[net-next-2.6.git] / net / ipv4 / ipvs / ip_vs_proto.c

/*
 * ip_vs_proto.c: transport protocol load balancing support for IPVS
 *
 * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
 *              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/kernel.h>
#include <linux/skbuff.h>
#include <linux/in.h>
#include <linux/ip.h>
#include <net/protocol.h>
#include <net/tcp.h>
#include <net/udp.h>
#include <asm/system.h>
#include <linux/stat.h>
#include <linux/proc_fs.h>

#include <net/ip_vs.h>


/*
 * IPVS protocols can only be registered/unregistered when the ipvs
 * module is loaded/unloaded, so no lock is needed in accessing the
 * ipvs protocol table.
 */

#define IP_VS_PROTO_TAB_SIZE		32	/* must be power of 2 */
#define IP_VS_PROTO_HASH(proto)		((proto) & (IP_VS_PROTO_TAB_SIZE-1))

static struct ip_vs_protocol *ip_vs_proto_table[IP_VS_PROTO_TAB_SIZE];


/*
 *	register an ipvs protocol
 */
static int __used __init register_ip_vs_protocol(struct ip_vs_protocol *pp)
{
	unsigned hash = IP_VS_PROTO_HASH(pp->protocol);

	pp->next = ip_vs_proto_table[hash];
	ip_vs_proto_table[hash] = pp;

	if (pp->init != NULL)
		pp->init(pp);

	return 0;
}


/*
 *	unregister an ipvs protocol
 */
static int unregister_ip_vs_protocol(struct ip_vs_protocol *pp)
{
	struct ip_vs_protocol **pp_p;
	unsigned hash = IP_VS_PROTO_HASH(pp->protocol);

	pp_p = &ip_vs_proto_table[hash];
	for (; *pp_p; pp_p = &(*pp_p)->next) {
		if (*pp_p == pp) {
			*pp_p = pp->next;
			if (pp->exit != NULL)
				pp->exit(pp);
			return 0;
		}
	}

	return -ESRCH;
}


/*
 *	get ip_vs_protocol object by its proto.
 */
struct ip_vs_protocol * ip_vs_proto_get(unsigned short proto)
{
	struct ip_vs_protocol *pp;
	unsigned hash = IP_VS_PROTO_HASH(proto);

	for (pp = ip_vs_proto_table[hash]; pp; pp = pp->next) {
		if (pp->protocol == proto)
			return pp;
	}

	return NULL;
}


/*
 *	Propagate event for state change to all protocols
 */
void ip_vs_protocol_timeout_change(int flags)
{
	struct ip_vs_protocol *pp;
	int i;

	for (i = 0; i < IP_VS_PROTO_TAB_SIZE; i++) {
		for (pp = ip_vs_proto_table[i]; pp; pp = pp->next) {
			if (pp->timeout_change)
				pp->timeout_change(pp, flags);
		}
	}
}


int *
ip_vs_create_timeout_table(int *table, int size)
{
	return kmemdup(table, size, GFP_ATOMIC);
}


/*
 *	Set timeout value for state specified by name
 */
int
ip_vs_set_state_timeout(int *table, int num, char **names, char *name, int to)
{
	int i;

	if (!table || !name || !to)
		return -EINVAL;

	for (i = 0; i < num; i++) {
		if (strcmp(names[i], name))
			continue;
		table[i] = to * HZ;
		return 0;
	}
	return -ENOENT;
}


const char * ip_vs_state_name(__u16 proto, int state)
{
	struct ip_vs_protocol *pp = ip_vs_proto_get(proto);

	if (pp == NULL || pp->state_name == NULL)
		return (IPPROTO_IP == proto) ? "NONE" : "ERR!";
	return pp->state_name(state);
}


void
ip_vs_tcpudp_debug_packet_v4(struct ip_vs_protocol *pp,
			     const struct sk_buff *skb,
			     int offset,
			     const char *msg)
{
	char buf[128];
	struct iphdr _iph, *ih;

