icmp.c

嵌入式系统设计与实例开发实验教材二源码 多线程应用程序设计 串行端口程序设计 AD接口实验 CAN总线通信实验 GPS通信实验 Linux内核移植与编译实验 IC卡读写实验 SD驱动使

/*
 *	NET3:	Implementation of the ICMP protocol layer. 
 *	
 *		Alan Cox, <[e-mail]>
 *
 *	Version: $Id: icmp.c,v 1.82.2.1 2001/12/13 08:59:27 davem Exp $
 *
 *	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.
 *
 *	Some of the function names and the icmp unreach table for this
 *	module were derived from [icmp.c 1.0.11 06/02/93] by
 *	Ross Biro, Fred N. van Kempen, Mark Evans, Alan Cox, Gerhard Koerting.
 *	Other than that this module is a complete rewrite.
 *
 *	Fixes:
 *	Clemens Fruhwirth	:	introduce global icmp rate limiting
 *					with icmp type masking ability instead
 *					of broken per type icmp timeouts.
 *		Mike Shaver	:	RFC1122 checks.
 *		Alan Cox	:	Multicast ping reply as self.
 *		Alan Cox	:	Fix atomicity lockup in ip_build_xmit 
 *					call.
 *		Alan Cox	:	Added 216,128 byte paths to the MTU 
 *					code.
 *		Martin Mares	:	RFC1812 checks.
 *		Martin Mares	:	Can be configured to follow redirects 
 *					if acting as a router _without_ a
 *					routing protocol (RFC 1812).
 *		Martin Mares	:	Echo requests may be configured to 
 *					be ignored (RFC 1812).
 *		Martin Mares	:	Limitation of ICMP error message 
 *					transmit rate (RFC 1812).
 *		Martin Mares	:	TOS and Precedence set correctly 
 *					(RFC 1812).
 *		Martin Mares	:	Now copying as much data from the 
 *					original packet as we can without
 *					exceeding 576 bytes (RFC 1812).
 *	Willy Konynenberg	:	Transparent proxying support.
 *		Keith Owens	:	RFC1191 correction for 4.2BSD based 
 *					path MTU bug.
 *		Thomas Quinot	:	ICMP Dest Unreach codes up to 15 are
 *					valid (RFC 1812).
 *		Andi Kleen	:	Check all packet lengths properly
 *					and moved all kfree_skb() up to
 *					icmp_rcv.
 *		Andi Kleen	:	Move the rate limit bookkeeping
 *					into the dest entry and use a token
 *					bucket filter (thanks to ANK). Make
 *					the rates sysctl configurable.
 *		Yu Tianli	:	Fixed two ugly bugs in icmp_send
 *					- IP option length was accounted wrongly
 *					- ICMP header length was not accounted at all.
 *              Tristan Greaves :       Added sysctl option to ignore bogus broadcast
 *                                      responses from broken routers.
 *
 * To Fix:
 *
 *	- Should use skb_pull() instead of all the manual checking.
 *	  This would also greatly simply some upper layer error handlers. --AK
 *
 */

#include <linux/config.h>
#include <linux/types.h>
#include <linux/sched.h>
#include <linux/kernel.h>
#include <linux/fcntl.h>
#include <linux/socket.h>
#include <linux/in.h>
#include <linux/inet.h>
#include <linux/netdevice.h>
#include <linux/string.h>
#include <linux/netfilter_ipv4.h>
#include <net/snmp.h>
#include <net/ip.h>
#include <net/route.h>
#include <net/protocol.h>
#include <net/icmp.h>
#include <net/tcp.h>
#include <net/udp.h>
#include <net/raw.h>
#include <linux/skbuff.h>
#include <net/sock.h>
#include <linux/errno.h>
#include <linux/timer.h>
#include <linux/init.h>
#include <asm/system.h>
#include <asm/uaccess.h>
#include <net/checksum.h>

/*
 *	Build xmit assembly blocks
 */

struct icmp_bxm
{
	struct sk_buff *skb;
	int offset;
	int data_len;

	unsigned int csum;
	struct {
		struct icmphdr icmph;
		__u32	       times[3];
	} data;
	int head_len;
	struct ip_options replyopts;
	unsigned char  optbuf[40];
};

/*
 *	Statistics
 */
 
struct icmp_mib icmp_statistics[NR_CPUS*2];

