[linux-2.6-block.git] / net / ipv4 / ip_forward.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 IP forwarding functionality.
 *
 * Authors:	see ip.c
 *
 * Fixes:
 *		Many		:	Split from ip.c , see ip_input.c for
 *					history.
 *		Dave Gregorich	:	NULL ip_rt_put fix for multicast
 *					routing.
 *		Jos Vos		:	Add call_out_firewall before sending,
 *					use output device for accounting.
 *		Jos Vos		:	Call forward firewall after routing
 *					(always use output device).
 *		Mike McLagan	:	Routing by source
 */

#include <linux/types.h>
#include <linux/mm.h>
#include <linux/skbuff.h>
#include <linux/ip.h>
#include <linux/icmp.h>
#include <linux/netdevice.h>
#include <linux/slab.h>
#include <net/sock.h>
#include <net/ip.h>
#include <net/tcp.h>
#include <net/udp.h>
#include <net/icmp.h>
#include <linux/tcp.h>
#include <linux/udp.h>
#include <linux/netfilter_ipv4.h>
#include <net/checksum.h>
#include <linux/route.h>
#include <net/route.h>
#include <net/xfrm.h>

static bool ip_exceeds_mtu(const struct sk_buff *skb, unsigned int mtu)
{
	if (skb->len <= mtu)
		return false;

	if (unlikely((ip_hdr(skb)->frag_off & htons(IP_DF)) == 0))
		return false;

	/* original fragment exceeds mtu and DF is set */
	if (unlikely(IPCB(skb)->frag_max_size > mtu))
		return true;

	if (skb->ignore_df)
		return false;

	if (skb_is_gso(skb) && skb_gso_network_seglen(skb) <= mtu)
		return false;

	return true;
}


static int ip_forward_finish(struct sock *sk, struct sk_buff *skb)
{
	struct ip_options *opt	= &(IPCB(skb)->opt);

	IP_INC_STATS_BH(dev_net(skb_dst(skb)->dev), IPSTATS_MIB_OUTFORWDATAGRAMS);
	IP_ADD_STATS_BH(dev_net(skb_dst(skb)->dev), IPSTATS_MIB_OUTOCTETS, skb->len);

	if (unlikely(opt->optlen))
		ip_forward_options(skb);

	skb_sender_cpu_clear(skb);
	return dst_output_sk(sk, skb);
}

int ip_forward(struct sk_buff *skb)
{
	u32 mtu;
	struct iphdr *iph;	/* Our header */
	struct rtable *rt;	/* Route we use */
	struct ip_options *opt	= &(IPCB(skb)->opt);

	/* that should never happen */
	if (skb->pkt_type != PACKET_HOST)
		goto drop;

	if (unlikely(skb->sk))
		goto drop;

	if (skb_warn_if_lro(skb))
		goto drop;

	if (!xfrm4_policy_check(NULL, XFRM_POLICY_FWD, skb))
		goto drop;

	if (IPCB(skb)->opt.router_alert && ip_call_ra_chain(skb))
		return NET_RX_SUCCESS;

	skb_forward_csum(skb);

	/*
	 *	According to the RFC, we must first decrease the TTL field. If
	 *	that reaches zero, we must reply an ICMP control message telling
	 *	that the packet's lifetime expired.
	 */
	if (ip_hdr(skb)->ttl <= 1)
		goto too_many_hops;

	if (!xfrm4_route_forward(skb))
		goto drop;

	rt = skb_rtable(skb);

	if (opt->is_strictroute && rt->rt_uses_gateway)
		goto sr_failed;

	IPCB(skb)->flags |= IPSKB_FORWARDED;
	mtu = ip_dst_mtu_maybe_forward(&rt->dst, true);
	if (ip_exceeds_mtu(skb, mtu)) {
		IP_INC_STATS(dev_net(rt->dst.dev), IPSTATS_MIB_FRAGFAILS);
		icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED,
			  htonl(mtu));
		goto drop;
	}

	/* We are about to mangle packet. Copy it! */
	if (skb_cow(skb, LL_RESERVED_SPACE(rt->dst.dev)+rt->dst.header_len))
		goto drop;
	iph = ip_hdr(skb);

	/* Decrease ttl after skb cow done */
	ip_decrease_ttl(iph);

