aboutsummaryrefslogblamecommitdiffstats
path: root/net/ipv4/ipvs/ip_vs_lblcr.c
blob: 32ba37ba72d855cadbd92979df7edf1734f388ca (plain) (tree)
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863








































                                                                             
                     

                         
                         






















































































































































































































































































                                                                                
                                      













                                           
                                                 













































































                                                                             




























































































































































































































































































































































































































































                                                                                 
/*
 * IPVS:        Locality-Based Least-Connection with Replication scheduler
 *
 * Version:     $Id: ip_vs_lblcr.c,v 1.11 2002/09/15 08:14:08 wensong Exp $
 *
 * Authors:     Wensong Zhang <wensong@gnuchina.org>
 *
 *              This program is free software; you can redistribute it and/or
 *              modify it under the terms of the GNU General Public License
 *              as published by the Free Software Foundation; either version
 *              2 of the License, or (at your option) any later version.
 *
 * Changes:
 *     Julian Anastasov        :    Added the missing (dest->weight>0)
 *                                  condition in the ip_vs_dest_set_max.
 *
 */

/*
 * The lblc/r algorithm is as follows (pseudo code):
 *
 *       if serverSet[dest_ip] is null then
 *               n, serverSet[dest_ip] <- {weighted least-conn node};
 *       else
 *               n <- {least-conn (alive) node in serverSet[dest_ip]};
 *               if (n is null) OR
 *                  (n.conns>n.weight AND
 *                   there is a node m with m.conns<m.weight/2) then
 *                   n <- {weighted least-conn node};
 *                   add n to serverSet[dest_ip];
 *               if |serverSet[dest_ip]| > 1 AND
 *                   now - serverSet[dest_ip].lastMod > T then
 *                   m <- {most conn node in serverSet[dest_ip]};
 *                   remove m from serverSet[dest_ip];
 *       if serverSet[dest_ip] changed then
 *               serverSet[dest_ip].lastMod <- now;
 *
 *       return n;
 *
 */

#include <linux/ip.h>
#include <linux/module.h>
#include <linux/kernel.h>
#include <linux/skbuff.h>

/* for sysctl */
#include <linux/fs.h>
#include <linux/sysctl.h>
/* for proc_net_create/proc_net_remove */
#include <linux/proc_fs.h>

#include <net/ip_vs.h>


/*
 *    It is for garbage collection of stale IPVS lblcr entries,
 *    when the table is full.
 */
#define CHECK_EXPIRE_INTERVAL   (60*HZ)
#define ENTRY_TIMEOUT           (6*60*HZ)

/*
 *    It is for full expiration check.
 *    When there is no partial expiration check (garbage collection)
 *    in a half hour, do a full expiration check to collect stale
 *    entries that haven't been touched for a day.
 */
#define COUNT_FOR_FULL_EXPIRATION   30
static int sysctl_ip_vs_lblcr_expiration = 24*60*60*HZ;


/*
 *     for IPVS lblcr entry hash table
 */
#ifndef CONFIG_IP_VS_LBLCR_TAB_BITS
#define CONFIG_IP_VS_LBLCR_TAB_BITS      10
#endif
#define IP_VS_LBLCR_TAB_BITS     CONFIG_IP_VS_LBLCR_TAB_BITS
#define IP_VS_LBLCR_TAB_SIZE     (1 << IP_VS_LBLCR_TAB_BITS)
#define IP_VS_LBLCR_TAB_MASK     (IP_VS_LBLCR_TAB_SIZE - 1)


