interface-ip: replace prefix_assignment_cmp with avl_strcmp
[project/netifd.git] / interface-ip.c
index 51612fa..183935e 100644 (file)
@@ -1,3 +1,16 @@
+/*
+ * netifd - network interface daemon
+ * Copyright (C) 2012 Felix Fietkau <nbd@openwrt.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2
+ * as published by the Free Software Foundation
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
 #include <string.h>
 #include <stdlib.h>
 #include <stdio.h>
@@ -37,6 +50,145 @@ const struct config_param_list route_attr_list = {
        .params = route_attr,
 };
 
+
+struct list_head prefixes = LIST_HEAD_INIT(prefixes);
+static struct device_prefix *ula_prefix = NULL;
+
+
+static void
+clear_if_addr(union if_addr *a, int mask)
+{
+       int m_bytes = (mask + 7) / 8;
+       uint8_t m_clear = (1 << (m_bytes * 8 - mask)) - 1;
+       uint8_t *p = (uint8_t *) a;
+
+       if (m_bytes < sizeof(a))
+               memset(p + m_bytes, 0, sizeof(a) - m_bytes);
+
+       p[m_bytes - 1] &= ~m_clear;
+}
+
+static bool
+match_if_addr(union if_addr *a1, union if_addr *a2, int mask)
+{
+       union if_addr *p1, *p2;
+
+       p1 = alloca(sizeof(*a1));
+       p2 = alloca(sizeof(*a2));
+
+       memcpy(p1, a1, sizeof(*a1));
+       clear_if_addr(p1, mask);
+       memcpy(p2, a2, sizeof(*a2));
+       clear_if_addr(p2, mask);
+
+       return !memcmp(p1, p2, sizeof(*p1));
+}
+
+static bool
+__find_ip_addr_target(struct interface_ip_settings *ip, union if_addr *a, bool v6)
+{
+       struct device_addr *addr;
+
+       vlist_for_each_element(&ip->addr, addr, node) {
+               if (!addr->enabled)
+                       continue;
+
+               if (v6 != ((addr->flags & DEVADDR_FAMILY) == DEVADDR_INET6))
+                       continue;
+
+               if (!match_if_addr(&addr->addr, a, addr->mask))
+                       continue;
+
+               return true;
+       }
+
+       return false;
+}
+
+static void
+__find_ip_route_target(struct interface_ip_settings *ip, union if_addr *a,
+                      bool v6, struct device_route **res)
+{
+       struct device_route *route;
+
+       vlist_for_each_element(&ip->route, route, node) {
+               if (!route->enabled)
+                       continue;
+
+               if (v6 != ((route->flags & DEVADDR_FAMILY) == DEVADDR_INET6))
+                       continue;
+
+               if (!match_if_addr(&route->addr, a, route->mask))
+                       continue;
+
+               if (!*res || route->mask < (*res)->mask)
+                       *res = route;
+       }
+}
+
+static bool
+interface_ip_find_addr_target(struct interface *iface, union if_addr *a, bool v6)
+{
+       return __find_ip_addr_target(&iface->proto_ip, a, v6) ||
+              __find_ip_addr_target(&iface->config_ip, a, v6);
+}
+
+static void
+interface_ip_find_route_target(struct interface *iface, union if_addr *a,
+                              bool v6, struct device_route **route)
+{
+       __find_ip_route_target(&iface->proto_ip, a, v6, route);
+       __find_ip_route_target(&iface->config_ip, a, v6, route);
+}
+
+struct interface *
+interface_ip_add_target_route(union if_addr *addr, bool v6)
+{
+       struct interface *iface;
+       struct device_route *route, *r_next = NULL;
+       bool defaultroute_target = false;
+       int addrsize = v6 ? sizeof(addr->in6) : sizeof(addr->in);
+
+       route = calloc(1, sizeof(*route));
+       if (!route)
+               return NULL;
+
+       route->flags = v6 ? DEVADDR_INET6 : DEVADDR_INET4;
+       route->mask = v6 ? 128 : 32;
+       if (memcmp(&route->addr, addr, addrsize) == 0)
+               defaultroute_target = true;
+       else
+               memcpy(&route->addr, addr, addrsize);
+
+       vlist_for_each_element(&interfaces, iface, node) {
+               /* look for locally addressable target first */
+               if (interface_ip_find_addr_target(iface, addr, v6))
+                       goto done;
+
+               /* do not stop at the first route, let the lookup compare
+                * masks to find the best match */
+               interface_ip_find_route_target(iface, addr, v6, &r_next);
+       }
+
+       if (!r_next) {
+               free(route);
+               return NULL;
+       }
+
+       iface = r_next->iface;
+       memcpy(&route->nexthop, &r_next->nexthop, sizeof(route->nexthop));
+       route->mtu = r_next->mtu;
+       route->metric = r_next->metric;
+
+done:
+       route->iface = iface;
+       if (defaultroute_target)
+               free(route);
+       else
+               vlist_add(&iface->host_routes, &route->node, &route->flags);
+       return iface;
+}
+
 void
 interface_ip_add_route(struct interface *iface, struct blob_attr *attr, bool v6)
 {
@@ -64,6 +216,7 @@ interface_ip_add_route(struct interface *iface, struct blob_attr *attr, bool v6)
        if (!route)
                return;
 
