udp.c

嵌入式系统设计与实验教材二源码linux内核移植与编译

/*
 * 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 User Datagram Protocol (UDP).
 *
 * Version:	$Id: udp.c,v 1.100.2.1 2002/01/12 07:39:23 davem Exp $
 *
 * Authors:	Ross Biro, <[e-mail]>
 *		Fred N. van Kempen, <[e-mail]>
 *		Arnt Gulbrandsen, <[e-mail]>
 *		Alan Cox, <[e-mail]>
 *
 * Fixes:
 *		Alan Cox	:	verify_area() calls
 *		Alan Cox	: 	stopped close while in use off icmp
 *					messages. Not a fix but a botch that
 *					for udp at least is 'valid'.
 *		Alan Cox	:	Fixed icmp handling properly
 *		Alan Cox	: 	Correct error for oversized datagrams
 *		Alan Cox	:	Tidied select() semantics. 
 *		Alan Cox	:	udp_err() fixed properly, also now 
 *					select and read wake correctly on errors
 *		Alan Cox	:	udp_send verify_area moved to avoid mem leak
 *		Alan Cox	:	UDP can count its memory
 *		Alan Cox	:	send to an unknown connection causes
 *					an ECONNREFUSED off the icmp, but
 *					does NOT close.
 *		Alan Cox	:	Switched to new sk_buff handlers. No more backlog!
 *		Alan Cox	:	Using generic datagram code. Even smaller and the PEEK
 *					bug no longer crashes it.
 *		Fred Van Kempen	: 	Net2e support for sk->broadcast.
 *		Alan Cox	:	Uses skb_free_datagram
 *		Alan Cox	:	Added get/set sockopt support.
 *		Alan Cox	:	Broadcasting without option set returns EACCES.
 *		Alan Cox	:	No wakeup calls. Instead we now use the callbacks.
 *		Alan Cox	:	Use ip_tos and ip_ttl
 *		Alan Cox	:	SNMP Mibs
 *		Alan Cox	:	MSG_DONTROUTE, and 0.0.0.0 support.
 *		Matt Dillon	:	UDP length checks.
 *		Alan Cox	:	Smarter af_inet used properly.
 *		Alan Cox	:	Use new kernel side addressing.
 *		Alan Cox	:	Incorrect return on truncated datagram receive.
 *	Arnt Gulbrandsen 	:	New udp_send and stuff
 *		Alan Cox	:	Cache last socket
 *		Alan Cox	:	Route cache
 *		Jon Peatfield	:	Minor efficiency fix to sendto().
 *		Mike Shaver	:	RFC1122 checks.
 *		Alan Cox	:	Nonblocking error fix.
 *	Willy Konynenberg	:	Transparent proxying support.
 *		Mike McLagan	:	Routing by source
 *		David S. Miller	:	New socket lookup architecture.
 *					Last socket cache retained as it
 *					does have a high hit rate.
 *		Olaf Kirch	:	Don't linearise iovec on sendmsg.
 *		Andi Kleen	:	Some cleanups, cache destination entry
 *					for connect. 
 *	Vitaly E. Lavrov	:	Transparent proxy revived after year coma.
 *		Melvin Smith	:	Check msg_name not msg_namelen in sendto(),
 *					return ENOTCONN for unconnected sockets (POSIX)
 *		Janos Farkas	:	don't deliver multi/broadcasts to a different
 *					bound-to-device socket
 *
 *
 *		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.
 */
 
#include <asm/system.h>
#include <asm/uaccess.h>
#include <asm/ioctls.h>
#include <linux/types.h>
#include <linux/fcntl.h>
#include <linux/socket.h>
#include <linux/sockios.h>
#include <linux/in.h>
#include <linux/errno.h>
#include <linux/timer.h>
#include <linux/mm.h>
#include <linux/config.h>
#include <linux/inet.h>
#include <linux/netdevice.h>
#include <net/snmp.h>
#include <net/ip.h>
#include <net/protocol.h>
#include <linux/skbuff.h>
#include <net/sock.h>
#include <net/udp.h>
#include <net/icmp.h>
#include <net/route.h>
#include <net/inet_common.h>
#include <net/checksum.h>

/*
 *	Snmp MIB for the UDP layer
 */

struct udp_mib		udp_statistics[NR_CPUS*2];

struct sock *udp_hash[UDP_HTABLE_SIZE];
rwlock_t udp_hash_lock = RW_LOCK_UNLOCKED;

