Linux-2.6.12-rc2v2.6.12-rc2

Initial git repository build. I'm not bothering with the full history,
even though we have it. We can create a separate "historical" git
archive of that later if we want to, and in the meantime it's about
3.2GB when imported into git - space that would just make the early
git days unnecessarily complicated, when we don't have a lot of good
infrastructure for it.

Let it rip!

1 files changed, 2391 insertions, 0 deletions

diff --git a/net/ipv4/ipvs/ip_vs_ctl.c b/net/ipv4/ipvs/ip_vs_ctl.c
new file mode 100644
index 00000000000..218d9701036
--- /dev/null
+++ b/net/ipv4/ipvs/ip_vs_ctl.c
@@ -0,0 +1,2391 @@
+/*
+ * IPVS         An implementation of the IP virtual server support for the
+ *              LINUX operating system.  IPVS is now implemented as a module
+ *              over the NetFilter framework. IPVS can be used to build a
+ *              high-performance and highly available server based on a
+ *              cluster of servers.
+ *
+ * Version:     $Id: ip_vs_ctl.c,v 1.36 2003/06/08 09:31:19 wensong Exp $
+ *
+ * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
+ *              Peter Kese <peter.kese@ijs.si>
+ *              Julian Anastasov <ja@ssi.bg>
+ *
+ *              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:
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/types.h>
+#include <linux/fs.h>
+#include <linux/sysctl.h>
+#include <linux/proc_fs.h>
+#include <linux/workqueue.h>
+#include <linux/swap.h>
+#include <linux/proc_fs.h>
+#include <linux/seq_file.h>
+
+#include <linux/netfilter.h>
+#include <linux/netfilter_ipv4.h>
+
+#include <net/ip.h>
+#include <net/sock.h>
+
+#include <asm/uaccess.h>
+
+#include <net/ip_vs.h>
+
+/* semaphore for IPVS sockopts. And, [gs]etsockopt may sleep. */
+static DECLARE_MUTEX(__ip_vs_mutex);
+
+/* lock for service table */
+static DEFINE_RWLOCK(__ip_vs_svc_lock);
+
+/* lock for table with the real services */
+static DEFINE_RWLOCK(__ip_vs_rs_lock);
+
+/* lock for state and timeout tables */
+static DEFINE_RWLOCK(__ip_vs_securetcp_lock);
+
+/* lock for drop entry handling */
+static DEFINE_SPINLOCK(__ip_vs_dropentry_lock);
+
+/* lock for drop packet handling */
+static DEFINE_SPINLOCK(__ip_vs_droppacket_lock);
+
+/* 1/rate drop and drop-entry variables */
+int ip_vs_drop_rate = 0;
+int ip_vs_drop_counter = 0;
+static atomic_t ip_vs_dropentry = ATOMIC_INIT(0);
+
+/* number of virtual services */
+static int ip_vs_num_services = 0;
+
+/* sysctl variables */
+static int sysctl_ip_vs_drop_entry = 0;
+static int sysctl_ip_vs_drop_packet = 0;
+static int sysctl_ip_vs_secure_tcp = 0;
+static int sysctl_ip_vs_amemthresh = 1024;
+static int sysctl_ip_vs_am_droprate = 10;
+int sysctl_ip_vs_cache_bypass = 0;
+int sysctl_ip_vs_expire_nodest_conn = 0;
+int sysctl_ip_vs_expire_quiescent_template = 0;
+int sysctl_ip_vs_sync_threshold[2] = { 3, 50 };
+int sysctl_ip_vs_nat_icmp_send = 0;
+
+
+#ifdef CONFIG_IP_VS_DEBUG
+static int sysctl_ip_vs_debug_level = 0;
+
+int ip_vs_get_debug_level(void)
+{
+	return sysctl_ip_vs_debug_level;
+}
+#endif
+
+/*
+ *	update_defense_level is called from keventd and from sysctl.