/*
 *      IPVS destination set structure and operations
 */
struct ip_vs_dest_list {
	struct ip_vs_dest_list  *next;          /* list link */
	struct ip_vs_dest       *dest;          /* destination server */
};

struct ip_vs_dest_set {
	atomic_t                size;           /* set size */
	unsigned long           lastmod;        /* last modified time */
	struct ip_vs_dest_list  *list;          /* destination list */
	rwlock_t	        lock;           /* lock for this list */
};


static struct ip_vs_dest_list *
ip_vs_dest_set_insert(struct ip_vs_dest_set *set, struct ip_vs_dest *dest)
{
	struct ip_vs_dest_list *e;

	for (e=set->list; e!=NULL; e=e->next) {
		if (e->dest == dest)
			/* already existed */
			return NULL;
	}

	e = kmalloc(sizeof(struct ip_vs_dest_list), GFP_ATOMIC);
	if (e == NULL) {
		IP_VS_ERR("ip_vs_dest_set_insert(): no memory\n");
		return NULL;
	}

	atomic_inc(&dest->refcnt);
	e->dest = dest;

	/* link it to the list */
	write_lock(&set->lock);
	e->next = set->list;
	set->list = e;
	atomic_inc(&set->size);
	write_unlock(&set->lock);

	set->lastmod = jiffies;
	return e;
}

static void
ip_vs_dest_set_erase(struct ip_vs_dest_set *set, struct ip_vs_dest *dest)
{
	struct ip_vs_dest_list *e, **ep;

	write_lock(&set->lock);
	for (ep=&set->list, e=*ep; e!=NULL; e=*ep) {
		if (e->dest == dest) {
			/* HIT */
			*ep = e->next;
			atomic_dec(&set->size);
			set->lastmod = jiffies;
			atomic_dec(&e->dest->refcnt);
			kfree(e);
			break;
		}
		ep = &e->next;
	}
	write_unlock(&set->lock);
}

static void ip_vs_dest_set_eraseall(struct ip_vs_dest_set *set)
{
	struct ip_vs_dest_list *e, **ep;

	write_lock(&set->lock);
	for (ep=&set->list, e=*ep; e!=NULL; e=*ep) {
		*ep = e->next;
		/*
		 * We don't kfree dest because it is refered either
		 * by its service or by the trash dest list.
		 */
		atomic_dec(&e->dest->refcnt);
		kfree(e);
	}
	write_unlock(&set->lock);
}

/* get weighted least-connection node in the destination set */
static inline struct ip_vs_dest *ip_vs_dest_set_min(struct ip_vs_dest_set *set)
{
	register struct ip_vs_dest_list *e;
	struct ip_vs_dest *dest, *least;
	int loh, doh;

	if (set == NULL)
		return NULL;

	read_lock(&set->lock);
	/* select the first destination server, whose weight > 0 */
	for (e=set->list; e!=NULL; e=e->next) {
		least = e->dest;
		if (least->flags & IP_VS_DEST_F_OVERLOAD)
			continue;

		if ((atomic_read(&least->weight) > 0)
		    && (least->flags & IP_VS_DEST_F_AVAILABLE)) {
			loh = atomic_read(&least->activeconns) * 50
				+ atomic_read(&least->inactconns);
			goto nextstage;
		}
	}
	read_unlock(&set->lock);
	return NULL;

	/* find the destination with the weighted least load */
  nextstage:
	for (e=e->next; e!=NULL; e=e->next) {
		dest = e->dest;
		if (dest->flags & IP_VS_DEST_F_OVERLOAD)
			continue;

		doh = atomic_read(&dest->activeconns) * 50
			+ atomic_read(&dest->inactconns);
		if ((loh * atomic_read(&dest->weight) >
		     doh * atomic_read(&least->weight))
		    && (dest->flags & IP_VS_DEST_F_AVAILABLE)) {
			least = dest;
			loh = doh;
		}
	}
	read_unlock(&set->lock);

	IP_VS_DBG(6, "ip_vs_dest_set_min: server %d.%d.%d.%d:%d "
		  "activeconns %d refcnt %d weight %d overhead %d\n",
		  NIPQUAD(least->addr), ntohs(least->port),
		  atomic_read(&least->activeconns),
		  atomic_read(&least->refcnt),
		  atomic_read(&least->weight), loh);
	return least;
}


