icmp.c

Linux内核源代码 为压缩文件 是<<Linux内核>>一书中的源代码

	{
	
		if (inet_addr_type(iph->daddr) == RTN_BROADCAST)
		{
			if (net_ratelimit())
				printk(KERN_WARNING "%u.%u.%u.%u 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);
	read_lock(&raw_v4_lock);
	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;
		}
	}
	read_unlock(&raw_v4_lock);

	/*
	 *	This can't change while we are doing it. 
	 *	Callers have obtained BR_NETPROTO_LOCK so
	 *	we are OK.
	 */

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

		nextip = (struct inet_protocol *) ipprot->next;
	
		/* 
		 *	Pass it off to everyone who wants it. 
		 */

		/* RFC1122: OK. Passes appropriate ICMP errors to the */
		/* appropriate protocol layer (MUST), as per 3.2.2. */

		if (iph->protocol == ipprot->protocol && ipprot->err_handler)
 			ipprot->err_handler(skb, dp, len);

		ipprot = nextip;
  	}
}


/*
 *	Handle ICMP_REDIRECT. 
 */

static void icmp_redirect(struct icmphdr *icmph, struct sk_buff *skb, int len)
{
	struct iphdr *iph;
	unsigned long ip;

	if (len < sizeof(struct iphdr)) {
		ICMP_INC_STATS_BH(IcmpInErrors);
		return; 
	}
		
	/*
	 *	Get the copied header of the packet that caused the redirect
	 */
	 
	iph = (struct iphdr *) (icmph + 1);
	ip = iph->daddr;

	switch(icmph->code & 7) {
		case ICMP_REDIR_NET:
		case ICMP_REDIR_NETTOS:
			/*
			 *	As per RFC recommendations now handle it as
			 *	a host redirect.
			 */
			 
		case ICMP_REDIR_HOST:
		case ICMP_REDIR_HOSTTOS:
			ip_rt_redirect(skb->nh.iph->saddr, ip, icmph->un.gateway, iph->saddr, iph->tos, skb->dev);
			break;
		default:
			break;
  	}
}

/*
 *	Handle ICMP_ECHO ("ping") requests. 
 *
 *	RFC 1122: 3.2.2.6 MUST have an echo server that answers ICMP echo requests.
 *	RFC 1122: 3.2.2.6 Data received in the ICMP_ECHO request MUST be included in the reply.
 *	RFC 1812: 4.3.3.6 SHOULD have a config option for silently ignoring echo requests, MUST have default=NOT.
 *	See also WRT handling of options once they are done and working.
 */

static void icmp_echo(struct icmphdr *icmph, struct sk_buff *skb, int len)
{
	if (!sysctl_icmp_echo_ignore_all) {
		struct icmp_bxm icmp_param;

		icmp_param.icmph=*icmph;
		icmp_param.icmph.type=ICMP_ECHOREPLY;
		icmp_param.data_ptr=(icmph+1);
		icmp_param.data_len=len;
		icmp_reply(&icmp_param, skb);
	}
}

/*
 *	Handle ICMP Timestamp requests. 
 *	RFC 1122: 3.2.2.8 MAY implement ICMP timestamp requests.
 *		  SHOULD be in the kernel for minimum random latency.
 *		  MUST be accurate to a few minutes.
 *		  MUST be updated at least at 15Hz.
 */
 
static void icmp_timestamp(struct icmphdr *icmph, struct sk_buff *skb, int len)
{
	struct timeval tv;
	__u32 times[3];		/* So the new timestamp works on ALPHA's.. */
	struct icmp_bxm icmp_param;
	
	/*
	 *	Too short.
	 */
	 
	if(len<12) {
		ICMP_INC_STATS_BH(IcmpInErrors);
		return;
	}
	
	/*
	 *	Fill in the current time as ms since midnight UT: 
	 */
	 
	do_gettimeofday(&tv);
	times[1] = htonl((tv.tv_sec % 86400) * 1000 + tv.tv_usec / 1000);
	times[2] = times[1];
	memcpy((void *)&times[0], icmph+1, 4);		/* Incoming stamp */
	icmp_param.icmph=*icmph;
	icmp_param.icmph.type=ICMP_TIMESTAMPREPLY;
	icmp_param.icmph.code=0;
	icmp_param.data_ptr=&times;
	icmp_param.data_len=12;
	icmp_reply(&icmp_param, skb);
}


