/*
 *	Linux NET3:	Internet Group Management Protocol  [IGMP]
 *
 *	This code implements the IGMP protocol as defined in RFC1112. There has
 *	been a further revision of this protocol since which is now supported.
 *
 *	If you have trouble with this module be careful what gcc you have used,
 *	the older version didn't come out right using gcc 2.5.8, the newer one
 *	seems to fall out with gcc 2.6.2.
 *
 *	Version: $Id: igmp.c,v 1.22 1997/10/29 20:27:24 kuznet Exp $
 *
 *	Authors:
 *		Alan Cox <Alan.Cox@linux.org>
 *
 *	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.
 *
 *	Fixes:
 *
 *		Alan Cox	:	Added lots of __inline__ to optimise
 *					the memory usage of all the tiny little
 *					functions.
 *		Alan Cox	:	Dumped the header building experiment.
 *		Alan Cox	:	Minor tweaks ready for multicast routing
 *					and extended IGMP protocol.
 *		Alan Cox	:	Removed a load of inline directives. Gcc 2.5.8
 *					writes utterly bogus code otherwise (sigh)
 *					fixed IGMP loopback to behave in the manner
 *					desired by mrouted, fixed the fact it has been
 *					broken since 1.3.6 and cleaned up a few minor
 *					points.
 *
 *		Chih-Jen Chang	:	Tried to revise IGMP to Version 2
 *		Tsu-Sheng Tsao		E-mail: chihjenc@scf.usc.edu and tsusheng@scf.usc.edu
 *					The enhancements are mainly based on Steve Deering's 
 * 					ipmulti-3.5 source code.
 *		Chih-Jen Chang	:	Added the igmp_get_mrouter_info and
 *		Tsu-Sheng Tsao		igmp_set_mrouter_info to keep track of
 *					the mrouted version on that device.
 *		Chih-Jen Chang	:	Added the max_resp_time parameter to
 *		Tsu-Sheng Tsao		igmp_heard_query(). Using this parameter
 *					to identify the multicast router version
 *					and do what the IGMP version 2 specified.
 *		Chih-Jen Chang	:	Added a timer to revert to IGMP V2 router
 *		Tsu-Sheng Tsao		if the specified time expired.
 *		Alan Cox	:	Stop IGMP from 0.0.0.0 being accepted.
 *		Alan Cox	:	Use GFP_ATOMIC in the right places.
 *		Christian Daudt :	igmp timer wasn't set for local group
 *					memberships but was being deleted, 
 *					which caused a "del_timer() called 
 *					from %p with timer not initialized\n"
 *					message (960131).
 *		Christian Daudt :	removed del_timer from 
 *					igmp_timer_expire function (960205).
 *             Christian Daudt :       igmp_heard_report now only calls
 *                                     igmp_timer_expire if tm->running is
 *                                     true (960216).
 *		Malcolm Beattie :	ttl comparison wrong in igmp_rcv made
 *					igmp_heard_query never trigger. Expiry
 *					miscalculation fixed in igmp_heard_query
 *					and random() made to return unsigned to
 *					prevent negative expiry times.
 *		Alexey Kuznetsov:	Wrong group leaving behaviour, backport
 *					fix from pending 2.1.x patches.
 *		Alan Cox:		Forget to enable FDDI support earlier.
 *		Alexey Kuznetsov:	Fixed leaving groups on device down.
 *		Alexey Kuznetsov:	Accordance to igmp-v2-06 draft.
 */


#include <linux/config.h>
#include <asm/uaccess.h>
#include <asm/system.h>
#include <linux/types.h>
#include <linux/kernel.h>
#include <linux/sched.h>
#include <linux/string.h>
#include <linux/socket.h>
#include <linux/sockios.h>
#include <linux/in.h>
#include <linux/inet.h>
#include <linux/netdevice.h>
#include <linux/skbuff.h>
#include <linux/inetdevice.h>
#include <linux/igmp.h>
#include <linux/if_arp.h>
#include <linux/rtnetlink.h>
#include <net/ip.h>
#include <net/protocol.h>
#include <net/route.h>
#include <net/sock.h>
#include <net/checksum.h>
#ifdef CONFIG_IP_MROUTE
#include <linux/mroute.h>
#endif


#ifdef CONFIG_IP_MULTICAST

/* Parameter names and values are taken from igmp-v2-06 draft */

#define IGMP_V1_Router_Present_Timeout		(400*HZ)
#define IGMP_Unsolicited_Report_Interval	(10*HZ)
#define IGMP_Query_Response_Interval		(10*HZ)
#define IGMP_Unsolicited_Report_Count		2


