int start_sync_thread(int state, char *mcast_ifn)
{
DECLARE_COMPLETION(startup);
pid_t pid;
if (sync_pid)
return -EEXIST;
IP_VS_DBG(7, "%s: pid %d\n", __FUNCTION__, current->pid);
IP_VS_DBG(7, "Each ip_vs_sync_conn entry need %d bytes\n",
sizeof(struct ip_vs_sync_conn));
ip_vs_sync_state = state;
strcpy(ip_vs_mcast_ifn, mcast_ifn);
repeat:
if ((pid = kernel_thread(fork_sync_thread, &startup, 0)) < 0) {
IP_VS_ERR("could not create fork_sync_thread due to %d... "
"retrying.\n", pid);
current->state = TASK_UNINTERRUPTIBLE;
schedule_timeout(HZ);
goto repeat;
}
wait_for_completion(&startup);
return 0;
}
/*
* Source Hashing scheduling
*/
static struct ip_vs_dest *
ip_vs_sh_schedule(struct ip_vs_service *svc, struct iphdr *iph)
{
struct ip_vs_dest *dest;
struct ip_vs_sh_bucket *tbl;
IP_VS_DBG(6, "ip_vs_sh_schedule(): Scheduling...\n");
tbl = (struct ip_vs_sh_bucket *)svc->sched_data;
dest = ip_vs_sh_get(tbl, iph->saddr);
if (!dest
|| !(dest->flags & IP_VS_DEST_F_AVAILABLE)
|| atomic_read(&dest->weight) <= 0
|| is_overloaded(dest)) {
return NULL;
}
IP_VS_DBG(6, "SH: source IP address %u.%u.%u.%u "
"--> server %u.%u.%u.%u:%d\n",
NIPQUAD(iph->saddr),
NIPQUAD(dest->addr),
ntohs(dest->port));
return dest;
}
/*
* IPVS main scheduling function
* It selects a server according to the virtual service, and
* creates a connection entry.
*/
static struct ip_vs_conn *
ip_vs_schedule(struct ip_vs_service *svc, struct iphdr *iph)
{
struct ip_vs_conn *cp = NULL;
struct ip_vs_dest *dest;
const __u16 *portp;
/*
* Persistent service
*/
if (svc->flags & IP_VS_SVC_F_PERSISTENT)
return ip_vs_sched_persist(svc, iph);
/*
* Non-persistent service
*/
portp = (__u16 *)&(((char *)iph)[iph->ihl*4]);
if (!svc->fwmark && portp[1] != svc->port) {
if (!svc->port)
IP_VS_ERR("Schedule: port zero only supported "
"in persistent services, "
"check your ipvs configuration\n");
return NULL;
}
dest = svc->scheduler->schedule(svc, iph);
if (dest == NULL) {
IP_VS_DBG(1, "Schedule: no dest found.\n");
return NULL;
}
/*
* Create a connection entry.
*/
cp = ip_vs_conn_new(iph->protocol,
iph->saddr, portp[0],
iph->daddr, portp[1],
dest->addr, dest->port?dest->port:portp[1],
0,
dest);
if (cp == NULL)
return NULL;
/*
* Increase the inactive connection counter because it is in
* Syn-Received state (inactive) when the connection is created.
*/
atomic_inc(&dest->inactconns);
IP_VS_DBG(6, "Schedule fwd:%c s:%s c:%u.%u.%u.%u:%u v:%u.%u.%u.%u:%u "
"d:%u.%u.%u.%u:%u flg:%X cnt:%d\n",
ip_vs_fwd_tag(cp), ip_vs_state_name(cp->state),
NIPQUAD(cp->caddr), ntohs(cp->cport),
NIPQUAD(cp->vaddr), ntohs(cp->vport),
NIPQUAD(cp->daddr), ntohs(cp->dport),
cp->flags, atomic_read(&cp->refcnt));
return cp;
}
/*
* IPVS main scheduling function
* It selects a server according to the virtual service, and
* creates a connection entry.
