all migratable of the network devices back to the
	 * initial network namespace
	 */
	rtnl_lock();
	for_each_netdev_safe(net, dev, next) {
		int err;
		char fb_name[IFNAMSIZ];

		/* Ignore unmoveable devices (i.e. loopback) */
		if (dev->features & NETIF_F_NETNS_LOCAL)
			continue;

		/* Push remaing network devices to init_net */
		snprintf(fb_name, IFNAMSIZ, "dev%d", dev->ifindex);
		err = dev_change_net_namespace(dev, &init_net, fb_name);
		if (err) {
			printk(KERN_EMERG "%s: failed to move %s to init_net: %d\n",
				__func__, dev->name, err);
			BUG();
		}
	}
	rtnl_unlock();
}

static struct pernet_operations __net_initdata default_device_ops = {
	.exit = default_device_exit,
};

/*
 *	Initialize the DEV module. At boot time this walks the device list and
 *	unhooks any devices that fail to initialise (normally hardware not
 *	present) and leaves us with a valid list of present and active devices.
 *
 */

/*
 *       This is called single threaded during boot, so no need
 *       to take the rtnl semaphore.
 */
static int __init net_dev_init(void)
{
	int i, rc = -ENOMEM;

	BUG_ON(!dev_boot_phase);

	if (dev_proc_init())
		goto out;

	if (netdev_kobject_init())
		goto out;

	INIT_LIST_HEAD(&ptype_all);
	for (i = 0; i < PTYPE_HASH_SIZE; i++)
		INIT_LIST_HEAD(&ptype_base[i]);

	if (register_pernet_subsys(&netdev_net_ops))
		goto out;

	if (register_pernet_device(&default_device_ops))
		goto out;

	/*
	 *	Initialise the packet receive queues.
	 */

	for_each_possible_cpu(i) {
		struct softnet_data *queue;

		queue = &per_cpu(softnet_data, i);
		skb_queue_head_init(&queue->input_pkt_queue);
		queue->completion_queue = NULL;
		INIT_LIST_HEAD(&queue->poll_list);

		queue->backlog.poll = process_backlog;
		queue->backlog.weight = weight_p;
	}

	netdev_dma_register();

	dev_boot_phase = 0;

	open_softirq(NET_TX_SOFTIRQ, net_tx_action, NULL);
	open_softirq(NET_RX_SOFTIRQ, net_rx_action, NULL);

	hotcpu_notifier(dev_cpu_callback, 0);
	dst_init();
	dev_mcast_init();
	rc = 0;
out:
	return rc;
}

subsys_initcall(net_dev_init);

EXPORT_SYMBOL(__dev_get_by_index);
EXPORT_SYMBOL(__dev_get_by_name);
EXPORT_SYMBOL(__dev_remove_pack);
EXPORT_SYMBOL(dev_valid_name);
EXPORT_SYMBOL(dev_add_pack);
EXPORT_SYMBOL(dev_alloc_name);
EXPORT_SYMBOL(dev_close);
EXPORT_SYMBOL(dev_get_by_flags);
EXPORT_SYMBOL(dev_get_by_index);
EXPORT_SYMBOL(dev_get_by_name);
EXPORT_SYMBOL(dev_open);
EXPORT_SYMBOL(dev_queue_xmit);
EXPORT_SYMBOL(dev_remove_pack);
EXPORT_SYMBOL(dev_set_allmulti);
EXPORT_SYMBOL(dev_set_promiscuity);
EXPORT_SYMBOL(dev_change_flags);
EXPORT_SYMBOL(dev_set_mtu);
EXPORT_SYMBOL(dev_set_mac_address);
EXPORT_SYMBOL(free_netdev);
EXPORT_SYMBOL(netdev_boot_setup_check);
EXPORT_SYMBOL(netdev_set_master);
EXPORT_SYMBOL(netdev_state_change);
EXPORT_SYMBOL(netif_receive_skb);
EXPORT_SYMBOL(netif_rx);
EXPORT_SYMBOL(register_gifconf);
EXPORT_SYMBOL(register_netdevice);
EXPORT_SYMBOL(register_netdevice_notifier);
EXPORT_SYMBOL(skb_checksum_help);
EXPORT_SYMBOL(synchronize_net);
EXPORT_SYMBOL(unregister_netdevice);
EXPORT_SYMBOL(unregister_netdevice_notifier);
EXPORT_SYMBOL(net_enable_timestamp);
EXPORT_SYMBOL(net_disable_timestamp);
EXPORT_SYMBOL(dev_get_flags);

#if defined(CONFIG_BRIDGE) || defined(CONFIG_BRIDGE_MODULE)
EXPORT_SYMBOL(br_handle_frame_hook);
EXPORT_SYMBOL(br_fdb_get_hook);
EXPORT_SYMBOL(br_fdb_put_hook);
#endif

#ifdef CONFIG_KMOD
EXPORT_SYMBOL(dev_load);
#endif

EXPORT_PER_CPU_SYMBOL(softnet_data);