From 7bc570c8b4f75ddb3fd5dbeb38127cdc4acbcc9c Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Thu, 3 Apr 2008 09:22:53 +0900 Subject: [PATCH] [IPV6] MROUTE: Support multicast forwarding. Based on ancient patch by Mickael Hoerdt , which is available at . Signed-off-by: YOSHIFUJI Hideaki --- include/linux/Kbuild | 1 + include/linux/ipv6.h | 5 + include/linux/mroute6.h | 227 +++++++ net/ipv6/Kconfig | 7 + net/ipv6/Makefile | 2 + net/ipv6/addrconf.c | 15 +- net/ipv6/af_inet6.c | 6 + net/ipv6/ip6_input.c | 81 ++- net/ipv6/ip6_output.c | 6 +- net/ipv6/ip6mr.c | 1384 ++++++++++++++++++++++++++++++++++++++ net/ipv6/ipv6_sockglue.c | 7 + net/ipv6/raw.c | 7 +- net/ipv6/route.c | 30 +- 13 files changed, 1751 insertions(+), 27 deletions(-) create mode 100644 include/linux/mroute6.h create mode 100644 net/ipv6/ip6mr.c diff --git a/include/linux/Kbuild b/include/linux/Kbuild index 84736acb4b99..29ab9b95d376 100644 --- a/include/linux/Kbuild +++ b/include/linux/Kbuild @@ -261,6 +261,7 @@ unifdef-y += mempolicy.h unifdef-y += mii.h unifdef-y += mman.h unifdef-y += mroute.h +unifdef-y += mroute6.h unifdef-y += msdos_fs.h unifdef-y += msg.h unifdef-y += nbd.h diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h index b90d3d461d4e..f53e4764fc05 100644 --- a/include/linux/ipv6.h +++ b/include/linux/ipv6.h @@ -159,6 +159,9 @@ struct ipv6_devconf { __s32 accept_source_route; #ifdef CONFIG_IPV6_OPTIMISTIC_DAD __s32 optimistic_dad; +#endif +#ifdef CONFIG_IPV6_MROUTE + __s32 mc_forwarding; #endif void *sysctl; }; @@ -190,6 +193,7 @@ enum { DEVCONF_PROXY_NDP, DEVCONF_OPTIMISTIC_DAD, DEVCONF_ACCEPT_SOURCE_ROUTE, + DEVCONF_MC_FORWARDING, DEVCONF_MAX }; @@ -230,6 +234,7 @@ struct inet6_skb_parm { #endif #define IP6SKB_XFRM_TRANSFORMED 1 +#define IP6SKB_FORWARDED 2 }; #define IP6CB(skb) ((struct inet6_skb_parm*)((skb)->cb)) diff --git a/include/linux/mroute6.h b/include/linux/mroute6.h new file mode 100644 index 000000000000..b92190304e0b --- /dev/null +++ b/include/linux/mroute6.h @@ -0,0 +1,227 @@ +#ifndef __LINUX_MROUTE6_H +#define __LINUX_MROUTE6_H + +#include +#include + +/* + * Based on the MROUTING 3.5 defines primarily to keep + * source compatibility with BSD. + * + * See the pim6sd code for the original history. + * + * Protocol Independent Multicast (PIM) data structures included + * Carlos Picoto (cap@di.fc.ul.pt) + * + */ + +#define MRT6_BASE 200 +#define MRT6_INIT (MRT6_BASE) /* Activate the kernel mroute code */ +#define MRT6_DONE (MRT6_BASE+1) /* Shutdown the kernel mroute */ +#define MRT6_ADD_MIF (MRT6_BASE+2) /* Add a virtual interface */ +#define MRT6_DEL_MIF (MRT6_BASE+3) /* Delete a virtual interface */ +#define MRT6_ADD_MFC (MRT6_BASE+4) /* Add a multicast forwarding entry */ +#define MRT6_DEL_MFC (MRT6_BASE+5) /* Delete a multicast forwarding entry */ +#define MRT6_VERSION (MRT6_BASE+6) /* Get the kernel multicast version */ + +#define SIOCGETMIFCNT_IN6 SIOCPROTOPRIVATE /* IP protocol privates */ +#define SIOCGETSGCNT_IN6 (SIOCPROTOPRIVATE+1) +#define SIOCGETRPF (SIOCPROTOPRIVATE+2) + +#define MAXMIFS 32 +typedef unsigned long mifbitmap_t; /* User mode code depends on this lot */ +typedef unsigned short mifi_t; +#define ALL_MIFS ((mifi_t)(-1)) + +#ifndef IF_SETSIZE +#define IF_SETSIZE 256 +#endif + +typedef __u32 if_mask; +#define NIFBITS (sizeof(if_mask) * 8) /* bits per mask */ + +#if !defined(__KERNEL__) && !defined(DIV_ROUND_UP) +#define DIV_ROUND_UP(x,y) (((x) + ((y) - 1)) / (y)) +#endif + +typedef struct if_set { + if_mask ifs_bits[DIV_ROUND_UP(IF_SETSIZE, NIFBITS)]; +} if_set; + +#define IF_SET(n, p) ((p)->ifs_bits[(n)/NIFBITS] |= (1 << ((n) % NIFBITS))) +#define IF_CLR(n, p) ((p)->ifs_bits[(n)/NIFBITS] &= ~(1 << ((n) % NIFBITS))) +#define IF_ISSET(n, p) ((p)->ifs_bits[(n)/NIFBITS] & (1 << ((n) % NIFBITS))) +#define IF_COPY(f, t) bcopy(f, t, sizeof(*(f))) +#define IF_ZERO(p) bzero(p, sizeof(*(p))) + +/* + * Passed by mrouted for an MRT_ADD_MIF - again we use the + * mrouted 3.6 structures for compatibility + */ + +struct mif6ctl { + mifi_t mif6c_mifi; /* Index of MIF */ + unsigned char mif6c_flags; /* MIFF_ flags */ + unsigned char vifc_threshold; /* ttl limit */ + u_short mif6c_pifi; /* the index of the physical IF */ + unsigned int vifc_rate_limit; /* Rate limiter values (NI) */ +}; + +#define MIFF_REGISTER 0x1 /* register vif */ + +/* + * Cache manipulation structures for mrouted and PIMd + */ + +struct mf6cctl +{ + struct sockaddr_in6 mf6cc_origin; /* Origin of mcast */ + struct sockaddr_in6 mf6cc_mcastgrp; /* Group in question */ + mifi_t mf6cc_parent; /* Where it arrived */ + struct if_set mf6cc_ifset; /* Where it is going */ +}; + +/* + * Group count retrieval for pim6sd + */ + +struct sioc_sg_req6 +{ + struct sockaddr_in6 src; + struct sockaddr_in6 grp; + unsigned long pktcnt; + unsigned long bytecnt; + unsigned long wrong_if; +}; + +/* + * To get vif packet counts + */ + +struct sioc_mif_req6 +{ + mifi_t mifi; /* Which iface */ + unsigned long icount; /* In packets */ + unsigned long ocount; /* Out packets */ + unsigned long ibytes; /* In bytes */ + unsigned long obytes; /* Out bytes */ +}; + +/* + * That's all usermode folks + */ + +#ifdef __KERNEL__ + +#include /* for struct sk_buff_head */ + +struct net_device; +struct inet6_dev *ipv6_find_idev(struct net_device *dev); + +#ifdef CONFIG_IPV6_MROUTE +static inline int ip6_mroute_opt(int opt) +{ + return (opt >= MRT6_BASE) && (opt <= MRT6_BASE + 10); +} +#else +static inline int ip6_mroute_opt(int opt) +{ + return 0; +} +#endif + +struct sock; + +extern int ip6_mroute_setsockopt(struct sock *, int, char __user *, int); +extern int ip6_mroute_getsockopt(struct sock *, int, char __user *, int __user *); +extern int ip6_mr_input(struct sk_buff *skb); +extern int ip6mr_ioctl(struct sock *sk, int cmd, void __user *arg); +extern void ip6_mr_init(void); + +struct mif_device +{ + struct net_device *dev; /* Device we are using */ + unsigned long