Merge nuts.ninka.net:/home/davem/src/BK/network-2.5

into nuts.ninka.net:/home/davem/src/BK/net-2.5

include/linux/in.h

@@ -70,6 +70,7 @@ struct in_addr {
 #define IP_MTU		14
 #define IP_FREEBIND	15
 #define IP_IPSEC_POLICY	16
+#define IP_XFRM_POLICY	17
 
 /* BSD compatibility */
 #define IP_RECVRETOPTS	IP_RETOPTS

include/linux/xfrm.h

@@ -91,6 +91,22 @@ struct xfrm_stats {
 	__u32	integrity_failed;
 };
 
+enum
+{
+	XFRM_POLICY_IN	= 0,
+	XFRM_POLICY_OUT	= 1,
+	XFRM_POLICY_FWD	= 2,
+	XFRM_POLICY_MAX	= 3
+};
+
+enum
+{
+	XFRM_SHARE_ANY,		/* No limitations */
+	XFRM_SHARE_SESSION,	/* For this session only */
+	XFRM_SHARE_USER,	/* For this user only */
+	XFRM_SHARE_UNIQUE	/* Use once */
+};
+
 /* Netlink configuration messages.  */
 #define XFRM_MSG_BASE		0x10
 
@@ -104,8 +120,9 @@ struct xfrm_stats {
 
 #define XFRM_MSG_ALLOCSPI	(RTM_BASE + 6)
 #define XFRM_MSG_ACQUIRE	(RTM_BASE + 7)
+#define XFRM_MSG_EXPIRE		(RTM_BASE + 8)
 
-#define XFRM_MSG_MAX		(XFRM_MSG_ACQUIRE+1)
+#define XFRM_MSG_MAX		(XFRM_MSG_EXPIRE+1)
 
 struct xfrm_user_tmpl {
 	struct xfrm_id		id;
@@ -113,6 +130,7 @@ struct xfrm_user_tmpl {
 	__u16			reqid;
 	__u8			mode;
 	__u8			share;
+	__u8			optional;
 	__u32			aalgos;
 	__u32			ealgos;
 	__u32			calgos;
@@ -135,9 +153,9 @@ struct xfrm_usersa_info {
 	struct xfrm_lifetime_cfg	lft;
 	struct xfrm_lifetime_cur	curlft;
 	struct xfrm_stats		stats;
+	__u32				seq;
 	__u16				family;
 	__u16				reqid;
-	__u8				sa_type;
 	__u8				mode; /* 0=transport,1=tunnel */
 	__u8				replay_window;
 };
@@ -148,15 +166,26 @@ struct xfrm_usersa_id {
 	__u8				proto;
 };
 
+struct xfrm_userspi_info {
+	struct xfrm_usersa_info		info;
+	u32				min;
+	u32				max;
+};
+
 struct xfrm_userpolicy_info {
 	struct xfrm_selector		sel;
-	struct xfrm_id			id;
 	struct xfrm_lifetime_cfg	lft;
 	struct xfrm_lifetime_cur	curlft;
+	__u32				priority;
 	__u32				index;
 	__u16				family;
 	__u8				dir;
 	__u8				action;
+#define XFRM_POLICY_ALLOW	0
+#define XFRM_POLICY_BLOCK	1
+	__u8				flags;
+#define XFRM_POLICY_LOCALOK	1	/* Allow user to override global policy */
+	__u8				share;
 };
 
 struct xfrm_userpolicy_id {
@@ -165,4 +194,22 @@ struct xfrm_userpolicy_id {
 	__u8				dir;
 };
 
+struct xfrm_user_acquire {
+	struct xfrm_id			id;
+	xfrm_address_t			saddr;
+	struct xfrm_userpolicy_info	policy;
+	__u32				aalgos;
+	__u32				ealgos;
+	__u32				calgos;
+	__u32				seq;
+};
+
+struct xfrm_user_expire {
+	struct xfrm_usersa_info		state;
+	__u8				hard;
+};
+
+#define XFRMGRP_ACQUIRE		1
+#define XFRMGRP_EXPIRE		2
+
 #endif /* _LINUX_XFRM_H */

include/net/xfrm.h

@@ -187,22 +187,6 @@ struct xfrm_tmpl
 
 #define XFRM_MAX_DEPTH		3
 
-enum
-{
-	XFRM_SHARE_ANY,		/* No limitations */
-	XFRM_SHARE_SESSION,	/* For this session only */
-	XFRM_SHARE_USER,	/* For this user only */
-	XFRM_SHARE_UNIQUE	/* Use once */
-};
-
-enum
-{
-	XFRM_POLICY_IN	= 0,
-	XFRM_POLICY_OUT	= 1,
-	XFRM_POLICY_FWD	= 2,
-	XFRM_POLICY_MAX	= 3
-};
-
 struct xfrm_policy
 {
 	struct xfrm_policy	*next;
@@ -218,10 +202,7 @@ struct xfrm_policy
 	struct xfrm_lifetime_cur curlft;
 	struct dst_entry       *bundles;
 	__u8			action;
-#define XFRM_POLICY_ALLOW	0
-#define XFRM_POLICY_BLOCK	1
 	__u8			flags;
-#define XFRM_POLICY_LOCALOK	1	/* Allow user to override global policy */
 	__u8			dead;
 	__u8			xfrm_nr;
 	struct xfrm_tmpl       	xfrm_vec[XFRM_MAX_DEPTH];

net/ipv4/Kconfig

@@ -362,5 +362,13 @@ config INET_ESP
 
 	  If unsure, say Y.
 
