Skip to content
Snippets Groups Projects
dev.c 104 KiB
Newer Older
Linus Torvalds's avatar
Linus Torvalds committed
	ret = 0;
	if ((old_flags ^ flags) & IFF_UP) {	/* Bit is different  ? */
		ret = ((old_flags & IFF_UP) ? dev_close : dev_open)(dev);

		if (!ret)
			dev_set_rx_mode(dev);
Linus Torvalds's avatar
Linus Torvalds committed
	}

	if (dev->flags & IFF_UP &&
	    ((old_flags ^ dev->flags) &~ (IFF_UP | IFF_PROMISC | IFF_ALLMULTI |
					  IFF_VOLATILE)))
		raw_notifier_call_chain(&netdev_chain,
				NETDEV_CHANGE, dev);
Linus Torvalds's avatar
Linus Torvalds committed

	if ((flags ^ dev->gflags) & IFF_PROMISC) {
		int inc = (flags & IFF_PROMISC) ? +1 : -1;
		dev->gflags ^= IFF_PROMISC;
		dev_set_promiscuity(dev, inc);
	}

	/* NOTE: order of synchronization of IFF_PROMISC and IFF_ALLMULTI
	   is important. Some (broken) drivers set IFF_PROMISC, when
	   IFF_ALLMULTI is requested not asking us and not reporting.
	 */
	if ((flags ^ dev->gflags) & IFF_ALLMULTI) {
		int inc = (flags & IFF_ALLMULTI) ? +1 : -1;
		dev->gflags ^= IFF_ALLMULTI;
		dev_set_allmulti(dev, inc);
	}

	/* Exclude state transition flags, already notified */
	changes = (old_flags ^ dev->flags) & ~(IFF_UP | IFF_RUNNING);
	if (changes)
		rtmsg_ifinfo(RTM_NEWLINK, dev, changes);
Linus Torvalds's avatar
Linus Torvalds committed

	return ret;
}

int dev_set_mtu(struct net_device *dev, int new_mtu)
{
	int err;

	if (new_mtu == dev->mtu)
		return 0;

	/*	MTU must be positive.	 */
	if (new_mtu < 0)
		return -EINVAL;

	if (!netif_device_present(dev))
		return -ENODEV;

	err = 0;
	if (dev->change_mtu)
		err = dev->change_mtu(dev, new_mtu);
	else
		dev->mtu = new_mtu;
	if (!err && dev->flags & IFF_UP)
		raw_notifier_call_chain(&netdev_chain,
				NETDEV_CHANGEMTU, dev);
Linus Torvalds's avatar
Linus Torvalds committed
	return err;
}

int dev_set_mac_address(struct net_device *dev, struct sockaddr *sa)
{
	int err;

	if (!dev->set_mac_address)
		return -EOPNOTSUPP;
	if (sa->sa_family != dev->type)
		return -EINVAL;
	if (!netif_device_present(dev))
		return -ENODEV;
	err = dev->set_mac_address(dev, sa);
	if (!err)
		raw_notifier_call_chain(&netdev_chain,
				NETDEV_CHANGEADDR, dev);
Linus Torvalds's avatar
Linus Torvalds committed
	return err;
}

/*
 *	Perform the SIOCxIFxxx calls.
 */
static int dev_ifsioc(struct net *net, struct ifreq *ifr, unsigned int cmd)
Linus Torvalds's avatar
Linus Torvalds committed
{
	int err;
	struct net_device *dev = __dev_get_by_name(net, ifr->ifr_name);
Linus Torvalds's avatar
Linus Torvalds committed

	if (!dev)
		return -ENODEV;

	switch (cmd) {
		case SIOCGIFFLAGS:	/* Get interface flags */
			ifr->ifr_flags = dev_get_flags(dev);
			return 0;

		case SIOCSIFFLAGS:	/* Set interface flags */
			return dev_change_flags(dev, ifr->ifr_flags);

		case SIOCGIFMETRIC:	/* Get the metric on the interface
					   (currently unused) */
			ifr->ifr_metric = 0;
			return 0;

