bonding.c

linux和2410结合开发 用他可以生成2410所需的zImage文件

/*
 * originally based on the dummy device.
 *
 * Copyright 1999, Thomas Davis, [e-mail].  
 * Licensed under the GPL. Based on dummy.c, and eql.c devices.
 *
 * bonding.c: an Ethernet Bonding driver
 *
 * This is useful to talk to a Cisco EtherChannel compatible equipment:
 *	Cisco 5500
 *	Sun Trunking (Solaris)
 *	Alteon AceDirector Trunks
 *	Linux Bonding
 *	and probably many L2 switches ...
 *
 * How it works:
 *    ifconfig bond0 ipaddress netmask up
 *      will setup a network device, with an ip address.  No mac address 
 *	will be assigned at this time.  The hw mac address will come from 
 *	the first slave bonded to the channel.  All slaves will then use 
 *	this hw mac address.
 *
 *    ifconfig bond0 down
 *         will release all slaves, marking them as down.
 *
 *    ifenslave bond0 eth0
 *	will attach eth0 to bond0 as a slave.  eth0 hw mac address will either
 *	a: be used as initial mac address
 *	b: if a hw mac address already is there, eth0's hw mac address 
 *	   will then be set from bond0.
 *
 * v0.1 - first working version.
 * v0.2 - changed stats to be calculated by summing slaves stats.
 *
 * Changes:
 * Arnaldo Carvalho de Melo <[e-mail]>
 * - fix leaks on failure at bond_init
 *
 * 2000/09/30 - Willy Tarreau <willy at meta-x.org>
 *     - added trivial code to release a slave device.
 *     - fixed security bug (CAP_NET_ADMIN not checked)
 *     - implemented MII link monitoring to disable dead links :
 *       All MII capable slaves are checked every <miimon> milliseconds
 *       (100 ms seems good). This value can be changed by passing it to
 *       insmod. A value of zero disables the monitoring (default).
 *     - fixed an infinite loop in bond_xmit_roundrobin() when there's no
 *       good slave.
 *     - made the code hopefully SMP safe
 *
 * 2000/10/03 - Willy Tarreau <willy at meta-x.org>
 *     - optimized slave lists based on relevant suggestions from Thomas Davis
 *     - implemented active-backup method to obtain HA with two switches:
 *       stay as long as possible on the same active interface, while we
 *       also monitor the backup one (MII link status) because we want to know
 *       if we are able to switch at any time. ( pass "mode=1" to insmod )
 *     - lots of stress testings because we need it to be more robust than the
 *       wires ! :->
 *
 * 2000/10/09 - Willy Tarreau <willy at meta-x.org>
 *     - added up and down delays after link state change.
 *     - optimized the slaves chaining so that when we run forward, we never
 *       repass through the bond itself, but we can find it by searching
 *       backwards. Renders the deletion more difficult, but accelerates the
 *       scan.
 *     - smarter enslaving and releasing.
 *     - finer and more robust SMP locking
 *
 * 2000/10/17 - Willy Tarreau <willy at meta-x.org>
 *     - fixed two potential SMP race conditions
 *
 * 2000/10/18 - Willy Tarreau <willy at meta-x.org>
 *     - small fixes to the monitoring FSM in case of zero delays
 * 2000/11/01 - Willy Tarreau <willy at meta-x.org>
 *     - fixed first slave not automatically used in trunk mode.
 * 2000/11/10 : spelling of "EtherChannel" corrected.
 * 2000/11/13 : fixed a race condition in case of concurrent accesses to ioctl().
 * 2000/12/16 : fixed improper usage of rtnl_exlock_nowait().
 *
 * 2001/1/3 - Chad N. Tindel <ctindel at ieee dot org>
 *     - The bonding driver now simulates MII status monitoring, just like
 *       a normal network device.  It will show that the link is down iff
 *       every slave in the bond shows that their links are down.  If at least
 *       one slave is up, the bond's MII status will appear as up.
 *
 * 2001/2/7 - Chad N. Tindel <ctindel at ieee dot org>