/* get weighted most-connection node in the destination set */
static inline struct ip_vs_dest *ip_vs_dest_set_max(struct ip_vs_dest_set *set)
{
	register struct ip_vs_dest_list *e;
	struct ip_vs_dest *dest, *most;
	int moh, doh;

	if (set == NULL)
		return NULL;

	read_lock(&set->lock);
	/* select the first destination server, whose weight > 0 */
	for (e=set->list; e!=NULL; e=e->next) {
		most = e->dest;
		if (atomic_read(&most->weight) > 0) {
			moh = atomic_read(&most->activeconns) * 50
				+ atomic_read(&most->inactconns);
			goto nextstage;
		}
	}
	read_unlock(&set->lock);
	return NULL;

	/* find the destination with the weighted most load */
  nextstage:
	for (e=e->next; e!=NULL; e=e->next) {
		dest = e->dest;
		doh = atomic_read(&dest->activeconns) * 50
			+ atomic_read(&dest->inactconns);
		/* moh/mw < doh/dw ==> moh*dw < doh*mw, where mw,dw>0 */
		if ((moh * atomic_read(&dest->weight) <
		     doh * atomic_read(&most->weight))
		    && (atomic_read(&dest->weight) > 0)) {
			most = dest;
			moh = doh;
		}
	}
	read_unlock(&set->lock);

	IP_VS_DBG(6, "ip_vs_dest_set_max: server %d.%d.%d.%d:%d "
		  "activeconns %d refcnt %d weight %d overhead %d\n",
		  NIPQUAD(most->addr), ntohs(most->port),
		  atomic_read(&most->activeconns),
		  atomic_read(&most->refcnt),
		  atomic_read(&most->weight), moh);
	return most;
}


/*
 *      IPVS lblcr entry represents an association between destination
 *      IP address and its destination server set
 */
struct ip_vs_lblcr_entry {
	struct list_head        list;
	__u32                   addr;           /* destination IP address */
	struct ip_vs_dest_set   set;            /* destination server set */
	unsigned long           lastuse;        /* last used time */
};


/*
 *      IPVS lblcr hash table
 */
struct ip_vs_lblcr_table {
	rwlock_t	        lock;           /* lock for this table */
	struct list_head        bucket[IP_VS_LBLCR_TAB_SIZE];  /* hash bucket */
	atomic_t                entries;        /* number of entries */
	int                     max_size;       /* maximum size of entries */
	struct timer_list       periodic_timer; /* collect stale entries */
	int                     rover;          /* rover for expire check */
	int                     counter;        /* counter for no expire */
};


/*
 *      IPVS LBLCR sysctl table
 */

static ctl_table vs_vars_table[] = {
	{
		.ctl_name	= NET_IPV4_VS_LBLCR_EXPIRE,
		.procname	= "lblcr_expiration",
		.data		= &sysctl_ip_vs_lblcr_expiration,
		.maxlen		= sizeof(int),
		.mode		= 0644, 
		.proc_handler	= &proc_dointvec_jiffies,
	},
	{ .ctl_name = 0 }
};

static ctl_table vs_table[] = {
	{
		.ctl_name	= NET_IPV4_VS,
		.procname	= "vs",
		.mode		= 0555,
		.child		= vs_vars_table
	},
	{ .ctl_name = 0 }
};

static ctl_table ipvs_ipv4_table[] = {
	{
		.ctl_name	= NET_IPV4,
		.procname	= "ipv4", 
		.mode		= 0555,
		.child		= vs_table
	},
	{ .ctl_name = 0 }
};

static ctl_table lblcr_root_table[] = {
	{
		.ctl_name	= CTL_NET,
		.procname	= "net", 
		.mode		= 0555, 
		.child		= ipvs_ipv4_table
	},
	{ .ctl_name = 0 }
};

static struct ctl_table_header * sysctl_header;

/*
 *      new/free a ip_vs_lblcr_entry, which is a mapping of a destination
 *      IP address to a server.
 */
static inline struct ip_vs_lblcr_entry *ip_vs_lblcr_new(__u32 daddr)
{
	struct ip_vs_lblcr_entry *en;

	en = kmalloc(sizeof(struct ip_vs_lblcr_entry), GFP_ATOMIC);
	if (en == NULL) {
		IP_VS_ERR("ip_vs_lblcr_new(): no memory\n");
		return NULL;
	}

