diff options
Diffstat (limited to 'kernel/net/netfilter')
-rw-r--r-- | kernel/net/netfilter/ipset/Kconfig | 2 | ||||
-rw-r--r-- | kernel/net/netfilter/ipset/ip_set_bitmap_gen.h | 14 | ||||
-rw-r--r-- | kernel/net/netfilter/ipset/ip_set_core.c | 129 | ||||
-rw-r--r-- | kernel/net/netfilter/ipset/ip_set_hash_gen.h | 19 | ||||
-rw-r--r-- | kernel/net/netfilter/ipset/ip_set_hash_netiface.c | 10 | ||||
-rw-r--r-- | kernel/net/netfilter/ipset/ip_set_hash_netportnet.c | 1 | ||||
-rw-r--r-- | kernel/net/netfilter/ipset/ip_set_list_set.c | 13 |
7 files changed, 124 insertions, 64 deletions
diff --git a/kernel/net/netfilter/ipset/Kconfig b/kernel/net/netfilter/ipset/Kconfig index 861659f..8772af5 100644 --- a/kernel/net/netfilter/ipset/Kconfig +++ b/kernel/net/netfilter/ipset/Kconfig @@ -29,7 +29,7 @@ config IP_SET_BITMAP_IP depends on IP_SET help This option adds the bitmap:ip set type support, by which one - can store IPv4 addresses (or network addresse) from a range. + can store IPv4 addresses (or network addresses) from a range. To compile it as a module, choose M here. If unsure, say N. diff --git a/kernel/net/netfilter/ipset/ip_set_bitmap_gen.h b/kernel/net/netfilter/ipset/ip_set_bitmap_gen.h index 0479750..3245b6b 100644 --- a/kernel/net/netfilter/ipset/ip_set_bitmap_gen.h +++ b/kernel/net/netfilter/ipset/ip_set_bitmap_gen.h @@ -29,6 +29,7 @@ #define mtype_del IPSET_TOKEN(MTYPE, _del) #define mtype_list IPSET_TOKEN(MTYPE, _list) #define mtype_gc IPSET_TOKEN(MTYPE, _gc) +#define mtype_cancel_gc IPSET_TOKEN(MTYPE, _cancel_gc) #define mtype MTYPE #define get_ext(set, map, id) ((map)->extensions + ((set)->dsize * (id))) @@ -58,9 +59,6 @@ mtype_destroy(struct ip_set *set) { struct mtype *map = set->data; - if (SET_WITH_TIMEOUT(set)) - del_timer_sync(&map->gc); - if (set->dsize && set->extensions & IPSET_EXT_DESTROY) mtype_ext_cleanup(set); ip_set_free(map->members); @@ -290,6 +288,15 @@ mtype_gc(GC_ARG) add_timer(&map->gc); } +static void +mtype_cancel_gc(struct ip_set *set) +{ + struct mtype *map = set->data; + + if (SET_WITH_TIMEOUT(set)) + del_timer_sync(&map->gc); +} + static const struct ip_set_type_variant mtype = { .kadt = mtype_kadt, .uadt = mtype_uadt, @@ -303,6 +310,7 @@ static const struct ip_set_type_variant mtype = { .head = mtype_head, .list = mtype_list, .same_set = mtype_same_set, + .cancel_gc = mtype_cancel_gc, }; #endif /* __IP_SET_BITMAP_IP_GEN_H */ diff --git a/kernel/net/netfilter/ipset/ip_set_core.c b/kernel/net/netfilter/ipset/ip_set_core.c index a80b999..c31dbc3 100644 --- a/kernel/net/netfilter/ipset/ip_set_core.c +++ b/kernel/net/netfilter/ipset/ip_set_core.c @@ -30,7 +30,6 @@ static DEFINE_RWLOCK(ip_set_ref_lock); /* protects the set refs */ struct ip_set_net { struct ip_set * __rcu *ip_set_list; /* all individual sets */ ip_set_id_t ip_set_max; /* max number of sets */ - bool is_deleted; /* deleted by ip_set_net_exit */ bool is_destroyed; /* all sets are destroyed */ }; @@ -62,6 +61,8 @@ MODULE_ALIAS_NFNL_SUBSYS(NFNL_SUBSYS_IPSET); ip_set_dereference((inst)->ip_set_list)[id] #define ip_set_ref_netlink(inst,id) \ rcu_dereference_raw((inst)->ip_set_list)[id] +#define