+config XFRM_USER
+	tristate "IP: IPsec user configuration interface"
+	---help---
+	  Support for IPsec user configuration interface used
+	  by native Linux tools.
+
+	  If unsure, say Y.
+
 source "net/ipv4/netfilter/Kconfig"
 

net/ipv4/Makefile

@@ -20,6 +20,7 @@ obj-$(CONFIG_INET_AH) += ah.o
 obj-$(CONFIG_INET_ESP) += esp.o
 obj-$(CONFIG_IP_PNP) += ipconfig.o
 obj-$(CONFIG_NETFILTER)	+= netfilter/
+obj-$(CONFIG_XFRM_USER) += xfrm_user.o
 
 obj-y += xfrm_policy.o xfrm_state.o xfrm_input.o
 

net/ipv4/ah.c

@@ -189,7 +189,7 @@ int ah_output(struct sk_buff *skb)
 		top_iph->saddr = x->props.saddr.xfrm4_addr;
 		top_iph->daddr = x->id.daddr.xfrm4_addr;
 		ah = (struct ip_auth_hdr*)(top_iph+1);
-		ah->nexthdr = IPPROTO_IP;
+		ah->nexthdr = IPPROTO_IPIP;
 	} else {
 		memcpy(&tmp_iph, skb->data, iph->ihl*4);
 		top_iph = (struct iphdr*)skb_push(skb, x->props.header_len);

net/ipv4/esp.c

@@ -190,11 +190,10 @@ esp_hmac_digest(struct esp_data *esp, struct sk_buff *skb, int offset,
 	struct crypto_tfm *tfm = esp->auth.tfm;
 	char *digest = esp->auth.work_digest;
 
-	memset(auth_data, 0, esp->auth.authlen);
  	crypto_hmac_init(tfm, esp->auth.key, &esp->auth.key_len);
 	skb_digest_walk(skb, tfm, offset, len);
 	crypto_hmac_final(tfm, esp->auth.key, &esp->auth.key_len, digest);
-	memcpy(auth_data, digest, crypto_tfm_alg_digestsize(tfm));
+	memcpy(auth_data, digest, esp->auth.authlen);
 }
 
 /* Check that skb data bits are writable. If they are not, copy data
@@ -370,7 +369,7 @@ int esp_output(struct sk_buff *skb)
 	if (x->props.mode) {
 		top_iph = (struct iphdr*)skb_push(skb, x->props.header_len);
 		esph = (struct ip_esp_hdr*)(top_iph+1);
-		*(u8*)(trailer->tail - 1) = IPPROTO_IP;
+		*(u8*)(trailer->tail - 1) = IPPROTO_IPIP;
 		top_iph->ihl = 5;
 		top_iph->version = 4;
 		top_iph->tos = iph->tos;	/* DS disclosed */
@@ -463,16 +462,16 @@ int esp_input(struct xfrm_state *x, struct sk_buff *skb)
 
 	/* If integrity check is required, do this. */
 	if (esp->auth.authlen) {
-		int icvsize = crypto_tfm_alg_digestsize(esp->auth.tfm);
-		u8 sum[icvsize];
-		u8 sum1[icvsize];
+		u8 sum[esp->auth.authlen];
+		u8 sum1[esp->auth.authlen];
 
 		esp->auth.digest(esp, skb, 0, skb->len-esp->auth.authlen, sum);
 
-		if (skb_copy_bits(skb, skb->len-esp->auth.authlen, sum1, icvsize))
+		if (skb_copy_bits(skb, skb->len-esp->auth.authlen, sum1, 
+				esp->auth.authlen))
 			BUG();
 
-		if (unlikely(memcmp(sum, sum1, icvsize))) {
+		if (unlikely(memcmp(sum, sum1, esp->auth.authlen))) {
 			x->stats.integrity_failed++;
 			goto out;
 		}
@@ -605,14 +604,20 @@ int esp_init_state(struct xfrm_state *x, void *args)
 	memset(esp, 0, sizeof(*esp));
 
 	if (x->aalg) {
+		int digestsize;
+
 		esp->auth.key = x->aalg->alg_key;
 		esp->auth.key_len = (x->aalg->alg_key_len+7)/8;
 		esp->auth.tfm = crypto_alloc_tfm(x->aalg->alg_name, 0);
 		if (esp->auth.tfm == NULL)
 			goto error;
 		esp->auth.digest = esp_hmac_digest;
-		esp->auth.authlen = crypto_tfm_alg_digestsize(esp->auth.tfm);
-		esp->auth.work_digest = kmalloc(esp->auth.authlen, GFP_KERNEL);
+		digestsize = crypto_tfm_alg_digestsize(esp->auth.tfm);
+		/* XXX RFC2403 and RFC 2404 truncate auth to 96 bit */
+		esp->auth.authlen = 12;
+		if (esp->auth.authlen > digestsize) /* XXX */
+			BUG();
+		esp->auth.work_digest = kmalloc(digestsize, GFP_KERNEL);
 		if (!esp->auth.work_digest)
 			goto error;
 	}

net/ipv4/ip_sockglue.c

@@ -626,6 +626,7 @@ int ip_setsockopt(struct sock *sk, int level, int optname, char *optval, int opt
 	                break;			
  
 		case IP_IPSEC_POLICY:
+		case IP_XFRM_POLICY:
 			err = xfrm_user_policy(sk, optname, optval, optlen);
 			break;
 

net/ipv4/udp.c

@@ -944,7 +944,7 @@ static int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb)
 	/*
 	 *	Charge it to the socket, dropping if the queue is full.
 	 */
-	if (!xfrm_policy_check(NULL, XFRM_POLICY_IN, skb)) {
+	if (!xfrm_policy_check(sk, XFRM_POLICY_IN, skb)) {
 		kfree_skb(skb);
 		return -1;
 	}

net/ipv4/xfrm_input.c

@@ -91,7 +91,7 @@ int xfrm4_rcv(struct sk_buff *skb)
 		iph = skb->nh.iph;
 
 		if (x->props.mode) {
-			if (iph->protocol != IPPROTO_IP)
+			if (iph->protocol != IPPROTO_IPIP)
 				goto drop;
 			skb->nh.raw = skb->data;
 			iph = skb->nh.iph;

