[linux-2.6-block.git] / net / ipv4 / ip_options.c

/*
 * INET		An implementation of the TCP/IP protocol suite for the LINUX
 *		operating system.  INET is implemented using the  BSD Socket
 *		interface as the means of communication with the user level.
 *
 *		The options processing module for ip.c
 *
 * Authors:	A.N.Kuznetsov
 *
 */

#define pr_fmt(fmt) "IPv4: " fmt

#include <linux/capability.h>
#include <linux/module.h>
#include <linux/slab.h>
#include <linux/types.h>
#include <asm/uaccess.h>
#include <asm/unaligned.h>
#include <linux/skbuff.h>
#include <linux/ip.h>
#include <linux/icmp.h>
#include <linux/netdevice.h>
#include <linux/rtnetlink.h>
#include <net/sock.h>
#include <net/ip.h>
#include <net/icmp.h>
#include <net/route.h>
#include <net/cipso_ipv4.h>

/*
 * Write options to IP header, record destination address to
 * source route option, address of outgoing interface
 * (we should already know it, so that this  function is allowed be
 * called only after routing decision) and timestamp,
 * if we originate this datagram.
 *
 * daddr is real destination address, next hop is recorded in IP header.
 * saddr is address of outgoing interface.
 */

void ip_options_build(struct sk_buff *skb, struct ip_options *opt,
		      __be32 daddr, struct rtable *rt, int is_frag)
{
	unsigned char *iph = skb_network_header(skb);

	memcpy(&(IPCB(skb)->opt), opt, sizeof(struct ip_options));
	memcpy(iph+sizeof(struct iphdr), opt->__data, opt->optlen);
	opt = &(IPCB(skb)->opt);

	if (opt->srr)
		memcpy(iph+opt->srr+iph[opt->srr+1]-4, &daddr, 4);

	if (!is_frag) {
		if (opt->rr_needaddr)
			ip_rt_get_source(iph+opt->rr+iph[opt->rr+2]-5, skb, rt);
		if (opt->ts_needaddr)
			ip_rt_get_source(iph+opt->ts+iph[opt->ts+2]-9, skb, rt);
		if (opt->ts_needtime) {
			struct timespec tv;
			__be32 midtime;
			getnstimeofday(&tv);
			midtime = htonl((tv.tv_sec % 86400) * MSEC_PER_SEC + tv.tv_nsec / NSEC_PER_MSEC);
			memcpy(iph+opt->ts+iph[opt->ts+2]-5, &midtime, 4);
		}
		return;
	}
	if (opt->rr) {
		memset(iph+opt->rr, IPOPT_NOP, iph[opt->rr+1]);
		opt->rr = 0;
		opt->rr_needaddr = 0;
	}
	if (opt->ts) {
		memset(iph+opt->ts, IPOPT_NOP, iph[opt->ts+1]);
		opt->ts = 0;
		opt->ts_needaddr = opt->ts_needtime = 0;
	}
}

/*
 * Provided (sopt, skb) points to received options,
 * build in dopt compiled option set appropriate for answering.
 * i.e. invert SRR option, copy anothers,
 * and grab room in RR/TS options.
 *
 * NOTE: dopt cannot point to skb.
 */

int ip_options_echo(struct ip_options *dopt, struct sk_buff *skb)
{
	const struct ip_options *sopt;
	unsigned char *sptr, *dptr;
	int soffset, doffset;
	int	optlen;
	__be32	daddr;

	memset(dopt, 0, sizeof(struct ip_options));

	sopt = &(IPCB(skb)->opt);

	if (sopt->optlen == 0)
		return 0;

	sptr = skb_network_header(skb);
	dptr = dopt->__data;

	daddr = skb_rtable(skb)->rt_spec_dst;

	if (sopt->rr) {
		optlen  = sptr[sopt->rr+1];
		soffset = sptr[sopt->rr+2];
		dopt->rr = dopt->optlen + sizeof(struct iphdr);
		memcpy(dptr, sptr+sopt->rr, optlen);
		if (sopt->rr_needaddr && soffset <= optlen) {
			if (soffset + 3 > optlen)
				return -EINVAL;
			dptr[2] = soffset + 4;
			dopt->rr_needaddr = 1;
		}
		dptr += optlen;
		dopt->optlen += optlen;
	}
	if (sopt->ts) {
		optlen = sptr[sopt->ts+1];
		soffset = sptr[sopt->ts+2];
		dopt->ts = dopt->optlen + sizeof(struct iphdr);
		memcpy(dptr, sptr+sopt->ts, optlen);
		if (soffset <= optlen) {
			if (sopt->ts_needaddr) {
				if (soffset + 3 > optlen)
					return -EINVAL;
				dopt->ts_needaddr = 1;
				soffset += 4;
			}
			if (sopt->ts_needtime) {
				if (soffset + 3 > optlen)
					return -EINVAL;
				if ((dptr[3]&0xF) != IPOPT_TS_PRESPEC) {
					dopt->ts_needtime = 1;
					soffset += 4;
				} else {
					dopt->ts_needtime = 0;

					if (soffset + 7 <= optlen) {
						__be32 addr;

						memcpy(&addr, dptr+soffset-1, 4);
						if (inet_addr_type(dev_net(skb_dst(skb)->dev), addr) != RTN_UNICAST) {
							dopt->ts_needtime = 1;
							soffset += 8;
						}
					}
				}
			}
			dptr[2] = soffset;
		}
		dptr += optlen;
		dopt->optlen += optlen;
	}
	if (sopt->srr) {
		unsigned char *start = sptr+sopt->srr;
		__be32 faddr;

		optlen  = start[1];
		soffset = start[2];
		doffset = 0;
		if (soffset > optlen)
			soffset = optlen + 1;
		soffset -= 4;
		if (soffset > 3) {
			memcpy(&faddr, &start[soffset-1], 4);
			for (soffset-=4, doffset=4; soffset > 3; soffset-=4, doffset+=4)
				memcpy(&dptr[doffset-1], &start[soffset-1], 4);
			/*
			 * RFC1812 requires to fix illegal source routes.
			 */
			if (memcmp(&ip_hdr(skb)->saddr,
				   &start[soffset + 3], 4) == 0)
				doffset -= 4;
		}
		if (doffset > 3) {
			memcpy(&start[doffset-1], &daddr, 4);
			dopt->faddr = faddr;
			dptr[0] = start[0];
			dptr[1] = doffset+3;
			dptr[2] = 4;
			dptr += doffset+3;
			dopt->srr = dopt->optlen + sizeof(struct iphdr);
			dopt->optlen += doffset+3;
			dopt->is_strictroute = sopt->is_strictroute;
		}
	}
	if (sopt->cipso) {
		optlen  = sptr[sopt->cipso+1];
		dopt->cipso = dopt->optlen+sizeof(struct iphdr);
		memcpy(dptr, sptr+sopt->cipso, optlen);
		dptr += optlen;
		dopt->optlen += optlen;
	}
	while (dopt->optlen & 3) {
		*dptr++ = IPOPT_END;
		dopt->optlen++;
	}
	return 0;
}