ip_set_dereference_nfnl(p) \ + rcu_dereference_check(p, lockdep_nfnl_is_held(NFNL_SUBSYS_IPSET)) /* The set types are implemented in modules and registered set types * can be found in ip_set_type_list. Adding/deleting types is @@ -684,6 +685,14 @@ __ip_set_put(struct ip_set *set) * a separate reference counter */ static void +__ip_set_get_netlink(struct ip_set *set) +{ + write_lock_bh(&ip_set_ref_lock); + set->ref_netlink++; + write_unlock_bh(&ip_set_ref_lock); +} + +static void __ip_set_put_netlink(struct ip_set *set) { write_lock_bh(&ip_set_ref_lock); @@ -701,15 +710,10 @@ __ip_set_put_netlink(struct ip_set *set) static struct ip_set * ip_set_rcu_get(struct net *net, ip_set_id_t index) { - struct ip_set *set; struct ip_set_net *inst = ip_set_pernet(net); - rcu_read_lock(); - /* ip_set_list itself needs to be protected */ - set = rcu_dereference(inst->ip_set_list)[index]; - rcu_read_unlock(); - - return set; + /* ip_set_list and the set pointer need to be protected */ + return ip_set_dereference_nfnl(inst->ip_set_list)[index]; } static inline void @@ -740,9 +744,7 @@ ip_set_test(ip_set_id_t index, const struct sk_buff *skb, !(opt->family == set->family || set->family == NFPROTO_UNSPEC)) return 0; - rcu_read_lock_bh(); ret = set->variant->kadt(set, skb, par, IPSET_TEST, opt); - rcu_read_unlock_bh(); if (ret == -EAGAIN) { /* Type requests element to be completed */ @@ -875,7 +877,7 @@ ip_set_name_byindex(struct net *net, ip_set_id_t index, char *name) BUG_ON(!set); read_lock_bh(&ip_set_ref_lock); - strncpy(name, set->name, IPSET_MAXNAMELEN); + strscpy_pad(name, set->name, IPSET_MAXNAMELEN); read_unlock_bh(&ip_set_ref_lock); } EXPORT_SYMBOL_GPL(ip_set_name_byindex); @@ -923,11 +925,9 @@ ip_set_nfnl_put(struct net *net, ip_set_id_t index) struct ip_set_net *inst = ip_set_pernet(net); nfnl_lock(NFNL_SUBSYS_IPSET); - if (!inst->is_deleted) { /* already deleted from ip_set_net_exit() */ - set = ip_set(inst, index); - if (set) - __ip_set_put(set); - } + set = ip_set(inst, index); + if (set) + __ip_set_put(set); nfnl_unlock(NFNL_SUBSYS_IPSET); } EXPORT_SYMBOL_GPL(ip_set_nfnl_put); @@ -1161,6 +1161,7 @@ IPSET_CBFN(ip_set_create, struct net *n, struct sock *ctnl, return ret; cleanup: + set->variant->cancel_gc(set); set->variant->destroy(set); put_out: module_put(set->type->me); @@ -1178,17 +1179,50 @@ ip_set_setname_policy[IPSET_ATTR_CMD_MAX + 1] = { .len = IPSET_MAXNAMELEN - 1 }, }; +/* Destroying a set is split into two stages when a DESTROY command issued: + * - Cancel garbage collectors and decrement the module reference counter: + * - Cancelling may wait and we are allowed to do it at this stage. + * - Module remove is protected by rcu_barrier() which waits for + * the second stage to be finished. + * - In order to prevent the race between kernel side add/del/test element + * operations and destroy, the destroying of the set data areas are + * performed via a call_rcu() call. + */ + +/* Call set variant specific destroy function and reclaim the set data. */ static void -ip_set_destroy_set(struct ip_set *set) +ip_set_destroy_set_variant(struct ip_set *set) { - pr_debug("set: %s\n", set->name); - /* Must