/* 
 *	Handle ICMP_ADDRESS_MASK requests.  (RFC950)
 *
 * RFC1122 (3.2.2.9).  A host MUST only send replies to 
 * ADDRESS_MASK requests if it's been configured as an address mask 
 * agent.  Receiving a request doesn't constitute implicit permission to 
 * act as one. Of course, implementing this correctly requires (SHOULD) 
 * a way to turn the functionality on and off.  Another one for sysctl(), 
 * I guess. -- MS
 *
 * RFC1812 (4.3.3.9).	A router MUST implement it.
 *			A router SHOULD have switch turning it on/off.
 *		      	This switch MUST be ON by default.
 *
 * Gratuitous replies, zero-source replies are not implemented,
 * that complies with RFC. DO NOT implement them!!! All the idea
 * of broadcast addrmask replies as specified in RFC950 is broken.
 * The problem is that it is not uncommon to have several prefixes
 * on one physical interface. Moreover, addrmask agent can even be
 * not aware of existing another prefixes.
 * If source is zero, addrmask agent cannot choose correct prefix.
 * Gratuitous mask announcements suffer from the same problem.
 * RFC1812 explains it, but still allows to use ADDRMASK,
 * that is pretty silly. --ANK
 *
 * All these rules are so bizarre, that I removed kernel addrmask
 * support at all. It is wrong, it is obsolete, nobody uses it in
 * any case. --ANK
 *
 * Furthermore you can do it with a usermode address agent program
 * anyway...
 */

static void icmp_address(struct icmphdr *icmph, struct sk_buff *skb, int len)
{
#if 0
	if (net_ratelimit())
		printk(KERN_DEBUG "a guy asks for address mask. Who is it?\n");
#endif		
}

/*
 * RFC1812 (4.3.3.9).	A router SHOULD listen all replies, and complain
 *			loudly if an inconsistency is found.
 */

static void icmp_address_reply(struct icmphdr *icmph, struct sk_buff *skb, int len)
{
	struct rtable *rt = (struct rtable*)skb->dst;
	struct net_device *dev = skb->dev;
	struct in_device *in_dev;
	struct in_ifaddr *ifa;
	u32 mask;

	if (len < 4 || !(rt->rt_flags&RTCF_DIRECTSRC))
		return;

	in_dev = in_dev_get(dev);
	if (!in_dev)
		return;
	read_lock(&in_dev->lock);
	if (in_dev->ifa_list &&
	    IN_DEV_LOG_MARTIANS(in_dev) &&
	    IN_DEV_FORWARD(in_dev)) {

		mask = *(u32*)&icmph[1];
		for (ifa=in_dev->ifa_list; ifa; ifa = ifa->ifa_next) {
			if (mask == ifa->ifa_mask && inet_ifa_match(rt->rt_src, ifa))
				break;
		}
		if (!ifa && net_ratelimit()) {
			printk(KERN_INFO "Wrong address mask %u.%u.%u.%u from %s/%u.%u.%u.%u\n",
			       NIPQUAD(mask), dev->name, NIPQUAD(rt->rt_src));
		}
	}
	read_unlock(&in_dev->lock);
	in_dev_put(in_dev);
}

static void icmp_discard(struct icmphdr *icmph, struct sk_buff *skb, int len)
{
}

/* 
 *	Deal with incoming ICMP packets.
 */
 
int icmp_rcv(struct sk_buff *skb, unsigned short len)
{
	struct icmphdr *icmph = skb->h.icmph;
	struct rtable *rt = (struct rtable*)skb->dst;

	ICMP_INC_STATS_BH(IcmpInMsgs);

	/*
	 *	18 is the highest 'known' ICMP type. Anything else is a mystery
	 *
	 *	RFC 1122: 3.2.2  Unknown ICMP messages types MUST be silently discarded.
	 */
	if(len < sizeof(struct icmphdr) ||
	   ip_compute_csum((unsigned char *) icmph, len) ||
	   icmph->type > NR_ICMP_TYPES)
		goto error;
	 
	/*
	 *	Parse the ICMP message 
	 */

 	if (rt->rt_flags&(RTCF_BROADCAST|RTCF_MULTICAST)) {
		/*
		 *	RFC 1122: 3.2.2.6 An ICMP_ECHO to broadcast MAY be
		 *	  silently ignored (we let user decide with a sysctl).
		 *	RFC 1122: 3.2.2.8 An ICMP_TIMESTAMP MAY be silently
		 *	  discarded if to broadcast/multicast.
		 */
		if (icmph->type == ICMP_ECHO &&
		    sysctl_icmp_echo_ignore_broadcasts) {
			goto error;
		}
		if (icmph->type != ICMP_ECHO &&
		    icmph->type != ICMP_TIMESTAMP &&
		    icmph->type != ICMP_ADDRESS &&
		    icmph->type != ICMP_ADDRESSREPLY) {
			goto error;
  		}
	}

	len -= sizeof(struct icmphdr);
	icmp_pointers[icmph->type].input[smp_processor_id()*2*sizeof(struct icmp_mib)/sizeof(unsigned long)]++;
	(icmp_pointers[icmph->type].handler)(icmph, skb, len);

drop:
	kfree_skb(skb);
	return 0;
error:
	ICMP_INC_STATS_BH(IcmpInErrors);
	goto drop;
}

/*
 *	A spare long used to speed up statistics updating
 */
 
static unsigned long dummy;

