1 files changed, 143 insertions, 59 deletions

diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index c26a5663795e..7e63b481cc86 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -22,6 +22,8 @@
 #include <net/net_namespace.h>
 #include <net/sock.h>
 
+#define NFT_MODULE_AUTOLOAD_LIMIT (MODULE_NAME_LEN - sizeof("nft-expr-255-"))
+
 static LIST_HEAD(nf_tables_expressions);
 static LIST_HEAD(nf_tables_objects);
 static LIST_HEAD(nf_tables_flowtables);
@@ -551,46 +553,70 @@ static inline u64 nf_tables_alloc_handle(struct nft_table *table)
 static const struct nft_chain_type *chain_type[NFPROTO_NUMPROTO][NFT_CHAIN_T_MAX];
 
 static const struct nft_chain_type *
+__nft_chain_type_get(u8 family, enum nft_chain_types type)
+{
+	if (family >= NFPROTO_NUMPROTO ||
+	    type >= NFT_CHAIN_T_MAX)
+		return NULL;
+
+	return chain_type[family][type];
+}
+
+static const struct nft_chain_type *
 __nf_tables_chain_type_lookup(const struct nlattr *nla, u8 family)
 {
+	const struct nft_chain_type *type;
 	int i;
 
 	for (i = 0; i < NFT_CHAIN_T_MAX; i++) {
-		if (chain_type[family][i] != NULL &&
-		    !nla_strcmp(nla, chain_type[family][i]->name))
-			return chain_type[family][i];
+		type = __nft_chain_type_get(family, i);
+		if (!type)
+			continue;
+		if (!nla_strcmp(nla, type->name))
+			return type;
 	}
 	return NULL;
 }
 
-/*
- * Loading a module requires dropping mutex that guards the
- * transaction.
- * We first need to abort any pending transactions as once
- * mutex is unlocked a different client could start a new
- * transaction.  It must not see any 'future generation'
- * changes * as these changes will never happen.
- */
-#ifdef CONFIG_MODULES
-static int __nf_tables_abort(struct net *net);
+struct nft_module_request {
+	struct list_head	list;
+	char			module[MODULE_NAME_LEN];
+	bool			done;
+};
 
-static void nft_request_module(struct net *net, const char *fmt, ...)
+#ifdef CONFIG_MODULES
+static int nft_request_module(struct net *net, const char *fmt, ...)
 {
 	char module_name[MODULE_NAME_LEN];
+	struct nft_module_request *req;
 	va_list args;
 	int ret;
 
-	__nf_tables_abort(net);
-
 	va_start(args, fmt);
 	ret = vsnprintf(module_name, MODULE_NAME_LEN, fmt, args);
 	va_end(args);
-	if (WARN(ret >= MODULE_NAME_LEN, "truncated: '%s' (len %d)", module_name, ret))
-		return;
+	if (ret >= MODULE_NAME_LEN)
+		return 0;
 
-	mutex_unlock(&net->nft.commit_mutex);
-	request_module("%s", module_name);
-	mutex_lock(&net->nft.commit_mutex);
+	list_for_each_entry(req, &net->nft.module_list, list) {
+		if (!strcmp(req->module, module_name)) {
+			if (req->done)
+				return 0;
+
+			/* A request to load this module already exists. */
+			return -EAGAIN;
+		}
+	}
+
+	req = kmalloc(sizeof(*req), GFP_KERNEL);
+	if (!req)
+		return -ENOMEM;
+
+	req->done = false;
+	strlcpy(req->module, module_name, MODULE_NAME_LEN);
+	list_add_tail(&req->list, &net->nft.module_list);
+
+	return -EAGAIN;
 }
 #endif
 