* Protocols supported: TCP, UDP
*/
struct ip_vs_conn *
ip_vs_schedule(struct ip_vs_service *svc, const struct sk_buff *skb)
{
struct ip_vs_conn *cp = NULL;
struct iphdr *iph = ip_hdr(skb);
struct ip_vs_dest *dest;
__be16 _ports[2], *pptr;
pptr = skb_header_pointer(skb, iph->ihl*4,
sizeof(_ports), _ports);
if (pptr == NULL)
return NULL;
/*
* Persistent service
*/
if (svc->flags & IP_VS_SVC_F_PERSISTENT)
return ip_vs_sched_persist(svc, skb, pptr);
/*
* Non-persistent service
*/
if (!svc->fwmark && pptr[1] != svc->port) {
if (!svc->port)
IP_VS_ERR("Schedule: port zero only supported "
"in persistent services, "
"check your ipvs configuration\n");
return NULL;
}
dest = svc->scheduler->schedule(svc, skb);
if (dest == NULL) {
IP_VS_DBG(1, "Schedule: no dest found.\n");
return NULL;
}
/*
* Create a connection entry.
*/
cp = ip_vs_conn_new(iph->protocol,
iph->saddr, pptr[0],
iph->daddr, pptr[1],
dest->addr, dest->port?dest->port:pptr[1],
0,
dest);
if (cp == NULL)
return NULL;
IP_VS_DBG(6, "Schedule fwd:%c c:%u.%u.%u.%u:%u v:%u.%u.%u.%u:%u "
"d:%u.%u.%u.%u:%u conn->flags:%X conn->refcnt:%d\n",
ip_vs_fwd_tag(cp),
NIPQUAD(cp->caddr), ntohs(cp->cport),
NIPQUAD(cp->vaddr), ntohs(cp->vport),
NIPQUAD(cp->daddr), ntohs(cp->dport),
cp->flags, atomic_read(&cp->refcnt));
ip_vs_conn_stats(cp, svc);
return cp;
}
/*
* Weighted Least Connection scheduling
*/
static struct ip_vs_dest *
ip_vs_wlc_schedule(struct ip_vs_service *svc, struct iphdr *iph)
{
register struct list_head *l, *e;
struct ip_vs_dest *dest, *least;
unsigned int loh, doh;
IP_VS_DBG(6, "ip_vs_wlc_schedule(): Scheduling...\n");
/*
* We calculate the load of each dest server as follows:
* (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 connections.
*/
l = &svc->destinations;
for (e=l->next; e!=l; e=e->next) {
least = list_entry(e, struct ip_vs_dest, n_list);
if (atomic_read(&least->weight) > 0) {
loh = ip_vs_wlc_dest_overhead(least);
goto nextstage;
}
}
return NULL;
/*
* Find the destination with the least load.
*/
nextstage:
for (e=e->next; e!=l; e=e->next) {
dest = list_entry(e, struct ip_vs_dest, n_list);
doh = ip_vs_wlc_dest_overhead(dest);
if (loh * atomic_read(&dest->weight) >
doh * atomic_read(&least->weight)) {
least = dest;
loh = doh;
}
}
IP_VS_DBG(6, "WLC: server %u.%u.%u.%u:%u "
"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;
}
/*
* Locality-Based (weighted) Least-Connection scheduling
*/
static struct ip_vs_dest *
ip_vs_lblcr_schedule(struct ip_vs_service *svc, struct iphdr *iph)
{
struct ip_vs_dest *dest;
struct ip_vs_lblcr_table *tbl;
struct ip_vs_lblcr_entry *en;
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;
}
/*
* Least Connection scheduling
*/
static struct ip_vs_dest *
ip_vs_lc_schedule(struct ip_vs_service *svc, struct iphdr *iph)
{
struct list_head *l, *e;
struct ip_vs_dest *dest, *least;
unsigned int loh, doh;
IP_VS_DBG(6, "ip_vs_lc_schedule(): Scheduling...\n");
/*
* Simply select the server with the least number of
* (activeconns<<5) + inactconns
* Except whose weight is equal to zero.
* If the weight is equal to zero, it means that the server is
* quiesced, the existing connections to the server still get
* served, but no new connection is assigned to the server.
