From 5d0bbeeb144f631150881712607345c532e38e7e Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Fri, 4 Aug 2006 03:37:36 -0700 Subject: [IPV6]: Remove ndiscs rt6_lock dependency (Ab)using rt6_lock wouldn't work anymore if rt6_lock is converted into a per table lock. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- net/ipv6/route.c | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-) (limited to 'net/ipv6/route.c') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index d9baca0..ce1f49b 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -747,8 +747,6 @@ static void ip6_rt_update_pmtu(struct dst_entry *dst, u32 mtu) } } -/* Protected by rt6_lock. */ -static struct dst_entry *ndisc_dst_gc_list; static int ipv6_get_mtu(struct net_device *dev); static inline unsigned int ipv6_advmss(unsigned int mtu) @@ -769,6 +767,9 @@ static inline unsigned int ipv6_advmss(unsigned int mtu) return mtu; } +static struct dst_entry *ndisc_dst_gc_list; +DEFINE_SPINLOCK(ndisc_lock); + struct dst_entry *ndisc_dst_alloc(struct net_device *dev, struct neighbour *neigh, struct in6_addr *addr, @@ -809,10 +810,10 @@ struct dst_entry *ndisc_dst_alloc(struct net_device *dev, rt->rt6i_dst.plen = 128; #endif - write_lock_bh(&rt6_lock); + spin_lock_bh(&ndisc_lock); rt->u.dst.next = ndisc_dst_gc_list; ndisc_dst_gc_list = &rt->u.dst; - write_unlock_bh(&rt6_lock); + spin_unlock_bh(&ndisc_lock); fib6_force_start_gc(); @@ -826,8 +827,11 @@ int ndisc_dst_gc(int *more) int freed; next = NULL; + freed = 0; + + spin_lock_bh(&ndisc_lock); pprev = &ndisc_dst_gc_list; - freed = 0; + while ((dst = *pprev) != NULL) { if (!atomic_read(&dst->__refcnt)) { *pprev = dst->next; @@ -839,6 +843,8 @@ int ndisc_dst_gc(int *more) } } + spin_unlock_bh(&ndisc_lock); + return freed; } -- cgit v1.1 From c71099acce933455123ee505cc75964610a209ad Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Fri, 4 Aug 2006 23:20:06 -0700 Subject: [IPV6]: Multiple Routing Tables Adds the framework to support multiple IPv6 routing tables. Currently all automatically generated routes are put into the same table. This could be changed at a later point after considering the produced locking overhead. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- net/ipv6/route.c | 380 ++++++++++++++++++++++++++++++++++++------------------- 1 file changed, 252 insertions(+), 128 deletions(-) (limited to 'net/ipv6/route.c') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index ce1f49b..73efdad 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -140,16 +140,6 @@ struct rt6_info ip6_null_entry = { .rt6i_ref = ATOMIC_INIT(1), }; -struct fib6_node ip6_routing_table = { - .leaf = &ip6_null_entry, - .fn_flags = RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO, -}; - -/* Protects all the ip6 fib */ - -DEFINE_RWLOCK(rt6_lock); - - /* allocate dst with ip6_dst_ops */ static __inline__ struct rt6_info *ip6_dst_alloc(void) { @@ -188,8 +178,14 @@ static __inline__ int rt6_check_expired(const struct rt6_info *rt) time_after(jiffies, rt->rt6i_expires)); } +static inline int rt6_need_strict(struct in6_addr *daddr) +{ + return (ipv6_addr_type(daddr) & + (IPV6_ADDR_MULTICAST | IPV6_ADDR_LINKLOCAL)); +} + /* - * Route lookup. Any rt6_lock is implied. + * Route lookup. Any table->tb6_lock is implied. */ static __inline__ struct rt6_info *rt6_device_match(struct rt6_info *rt, @@ -441,27 +437,66 @@ int rt6_route_rcv(struct net_device *dev, u8 *opt, int len, } #endif -struct rt6_info *rt6_lookup(struct in6_addr *daddr, struct in6_addr *saddr, - int oif, int strict) +#define BACKTRACK() \ +if (rt == &ip6_null_entry && flags & RT6_F_STRICT) { \ + while ((fn = fn->parent) != NULL) { \ + if (fn->fn_flags & RTN_TL_ROOT) { \ + dst_hold(&rt->u.dst); \ + goto out; \ + } \ + if (fn->fn_flags & RTN_RTINFO) \ + goto restart; \ + } \ +} + +static struct rt6_info *ip6_pol_route_lookup(struct fib6_table *table, + struct flowi *fl, int flags) { struct fib6_node *fn; struct rt6_info *rt; - read_lock_bh(&rt6_lock); - fn = fib6_lookup(&ip6_routing_table, daddr, saddr); - rt = rt6_device_match(fn->leaf, oif, strict); + read_lock_bh(&table->tb6_lock); + fn = fib6_lookup(&table->tb6_root, &fl->fl6_dst, &fl->fl6_src); +restart: + rt = fn->leaf; + rt = rt6_device_match(rt, fl->oif, flags & RT6_F_STRICT); + BACKTRACK(); dst_hold(&rt->u.dst); - rt->u.dst.__use++; - read_unlock_bh(&rt6_lock); +out: + read_unlock_bh(&table->tb6_lock); rt->u.dst.lastuse = jiffies; - if (rt->u.dst.error == 0) - return rt; - dst_release(&rt->u.dst); + rt->u.dst.__use++; + + return rt; + +} + +struct rt6_info *rt6_lookup(struct in6_addr *daddr, struct in6_addr *saddr, + int oif, int strict) +{ + struct flowi fl = { + .oif = oif, + .nl_u = { + .ip6_u = { + .daddr = *daddr, + /* TODO: saddr */ + }, + }, + }; + struct dst_entry *dst; + int flags = strict ? RT6_F_STRICT : 0; + + dst = fib6_rule_lookup(&fl, flags, ip6_pol_route_lookup); + if (dst->error == 0) + return (struct rt6_info *) dst; + + dst_release(dst); + return NULL; } -/* ip6_ins_rt is called with FREE rt6_lock. +/* ip6_ins_rt is called with FREE table->tb6_lock. It takes new route entry, the addition fails by any reason the route is freed. In any case, if caller does not hold it, it may be destroyed. @@ -471,10 +506,12 @@ int ip6_ins_rt(struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr, struct netlink_skb_parms *req) { int err; + struct fib6_table *table; - write_lock_bh(&rt6_lock); - err = fib6_add(&ip6_routing_table, rt, nlh, _rtattr, req); - write_unlock_bh(&rt6_lock); + table = rt->rt6i_table; + write_lock_bh(&table->tb6_lock); + err = fib6_add(&table->tb6_root, rt, nlh, _rtattr, req); + write_unlock_bh(&table->tb6_lock); return err; } @@ -532,51 +569,40 @@ static struct rt6_info *rt6_alloc_clone(struct rt6_info *ort, struct in6_addr *d return rt; } -#define BACKTRACK() \ -if (rt == &ip6_null_entry) { \ - while ((fn = fn->parent) != NULL) { \ - if (fn->fn_flags & RTN_ROOT) { \ - goto out; \ - } \ - if (fn->fn_flags & RTN_RTINFO) \ - goto restart; \ - } \ -} - - -void ip6_route_input(struct sk_buff *skb) +struct rt6_info *ip6_pol_route_input(struct fib6_table *table, struct flowi *fl, + int flags) { struct fib6_node *fn; struct rt6_info *rt, *nrt; - int strict; + int strict = 0; int attempts = 3; int err; int reachable = RT6_SELECT_F_REACHABLE; - strict = ipv6_addr_type(&skb->nh.ipv6h->daddr) & (IPV6_ADDR_MULTICAST|IPV6_ADDR_LINKLOCAL) ? RT6_SELECT_F_IFACE : 0; + if (flags & RT6_F_STRICT) + strict = RT6_SELECT_F_IFACE; relookup: - read_lock_bh(&rt6_lock); + read_lock_bh(&table->tb6_lock); restart_2: - fn = fib6_lookup(&ip6_routing_table, &skb->nh.ipv6h->daddr, - &skb->nh.ipv6h->saddr); + fn = fib6_lookup(&table->tb6_root, &fl->fl6_dst, &fl->fl6_src); restart: - rt = rt6_select(&fn->leaf, skb->dev->ifindex, strict | reachable); + rt = rt6_select(&fn->leaf, fl->iif, strict | reachable); BACKTRACK(); if (rt == &ip6_null_entry || rt->rt6i_flags & RTF_CACHE) goto out; dst_hold(&rt->u.dst); - read_unlock_bh(&rt6_lock); + read_unlock_bh(&table->tb6_lock); if (!rt->rt6i_nexthop && !(rt->rt6i_flags & RTF_NONEXTHOP)) - nrt = rt6_alloc_cow(rt, &skb->nh.ipv6h->daddr, &skb->nh.ipv6h->saddr); + nrt = rt6_alloc_cow(rt, &fl->fl6_dst, &fl->fl6_src); else { #if CLONE_OFFLINK_ROUTE - nrt = rt6_alloc_clone(rt, &skb->nh.ipv6h->daddr); + nrt = rt6_alloc_clone(rt, &fl->fl6_dst); #else goto out2; #endif @@ -587,7 +613,7 @@ restart: dst_hold(&rt->u.dst); if (nrt) { - err = ip6_ins_rt(nrt, NULL, NULL, &NETLINK_CB(skb)); + err = ip6_ins_rt(nrt, NULL, NULL, NULL); if (!err) goto out2; } @@ -596,7 +622,7 @@ restart: goto out2; /* - * Race condition! In the gap, when rt6_lock was + * Race condition! In the gap, when table->tb6_lock was * released someone could insert this route. Relookup. */ dst_release(&rt->u.dst); @@ -608,30 +634,54 @@ out: goto restart_2; } dst_hold(&rt->u.dst); - read_unlock_bh(&rt6_lock); + read_unlock_bh(&table->tb6_lock); out2: rt->u.dst.lastuse = jiffies; rt->u.dst.__use++; - skb->dst = (struct dst_entry *) rt; - return; + + return rt; } -struct dst_entry * ip6_route_output(struct sock *sk, struct flowi *fl) +void ip6_route_input(struct sk_buff *skb) +{ + struct ipv6hdr *iph = skb->nh.ipv6h; + struct flowi fl = { + .iif = skb->dev->ifindex, + .nl_u = { + .ip6_u = { + .daddr = iph->daddr, + .saddr = iph->saddr, + .flowlabel = (* (u32 *) iph)&IPV6_FLOWINFO_MASK, + }, + }, + .proto = iph->nexthdr, + }; + int flags = 0; + + if (rt6_need_strict(&iph->daddr)) + flags |= RT6_F_STRICT; + + skb->dst = fib6_rule_lookup(&fl, flags, ip6_pol_route_input); +} + +static struct rt6_info *ip6_pol_route_output(struct fib6_table *table, + struct flowi *fl, int flags) { struct fib6_node *fn; struct rt6_info *rt, *nrt; - int strict; + int strict = 0; int attempts = 3; int err; int reachable = RT6_SELECT_F_REACHABLE; - strict = ipv6_addr_type(&fl->fl6_dst) & (IPV6_ADDR_MULTICAST|IPV6_ADDR_LINKLOCAL) ? RT6_SELECT_F_IFACE : 0; + if (flags & RT6_F_STRICT) + strict = RT6_SELECT_F_IFACE; relookup: - read_lock_bh(&rt6_lock); + read_lock_bh(&table->tb6_lock); restart_2: - fn = fib6_lookup(&ip6_routing_table, &fl->fl6_dst, &fl->fl6_src); + fn = fib6_lookup(&table->tb6_root, &fl->fl6_dst, &fl->fl6_src); restart: rt = rt6_select(&fn->leaf, fl->oif, strict | reachable); @@ -641,7 +691,7 @@ restart: goto out; dst_hold(&rt->u.dst); - read_unlock_bh(&rt6_lock); + read_unlock_bh(&table->tb6_lock); if (!rt->rt6i_nexthop && !(rt->rt6i_flags & RTF_NONEXTHOP)) nrt = rt6_alloc_cow(rt, &fl->fl6_dst, &fl->fl6_src); @@ -667,7 +717,7 @@ restart: goto out2; /* - * Race condition! In the gap, when rt6_lock was + * Race condition! In the gap, when table->tb6_lock was * released someone could insert this route. Relookup. */ dst_release(&rt->u.dst); @@ -679,11 +729,21 @@ out: goto restart_2; } dst_hold(&rt->u.dst); - read_unlock_bh(&rt6_lock); + read_unlock_bh(&table->tb6_lock); out2: rt->u.dst.lastuse = jiffies; rt->u.dst.