		case SIOCSIFMETRIC:	/* Set the metric on the interface
					   (currently unused) */
			return -EOPNOTSUPP;

		case SIOCGIFMTU:	/* Get the MTU of a device */
			ifr->ifr_mtu = dev->mtu;
			return 0;

		case SIOCSIFMTU:	/* Set the MTU of a device */
			return dev_set_mtu(dev, ifr->ifr_mtu);

		case SIOCGIFHWADDR:
			if (!dev->addr_len)
				memset(ifr->ifr_hwaddr.sa_data, 0, sizeof ifr->ifr_hwaddr.sa_data);
			else
				memcpy(ifr->ifr_hwaddr.sa_data, dev->dev_addr,
				       min(sizeof ifr->ifr_hwaddr.sa_data, (size_t) dev->addr_len));
			ifr->ifr_hwaddr.sa_family = dev->type;
			return 0;

		case SIOCSIFHWADDR:
			return dev_set_mac_address(dev, &ifr->ifr_hwaddr);

		case SIOCSIFHWBROADCAST:
			if (ifr->ifr_hwaddr.sa_family != dev->type)
				return -EINVAL;
			memcpy(dev->broadcast, ifr->ifr_hwaddr.sa_data,
			       min(sizeof ifr->ifr_hwaddr.sa_data, (size_t) dev->addr_len));
			raw_notifier_call_chain(&netdev_chain,
Linus Torvalds's avatar
Linus Torvalds committed
					    NETDEV_CHANGEADDR, dev);
			return 0;

		case SIOCGIFMAP:
			ifr->ifr_map.mem_start = dev->mem_start;
			ifr->ifr_map.mem_end   = dev->mem_end;
			ifr->ifr_map.base_addr = dev->base_addr;
			ifr->ifr_map.irq       = dev->irq;
			ifr->ifr_map.dma       = dev->dma;
			ifr->ifr_map.port      = dev->if_port;
			return 0;

		case SIOCSIFMAP:
			if (dev->set_config) {
				if (!netif_device_present(dev))
					return -ENODEV;
				return dev->set_config(dev, &ifr->ifr_map);
			}
			return -EOPNOTSUPP;

		case SIOCADDMULTI:
			if (!dev->set_multicast_list ||
			    ifr->ifr_hwaddr.sa_family != AF_UNSPEC)
				return -EINVAL;
			if (!netif_device_present(dev))
				return -ENODEV;
			return dev_mc_add(dev, ifr->ifr_hwaddr.sa_data,
					  dev->addr_len, 1);

		case SIOCDELMULTI:
			if (!dev->set_multicast_list ||
			    ifr->ifr_hwaddr.sa_family != AF_UNSPEC)
				return -EINVAL;
			if (!netif_device_present(dev))
				return -ENODEV;
			return dev_mc_delete(dev, ifr->ifr_hwaddr.sa_data,
					     dev->addr_len, 1);

		case SIOCGIFINDEX:
			ifr->ifr_ifindex = dev->ifindex;
			return 0;

		case SIOCGIFTXQLEN:
			ifr->ifr_qlen = dev->tx_queue_len;
			return 0;

		case SIOCSIFTXQLEN:
			if (ifr->ifr_qlen < 0)
				return -EINVAL;
			dev->tx_queue_len = ifr->ifr_qlen;
			return 0;

		case SIOCSIFNAME:
			ifr->ifr_newname[IFNAMSIZ-1] = '\0';
			return dev_change_name(dev, ifr->ifr_newname);