#define IGMP_Initial_Report_Delay		(1*HZ)

/* IGMP_Initial_Report_Delay is not from IGMP specs!
 * IGMP specs require to report membership immediately after
 * joining a group, but we delay the first report by a
 * small interval. It seems more natural and still does not
 * contradict to specs provided this delay is small enough.
 */

#define IGMP_V1_SEEN(in_dev) ((in_dev)->mr_v1_seen && jiffies - (in_dev)->mr_v1_seen < 0)

/*
 *	Timer management
 */

static __inline__ void igmp_stop_timer(struct ip_mc_list *im)
{
	if (im->tm_running) {
		del_timer(&im->timer);
		im->tm_running=0;
	}
}

extern __inline__ unsigned int random(void)
{
	static unsigned long seed=152L;
	seed=seed*69069L+1;
	return seed^jiffies;
}

static __inline__ void igmp_start_timer(struct ip_mc_list *im, int max_delay)
{
	int tv;
	if (im->tm_running)
		return;
	tv=random() % max_delay;
	im->timer.expires=jiffies+tv+2;
	im->tm_running=1;
	add_timer(&im->timer);
}

/*
 *	Send an IGMP report.
 */

#define IGMP_SIZE (sizeof(struct igmphdr)+sizeof(struct iphdr)+4)

static int igmp_send_report(struct device *dev, u32 group, int type)
{
	struct sk_buff *skb;
	struct iphdr *iph;
	struct igmphdr *ih;
	struct rtable *rt;
	u32	dst;

	/* According to IGMPv2 specs, LEAVE messages are
	 * sent to all-routers group.
	 */
	dst = group;
	if (type == IGMP_HOST_LEAVE_MESSAGE)
		dst = IGMP_ALL_ROUTER;

	if (ip_route_output(&rt, dst, 0, 0, dev->ifindex))
		return -1;
	if (rt->rt_src == 0) {
		ip_rt_put(rt);
		return -1;
	}

	skb=alloc_skb(IGMP_SIZE+dev->hard_header_len+15, GFP_ATOMIC);
	if (skb == NULL) {
		ip_rt_put(rt);
		return -1;
	}

	skb->dst = &rt->u.dst;

	skb_reserve(skb, (dev->hard_header_len+15)&~15);
	ip_ll_header(skb);

	skb->nh.iph = iph = (struct iphdr *)skb_put(skb, sizeof(struct iphdr)+4);

	iph->version  = 4;
	iph->ihl      = (sizeof(struct iphdr)+4)>>2;
	iph->tos      = 0;
	iph->frag_off = 0;
	iph->ttl      = 1;
	iph->daddr    = dst;
	iph->saddr    = rt->rt_src;
	iph->protocol = IPPROTO_IGMP;
	iph->tot_len  = htons(IGMP_SIZE);
	iph->id	      = htons(ip_id_count++);
	((u8*)&iph[1])[0] = IPOPT_RA;
	((u8*)&iph[1])[1] = 4;
	((u8*)&iph[1])[2] = 0;
	((u8*)&iph[1])[3] = 0;
	ip_send_check(iph);

	ih = (struct igmphdr *)skb_put(skb, sizeof(struct igmphdr));
	ih->type=type;
	ih->code=0;
	ih->csum=0;
	ih->group=group;
	ih->csum=ip_compute_csum((void *)ih, sizeof(struct igmphdr));

	return skb->dst->output(skb);
}


static void igmp_timer_expire(unsigned long data)
{
	struct ip_mc_list *im=(struct ip_mc_list *)data;
	struct in_device *in_dev = im->interface;
	int err;

	im->tm_running=0;

	if (IGMP_V1_SEEN(in_dev))
		err = igmp_send_report(in_dev->dev, im->multiaddr, IGMP_HOST_MEMBERSHIP_REPORT);
	else
		err = igmp_send_report(in_dev->dev, im->multiaddr, IGMP_HOST_NEW_MEMBERSHIP_REPORT);

	/* Failed. Retry later. */
	if (err) {
		igmp_start_timer(im, IGMP_Unsolicited_Report_Interval);
		return;
	}

	if (im->unsolicit_count) {
		im->unsolicit_count--;
		igmp_start_timer(im, IGMP_Unsolicited_Report_Interval);
	}
	im->reporter = 1;
}

static void igmp_heard_report(struct in_device *in_dev, u32 group)
{
	struct ip_mc_list *im;