	INIT_LIST_HEAD(&en->list);
	en->addr = daddr;

	/* initilize its dest set */
	atomic_set(&(en->set.size), 0);
	en->set.list = NULL;
	rwlock_init(&en->set.lock);

	return en;
}


static inline void ip_vs_lblcr_free(struct ip_vs_lblcr_entry *en)
{
	list_del(&en->list);
	ip_vs_dest_set_eraseall(&en->set);
	kfree(en);
}


/*
 *	Returns hash value for IPVS LBLCR entry
 */
static inline unsigned ip_vs_lblcr_hashkey(__u32 addr)
{
	return (ntohl(addr)*2654435761UL) & IP_VS_LBLCR_TAB_MASK;
}


/*
 *	Hash an entry in the ip_vs_lblcr_table.
 *	returns bool success.
 */
static int
ip_vs_lblcr_hash(struct ip_vs_lblcr_table *tbl, struct ip_vs_lblcr_entry *en)
{
	unsigned hash;

	if (!list_empty(&en->list)) {
		IP_VS_ERR("ip_vs_lblcr_hash(): request for already hashed, "
			  "called from %p\n", __builtin_return_address(0));
		return 0;
	}

	/*
	 *	Hash by destination IP address
	 */
	hash = ip_vs_lblcr_hashkey(en->addr);

	write_lock(&tbl->lock);
	list_add(&en->list, &tbl->bucket[hash]);
	atomic_inc(&tbl->entries);
	write_unlock(&tbl->lock);

	return 1;
}


/*
 *  Get ip_vs_lblcr_entry associated with supplied parameters.
 */
static inline struct ip_vs_lblcr_entry *
ip_vs_lblcr_get(struct ip_vs_lblcr_table *tbl, __u32 addr)
{
	unsigned hash;
	struct ip_vs_lblcr_entry *en;

	hash = ip_vs_lblcr_hashkey(addr);

	read_lock(&tbl->lock);

	list_for_each_entry(en, &tbl->bucket[hash], list) {
		if (en->addr == addr) {
			/* HIT */
			read_unlock(&tbl->lock);
			return en;
		}
	}

	read_unlock(&tbl->lock);

	return NULL;
}


/*
 *      Flush all the entries of the specified table.
 */
static void ip_vs_lblcr_flush(struct ip_vs_lblcr_table *tbl)
{
	int i;
	struct ip_vs_lblcr_entry *en, *nxt;

	for (i=0; i<IP_VS_LBLCR_TAB_SIZE; i++) {
		write_lock(&tbl->lock);
		list_for_each_entry_safe(en, nxt, &tbl->bucket[i], list) {
			ip_vs_lblcr_free(en);
			atomic_dec(&tbl->entries);
		}
		write_unlock(&tbl->lock);
	}
}


static inline void ip_vs_lblcr_full_check(struct ip_vs_lblcr_table *tbl)
{
	unsigned long now = jiffies;
	int i, j;
	struct ip_vs_lblcr_entry *en, *nxt;

	for (i=0, j=tbl->rover; i<IP_VS_LBLCR_TAB_SIZE; i++) {
		j = (j + 1) & IP_VS_LBLCR_TAB_MASK;

		write_lock(&tbl->lock);
		list_for_each_entry_safe(en, nxt, &tbl->bucket[j], list) {
			if (time_after(en->lastuse+sysctl_ip_vs_lblcr_expiration,
				       now))
				continue;

			ip_vs_lblcr_free(en);
			atomic_dec(&tbl->entries);
		}
		write_unlock(&tbl->lock);
	}
	tbl->rover = j;
}