call it without holding any lock */ set->variant->destroy(set); - module_put(set->type->me); kfree(set); } +static void +ip_set_destroy_set_variant_rcu(struct rcu_head *head) +{ + struct ip_set *set = container_of(head, struct ip_set, rcu); + + ip_set_destroy_set_variant(set); +} + +/* Cancel the garbage collectors and decrement module references */ +static void +ip_set_destroy_cancel_gc(struct ip_set *set) +{ + set->variant->cancel_gc(set); + module_put(set->type->me); +} + +/* Use when we may wait for the complete destroy to be finished. + */ +static void +ip_set_destroy_set(struct ip_set *set) +{ + ip_set_destroy_cancel_gc(set); + ip_set_destroy_set_variant(set); +} + static int IPSET_CBFN(ip_set_destroy, struct net *net, struct sock *ctnl, struct sk_buff *skb, const struct nlmsghdr *nlh, @@ -1204,9 +1238,6 @@ IPSET_CBFN(ip_set_destroy, struct net *net, struct sock *ctnl, if (unlikely(protocol_min_failed(attr))) return -IPSET_ERR_PROTOCOL; - /* Must wait for flush to be really finished in list:set */ - rcu_barrier(); - /* Commands are serialized and references are * protected by the ip_set_ref_lock. * External systems (i.e. xt_set) must call @@ -1217,8 +1248,10 @@ IPSET_CBFN(ip_set_destroy, struct net *net, struct sock *ctnl, * counter, so if it's already zero, we can proceed * without holding the lock. */ - read_lock_bh(&ip_set_ref_lock); if (!attr[IPSET_ATTR_SETNAME]) { + /* Must wait for flush to be really finished in list:set */ + rcu_barrier(); + read_lock_bh(&ip_set_ref_lock); for (i = 0; i < inst->ip_set_max; i++) { s = ip_set(inst, i); if (s && (s->ref || s->ref_netlink)) { @@ -1239,6 +1272,9 @@ IPSET_CBFN(ip_set_destroy, struct net *net, struct sock *ctnl, inst->is_destroyed = false; } else { u32 flags = flag_exist(INFO_NLH(info, nlh)); + u16 features = 0; + + read_lock_bh(&ip_set_ref_lock); s = find_set_and_id(inst, nla_data(attr[IPSET_ATTR_SETNAME]), &i); if (!s) { @@ -1249,10 +1285,16 @@ IPSET_CBFN(ip_set_destroy, struct net *net, struct sock *ctnl, ret = -IPSET_ERR_BUSY; goto out; } + features = s->type->features; ip_set(inst, i) = NULL; read_unlock_bh(&ip_set_ref_lock); - - ip_set_destroy_set(s); + if (features & IPSET_TYPE_NAME) { + /* Must wait for flush to be really finished */ + rcu_barrier(); + } + /* Must cancel garbage collectors */ + ip_set_destroy_cancel_gc(s); + call_rcu(&s->rcu, ip_set_destroy_set_variant_rcu); } return 0; out: @@ -1350,11 +1392,11 @@ IPSET_CBFN(ip_set_rename, struct net *net, struct sock *ctnl, goto out; } } - ret = strscpy(set->name, name2, IPSET_MAXNAMELEN); + strscpy_pad(set->name, name2, IPSET_MAXNAMELEN); out: write_unlock_bh(&ip_set_ref_lock); - return ret < 0 ? ret : 0; + return ret; } /* Swap two sets so that name/index points to the other. @@ -1408,9 +1450,9 @@ IPSET_CBFN(ip_set_swap, struct net *net, struct sock *ctnl, return -EBUSY; } - strncpy(from_name, from->name, IPSET_MAXNAMELEN); - strncpy(from->name, to->name, IPSET_MAXNAMELEN); - strncpy(to->name, from_name, IPSET_MAXNAMELEN); + strscpy_pad(from_name, from->name, IPSET_MAXNAMELEN); + strscpy_pad(from->name, to->name, IPSET_MAXNAMELEN); + strscpy_pad(to->name, from_name, IPSET_MAXNAMELEN); swap(from->ref, to->ref); ip_set(inst, from_id) = to; @@ -1750,6 +1792,14 @@ CALL_AD(struct net *net, struct sock *ctnl, struct sk_buff *skb, bool eexist = flags & IPSET_FLAG_EXIST, retried = false; do { + if (retried) { + __ip_set_get_netlink(set); + nfnl_unlock(NFNL_SUBSYS_IPSET); + cond_resched(); + nfnl_lock(NFNL_SUBSYS_IPSET); + __ip_set_put_netlink(set); + } + ip_set_lock(set); ret = set->variant->uadt(set, tb, adt, &lineno, flags, retried); ip_set_unlock(set); @@ -2431,7 +2481,6 @@ ip_set_net_init(struct net *net) #else goto err_alloc; #endif - inst->is_deleted = false; inst->is_destroyed = false; rcu_assign_pointer(inst->ip_set_list, list); return 0; @@ -2448,20 +2497,6 @@ ip_set_net_exit(struct net *net) { struct ip_set_net *inst = ip_set_pernet(net); - struct ip_set *set = NULL; - ip_set_id_t i; - - inst->is_deleted = true; /* flag for ip_set_nfnl_put */ - - nfnl_lock(NFNL_SUBSYS_IPSET); - for (i = 0; i < inst->ip_set_max; i++) { - set = ip_set(inst, i); - if (set) { - ip_set(inst, i) = NULL; - ip_set_destroy_set(set); - } - } - nfnl_unlock(NFNL_SUBSYS_IPSET); kvfree(rcu_dereference_protected(inst->ip_set_list, 1)); #ifndef HAVE_NET_OPS_ID kvfree(inst); @@ -2526,8 +2561,8 @@ ip_set_fini(void) { nf_unregister_sockopt(&so_set); nfnetlink_subsys_unregister(&ip_set_netlink_subsys); - UNREGISTER_PERNET_SUBSYS(&ip_set_net_ops); + pr_debug("these are the famous last words\n"); } diff --git a/kernel/net/netfilter/ipset/ip_set_hash_gen.h b/kernel/net/netfilter/ipset/ip_set_hash_gen.h index af38991..76c3894 100644 --- a/kernel/net/netfilter/ipset/ip_set_hash_gen.h +++ b/kernel/net/netfilter/ipset/ip_set_hash_gen.h @@ -222,6 +222,7 @@ static const union nf_inet_addr zeromask = {}; #undef mtype_gc_do #undef mtype_gc #undef mtype_gc_init +#undef mtype_cancel_gc #undef mtype_variant #undef mtype_data_match @@ -266,6 +267,7 @@ static const union nf_inet_addr zeromask = {}; #define mtype_gc_do IPSET_TOKEN(MTYPE, _gc_do) #define mtype_gc IPSET_TOKEN(MTYPE, _gc) #define mtype_gc_init IPSET_TOKEN(MTYPE, _gc_init) +#define mtype_cancel_gc IPSET_TOKEN(MTYPE, _cancel_gc) #define mtype_variant IPSET_TOKEN(MTYPE, _variant) #define mtype_data_match IPSET_TOKEN(MTYPE, _data_match) @@ -430,7 +432,7 @@ mtype_ahash_destroy(struct ip_set *set, struct htable *t, bool ext_destroy) u32 i; for (i = 0; i < jhash_size(t->htable_bits); i++) { - n = __ipset_dereference(hbucket(t, i)); + n = (__force struct hbucket *)hbucket(t, i); if (!n) continue; if (set->extensions & IPSET_EXT_DESTROY && ext_destroy) @@ -450,10 +452,7 @@ mtype_destroy(struct ip_set *set) struct htype *h = set->data; struct list_head *l, *lt; - if (SET_WITH_TIMEOUT(set)) - cancel_delayed_work_sync(&h->gc.dwork); - - mtype_ahash_destroy(set, ipset_dereference_nfnl(h->table), true); + mtype_ahash_destroy(set, (__force struct htable *)h->table, true); list_for_each_safe(l, lt, &h->ad) { list_del(l); kfree(l); @@ -598,6 +597,15 @@ mtype_gc_init(struct htable_gc *gc) queue_delayed_work(system_power_efficient_wq, &gc->dwork, HZ); } +static void +mtype_cancel_gc(struct ip_set *set) +{ + struct htype *h = set->data; + + if (SET_WITH_TIMEOUT(set)) + cancel_delayed_work_sync(&h->gc.dwork); +} + static int mtype_add(struct ip_set *set, void *value, const struct ip_set_ext *ext, struct ip_set_ext *mext, u32 flags); @@ -1441,6 +1449,7 @@ static const struct ip_set_type_variant mtype_variant = { .uref = mtype_uref, .resize = mtype_resize, .same_set = mtype_same_set, + .cancel_gc = mtype_cancel_gc, .region_lock = true, }; diff --git a/kernel/net/netfilter/ipset/ip_set_hash_netiface.c b/kernel/net/netfilter/ipset/ip_set_hash_netiface.c index af210c7..5baa852 100644 --- a/kernel/net/netfilter/ipset/ip_set_hash_netiface.c +++ b/kernel/net/netfilter/ipset/ip_set_hash_netiface.c @@ -41,7 +41,7 @@ MODULE_ALIAS("ip_set_hash:net,iface"); #define IP_SET_HASH_WITH_MULTI #define IP_SET_HASH_WITH_NET0 -#define STRLCPY(a, b) strlcpy(a, b, IFNAMSIZ) +#define STRSCPY(a, b) strscpy(a, b, IFNAMSIZ) /* IPv4 variant */ @@ -183,11 +183,11 @@ hash_netiface4_kadt(struct ip_set *set, const struct sk_buff *skb, if (!eiface) return -EINVAL; - STRLCPY(e.iface, eiface); + STRSCPY(e.iface, eiface); e.physdev = 1; #endif } else { - STRLCPY(e.iface, SRCDIR ? IFACE(in) : IFACE(out)); + STRSCPY(e.iface, SRCDIR ? IFACE(in) : IFACE(out)); } if (strlen(e.iface) == 0) @@ -401,11 +401,11 @@ hash_netiface6_kadt(struct ip_set *set, const struct sk_buff *skb, if (!eiface) return -EINVAL; - STRLCPY(e.iface, eiface); + STRSCPY(e.iface, eiface); e.physdev = 1; #endif } else { - STRLCPY(e.iface, SRCDIR ? IFACE(in) : IFACE(out)); + STRSCPY(e.iface, SRCDIR ? IFACE(in) : IFACE(out)); } if (strlen(e.iface) == 0) diff --git a/kernel/net/netfilter/ipset/ip_set_hash_netportnet.c b/kernel/net/netfilter/ipset/ip_set_hash_netportnet.c index fd080ab..d5774a1 100644 --- a/kernel/net/netfilter/ipset/ip_set_hash_netportnet.c +++ b/kernel/net/netfilter/ipset/ip_set_hash_netportnet.c @@ -37,6 +37,7 @@ MODULE_ALIAS("ip_set_hash:net,port,net"); #define IP_SET_HASH_WITH_PROTO #define IP_SET_HASH_WITH_NETS #define IPSET_NET_COUNT 2 +#define IP_SET_HASH_WITH_NET0 /* IPv4 variant */ diff --git a/kernel/net/netfilter/ipset/ip_set_list_set.c b/kernel/net/netfilter/ipset/ip_set_list_set.c index 8c7fef8..cc2e5b9 100644 --- a/kernel/net/netfilter/ipset/ip_set_list_set.c +++ b/kernel/net/netfilter/ipset/ip_set_list_set.c @@ -429,9 +429,6 @@ list_set_destroy(struct ip_set *set) struct list_set *map = set->data; struct set_elem *e, *n; - if (SET_WITH_TIMEOUT(set)) - del_timer_sync(&map->gc); - list_for_each_entry_safe(e, n, &map->members, list) { list_del(&e->list); ip_set_put_byindex(map->net, e->id); @@ -548,6 +545,15 @@ list_set_same_set(const struct ip_set *a, const struct ip_set *b) a->extensions == b->extensions; } +static void +list_set_cancel_gc(struct ip_set *set) +{ + struct list_set *map = set->data; + + if (SET_WITH_TIMEOUT(set)) + timer_shutdown_sync(&map->gc); +} + static const struct ip_set_type_variant set_variant = { .kadt = list_set_kadt, .uadt = list_set_uadt, @@ -561,6 +567,7 @@ static const struct ip_set_type_variant set_variant = { .head = list_set_head, .list = list_set_list, .same_set = list_set_same_set, + .cancel_gc = list_set_cancel_gc, }; static void |