X-Git-Url: http://ftp.safe.ca/?a=blobdiff_plain;f=net%2Fipv6%2Fip6_tunnel.c;h=138980eec214df5aca51a2933a9b901f6606e231;hb=e213e9cf707c51808e372dabd1070a61af17e77b;hp=92ead3cf956b1d27817f98a084cdd43d3cc7fb20;hpb=4fc268d24ceb9f4150777c1b5b2b8e6214e56b2b;p=safe%2Fjmp%2Flinux-2.6 diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index 92ead3c..138980e 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -1,14 +1,13 @@ /* - * IPv6 over IPv6 tunnel device + * IPv6 tunneling device * Linux INET6 implementation * * Authors: - * Ville Nuorvala - * - * $Id$ + * Ville Nuorvala + * Yasuyuki Kozakai * * Based on: - * linux/net/ipv6/sit.c + * linux/net/ipv6/sit.c and linux/net/ipv4/ipip.c * * RFC 2473 * @@ -19,12 +18,12 @@ * */ -#include #include #include #include #include #include +#include #include #include #include @@ -42,57 +41,62 @@ #include #include +#include #include #include -#include #include #include #include #include #include #include +#include +#include MODULE_AUTHOR("Ville Nuorvala"); -MODULE_DESCRIPTION("IPv6-in-IPv6 tunnel"); +MODULE_DESCRIPTION("IPv6 tunneling device"); MODULE_LICENSE("GPL"); #define IPV6_TLV_TEL_DST_SIZE 8 #ifdef IP6_TNL_DEBUG -#define IP6_TNL_TRACE(x...) printk(KERN_DEBUG "%s:" x "\n", __FUNCTION__) +#define IP6_TNL_TRACE(x...) printk(KERN_DEBUG "%s:" x "\n", __func__) #else #define IP6_TNL_TRACE(x...) do {;} while(0) #endif #define IPV6_TCLASS_MASK (IPV6_FLOWINFO_MASK & ~IPV6_FLOWLABEL_MASK) +#define IPV6_TCLASS_SHIFT 20 #define HASH_SIZE 32 -#define HASH(addr) (((addr)->s6_addr32[0] ^ (addr)->s6_addr32[1] ^ \ - (addr)->s6_addr32[2] ^ (addr)->s6_addr32[3]) & \ - (HASH_SIZE - 1)) - -static int ip6ip6_fb_tnl_dev_init(struct net_device *dev); -static int ip6ip6_tnl_dev_init(struct net_device *dev); -static void ip6ip6_tnl_dev_setup(struct net_device *dev); - -/* the IPv6 tunnel fallback device */ -static struct net_device *ip6ip6_fb_tnl_dev; - - -/* lists for storing tunnels in use */ -static struct ip6_tnl *tnls_r_l[HASH_SIZE]; -static struct ip6_tnl *tnls_wc[1]; -static struct ip6_tnl **tnls[2] = { tnls_wc, tnls_r_l }; +#define HASH(addr) ((__force u32)((addr)->s6_addr32[0] ^ (addr)->s6_addr32[1] ^ \ + (addr)->s6_addr32[2] ^ (addr)->s6_addr32[3]) & \ + (HASH_SIZE - 1)) + +static void ip6_tnl_dev_init(struct net_device *dev); +static void ip6_tnl_dev_setup(struct net_device *dev); + +static int ip6_tnl_net_id __read_mostly; +struct ip6_tnl_net { + /* the IPv6 tunnel fallback device */ + struct net_device *fb_tnl_dev; + /* lists for storing tunnels in use */ + struct ip6_tnl *tnls_r_l[HASH_SIZE]; + struct ip6_tnl *tnls_wc[1]; + struct ip6_tnl **tnls[2]; +}; -/* lock for the tunnel lists */ -static DEFINE_RWLOCK(ip6ip6_lock); +/* + * Locking : hash tables are protected by RCU and a spinlock + */ +static DEFINE_SPINLOCK(ip6_tnl_lock); static inline struct dst_entry *ip6_tnl_dst_check(struct ip6_tnl *t) { struct dst_entry *dst = t->dst_cache; - if (dst && dst->obsolete && + if (dst && dst->obsolete && dst->ops->check(dst, t->dst_cookie) == NULL) { t->dst_cache = NULL; dst_release(dst); @@ -117,48 +121,53 @@ static inline void ip6_tnl_dst_store(struct ip6_tnl *t, struct dst_entry *dst) } /** - * ip6ip6_tnl_lookup - fetch tunnel matching the end-point addresses - * @remote: the address of the tunnel exit-point - * @local: the address of the tunnel entry-point + * ip6_tnl_lookup - fetch tunnel matching the end-point addresses + * @remote: the address of the tunnel exit-point + * @local: the address of the tunnel entry-point * - * Return: + * Return: * tunnel matching given end-points if found, - * else fallback tunnel if its device is up, + * else fallback tunnel if its device is up, * else %NULL **/ +#define for_each_ip6_tunnel_rcu(start) \ + for (t = rcu_dereference(start); t; t = rcu_dereference(t->next)) + static struct ip6_tnl * -ip6ip6_tnl_lookup(struct in6_addr *remote, struct in6_addr *local) +ip6_tnl_lookup(struct net *net, struct in6_addr *remote, struct in6_addr *local) { unsigned h0 = HASH(remote); unsigned h1 = HASH(local); struct ip6_tnl *t; + struct ip6_tnl_net *ip6n = net_generic(net, ip6_tnl_net_id); - for (t = tnls_r_l[h0 ^ h1]; t; t = t->next) { + for_each_ip6_tunnel_rcu(ip6n->tnls_r_l[h0 ^ h1]) { if (ipv6_addr_equal(local, &t->parms.laddr) && ipv6_addr_equal(remote, &t->parms.raddr) && (t->dev->flags & IFF_UP)) return t; } - if ((t = tnls_wc[0]) != NULL && (t->dev->flags & IFF_UP)) + t = rcu_dereference(ip6n->tnls_wc[0]); + if (t && (t->dev->flags & IFF_UP)) return t; return NULL; } /** - * ip6ip6_bucket - get head of list matching given tunnel parameters - * @p: parameters containing tunnel end-points + * ip6_tnl_bucket - get head of list matching given tunnel parameters + * @p: parameters containing tunnel end-points * * Description: - * ip6ip6_bucket() returns the head of the list matching the + * ip6_tnl_bucket() returns the head of the list matching the * &struct in6_addr entries laddr and raddr in @p. * - * Return: head of IPv6 tunnel list + * Return: head of IPv6 tunnel list **/ static struct ip6_tnl ** -ip6ip6_bucket(struct ip6_tnl_parm *p) +ip6_tnl_bucket(struct ip6_tnl_net *ip6n, struct ip6_tnl_parm *p) { struct in6_addr *remote = &p->raddr; struct in6_addr *local = &p->laddr; @@ -169,40 +178,40 @@ ip6ip6_bucket(struct ip6_tnl_parm *p) prio = 1; h = HASH(remote) ^ HASH(local); } - return &tnls[prio][h]; + return &ip6n->tnls[prio][h]; } /** - * ip6ip6_tnl_link - add tunnel to hash table + * ip6_tnl_link - add tunnel to hash table * @t: tunnel to be added **/ static void -ip6ip6_tnl_link(struct ip6_tnl *t) +ip6_tnl_link(struct ip6_tnl_net *ip6n, struct ip6_tnl *t) { - struct ip6_tnl **tp = ip6ip6_bucket(&t->parms); + struct ip6_tnl **tp = ip6_tnl_bucket(ip6n, &t->parms); + spin_lock_bh(&ip6_tnl_lock); t->next = *tp; - write_lock_bh(&ip6ip6_lock); - *tp = t; - write_unlock_bh(&ip6ip6_lock); + rcu_assign_pointer(*tp, t); + spin_unlock_bh(&ip6_tnl_lock); } /** - * ip6ip6_tnl_unlink - remove tunnel from hash table + * ip6_tnl_unlink - remove tunnel from hash table * @t: tunnel to be removed **/ static void -ip6ip6_tnl_unlink(struct ip6_tnl *t) +ip6_tnl_unlink(struct ip6_tnl_net *ip6n, struct ip6_tnl *t) { struct ip6_tnl **tp; - for (tp = ip6ip6_bucket(&t->parms); *tp; tp = &(*tp)->next) { + for (tp = ip6_tnl_bucket(ip6n, &t->parms); *tp; tp = &(*tp)->next) { if (t == *tp) { - write_lock_bh(&ip6ip6_lock); + spin_lock_bh(&ip6_tnl_lock); *tp = t->next; - write_unlock_bh(&ip6ip6_lock); + spin_unlock_bh(&ip6_tnl_lock); break; } } @@ -215,108 +224,105 @@ ip6ip6_tnl_unlink(struct ip6_tnl *t) * * Description: * Create tunnel matching given parameters. - * - * Return: - * 0 on success + * + * Return: + * created tunnel or NULL **/ -static int -ip6_tnl_create(struct ip6_tnl_parm *p, struct ip6_tnl **pt) +static struct ip6_tnl *ip6_tnl_create(struct net *net, struct ip6_tnl_parm *p) { struct net_device *dev; struct ip6_tnl *t; char name[IFNAMSIZ]; int err; + struct ip6_tnl_net *ip6n = net_generic(net, ip6_tnl_net_id); - if (p->name[0]) { + if (p->name[0]) strlcpy(name, p->name, IFNAMSIZ); - } else { - int i; - for (i = 1; i < IP6_TNL_MAX; i++) { - sprintf(name, "ip6tnl%d", i); - if (__dev_get_by_name(name) == NULL) - break; - } - if (i == IP6_TNL_MAX) - return -ENOBUFS; - } - dev = alloc_netdev(sizeof (*t), name, ip6ip6_tnl_dev_setup); + else + sprintf(name, "ip6tnl%%d"); + + dev = alloc_netdev(sizeof (*t), name, ip6_tnl_dev_setup); if (dev == NULL) - return -ENOMEM; + goto failed; + + dev_net_set(dev, net); + + if (strchr(name, '%')) { + if (dev_alloc_name(dev, name) < 0) + goto failed_free; + } t = netdev_priv(dev); - dev->init = ip6ip6_tnl_dev_init; t->parms = *p; + ip6_tnl_dev_init(dev); + + if ((err = register_netdevice(dev)) < 0) + goto failed_free; - if ((err = register_netdevice(dev)) < 0) { - free_netdev(dev); - return err; - } dev_hold(dev); + ip6_tnl_link(ip6n, t); + return t; - ip6ip6_tnl_link(t); - *pt = t; - return 0; +failed_free: + free_netdev(dev); +failed: + return NULL; } /** - * ip6ip6_tnl_locate - find or create tunnel matching given parameters - * @p: tunnel parameters + * ip6_tnl_locate - find or create tunnel matching given parameters + * @p: tunnel parameters * @create: != 0 if allowed to create new tunnel if no match found * * Description: - * ip6ip6_tnl_locate() first tries to locate an existing tunnel + * ip6_tnl_locate() first tries to locate an existing tunnel * based on @parms. If this is unsuccessful, but @create is set a new * tunnel device is created and registered for use. * * Return: - * 0 if tunnel located or created, - * -EINVAL if parameters incorrect, - * -ENODEV if no matching tunnel available + * matching tunnel or NULL **/ -static int -ip6ip6_tnl_locate(struct ip6_tnl_parm *p, struct ip6_tnl **pt, int create) +static struct ip6_tnl *ip6_tnl_locate(struct net *net, + struct ip6_tnl_parm *p, int create) { struct in6_addr *remote = &p->raddr; struct in6_addr *local = &p->laddr; struct ip6_tnl *t; + struct ip6_tnl_net *ip6n = net_generic(net, ip6_tnl_net_id); - if (p->proto != IPPROTO_IPV6) - return -EINVAL; - - for (t = *ip6ip6_bucket(p); t; t = t->next) { + for (t = *ip6_tnl_bucket(ip6n, p); t; t = t->next) { if (ipv6_addr_equal(local, &t->parms.laddr) && - ipv6_addr_equal(remote, &t->parms.raddr)) { - *pt = t; - return (create ? -EEXIST : 0); - } + ipv6_addr_equal(remote, &t->parms.raddr)) + return t; } if (!create) - return -ENODEV; - - return ip6_tnl_create(p, pt); + return NULL; + return ip6_tnl_create(net, p); } /** - * ip6ip6_tnl_dev_uninit - tunnel device uninitializer + * ip6_tnl_dev_uninit - tunnel device uninitializer * @dev: the device to be destroyed - * + * * Description: - * ip6ip6_tnl_dev_uninit() removes tunnel from its list + * ip6_tnl_dev_uninit() removes tunnel from its list **/ static void -ip6ip6_tnl_dev_uninit(struct net_device *dev) +ip6_tnl_dev_uninit(struct net_device *dev) { struct ip6_tnl *t = netdev_priv(dev); + struct net *net = dev_net(dev); + struct ip6_tnl_net *ip6n = net_generic(net, ip6_tnl_net_id); - if (dev == ip6ip6_fb_tnl_dev) { - write_lock_bh(&ip6ip6_lock); - tnls_wc[0] = NULL; - write_unlock_bh(&ip6ip6_lock); + if (dev == ip6n->fb_tnl_dev) { + spin_lock_bh(&ip6_tnl_lock); + ip6n->tnls_wc[0] = NULL; + spin_unlock_bh(&ip6_tnl_lock); } else { - ip6ip6_tnl_unlink(t); + ip6_tnl_unlink(ip6n, t); } ip6_tnl_dst_reset(t); dev_put(dev); @@ -326,8 +332,8 @@ ip6ip6_tnl_dev_uninit(struct net_device *dev) * parse_tvl_tnl_enc_lim - handle encapsulation limit option * @skb: received socket buffer * - * Return: - * 0 if none was found, + * Return: + * 0 if none was found, * else index to encapsulation limit **/ @@ -384,34 +390,41 @@ parse_tlv_tnl_enc_lim(struct sk_buff *skb, __u8 * raw) } /** - * ip6ip6_err - tunnel error handler + * ip6_tnl_err - tunnel error handler * * Description: - * ip6ip6_err() should handle errors in the tunnel according + * ip6_tnl_err() should handle errors in the tunnel according * to the specifications in RFC 2473. **/ -static void -ip6ip6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, - int type, int code, int offset, __u32 info) +static int +ip6_tnl_err(struct sk_buff *skb, __u8 ipproto, struct inet6_skb_parm *opt, + u8 *type, u8 *code, int *msg, __u32 *info, int offset) { struct ipv6hdr *ipv6h = (struct ipv6hdr *) skb->data; struct ip6_tnl *t; int rel_msg = 0; - int rel_type = ICMPV6_DEST_UNREACH; - int rel_code = ICMPV6_ADDR_UNREACH; + u8 rel_type = ICMPV6_DEST_UNREACH; + u8 rel_code = ICMPV6_ADDR_UNREACH; __u32 rel_info = 0; __u16 len; + int err = -ENOENT; - /* If the packet doesn't contain the original IPv6 header we are - in trouble since we might need the source address for further + /* If the packet doesn't contain the original IPv6 header we are + in trouble since we might need the source address for further processing of the error. */ - read_lock(&ip6ip6_lock); - if ((t = ip6ip6_tnl_lookup(&ipv6h->daddr, &ipv6h->saddr)) == NULL) + rcu_read_lock(); + if ((t = ip6_tnl_lookup(dev_net(skb->dev), &ipv6h->daddr, + &ipv6h->saddr)) == NULL) goto out; - switch (type) { + if (t->parms.proto != ipproto && t->parms.proto != 0) + goto out; + + err = 0; + + switch (*type) { __u32 teli; struct ipv6_tlv_tnl_enc_lim *tel; __u32 mtu; @@ -423,24 +436,21 @@ ip6ip6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, rel_msg = 1; break; case ICMPV6_TIME_EXCEED: - if (code == ICMPV6_EXC_HOPLIMIT) { + if ((*code) == ICMPV6_EXC_HOPLIMIT) { if (net_ratelimit()) printk(KERN_WARNING "%s: Too small hop limit or " - "routing loop in tunnel!