	/* Timers are only set for non-local groups */

	if (LOCAL_MCAST(group))
		return;

	for (im=in_dev->mc_list; im!=NULL; im=im->next) {
		if (im->multiaddr == group) {
			igmp_stop_timer(im);
			im->reporter = 0;
			im->unsolicit_count = 0;
			return;
		}
	}
}

static void igmp_heard_query(struct in_device *in_dev, unsigned char max_resp_time,
			     u32 group)
{
	struct ip_mc_list	*im;
	int			max_delay;

	max_delay = max_resp_time*(HZ/IGMP_TIMER_SCALE);

	if (max_resp_time == 0) {
		/* Alas, old v1 router presents here. */

		max_delay = IGMP_Query_Response_Interval;
		in_dev->mr_v1_seen = jiffies + IGMP_V1_Router_Present_Timeout;
		group = 0;
	}
		
	/*
	 * - Start the timers in all of our membership records
	 *   that the query applies to for the interface on
	 *   which the query arrived excl. those that belong
	 *   to a "local" group (224.0.0.X)
	 * - For timers already running check if they need to
	 *   be reset.
	 * - Use the igmp->igmp_code field as the maximum
	 *   delay possible
	 */
	for (im=in_dev->mc_list; im!=NULL; im=im->next) {
		if (group && group != im->multiaddr)
			continue;
		if (LOCAL_MCAST(im->multiaddr))
			continue;
		im->unsolicit_count = 0;
		if (im->tm_running && im->timer.expires-jiffies > max_delay)
			igmp_stop_timer(im);
		igmp_start_timer(im, max_delay);
	}
}

int igmp_rcv(struct sk_buff *skb, unsigned short len)
{
	/* This basically follows the spec line by line -- see RFC1112 */
	struct igmphdr *ih = skb->h.igmph;
	struct in_device *in_dev = skb->dev->ip_ptr;

	if (len < sizeof(struct igmphdr) || ip_compute_csum((void *)ih, len)
	    || in_dev==NULL) {
		kfree_skb(skb, FREE_READ);
		return 0;
	}
	
	switch (ih->type) {
	case IGMP_HOST_MEMBERSHIP_QUERY:
		igmp_heard_query(in_dev, ih->code, ih->group);
		break;
	case IGMP_HOST_MEMBERSHIP_REPORT:
	case IGMP_HOST_NEW_MEMBERSHIP_REPORT:
		/* Is it our report looped back? */
		if (((struct rtable*)skb->dst)->key.iif == 0)
			break;
		igmp_heard_report(in_dev, ih->group);
		break;
	case IGMP_PIM:
#ifdef CONFIG_IP_PIMSM_V1
		return pim_rcv_v1(skb, len);
#endif
	case IGMP_DVMRP:
	case IGMP_TRACE:
	case IGMP_HOST_LEAVE_MESSAGE:
	case IGMP_MTRACE:
	case IGMP_MTRACE_RESP:
		break;
	default:
		NETDEBUG(printk(KERN_DEBUG "New IGMP type=%d, why we do not know about it?\n", ih->type));
	}
	kfree_skb(skb, FREE_READ);
	return 0;
}

#endif

/*
 *	Map a multicast IP onto multicast MAC for type ethernet.
 */

extern __inline__ void ip_mc_map(u32 addr, char *buf)
{
	addr=ntohl(addr);
	buf[0]=0x01;
	buf[1]=0x00;
	buf[2]=0x5e;
	buf[5]=addr&0xFF;
	addr>>=8;
	buf[4]=addr&0xFF;
	addr>>=8;
	buf[3]=addr&0x7F;
}

/*
 *	Add a filter to a device
 */

static void ip_mc_filter_add(struct in_device *in_dev, u32 addr)
{
	char buf[6];
	struct device *dev = in_dev->dev;

	if (!(dev->flags & IFF_MULTICAST))
		return;
	if (dev->type!=ARPHRD_ETHER && dev->type!=ARPHRD_FDDI)
		return; /* Only do ethernet or FDDI for now */
	ip_mc_map(addr, buf);
	dev_mc_add(dev,buf,ETH_ALEN,0);
}

/*
 *	Remove a filter from a device
 */

static void ip_mc_filter_del(struct in_device *in_dev, u32 addr)
{
	char buf[6];
	struct device *dev = in_dev->dev;

	if (dev->type!=ARPHRD_ETHER && dev->type!=ARPHRD_FDDI)
		return; /* Only do ethernet or FDDI for now */
	ip_mc_map(addr,buf);
	dev_mc_delete(dev,buf,ETH_ALEN,0);
}

static void igmp_group_dropped(struct ip_mc_list *im)
{
	ip_mc_filter_del(im->interface, im->multiaddr);