+       route->flags = v6 ? DEVADDR_INET6 : DEVADDR_INET4;
        route->mask = v6 ? 128 : 32;
        if ((cur = tb[ROUTE_MASK]) != NULL) {
                route->mask = parse_netmask_string(blobmsg_data(cur), v6);
@@ -72,7 +225,7 @@ interface_ip_add_route(struct interface *iface, struct blob_attr *attr, bool v6)
        }
 
        if ((cur = tb[ROUTE_TARGET]) != NULL) {
-               if (!inet_pton(af, blobmsg_data(cur), &route->addr)) {
+               if (!parse_ip_and_netmask(af, blobmsg_data(cur), &route->addr, &route->mask)) {
                        DPRINTF("Failed to parse route target: %s\n", (char *) blobmsg_data(cur));
                        goto error;
                }
@@ -83,17 +236,19 @@ interface_ip_add_route(struct interface *iface, struct blob_attr *attr, bool v6)
                        DPRINTF("Failed to parse route gateway: %s\n", (char *) blobmsg_data(cur));
                        goto error;
                }
-       } else {
-               route->flags |= DEVADDR_DEVICE;
        }
 
-       if ((cur = tb[ROUTE_METRIC]) != NULL)
+       if ((cur = tb[ROUTE_METRIC]) != NULL) {
                route->metric = blobmsg_get_u32(cur);
+               route->flags |= DEVROUTE_METRIC;
+       }
 
-       if ((cur = tb[ROUTE_MTU]) != NULL)
+       if ((cur = tb[ROUTE_MTU]) != NULL) {
                route->mtu = blobmsg_get_u32(cur);
+               route->flags |= DEVROUTE_MTU;
+       }
 
-       vlist_add(&ip->route, &route->node, &route->mask);
+       vlist_add(&ip->route, &route->node, &route->flags);
        return;
 
 error:
@@ -104,14 +259,46 @@ static int
 addr_cmp(const void *k1, const void *k2, void *ptr)
 {
        return memcmp(k1, k2, sizeof(struct device_addr) -
-                     offsetof(struct device_addr, mask));
+                     offsetof(struct device_addr, flags));
 }
 
