icmp.c

GNU Hurd 源代码

 * 	Shared between ICMPv4 and ICMPv6.
 */
#define XRLIM_BURST_FACTOR 6
int xrlim_allow(struct dst_entry *dst, int timeout)
{
	unsigned long now;

	now = jiffies;
	dst->rate_tokens += now - dst->rate_last;
	dst->rate_last = now;
	if (dst->rate_tokens > XRLIM_BURST_FACTOR*timeout)
		dst->rate_tokens = XRLIM_BURST_FACTOR*timeout;
	if (dst->rate_tokens >= timeout) {
		dst->rate_tokens -= timeout;
		return 1;
	}
	return 0; 
}

static inline int icmpv4_xrlim_allow(struct rtable *rt, int type, int code)
{
	struct dst_entry *dst = &rt->u.dst; 

	if (type > NR_ICMP_TYPES || !icmp_pointers[type].timeout)
		return 1;

	/* Don't limit PMTU discovery. */
	if (type == ICMP_DEST_UNREACH && code == ICMP_FRAG_NEEDED)
		return 1;

	/* Redirect has its own rate limit mechanism */
	if (type == ICMP_REDIRECT)
		return 1;

	/* No rate limit on loopback */
	if (dst->dev && (dst->dev->flags&IFF_LOOPBACK))
 		return 1;

	return xrlim_allow(dst, *(icmp_pointers[type].timeout));
}

/*
 *	Maintain the counters used in the SNMP statistics for outgoing ICMP
 */
 
static void icmp_out_count(int type)
{
	if (type>NR_ICMP_TYPES)
		return;
	(*icmp_pointers[type].output)++;
	icmp_statistics.IcmpOutMsgs++;
}
 
/*
 *	Checksum each fragment, and on the first include the headers and final checksum.
 */
 
static int icmp_glue_bits(const void *p, char *to, unsigned int offset, unsigned int fraglen)
{
	struct icmp_bxm *icmp_param = (struct icmp_bxm *)p;
	struct icmphdr *icmph;
	unsigned long csum;

	if (offset) {
		icmp_param->csum=csum_partial_copy(icmp_param->data_ptr+offset-sizeof(struct icmphdr), 
				to, fraglen,icmp_param->csum);
		return 0;
	}

	/*
	 *	First fragment includes header. Note that we've done
	 *	the other fragments first, so that we get the checksum
	 *	for the whole packet here.
	 */
	csum = csum_partial_copy((void *)&icmp_param->icmph,
		to, sizeof(struct icmphdr), 
		icmp_param->csum);
	csum = csum_partial_copy(icmp_param->data_ptr,
		to+sizeof(struct icmphdr),
		fraglen-sizeof(struct icmphdr), csum);
	icmph=(struct icmphdr *)to;
	icmph->checksum = csum_fold(csum);
	return 0;
}
 
/*
 *	Driving logic for building and sending ICMP messages.
 */

static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb)
{
	struct sock *sk=icmp_socket->sk;
	struct ipcm_cookie ipc;
	struct rtable *rt = (struct rtable*)skb->dst;
	u32 daddr;

	if (ip_options_echo(&icmp_param->replyopts, skb))
		return;

	icmp_param->icmph.checksum=0;
	icmp_param->csum=0;
	icmp_out_count(icmp_param->icmph.type);

	sk->ip_tos = skb->nh.iph->tos;
	daddr = ipc.addr = rt->rt_src;
	ipc.opt = &icmp_param->replyopts;
	if (ipc.opt->srr)
		daddr = icmp_param->replyopts.faddr;
	if (ip_route_output(&rt, daddr, rt->rt_spec_dst, RT_TOS(skb->nh.iph->tos), 0))
		return;
	ip_build_xmit(sk, icmp_glue_bits, icmp_param, 
		icmp_param->data_len+sizeof(struct icmphdr),
		&ipc, rt, MSG_DONTWAIT);
	ip_rt_put(rt);
}


/*
 *	Send an ICMP message in response to a situation
 *
 *	RFC 1122: 3.2.2	MUST send at least the IP header and 8 bytes of header. MAY send more (we do).
 *			MUST NOT change this header information.
 *			MUST NOT reply to a multicast/broadcast IP address.
 *			MUST NOT reply to a multicast/broadcast MAC address.
 *			MUST reply to only the first fragment.
 */

void icmp_send(struct sk_buff *skb_in, int type, int code, unsigned long info)
{
	struct iphdr *iph;
	struct icmphdr *icmph;
	int room;
	struct icmp_bxm icmp_param;
	struct rtable *rt = (struct rtable*)skb_in->dst;
	struct ipcm_cookie ipc;
	u32 saddr;
	u8  tos;
	