		/*
		 *	Unknown or private ioctl
		 */

		default:
			if ((cmd >= SIOCDEVPRIVATE &&
			    cmd <= SIOCDEVPRIVATE + 15) ||
			    cmd == SIOCBONDENSLAVE ||
			    cmd == SIOCBONDRELEASE ||
			    cmd == SIOCBONDSETHWADDR ||
			    cmd == SIOCBONDSLAVEINFOQUERY ||
			    cmd == SIOCBONDINFOQUERY ||
			    cmd == SIOCBONDCHANGEACTIVE ||
			    cmd == SIOCGMIIPHY ||
			    cmd == SIOCGMIIREG ||
			    cmd == SIOCSMIIREG ||
			    cmd == SIOCBRADDIF ||
			    cmd == SIOCBRDELIF ||
			    cmd == SIOCWANDEV) {
				err = -EOPNOTSUPP;
				if (dev->do_ioctl) {
					if (netif_device_present(dev))
						err = dev->do_ioctl(dev, ifr,
								    cmd);
					else
						err = -ENODEV;
				}
			} else
				err = -EINVAL;

	}
	return err;
}

/*
 *	This function handles all "interface"-type I/O control requests. The actual
 *	'doing' part of this is dev_ifsioc above.
 */

/**
 *	dev_ioctl	-	network device ioctl
 *	@cmd: command to issue
 *	@arg: pointer to a struct ifreq in user space
 *
 *	Issue ioctl functions to devices. This is normally called by the
 *	user space syscall interfaces but can sometimes be useful for
 *	other purposes. The return value is the return from the syscall if
 *	positive or a negative errno code on error.
 */

int dev_ioctl(struct net *net, unsigned int cmd, void __user *arg)
Linus Torvalds's avatar
Linus Torvalds committed
{
	struct ifreq ifr;
	int ret;
	char *colon;

	/* One special case: SIOCGIFCONF takes ifconf argument
	   and requires shared lock, because it sleeps writing
	   to user space.
	 */

	if (cmd == SIOCGIFCONF) {
		rtnl_lock();
		ret = dev_ifconf(net, (char __user *) arg);
		rtnl_unlock();
Linus Torvalds's avatar
Linus Torvalds committed
		return ret;
	}
	if (cmd == SIOCGIFNAME)
		return dev_ifname(net, (struct ifreq __user *)arg);
Linus Torvalds's avatar
Linus Torvalds committed

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

	ifr.ifr_name[IFNAMSIZ-1] = 0;

	colon = strchr(ifr.ifr_name, ':');
	if (colon)
		*colon = 0;

	/*
	 *	See which interface the caller is talking about.
	 */

	switch (cmd) {
		/*
		 *	These ioctl calls:
		 *	- can be done by all.
		 *	- atomic and do not require locking.
		 *	- return a value
		 */
		case SIOCGIFFLAGS:
		case SIOCGIFMETRIC:
		case SIOCGIFMTU:
		case SIOCGIFHWADDR:
		case SIOCGIFSLAVE:
		case SIOCGIFMAP:
		case SIOCGIFINDEX:
		case SIOCGIFTXQLEN:
Linus Torvalds's avatar
Linus Torvalds committed
			read_lock(&dev_base_lock);
			ret = dev_ifsioc(net, &ifr, cmd);
Linus Torvalds's avatar
Linus Torvalds committed
			read_unlock(&dev_base_lock);
			if (!ret) {
				if (colon)
					*colon = ':';
				if (copy_to_user(arg, &ifr,
						 sizeof(struct ifreq)))
					ret = -EFAULT;
			}
			return ret;

		case SIOCETHTOOL:
Linus Torvalds's avatar
Linus Torvalds committed
			rtnl_lock();
Linus Torvalds's avatar
Linus Torvalds committed
			rtnl_unlock();
			if (!ret) {
				if (colon)
					*colon = ':';
				if (copy_to_user(arg, &ifr,
						 sizeof(struct ifreq)))
					ret = -EFAULT;
			}
			return ret;

		/*
		 *	These ioctl calls:
		 *	- require superuser power.
		 *	- require strict serialization.
		 *	- return a value
		 */
		case SIOCGMIIPHY:
		case SIOCGMIIREG:
		case SIOCSIFNAME:
			if (!capable(CAP_NET_ADMIN))
				return -EPERM;
Linus Torvalds's avatar
Linus Torvalds committed
			rtnl_lock();
			ret = dev_ifsioc(net, &ifr, cmd);
Linus Torvalds's avatar
Linus Torvalds committed
			rtnl_unlock();
			if (!ret) {
				if (colon)
					*colon = ':';
				if (copy_to_user(arg, &ifr,
						 sizeof(struct ifreq)))
					ret = -EFAULT;
			}
			return ret;