 static int
 route_cmp(const void *k1, const void *k2, void *ptr)
 {
        return memcmp(k1, k2, sizeof(struct device_route) -
-                     offsetof(struct device_route, mask));
+                     offsetof(struct device_route, flags));
+}
+
+static int
+prefix_cmp(const void *k1, const void *k2, void *ptr)
+{
+       return memcmp(k1, k2, sizeof(struct device_prefix) -
+                       offsetof(struct device_prefix, addr));
+}
+
+static void
+interface_handle_subnet_route(struct interface *iface, struct device_addr *addr, bool add)
+{
+       struct device *dev = iface->l3_dev.dev;
+       struct device_route route;
+
+       memset(&route, 0, sizeof(route));
+       route.iface = iface;
+       route.flags = addr->flags;
+       route.mask = addr->mask;
+       memcpy(&route.addr, &addr->addr, sizeof(route.addr));
+       clear_if_addr(&route.addr, route.mask);
+
+       if (add) {
+               route.flags |= DEVADDR_KERNEL;
+               system_del_route(dev, &route);
+
+               route.flags &= ~DEVADDR_KERNEL;
+               route.metric = iface->metric;
+               system_add_route(dev, &route);
+       } else {
+               system_del_route(dev, &route);
+       }
 }
 
 static void
@@ -127,7 +314,7 @@ interface_update_proto_addr(struct vlist_tree *tree,
 
        ip = container_of(tree, struct interface_ip_settings, addr);
        iface = ip->iface;
-       dev = iface->l3_dev->dev;
+       dev = iface->l3_dev.dev;
 
        if (node_new) {
                a_new = container_of(node_new, struct device_addr, node);
@@ -139,7 +326,7 @@ interface_update_proto_addr(struct vlist_tree *tree,
                        uint32_t *a = (uint32_t *) &a_new->addr;
 
                        mask >>= a_new->mask;
-                       a_new->broadcast = *a | mask;
+                       a_new->broadcast = *a | htonl(mask);
                }
        }
 
@@ -158,15 +345,20 @@ interface_update_proto_addr(struct vlist_tree *tree,
        }
 
        if (node_old) {
-               if (!(a_old->flags & DEVADDR_EXTERNAL) && a_old->enabled && !keep)
+               if (!(a_old->flags & DEVADDR_EXTERNAL) && a_old->enabled && !keep) {
+                       interface_handle_subnet_route(iface, a_old, false);
                        system_del_address(dev, a_old);
+               }
                free(a_old);
        }
 
        if (node_new) {
-               if (!(a_new->flags & DEVADDR_EXTERNAL) && !keep)
-                       system_add_address(dev, a_new);
                a_new->enabled = true;
+               if (!(a_new->flags & DEVADDR_EXTERNAL) && !keep) {
+                       system_add_address(dev, a_new);
+                       if (iface->metric)
+                               interface_handle_subnet_route(iface, a_new, true);
+               }
        }
 }
 
@@ -176,7 +368,7 @@ enable_route(struct interface_ip_settings *ip, struct device_route *route)
        if (ip->no_defaultroute && !route->mask)
                return false;
 
-       return true;
+       return ip->enabled;
 }
 
 static void
@@ -192,7 +384,7 @@ interface_update_proto_route(struct vlist_tree *tree,
 
        ip = container_of(tree, struct interface_ip_settings, route);
        iface = ip->iface;
-       dev = iface->l3_dev->dev;
+       dev = iface->l3_dev.dev;
 
        route_old = container_of(node_old, struct device_route, node);
        route_new = container_of(node_new, struct device_route, node);
@@ -209,19 +401,265 @@ interface_update_proto_route(struct vlist_tree *tree,
        if (node_new) {
                bool _enabled = enable_route(ip, route_new);
 
+               if (!(route_new->flags & DEVROUTE_METRIC))
+                       route_new->metric = iface->metric;
+
                if (!(route_new->flags & DEVADDR_EXTERNAL) && !keep && _enabled)
                        system_add_route(dev, route_new);
 
+               route_new->iface = iface;
                route_new->enabled = _enabled;
        }
 }
 