	ih = skb_header_pointer(skb, offset, sizeof(_iph), &_iph);
	if (ih == NULL)
		sprintf(buf, "%s TRUNCATED", pp->name);
	else if (ih->frag_off & htons(IP_OFFSET))
		sprintf(buf, "%s %u.%u.%u.%u->%u.%u.%u.%u frag",
			pp->name, NIPQUAD(ih->saddr),
			NIPQUAD(ih->daddr));
	else {
		__be16 _ports[2], *pptr
;
		pptr = skb_header_pointer(skb, offset + ih->ihl*4,
					  sizeof(_ports), _ports);
		if (pptr == NULL)
			sprintf(buf, "%s TRUNCATED %u.%u.%u.%u->%u.%u.%u.%u",
				pp->name,
				NIPQUAD(ih->saddr),
				NIPQUAD(ih->daddr));
		else
			sprintf(buf, "%s %u.%u.%u.%u:%u->%u.%u.%u.%u:%u",
				pp->name,
				NIPQUAD(ih->saddr),
				ntohs(pptr[0]),
				NIPQUAD(ih->daddr),
				ntohs(pptr[1]));
	}

	printk(KERN_DEBUG "IPVS: %s: %s\n", msg, buf);
}

#ifdef CONFIG_IP_VS_IPV6
void
ip_vs_tcpudp_debug_packet_v6(struct ip_vs_protocol *pp,
			     const struct sk_buff *skb,
			     int offset,
			     const char *msg)
{
	char buf[192];
	struct ipv6hdr _iph, *ih;

	ih = skb_header_pointer(skb, offset, sizeof(_iph), &_iph);
	if (ih == NULL)
		sprintf(buf, "%s TRUNCATED", pp->name);
	else if (ih->nexthdr == IPPROTO_FRAGMENT)
		sprintf(buf, "%s " NIP6_FMT "->" NIP6_FMT " frag",
			pp->name, NIP6(ih->saddr),
			NIP6(ih->daddr));
	else {
		__be16 _ports[2], *pptr;

		pptr = skb_header_pointer(skb, offset + sizeof(struct ipv6hdr),
					  sizeof(_ports), _ports);
		if (pptr == NULL)
			sprintf(buf, "%s TRUNCATED " NIP6_FMT "->" NIP6_FMT,
				pp->name,
				NIP6(ih->saddr),
				NIP6(ih->daddr));
		else
			sprintf(buf, "%s " NIP6_FMT ":%u->" NIP6_FMT ":%u",
				pp->name,
				NIP6(ih->saddr),
				ntohs(pptr[0]),
				NIP6(ih->daddr),
				ntohs(pptr[1]));
	}

	printk(KERN_DEBUG "IPVS: %s: %s\n", msg, buf);
}
#endif


void
ip_vs_tcpudp_debug_packet(struct ip_vs_protocol *pp,
			  const struct sk_buff *skb,
			  int offset,
			  const char *msg)
{
#ifdef CONFIG_IP_VS_IPV6
	if (skb->protocol == __constant_htons(ETH_P_IPV6))
		ip_vs_tcpudp_debug_packet_v6(pp, skb, offset, msg);
	else
#endif
		ip_vs_tcpudp_debug_packet_v4(pp, skb, offset, msg);
}


int __init ip_vs_protocol_init(void)
{
	char protocols[64];
#define REGISTER_PROTOCOL(p)			\
	do {					\
		register_ip_vs_protocol(p);	\
		strcat(protocols, ", ");	\
		strcat(protocols, (p)->name);	\
	} while (0)

	protocols[0] = '\0';
	protocols[2] = '\0';
#ifdef CONFIG_IP_VS_PROTO_TCP
	REGISTER_PROTOCOL(&ip_vs_protocol_tcp);
#endif
#ifdef CONFIG_IP_VS_PROTO_UDP
	REGISTER_PROTOCOL(&ip_vs_protocol_udp);
#endif
#ifdef CONFIG_IP_VS_PROTO_AH
	REGISTER_PROTOCOL(&ip_vs_protocol_ah);
#endif
#ifdef CONFIG_IP_VS_PROTO_ESP
	REGISTER_PROTOCOL(&ip_vs_protocol_esp);
#endif
	IP_VS_INFO("Registered protocols (%s)\n", &protocols[2]);

	return 0;
}


void ip_vs_protocol_cleanup(void)
{
	struct ip_vs_protocol *pp;
	int i;