net/ipv4/xfrm_user.c

+/* xfrm_user.c: User interface to configure xfrm engine.
+ *
+ * Copyright (C) 2002 David S. Miller (davem@redhat.com)
+ */
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/slab.h>
+#include <linux/socket.h>
+#include <linux/string.h>
+#include <linux/net.h>
+#include <linux/skbuff.h>
+#include <linux/netlink.h>
+#include <linux/rtnetlink.h>
+#include <linux/pfkeyv2.h>
+#include <linux/ipsec.h>
+#include <linux/init.h>
+#include <net/sock.h>
+#include <net/xfrm.h>
+
+static struct sock *xfrm_nl;
+
+static int verify_one_alg(struct rtattr **xfrma, enum xfrm_attr_type_t type)
+{
+	struct rtattr *rt = xfrma[type];
+	struct xfrm_algo *algp;
+
+	if (!rt)
+		return 0;
+
+	if ((rt->rta_len - sizeof(*rt)) < sizeof(*algp))
+		return -EINVAL;
+
+	algp = RTA_DATA(rt);
+	switch (type) {
+	case XFRMA_ALG_AUTH:
+	case XFRMA_ALG_CRYPT:
+		if (!algp->alg_key_len)
+			return -EINVAL;
+		break;
+
+	case XFRMA_ALG_COMP:
+		/* Zero length keys are legal.  */
+		break;
+
+	default:
+		return -EINVAL;
+	};
+
+	algp->alg_name[CRYPTO_MAX_ALG_NAME - 1] = '\0';
+	return 0;
+}
+
+static int verify_newsa_info(struct xfrm_usersa_info *p,
+			     struct rtattr **xfrma)
+{
+	int err;
+
+	err = -EINVAL;
+	switch (p->family) {
+	case AF_INET:
+		break;
+
+	case AF_INET6: /* XXX */
+		err = -EAFNOSUPPORT;
+
+		/* fallthru */
+	default:
+		goto out;
+	};
+
+	err = -EINVAL;
+	switch (p->id.proto) {
+	case IPPROTO_AH:
+		if (!xfrma[XFRMA_ALG_AUTH]	||
+		    xfrma[XFRMA_ALG_CRYPT]	||
+		    xfrma[XFRMA_ALG_COMP])
+			goto out;
+		break;
+
+	case IPPROTO_ESP:
+		if ((!xfrma[XFRMA_ALG_AUTH] &&
+		     !xfrma[XFRMA_ALG_CRYPT])	||
+		    xfrma[XFRMA_ALG_COMP])
+			goto out;
+		break;
+
+	case IPPROTO_COMP:
+		if (!xfrma[XFRMA_ALG_COMP]	||
+		    xfrma[XFRMA_ALG_AUTH]	||
+		    xfrma[XFRMA_ALG_CRYPT])
+			goto out;
+		break;
+
+	default:
+		goto out;
+	};
+
+	if ((err = verify_one_alg(xfrma, XFRMA_ALG_AUTH)))
+		goto out;
+	if ((err = verify_one_alg(xfrma, XFRMA_ALG_CRYPT)))
+		goto out;
+	if ((err = verify_one_alg(xfrma, XFRMA_ALG_COMP)))
+		goto out;
+
+	err = -EINVAL;
+	switch (p->mode) {
+	case 0:
+	case 1:
+		break;
+
+	default:
+		goto out;
+	};
+
+	err = 0;
+
+out:
+	return err;
+}
+
+static int attach_one_algo(struct xfrm_algo **algpp, struct rtattr *u_arg)
+{
+	struct rtattr *rta = u_arg;
+	struct xfrm_algo *p, *ualg;
+
+	if (!rta)
+		return 0;
+
+	ualg = RTA_DATA(rta);
+	p = kmalloc(sizeof(*ualg) + ualg->alg_key_len, GFP_KERNEL);
+	if (!p)
+		return -ENOMEM;
+
+	memcpy(p, ualg, sizeof(*ualg) + ualg->alg_key_len);
+	*algpp = p;
+	return 0;
+}
+
+static void copy_from_user_state(struct xfrm_state *x, struct xfrm_usersa_info *p)
+{
+	memcpy(&x->id, &p->id, sizeof(x->id));
+	memcpy(&x->sel, &p->sel, sizeof(x->sel));
+	memcpy(&x->lft, &p->lft, sizeof(x->lft));
+	x->props.mode = p->mode;
+	x->props.replay_window = p->replay_window;
+	x->props.reqid = p->reqid;
+	x->props.saddr = x->sel.saddr;
+}
+
+static struct xfrm_state *xfrm_state_construct(struct xfrm_usersa_info *p,
+					       struct rtattr **xfrma,
+					       int *errp)
+{
+	struct xfrm_state *x = xfrm_state_alloc();
+	int err = -ENOMEM;
+
+	if (!x)
+		goto error_no_put;
+
+	copy_from_user_state(x, p);
+
+	if ((err = attach_one_algo(&x->aalg, xfrma[XFRMA_ALG_AUTH])))
+		goto error;
+	if ((err = attach_one_algo(&x->ealg, xfrma[XFRMA_ALG_CRYPT])))
+		goto error;
+	if ((err = attach_one_algo(&x->calg, xfrma[XFRMA_ALG_COMP])))
+		goto error;
+
+	err = -ENOENT;
+	x->type = xfrm_get_type(x->id.proto);
+	if (x->type == NULL)
+		goto error;
+
+	err = x->type->init_state(x, NULL);
+	if (err)
+		goto error;
+
+	x->curlft.add_time = (unsigned long) xtime.tv_sec;
+	x->km.state = XFRM_STATE_VALID;
+	x->km.seq = p->seq;
+
+	return x;
+
+error:
+	xfrm_state_put(x);
+error_no_put:
+	*errp = err;
+	return NULL;
+}
+
