diff options
Diffstat (limited to 'kernel/net/netfilter/ipset/ip_set_core.c')
| -rw-r--r-- | kernel/net/netfilter/ipset/ip_set_core.c | 147 |
1 files changed, 92 insertions, 55 deletions
diff --git a/kernel/net/netfilter/ipset/ip_set_core.c b/kernel/net/netfilter/ipset/ip_set_core.c index 0fdafb7..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 @@ -354,7 +355,7 @@ ip_set_init_comment(struct ip_set *set, struct ip_set_comment *comment, c = kmalloc(sizeof(*c) + len + 1, GFP_ATOMIC); if (unlikely(!c)) return; - strlcpy(c->str, ext->comment, len + 1); + strscpy(c->str, ext->comment, len + 1); set->ext_size += sizeof(*c) + strlen(c->str) + 1; rcu_assign_pointer(comment->c, c); } @@ -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); @@ -1082,7 +1082,7 @@ IPSET_CBFN(ip_set_create, struct net *n, struct sock *ctnl, if (!set) return -ENOMEM; spin_lock_init(&set->lock); - strlcpy(set->name, name, IPSET_MAXNAMELEN); + strscpy(set->name, name, IPSET_MAXNAMELEN); set->family = family; set->revision = revision; @@ -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; @@ -1750,13 +1792,22 @@ 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); retried = true; - } while (ret == -EAGAIN && - set->variant->resize && - (ret = set->variant->resize(set, retried)) == 0); + } while (ret == -ERANGE || + (ret == -EAGAIN && + set->variant->resize && + (ret = set->variant->resize(set, retried)) == 0)); if (!ret || (ret == -IPSET_ERR_EXIST && eexist)) return 0; @@ -1775,11 +1826,12 @@ CALL_AD(struct net *net, struct sock *ctnl, struct sk_buff *skb, skb2 = nlmsg_new(payload, GFP_KERNEL); if (!skb2) return -ENOMEM; - rep = __nlmsg_put(skb2, NETLINK_PORTID(skb), - nlh->nlmsg_seq, NLMSG_ERROR, payload, 0); + rep = nlmsg_put(skb2, NETLINK_PORTID(skb), + nlh->nlmsg_seq, NLMSG_ERROR, payload, 0); errmsg = nlmsg_data(rep); errmsg->error = ret; - memcpy(&errmsg->msg, nlh, nlh->nlmsg_len); + unsafe_memcpy(&errmsg->msg, nlh, nlh->nlmsg_len, + /* Bounds checked by the skb layer. */); cmdattr = (void *)&errmsg->msg + min_len; ret = NLA_PARSE(cda, IPSET_ATTR_CMD_MAX, cmdattr, @@ -2429,7 +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; @@ -2446,20 +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); @@ -2524,8 +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"); } |