/* * INET An implementation of the TCP/IP protocol suite for the LINUX * operating system. INET is implemented using the BSD Socket * interface as the means of communication with the user level. * * IPv4 Forwarding Information Base: FIB frontend. * * Version: $Id: fib_frontend.c,v 1.21 1999/12/15 22:39:07 davem Exp $ * * Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru> * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License * as published by the Free Software Foundation; either version * 2 of the License, or (at your option) any later version. */ #include <linux/config.h> #include <asm/uaccess.h> #include <asm/system.h> #include <asm/bitops.h> #include <linux/types.h> #include <linux/kernel.h> #include <linux/sched.h> #include <linux/mm.h> #include <linux/string.h> #include <linux/socket.h> #include <linux/sockios.h> #include <linux/errno.h> #include <linux/in.h> #include <linux/inet.h> #include <linux/netdevice.h> #include <linux/if_arp.h> #include <linux/proc_fs.h> #include <linux/skbuff.h> #include <linux/netlink.h> #include <linux/init.h> #include <net/ip.h> #include <net/protocol.h> #include <net/route.h> #include <net/tcp.h> #include <net/sock.h> #include <net/icmp.h> #include <net/arp.h> #include <net/ip_fib.h> #define FFprint(a...) printk(KERN_DEBUG a) #ifndef CONFIG_IP_MULTIPLE_TABLES #define RT_TABLE_MIN RT_TABLE_MAIN struct fib_table *local_table; struct fib_table *main_table; #else #define RT_TABLE_MIN 1 struct fib_table *fib_tables[RT_TABLE_MAX+1]; struct fib_table *__fib_new_table(int id) { struct fib_table *tb; tb = fib_hash_init(id); if (!tb) return NULL; fib_tables[id] = tb; return tb; } #endif /* CONFIG_IP_MULTIPLE_TABLES */ 78 void fib_flush(void) { int flushed = 0; #ifdef CONFIG_IP_MULTIPLE_TABLES struct fib_table *tb; int id; for (id = RT_TABLE_MAX; id>0; id--) { if ((tb = fib_get_table(id))==NULL) continue; flushed += tb->tb_flush(tb); } #else /* CONFIG_IP_MULTIPLE_TABLES */ flushed += main_table->tb_flush(main_table); flushed += local_table->tb_flush(local_table); #endif /* CONFIG_IP_MULTIPLE_TABLES */ 95 if (flushed) rt_cache_flush(-1); } #ifdef CONFIG_PROC_FS /* * Called from the PROCfs module. This outputs /proc/net/route. * * It always works in backward compatibility mode. * The format of the file is not supposed to be changed. */ static int 110 fib_get_procinfo(char *buffer, char **start, off_t offset, int length) { int first = offset/128; char *ptr = buffer; int count = (length+127)/128; int len; *start = buffer + offset%128; 119 if (--first < 0) { sprintf(buffer, "%-127s\n", "Iface\tDestination\tGateway \tFlags\tRefCnt\tUse\tMetric\tMask\t\tMTU\tWindow\tIRTT"); --count; ptr += 128; first = 0; } 126 if (main_table && count > 0) { int n = main_table->tb_get_info(main_table, ptr, first, count); count -= n; ptr += n*128; } len = ptr - *start; 132 if (len >= length) 133 return length; 134 if (len >= 0) 135 return len; 136 