		/*
		 *	These ioctl calls:
		 *	- require superuser power.
		 *	- require strict serialization.
		 *	- do not return a value
		 */
		case SIOCSIFFLAGS:
		case SIOCSIFMETRIC:
		case SIOCSIFMTU:
		case SIOCSIFMAP:
		case SIOCSIFHWADDR:
		case SIOCSIFSLAVE:
		case SIOCADDMULTI:
		case SIOCDELMULTI:
		case SIOCSIFHWBROADCAST:
		case SIOCSIFTXQLEN:
		case SIOCSMIIREG:
		case SIOCBONDENSLAVE:
		case SIOCBONDRELEASE:
		case SIOCBONDSETHWADDR:
		case SIOCBONDCHANGEACTIVE:
		case SIOCBRADDIF:
		case SIOCBRDELIF:
			if (!capable(CAP_NET_ADMIN))
				return -EPERM;
			/* fall through */
		case SIOCBONDSLAVEINFOQUERY:
		case SIOCBONDINFOQUERY:
Linus Torvalds's avatar
Linus Torvalds committed
			rtnl_lock();
			ret = dev_ifsioc(net, &ifr, cmd);
Linus Torvalds's avatar
Linus Torvalds committed
			rtnl_unlock();
			return ret;

		case SIOCGIFMEM:
			/* Get the per device memory space. We can add this but
			 * currently do not support it */
		case SIOCSIFMEM:
			/* Set the per device memory buffer space.
			 * Not applicable in our case */
		case SIOCSIFLINK:
			return -EINVAL;

		/*
		 *	Unknown or private ioctl.
		 */
		default:
			if (cmd == SIOCWANDEV ||
			    (cmd >= SIOCDEVPRIVATE &&
			     cmd <= SIOCDEVPRIVATE + 15)) {
Linus Torvalds's avatar
Linus Torvalds committed
				rtnl_lock();
				ret = dev_ifsioc(net, &ifr, cmd);
Linus Torvalds's avatar
Linus Torvalds committed
				rtnl_unlock();
				if (!ret && copy_to_user(arg, &ifr,
							 sizeof(struct ifreq)))
					ret = -EFAULT;
				return ret;
			}
			/* Take care of Wireless Extensions */
			if (cmd >= SIOCIWFIRST && cmd <= SIOCIWLAST)
				return wext_handle_ioctl(net, &ifr, cmd, arg);
Linus Torvalds's avatar
Linus Torvalds committed
			return -EINVAL;
	}
}


/**
 *	dev_new_index	-	allocate an ifindex
 *
 *	Returns a suitable unique value for a new device interface
 *	number.  The caller must hold the rtnl semaphore or the
 *	dev_base_lock to be sure it remains unique.
 */
static int dev_new_index(struct net *net)
Linus Torvalds's avatar
Linus Torvalds committed
{
	static int ifindex;
	for (;;) {
		if (++ifindex <= 0)
			ifindex = 1;
		if (!__dev_get_by_index(net, ifindex))
Linus Torvalds's avatar
Linus Torvalds committed
			return ifindex;
	}
}

/* Delayed registration/unregisteration */
static DEFINE_SPINLOCK(net_todo_list_lock);
static struct list_head net_todo_list = LIST_HEAD_INIT(net_todo_list);

static void net_set_todo(struct net_device *dev)
Linus Torvalds's avatar
Linus Torvalds committed
{
	spin_lock(&net_todo_list_lock);
	list_add_tail(&dev->todo_list, &net_todo_list);
	spin_unlock(&net_todo_list_lock);
}