/* 
 * 	Configurable rate limits.
 *	Someone should check if these default values are correct.
 *	Note that these values interact with the routing cache GC timeout.
 *	If you chose them too high they won't take effect, because the
 *	dst_entry gets expired too early. The same should happen when
 *	the cache grows too big.
 */
int sysctl_icmp_destunreach_time = 1*HZ;
int sysctl_icmp_timeexceed_time = 1*HZ;
int sysctl_icmp_paramprob_time = 1*HZ;
int sysctl_icmp_echoreply_time = 0; /* don't limit it per default. */

/*
 *	This table is the definition of how we handle ICMP.
 */
 
static struct icmp_control icmp_pointers[NR_ICMP_TYPES+1] = {
/* ECHO REPLY (0) */
 { &icmp_statistics[0].IcmpOutEchoReps, &icmp_statistics[0].IcmpInEchoReps, icmp_discard, 0, &sysctl_icmp_echoreply_time},
 { &dummy, &icmp_statistics[0].IcmpInErrors, icmp_discard, 1, },
 { &dummy, &icmp_statistics[0].IcmpInErrors, icmp_discard, 1, },
/* DEST UNREACH (3) */
 { &icmp_statistics[0].IcmpOutDestUnreachs, &icmp_statistics[0].IcmpInDestUnreachs, icmp_unreach, 1, &sysctl_icmp_destunreach_time },
/* SOURCE QUENCH (4) */
 { &icmp_statistics[0].IcmpOutSrcQuenchs, &icmp_statistics[0].IcmpInSrcQuenchs, icmp_unreach, 1, },
/* REDIRECT (5) */
 { &icmp_statistics[0].IcmpOutRedirects, &icmp_statistics[0].IcmpInRedirects, icmp_redirect, 1, },
 { &dummy, &icmp_statistics[0].IcmpInErrors, icmp_discard, 1, },
 { &dummy, &icmp_statistics[0].IcmpInErrors, icmp_discard, 1, },
/* ECHO (8) */
 { &icmp_statistics[0].IcmpOutEchos, &icmp_statistics[0].IcmpInEchos, icmp_echo, 0, },
 { &dummy, &icmp_statistics[0].IcmpInErrors, icmp_discard, 1, },
 { &dummy, &icmp_statistics[0].IcmpInErrors, icmp_discard, 1, },
/* TIME EXCEEDED (11) */
 { &icmp_statistics[0].IcmpOutTimeExcds, &icmp_statistics[0].IcmpInTimeExcds, icmp_unreach, 1, &sysctl_icmp_timeexceed_time },
/* PARAMETER PROBLEM (12) */
 { &icmp_statistics[0].IcmpOutParmProbs, &icmp_statistics[0].IcmpInParmProbs, icmp_unreach, 1, &sysctl_icmp_paramprob_time },
/* TIMESTAMP (13) */
 { &icmp_statistics[0].IcmpOutTimestamps, &icmp_statistics[0].IcmpInTimestamps, icmp_timestamp, 0,  },
/* TIMESTAMP REPLY (14) */
 { &icmp_statistics[0].IcmpOutTimestampReps, &icmp_statistics[0].IcmpInTimestampReps, icmp_discard, 0, },
/* INFO (15) */
 { &dummy, &dummy, icmp_discard, 0, },
/* INFO REPLY (16) */
 { &dummy, &dummy, icmp_discard, 0, },
/* ADDR MASK (17) */
 { &icmp_statistics[0].IcmpOutAddrMasks, &icmp_statistics[0].IcmpInAddrMasks, icmp_address, 0,  },
/* ADDR MASK REPLY (18) */
 { &icmp_statistics[0].IcmpOutAddrMaskReps, &icmp_statistics[0].IcmpInAddrMaskReps, icmp_address_reply, 0, }
};

void __init icmp_init(struct net_proto_family *ops)
{
	int err;

	icmp_inode.i_mode = S_IFSOCK;
	icmp_inode.i_sock = 1;
	icmp_inode.i_uid = 0;
	icmp_inode.i_gid = 0;
	init_waitqueue_head(&icmp_inode.i_wait);
	init_waitqueue_head(&icmp_inode.u.socket_i.wait);

	icmp_socket->inode = &icmp_inode;
	icmp_socket->state = SS_UNCONNECTED;
	icmp_socket->type=SOCK_RAW;

	if ((err=ops->create(icmp_socket, IPPROTO_ICMP))<0)
		panic("Failed to create the ICMP control socket.\n");
	icmp_socket->sk->allocation=GFP_ATOMIC;
	icmp_socket->sk->sndbuf = SK_WMEM_MAX*2;
	icmp_socket->sk->protinfo.af_inet.ttl = MAXTTL;

	/* Unhash it so that IP input processing does not even
	 * see it, we do not wish this socket to see incoming
	 * packets.
	 */
	icmp_socket->sk->prot->unhash(icmp_socket->sk);
}