/*
 *	Options "fragmenting", just fill options not
 *	allowed in fragments with NOOPs.
 *	Simple and stupid 8), but the most efficient way.
 */

void ip_options_fragment(struct sk_buff *skb)
{
	unsigned char *optptr = skb_network_header(skb) + sizeof(struct iphdr);
	struct ip_options *opt = &(IPCB(skb)->opt);
	int  l = opt->optlen;
	int  optlen;

	while (l > 0) {
		switch (*optptr) {
		case IPOPT_END:
			return;
		case IPOPT_NOOP:
			l--;
			optptr++;
			continue;
		}
		optlen = optptr[1];
		if (optlen<2 || optlen>l)
		  return;
		if (!IPOPT_COPIED(*optptr))
			memset(optptr, IPOPT_NOOP, optlen);
		l -= optlen;
		optptr += optlen;
	}
	opt->ts = 0;
	opt->rr = 0;
	opt->rr_needaddr = 0;
	opt->ts_needaddr = 0;
	opt->ts_needtime = 0;
}

/*
 * Verify options and fill pointers in struct options.
 * Caller should clear *opt, and set opt->data.
 * If opt == NULL, then skb->data should point to IP header.
 */

int ip_options_compile(struct net *net,
		       struct ip_options *opt, struct sk_buff *skb)
{
	int l;
	unsigned char *iph;
	unsigned char *optptr;
	int optlen;
	unsigned char *pp_ptr = NULL;
	struct rtable *rt = NULL;

	if (skb != NULL) {
		rt = skb_rtable(skb);
		optptr = (unsigned char *)&(ip_hdr(skb)[1]);
	} else
		optptr = opt->__data;
	iph = optptr - sizeof(struct iphdr);

	for (l = opt->optlen; l > 0; ) {
		switch (*optptr) {
		      case IPOPT_END:
			for (optptr++, l--; l>0; optptr++, l--) {
				if (*optptr != IPOPT_END) {
					*optptr = IPOPT_END;
					opt->is_changed = 1;
				}
			}
			goto eol;
		      case IPOPT_NOOP:
			l--;
			optptr++;
			continue;
		}
		optlen = optptr[1];
		if (optlen<2 || optlen>l) {
			pp_ptr = optptr;
			goto error;
		}
		switch (*optptr) {
		      case IPOPT_SSRR:
		      case IPOPT_LSRR:
			if (optlen < 3) {
				pp_ptr = optptr + 1;
				goto error;
			}
			if (optptr[2] < 4) {
				pp_ptr = optptr + 2;
				goto error;
			}
			/* NB: cf RFC-1812 5.2.4.1 */
			if (opt->srr) {
				pp_ptr = optptr;
				goto error;
			}
			if (!skb) {
				if (optptr[2] != 4 || optlen < 7 || ((optlen-3) & 3)) {
					pp_ptr = optptr + 1;
					goto error;
				}
				memcpy(&opt->faddr, &optptr[3], 4);
				if (optlen > 7)
					memmove(&optptr[3], &optptr[7], optlen-7);
			}
			opt->is_strictroute = (optptr[0] == IPOPT_SSRR);
			opt->srr = optptr - iph;
			break;
		      case IPOPT_RR:
			if (opt->rr) {
				pp_ptr = optptr;
				goto error;
			}
			if (optlen < 3) {
				pp_ptr = optptr + 1;
				goto error;
			}
			if (optptr[2] < 4) {
				pp_ptr = optptr + 2;
				goto error;
			}
			if (optptr[2] <= optlen) {
				if (optptr[2]+3 > optlen) {
					pp_ptr = optptr + 2;
					goto error;
				}
				if (rt) {
					memcpy(&optptr[optptr[2]-1], &rt->rt_spec_dst, 4);
					opt->is_changed = 1;
				}
				optptr[2] += 4;
				opt->rr_needaddr = 1;
			}
			opt->rr = optptr - iph;
			break;
		      case IPOPT_TIMESTAMP:
			if (opt->ts) {
				pp_ptr = optptr;
				goto error;
			}
			if (optlen < 4) {
				pp_ptr = optptr + 1;
				goto error;
			}
			if (optptr[2] < 5) {
				pp_ptr = optptr + 2;
				goto error;
			}
			if (optptr[2] <= optlen) {
				unsigned char *timeptr = NULL;
				if (optptr[2]+3 > optptr[1]) {
					pp_ptr = optptr + 2;
					goto error;
				}
				switch (optptr[3]&0xF) {
				      case IPOPT_TS_TSONLY:
					opt->ts = optptr - iph;
					if (skb)
						timeptr = &optptr[optptr[2]-1];
					opt->ts_needtime = 1;
					optptr[2] += 4;
					break;
				      case IPOPT_TS_TSANDADDR:
					if (optptr[2]+7 > optptr[1]) {
						pp_ptr = optptr + 2;
						goto error;
					}
					opt->ts = optptr - iph;
					if (rt)  {
						memcpy(&optptr[optptr[2]-1], &rt->rt_spec_dst, 4);
						timeptr = &optptr[optptr[2]+3];
					}
					opt->ts_needaddr = 1;
					opt->ts_needtime = 1;
					optptr[2] += 8;
					break;
				      case IPOPT_TS_PRESPEC:
					if (optptr[2]+7 > optptr[1]) {
						pp_ptr = optptr + 2;
						goto error;
					}
					opt->ts = optptr - iph;
					{
						__be32 addr;
						memcpy(&addr, &optptr[optptr[2]-1], 4);
						if (inet_addr_type(net, addr) == RTN_UNICAST)
							break;
						if (skb)
							timeptr = &optptr[optptr[2]+3];
					}
					opt->ts_needtime = 1;
					optptr[2] += 8;
					break;
				      default:
					if (!skb && !capable(CAP_NET_RAW)) {
						pp_ptr = optptr + 3;
						goto error;
					}
					break;
				}
				if (timeptr) {
					struct timespec tv;
					u32  midtime;
					getnstimeofday(&tv);
					midtime = (tv.tv_sec % 86400) * MSEC_PER_SEC + tv.tv_nsec / NSEC_PER_MSEC;
					put_unaligned_be32(midtime, timeptr);
					opt->is_changed = 1;
				}
			} else {
				unsigned int overflow = optptr[3]>>4;
				if (overflow == 15) {
					pp_ptr = optptr + 3;
					goto error;
				}
				opt->ts = optptr - iph;
				if (skb) {
					optptr[3] = (optptr[3]&0xF)|((overflow+1)<<4);
					opt->is_changed = 1;
				}
			}
			break;
		      case IPOPT_RA:
			if (optlen < 4) {
				pp_ptr = optptr + 1;
				goto error;
			}
			if (optptr[2] == 0 && optptr[3] == 0)
				opt->router_alert = optptr - iph;
			break;
		      case IPOPT_CIPSO:
			if ((!skb && !capable(CAP_NET_RAW)) || opt->cipso) {
				pp_ptr = optptr;
				goto error;
			}
			opt->cipso = optptr - iph;
			if (cipso_v4_validate(skb, &optptr)) {
				pp_ptr = optptr;
				goto error;
			}
			break;
		      case IPOPT_SEC:
		      case IPOPT_SID:
		      default:
			if (!skb && !capable(CAP_NET_RAW)) {
				pp_ptr = optptr;
				goto error;
			}
			break;
		}
		l -= optlen;
		optptr += optlen;
	}