	/*
	 *	Find the original header
	 */
	 
	iph = skb_in->nh.iph;
	
	/*
	 *	No replies to physical multicast/broadcast
	 */
	 
	if (skb_in->pkt_type!=PACKET_HOST)
		return;
		
	/*
	 *	Now check at the protocol level
	 */
	if (!rt) {
                if (sysctl_ip_always_defrag == 0 &&
                    net_ratelimit())
			printk(KERN_DEBUG "icmp_send: destinationless packet\n");
		return;
	}
	if (rt->rt_flags&(RTCF_BROADCAST|RTCF_MULTICAST))
		return;
	 
		
	/*
	 *	Only reply to fragment 0. We byte re-order the constant
	 *	mask for efficiency.
	 */
	 
	if (iph->frag_off&htons(IP_OFFSET))
		return;
		
	/* 
	 *	If we send an ICMP error to an ICMP error a mess would result..
	 */
	 
	if (icmp_pointers[type].error) {
		/*
		 *	We are an error, check if we are replying to an ICMP error
		 */
		 
		if (iph->protocol==IPPROTO_ICMP) {
			icmph = (struct icmphdr *)((char *)iph + (iph->ihl<<2));
			/*
			 *	Assume any unknown ICMP type is an error. This isn't
			 *	specified by the RFC, but think about it..
			 */
			if (icmph->type>NR_ICMP_TYPES || icmp_pointers[icmph->type].error)
				return;
		}
	}


	/*
	 *	Construct source address and options.
	 */

#ifdef CONFIG_IP_ROUTE_NAT	
	/*
	 *	Restore original addresses if packet has been translated.
	 */
	if (rt->rt_flags&RTCF_NAT && IPCB(skb_in)->flags&IPSKB_TRANSLATED) {
		iph->daddr = rt->key.dst;
		iph->saddr = rt->key.src;
	}
#endif
#ifdef CONFIG_IP_MASQUERADE
	if (type==ICMP_DEST_UNREACH && IPCB(skb_in)->flags&IPSKB_MASQUERADED) {
			ip_fw_unmasq_icmp(skb_in);
	}
#endif

	saddr = iph->daddr;
	if (!(rt->rt_flags & RTCF_LOCAL))
		saddr = 0;

	tos = icmp_pointers[type].error ?
		((iph->tos & IPTOS_TOS_MASK) | IPTOS_PREC_INTERNETCONTROL) :
			iph->tos;

	/* XXX: use a more aggressive expire for routes created by 
	 * this call (not longer than the rate limit timeout). 
	 * It could be also worthwhile to not put them into ipv4
	 * fast routing cache at first. Otherwise an attacker can
	 * grow the routing table.
	 */
	if (ip_route_output(&rt, iph->saddr, saddr, RT_TOS(tos), 0))
		return;
	
	if (ip_options_echo(&icmp_param.replyopts, skb_in)) 
		goto ende;


	/*
	 *	Prepare data for ICMP header.
	 */

	icmp_param.icmph.type=type;
	icmp_param.icmph.code=code;
	icmp_param.icmph.un.gateway = info;
	icmp_param.icmph.checksum=0;
	icmp_param.csum=0;
	icmp_param.data_ptr=iph;
	icmp_out_count(icmp_param.icmph.type);
	icmp_socket->sk->ip_tos = tos;
	ipc.addr = iph->saddr;
	ipc.opt = &icmp_param.replyopts;
	if (icmp_param.replyopts.srr) {
		ip_rt_put(rt);
		if (ip_route_output(&rt, icmp_param.replyopts.faddr, saddr, RT_TOS(tos), 0))
			return;
	}

	if (!icmpv4_xrlim_allow(rt, type, code))
		goto ende;