bytes_in,bytes_out; + unsigned long pkt_in,pkt_out; /* Statistics */ + unsigned long rate_limit; /* Traffic shaping (NI) */ + unsigned char threshold; /* TTL threshold */ + unsigned short flags; /* Control flags */ + int link; /* Physical interface index */ +}; + +#define VIFF_STATIC 0x8000 + +struct mfc6_cache +{ + struct mfc6_cache *next; /* Next entry on cache line */ + struct in6_addr mf6c_mcastgrp; /* Group the entry belongs to */ + struct in6_addr mf6c_origin; /* Source of packet */ + mifi_t mf6c_parent; /* Source interface */ + int mfc_flags; /* Flags on line */ + + union { + struct { + unsigned long expires; + struct sk_buff_head unresolved; /* Unresolved buffers */ + } unres; + struct { + unsigned long last_assert; + int minvif; + int maxvif; + unsigned long bytes; + unsigned long pkt; + unsigned long wrong_if; + unsigned char ttls[MAXMIFS]; /* TTL thresholds */ + } res; + } mfc_un; +}; + +#define MFC_STATIC 1 +#define MFC_NOTIFY 2 + +#define MFC6_LINES 64 + +#define MFC6_HASH(a, g) (((__force u32)(a)->s6_addr32[0] ^ \ + (__force u32)(a)->s6_addr32[1] ^ \ + (__force u32)(a)->s6_addr32[2] ^ \ + (__force u32)(a)->s6_addr32[3] ^ \ + (__force u32)(g)->s6_addr32[0] ^ \ + (__force u32)(g)->s6_addr32[1] ^ \ + (__force u32)(g)->s6_addr32[2] ^ \ + (__force u32)(g)->s6_addr32[3]) % MFC6_LINES) + +#define MFC_ASSERT_THRESH (3*HZ) /* Maximal freq. of asserts */ + +#endif + +#ifdef __KERNEL__ +struct rtmsg; +extern int ip6mr_get_route(struct sk_buff *skb, struct rtmsg *rtm, int nowait); + +#ifdef CONFIG_IPV6_MROUTE +extern struct sock *mroute6_socket; +extern int ip6mr_sk_done(struct sock *sk); +#else +#define mroute6_socket NULL +static inline int ip6mr_sk_done(struct sock *sk) { return 0; } +#endif +#endif + +/* + * Structure used to communicate from kernel to multicast router. + * We'll overlay the structure onto an MLD header (not an IPv6 heder like igmpmsg{} + * used for IPv4 implementation). This is because this structure will be passed via an + * IPv6 raw socket, on wich an application will only receiver the payload i.e the data after + * the IPv6 header and all the extension headers. (See section 3 of RFC 3542) + */ + +struct mrt6msg { +#define MRT6MSG_NOCACHE 1 + __u8 im6_mbz; /* must be zero */ + __u8 im6_msgtype; /* what type of message */ + __u16 im6_mif; /* mif rec'd on */ + __u32 im6_pad; /* padding for 64 bit arch */ + struct in6_addr im6_src, im6_dst; +}; + +#endif diff --git a/net/ipv6/Kconfig b/net/ipv6/Kconfig index 47263e45bacb..9a2ea81e499f 100644 --- a/net/ipv6/Kconfig +++ b/net/ipv6/Kconfig @@ -209,3 +209,10 @@ config IPV6_SUBTREES If unsure, say N. +config IPV6_MROUTE + bool "IPv6: multicast routing (EXPERIMENTAL)" + depends on IPV6 && EXPERIMENTAL + ---help--- + Experimental support for IPv6 multicast forwarding. + If unsure, say N. + diff --git a/net/ipv6/Makefile b/net/ipv6/Makefile index ae14617e607f..686934acfac1 100644 --- a/net/ipv6/Makefile +++ b/net/ipv6/Makefile @@ -11,6 +11,8 @@ ipv6-objs := af_inet6.o anycast.o ip6_output.o ip6_input.o addrconf.o \ exthdrs.o datagram.o ip6_flowlabel.o inet6_connection_sock.o ipv6-$(CONFIG_SYSCTL) = sysctl_net_ipv6.o +ipv6-$(CONFIG_IPV6_MROUTE) += ip6mr.o + ipv6-$(CONFIG_XFRM) += xfrm6_policy.o xfrm6_state.o xfrm6_input.o \ xfrm6_output.o ipv6-$(CONFIG_NETFILTER) += netfilter.o diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 11037615bc73..dbc51af69017 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -412,7 +412,7 @@ static struct inet6_dev * ipv6_add_dev(struct net_device *dev) return ndev; } -static struct inet6_dev * ipv6_find_idev(struct net_device *dev) +struct inet6_dev * ipv6_find_idev(struct net_device *dev) { struct inet6_dev *idev; @@ -3547,6 +3547,9 @@ static inline void ipv6_store_devconf(struct ipv6_devconf *cnf, #ifdef CONFIG_IPV6_OPTIMISTIC_DAD array[DEVCONF_OPTIMISTIC_DAD] = cnf->optimistic_dad; #endif +#ifdef CONFIG_IPV6_MROUTE + array[DEVCONF_MC_FORWARDING] = cnf->mc_forwarding; +#endif } static inline size_t inet6_if_nlmsg_size(void) @@ -4094,6 +4097,16 @@ static struct addrconf_sysctl_table .proc_handler = &proc_dointvec, }, +#endif +#ifdef CONFIG_IPV6_MROUTE + { + .ctl_name = CTL_UNNUMBERED, + .procname = "mc_forwarding", + .data = &ipv6_devconf.mc_forwarding, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, #endif { .ctl_name = 0, /* sentinel */ diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index 1731b0abf7f5..3c6aafb02183 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -61,6 +61,9 @@ #include #include +#ifdef CONFIG_IPV6_MROUTE +#include +#endif MODULE_AUTHOR("Cast of dozens"); MODULE_DESCRIPTION("IPv6 protocol stack for Linux"); @@ -953,6 +956,9 @@ static int __init inet6_init(void) err = icmpv6_init(); if (err) goto icmp_fail; +#ifdef CONFIG_IPV6_MROUTE + ip6_mr_init(); +#endif err = ndisc_init(); if (err) goto ndisc_fail; diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c index 43a617e2268b..09a3201e408a 100644 --- a/net/ipv6/ip6_input.c +++ b/net/ipv6/ip6_input.c @@ -29,6 +29,7 @@ #include #include #include +#include #include #include @@ -236,36 +237,84 @@ int ip6_mc_input(struct sk_buff *skb) hdr = ipv6_hdr(skb); deliver = ipv6_chk_mcast_addr(skb->dev, &hdr->daddr, NULL); +#ifdef CONFIG_IPV6_MROUTE /* - * IPv6 multicast router mode isnt currently supported. + * IPv6 multicast router mode is now supported ;) */ -#if 0 - if (ipv6_config.multicast_route) { - int addr_type; + if (ipv6_devconf.mc_forwarding && + likely(!(IP6CB(skb)->flags & IP6SKB_FORWARDED))) { + /* + * Okay, we try to forward - split and duplicate + * packets. + */ + struct sk_buff *skb2; + struct inet6_skb_parm *opt = IP6CB(skb); - addr_type = ipv6_addr_type(&hdr->daddr); + /* Check for MLD */ + if (unlikely(opt->ra)) { + /* Check if this is a mld message */ + u8 *ptr = skb_network_header(skb) + opt->ra; + struct icmp6hdr *icmp6; + u8 nexthdr = hdr->nexthdr; + int offset; - if (!