/*
 *      Periodical timer handler for IPVS lblcr table
 *      It is used to collect stale entries when the number of entries
 *      exceeds the maximum size of the table.
 *
 *      Fixme: we probably need more complicated algorithm to collect
 *             entries that have not been used for a long time even
 *             if the number of entries doesn't exceed the maximum size
 *             of the table.
 *      The full expiration check is for this purpose now.
 */
static void ip_vs_lblcr_check_expire(unsigned long data)
{
	struct ip_vs_lblcr_table *tbl;
	unsigned long now = jiffies;
	int goal;
	int i, j;
	struct ip_vs_lblcr_entry *en, *nxt;

	tbl = (struct ip_vs_lblcr_table *)data;

	if ((tbl->counter % COUNT_FOR_FULL_EXPIRATION) == 0) {
		/* do full expiration check */
		ip_vs_lblcr_full_check(tbl);
		tbl->counter = 1;
		goto out;
	}

	if (atomic_read(&tbl->entries) <= tbl->max_size) {
		tbl->counter++;
		goto out;
	}

	goal = (atomic_read(&tbl->entries) - tbl->max_size)*4/3;
	if (goal > tbl->max_size/2)
		goal = tbl->max_size/2;

	for (i=0, j=tbl->rover; i<IP_VS_LBLCR_TAB_SIZE; i++) {
		j = (j + 1) & IP_VS_LBLCR_TAB_MASK;

		write_lock(&tbl->lock);
		list_for_each_entry_safe(en, nxt, &tbl->bucket[j], list) {
			if (time_before(now, en->lastuse+ENTRY_TIMEOUT))
				continue;

			ip_vs_lblcr_free(en);
			atomic_dec(&tbl->entries);
			goal--;
		}
		write_unlock(&tbl->lock);
		if (goal <= 0)
			break;
	}
	tbl->rover = j;

  out:
	mod_timer(&tbl->periodic_timer, jiffies+CHECK_EXPIRE_INTERVAL);
}


#ifdef CONFIG_IP_VS_LBLCR_DEBUG
static struct ip_vs_lblcr_table *lblcr_table_list;

/*
 *	/proc/net/ip_vs_lblcr to display the mappings of
 *                  destination IP address <==> its serverSet
 */
static int
ip_vs_lblcr_getinfo(char *buffer, char **start, off_t offset, int length)
{
	off_t pos=0, begin;
	int len=0, size;
	struct ip_vs_lblcr_table *tbl;
	unsigned long now = jiffies;
	int i;
	struct ip_vs_lblcr_entry *en;

	tbl = lblcr_table_list;

	size = sprintf(buffer, "LastTime Dest IP address  Server set\n");
	pos += size;
	len += size;

	for (i=0; i<IP_VS_LBLCR_TAB_SIZE; i++) {
		read_lock_bh(&tbl->lock);
		list_for_each_entry(en, &tbl->bucket[i], list) {
			char tbuf[16];
			struct ip_vs_dest_list *d;

			sprintf(tbuf, "%u.%u.%u.%u", NIPQUAD(en->addr));
			size = sprintf(buffer+len, "%8lu %-16s ",
				       now-en->lastuse, tbuf);

			read_lock(&en->set.lock);
			for (d=en->set.list; d!=NULL; d=d->next) {
				size += sprintf(buffer+len+size,
						"%u.%u.%u.%u ",
						NIPQUAD(d->dest->addr));
			}
			read_unlock(&en->set.lock);
			size += sprintf(buffer+len+size, "\n");
			len += size;
			pos += size;
			if (pos <= offset)
				len=0;
			if (pos >= offset+length) {
				read_unlock_bh(&tbl->lock);
				goto done;
			}
		}
		read_unlock_bh(&tbl->lock);
	}

  done:
	begin = len - (pos - offset);
	*start = buffer + begin;
	len -= begin;
	if(len>length)
		len = length;
	return len;
}
#endif


static int ip_vs_lblcr_init_svc(struct ip_vs_service *svc)
{
	int i;
	struct ip_vs_lblcr_table *tbl;

	/*
	 *    Allocate the ip_vs_lblcr_table for this service
	 */
	tbl = kmalloc(sizeof(struct ip_vs_lblcr_table), GFP_ATOMIC);
	if (tbl == NULL) {
		IP_VS_ERR("ip_vs_lblcr_init_svc(): no memory\n");
		return -ENOMEM;
	}
	svc->sched_data = tbl;
	IP_VS_DBG(6, "LBLCR hash table (memory=%Zdbytes) allocated for "
		  "current service\n",
		  sizeof(struct ip_vs_lblcr_table));