eol:
	if (!pp_ptr)
		return 0;

error:
	if (skb) {
		icmp_send(skb, ICMP_PARAMETERPROB, 0, htonl((pp_ptr-iph)<<24));
	}
	return -EINVAL;
}
EXPORT_SYMBOL(ip_options_compile);

/*
 *	Undo all the changes done by ip_options_compile().
 */

void ip_options_undo(struct ip_options *opt)
{
	if (opt->srr) {
		unsigned  char *optptr = opt->__data+opt->srr-sizeof(struct  iphdr);
		memmove(optptr+7, optptr+3, optptr[1]-7);
		memcpy(optptr+3, &opt->faddr, 4);
	}
	if (opt->rr_needaddr) {
		unsigned  char *optptr = opt->__data+opt->rr-sizeof(struct  iphdr);
		optptr[2] -= 4;
		memset(&optptr[optptr[2]-1], 0, 4);
	}
	if (opt->ts) {
		unsigned  char *optptr = opt->__data+opt->ts-sizeof(struct  iphdr);
		if (opt->ts_needtime) {
			optptr[2] -= 4;
			memset(&optptr[optptr[2]-1], 0, 4);
			if ((optptr[3]&0xF) == IPOPT_TS_PRESPEC)
				optptr[2] -= 4;
		}
		if (opt->ts_needaddr) {
			optptr[2] -= 4;
			memset(&optptr[optptr[2]-1], 0, 4);
		}
	}
}

static struct ip_options_rcu *ip_options_get_alloc(const int optlen)
{
	return kzalloc(sizeof(struct ip_options_rcu) + ((optlen + 3) & ~3),
		       GFP_KERNEL);
}

static int ip_options_get_finish(struct net *net, struct ip_options_rcu **optp,
				 struct ip_options_rcu *opt, int optlen)
{
	while (optlen & 3)
		opt->opt.__data[optlen++] = IPOPT_END;
	opt->opt.optlen = optlen;
	if (optlen && ip_options_compile(net, &opt->opt, NULL)) {
		kfree(opt);
		return -EINVAL;
	}
	kfree(*optp);
	*optp = opt;
	return 0;
}

int ip_options_get_from_user(struct net *net, struct ip_options_rcu **optp,
			     unsigned char __user *data, int optlen)
{
	struct ip_options_rcu *opt = ip_options_get_alloc(optlen);

	if (!opt)
		return -ENOMEM;
	if (optlen && copy_from_user(opt->opt.__data, data, optlen)) {
		kfree(opt);
		return -EFAULT;
	}
	return ip_options_get_finish(net, optp, opt, optlen);
}

int ip_options_get(struct net *net, struct ip_options_rcu **optp,
		   unsigned char *data, int optlen)
{
	struct ip_options_rcu *opt = ip_options_get_alloc(optlen);

	if (!opt)
		return -ENOMEM;
	if (optlen)
		memcpy(opt->opt.__data, data, optlen);
	return ip_options_get_finish(net, optp, opt, optlen);
}

void ip_forward_options(struct sk_buff *skb)
{
	struct   ip_options *opt	= &(IPCB(skb)->opt);
	unsigned char *optptr;
	struct rtable *rt = skb_rtable(skb);
	unsigned char *raw = skb_network_header(skb);

	if (opt->rr_needaddr) {
		optptr = (unsigned char *)raw + opt->rr;
		ip_rt_get_source(&optptr[optptr[2]-5], skb, rt);
		opt->is_changed = 1;
	}
	if (opt->srr_is_hit) {
		int srrptr, srrspace;

		optptr = raw + opt->srr;