+static int xfrm_add_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma)
+{
+	struct xfrm_usersa_info *p = NLMSG_DATA(nlh);
+	struct xfrm_state *x, *x1;
+	int err;
+
+	err = verify_newsa_info(p, (struct rtattr **) xfrma);
+	if (err)
+		return err;
+
+	x = xfrm_state_construct(p, (struct rtattr **) xfrma, &err);
+	if (!x)
+		return err;
+
+	x1 = xfrm_state_lookup(x->props.saddr.xfrm4_addr,
+			       x->id.spi, x->id.proto);
+	if (x1) {
+		xfrm_state_put(x);
+		xfrm_state_put(x1);
+		return -EEXIST;
+	}
+
+	xfrm_state_insert(x);
+
+	return 0;
+}
+
+static int xfrm_del_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma)
+{
+	struct xfrm_state *x;
+	struct xfrm_usersa_id *p = NLMSG_DATA(nlh);
+
+	x = xfrm_state_lookup(p->saddr.xfrm4_addr, p->spi, p->proto);
+	if (x == NULL)
+		return -ESRCH;
+
+	xfrm_state_delete(x);
+	xfrm_state_put(x);
+
+	return 0;
+}
+
+static void copy_to_user_state(struct xfrm_state *x, struct xfrm_usersa_info *p)
+{
+	memcpy(&p->id, &x->id, sizeof(p->id));
+	memcpy(&p->sel, &x->sel, sizeof(p->sel));
+	memcpy(&p->lft, &x->lft, sizeof(p->lft));
+	memcpy(&p->curlft, &x->curlft, sizeof(p->curlft));
+	memcpy(&p->stats, &x->stats, sizeof(p->stats));
+	p->mode = x->props.mode;
+	p->replay_window = x->props.replay_window;
+	p->reqid = x->props.reqid;
+	p->seq = x->km.seq;
+}
+
+struct xfrm_dump_info {
+	struct sk_buff *in_skb;
+	struct sk_buff *out_skb;
+	u32 nlmsg_seq;
+	int start_idx;
+	int this_idx;
+};
+
+static int dump_one_state(struct xfrm_state *x, int count, void *ptr)
+{
+	struct xfrm_dump_info *sp = ptr;
+	struct sk_buff *in_skb = sp->in_skb;
+	struct sk_buff *skb = sp->out_skb;
+	struct xfrm_usersa_info *p;
+	struct nlmsghdr *nlh;
+	unsigned char *b = skb->tail;
+
+	if (sp->this_idx < sp->start_idx)
+		goto out;
+
+	nlh = NLMSG_PUT(skb, NETLINK_CB(in_skb).pid,
+			sp->nlmsg_seq,
+			XFRM_MSG_NEWSA, sizeof(*p));
+	nlh->nlmsg_flags = 0;
+
+	p = NLMSG_DATA(nlh);
+	copy_to_user_state(x, p);
+
+	if (x->aalg)
+		RTA_PUT(skb, XFRMA_ALG_AUTH, sizeof(*(x->aalg)), x->aalg);
+	if (x->ealg)
+		RTA_PUT(skb, XFRMA_ALG_CRYPT, sizeof(*(x->ealg)), x->ealg);
+	if (x->calg)
+		RTA_PUT(skb, XFRMA_ALG_COMP, sizeof(*(x->calg)), x->calg);
+
+	nlh->nlmsg_len = skb->tail - b;
+out:
+	sp->this_idx++;
+	return 0;
+
+nlmsg_failure:
+rtattr_failure:
+	skb_trim(skb, b - skb->data);
+	return -1;
+}
+
+static int xfrm_dump_sa(struct sk_buff *skb, struct netlink_callback *cb)
+{
+	struct xfrm_dump_info info;
+
+	info.in_skb = cb->skb;
+	info.out_skb = skb;
+	info.nlmsg_seq = cb->nlh->nlmsg_seq;
+	info.this_idx = 0;
+	info.start_idx = cb->args[0];
+	(void) xfrm_state_walk(IPSEC_PROTO_ANY, dump_one_state, &info);
+	cb->args[0] = info.this_idx;
+
+	return skb->len;
+}
+
+static struct sk_buff *xfrm_state_netlink(struct sk_buff *in_skb,
+					  struct xfrm_state *x, u32 seq)
+{
+	struct xfrm_dump_info info;
+	struct sk_buff *skb;
+
+	skb = alloc_skb(NLMSG_GOODSIZE, GFP_ATOMIC);
+	if (!skb)
+		return ERR_PTR(-ENOMEM);
+
+	NETLINK_CB(skb).dst_pid = NETLINK_CB(in_skb).pid;
+	info.in_skb = in_skb;
+	info.out_skb = skb;
+	info.nlmsg_seq = seq;
+	info.this_idx = info.start_idx = 0;
+
+	if (dump_one_state(x, 0, &info)) {
+		kfree_skb(skb);
+		return NULL;
+	}
+
+	return skb;
+}
+
+static int xfrm_get_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma)
+{
+	struct xfrm_usersa_id *p = NLMSG_DATA(nlh);
+	struct xfrm_state *x;
+	struct sk_buff *resp_skb;
+	int err;
+
+	x = xfrm_state_lookup(p->saddr.xfrm4_addr, p->spi, p->proto);
+	err = -ESRCH;
+	if (x == NULL)
+		goto out_noput;
+
+	resp_skb = xfrm_state_netlink(skb, x, nlh->nlmsg_seq);
+	if (IS_ERR(resp_skb)) {
+		err = PTR_ERR(resp_skb);
+	} else {
+		err = netlink_unicast(xfrm_nl, resp_skb,
+				      NETLINK_CB(skb).pid, MSG_DONTWAIT);
+	}
+	xfrm_state_put(x);
+out_noput:
+	return err;
+}
+
+static int verify_userspi_info(struct xfrm_userspi_info *p)