/**
 *	register_netdevice	- register a network device
 *	@dev: device to register
 *
 *	Take a completed network device structure and add it to the kernel
 *	interfaces. A %NETDEV_REGISTER message is sent to the netdev notifier
 *	chain. 0 is returned on success. A negative errno code is returned
 *	on a failure to set up the device, or if the name is a duplicate.
 *
 *	Callers must hold the rtnl semaphore. You may want
 *	register_netdev() instead of this.
 *
 *	BUGS:
 *	The locking appears insufficient to guarantee two parallel registers
 *	will not get the same name.
 */

int register_netdevice(struct net_device *dev)
{
	struct hlist_head *head;
	struct hlist_node *p;
	int ret;
Linus Torvalds's avatar
Linus Torvalds committed

	BUG_ON(dev_boot_phase);
	ASSERT_RTNL();

Linus Torvalds's avatar
Linus Torvalds committed
	/* When net_device's are persistent, this will be fatal. */
	BUG_ON(dev->reg_state != NETREG_UNINITIALIZED);
	BUG_ON(!dev->nd_net);
	net = dev->nd_net;
Linus Torvalds's avatar
Linus Torvalds committed

	spin_lock_init(&dev->queue_lock);
Herbert Xu's avatar
Herbert Xu committed
	spin_lock_init(&dev->_xmit_lock);
	netdev_set_lockdep_class(&dev->_xmit_lock, dev->type);
Linus Torvalds's avatar
Linus Torvalds committed
	dev->xmit_lock_owner = -1;
	spin_lock_init(&dev->ingress_lock);

	dev->iflink = -1;

	/* Init, if this function is available */
	if (dev->init) {
		ret = dev->init(dev);
		if (ret) {
			if (ret > 0)
				ret = -EIO;
Linus Torvalds's avatar
Linus Torvalds committed
		}
	}
Linus Torvalds's avatar
Linus Torvalds committed
	if (!dev_valid_name(dev->name)) {
		ret = -EINVAL;
	dev->ifindex = dev_new_index(net);
Linus Torvalds's avatar
Linus Torvalds committed
	if (dev->iflink == -1)
		dev->iflink = dev->ifindex;

	/* Check for existence of name */
	head = dev_name_hash(net, dev->name);
Linus Torvalds's avatar
Linus Torvalds committed
	hlist_for_each(p, head) {
		struct net_device *d
			= hlist_entry(p, struct net_device, name_hlist);
		if (!strncmp(d->name, dev->name, IFNAMSIZ)) {
			ret = -EEXIST;
Linus Torvalds's avatar
Linus Torvalds committed
		}
Linus Torvalds's avatar
Linus Torvalds committed

	/* Fix illegal checksum combinations */
	if ((dev->features & NETIF_F_HW_CSUM) &&
	    (dev->features & (NETIF_F_IP_CSUM|NETIF_F_IPV6_CSUM))) {
		printk(KERN_NOTICE "%s: mixed HW and IP checksum settings.\n",
		       dev->name);
		dev->features &= ~(NETIF_F_IP_CSUM|NETIF_F_IPV6_CSUM);
	}

	if ((dev->features & NETIF_F_NO_CSUM) &&
	    (dev->features & (NETIF_F_HW_CSUM|NETIF_F_IP_CSUM|NETIF_F_IPV6_CSUM))) {
		printk(KERN_NOTICE "%s: mixed no checksumming and other settings.\n",
		       dev->name);
		dev->features &= ~(NETIF_F_IP_CSUM|NETIF_F_IPV6_CSUM|NETIF_F_HW_CSUM);
	}


Linus Torvalds's avatar
Linus Torvalds committed
	/* Fix illegal SG+CSUM combinations. */
	if ((dev->features & NETIF_F_SG) &&
	    !(dev->features & NETIF_F_ALL_CSUM)) {
		printk(KERN_NOTICE "%s: Dropping NETIF_F_SG since no checksum feature.\n",
Linus Torvalds's avatar
Linus Torvalds committed
		       dev->name);
		dev->features &= ~NETIF_F_SG;
	}

