[deliverable/linux.git] / net / ipv4 / ipcomp.c

/*
 * IP Payload Compression Protocol (IPComp) - RFC3173.
 *
 * Copyright (c) 2003 James Morris <jmorris@intercode.com.au>
 *
 * 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.
 *
 * Todo:
 *   - Tunable compression parameters.
 *   - Compression stats.
 *   - Adaptive compression.
 */
#include <linux/config.h>
#include <linux/module.h>
#include <asm/scatterlist.h>
#include <asm/semaphore.h>
#include <linux/crypto.h>
#include <linux/pfkeyv2.h>
#include <linux/percpu.h>
#include <linux/smp.h>
#include <linux/list.h>
#include <linux/vmalloc.h>
#include <linux/rtnetlink.h>
#include <linux/mutex.h>
#include <net/ip.h>
#include <net/xfrm.h>
#include <net/icmp.h>
#include <net/ipcomp.h>
#include <net/protocol.h>

struct ipcomp_tfms {
	struct list_head list;
	struct crypto_tfm **tfms;
	int users;
};

static DEFINE_MUTEX(ipcomp_resource_mutex);
static void **ipcomp_scratches;
static int ipcomp_scratch_users;
static LIST_HEAD(ipcomp_tfms_list);

static int ipcomp_decompress(struct xfrm_state *x, struct sk_buff *skb)
{
	int err, plen, dlen;
	struct ipcomp_data *ipcd = x->data;
	u8 *start, *scratch;
	struct crypto_tfm *tfm;
	int cpu;
	
	plen = skb->len;
	dlen = IPCOMP_SCRATCH_SIZE;
	start = skb->data;

	cpu = get_cpu();
	scratch = *per_cpu_ptr(ipcomp_scratches, cpu);
	tfm = *per_cpu_ptr(ipcd->tfms, cpu);

	err = crypto_comp_decompress(tfm, start, plen, scratch, &dlen);
	if (err)
		goto out;

	if (dlen < (plen + sizeof(struct ip_comp_hdr))) {
		err = -EINVAL;
		goto out;
	}

	err = pskb_expand_head(skb, 0, dlen - plen, GFP_ATOMIC);
	if (err)
		goto out;
		
	skb_put(skb, dlen - plen);
	memcpy(skb->data, scratch, dlen);
out:	
	put_cpu();
	return err;
}

static int ipcomp_input(struct xfrm_state *x, struct sk_buff *skb)
{
	int err = 0;
	struct iphdr *iph;
	struct ip_comp_hdr *ipch;

	if ((skb_is_nonlinear(skb) || skb_cloned(skb)) &&
	    skb_linearize(skb, GFP_ATOMIC) != 0) {
	    	err = -ENOMEM;
	    	goto out;
	}

	skb->ip_summed = CHECKSUM_NONE;

	/* Remove ipcomp header and decompress original payload */	
	iph = skb->nh.iph;
	ipch = (void *)skb->data;
	iph->protocol = ipch->nexthdr;
	skb->h.raw = skb->nh.raw + sizeof(*ipch);
	__skb_pull(skb, sizeof(*ipch));
	err = ipcomp_decompress(x, skb);

out:	
	return err;
}

static int ipcomp_compress(struct xfrm_state *x, struct sk_buff *skb)
{
	int err, plen, dlen, ihlen;
	struct iphdr *iph = skb->nh.iph;
	struct ipcomp_data *ipcd = x->data;
	u8 *start, *scratch;
	struct crypto_tfm *tfm;
	int cpu;
	
	ihlen = iph->ihl * 4;
	plen = skb->len - ihlen;
	dlen = IPCOMP_SCRATCH_SIZE;
	start = skb->data + ihlen;

	cpu = get_cpu();
	scratch = *per_cpu_ptr(ipcomp_scratches, cpu);
	tfm = *per_cpu_ptr(ipcd->tfms, cpu);

	err = crypto_comp_compress(tfm, start, plen, scratch, &dlen);
	if (err)
		goto out;

	if ((dlen + sizeof(struct ip_comp_hdr)) >= plen) {
		err = -EMSGSIZE;
		goto out;
	}
	
	memcpy(start + sizeof(struct ip_comp_hdr), scratch, dlen);
	put_cpu();