*/
l = &svc->destinations;
for (e=l->next; e!=l; e=e->next) {
least = list_entry (e, struct ip_vs_dest, n_list);
if (atomic_read(&least->weight) > 0) {
loh = ip_vs_lc_dest_overhead(least);
goto nextstage;
}
}
return NULL;
/*
* Find the destination with the least load.
*/
nextstage:
for (e=e->next; e!=l; e=e->next) {
dest = list_entry(e, struct ip_vs_dest, n_list);
if (atomic_read(&dest->weight) == 0)
continue;
doh = ip_vs_lc_dest_overhead(dest);
if (doh < loh) {
least = dest;
loh = doh;
}
}
IP_VS_DBG(6, "LC: server %u.%u.%u.%u:%u activeconns %d inactconns %d\n",
NIPQUAD(least->addr), ntohs(least->port),
atomic_read(&least->activeconns),
atomic_read(&least->inactconns));
return least;
}
int stop_sync_thread(void)
{
DECLARE_WAITQUEUE(wait, current);
if (!sync_pid)
return -ESRCH;
IP_VS_DBG(7, "%s: pid %d\n", __FUNCTION__, current->pid);
IP_VS_INFO("stopping sync thread %d ...\n", sync_pid);
__set_current_state(TASK_UNINTERRUPTIBLE);
add_wait_queue(&stop_sync_wait, &wait);
ip_vs_sync_state = IP_VS_STATE_NONE;
stop_sync = 1;
wake_up(&sync_wait);
schedule();
__set_current_state(TASK_RUNNING);
remove_wait_queue(&stop_sync_wait, &wait);
/* Note: no need to reap the sync thread, because its parent
process is the init process */
if (stop_sync)
IP_VS_BUG();
return 0;
}
/*
* Source Hashing scheduling
*/
static struct ip_vs_dest *
ip_vs_sh_schedule(struct ip_vs_service *svc, const struct sk_buff *skb)
{
struct ip_vs_dest *dest;
struct ip_vs_sh_state *s;
struct ip_vs_iphdr iph;
ip_vs_fill_iph_addr_only(svc->af, skb, &iph);
IP_VS_DBG(6, "ip_vs_sh_schedule(): Scheduling...\n");
s = (struct ip_vs_sh_state *) svc->sched_data;
dest = ip_vs_sh_get(svc->af, s, &iph.saddr);
if (!dest
|| !(dest->flags & IP_VS_DEST_F_AVAILABLE)
|| atomic_read(&dest->weight) <= 0
|| is_overloaded(dest)) {
ip_vs_scheduler_err(svc, "no destination available");
return NULL;
}
IP_VS_DBG_BUF(6, "SH: source IP address %s --> server %s:%d\n",
IP_VS_DBG_ADDR(svc->af, &iph.saddr),
IP_VS_DBG_ADDR(svc->af, &dest->addr),
ntohs(dest->port));
return dest;
}
static inline struct ip_vs_dest *
__ip_vs_wlc_schedule(struct ip_vs_service *svc, struct iphdr *iph)
{
register struct list_head *l, *e;
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.
*/
l = &svc->destinations;
for (e=l->next; e!=l; e=e->next) {
least = list_entry(e, struct ip_vs_dest, n_list);
if (atomic_read(&least->weight) > 0) {
loh = atomic_read(&least->activeconns) * 50
+ atomic_read(&least->inactconns);
goto nextstage;
}
}
return NULL;
/*
* Find the destination with the least load.