+ */
+static void update_defense_level(void)
+{
+	struct sysinfo i;
+	static int old_secure_tcp = 0;
+	int availmem;
+	int nomem;
+	int to_change = -1;
+
+	/* we only count free and buffered memory (in pages) */
+	si_meminfo(&i);
+	availmem = i.freeram + i.bufferram;
+	/* however in linux 2.5 the i.bufferram is total page cache size,
+	   we need adjust it */
+	/* si_swapinfo(&i); */
+	/* availmem = availmem - (i.totalswap - i.freeswap); */
+
+	nomem = (availmem < sysctl_ip_vs_amemthresh);
+
+	/* drop_entry */
+	spin_lock(&__ip_vs_dropentry_lock);
+	switch (sysctl_ip_vs_drop_entry) {
+	case 0:
+		atomic_set(&ip_vs_dropentry, 0);
+		break;
+	case 1:
+		if (nomem) {
+			atomic_set(&ip_vs_dropentry, 1);
+			sysctl_ip_vs_drop_entry = 2;
+		} else {
+			atomic_set(&ip_vs_dropentry, 0);
+		}
+		break;
+	case 2:
+		if (nomem) {
+			atomic_set(&ip_vs_dropentry, 1);
+		} else {
+			atomic_set(&ip_vs_dropentry, 0);
+			sysctl_ip_vs_drop_entry = 1;
+		};
+		break;
+	case 3:
+		atomic_set(&ip_vs_dropentry, 1);
+		break;
+	}
+	spin_unlock(&__ip_vs_dropentry_lock);
+
+	/* drop_packet */
+	spin_lock(&__ip_vs_droppacket_lock);
+	switch (sysctl_ip_vs_drop_packet) {
+	case 0:
+		ip_vs_drop_rate = 0;
+		break;
+	case 1:
+		if (nomem) {
+			ip_vs_drop_rate = ip_vs_drop_counter
+				= sysctl_ip_vs_amemthresh /
+				(sysctl_ip_vs_amemthresh-availmem);
+			sysctl_ip_vs_drop_packet = 2;
+		} else {
+			ip_vs_drop_rate = 0;
+		}
+		break;
+	case 2:
+		if (nomem) {
+			ip_vs_drop_rate = ip_vs_drop_counter
+				= sysctl_ip_vs_amemthresh /
+				(sysctl_ip_vs_amemthresh-availmem);
+		} else {
+			ip_vs_drop_rate = 0;
+			sysctl_ip_vs_drop_packet = 1;
+		}
+		break;
+	case 3:
+		ip_vs_drop_rate = sysctl_ip_vs_am_droprate;
+		break;
+	}
+	spin_unlock(&__ip_vs_droppacket_lock);
+
+	/* secure_tcp */
+	write_lock(&__ip_vs_securetcp_lock);
+	switch (sysctl_ip_vs_secure_tcp) {
+	case 0:
+		if (old_secure_tcp >= 2)
+			to_change = 0;
+		break;
+	case 1:
+		if (nomem) {
+			if (old_secure_tcp < 2)
+				to_change = 1;
+			sysctl_ip_vs_secure_tcp = 2;
+		} else {
+			if (old_secure_tcp >= 2)
+				to_change = 0;
+		}
+		break;
+	case 2:
+		if (nomem) {
+			if (old_secure_tcp < 2)
+				to_change = 1;
+		} else {
+			if (old_secure_tcp >= 2)
+				to_change = 0;
+			sysctl_ip_vs_secure_tcp = 1;
+		}
+		break;
+	case 3:
+		if (old_secure_tcp < 2)
+			to_change = 1;
+		break;
+	}
+	old_secure_tcp = sysctl_ip_vs_secure_tcp;
+	if (to_change >= 0)
+		ip_vs_protocol_timeout_change(sysctl_ip_vs_secure_tcp>1);
+	write_unlock(&__ip_vs_securetcp_lock);
+}
+
+
+/*
+ *	Timer for checking the defense
+ */
+#define DEFENSE_TIMER_PERIOD	1*HZ
+static void defense_work_handler(void *data);
+static DECLARE_WORK(defense_work, defense_work_handler, NULL);
+
+static void defense_work_handler(void *data)
+{
+	update_defense_level();
+	if (atomic_read(&ip_vs_dropentry))
+		ip_vs_random_dropentry();
+
+	schedule_delayed_work(&defense_work, DEFENSE_TIMER_PERIOD);
+}
+
+int
+ip_vs_use_count_inc(void)
+{
+	return try_module_get(THIS_MODULE);
+}
+
+void
+ip_vs_use_count_dec(void)
+{
+	module_put(THIS_MODULE);
+}
+
+
+/*
+ *	Hash table: for virtual service lookups
+ */
+#define IP_VS_SVC_TAB_BITS 8
+#define IP_VS_SVC_TAB_SIZE (1 << IP_VS_SVC_TAB_BITS)
+#define IP_VS_SVC_TAB_MASK (IP_VS_SVC_TAB_SIZE - 1)
+
+/* the service table hashed by <protocol, addr, port> */
+static struct list_head ip_vs_svc_table[IP_VS_SVC_TAB_SIZE];
+/* the service table hashed by fwmark */
+static struct list_head ip_vs_svc_fwm_table[IP_VS_SVC_TAB_SIZE];
+
+/*
+ *	Hash table: for real service lookups
+ */
+#define IP_VS_RTAB_BITS 4
+#define IP_VS_RTAB_SIZE (1 << IP_VS_RTAB_BITS)
+#define IP_VS_RTAB_MASK (IP_VS_RTAB_SIZE - 1)
+
+static struct list_head ip_vs_rtable[IP_VS_RTAB_SIZE];
+
+/*
+ *	Trash for destinations
+ */
+static LIST_HEAD(ip_vs_dest_trash);
+
+/*
+ *	FTP & NULL virtual service counters
+ */
+static atomic_t ip_vs_ftpsvc_counter = ATOMIC_INIT(0);
+static atomic_t ip_vs_nullsvc_counter = ATOMIC_INIT(0);
+
+
+/*
+ *	Returns hash value for virtual service
+ */
+static __inline__ unsigned
+ip_vs_svc_hashkey(unsigned proto, __u32 addr, __u16 port)
+{
+	register unsigned porth = ntohs(port);
+
+	return (proto^ntohl(addr)^(porth>>IP_VS_SVC_TAB_BITS)^porth)
+		& IP_VS_SVC_TAB_MASK;
+}
+
+/*
+ *	Returns hash value of fwmark for virtual service lookup
+ */
+static __inline__ unsigned ip_vs_svc_fwm_hashkey(__u32 fwmark)
+{
+	return fwmark & IP_VS_SVC_TAB_MASK;
+}
+
+/*
+ *	Hashes a service in the ip_vs_svc_table by <proto,addr,port>
+ *	or in the ip_vs_svc_fwm_table by fwmark.