	/*
	 *    Initialize the hash buckets
	 */
	for (i=0; i<IP_VS_LBLCR_TAB_SIZE; i++) {
		INIT_LIST_HEAD(&tbl->bucket[i]);
	}
	rwlock_init(&tbl->lock);
	tbl->max_size = IP_VS_LBLCR_TAB_SIZE*16;
	tbl->rover = 0;
	tbl->counter = 1;

	/*
	 *    Hook periodic timer for garbage collection
	 */
	init_timer(&tbl->periodic_timer);
	tbl->periodic_timer.data = (unsigned long)tbl;
	tbl->periodic_timer.function = ip_vs_lblcr_check_expire;
	tbl->periodic_timer.expires = jiffies+CHECK_EXPIRE_INTERVAL;
	add_timer(&tbl->periodic_timer);

#ifdef CONFIG_IP_VS_LBLCR_DEBUG
	lblcr_table_list = tbl;
#endif
	return 0;
}


static int ip_vs_lblcr_done_svc(struct ip_vs_service *svc)
{
	struct ip_vs_lblcr_table *tbl = svc->sched_data;

	/* remove periodic timer */
	del_timer_sync(&tbl->periodic_timer);

	/* got to clean up table entries here */
	ip_vs_lblcr_flush(tbl);

	/* release the table itself */
	kfree(svc->sched_data);
	IP_VS_DBG(6, "LBLCR hash table (memory=%Zdbytes) released\n",
		  sizeof(struct ip_vs_lblcr_table));

	return 0;
}


static int ip_vs_lblcr_update_svc(struct ip_vs_service *svc)
{
	return 0;
}


static inline struct ip_vs_dest *
__ip_vs_wlc_schedule(struct ip_vs_service *svc, struct iphdr *iph)
{
	struct ip_vs_dest *dest, *least;
	int loh, doh;

	/*
	 * We think the overhead of processing active connections is fifty
	 * times higher than that of inactive connections in average. (This
	 * fifty times might not be accurate, we will change it later.) We
	 * use the following formula to estimate the overhead:
	 *                dest->activeconns*50 + dest->inactconns
	 * and the load:
	 *                (dest overhead) / dest->weight
	 *
	 * Remember -- no floats in kernel mode!!!
	 * The comparison of h1*w2 > h2*w1 is equivalent to that of
	 *                h1/w1 > h2/w2
	 * if every weight is larger than zero.
	 *
	 * The server with weight=0 is quiesced and will not receive any
	 * new connection.
	 */
	list_for_each_entry(dest, &svc->destinations, n_list) {
		if (dest->flags & IP_VS_DEST_F_OVERLOAD)
			continue;

		if (atomic_read(&dest->weight) > 0) {
			least = dest;
			loh = atomic_read(&least->activeconns) * 50
				+ atomic_read(&least->inactconns);
			goto nextstage;
		}
	}
	return NULL;

	/*
	 *    Find the destination with the least load.
	 */
  nextstage:
	list_for_each_entry_continue(dest, &svc->destinations, n_list) {
		if (dest->flags & IP_VS_DEST_F_OVERLOAD)
			continue;

		doh = atomic_read(&dest->activeconns) * 50
			+ atomic_read(&dest->inactconns);
		if (loh * atomic_read(&dest->weight) >
		    doh * atomic_read(&least->weight)) {
			least = dest;
			loh = doh;
		}
	}

	IP_VS_DBG(6, "LBLCR: server %d.%d.%d.%d:%d "
		  "activeconns %d refcnt %d weight %d overhead %d\n",
		  NIPQUAD(least->addr), ntohs(least->port),
		  atomic_read(&least->activeconns),
		  atomic_read(&least->refcnt),
		  atomic_read(&least->weight), loh);

	return least;
}


/*
 *   If this destination server is overloaded and there is a less loaded
 *   server, then return true.
 */
static inline int
is_overloaded(struct ip_vs_dest *dest, struct ip_vs_service *svc)
{
	if (atomic_read(&dest->activeconns) > atomic_read(&dest->weight)) {
		struct ip_vs_dest *d;

		list_for_each_entry(d, &svc->destinations, n_list) {
			if (atomic_read(&d->activeconns)*2
			    < atomic_read(&d->weight)) {
				return 1;
			}
		}
	}
	return 0;
}