	pskb_trim(skb, ihlen + dlen + sizeof(struct ip_comp_hdr));
	return 0;
	
out:	
	put_cpu();
	return err;
}

static int ipcomp_output(struct xfrm_state *x, struct sk_buff *skb)
{
	int err;
	struct iphdr *iph;
	struct ip_comp_hdr *ipch;
	struct ipcomp_data *ipcd = x->data;
	int hdr_len = 0;

	iph = skb->nh.iph;
	iph->tot_len = htons(skb->len);
	hdr_len = iph->ihl * 4;
	if ((skb->len - hdr_len) < ipcd->threshold) {
		/* Don't bother compressing */
		goto out_ok;
	}

	if ((skb_is_nonlinear(skb) || skb_cloned(skb)) &&
	    skb_linearize(skb, GFP_ATOMIC) != 0) {
		goto out_ok;
	}
	
	err = ipcomp_compress(x, skb);
	iph = skb->nh.iph;

	if (err) {
		goto out_ok;
	}

	/* Install ipcomp header, convert into ipcomp datagram. */
	iph->tot_len = htons(skb->len);
	ipch = (struct ip_comp_hdr *)((char *)iph + iph->ihl * 4);
	ipch->nexthdr = iph->protocol;
	ipch->flags = 0;
	ipch->cpi = htons((u16 )ntohl(x->id.spi));
	iph->protocol = IPPROTO_COMP;
	ip_send_check(iph);
	return 0;

out_ok:
	if (x->props.mode)
		ip_send_check(iph);
	return 0;
}

static void ipcomp4_err(struct sk_buff *skb, u32 info)
{
	u32 spi;
	struct iphdr *iph = (struct iphdr *)skb->data;
	struct ip_comp_hdr *ipch = (struct ip_comp_hdr *)(skb->data+(iph->ihl<<2));
	struct xfrm_state *x;

	if (skb->h.icmph->type != ICMP_DEST_UNREACH ||
	    skb->h.icmph->code != ICMP_FRAG_NEEDED)
		return;

	spi = htonl(ntohs(ipch->cpi));
	x = xfrm_state_lookup((xfrm_address_t *)&iph->daddr,
	                      spi, IPPROTO_COMP, AF_INET);
	if (!x)
		return;
	NETDEBUG(KERN_DEBUG "pmtu discovery on SA IPCOMP/%08x/%u.%u.%u.%u\n",
		 spi, NIPQUAD(iph->daddr));
	xfrm_state_put(x);
}

/* We always hold one tunnel user reference to indicate a tunnel */ 
static struct xfrm_state *ipcomp_tunnel_create(struct xfrm_state *x)
{
	struct xfrm_state *t;
	
	t = xfrm_state_alloc();
	if (t == NULL)
		goto out;

	t->id.proto = IPPROTO_IPIP;
	t->id.spi = x->props.saddr.a4;
	t->id.daddr.a4 = x->id.daddr.a4;
	memcpy(&t->sel, &x->sel, sizeof(t->sel));
	t->props.family = AF_INET;
	t->props.mode = 1;
	t->props.saddr.a4 = x->props.saddr.a4;
	t->props.flags = x->props.flags;

	if (xfrm_init_state(t))
		goto error;

	atomic_set(&t->tunnel_users, 1);
out:
	return t;

error:
	t->km.state = XFRM_STATE_DEAD;
	xfrm_state_put(t);
	t = NULL;
	goto out;
}

/*
 * Must be protected by xfrm_cfg_mutex.  State and tunnel user references are
 * always incremented on success.
 */
static int ipcomp_tunnel_attach(struct xfrm_state *x)
{
	int err = 0;
	struct xfrm_state *t;

	t = xfrm_state_lookup((xfrm_address_t *)&x->id.daddr.a4,
	                      x->props.saddr.a4, IPPROTO_IPIP, AF_INET);
	if (!t) {
		t = ipcomp_tunnel_create(x);
		if (!t) {
			err = -EINVAL;
			goto out;
		}
		xfrm_state_insert(t);
		xfrm_state_hold(t);
	}
	x->tunnel = t;
	atomic_inc(&t->tunnel_users);
out:
	return err;
}

static void ipcomp_free_scratches(void)
{
	int i;
	void **scratches;

	if (--ipcomp_scratch_users)
		return;

	scratches = ipcomp_scratches;
	if (!scratches)
		return;