+static void
+interface_update_host_route(struct vlist_tree *tree,
+                            struct vlist_node *node_new,
+                            struct vlist_node *node_old)
+{
+       struct interface *iface;
+       struct device *dev;
+       struct device_route *route_old, *route_new;
+
+       iface = container_of(tree, struct interface, host_routes);
+       dev = iface->l3_dev.dev;
+
+       route_old = container_of(node_old, struct device_route, node);
+       route_new = container_of(node_new, struct device_route, node);
+
+       if (node_old) {
+               system_del_route(dev, route_old);
+               free(route_old);
+       }
+
+       if (node_new)
+               system_add_route(dev, route_new);
+}
+
+
+static void
+interface_set_prefix_address(struct interface *iface, bool add,
+               struct device_prefix_assignment *assignment)
+{
+       struct interface *uplink = assignment->prefix->iface;
+       if (!iface->l3_dev.dev)
+               return;
+
+       struct device *l3_downlink = iface->l3_dev.dev;
+
+       struct device_addr addr;
+       memset(&addr, 0, sizeof(addr));
+       addr.addr.in6 = assignment->addr;
+       addr.mask = assignment->length;
+       addr.flags = DEVADDR_INET6;
+       addr.preferred_until = assignment->prefix->preferred_until;
+       addr.valid_until = assignment->prefix->valid_until;
+
+       if (!add) {
+               if (assignment->enabled)
+                       system_del_address(l3_downlink, &addr);
+       } else {
+               system_add_address(l3_downlink, &addr);
+
+               if (uplink && uplink->l3_dev.dev) {
+                       int mtu = system_update_ipv6_mtu(
+                                       uplink->l3_dev.dev, 0);
+                       if (mtu > 0)
+                               system_update_ipv6_mtu(l3_downlink, mtu);
+               }
+       }
+       assignment->enabled = add;
+}
+
+
+static void
+interface_update_prefix_assignments(struct vlist_tree *tree,
+                            struct vlist_node *node_new,
+                            struct vlist_node *node_old)
+{
+       struct device_prefix_assignment *old, *new;
+       old = container_of(node_old, struct device_prefix_assignment, node);
+       new = container_of(node_new, struct device_prefix_assignment, node);
+
+       // Assignments persist across interface reloads etc.
+       // so use indirection to avoid dangling pointers
+       struct interface *iface = vlist_find(&interfaces,
+                       (node_new) ? new->name : old->name, iface, node);
+
+       if (node_old && node_new) {
+               new->addr = old->addr;
+               new->length = old->length;
+       } else if (node_old) {
+               if (iface)
+                       interface_set_prefix_address(iface, false, old);
+               free(old->name);
+               free(old);
+       } else if (node_new) {
+               struct device_prefix *prefix = new->prefix;
+               uint64_t want = 1ULL << (64 - new->length);
+               prefix->avail &= ~(want - 1);
+               prefix->avail -= want;
+
+               // Invert assignment
+               uint64_t assigned = ~prefix->avail;
+               assigned &= (1ULL << (64 - prefix->length)) - 1;
+               assigned &= ~(want - 1);
+
+               // Assignment
+               new->addr = prefix->addr;
+               new->addr.s6_addr32[0] |=
+                               htonl(assigned >> 32);
+               new->addr.s6_addr32[1] |=
+                               htonl(assigned & 0xffffffffU);
+               new->addr.s6_addr[15] += 1;
+       }
+
+       if (node_new && (iface->state == IFS_UP || iface->state == IFS_SETUP))
+               interface_set_prefix_address(iface, true, new);
+}
+
+
+void
+interface_ip_set_prefix_assignment(struct device_prefix *prefix,
+               struct interface *iface, uint8_t length)
+{
+       if (!length || length > 64) {
+               struct device_prefix_assignment *assignment = vlist_find(
+                               prefix->assignments, &iface, assignment, node);
+               if (assignment)
+                       interface_set_prefix_address(iface, false, assignment);
+       } else {
+               uint8_t length = iface->proto_ip.assignment_length;
+               uint64_t want = 1ULL << (64 - length);
+               if (prefix->avail < want && prefix->avail > 0) {
+                       do {
+                               want = 1ULL << (64 - ++length);