+ *	Should be called with locked tables.
+ */
+static int ip_vs_svc_hash(struct ip_vs_service *svc)
+{
+	unsigned hash;
+
+	if (svc->flags & IP_VS_SVC_F_HASHED) {
+		IP_VS_ERR("ip_vs_svc_hash(): request for already hashed, "
+			  "called from %p\n", __builtin_return_address(0));
+		return 0;
+	}
+
+	if (svc->fwmark == 0) {
+		/*
+		 *  Hash it by <protocol,addr,port> in ip_vs_svc_table
+		 */
+		hash = ip_vs_svc_hashkey(svc->protocol, svc->addr, svc->port);
+		list_add(&svc->s_list, &ip_vs_svc_table[hash]);
+	} else {
+		/*
+		 *  Hash it by fwmark in ip_vs_svc_fwm_table
+		 */
+		hash = ip_vs_svc_fwm_hashkey(svc->fwmark);
+		list_add(&svc->f_list, &ip_vs_svc_fwm_table[hash]);
+	}
+
+	svc->flags |= IP_VS_SVC_F_HASHED;
+	/* increase its refcnt because it is referenced by the svc table */
+	atomic_inc(&svc->refcnt);
+	return 1;
+}
+
+
+/*
+ *	Unhashes a service from ip_vs_svc_table/ip_vs_svc_fwm_table.
+ *	Should be called with locked tables.
+ */
+static int ip_vs_svc_unhash(struct ip_vs_service *svc)
+{
+	if (!(svc->flags & IP_VS_SVC_F_HASHED)) {
+		IP_VS_ERR("ip_vs_svc_unhash(): request for unhash flagged, "
+			  "called from %p\n", __builtin_return_address(0));
+		return 0;
+	}
+
+	if (svc->fwmark == 0) {
+		/* Remove it from the ip_vs_svc_table table */
+		list_del(&svc->s_list);
+	} else {
+		/* Remove it from the ip_vs_svc_fwm_table table */
+		list_del(&svc->f_list);
+	}
+
+	svc->flags &= ~IP_VS_SVC_F_HASHED;
+	atomic_dec(&svc->refcnt);
+	return 1;
+}
+
+
+/*
+ *	Get service by {proto,addr,port} in the service table.
+ */
+static __inline__ struct ip_vs_service *
+__ip_vs_service_get(__u16 protocol, __u32 vaddr, __u16 vport)
+{
+	unsigned hash;
+	struct ip_vs_service *svc;
+
+	/* Check for "full" addressed entries */
+	hash = ip_vs_svc_hashkey(protocol, vaddr, vport);
+
+	list_for_each_entry(svc, &ip_vs_svc_table[hash], s_list){
+		if ((svc->addr == vaddr)
+		    && (svc->port == vport)
+		    && (svc->protocol == protocol)) {
+			/* HIT */
+			atomic_inc(&svc->usecnt);
+			return svc;
+		}
+	}
+
+	return NULL;
+}
+
+
+/*
+ *	Get service by {fwmark} in the service table.
+ */
+static __inline__ struct ip_vs_service *__ip_vs_svc_fwm_get(__u32 fwmark)
+{
+	unsigned hash;
+	struct ip_vs_service *svc;
+
+	/* Check for fwmark addressed entries */
+	hash = ip_vs_svc_fwm_hashkey(fwmark);
+
+	list_for_each_entry(svc, &ip_vs_svc_fwm_table[hash], f_list) {
+		if (svc->fwmark == fwmark) {
+			/* HIT */
+			atomic_inc(&svc->usecnt);
+			return svc;
+		}
+	}
+
+	return NULL;
+}
+
+struct ip_vs_service *
+ip_vs_service_get(__u32 fwmark, __u16 protocol, __u32 vaddr, __u16 vport)
+{
+	struct ip_vs_service *svc;
+
+	read_lock(&__ip_vs_svc_lock);
+
+	/*
+	 *	Check the table hashed by fwmark first
+	 */
+	if (fwmark && (svc = __ip_vs_svc_fwm_get(fwmark)))
+		goto out;
+
+	/*
+	 *	Check the table hashed by <protocol,addr,port>
+	 *	for "full" addressed entries
+	 */
+	svc = __ip_vs_service_get(protocol, vaddr, vport);
+
+	if (svc == NULL
+	    && protocol == IPPROTO_TCP
+	    && atomic_read(&ip_vs_ftpsvc_counter)
+	    && (vport == FTPDATA || ntohs(vport) >= PROT_SOCK)) {
+		/*
+		 * Check if ftp service entry exists, the packet
+		 * might belong to FTP data connections.