	/* unregister all the ipvs protocols */
	for (i = 0; i < IP_VS_PROTO_TAB_SIZE; i++) {
		while ((pp = ip_vs_proto_table[i]) != NULL)
			unregister_ip_vs_protocol(pp);
	}
}
Commit	Line	Data
1da177e4 LT	1	/*
	2	* ip_vs_proto.c: transport protocol load balancing support for IPVS
	3	*
1da177e4 LT	4	* Authors: Wensong Zhang <wensong@linuxvirtualserver.org>
	5	* Julian Anastasov <ja@ssi.bg>
	6	*
	7	* This program is free software; you can redistribute it and/or
	8	* modify it under the terms of the GNU General Public License
	9	* as published by the Free Software Foundation; either version
	10	* 2 of the License, or (at your option) any later version.
	11	*
	12	* Changes:
	13	*
	14	*/
	15
	16	#include <linux/module.h>
	17	#include <linux/kernel.h>
	18	#include <linux/skbuff.h>
	19	#include <linux/in.h>
	20	#include <linux/ip.h>
	21	#include <net/protocol.h>
	22	#include <net/tcp.h>
	23	#include <net/udp.h>
	24	#include <asm/system.h>
	25	#include <linux/stat.h>
	26	#include <linux/proc_fs.h>
	27
	28	#include <net/ip_vs.h>
	29
	30
	31	/*
	32	* IPVS protocols can only be registered/unregistered when the ipvs
	33	* module is loaded/unloaded, so no lock is needed in accessing the
	34	* ipvs protocol table.
	35	*/
	36
	37	#define IP_VS_PROTO_TAB_SIZE 32 /* must be power of 2 */
	38	#define IP_VS_PROTO_HASH(proto) ((proto) & (IP_VS_PROTO_TAB_SIZE-1))
	39
	40	static struct ip_vs_protocol *ip_vs_proto_table[IP_VS_PROTO_TAB_SIZE];
	41
	42
	43	/*
	44	* register an ipvs protocol
	45	*/
048cf48b	46	static int __used __init register_ip_vs_protocol(struct ip_vs_protocol *pp)
1da177e4 LT	47	{
	48	unsigned hash = IP_VS_PROTO_HASH(pp->protocol);
	49
	50	pp->next = ip_vs_proto_table[hash];
	51	ip_vs_proto_table[hash] = pp;
	52
	53	if (pp->init != NULL)
	54	pp->init(pp);
	55
	56	return 0;
	57	}
	58
	59
	60	/*
	61	* unregister an ipvs protocol
	62	*/
	63	static int unregister_ip_vs_protocol(struct ip_vs_protocol *pp)
	64	{
	65	struct ip_vs_protocol **pp_p;
	66	unsigned hash = IP_VS_PROTO_HASH(pp->protocol);
	67
	68	pp_p = &ip_vs_proto_table[hash];
	69	for (; pp_p; pp_p = &(pp_p)->next) {
	70	if (*pp_p == pp) {
	71	*pp_p = pp->next;
	72	if (pp->exit != NULL)
	73	pp->exit(pp);
	74	return 0;
	75	}
	76	}
	77
	78	return -ESRCH;
	79	}
	80
	81
	82	/*
	83	* get ip_vs_protocol object by its proto.
	84	*/
	85	struct ip_vs_protocol * ip_vs_proto_get(unsigned short proto)
	86	{
	87	struct ip_vs_protocol *pp;
	88	unsigned hash = IP_VS_PROTO_HASH(proto);
	89
	90	for (pp = ip_vs_proto_table[hash]; pp; pp = pp->next) {