+                       } while (want > prefix->avail);
+               }
+
+               if (prefix->avail < want)
+                       return;
+
+               // Assignment
+               struct device_prefix_assignment *assignment = calloc(1, sizeof(*assignment));
+               assignment->prefix = prefix;
+               assignment->length = length;
+               assignment->name = strdup(iface->name);
+
+               vlist_add(prefix->assignments, &assignment->node, assignment->name);
+       }
+}
+
+static void
+interface_update_prefix(struct vlist_tree *tree,
+                            struct vlist_node *node_new,
+                            struct vlist_node *node_old)
+{
+       struct device_prefix *prefix_old, *prefix_new;
+       prefix_old = container_of(node_old, struct device_prefix, node);
+       prefix_new = container_of(node_new, struct device_prefix, node);
+
+       struct device_route route;
+       memset(&route, 0, sizeof(route));
+       route.flags = DEVADDR_INET6;
+       route.metric = INT32_MAX;
+       route.mask = (node_new) ? prefix_new->length : prefix_old->length;
+       route.addr.in6 = (node_new) ? prefix_new->addr : prefix_old->addr;
+
+       if (node_old && node_new) {
+               prefix_new->avail = prefix_old->avail;
+               prefix_new->assignments = prefix_old->assignments;
+               prefix_old->assignments = NULL;
+
+               // Update all assignments
+               struct device_prefix_assignment *assignment;
+               struct vlist_tree *assignments = prefix_new->assignments;
+               vlist_for_each_element(assignments, assignment, node)
+                       assignments->update(assignments,
+                                       &assignment->node, &assignment->node);
+       } else if (node_new) {
+               prefix_new->avail = 1ULL << (64 - prefix_new->length);
+               prefix_new->assignments = calloc(1, sizeof(*prefix_new->assignments));
+               vlist_init(prefix_new->assignments, avl_strcmp,
+                               interface_update_prefix_assignments);
+
+               // Create initial assignments for interfaces
+               struct interface *iface;
+               vlist_for_each_element(&interfaces, iface, node)
+                       interface_ip_set_prefix_assignment(prefix_new, iface,
+                                       iface->proto_ip.assignment_length);
+
+               list_add(&prefix_new->head, &prefixes);
+
+               // Set null-route to avoid routing loops
+               system_add_route(NULL, &route);
+       }
+
+       if (node_old) {
+               // Remove null-route
+               system_del_route(NULL, &route);
+
+               list_del(&prefix_old->head);
+
+               if (prefix_old->assignments) {
+                       vlist_flush_all(prefix_old->assignments);
+                       free(prefix_old->assignments);
+               }
+               free(prefix_old);
+       }
+}
+
+void
+interface_ip_add_device_prefix(struct interface *iface, struct in6_addr *addr,
+               uint8_t length, time_t valid_until, time_t preferred_until)
+{
+       struct device_prefix *prefix = calloc(1, sizeof(*prefix));
+       prefix->length = length;
+       prefix->addr = *addr;
+       prefix->preferred_until = preferred_until;
+       prefix->valid_until = valid_until;
+       prefix->iface = iface;
+
+       if (iface)
+               vlist_add(&iface->proto_ip.prefix, &prefix->node, &prefix->addr);
+       else
+               interface_update_prefix(NULL, &prefix->node, NULL);
+}
+
+void
+interface_ip_set_ula_prefix(const char *prefix)
+{
+       char buf[INET6_ADDRSTRLEN + 4] = {0}, *saveptr;
+       strncpy(buf, prefix, sizeof(buf) - 1);
+       char *prefixaddr = strtok_r(buf, "/", &saveptr);
+
+       struct in6_addr addr;
+       if (!prefixaddr || inet_pton(AF_INET6, prefixaddr, &addr) < 1)
+               return;
+
+       int length;
+       char *prefixlen = strtok_r(NULL, ",", &saveptr);
+       if (!prefixlen || (length = atoi(prefixlen)) < 1 || length > 64)
+               return;
+
+       if (ula_prefix && (!IN6_ARE_ADDR_EQUAL(&addr, &ula_prefix->addr) ||
+                       ula_prefix->length != length)) {
+               interface_update_prefix(NULL, NULL, &ula_prefix->node);
+               ula_prefix = NULL;
+       }
+
+       interface_ip_add_device_prefix(NULL, &addr, length, 0, 0);
+}
+
 void
 interface_add_dns_server(struct interface_ip_settings *ip, const char *str)
 {
        struct dns_server *s;
 