+		 */
+		svc = __ip_vs_service_get(protocol, vaddr, FTPPORT);
+	}
+
+	if (svc == NULL
+	    && atomic_read(&ip_vs_nullsvc_counter)) {
+		/*
+		 * Check if the catch-all port (port zero) exists
+		 */
+		svc = __ip_vs_service_get(protocol, vaddr, 0);
+	}
+
+  out:
+	read_unlock(&__ip_vs_svc_lock);
+
+	IP_VS_DBG(6, "lookup service: fwm %u %s %u.%u.%u.%u:%u %s\n",
+		  fwmark, ip_vs_proto_name(protocol),
+		  NIPQUAD(vaddr), ntohs(vport),
+		  svc?"hit":"not hit");
+
+	return svc;
+}
+
+
+static inline void
+__ip_vs_bind_svc(struct ip_vs_dest *dest, struct ip_vs_service *svc)
+{
+	atomic_inc(&svc->refcnt);
+	dest->svc = svc;
+}
+
+static inline void
+__ip_vs_unbind_svc(struct ip_vs_dest *dest)
+{
+	struct ip_vs_service *svc = dest->svc;
+
+	dest->svc = NULL;
+	if (atomic_dec_and_test(&svc->refcnt))
+		kfree(svc);
+}
+
+
+/*
+ *	Returns hash value for real service
+ */
+static __inline__ unsigned ip_vs_rs_hashkey(__u32 addr, __u16 port)
+{
+	register unsigned porth = ntohs(port);
+
+	return (ntohl(addr)^(porth>>IP_VS_RTAB_BITS)^porth)
+		& IP_VS_RTAB_MASK;
+}
+
+/*
+ *	Hashes ip_vs_dest in ip_vs_rtable by <proto,addr,port>.
+ *	should be called with locked tables.
+ */
+static int ip_vs_rs_hash(struct ip_vs_dest *dest)
+{
+	unsigned hash;
+
+	if (!list_empty(&dest->d_list)) {
+		return 0;
+	}
+
+	/*
+	 *	Hash by proto,addr,port,
+	 *	which are the parameters of the real service.
+	 */
+	hash = ip_vs_rs_hashkey(dest->addr, dest->port);
+	list_add(&dest->d_list, &ip_vs_rtable[hash]);
+
+	return 1;
+}
+
+/*
+ *	UNhashes ip_vs_dest from ip_vs_rtable.
+ *	should be called with locked tables.
+ */
+static int ip_vs_rs_unhash(struct ip_vs_dest *dest)
+{
+	/*
+	 * Remove it from the ip_vs_rtable table.
+	 */
+	if (!list_empty(&dest->d_list)) {
+		list_del(&dest->d_list);
+		INIT_LIST_HEAD(&dest->d_list);
+	}
+
+	return 1;
+}
+
+/*
+ *	Lookup real service by <proto,addr,port> in the real service table.
+ */
+struct ip_vs_dest *
+ip_vs_lookup_real_service(__u16 protocol, __u32 daddr, __u16 dport)
+{
+	unsigned hash;
+	struct ip_vs_dest *dest;
+
+	/*
+	 *	Check for "full" addressed entries
+	 *	Return the first found entry
+	 */
+	hash = ip_vs_rs_hashkey(daddr, dport);
+
+	read_lock(&__ip_vs_rs_lock);
+	list_for_each_entry(dest, &ip_vs_rtable[hash], d_list) {
+		if ((dest->addr == daddr)
+		    && (dest->port == dport)
+		    && ((dest->protocol == protocol) ||
+			dest->vfwmark)) {
+			/* HIT */
+			read_unlock(&__ip_vs_rs_lock);
+			return dest;
+		}
+	}
+	read_unlock(&__ip_vs_rs_lock);
+
+	return NULL;
+}
+
+/*
+ *	Lookup destination by {addr,port} in the given service
+ */
+static struct ip_vs_dest *
+ip_vs_lookup_dest(struct ip_vs_service *svc, __u32 daddr, __u16 dport)
+{
+	struct ip_vs_dest *dest;
+
+	/*
+	 * Find the destination for the given service
+	 */
+	list_for_each_entry(dest, &svc->destinations, n_list) {
+		if ((dest->addr == daddr) && (dest->port == dport)) {
+			/* HIT */
+			return dest;
+		}
+	}
+
+	return NULL;
+}
+
+
+/*
+ *  Lookup dest by {svc,addr,port} in the destination trash.
+ *  The destination trash is used to hold the destinations that are removed
+ *  from the service table but are still referenced by some conn entries.
+ *  The reason to add the destination trash is when the dest is temporary
+ *  down (either by administrator or by monitor program), the dest can be
+ *  picked back from the trash, the remaining connections to the dest can
+ *  continue, and the counting information of the dest is also useful for
+ *  scheduling.