	/* TSO requires that SG is present as well. */
	if ((dev->features & NETIF_F_TSO) &&
	    !(dev->features & NETIF_F_SG)) {
		printk(KERN_NOTICE "%s: Dropping NETIF_F_TSO since no SG feature.\n",
Linus Torvalds's avatar
Linus Torvalds committed
		       dev->name);
		dev->features &= ~NETIF_F_TSO;
	}
	if (dev->features & NETIF_F_UFO) {
		if (!(dev->features & NETIF_F_HW_CSUM)) {
			printk(KERN_ERR "%s: Dropping NETIF_F_UFO since no "
					"NETIF_F_HW_CSUM feature.\n",
							dev->name);
			dev->features &= ~NETIF_F_UFO;
		}
		if (!(dev->features & NETIF_F_SG)) {
			printk(KERN_ERR "%s: Dropping NETIF_F_UFO since no "
					"NETIF_F_SG feature.\n",
					dev->name);
			dev->features &= ~NETIF_F_UFO;
		}
	}
Linus Torvalds's avatar
Linus Torvalds committed

	/*
	 *	nil rebuild_header routine,
	 *	that should be never called and used as just bug trap.
	 */

	if (!dev->rebuild_header)
		dev->rebuild_header = default_rebuild_header;

	ret = netdev_register_sysfs(dev);
	if (ret)
	dev->reg_state = NETREG_REGISTERED;

Linus Torvalds's avatar
Linus Torvalds committed
	/*
	 *	Default initial state at registry is that the
	 *	device is present.
	 */

	set_bit(__LINK_STATE_PRESENT, &dev->state);

	dev_init_scheduler(dev);
	write_lock_bh(&dev_base_lock);
	list_add_tail(&dev->dev_list, &net->dev_base_head);
Linus Torvalds's avatar
Linus Torvalds committed
	hlist_add_head(&dev->name_hlist, head);
	hlist_add_head(&dev->index_hlist, dev_index_hash(net, dev->ifindex));
Linus Torvalds's avatar
Linus Torvalds committed
	dev_hold(dev);
	write_unlock_bh(&dev_base_lock);

	/* Notify protocols, that a new device appeared. */
	ret = raw_notifier_call_chain(&netdev_chain, NETDEV_REGISTER, dev);
	ret = notifier_to_errno(ret);
	if (ret)
		unregister_netdevice(dev);
Linus Torvalds's avatar
Linus Torvalds committed

out:
	return ret;

err_uninit:
	if (dev->uninit)
		dev->uninit(dev);
	goto out;
Linus Torvalds's avatar
Linus Torvalds committed
}

/**
 *	register_netdev	- register a network device
 *	@dev: device to register
 *
 *	Take a completed network device structure and add it to the kernel
 *	interfaces. A %NETDEV_REGISTER message is sent to the netdev notifier
 *	chain. 0 is returned on success. A negative errno code is returned
 *	on a failure to set up the device, or if the name is a duplicate.
 *
 *	This is a wrapper around register_netdevice that takes the rtnl semaphore
Linus Torvalds's avatar
Linus Torvalds committed
 *	and expands the device name if you passed a format string to
 *	alloc_netdev.
 */
int register_netdev(struct net_device *dev)
{
	int err;

	rtnl_lock();

	/*
	 * If the name is a format string the caller wants us to do a
	 * name allocation.
	 */
	if (strchr(dev->name, '%')) {
		err = dev_alloc_name(dev, dev->name);
		if (err < 0)
			goto out;
	}
Linus Torvalds's avatar
Linus Torvalds committed
	err = register_netdevice(dev);
out:
	rtnl_unlock();
	return err;
}
EXPORT_SYMBOL(register_netdev);

/*
 * netdev_wait_allrefs - wait until all references are gone.
 *
 * This is called when unregistering network devices.
 *
 * Any protocol or device that holds a reference should register
 * for netdevice notification, and cleanup and put back the
 * reference if they receive an UNREGISTER event.
 * We can get stuck here if buggy protocols don't correctly
 * call dev_put.
Linus Torvalds's avatar
Linus Torvalds committed
 */
static void netdev_wait_allrefs(struct net_device *dev)
{
	unsigned long rebroadcast_time, warning_time;

	rebroadcast_time = warning_time = jiffies;
	while (atomic_read(&dev->refcnt) != 0) {
		if (time_after(jiffies, rebroadcast_time + 1 * HZ)) {
			rtnl_lock();
Linus Torvalds's avatar
Linus Torvalds committed