(addr_type & (IPV6_ADDR_LOOPBACK | IPV6_ADDR_LINKLOCAL))) { - struct sk_buff *skb2; - struct dst_entry *dst; + /* Check if the value of Router Alert + * is for MLD (0x0000). + */ + if ((ptr[2] | ptr[3]) == 0) { + if (!ipv6_ext_hdr(nexthdr)) { + /* BUG */ + goto discard; + } + offset = ipv6_skip_exthdr(skb, sizeof(*hdr), + &nexthdr); + if (offset < 0) + goto discard; - dst = skb->dst; + if (nexthdr != IPPROTO_ICMPV6) + goto discard; - if (deliver) { - skb2 = skb_clone(skb, GFP_ATOMIC); - dst_output(skb2); - } else { - dst_output(skb); - return 0; + if (!pskb_may_pull(skb, (skb_network_header(skb) + + offset + 1 - skb->data))) + goto discard; + + icmp6 = (struct icmp6hdr *)(skb_network_header(skb) + offset); + + switch (icmp6->icmp6_type) { + case ICMPV6_MGM_QUERY: + case ICMPV6_MGM_REPORT: + case ICMPV6_MGM_REDUCTION: + case ICMPV6_MLD2_REPORT: + break; + default: + /* Bogus */ + goto discard; + } + deliver = 1; + goto out; } + /* unknown RA - process it normally */ + } + + if (deliver) + skb2 = skb_clone(skb, GFP_ATOMIC); + else { + skb2 = skb; + skb = NULL; + } + + if (skb2) { + skb2->dev = skb2->dst->dev; + ip6_mr_input(skb2); } } #endif - +out: if (likely(deliver)) { ip6_input(skb); return 0; } +discard: /* discard */ kfree_skb(skb); diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index a8b4da25b0a7..c0dbe549cc42 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -55,6 +55,7 @@ #include #include #include +#include static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)); @@ -137,8 +138,9 @@ static int ip6_output2(struct sk_buff *skb) struct inet6_dev *idev = ip6_dst_idev(skb->dst); if (!(dev->flags & IFF_LOOPBACK) && (!np || np->mc_loop) && - ipv6_chk_mcast_addr(dev, &ipv6_hdr(skb)->daddr, - &ipv6_hdr(skb)->saddr)) { + ((mroute6_socket && !(IP6CB(skb)->flags & IP6SKB_FORWARDED)) || + ipv6_chk_mcast_addr(dev, &ipv6_hdr(skb)->daddr, + &ipv6_hdr(skb)->saddr))) { struct sk_buff *newskb = skb_clone(skb, GFP_ATOMIC); /* Do not check for IFF_ALLMULTI; multicast routing diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c new file mode 100644 index 000000000000..1bdf3c177d58 --- /dev/null +++ b/net/ipv6/ip6mr.c @@ -0,0 +1,1384 @@ +/* + * Linux IPv6 multicast routing support for BSD pim6sd + * Based on net/ipv4/ipmr.c. + * + * (c) 2004 Mickael Hoerdt, + * LSIIT Laboratory, Strasbourg, France + * (c) 2004 Jean-Philippe Andriot, + * 6WIND, Paris, France + * Copyright (C)2007,2008 USAGI/WIDE Project + * YOSHIFUJI Hideaki + * + * 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 +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include + +struct sock *mroute6_socket; + + +/* Big lock, protecting vif table, mrt cache and mroute socket state. + Note that the changes are semaphored via rtnl_lock. + */ + +static DEFINE_RWLOCK(mrt_lock); + +/* + * Multicast router control variables + */ + +static struct mif_device vif6_table[MAXMIFS]; /* Devices */ +static int maxvif; + +#define MIF_EXISTS(idx) (vif6_table[idx].dev != NULL) + +static struct mfc6_cache *mfc6_cache_array[MFC_LINES]; /* Forwarding cache */ + +static struct mfc6_cache *mfc_unres_queue; /* Queue of unresolved entries */ +static atomic_t cache_resolve_queue_len; /* Size of unresolved */ + +/* Special spinlock for queue of unresolved entries */ +static DEFINE_SPINLOCK(mfc_unres_lock); + +/* We return to original Alan's scheme. Hash table of resolved + entries is changed only in process context and protected + with weak lock mrt_lock. Queue of unresolved entries is protected + with strong spinlock mfc_unres_lock. + + In this case data path is free of exclusive locks at all. + */ + +static struct kmem_cache *mrt_cachep __read_mostly; + +static int ip6_mr_forward(struct sk_buff *skb, struct mfc6_cache *cache); +static int ip6mr_cache_report(struct sk_buff *pkt, vifi_t vifi, int assert); +static int ip6mr_fill_mroute(struct sk_buff *skb, struct mfc6_cache *c, struct rtmsg *rtm); + +static struct timer_list ipmr_expire_timer; + + +#ifdef CONFIG_PROC_FS + +struct ipmr_mfc_iter { + struct mfc6_cache **cache; + int ct; +}; + + +static struct mfc6_cache *ipmr_mfc_seq_idx(struct ipmr_mfc_iter *it, loff_t pos) +{ + struct mfc6_cache *mfc; + + it->cache = mfc6_cache_array; + read_lock(&mrt_lock); + for (it->ct = 0; it->ct < ARRAY_SIZE(mfc6_cache_array); it->ct++) + for (mfc = mfc6_cache_array[it->ct]; mfc; mfc = mfc->next) + if (pos-- == 0) + return mfc; + read_unlock(&mrt_lock); + + it->cache = &mfc_unres_queue; + spin_lock_bh(&mfc_unres_lock); + for (mfc = mfc_unres_queue; mfc; mfc = mfc->next) + if (pos-- == 0) + return mfc; + spin_unlock_bh(&mfc_unres_lock); + + it->cache = NULL; + return NULL; +} + + + + +/* + * The /proc interfaces to multicast routing /proc/ip6_mr_cache /proc/ip6_mr_vif + */ + +struct ipmr_vif_iter { + int ct; +}; + +static struct mif_device *ip6mr_vif_seq_idx(struct ipmr_vif_iter *iter, + loff_t pos) +{ + for (iter->ct = 0; iter->ct < maxvif; ++iter->ct) { + if (!MIF_EXISTS(iter->ct)) + continue; + if (pos-- == 0) + return &vif6_table[iter->ct]; + } + return NULL; +} + +static void *ip6mr_vif_seq_start(struct seq_file *seq, loff_t *pos) + __acquires(mrt_lock) +{ + read_lock(&mrt_lock); + return (*pos ? ip6mr_vif_seq_idx(seq->private, *pos - 1) + : SEQ_START_TOKEN); +} + +static void *ip6mr_vif_seq_next(struct seq_file *seq, void *v, loff_t *pos) +{ + struct ipmr_vif_iter *iter = seq->private; + + ++*pos; + if (v == SEQ_START_TOKEN) + return ip6mr_vif_seq_idx(iter, 0); + + while (++iter->ct < maxvif) { + if (!MIF_EXISTS(iter->ct)) + continue; + return &vif6_table[iter->ct]; + } + return NULL; +} + +static void ip6mr_vif_seq_stop(struct seq_file *seq, void *v) + __releases(mrt_lock) +{ + read_unlock(&mrt_lock); +} + +static int ip6mr_vif_seq_show(struct seq_file *seq, void *v) +{ + if (v == SEQ_START_TOKEN) { + seq_puts(seq, + "Interface BytesIn PktsIn BytesOut PktsOut Flags\n"); + } else { + const struct mif_device *vif = v; + const char *name = vif->dev ? vif->dev->name : "none"; + + seq_printf(seq, + "%2Zd %-10s %8ld %7ld %8ld %7ld %05X\n", + vif - vif6_table, + name, vif->bytes_in, vif->pkt_in, + vif->bytes_out, vif->pkt_out, + vif->flags); + } + return 0; +} + +static struct seq_operations ip6mr_vif_seq_ops = { + .start = ip6mr_vif_seq_start, + .next = ip6mr_vif_seq_next, + .stop = ip6mr_vif_seq_stop, + .show = ip6mr_vif_seq_show, +}; + +static int ip6mr_vif_open(struct inode *inode, struct file *file) +{ + return seq_open_private(file, &ip6mr_vif_seq_ops, + sizeof(struct ipmr_vif_iter)); +} + +static struct file_operations ip6mr_vif_fops = { + .owner = THIS_MODULE, + .open = ip6mr_vif_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; + +static void *ipmr_mfc_seq_start(struct seq_file *seq, loff_t *pos) +{ + return (*pos ? ipmr_mfc_seq_idx(seq->private, *pos - 1) + : SEQ_START_TOKEN); +} + +static void *ipmr_mfc_seq_next(struct seq_file *seq, void *v, loff_t *pos) +{ + struct mfc6_cache *mfc = v; + struct ipmr_mfc_iter *it = seq->private; + + ++*pos; + + if (v == SEQ_START_TOKEN) + return ipmr_mfc_seq_idx(seq->private, 0); + + if (mfc->next) + return mfc->next; + + if (it->cache == &mfc_unres_queue) + goto end_of_list; + + BUG_ON(it->cache != mfc6_cache_array); + + while (++it->ct < ARRAY_SIZE(mfc6_cache_array)) { + mfc = mfc6_cache_array[it->ct]; + if (mfc) + return mfc; + } + + /* exhausted cache_array, show unresolved */ + read_unlock(&mrt_lock); + it->cache = &mfc_unres_queue; + it->ct = 0; + + spin_lock_bh(&mfc_unres_lock); + mfc = mfc_unres_queue; + if (mfc) + return mfc; + + end_of_list: + spin_unlock_bh(&mfc_unres_lock); + it->cache = NULL; + + return NULL; +} + +static void ipmr_mfc_seq_stop(struct seq_file *seq, void *v) +{ + struct ipmr_mfc_iter *it = seq->private; + + if (it->cache == &mfc_unres_queue) + spin_unlock_bh(&mfc_unres_lock); + else if (it->cache == mfc6_cache_array) + read_unlock(&mrt_lock); +} + +static int ipmr_mfc_seq_show(struct seq_file *seq, void *v) +{ + int n; + + if (v == SEQ_START_TOKEN) { + seq_puts(seq, + "Group " + "Origin " + "Iif Pkts Bytes Wrong Oifs\n"); + } else { + const struct mfc6_cache *mfc = v; + const struct ipmr_mfc_iter *it = seq->private; + + seq_printf(seq, + NIP6_FMT " " NIP6_FMT " %-3d %8ld %8ld %8ld", + NIP6(mfc->mf6c_mcastgrp), NIP6(mfc->mf6c_origin), + mfc->mf6c_parent, + mfc->mfc_un.res.pkt, + mfc->mfc_un.res.bytes, + mfc->mfc_un.res.wrong_if); + + if (it->cache != &mfc_unres_queue) { + for (n = mfc->mfc_un.res.minvif; + n < mfc->mfc_un.res.maxvif; n++) { + if (MIF_EXISTS(n) && + mfc->mfc_un.res.ttls[n] < 255) + seq_printf(seq, + " %2d:%-3d", + n, mfc->mfc_un.res.ttls[n]); + } + } + seq_putc(seq, '\n'); + } + return 0; +} + +static struct seq_operations ipmr_mfc_seq_ops = { + .start = ipmr_mfc_seq_start, + .next = ipmr_mfc_seq_next, + .stop = ipmr_mfc_seq_stop, + .show = ipmr_mfc_seq_show, +}; + +static int ipmr_mfc_open(struct inode *inode, struct file *file) +{ + return seq_open_private(file, &ipmr_mfc_seq_ops, + sizeof(struct ipmr_mfc_iter)); +} + +static struct file_operations ip6mr_mfc_fops = { + .owner = THIS_MODULE, + .open = ipmr_mfc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; +#endif + +/* + * Delete a VIF entry + */ + +static int mif6_delete(int vifi) +{ + struct mif_device *v; + struct net_device *dev; + if (vifi < 0 || vifi >= maxvif) + return -EADDRNOTAVAIL; + + v = &vif6_table[vifi]; + + write_lock_bh(&mrt_lock); + dev = v->dev; + v->dev = NULL; + + if (!dev) { + write_unlock_bh(&mrt_lock); + return -EADDRNOTAVAIL; + } + + if (vifi + 1 == maxvif) { + int tmp; + for (tmp = vifi - 1; tmp >= 0; tmp--) { + if (MIF_EXISTS(tmp)) + break; + } + maxvif = tmp + 1; + } + + write_unlock_bh(&mrt_lock); + + dev_set_allmulti(dev, -1); + + if (v->flags & MIFF_REGISTER) + unregister_netdevice(dev); + + dev_put(dev); + return 0; +} + +/* Destroy an unresolved cache entry, killing queued skbs + and reporting error to netlink readers. + */ + +static void ip6mr_destroy_unres(struct mfc6_cache *c) +{ + struct sk_buff *skb; + + atomic_dec(&cache_resolve_queue_len); + + while((skb = skb_dequeue(&c->mfc_un.unres.unresolved)) != NULL) { + if (ipv6_hdr(skb)->version == 0) { + struct nlmsghdr *nlh = (struct nlmsghdr *)skb_pull(skb, sizeof(struct ipv6hdr)); + nlh->nlmsg_type = NLMSG_ERROR; + nlh->nlmsg_len = NLMSG_LENGTH(sizeof(struct nlmsgerr)); + skb_trim(skb, nlh->nlmsg_len); + ((struct nlmsgerr *)NLMSG_DATA(nlh))->error = -ETIMEDOUT; + rtnl_unicast(skb, &init_net, NETLINK_CB(skb).pid); + } else + kfree_skb(skb); + } + + kmem_cache_free(mrt_cachep, c); +} + + +/* Single timer process for all the unresolved queue. */ + +static void ipmr_do_expire_process(unsigned long dummy) +{ + unsigned long now = jiffies; + unsigned long expires = 10 * HZ; + struct mfc6_cache *c, **cp; + + cp = &mfc_unres_queue; + + while ((c = *cp) != NULL) { + if (time_after(c->mfc_un.unres.expires, now)) { + /* not yet... */ + unsigned long interval = c->mfc_un.unres.expires - now; + if (interval < expires) + expires = interval; + cp = &c->next; + continue; + } + + *cp = c->next; + ip6mr_destroy_unres(c); + } + + if (atomic_read(&cache_resolve_queue_len)) + mod_timer(&ipmr_expire_timer, jiffies + expires); +} + +static void ipmr_expire_process(unsigned long dummy) +{ + if (!spin_trylock(&mfc_unres_lock)) { + mod_timer(&ipmr_expire_timer, jiffies + 1); + return; + } + + if (atomic_read(&cache_resolve_queue_len)) + ipmr_do_expire_process(dummy); + + spin_unlock(&mfc_unres_lock); +} + +/* Fill oifs list. It is called under write locked mrt_lock. */ + +static void ip6mr_update_thresholds(struct mfc6_cache *cache, unsigned char *ttls) +{ + int vifi; + + cache->mfc_un.res.minvif = MAXVIFS; + cache->mfc_un.res.maxvif = 0; + memset(cache->mfc_un.res.ttls, 255, MAXVIFS); + + for (vifi = 0; vifi < maxvif; vifi++) { + if (MIF_EXISTS(vifi) && ttls[vifi] && ttls[vifi] < 255) { + cache->mfc_un.res.ttls[vifi] = ttls[vifi]; + if (cache->mfc_un.res.minvif > vifi) + cache->mfc_un.res.minvif = vifi; + if (cache->mfc_un.res.maxvif <= vifi) + cache->mfc_un.res.maxvif = vifi + 1; + } + } +} + +static int mif6_add(struct mif6ctl *vifc, int mrtsock) +{ + int vifi = vifc->mif6c_mifi; + struct mif_device *v = &vif6_table[vifi]; + struct net_device *dev; + + /* Is vif busy ? */ + if (MIF_EXISTS(vifi)) + return -EADDRINUSE; + + switch (vifc->mif6c_flags) { + case 0: + dev = dev_get_by_index(&init_net, vifc->mif6c_pifi); + if (!dev) + return -EADDRNOTAVAIL; + dev_put(dev); + break; + default: + return -EINVAL; + } + + dev_set_allmulti(dev, 1); + + /* + * Fill in the VIF structures + */ + v->rate_limit = vifc->vifc_rate_limit; + v->flags = vifc->mif6c_flags; + if (!mrtsock) + v->flags |= VIFF_STATIC; + v->threshold = vifc->vifc_threshold; + v->bytes_in = 0; + v->bytes_out = 0; + v->pkt_in = 0; + v->pkt_out = 0; + v->link = dev->ifindex; + if (v->flags & MIFF_REGISTER) + v->link = dev->iflink; + + /* And finish update writing critical data */ + write_lock_bh(&mrt_lock); + dev_hold(dev); + v->dev = dev; + if (vifi + 1 > maxvif) + maxvif = vifi + 1; + write_unlock_bh(&mrt_lock); + return 0; +} + +static struct mfc6_cache *ip6mr_cache_find(struct in6_addr *origin, struct in6_addr *mcastgrp) +{ + int line = MFC6_HASH(mcastgrp, origin); + struct mfc6_cache *c; + + for (c = mfc6_cache_array[line]; c; c = c->next) { + if (ipv6_addr_equal(&c->mf6c_origin, origin) && + ipv6_addr_equal(&c->mf6c_mcastgrp, mcastgrp)) + break; + } + return c; +} + +/* + * Allocate a multicast cache entry + */ +static struct mfc6_cache *ip6mr_cache_alloc(void) +{ + struct mfc6_cache *c = kmem_cache_alloc(mrt_cachep, GFP_KERNEL); + if (c == NULL) + return NULL; + memset(c, 0, sizeof(*c)); + c->mfc_un.res.minvif = MAXVIFS; + return c; +} + +static struct mfc6_cache *ip6mr_cache_alloc_unres(void) +{ + struct mfc6_cache *c = kmem_cache_alloc(mrt_cachep, GFP_ATOMIC); + if (c == NULL) + return NULL; + memset(c, 0, sizeof(*c)); + skb_queue_head_init(&c->mfc_un.unres.unresolved); + c->mfc_un.unres.expires = jiffies + 10 * HZ; + return c; +} + +/* + * A cache entry has gone into a resolved state from queued + */ + +static void ip6mr_cache_resolve(struct mfc6_cache *uc, struct mfc6_cache *c) +{ + struct sk_buff *skb; + + /* + * Play the pending entries through our router + */ + + while((skb = __skb_dequeue(&uc->mfc_un.unres.unresolved))) { + if (ipv6_hdr(skb)->version == 0) { + int err; + struct nlmsghdr *nlh = (struct nlmsghdr *)skb_pull(skb, sizeof(struct ipv6hdr)); + + if (ip6mr_fill_mroute(skb, c, NLMSG_DATA(nlh)) > 0) { + nlh->nlmsg_len = skb->tail - (u8 *)nlh; + } else { + nlh->nlmsg_type = NLMSG_ERROR; + nlh->nlmsg_len = NLMSG_LENGTH(sizeof(struct nlmsgerr)); + skb_trim(skb, nlh->nlmsg_len); + ((struct nlmsgerr *)NLMSG_DATA(nlh))->error = -EMSGSIZE; + } + err = rtnl_unicast(skb, &init_net, NETLINK_CB(skb).pid); + } else + ip6_mr_forward(skb, c); + } +} + +/* + * Bounce a cache query up to pim6sd. We could use netlink for this but pim6sd + * expects the following bizarre scheme. + * + * Called under mrt_lock. + */ + +static int ip6mr_cache_report(struct sk_buff *pkt, vifi_t vifi, int assert) +{ + struct sk_buff *skb; + struct mrt6msg *msg; + int ret; + + skb = alloc_skb(sizeof(struct ipv6hdr) + sizeof(*msg), GFP_ATOMIC); + + if (!skb) + return -ENOBUFS; + + /* I suppose that internal messages + * do not require checksums */ + + skb->ip_summed = CHECKSUM_UNNECESSARY; + + /* + * Copy the IP header + */ + + skb_put(skb, sizeof(struct ipv6hdr)); + skb_reset_network_header(skb); + skb_copy_to_linear_data(skb, ipv6_hdr(pkt), sizeof(struct ipv6hdr)); + + /* + * Add our header + */ + skb_put(skb, sizeof(*msg)); + skb_reset_transport_header(skb); + msg = (struct mrt6msg *)skb_transport_header(skb); + + msg->im6_mbz = 0; + msg->im6_msgtype = assert; + msg->im6_mif = vifi; + msg->im6_pad = 0; + ipv6_addr_copy(&msg->im6_src, &ipv6_hdr(pkt)->saddr); + ipv6_addr_copy(&msg->im6_dst, &ipv6_hdr(pkt)->daddr); + + skb->dst = dst_clone(pkt->dst); + skb->ip_summed = CHECKSUM_UNNECESSARY; + + skb_pull(skb, sizeof(struct ipv6hdr)); + + if (mroute6_socket == NULL) { + kfree_skb(skb); + return -EINVAL; + } + + /* + * Deliver to user space multicast routing algorithms + */ + if ((ret = sock_queue_rcv_skb(mroute6_socket, skb)) < 0) { + if (net_ratelimit()) + printk(KERN_WARNING "mroute6: pending queue full, dropping entries.\n"); + kfree_skb(skb); + } + + return ret; +} + +/* + * Queue a packet for resolution. It gets locked cache entry! + */ + +static int +ip6mr_cache_unresolved(vifi_t vifi, struct sk_buff *skb) +{ + int err; + struct mfc6_cache *c; + + spin_lock_bh(&mfc_unres_lock); + for (c = mfc_unres_queue; c; c = c->next) { + if (ipv6_addr_equal(&c->mf6c_mcastgrp, &ipv6_hdr(skb)->daddr) && + ipv6_addr_equal(&c->mf6c_origin, &ipv6_hdr(skb)->saddr)) + break; + } + + if (c == NULL) { + /* + * Create a new entry if allowable + */ + + if (atomic_read(&cache_resolve_queue_len) >= 10 || + (c = ip6mr_cache_alloc_unres()) == NULL) { + spin_unlock_bh(&mfc_unres_lock); + + kfree_skb(skb); + return -ENOBUFS; + } + + /* + * Fill in the new cache entry + */ + c->mf6c_parent = -1; + c->mf6c_origin = ipv6_hdr(skb)->saddr; + c->mf6c_mcastgrp = ipv6_hdr(skb)->daddr; + + /* + * Reflect first query at pim6sd + */ + if ((err = ip6mr_cache_report(skb, vifi, MRT6MSG_NOCACHE)) < 0) { + /* If the report failed throw the cache entry + out - Brad Parker + */ + spin_unlock_bh(&mfc_unres_lock); + + kmem_cache_free(mrt_cachep, c); + kfree_skb(skb); + return err; + } + + atomic_inc(&cache_resolve_queue_len); + c->next = mfc_unres_queue; + mfc_unres_queue = c; + + ipmr_do_expire_process(1); + } + + /* + * See if we can append the packet + */ + if (c->mfc_un.unres.unresolved.qlen > 3) { + kfree_skb(skb); + err = -ENOBUFS; + } else { + skb_queue_tail(&c->mfc_un.unres.unresolved, skb); + err = 0; + } + + spin_unlock_bh(&mfc_unres_lock); + return err; +} + +/* + * MFC6 