__use++; - return &rt->u.dst; + return rt; +} + +struct dst_entry * ip6_route_output(struct sock *sk, struct flowi *fl) +{ + int flags = 0; + + if (rt6_need_strict(&fl->fl6_dst)) + flags |= RT6_F_STRICT; + + return fib6_rule_lookup(fl, flags, ip6_pol_route_output); } @@ -906,7 +966,8 @@ int ipv6_get_hoplimit(struct net_device *dev) */ int ip6_route_add(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, - void *_rtattr, struct netlink_skb_parms *req) + void *_rtattr, struct netlink_skb_parms *req, + u32 table_id) { int err; struct rtmsg *r; @@ -914,6 +975,7 @@ int ip6_route_add(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, struct rt6_info *rt = NULL; struct net_device *dev = NULL; struct inet6_dev *idev = NULL; + struct fib6_table *table; int addr_type; rta = (struct rtattr **) _rtattr; @@ -937,6 +999,12 @@ int ip6_route_add(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, if (rtmsg->rtmsg_metric == 0) rtmsg->rtmsg_metric = IP6_RT_PRIO_USER; + table = fib6_new_table(table_id); + if (table == NULL) { + err = -ENOBUFS; + goto out; + } + rt = ip6_dst_alloc(); if (rt == NULL) { @@ -1093,6 +1161,7 @@ install_route: rt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(dst_mtu(&rt->u.dst)); rt->u.dst.dev = dev; rt->rt6i_idev = idev; + rt->rt6i_table = table; return ip6_ins_rt(rt, nlh, _rtattr, req); out: @@ -1108,26 +1177,35 @@ out: int ip6_del_rt(struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr, struct netlink_skb_parms *req) { int err; + struct fib6_table *table; - write_lock_bh(&rt6_lock); + table = rt->rt6i_table; + write_lock_bh(&table->tb6_lock); err = fib6_del(rt, nlh, _rtattr, req); dst_release(&rt->u.dst); - write_unlock_bh(&rt6_lock); + write_unlock_bh(&table->tb6_lock); return err; } -static int ip6_route_del(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, void *_rtattr, struct netlink_skb_parms *req) +static int ip6_route_del(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, + void *_rtattr, struct netlink_skb_parms *req, + u32 table_id) { + struct fib6_table *table; struct fib6_node *fn; struct rt6_info *rt; int err = -ESRCH; - read_lock_bh(&rt6_lock); + table = fib6_get_table(table_id); + if (table == NULL) + return err; + + read_lock_bh(&table->tb6_lock); - fn = fib6_locate(&ip6_routing_table, + fn = fib6_locate(&table->tb6_root, &rtmsg->rtmsg_dst, rtmsg->rtmsg_dst_len, &rtmsg->rtmsg_src, rtmsg->rtmsg_src_len); @@ -1144,12 +1222,12 @@ static int ip6_route_del(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, void *_r rtmsg->rtmsg_metric != rt->rt6i_metric) continue; dst_hold(&rt->u.dst); - read_unlock_bh(&rt6_lock); + read_unlock_bh(&table->tb6_lock); return ip6_del_rt(rt, nlh, _rtattr, req); } } - read_unlock_bh(&rt6_lock); + read_unlock_bh(&table->tb6_lock); return err; } @@ -1161,10 +1239,15 @@ void rt6_redirect(struct in6_addr *dest, struct in6_addr *saddr, struct neighbour *neigh, u8 *lladdr, int on_link) { struct rt6_info *rt, *nrt = NULL; - int strict; struct fib6_node *fn; + struct fib6_table *table; struct netevent_redirect netevent; + /* TODO: Very lazy, might need to check all tables */ + table = fib6_get_table(RT6_TABLE_MAIN); + if (table == NULL) + return; + /* * Get the "current" route for this destination and * check if the redirect has come from approriate router. @@ -1175,10 +1258,9 @@ void rt6_redirect(struct in6_addr *dest, struct in6_addr *saddr, * is a bit fuzzy and one might need to check all possible * routes. */ - strict = ipv6_addr_type(dest) & (IPV6_ADDR_MULTICAST | IPV6_ADDR_LINKLOCAL); - read_lock_bh(&rt6_lock); - fn = fib6_lookup(&ip6_routing_table, dest, NULL); + read_lock_bh(&table->tb6_lock); + fn = fib6_lookup(&table->tb6_root, dest, NULL); restart: for (rt = fn->leaf; rt; rt = rt->u.next) { /* @@ -1201,7 +1283,7 @@ restart: } if (rt) dst_hold(&rt->u.dst); - else if (strict) { + else if (rt6_need_strict(dest)) { while ((fn = fn->parent) != NULL) { if (fn->fn_flags & RTN_ROOT) break; @@ -1209,7 +1291,7 @@ restart: goto restart; } } - read_unlock_bh(&rt6_lock); + read_unlock_bh(&table->tb6_lock); if (!rt) { if (net_ratelimit()) @@ -1384,6 +1466,7 @@ static struct rt6_info * ip6_rt_copy(struct rt6_info *ort) #ifdef CONFIG_IPV6_SUBTREES memcpy(&rt->rt6i_src, &ort->rt6i_src, sizeof(struct rt6key)); #endif + rt->rt6i_table = ort->rt6i_table; } return rt; } @@ -1394,9 +1477,14 @@ static struct rt6_info *rt6_get_route_info(struct in6_addr *prefix, int prefixle { struct fib6_node *fn; struct rt6_info *rt = NULL; + struct fib6_table *table; + + table = fib6_get_table(RT6_TABLE_INFO); + if (table == NULL) + return NULL; - write_lock_bh(&rt6_lock); - fn = fib6_locate(&ip6_routing_table, prefix ,prefixlen, NULL, 0); + write_lock_bh(&table->tb6_lock); + fn = fib6_locate(&table->tb6_root, prefix ,prefixlen, NULL, 0); if (!fn) goto out; @@ -1411,7 +1499,7 @@ static struct rt6_info *rt6_get_route_info(struct in6_addr *prefix, int prefixle break; } out: - write_unlock_bh(&rt6_lock); + write_unlock_bh(&table->tb6_lock); return rt; } @@ -1433,7 +1521,7 @@ static struct rt6_info *rt6_add_route_info(struct in6_addr *prefix, int prefixle rtmsg.rtmsg_flags |= RTF_DEFAULT; rtmsg.rtmsg_ifindex = ifindex; - ip6_route_add(&rtmsg, NULL, NULL, NULL); + ip6_route_add(&rtmsg, NULL, NULL, NULL, RT6_TABLE_INFO); return rt6_get_route_info(prefix, prefixlen, gwaddr, ifindex); } @@ -1442,12 +1530,14 @@ static struct rt6_info *rt6_add_route_info(struct in6_addr *prefix, int prefixle struct rt6_info *rt6_get_dflt_router(struct in6_addr *addr, struct net_device *dev) { struct rt6_info *rt; - struct fib6_node *fn; + struct fib6_table *table; - fn = &ip6_routing_table; + table = fib6_get_table(RT6_TABLE_DFLT); + if (table == NULL) + return NULL; - write_lock_bh(&rt6_lock); - for (rt = fn->leaf; rt; rt=rt->u.next) { + write_lock_bh(&table->tb6_lock); + for (rt = table->tb6_root.leaf; rt; rt=rt->u.next) { if (dev == rt->rt6i_dev && ((rt->rt6i_flags & (RTF_ADDRCONF | RTF_DEFAULT)) == (RTF_ADDRCONF | RTF_DEFAULT)) && ipv6_addr_equal(&rt->rt6i_gateway, addr)) @@ -1455,7 +1545,7 @@ struct rt6_info *rt6_get_dflt_router(struct in6_addr *addr, struct net_device *d } if (rt) dst_hold(&rt->u.dst); - write_unlock_bh(&rt6_lock); + write_unlock_bh(&table->tb6_lock); return rt; } @@ -1474,28 +1564,31 @@ struct rt6_info *rt6_add_dflt_router(struct in6_addr *gwaddr, rtmsg.rtmsg_ifindex = dev->ifindex; - ip6_route_add(&rtmsg, NULL, NULL, NULL); + ip6_route_add(&rtmsg, NULL, NULL, NULL, RT6_TABLE_DFLT); return rt6_get_dflt_router(gwaddr, dev); } void rt6_purge_dflt_routers(void) { struct rt6_info *rt; + struct fib6_table *table; + + /* NOTE: Keep consistent with rt6_get_dflt_router */ + table = fib6_get_table(RT6_TABLE_DFLT); + if (table == NULL) + return; restart: - read_lock_bh(&rt6_lock); - for (rt = ip6_routing_table.leaf; rt; rt = rt->u.next) { + read_lock_bh(&table->tb6_lock); + for (rt = table->tb6_root.leaf; rt; rt = rt->u.next) { if (rt->rt6i_flags & (RTF_DEFAULT | RTF_ADDRCONF)) { dst_hold(&rt->u.dst); - - read_unlock_bh(&rt6_lock); - + read_unlock_bh(&table->tb6_lock); ip6_del_rt(rt, NULL, NULL, NULL); - goto restart; } } - read_unlock_bh(&rt6_lock); + read_unlock_bh(&table->tb6_lock); } int ipv6_route_ioctl(unsigned int cmd, void __user *arg) @@ -1516,10 +1609,12 @@ int ipv6_route_ioctl(unsigned int cmd, void __user *arg) rtnl_lock(); switch (cmd) { case SIOCADDRT: - err = ip6_route_add(&rtmsg, NULL, NULL, NULL); + err = ip6_route_add(&rtmsg, NULL, NULL, NULL, + RT6_TABLE_MAIN); break; case SIOCDELRT: - err = ip6_route_del(&rtmsg, NULL, NULL, NULL); + err = ip6_route_del(&rtmsg, NULL, NULL, NULL, + RT6_TABLE_MAIN); break; default: err = -EINVAL; @@ -1593,6 +1688,7 @@ struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev, ipv6_addr_copy(&rt->rt6i_dst.addr, addr); rt->rt6i_dst.plen = 128; + rt->rt6i_table = fib6_get_table(RT6_TABLE_LOCAL); atomic_set(&rt->u.dst.__refcnt, 1); @@ -1611,9 +1707,7 @@ static int fib6_ifdown(struct rt6_info *rt, void *arg) void rt6_ifdown(struct net_device *dev) { - write_lock_bh(&rt6_lock); - fib6_clean_tree(&ip6_routing_table, fib6_ifdown, 0, dev); - write_unlock_bh(&rt6_lock); + fib6_clean_all(fib6_ifdown, 0, dev); } struct rt6_mtu_change_arg @@ -1663,13 +1757,12 @@ static int rt6_mtu_change_route(struct rt6_info *rt, void *p_arg) void rt6_mtu_change(struct net_device *dev, unsigned mtu) { - struct rt6_mtu_change_arg arg; + struct rt6_mtu_change_arg arg = { + .dev = dev, + .mtu = mtu, + }; - arg.dev = dev; - arg.mtu = mtu; - read_lock_bh(&rt6_lock); - fib6_clean_tree(&ip6_routing_table, rt6_mtu_change_route, 0, &arg); - read_unlock_bh(&rt6_lock); + fib6_clean_all(rt6_mtu_change_route, 0, &arg); } static int inet6_rtm_to_rtmsg(struct rtmsg *r, struct rtattr **rta, @@ -1719,7 +1812,7 @@ int inet6_rtm_delroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) if (inet6_rtm_to_rtmsg(r, arg, &rtmsg)) return -EINVAL; - return ip6_route_del(&rtmsg, nlh, arg, &NETLINK_CB(skb)); + return ip6_route_del(&rtmsg, nlh, arg, &NETLINK_CB(skb), r->rtm_table); } int inet6_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) @@ -1729,7 +1822,7 @@ int inet6_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) if (inet6_rtm_to_rtmsg(r, arg, &rtmsg)) return -EINVAL; - return ip6_route_add(&rtmsg, nlh, arg, &NETLINK_CB(skb)); + return ip6_route_add(&rtmsg, nlh, arg, &NETLINK_CB(skb), r->rtm_table); } struct rt6_rtnl_dump_arg @@ -1761,6 +1854,10 @@ static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt, rtm->rtm_dst_len = rt->rt6i_dst.plen; rtm->rtm_src_len = rt->rt6i_src.plen; rtm->rtm_tos = 0; + if (rt->rt6i_table) + rtm->rtm_table = rt->rt6i_table->tb6_id; + else + rtm->rtm_table = RT6_TABLE_UNSPEC; rtm->rtm_table = RT_TABLE_MAIN; if (rt->rt6i_flags&RTF_REJECT) rtm->rtm_type = RTN_UNREACHABLE; @@ -1868,7 +1965,6 @@ static void fib6_dump_end(struct netlink_callback *cb) if (w) { cb->args[0] = 0; - fib6_walker_unlink(w); kfree(w); } cb->done = (void*)cb->args[1]; @@ -1883,13 +1979,20 @@ static int fib6_dump_done(struct netlink_callback *cb) int inet6_dump_fib(struct sk_buff *skb, struct netlink_callback *cb) { + struct fib6_table *table; struct rt6_rtnl_dump_arg arg; struct fib6_walker_t *w; - int res; + int i, res = 0; arg.skb = skb; arg.cb = cb; + /* + * cb->args[0] = pointer to walker structure + * cb->args[1] = saved cb->done() pointer + * cb->args[2] = current table being dumped + */ + w = (void*)cb->args[0]; if (w == NULL) { /* New dump: @@ -1905,24 +2008,48 @@ int inet6_dump_fib(struct sk_buff *skb, struct netlink_callback *cb) w = kzalloc(sizeof(*w), GFP_ATOMIC); if (w == NULL) return -ENOMEM; - RT6_TRACE("dump<%p", w); - w->root = &ip6_routing_table; w->func = fib6_dump_node; w->args = &arg; cb->args[0] = (long)w; - read_lock_bh(&rt6_lock); - res = fib6_walk(w); - read_unlock_bh(&rt6_lock); + cb->args[2] = FIB6_TABLE_MIN; } else { w->args = &arg; - read_lock_bh(&rt6_lock); - res = fib6_walk_continue(w); - read_unlock_bh(&rt6_lock); + i = cb->args[2]; + if (i > FIB6_TABLE_MAX) + goto end; + + table = fib6_get_table(i); + if (table != NULL) { + read_lock_bh(&table->tb6_lock); + w->root = &table->tb6_root; + res = fib6_walk_continue(w); + read_unlock_bh(&table->tb6_lock); + if (res != 0) { + if (res < 0) + fib6_walker_unlink(w); + goto end; + } + } + + fib6_walker_unlink(w); + cb->args[2] = ++i; } -#if RT6_DEBUG >= 3 - if (res <= 0 && skb->len == 0) - RT6_TRACE("%p>dump end\n", w); -#endif + + for (i = cb->args[2]; i <= FIB6_TABLE_MAX; i++) { + table = fib6_get_table(i); + if (table == NULL) + continue; + + read_lock_bh(&table->tb6_lock); + w->root = &table->tb6_root; + res = fib6_walk(w); + read_unlock_bh(&table->tb6_lock); + if (res) + break; + } +end: + cb->args[2] = i; + res = res < 0 ? res : skb->len; /* res < 0 is an error. (really, impossible) res == 0 means that dump is complete, but skb still can contain data. @@ -2102,16 +2229,13 @@ static int rt6_info_route(struct rt6_info *rt, void *p_arg) static int rt6_proc_info(char *buffer, char **start, off_t offset, int length) { - struct rt6_proc_arg arg; - arg.buffer = buffer; - arg.offset = offset; - arg.length = length; - arg.skip = 0; - arg.len = 0; + struct rt6_proc_arg arg = { + .buffer = buffer, + .offset = offset, + .length = length, + }; - read_lock_bh(&rt6_lock); - fib6_clean_tree(&ip6_routing_table, rt6_info_route, 0, &arg); - read_unlock_bh(&rt6_lock); + fib6_clean_all(rt6_info_route, 0, &arg); *start = buffer; if (offset) -- cgit v1.1 From 101367c2f8c464ea96643192673aa18d88e6336d Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Fri, 4 Aug 2006 03:39:02 -0700 Subject: [IPV6]: Policy Routing Rules Adds support for policy routing rules including a new local table for routes with a local destination. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- net/ipv6/route.c | 50 ++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 50 insertions(+) (limited to 'net/ipv6/route.c') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 73efdad..438977e 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -140,6 +140,50 @@ struct rt6_info ip6_null_entry = { .rt6i_ref = ATOMIC_INIT(1), }; +#ifdef CONFIG_IPV6_MULTIPLE_TABLES + +struct rt6_info ip6_prohibit_entry = { + .u = { + .dst = { + .__refcnt = ATOMIC_INIT(1), + .__use = 1, + .dev = &loopback_dev, + .obsolete = -1, + .error = -EACCES, + .metrics = { [RTAX_HOPLIMIT - 1] = 255, }, + .input = ip6_pkt_discard, + .output = ip6_pkt_discard_out, + .ops = &ip6_dst_ops, + .path = (struct dst_entry*)&ip6_prohibit_entry, + } + }, + .rt6i_flags = (RTF_REJECT | RTF_NONEXTHOP), + .rt6i_metric = ~(u32) 0, + .rt6i_ref = ATOMIC_INIT(1), +}; + +struct rt6_info ip6_blk_hole_entry = { + .u = { + .dst = { + .__refcnt = ATOMIC_INIT(1), + .__use = 1, + .dev = &loopback_dev, + .obsolete = -1, + .error = -EINVAL, + .metrics = { [RTAX_HOPLIMIT - 1] = 255, }, + .input = ip6_pkt_discard, + .output = ip6_pkt_discard_out, + .ops = &ip6_dst_ops, + .path = (struct dst_entry*)&ip6_blk_hole_entry, + } + }, + .rt6i_flags = (RTF_REJECT | RTF_NONEXTHOP), + .rt6i_metric = ~(u32) 0, + .rt6i_ref = ATOMIC_INIT(1), +}; + +#endif + /* allocate dst with ip6_dst_ops */ static __inline__ struct rt6_info *ip6_dst_alloc(void) { @@ -2408,10 +2452,16 @@ void __init ip6_route_init(void) #ifdef CONFIG_XFRM xfrm6_init(); #endif +#ifdef CONFIG_IPV6_MULTIPLE_TABLES + fib6_rules_init(); +#endif } void ip6_route_cleanup(void) { +#ifdef CONFIG_IPV6_MULTIPLE_TABLES + fib6_rules_cleanup(); +#endif #ifdef CONFIG_PROC_FS proc_net_remove("ipv6_route"); proc_net_remove("rt6_stats"); -- cgit v1.1 From d7aba67f814729647c938ac6da2d5224b790f926 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Sat, 5 Aug 2006 02:20:42 -0700 Subject: [IPV6]: Fix thinko in rt6_fill_node This looks like a mistake, the table ID is overwritten again. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/ipv6/route.c | 1 - 1 file changed, 1 deletion(-) (limited to 'net/ipv6/route.c') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 438977e..ff5affe 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1902,7 +1902,6 @@ static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt, rtm->rtm_table = rt->rt6i_table->tb6_id; else rtm->rtm_table = RT6_TABLE_UNSPEC; - rtm->rtm_table = RT_TABLE_MAIN; if (rt->rt6i_flags&RTF_REJECT) rtm->rtm_type = RTN_UNREACHABLE; else if (rt->rt6i_dev && (rt->rt6i_dev->flags&IFF_LOOPBACK)) -- cgit v1.1 From 6c813a7297e3af4cd7c3458e09e9ee3d161c6830 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Sun, 6 Aug 2006 22:22:47 -0700 Subject: [IPV6]: Fix crash in ip6_del_rt ip6_null_entry doesn't have rt6i_table set, when trying to delete it the kernel crashes dereferencing table->tb6_lock. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/ipv6/route.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net/ipv6/route.c') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index ff5affe..41c5905 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1223,6 +1223,9 @@ int ip6_del_rt(struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr, struct int err; struct fib6_table *table; + if (rt == &ip6_null_entry) + return -ENOENT; + table = rt->rt6i_table; write_lock_bh(&table->tb6_lock); -- cgit v1.1 From 8ce11e6a9faf1f1c849b77104adc1642c46aee95 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Mon, 7 Aug 2006 21:50:48 -0700 Subject: [NET]: Make code static. This patch makes needlessly global code static. Signed-off-by: Adrian Bunk Signed-off-by: David S. Miller --- net/ipv6/route.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'net/ipv6/route.c') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 41c5905..e08d840 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -613,8 +613,8 @@ static struct rt6_info *rt6_alloc_clone(struct rt6_info *ort, struct in6_addr *d return rt; } -struct rt6_info *ip6_pol_route_input(struct fib6_table *table, struct flowi *fl, - int flags) +static struct rt6_info *ip6_pol_route_input(struct fib6_table *table, + struct flowi *fl, int flags) { struct fib6_node *fn; struct rt6_info *rt, *nrt; @@ -872,7 +872,7 @@ static inline unsigned int ipv6_advmss(unsigned int mtu) } static struct dst_entry *ndisc_dst_gc_list; -DEFINE_SPINLOCK(ndisc_lock); +static DEFINE_SPINLOCK(ndisc_lock); struct dst_entry *ndisc_dst_alloc(struct net_device *dev, struct neighbour *neigh, -- cgit v1.1 From 9e762a4a89b302cb3b26a1f9bb33eff459eaeca9 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 10 Aug 2006 23:09:48 -0700 Subject: [NET]: Introduce RTA_TABLE/FRA_TABLE attributes Introduce RTA_TABLE route attribute and FRA_TABLE routing rule attribute to hold 32 bit routing table IDs. Usespace compatibility is provided by continuing to accept and send the rtm_table field, but because of its limited size it can only carry the low 8 bits of the table ID. This implies that if larger IDs are used, _all_ userspace programs using them need to use RTA_TABLE. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/ipv6/route.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) (limited to 'net/ipv6/route.c') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index e08d840..843c550 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1859,7 +1859,8 @@ int inet6_rtm_delroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) if (inet6_rtm_to_rtmsg(r, arg, &rtmsg)) return -EINVAL; - return ip6_route_del(&rtmsg, nlh, arg, &NETLINK_CB(skb), r->rtm_table); + return ip6_route_del(&rtmsg, nlh, arg, &NETLINK_CB(skb), + rtm_get_table(arg, r->rtm_table)); } int inet6_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) @@ -1869,7 +1870,8 @@ int inet6_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) if (inet6_rtm_to_rtmsg(r, arg, &rtmsg)) return -EINVAL; - return ip6_route_add(&rtmsg, nlh, arg, &NETLINK_CB(skb), r->rtm_table); + return ip6_route_add(&rtmsg, nlh, arg, &NETLINK_CB(skb), + rtm_get_table(arg, r->rtm_table)); } struct rt6_rtnl_dump_arg @@ -1887,6 +1889,7 @@ static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt, struct nlmsghdr *nlh; unsigned char *b = skb->tail; struct rta_cacheinfo ci; + u32 table; if (prefix) { /* user wants prefix routes only */ if (!(rt->rt6i_flags & RTF_PREFIX_RT)) { @@ -1902,9 +1905,11 @@ static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt, rtm->rtm_src_len = rt->rt6i_src.plen; rtm->rtm_tos = 0; if (rt->rt6i_table) - rtm->rtm_table = rt->rt6i_table->tb6_id; + table = rt->rt6i_table->tb6_id; else - rtm->rtm_table = RT6_TABLE_UNSPEC; + table = RT6_TABLE_UNSPEC; + rtm->rtm_table = table; + RTA_PUT_U32(skb, RTA_TABLE, table); if (rt->rt6i_flags&RTF_REJECT) rtm->rtm_type = RTN_UNREACHABLE; else if (rt->rt6i_dev && (rt->rt6i_dev->flags&IFF_LOOPBACK)) -- cgit v1.1 From 1b43af5480c351dbcb2eef478bafe179cbeb6e83 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 10 Aug 2006 23:11:17 -0700 Subject: [IPV6]: Increase number of possible routing tables to 2^32 Increase number of possible routing tables to 2^32 by replacing iterations over all possible table IDs by hash table walking. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/ipv6/route.c | 128 +------------------------------------------------------ 1 file changed, 1 insertion(+), 127 deletions(-) (limited to 'net/ipv6/route.c') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 843c550..9ce28277 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1874,12 +1874,6 @@ int inet6_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) rtm_get_table(arg, r->rtm_table)); } -struct rt6_rtnl_dump_arg -{ - struct sk_buff *skb; - struct netlink_callback *cb; -}; - 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, @@ -1976,7 +1970,7 @@ rtattr_failure: return -1; } -static int rt6_dump_route(struct rt6_info *rt, void *p_arg) +int rt6_dump_route(struct rt6_info *rt, void *p_arg) { struct rt6_rtnl_dump_arg *arg = (struct rt6_rtnl_dump_arg *) p_arg; int prefix; @@ -1992,126 +1986,6 @@ static int rt6_dump_route(struct rt6_info *rt, void *p_arg) prefix, NLM_F_MULTI); } -static int fib6_dump_node(struct fib6_walker_t *w) -{ - int res; - struct rt6_info *rt; - - for (rt = w->leaf; rt; rt = rt->u.next) { - res = rt6_dump_route(rt, w->args); - if (res < 0) { - /* Frame is full, suspend walking */ - w->leaf = rt; - return 1; - } - BUG_TRAP(res!