+{
+	switch (p->info.id.proto) {
+	case IPPROTO_AH:
+	case IPPROTO_ESP:
+		break;
+
+	case IPPROTO_COMP:
+		/* IPCOMP spi is 16-bits. */
+		if (p->min >= 0x10000 ||
+		    p->max >= 0x10000)
+			return -EINVAL;
+
+	default:
+		return -EINVAL;
+	};
+
+	if (p->min > p->max)
+		return -EINVAL;
+
+	return 0;
+}
+
+static int xfrm_alloc_userspi(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma)
+{
+	struct xfrm_state *x;
+	struct xfrm_userspi_info *p;
+	struct sk_buff *resp_skb;
+	int err;
+
+	p = NLMSG_DATA(nlh);
+	err = verify_userspi_info(p);
+	if (err)
+		goto out_noput;
+	x = xfrm_find_acq(p->info.mode, p->info.reqid, p->info.id.proto,
+			  p->info.sel.daddr.xfrm4_addr,
+			  p->info.sel.saddr.xfrm4_addr);
+	err = -ENOENT;
+	if (x == NULL)
+		goto out_noput;
+
+	resp_skb = ERR_PTR(-ENOENT);
+
+	spin_lock_bh(&x->lock);
+	if (x->km.state != XFRM_STATE_DEAD) {
+		xfrm_alloc_spi(x, p->min, p->max);
+		if (x->id.spi)
+			resp_skb = xfrm_state_netlink(skb, x, nlh->nlmsg_seq);
+	}
+	spin_unlock_bh(&x->lock);
+
+	if (IS_ERR(resp_skb)) {
+		err = PTR_ERR(resp_skb);
+		goto out;
+	}
+
+	err = netlink_unicast(xfrm_nl, resp_skb,
+			      NETLINK_CB(skb).pid, MSG_DONTWAIT);
+
+out:
+	xfrm_state_put(x);
+out_noput:
+	return err;
+}
+
+static int verify_policy_dir(__u8 dir)
+{
+	switch (dir) {
+	case XFRM_POLICY_IN:
+	case XFRM_POLICY_OUT:
+	case XFRM_POLICY_FWD:
+		break;
+
+	default:
+		return -EINVAL;
+	};
+
+	return 0;
+}
+
+static int verify_newpolicy_info(struct xfrm_userpolicy_info *p)
+{
+	switch (p->share) {
+	case XFRM_SHARE_ANY:
+	case XFRM_SHARE_SESSION:
+	case XFRM_SHARE_USER:
+	case XFRM_SHARE_UNIQUE:
+		break;
+
+	default:
+		return -EINVAL;
+	};
+
+	switch (p->action) {
+	case XFRM_POLICY_ALLOW:
+	case XFRM_POLICY_BLOCK:
+		break;
+
+	default:
+		return -EINVAL;
+	};
+
+	return verify_policy_dir(p->dir);
+}
+
+static void copy_templates(struct xfrm_policy *xp, struct xfrm_user_tmpl *ut,
+			   int nr)
+{
+	int i;
+
+	xp->xfrm_nr = nr;
+	for (i = 0; i < nr; i++, ut++) {
+		struct xfrm_tmpl *t = &xp->xfrm_vec[i];
+
+		memcpy(&t->id, &ut->id, sizeof(struct xfrm_id));
+		memcpy(&t->saddr, &ut->saddr,
+		       sizeof(xfrm_address_t));
+		t->reqid = ut->reqid;
+		t->mode = ut->mode;
+		t->share = ut->share;
+		t->optional = ut->optional;
+		t->aalgos = ut->aalgos;
+		t->ealgos = ut->ealgos;
+		t->calgos = ut->calgos;
+	}
+}
+
+static int copy_user_tmpl(struct xfrm_policy *pol, struct rtattr **xfrma)
+{
+	struct rtattr *rt = xfrma[XFRMA_TMPL];
+	struct xfrm_user_tmpl *utmpl;
+	int nr;
+
+	if (!rt) {
+		pol->xfrm_nr = 0;
+	} else {
+		nr = (rt->rta_len - sizeof(*rt)) / sizeof(*utmpl);
+
+		if (nr > XFRM_MAX_DEPTH)
+			return -EINVAL;
+
+		copy_templates(pol, RTA_DATA(rt), nr);
+	}
+	return 0;
+}
+
+static void copy_from_user_policy(struct xfrm_policy *xp, struct xfrm_userpolicy_info *p)
+{
+	xp->priority = p->priority;
+	xp->index = p->index;
+	memcpy(&xp->selector, &p->sel, sizeof(xp->selector));
+	memcpy(&xp->lft, &p->lft, sizeof(xp->lft));
+	xp->action = p->action;
+	xp->flags = p->flags;
+	/* XXX xp->family = p->family; */
+	/* XXX xp->share = p->share; */
+}
+
+static void copy_to_user_policy(struct xfrm_policy *xp, struct xfrm_userpolicy_info *p, int dir)
+{
+	memcpy(&p->sel, &xp->selector, sizeof(p->sel));
+	memcpy(&p->lft, &xp->lft, sizeof(p->lft));
+	memcpy(&p->curlft, &xp->curlft, sizeof(p->curlft));
+	p->priority = xp->priority;
+	p->index = xp->index;
+	p->family = AF_INET; /* XXX xp->family */
+	p->dir = dir;
+	p->action = xp->action;
+	p->flags = xp->flags;
+	p->share = XFRM_SHARE_ANY; /* XXX xp->share */
+}
+
+static struct xfrm_policy *xfrm_policy_construct(struct xfrm_userpolicy_info *p, struct rtattr **xfrma, int *errp)
+{
+	struct xfrm_policy *xp = xfrm_policy_alloc(GFP_KERNEL);
+	int err;
+
+	if (!xp) {
+		*errp = -ENOMEM;