		for ( srrptr=optptr[2], srrspace = optptr[1];
		     srrptr <= srrspace;
		     srrptr += 4
		     ) {
			if (srrptr + 3 > srrspace)
				break;
			if (memcmp(&opt->nexthop, &optptr[srrptr-1], 4) == 0)
				break;
		}
		if (srrptr + 3 <= srrspace) {
			opt->is_changed = 1;
			ip_hdr(skb)->daddr = opt->nexthop;
			ip_rt_get_source(&optptr[srrptr-1], skb, rt);
			optptr[2] = srrptr+4;
		} else if (net_ratelimit())
			pr_crit("%s(): Argh! Destination lost!\n", __func__);
		if (opt->ts_needaddr) {
			optptr = raw + opt->ts;
			ip_rt_get_source(&optptr[optptr[2]-9], skb, rt);
			opt->is_changed = 1;
		}
	}
	if (opt->is_changed) {
		opt->is_changed = 0;
		ip_send_check(ip_hdr(skb));
	}
}

int ip_options_rcv_srr(struct sk_buff *skb)
{
	struct ip_options *opt = &(IPCB(skb)->opt);
	int srrspace, srrptr;
	__be32 nexthop;
	struct iphdr *iph = ip_hdr(skb);
	unsigned char *optptr = skb_network_header(skb) + opt->srr;
	struct rtable *rt = skb_rtable(skb);
	struct rtable *rt2;
	unsigned long orefdst;
	int err;

	if (!rt)
		return 0;

	if (skb->pkt_type != PACKET_HOST)
		return -EINVAL;
	if (rt->rt_type == RTN_UNICAST) {
		if (!opt->is_strictroute)
			return 0;
		icmp_send(skb, ICMP_PARAMETERPROB, 0, htonl(16<<24));
		return -EINVAL;
	}
	if (rt->rt_type != RTN_LOCAL)
		return -EINVAL;

	for (srrptr=optptr[2], srrspace = optptr[1]; srrptr <= srrspace; srrptr += 4) {
		if (srrptr + 3 > srrspace) {
			icmp_send(skb, ICMP_PARAMETERPROB, 0, htonl((opt->srr+2)<<24));
			return -EINVAL;
		}
		memcpy(&nexthop, &optptr[srrptr-1], 4);