#ifdef CONFIG_IP_MULTICAST
	if (LOCAL_MCAST(im->multiaddr))
		return;

	start_bh_atomic();
	igmp_stop_timer(im);
	end_bh_atomic();

	if (im->reporter && !IGMP_V1_SEEN(im->interface))
		igmp_send_report(im->interface->dev, im->multiaddr, IGMP_HOST_LEAVE_MESSAGE);
#endif
}

static void igmp_group_added(struct ip_mc_list *im)
{
	ip_mc_filter_add(im->interface, im->multiaddr);

#ifdef CONFIG_IP_MULTICAST
	if (LOCAL_MCAST(im->multiaddr))
		return;

	start_bh_atomic();
	igmp_start_timer(im, IGMP_Initial_Report_Delay);
	end_bh_atomic();
#endif
}


/*
 *	Multicast list managers
 */


/*
 *	A socket has joined a multicast group on device dev.
 */

void ip_mc_inc_group(struct in_device *in_dev, u32 addr)
{
	struct ip_mc_list *i, *im;

	im = (struct ip_mc_list *)kmalloc(sizeof(*im), GFP_KERNEL);

	for (i=in_dev->mc_list; i; i=i->next) {
		if (i->multiaddr == addr) {
			i->users++;
			if (im)
				kfree(im);
			return;
		}
	}
	if (!im)
		return;
	im->users=1;
	im->interface=in_dev;
	im->multiaddr=addr;
#ifdef  CONFIG_IP_MULTICAST
	im->tm_running=0;
	init_timer(&im->timer);
	im->timer.data=(unsigned long)im;
	im->timer.function=&igmp_timer_expire;
	im->unsolicit_count = IGMP_Unsolicited_Report_Count;
	im->reporter = 0;
#endif
	im->next=in_dev->mc_list;
	in_dev->mc_list=im;
	if (in_dev->dev->flags & IFF_UP) {
		igmp_group_added(im);
		ip_rt_multicast_event(in_dev);
	}
	return;
}

/*
 *	A socket has left a multicast group on device dev
 */

int ip_mc_dec_group(struct in_device *in_dev, u32 addr)
{
	struct ip_mc_list *i, **ip;

	for (ip=&in_dev->mc_list; (i=*ip)!=NULL; ip=&i->next) {
		if (i->multiaddr==addr) {
			if (--i->users == 0) {
				*ip = i->next;
				if (in_dev->dev->flags & IFF_UP) {
					igmp_group_dropped(i);
					ip_rt_multicast_event(in_dev);
				}
				kfree_s(i, sizeof(*i));
			}
			return 0;
		}
	}
	return -ESRCH;
}

/* Device going down */

void ip_mc_down(struct in_device *in_dev)
{
	struct ip_mc_list *i;

	for (i=in_dev->mc_list; i; i=i->next)
		igmp_group_dropped(i);
}

/* Device going up */

void ip_mc_up(struct in_device *in_dev)
{
	struct ip_mc_list *i;

	for (i=in_dev->mc_list; i; i=i->next)
		igmp_group_added(i);
}

/*
 *	Device is about to be destroyed: clean up.
 */

void ip_mc_destroy_dev(struct in_device *in_dev)
{
	struct ip_mc_list *i;

	while ((i = in_dev->mc_list) != NULL) {
		in_dev->mc_list = i->next;
		kfree_s(i, sizeof(*i));
	}
}

/* Initialize multicasting on an IP interface */

void ip_mc_init_dev(struct in_device *in_dev)
{
	in_dev->mc_list = NULL;
	in_dev->mr_v1_seen = 0;
	ip_mc_inc_group(in_dev, IGMP_ALL_HOSTS);
}

static struct in_device * ip_mc_find_dev(struct ip_mreqn *imr)
{
	struct rtable *rt;
	struct device *dev = NULL;

	if (imr->imr_address.s_addr) {
		dev = ip_dev_find(imr->imr_address.s_addr);
		if (!dev)
			return NULL;
	}

	if (!dev && !ip_route_output(&rt, imr->imr_multiaddr.s_addr, 0, 0, 0)) {
		dev = rt->u.dst.dev;
		ip_rt_put(rt);
	}
	if (dev) {
		imr->imr_ifindex = dev->ifindex;
		return dev->ip_ptr;
	}
	return NULL;
}