/*
 *    Locality-Based (weighted) Least-Connection scheduling
 */
static struct ip_vs_dest *
ip_vs_lblcr_schedule(struct ip_vs_service *svc, const struct sk_buff *skb)
{
	struct ip_vs_dest *dest;
	struct ip_vs_lblcr_table *tbl;
	struct ip_vs_lblcr_entry *en;
	struct iphdr *iph = skb->nh.iph;

	IP_VS_DBG(6, "ip_vs_lblcr_schedule(): Scheduling...\n");

	tbl = (struct ip_vs_lblcr_table *)svc->sched_data;
	en = ip_vs_lblcr_get(tbl, iph->daddr);
	if (en == NULL) {
		dest = __ip_vs_wlc_schedule(svc, iph);
		if (dest == NULL) {
			IP_VS_DBG(1, "no destination available\n");
			return NULL;
		}
		en = ip_vs_lblcr_new(iph->daddr);
		if (en == NULL) {
			return NULL;
		}
		ip_vs_dest_set_insert(&en->set, dest);
		ip_vs_lblcr_hash(tbl, en);
	} else {
		dest = ip_vs_dest_set_min(&en->set);
		if (!dest || is_overloaded(dest, svc)) {
			dest = __ip_vs_wlc_schedule(svc, iph);
			if (dest == NULL) {
				IP_VS_DBG(1, "no destination available\n");
				return NULL;
			}
			ip_vs_dest_set_insert(&en->set, dest);
		}
		if (atomic_read(&en->set.size) > 1 &&
		    jiffies-en->set.lastmod > sysctl_ip_vs_lblcr_expiration) {
			struct ip_vs_dest *m;
			m = ip_vs_dest_set_max(&en->set);
			if (m)
				ip_vs_dest_set_erase(&en->set, m);
		}
	}
	en->lastuse = jiffies;

	IP_VS_DBG(6, "LBLCR: destination IP address %u.%u.%u.%u "
		  "--> server %u.%u.%u.%u:%d\n",
		  NIPQUAD(en->addr),
		  NIPQUAD(dest->addr),
		  ntohs(dest->port));

	return dest;
}


/*
 *      IPVS LBLCR Scheduler structure
 */
static struct ip_vs_scheduler ip_vs_lblcr_scheduler =
{
	.name =			"lblcr",
	.refcnt =		ATOMIC_INIT(0),
	.module =		THIS_MODULE,
	.init_service =		ip_vs_lblcr_init_svc,
	.done_service =		ip_vs_lblcr_done_svc,
	.update_service =	ip_vs_lblcr_update_svc,
	.schedule =		ip_vs_lblcr_schedule,
};


static int __init ip_vs_lblcr_init(void)
{
	INIT_LIST_HEAD(&ip_vs_lblcr_scheduler.n_list);
	sysctl_header = register_sysctl_table(lblcr_root_table, 0);
#ifdef CONFIG_IP_VS_LBLCR_DEBUG
	proc_net_create("ip_vs_lblcr", 0, ip_vs_lblcr_getinfo);
#endif
	return register_ip_vs_scheduler(&ip_vs_lblcr_scheduler);
}


static void __exit ip_vs_lblcr_cleanup(void)
{
#ifdef CONFIG_IP_VS_LBLCR_DEBUG
	proc_net_remove("ip_vs_lblcr");
#endif
	unregister_sysctl_table(sysctl_header);
	unregister_ip_vs_scheduler(&ip_vs_lblcr_scheduler);
}


module_init(ip_vs_lblcr_init);
module_exit(ip_vs_lblcr_cleanup);
MODULE_LICENSE("GPL");