	/* RFC says return as much as we can without exceeding 576 bytes. */

	room = rt->u.dst.pmtu;
	if (room > 576)
		room = 576;
	room -= sizeof(struct iphdr) + icmp_param.replyopts.optlen;
	room -= sizeof(struct icmphdr);

	icmp_param.data_len=(iph->ihl<<2)+skb_in->len;
	if (icmp_param.data_len > room)
		icmp_param.data_len = room;
	
	ip_build_xmit(icmp_socket->sk, icmp_glue_bits, &icmp_param, 
		icmp_param.data_len+sizeof(struct icmphdr),
		&ipc, rt, MSG_DONTWAIT);

ende:
	ip_rt_put(rt);
}


/* 
 *	Handle ICMP_DEST_UNREACH, ICMP_TIME_EXCEED, and ICMP_QUENCH. 
 */

static void icmp_unreach(struct icmphdr *icmph, struct sk_buff *skb, int len)
{
	struct iphdr *iph;
	int hash;
	struct inet_protocol *ipprot;
	unsigned char *dp;
	struct sock *raw_sk;
	
	/*
	 *	Incomplete header ?
	 * 	Only checks for the IP header, there should be an
	 *	additional check for longer headers in upper levels.
	 */

	if(len<sizeof(struct iphdr)) {
		icmp_statistics.IcmpInErrors++;
		return;
	}
		
	iph = (struct iphdr *) (icmph + 1);
	dp = (unsigned char*)iph;
	
	if(icmph->type==ICMP_DEST_UNREACH) {
		switch(icmph->code & 15) {
			case ICMP_NET_UNREACH:
				break;
			case ICMP_HOST_UNREACH:
				break;
			case ICMP_PROT_UNREACH:
				break;
			case ICMP_PORT_UNREACH:
				break;
			case ICMP_FRAG_NEEDED:
				if (ipv4_config.no_pmtu_disc) {
					if (sysctl_ip_always_defrag == 0 && net_ratelimit())
						printk(KERN_INFO "ICMP: %d.%d.%d.%d: fragmentation needed and DF set.\n",
					       NIPQUAD(iph->daddr));
				} else {
					unsigned short new_mtu;
					new_mtu = ip_rt_frag_needed(iph, ntohs(icmph->un.frag.mtu));
					if (!new_mtu) 
						return;
					icmph->un.frag.mtu = htons(new_mtu);
				}
				break;
			case ICMP_SR_FAILED:
				if (sysctl_ip_always_defrag == 0 && net_ratelimit())
					printk(KERN_INFO "ICMP: %d.%d.%d.%d: Source Route Failed.\n", NIPQUAD(iph->daddr));
				break;
			default:
				break;
		}
		if (icmph->code>NR_ICMP_UNREACH) 
			return;
	}
	
	/*
	 *	Throw it at our lower layers
	 *
	 *	RFC 1122: 3.2.2 MUST extract the protocol ID from the passed header.
	 *	RFC 1122: 3.2.2.1 MUST pass ICMP unreach messages to the transport layer.
	 *	RFC 1122: 3.2.2.2 MUST pass ICMP time expired messages to transport layer.
	 */
	 
	/*
	 *	Check the other end isnt violating RFC 1122. Some routers send
	 *	bogus responses to broadcast frames. If you see this message
	 *	first check your netmask matches at both ends, if it does then
	 *	get the other vendor to fix their kit.
	 */

	if (!sysctl_icmp_ignore_bogus_error_responses)
	{
	
		if (inet_addr_type(iph->daddr) == RTN_BROADCAST)
		{
			if (net_ratelimit())
				printk(KERN_WARNING "%d.%d.%d.%d sent an invalid ICMP error to a broadcast.\n",
			       	NIPQUAD(skb->nh.iph->saddr));
			return; 
		}
	}

	/*
	 *	Deliver ICMP message to raw sockets. Pretty useless feature?
	 */

	/* Note: See raw.c and net/raw.h, RAWV4_HTABLE_SIZE==MAX_INET_PROTOS */
	hash = iph->protocol & (MAX_INET_PROTOS - 1);
	if ((raw_sk = raw_v4_htable[hash]) != NULL) 
	{
		while ((raw_sk = raw_v4_lookup(raw_sk, iph->protocol, iph->saddr,
					       iph->daddr, skb->dev->ifindex)) != NULL) {
			raw_err(raw_sk, skb);
			raw_sk = raw_sk->next;
		}
	}

	/*
	 *	This can't change while we are doing it. 
	 */

	ipprot = (struct inet_protocol *) inet_protos[hash];
	while(ipprot != NULL) {
		struct inet_protocol *nextip;