 *     - Applications can now query the bond from user space to get
 *       information which may be useful.  They do this by calling
 *       the BOND_INFO_QUERY ioctl.  Once the app knows how many slaves
 *       are in the bond, it can call the BOND_SLAVE_INFO_QUERY ioctl to
 *       get slave specific information (# link failures, etc).  See
 *       <linux/if_bonding.h> for more details.  The structs of interest
 *       are ifbond and ifslave.
 *
 * 2001/4/5 - Chad N. Tindel <ctindel at ieee dot org>
 *     - Ported to 2.4 Kernel
 * 
 * 2001/5/2 - Jeffrey E. Mast <jeff at mastfamily dot com>
 *     - When a device is detached from a bond, the slave device is no longer
 *       left thinking that is has a master.
 *
 * 2001/5/16 - Jeffrey E. Mast <jeff at mastfamily dot com>
 *     - memset did not appropriately initialized the bond rw_locks. Used 
 *       rwlock_init to initialize to unlocked state to prevent deadlock when 
 *       first attempting a lock
 *     - Called SET_MODULE_OWNER for bond device
 *
 * 2001/5/17 - Tim Anderson <tsa at mvista.com>
 *     - 2 paths for releasing for slave release; 1 through ioctl
 *       and 2) through close. Both paths need to release the same way.
 *     - the free slave in bond release is changing slave status before
 *       the free. The netdev_set_master() is intended to change slave state
 *       so it should not be done as part of the release process.
 *     - Simple rule for slave state at release: only the active in A/B and
 *       only one in the trunked case.
 *
 * 2001/6/01 - Tim Anderson <tsa at mvista.com>
 *     - Now call dev_close when releasing a slave so it doesn't screw up
 *       out routing table.
 *
 * 2001/6/01 - Chad N. Tindel <ctindel at ieee dot org>
 *     - Added /proc support for getting bond and slave information.
 *       Information is in /proc/net/<bond device>/info. 
 *     - Changed the locking when calling bond_close to prevent deadlock.
 *
 * 2001/8/05 - Janice Girouard <girouard at us.ibm.com>
 *     - correct problem where refcnt of slave is not incremented in bond_ioctl
 *       so the system hangs when halting.
 *     - correct locking problem when unable to malloc in bond_enslave.
 *     - adding bond_xmit_xor logic.
 *     - adding multiple bond device support.
 *
 * 2001/8/13 - Erik Habbinga <erik_habbinga at hp dot com>
 *     - correct locking problem with rtnl_exlock_nowait
 *
 * 2001/8/23 - Janice Girouard <girouard at us.ibm.com>
 *     - bzero initial dev_bonds, to correct oops
 *     - convert SIOCDEVPRIVATE to new MII ioctl calls
 *
 * 2001/9/13 - Takao Indoh <indou dot takao at jp dot fujitsu dot com>
 *     - Add the BOND_CHANGE_ACTIVE ioctl implementation
 *
 * 2001/9/14 - Mark Huth <mhuth at mvista dot com>
 *     - Change MII_LINK_READY to not check for end of auto-negotiation,
 *       but only for an up link.
 *
 * 2001/9/20 - Chad N. Tindel <ctindel at ieee dot org>
 *     - Add the device field to bonding_t.  Previously the net_device 
 *       corresponding to a bond wasn't available from the bonding_t 
 *       structure.
 *
 * 2001/9/25 - Janice Girouard <girouard at us.ibm.com>
 *     - add arp_monitor for active backup mode
 *
 * 2001/10/23 - Takao Indoh <indou dot takao at jp dot fujitsu dot com>
 *     - Various memory leak fixes
 *
 * 2001/11/5 - Mark Huth <mark dot huth at mvista dot com>
 *     - Don't take rtnl lock in bond_mii_monitor as it deadlocks under 
 *       certain hotswap conditions.  
 *       Note:  this same change may be required in bond_arp_monitor ???
 *     - Remove possibility of calling bond_sethwaddr with NULL slave_dev ptr 
 *     - Handle hot swap ethernet interface deregistration events to remove
 *       kernel oops following hot swap of enslaved interface
 */