cache manipulation by user space + */ + +static int ip6mr_mfc_delete(struct mf6cctl *mfc) +{ + int line; + struct mfc6_cache *c, **cp; + + line = MFC6_HASH(&mfc->mf6cc_mcastgrp.sin6_addr, &mfc->mf6cc_origin.sin6_addr); + + for (cp = &mfc6_cache_array[line]; (c = *cp) != NULL; cp = &c->next) { + if (ipv6_addr_equal(&c->mf6c_origin, &mfc->mf6cc_origin.sin6_addr) && + ipv6_addr_equal(&c->mf6c_mcastgrp, &mfc->mf6cc_mcastgrp.sin6_addr)) { + write_lock_bh(&mrt_lock); + *cp = c->next; + write_unlock_bh(&mrt_lock); + + kmem_cache_free(mrt_cachep, c); + return 0; + } + } + return -ENOENT; +} + +static int ip6mr_device_event(struct notifier_block *this, + unsigned long event, void *ptr) +{ + struct net_device *dev = ptr; + struct mif_device *v; + int ct; + + if (dev_net(dev) != &init_net) + return NOTIFY_DONE; + + if (event != NETDEV_UNREGISTER) + return NOTIFY_DONE; + + v = &vif6_table[0]; + for (ct = 0; ct < maxvif; ct++, v++) { + if (v->dev == dev) + mif6_delete(ct); + } + return NOTIFY_DONE; +} + +static struct notifier_block ip6_mr_notifier = { + .notifier_call = ip6mr_device_event +}; + +/* + * Setup for IP multicast routing + */ + +void __init ip6_mr_init(void) +{ + mrt_cachep = kmem_cache_create("ip6_mrt_cache", + sizeof(struct mfc6_cache), + 0, SLAB_HWCACHE_ALIGN, + NULL); + if (!mrt_cachep) + panic("cannot allocate ip6_mrt_cache"); + + setup_timer(&ipmr_expire_timer, ipmr_expire_process, 0); + register_netdevice_notifier(&ip6_mr_notifier); +#ifdef CONFIG_PROC_FS + proc_net_fops_create(&init_net, "ip6_mr_vif", 0, &ip6mr_vif_fops); + proc_net_fops_create(&init_net, "ip6_mr_cache", 0, &ip6mr_mfc_fops); +#endif +} + + +static int ip6mr_mfc_add(struct mf6cctl *mfc, int mrtsock) +{ + int line; + struct mfc6_cache *uc, *c, **cp; + unsigned char ttls[MAXVIFS]; + int i; + + memset(ttls, 255, MAXVIFS); + for (i = 0; i < MAXVIFS; i++) { + if (IF_ISSET(i, &mfc->mf6cc_ifset)) + ttls[i] = 1; + + } + + line = MFC6_HASH(&mfc->mf6cc_mcastgrp.sin6_addr, &mfc->mf6cc_origin.sin6_addr); + + for (cp = &mfc6_cache_array[line]; (c = *cp) != NULL; cp = &c->next) { + if (ipv6_addr_equal(&c->mf6c_origin, &mfc->mf6cc_origin.sin6_addr) && + ipv6_addr_equal(&c->mf6c_mcastgrp, &mfc->mf6cc_mcastgrp.sin6_addr)) + break; + } + + if (c != NULL) { + write_lock_bh(&mrt_lock); + c->mf6c_parent = mfc->mf6cc_parent; + ip6mr_update_thresholds(c, ttls); + if (!mrtsock) + c->mfc_flags |= MFC_STATIC; + write_unlock_bh(&mrt_lock); + return 0; + } + + if (!ipv6_addr_is_multicast(&mfc->mf6cc_mcastgrp.sin6_addr)) + return -EINVAL; + + c = ip6mr_cache_alloc(); + if (c == NULL) + return -ENOMEM; + + c->mf6c_origin = mfc->mf6cc_origin.sin6_addr; + c->mf6c_mcastgrp = mfc->mf6cc_mcastgrp.sin6_addr; + c->mf6c_parent = mfc->mf6cc_parent; + ip6mr_update_thresholds(c, ttls); + if (!mrtsock) + c->mfc_flags |= MFC_STATIC; + + write_lock_bh(&mrt_lock); + c->next = mfc6_cache_array[line]; + mfc6_cache_array[line] = c; + write_unlock_bh(&mrt_lock); + + /* + * Check to see if we resolved a queued list. If so we + * need to send on the frames and tidy up. + */ + spin_lock_bh(&mfc_unres_lock); + for (cp = &mfc_unres_queue; (uc = *cp) != NULL; + cp = &uc->next) { + if (ipv6_addr_equal(&uc->mf6c_origin, &c->mf6c_origin) && + ipv6_addr_equal(&uc->mf6c_mcastgrp, &c->mf6c_mcastgrp)) { + *cp = uc->next; + if (atomic_dec_and_test(&cache_resolve_queue_len)) + del_timer(&ipmr_expire_timer); + break; + } + } + spin_unlock_bh(&mfc_unres_lock); + + if (uc) { + ip6mr_cache_resolve(uc, c); + kmem_cache_free(mrt_cachep, uc); + } + return 0; +} + +/* + * Close the multicast socket, and clear the vif tables etc + */ + +static void mroute_clean_tables(struct sock *sk) +{ + int i; + + /* + * Shut down all active vif entries + */ + for (i = 0; i < maxvif; i++) { + if (!(vif6_table[i].flags & VIFF_STATIC)) + mif6_delete(i); + } + + /* + * Wipe the cache + */ + for (i = 0; i < ARRAY_SIZE(mfc6_cache_array); i++) { + struct mfc6_cache *c, **cp; + + cp = &mfc6_cache_array[i]; + while ((c = *cp) != NULL) { + if (c->mfc_flags & MFC_STATIC) { + cp = &c->next; + continue; + } + write_lock_bh(&mrt_lock); + *cp = c->next; + write_unlock_bh(&mrt_lock); + + kmem_cache_free(mrt_cachep, c); + } + } + + if (atomic_read(&cache_resolve_queue_len) != 0) { + struct mfc6_cache *c; + + spin_lock_bh(&mfc_unres_lock); + while (mfc_unres_queue != NULL) { + c = mfc_unres_queue; + mfc_unres_queue = c->next; + spin_unlock_bh(&mfc_unres_lock); + + ip6mr_destroy_unres(c); + + spin_lock_bh(&mfc_unres_lock); + } + spin_unlock_bh(&mfc_unres_lock); + } +} + +static int ip6mr_sk_init(struct sock *sk) +{ + int err = 0; + + rtnl_lock(); + write_lock_bh(&mrt_lock); + if (likely(mroute6_socket == NULL)) + mroute6_socket = sk; + else + err = -EADDRINUSE; + write_unlock_bh(&mrt_lock); + + rtnl_unlock(); + + return err; +} + +int ip6mr_sk_done(struct sock *sk) +{ + int err = 0; + + rtnl_lock(); + if (sk == mroute6_socket) { + write_lock_bh(&mrt_lock); + mroute6_socket = NULL; + write_unlock_bh(&mrt_lock); + + mroute_clean_tables(sk); + } else + err = -EACCES; + rtnl_unlock(); + + return err; +} + +/* + * Socket options and virtual interface manipulation. The whole + * virtual interface system is a complete heap, but unfortunately + * that's how BSD mrouted happens to think. Maybe one day with a proper + * MOSPF/PIM router set up we can clean this up. + */ + +int ip6_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, int optlen) +{ + int ret; + struct mif6ctl vif; + struct mf6cctl mfc; + mifi_t mifi; + + if (optname != MRT6_INIT) { + if (sk != mroute6_socket && !capable(CAP_NET_ADMIN)) + return -EACCES; + } + + switch (optname) { + case MRT6_INIT: + if (sk->sk_type != SOCK_RAW || + inet_sk(sk)->num != IPPROTO_ICMPV6) + return -EOPNOTSUPP; + if (optlen < sizeof(int)) + return -EINVAL; + + return ip6mr_sk_init(sk); + + case MRT6_DONE: + return ip6mr_sk_done(sk); + + case MRT6_ADD_MIF: + if (optlen < sizeof(vif)) + return -EINVAL; + if (copy_from_user(&vif, optval, sizeof(vif))) + return -EFAULT; + if (vif.mif6c_mifi >= MAXVIFS) + return -ENFILE; + rtnl_lock(); + ret = mif6_add(&vif, sk == mroute6_socket); + rtnl_unlock(); + return ret; + + case MRT6_DEL_MIF: + if (optlen < sizeof(mifi_t)) + return -EINVAL; + if (copy_from_user(&mifi, optval, sizeof(mifi_t))) + return -EFAULT; + rtnl_lock(); + ret = mif6_delete(mifi); + rtnl_unlock(); + return ret; + + /* + * Manipulate the forwarding caches. These live + * in a sort of kernel/user symbiosis. + */ + case MRT6_ADD_MFC: + case MRT6_DEL_MFC: + if (optlen < sizeof(mfc)) + return -EINVAL; + if (copy_from_user(&mfc, optval, sizeof(mfc))) + return -EFAULT; + rtnl_lock(); + if (optname == MRT6_DEL_MFC) + ret = ip6mr_mfc_delete(&mfc); + else + ret = ip6mr_mfc_add(&mfc, sk == mroute6_socket); + rtnl_unlock(); + return ret; + + /* + * Spurious command, or MRT_VERSION which you cannot + * set. + */ + default: + return -ENOPROTOOPT; + } +} + +/* + * Getsock opt support for the multicast routing system. + */ + +int ip6_mroute_getsockopt(struct sock *sk, int optname, char __user *optval, + int __user *optlen) +{ + int olr; + int val; + + switch (optname) { + case MRT6_VERSION: + val = 0x0305; + break; + default: + return -ENOPROTOOPT; + } + + if (get_user(olr, optlen)) + return -EFAULT; + + olr = min_t(int, olr, sizeof(int)); + if (olr < 0) + return -EINVAL; + + if (put_user(olr, optlen)) + return -EFAULT; + if (copy_to_user(optval, &val, olr)) + return -EFAULT; + return 0; +} + +/* + * The IP multicast ioctl support routines. + */ + +int ip6mr_ioctl(struct sock *sk, int cmd, void __user *arg) +{ + struct sioc_sg_req6 sr; + struct sioc_mif_req6 vr; + struct mif_device *vif; + struct mfc6_cache *c; + + switch (cmd) { + case SIOCGETMIFCNT_IN6: + if (copy_from_user(&vr, arg, sizeof(vr))) + return -EFAULT; + if (vr.mifi >= maxvif) + return -EINVAL; + read_lock(&mrt_lock); + vif = &vif6_table[vr.mifi]; + if (MIF_EXISTS(vr.mifi)) { + vr.icount = vif->pkt_in; + vr.ocount = vif->pkt_out; + vr.ibytes = vif->bytes_in; + vr.obytes = vif->bytes_out; + read_unlock(&mrt_lock); + + if (copy_to_user(arg, &vr, sizeof(vr))) + return -EFAULT; + return 0; + } + read_unlock(&mrt_lock); + return -EADDRNOTAVAIL; + case SIOCGETSGCNT_IN6: + if (copy_from_user(&sr, arg, sizeof(sr))) + return -EFAULT; + + read_lock(&mrt_lock); + c = ip6mr_cache_find(&sr.src.sin6_addr, &sr.grp.sin6_addr); + if (c) { + sr.pktcnt = c->mfc_un.res.pkt; + sr.bytecnt = c->mfc_un.res.bytes; + sr.wrong_if = c->mfc_un.res.wrong_if; + read_unlock(&mrt_lock); + + if (copy_to_user(arg, &sr, sizeof(sr))) + return -EFAULT; + return 0; + } + read_unlock(&mrt_lock); + return -EADDRNOTAVAIL; + default: + return -ENOIOCTLCMD; + } +} + + +static inline int ip6mr_forward2_finish(struct sk_buff *skb) +{ + /* XXX stats */ + return dst_output(skb); +} + +/* + * Processing handlers for ip6mr_forward + */ + +static int ip6mr_forward2(struct sk_buff *skb, struct mfc6_cache *c, int vifi) +{ + struct ipv6hdr *ipv6h; + struct mif_device *vif = &vif6_table[vifi]; + struct net_device *dev; + struct dst_entry *dst; + struct flowi fl; + + if (vif->dev == NULL) + goto out_free; + + ipv6h = ipv6_hdr(skb); + + fl = (struct flowi) { + .oif = vif->link, + .nl_u = { .ip6_u = + { .daddr = ipv6h->daddr, } + } + }; + + dst = ip6_route_output(&init_net, NULL, &fl); + if (!dst) + goto out_free; + + dst_release(skb->dst); + skb->dst = dst; + + /* + * RFC1584 teaches, that DVMRP/PIM router must deliver packets locally + * not only before forwarding, but after forwarding on all output + * interfaces. It is clear, if mrouter runs a multicasting + * program, it should receive packets not depending to what interface + * program is joined. + * If we will not make it, the program will have to join on all + * interfaces. On the other hand, multihoming host (or router, but + * not mrouter) cannot join to more than one interface - it will + * result in receiving multiple packets. + */ + dev = vif->dev; + skb->dev = dev; + vif->pkt_out++; + vif->bytes_out += skb->len; + + /* We are about to write */ + /* XXX: extension headers? */ + if (skb_cow(skb, sizeof(*ipv6h) + LL_RESERVED_SPACE(dev))) + goto out_free; + + ipv6h = ipv6_hdr(skb); + ipv6h->hop_limit--; + + IP6CB(skb)->flags |= IP6SKB_FORWARDED; + + return NF_HOOK(PF_INET6, NF_INET_FORWARD, skb, skb->dev, dev, + ip6mr_forward2_finish); + +out_free: + kfree_skb(skb); + return 0; +} + +static int ip6mr_find_vif(struct net_device *dev) +{ + int ct; + for (ct = maxvif - 1; ct >= 0; ct--) { + if (vif6_table[ct].dev == dev) + break; + } + return ct; +} + +static int ip6_mr_forward(struct sk_buff *skb, struct mfc6_cache *cache) +{ + int psend = -1; + int vif, ct; + + vif = cache->mf6c_parent; + cache->mfc_un.res.pkt++; + cache->mfc_un.res.bytes += skb->len; + + vif6_table[vif].pkt_in++; + vif6_table[vif].bytes_in += skb->len; + + /* + * Forward the frame + */ + for (ct = cache->mfc_un.res.maxvif - 1; ct >= cache->mfc_un.res.minvif; ct--) { + if (ipv6_hdr(skb)->hop_limit > cache->mfc_un.res.ttls[ct]) { + if (psend != -1) { + struct sk_buff *skb2 = skb_clone(skb, GFP_ATOMIC); + if (skb2) + ip6mr_forward2(skb2, cache, psend); + } + psend = ct; + } + } + if (psend != -1) { + ip6mr_forward2(skb, cache, psend); + return 0; + } + + kfree_skb(skb); + return 0; +} + + +/* + * Multicast packets for forwarding arrive here + */ + +int ip6_mr_input(struct sk_buff *skb) +{ + struct mfc6_cache *cache; + + read_lock(&mrt_lock); + cache = ip6mr_cache_find(&ipv6_hdr(skb)->saddr, &ipv6_hdr(skb)->daddr); + + /* + * No usable cache entry + */ + if (cache == NULL) { + int vif; + + vif = ip6mr_find_vif(skb->dev); + if (vif >= 0) { + int err = ip6mr_cache_unresolved(vif, skb); + read_unlock(&mrt_lock); + + return err; + } + read_unlock(&mrt_lock); + kfree_skb(skb); + return -ENODEV; + } + + ip6_mr_forward(skb, cache); + + read_unlock(&mrt_lock); + + return 0; +} + + +static int +ip6mr_fill_mroute(struct sk_buff *skb, struct mfc6_cache *c, struct rtmsg *rtm) +{ + int ct; + struct rtnexthop *nhp; + struct net_device *dev = vif6_table[c->mf6c_parent].dev; + u8 *b = skb->tail; + struct rtattr *mp_head; + + if (dev) + RTA_PUT(skb, RTA_IIF, 4, &dev->ifindex); + + mp_head = (struct rtattr *)skb_put(skb, RTA_LENGTH(0)); + + for (ct = c->mfc_un.res.minvif; ct < c->mfc_un.res.maxvif; ct++) { + if (c->mfc_un.res.ttls[ct] < 255) { + if (skb_tailroom(skb) < RTA_ALIGN(RTA_ALIGN(sizeof(*nhp)) + 4)) + goto rtattr_failure; + nhp = (struct rtnexthop *)skb_put(skb, RTA_ALIGN(sizeof(*nhp))); + nhp->rtnh_flags = 0; + nhp->rtnh_hops = c->mfc_un.res.ttls[ct]; + nhp->rtnh_ifindex = vif6_table[ct].dev->ifindex; + nhp->rtnh_len = sizeof(*nhp); + } + } + mp_head->rta_type = RTA_MULTIPATH; + mp_head->rta_len = skb->tail - (u8 *)mp_head; + rtm->rtm_type = RTN_MULTICAST; + return 1; + +rtattr_failure: + nlmsg_trim(skb, b); + return -EMSGSIZE; +} + +int ip6mr_get_route(struct sk_buff *skb, struct rtmsg *rtm, int nowait) +{ + int err; + struct mfc6_cache *cache; + struct rt6_info *rt = (struct rt6_info *)skb->dst; + + read_lock(&mrt_lock); + cache = ip6mr_cache_find(&rt->rt6i_src.addr, &rt->rt6i_dst.addr); + + if (!cache) { + struct sk_buff *skb2; + struct ipv6hdr *iph; + struct net_device *dev; + int vif; + + if (nowait) { + read_unlock(&mrt_lock); + return -EAGAIN; + } + + dev = skb->dev; + if (dev == NULL || (vif = ip6mr_find_vif(dev)) < 0) { + read_unlock(&mrt_lock); + return -ENODEV; + } + + /* really correct? */ + skb2 = alloc_skb(sizeof(struct ipv6hdr), GFP_ATOMIC); + if (!skb2) { + read_unlock(&mrt_lock); + return -ENOMEM; + } + + skb_reset_transport_header(skb2); + + skb_put(skb2, sizeof(struct ipv6hdr)); + skb_reset_network_header(skb2); + + iph = ipv6_hdr(skb2); + iph->version = 0; + iph->priority = 0; + iph->flow_lbl[0] = 0; + iph->flow_lbl[1] = 0; + iph->flow_lbl[2] = 0; + iph->payload_len = 0; + iph->nexthdr = IPPROTO_NONE; + iph->hop_limit = 0; + ipv6_addr_copy(&iph->saddr, &rt->rt6i_src.addr); + ipv6_addr_copy(&iph->daddr, &rt->rt6i_dst.addr); + + err = ip6mr_cache_unresolved(vif, skb2); + read_unlock(&mrt_lock); + + return err; + } + + if (!nowait && (rtm->rtm_flags&RTM_F_NOTIFY)) + cache->mfc_flags |= MFC_NOTIFY; + + err = ip6mr_fill_mroute(skb, cache, rtm); + read_unlock(&mrt_lock); + return err; +} + diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c index 4195ac92345e..99624109c010 100644 --- a/net/ipv6/ipv6_sockglue.c +++ b/net/ipv6/ipv6_sockglue.c @@ -33,6 +33,7 @@ #include #include #include +#include #include #include #include @@ -118,6 +119,9 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname, valbool = (val!=0); + if (ip6_mroute_opt(optname)) + return ip6_mroute_setsockopt(sk, optname, optval, optlen); + lock_sock(sk); switch (optname) { @@ -790,6 +794,9 @@ static int do_ipv6_getsockopt(struct sock *sk, int level, int optname, int len; int val; + if (ip6_mroute_opt(optname)) + return ip6_mroute_getsockopt(sk, optname, optval, optlen); + if (get_user(len, optlen)) return -EFAULT; switch (optname) { diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index aae6cedf1709..088b80b4ce74 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -53,6 +53,7 @@ #if defined(CONFIG_IPV6_MIP6) || defined(CONFIG_IPV6_MIP6_MODULE) #include #endif +#include #include #include @@ -1135,7 +1136,11 @@ static int rawv6_ioctl(struct sock *sk, int cmd, unsigned long arg) } default: +#ifdef CONFIG_IPV6_MROUTE + return ip6mr_ioctl(sk, cmd, (void __user *)arg); +#else return -ENOIOCTLCMD; +#endif } } @@ -1143,7 +1148,7 @@ static void rawv6_close(struct sock *sk, long timeout) { if (inet_sk(sk)->num == IPPROTO_RAW) ip6_ra_control(sk, -1, NULL); - + ip6mr_sk_done(sk); sk_common_release(sk); } diff --git a/net/ipv6/route.c b/net/ipv6/route.c index cd82b6db35ff..3c314d5f46c6 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -36,6 +36,7 @@ #include #include #include +#include #include #include #include @@ -2106,7 +2107,7 @@ static inline size_t rt6_nlmsg_size(void) static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt, struct in6_addr *dst, struct in6_addr *src, int iif, int type, u32 pid, u32 seq, - int prefix, unsigned int flags) + int prefix, int nowait, unsigned int flags) { struct rtmsg *rtm; struct nlmsghdr *nlh; @@ -2166,9 +2167,24 @@ static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt, } else if (rtm->rtm_src_len) NLA_PUT(skb, RTA_SRC, 16, &rt->rt6i_src.addr); #endif - if (iif) - NLA_PUT_U32(skb, RTA_IIF, iif); - else if (dst) { + if (iif) { +#ifdef CONFIG_IPV6_MROUTE + if (ipv6_addr_is_multicast(&rt->rt6i_dst.addr)) { + int err = ip6mr_get_route(skb, rtm, nowait); + if (err <= 0) { + if (!nowait) { + if (err == 0) + return 0; + goto nla_put_failure; + } else { + if (err == -EMSGSIZE) + goto nla_put_failure; + } + } + } else +#endif + NLA_PUT_U32(skb, RTA_IIF, iif); + } else if (dst) { struct in6_addr saddr_buf; if (ipv6_dev_get_saddr(ip6_dst_idev(&rt->u.dst)->dev, dst, 0, &saddr_buf) == 0) @@ -2211,7 +2227,7 @@ int rt6_dump_route(struct rt6_info *rt, void *p_arg) return rt6_fill_node(arg->skb, rt, NULL, NULL, 0, RTM_NEWROUTE, NETLINK_CB(arg->cb->skb).pid, arg->cb->nlh->nlmsg_seq, - prefix, NLM_F_MULTI); + prefix, 0, NLM_F_MULTI); } static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void *arg) @@ -2277,7 +2293,7 @@ static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void err = rt6_fill_node(skb, rt, &fl.fl6_dst, &fl.fl6_src, iif, RTM_NEWROUTE, NETLINK_CB(in_skb).pid, - nlh->nlmsg_seq, 0, 0); + nlh->nlmsg_seq, 0, 0, 0); if (err < 0) { kfree_skb(skb); goto errout; @@ -2303,7 +2319,7 @@ void inet6_rt_notify(int event, struct rt6_info *rt, struct nl_info *info) goto errout; err = rt6_fill_node(skb, rt, NULL, NULL, 0, - event, info->pid, seq, 0, 0); + event, info->pid, seq, 0, 0, 0); if (err < 0) { /* -EMSGSIZE implies BUG in rt6_nlmsg_size() */ WARN_ON(err == -EMSGSIZE);