	for_each_possible_cpu(i)
		vfree(*per_cpu_ptr(scratches, i));

	free_percpu(scratches);
}

static void **ipcomp_alloc_scratches(void)
{
	int i;
	void **scratches;

	if (ipcomp_scratch_users++)
		return ipcomp_scratches;

	scratches = alloc_percpu(void *);
	if (!scratches)
		return NULL;

	ipcomp_scratches = scratches;

	for_each_possible_cpu(i) {
		void *scratch = vmalloc(IPCOMP_SCRATCH_SIZE);
		if (!scratch)
			return NULL;
		*per_cpu_ptr(scratches, i) = scratch;
	}

	return scratches;
}

static void ipcomp_free_tfms(struct crypto_tfm **tfms)
{
	struct ipcomp_tfms *pos;
	int cpu;

	list_for_each_entry(pos, &ipcomp_tfms_list, list) {
		if (pos->tfms == tfms)
			break;
	}

	BUG_TRAP(pos);

	if (--pos->users)
		return;

	list_del(&pos->list);
	kfree(pos);

	if (!tfms)
		return;

	for_each_possible_cpu(cpu) {
		struct crypto_tfm *tfm = *per_cpu_ptr(tfms, cpu);
		crypto_free_tfm(tfm);
	}
	free_percpu(tfms);
}

static struct crypto_tfm **ipcomp_alloc_tfms(const char *alg_name)
{
	struct ipcomp_tfms *pos;
	struct crypto_tfm **tfms;
	int cpu;

	/* This can be any valid CPU ID so we don't need locking. */
	cpu = raw_smp_processor_id();

	list_for_each_entry(pos, &ipcomp_tfms_list, list) {
		struct crypto_tfm *tfm;

		tfms = pos->tfms;
		tfm = *per_cpu_ptr(tfms, cpu);

		if (!strcmp(crypto_tfm_alg_name(tfm), alg_name)) {
			pos->users++;
			return tfms;
		}
	}

	pos = kmalloc(sizeof(*pos), GFP_KERNEL);
	if (!pos)
		return NULL;

	pos->users = 1;
	INIT_LIST_HEAD(&pos->list);
	list_add(&pos->list, &ipcomp_tfms_list);

	pos->tfms = tfms = alloc_percpu(struct crypto_tfm *);
	if (!tfms)
		goto error;

	for_each_possible_cpu(cpu) {
		struct crypto_tfm *tfm = crypto_alloc_tfm(alg_name, 0);
		if (!tfm)
			goto error;
		*per_cpu_ptr(tfms, cpu) = tfm;
	}

	return tfms;

error:
	ipcomp_free_tfms(tfms);
	return NULL;
}

static void ipcomp_free_data(struct ipcomp_data *ipcd)
{
	if (ipcd->tfms)
		ipcomp_free_tfms(ipcd->tfms);
	ipcomp_free_scratches();
}

static void ipcomp_destroy(struct xfrm_state *x)
{
	struct ipcomp_data *ipcd = x->data;
	if (!ipcd)
		return;
	xfrm_state_delete_tunnel(x);
	mutex_lock(&ipcomp_resource_mutex);
	ipcomp_free_data(ipcd);
	mutex_unlock(&ipcomp_resource_mutex);
	kfree(ipcd);
}

static int ipcomp_init_state(struct xfrm_state *x)
{
	int err;
	struct ipcomp_data *ipcd;
	struct xfrm_algo_desc *calg_desc;

	err = -EINVAL;
	if (!x->calg)
		goto out;

	if (x->encap)
		goto out;

	err = -ENOMEM;
	ipcd = kmalloc(sizeof(*ipcd), GFP_KERNEL);
	if (!ipcd)
		goto out;

	memset(ipcd, 0, sizeof(*ipcd));
	x->props.header_len = 0;
	if (x->props.mode)
		x->props.header_len += sizeof(struct iphdr);

	mutex_lock(&ipcomp_resource_mutex);
	if (!ipcomp_alloc_scratches())
		goto error;

	ipcd->tfms = ipcomp_alloc_tfms(x->calg->alg_name);
	if (!ipcd->tfms)
		goto error;
	mutex_unlock(&ipcomp_resource_mutex);