\n", + "routing loop in tunnel!\n", t->parms.name); rel_msg = 1; } break; case ICMPV6_PARAMPROB: - /* ignore if parameter problem not caused by a tunnel - encapsulation limit sub-option */ - if (code != ICMPV6_HDR_FIELD) { - break; - } - teli = parse_tlv_tnl_enc_lim(skb, skb->data); + teli = 0; + if ((*code) == ICMPV6_HDR_FIELD) + teli = parse_tlv_tnl_enc_lim(skb, skb->data); - if (teli && teli == ntohl(info) - 2) { + if (teli && teli == *info - 2) { tel = (struct ipv6_tlv_tnl_enc_lim *) &skb->data[teli]; if (tel->encap_limit == 0) { if (net_ratelimit()) @@ -450,15 +460,19 @@ ip6ip6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, "tunnel!\n", t->parms.name); rel_msg = 1; } + } else if (net_ratelimit()) { + printk(KERN_WARNING + "%s: Recipient unable to parse tunneled " + "packet!\n ", t->parms.name); } break; case ICMPV6_PKT_TOOBIG: - mtu = ntohl(info) - offset; + mtu = *info - offset; if (mtu < IPV6_MIN_MTU) mtu = IPV6_MIN_MTU; t->dev->mtu = mtu; - if ((len = sizeof (*ipv6h) + ipv6h->payload_len) > mtu) { + if ((len = sizeof (*ipv6h) + ntohs(ipv6h->payload_len)) > mtu) { rel_type = ICMPV6_PKT_TOOBIG; rel_code = 0; rel_info = mtu; @@ -466,219 +480,393 @@ ip6ip6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, } break; } - if (rel_msg && pskb_may_pull(skb, offset + sizeof (*ipv6h))) { + + *type = rel_type; + *code = rel_code; + *info = rel_info; + *msg = rel_msg; + +out: + rcu_read_unlock(); + return err; +} + +static int +ip4ip6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, + u8 type, u8 code, int offset, __be32 info) +{ + int rel_msg = 0; + u8 rel_type = type; + u8 rel_code = code; + __u32 rel_info = ntohl(info); + int err; + struct sk_buff *skb2; + struct iphdr *eiph; + struct flowi fl; + struct rtable *rt; + + err = ip6_tnl_err(skb, IPPROTO_IPIP, opt, &rel_type, &rel_code, + &rel_msg, &rel_info, offset); + if (err < 0) + return err; + + if (rel_msg == 0) + return 0; + + switch (rel_type) { + case ICMPV6_DEST_UNREACH: + if (rel_code != ICMPV6_ADDR_UNREACH) + return 0; + rel_type = ICMP_DEST_UNREACH; + rel_code = ICMP_HOST_UNREACH; + break; + case ICMPV6_PKT_TOOBIG: + if (rel_code != 0) + return 0; + rel_type = ICMP_DEST_UNREACH; + rel_code = ICMP_FRAG_NEEDED; + break; + default: + return 0; + } + + if (!pskb_may_pull(skb, offset + sizeof(struct iphdr))) + return 0; + + skb2 = skb_clone(skb, GFP_ATOMIC); + if (!skb2) + return 0; + + skb_dst_drop(skb2); + + skb_pull(skb2, offset); + skb_reset_network_header(skb2); + eiph = ip_hdr(skb2); + + /* Try to guess incoming interface */ + memset(&fl, 0, sizeof(fl)); + fl.fl4_dst = eiph->saddr; + fl.fl4_tos = RT_TOS(eiph->tos); + fl.proto = IPPROTO_IPIP; + if (ip_route_output_key(dev_net(skb->dev), &rt, &fl)) + goto out; + + skb2->dev = rt->u.dst.dev; + + /* route "incoming" packet */ + if (rt->rt_flags & RTCF_LOCAL) { + ip_rt_put(rt); + rt = NULL; + fl.fl4_dst = eiph->daddr; + fl.fl4_src = eiph->saddr; + fl.fl4_tos = eiph->tos; + if (ip_route_output_key(dev_net(skb->dev), &rt, &fl) || + rt->u.dst.dev->type != ARPHRD_TUNNEL) { + ip_rt_put(rt); + goto out; + } + skb_dst_set(skb2, (struct dst_entry *)rt); + } else { + ip_rt_put(rt); + if (ip_route_input(skb2, eiph->daddr, eiph->saddr, eiph->tos, + skb2->dev) || + skb_dst(skb2)->dev->type != ARPHRD_TUNNEL) + goto out; + } + + /* change mtu on this route */ + if (rel_type == ICMP_DEST_UNREACH && rel_code == ICMP_FRAG_NEEDED) { + if (rel_info > dst_mtu(skb_dst(skb2))) + goto out; + + skb_dst(skb2)->ops->update_pmtu(skb_dst(skb2), rel_info); + } + + icmp_send(skb2, rel_type, rel_code, htonl(rel_info)); + +out: + kfree_skb(skb2); + return 0; +} + +static int +ip6ip6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, + u8 type, u8 code, int offset, __be32 info) +{ + int rel_msg = 0; + u8 rel_type = type; + u8 rel_code = code; + __u32 rel_info = ntohl(info); + int err; + + err = ip6_tnl_err(skb, IPPROTO_IPV6, opt, &rel_type, &rel_code, + &rel_msg, &rel_info, offset); + if (err < 0) + return err; + + if (rel_msg && pskb_may_pull(skb, offset + sizeof(struct ipv6hdr))) { struct rt6_info *rt; struct sk_buff *skb2 = skb_clone(skb, GFP_ATOMIC); + if (!skb2) - goto out; + return 0; - dst_release(skb2->dst); - skb2->dst = NULL; + skb_dst_drop(skb2); skb_pull(skb2, offset); - skb2->nh.raw = skb2->data; + skb_reset_network_header(skb2); /* Try to guess incoming interface */ - rt = rt6_lookup(&skb2->nh.ipv6h->saddr, NULL, 0, 0); + rt = rt6_lookup(dev_net(skb->dev), &ipv6_hdr(skb2)->saddr, + NULL, 0, 0); if (rt && rt->rt6i_dev) skb2->dev = rt->rt6i_dev; - icmpv6_send(skb2, rel_type, rel_code, rel_info, skb2->dev); + icmpv6_send(skb2, rel_type, rel_code, rel_info); if (rt) dst_release(&rt->u.dst); kfree_skb(skb2); } -out: - read_unlock(&ip6ip6_lock); + + return 0; } -static inline void ip6ip6_ecn_decapsulate(struct ipv6hdr *outer_iph, - struct sk_buff *skb) +static void ip4ip6_dscp_ecn_decapsulate(struct ip6_tnl *t, + struct ipv6hdr *ipv6h, + struct sk_buff *skb) { - struct ipv6hdr *inner_iph = skb->nh.ipv6h; + __u8 dsfield = ipv6_get_dsfield(ipv6h) & ~INET_ECN_MASK; + + if (t->parms.flags & IP6_TNL_F_RCV_DSCP_COPY) + ipv4_change_dsfield(ip_hdr(skb), INET_ECN_MASK, dsfield); - if (INET_ECN_is_ce(ipv6_get_dsfield(outer_iph))) - IP6_ECN_set_ce(inner_iph); + if (INET_ECN_is_ce(dsfield)) + IP_ECN_set_ce(ip_hdr(skb)); +} + +static void ip6ip6_dscp_ecn_decapsulate(struct ip6_tnl *t, + struct ipv6hdr *ipv6h, + struct sk_buff *skb) +{ + if (t->parms.flags & IP6_TNL_F_RCV_DSCP_COPY) + ipv6_copy_dscp(ipv6_get_dsfield(ipv6h), ipv6_hdr(skb)); + + if (INET_ECN_is_ce(ipv6_get_dsfield(ipv6h))) + IP6_ECN_set_ce(ipv6_hdr(skb)); +} + +/* called with rcu_read_lock() */ +static inline int ip6_tnl_rcv_ctl(struct ip6_tnl *t) +{ + struct ip6_tnl_parm *p = &t->parms; + int ret = 0; + struct net *net = dev_net(t->dev); + + if (p->flags & IP6_TNL_F_CAP_RCV) { + struct net_device *ldev = NULL; + + if (p->link) + ldev = dev_get_by_index_rcu(net, p->link); + + if ((ipv6_addr_is_multicast(&p->laddr) || + likely(ipv6_chk_addr(net, &p->laddr, ldev, 0))) && + likely(!ipv6_chk_addr(net, &p->raddr, NULL, 0))) + ret = 1; + + } + return ret; } /** - * ip6ip6_rcv - decapsulate IPv6 packet and retransmit it locally + * ip6_tnl_rcv - decapsulate IPv6 packet and retransmit it locally * @skb: received socket buffer + * @protocol: ethernet protocol ID + * @dscp_ecn_decapsulate: the function to decapsulate DSCP code and ECN * * Return: 0 **/ -static int -ip6ip6_rcv(struct sk_buff **pskb) +static int ip6_tnl_rcv(struct sk_buff *skb, __u16 protocol, + __u8 ipproto, + void (*dscp_ecn_decapsulate)(struct ip6_tnl *t, + struct ipv6hdr *ipv6h, + struct sk_buff *skb)) { - struct sk_buff *skb = *pskb; - struct ipv6hdr *ipv6h; struct ip6_tnl *t; + struct ipv6hdr *ipv6h = ipv6_hdr(skb); - if (!pskb_may_pull(skb, sizeof (*ipv6h))) - goto discard; - - ipv6h = skb->nh.ipv6h; + rcu_read_lock(); - read_lock(&ip6ip6_lock); + if ((t = ip6_tnl_lookup(dev_net(skb->dev), &ipv6h->saddr, + &ipv6h->daddr)) != NULL) { + if (t->parms.proto != ipproto && t->parms.proto != 0) { + rcu_read_unlock(); + goto discard; + } - if ((t = ip6ip6_tnl_lookup(&ipv6h->saddr, &ipv6h->daddr)) != NULL) { if (!xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb)) { - read_unlock(&ip6ip6_lock); - kfree_skb(skb); - return 0; + rcu_read_unlock(); + goto discard; } - if (!(t->parms.flags & IP6_TNL_F_CAP_RCV)) { - t->stat.rx_dropped++; - read_unlock(&ip6ip6_lock); + if (!ip6_tnl_rcv_ctl(t)) { + t->dev->stats.rx_dropped++; + rcu_read_unlock(); goto discard; } secpath_reset(skb); - skb->mac.raw = skb->nh.raw; - skb->nh.raw = skb->data; - skb->protocol = htons(ETH_P_IPV6); + skb->mac_header = skb->network_header; + skb_reset_network_header(skb); + skb->protocol = htons(protocol); skb->pkt_type = PACKET_HOST; memset(skb->cb, 0, sizeof(struct inet6_skb_parm)); skb->dev = t->dev; - dst_release(skb->dst); - skb->dst = NULL; - if (t->parms.flags & IP6_TNL_F_RCV_DSCP_COPY) - ipv6_copy_dscp(ipv6h, skb->nh.ipv6h); - ip6ip6_ecn_decapsulate(ipv6h, skb); - t->stat.rx_packets++; - t->stat.rx_bytes += skb->len; + skb_dst_drop(skb); + nf_reset(skb); + + dscp_ecn_decapsulate(t, ipv6h, skb); + + t->dev->stats.rx_packets++; + t->dev->stats.rx_bytes += skb->len; netif_rx(skb); - read_unlock(&ip6ip6_lock); + rcu_read_unlock(); return 0; } - read_unlock(&ip6ip6_lock); - icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_ADDR_UNREACH, 0, skb->dev); -discard: + rcu_read_unlock(); return 1; + +discard: + kfree_skb(skb); + return 0; } -static inline struct ipv6_txoptions *create_tel(__u8 encap_limit) +static int ip4ip6_rcv(struct sk_buff *skb) { - struct ipv6_tlv_tnl_enc_lim *tel; - struct ipv6_txoptions *opt; - __u8 *raw; + return ip6_tnl_rcv(skb, ETH_P_IP, IPPROTO_IPIP, + ip4ip6_dscp_ecn_decapsulate); +} - int opt_len = sizeof(*opt) + 8; +static int ip6ip6_rcv(struct sk_buff *skb) +{ + return ip6_tnl_rcv(skb, ETH_P_IPV6, IPPROTO_IPV6, + ip6ip6_dscp_ecn_decapsulate); +} - if (!(opt = kmalloc(opt_len, GFP_ATOMIC))) { - return NULL; - } - memset(opt, 0, opt_len); - opt->tot_len = opt_len; - opt->dst0opt = (struct ipv6_opt_hdr *) (opt + 1); - opt->opt_nflen = 8; +struct ipv6_tel_txoption { + struct ipv6_txoptions ops; + __u8 dst_opt[8]; +}; - tel = (struct ipv6_tlv_tnl_enc_lim *) (opt->dst0opt + 1); - tel->type = IPV6_TLV_TNL_ENCAP_LIMIT; - tel->length = 1; - tel->encap_limit = encap_limit; +static void init_tel_txopt(struct ipv6_tel_txoption *opt, __u8 encap_limit) +{ + memset(opt, 0, sizeof(struct ipv6_tel_txoption)); - raw = (__u8 *) opt->dst0opt; - raw[5] = IPV6_TLV_PADN; - raw[6] = 1; + opt->dst_opt[2] = IPV6_TLV_TNL_ENCAP_LIMIT; + opt->dst_opt[3] = 1; + opt->dst_opt[4] = encap_limit; + opt->dst_opt[5] = IPV6_TLV_PADN; + opt->dst_opt[6] = 1; - return opt; + opt->ops.dst0opt = (struct ipv6_opt_hdr *) opt->dst_opt; + opt->ops.opt_nflen = 8; } /** - * ip6ip6_tnl_addr_conflict - compare packet addresses to tunnel's own + * ip6_tnl_addr_conflict - compare packet addresses to tunnel's own * @t: the outgoing tunnel device - * @hdr: IPv6 header from the incoming packet + * @hdr: IPv6 header from the incoming packet * * Description: - * Avoid trivial tunneling loop by checking that tunnel exit-point + * Avoid trivial tunneling loop by checking that tunnel exit-point * doesn't match source of incoming packet. * - * Return: + * Return: * 1 if conflict, * 0 else **/ static inline int -ip6ip6_tnl_addr_conflict(struct ip6_tnl *t, struct ipv6hdr *hdr) +ip6_tnl_addr_conflict(struct ip6_tnl *t, struct ipv6hdr *hdr) { return ipv6_addr_equal(&t->parms.raddr, &hdr->saddr); } +static inline int ip6_tnl_xmit_ctl(struct ip6_tnl *t) +{ + struct ip6_tnl_parm *p = &t->parms; + int ret = 0; + struct net *net = dev_net(t->dev); + + if (p->flags & IP6_TNL_F_CAP_XMIT) { + struct net_device *ldev = NULL; + + rcu_read_lock(); + if (p->link) + ldev = dev_get_by_index_rcu(net, p->link); + + if (unlikely(!ipv6_chk_addr(net, &p->laddr, ldev, 0))) + printk(KERN_WARNING + "%s xmit: Local address not yet configured!\n", + p->name); + else if (!ipv6_addr_is_multicast(&p->raddr) && + unlikely(ipv6_chk_addr(net, &p->raddr, NULL, 0))) + printk(KERN_WARNING + "%s xmit: Routing loop! " + "Remote address found on this node!