return 0; } #endif /* CONFIG_PROC_FS */ /* * Find the first device with a given source address. */ 145 struct net_device * ip_dev_find(u32 addr) { struct rt_key key; struct fib_result res; struct net_device *dev = NULL; memset(&key, 0, sizeof(key)); key.dst = addr; #ifdef CONFIG_IP_MULTIPLE_TABLES res.r = NULL; #endif 157 if (!local_table || local_table->tb_lookup(local_table, &key, &res)) { 158 return NULL; } 160 if (res.type != RTN_LOCAL) 161 goto out; dev = FIB_RES_DEV(res); 163 if (dev) atomic_inc(&dev->refcnt); out: fib_res_put(&res); 168 return dev; } 171 unsigned inet_addr_type(u32 addr) { struct rt_key key; struct fib_result res; unsigned ret = RTN_BROADCAST; 177 if (ZERONET(addr) || BADCLASS(addr)) 178 return RTN_BROADCAST; 179 if (MULTICAST(addr)) 180 return RTN_MULTICAST; memset(&key, 0, sizeof(key)); key.dst = addr; #ifdef CONFIG_IP_MULTIPLE_TABLES res.r = NULL; #endif 188 if (local_table) { ret = RTN_UNICAST; 190 if (local_table->tb_lookup(local_table, &key, &res) == 0) { ret = res.type; fib_res_put(&res); } } 195 return ret; } /* Given (packet source, input interface) and optional (dst, oif, tos): - (main) check, that source is valid i.e. not broadcast or our local address. - figure out what "logical" interface this packet arrived and calculate "specific destination" address. - check, that packet arrived from expected physical interface. */ 206 int fib_validate_source(u32 src, u32 dst, u8 tos, int oif, struct net_device *dev, u32 *spec_dst, u32 *itag) { struct in_device *in_dev; struct rt_key key; struct fib_result res; int no_addr, rpf; int ret; key.dst = src; key.src = dst; key.tos = tos; key.oif = 0; key.iif = oif; key.scope = RT_SCOPE_UNIVERSE; no_addr = rpf = 0; read_lock(&inetdev_lock); in_dev = __in_dev_get(dev); 225 if (in_dev) { no_addr = in_dev->ifa_list == NULL; rpf = IN_DEV_RPFILTER(in_dev); } 229 read_unlock(&inetdev_lock); 231 if (in_dev == NULL) 232 goto e_inval; 234 if (fib_lookup(&key, &res)) 235 goto last_resort; 236 if (res.type != RTN_UNICAST) 237 goto e_inval_res; *spec_dst = FIB_RES_PREFSRC(res); 239 if (itag) fib_combine_itag(itag, &res); #ifdef CONFIG_IP_ROUTE_MULTIPATH if (FIB_RES_DEV(res) == dev || res.fi->fib_nhs > 1) #else 244 if (FIB_RES_DEV(res) == dev) #endif { ret = FIB_RES_NH(res).nh_scope >= RT_SCOPE_HOST; fib_res_put(&res); 249 return ret; } fib_res_put(&res); 252 if (no_addr) 253 goto last_resort; 254 if (rpf) 255 goto e_inval; key.oif = dev->ifindex; ret = 0; 259 if (fib_lookup(&key, &res) == 0) { 260 if (res.type == RTN_UNICAST) { *spec_dst = FIB_RES_PREFSRC(res); ret = FIB_RES_NH(res).nh_scope >= RT_SCOPE_HOST; } fib_res_put(&res); } 266 return ret; last_resort: 269 if (rpf) 270 goto e_inval; *spec_dst = inet_select_addr(dev, 0, RT_SCOPE_UNIVERSE); *itag = 0; 273 return 0; e_inval_res: fib_res_put(&res); e_inval: 278 return -EINVAL; } #ifndef CONFIG_IP_NOSIOCRT /* * Handle IP routing ioctl calls. These are used