	if (x->props.mode) {
		err = ipcomp_tunnel_attach(x);
		if (err)
			goto error_tunnel;
	}

	calg_desc = xfrm_calg_get_byname(x->calg->alg_name, 0);
	BUG_ON(!calg_desc);
	ipcd->threshold = calg_desc->uinfo.comp.threshold;
	x->data = ipcd;
	err = 0;
out:
	return err;

error_tunnel:
	mutex_lock(&ipcomp_resource_mutex);
error:
	ipcomp_free_data(ipcd);
	mutex_unlock(&ipcomp_resource_mutex);
	kfree(ipcd);
	goto out;
}

static struct xfrm_type ipcomp_type = {
	.description	= "IPCOMP4",
	.owner		= THIS_MODULE,
	.proto	     	= IPPROTO_COMP,
	.init_state	= ipcomp_init_state,
	.destructor	= ipcomp_destroy,
	.input		= ipcomp_input,
	.output		= ipcomp_output
};

static struct net_protocol ipcomp4_protocol = {
	.handler	=	xfrm4_rcv,
	.err_handler	=	ipcomp4_err,
	.no_policy	=	1,
};

static int __init ipcomp4_init(void)
{
	if (xfrm_register_type(&ipcomp_type, AF_INET) < 0) {
		printk(KERN_INFO "ipcomp init: can't add xfrm type\n");
		return -EAGAIN;
	}
	if (inet_add_protocol(&ipcomp4_protocol, IPPROTO_COMP) < 0) {
		printk(KERN_INFO "ipcomp init: can't add protocol\n");
		xfrm_unregister_type(&ipcomp_type, AF_INET);
		return -EAGAIN;
	}
	return 0;
}

static void __exit ipcomp4_fini(void)
{
	if (inet_del_protocol(&ipcomp4_protocol, IPPROTO_COMP) < 0)
		printk(KERN_INFO "ip ipcomp close: can't remove protocol\n");
	if (xfrm_unregister_type(&ipcomp_type, AF_INET) < 0)
		printk(KERN_INFO "ip ipcomp close: can't remove xfrm type\n");
}

module_init(ipcomp4_init);
module_exit(ipcomp4_fini);