+		return NULL;
+	}
+
+	copy_from_user_policy(xp, p);
+	err = copy_user_tmpl(xp, xfrma);
+	if (err) {
+		*errp = err;
+		kfree(xp);
+		xp = NULL;
+	}
+
+	return xp;
+}
+
+static int xfrm_add_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma)
+{
+	struct xfrm_userpolicy_info *p = NLMSG_DATA(nlh);
+	struct xfrm_policy *xp;
+	int err;
+
+	err = verify_newpolicy_info(p);
+	if (err)
+		return err;
+
+	xp = xfrm_policy_construct(p, (struct rtattr **) xfrma, &err);
+	if (!xp)
+		return err;
+
+	err = xfrm_policy_insert(p->dir, xp, 1);
+	if (err) {
+		kfree(xp);
+		return err;
+	}
+
+	xfrm_pol_put(xp);
+
+	return 0;
+}
+
+static int xfrm_del_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma)
+{
+	struct xfrm_policy *xp;
+	struct xfrm_userpolicy_id *p;
+	int err;
+
+	p = NLMSG_DATA(nlh);
+
+	err = verify_policy_dir(p->dir);
+	if (err)
+		return err;
+
+	xp = xfrm_policy_delete(p->dir, &p->sel);
+	if (xp == NULL)
+		return -ENOENT;
+	xfrm_policy_kill(xp);
+	xfrm_pol_put(xp);
+	return 0;
+}
+
+static int dump_one_policy(struct xfrm_policy *xp, int dir, int count, void *ptr)
+{
+	struct xfrm_dump_info *sp = ptr;
+	struct xfrm_userpolicy_info *p;
+	struct sk_buff *in_skb = sp->in_skb;
+	struct sk_buff *skb = sp->out_skb;
+	struct nlmsghdr *nlh;
+	unsigned char *b = skb->tail;
+
+	if (sp->this_idx < sp->start_idx)
+		goto out;
+
+	nlh = NLMSG_PUT(skb, NETLINK_CB(in_skb).pid,
+			sp->nlmsg_seq,
+			XFRM_MSG_NEWPOLICY, sizeof(*p));
+	p = NLMSG_DATA(nlh);
+	nlh->nlmsg_flags = 0;
+
+	copy_to_user_policy(xp, p, dir);
+
+	if (xp->xfrm_nr) {
+		struct xfrm_user_tmpl vec[XFRM_MAX_DEPTH];
+		int i;
+
+		for (i = 0; i < xp->xfrm_nr; i++) {
+			struct xfrm_user_tmpl *up = &vec[i];
+			struct xfrm_tmpl *kp = &xp->xfrm_vec[i];
+
+			memcpy(&up->id, &kp->id, sizeof(up->id));
+			memcpy(&up->saddr, &kp->saddr, sizeof(up->saddr));
+			up->reqid = kp->reqid;
+			up->mode = kp->mode;
+			up->share = kp->share;
+			up->optional = kp->optional;
+			up->aalgos = kp->aalgos;
+			up->ealgos = kp->ealgos;
+			up->calgos = kp->calgos;
+		}
+		RTA_PUT(skb, XFRMA_TMPL,
+			(sizeof(struct xfrm_user_tmpl) * xp->xfrm_nr),
+			vec);
+	}
+
+	nlh->nlmsg_len = skb->tail - b;
+out:
+	sp->this_idx++;
+	return 0;
+
+nlmsg_failure:
+rtattr_failure:
+	skb_trim(skb, b - skb->data);
+	return -1;
+}
+
+static int xfrm_dump_policy(struct sk_buff *skb, struct netlink_callback *cb)
+{
+	struct xfrm_dump_info info;
+
+	info.in_skb = cb->skb;
+	info.out_skb = skb;
+	info.nlmsg_seq = cb->nlh->nlmsg_seq;
+	info.start_idx = cb->args[0];
+	(void) xfrm_policy_walk(dump_one_policy, &info);
+	cb->args[0] = info.this_idx;
+
+	return skb->len;
+}
+
+static struct sk_buff *xfrm_policy_netlink(struct sk_buff *in_skb,
+					  struct xfrm_policy *xp,
+					  int dir, u32 seq)
+{
+	struct xfrm_dump_info info;
+	struct sk_buff *skb;
+
+	skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL);
+	if (!skb)
+		return ERR_PTR(-ENOMEM);
+
+	NETLINK_CB(skb).dst_pid = NETLINK_CB(in_skb).pid;
+	info.in_skb = in_skb;
+	info.out_skb = skb;
+	info.nlmsg_seq = seq;
+	info.this_idx = info.start_idx = 0;
+
+	if (dump_one_policy(xp, dir, 0, &info) < 0) {
+		kfree_skb(skb);
+		return NULL;
+	}
+
+	return skb;
+}
+
+static int xfrm_get_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma)
+{
+	struct xfrm_policy *xp;
+	struct xfrm_userpolicy_id *p;
+	struct sk_buff *resp_skb;
+	int err;
+
+	p = NLMSG_DATA(nlh);
+	xp = xfrm_policy_byid(p->dir, p->index, 0);
+	if (xp == NULL)
+		return -ENOENT;
+
+	resp_skb = xfrm_policy_netlink(skb, xp, p->dir, nlh->nlmsg_seq);
+	if (IS_ERR(resp_skb)) {
+		err = PTR_ERR(resp_skb);
+	} else {
+		err = netlink_unicast(xfrm_nl, resp_skb,
+				      NETLINK_CB(skb).pid, MSG_DONTWAIT);
+	}
+
+	xfrm_pol_put(xp);
+
+	return err;
+}
+
+static const int xfrm_msg_min[(XFRM_MSG_MAX + 1 - XFRM_MSG_BASE)] = {