*/
nextstage:
for (e=e->next; e!=l; e=e->next) {
dest = list_entry(e, struct ip_vs_dest, n_list);
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, "LBLC: 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;
}
static struct ip_vs_dest *
ip_vs_dh_schedule(struct ip_vs_service *svc, const struct sk_buff *skb)
{
struct ip_vs_dest *dest;
struct ip_vs_dh_bucket *tbl;
struct ip_vs_iphdr iph;
ip_vs_fill_iphdr(svc->af, skb_network_header(skb), &iph);
IP_VS_DBG(6, "%s(): Scheduling...\n", __func__);
tbl = (struct ip_vs_dh_bucket *)svc->sched_data;
dest = ip_vs_dh_get(svc->af, tbl, &iph.daddr);
if (!dest
|| !(dest->flags & IP_VS_DEST_F_AVAILABLE)
|| atomic_read(&dest->weight) <= 0
|| is_overloaded(dest)) {
return NULL;
}
IP_VS_DBG_BUF(6, "DH: destination IP address %s --> server %s:%d\n",
IP_VS_DBG_ADDR(svc->af, &iph.daddr),
IP_VS_DBG_ADDR(svc->af, &dest->addr),
ntohs(dest->port));
return dest;
}
static struct ip_vs_conn *
ah_conn_out_get(const struct sk_buff *skb, struct ip_vs_protocol *pp,
const struct iphdr *iph, unsigned int proto_off, int inverse)
{
struct ip_vs_conn *cp;
if (likely(!inverse)) {
cp = ip_vs_conn_out_get(IPPROTO_UDP,
iph->saddr,
htons(PORT_ISAKMP),
iph->daddr,
htons(PORT_ISAKMP));
} else {
cp = ip_vs_conn_out_get(IPPROTO_UDP,
iph->daddr,
htons(PORT_ISAKMP),
iph->saddr,
htons(PORT_ISAKMP));
}
if (!cp) {
IP_VS_DBG(12, "Unknown ISAKMP entry for inout packet "
"%s%s %u.%u.%u.%u->%u.%u.%u.%u\n",
inverse ? "ICMP+" : "",
pp->name,
NIPQUAD(iph->saddr),
NIPQUAD(iph->daddr));
}
return cp;
}
/*
* Source Hashing scheduling
*/
static struct ip_vs_dest *
ip_vs_sh_schedule(struct ip_vs_service *svc, const struct sk_buff *skb,
struct ip_vs_iphdr *iph)
{
struct ip_vs_dest *dest;
struct ip_vs_sh_state *s;
__be16 port = 0;
IP_VS_DBG(6, "ip_vs_sh_schedule(): Scheduling...\n");
if (svc->flags & IP_VS_SVC_F_SCHED_SH_PORT)
port = ip_vs_sh_get_port(skb, iph);
s = (struct ip_vs_sh_state *) svc->sched_data;
if (svc->flags & IP_VS_SVC_F_SCHED_SH_FALLBACK)
dest = ip_vs_sh_get_fallback(svc, s, &iph->saddr, port);
else
dest = ip_vs_sh_get(svc, s, &iph->saddr, port);
if (!dest) {
ip_vs_scheduler_err(svc, "no destination available");
return NULL;
}
IP_VS_DBG_BUF(6, "SH: source IP address %s --> server %s:%d\n",
IP_VS_DBG_ADDR(svc->af, &iph->saddr),
IP_VS_DBG_ADDR(svc->af, &dest->addr),
ntohs(dest->port));
return dest;
}
static struct rtable *
__ip_vs_get_out_rt(struct ip_vs_conn *cp, u32 rtos)
{
struct rtable *rt; /* Route to the other host */
struct ip_vs_dest *dest = cp->dest;
if (dest) {
spin_lock(&dest->dst_lock);
if (!(rt = (struct rtable *)
__ip_vs_dst_check(dest, rtos, 0))) {
struct flowi fl = {
.oif = 0,
.nl_u = {
.ip4_u = {
.daddr = dest->addr.ip,
.saddr = 0,
.tos = rtos, } },
};
if (ip_route_output_key(&init_net, &rt, &fl)) {
spin_unlock(&dest->dst_lock);
IP_VS_DBG_RL("ip_route_output error, dest: %pI4\n",
&dest->addr.ip);
return NULL;
}
__ip_vs_dst_set(dest, rtos, dst_clone(&rt->u.dst));
IP_VS_DBG(10, "new dst %pI4, refcnt=%d, rtos=%X\n",
&dest->addr.ip,
atomic_read(&rt->u.dst.__refcnt), rtos);
}
spin_unlock(&dest->dst_lock);
} else {
/*