MODULE_LICENSE("GPL");
MODULE_DESCRIPTION("IP Payload Compression Protocol (IPComp) - RFC3173");
MODULE_AUTHOR("James Morris <jmorris@intercode.com.au>");
Commit	Line	Data
1da177e4 LT	1	/*
	2	* IP Payload Compression Protocol (IPComp) - RFC3173.
	3	*
	4	* Copyright (c) 2003 James Morris <jmorris@intercode.com.au>
	5	*
	6	* This program is free software; you can redistribute it and/or modify it
	7	* under the terms of the GNU General Public License as published by the Free
	8	* Software Foundation; either version 2 of the License, or (at your option)
	9	* any later version.
	10	*
	11	* Todo:
	12	* - Tunable compression parameters.
	13	* - Compression stats.
	14	* - Adaptive compression.
	15	*/
	16	#include <linux/config.h>
	17	#include <linux/module.h>
	18	#include <asm/scatterlist.h>
	19	#include <asm/semaphore.h>
	20	#include <linux/crypto.h>
	21	#include <linux/pfkeyv2.h>
	22	#include <linux/percpu.h>
	23	#include <linux/smp.h>
	24	#include <linux/list.h>
	25	#include <linux/vmalloc.h>
	26	#include <linux/rtnetlink.h>
4a3e2f71	27	#include <linux/mutex.h>
1da177e4 LT	28	#include <net/ip.h>
	29	#include <net/xfrm.h>
	30	#include <net/icmp.h>
	31	#include <net/ipcomp.h>
14c85021	32	#include <net/protocol.h>
1da177e4 LT	33
	34	struct ipcomp_tfms {
	35	struct list_head list;
	36	struct crypto_tfm **tfms;
	37	int users;
	38	};
	39
4a3e2f71	40	static DEFINE_MUTEX(ipcomp_resource_mutex);
1da177e4 LT	41	static void **ipcomp_scratches;
	42	static int ipcomp_scratch_users;
	43	static LIST_HEAD(ipcomp_tfms_list);
	44
	45	static int ipcomp_decompress(struct xfrm_state x, struct sk_buff skb)
	46	{
	47	int err, plen, dlen;
1da177e4 LT	48	struct ipcomp_data *ipcd = x->data;
	49	u8 start, scratch;
	50	struct crypto_tfm *tfm;
	51	int cpu;
	52
	53	plen = skb->len;
	54	dlen = IPCOMP_SCRATCH_SIZE;
	55	start = skb->data;
	56
	57	cpu = get_cpu();
	58	scratch = *per_cpu_ptr(ipcomp_scratches, cpu);
	59	tfm = *per_cpu_ptr(ipcd->tfms, cpu);
	60
	61	err = crypto_comp_decompress(tfm, start, plen, scratch, &dlen);
	62	if (err)
	63	goto out;
	64
	65	if (dlen < (plen + sizeof(struct ip_comp_hdr))) {
	66	err = -EINVAL;
	67	goto out;
	68	}
	69
	70	err = pskb_expand_head(skb, 0, dlen - plen, GFP_ATOMIC);
	71	if (err)
	72	goto out;
	73
	74	skb_put(skb, dlen - plen);
	75	memcpy(skb->data, scratch, dlen);
1da177e4 LT	76	out:
	77	put_cpu();
	78	return err;
	79	}
	80
e695633e	81	static int ipcomp_input(struct xfrm_state x, struct sk_buff skb)
1da177e4	82	{
1da177e4 LT	83	int err = 0;
1da177e4 LT	84	struct iphdr *iph;
31a4ab93	85	struct ip_comp_hdr *ipch;
1da177e4 LT	86
	87	if ((skb_is_nonlinear(skb) \|\| skb_cloned(skb)) &&
	88	skb_linearize(skb, GFP_ATOMIC) != 0) {
	89	err = -ENOMEM;
	90	goto out;
	91	}
	92
	93	skb->ip_summed = CHECKSUM_NONE;
	94
	95	/* Remove ipcomp header and decompress original payload */
	96	iph = skb->nh.iph;
31a4ab93 HX	97	ipch = (void *)skb->data;
	98	iph->protocol = ipch->nexthdr;
	99	skb->h.raw = skb->nh.raw + sizeof(*ipch);
	100	__skb_pull(skb, sizeof(*ipch));
1da177e4 LT	101	err = ipcomp_decompress(x, skb);
	102
	103	out:
	104	return err;
	105	}
	106
	107	static int ipcomp_compress(struct xfrm_state x, struct sk_buff skb)
	108	{
	109	int err, plen, dlen, ihlen;
	110	struct iphdr *iph = skb->nh.iph;
	111	struct ipcomp_data *ipcd = x->data;