to manipulate the routing tables */ 287 int ip_rt_ioctl(unsigned int cmd, void *arg) { int err; struct kern_rta rta; struct rtentry r; struct { struct nlmsghdr nlh; struct rtmsg rtm; } req; 297 switch (cmd) { 298 case SIOCADDRT: /* Add a route */ 299 case SIOCDELRT: /* Delete a route */ 300 if (!capable(CAP_NET_ADMIN)) 301 return -EPERM; 302 if (copy_from_user(&r, arg, sizeof(struct rtentry))) 303 return -EFAULT; rtnl_lock(); err = fib_convert_rtentry(cmd, &req.nlh, &req.rtm, &rta, &r); 306 if (err == 0) { 307 if (cmd == SIOCDELRT) { struct fib_table *tb = fib_get_table(req.rtm.rtm_table); err = -ESRCH; 310 if (tb) err = tb->tb_delete(tb, &req.rtm, &rta, &req.nlh, NULL); 312 } else { struct fib_table *tb = fib_new_table(req.rtm.rtm_table); err = -ENOBUFS; 315 if (tb) err = tb->tb_insert(tb, &req.rtm, &rta, &req.nlh, NULL); } 318 if (rta.rta_mx) kfree(rta.rta_mx); } rtnl_unlock(); 322 return err; } 324 return -EINVAL; } #else int ip_rt_ioctl(unsigned int cmd, void *arg) { return -EINVAL; } #endif #ifdef CONFIG_RTNETLINK static int inet_check_attr(struct rtmsg *r, struct rtattr **rta) { int i; for (i=1; i<=RTA_MAX; i++) { struct rtattr *attr = rta[i-1]; if (attr) { if (RTA_PAYLOAD(attr) < 4) return -EINVAL; if (i != RTA_MULTIPATH && i != RTA_METRICS) rta[i-1] = (struct rtattr*)RTA_DATA(attr); } } return 0; } int inet_rtm_delroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) { struct fib_table * tb; struct rtattr **rta = arg; struct rtmsg *r = NLMSG_DATA(nlh); if (inet_check_attr(r, rta)) return -EINVAL; tb = fib_get_table(r->rtm_table); if (tb) return tb->tb_delete(tb, r, (struct kern_rta*)rta, nlh, &NETLINK_CB(skb)); return -ESRCH; } int inet_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) { struct fib_table * tb; struct rtattr **rta = arg; struct rtmsg *r = NLMSG_DATA(nlh); if (inet_check_attr(r, rta)) return -EINVAL; tb = fib_new_table(r->rtm_table); if (tb) return tb->tb_insert(tb, r, (struct kern_rta*)rta, nlh, &NETLINK_CB(skb)); return -ENOBUFS; } int inet_dump_fib(struct sk_buff *skb, struct netlink_callback *cb) { int t; int s_t; struct fib_table *tb; if (NLMSG_PAYLOAD(cb->nlh, 0) >= sizeof(struct rtmsg) && ((struct rtmsg*)NLMSG_DATA(cb->nlh))->rtm_flags&RTM_F_CLONED) return ip_rt_dump(skb, cb); s_t = cb->args[0]; if (s_t == 0) s_t = cb->args[0] = RT_TABLE_MIN; for (t=s_t; t<=RT_TABLE_MAX; t++) { if (t < s_t) continue; if (t > s_t) memset(&cb->args[1], 0, sizeof(cb->args)-sizeof(cb->args[0])); if ((tb = fib_get_table(t))==NULL) continue; if (tb->tb_dump(tb, skb, cb) < 0) break; } cb->args[0] = t; return skb->len; } #endif /* Prepare and feed intra-kernel routing request. Really, it should be netlink message, but :-( netlink can be not configured, so that we feed it directly to fib engine. It is legal, because all events occur only when netlink is already locked. */ 422 static void fib_magic(int cmd, int type, u32 dst, int dst_len, struct