+ */
+static struct ip_vs_dest *
+ip_vs_trash_get_dest(struct ip_vs_service *svc, __u32 daddr, __u16 dport)
+{
+	struct ip_vs_dest *dest, *nxt;
+
+	/*
+	 * Find the destination in trash
+	 */
+	list_for_each_entry_safe(dest, nxt, &ip_vs_dest_trash, n_list) {
+		IP_VS_DBG(3, "Destination %u/%u.%u.%u.%u:%u still in trash, "
+			  "refcnt=%d\n",
+			  dest->vfwmark,
+			  NIPQUAD(dest->addr), ntohs(dest->port),
+			  atomic_read(&dest->refcnt));
+		if (dest->addr == daddr &&
+		    dest->port == dport &&
+		    dest->vfwmark == svc->fwmark &&
+		    dest->protocol == svc->protocol &&
+		    (svc->fwmark ||
+		     (dest->vaddr == svc->addr &&
+		      dest->vport == svc->port))) {
+			/* HIT */
+			return dest;
+		}
+
+		/*
+		 * Try to purge the destination from trash if not referenced
+		 */
+		if (atomic_read(&dest->refcnt) == 1) {
+			IP_VS_DBG(3, "Removing destination %u/%u.%u.%u.%u:%u "
+				  "from trash\n",
+				  dest->vfwmark,
+				  NIPQUAD(dest->addr), ntohs(dest->port));
+			list_del(&dest->n_list);
+			ip_vs_dst_reset(dest);
+			__ip_vs_unbind_svc(dest);
+			kfree(dest);
+		}
+	}
+
+	return NULL;
+}
+
+
+/*
+ *  Clean up all the destinations in the trash
+ *  Called by the ip_vs_control_cleanup()
+ *
+ *  When the ip_vs_control_clearup is activated by ipvs module exit,
+ *  the service tables must have been flushed and all the connections
+ *  are expired, and the refcnt of each destination in the trash must
+ *  be 1, so we simply release them here.
+ */
+static void ip_vs_trash_cleanup(void)
+{
+	struct ip_vs_dest *dest, *nxt;
+
+	list_for_each_entry_safe(dest, nxt, &ip_vs_dest_trash, n_list) {
+		list_del(&dest->n_list);
+		ip_vs_dst_reset(dest);
+		__ip_vs_unbind_svc(dest);
+		kfree(dest);
+	}
+}
+
+
+static void
+ip_vs_zero_stats(struct ip_vs_stats *stats)
+{
+	spin_lock_bh(&stats->lock);
+	memset(stats, 0, (char *)&stats->lock - (char *)stats);
+	spin_unlock_bh(&stats->lock);
+	ip_vs_zero_estimator(stats);
+}
+
+/*
+ *	Update a destination in the given service
+ */
+static void
+__ip_vs_update_dest(struct ip_vs_service *svc,
+		    struct ip_vs_dest *dest, struct ip_vs_dest_user *udest)
+{
+	int conn_flags;
+
+	/* set the weight and the flags */
+	atomic_set(&dest->weight, udest->weight);
+	conn_flags = udest->conn_flags | IP_VS_CONN_F_INACTIVE;
+
+	/* check if local node and update the flags */
+	if (inet_addr_type(udest->addr) == RTN_LOCAL) {
+		conn_flags = (conn_flags & ~IP_VS_CONN_F_FWD_MASK)
+			| IP_VS_CONN_F_LOCALNODE;
+	}
+
+	/* set the IP_VS_CONN_F_NOOUTPUT flag if not masquerading/NAT */
+	if ((conn_flags & IP_VS_CONN_F_FWD_MASK) != 0) {
+		conn_flags |= IP_VS_CONN_F_NOOUTPUT;
+	} else {
+		/*
+		 *    Put the real service in ip_vs_rtable if not present.
+		 *    For now only for NAT!
+		 */
+		write_lock_bh(&__ip_vs_rs_lock);
+		ip_vs_rs_hash(dest);
+		write_unlock_bh(&__ip_vs_rs_lock);
+	}
+	atomic_set(&dest->conn_flags, conn_flags);
+
+	/* bind the service */
+	if (!dest->svc) {
+		__ip_vs_bind_svc(dest, svc);
+	} else {
+		if (dest->svc != svc) {
+			__ip_vs_unbind_svc(dest);
+			ip_vs_zero_stats(&dest->stats);
+			__ip_vs_bind_svc(dest, svc);
+		}
+	}
+
+	/* set the dest status flags */
+	dest->flags |= IP_VS_DEST_F_AVAILABLE;
+
+	if (udest->u_threshold == 0 || udest->u_threshold > dest->u_threshold)
+		dest->flags &= ~IP_VS_DEST_F_OVERLOAD;
+	dest->u_threshold = udest->u_threshold;
+	dest->l_threshold = udest->l_threshold;
+}
+
+
+/*
+ *	Create a destination for the given service
+ */
+static int
+ip_vs_new_dest(struct ip_vs_service *svc, struct ip_vs_dest_user *udest,
+	       struct ip_vs_dest **dest_p)
+{
+	struct ip_vs_dest *dest;
+	unsigned atype;
+
+	EnterFunction(2);
+
+	atype = inet_addr_type(udest->addr);
+	if (atype != RTN_LOCAL && atype != RTN_UNICAST)
+		return -EINVAL;
+
+	dest = kmalloc(sizeof(struct ip_vs_dest), GFP_ATOMIC);
+	if (dest == NULL) {
+		IP_VS_ERR("ip_vs_new_dest: kmalloc failed.\n");
+		return -ENOMEM;
+	}
+	memset(dest, 0, sizeof(struct ip_vs_dest));
+
+	dest->protocol = svc->protocol;
+	dest->vaddr = svc->addr;
+	dest->vport = svc->port;
+	dest->vfwmark = svc->fwmark;
+	dest->addr = udest->addr;