		orefdst = skb->_skb_refdst;
		skb_dst_set(skb, NULL);
		err = ip_route_input(skb, nexthop, iph->saddr, iph->tos, skb->dev);
		rt2 = skb_rtable(skb);
		if (err || (rt2->rt_type != RTN_UNICAST && rt2->rt_type != RTN_LOCAL)) {
			skb_dst_drop(skb);
			skb->_skb_refdst = orefdst;
			return -EINVAL;
		}
		refdst_drop(orefdst);
		if (rt2->rt_type != RTN_LOCAL)
			break;
		/* Superfast 8) loopback forward */
		iph->daddr = nexthop;
		opt->is_changed = 1;
	}
	if (srrptr <= srrspace) {
		opt->srr_is_hit = 1;
		opt->nexthop = nexthop;
		opt->is_changed = 1;
	}
	return 0;
}
EXPORT_SYMBOL(ip_options_rcv_srr);
Commit	Line	Data
1da177e4 LT	1	/*
	2	* INET An implementation of the TCP/IP protocol suite for the LINUX
	3	* operating system. INET is implemented using the BSD Socket
	4	* interface as the means of communication with the user level.
	5	*
	6	* The options processing module for ip.c
	7	*
1da177e4	8	* Authors: A.N.Kuznetsov
e905a9ed	9	*
1da177e4 LT	10	*/
1da177e4 LT	11
afd46503 JP	12	#define pr_fmt(fmt) "IPv4: " fmt
afd46503 JP	13
4fc268d2	14	#include <linux/capability.h>
1da177e4	15	#include <linux/module.h>
5a0e3ad6	16	#include <linux/slab.h>
1da177e4 LT	17	#include <linux/types.h>
1da177e4 LT	18	#include <asm/uaccess.h>
48bdf072	19	#include <asm/unaligned.h>
1da177e4 LT	20	#include <linux/skbuff.h>
	21	#include <linux/ip.h>
	22	#include <linux/icmp.h>
	23	#include <linux/netdevice.h>
	24	#include <linux/rtnetlink.h>
	25	#include <net/sock.h>
	26	#include <net/ip.h>
	27	#include <net/icmp.h>
14c85021	28	#include <net/route.h>
11a03f78	29	#include <net/cipso_ipv4.h>
1da177e4	30
e905a9ed	31	/*
1da177e4 LT	32	* Write options to IP header, record destination address to
	33	* source route option, address of outgoing interface
	34	* (we should already know it, so that this function is allowed be
	35	* called only after routing decision) and timestamp,
	36	* if we originate this datagram.
	37	*
	38	* daddr is real destination address, next hop is recorded in IP header.
	39	* saddr is address of outgoing interface.
	40	*/
	41
f6d8bd05	42	void ip_options_build(struct sk_buff skb, struct ip_options opt,
8e36360a	43	__be32 daddr, struct rtable *rt, int is_frag)
1da177e4	44	{
d56f90a7	45	unsigned char *iph = skb_network_header(skb);
1da177e4 LT	46
	47	memcpy(&(IPCB(skb)->opt), opt, sizeof(struct ip_options));
	48	memcpy(iph+sizeof(struct iphdr), opt->__data, opt->optlen);
	49	opt = &(IPCB(skb)->opt);
1da177e4 LT	50
	51	if (opt->srr)
	52	memcpy(iph+opt->srr+iph[opt->srr+1]-4, &daddr, 4);
	53
	54	if (!is_frag) {
	55	if (opt->rr_needaddr)
8e36360a	56	ip_rt_get_source(iph+opt->rr+iph[opt->rr+2]-5, skb, rt);
1da177e4	57	if (opt->ts_needaddr)
8e36360a	58	ip_rt_get_source(iph+opt->ts+iph[opt->ts+2]-9, skb, rt);
1da177e4	59	if (opt->ts_needtime) {
f25c3d61	60	struct timespec tv;
e25d2ca6	61	__be32 midtime;
f25c3d61 YH	62	getnstimeofday(&tv);
f25c3d61 YH	63	midtime = htonl((tv.tv_sec % 86400) * MSEC_PER_SEC + tv.tv_nsec / NSEC_PER_MSEC);
1da177e4 LT	64	memcpy(iph+opt->ts+iph[opt->ts+2]-5, &midtime, 4);
	65	}
	66	return;
	67	}
	68	if (opt->rr) {
	69	memset(iph+opt->rr, IPOPT_NOP, iph[opt->rr+1]);
	70	opt->rr = 0;
	71	opt->rr_needaddr = 0;
	72	}
	73	if (opt->ts) {
	74	memset(iph+opt->ts, IPOPT_NOP, iph[opt->ts+1]);
	75	opt->ts = 0;
	76	opt->ts_needaddr = opt->ts_needtime = 0;
	77	}
	78	}
	79
e905a9ed	80	/*
1da177e4 LT	81	* Provided (sopt, skb) points to received options,
	82	* build in dopt compiled option set appropriate for answering.
	83	* i.e. invert SRR option, copy anothers,
	84	* and grab room in RR/TS options.
	85	*
	86	* NOTE: dopt cannot point to skb.
	87	*/
	88
f6d8bd05	89	int ip_options_echo(struct ip_options dopt, struct sk_buff skb)
1da177e4	90	{
f6d8bd05	91	const struct ip_options *sopt;
1da177e4 LT	92	unsigned char sptr, dptr;
	93	int soffset, doffset;
	94	int optlen;
e25d2ca6	95	__be32 daddr;
1da177e4 LT	96
	97	memset(dopt, 0, sizeof(struct ip_options));
	98
1da177e4 LT	99	sopt = &(IPCB(skb)->opt);
1da177e4 LT	100
f6d8bd05	101	if (sopt->optlen == 0)
1da177e4	102	return 0;
1da177e4	103
d56f90a7	104	sptr = skb_network_header(skb);
1da177e4 LT	105	dptr = dopt->__data;
1da177e4 LT	106
511c3f92	107	daddr = skb_rtable(skb)->rt_spec_dst;
1da177e4 LT	108
	109	if (sopt->rr) {
	110	optlen = sptr[sopt->rr+1];
	111	soffset = sptr[sopt->rr+2];
	112	dopt->rr = dopt->optlen + sizeof(struct iphdr);
	113	memcpy(dptr, sptr+sopt->rr, optlen);
	114	if (sopt->rr_needaddr && soffset <= optlen) {
	115	if (soffset + 3 > optlen)
	116	return -EINVAL;
	117	dptr[2] = soffset + 4;
	118	dopt->rr_needaddr = 1;
	119	}
	120	dptr += optlen;
	121	dopt->optlen += optlen;
	122	}
	123	if (sopt->ts) {
	124	optlen = sptr[sopt->ts+1];
	125	soffset = sptr[sopt->ts+2];
	126	dopt->ts = dopt->optlen + sizeof(struct iphdr);
	127	memcpy(dptr, sptr+sopt->ts, optlen);
	128	if (soffset <= optlen) {
	129	if (sopt->ts_needaddr) {