	112	u8 start, scratch;
	113	struct crypto_tfm *tfm;
	114	int cpu;
	115
	116	ihlen = iph->ihl * 4;
	117	plen = skb->len - ihlen;
	118	dlen = IPCOMP_SCRATCH_SIZE;
	119	start = skb->data + ihlen;
	120
	121	cpu = get_cpu();
	122	scratch = *per_cpu_ptr(ipcomp_scratches, cpu);
	123	tfm = *per_cpu_ptr(ipcd->tfms, cpu);
	124
	125	err = crypto_comp_compress(tfm, start, plen, scratch, &dlen);
	126	if (err)
	127	goto out;
	128
	129	if ((dlen + sizeof(struct ip_comp_hdr)) >= plen) {
	130	err = -EMSGSIZE;
	131	goto out;
	132	}
	133
	134	memcpy(start + sizeof(struct ip_comp_hdr), scratch, dlen);
	135	put_cpu();
	136
	137	pskb_trim(skb, ihlen + dlen + sizeof(struct ip_comp_hdr));
	138	return 0;
	139
	140	out:
	141	put_cpu();
	142	return err;
	143	}
	144
	145	static int ipcomp_output(struct xfrm_state x, struct sk_buff skb)
	146	{
	147	int err;
	148	struct iphdr *iph;
	149	struct ip_comp_hdr *ipch;
	150	struct ipcomp_data *ipcd = x->data;
	151	int hdr_len = 0;
	152
	153	iph = skb->nh.iph;
	154	iph->tot_len = htons(skb->len);
	155	hdr_len = iph->ihl * 4;
	156	if ((skb->len - hdr_len) < ipcd->threshold) {
	157	/* Don't bother compressing */
	158	goto out_ok;
	159	}
	160
	161	if ((skb_is_nonlinear(skb) \|\| skb_cloned(skb)) &&
	162	skb_linearize(skb, GFP_ATOMIC) != 0) {
	163	goto out_ok;
	164	}
165
166	err = ipcomp_compress(x, skb);
167	iph = skb->nh.iph;
168
169	if (err) {
170	goto out_ok;
171	}
172
173	/* Install ipcomp header, convert into ipcomp datagram. */
174	iph->tot_len = htons(skb->len);
175	ipch = (struct ip_comp_hdr )((char )iph + iph->ihl * 4);
176	ipch->nexthdr = iph->protocol;
177	ipch->flags = 0;
178	ipch->cpi = htons((u16 )ntohl(x->id.spi));
179	iph->protocol = IPPROTO_COMP;
180	ip_send_check(iph);
181	return 0;
182
183	out_ok:
184	if (x->props.mode)
185	ip_send_check(iph);
186	return 0;
187	}
188
189	static void ipcomp4_err(struct sk_buff *skb, u32 info)
190	{
191	u32 spi;
192	struct iphdr iph = (struct iphdr )skb->data;
193	struct ip_comp_hdr ipch = (struct ip_comp_hdr )(skb->data+(iph->ihl<<2));
194	struct xfrm_state *x;
195
196	if (skb->h.icmph->type != ICMP_DEST_UNREACH \|\|
197	skb->h.icmph->code != ICMP_FRAG_NEEDED)
198	return;
199
4195f814	200	spi = htonl(ntohs(ipch->cpi));
1da177e4 LT	201	x = xfrm_state_lookup((xfrm_address_t *)&iph->daddr,
	202	spi, IPPROTO_COMP, AF_INET);
	203	if (!x)
	204	return;
64ce2073 PM	205	NETDEBUG(KERN_DEBUG "pmtu discovery on SA IPCOMP/%08x/%u.%u.%u.%u\n",
64ce2073 PM	206	spi, NIPQUAD(iph->daddr));
1da177e4 LT	207	xfrm_state_put(x);
	208	}
	209
	210	/* We always hold one tunnel user reference to indicate a tunnel */
	211	static struct xfrm_state ipcomp_tunnel_create(struct xfrm_state x)
	212	{
	213	struct xfrm_state *t;
	214
	215	t = xfrm_state_alloc();
	216	if (t == NULL)
	217	goto out;
	218
	219	t->id.proto = IPPROTO_IPIP;
	220	t->id.spi = x->props.saddr.a4;
	221	t->id.daddr.a4 = x->id.daddr.a4;
	222	memcpy(&t->sel, &x->sel, sizeof(t->sel));
	223	t->props.family = AF_INET;
	224	t->props.mode = 1;
	225	t->props.saddr.a4 = x->props.saddr.a4;
	226	t->props.flags = x->props.flags;
72cb6962 HX	227
72cb6962 HX	228	if (xfrm_init_state(t))
1da177e4 LT	229	goto error;
1da177e4 LT	230
1da177e4 LT	231	atomic_set(&t->tunnel_users, 1);
	232	out:
	233	return t;
	234
	235	error:
	236	t->km.state = XFRM_STATE_DEAD;