+	dest->port = udest->port;
+
+	atomic_set(&dest->activeconns, 0);
+	atomic_set(&dest->inactconns, 0);
+	atomic_set(&dest->persistconns, 0);
+	atomic_set(&dest->refcnt, 0);
+
+	INIT_LIST_HEAD(&dest->d_list);
+	spin_lock_init(&dest->dst_lock);
+	spin_lock_init(&dest->stats.lock);
+	__ip_vs_update_dest(svc, dest, udest);
+	ip_vs_new_estimator(&dest->stats);
+
+	*dest_p = dest;
+
+	LeaveFunction(2);
+	return 0;
+}
+
+
+/*
+ *	Add a destination into an existing service
+ */
+static int
+ip_vs_add_dest(struct ip_vs_service *svc, struct ip_vs_dest_user *udest)
+{
+	struct ip_vs_dest *dest;
+	__u32 daddr = udest->addr;
+	__u16 dport = udest->port;
+	int ret;
+
+	EnterFunction(2);
+
+	if (udest->weight < 0) {
+		IP_VS_ERR("ip_vs_add_dest(): server weight less than zero\n");
+		return -ERANGE;
+	}
+
+	if (udest->l_threshold > udest->u_threshold) {
+		IP_VS_ERR("ip_vs_add_dest(): lower threshold is higher than "
+			  "upper threshold\n");
+		return -ERANGE;
+	}
+
+	/*
+	 * Check if the dest already exists in the list
+	 */
+	dest = ip_vs_lookup_dest(svc, daddr, dport);
+	if (dest != NULL) {
+		IP_VS_DBG(1, "ip_vs_add_dest(): dest already exists\n");
+		return -EEXIST;
+	}
+
+	/*
+	 * Check if the dest already exists in the trash and
+	 * is from the same service
+	 */
+	dest = ip_vs_trash_get_dest(svc, daddr, dport);
+	if (dest != NULL) {
+		IP_VS_DBG(3, "Get destination %u.%u.%u.%u:%u from trash, "
+			  "refcnt=%d, service %u/%u.%u.%u.%u:%u\n",
+			  NIPQUAD(daddr), ntohs(dport),
+			  atomic_read(&dest->refcnt),
+			  dest->vfwmark,
+			  NIPQUAD(dest->vaddr),
+			  ntohs(dest->vport));
+		__ip_vs_update_dest(svc, dest, udest);
+
+		/*
+		 * Get the destination from the trash
+		 */
+		list_del(&dest->n_list);
+
+		ip_vs_new_estimator(&dest->stats);
+
+		write_lock_bh(&__ip_vs_svc_lock);
+
+		/*
+		 * Wait until all other svc users go away.
+		 */
+		IP_VS_WAIT_WHILE(atomic_read(&svc->usecnt) > 1);
+
+		list_add(&dest->n_list, &svc->destinations);
+		svc->num_dests++;
+
+		/* call the update_service function of its scheduler */
+		svc->scheduler->update_service(svc);
+
+		write_unlock_bh(&__ip_vs_svc_lock);
+		return 0;
+	}
+
+	/*
+	 * Allocate and initialize the dest structure
+	 */
+	ret = ip_vs_new_dest(svc, udest, &dest);
+	if (ret) {
+		return ret;
+	}
+
+	/*
+	 * Add the dest entry into the list
+	 */
+	atomic_inc(&dest->refcnt);
+
+	write_lock_bh(&__ip_vs_svc_lock);
+
+	/*
+	 * Wait until all other svc users go away.
+	 */
+	IP_VS_WAIT_WHILE(atomic_read(&svc->usecnt) > 1);
+
+	list_add(&dest->n_list, &svc->destinations);
+	svc->num_dests++;
+
+	/* call the update_service function of its scheduler */
+	svc->scheduler->update_service(svc);
+
+	write_unlock_bh(&__ip_vs_svc_lock);
+
+	LeaveFunction(2);
+
+	return 0;
+}
+
+
+/*
+ *	Edit a destination in the given service
+ */
+static int
+ip_vs_edit_dest(struct ip_vs_service *svc, struct ip_vs_dest_user *udest)
+{
+	struct ip_vs_dest *dest;
+	__u32 daddr = udest->addr;
+	__u16 dport = udest->port;
+
+	EnterFunction(2);
+
+	if (udest->weight < 0) {
+		IP_VS_ERR("ip_vs_edit_dest(): server weight less than zero\n");
+		return -ERANGE;
+	}
+
+	if (udest->l_threshold > udest->u_threshold) {
+		IP_VS_ERR("ip_vs_edit_dest(): lower threshold is higher than "
+			  "upper threshold\n");
+		return -ERANGE;
+	}
+
+	/*
+	 *  Lookup the destination list
+	 */
+	dest = ip_vs_lookup_dest(svc, daddr, dport);
+	if (dest == NULL) {
+		IP_VS_DBG(1, "ip_vs_edit_dest(): dest doesn't exist\n");
+		return -ENOENT;
+	}
+
+	__ip_vs_update_dest(svc, dest, udest);
+
+	write_lock_bh(&__ip_vs_svc_lock);
+
+	/* Wait until all other svc users go away */
+	while (atomic_read(&svc->usecnt) > 1) {};
+
+	/* call the update_service, because server weight may be changed */
+	svc->scheduler->update_service(svc);
+
+	write_unlock_bh(&__ip_vs_svc_lock);
+
+	LeaveFunction(2);
+
+	return 0;
+}
+
+
+/*
+ *	Delete a destination (must be already unlinked from the service)
+ */
+static void __ip_vs_del_dest(struct ip_vs_dest *dest)
+{
+	ip_vs_kill_estimator(&dest->stats);
+
+	/*
+	 *  Remove it from the d-linked list with the real services.