/* An array of errno for error messages from dest unreach. */
/* RFC 1122: 3.2.2.1 States that NET_UNREACH, HOS_UNREACH and SR_FAIELD MUST be considered 'transient errs'. */

struct icmp_err icmp_err_convert[] = {
  { ENETUNREACH,	0 },	/*	ICMP_NET_UNREACH	*/
  { EHOSTUNREACH,	0 },	/*	ICMP_HOST_UNREACH	*/
  { ENOPROTOOPT,	1 },	/*	ICMP_PROT_UNREACH	*/
  { ECONNREFUSED,	1 },	/*	ICMP_PORT_UNREACH	*/
  { EMSGSIZE,		0 },	/*	ICMP_FRAG_NEEDED	*/
  { EOPNOTSUPP,		0 },	/*	ICMP_SR_FAILED		*/
  { ENETUNREACH,	1 },	/* 	ICMP_NET_UNKNOWN	*/
  { EHOSTDOWN,		1 },	/*	ICMP_HOST_UNKNOWN	*/
  { ENONET,		1 },	/*	ICMP_HOST_ISOLATED	*/
  { ENETUNREACH,	1 },	/*	ICMP_NET_ANO		*/
  { EHOSTUNREACH,	1 },	/*	ICMP_HOST_ANO		*/
  { ENETUNREACH,	0 },	/*	ICMP_NET_UNR_TOS	*/
  { EHOSTUNREACH,	0 },	/*	ICMP_HOST_UNR_TOS	*/
  { EHOSTUNREACH,	1 },	/*	ICMP_PKT_FILTERED	*/
  { EHOSTUNREACH,	1 },	/*	ICMP_PREC_VIOLATION	*/
  { EHOSTUNREACH,	1 }	/*	ICMP_PREC_CUTOFF	*/
};

/* Control parameters for ECHO replies. */
int sysctl_icmp_echo_ignore_all;
int sysctl_icmp_echo_ignore_broadcasts;

/* Control parameter - ignore bogus broadcast responses? */
int sysctl_icmp_ignore_bogus_error_responses;

/* 
 * 	Configurable global rate limit.
 *
 *	ratelimit defines tokens/packet consumed for dst->rate_token bucket
 *	ratemask defines which icmp types are ratelimited by setting
 * 	it's bit position.
 *
 *	default: 
 *	dest unreachable (3), source quench (4),
 *	time exceeded (11), parameter problem (12)
 */

int sysctl_icmp_ratelimit = 1*HZ;
int sysctl_icmp_ratemask = 0x1818;

/*
 *	ICMP control array. This specifies what to do with each ICMP.
 */

struct icmp_control
{
	unsigned long *output;		/* Address to increment on output */
	unsigned long *input;		/* Address to increment on input */
	void (*handler)(struct sk_buff *skb);
	short	error;		/* This ICMP is classed as an error message */
};

static struct icmp_control icmp_pointers[NR_ICMP_TYPES+1];

/*
 *	The ICMP socket. This is the most convenient way to flow control
 *	our ICMP output as well as maintain a clean interface throughout
 *	all layers. All Socketless IP sends will soon be gone.
 */
	
struct inode icmp_inode;
struct socket *icmp_socket = &icmp_inode.u.socket_i;

/* ICMPv4 socket is only a bit non-reenterable (unlike ICMPv6,
   which is strongly non-reenterable). A bit later it will be made
   reenterable and the lock may be removed then.
 */

static int icmp_xmit_holder = -1;

static int icmp_xmit_lock_bh(void)
{
	if (!spin_trylock(&icmp_socket->sk->lock.slock)) {
		if (icmp_xmit_holder == smp_processor_id())
			return -EAGAIN;
		spin_lock(&icmp_socket->sk->lock.slock);
	}
	icmp_xmit_holder = smp_processor_id();
	return 0;
}

static __inline__ int icmp_xmit_lock(void)
{
	int ret;
	local_bh_disable();
	ret = icmp_xmit_lock_bh();
	if (ret)
		local_bh_enable();
	return ret;
}

static void icmp_xmit_unlock_bh(void)
{
	icmp_xmit_holder = -1;
	spin_unlock(&icmp_socket->sk->lock.slock);
}

static __inline__ void icmp_xmit_unlock(void)
{
	icmp_xmit_unlock_bh();
	local_bh_enable();
}