/* Shared by v4/v6 udp. */
int udp_port_rover;

static int udp_v4_get_port(struct sock *sk, unsigned short snum)
{
	write_lock_bh(&udp_hash_lock);
	if (snum == 0) {
		int best_size_so_far, best, result, i;

		if (udp_port_rover > sysctl_local_port_range[1] ||
		    udp_port_rover < sysctl_local_port_range[0])
			udp_port_rover = sysctl_local_port_range[0];
		best_size_so_far = 32767;
		best = result = udp_port_rover;
		for (i = 0; i < UDP_HTABLE_SIZE; i++, result++) {
			struct sock *sk;
			int size;

			sk = udp_hash[result & (UDP_HTABLE_SIZE - 1)];
			if (!sk) {
				if (result > sysctl_local_port_range[1])
					result = sysctl_local_port_range[0] +
						((result - sysctl_local_port_range[0]) &
						 (UDP_HTABLE_SIZE - 1));
				goto gotit;
			}
			size = 0;
			do {
				if (++size >= best_size_so_far)
					goto next;
			} while ((sk = sk->next) != NULL);
			best_size_so_far = size;
			best = result;
		next:;
		}
		result = best;
		for(i = 0; i < (1 << 16) / UDP_HTABLE_SIZE; i++, result += UDP_HTABLE_SIZE) {
			if (result > sysctl_local_port_range[1])
				result = sysctl_local_port_range[0]
					+ ((result - sysctl_local_port_range[0]) &
					   (UDP_HTABLE_SIZE - 1));
			if (!udp_lport_inuse(result))
				break;
		}
		if (i >= (1 << 16) / UDP_HTABLE_SIZE)
			goto fail;
gotit:
		udp_port_rover = snum = result;
	} else {
		struct sock *sk2;

		for (sk2 = udp_hash[snum & (UDP_HTABLE_SIZE - 1)];
		     sk2 != NULL;
		     sk2 = sk2->next) {
			if (sk2->num == snum &&
			    sk2 != sk &&
			    sk2->bound_dev_if == sk->bound_dev_if &&
			    (!sk2->rcv_saddr ||
			     !sk->rcv_saddr ||
			     sk2->rcv_saddr == sk->rcv_saddr) &&
			    (!sk2->reuse || !sk->reuse))
				goto fail;
		}
	}
	sk->num = snum;
	if (sk->pprev == NULL) {
		struct sock **skp = &udp_hash[snum & (UDP_HTABLE_SIZE - 1)];
		if ((sk->next = *skp) != NULL)
			(*skp)->pprev = &sk->next;
		*skp = sk;
		sk->pprev = skp;
		sock_prot_inc_use(sk->prot);
		sock_hold(sk);
	}
	write_unlock_bh(&udp_hash_lock);
	return 0;

fail:
	write_unlock_bh(&udp_hash_lock);
	return 1;
}

static void udp_v4_hash(struct sock *sk)
{
	BUG();
}

static void udp_v4_unhash(struct sock *sk)
{
	write_lock_bh(&udp_hash_lock);
	if (sk->pprev) {
		if (sk->next)
			sk->next->pprev = sk->pprev;
		*sk->pprev = sk->next;
		sk->pprev = NULL;
		sk->num = 0;
		sock_prot_dec_use(sk->prot);
		__sock_put(sk);
	}
	write_unlock_bh(&udp_hash_lock);
}

/* UDP is nearly always wildcards out the wazoo, it makes no sense to try
 * harder than this. -DaveM
 */
struct sock *udp_v4_lookup_longway(u32 saddr, u16 sport, u32 daddr, u16 dport, int dif)
{
	struct sock *sk, *result = NULL;
	unsigned short hnum = ntohs(dport);
	int badness = -1;

	for(sk = udp_hash[hnum & (UDP_HTABLE_SIZE - 1)]; sk != NULL; sk = sk->next) {
		if(sk->num == hnum) {
			int score = 0;
			if(sk->rcv_saddr) {
				if(sk->rcv_saddr != daddr)
					continue;
				score++;
			}
			if(sk->daddr) {
				if(sk->daddr != saddr)
					continue;
				score++;
			}
			if(sk->dport) {
				if(sk->dport != sport)
					continue;
				score++;
			}
			if(sk->bound_dev_if) {
				if(sk->bound_dev_if != dif)
					continue;
				score++;
			}
			if(score == 4) {
				result = sk;
				break;
			} else if(score > badness) {
				result = sk;
				badness = score;
			}
		}
	}
	return result;
}