\n", + p->name); + else + ret = 1; + rcu_read_unlock(); + } + return ret; +} /** - * ip6ip6_tnl_xmit - encapsulate packet and send + * ip6_tnl_xmit2 - encapsulate packet and send * @skb: the outgoing socket buffer - * @dev: the outgoing tunnel device + * @dev: the outgoing tunnel device + * @dsfield: dscp code for outer header + * @fl: flow of tunneled packet + * @encap_limit: encapsulation limit + * @pmtu: Path MTU is stored if packet is too big * * Description: * Build new header and do some sanity checks on the packet before sending * it. * - * Return: - * 0 + * Return: + * 0 on success + * -1 fail + * %-EMSGSIZE message too big. return mtu in this case. **/ -static int -ip6ip6_tnl_xmit(struct sk_buff *skb, struct net_device *dev) +static int ip6_tnl_xmit2(struct sk_buff *skb, + struct net_device *dev, + __u8 dsfield, + struct flowi *fl, + int encap_limit, + __u32 *pmtu) { + struct net *net = dev_net(dev); struct ip6_tnl *t = netdev_priv(dev); - struct net_device_stats *stats = &t->stat; - struct ipv6hdr *ipv6h = skb->nh.ipv6h; - struct ipv6_txoptions *opt = NULL; - int encap_limit = -1; - __u16 offset; - struct flowi fl; + struct net_device_stats *stats = &t->dev->stats; + struct ipv6hdr *ipv6h = ipv6_hdr(skb); + struct ipv6_tel_txoption opt; struct dst_entry *dst; struct net_device *tdev; int mtu; - int max_headroom = sizeof(struct ipv6hdr); + unsigned int max_headroom = sizeof(struct ipv6hdr); u8 proto; - int err; + int err = -1; int pkt_len; - int dsfield; - - if (t->recursion++) { - stats->collisions++; - goto tx_err; - } - if (skb->protocol != htons(ETH_P_IPV6) || - !(t->parms.flags & IP6_TNL_F_CAP_XMIT) || - ip6ip6_tnl_addr_conflict(t, ipv6h)) { - goto tx_err; - } - if ((offset = parse_tlv_tnl_enc_lim(skb, skb->nh.raw)) > 0) { - struct ipv6_tlv_tnl_enc_lim *tel; - tel = (struct ipv6_tlv_tnl_enc_lim *) &skb->nh.raw[offset]; - if (tel->encap_limit == 0) { - icmpv6_send(skb, ICMPV6_PARAMPROB, - ICMPV6_HDR_FIELD, offset + 2, skb->dev); - goto tx_err; - } - encap_limit = tel->encap_limit - 1; - } else if (!(t->parms.flags & IP6_TNL_F_IGN_ENCAP_LIMIT)) { - encap_limit = t->parms.encap_limit; - } - memcpy(&fl, &t->fl, sizeof (fl)); - proto = fl.proto; - - dsfield = ipv6_get_dsfield(ipv6h); - if ((t->parms.flags & IP6_TNL_F_USE_ORIG_TCLASS)) - fl.fl6_flowlabel |= (*(__u32 *) ipv6h & IPV6_TCLASS_MASK); - if ((t->parms.flags & IP6_TNL_F_USE_ORIG_FLOWLABEL)) - fl.fl6_flowlabel |= (*(__u32 *) ipv6h & IPV6_FLOWLABEL_MASK); - - if (encap_limit >= 0 && (opt = create_tel(encap_limit)) == NULL) - goto tx_err; if ((dst = ip6_tnl_dst_check(t)) != NULL) dst_hold(dst); else { - dst = ip6_route_output(NULL, &fl); + dst = ip6_route_output(net, NULL, fl); - if (dst->error || xfrm_lookup(&dst, &fl, NULL, 0) < 0) + if (dst->error || xfrm_lookup(net, &dst, fl, NULL, 0) < 0) goto tx_err_link_failure; } @@ -687,25 +875,23 @@ ip6ip6_tnl_xmit(struct sk_buff *skb, struct net_device *dev) if (tdev == dev) { stats->collisions++; if (net_ratelimit()) - printk(KERN_WARNING + printk(KERN_WARNING "%s: Local routing loop detected!\n", t->parms.name); goto tx_err_dst_release; } mtu = dst_mtu(dst) - sizeof (*ipv6h); - if (opt) { + if (encap_limit >= 0) { max_headroom += 8; mtu -= 8; } if (mtu < IPV6_MIN_MTU) mtu = IPV6_MIN_MTU; - if (skb->dst && mtu < dst_mtu(skb->dst)) { - struct rt6_info *rt = (struct rt6_info *) skb->dst; - rt->rt6i_flags |= RTF_MODIFIED; - rt->u.dst.metrics[RTAX_MTU-1] = mtu; - } + if (skb_dst(skb)) + skb_dst(skb)->ops->update_pmtu(skb_dst(skb), mtu); if (skb->len > mtu) { - icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, dev); + *pmtu = mtu; + err = -EMSGSIZE; goto tx_err_dst_release; } @@ -713,11 +899,11 @@ ip6ip6_tnl_xmit(struct sk_buff *skb, struct net_device *dev) * Okay, now see if we can stuff it in the buffer as-is. */ max_headroom += LL_RESERVED_SPACE(tdev); - - if (skb_headroom(skb) < max_headroom || - skb_cloned(skb) || skb_shared(skb)) { + + if (skb_headroom(skb) < max_headroom || skb_shared(skb) || + (skb_cloned(skb) && !skb_clone_writable(skb, 0))) { struct sk_buff *new_skb; - + if (!(new_skb = skb_realloc_headroom(skb, max_headroom))) goto tx_err_dst_release; @@ -726,30 +912,31 @@ ip6ip6_tnl_xmit(struct sk_buff *skb, struct net_device *dev) kfree_skb(skb); skb = new_skb; } - dst_release(skb->dst); - skb->dst = dst_clone(dst); + skb_dst_drop(skb); + skb_dst_set(skb, dst_clone(dst)); - skb->h.raw = skb->nh.raw; + skb->transport_header = skb->network_header; - if (opt) - ipv6_push_nfrag_opts(skb, opt, &proto, NULL); - - skb->nh.raw = skb_push(skb, sizeof(struct ipv6hdr)); - ipv6h = skb->nh.ipv6h; - *(u32*)ipv6h = fl.fl6_flowlabel | htonl(0x60000000); + proto = fl->proto; + if (encap_limit >= 0) { + init_tel_txopt(&opt, encap_limit); + ipv6_push_nfrag_opts(skb, &opt.ops, &proto, NULL); + } + skb_push(skb, sizeof(struct ipv6hdr)); + skb_reset_network_header(skb); + ipv6h = ipv6_hdr(skb); + *(__be32*)ipv6h = fl->fl6_flowlabel | htonl(0x60000000); dsfield = INET_ECN_encapsulate(0, dsfield); ipv6_change_dsfield(ipv6h, ~INET_ECN_MASK, dsfield); - ipv6h->payload_len = htons(skb->len - sizeof(struct ipv6hdr)); ipv6h->hop_limit = t->parms.hop_limit; ipv6h->nexthdr = proto; - ipv6_addr_copy(&ipv6h->saddr, &fl.fl6_src); - ipv6_addr_copy(&ipv6h->daddr, &fl.fl6_dst); + ipv6_addr_copy(&ipv6h->saddr, &fl->fl6_src); + ipv6_addr_copy(&ipv6h->daddr, &fl->fl6_dst); nf_reset(skb); pkt_len = skb->len; - err = NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, skb, NULL, - skb->dst->dev, dst_output); + err = ip6_local_out(skb); - if (err == NET_XMIT_SUCCESS || err == NET_XMIT_CN) { + if (net_xmit_eval(err) == 0) { stats->tx_bytes += pkt_len; stats->tx_packets++; } else { @@ -757,72 +944,159 @@ ip6ip6_tnl_xmit(struct sk_buff *skb, struct net_device *dev) stats->tx_aborted_errors++; } ip6_tnl_dst_store(t, dst); - - kfree(opt); - - t->recursion--; return 0; tx_err_link_failure: stats->tx_carrier_errors++; dst_link_failure(skb); tx_err_dst_release: dst_release(dst); - kfree(opt); + return err; +} + +static inline int +ip4ip6_tnl_xmit(struct sk_buff *skb, struct net_device *dev) +{ + struct ip6_tnl *t = netdev_priv(dev); + struct iphdr *iph = ip_hdr(skb); + int encap_limit = -1; + struct flowi fl; + __u8 dsfield; + __u32 mtu; + int err; + + if ((t->parms.proto != IPPROTO_IPIP && t->parms.proto != 0) || + !ip6_tnl_xmit_ctl(t)) + return -1; + + if (!