* Get route to destination or remote server
* rt_mode: flags, &1=Allow local dest, &2=Allow non-local dest,
* &4=Allow redirect from remote daddr to local
*/
static struct rtable *
__ip_vs_get_out_rt(struct sk_buff *skb, struct ip_vs_dest *dest,
__be32 daddr, u32 rtos, int rt_mode)
{
struct net *net = dev_net(skb_dst(skb)->dev);
struct rtable *rt; /* Route to the other host */
struct rtable *ort; /* Original route */
int local;
if (dest) {
spin_lock(&dest->dst_lock);
if (!(rt = (struct rtable *)
__ip_vs_dst_check(dest, rtos))) {
struct flowi fl = {
.oif = 0,
.nl_u = {
.ip4_u = {
.daddr = dest->addr.ip,
.saddr = 0,
.tos = rtos, } },
};
if (ip_route_output_key(net, &rt, &fl)) {
spin_unlock(&dest->dst_lock);
IP_VS_DBG_RL("ip_route_output error, dest: %pI4\n",
&dest->addr.ip);
return NULL;
}
__ip_vs_dst_set(dest, rtos, dst_clone(&rt->dst), 0);
IP_VS_DBG(10, "new dst %pI4, refcnt=%d, rtos=%X\n",
&dest->addr.ip,
atomic_read(&rt->dst.__refcnt), rtos);
}
spin_unlock(&dest->dst_lock);
} else {
/*
* Locality-Based (weighted) Least-Connection scheduling
*/
static struct ip_vs_dest *
ip_vs_lblc_schedule(struct ip_vs_service *svc, struct iphdr *iph)
{
struct ip_vs_dest *dest;
struct ip_vs_lblc_table *tbl;
struct ip_vs_lblc_entry *en;
IP_VS_DBG(6, "ip_vs_lblc_schedule(): Scheduling...\n");
tbl = (struct ip_vs_lblc_table *)svc->sched_data;
en = ip_vs_lblc_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_lblc_new(iph->daddr, dest);
if (en == NULL) {
return NULL;
}
ip_vs_lblc_hash(tbl, en);
} else {
dest = en->dest;
if (!(dest->flags & IP_VS_DEST_F_AVAILABLE)
|| atomic_read(&dest->weight) <= 0
|| is_overloaded(dest, svc)) {
dest = __ip_vs_wlc_schedule(svc, iph);
if (dest == NULL) {
IP_VS_DBG(1, "no destination available\n");
return NULL;
}
atomic_dec(&en->dest->refcnt);
atomic_inc(&dest->refcnt);
en->dest = dest;
}
}
en->lastuse = jiffies;
IP_VS_DBG(6, "LBLC: 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;
}
/*
* Check if skb has user data.
* Attention: decrease iph len also.
*/
static inline int
syn_proxy_ack_has_data(struct sk_buff *skb, struct ip_vs_iphdr *iph,
struct tcphdr *th)
{
IP_VS_DBG(6, "tot_len = %u, iph_len = %u, tcph_len = %u\n",
skb->len, iph->len, th->doff * 4);
return (skb->len - iph->len - th->doff * 4) != 0;
}
static inline bool ensure_mtu_is_adequate(int skb_af, int rt_mode,
struct ip_vs_iphdr *ipvsh,
struct sk_buff *skb, int mtu)
{
#ifdef CONFIG_IP_VS_IPV6
if (skb_af == AF_INET6) {
struct net *net = dev_net(skb_dst(skb)->dev);
if (unlikely(__mtu_check_toobig_v6(skb, mtu))) {
if (!skb->dev)
skb->dev = net->loopback_dev;
/* only send ICMP too big on first fragment */
if (!ipvsh->fragoffs)
icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu);
IP_VS_DBG(1, "frag needed for %pI6c\n",
&ipv6_hdr(skb)->saddr);
return false;
}
} else
#endif
{
struct netns_ipvs *ipvs = net_ipvs(skb_net(skb));
/* If we're going to tunnel the packet and pmtu discovery
* is disabled, we'll just fragment it anyway
*/
if ((rt_mode & IP_VS_RT_MODE_TUNNEL) && !sysctl_pmtu_disc(ipvs))
return true;
if (unlikely(ip_hdr(skb)->frag_off & htons(IP_DF) &&
skb->len > mtu && !skb_is_gso(skb))) {
icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED,
htonl(mtu));
IP_VS_DBG(1, "frag needed for %pI4\n",
&ip_hdr(skb)->saddr);
return false;
}
}
return true;
}
static void ip_vs_sh_done_svc(struct ip_vs_service *svc)
{
struct ip_vs_sh_state *s = svc->sched_data;
/* got to clean up hash buckets here */
ip_vs_sh_flush(s);
/* release the table itself */
kfree_rcu(s, rcu_head);
IP_VS_DBG(6, "SH hash table (memory=%Zdbytes) released\n",
sizeof(struct ip_vs_sh_bucket)*IP_VS_SH_TAB_SIZE);
}
/*
* Round-Robin Scheduling
*/
static struct ip_vs_dest *
ip_vs_rr_schedule(struct ip_vs_service *svc, const struct sk_buff *skb)
{
struct list_head *p, *q;
struct ip_vs_dest *dest;
IP_VS_DBG(6, "ip_vs_rr_schedule(): Scheduling...\n");
write_lock(&svc->sched_lock);
p = (struct list_head *)svc->sched_data;
p = p->next;
q = p;
do {
/* skip list head */
if (q == &svc->destinations) {
q = q->next;
continue;
}
dest = list_entry(q, struct ip_vs_dest, n_list);
if (!(dest->flags & IP_VS_DEST_F_OVERLOAD) &&
atomic_read(&dest->weight) > 0)
/* HIT */
goto out;
q = q->next;
} while (q != p);
write_unlock(&svc->sched_lock);
return NULL;
out:
svc->sched_data = q;
write_unlock(&svc->sched_lock);
IP_VS_DBG(6, "RR: server %u.%u.%u.%u:%u "
"activeconns %d refcnt %d weight %d\n",
NIPQUAD(dest->addr), ntohs(dest->port),
atomic_read(&dest->activeconns),
atomic_read(&dest->refcnt), atomic_read(&dest->weight));
return dest;
}
/* 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;
}
/*
* Source Hashing scheduling
*/
static struct ip_vs_dest *
ip_vs_sh_schedule(struct ip_vs_service *svc, const struct sk_buff *skb)
{
struct ip_vs_dest *dest;
struct ip_vs_sh_bucket *tbl;
struct iphdr *iph = ip_hdr(skb);
IP_VS_DBG(6, "ip_vs_sh_schedule(): Scheduling...\n");
tbl = (struct ip_vs_sh_bucket *)svc->sched_data;
dest = ip_vs_sh_get(tbl, iph->saddr);
if (!dest
|| !(dest->flags & IP_VS_DEST_F_AVAILABLE)
|| atomic_read(&dest->weight) <= 0
|| is_overloaded(dest)) {
return NULL;
}
IP_VS_DBG(6, "SH: source IP address %pI4 --> server %pI4:%u\n",
&iph->saddr, &dest->addr, ntohs(dest->port));
return dest;
}
static int ip_vs_sh_done_svc(struct ip_vs_service *svc)
{
struct ip_vs_sh_bucket *tbl = svc->sched_data;
/* got to clean up hash buckets here */
ip_vs_sh_flush(tbl);
/* release the table itself */
kfree(svc->sched_data);
IP_VS_DBG(6, "SH hash table (memory=%dbytes) released\n",
sizeof(struct ip_vs_sh_bucket)*IP_VS_SH_TAB_SIZE);
return 0;
}
static int ip_vs_dh_done_svc(struct ip_vs_service *svc)
{
struct ip_vs_dh_bucket *tbl = svc->sched_data;
ip_vs_dh_flush(tbl);
kfree(svc->sched_data);
IP_VS_DBG(6, "DH hash table (memory=%Zdbytes) released\n",
sizeof(struct ip_vs_dh_bucket)*IP_VS_DH_TAB_SIZE);
return 0;
}
/*
* Weighted Round-Robin Scheduling
*/
static struct ip_vs_dest *
ip_vs_wrr_schedule(struct ip_vs_service *svc, struct iphdr *iph)
{
struct ip_vs_dest *dest;
struct ip_vs_wrr_mark *mark = svc->sched_data;
IP_VS_DBG(6, "ip_vs_wrr_schedule(): Scheduling...\n");
/*
* This loop will always terminate, because 0<mark->cw<max_weight,
* and at least one server has its weight equal to max_weight.