__inline__ struct sock *udp_v4_lookup(u32 saddr, u16 sport, u32 daddr, u16 dport, int dif)
{
	struct sock *sk;

	read_lock(&udp_hash_lock);
	sk = udp_v4_lookup_longway(saddr, sport, daddr, dport, dif);
	if (sk)
		sock_hold(sk);
	read_unlock(&udp_hash_lock);
	return sk;
}

static inline struct sock *udp_v4_mcast_next(struct sock *sk,
					     u16 loc_port, u32 loc_addr,
					     u16 rmt_port, u32 rmt_addr,
					     int dif)
{
	struct sock *s = sk;
	unsigned short hnum = ntohs(loc_port);
	for(; s; s = s->next) {
		if ((s->num != hnum)					||
		    (s->daddr && s->daddr!=rmt_addr)			||
		    (s->dport != rmt_port && s->dport != 0)			||
		    (s->rcv_saddr  && s->rcv_saddr != loc_addr)		||
		    (s->bound_dev_if && s->bound_dev_if != dif))
			continue;
		break;
  	}
  	return s;
}

/*
 * This routine is called by the ICMP module when it gets some
 * sort of error condition.  If err < 0 then the socket should
 * be closed and the error returned to the user.  If err > 0
 * it's just the icmp type << 8 | icmp code.  
 * Header points to the ip header of the error packet. We move
 * on past this. Then (as it used to claim before adjustment)
 * header points to the first 8 bytes of the udp header.  We need
 * to find the appropriate port.
 */

void udp_err(struct sk_buff *skb, u32 info)
{
	struct iphdr *iph = (struct iphdr*)skb->data;
	struct udphdr *uh = (struct udphdr*)(skb->data+(iph->ihl<<2));
	int type = skb->h.icmph->type;
	int code = skb->h.icmph->code;
	struct sock *sk;
	int harderr;
	int err;

	sk = udp_v4_lookup(iph->daddr, uh->dest, iph->saddr, uh->source, skb->dev->ifindex);
	if (sk == NULL) {
		ICMP_INC_STATS_BH(IcmpInErrors);
    	  	return;	/* No socket for error */
	}

	err = 0;
	harderr = 0;

	switch (type) {
	default:
	case ICMP_TIME_EXCEEDED:
		err = EHOSTUNREACH;
		break;
	case ICMP_SOURCE_QUENCH:
		goto out;
	case ICMP_PARAMETERPROB:
		err = EPROTO;
		harderr = 1;
		break;
	case ICMP_DEST_UNREACH:
		if (code == ICMP_FRAG_NEEDED) { /* Path MTU discovery */
			if (sk->protinfo.af_inet.pmtudisc != IP_PMTUDISC_DONT) {
				err = EMSGSIZE;
				harderr = 1;
				break;
			}
			goto out;
		}
		err = EHOSTUNREACH;
		if (code <= NR_ICMP_UNREACH) {
			harderr = icmp_err_convert[code].fatal;
			err = icmp_err_convert[code].errno;
		}
		break;
	}

	/*
	 *      RFC1122: OK.  Passes ICMP errors back to application, as per 
	 *	4.1.3.3.
	 */
	if (!sk->protinfo.af_inet.recverr) {
		if (!harderr || sk->state != TCP_ESTABLISHED)
			goto out;
	} else {
		ip_icmp_error(sk, skb, err, uh->dest, info, (u8*)(uh+1));
	}
	sk->err = err;
	sk->error_report(sk);
out:
	sock_put(sk);
}


static unsigned short udp_check(struct udphdr *uh, int len, unsigned long saddr, unsigned long daddr, unsigned long base)
{
	return(csum_tcpudp_magic(saddr, daddr, len, IPPROTO_UDP, base));
}

struct udpfakehdr 
{
	struct udphdr uh;
	u32 saddr;
	u32 daddr;
	struct iovec *iov;
	u32 wcheck;
};

/*
 *	Copy and checksum a UDP packet from user space into a buffer.
 */
 
static int udp_getfrag(const void *p, char * to, unsigned int offset, unsigned int fraglen) 
{
	struct udpfakehdr *ufh = (struct udpfakehdr *)p;
	if (offset==0) {
		if (csum_partial_copy_fromiovecend(to+sizeof(struct udphdr), ufh->iov, offset,
						   fraglen-sizeof(struct udphdr), &ufh->wcheck))
			return -EFAULT;
 		ufh->wcheck = csum_partial((char *)ufh, sizeof(struct udphdr),
					   ufh->wcheck);
		ufh->uh.check = csum_tcpudp_magic(ufh->saddr, ufh->daddr, 
					  ntohs(ufh->uh.len),
					  IPPROTO_UDP, ufh->wcheck);
		if (ufh->uh.check == 0)
			ufh->uh.check = -1;
		memcpy(to, ufh, sizeof(struct udphdr));
		return 0;
	}
	if (csum_partial_copy_fromiovecend(to, ufh->iov, offset-sizeof(struct udphdr),
					   fraglen, &ufh->wcheck))
		return -EFAULT;
	return 0;
}