+	 */
+	write_lock_bh(&__ip_vs_rs_lock);
+	ip_vs_rs_unhash(dest);
+	write_unlock_bh(&__ip_vs_rs_lock);
+
+	/*
+	 *  Decrease the refcnt of the dest, and free the dest
+	 *  if nobody refers to it (refcnt=0). Otherwise, throw
+	 *  the destination into the trash.
+	 */
+	if (atomic_dec_and_test(&dest->refcnt)) {
+		ip_vs_dst_reset(dest);
+		/* simply decrease svc->refcnt here, let the caller check
+		   and release the service if nobody refers to it.
+		   Only user context can release destination and service,
+		   and only one user context can update virtual service at a
+		   time, so the operation here is OK */
+		atomic_dec(&dest->svc->refcnt);
+		kfree(dest);
+	} else {
+		IP_VS_DBG(3, "Moving dest %u.%u.%u.%u:%u into trash, refcnt=%d\n",
+			  NIPQUAD(dest->addr), ntohs(dest->port),
+			  atomic_read(&dest->refcnt));
+		list_add(&dest->n_list, &ip_vs_dest_trash);
+		atomic_inc(&dest->refcnt);
+	}
+}
+
+
+/*
+ *	Unlink a destination from the given service
+ */
+static void __ip_vs_unlink_dest(struct ip_vs_service *svc,
+				struct ip_vs_dest *dest,
+				int svcupd)
+{
+	dest->flags &= ~IP_VS_DEST_F_AVAILABLE;
+
+	/*
+	 *  Remove it from the d-linked destination list.
+	 */
+	list_del(&dest->n_list);
+	svc->num_dests--;
+	if (svcupd) {
+		/*
+		 *  Call the update_service function of its scheduler
+		 */
+		svc->scheduler->update_service(svc);
+	}
+}
+
+
+/*
+ *	Delete a destination server in the given service
+ */
+static int
+ip_vs_del_dest(struct ip_vs_service *svc,struct ip_vs_dest_user *udest)
+{
+	struct ip_vs_dest *dest;
+	__u32 daddr = udest->addr;
+	__u16 dport = udest->port;
+
+	EnterFunction(2);
+
+	dest = ip_vs_lookup_dest(svc, daddr, dport);
+	if (dest == NULL) {
+		IP_VS_DBG(1, "ip_vs_del_dest(): destination not found!\n");
+		return -ENOENT;
+	}
+
+	write_lock_bh(&__ip_vs_svc_lock);
+
+	/*
+	 *	Wait until all other svc users go away.
+	 */
+	IP_VS_WAIT_WHILE(atomic_read(&svc->usecnt) > 1);
+
+	/*
+	 *	Unlink dest from the service
+	 */
+	__ip_vs_unlink_dest(svc, dest, 1);
+
+	write_unlock_bh(&__ip_vs_svc_lock);
+
+	/*
+	 *	Delete the destination
+	 */
+	__ip_vs_del_dest(dest);
+
+	LeaveFunction(2);
+
+	return 0;
+}
+
+
+/*
+ *	Add a service into the service hash table
+ */
+static int
+ip_vs_add_service(struct ip_vs_service_user *u, struct ip_vs_service **svc_p)
+{
+	int ret = 0;
+	struct ip_vs_scheduler *sched = NULL;
+	struct ip_vs_service *svc = NULL;
+
+	/* increase the module use count */
+	ip_vs_use_count_inc();
+
+	/* Lookup the scheduler by 'u->sched_name' */
+	sched = ip_vs_scheduler_get(u->sched_name);
+	if (sched == NULL) {
+		IP_VS_INFO("Scheduler module ip_vs_%s not found\n",
+			   u->sched_name);
+		ret = -ENOENT;
+		goto out_mod_dec;
+	}
+
+	svc = (struct ip_vs_service *)
+		kmalloc(sizeof(struct ip_vs_service), GFP_ATOMIC);
+	if (svc == NULL) {
+		IP_VS_DBG(1, "ip_vs_add_service: kmalloc failed.\n");
+		ret = -ENOMEM;
+		goto out_err;
+	}
+	memset(svc, 0, sizeof(struct ip_vs_service));
+
+	/* I'm the first user of the service */
+	atomic_set(&svc->usecnt, 1);
+	atomic_set(&svc->refcnt, 0);
+
+	svc->protocol = u->protocol;
+	svc->addr = u->addr;
+	svc->port = u->port;
+	svc->fwmark = u->fwmark;
+	svc->flags = u->flags;
+	svc->timeout = u->timeout * HZ;
+	svc->netmask = u->netmask;
+
+	INIT_LIST_HEAD(&svc->destinations);
+	rwlock_init(&svc->sched_lock);
+	spin_lock_init(&svc->stats.lock);
+
+	/* Bind the scheduler */
+	ret = ip_vs_bind_scheduler(svc, sched);
+	if (ret)
+		goto out_err;
+	sched = NULL;
+
+	/* Update the virtual service counters */
+	if (svc->port == FTPPORT)