	/*
	 *	We now generate an ICMP HOST REDIRECT giving the route
	 *	we calculated.
	 */
	if (IPCB(skb)->flags & IPSKB_DOREDIRECT && !opt->srr &&
	    !skb_sec_path(skb))
		ip_rt_send_redirect(skb);

	skb->priority = rt_tos2priority(iph->tos);

	return NF_HOOK(NFPROTO_IPV4, NF_INET_FORWARD, NULL, skb,
		       skb->dev, rt->dst.dev, ip_forward_finish);

sr_failed:
	/*
	 *	Strict routing permits no gatewaying
	 */
	 icmp_send(skb, ICMP_DEST_UNREACH, ICMP_SR_FAILED, 0);
	 goto drop;

too_many_hops:
	/* Tell the sender its packet died... */
	IP_INC_STATS_BH(dev_net(skb_dst(skb)->dev), IPSTATS_MIB_INHDRERRORS);
	icmp_send(skb, ICMP_TIME_EXCEEDED, ICMP_EXC_TTL, 0);
drop:
	kfree_skb(skb);
	return NET_RX_DROP;
}
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 IP forwarding functionality.
e905a9ed	7	*
1da177e4 LT	8	* Authors: see ip.c
	9	*
	10	* Fixes:
e905a9ed	11	* Many : Split from ip.c , see ip_input.c for
1da177e4	12	* history.
e905a9ed	13	* Dave Gregorich : NULL ip_rt_put fix for multicast
1da177e4 LT	14	* routing.
	15	* Jos Vos : Add call_out_firewall before sending,
	16	* use output device for accounting.
	17	* Jos Vos : Call forward firewall after routing
	18	* (always use output device).
	19	* Mike McLagan : Routing by source
	20	*/
	21
1da177e4 LT	22	#include <linux/types.h>
1da177e4 LT	23	#include <linux/mm.h>
1da177e4 LT	24	#include <linux/skbuff.h>
	25	#include <linux/ip.h>
	26	#include <linux/icmp.h>
	27	#include <linux/netdevice.h>
5a0e3ad6	28	#include <linux/slab.h>
1da177e4 LT	29	#include <net/sock.h>
	30	#include <net/ip.h>
	31	#include <net/tcp.h>
	32	#include <net/udp.h>
	33	#include <net/icmp.h>
	34	#include <linux/tcp.h>
	35	#include <linux/udp.h>
	36	#include <linux/netfilter_ipv4.h>
	37	#include <net/checksum.h>
	38	#include <linux/route.h>
	39	#include <net/route.h>
	40	#include <net/xfrm.h>
	41
fe6cc55f FW	42	static bool ip_exceeds_mtu(const struct sk_buff *skb, unsigned int mtu)
fe6cc55f FW	43	{
ca6c5d4a	44	if (skb->len <= mtu)
fe6cc55f FW	45	return false;
fe6cc55f FW	46
cf826244 FW	47	if (unlikely((ip_hdr(skb)->frag_off & htons(IP_DF)) == 0))
	48	return false;
	49
	50	/* original fragment exceeds mtu and DF is set */
	51	if (unlikely(IPCB(skb)->frag_max_size > mtu))
	52	return true;
	53
	54	if (skb->ignore_df)
	55	return false;
	56
fe6cc55f FW	57	if (skb_is_gso(skb) && skb_gso_network_seglen(skb) <= mtu)
	58	return false;
	59
	60	return true;
	61	}
	62
fe6cc55f	63
7026b1dd	64	static int ip_forward_finish(struct sock sk, struct sk_buff skb)
1da177e4	65	{
5e73ea1a	66	struct ip_options *opt = &(IPCB(skb)->opt);
1da177e4	67
adf30907	68	IP_INC_STATS_BH(dev_net(skb_dst(skb)->dev), IPSTATS_MIB_OUTFORWDATAGRAMS);
2d8dbb04	69	IP_ADD_STATS_BH(dev_net(skb_dst(skb)->dev), IPSTATS_MIB_OUTOCTETS, skb->len);
1da177e4 LT	70
	71	if (unlikely(opt->optlen))
	72	ip_forward_options(skb);
	73
c29390c6	74	skb_sender_cpu_clear(skb);