/*
 *	Copy a UDP packet from user space into a buffer without checksumming.
 */
 
static int udp_getfrag_nosum(const void *p, char * to, unsigned int offset, unsigned int fraglen) 
{
	struct udpfakehdr *ufh = (struct udpfakehdr *)p;

	if (offset==0) {
		memcpy(to, ufh, sizeof(struct udphdr));
		return memcpy_fromiovecend(to+sizeof(struct udphdr), ufh->iov, offset,
					   fraglen-sizeof(struct udphdr));
	}
	return memcpy_fromiovecend(to, ufh->iov, offset-sizeof(struct udphdr),
				   fraglen);
}

int udp_sendmsg(struct sock *sk, struct msghdr *msg, int len)
{
	int ulen = len + sizeof(struct udphdr);
	struct ipcm_cookie ipc;
	struct udpfakehdr ufh;
	struct rtable *rt = NULL;
	int free = 0;
	int connected = 0;
	u32 daddr;
	u8  tos;
	int err;

	/* This check is ONLY to check for arithmetic overflow
	   on integer(!) len. Not more! Real check will be made
	   in ip_build_xmit --ANK

	   BTW socket.c -> af_*.c -> ... make multiple
	   invalid conversions size_t -> int. We MUST repair it f.e.
	   by replacing all of them with size_t and revise all
	   the places sort of len += sizeof(struct iphdr)
	   If len was ULONG_MAX-10 it would be cathastrophe  --ANK
	 */

	if (len < 0 || len > 0xFFFF)
		return -EMSGSIZE;

	/* 
	 *	Check the flags.
	 */

	if (msg->msg_flags&MSG_OOB)	/* Mirror BSD error message compatibility */
		return -EOPNOTSUPP;

	/*
	 *	Get and verify the address. 
	 */
	 
	if (msg->msg_name) {
		struct sockaddr_in * usin = (struct sockaddr_in*)msg->msg_name;
		if (msg->msg_namelen < sizeof(*usin))
			return -EINVAL;
		if (usin->sin_family != AF_INET) {
			if (usin->sin_family != AF_UNSPEC)
				return -EINVAL;
		}

		ufh.daddr = usin->sin_addr.s_addr;
		ufh.uh.dest = usin->sin_port;
		if (ufh.uh.dest == 0)
			return -EINVAL;
	} else {
		if (sk->state != TCP_ESTABLISHED)
			return -ENOTCONN;
		ufh.daddr = sk->daddr;
		ufh.uh.dest = sk->dport;
		/* Open fast path for connected socket.
		   Route will not be used, if at least one option is set.
		 */
		connected = 1;
  	}
	ipc.addr = sk->saddr;
	ufh.uh.source = sk->sport;

	ipc.opt = NULL;
	ipc.oif = sk->bound_dev_if;
	if (msg->msg_controllen) {
		err = ip_cmsg_send(msg, &ipc);
		if (err)
			return err;
		if (ipc.opt)
			free = 1;
		connected = 0;
	}
	if (!ipc.opt)
		ipc.opt = sk->protinfo.af_inet.opt;

	ufh.saddr = ipc.addr;
	ipc.addr = daddr = ufh.daddr;

	if (ipc.opt && ipc.opt->srr) {
		if (!daddr)
			return -EINVAL;
		daddr = ipc.opt->faddr;
		connected = 0;
	}
	tos = RT_TOS(sk->protinfo.af_inet.tos);
	if (sk->localroute || (msg->msg_flags&MSG_DONTROUTE) || 
	    (ipc.opt && ipc.opt->is_strictroute)) {
		tos |= RTO_ONLINK;
		connected = 0;
	}

	if (MULTICAST(daddr)) {
		if (!ipc.oif)
			ipc.oif = sk->protinfo.af_inet.mc_index;
		if (!ufh.saddr)
			ufh.saddr = sk->protinfo.af_inet.mc_addr;
		connected = 0;
	}