*/
write_lock(&svc->sched_lock);
while (1) {
if (mark->cl == &svc->destinations) {
/* it is at the head of the destination list */
if (mark->cl == mark->cl->next) {
/* no dest entry */
write_unlock(&svc->sched_lock);
return NULL;
}
mark->cl = svc->destinations.next;
mark->cw -= mark->di;
if (mark->cw <= 0) {
mark->cw = mark->mw;
/*
* Still zero, which means no available servers.
*/
if (mark->cw == 0) {
mark->cl = &svc->destinations;
write_unlock(&svc->sched_lock);
IP_VS_INFO("ip_vs_wrr_schedule(): "
"no available servers\n");
return NULL;
}
}
}
else mark->cl = mark->cl->next;
if (mark->cl != &svc->destinations) {
/* not at the head of the list */
dest = list_entry(mark->cl, struct ip_vs_dest, n_list);
if (atomic_read(&dest->weight) >= mark->cw) {
write_unlock(&svc->sched_lock);
break;
}
}
}
/*
* Update out-in ack_seqs: include th->ack_seq, sack opt
* and also correct tcph->check.
*/
void ip_vs_synproxy_dnat_handler(struct tcphdr *tcph, struct ip_vs_seq *sp_seq)
{
__u32 old_ack_seq;
if (sp_seq->delta != 0) {
old_ack_seq = ntohl(tcph->ack_seq);
tcph->ack_seq = htonl((__u32) (old_ack_seq - sp_seq->delta));
syn_proxy_seq_csum_update(tcph, htonl(old_ack_seq),
tcph->ack_seq);
syn_proxy_filter_opt_outin(tcph, sp_seq);
IP_VS_DBG(6,
"tcp_dnat_handler: tcph->ack_seq %u => %u, delta = %u \n",
old_ack_seq, htonl(tcph->ack_seq), sp_seq->delta);
}
}
/* 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;
}
static int ip_vs_lblc_done_svc(struct ip_vs_service *svc)
{
struct ip_vs_lblc_table *tbl = svc->sched_data;
/* remove periodic timer */
del_timer_sync(&tbl->periodic_timer);
/* got to clean up table entries here */
ip_vs_lblc_flush(tbl);
/* release the table itself */
kfree(svc->sched_data);
IP_VS_DBG(6, "LBLC hash table (memory=%dbytes) released\n",
sizeof(struct ip_vs_lblc_table));
return 0;
}
/*
* Syn-proxy snat handler:
* 1) check and stop ack storm.
* 2)Update in-out seqs: include th->seq
* and also correct tcph->check.
*
* Return 0 if ack storm is found and stoped.
*/
int ip_vs_synproxy_snat_handler(struct tcphdr *tcph, struct ip_vs_conn *cp)
{
__u32 old_seq;
if (syn_proxy_is_ack_storm(tcph, cp) == 0) {
return 0;
}
if (cp->syn_proxy_seq.delta != 0) {
old_seq = ntohl(tcph->seq);
tcph->seq = htonl((__u32) (old_seq + cp->syn_proxy_seq.delta));
syn_proxy_seq_csum_update(tcph, htonl(old_seq), tcph->seq);
IP_VS_DBG(6,
"tcp_snat_handler: tcph->seq %u => %u, delta = %u \n",
old_seq, htonl(tcph->seq), cp->syn_proxy_seq.delta);
}
return 1;
}
static int ip_vs_sh_init_svc(struct ip_vs_service *svc)
{
struct ip_vs_sh_state *s;
/* allocate the SH table for this service */
s = kzalloc(sizeof(struct ip_vs_sh_state), GFP_KERNEL);
if (s == NULL)
return -ENOMEM;
svc->sched_data = s;
IP_VS_DBG(6, "SH hash table (memory=%Zdbytes) allocated for "
"current service\n",
sizeof(struct ip_vs_sh_bucket)*IP_VS_SH_TAB_SIZE);
/* assign the hash buckets with current dests */
ip_vs_sh_reassign(s, svc);
return 0;
}