=0); - } - w->leaf = NULL; - return 0; -} - -static void fib6_dump_end(struct netlink_callback *cb) -{ - struct fib6_walker_t *w = (void*)cb->args[0]; - - if (w) { - cb->args[0] = 0; - kfree(w); - } - cb->done = (void*)cb->args[1]; - cb->args[1] = 0; -} - -static int fib6_dump_done(struct netlink_callback *cb) -{ - fib6_dump_end(cb); - return cb->done ? cb->done(cb) : 0; -} - -int inet6_dump_fib(struct sk_buff *skb, struct netlink_callback *cb) -{ - struct fib6_table *table; - struct rt6_rtnl_dump_arg arg; - struct fib6_walker_t *w; - int i, res = 0; - - arg.skb = skb; - arg.cb = cb; - - /* - * cb->args[0] = pointer to walker structure - * cb->args[1] = saved cb->done() pointer - * cb->args[2] = current table being dumped - */ - - w = (void*)cb->args[0]; - if (w == NULL) { - /* New dump: - * - * 1. hook callback destructor. - */ - cb->args[1] = (long)cb->done; - cb->done = fib6_dump_done; - - /* - * 2. allocate and initialize walker. - */ - w = kzalloc(sizeof(*w), GFP_ATOMIC); - if (w == NULL) - return -ENOMEM; - w->func = fib6_dump_node; - w->args = &arg; - cb->args[0] = (long)w; - cb->args[2] = FIB6_TABLE_MIN; - } else { - w->args = &arg; - i = cb->args[2]; - if (i > FIB6_TABLE_MAX) - goto end; - - table = fib6_get_table(i); - if (table != NULL) { - read_lock_bh(&table->tb6_lock); - w->root = &table->tb6_root; - res = fib6_walk_continue(w); - read_unlock_bh(&table->tb6_lock); - if (res != 0) { - if (res < 0) - fib6_walker_unlink(w); - goto end; - } - } - - fib6_walker_unlink(w); - cb->args[2] = ++i; - } - - for (i = cb->args[2]; i <= FIB6_TABLE_MAX; i++) { - table = fib6_get_table(i); - if (table == NULL) - continue; - - read_lock_bh(&table->tb6_lock); - w->root = &table->tb6_root; - res = fib6_walk(w); - read_unlock_bh(&table->tb6_lock); - if (res) - break; - } -end: - cb->args[2] = i; - - res = res < 0 ? res : skb->len; - /* res < 0 is an error. (really, impossible) - res == 0 means that dump is complete, but skb still can contain data. - res > 0 dump is not complete, but frame is full. - */ - /* Destroy walker, if dump of this table is complete. */ - if (res <= 0) - fib6_dump_end(cb); - return res; -} - int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void *arg) { struct rtattr **rta = arg; -- cgit v1.1 From 2942e90050569525628a9f34e0daaa9b661b49cc Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Tue, 15 Aug 2006 00:30:25 -0700 Subject: [RTNETLINK]: Use rtnl_unicast() for rtnetlink unicasts Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- net/ipv6/route.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'net/ipv6/route.c') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 9ce28277..024c8e2 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -2044,9 +2044,7 @@ int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void *arg) goto out_free; } - err = netlink_unicast(rtnl, skb, NETLINK_CB(in_skb).pid, MSG_DONTWAIT); - if (err > 0) - err = 0; + err = rtnl_unicast(skb, NETLINK_CB(in_skb).pid); out: return err; out_free: -- cgit v1.1 From 21713ebc4f119950e87d21c4637d5a750eea20e8 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Tue, 15 Aug 2006 00:35:24 -0700 Subject: [IPv6] route: Convert route notifications to use rtnl_notify() Fixes a wrong use of current->pid as netlink pid. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- net/ipv6/route.c | 38 ++++++++++++++++++-------------------- 1 file changed, 18 insertions(+), 20 deletions(-) (limited to 'net/ipv6/route.c') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 024c8e2..1aca787 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -35,7 +35,6 @@ #include #include #include -#include #include #ifdef CONFIG_PROC_FS @@ -54,6 +53,7 @@ #include #include #include +#include #include @@ -2056,27 +2056,25 @@ void inet6_rt_notify(int event, struct rt6_info *rt, struct nlmsghdr *nlh, struct netlink_skb_parms *req) { struct sk_buff *skb; - int size = NLMSG_SPACE(sizeof(struct rtmsg)+256); - u32 pid = current->pid; - u32 seq = 0; - - if (req) - pid = req->pid; - if (nlh) - seq = nlh->nlmsg_seq; - - skb = alloc_skb(size, gfp_any()); - if (!skb) { - netlink_set_err(rtnl, 0, RTNLGRP_IPV6_ROUTE, ENOBUFS); - return; - } - if (rt6_fill_node(skb, rt, NULL, NULL, 0, event, pid, seq, 0, 0) < 0) { + u32 pid = req ? req->pid : 0; + u32 seq = nlh ? nlh->nlmsg_seq : 0; + int payload = sizeof(struct rtmsg) + 256; + int err = -ENOBUFS; + + skb = nlmsg_new(nlmsg_total_size(payload), gfp_any()); + if (skb == NULL) + goto errout; + + err = rt6_fill_node(skb, rt, NULL, NULL, 0, event, pid, seq, 0, 0); + if (err < 0) { kfree_skb(skb); - netlink_set_err(rtnl, 0, RTNLGRP_IPV6_ROUTE, EINVAL); - return; + goto errout; } - NETLINK_CB(skb).dst_group = RTNLGRP_IPV6_ROUTE; - netlink_broadcast(rtnl, skb, 0, RTNLGRP_IPV6_ROUTE, gfp_any()); + + err = rtnl_notify(skb, pid, RTNLGRP_IPV6_ROUTE, nlh, gfp_any()); +errout: + if (err < 0) + rtnl_set_sk_err(RTNLGRP_IPV6_ROUTE, err); } /* -- cgit v1.1 From e0a1ad73d34fd6dfdb630479400511e9879069c0 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Tue, 22 Aug 2006 00:00:21 -0700 Subject: [IPv6] route: Simplify ip6_del_rt() Provide a simple ip6_del_rt() for the majority of users and an alternative for the exception via netlink. Avoids code obfuscation. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- net/ipv6/route.c | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) (limited to 'net/ipv6/route.c') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 1aca787..8d511de 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -457,7 +457,7 @@ int rt6_route_rcv(struct net_device *dev, u8 *opt, int len, rt = rt6_get_route_info(prefix, rinfo->prefix_len, gwaddr, dev->ifindex); if (rt && !lifetime) { - ip6_del_rt(rt, NULL, NULL, NULL); + ip6_del_rt(rt); rt = NULL; } @@ -813,7 +813,7 @@ static struct dst_entry *ip6_negative_advice(struct dst_entry *dst) if (rt) { if (rt->rt6i_flags & RTF_CACHE) - ip6_del_rt(rt, NULL, NULL, NULL); + ip6_del_rt(rt); else dst_release(dst); } @@ -1218,7 +1218,8 @@ out: return err; } -int ip6_del_rt(struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr, struct netlink_skb_parms *req) +static int __ip6_del_rt(struct rt6_info *rt, struct nlmsghdr *nlh, + void *_rtattr, struct netlink_skb_parms *req) { int err; struct fib6_table *table; @@ -1237,6 +1238,11 @@ int ip6_del_rt(struct rt6_info *rt, struct nlmsghdr *nlh, void *_rtattr, struct return err; } +int ip6_del_rt(struct rt6_info *rt) +{ + return __ip6_del_rt(rt, NULL, NULL, NULL); +} + static int ip6_route_del(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, void *_rtattr, struct netlink_skb_parms *req, u32 table_id) @@ -1271,7 +1277,7 @@ static int ip6_route_del(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, dst_hold(&rt->u.dst); read_unlock_bh(&table->tb6_lock); - return ip6_del_rt(rt, nlh, _rtattr, req); + return __ip6_del_rt(rt, nlh, _rtattr, req); } } read_unlock_bh(&table->tb6_lock); @@ -1395,7 +1401,7 @@ restart: call_netevent_notifiers(NETEVENT_REDIRECT, &netevent); if (rt->rt6i_flags&RTF_CACHE) { - ip6_del_rt(rt, NULL, NULL, NULL); + ip6_del_rt(rt); return; } @@ -1631,7 +1637,7 @@ restart: if (rt->rt6i_flags & (RTF_DEFAULT | RTF_ADDRCONF)) { dst_hold(&rt->u.dst); read_unlock_bh(&table->tb6_lock); - ip6_del_rt(rt, NULL, NULL, NULL); + ip6_del_rt(rt); goto restart; } } -- cgit v1.1 From 40e22e8f3d4d4f1ff68fb03683f007c53ee8b348 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Tue, 22 Aug 2006 00:00:45 -0700 Subject: [IPv6] route: Simplify ip6_ins_rt() Provide a simple ip6_ins_rt() for the majority of users and an alternative for the exception via netlink. Avoids code obfuscation. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- net/ipv6/route.c | 19 ++++++++++++------- 1 file changed, 12 insertions(+), 7 deletions(-) (limited to 'net/ipv6/route.c') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 8d511de..9ec348a 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -546,8 +546,8 @@ struct rt6_info *rt6_lookup(struct in6_addr *daddr, struct in6_addr *saddr, be destroyed. */ -int ip6_ins_rt(struct rt6_info *rt, struct nlmsghdr *nlh, - void *_rtattr, struct netlink_skb_parms *req) +static int __ip6_ins_rt(struct rt6_info *rt, struct nlmsghdr *nlh, + void *_rtattr, struct netlink_skb_parms *req) { int err; struct fib6_table *table; @@ -560,6 +560,11 @@ int ip6_ins_rt(struct rt6_info *rt, struct nlmsghdr *nlh, return err; } +int ip6_ins_rt(struct rt6_info *rt) +{ + return __ip6_ins_rt(rt, NULL, NULL, NULL); +} + static struct rt6_info *rt6_alloc_cow(struct rt6_info *ort, struct in6_addr *daddr, struct in6_addr *saddr) { @@ -657,7 +662,7 @@ restart: dst_hold(&rt->u.dst); if (nrt) { - err = ip6_ins_rt(nrt, NULL, NULL, NULL); + err = ip6_ins_rt(nrt); if (!err) goto out2; } @@ -752,7 +757,7 @@ restart: dst_hold(&rt->u.dst); if (nrt) { - err = ip6_ins_rt(nrt, NULL, NULL, NULL); + err = ip6_ins_rt(nrt); if (!err) goto out2; } @@ -1206,7 +1211,7 @@ install_route: rt->u.dst.dev = dev; rt->rt6i_idev = idev; rt->rt6i_table = table; - return ip6_ins_rt(rt, nlh, _rtattr, req); + return __ip6_ins_rt(rt, nlh, _rtattr, req); out: if (dev) @@ -1393,7 +1398,7 @@ restart: nrt->u.dst.metrics[RTAX_MTU-1] = ipv6_get_mtu(neigh->dev); nrt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(dst_mtu(&nrt->u.dst)); - if (ip6_ins_rt(nrt, NULL, NULL, NULL)) + if (ip6_ins_rt(nrt)) goto out; netevent.old = &rt->u.dst; @@ -1483,7 +1488,7 @@ void rt6_pmtu_discovery(struct in6_addr *daddr, struct in6_addr *saddr, dst_set_expires(&nrt->u.dst, ip6_rt_mtu_expires); nrt->rt6i_flags |= RTF_DYNAMIC|RTF_EXPIRES; - ip6_ins_rt(nrt, NULL, NULL, NULL); + ip6_ins_rt(nrt); } out: dst_release(&rt->u.dst); -- cgit v1.1 From 86872cb57925c46a6499887d77afb880a892c0ec Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Tue, 22 Aug 2006 00:01:08 -0700 Subject: [IPv6] route: FIB6 configuration using struct fib6_config Replaces the struct in6_rtmsg based interface orignating from the ioctl interface with a struct fib6_config based on. Allows changing the interface without breaking the ioctl interface and avoids passing on tons of parameters. The recently introduced struct nl_info is used to pass on netlink authorship information for notifications. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- net/ipv6/route.c | 331 ++++++++++++++++++++++++++++++++----------------------- 1 file changed, 191 insertions(+), 140 deletions(-) (limited to 'net/ipv6/route.c') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 9ec348a..7bcffa6 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -546,15 +546,14 @@ struct rt6_info *rt6_lookup(struct in6_addr *daddr, struct in6_addr *saddr, be destroyed. */ -static int __ip6_ins_rt(struct rt6_info *rt, struct nlmsghdr *nlh, - void *_rtattr, struct netlink_skb_parms *req) +static int __ip6_ins_rt(struct rt6_info *rt, struct nl_info *info) { int err; struct fib6_table *table; table = rt->rt6i_table; write_lock_bh(&table->tb6_lock); - err = fib6_add(&table->tb6_root, rt, nlh, _rtattr, req); + err = fib6_add(&table->tb6_root, rt, info); write_unlock_bh(&table->tb6_lock); return err; @@ -562,7 +561,7 @@ static int __ip6_ins_rt(struct rt6_info *rt, struct nlmsghdr *nlh, int ip6_ins_rt(struct rt6_info *rt) { - return __ip6_ins_rt(rt, NULL, NULL, NULL); + return __ip6_ins_rt(rt, NULL); } static struct rt6_info *rt6_alloc_cow(struct rt6_info *ort, struct in6_addr *daddr, @@ -1014,30 +1013,24 @@ int ipv6_get_hoplimit(struct net_device *dev) * */ -int ip6_route_add(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, - void *_rtattr, struct netlink_skb_parms *req, - u32 table_id) +int ip6_route_add(struct fib6_config *cfg) { int err; - struct rtmsg *r; - struct rtattr **rta; struct rt6_info *rt = NULL; struct net_device *dev = NULL; struct inet6_dev *idev = NULL; struct fib6_table *table; int addr_type; - rta = (struct rtattr **) _rtattr; - - if (rtmsg->rtmsg_dst_len > 128 || rtmsg->rtmsg_src_len > 128) + if (cfg->fc_dst_len > 128 || cfg->fc_src_len > 128) return -EINVAL; #ifndef CONFIG_IPV6_SUBTREES - if (rtmsg->rtmsg_src_len) + if (cfg->fc_src_len) return -EINVAL; #endif - if (rtmsg->rtmsg_ifindex) { + if (cfg->fc_ifindex) { err = -ENODEV; - dev = dev_get_by_index(rtmsg->rtmsg_ifindex); + dev = dev_get_by_index(cfg->fc_ifindex); if (!dev) goto out; idev = in6_dev_get(dev); @@ -1045,10 +1038,10 @@ int ip6_route_add(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, goto out; } - if (rtmsg->rtmsg_metric == 0) - rtmsg->rtmsg_metric = IP6_RT_PRIO_USER; + if (cfg->fc_metric == 0) + cfg->fc_metric = IP6_RT_PRIO_USER; - table = fib6_new_table(table_id); + table = fib6_new_table(cfg->fc_table); if (table == NULL) { err = -ENOBUFS; goto out; @@ -1062,14 +1055,13 @@ int ip6_route_add(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, } rt->u.dst.obsolete = -1; - rt->rt6i_expires = jiffies + clock_t_to_jiffies(rtmsg->rtmsg_info); - if (nlh && (r = NLMSG_DATA(nlh))) { - rt->rt6i_protocol = r->rtm_protocol; - } else { - rt->rt6i_protocol = RTPROT_BOOT; - } + rt->rt6i_expires = jiffies + clock_t_to_jiffies(cfg->fc_expires); + + if (cfg->fc_protocol == RTPROT_UNSPEC) + cfg->fc_protocol = RTPROT_BOOT; + rt->rt6i_protocol = cfg->fc_protocol; - addr_type = ipv6_addr_type(&rtmsg->rtmsg_dst); + addr_type = ipv6_addr_type(&cfg->fc_dst); if (addr_type & IPV6_ADDR_MULTICAST) rt->u.dst.input = ip6_mc_input; @@ -1078,24 +1070,22 @@ int ip6_route_add(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, rt->u.dst.output = ip6_output; - ipv6_addr_prefix(&rt->rt6i_dst.addr, - &rtmsg->rtmsg_dst, rtmsg->rtmsg_dst_len); - rt->rt6i_dst.plen = rtmsg->rtmsg_dst_len; + ipv6_addr_prefix(&rt->rt6i_dst.addr, &cfg->fc_dst, cfg->fc_dst_len); + rt->rt6i_dst.plen = cfg->fc_dst_len; if (rt->rt6i_dst.plen == 128) rt->u.dst.flags = DST_HOST; #ifdef CONFIG_IPV6_SUBTREES - ipv6_addr_prefix(&rt->rt6i_src.addr, - &rtmsg->rtmsg_src, rtmsg->rtmsg_src_len); - rt->rt6i_src.plen = rtmsg->rtmsg_src_len; + ipv6_addr_prefix(&rt->rt6i_src.addr, &cfg->fc_src, cfg->fc_src_len); + rt->rt6i_src.plen = cfg->fc_src_len; #endif - rt->rt6i_metric = rtmsg->rtmsg_metric; + rt->rt6i_metric = cfg->fc_metric; /* We cannot add true routes via loopback here, they would result in kernel looping; promote them to reject routes */ - if ((rtmsg->rtmsg_flags&RTF_REJECT) || + if ((cfg->fc_flags & RTF_REJECT) || (dev && (dev->flags&IFF_LOOPBACK) && !(addr_type&IPV6_ADDR_LOOPBACK))) { /* hold loopback dev/idev if we haven't done so. */ if (dev != &loopback_dev) { @@ -1118,12 +1108,12 @@ int ip6_route_add(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, goto install_route; } - if (rtmsg->rtmsg_flags & RTF_GATEWAY) { + if (cfg->fc_flags & RTF_GATEWAY) { struct in6_addr *gw_addr; int gwa_type; - gw_addr = &rtmsg->rtmsg_gateway; - ipv6_addr_copy(&rt->rt6i_gateway, &rtmsg->rtmsg_gateway); + gw_addr = &cfg->fc_gateway; + ipv6_addr_copy(&rt->rt6i_gateway, gw_addr); gwa_type = ipv6_addr_type(gw_addr); if (gwa_type != (IPV6_ADDR_LINKLOCAL|IPV6_ADDR_UNICAST)) { @@ -1140,7 +1130,7 @@ int ip6_route_add(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, if (!(gwa_type&IPV6_ADDR_UNICAST)) goto out; - grt = rt6_lookup(gw_addr, NULL, rtmsg->rtmsg_ifindex, 1); + grt = rt6_lookup(gw_addr, NULL, cfg->fc_ifindex, 1); err = -EHOSTUNREACH; if (grt == NULL) @@ -1172,7 +1162,7 @@ int ip6_route_add(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, if (dev == NULL) goto out; - if (rtmsg->rtmsg_flags & (RTF_GATEWAY|RTF_NONEXTHOP)) { + if (cfg->fc_flags & (RTF_GATEWAY | RTF_NONEXTHOP)) { rt->rt6i_nexthop = __neigh_lookup_errno(&nd_tbl, &rt->rt6i_gateway, dev); if (IS_ERR(rt->rt6i_nexthop)) { err = PTR_ERR(rt->rt6i_nexthop); @@ -1181,24 +1171,24 @@ int ip6_route_add(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, } } - rt->rt6i_flags = rtmsg->rtmsg_flags; + rt->rt6i_flags = cfg->fc_flags; install_route: - if (rta && rta[RTA_METRICS-1]) { - int attrlen = RTA_PAYLOAD(rta[RTA_METRICS-1]); - struct rtattr *attr = RTA_DATA(rta[RTA_METRICS-1]); - - while (RTA_OK(attr, attrlen)) { - unsigned flavor = attr->rta_type; - if (flavor) { - if (flavor > RTAX_MAX) { + if (cfg->fc_mx) { + struct nlattr *nla; + int remaining; + + nla_for_each_attr(nla, cfg->fc_mx, cfg->fc_mx_len, remaining) { + int type = nla->nla_type; + + if (type) { + if (type > RTAX_MAX) { err = -EINVAL; goto out; } - rt->u.dst.metrics[flavor-1] = - *(u32 *)RTA_DATA(attr); + + rt->u.dst.metrics[type - 1] = nla_get_u32(nla); } - attr = RTA_NEXT(attr, attrlen); } } @@ -1211,7 +1201,7 @@ install_route: rt->u.dst.dev = dev; rt->rt6i_idev = idev; rt->rt6i_table = table; - return __ip6_ins_rt(rt, nlh, _rtattr, req); + return __ip6_ins_rt(rt, &cfg->fc_nlinfo); out: if (dev) @@ -1223,8 +1213,7 @@ out: return err; } -static int __ip6_del_rt(struct rt6_info *rt, struct nlmsghdr *nlh, - void *_rtattr, struct netlink_skb_parms *req) +static int __ip6_del_rt(struct rt6_info *rt, struct nl_info *info) { int err; struct fib6_table *table; @@ -1235,7 +1224,7 @@ static int __ip6_del_rt(struct rt6_info *rt, struct nlmsghdr *nlh, table = rt->rt6i_table; write_lock_bh(&table->tb6_lock); - err = fib6_del(rt, nlh, _rtattr, req); + err = fib6_del(rt, info); dst_release(&rt->u.dst); write_unlock_bh(&table->tb6_lock); @@ -1245,44 +1234,41 @@ static int __ip6_del_rt(struct rt6_info *rt, struct nlmsghdr *nlh, int ip6_del_rt(struct rt6_info *rt) { - return __ip6_del_rt(rt, NULL, NULL, NULL); + return __ip6_del_rt(rt, NULL); } -static int ip6_route_del(struct in6_rtmsg *rtmsg, struct nlmsghdr *nlh, - void *_rtattr, struct netlink_skb_parms *req, - u32 table_id) +static int ip6_route_del(struct fib6_config *cfg) { struct fib6_table *table; struct fib6_node *fn; struct rt6_info *rt; int err = -ESRCH; - table = fib6_get_table(table_id); + table = fib6_get_table(cfg->fc_table); if (table == NULL) return err; read_lock_bh(&table->tb6_lock); fn = fib6_locate(&table->tb6_root, - &rtmsg->rtmsg_dst, rtmsg->rtmsg_dst_len, - &rtmsg->rtmsg_src, rtmsg->rtmsg_src_len); + &cfg->fc_dst, cfg->fc_dst_len, + &cfg->fc_src, cfg->fc_src_len); if (fn) { for (rt = fn->leaf; rt; rt = rt->u.next) { - if (rtmsg->rtmsg_ifindex && + if (cfg->fc_ifindex && (rt->rt6i_dev == NULL || - rt->rt6i_dev->ifindex != rtmsg->rtmsg_ifindex)) + rt->rt6i_dev->ifindex != cfg->fc_ifindex)) continue; - if (rtmsg->rtmsg_flags&RTF_GATEWAY && - !ipv6_addr_equal(&rtmsg->rtmsg_gateway, &rt->rt6i_gateway)) + if (cfg->fc_flags & RTF_GATEWAY && + !ipv6_addr_equal(&cfg->fc_gateway, &rt->rt6i_gateway)) continue; - if (rtmsg->rtmsg_metric && - rtmsg->rtmsg_metric != rt->rt6i_metric) + if (cfg->fc_metric && cfg->fc_metric != rt->rt6i_metric) continue; dst_hold(&rt->u.dst); read_unlock_bh(&table->tb6_lock); - return __ip6_del_rt(rt, nlh, _rtattr, req); + return __ip6_del_rt(rt, &cfg->fc_nlinfo); } } read_unlock_bh(&table->tb6_lock); @@ -1565,21 +1551,23 @@ static struct rt6_info *rt6_add_route_info(struct in6_addr *prefix, int prefixle struct in6_addr *gwaddr, int ifindex, unsigned pref) { - struct in6_rtmsg rtmsg; + struct fib6_config cfg = { + .fc_table = RT6_TABLE_INFO, + .fc_metric = 1024, + .fc_ifindex = ifindex, + .fc_dst_len = prefixlen, + .fc_flags = RTF_GATEWAY | RTF_ADDRCONF | RTF_ROUTEINFO | + RTF_UP | RTF_PREF(pref), + }; + + ipv6_addr_copy(&cfg.fc_dst, prefix); + ipv6_addr_copy(&cfg.fc_gateway, gwaddr); - memset(&rtmsg, 0, sizeof(rtmsg)); - rtmsg.rtmsg_type = RTMSG_NEWROUTE; - ipv6_addr_copy(&rtmsg.rtmsg_dst, prefix); - rtmsg.rtmsg_dst_len = prefixlen; - ipv6_addr_copy(&rtmsg.rtmsg_gateway, gwaddr); - rtmsg.rtmsg_metric = 1024; - rtmsg.rtmsg_flags = RTF_GATEWAY | RTF_ADDRCONF | RTF_ROUTEINFO | RTF_UP | RTF_PREF(pref); /* We should treat it as a default route if prefix length is 0. */ if (!prefixlen) - rtmsg.rtmsg_flags |= RTF_DEFAULT; - rtmsg.rtmsg_ifindex = ifindex; + cfg.fc_flags |= RTF_DEFAULT; - ip6_route_add(&rtmsg, NULL, NULL, NULL, RT6_TABLE_INFO); + ip6_route_add(&cfg); return rt6_get_route_info(prefix, prefixlen, gwaddr, ifindex); } @@ -1611,18 +1599,18 @@ struct rt6_info *rt6_add_dflt_router(struct in6_addr *gwaddr, struct net_device *dev, unsigned int pref) { - struct in6_rtmsg