in_ifaddr *ifa) { struct fib_table * tb; struct { struct nlmsghdr nlh; struct rtmsg rtm; } req; struct kern_rta rta; memset(&req.rtm, 0, sizeof(req.rtm)); memset(&rta, 0, sizeof(rta)); 434 if (type == RTN_UNICAST) tb = fib_new_table(RT_TABLE_MAIN); 436 else tb = fib_new_table(RT_TABLE_LOCAL); 439 if (tb == NULL) 440 return; req.nlh.nlmsg_len = sizeof(req); req.nlh.nlmsg_type = cmd; req.nlh.nlmsg_flags = NLM_F_REQUEST|NLM_F_CREATE|NLM_F_APPEND; req.nlh.nlmsg_pid = 0; req.nlh.nlmsg_seq = 0; req.rtm.rtm_dst_len = dst_len; req.rtm.rtm_table = tb->tb_id; req.rtm.rtm_protocol = RTPROT_KERNEL; req.rtm.rtm_scope = (type != RTN_LOCAL ? RT_SCOPE_LINK : RT_SCOPE_HOST); req.rtm.rtm_type = type; rta.rta_dst = &dst; rta.rta_prefsrc = &ifa->ifa_local; rta.rta_oif = &ifa->ifa_dev->dev->ifindex; 458 if (cmd == RTM_NEWROUTE) tb->tb_insert(tb, &req.rtm, &rta, &req.nlh, NULL); 460 else tb->tb_delete(tb, &req.rtm, &rta, &req.nlh, NULL); } 464 static void fib_add_ifaddr(struct in_ifaddr *ifa) { struct in_device *in_dev = ifa->ifa_dev; struct net_device *dev = in_dev->dev; struct in_ifaddr *prim = ifa; u32 mask = ifa->ifa_mask; u32 addr = ifa->ifa_local; u32 prefix = ifa->ifa_address&mask; 473 if (ifa->ifa_flags&IFA_F_SECONDARY) { prim = inet_ifa_byprefix(in_dev, prefix, mask); 475 if (prim == NULL) { printk(KERN_DEBUG "fib_add_ifaddr: bug: prim == NULL\n"); 477 return; } } fib_magic(RTM_NEWROUTE, RTN_LOCAL, addr, 32, prim); 483 if (!(dev->flags&IFF_UP)) 484 return; /* Add broadcast address, if it is explicitly assigned. */ 487 if (ifa->ifa_broadcast && ifa->ifa_broadcast != 0xFFFFFFFF) fib_magic(RTM_NEWROUTE, RTN_BROADCAST, ifa->ifa_broadcast, 32, prim); if (!ZERONET(prefix) && !(ifa->ifa_flags&IFA_F_SECONDARY) && 491 (prefix != addr || ifa->ifa_prefixlen < 32)) { fib_magic(RTM_NEWROUTE, dev->flags&IFF_LOOPBACK ? RTN_LOCAL : RTN_UNICAST, prefix, ifa->ifa_prefixlen, prim); /* Add network specific broadcasts, when it takes a sense */ 496 if (ifa->ifa_prefixlen < 31) { fib_magic(RTM_NEWROUTE, RTN_BROADCAST, prefix, 32, prim); fib_magic(RTM_NEWROUTE, RTN_BROADCAST, prefix|~mask, 32, prim); } } } 503 static void fib_del_ifaddr(struct in_ifaddr *ifa) { struct in_device *in_dev = ifa->ifa_dev; struct net_device *dev = in_dev->dev; struct in_ifaddr *ifa1; struct in_ifaddr *prim = ifa; u32 brd = ifa->ifa_address|~ifa->ifa_mask; u32 any = ifa->ifa_address&ifa->ifa_mask; #define LOCAL_OK 1 #define BRD_OK 2 #define BRD0_OK 4 #define BRD1_OK 8 unsigned ok = 0; 517 if (!(ifa->ifa_flags&IFA_F_SECONDARY)) fib_magic(RTM_DELROUTE, dev->flags&IFF_LOOPBACK ? RTN_LOCAL : RTN_UNICAST, any, ifa->ifa_prefixlen, prim); 520 else { prim = inet_ifa_byprefix(in_dev, any, ifa->ifa_mask); 522 if (prim == NULL) { printk(KERN_DEBUG "fib_del_ifaddr: bug: prim == NULL\n"); 524 return; } } /* Deletion is more complicated than add. We should take care of not to delete too much :-) Scan address list to be sure that addresses are really gone. */ 534 for (ifa1 = in_dev->ifa_list; ifa1; ifa1 = ifa1->ifa_next) { 535 if (ifa->ifa_local == ifa1->ifa_local) ok |= LOCAL_OK; 537 if (ifa->ifa_broadcast == ifa1->ifa_broadcast) ok |= BRD_OK; 539 if (brd == ifa1->ifa_broadcast) ok |= BRD1_OK; 541 if (any == ifa1->ifa_broadcast) ok |= BRD0_OK; } 545 if (!