From 58c4fb86eabcbc385d954843a635b7f4327be6b0 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Wed, 21 Dec 2005 22:56:42 +0900 Subject: [IPV6]: Flag RTF_ANYCAST for anycast routes. Signed-off-by: YOSHIFUJI Hideaki --- net/ipv6/route.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) (limited to 'net/ipv6') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 7c68bfbee36..66140f13d11 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -413,11 +413,14 @@ static struct rt6_info *rt6_cow(struct rt6_info *ort, struct in6_addr *daddr, rt = ip6_rt_copy(ort); if (rt) { - ipv6_addr_copy(&rt->rt6i_dst.addr, daddr); - - if (!(rt->rt6i_flags&RTF_GATEWAY)) + if (!(rt->rt6i_flags&RTF_GATEWAY)) { + if (rt->rt6i_dst.plen != 128 && + ipv6_addr_equal(&rt->rt6i_dst.addr, daddr)) + rt->rt6i_flags |= RTF_ANYCAST; ipv6_addr_copy(&rt->rt6i_gateway, daddr); + } + ipv6_addr_copy(&rt->rt6i_dst.addr, daddr); rt->rt6i_dst.plen = 128; rt->rt6i_flags |= RTF_CACHE; rt->u.dst.flags |= DST_HOST; @@ -1413,7 +1416,9 @@ struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev, rt->u.dst.obsolete = -1; rt->rt6i_flags = RTF_UP | RTF_NONEXTHOP; - if (!anycast) + if (anycast) + rt->rt6i_flags |= RTF_ANYCAST; + else rt->rt6i_flags |= RTF_LOCAL; rt->rt6i_nexthop = ndisc_get_neigh(rt->rt6i_dev, &rt->rt6i_gateway); if (rt->rt6i_nexthop == NULL) { -- cgit v1.2.3-70-g09d2 From 8de3351e6e0a1081fbf6864ae37839e327699a08 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Wed, 21 Dec 2005 22:57:06 +0900 Subject: [IPV6]: Try not to send icmp to anycast address. Signed-off-by: YOSHIFUJI Hideaki --- net/ipv6/icmp.c | 16 ++++++++++++++-- 1 file changed, 14 insertions(+), 2 deletions(-) (limited to 'net/ipv6') diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 34a332225c1..6ec6a2b549b 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -328,8 +328,10 @@ void icmpv6_send(struct sk_buff *skb, int type, int code, __u32 info, iif = skb->dev->ifindex; /* - * Must not send if we know that source is Anycast also. - * for now we don't know that. + * Must not send error if the source does not uniquely + * identify a single node (RFC2463 Section 2.4). + * We check unspecified / multicast addresses here, + * and anycast addresses will be checked later. */ if ((addr_type == IPV6_ADDR_ANY) || (addr_type & IPV6_ADDR_MULTICAST)) { LIMIT_NETDEBUG(KERN_DEBUG "icmpv6_send: addr_any/mcast source\n"); @@ -373,6 +375,16 @@ void icmpv6_send(struct sk_buff *skb, int type, int code, __u32 info, err = ip6_dst_lookup(sk, &dst, &fl); if (err) goto out; + + /* + * We won't send icmp if the destination is known + * anycast. + */ + if (((struct rt6_info *)dst)->rt6i_flags & RTF_ANYCAST) { + LIMIT_NETDEBUG(KERN_DEBUG "icmpv6_send: acast source\n"); + goto out_dst_release; + } + if ((err = xfrm_lookup(&dst, &fl, sk, 0)) < 0) goto out; -- cgit v1.2.3-70-g09d2 From 3c21edbd113788b110116141c8078623a0900b6a Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Wed, 21 Dec 2005 22:57:24 +0900 Subject: [IPV6]: Defer IPv6 device initialization until the link becomes ready. NETDEV_UP might be sent even if the link attached to the interface was not ready. DAD does not make sense in such case, so we won't do so. After interface Signed-off-by: YOSHIFUJI Hideaki --- include/net/if_inet6.h | 1 + net/ipv6/addrconf.c | 74 +++++++++++++++++++++++++++++++++++++++++++------- 2 files changed, 65 insertions(+), 10 deletions(-) (limited to 'net/ipv6') diff --git a/include/net/if_inet6.h b/include/net/if_inet6.h index e97a9accb71..d8234f9bd4c 100644 --- a/include/net/if_inet6.h +++ b/include/net/if_inet6.h @@ -24,6 +24,7 @@ #define IF_RA_MANAGED 0x40 #define IF_RA_RCVD 0x20 #define IF_RS_SENT 0x10 +#define IF_READY 0x80000000 /* prefix flags */ #define IF_PREFIX_ONLINK 0x01 diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 4ea8cf7c0cc..d012f6ac704 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -388,6 +388,9 @@ static struct inet6_dev * ipv6_add_dev(struct net_device *dev) } #endif + if (netif_carrier_ok(dev)) + ndev->if_flags |= IF_READY; + write_lock_bh(&addrconf_lock); dev->ip6_ptr = ndev; write_unlock_bh(&addrconf_lock); @@ -1215,10 +1218,8 @@ int ipv6_rcv_saddr_equal(const struct sock *sk, const struct sock *sk2) /* Gets referenced address, destroys ifaddr */ -void addrconf_dad_failure(struct inet6_ifaddr *ifp) +void addrconf_dad_stop(struct inet6_ifaddr *ifp) { - if (net_ratelimit()) - printk(KERN_INFO "%s: duplicate address detected!\n", ifp->idev->dev->name); if (ifp->flags&IFA_F_PERMANENT) { spin_lock_bh(&ifp->lock); addrconf_del_timer(ifp); @@ -1244,6 +1245,12 @@ void addrconf_dad_failure(struct inet6_ifaddr *ifp) ipv6_del_addr(ifp); } +void addrconf_dad_failure(struct inet6_ifaddr *ifp) +{ + if (net_ratelimit()) + printk(KERN_INFO "%s: duplicate address detected!\n", ifp->idev->dev->name); + addrconf_dad_stop(ifp); +} /* Join to solicited addr multicast group. */ @@ -2136,6 +2143,37 @@ static int addrconf_notify(struct notifier_block *this, unsigned long event, switch(event) { case NETDEV_UP: + case NETDEV_CHANGE: + if (event == NETDEV_UP) { + if (!netif_carrier_ok(dev)) { + /* device is not ready yet. */ + printk(KERN_INFO + "ADDRCONF(NETDEV_UP): %s: " + "link is not ready\n", + dev->name); + break; + } + } else { + if (!netif_carrier_ok(dev)) { + /* device is still not ready. */ + break; + } + + if (idev) { + if (idev->if_flags & IF_READY) { + /* device is already configured. */ + break; + } + idev->if_flags |= IF_READY; + } + + printk(KERN_INFO + "ADDRCONF(NETDEV_CHANGE): %s: " + "link becomes ready\n", + dev->name); + + } + switch(dev->type) { case ARPHRD_SIT: addrconf_sit_config(dev); @@ -2186,8 +2224,7 @@ static int addrconf_notify(struct notifier_block *this, unsigned long event, */ addrconf_ifdown(dev, event != NETDEV_DOWN); break; - case NETDEV_CHANGE: - break; + case NETDEV_CHANGENAME: #ifdef CONFIG_SYSCTL if (idev) { @@ -2268,7 +2305,7 @@ static int addrconf_ifdown(struct net_device *dev, int how) /* Step 3: clear flags for stateless addrconf */ if (how != 1) - idev->if_flags &= ~(IF_RS_SENT|IF_RA_RCVD); + idev->if_flags &= ~(IF_RS_SENT|IF_RA_RCVD|IF_READY); /* Step 4: clear address list */ #ifdef CONFIG_IPV6_PRIVACY @@ -2377,11 +2414,20 @@ out: /* * Duplicate Address Detection */ +static void addrconf_dad_kick(struct inet6_ifaddr *ifp) +{ + unsigned long rand_num; + struct inet6_dev *idev = ifp->idev; + + rand_num = net_random() % (idev->cnf.rtr_solicit_delay ? : 1); + ifp->probes = idev->cnf.dad_transmits; + addrconf_mod_timer(ifp, AC_DAD, rand_num); +} + static void addrconf_dad_start(struct inet6_ifaddr *ifp, u32 flags) { struct inet6_dev *idev = ifp->idev; struct net_device *dev = idev->dev; - unsigned long rand_num; addrconf_join_solict(dev, &ifp->addr); @@ -2390,7 +2436,6 @@ static void addrconf_dad_start(struct inet6_ifaddr *ifp, u32 flags) flags); net_srandom(ifp->addr.s6_addr32[3]); - rand_num = net_random() % (idev->cnf.rtr_solicit_delay ? : 1); read_lock_bh(&idev->lock); if (ifp->dead) @@ -2407,8 +2452,17 @@ static void addrconf_dad_start(struct inet6_ifaddr *ifp, u32 flags) return; } - ifp->probes = idev->cnf.dad_transmits; - addrconf_mod_timer(ifp, AC_DAD, rand_num); + if (idev->if_flags & IF_READY) + addrconf_dad_kick(ifp); + else { + /* + * If the defice is not ready: + * - keep it tentative if it is a permanent address. + * - otherwise, kill it. + */ + in6_ifa_hold(ifp); + addrconf_dad_stop(ifp); + } spin_unlock_bh(&ifp->lock); out: -- cgit v1.2.3-70-g09d2 From c5e33bddd3c798f681f8f3027270127be6b61a3b Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Wed, 21 Dec 2005 22:57:44 +0900 Subject: [IPV6]: Run DAD when the link becomes ready. If the link was not available when the interface was created, run DAD for pending tentative addresses when the link becomes ready. Signed-off-by: YOSHIFUJI Hideaki --- net/ipv6/addrconf.c | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+) (limited to 'net/ipv6') diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index d012f6ac704..f6ead6a843e 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -137,6 +137,7 @@ static int addrconf_ifdown(struct net_device *dev, int how); static void addrconf_dad_start(struct inet6_ifaddr *ifp, u32 flags); static void addrconf_dad_timer(unsigned long data); static void addrconf_dad_completed(struct inet6_ifaddr *ifp); +static void addrconf_dad_run(struct inet6_dev *idev); static void addrconf_rs_timer(unsigned long data); static void __ipv6_ifa_notify(int event, struct inet6_ifaddr *ifa); static void ipv6_ifa_notify(int event, struct inet6_ifaddr *ifa); @@ -418,6 +419,7 @@ static struct inet6_dev * ipv6_find_idev(struct net_device *dev) if ((idev = ipv6_add_dev(dev)) == NULL) return NULL; } + if (dev->flags&IFF_UP) ipv6_mc_up(idev); return idev; @@ -2140,6 +2142,7 @@ static int addrconf_notify(struct notifier_block *this, unsigned long event, { struct net_device *dev = (struct net_device *) data; struct inet6_dev *idev = __in6_dev_get(dev); + int run_pending = 0; switch(event) { case NETDEV_UP: @@ -2172,6 +2175,7 @@ static int addrconf_notify(struct notifier_block *this, unsigned long event, "link becomes ready\n", dev->name); + run_pending = 1; } switch(dev->type) { @@ -2190,6 +2194,9 @@ static int addrconf_notify(struct notifier_block *this, unsigned long event, break; }; if (idev) { + if (run_pending) + addrconf_dad_run(idev); + /* If the MTU changed during the interface down, when the interface up, the changed MTU must be reflected in the idev as well as routers. @@ -2546,6 +2553,22 @@ static void addrconf_dad_completed(struct inet6_ifaddr *ifp) } } +static void addrconf_dad_run(struct inet6_dev *idev) { + struct inet6_ifaddr *ifp; + + read_lock_bh(&idev->lock); + for (ifp = idev->addr_list; ifp; ifp = ifp->if_next) { + spin_lock_bh(&ifp->lock); + if (!