rtmsg; + struct fib6_config cfg = { + .fc_table = RT6_TABLE_DFLT, + .fc_metric = 1024, + .fc_ifindex = dev->ifindex, + .fc_flags = RTF_GATEWAY | RTF_ADDRCONF | RTF_DEFAULT | + RTF_UP | RTF_EXPIRES | RTF_PREF(pref), + }; - memset(&rtmsg, 0, sizeof(struct in6_rtmsg)); - rtmsg.rtmsg_type = RTMSG_NEWROUTE; - ipv6_addr_copy(&rtmsg.rtmsg_gateway, gwaddr); - rtmsg.rtmsg_metric = 1024; - rtmsg.rtmsg_flags = RTF_GATEWAY | RTF_ADDRCONF | RTF_DEFAULT | RTF_UP | RTF_EXPIRES | - RTF_PREF(pref); + ipv6_addr_copy(&cfg.fc_gateway, gwaddr); - rtmsg.rtmsg_ifindex = dev->ifindex; + ip6_route_add(&cfg); - ip6_route_add(&rtmsg, NULL, NULL, NULL, RT6_TABLE_DFLT); return rt6_get_dflt_router(gwaddr, dev); } @@ -1649,8 +1637,27 @@ restart: read_unlock_bh(&table->tb6_lock); } +static void rtmsg_to_fib6_config(struct in6_rtmsg *rtmsg, + struct fib6_config *cfg) +{ + memset(cfg, 0, sizeof(*cfg)); + + cfg->fc_table = RT6_TABLE_MAIN; + cfg->fc_ifindex = rtmsg->rtmsg_ifindex; + cfg->fc_metric = rtmsg->rtmsg_metric; + cfg->fc_expires = rtmsg->rtmsg_info; + cfg->fc_dst_len = rtmsg->rtmsg_dst_len; + cfg->fc_src_len = rtmsg->rtmsg_src_len; + cfg->fc_flags = rtmsg->rtmsg_flags; + + ipv6_addr_copy(&cfg->fc_dst, &rtmsg->rtmsg_dst); + ipv6_addr_copy(&cfg->fc_src, &rtmsg->rtmsg_src); + ipv6_addr_copy(&cfg->fc_gateway, &rtmsg->rtmsg_gateway); +} + int ipv6_route_ioctl(unsigned int cmd, void __user *arg) { + struct fib6_config cfg; struct in6_rtmsg rtmsg; int err; @@ -1663,16 +1670,16 @@ int ipv6_route_ioctl(unsigned int cmd, void __user *arg) sizeof(struct in6_rtmsg)); if (err) return -EFAULT; - + + rtmsg_to_fib6_config(&rtmsg, &cfg); + rtnl_lock(); switch (cmd) { case SIOCADDRT: - err = ip6_route_add(&rtmsg, NULL, NULL, NULL, - RT6_TABLE_MAIN); + err = ip6_route_add(&cfg); break; case SIOCDELRT: - err = ip6_route_del(&rtmsg, NULL, NULL, NULL, - RT6_TABLE_MAIN); + err = ip6_route_del(&cfg); break; default: err = -EINVAL; @@ -1823,66 +1830,104 @@ void rt6_mtu_change(struct net_device *dev, unsigned mtu) fib6_clean_all(rt6_mtu_change_route, 0, &arg); } -static int inet6_rtm_to_rtmsg(struct rtmsg *r, struct rtattr **rta, - struct in6_rtmsg *rtmsg) +static struct nla_policy rtm_ipv6_policy[RTA_MAX+1] __read_mostly = { + [RTA_GATEWAY] = { .minlen = sizeof(struct in6_addr) }, + [RTA_OIF] = { .type = NLA_U32 }, + [RTA_PRIORITY] = { .type = NLA_U32 }, + [RTA_METRICS] = { .type = NLA_NESTED }, +}; + +static int rtm_to_fib6_config(struct sk_buff *skb, struct nlmsghdr *nlh, + struct fib6_config *cfg) { - memset(rtmsg, 0, sizeof(*rtmsg)); + struct rtmsg *rtm; + struct nlattr *tb[RTA_MAX+1]; + int err; + + err = nlmsg_parse(nlh, sizeof(*rtm), tb, RTA_MAX, rtm_ipv6_policy); + if (err < 0) + goto errout; - rtmsg->rtmsg_dst_len = r->rtm_dst_len; - rtmsg->rtmsg_src_len = r->rtm_src_len; - rtmsg->rtmsg_flags = RTF_UP; - if (r->rtm_type == RTN_UNREACHABLE) - rtmsg->rtmsg_flags |= RTF_REJECT; + err = -EINVAL; + rtm = nlmsg_data(nlh); + memset(cfg, 0, sizeof(*cfg)); - if (rta[RTA_GATEWAY-1]) { - if (rta[RTA_GATEWAY-1]->rta_len != RTA_LENGTH(16)) - return -EINVAL; - memcpy(&rtmsg->rtmsg_gateway, RTA_DATA(rta[RTA_GATEWAY-1]), 16); - rtmsg->rtmsg_flags |= RTF_GATEWAY; - } - if (rta[RTA_DST-1]) { - if (RTA_PAYLOAD(rta[RTA_DST-1]) < ((r->rtm_dst_len+7)>>3)) - return -EINVAL; - memcpy(&rtmsg->rtmsg_dst, RTA_DATA(rta[RTA_DST-1]), ((r->rtm_dst_len+7)>>3)); + cfg->fc_table = rtm->rtm_table; + cfg->fc_dst_len = rtm->rtm_dst_len; + cfg->fc_src_len = rtm->rtm_src_len; + cfg->fc_flags = RTF_UP; + cfg->fc_protocol = rtm->rtm_protocol; + + if (rtm->rtm_type == RTN_UNREACHABLE) + cfg->fc_flags |= RTF_REJECT; + + cfg->fc_nlinfo.pid = NETLINK_CB(skb).pid; + cfg->fc_nlinfo.nlh = nlh; + + if (tb[RTA_GATEWAY]) { + nla_memcpy(&cfg->fc_gateway, tb[RTA_GATEWAY], 16); + cfg->fc_flags |= RTF_GATEWAY; } - if (rta[RTA_SRC-1]) { - if (RTA_PAYLOAD(rta[RTA_SRC-1]) < ((r->rtm_src_len+7)>>3)) - return -EINVAL; - memcpy(&rtmsg->rtmsg_src, RTA_DATA(rta[RTA_SRC-1]), ((r->rtm_src_len+7)>>3)); + + if (tb[RTA_DST]) { + int plen = (rtm->rtm_dst_len + 7) >> 3; + + if (nla_len(tb[RTA_DST]) < plen) + goto errout; + + nla_memcpy(&cfg->fc_dst, tb[RTA_DST], plen); } - if (rta[RTA_OIF-1]) { - if (rta[RTA_OIF-1]->rta_len != RTA_LENGTH(sizeof(int))) - return -EINVAL; - memcpy(&rtmsg->rtmsg_ifindex, RTA_DATA(rta[RTA_OIF-1]), sizeof(int)); + + if (tb[RTA_SRC]) { + int plen = (rtm->rtm_src_len + 7) >> 3; + + if (nla_len(tb[RTA_SRC]) < plen) + goto errout; + + nla_memcpy(&cfg->fc_src, tb[RTA_SRC], plen); } - if (rta[RTA_PRIORITY-1]) { - if (rta[RTA_PRIORITY-1]->rta_len != RTA_LENGTH(4)) - return -EINVAL; - memcpy(&rtmsg->rtmsg_metric, RTA_DATA(rta[RTA_PRIORITY-1]), 4); + + if (tb[RTA_OIF]) + cfg->fc_ifindex = nla_get_u32(tb[RTA_OIF]); + + if (tb[RTA_PRIORITY]) + cfg->fc_metric = nla_get_u32(tb[RTA_PRIORITY]); + + if (tb[RTA_METRICS]) { + cfg->fc_mx = nla_data(tb[RTA_METRICS]); + cfg->fc_mx_len = nla_len(tb[RTA_METRICS]); } - return 0; + + if (tb[RTA_TABLE]) + cfg->fc_table = nla_get_u32(tb[RTA_TABLE]); + + err = 0; +errout: + return err; } int inet6_rtm_delroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) { - struct rtmsg *r = NLMSG_DATA(nlh); - struct in6_rtmsg rtmsg; + struct fib6_config cfg; + int err; - if (inet6_rtm_to_rtmsg(r, arg, &rtmsg)) - return -EINVAL; - return ip6_route_del(&rtmsg, nlh, arg, &NETLINK_CB(skb), - rtm_get_table(arg, r->rtm_table)); + err = rtm_to_fib6_config(skb, nlh, &cfg); + if (err < 0) + return err; + + return ip6_route_del(&cfg); } int inet6_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) { - struct rtmsg *r = NLMSG_DATA(nlh); - struct in6_rtmsg rtmsg; + struct fib6_config cfg; + int err; - if (inet6_rtm_to_rtmsg(r, arg, &rtmsg)) - return -EINVAL; - return ip6_route_add(&rtmsg, nlh, arg, &NETLINK_CB(skb), - rtm_get_table(arg, r->rtm_table)); + err = rtm_to_fib6_config(skb, nlh, &cfg); + if (err < 0) + return err; + + return ip6_route_add(&cfg); } static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt, @@ -2063,15 +2108,21 @@ out_free: goto out; } -void inet6_rt_notify(int event, struct rt6_info *rt, struct nlmsghdr *nlh, - struct netlink_skb_parms *req) +void inet6_rt_notify(int event, struct rt6_info *rt, struct nl_info *info) { struct sk_buff *skb; - u32 pid = req ? req->pid : 0; - u32 seq = nlh ? nlh->nlmsg_seq : 0; + u32 pid = 0, seq = 0; + struct nlmsghdr *nlh = NULL; int payload = sizeof(struct rtmsg) + 256; int err = -ENOBUFS; + if (info) { + pid = info->pid; + nlh = info->nlh; + if (nlh) + seq = nlh->nlmsg_seq; + } + skb = nlmsg_new(nlmsg_total_size(payload), gfp_any()); if (skb == NULL) goto errout; -- cgit v1.1 From 2d7202bfdd28687073f5efef8d2f51bbab0af867 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Tue, 22 Aug 2006 00:01:27 -0700 Subject: [IPv6] route: Convert FIB6 dumping to use new netlink api Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- net/ipv6/route.c | 52 ++++++++++++++++++++++++++++------------------------ 1 file changed, 28 insertions(+), 24 deletions(-) (limited to 'net/ipv6/route.c') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 7bcffa6..f0a66de 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1936,8 +1936,7 @@ static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt, int prefix, unsigned int flags) { struct rtmsg *rtm; - struct nlmsghdr *nlh; - unsigned char *b = skb->tail; + struct nlmsghdr *nlh; struct rta_cacheinfo ci; u32 table; @@ -1948,8 +1947,11 @@ static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt, } } - nlh = NLMSG_NEW(skb, pid, seq, type, sizeof(*rtm), flags); - rtm = NLMSG_DATA(nlh); + nlh = nlmsg_put(skb, pid, seq, type, sizeof(*rtm), flags); + if (nlh == NULL) + return -ENOBUFS; + + rtm = nlmsg_data(nlh); rtm->rtm_family = AF_INET6; rtm->rtm_dst_len = rt->rt6i_dst.plen; rtm->rtm_src_len = rt->rt6i_src.plen; @@ -1959,7 +1961,7 @@ static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt, else table = RT6_TABLE_UNSPEC; rtm->rtm_table = table; - RTA_PUT_U32(skb, RTA_TABLE, table); + NLA_PUT_U32(skb, RTA_TABLE, table); if (rt->rt6i_flags&RTF_REJECT) rtm->rtm_type = RTN_UNREACHABLE; else if (rt->rt6i_dev && (rt->rt6i_dev->flags&IFF_LOOPBACK)) @@ -1980,31 +1982,35 @@ static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt, rtm->rtm_flags |= RTM_F_CLONED; if (dst) { - RTA_PUT(skb, RTA_DST, 16, dst); + NLA_PUT(skb, RTA_DST, 16, dst); rtm->rtm_dst_len = 128; } else if (rtm->rtm_dst_len) - RTA_PUT(skb, RTA_DST, 16, &rt->rt6i_dst.addr); + NLA_PUT(skb, RTA_DST, 16, &rt->rt6i_dst.addr); #ifdef CONFIG_IPV6_SUBTREES if (src) { - RTA_PUT(skb, RTA_SRC, 16, src); + NLA_PUT(skb, RTA_SRC, 16, src); rtm->rtm_src_len = 128; } else if (rtm->rtm_src_len) - RTA_PUT(skb, RTA_SRC, 16, &rt->rt6i_src.addr); + NLA_PUT(skb, RTA_SRC, 16, &rt->rt6i_src.addr); #endif if (iif) - RTA_PUT(skb, RTA_IIF, 4, &iif); + NLA_PUT_U32(skb, RTA_IIF, iif); else if (dst) { struct in6_addr saddr_buf; if (ipv6_get_saddr(&rt->u.dst, dst, &saddr_buf) == 0) - RTA_PUT(skb, RTA_PREFSRC, 16, &saddr_buf); + NLA_PUT(skb, RTA_PREFSRC, 16, &saddr_buf); } + if (rtnetlink_put_metrics(skb, rt->u.dst.metrics) < 0) - goto rtattr_failure; + goto nla_put_failure; + if (rt->u.dst.neighbour) - RTA_PUT(skb, RTA_GATEWAY, 16, &rt->u.dst.neighbour->primary_key); + NLA_PUT(skb, RTA_GATEWAY, 16, &rt->u.dst.neighbour->primary_key); + if (rt->u.dst.dev) - RTA_PUT(skb, RTA_OIF, sizeof(int), &rt->rt6i_dev->ifindex); - RTA_PUT(skb, RTA_PRIORITY, 4, &rt->rt6i_metric); + NLA_PUT_U32(skb, RTA_OIF, rt->rt6i_dev->ifindex); + + NLA_PUT_U32(skb, RTA_PRIORITY, rt->rt6i_metric); ci.rta_lastuse = jiffies_to_clock_t(jiffies - rt->u.dst.lastuse); if (rt->rt6i_expires) ci.rta_expires = jiffies_to_clock_t(rt->rt6i_expires - jiffies); @@ -2016,14 +2022,12 @@ static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt, ci.rta_id = 0; ci.rta_ts = 0; ci.rta_tsage = 0; - RTA_PUT(skb, RTA_CACHEINFO, sizeof(ci), &ci); - nlh->nlmsg_len = skb->tail - b; - return skb->len; + NLA_PUT(skb, RTA_CACHEINFO, sizeof(ci), &ci); + + return nlmsg_end(skb, nlh); -nlmsg_failure: -rtattr_failure: - skb_trim(skb, b - skb->data); - return -1; +nla_put_failure: + return nlmsg_cancel(skb, nlh); } int rt6_dump_route(struct rt6_info *rt, void *p_arg) @@ -2031,8 +2035,8 @@ int rt6_dump_route(struct rt6_info *rt, void *p_arg) struct rt6_rtnl_dump_arg *arg = (struct rt6_rtnl_dump_arg *) p_arg; int prefix; - if (arg->cb->nlh->nlmsg_len >= NLMSG_LENGTH(sizeof(struct rtmsg))) { - struct rtmsg *rtm = NLMSG_DATA(arg->cb->nlh); + if (nlmsg_len(arg->cb->nlh) >= sizeof(struct rtmsg)) { + struct rtmsg *rtm = nlmsg_data(arg->cb->nlh); prefix = (rtm->rtm_flags & RTM_F_PREFIX) != 0; } else prefix = 0; -- cgit v1.1 From ab364a6f96bad9625bdb97b5688c76c44eb1e96e Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Tue, 22 Aug 2006 00:01:47 -0700 Subject: [IPv6] route: Convert GETROUTE to use new netlink api Fixes various unvalidated netlink attributes causing memory corruptions when left empty by userspace applications. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- net/ipv6/route.c | 80 +++++++++++++++++++++++++++++++------------------------- 1 file changed, 44 insertions(+), 36 deletions(-) (limited to 'net/ipv6/route.c') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index f0a66de..5d6e908 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1833,6 +1833,7 @@ void rt6_mtu_change(struct net_device *dev, unsigned mtu) static struct nla_policy rtm_ipv6_policy[RTA_MAX+1] __read_mostly = { [RTA_GATEWAY] = { .minlen = sizeof(struct in6_addr) }, [RTA_OIF] = { .type = NLA_U32 }, + [RTA_IIF] = { .type = NLA_U32 }, [RTA_PRIORITY] = { .type = NLA_U32 }, [RTA_METRICS] = { .type = NLA_NESTED }, }; @@ -2048,68 +2049,75 @@ int rt6_dump_route(struct rt6_info *rt, void *p_arg) int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void *arg) { - struct rtattr **rta = arg; - int iif = 0; - int err = -ENOBUFS; + struct nlattr *tb[RTA_MAX+1]; + struct rt6_info *rt; struct sk_buff *skb; + struct rtmsg *rtm; struct flowi fl; - struct rt6_info *rt; - - skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); - if (skb == NULL) - goto out; + int err, iif = 0; - /* Reserve room for dummy headers, this skb can pass - through good chunk of routing engine. - */ - skb->mac.raw = skb->data; - skb_reserve(skb, MAX_HEADER + sizeof(struct ipv6hdr)); + err = nlmsg_parse(nlh, sizeof(*rtm), tb, RTA_MAX, rtm_ipv6_policy); + if (err < 0) + goto errout; + err = -EINVAL; memset(&fl, 0, sizeof(fl)); - if (rta[RTA_SRC-1]) - ipv6_addr_copy(&fl.fl6_src, - (struct in6_addr*)RTA_DATA(rta[RTA_SRC-1])); - if (rta[RTA_DST-1]) - ipv6_addr_copy(&fl.fl6_dst, - (struct in6_addr*)RTA_DATA(rta[RTA_DST-1])); - if (rta[RTA_IIF-1]) - memcpy(&iif, RTA_DATA(rta[RTA_IIF-1]), sizeof(int)); + if (tb[RTA_SRC]) { + if (nla_len(tb[RTA_SRC]) < sizeof(struct in6_addr)) + goto errout; + + ipv6_addr_copy(&fl.fl6_src, nla_data(tb[RTA_SRC])); + } + + if (tb[RTA_DST]) { + if (nla_len(tb[RTA_DST]) < sizeof(struct in6_addr)) + goto errout; + + ipv6_addr_copy(&fl.fl6_dst, nla_data(tb[RTA_DST])); + } + + if (tb[RTA_IIF]) + iif = nla_get_u32(tb[RTA_IIF]); + + if (tb[RTA_OIF]) + fl.oif = nla_get_u32(tb[RTA_OIF]); if (iif) { struct net_device *dev; dev = __dev_get_by_index(iif); if (!dev) { err = -ENODEV; - goto out_free; + goto errout; } } - fl.oif = 0; - if (rta[RTA_OIF-1]) - memcpy(&fl.oif, RTA_DATA(rta[RTA_OIF-1]), sizeof(int)); + skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); + if (skb == NULL) { + err = -ENOBUFS; + goto errout; + } - rt = (struct rt6_info*)ip6_route_output(NULL, &fl); + /* Reserve room for dummy headers, this skb can pass + through good chunk of routing engine. + */ + skb->mac.raw = skb->data; + skb_reserve(skb, MAX_HEADER + sizeof(struct ipv6hdr)); + rt = (struct rt6_info*) ip6_route_output(NULL, &fl); skb->dst = &rt->u.dst; - NETLINK_CB(skb).dst_pid = NETLINK_CB(in_skb).pid; - err = rt6_fill_node(skb, rt, - &fl.fl6_dst, &fl.fl6_src, - iif, + 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); if (err < 0) { - err = -EMSGSIZE; - goto out_free; + kfree_skb(skb); + goto errout; } err = rtnl_unicast(skb, NETLINK_CB(in_skb).pid); -out: +errout: return err; -out_free: - kfree_skb(skb); - goto out; } void inet6_rt_notify(int event, struct rt6_info *rt, struct nl_info *info) -- cgit v1.1 From 5e032e32ecc2e6cb0385dc115ca9bfe5e19a9539 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Wed, 23 Aug 2006 17:12:24 -0700 Subject: [IPV6] NDISC: Take source address into account for redirects. Signed-off-by: YOSHIFUJI Hideaki Signed-off-by: Ville Nuorvala Signed-off-by: David S. Miller --- net/ipv6/route.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'net/ipv6/route.c') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 5d6e908..a9b08a2 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1279,7 +1279,8 @@ static int ip6_route_del(struct fib6_config *cfg) /* * Handle redirects */ -void rt6_redirect(struct in6_addr *dest, struct in6_addr *saddr, +void rt6_redirect(struct in6_addr *dest, struct in6_addr *src, + struct in6_addr *saddr, struct neighbour *neigh, u8 *lladdr, int on_link) { struct rt6_info *rt, *nrt = NULL; @@ -1304,7 +1305,7 @@ void rt6_redirect(struct in6_addr *dest, struct in6_addr *saddr, */ read_lock_bh(&table->tb6_lock); - fn = fib6_lookup(&table->tb6_root, dest, NULL); + fn = fib6_lookup(&table->tb6_root, dest, src); restart: for (rt = fn->leaf; rt; rt = rt->u.next) { /* -- cgit v1.1 From a6279458c534d01ccc39498aba61c93083ee0372 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Wed, 23 Aug 2006 17:18:26 -0700 Subject: [IPV6] NDISC: Search over all possible rules on receipt of redirect. Split up function for finding routes for redirects. Signed-off-by: YOSHIFUJI Hideaki Signed-off-by: David S. Miller --- net/ipv6/route.c | 85 ++++++++++++++++++++++++++++++++++++++++---------------- 1 file changed, 61 insertions(+), 24 deletions(-) (limited to 'net/ipv6/route.c') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index a9b08a2..8d00a9d 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1279,19 +1279,18 @@ static int ip6_route_del(struct fib6_config *cfg) /* * Handle redirects */ -void rt6_redirect(struct in6_addr *dest, struct in6_addr *src, - struct in6_addr *saddr, - struct neighbour *neigh, u8 *lladdr, int on_link) +struct ip6rd_flowi { + struct flowi fl; + struct in6_addr gateway; +}; + +static struct rt6_info *__ip6_route_redirect(struct fib6_table *table, + struct flowi *fl, + int flags) { - struct rt6_info *rt, *nrt = NULL; + struct ip6rd_flowi *rdfl = (struct ip6rd_flowi *)fl; + struct rt6_info *rt; struct fib6_node *fn; - struct fib6_table *table; - struct netevent_redirect netevent; - - /* TODO: Very lazy, might need to check all tables */ - table = fib6_get_table(RT6_TABLE_MAIN); - if (table == NULL) - return; /* * Get the "current" route for this destination and @@ -1305,7 +1304,7 @@ void rt6_redirect(struct in6_addr *dest, struct in6_addr *src, */ read_lock_bh(&table->tb6_lock); - fn = fib6_lookup(&table->tb6_root, dest, src); + fn = fib6_lookup(&table->tb6_root, &fl->fl6_dst, &fl->fl6_src); restart: for (rt = fn->leaf; rt; rt = rt->u.next) { /* @@ -1320,29 +1319,67 @@ restart: continue; if (!(rt->rt6i_flags & RTF_GATEWAY)) continue; - if (neigh->dev != rt->rt6i_dev) + if (fl->oif != rt->rt6i_dev->ifindex) continue; - if (!ipv6_addr_equal(saddr, &rt->rt6i_gateway)) + if (!ipv6_addr_equal(&rdfl->gateway, &rt->rt6i_gateway)) continue; break; } - if (rt) - dst_hold(&rt->u.dst); - else if (rt6_need_strict(dest)) { - while ((fn = fn->parent) != NULL) { - if (fn->fn_flags & RTN_ROOT) - break; - if (fn->fn_flags & RTN_RTINFO) - goto restart; + + if (!rt) { + if (rt6_need_strict(&fl->fl6_dst)) { + while ((fn = fn->parent) != NULL) { + if (fn->fn_flags & RTN_ROOT) + break; + if (fn->fn_flags & RTN_RTINFO) + goto restart; + } } + rt = &ip6_null_entry; } + dst_hold(&rt->u.dst); + read_unlock_bh(&table->tb6_lock); - if (!rt) { + return rt; +}; + +static struct rt6_info *ip6_route_redirect(struct in6_addr *dest, + struct in6_addr *src, + struct in6_addr *gateway, + struct net_device *dev) +{ + struct ip6rd_flowi rdfl = { + .fl = { + .oif = dev->ifindex, + .nl_u = { + .ip6_u = { + .daddr = *dest, + .saddr = *src, + }, + }, + }, + .gateway = *gateway, + }; + int flags = rt6_need_strict(dest) ? RT6_F_STRICT : 0; + + return (struct rt6_info *)fib6_rule_lookup((struct flowi *)&rdfl, flags, __ip6_route_redirect); +} + +void rt6_redirect(struct in6_addr *dest, struct in6_addr *src, + struct in6_addr *saddr, + struct neighbour *neigh, u8 *lladdr, int on_link) +{ + struct rt6_info *rt, *nrt = NULL; + struct netevent_redirect netevent; + + rt = ip6_route_redirect(dest, src, saddr, neigh->dev); + + if (rt == &ip6_null_entry) { if (net_ratelimit()) printk(KERN_DEBUG "rt6_redirect: source isn't a valid nexthop " "for redirect target\n"); - return; + goto out; } /* -- cgit v1.1 From 982f56f3a9be4651520c0fdd3d80a5d02e95a178 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Wed, 23 Aug 2006 17:22:39 -0700 Subject: [IPV6] ROUTE: Search subtree when backtracking. Based on MIPL2 kernel patch. Signed-off-by: YOSHIFUJI Hideaki Signed-off-by: Ville Nuorvala Signed-off-by: David S. Miller --- net/ipv6/route.c | 30 ++++++++++++++++++------------ 1 file changed, 18 insertions(+), 12 deletions(-) (limited to 'net/ipv6/route.c') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 8d00a9d..bd4cf17 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -481,17 +481,23 @@ int rt6_route_rcv(struct net_device *dev, u8 *opt, int len, } #endif -#define BACKTRACK() \ -if (rt == &ip6_null_entry && flags & RT6_F_STRICT) { \ - while ((fn = fn->parent) != NULL) { \ - if (fn->fn_flags & RTN_TL_ROOT) { \ - dst_hold(&rt->u.dst); \ - goto out; \ +#define BACKTRACK(saddr) \ +do { \ + if (rt == &ip6_null_entry) { \ + struct fib6_node *pn; \ + while (fn) { \ + if (fn->fn_flags & RTN_TL_ROOT) \ + goto out; \ + pn = fn->parent; \ + if (FIB6_SUBTREE(pn) && FIB6_SUBTREE(pn) != fn) \ + fn = fib6_lookup(pn->subtree, NULL, saddr); \ + else \ + fn = pn; \ + if (fn->fn_flags & RTN_RTINFO) \ + goto restart; \ } \ - if (fn->fn_flags & RTN_RTINFO) \ - goto restart; \ } \ -} +} while(0) static struct rt6_info *ip6_pol_route_lookup(struct fib6_table *table, struct flowi *fl, int flags) @@ -504,7 +510,7 @@ static struct rt6_info *ip6_pol_route_lookup(struct fib6_table *table, restart: rt = fn->leaf; rt = rt6_device_match(rt, fl->oif, flags & RT6_F_STRICT); - BACKTRACK(); + BACKTRACK(&fl->fl6_src); dst_hold(&rt->u.dst); out: read_unlock_bh(&table->tb6_lock); @@ -638,7 +644,7 @@ restart_2: restart: rt = rt6_select(&fn->leaf, fl->iif, strict | reachable); - BACKTRACK(); + BACKTRACK(&fl->fl6_src); if (rt == &ip6_null_entry || rt->rt6i_flags & RTF_CACHE) goto out; @@ -733,7 +739,7 @@ restart_2: restart: rt = rt6_select(&fn->leaf, fl->oif, strict | reachable); - BACKTRACK(); + BACKTRACK(&fl->fl6_src); if (rt == &ip6_null_entry || rt->rt6i_flags & RTF_CACHE) goto out; -- cgit v1.1 From cb15d9c224fcc03b32396c1c7416e777c2dcca34 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Wed, 23 Aug 2006 17:23:11 -0700 Subject: [IPV6] NDISC: Search subtrees when backtracking on receipt of redirects. Signed-off-by: YOSHIFUJI Hideaki Acked-by: Ville Nuorvala --- net/ipv6/route.c | 13 +++---------- 1 file changed, 3 insertions(+), 10 deletions(-) (limited to 'net/ipv6/route.c') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index bd4cf17..fd626d4 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1332,17 +1332,10 @@ restart: break; } - if (!rt) { - if (rt6_need_strict(&fl->fl6_dst)) { - while ((fn = fn->parent) != NULL) { - if (fn->fn_flags & RTN_ROOT) - break; - if (fn->fn_flags & RTN_RTINFO) - goto restart; - } - } + if (!rt) rt = &ip6_null_entry; - } + BACKTRACK(&fl->fl6_src); +out: dst_hold(&rt->u.dst); read_unlock_bh(&table->tb6_lock); -- cgit v1.1 From c0bece9f2aec546c3750ae3972f80e024a923f34 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Wed, 23 Aug 2006 17:23:25 -0700 Subject: [IPV6] ROUTE: Add credits about subtree fixes. Based on MIPL2 kernel patch. Signed-off-by: YOSHIFUJI Hideaki Signed-off-by: David S. Miller --- net/ipv6/route.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net/ipv6/route.c') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index fd626d4..fd6f2ec 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -22,6 +22,8 @@ * routers in REACHABLE, STALE, DELAY or PROBE states). * - always select the same router if it is (probably) * reachable. otherwise, round-robin the list. + * Ville Nuorvala + * Fixed routing subtrees. */ #include -- cgit v1.1 From 77d16f450ae0452d7d4b009f78debb1294fb435c Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Wed, 23 Aug 2006 17:25:05 -0700 Subject: [IPV6] ROUTE: Unify RT6_F_xxx and RT6_SELECT_F_xxx flags Unify RT6_F_xxx and RT6_SELECT_F_xxx flags into RT6_LOOKUP_F_xxx flags, and put them into ip6_route.h Signed-off-by: YOSHIFUJI Hideaki Acked-by: Ville Nuorvala --- net/ipv6/route.c | 32 ++++++++++++-------------------- 1 file changed, 12 insertions(+), 20 deletions(-) (limited to 'net/ipv6/route.c') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index fd6f2ec..2069128 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -76,9 +76,6 @@ #define CLONE_OFFLINK_ROUTE 0 -#define RT6_SELECT_F_IFACE 0x1 -#define RT6_SELECT_F_REACHABLE 0x2 - static int ip6_rt_max_size = 4096; static int ip6_rt_gc_min_interval = HZ / 2; static int ip6_rt_gc_timeout = 60*HZ; @@ -340,7 +337,7 @@ static int rt6_score_route(struct rt6_info *rt, int oif, int m, n; m = rt6_check_dev(rt, oif); - if (!m && (strict & RT6_SELECT_F_IFACE)) + if (!m && (strict & RT6_LOOKUP_F_IFACE)) return -1; #ifdef CONFIG_IPV6_ROUTER_PREF m |= IPV6_DECODE_PREF(IPV6_EXTRACT_PREF(rt->rt6i_flags)) << 2; @@ -348,7 +345,7 @@ static int rt6_score_route(struct rt6_info *rt, int oif, n = rt6_check_neigh(rt); if (n > 1) m |= 16; - else if (!n && strict & RT6_SELECT_F_REACHABLE) + else if (!n && strict & RT6_LOOKUP_F_REACHABLE) return -1; return m; } @@ -388,7 +385,7 @@ static struct rt6_info *rt6_select(struct rt6_info **head, int oif, } if (!match && - (strict & RT6_SELECT_F_REACHABLE) && + (strict & RT6_LOOKUP_F_REACHABLE) && last && last != rt0) { /* no entries matched; do round-robin */ static DEFINE_SPINLOCK(lock); @@ -511,7 +508,7 @@ static struct rt6_info *ip6_pol_route_lookup(struct fib6_table *table, fn = fib6_lookup(&table->tb6_root, &fl->fl6_dst, &fl->fl6_src); restart: rt = fn->leaf; - rt = rt6_device_match(rt, fl->oif, flags & RT6_F_STRICT); + rt = rt6_device_match(rt, fl->oif, flags); BACKTRACK(&fl->fl6_src); dst_hold(&rt->u.dst); out: @@ -537,7 +534,7 @@ struct rt6_info *rt6_lookup(struct in6_addr *daddr, struct in6_addr *saddr, }, }; struct dst_entry *dst; - int flags = strict ? RT6_F_STRICT : 0; + int flags = strict ? RT6_LOOKUP_F_IFACE : 0; dst = fib6_rule_lookup(&fl, flags, ip6_pol_route_lookup); if (dst->error == 0) @@ -633,10 +630,9 @@ static struct rt6_info *ip6_pol_route_input(struct fib6_table *table, int strict = 0; int attempts = 3; int err; - int reachable = RT6_SELECT_F_REACHABLE; + int reachable = RT6_LOOKUP_F_REACHABLE; - if (flags & RT6_F_STRICT) - strict = RT6_SELECT_F_IFACE; + strict |= flags & RT6_LOOKUP_F_IFACE; relookup: read_lock_bh(&table->tb6_lock); @@ -712,10 +708,7 @@ void ip6_route_input(struct sk_buff *skb) }, .proto = iph->nexthdr, }; - int flags = 0; - - if (rt6_need_strict(&iph->daddr)) - flags |= RT6_F_STRICT; + int flags = rt6_need_strict(&iph->daddr) ? RT6_LOOKUP_F_IFACE : 0; skb->dst = fib6_rule_lookup(&fl, flags, ip6_pol_route_input); } @@ -728,10 +721,9 @@ static struct rt6_info *ip6_pol_route_output(struct fib6_table *table, int strict = 0; int attempts = 3; int err; - int reachable = RT6_SELECT_F_REACHABLE; + int reachable = RT6_LOOKUP_F_REACHABLE; - if (flags & RT6_F_STRICT) - strict = RT6_SELECT_F_IFACE; + strict |= flags & RT6_LOOKUP_F_IFACE; relookup: read_lock_bh(&table->tb6_lock); @@ -797,7 +789,7 @@ struct dst_entry * ip6_route_output(struct sock *sk, struct flowi *fl) int flags = 0; if (rt6_need_strict(&fl->fl6_dst)) - flags |= RT6_F_STRICT; + flags |= RT6_LOOKUP_F_IFACE; return fib6_rule_lookup(fl, flags, ip6_pol_route_output); } @@ -1362,7 +1354,7 @@ static struct rt6_info *ip6_route_redirect(struct in6_addr *dest, }, .gateway = *gateway, }; - int flags = rt6_need_strict(dest) ? RT6_F_STRICT : 0; + int flags = rt6_need_strict(dest) ? RT6_LOOKUP_F_IFACE : 0; return (struct rt6_info *)fib6_rule_lookup((struct flowi *)&rdfl, flags, __ip6_route_redirect); } -- cgit v1.1 From 75bff8f023e02b045a8f68f36fa7da98dca124b8 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Mon, 21 Aug 2006 19:22:01 +0900 Subject: [IPV6] ROUTE: Routing by FWMARK. Based on patch by Jean Lorchat . Signed-off-by: YOSHIFUJI Hideaki --- net/ipv6/route.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net/ipv6/route.c') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 2069128..649350b 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -703,6 +703,7 @@ void ip6_route_input(struct sk_buff *skb) .ip6_u = { .daddr = iph->daddr, .saddr = iph->saddr, + .fwmark = skb->nfmark, .flowlabel = (* (u32 *) iph)&IPV6_FLOWINFO_MASK, }, }, -- cgit v1.1 From 267935b197d2a6e6924f9de2841f0470bfe63acd Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 25 Aug 2006 16:07:48 -0700 Subject: [IPV6]: Fix build with fwmark disabled. Based upon a patch by Brian Haley. Signed-off-by: David S. Miller --- net/ipv6/route.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net/ipv6/route.c') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 649350b..d83844d 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -703,7 +703,9 @@ void ip6_route_input(struct sk_buff *skb) .ip6_u = { .daddr = iph->daddr, .saddr = iph->saddr, +#ifdef CONFIG_IPV6_ROUTE_FWMARK .fwmark = skb->nfmark, +#endif .flowlabel = (* (u32 *) iph)&IPV6_FLOWINFO_MASK, }, }, -- cgit v1.1 From e5d679f33900c71d1a76ba07c5b04055abd34480 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Sat, 26 Aug 2006 19:25:52 -0700 Subject: [NET]: Use SLAB_PANIC Signed-off-by: Alexey Dobriyan Signed-off-by: David S. Miller --- net/ipv6/route.c | 10 +++------- 1 file changed, 3 insertions(+), 7 deletions(-) (limited to 'net/ipv6/route.c') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index d83844d..ba1b3d1 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -2419,13 +2419,9 @@ void __init ip6_route_init(void) { struct proc_dir_entry *p; - ip6_dst_ops.kmem_cachep = kmem_cache_create("ip6_dst_cache", - sizeof(struct rt6_info), - 0, SLAB_HWCACHE_ALIGN, - NULL, NULL); - if (!ip6_dst_ops.kmem_cachep) - panic("cannot create ip6_dst_cache"); - + ip6_dst_ops.kmem_cachep = + kmem_cache_create("ip6_dst_cache", sizeof(struct rt6_info), 0, + SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL, NULL); fib6_init(); #ifdef CONFIG_PROC_FS p = proc_net_create("ipv6_route", 0, rt6_proc_info); -- cgit v1.1 From 5176f91ea83f1a59eba4dba88634a4729d51d1ac Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Sat, 26 Aug 2006 20:13:18 -0700 Subject: [NETLINK]: Make use of NLA_STRING/NLA_NUL_STRING attribute validation Converts existing NLA_STRING attributes to use the new validation features, saving a couple of temporary buffers. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- net/ipv6/route.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/ipv6/route.c') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index ba1b3d1..75f4bb9 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1865,7 +1865,7 @@ void rt6_mtu_change(struct net_device *dev, unsigned mtu) } static struct nla_policy rtm_ipv6_policy[RTA_MAX+1] __read_mostly = { - [RTA_GATEWAY] = { .minlen = sizeof(struct in6_addr) }, + [RTA_GATEWAY] = { .len = sizeof(struct in6_addr) }, [RTA_OIF] = { .type = NLA_U32 }, [RTA_IIF] = { .type = NLA_U32 }, [RTA_PRIORITY] = { .type = NLA_U32 }, -- cgit v1.1 From 33cc48966827165e49de1cb8ff4fb57c127d4be0 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Mon, 28 Aug 2006 13:19:30 -0700 Subject: [IPV6] ROUTE: Fix dst reference counting in ip6_pol_route_lookup(). In ip6_pol_route_lookup(), when we finish backtracking at the top-level root entry, we need to hold it. Bug noticed by Mitsuru Chinen . Signed-off-by: YOSHIFUJI Hideaki Signed-off-by: David S. Miller --- net/ipv6/route.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/ipv6/route.c') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 75f4bb9..d6b4b4f 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -510,8 +510,8 @@ restart: rt = fn->leaf; rt = rt6_device_match(rt, fl->oif, flags); BACKTRACK(&fl->fl6_src); - dst_hold(&rt->u.dst); out: + dst_hold(&rt->u.dst); read_unlock_bh(&table->tb6_lock); rt->u.dst.lastuse = jiffies; -- cgit v1.1