	91	if (pp->protocol == proto)
	92	return pp;
	93	}
	94
	95	return NULL;
	96	}
	97
	98
	99	/*
	100	* Propagate event for state change to all protocols
	101	*/
	102	void ip_vs_protocol_timeout_change(int flags)
	103	{
	104	struct ip_vs_protocol *pp;
	105	int i;
	106
	107	for (i = 0; i < IP_VS_PROTO_TAB_SIZE; i++) {
	108	for (pp = ip_vs_proto_table[i]; pp; pp = pp->next) {
	109	if (pp->timeout_change)
	110	pp->timeout_change(pp, flags);
111	}
112	}
113	}
114
115
116	int *
117	ip_vs_create_timeout_table(int *table, int size)
118	{
8b2ed4bb	119	return kmemdup(table, size, GFP_ATOMIC);
1da177e4 LT	120	}
	121
	122
	123	/*
	124	* Set timeout value for state specified by name
	125	*/
	126	int
	127	ip_vs_set_state_timeout(int table, int num, char names, char name, int to)
	128	{
	129	int i;
	130
	131	if (!table \|\| !name \|\| !to)
	132	return -EINVAL;
	133
	134	for (i = 0; i < num; i++) {
	135	if (strcmp(names[i], name))
	136	continue;
	137	table[i] = to * HZ;
	138	return 0;
	139	}
	140	return -ENOENT;
	141	}
	142
	143
	144	const char * ip_vs_state_name(__u16 proto, int state)
	145	{
	146	struct ip_vs_protocol *pp = ip_vs_proto_get(proto);
	147
	148	if (pp == NULL \|\| pp->state_name == NULL)
2ad17def	149	return (IPPROTO_IP == proto) ? "NONE" : "ERR!";
1da177e4 LT	150	return pp->state_name(state);
	151	}
	152
	153
	154	void
3b047d9d JV	155	ip_vs_tcpudp_debug_packet_v4(struct ip_vs_protocol *pp,
	156	const struct sk_buff *skb,
	157	int offset,
	158	const char *msg)
1da177e4 LT	159	{
	160	char buf[128];
	161	struct iphdr _iph, *ih;
	162
	163	ih = skb_header_pointer(skb, offset, sizeof(_iph), &_iph);
	164	if (ih == NULL)
	165	sprintf(buf, "%s TRUNCATED", pp->name);
5661df7b	166	else if (ih->frag_off & htons(IP_OFFSET))
1da177e4 LT	167	sprintf(buf, "%s %u.%u.%u.%u->%u.%u.%u.%u frag",
	168	pp->name, NIPQUAD(ih->saddr),
	169	NIPQUAD(ih->daddr));
	170	else {
014d730d	171	__be16 _ports[2], *pptr
1da177e4 LT	172	;
	173	pptr = skb_header_pointer(skb, offset + ih->ihl*4,
	174	sizeof(_ports), _ports);
	175	if (pptr == NULL)
	176	sprintf(buf, "%s TRUNCATED %u.%u.%u.%u->%u.%u.%u.%u",
	177	pp->name,
	178	NIPQUAD(ih->saddr),
	179	NIPQUAD(ih->daddr));
	180	else
	181	sprintf(buf, "%s %u.%u.%u.%u:%u->%u.%u.%u.%u:%u",
	182	pp->name,
	183	NIPQUAD(ih->saddr),
	184	ntohs(pptr[0]),
	185	NIPQUAD(ih->daddr),
	186	ntohs(pptr[1]));
	187	}
	188
	189	printk(KERN_DEBUG "IPVS: %s: %s\n", msg, buf);
	190	}
	191
3b047d9d JV	192	#ifdef CONFIG_IP_VS_IPV6
	193	void