(ifp->flags & IFA_F_TENTATIVE)) { + spin_unlock_bh(&ifp->lock); + continue; + } + spin_unlock_bh(&ifp->lock); + addrconf_dad_kick(ifp); + } + read_unlock_bh(&idev->lock); +} + #ifdef CONFIG_PROC_FS struct if6_iter_state { int bucket; -- cgit v1.2.3-70-g09d2 From 6b3ae80a63e47f6e97d68a1ddd520e3509e62821 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Wed, 21 Dec 2005 22:58:01 +0900 Subject: [IPV6]: Don't select a tentative address as a source address. A tentative address is not considered "assigned to an interface" in the traditional sense (RFC2462 Section 4). Don't try to select such an address for the source address. Signed-off-by: YOSHIFUJI Hideaki --- net/ipv6/addrconf.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) (limited to 'net/ipv6') diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index f6ead6a843e..fd03c394436 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -908,11 +908,18 @@ int ipv6_dev_get_saddr(struct net_device *daddr_dev, score.addr_type = __ipv6_addr_type(&ifa->addr); - /* Rule 0: Candidate Source Address (section 4) + /* Rule 0: + * - Tentative Address (RFC2462 section 5.4) + * - A tentative address is not considered + * "assigned to an interface" in the traditional + * sense. + * - Candidate Source Address (section 4) * - In any case, anycast addresses, multicast * addresses, and the unspecified address MUST * NOT be included in a candidate set. */ + if (ifa->flags & IFA_F_TENTATIVE) + continue; if (unlikely(score.addr_type == IPV6_ADDR_ANY || score.addr_type & IPV6_ADDR_MULTICAST)) { LIMIT_NETDEBUG(KERN_DEBUG -- cgit v1.2.3-70-g09d2 From 1d1428045c54ef3d172d480806e2066dde0b4b76 Mon Sep 17 00:00:00 2001 From: Kristian Slavov Date: Wed, 21 Dec 2005 18:47:24 -0800 Subject: [IPV6]: Fix address deletion If you add more than one IPv6 address belonging to the same prefix and delete the address that was last added, routing table entry for that prefix is also deleted. Tested on 2.6.14.4 To reproduce: ip addr add 3ffe::1/64 dev eth0 ip addr add 3ffe::2/64 dev eth0 /* wait DAD */ sleep 1 ip addr del 3ffe::2/64 dev eth0 ip -6 route (route to 3ffe::/64 should be gone) In ipv6_del_addr(), if ifa == ifp, we set ifa->if_next to NULL, and later assign ifap = &ifa->if_next, effectively terminating the for-loop. This prevents us from checking if there are other addresses using the same prefix that are valid, and thus resulting in deletion of the prefix. This applies only if the first entry in idev->addr_list is the address to be deleted. Signed-off-by: Kristian Slavov Acked-by: YOSHIFUJI Hideaki Signed-off-by: David S. Miller --- net/ipv6/addrconf.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'net/ipv6') diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 4ea8cf7c0cc..e717a034c95 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -634,8 +634,7 @@ static void ipv6_del_addr(struct inet6_ifaddr *ifp) } #endif - for (ifap = &idev->addr_list; (ifa=*ifap) != NULL; - ifap = &ifa->if_next) { + for (ifap = &idev->addr_list; (ifa=*ifap) != NULL;) { if (ifa == ifp) { *ifap = ifa->if_next; __in6_ifa_put(ifp); @@ -643,6 +642,7 @@ static void ipv6_del_addr(struct inet6_ifaddr *ifp) if (!(ifp->flags & IFA_F_PERMANENT) || onlink > 0) break; deleted = 1; + continue; } else if (ifp->flags & IFA_F_PERMANENT) { if (ipv6_prefix_equal(&ifa->addr, &ifp->addr, ifp->prefix_len)) { @@ -666,6 +666,7 @@ static void ipv6_del_addr(struct inet6_ifaddr *ifp) } } } + ifap = &ifa->if_next; } write_unlock_bh(&idev->lock); -- cgit v1.2.3-70-g09d2