        s = calloc(1, sizeof(*s));
+       if (!s)
+               return;
+
        s->af = AF_INET;
        if (inet_pton(s->af, str, &s->addr.in))
                goto add;
@@ -336,7 +774,8 @@ interface_write_resolv_conf(void)
 
                fprintf(f, "# Interface %s\n", iface->name);
                write_resolv_conf_entries(f, &iface->config_ip);
-               write_resolv_conf_entries(f, &iface->proto_ip);
+               if (!iface->proto_ip.no_dns)
+                       write_resolv_conf_entries(f, &iface->proto_ip);
        }
        fclose(f);
        if (rename(path, resolv_conf) < 0) {
@@ -352,7 +791,7 @@ void interface_ip_set_enabled(struct interface_ip_settings *ip, bool enabled)
        struct device *dev;
 
        ip->enabled = enabled;
-       dev = ip->iface->l3_dev->dev;
+       dev = ip->iface->l3_dev.dev;
        if (!dev)
                return;
 
@@ -376,9 +815,12 @@ void interface_ip_set_enabled(struct interface_ip_settings *ip, bool enabled)
                if (route->enabled == _enabled)
                        continue;
 
-               if (_enabled)
+               if (_enabled) {
+                       if (!(route->flags & DEVROUTE_METRIC))
+                               route->metric = ip->iface->metric;
+
                        system_add_route(dev, route);
-               else
+               else
                        system_del_route(dev, route);
                route->enabled = _enabled;
        }
@@ -387,10 +829,13 @@ void interface_ip_set_enabled(struct interface_ip_settings *ip, bool enabled)
 void
 interface_ip_update_start(struct interface_ip_settings *ip)
 {
-       vlist_simple_update(&ip->dns_servers);
-       vlist_simple_update(&ip->dns_search);
+       if (ip != &ip->iface->config_ip) {
+               vlist_simple_update(&ip->dns_servers);
+               vlist_simple_update(&ip->dns_search);
+       }
        vlist_update(&ip->route);
        vlist_update(&ip->addr);
+       vlist_update(&ip->prefix);
 }
 
 void
@@ -400,19 +845,24 @@ interface_ip_update_complete(struct interface_ip_settings *ip)
        vlist_simple_flush(&ip->dns_search);
        vlist_flush(&ip->route);
        vlist_flush(&ip->addr);
+       vlist_flush(&ip->prefix);
+       interface_write_resolv_conf();
 }
 
 void
 interface_ip_flush(struct interface_ip_settings *ip)
 {
+       if (ip == &ip->iface->proto_ip)
+               vlist_flush_all(&ip->iface->host_routes);
        vlist_simple_flush_all(&ip->dns_servers);
        vlist_simple_flush_all(&ip->dns_search);
        vlist_flush_all(&ip->route);
        vlist_flush_all(&ip->addr);
+       vlist_flush_all(&ip->prefix);
 }
 
-void
-interface_ip_init(struct interface_ip_settings *ip, struct interface *iface)
+static void
+__interface_ip_init(struct interface_ip_settings *ip, struct interface *iface)
 {
        ip->iface = iface;
        ip->enabled = true;
@@ -420,4 +870,13 @@ interface_ip_init(struct interface_ip_settings *ip, struct interface *iface)
        vlist_simple_init(&ip->dns_servers, struct dns_server, node);
        vlist_init(&ip->route, route_cmp, interface_update_proto_route);
        vlist_init(&ip->addr, addr_cmp, interface_update_proto_addr);
+       vlist_init(&ip->prefix, prefix_cmp, interface_update_prefix);
+}
+
+void
+interface_ip_init(struct interface *iface)
+{
+       __interface_ip_init(&iface->proto_ip, iface);
+       __interface_ip_init(&iface->config_ip, iface);
+       vlist_init(&iface->host_routes, route_cmp, interface_update_host_route);
 }