#include <linux/config.h>
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/sched.h>
#include <linux/types.h>
#include <linux/fcntl.h>
#include <linux/interrupt.h>
#include <linux/ptrace.h>
#include <linux/ioport.h>
#include <linux/in.h>
#include <linux/slab.h>
#include <linux/string.h>
#include <linux/init.h>
#include <linux/timer.h>
#include <linux/socket.h>
#include <asm/system.h>
#include <asm/bitops.h>
#include <asm/io.h>
#include <asm/dma.h>
#include <asm/uaccess.h>
#include <linux/errno.h>

#include <linux/netdevice.h>
#include <linux/etherdevice.h>
#include <linux/skbuff.h>
#include <net/sock.h>
#include <linux/rtnetlink.h>

#include <linux/if_bonding.h>
#include <linux/smp.h>
#include <linux/if_ether.h>
#include <linux/if_arp.h>

/* monitor all links that often (in milliseconds). <=0 disables monitoring */
#ifndef BOND_LINK_MON_INTERV
#define BOND_LINK_MON_INTERV	0
#endif

#undef  MII_LINK_UP
#define MII_LINK_UP	0x04

#undef  MII_ENDOF_NWAY
#define MII_ENDOF_NWAY	0x20

#undef  MII_LINK_READY
/*#define MII_LINK_READY	(MII_LINK_UP | MII_ENDOF_NWAY)*/
#define MII_LINK_READY	(MII_LINK_UP)

#define MAX_BOND_ADDR 256

#ifndef BOND_LINK_ARP_INTERV
#define BOND_LINK_ARP_INTERV	0
#endif

static int arp_interval = BOND_LINK_ARP_INTERV;
static char *arp_ip_target = NULL;
static unsigned long arp_target = 0;
static u32 my_ip = 0;
char *arp_target_hw_addr = NULL;

static int max_bonds	= MAX_BONDS;
static int miimon	= BOND_LINK_MON_INTERV;
static int mode		= BOND_MODE_ROUNDROBIN;
static int updelay	= 0;
static int downdelay	= 0;

static int first_pass	= 1;
int bond_cnt;
static struct bonding *these_bonds =  NULL;
static struct net_device *dev_bonds = NULL;

MODULE_PARM(max_bonds, "1-" __MODULE_STRING(INT_MAX) "i");
MODULE_PARM_DESC(max_bonds, "Max number of bonded devices");
MODULE_PARM(miimon, "i");
MODULE_PARM_DESC(miimon, "Link check interval in milliseconds");
MODULE_PARM(mode, "i");
MODULE_PARM(arp_interval, "i");
MODULE_PARM_DESC(arp_interval, "arp interval in milliseconds");
MODULE_PARM(arp_ip_target, "1-12s");
MODULE_PARM_DESC(arp_ip_target, "arp target in n.n.n.n form");
MODULE_PARM_DESC(mode, "Mode of operation : 0 for round robin, 1 for active-backup, 2 for xor");
MODULE_PARM(updelay, "i");
MODULE_PARM_DESC(updelay, "Delay before considering link up, in milliseconds");
MODULE_PARM(downdelay, "i");
MODULE_PARM_DESC(downdelay, "Delay before considering link down, in milliseconds");

extern void arp_send( int type, int ptype, u32 dest_ip, struct net_device *dev,
	u32 src_ip, unsigned char *dest_hw, unsigned char *src_hw, 
	unsigned char *target_hw);

static int bond_xmit_roundrobin(struct sk_buff *skb, struct net_device *dev);
static int bond_xmit_xor(struct sk_buff *skb, struct net_device *dev);
static int bond_xmit_activebackup(struct sk_buff *skb, struct net_device *dev);
static struct net_device_stats *bond_get_stats(struct net_device *dev);
static void bond_mii_monitor(struct net_device *dev);
static void bond_arp_monitor(struct net_device *dev);
static int bond_event(struct notifier_block *this, unsigned long event, void *ptr);
static void bond_set_slave_inactive_flags(slave_t *slave);
static void bond_set_slave_active_flags(slave_t *slave);
static int bond_enslave(struct net_device *master, struct net_device *slave);
static int bond_release(struct net_device *master, struct net_device *slave);
static int bond_release_all(struct net_device *master);
static int bond_sethwaddr(struct net_device *master, struct net_device *slave);

/*
 * bond_get_info is the interface into the /proc filesystem.  This is
 * a different interface than the BOND_INFO_QUERY ioctl.  That is done
 * through the generic networking ioctl interface, and bond_info_query
 * is the internal function which provides that information.
 */
static int bond_get_info(char *buf, char **start, off_t offset, int length);