@@ -614,10 +640,9 @@ nf_tables_chain_type_lookup(struct net *net, const struct nlattr *nla,
 	lockdep_nfnl_nft_mutex_not_held();
 #ifdef CONFIG_MODULES
 	if (autoload) {
-		nft_request_module(net, "nft-chain-%u-%.*s", family,
-				   nla_len(nla), (const char *)nla_data(nla));
-		type = __nf_tables_chain_type_lookup(nla, family);
-		if (type != NULL)
+		if (nft_request_module(net, "nft-chain-%u-%.*s", family,
+				       nla_len(nla),
+				       (const char *)nla_data(nla)) == -EAGAIN)
 			return ERR_PTR(-EAGAIN);
 	}
 #endif
@@ -1045,12 +1070,18 @@ static int nft_flush_table(struct nft_ctx *ctx)
 	}
 
 	list_for_each_entry_safe(flowtable, nft, &ctx->table->flowtables, list) {
+		if (!nft_is_active_next(ctx->net, flowtable))
+			continue;
+
 		err = nft_delflowtable(ctx, flowtable);
 		if (err < 0)
 			goto out;
 	}
 
 	list_for_each_entry_safe(obj, ne, &ctx->table->objects, list) {
+		if (!nft_is_active_next(ctx->net, obj))
+			continue;
+
 		err = nft_delobj(ctx, obj);
 		if (err < 0)
 			goto out;
@@ -1153,11 +1184,8 @@ static void nf_tables_table_destroy(struct nft_ctx *ctx)
 
 void nft_register_chain_type(const struct nft_chain_type *ctype)
 {
-	if (WARN_ON(ctype->family >= NFPROTO_NUMPROTO))
-		return;
-
 	nfnl_lock(NFNL_SUBSYS_NFTABLES);
-	if (WARN_ON(chain_type[ctype->family][ctype->type] != NULL)) {
+	if (WARN_ON(__nft_chain_type_get(ctype->family, ctype->type))) {
 		nfnl_unlock(NFNL_SUBSYS_NFTABLES);
 		return;
 	}
@@ -1241,7 +1269,8 @@ static const struct nla_policy nft_chain_policy[NFTA_CHAIN_MAX + 1] = {
 				    .len = NFT_CHAIN_MAXNAMELEN - 1 },
 	[NFTA_CHAIN_HOOK]	= { .type = NLA_NESTED },
 	[NFTA_CHAIN_POLICY]	= { .type = NLA_U32 },
-	[NFTA_CHAIN_TYPE]	= { .type = NLA_STRING },
+	[NFTA_CHAIN_TYPE]	= { .type = NLA_STRING,
+				    .len = NFT_MODULE_AUTOLOAD_LIMIT },
 	[NFTA_CHAIN_COUNTERS]	= { .type = NLA_NESTED },
 	[NFTA_CHAIN_FLAGS]	= { .type = NLA_U32 },
 };
@@ -1676,6 +1705,7 @@ static int nf_tables_parse_netdev_hooks(struct net *net,
 			goto err_hook;
 		}
 		if (nft_hook_list_find(hook_list, hook)) {
+			kfree(hook);
 			err = -EEXIST;
 			goto err_hook;
 		}
@@ -1757,7 +1787,10 @@ static int nft_chain_parse_hook(struct net *net,
 	hook->num = ntohl(nla_get_be32(ha[NFTA_HOOK_HOOKNUM]));
 	hook->priority = ntohl(nla_get_be32(ha[NFTA_HOOK_PRIORITY]));
 
-	type = chain_type[family][NFT_CHAIN_T_DEFAULT];
+	type = __nft_chain_type_get(family, NFT_CHAIN_T_DEFAULT);
+	if (!type)
+		return -EOPNOTSUPP;
+
 	if (nla[NFTA_CHAIN_TYPE]) {
 		type = nf_tables_chain_type_lookup(net, nla[NFTA_CHAIN_TYPE],
 						   family, autoload);
@@ -2317,9 +2350,8 @@ static const struct nft_expr_type *__nft_expr_type_get(u8 family,
 static int nft_expr_type_request_module(struct net *net, u8 family,
 					struct nlattr *nla)
 {
-	nft_request_module(net, "nft-expr-%u-%.*s", family,
-			   nla_len(nla), (char *)nla_data(nla));
-	if (__nft_expr_type_get(family, nla))
+	if (nft_request_module(net, "nft-expr-%u-%.*s", family,
+			       nla_len(nla), (char *)nla_data(nla)) == -EAGAIN)
 		return -EAGAIN;
 
 	return 0;
@@ -2345,9 +2377,9 @@ static const struct nft_expr_type *nft_expr_type_get(struct net *net,
 		if (nft_expr_type_request_module(net, family, nla) == -EAGAIN)
 			return ERR_PTR(-EAGAIN);
 