	130	if (soffset + 3 > optlen)
	131	return -EINVAL;
	132	dopt->ts_needaddr = 1;
	133	soffset += 4;
	134	}
	135	if (sopt->ts_needtime) {
	136	if (soffset + 3 > optlen)
	137	return -EINVAL;
	138	if ((dptr[3]&0xF) != IPOPT_TS_PRESPEC) {
	139	dopt->ts_needtime = 1;
	140	soffset += 4;
	141	} else {
	142	dopt->ts_needtime = 0;
	143
8628bd8a	144	if (soffset + 7 <= optlen) {
fd683222	145	__be32 addr;
1da177e4	146
8628bd8a JL	147	memcpy(&addr, dptr+soffset-1, 4);
8628bd8a JL	148	if (inet_addr_type(dev_net(skb_dst(skb)->dev), addr) != RTN_UNICAST) {
1da177e4 LT	149	dopt->ts_needtime = 1;
	150	soffset += 8;
	151	}
	152	}
	153	}
	154	}
	155	dptr[2] = soffset;
	156	}
	157	dptr += optlen;
	158	dopt->optlen += optlen;
	159	}
	160	if (sopt->srr) {
f6d8bd05	161	unsigned char *start = sptr+sopt->srr;
3ca3c68e	162	__be32 faddr;
1da177e4 LT	163
	164	optlen = start[1];
	165	soffset = start[2];
	166	doffset = 0;
	167	if (soffset > optlen)
	168	soffset = optlen + 1;
	169	soffset -= 4;
	170	if (soffset > 3) {
	171	memcpy(&faddr, &start[soffset-1], 4);
	172	for (soffset-=4, doffset=4; soffset > 3; soffset-=4, doffset+=4)
	173	memcpy(&dptr[doffset-1], &start[soffset-1], 4);
	174	/*
	175	* RFC1812 requires to fix illegal source routes.
	176	*/
eddc9ec5 ACM	177	if (memcmp(&ip_hdr(skb)->saddr,
eddc9ec5 ACM	178	&start[soffset + 3], 4) == 0)
1da177e4 LT	179	doffset -= 4;
	180	}
	181	if (doffset > 3) {
	182	memcpy(&start[doffset-1], &daddr, 4);
	183	dopt->faddr = faddr;
	184	dptr[0] = start[0];
	185	dptr[1] = doffset+3;
	186	dptr[2] = 4;
	187	dptr += doffset+3;
	188	dopt->srr = dopt->optlen + sizeof(struct iphdr);
	189	dopt->optlen += doffset+3;
	190	dopt->is_strictroute = sopt->is_strictroute;
	191	}
	192	}
11a03f78 PM	193	if (sopt->cipso) {
	194	optlen = sptr[sopt->cipso+1];
	195	dopt->cipso = dopt->optlen+sizeof(struct iphdr);
	196	memcpy(dptr, sptr+sopt->cipso, optlen);
	197	dptr += optlen;
	198	dopt->optlen += optlen;
	199	}
1da177e4 LT	200	while (dopt->optlen & 3) {
	201	*dptr++ = IPOPT_END;
	202	dopt->optlen++;
	203	}
	204	return 0;
	205	}
	206
	207	/*
	208	* Options "fragmenting", just fill options not
	209	* allowed in fragments with NOOPs.
	210	* Simple and stupid 8), but the most efficient way.
	211	*/
	212
5e73ea1a	213	void ip_options_fragment(struct sk_buff *skb)
1da177e4	214	{
d56f90a7	215	unsigned char *optptr = skb_network_header(skb) + sizeof(struct iphdr);
5e73ea1a	216	struct ip_options *opt = &(IPCB(skb)->opt);
1da177e4 LT	217	int l = opt->optlen;
	218	int optlen;
	219
	220	while (l > 0) {
	221	switch (*optptr) {
	222	case IPOPT_END:
	223	return;
	224	case IPOPT_NOOP:
	225	l--;
	226	optptr++;
	227	continue;
	228	}
	229	optlen = optptr[1];
	230	if (optlen<2 \|\| optlen>l)
	231	return;
	232	if (!IPOPT_COPIED(*optptr))
	233	memset(optptr, IPOPT_NOOP, optlen);
	234	l -= optlen;
	235	optptr += optlen;
	236	}
	237	opt->ts = 0;
	238	opt->rr = 0;
	239	opt->rr_needaddr = 0;
	240	opt->ts_needaddr = 0;
	241	opt->ts_needtime = 0;
1da177e4 LT	242	}
	243
	244	/*
	245	* Verify options and fill pointers in struct options.
	246	* Caller should clear *opt, and set opt->data.
	247	* If opt == NULL, then skb->data should point to IP header.
	248	*/
	249
0e6bd4a1	250	int ip_options_compile(struct net *net,
5e73ea1a	251	struct ip_options opt, struct sk_buff skb)
1da177e4 LT	252	{
1da177e4 LT	253	int l;
5e73ea1a DB	254	unsigned char *iph;
5e73ea1a DB	255	unsigned char *optptr;
1da177e4	256	int optlen;
5e73ea1a	257	unsigned char *pp_ptr = NULL;
22aba383	258	struct rtable *rt = NULL;
1da177e4	259
22aba383	260	if (skb != NULL) {
511c3f92	261	rt = skb_rtable(skb);
22aba383 DL	262	optptr = (unsigned char *)&(ip_hdr(skb)[1]);
22aba383 DL	263	} else
10fe7d85	264	optptr = opt->__data;
22aba383	265	iph = optptr - sizeof(struct iphdr);
1da177e4 LT	266
	267	for (l = opt->optlen; l > 0; ) {
	268	switch (*optptr) {
	269	case IPOPT_END:
	270	for (optptr++, l--; l>0; optptr++, l--) {
	271	if (*optptr != IPOPT_END) {
	272	*optptr = IPOPT_END;
	273	opt->is_changed = 1;
	274	}
	275	}
	276	goto eol;
	277	case IPOPT_NOOP:
	278	l--;
	279	optptr++;
	280	continue;
	281	}
	282	optlen = optptr[1];
	283	if (optlen<2 \|\| optlen>l) {
	284	pp_ptr = optptr;
	285	goto error;
	286	}
	287	switch (*optptr) {
	288	case IPOPT_SSRR:
	289	case IPOPT_LSRR:
	290	if (optlen < 3) {
	291	pp_ptr = optptr + 1;
	292	goto error;
	293	}
	294	if (optptr[2] < 4) {
	295	pp_ptr = optptr + 2;
	296	goto error;
	297	}
	298	/* NB: cf RFC-1812 5.2.4.1 */
	299	if (opt->srr) {
	300	pp_ptr = optptr;
	301	goto error;
	302	}
	303	if (!skb) {
	304	if (optptr[2] != 4 \|\| optlen < 7 \|\| ((optlen-3) & 3)) {
	305	pp_ptr = optptr + 1;
	306	goto error;
	307	}
	308	memcpy(&opt->faddr, &optptr[3], 4);
	309	if (optlen > 7)
	310	memmove(&optptr[3], &optptr[7], optlen-7);
	311	}
	312	opt->is_strictroute = (optptr[0] == IPOPT_SSRR);
	313	opt->srr = optptr - iph;