(ok&BRD_OK)) fib_magic(RTM_DELROUTE, RTN_BROADCAST, ifa->ifa_broadcast, 32, prim); 547 if (!(ok&BRD1_OK)) fib_magic(RTM_DELROUTE, RTN_BROADCAST, brd, 32, prim); 549 if (!(ok&BRD0_OK)) fib_magic(RTM_DELROUTE, RTN_BROADCAST, any, 32, prim); 551 if (!(ok&LOCAL_OK)) { fib_magic(RTM_DELROUTE, RTN_LOCAL, ifa->ifa_local, 32, prim); /* Check, that this local address finally disappeared. */ 555 if (inet_addr_type(ifa->ifa_local) != RTN_LOCAL) { /* And the last, but not the least thing. We must flush stray FIB entries. First of all, we scan fib_info list searching for stray nexthop entries, then ignite fib_flush. */ 562 if (fib_sync_down(ifa->ifa_local, NULL, 0)) fib_flush(); } } #undef LOCAL_OK #undef BRD_OK #undef BRD0_OK #undef BRD1_OK } 572 static void fib_disable_ip(struct net_device *dev, int force) { 574 if (fib_sync_down(0, dev, force)) fib_flush(); rt_cache_flush(0); arp_ifdown(dev); } 580 static int fib_inetaddr_event(struct notifier_block *this, unsigned long event, void *ptr) { struct in_ifaddr *ifa = (struct in_ifaddr*)ptr; 584 switch (event) { 585 case NETDEV_UP: fib_add_ifaddr(ifa); rt_cache_flush(-1); 588 break; 589 case NETDEV_DOWN: 590 if (ifa->ifa_dev && ifa->ifa_dev->ifa_list == NULL) { /* Last address was deleted from this interface. Disable IP. */ fib_disable_ip(ifa->ifa_dev->dev, 1); 595 } else { fib_del_ifaddr(ifa); rt_cache_flush(-1); } 599 break; } 601 return NOTIFY_DONE; } 604 static int fib_netdev_event(struct notifier_block *this, unsigned long event, void *ptr) { struct net_device *dev = ptr; struct in_device *in_dev = __in_dev_get(dev); 609 if (!in_dev) 610 return NOTIFY_DONE; 612 switch (event) { 613 case NETDEV_UP: 614 for_ifa(in_dev) { fib_add_ifaddr(ifa); } endfor_ifa(in_dev); #ifdef CONFIG_IP_ROUTE_MULTIPATH fib_sync_up(dev); #endif rt_cache_flush(-1); 621 break; 622 case NETDEV_DOWN: fib_disable_ip(dev, 0); 624 break; 625 case NETDEV_UNREGISTER: fib_disable_ip(dev, 1); 627 break; 628 case NETDEV_CHANGEMTU: 629 case NETDEV_CHANGE: rt_cache_flush(0); 631 break; } 633 return NOTIFY_DONE; } struct notifier_block fib_inetaddr_notifier = { fib_inetaddr_event, NULL, 0 }; struct notifier_block fib_netdev_notifier = { fib_netdev_event, NULL, 0 }; 648 void __init ip_fib_init(void) { #ifdef CONFIG_PROC_FS proc_net_create("route",0,fib_get_procinfo); #endif /* CONFIG_PROC_FS */ #ifndef CONFIG_IP_MULTIPLE_TABLES local_table = fib_hash_init(RT_TABLE_LOCAL); main_table = fib_hash_init(RT_TABLE_MAIN); #else fib_rules_init(); #endif register_netdevice_notifier(&fib_netdev_notifier); register_inetaddr_notifier(&fib_inetaddr_notifier); }