/* #define BONDING_DEBUG 1 */

/* several macros */

#define IS_UP(dev)	((((dev)->flags & (IFF_UP)) == (IFF_UP)) && \
			(netif_running(dev) && netif_carrier_ok(dev)))

static void bond_set_slave_inactive_flags(slave_t *slave)
{
	slave->state = BOND_STATE_BACKUP;
	slave->dev->flags |= IFF_NOARP;
}

static void bond_set_slave_active_flags(slave_t *slave)
{
	slave->state = BOND_STATE_ACTIVE;
	slave->dev->flags &= ~IFF_NOARP;
}

/* 
 * This function detaches the slave <slave> from the list <bond>.
 * WARNING: no check is made to verify if the slave effectively
 * belongs to <bond>. It returns <slave> in case it's needed.
 * Nothing is freed on return, structures are just unchained.
 * If the bond->current_slave pointer was pointing to <slave>,
 * it's replaced with slave->next, or <bond> if not applicable.
 */
static slave_t *bond_detach_slave(bonding_t *bond, slave_t *slave)
{
	if ((bond == NULL) || (slave == NULL) ||
	   ((void *)bond == (void *)slave)) {
		printk(KERN_ERR
			"bond_detach_slave(): trying to detach "
			"slave %p from bond %p\n", bond, slave);
		return slave;
	}

	if (bond->next == slave) {  /* is the slave at the head ? */
		if (bond->prev == slave) {  /* is the slave alone ? */
			write_lock(&bond->ptrlock);
			bond->current_slave = NULL; /* no slave anymore */
			write_unlock(&bond->ptrlock);
			bond->prev = bond->next = (slave_t *)bond;
		} else { /* not alone */
			bond->next        = slave->next;
			slave->next->prev = (slave_t *)bond;
			bond->prev->next  = slave->next;

			write_lock(&bond->ptrlock);
			if (bond->current_slave == slave) {
				bond->current_slave = slave->next;
			}
			write_unlock(&bond->ptrlock);
		}
	}
	else {
		slave->prev->next = slave->next;
		if (bond->prev == slave) {  /* is this slave the last one ? */
			bond->prev = slave->prev;
		} else {
			slave->next->prev = slave->prev;
		}

		write_lock(&bond->ptrlock);
		if (bond->current_slave == slave) {
			bond->current_slave = slave->next;
		}
		write_unlock(&bond->ptrlock);
	}

	return slave;
}

/* 
 * if <dev> supports MII link status reporting, check its link
 * and report it as a bit field in a short int :
 *   - 0x04 means link is up,
 *   - 0x20 means end of autonegociation
 * If the device doesn't support MII, then we only report 0x24,
 * meaning that the link is up and running since we can't check it.
 */
static u16 bond_check_dev_link(struct net_device *dev)
{
	static int (* ioctl)(struct net_device *, struct ifreq *, int);
	struct ifreq ifr;
	u16 *data = (u16 *)&ifr.ifr_data;
		
	/* data[0] automagically filled by the ioctl */
	data[1] = 1; /* MII location 1 reports Link Status */

	if (((ioctl = dev->do_ioctl) != NULL) &&  /* ioctl to access MII */
	    (ioctl(dev, &ifr, SIOCGMIIPHY) == 0)) {
		/* now, data[3] contains info about link status :
		   - data[3] & 0x04 means link up
		   - data[3] & 0x20 means end of auto-negociation
		*/
		return data[3];
	} else {
		return MII_LINK_READY;  /* spoof link up ( we can't check it) */
	}
}

static u16 bond_check_mii_link(bonding_t *bond)
{
	int has_active_interface = 0;
	unsigned long flags;

	read_lock_irqsave(&bond->lock, flags);
	read_lock(&bond->ptrlock);
	has_active_interface = (bond->current_slave != NULL);
	read_unlock(&bond->ptrlock);
	read_unlock_irqrestore(&bond->lock, flags);