			/* Rebroadcast unregister notification */
			raw_notifier_call_chain(&netdev_chain,
Linus Torvalds's avatar
Linus Torvalds committed
					    NETDEV_UNREGISTER, dev);

			if (test_bit(__LINK_STATE_LINKWATCH_PENDING,
				     &dev->state)) {
				/* We must not have linkwatch events
				 * pending on unregister. If this
				 * happens, we simply run the queue
				 * unscheduled, resulting in a noop
				 * for this device.
				 */
				linkwatch_run_queue();
			}

			__rtnl_unlock();
Linus Torvalds's avatar
Linus Torvalds committed

			rebroadcast_time = jiffies;
		}

		msleep(250);

		if (time_after(jiffies, warning_time + 10 * HZ)) {
			printk(KERN_EMERG "unregister_netdevice: "
			       "waiting for %s to become free. Usage "
			       "count = %d\n",
			       dev->name, atomic_read(&dev->refcnt));
			warning_time = jiffies;
		}
	}
}

/* The sequence is:
 *
 *	rtnl_lock();
 *	...
 *	register_netdevice(x1);
 *	register_netdevice(x2);
 *	...
 *	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
Linus Torvalds's avatar
Linus Torvalds committed
 *    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.
 */
Arjan van de Ven's avatar
Arjan van de Ven committed
static DEFINE_MUTEX(net_todo_run_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
void netdev_run_todo(void)
{
Linus Torvalds's avatar
Linus Torvalds committed

	/* Need to guard against multiple cpu's getting out of order. */
Arjan van de Ven's avatar
Arjan van de Ven committed
	mutex_lock(&net_todo_run_mutex);
Linus Torvalds's avatar
Linus Torvalds committed

	/* 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);
Linus Torvalds's avatar
Linus Torvalds committed
	spin_unlock(&net_todo_list_lock);
Linus Torvalds's avatar
Linus Torvalds committed
	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;
		}
Linus Torvalds's avatar
Linus Torvalds committed

		dev->reg_state = NETREG_UNREGISTERED;
Linus Torvalds's avatar
Linus Torvalds committed

Linus Torvalds's avatar
Linus Torvalds committed

		/* paranoia */
		BUG_ON(atomic_read(&dev->refcnt));
		BUG_TRAP(!dev->ip_ptr);
		BUG_TRAP(!dev->ip6_ptr);
		BUG_TRAP(!dev->dn_ptr);
Linus Torvalds's avatar
Linus Torvalds committed

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

		/* Free network device */
		kobject_put(&dev->dev.kobj);
Arjan van de Ven's avatar
Arjan van de Ven committed
	mutex_unlock(&net_todo_run_mutex);
static struct net_device_stats *internal_stats(struct net_device *dev)
Linus Torvalds's avatar
Linus Torvalds committed
/**
 *	alloc_netdev_mq - allocate network device
Linus Torvalds's avatar
Linus Torvalds committed
 *	@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
Linus Torvalds's avatar
Linus Torvalds committed
 *
 *	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.
Linus Torvalds's avatar
Linus Torvalds committed
 */
struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name,
		void (*setup)(struct net_device *), unsigned int queue_count)
Linus Torvalds's avatar
Linus Torvalds committed
{
	void *p;
	struct net_device *dev;
	int alloc_size;

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

Linus Torvalds's avatar
Linus Torvalds committed
	/* ensure 32-byte alignment of both the device and private area */
	alloc_size = (sizeof(*dev) + NETDEV_ALIGN_CONST +
		     (sizeof(struct net_device_subqueue) * (queue_count - 1))) &
Linus Torvalds's avatar
Linus Torvalds committed
	alloc_size += sizeof_priv + NETDEV_ALIGN_CONST;