-		nft_request_module(net, "nft-expr-%.*s",
-				   nla_len(nla), (char *)nla_data(nla));
-		if (__nft_expr_type_get(family, nla))
+		if (nft_request_module(net, "nft-expr-%.*s",
+				       nla_len(nla),
+				       (char *)nla_data(nla)) == -EAGAIN)
 			return ERR_PTR(-EAGAIN);
 	}
 #endif
@@ -2355,7 +2387,8 @@ static const struct nft_expr_type *nft_expr_type_get(struct net *net,
 }
 
 static const struct nla_policy nft_expr_policy[NFTA_EXPR_MAX + 1] = {
-	[NFTA_EXPR_NAME]	= { .type = NLA_STRING },
+	[NFTA_EXPR_NAME]	= { .type = NLA_STRING,
+				    .len = NFT_MODULE_AUTOLOAD_LIMIT },
 	[NFTA_EXPR_DATA]	= { .type = NLA_NESTED },
 };
 
@@ -2437,9 +2470,10 @@ static int nf_tables_expr_parse(const struct nft_ctx *ctx,
 			err = PTR_ERR(ops);
 #ifdef CONFIG_MODULES
 			if (err == -EAGAIN)
-				nft_expr_type_request_module(ctx->net,
-							     ctx->family,
-							     tb[NFTA_EXPR_NAME]);
+				if (nft_expr_type_request_module(ctx->net,
+								 ctx->family,
+								 tb[NFTA_EXPR_NAME]) != -EAGAIN)
+					err = -ENOENT;
 #endif
 			goto err1;
 		}
@@ -3276,8 +3310,7 @@ nft_select_set_ops(const struct nft_ctx *ctx,
 	lockdep_nfnl_nft_mutex_not_held();
 #ifdef CONFIG_MODULES
 	if (list_empty(&nf_tables_set_types)) {
-		nft_request_module(ctx->net, "nft-set");
-		if (!list_empty(&nf_tables_set_types))
+		if (nft_request_module(ctx->net, "nft-set") == -EAGAIN)
 			return ERR_PTR(-EAGAIN);
 	}
 #endif
@@ -4198,7 +4231,8 @@ static const struct nla_policy nft_set_elem_policy[NFTA_SET_ELEM_MAX + 1] = {
 	[NFTA_SET_ELEM_USERDATA]	= { .type = NLA_BINARY,
 					    .len = NFT_USERDATA_MAXLEN },
 	[NFTA_SET_ELEM_EXPR]		= { .type = NLA_NESTED },
-	[NFTA_SET_ELEM_OBJREF]		= { .type = NLA_STRING },
+	[NFTA_SET_ELEM_OBJREF]		= { .type = NLA_STRING,
+					    .len = NFT_OBJ_MAXNAMELEN - 1 },
 };
 
 static const struct nla_policy nft_set_elem_list_policy[NFTA_SET_ELEM_LIST_MAX + 1] = {
@@ -4519,8 +4553,10 @@ static int nft_get_set_elem(struct nft_ctx *ctx, struct nft_set *set,
 		return err;
 
 	err = -EINVAL;
-	if (desc.type != NFT_DATA_VALUE || desc.len != set->klen)
+	if (desc.type != NFT_DATA_VALUE || desc.len != set->klen) {
+		nft_data_release(&elem.key.val, desc.type);
 		return err;
+	}
 
 	priv = set->ops->get(ctx->net, set, &elem, flags);
 	if (IS_ERR(priv))
@@ -4756,14 +4792,20 @@ static int nft_add_set_elem(struct nft_ctx *ctx, struct nft_set *set,
 		if (nla[NFTA_SET_ELEM_DATA] == NULL &&
 		    !(flags & NFT_SET_ELEM_INTERVAL_END))
 			return -EINVAL;
-		if (nla[NFTA_SET_ELEM_DATA] != NULL &&
-		    flags & NFT_SET_ELEM_INTERVAL_END)
-			return -EINVAL;
 	} else {
 		if (nla[NFTA_SET_ELEM_DATA] != NULL)
 			return -EINVAL;
 	}
 