/*
 *	Join a socket to a group
 */

int ip_mc_join_group(struct sock *sk , struct ip_mreqn *imr)
{
	int err;
	u32 addr = imr->imr_multiaddr.s_addr;
	struct ip_mc_socklist *iml, *i;
	struct in_device *in_dev;
	int count = 0;

	if (!MULTICAST(addr))
		return -EINVAL;

	rtnl_shlock();

	if (!imr->imr_ifindex)
		in_dev = ip_mc_find_dev(imr);
	else
		in_dev = inetdev_by_index(imr->imr_ifindex);

	if (!in_dev) {
		iml = NULL;
		err = -ENODEV;
		goto done;
	}

	iml = (struct ip_mc_socklist *)kmalloc(sizeof(*iml), GFP_KERNEL);

	err = -EADDRINUSE;
	for (i=sk->ip_mc_list; i; i=i->next) {
		if (memcmp(&i->multi, imr, sizeof(*imr)) == 0) {
			/* New style additions are reference counted */
			if (imr->imr_address.s_addr == 0) {
				i->count++;
				err = 0;
			}
			goto done;
		}
		count++;
	}
	err = -ENOBUFS;
	if (iml == NULL || count >= IP_MAX_MEMBERSHIPS)
		goto done;
	memcpy(&iml->multi, imr, sizeof(*imr));
	iml->next = sk->ip_mc_list;
	iml->count = 1;
	sk->ip_mc_list = iml;
	ip_mc_inc_group(in_dev, addr);
	iml = NULL;
	err = 0;
done:
	rtnl_shunlock();
	if (iml)
		kfree(iml);
	return err;
}

/*
 *	Ask a socket to leave a group.
 */

int ip_mc_leave_group(struct sock *sk, struct ip_mreqn *imr)
{
	struct ip_mc_socklist *iml, **imlp;

	for (imlp=&sk->ip_mc_list; (iml=*imlp)!=NULL; imlp=&iml->next) {
		if (iml->multi.imr_multiaddr.s_addr==imr->imr_multiaddr.s_addr &&
		    iml->multi.imr_address.s_addr==imr->imr_address.s_addr &&
		    (!imr->imr_ifindex || iml->multi.imr_ifindex==imr->imr_ifindex)) {
			struct in_device *in_dev;
			if (--iml->count)
				return 0;
			*imlp = iml->next;
			in_dev = inetdev_by_index(iml->multi.imr_ifindex);
			if (in_dev)
				ip_mc_dec_group(in_dev, imr->imr_multiaddr.s_addr);
			kfree_s(iml, sizeof(*iml));
			return 0;
		}
	}
	return -EADDRNOTAVAIL;
}

/*
 *	A socket is closing.
 */

void ip_mc_drop_socket(struct sock *sk)
{
	struct ip_mc_socklist *iml;

	while ((iml=sk->ip_mc_list) != NULL) {
		struct in_device *in_dev;
		sk->ip_mc_list = iml->next;
		if ((in_dev = inetdev_by_index(iml->multi.imr_ifindex)) != NULL)
			ip_mc_dec_group(in_dev, iml->multi.imr_multiaddr.s_addr);
		kfree_s(iml, sizeof(*iml));
	}
}


#ifdef CONFIG_IP_MULTICAST
 
int ip_mc_procinfo(char *buffer, char **start, off_t offset, int length, int dummy)
{
	off_t pos=0, begin=0;
	struct ip_mc_list *im;
	int len=0;
	struct device *dev;
	
	len=sprintf(buffer,"Idx\tDevice    : Count Querier\tGroup    Users Timer\tReporter\n");  
	
	for(dev = dev_base; dev; dev = dev->next)
	{
		struct in_device *in_dev = dev->ip_ptr;
		char   *querier = "NONE";
		
		if (in_dev == NULL)
			continue;

		querier = IGMP_V1_SEEN(in_dev) ? "V1" : "V2";

		len+=sprintf(buffer+len,"%d\t%-10s: %5d %7s\n",
			     dev->ifindex, dev->name, dev->mc_count, querier);

		for (im = in_dev->mc_list; im; im = im->next) {
			len+=sprintf(buffer+len,
				     "\t\t\t\t%08lX %5d %d:%08lX\t\t%d\n",
				     im->multiaddr, im->users,
				     im->tm_running, im->timer.expires-jiffies, im->reporter);

			pos=begin+len;
			if(pos<offset)
			{
				len=0;
				begin=pos;
			}
			if(pos>offset+length)
				break;
		}
	}
	*start=buffer+(offset-begin);
	len-=(offset-begin);
	if(len>length)
		len=length;
	return len;
}
#endif