	194	ip_vs_tcpudp_debug_packet_v6(struct ip_vs_protocol *pp,
	195	const struct sk_buff *skb,
	196	int offset,
	197	const char *msg)
	198	{
	199	char buf[192];
	200	struct ipv6hdr _iph, *ih;
	201
	202	ih = skb_header_pointer(skb, offset, sizeof(_iph), &_iph);
	203	if (ih == NULL)
	204	sprintf(buf, "%s TRUNCATED", pp->name);
	205	else if (ih->nexthdr == IPPROTO_FRAGMENT)
	206	sprintf(buf, "%s " NIP6_FMT "->" NIP6_FMT " frag",
	207	pp->name, NIP6(ih->saddr),
	208	NIP6(ih->daddr));
	209	else {
	210	__be16 _ports[2], *pptr;
	211
	212	pptr = skb_header_pointer(skb, offset + sizeof(struct ipv6hdr),
	213	sizeof(_ports), _ports);
	214	if (pptr == NULL)
	215	sprintf(buf, "%s TRUNCATED " NIP6_FMT "->" NIP6_FMT,
	216	pp->name,
	217	NIP6(ih->saddr),
	218	NIP6(ih->daddr));
	219	else
	220	sprintf(buf, "%s " NIP6_FMT ":%u->" NIP6_FMT ":%u",
	221	pp->name,
	222	NIP6(ih->saddr),
	223	ntohs(pptr[0]),
	224	NIP6(ih->daddr),
	225	ntohs(pptr[1]));
	226	}
	227
	228	printk(KERN_DEBUG "IPVS: %s: %s\n", msg, buf);
	229	}
	230	#endif
	231
	232
	233	void
	234	ip_vs_tcpudp_debug_packet(struct ip_vs_protocol *pp,
	235	const struct sk_buff *skb,
	236	int offset,
	237	const char *msg)
	238	{
	239	#ifdef CONFIG_IP_VS_IPV6
	240	if (skb->protocol == __constant_htons(ETH_P_IPV6))
	241	ip_vs_tcpudp_debug_packet_v6(pp, skb, offset, msg);
	242	else
	243	#endif
	244	ip_vs_tcpudp_debug_packet_v4(pp, skb, offset, msg);
	245	}
	246
1da177e4	247
048cf48b	248	int __init ip_vs_protocol_init(void)
1da177e4 LT	249	{
	250	char protocols[64];
	251	#define REGISTER_PROTOCOL(p) \
	252	do { \
	253	register_ip_vs_protocol(p); \
	254	strcat(protocols, ", "); \
	255	strcat(protocols, (p)->name); \
	256	} while (0)
	257
	258	protocols[0] = '\0';
	259	protocols[2] = '\0';
	260	#ifdef CONFIG_IP_VS_PROTO_TCP
	261	REGISTER_PROTOCOL(&ip_vs_protocol_tcp);
	262	#endif
	263	#ifdef CONFIG_IP_VS_PROTO_UDP
	264	REGISTER_PROTOCOL(&ip_vs_protocol_udp);
	265	#endif
1da177e4 LT	266	#ifdef CONFIG_IP_VS_PROTO_AH
	267	REGISTER_PROTOCOL(&ip_vs_protocol_ah);
	268	#endif
	269	#ifdef CONFIG_IP_VS_PROTO_ESP
	270	REGISTER_PROTOCOL(&ip_vs_protocol_esp);
	271	#endif
	272	IP_VS_INFO("Registered protocols (%s)\n", &protocols[2]);
	273
	274	return 0;
	275	}
	276
	277
	278	void ip_vs_protocol_cleanup(void)
	279	{
	280	struct ip_vs_protocol *pp;
	281	int i;
	282
	283	/* unregister all the ipvs protocols */
	284	for (i = 0; i < IP_VS_PROTO_TAB_SIZE; i++) {
	285	while ((pp = ip_vs_proto_table[i]) != NULL)
	286	unregister_ip_vs_protocol(pp);
	287	}
	288	}