/*
 *	Send an ICMP frame.
 */

/*
 *	Check transmit rate limitation for given message.
 *	The rate information is held in the destination cache now.
 *	This function is generic and could be used for other purposes
 *	too. It uses a Token bucket filter as suggested by Alexey Kuznetsov.
 *
 *	Note that the same dst_entry fields are modified by functions in 
 *	route.c too, but these work for packet destinations while xrlim_allow
 *	works for icmp destinations. This means the rate limiting information
 *	for one "ip object" is shared - and these ICMPs are twice limited:
 *	by source and by destination.
 *
 *	RFC 1812: 4.3.2.8 SHOULD be able to limit error message rate
 *			  SHOULD allow setting of rate limits 
 *
 * 	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)
		return 1;

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

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

	/* Limit if icmp type is enabled in ratemask. */
	if((1 << type) & sysctl_icmp_ratemask)
		return xrlim_allow(dst, sysctl_icmp_ratelimit);
	else
		return 1;
}

/*
 *	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)[(smp_processor_id()*2+!in_softirq())*sizeof(struct icmp_mib)/sizeof(unsigned long)]++;
	ICMP_INC_STATS(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 int csum;

	if (offset) {
		icmp_param->csum=skb_copy_and_csum_bits(icmp_param->skb,
							icmp_param->offset+(offset-icmp_param->head_len), 
							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_nocheck((void *)&icmp_param->data,
		to, icmp_param->head_len,
		icmp_param->csum);
	csum=skb_copy_and_csum_bits(icmp_param->skb,
				    icmp_param->offset, 
				    to+icmp_param->head_len,
				    fraglen-icmp_param->head_len,
				    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;

	if (icmp_xmit_lock_bh())
		return;

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

	sk->protinfo.af_inet.tos = skb->nh.iph->tos;
	daddr = ipc.addr = rt->rt_src;
	ipc.opt = NULL;
	if (icmp_param->replyopts.optlen) {
		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))
		goto out;
	if (icmpv4_xrlim_allow(rt, icmp_param->data.icmph.type, 
			       icmp_param->data.icmph.code)) { 
		ip_build_xmit(sk, icmp_glue_bits, icmp_param, 
			      icmp_param->data_len+icmp_param->head_len,
			      &ipc, rt, MSG_DONTWAIT);
	}
	ip_rt_put(rt);
out:
	icmp_xmit_unlock_bh();
}


/*
 *	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, u32 info)
{
	struct iphdr *iph;
	int room;
	struct icmp_bxm icmp_param;
	struct rtable *rt = (struct rtable*)skb_in->dst;
	struct ipcm_cookie ipc;
	u32 saddr;
	u8  tos;

	if (!rt)
		return;

	/*
	 *	Find the original header. It is expected to be valid, of course.
	 *	Check this, icmp_send is called from the most obscure devices
	 *	sometimes.
	 */
	iph = skb_in->nh.iph;

	if ((u8*)iph < skb_in->head || (u8*)(iph+1) > skb_in->tail)
		return;

	/*
	 *	No replies to physical multicast/broadcast
	 */
	if (skb_in->pkt_type!=PACKET_HOST)
		return;

	/*
	 *	Now check at the protocol level
	 */
	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) {
			u8 inner_type;

			if (skb_copy_bits(skb_in,
					  skb_in->nh.raw + (iph->ihl<<2)
					  + offsetof(struct icmphdr, type)
					  - skb_in->data,
					  &inner_type, 1))
				return;

			/*
			 *	Assume any unknown ICMP type is an error. This isn't
			 *	specified by the RFC, but think about it..
			 */
			if (inner_type>NR_ICMP_TYPES || icmp_pointers[inner_type].error)
				return;
		}
	}

	if (icmp_xmit_lock())
		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

	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;

	if (ip_route_output(&rt, iph->saddr, saddr, RT_TOS(tos), 0))
		goto out;

	if (ip_options_echo(&icmp_param.replyopts, skb_in)) 
		goto ende;


	/*
	 *	Prepare data for ICMP header.
	 */

	icmp_param.data.icmph.type=type;
	icmp_param.data.icmph.code=code;
	icmp_param.data.icmph.un.gateway = info;
	icmp_param.data.icmph.checksum=0;
	icmp_param.csum=0;
	icmp_param.skb=skb_in;
	icmp_param.offset=skb_in->nh.raw - skb_in->data;
	icmp_out_count(icmp_param.data.icmph.type);
	icmp_socket->sk->protinfo.af_inet.tos = tos;
	ipc.addr = iph->saddr;
	ipc.opt = &icmp_param.replyopts;
	if (icmp_param.replyopts.srr) {
		ip_rt_put(rt);