7026b1dd	75	return dst_output_sk(sk, skb);
1da177e4 LT	76	}
	77
	78	int ip_forward(struct sk_buff *skb)
	79	{
f87c10a8	80	u32 mtu;
1da177e4 LT	81	struct iphdr iph; / Our header */
1da177e4 LT	82	struct rtable rt; / Route we use */
5e73ea1a	83	struct ip_options *opt = &(IPCB(skb)->opt);
1da177e4	84
d4f2fa6a DK	85	/* that should never happen */
	86	if (skb->pkt_type != PACKET_HOST)
	87	goto drop;
	88
2ab95749 SP	89	if (unlikely(skb->sk))
	90	goto drop;
	91
4497b076 BH	92	if (skb_warn_if_lro(skb))
	93	goto drop;
	94
1da177e4 LT	95	if (!xfrm4_policy_check(NULL, XFRM_POLICY_FWD, skb))
	96	goto drop;
	97
	98	if (IPCB(skb)->opt.router_alert && ip_call_ra_chain(skb))
	99	return NET_RX_SUCCESS;
	100
35fc92a9	101	skb_forward_csum(skb);
e905a9ed	102
1da177e4 LT	103	/*
	104	* According to the RFC, we must first decrease the TTL field. If
	105	* that reaches zero, we must reply an ICMP control message telling
	106	* that the packet's lifetime expired.
	107	*/
eddc9ec5	108	if (ip_hdr(skb)->ttl <= 1)
e905a9ed	109	goto too_many_hops;
1da177e4 LT	110
	111	if (!xfrm4_route_forward(skb))
	112	goto drop;
	113
511c3f92	114	rt = skb_rtable(skb);
1da177e4	115
155e8336	116	if (opt->is_strictroute && rt->rt_uses_gateway)
1da177e4 LT	117	goto sr_failed;
1da177e4 LT	118
f87c10a8 HFS	119	IPCB(skb)->flags \|= IPSKB_FORWARDED;
f87c10a8 HFS	120	mtu = ip_dst_mtu_maybe_forward(&rt->dst, true);
cf826244	121	if (ip_exceeds_mtu(skb, mtu)) {
d8d1f30b	122	IP_INC_STATS(dev_net(rt->dst.dev), IPSTATS_MIB_FRAGFAILS);
9af3912e	123	icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED,
f87c10a8	124	htonl(mtu));
9af3912e JH	125	goto drop;
	126	}
	127
1da177e4	128	/* We are about to mangle packet. Copy it! */
d8d1f30b	129	if (skb_cow(skb, LL_RESERVED_SPACE(rt->dst.dev)+rt->dst.header_len))
1da177e4	130	goto drop;
eddc9ec5	131	iph = ip_hdr(skb);
1da177e4 LT	132
	133	/* Decrease ttl after skb cow done */
	134	ip_decrease_ttl(iph);
	135
	136	/*
	137	* We now generate an ICMP HOST REDIRECT giving the route
	138	* we calculated.
	139	*/
df4d9254 HFS	140	if (IPCB(skb)->flags & IPSKB_DOREDIRECT && !opt->srr &&
df4d9254 HFS	141	!skb_sec_path(skb))
1da177e4 LT	142	ip_rt_send_redirect(skb);
	143
	144	skb->priority = rt_tos2priority(iph->tos);
	145
7026b1dd DM	146	return NF_HOOK(NFPROTO_IPV4, NF_INET_FORWARD, NULL, skb,
7026b1dd DM	147	skb->dev, rt->dst.dev, ip_forward_finish);
1da177e4 LT	148
1da177e4 LT	149	sr_failed:
e905a9ed	150	/*
1da177e4 LT	151	* Strict routing permits no gatewaying
1da177e4 LT	152	*/
e905a9ed YH	153	icmp_send(skb, ICMP_DEST_UNREACH, ICMP_SR_FAILED, 0);
e905a9ed YH	154	goto drop;
1da177e4 LT	155
1da177e4 LT	156	too_many_hops:
e905a9ed	157	/* Tell the sender its packet died... */
adf30907	158	IP_INC_STATS_BH(dev_net(skb_dst(skb)->dev), IPSTATS_MIB_INHDRERRORS);
e905a9ed	159	icmp_send(skb, ICMP_TIME_EXCEEDED, ICMP_EXC_TTL, 0);
1da177e4 LT	160	drop:
	161	kfree_skb(skb);
	162	return NET_RX_DROP;
	163	}