+	NLMSG_LENGTH(sizeof(struct xfrm_usersa_info)),	/* NEW SA */
+	NLMSG_LENGTH(sizeof(struct xfrm_usersa_id)),	/* DEL SA */
+	NLMSG_LENGTH(sizeof(struct xfrm_usersa_id)),	/* GET SA */
+	NLMSG_LENGTH(sizeof(struct xfrm_userpolicy_info)),/* NEW POLICY */
+	NLMSG_LENGTH(sizeof(struct xfrm_userpolicy_id)),  /* DEL POLICY */
+	NLMSG_LENGTH(sizeof(struct xfrm_userpolicy_id)),  /* GET POLICY */
+	NLMSG_LENGTH(sizeof(struct xfrm_userspi_info)),	/* ALLOC SPI */
+	NLMSG_LENGTH(sizeof(struct xfrm_user_acquire)),	/* ACQUIRE */
+	NLMSG_LENGTH(sizeof(struct xfrm_user_expire)),	/* EXPIRE */
+};
+
+static struct xfrm_link {
+	int (*doit)(struct sk_buff *, struct nlmsghdr *, void **);
+	int (*dump)(struct sk_buff *, struct netlink_callback *);
+} xfrm_dispatch[] = {
+	{	.doit	=	xfrm_add_sa, 		},
+	{	.doit	=	xfrm_del_sa, 		},
+	{
+		.doit	=	xfrm_get_sa,
+		.dump	=	xfrm_dump_sa,
+	},
+	{	.doit	=	xfrm_add_policy 	},
+	{	.doit	=	xfrm_del_policy 	},
+	{
+		.doit	=	xfrm_get_policy,
+		.dump	=	xfrm_dump_policy,
+	},
+	{	.doit	=	xfrm_alloc_userspi	},
+};
+
+static int xfrm_done(struct netlink_callback *cb)
+{
+	return 0;
+}
+
+static int xfrm_user_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh, int *errp)
+{
+	struct rtattr *xfrma[XFRMA_MAX];
+	struct xfrm_link *link;
+	int type, min_len, kind;
+
+	if (!(nlh->nlmsg_flags & NLM_F_REQUEST))
+		return 0;
+
+	type = nlh->nlmsg_type;
+
+	/* A control message: ignore them */
+	if (type < XFRM_MSG_BASE)
+		return 0;
+
+	/* Unknown message: reply with EINVAL */
+	if (type > XFRM_MSG_MAX)
+		goto err_einval;
+
+	type -= XFRM_MSG_BASE;
+	kind = (type & 3);
+	link = &xfrm_dispatch[type];
+
+	/* All operations require privileges, even GET */
+	if (!cap_raised(NETLINK_CB(skb).eff_cap, CAP_NET_ADMIN)) {
+		*errp = -EPERM;
+		return -1;
+	}
+
+	if (kind == 2 && (nlh->nlmsg_flags & NLM_F_DUMP)) {
+		u32 rlen;
+
+		if (link->dump == NULL)
+			goto err_einval;
+
+		if ((*errp = netlink_dump_start(xfrm_nl, skb, nlh,
+						link->dump,
+						xfrm_done)) != 0) {
+			return -1;
+		}
+		rlen = NLMSG_ALIGN(nlh->nlmsg_len);
+		if (rlen > skb->len)
+			rlen = skb->len;
+		skb_pull(skb, rlen);
+		return -1;
+	}
+
+	memset(xfrma, 0, sizeof(xfrma));
+
+	if (nlh->nlmsg_len < (min_len = xfrm_msg_min[type]))
+		goto err_einval;
+
+	if (nlh->nlmsg_len > min_len) {
+		int attrlen = nlh->nlmsg_len - NLMSG_ALIGN(min_len);
+		struct rtattr *attr = (void *) nlh + NLMSG_ALIGN(min_len);
+
+		while (RTA_OK(attr, attrlen)) {
+			unsigned short flavor = attr->rta_type;
+			if (flavor) {
+				if (flavor > XFRMA_MAX)
+					goto err_einval;
+				xfrma[flavor - 1] = attr;
+			}
+			attr = RTA_NEXT(attr, attrlen);
+		}
+	}
+
+	if (link->doit == NULL)
+		goto err_einval;
+	*errp = link->doit(skb, nlh, (void **) &xfrma);
+
+	return *errp;
+
+err_einval:
+	*errp = -EINVAL;
+	return -1;
+}
+
+static int xfrm_user_rcv_skb(struct sk_buff *skb)
+{
+	int err;
+	struct nlmsghdr *nlh;
+
+	while (skb->len >= NLMSG_SPACE(0)) {
+		u32 rlen;
+
+		nlh = (struct nlmsghdr *) skb->data;
+		if (nlh->nlmsg_len < sizeof(*nlh) ||
+		    skb->len < nlh->nlmsg_len)
+			return 0;
+		rlen = NLMSG_ALIGN(nlh->nlmsg_len);
+		if (rlen > skb->len)
+			rlen = skb->len;
+		if (xfrm_user_rcv_msg(skb, nlh, &err)) {
+			if (err == 0)
+				return -1;
+			netlink_ack(skb, nlh, err);
+		} else if (nlh->nlmsg_flags & NLM_F_ACK)
+			netlink_ack(skb, nlh, 0);
+		skb_pull(skb, rlen);
+	}
+
+	return 0;
+}
+
+static void xfrm_netlink_rcv(struct sock *sk, int len)
+{
+	do {
+		struct sk_buff *skb;
+
+		down(&xfrm_cfg_sem);
+
+		while ((skb = skb_dequeue(&sk->receive_queue)) != NULL) {
+			if (xfrm_user_rcv_skb(skb)) {
+				if (skb->len)
+					skb_queue_head(&sk->receive_queue, skb);
+				else
+					kfree_skb(skb);
+				break;
+			}
+			kfree_skb(skb);
+		}
+
+		up(&xfrm_cfg_sem);
+
+	} while (xfrm_nl && xfrm_nl->receive_queue.qlen);