+		atomic_inc(&ip_vs_ftpsvc_counter);
+	else if (svc->port == 0)
+		atomic_inc(&ip_vs_nullsvc_counter);
+
+	ip_vs_new_estimator(&svc->stats);
+	ip_vs_num_services++;
+
+	/* Hash the service into the service table */
+	write_lock_bh(&__ip_vs_svc_lock);
+	ip_vs_svc_hash(svc);
+	write_unlock_bh(&__ip_vs_svc_lock);
+
+	*svc_p = svc;
+	return 0;
+
+  out_err:
+	if (svc != NULL) {
+		if (svc->scheduler)
+			ip_vs_unbind_scheduler(svc);
+		if (svc->inc) {
+			local_bh_disable();
+			ip_vs_app_inc_put(svc->inc);
+			local_bh_enable();
+		}
+		kfree(svc);
+	}
+	ip_vs_scheduler_put(sched);
+
+  out_mod_dec:
+	/* decrease the module use count */
+	ip_vs_use_count_dec();
+
+	return ret;
+}
+
+
+/*
+ *	Edit a service and bind it with a new scheduler
+ */
+static int
+ip_vs_edit_service(struct ip_vs_service *svc, struct ip_vs_service_user *u)
+{
+	struct ip_vs_scheduler *sched, *old_sched;
+	int ret = 0;
+
+	/*
+	 * Lookup the scheduler, by 'u->sched_name'
+	 */
+	sched = ip_vs_scheduler_get(u->sched_name);
+	if (sched == NULL) {
+		IP_VS_INFO("Scheduler module ip_vs_%s not found\n",
+			   u->sched_name);
+		return -ENOENT;
+	}
+	old_sched = sched;
+
+	write_lock_bh(&__ip_vs_svc_lock);
+
+	/*
+	 * Wait until all other svc users go away.
+	 */
+	IP_VS_WAIT_WHILE(atomic_read(&svc->usecnt) > 1);
+
+	/*
+	 * Set the flags and timeout value
+	 */
+	svc->flags = u->flags | IP_VS_SVC_F_HASHED;
+	svc->timeout = u->timeout * HZ;
+	svc->netmask = u->netmask;
+
+	old_sched = svc->scheduler;
+	if (sched != old_sched) {
+		/*
+		 * Unbind the old scheduler
+		 */
+		if ((ret = ip_vs_unbind_scheduler(svc))) {
+			old_sched = sched;
+			goto out;
+		}
+
+		/*
+		 * Bind the new scheduler
+		 */
+		if ((ret = ip_vs_bind_scheduler(svc, sched))) {
+			/*
+			 * If ip_vs_bind_scheduler fails, restore the old
+			 * scheduler.
+			 * The main reason of failure is out of memory.
+			 *
+			 * The question is if the old scheduler can be
+			 * restored all the time. TODO: if it cannot be
+			 * restored some time, we must delete the service,
+			 * otherwise the system may crash.
+			 */
+			ip_vs_bind_scheduler(svc, old_sched);
+			old_sched = sched;
+			goto out;
+		}
+	}
+
+  out:
+	write_unlock_bh(&__ip_vs_svc_lock);
+
+	if (old_sched)
+		ip_vs_scheduler_put(old_sched);
+
+	return ret;
+}
+
+
+/*
+ *	Delete a service from the service list
+ *	- The service must be unlinked, unlocked and not referenced!
+ *	- We are called under _bh lock
+ */
+static void __ip_vs_del_service(struct ip_vs_service *svc)
+{
+	struct ip_vs_dest *dest, *nxt;
+	struct ip_vs_scheduler *old_sched;
+
+	ip_vs_num_services--;
+	ip_vs_kill_estimator(&svc->stats);
+
+	/* Unbind scheduler */
+	old_sched = svc->scheduler;
+	ip_vs_unbind_scheduler(svc);
+	if (old_sched)
+		ip_vs_scheduler_put(old_sched);
+
+	/* Unbind app inc */
+	if (svc->inc) {
+		ip_vs_app_inc_put(svc->inc);
+		svc->inc = NULL;
+	}
+
+	/*
+	 *    Unlink the whole destination list
+	 */
+	list_for_each_entry_safe(dest, nxt, &svc->destinations, n_list) {
+		__ip_vs_unlink_dest(svc, dest, 0);
+		__ip_vs_del_dest(dest);
+	}
+
+	/*
+	 *    Update the virtual service counters
+	 */
+	if (svc->port == FTPPORT)
+		atomic_dec(&ip_vs_ftpsvc_counter);
+	else if (svc->port == 0)
+		atomic_dec(&ip_vs_nullsvc_counter);
+
+	/*
+	 *    Free the service if nobody refers to it
+	 */
+	if (atomic_read(&svc->refcnt) == 0)
+		kfree(svc);
+
+	/* decrease the module use count */
+	ip_vs_use_count_dec();
+}
+
+/*
+ *	Delete a service from the service list
+ */
+static int ip_vs_del_service(struct ip_vs_service *svc)
+{
+	if (svc == NULL)
+		retu