	314	break;
	315	case IPOPT_RR:
	316	if (opt->rr) {
	317	pp_ptr = optptr;
	318	goto error;
	319	}
	320	if (optlen < 3) {
	321	pp_ptr = optptr + 1;
	322	goto error;
	323	}
	324	if (optptr[2] < 4) {
	325	pp_ptr = optptr + 2;
	326	goto error;
	327	}
	328	if (optptr[2] <= optlen) {
	329	if (optptr[2]+3 > optlen) {
330	pp_ptr = optptr + 2;
331	goto error;
332	}
c65353da	333	if (rt) {
1da177e4 LT	334	memcpy(&optptr[optptr[2]-1], &rt->rt_spec_dst, 4);
	335	opt->is_changed = 1;
	336	}
	337	optptr[2] += 4;
	338	opt->rr_needaddr = 1;
	339	}
	340	opt->rr = optptr - iph;
	341	break;
	342	case IPOPT_TIMESTAMP:
	343	if (opt->ts) {
	344	pp_ptr = optptr;
	345	goto error;
	346	}
	347	if (optlen < 4) {
	348	pp_ptr = optptr + 1;
	349	goto error;
	350	}
	351	if (optptr[2] < 5) {
	352	pp_ptr = optptr + 2;
	353	goto error;
	354	}
	355	if (optptr[2] <= optlen) {
48bdf072	356	unsigned char *timeptr = NULL;
1da177e4 LT	357	if (optptr[2]+3 > optptr[1]) {
	358	pp_ptr = optptr + 2;
	359	goto error;
	360	}
	361	switch (optptr[3]&0xF) {
	362	case IPOPT_TS_TSONLY:
	363	opt->ts = optptr - iph;
e905a9ed	364	if (skb)
48bdf072	365	timeptr = &optptr[optptr[2]-1];
1da177e4 LT	366	opt->ts_needtime = 1;
	367	optptr[2] += 4;
	368	break;
	369	case IPOPT_TS_TSANDADDR:
	370	if (optptr[2]+7 > optptr[1]) {
	371	pp_ptr = optptr + 2;
	372	goto error;
	373	}
	374	opt->ts = optptr - iph;
c65353da	375	if (rt) {
1da177e4	376	memcpy(&optptr[optptr[2]-1], &rt->rt_spec_dst, 4);
48bdf072	377	timeptr = &optptr[optptr[2]+3];
1da177e4 LT	378	}
	379	opt->ts_needaddr = 1;
	380	opt->ts_needtime = 1;
	381	optptr[2] += 8;
	382	break;
	383	case IPOPT_TS_PRESPEC:
	384	if (optptr[2]+7 > optptr[1]) {
	385	pp_ptr = optptr + 2;
	386	goto error;
	387	}
	388	opt->ts = optptr - iph;
	389	{
fd683222	390	__be32 addr;
1da177e4	391	memcpy(&addr, &optptr[optptr[2]-1], 4);
0e6bd4a1	392	if (inet_addr_type(net, addr) == RTN_UNICAST)
1da177e4 LT	393	break;
1da177e4 LT	394	if (skb)
48bdf072	395	timeptr = &optptr[optptr[2]+3];
1da177e4 LT	396	}
	397	opt->ts_needtime = 1;
	398	optptr[2] += 8;
	399	break;
	400	default:
	401	if (!skb && !capable(CAP_NET_RAW)) {
	402	pp_ptr = optptr + 3;
	403	goto error;
	404	}
	405	break;
	406	}
	407	if (timeptr) {
f25c3d61	408	struct timespec tv;
48bdf072	409	u32 midtime;
f25c3d61	410	getnstimeofday(&tv);
48bdf072 CM	411	midtime = (tv.tv_sec % 86400) * MSEC_PER_SEC + tv.tv_nsec / NSEC_PER_MSEC;
48bdf072 CM	412	put_unaligned_be32(midtime, timeptr);
1da177e4 LT	413	opt->is_changed = 1;
	414	}
	415	} else {
95c96174	416	unsigned int overflow = optptr[3]>>4;
1da177e4 LT	417	if (overflow == 15) {
	418	pp_ptr = optptr + 3;
	419	goto error;
	420	}
	421	opt->ts = optptr - iph;
	422	if (skb) {
	423	optptr[3] = (optptr[3]&0xF)\|((overflow+1)<<4);
	424	opt->is_changed = 1;
	425	}
	426	}
	427	break;
	428	case IPOPT_RA:
	429	if (optlen < 4) {
	430	pp_ptr = optptr + 1;
	431	goto error;
	432	}
	433	if (optptr[2] == 0 && optptr[3] == 0)
	434	opt->router_alert = optptr - iph;
	435	break;
11a03f78	436	case IPOPT_CIPSO:
f8687afe	437	if ((!skb && !capable(CAP_NET_RAW)) \|\| opt->cipso) {
11a03f78 PM	438	pp_ptr = optptr;
	439	goto error;
	440	}
	441	opt->cipso = optptr - iph;
15c45f7b	442	if (cipso_v4_validate(skb, &optptr)) {
11a03f78 PM	443	pp_ptr = optptr;
	444	goto error;
	445	}
	446	break;
1da177e4 LT	447	case IPOPT_SEC:
	448	case IPOPT_SID:
	449	default:
	450	if (!skb && !capable(CAP_NET_RAW)) {
	451	pp_ptr = optptr;
	452	goto error;
	453	}
	454	break;
	455	}
	456	l -= optlen;
	457	optptr += optlen;
	458	}
	459
	460	eol:
	461	if (!pp_ptr)
	462	return 0;
	463
	464	error:
	465	if (skb) {
	466	icmp_send(skb, ICMP_PARAMETERPROB, 0, htonl((pp_ptr-iph)<<24));
	467	}
	468	return -EINVAL;
	469	}
462fb2af	470	EXPORT_SYMBOL(ip_options_compile);
1da177e4 LT	471
	472	/*
	473	* Undo all the changes done by ip_options_compile().
	474	*/
	475
5e73ea1a	476	void ip_options_undo(struct ip_options *opt)
1da177e4 LT	477	{
1da177e4 LT	478	if (opt->srr) {
5e73ea1a	479	unsigned char *optptr = opt->__data+opt->srr-sizeof(struct iphdr);
1da177e4 LT	480	memmove(optptr+7, optptr+3, optptr[1]-7);
	481	memcpy(optptr+3, &opt->faddr, 4);
	482	}
	483	if (opt->rr_needaddr) {
5e73ea1a	484	unsigned char *optptr = opt->__data+opt->rr-sizeof(struct iphdr);
1da177e4 LT	485	optptr[2] -= 4;
	486	memset(&optptr[optptr[2]-1], 0, 4);
	487	}
	488	if (opt->ts) {
5e73ea1a	489	unsigned char *optptr = opt->__data+opt->ts-sizeof(struct iphdr);
1da177e4 LT	490	if (opt->ts_needtime) {
	491	optptr[2] -= 4;
	492	memset(&optptr[optptr[2]-1], 0, 4);
	493	if ((optptr[3]&0xF) == IPOPT_TS_PRESPEC)
	494	optptr[2] -= 4;
	495	}
	496	if (opt->ts_needaddr) {
	497	optptr[2] -= 4;
	498	memset(&optptr[optptr[2]-1], 0, 4);
	499	}
	500	}
	501	}
	502
f6d8bd05	503	static struct ip_options_rcu *ip_options_get_alloc(const int optlen)
1da177e4	504	{
f6d8bd05	505	return kzalloc(sizeof(struct ip_options_rcu) + ((optlen + 3) & ~3),