+	if ((flags & NFT_SET_ELEM_INTERVAL_END) &&
+	     (nla[NFTA_SET_ELEM_DATA] ||
+	      nla[NFTA_SET_ELEM_OBJREF] ||
+	      nla[NFTA_SET_ELEM_TIMEOUT] ||
+	      nla[NFTA_SET_ELEM_EXPIRATION] ||
+	      nla[NFTA_SET_ELEM_USERDATA] ||
+	      nla[NFTA_SET_ELEM_EXPR]))
+		return -EINVAL;
+
 	timeout = 0;
 	if (nla[NFTA_SET_ELEM_TIMEOUT] != NULL) {
 		if (!(set->flags & NFT_SET_TIMEOUT))
@@ -5394,8 +5436,7 @@ nft_obj_type_get(struct net *net, u32 objtype)
 	lockdep_nfnl_nft_mutex_not_held();
 #ifdef CONFIG_MODULES
 	if (type == NULL) {
-		nft_request_module(net, "nft-obj-%u", objtype);
-		if (__nft_obj_type_get(objtype))
+		if (nft_request_module(net, "nft-obj-%u", objtype) == -EAGAIN)
 			return ERR_PTR(-EAGAIN);
 	}
 #endif
@@ -5476,7 +5517,7 @@ static int nf_tables_newobj(struct net *net, struct sock *nlsk,
 		if (nlh->nlmsg_flags & NLM_F_REPLACE)
 			return -EOPNOTSUPP;
 
-		type = nft_obj_type_get(net, objtype);
+		type = __nft_obj_type_get(objtype);
 		nft_ctx_init(&ctx, net, skb, nlh, family, table, NULL, nla);
 
 		return nf_tables_updobj(&ctx, type, nla[NFTA_OBJ_DATA], obj);
@@ -5968,14 +6009,14 @@ nft_flowtable_type_get(struct net *net, u8 family)
 	lockdep_nfnl_nft_mutex_not_held();
 #ifdef CONFIG_MODULES
 	if (type == NULL) {
-		nft_request_module(net, "nf-flowtable-%u", family);
-		if (__nft_flowtable_type_get(family))
+		if (nft_request_module(net, "nf-flowtable-%u", family) == -EAGAIN)
 			return ERR_PTR(-EAGAIN);
 	}
 #endif
 	return ERR_PTR(-ENOENT);
 }
 
+/* Only called from error and netdev event paths. */
 static void nft_unregister_flowtable_hook(struct net *net,
 					  struct nft_flowtable *flowtable,
 					  struct nft_hook *hook)
@@ -5991,7 +6032,7 @@ static void nft_unregister_flowtable_net_hooks(struct net *net,
 	struct nft_hook *hook;
 
 	list_for_each_entry(hook, &flowtable->hook_list, list)
-		nft_unregister_flowtable_hook(net, flowtable, hook);
+		nf_unregister_net_hook(net, &hook->ops);
 }
 
 static int nft_register_flowtable_net_hooks(struct net *net,
@@ -6440,12 +6481,14 @@ static void nf_tables_flowtable_destroy(struct nft_flowtable *flowtable)
 {
 	struct nft_hook *hook, *next;
 
+	flowtable->data.type->free(&flowtable->data);
 	list_for_each_entry_safe(hook, next, &flowtable->hook_list, list) {
+		flowtable->data.type->setup(&flowtable->data, hook->ops.dev,
+					    FLOW_BLOCK_UNBIND);
 		list_del_rcu(&hook->list);
 		kfree(hook);
 	}
 	kfree(flowtable->name);
-	flowtable->data.type->free(&flowtable->data);
 	module_put(flowtable->data.type->owner);
 	kfree(flowtable);
 }
@@ -6489,6 +6532,7 @@ static void nft_flowtable_event(unsigned long event, struct net_device *dev,
 		if (hook->ops.dev != dev)
 			continue;
 
+		/* flow_offload_netdev_event() cleans up entries for us. */
 		nft_unregister_flowtable_hook(dev_net(dev), flowtable, hook);
 		list_del_rcu(&hook->list);
 		kfree_rcu(hook, rcu);
@@ -6967,6 +7011,18 @@ static void nft_chain_del(struct nft_chain *chain)
 	list_del_rcu(&chain->list);
 }
 