	return (has_active_interface ? MII_LINK_READY : 0);
}

static int bond_open(struct net_device *dev)
{
	struct timer_list *timer = &((struct bonding *)(dev->priv))->mii_timer;
	struct timer_list *arp_timer = &((struct bonding *)(dev->priv))->arp_timer;
	MOD_INC_USE_COUNT;

	if (miimon > 0) {  /* link check interval, in milliseconds. */
		init_timer(timer);
		timer->expires  = jiffies + (miimon * HZ / 1000);
		timer->data     = (unsigned long)dev;
		timer->function = (void *)&bond_mii_monitor;
		add_timer(timer);
	}

	if (arp_interval> 0) {  /* arp interval, in milliseconds. */
		init_timer(arp_timer);
		arp_timer->expires  = jiffies + (arp_interval * HZ / 1000);
		arp_timer->data     = (unsigned long)dev;
		arp_timer->function = (void *)&bond_arp_monitor;
		add_timer(arp_timer);
	}
	return 0;
}

static int bond_close(struct net_device *master)
{
	bonding_t *bond = (struct bonding *) master->priv;
	unsigned long flags;

	write_lock_irqsave(&bond->lock, flags);

	if (miimon > 0) {  /* link check interval, in milliseconds. */
		del_timer(&bond->mii_timer);
	}
	if (arp_interval> 0) {  /* arp interval, in milliseconds. */
		del_timer(&bond->arp_timer);
	}

	/* Release the bonded slaves */
	bond_release_all(master);

	write_unlock_irqrestore(&bond->lock, flags);

	MOD_DEC_USE_COUNT;
	return 0;
}

static void set_multicast_list(struct net_device *master)
{
/*
	bonding_t *bond = master->priv;
	slave_t *slave;

	for (slave = bond->next; slave != (slave_t*)bond; slave = slave->next) {
		slave->dev->mc_list = master->mc_list;
		slave->dev->mc_count = master->mc_count;
		slave->dev->flags = master->flags;
		slave->dev->set_multicast_list(slave->dev);
	}
 */
}

/*
 * This function counts the the number of attached 
 * slaves for use by bond_xmit_xor.
 */
static void update_slave_cnt(bonding_t *bond)
{
	slave_t *slave = NULL;

	bond->slave_cnt = 0;
	for (slave = bond->prev; slave != (slave_t*)bond; slave = slave->prev) {
		bond->slave_cnt++;
	}
}

/* enslave device <slave> to bond device <master> */
static int bond_enslave(struct net_device *master_dev, 
                        struct net_device *slave_dev)
{
	bonding_t *bond = NULL;
	slave_t *new_slave = NULL;
	unsigned long flags = 0;
	int ndx = 0;
	int err = 0;

	if (master_dev == NULL || slave_dev == NULL) {
		return -ENODEV;
	}
	bond = (struct bonding *) master_dev->priv;

	if (slave_dev->do_ioctl == NULL) {
		printk(KERN_DEBUG
			"Warning : no link monitoring support for %s\n",
			slave_dev->name);
	}
	write_lock_irqsave(&bond->lock, flags);

	/* not running. */
	if ((slave_dev->flags & IFF_UP) != IFF_UP) {
#ifdef BONDING_DEBUG
		printk(KERN_CRIT "Error, slave_dev is not running\n");
#endif
		write_unlock_irqrestore(&bond->lock, flags);
		return -EINVAL;
	}

	/* already enslaved */
	if (master_dev->flags & IFF_SLAVE || slave_dev->flags & IFF_SLAVE) {
#ifdef BONDING_DEBUG
		printk(KERN_CRIT "Error, Device was already enslaved\n");
#endif
		write_unlock_irqrestore(&bond->lock, flags);
		return -EBUSY;
	}
		   
	if ((new_slave = kmalloc(sizeof(slave_t), GFP_KERNEL)) == NULL) {
		write_unlock_irqrestore(&bond->lock, flags);
		return -ENOMEM;
	}
	memset(new_slave, 0, sizeof(slave_t));

	err = netdev_set_master(slave_dev, master_dev);

	if (err) {
#ifdef BONDING_DEBUG
		printk(KERN_CRIT "Error %d calling netdev_set_master\n", err);
#endif
		kfree(new_slave);
		write_unlock_irqrestore(&bond->lock, flags);
		return err;      
	}

	new_slave->dev = slave_dev;