	237	xfrm_state_put(t);
	238	t = NULL;
	239	goto out;
	240	}
	241
	242	/*
4a3e2f71	243	* Must be protected by xfrm_cfg_mutex. State and tunnel user references are
1da177e4 LT	244	* always incremented on success.
	245	*/
	246	static int ipcomp_tunnel_attach(struct xfrm_state *x)
	247	{
	248	int err = 0;
	249	struct xfrm_state *t;
	250
	251	t = xfrm_state_lookup((xfrm_address_t *)&x->id.daddr.a4,
	252	x->props.saddr.a4, IPPROTO_IPIP, AF_INET);
	253	if (!t) {
	254	t = ipcomp_tunnel_create(x);
	255	if (!t) {
	256	err = -EINVAL;
	257	goto out;
	258	}
	259	xfrm_state_insert(t);
	260	xfrm_state_hold(t);
	261	}
	262	x->tunnel = t;
	263	atomic_inc(&t->tunnel_users);
	264	out:
	265	return err;
	266	}
	267
	268	static void ipcomp_free_scratches(void)
	269	{
	270	int i;
	271	void **scratches;
	272
	273	if (--ipcomp_scratch_users)
	274	return;
	275
	276	scratches = ipcomp_scratches;
	277	if (!scratches)
	278	return;
	279
63903ca6 JJ	280	for_each_possible_cpu(i)
63903ca6 JJ	281	vfree(*per_cpu_ptr(scratches, i));
1da177e4 LT	282
	283	free_percpu(scratches);
	284	}
	285
	286	static void **ipcomp_alloc_scratches(void)
	287	{
	288	int i;
	289	void **scratches;
	290
	291	if (ipcomp_scratch_users++)
	292	return ipcomp_scratches;
	293
	294	scratches = alloc_percpu(void *);
	295	if (!scratches)
	296	return NULL;
	297
	298	ipcomp_scratches = scratches;
	299
6f912042	300	for_each_possible_cpu(i) {
1da177e4 LT	301	void *scratch = vmalloc(IPCOMP_SCRATCH_SIZE);
	302	if (!scratch)
	303	return NULL;
	304	*per_cpu_ptr(scratches, i) = scratch;
	305	}
	306
	307	return scratches;
	308	}
	309
	310	static void ipcomp_free_tfms(struct crypto_tfm **tfms)
	311	{
	312	struct ipcomp_tfms *pos;
	313	int cpu;
	314
	315	list_for_each_entry(pos, &ipcomp_tfms_list, list) {
	316	if (pos->tfms == tfms)
	317	break;
	318	}
	319
	320	BUG_TRAP(pos);
	321
	322	if (--pos->users)
	323	return;
	324
	325	list_del(&pos->list);
	326	kfree(pos);
	327
	328	if (!tfms)
	329	return;
	330
6f912042	331	for_each_possible_cpu(cpu) {
1da177e4	332	struct crypto_tfm tfm = per_cpu_ptr(tfms, cpu);
573dbd95	333	crypto_free_tfm(tfm);
1da177e4 LT	334	}
	335	free_percpu(tfms);
	336	}
	337
	338	static struct crypto_tfm *ipcomp_alloc_tfms(const char alg_name)
	339	{
	340	struct ipcomp_tfms *pos;
	341	struct crypto_tfm **tfms;
	342	int cpu;
	343
	344	/* This can be any valid CPU ID so we don't need locking. */
6fc8b9e7	345	cpu = raw_smp_processor_id();
1da177e4 LT	346
	347	list_for_each_entry(pos, &ipcomp_tfms_list, list) {
	348	struct crypto_tfm *tfm;
	349
	350	tfms = pos->tfms;
	351	tfm = *per_cpu_ptr(tfms, cpu);
	352
	353	if (!strcmp(crypto_tfm_alg_name(tfm), alg_name)) {
	354	pos->users++;
	355	return tfms;
	356	}
	357	}
	358
	359	pos = kmalloc(sizeof(*pos), GFP_KERNEL);
	360	if (!pos)
	361	return NULL;
	362
	363	pos->users = 1;
	364	INIT_LIST_HEAD(&pos->list);
	365	list_add(&pos->list, &ipcomp_tfms_list);
	366
	367	pos->tfms = tfms = alloc_percpu(struct crypto_tfm *);
	368	if (!tfms)
	369	goto error;
	370
6f912042	371	for_each_possible_cpu(cpu) {
1da177e4 LT	372	struct crypto_tfm *tfm = crypto_alloc_tfm(alg_name, 0);
	373	if (!tfm)
	374	goto error;
	375	*per_cpu_ptr(tfms, cpu) = tfm;
	376	}
	377
	378	return tfms;
	379
	380	error:
	381	ipcomp_free_tfms(tfms);
	382	return NULL;
	383	}