(t->parms.flags & IP6_TNL_F_IGN_ENCAP_LIMIT)) + encap_limit = t->parms.encap_limit; + + memcpy(&fl, &t->fl, sizeof (fl)); + fl.proto = IPPROTO_IPIP; + + dsfield = ipv4_get_dsfield(iph); + + if ((t->parms.flags & IP6_TNL_F_USE_ORIG_TCLASS)) + fl.fl6_flowlabel |= htonl((__u32)iph->tos << IPV6_TCLASS_SHIFT) + & IPV6_TCLASS_MASK; + + err = ip6_tnl_xmit2(skb, dev, dsfield, &fl, encap_limit, &mtu); + if (err != 0) { + /* XXX: send ICMP error even if DF is not set. */ + if (err == -EMSGSIZE) + icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED, + htonl(mtu)); + return -1; + } + + return 0; +} + +static inline int +ip6ip6_tnl_xmit(struct sk_buff *skb, struct net_device *dev) +{ + struct ip6_tnl *t = netdev_priv(dev); + struct ipv6hdr *ipv6h = ipv6_hdr(skb); + int encap_limit = -1; + __u16 offset; + struct flowi fl; + __u8 dsfield; + __u32 mtu; + int err; + + if ((t->parms.proto != IPPROTO_IPV6 && t->parms.proto != 0) || + !ip6_tnl_xmit_ctl(t) || ip6_tnl_addr_conflict(t, ipv6h)) + return -1; + + offset = parse_tlv_tnl_enc_lim(skb, skb_network_header(skb)); + if (offset > 0) { + struct ipv6_tlv_tnl_enc_lim *tel; + tel = (struct ipv6_tlv_tnl_enc_lim *)&skb_network_header(skb)[offset]; + if (tel->encap_limit == 0) { + icmpv6_send(skb, ICMPV6_PARAMPROB, + ICMPV6_HDR_FIELD, offset + 2); + return -1; + } + encap_limit = tel->encap_limit - 1; + } else if (!(t->parms.flags & IP6_TNL_F_IGN_ENCAP_LIMIT)) + encap_limit = t->parms.encap_limit; + + memcpy(&fl, &t->fl, sizeof (fl)); + fl.proto = IPPROTO_IPV6; + + dsfield = ipv6_get_dsfield(ipv6h); + if ((t->parms.flags & IP6_TNL_F_USE_ORIG_TCLASS)) + fl.fl6_flowlabel |= (*(__be32 *) ipv6h & IPV6_TCLASS_MASK); + if ((t->parms.flags & IP6_TNL_F_USE_ORIG_FLOWLABEL)) + fl.fl6_flowlabel |= (*(__be32 *) ipv6h & IPV6_FLOWLABEL_MASK); + + err = ip6_tnl_xmit2(skb, dev, dsfield, &fl, encap_limit, &mtu); + if (err != 0) { + if (err == -EMSGSIZE) + icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu); + return -1; + } + + return 0; +} + +static netdev_tx_t +ip6_tnl_xmit(struct sk_buff *skb, struct net_device *dev) +{ + struct ip6_tnl *t = netdev_priv(dev); + struct net_device_stats *stats = &t->dev->stats; + int ret; + + switch (skb->protocol) { + case htons(ETH_P_IP): + ret = ip4ip6_tnl_xmit(skb, dev); + break; + case htons(ETH_P_IPV6): + ret = ip6ip6_tnl_xmit(skb, dev); + break; + default: + goto tx_err; + } + + if (ret < 0) + goto tx_err; + + return NETDEV_TX_OK; + tx_err: stats->tx_errors++; stats->tx_dropped++; kfree_skb(skb); - t->recursion--; - return 0; + return NETDEV_TX_OK; } static void ip6_tnl_set_cap(struct ip6_tnl *t) { struct ip6_tnl_parm *p = &t->parms; - struct in6_addr *laddr = &p->laddr; - struct in6_addr *raddr = &p->raddr; - int ltype = ipv6_addr_type(laddr); - int rtype = ipv6_addr_type(raddr); + int ltype = ipv6_addr_type(&p->laddr); + int rtype = ipv6_addr_type(&p->raddr); p->flags &= ~(IP6_TNL_F_CAP_XMIT|IP6_TNL_F_CAP_RCV); - if (ltype != IPV6_ADDR_ANY && rtype != IPV6_ADDR_ANY && - ((ltype|rtype) & - (IPV6_ADDR_UNICAST| - IPV6_ADDR_LOOPBACK|IPV6_ADDR_LINKLOCAL| - IPV6_ADDR_MAPPED|IPV6_ADDR_RESERVED)) == IPV6_ADDR_UNICAST) { - struct net_device *ldev = NULL; - int l_ok = 1; - int r_ok = 1; - - if (p->link) - ldev = dev_get_by_index(p->link); - - if (ltype&IPV6_ADDR_UNICAST && !ipv6_chk_addr(laddr, ldev, 0)) - l_ok = 0; - - if (rtype&IPV6_ADDR_UNICAST && ipv6_chk_addr(raddr, NULL, 0)) - r_ok = 0; - - if (l_ok && r_ok) { - if (ltype&IPV6_ADDR_UNICAST) - p->flags |= IP6_TNL_F_CAP_XMIT; - if (rtype&IPV6_ADDR_UNICAST) - p->flags |= IP6_TNL_F_CAP_RCV; - } - if (ldev) - dev_put(ldev); + if (ltype & (IPV6_ADDR_UNICAST|IPV6_ADDR_MULTICAST) && + rtype & (IPV6_ADDR_UNICAST|IPV6_ADDR_MULTICAST) && + !((ltype|rtype) & IPV6_ADDR_LOOPBACK) && + (!((ltype|rtype) & IPV6_ADDR_LINKLOCAL) || p->link)) { + if (ltype&IPV6_ADDR_UNICAST) + p->flags |= IP6_TNL_F_CAP_XMIT; + if (rtype&IPV6_ADDR_UNICAST) + p->flags |= IP6_TNL_F_CAP_RCV; } } -static void ip6ip6_tnl_link_config(struct ip6_tnl *t) +static void ip6_tnl_link_config(struct ip6_tnl *t) { struct net_device *dev = t->dev; struct ip6_tnl_parm *p = &t->parms; struct flowi *fl = &t->fl; - memcpy(&dev->dev_addr, &p->laddr, sizeof(struct in6_addr)); - memcpy(&dev->broadcast, &p->raddr, sizeof(struct in6_addr)); + memcpy(dev->dev_addr, &p->laddr, sizeof(struct in6_addr)); + memcpy(dev->broadcast, &p->raddr, sizeof(struct in6_addr)); /* Set up flowi template */ ipv6_addr_copy(&fl->fl6_src, &p->laddr); @@ -845,8 +1119,12 @@ static void ip6ip6_tnl_link_config(struct ip6_tnl *t) dev->iflink = p->link; if (p->flags & IP6_TNL_F_CAP_XMIT) { - struct rt6_info *rt = rt6_lookup(&p->raddr, &p->laddr, - p->link, 0); + int strict = (ipv6_addr_type(&p->raddr) & + (IPV6_ADDR_MULTICAST|IPV6_ADDR_LINKLOCAL)); + + struct rt6_info *rt = rt6_lookup(dev_net(dev), + &p->raddr, &p->laddr, + p->link, strict); if (rt == NULL) return; @@ -865,17 +1143,16 @@ static void ip6ip6_tnl_link_config(struct ip6_tnl *t) } /** - * ip6ip6_tnl_change - update the tunnel parameters + * ip6_tnl_change - update the tunnel parameters * @t: tunnel to be changed * @p: tunnel configuration parameters - * @active: != 0 if tunnel is ready for use * * Description: - * ip6ip6_tnl_change() updates the tunnel parameters + * ip6_tnl_change() updates the tunnel parameters **/ static int -ip6ip6_tnl_change(struct ip6_tnl *t, struct ip6_tnl_parm *p) +ip6_tnl_change(struct ip6_tnl *t, struct ip6_tnl_parm *p) { ipv6_addr_copy(&t->parms.laddr, &p->laddr); ipv6_addr_copy(&t->parms.raddr, &p->raddr); @@ -884,19 +1161,21 @@ ip6ip6_tnl_change(struct ip6_tnl *t, struct ip6_tnl_parm *p) t->parms.encap_limit = p->encap_limit; t->parms.flowinfo = p->flowinfo; t->parms.link = p->link; - ip6ip6_tnl_link_config(t); + t->parms.proto = p->proto; + ip6_tnl_dst_reset(t); + ip6_tnl_link_config(t); return 0; } /** - * ip6ip6_tnl_ioctl - configure ipv6 tunnels from userspace + * ip6_tnl_ioctl - configure ipv6 tunnels from userspace * @dev: virtual device associated with tunnel * @ifr: parameters passed from userspace * @cmd: command to be performed * * Description: - * ip6ip6_tnl_ioctl() is used for managing IPv6 tunnels - * from userspace. + * ip6_tnl_ioctl() is used for managing IPv6 tunnels + * from userspace. * * The possible commands are the following: * %SIOCGETTUNNEL: get tunnel parameters for device @@ -904,7 +1183,7 @@ ip6ip6_tnl_change(struct ip6_tnl *t, struct ip6_tnl_parm *p) * %SIOCCHGTUNNEL: change tunnel parameters to those given * %SIOCDELTUNNEL: delete tunnel * - * The fallback device "ip6tnl0", created during module + * The fallback device "ip6tnl0", created during module * initialization, can be used for creating other tunnel devices. * * Return: @@ -917,29 +1196,25 @@ ip6ip6_tnl_change(struct ip6_tnl *t, struct ip6_tnl_parm *p) **/ static int -ip6ip6_tnl_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) +ip6_tnl_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) { int err = 0; - int create; struct ip6_tnl_parm p; struct ip6_tnl *t = NULL; + struct net *net = dev_net(dev); + struct ip6_tnl_net *ip6n = net_generic(net, ip6_tnl_net_id); switch (cmd) { case SIOCGETTUNNEL: - if (dev == ip6ip6_fb_tnl_dev) { - if (copy_from_user(&p, - ifr->ifr_ifru.ifru_data, - sizeof (p))) { + if (dev == ip6n->fb_tnl_dev) { + if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof (p))) { err = -EFAULT; break; } - if ((err = ip6ip6_tnl_locate(&p, &t, 0)) == -ENODEV) - t = netdev_priv(dev); - else if (err) - break; - } else + t = ip6_tnl_locate(net, &p, 0); + } + if (t == NULL) t = netdev_priv(dev); - memcpy(&p, &t->parms, sizeof (p)); if (copy_to_user(ifr->ifr_ifru.ifru_data, &p, sizeof (p))) { err = -EFAULT; @@ -948,58 +1223,57 @@ ip6ip6_tnl_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) case SIOCADDTUNNEL: case SIOCCHGTUNNEL: err = -EPERM; - create = (cmd == SIOCADDTUNNEL); if (!capable(CAP_NET_ADMIN)) break; - if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof (p))) { - err = -EFAULT; + err = -EFAULT; + if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof (p))) break; - } - if (!create && dev != ip6ip6_fb_tnl_dev) { - t = netdev_priv(dev); - } - if (!t && (err = ip6ip6_tnl_locate(&p, &t, create))) { + err = -EINVAL; + if (p.proto != IPPROTO_IPV6 && p.proto != IPPROTO_IPIP && + p.proto != 0) break; - } - if (cmd == SIOCCHGTUNNEL) { - if (t->dev != dev) { - err = -EEXIST; - break; - } - ip6ip6_tnl_unlink(t); - err = ip6ip6_tnl_change(t, &p); - ip6ip6_tnl_link(t); + t = ip6_tnl_locate(net, &p, cmd == SIOCADDTUNNEL); + if (dev != ip6n->fb_tnl_dev && cmd == SIOCCHGTUNNEL) { + if (t != NULL) { + if (t->dev != dev) { + err = -EEXIST; + break; + } + } else + t = netdev_priv(dev); + + ip6_tnl_unlink(ip6n, t); + err = ip6_tnl_change(t, &p); + ip6_tnl_link(ip6n, t); netdev_state_change(dev); } - if (copy_to_user(ifr->ifr_ifru.ifru_data, - &t->parms, sizeof (p))) { - err = -EFAULT; - } else { + if (t) { err = 0; - } + if (copy_to_user(ifr->ifr_ifru.ifru_data, &t->parms, sizeof (p))) + err = -EFAULT; + + } else + err = (cmd == SIOCADDTUNNEL ? -ENOBUFS : -ENOENT); break; case SIOCDELTUNNEL: err = -EPERM; if (!capable(CAP_NET_ADMIN)) break; - if (dev == ip6ip6_fb_tnl_dev) { - if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, - sizeof (p))) { - err = -EFAULT; + if (dev == ip6n->fb_tnl_dev) { + err = -EFAULT; + if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof (p))) break; - } - err = ip6ip6_tnl_locate(&p, &t, 0); - if (err) + err = -ENOENT; + if ((t = ip6_tnl_locate(net, &p, 0)) == NULL) break; - if (t == netdev_priv(ip6ip6_fb_tnl_dev)) { - err = -EPERM; + err = -EPERM; + if (t->dev == ip6n->fb_tnl_dev) break; - } - } else { - t = netdev_priv(dev); + dev = t->dev; } - err = unregister_netdevice(t->dev); + err = 0; + unregister_netdevice(dev); break; default: err = -EINVAL; @@ -1008,20 +1282,7 @@ ip6ip6_tnl_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) } /** - * ip6ip6_tnl_get_stats - return the stats for tunnel device - * @dev: virtual device associated with tunnel - * - * Return: stats for device - **/ - -static struct net_device_stats * -ip6ip6_tnl_get_stats(struct net_device *dev) -{ - return &(((struct ip6_tnl *)netdev_priv(dev))->stat); -} - -/** - * ip6ip6_tnl_change_mtu - change mtu manually for tunnel device + * ip6_tnl_change_mtu - change mtu manually for tunnel device * @dev: virtual device associated with tunnel * @new_mtu: the new mtu * @@ -1031,7 +1292,7 @@ ip6ip6_tnl_get_stats(struct net_device *dev) **/ static int -ip6ip6_tnl_change_mtu(struct net_device *dev, int new_mtu) +ip6_tnl_change_mtu(struct net_device *dev, int new_mtu) { if (new_mtu < IPV6_MIN_MTU) { return -EINVAL; @@ -1040,109 +1301,155 @@ ip6ip6_tnl_change_mtu(struct net_device *dev, int new_mtu) return 0; } + +static const struct net_device_ops ip6_tnl_netdev_ops = { + .ndo_uninit = ip6_tnl_dev_uninit, + .ndo_start_xmit = ip6_tnl_xmit, + .ndo_do_ioctl = ip6_tnl_ioctl, + .ndo_change_mtu = ip6_tnl_change_mtu, +}; + /** - * ip6ip6_tnl_dev_setup - setup virtual tunnel device + * ip6_tnl_dev_setup - setup virtual tunnel device * @dev: virtual device associated with tunnel * * Description: * Initialize function pointers and device parameters **/ -static void ip6ip6_tnl_dev_setup(struct net_device *dev) +static void ip6_tnl_dev_setup(struct net_device *dev) { - SET_MODULE_OWNER(dev); - dev->uninit = ip6ip6_tnl_dev_uninit; + dev->netdev_ops = &ip6_tnl_netdev_ops; dev->destructor = free_netdev; - dev->hard_start_xmit = ip6ip6_tnl_xmit; - dev->get_stats = ip6ip6_tnl_get_stats; - dev->do_ioctl = ip6ip6_tnl_ioctl; - dev->change_mtu = ip6ip6_tnl_change_mtu; dev->type = ARPHRD_TUNNEL6; dev->hard_header_len = LL_MAX_HEADER + sizeof (struct ipv6hdr); dev->mtu = ETH_DATA_LEN - sizeof (struct ipv6hdr); dev->flags |= IFF_NOARP; dev->addr_len = sizeof(struct in6_addr); + dev->features |= NETIF_F_NETNS_LOCAL; } /** - * ip6ip6_tnl_dev_init_gen - general initializer for all tunnel devices + * ip6_tnl_dev_init_gen - general initializer for all tunnel devices * @dev: virtual device associated with tunnel **/ static inline void -ip6ip6_tnl_dev_init_gen(struct net_device *dev) +ip6_tnl_dev_init_gen(struct net_device *dev) { struct ip6_tnl *t = netdev_priv(dev); - t->fl.proto = IPPROTO_IPV6; t->dev = dev; strcpy(t->parms.name, dev->name); } /** - * ip6ip6_tnl_dev_init - initializer for all non fallback tunnel devices + * ip6_tnl_dev_init - initializer for all non fallback tunnel devices * @dev: virtual device associated with tunnel **/ -static int -ip6ip6_tnl_dev_init(struct net_device *dev) +static void ip6_tnl_dev_init(struct net_device *dev) { struct ip6_tnl *t = netdev_priv(dev); - ip6ip6_tnl_dev_init_gen(dev); - ip6ip6_tnl_link_config(t); - return 0; + ip6_tnl_dev_init_gen(dev); + ip6_tnl_link_config(t); } /** - * ip6ip6_fb_tnl_dev_init - initializer for fallback tunnel device + * ip6_fb_tnl_dev_init - initializer for fallback tunnel device * @dev: fallback device * * Return: 0 **/ -static int -ip6ip6_fb_tnl_dev_init(struct net_device *dev) +static void __net_init ip6_fb_tnl_dev_init(struct net_device *dev) { struct ip6_tnl *t = netdev_priv(dev); - ip6ip6_tnl_dev_init_gen(dev); + struct net *net = dev_net(dev); + struct ip6_tnl_net *ip6n = net_generic(net, ip6_tnl_net_id); + + ip6_tnl_dev_init_gen(dev); + t->parms.proto = IPPROTO_IPV6; dev_hold(dev); - tnls_wc[0] = t; - return 0; + ip6n->tnls_wc[0] = t; } -#ifdef CONFIG_INET6_TUNNEL +static struct xfrm6_tunnel ip4ip6_handler = { + .handler = ip4ip6_rcv, + .err_handler = ip4ip6_err, + .priority = 1, +}; + static struct xfrm6_tunnel ip6ip6_handler = { .handler = ip6ip6_rcv, .err_handler = ip6ip6_err, + .priority = 1, }; -static inline int ip6ip6_register(void) +static void __net_exit ip6_tnl_destroy_tunnels(struct ip6_tnl_net *ip6n) { - return xfrm6_tunnel_register(&ip6ip6_handler); -} + int h; + struct ip6_tnl *t; + LIST_HEAD(list); -static inline int ip6ip6_unregister(void) -{ - return xfrm6_tunnel_deregister(&ip6ip6_handler); + for (h = 0; h < HASH_SIZE; h++) { + t = ip6n->tnls_r_l[h]; + while (t != NULL) { + unregister_netdevice_queue(t->dev, &list); + t = t->next; + } + } + + t = ip6n->tnls_wc[0]; + unregister_netdevice_queue(t->dev, &list); + unregister_netdevice_many(&list); } -#else -static struct inet6_protocol xfrm6_tunnel_protocol = { - .handler = ip6ip6_rcv, - .err_handler = ip6ip6_err, - .flags = INET6_PROTO_NOPOLICY|INET6_PROTO_FINAL, -}; -static inline int ip6ip6_register(void) +static int __net_init ip6_tnl_init_net(struct net *net) { - return inet6_add_protocol(&xfrm6_tunnel_protocol, IPPROTO_IPV6); + struct ip6_tnl_net *ip6n = net_generic(net, ip6_tnl_net_id); + int err; + + ip6n->tnls[0] = ip6n->tnls_wc; + ip6n->tnls[1] = ip6n->tnls_r_l; + + err = -ENOMEM; + ip6n->fb_tnl_dev = alloc_netdev(sizeof(struct ip6_tnl), "ip6tnl0", + ip6_tnl_dev_setup); + + if (!ip6n->fb_tnl_dev) + goto err_alloc_dev; + dev_net_set(ip6n->fb_tnl_dev, net); + + ip6_fb_tnl_dev_init(ip6n->fb_tnl_dev); + + err = register_netdev(ip6n->fb_tnl_dev); + if (err < 0) + goto err_register; + return 0; + +err_register: + free_netdev(ip6n->fb_tnl_dev); +err_alloc_dev: + return err; } -static inline int ip6ip6_unregister(void) +static void __net_exit ip6_tnl_exit_net(struct net *net) { - return inet6_del_protocol(&xfrm6_tunnel_protocol, IPPROTO_IPV6); + struct ip6_tnl_net *ip6n = net_generic(net, ip6_tnl_net_id); + + rtnl_lock(); + ip6_tnl_destroy_tunnels(ip6n); + rtnl_unlock(); } -#endif + +static struct pernet_operations ip6_tnl_net_ops = { + .init = ip6_tnl_init_net, + .exit = ip6_tnl_exit_net, + .id = &ip6_tnl_net_id, + .size = sizeof(struct ip6_tnl_net), +}; /** * ip6_tunnel_init - register protocol and reserve needed resources @@ -1154,26 +1461,29 @@ static int __init ip6_tunnel_init(void) { int err; - if (ip6ip6_register() < 0) { - printk(KERN_ERR "ip6ip6 init: can't register tunnel\n"); - return -EAGAIN; - } - ip6ip6_fb_tnl_dev = alloc_netdev(sizeof(struct ip6_tnl), "ip6tnl0", - ip6ip6_tnl_dev_setup); + err = register_pernet_device(&ip6_tnl_net_ops); + if (err < 0) + goto out_pernet; - if (!ip6ip6_fb_tnl_dev) { - err = -ENOMEM; - goto fail; + err = xfrm6_tunnel_register(&ip4ip6_handler, AF_INET); + if (err < 0) { + printk(KERN_ERR "ip6_tunnel init: can't register ip4ip6\n"); + goto out_ip4ip6; } - ip6ip6_fb_tnl_dev->init = ip6ip6_fb_tnl_dev_init; - if ((err = register_netdev(ip6ip6_fb_tnl_dev))) { - free_netdev(ip6ip6_fb_tnl_dev); - goto fail; + err = xfrm6_tunnel_register(&ip6ip6_handler, AF_INET6); + if (err < 0) { + printk(KERN_ERR "ip6_tunnel init: can't register ip6ip6\n"); + goto out_ip6ip6; } + return 0; -fail: - ip6ip6_unregister(); + +out_ip6ip6: + xfrm6_tunnel_deregister(&ip4ip6_handler, AF_INET); +out_ip4ip6: + unregister_pernet_device(&ip6_tnl_net_ops); +out_pernet: return err; } @@ -1183,10 +1493,13 @@ fail: static void __exit ip6_tunnel_cleanup(void) { - if (ip6ip6_unregister() < 0) - printk(KERN_INFO "ip6ip6 close: can't deregister tunnel\n"); + if (xfrm6_tunnel_deregister(&ip4ip6_handler, AF_INET)) + printk(KERN_INFO "ip6_tunnel close: can't deregister ip4ip6\n"); + + if (xfrm6_tunnel_deregister(&ip6ip6_handler, AF_INET6)) + printk(KERN_INFO "ip6_tunnel close: can't deregister ip6ip6\n"); - unregister_netdev(ip6ip6_fb_tnl_dev); + unregister_pernet_device(&ip6_tnl_net_ops); } module_init(ip6_tunnel_init);