37640703	506	GFP_KERNEL);
4c6ea29d	507	}
1da177e4	508
f6d8bd05 ED	509	static int ip_options_get_finish(struct net net, struct ip_options_rcu *optp,
f6d8bd05 ED	510	struct ip_options_rcu *opt, int optlen)
4c6ea29d	511	{
1da177e4	512	while (optlen & 3)
f6d8bd05 ED	513	opt->opt.__data[optlen++] = IPOPT_END;
	514	opt->opt.optlen = optlen;
	515	if (optlen && ip_options_compile(net, &opt->opt, NULL)) {
1da177e4 LT	516	kfree(opt);
	517	return -EINVAL;
	518	}
a51482bd	519	kfree(*optp);
1da177e4 LT	520	*optp = opt;
	521	return 0;
	522	}
	523
f6d8bd05	524	int ip_options_get_from_user(struct net net, struct ip_options_rcu *optp,
f2c4802b	525	unsigned char __user *data, int optlen)
4c6ea29d	526	{
f6d8bd05	527	struct ip_options_rcu *opt = ip_options_get_alloc(optlen);
4c6ea29d ACM	528
	529	if (!opt)
	530	return -ENOMEM;
f6d8bd05	531	if (optlen && copy_from_user(opt->opt.__data, data, optlen)) {
4c6ea29d ACM	532	kfree(opt);
	533	return -EFAULT;
	534	}
f2c4802b	535	return ip_options_get_finish(net, optp, opt, optlen);
4c6ea29d ACM	536	}
4c6ea29d ACM	537
f6d8bd05	538	int ip_options_get(struct net net, struct ip_options_rcu *optp,
f2c4802b	539	unsigned char *data, int optlen)
4c6ea29d	540	{
f6d8bd05	541	struct ip_options_rcu *opt = ip_options_get_alloc(optlen);
4c6ea29d ACM	542
	543	if (!opt)
	544	return -ENOMEM;
	545	if (optlen)
f6d8bd05	546	memcpy(opt->opt.__data, data, optlen);
f2c4802b	547	return ip_options_get_finish(net, optp, opt, optlen);
4c6ea29d ACM	548	}
4c6ea29d ACM	549
1da177e4 LT	550	void ip_forward_options(struct sk_buff *skb)
1da177e4 LT	551	{
5e73ea1a DB	552	struct ip_options *opt = &(IPCB(skb)->opt);
5e73ea1a DB	553	unsigned char *optptr;
511c3f92	554	struct rtable *rt = skb_rtable(skb);
d56f90a7	555	unsigned char *raw = skb_network_header(skb);
1da177e4 LT	556
	557	if (opt->rr_needaddr) {
	558	optptr = (unsigned char *)raw + opt->rr;
8e36360a	559	ip_rt_get_source(&optptr[optptr[2]-5], skb, rt);
1da177e4 LT	560	opt->is_changed = 1;
	561	}
	562	if (opt->srr_is_hit) {
	563	int srrptr, srrspace;
	564
	565	optptr = raw + opt->srr;
	566
	567	for ( srrptr=optptr[2], srrspace = optptr[1];
	568	srrptr <= srrspace;
	569	srrptr += 4
	570	) {
	571	if (srrptr + 3 > srrspace)
	572	break;
ac8a4810	573	if (memcmp(&opt->nexthop, &optptr[srrptr-1], 4) == 0)
1da177e4 LT	574	break;
	575	}
	576	if (srrptr + 3 <= srrspace) {
	577	opt->is_changed = 1;
ac8a4810	578	ip_hdr(skb)->daddr = opt->nexthop;
5dc7883f	579	ip_rt_get_source(&optptr[srrptr-1], skb, rt);
1da177e4 LT	580	optptr[2] = srrptr+4;
1da177e4 LT	581	} else if (net_ratelimit())
058bd4d2	582	pr_crit("%s(): Argh! Destination lost!\n", __func__);
1da177e4 LT	583	if (opt->ts_needaddr) {
1da177e4 LT	584	optptr = raw + opt->ts;
8e36360a	585	ip_rt_get_source(&optptr[optptr[2]-9], skb, rt);
1da177e4 LT	586	opt->is_changed = 1;
	587	}
	588	}
	589	if (opt->is_changed) {
	590	opt->is_changed = 0;
eddc9ec5	591	ip_send_check(ip_hdr(skb));
1da177e4 LT	592	}
	593	}
	594
	595	int ip_options_rcv_srr(struct sk_buff *skb)
	596	{
	597	struct ip_options *opt = &(IPCB(skb)->opt);
	598	int srrspace, srrptr;
9e12bb22	599	__be32 nexthop;
eddc9ec5	600	struct iphdr *iph = ip_hdr(skb);
d56f90a7	601	unsigned char *optptr = skb_network_header(skb) + opt->srr;
511c3f92	602	struct rtable *rt = skb_rtable(skb);
1da177e4	603	struct rtable *rt2;
7fee226a	604	unsigned long orefdst;
1da177e4 LT	605	int err;
1da177e4 LT	606
10949550	607	if (!rt)
1da177e4 LT	608	return 0;
	609
	610	if (skb->pkt_type != PACKET_HOST)
	611	return -EINVAL;
	612	if (rt->rt_type == RTN_UNICAST) {
	613	if (!opt->is_strictroute)
	614	return 0;
	615	icmp_send(skb, ICMP_PARAMETERPROB, 0, htonl(16<<24));
	616	return -EINVAL;
	617	}
	618	if (rt->rt_type != RTN_LOCAL)
	619	return -EINVAL;
	620
	621	for (srrptr=optptr[2], srrspace = optptr[1]; srrptr <= srrspace; srrptr += 4) {
	622	if (srrptr + 3 > srrspace) {
	623	icmp_send(skb, ICMP_PARAMETERPROB, 0, htonl((opt->srr+2)<<24));
	624	return -EINVAL;
	625	}
	626	memcpy(&nexthop, &optptr[srrptr-1], 4);
	627
7fee226a	628	orefdst = skb->_skb_refdst;
adf30907	629	skb_dst_set(skb, NULL);
1da177e4	630	err = ip_route_input(skb, nexthop, iph->saddr, iph->tos, skb->dev);
511c3f92	631	rt2 = skb_rtable(skb);
1da177e4	632	if (err \|\| (rt2->rt_type != RTN_UNICAST && rt2->rt_type != RTN_LOCAL)) {
7fee226a ED	633	skb_dst_drop(skb);
7fee226a ED	634	skb->_skb_refdst = orefdst;
1da177e4 LT	635	return -EINVAL;
1da177e4 LT	636	}
7fee226a	637	refdst_drop(orefdst);
1da177e4 LT	638	if (rt2->rt_type != RTN_LOCAL)
	639	break;
	640	/* Superfast 8) loopback forward */
c30883bd	641	iph->daddr = nexthop;
1da177e4 LT	642	opt->is_changed = 1;
	643	}
	644	if (srrptr <= srrspace) {
	645	opt->srr_is_hit = 1;
ac8a4810	646	opt->nexthop = nexthop;
1da177e4 LT	647	opt->is_changed = 1;
	648	}
	649	return 0;
	650	}
462fb2af	651	EXPORT_SYMBOL(ip_options_rcv_srr);