	384
	385	static void ipcomp_free_data(struct ipcomp_data *ipcd)
	386	{
	387	if (ipcd->tfms)
	388	ipcomp_free_tfms(ipcd->tfms);
	389	ipcomp_free_scratches();
	390	}
	391
	392	static void ipcomp_destroy(struct xfrm_state *x)
	393	{
	394	struct ipcomp_data *ipcd = x->data;
	395	if (!ipcd)
	396	return;
	397	xfrm_state_delete_tunnel(x);
4a3e2f71	398	mutex_lock(&ipcomp_resource_mutex);
1da177e4	399	ipcomp_free_data(ipcd);
4a3e2f71	400	mutex_unlock(&ipcomp_resource_mutex);
1da177e4 LT	401	kfree(ipcd);
	402	}
	403
72cb6962	404	static int ipcomp_init_state(struct xfrm_state *x)
1da177e4 LT	405	{
	406	int err;
	407	struct ipcomp_data *ipcd;
	408	struct xfrm_algo_desc *calg_desc;
	409
	410	err = -EINVAL;
	411	if (!x->calg)
	412	goto out;
	413
	414	if (x->encap)
	415	goto out;
	416
	417	err = -ENOMEM;
	418	ipcd = kmalloc(sizeof(*ipcd), GFP_KERNEL);
	419	if (!ipcd)
	420	goto out;
	421
	422	memset(ipcd, 0, sizeof(*ipcd));
	423	x->props.header_len = 0;
	424	if (x->props.mode)
	425	x->props.header_len += sizeof(struct iphdr);
	426
4a3e2f71	427	mutex_lock(&ipcomp_resource_mutex);
1da177e4 LT	428	if (!ipcomp_alloc_scratches())
	429	goto error;
	430
	431	ipcd->tfms = ipcomp_alloc_tfms(x->calg->alg_name);
	432	if (!ipcd->tfms)
	433	goto error;
4a3e2f71	434	mutex_unlock(&ipcomp_resource_mutex);
1da177e4 LT	435
	436	if (x->props.mode) {
	437	err = ipcomp_tunnel_attach(x);
	438	if (err)
	439	goto error_tunnel;
	440	}
	441
	442	calg_desc = xfrm_calg_get_byname(x->calg->alg_name, 0);
	443	BUG_ON(!calg_desc);
	444	ipcd->threshold = calg_desc->uinfo.comp.threshold;
	445	x->data = ipcd;
	446	err = 0;
	447	out:
	448	return err;
	449
	450	error_tunnel:
4a3e2f71	451	mutex_lock(&ipcomp_resource_mutex);
1da177e4 LT	452	error:
1da177e4 LT	453	ipcomp_free_data(ipcd);
4a3e2f71	454	mutex_unlock(&ipcomp_resource_mutex);
1da177e4 LT	455	kfree(ipcd);
	456	goto out;
	457	}
	458
	459	static struct xfrm_type ipcomp_type = {
	460	.description = "IPCOMP4",
	461	.owner = THIS_MODULE,
	462	.proto = IPPROTO_COMP,
	463	.init_state = ipcomp_init_state,
	464	.destructor = ipcomp_destroy,
	465	.input = ipcomp_input,
	466	.output = ipcomp_output
	467	};
	468
	469	static struct net_protocol ipcomp4_protocol = {
	470	.handler = xfrm4_rcv,
	471	.err_handler = ipcomp4_err,
	472	.no_policy = 1,
	473	};
	474
	475	static int __init ipcomp4_init(void)
	476	{
	477	if (xfrm_register_type(&ipcomp_type, AF_INET) < 0) {
	478	printk(KERN_INFO "ipcomp init: can't add xfrm type\n");
	479	return -EAGAIN;
	480	}
	481	if (inet_add_protocol(&ipcomp4_protocol, IPPROTO_COMP) < 0) {
	482	printk(KERN_INFO "ipcomp init: can't add protocol\n");
	483	xfrm_unregister_type(&ipcomp_type, AF_INET);
	484	return -EAGAIN;
	485	}
	486	return 0;
	487	}
	488
	489	static void __exit ipcomp4_fini(void)
	490	{
	491	if (inet_del_protocol(&ipcomp4_protocol, IPPROTO_COMP) < 0)
	492	printk(KERN_INFO "ip ipcomp close: can't remove protocol\n");
	493	if (xfrm_unregister_type(&ipcomp_type, AF_INET) < 0)
	494	printk(KERN_INFO "ip ipcomp close: can't remove xfrm type\n");
	495	}
	496
	497	module_init(ipcomp4_init);
	498	module_exit(ipcomp4_fini);
	499
	500	MODULE_LICENSE("GPL");
	501	MODULE_DESCRIPTION("IP Payload Compression Protocol (IPComp) - RFC3173");
	502	MODULE_AUTHOR("James Morris <jmorris@intercode.com.au>");
	503