+static void nf_tables_module_autoload_cleanup(struct net *net)
+{
+	struct nft_module_request *req, *next;
+
+	WARN_ON_ONCE(!list_empty(&net->nft.commit_list));
+	list_for_each_entry_safe(req, next, &net->nft.module_list, list) {
+		WARN_ON_ONCE(!req->done);
+		list_del(&req->list);
+		kfree(req);
+	}
+}
+
 static void nf_tables_commit_release(struct net *net)
 {
 	struct nft_trans *trans;
@@ -6979,6 +7035,7 @@ static void nf_tables_commit_release(struct net *net)
 	 * to prevent expensive synchronize_rcu() in commit phase.
 	 */
 	if (list_empty(&net->nft.commit_list)) {
+		nf_tables_module_autoload_cleanup(net);
 		mutex_unlock(&net->nft.commit_mutex);
 		return;
 	}
@@ -6993,6 +7050,7 @@ static void nf_tables_commit_release(struct net *net)
 	list_splice_tail_init(&net->nft.commit_list, &nf_tables_destroy_list);
 	spin_unlock(&nf_tables_destroy_list_lock);
 
+	nf_tables_module_autoload_cleanup(net);
 	mutex_unlock(&net->nft.commit_mutex);
 
 	schedule_work(&trans_destroy_work);
@@ -7184,6 +7242,26 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 	return 0;
 }
 
+static void nf_tables_module_autoload(struct net *net)
+{
+	struct nft_module_request *req, *next;
+	LIST_HEAD(module_list);
+
+	list_splice_init(&net->nft.module_list, &module_list);
+	mutex_unlock(&net->nft.commit_mutex);
+	list_for_each_entry_safe(req, next, &module_list, list) {
+		if (req->done) {
+			list_del(&req->list);
+			kfree(req);
+		} else {
+			request_module("%s", req->module);
+			req->done = true;
+		}
+	}
+	mutex_lock(&net->nft.commit_mutex);
+	list_splice(&module_list, &net->nft.module_list);
+}
+
 static void nf_tables_abort_release(struct nft_trans *trans)
 {
 	switch (trans->msg_type) {
@@ -7213,7 +7291,7 @@ static void nf_tables_abort_release(struct nft_trans *trans)
 	kfree(trans);
 }
 
-static int __nf_tables_abort(struct net *net)
+static int __nf_tables_abort(struct net *net, bool autoload)
 {
 	struct nft_trans *trans, *next;
 	struct nft_trans_elem *te;
@@ -7335,6 +7413,11 @@ static int __nf_tables_abort(struct net *net)
 		nf_tables_abort_release(trans);
 	}
 
+	if (autoload)
+		nf_tables_module_autoload(net);
+	else
+		nf_tables_module_autoload_cleanup(net);
+
 	return 0;
 }
 
@@ -7343,9 +7426,9 @@ static void nf_tables_cleanup(struct net *net)
 	nft_validate_state_update(net, NFT_VALIDATE_SKIP);
 }
 
-static int nf_tables_abort(struct net *net, struct sk_buff *skb)
+static int nf_tables_abort(struct net *net, struct sk_buff *skb, bool autoload)
 {
-	int ret = __nf_tables_abort(net);
+	int ret = __nf_tables_abort(net, autoload);
 
 	mutex_unlock(&net->nft.commit_mutex);
 
@@ -7940,6 +8023,7 @@ static int __net_init nf_tables_init_net(struct net *net)
 {
 	INIT_LIST_HEAD(&net->nft.tables);
 	INIT_LIST_HEAD(&net->nft.commit_list);
+	INIT_LIST_HEAD(&net->nft.module_list);
 	mutex_init(&net->nft.commit_mutex);
 	net->nft.base_seq = 1;
 	net->nft.validate_state = NFT_VALIDATE_SKIP;
@@ -7951,7 +8035,7 @@ static void __net_exit nf_tables_exit_net(struct net *net)
 {
 	mutex_lock(&net->nft.commit_mutex);
 	if (!list_empty(&net->nft.commit_list))
-		__nf_tables_abort(net);
+		__nf_tables_abort(net, false);
 	__nft_release_tables(net);
 	mutex_unlock(&net->nft.commit_mutex);
 	WARN_ON_ONCE(!list_empty(&net->nft.tables));