+}
+
+static int build_expire(struct sk_buff *skb, struct xfrm_state *x, int hard)
+{
+	struct xfrm_user_expire *ue;
+	struct nlmsghdr *nlh;
+	unsigned char *b = skb->tail;
+
+	nlh = NLMSG_PUT(skb, 0, 0, XFRM_MSG_EXPIRE,
+			sizeof(*ue));
+	ue = NLMSG_DATA(nlh);
+	nlh->nlmsg_flags = 0;
+
+	copy_to_user_state(x, &ue->state);
+	ue->hard = (hard != 0) ? 1 : 0;
+
+	nlh->nlmsg_len = skb->tail - b;
+	return skb->len;
+
+nlmsg_failure:
+	skb_trim(skb, b - skb->data);
+	return -1;
+}
+
+static int xfrm_send_notify(struct xfrm_state *x, int hard)
+{
+	struct sk_buff *skb;
+
+	skb = alloc_skb(sizeof(struct xfrm_user_expire) + 16, GFP_ATOMIC);
+	if (skb == NULL)
+		return -ENOMEM;
+
+	if (build_expire(skb, x, hard) < 0)
+		BUG();
+
+	NETLINK_CB(skb).dst_groups = XFRMGRP_EXPIRE;
+	netlink_broadcast(xfrm_nl, skb, 0, XFRMGRP_EXPIRE, GFP_ATOMIC);
+
+	return 0;
+}
+
+/* XXX Make this xfrm_state.c:xfrm_get_acqseq() */
+static u32 get_acqseq(void)
+{
+	u32 res;
+	static u32 acqseq;
+	static spinlock_t acqseq_lock = SPIN_LOCK_UNLOCKED;
+
+	spin_lock_bh(&acqseq_lock);
+	res = (++acqseq ? : ++acqseq);
+	spin_unlock_bh(&acqseq_lock);
+	return res;
+}
+
+static int build_acquire(struct sk_buff *skb, struct xfrm_state *x,
+			 struct xfrm_tmpl *xt, struct xfrm_policy *xp,
+			 int dir)
+{
+	struct xfrm_user_acquire *ua;
+	struct nlmsghdr *nlh;
+	unsigned char *b = skb->tail;
+	__u32 seq = get_acqseq();
+
+	nlh = NLMSG_PUT(skb, 0, 0, XFRM_MSG_ACQUIRE,
+			sizeof(*ua));
+	ua = NLMSG_DATA(nlh);
+	nlh->nlmsg_flags = 0;
+
+	memcpy(&ua->id, &x->id, sizeof(ua->id));
+	memcpy(&ua->saddr, &x->props.saddr, sizeof(ua->saddr));
+	copy_to_user_policy(xp, &ua->policy, dir);
+	ua->aalgos = xt->aalgos;
+	ua->ealgos = xt->ealgos;
+	ua->calgos = xt->calgos;
+	ua->seq = x->km.seq = seq;
+
+	nlh->nlmsg_len = skb->tail - b;
+	return skb->len;
+
+nlmsg_failure:
+	skb_trim(skb, b - skb->data);
+	return -1;
+}
+
+static int xfrm_send_acquire(struct xfrm_state *x, struct xfrm_tmpl *xt,
+			     struct xfrm_policy *xp, int dir)
+{
+	struct sk_buff *skb;
+
+	skb = alloc_skb(sizeof(struct xfrm_user_acquire) + 16, GFP_ATOMIC);
+	if (skb == NULL)
+		return -ENOMEM;
+
+	if (build_acquire(skb, x, xt, xp, dir) < 0)
+		BUG();
+
+	NETLINK_CB(skb).dst_groups = XFRMGRP_ACQUIRE;
+	netlink_broadcast(xfrm_nl, skb, 0, XFRMGRP_ACQUIRE, GFP_ATOMIC);
+
+	return 0;
+}
+
+/* User gives us xfrm_user_policy_info followed by an array of 0
+ * or more templates.
+ */
+struct xfrm_policy *xfrm_compile_policy(int opt, u8 *data, int len, int *dir)
+{
+	struct xfrm_userpolicy_info *p = (struct xfrm_userpolicy_info *)data;
+	struct xfrm_user_tmpl *ut = (struct xfrm_user_tmpl *) (p + 1);
+	struct xfrm_policy *xp;
+	int nr;
+
+	if (opt != IP_XFRM_POLICY) {
+		*dir = -EOPNOTSUPP;
+		return NULL;
+	}
+	*dir = -EINVAL;
+
+	if (len < sizeof(*p) ||
+	    verify_newpolicy_info(p))
+		return NULL;
+
+	nr = ((len - sizeof(*p)) / sizeof(*ut));
+	if (nr > XFRM_MAX_DEPTH)
+		return NULL;
+
+	xp = xfrm_policy_alloc(GFP_KERNEL);
+	if (xp == NULL) {
+		*dir = -ENOBUFS;
+		return NULL;
+	}
+
+	copy_from_user_policy(xp, p);
+	copy_templates(xp, ut, nr);
+
+	*dir = p->dir;
+
+	return xp;
+}
+
+static struct xfrm_mgr netlink_mgr = {
+	.id		= "netlink",
+	.notify		= xfrm_send_notify,
+	.acquire	= xfrm_send_acquire,
+	.compile_policy	= xfrm_compile_policy,
+};
+
+static int __init xfrm_user_init(void)
+{
+	printk(KERN_INFO "Initializing IPsec netlink socket\n");
+
+	xfrm_nl = netlink_kernel_create(NETLINK_XFRM, xfrm_netlink_rcv);
+	if (xfrm_nl == NULL)
+		panic("xfrm_user_init: cannot initialize xfrm_nl\n");
+
+
+	xfrm_register_km(&netlink_mgr);
+
+	return 0;
+}
+
+static void __exit xfrm_user_exit(void)
+{
+	xfrm_unregister_km(&netlink_mgr);
+
+	/* XXX need netlink_kernel_destroy XXX */
+}
+
+module_init(xfrm_user_init);
+module_exit(xfrm_user_exit);