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

	dev = (struct net_device *)
		(((long)p + NETDEV_ALIGN_CONST) & ~NETDEV_ALIGN_CONST);
	dev->padded = (char *)dev - (char *)p;
Linus Torvalds's avatar
Linus Torvalds committed

	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;
Linus Torvalds's avatar
Linus Torvalds committed

	dev->get_stats = internal_stats;
Linus Torvalds's avatar
Linus Torvalds committed
	setup(dev);
	strcpy(dev->name, name);
	return dev;
}
Linus Torvalds's avatar
Linus Torvalds committed

/**
 *	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.
Linus Torvalds's avatar
Linus Torvalds committed
 *	If this is the last reference then it will be freed.
 */
void free_netdev(struct net_device *dev)
{
#ifdef CONFIG_SYSFS
	/*  Compatibility with error handling in drivers */
Linus Torvalds's avatar
Linus Torvalds committed
	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);
Linus Torvalds's avatar
Linus Torvalds committed
#else
	kfree((char *)dev - dev->padded);
#endif
}
Linus Torvalds's avatar
Linus Torvalds committed
/* Synchronize with packet receive processing. */
void synchronize_net(void)
Linus Torvalds's avatar
Linus Torvalds committed
{
	might_sleep();
Linus Torvalds's avatar
Linus Torvalds committed
}

/**
 *	unregister_netdevice - remove device from the kernel
 *	@dev: device
 *
 *	This function shuts down a device interface and removes it
 *	from the kernel tables. On success 0 is returned, on a failure
 *	a negative errno code is returned.
 *
 *	Callers must hold the rtnl semaphore.  You may want
 *	unregister_netdev() instead of this.
 */

void unregister_netdevice(struct net_device *dev)
Linus Torvalds's avatar
Linus Torvalds committed
{
	BUG_ON(dev_boot_phase);
	ASSERT_RTNL();

	/* Some devices call without registering for initialization unwind. */
	if (dev->reg_state == NETREG_UNINITIALIZED) {
		printk(KERN_DEBUG "unregister_netdevice: device %s/%p never "
				  "was registered\n", dev->name, dev);
Linus Torvalds's avatar
Linus Torvalds committed
	}

	BUG_ON(dev->reg_state != NETREG_REGISTERED);

	/* If device is running, close it first. */
	if (dev->flags & IFF_UP)
		dev_close(dev);

	/* And unlink it from device chain. */
	write_lock_bh(&dev_base_lock);
	list_del(&dev->dev_list);
	hlist_del(&dev->name_hlist);
	hlist_del(&dev->index_hlist);
	write_unlock_bh(&dev_base_lock);
Linus Torvalds's avatar
Linus Torvalds committed

	dev->reg_state = NETREG_UNREGISTERING;

	synchronize_net();

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

Linus Torvalds's avatar
Linus Torvalds committed
	/* Notify protocols, that we are about to destroy
	   this device. They should clean all the things.
	*/
	raw_notifier_call_chain(&netdev_chain, NETDEV_UNREGISTER, dev);
Linus Torvalds's avatar
Linus Torvalds committed
	/*
	 *	Flush the unicast and multicast chains
Linus Torvalds's avatar
Linus Torvalds committed
	 */
Linus Torvalds's avatar
Linus Torvalds committed

	if (dev->uninit)
		dev->uninit(dev);

	/* Notifier chain MUST detach us from master device. */
	BUG_TRAP(!dev->master);

	/* Remove entries from sysfs */
	netdev_unregister_sysfs(dev);

Linus Torvalds's avatar
Linus Torvalds committed
	/* Finish processing unregister after unlock */
	net_set_todo(dev);

	synchronize_net();

	dev_put(dev);
}

/**
 *	unregister_netdev - remove device from the kernel
 *	@dev: device
 *
 *	This function shuts down a device interface and removes it
 *	from the kernel tables. On success 0 is returned, on a failure
 *	a negative errno code is returned.
 *
 *	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);

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)
Linus Torvalds's avatar
Linus Torvalds committed
		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;
}

Randy Dunlap's avatar
Randy Dunlap committed
 * 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.