2 files changed, 300 insertions, 12 deletions

diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h
index 640441a2f926..7256e9c80477 100644
--- a/include/net/netfilter/nf_tables.h
+++ b/include/net/netfilter/nf_tables.h
@@ -512,6 +512,7 @@ struct nft_set_elem_expr {
  *
  *	@list: table set list node
  *	@bindings: list of set bindings
+ *	@refs: internal refcounting for async set destruction
  *	@table: table this set belongs to
  *	@net: netnamespace this set belongs to
  * 	@name: name of the set
@@ -541,6 +542,7 @@ struct nft_set_elem_expr {
 struct nft_set {
 	struct list_head		list;
 	struct list_head		bindings;
+	refcount_t			refs;
 	struct nft_table		*table;
 	possible_net_t			net;
 	char				*name;
@@ -562,7 +564,8 @@ struct nft_set {
 	struct list_head		pending_update;
 	/* runtime data below here */
 	const struct nft_set_ops	*ops ____cacheline_aligned;
-	u16				flags:14,
+	u16				flags:13,
+					dead:1,
 					genmask:2;
 	u8				klen;
 	u8				dlen;
@@ -1592,6 +1595,32 @@ static inline void nft_set_elem_clear_busy(struct nft_set_ext *ext)
 	clear_bit(NFT_SET_ELEM_BUSY_BIT, word);
 }
 
+#define NFT_SET_ELEM_DEAD_MASK	(1 << 3)
+
+#if defined(__LITTLE_ENDIAN_BITFIELD)
+#define NFT_SET_ELEM_DEAD_BIT	3
+#elif defined(__BIG_ENDIAN_BITFIELD)
+#define NFT_SET_ELEM_DEAD_BIT	(BITS_PER_LONG - BITS_PER_BYTE + 3)
+#else
+#error
+#endif
+
+static inline void nft_set_elem_dead(struct nft_set_ext *ext)
+{
+	unsigned long *word = (unsigned long *)ext;
+
+	BUILD_BUG_ON(offsetof(struct nft_set_ext, genmask) != 0);
+	set_bit(NFT_SET_ELEM_DEAD_BIT, word);
+}
+
+static inline int nft_set_elem_is_dead(const struct nft_set_ext *ext)
+{
+	unsigned long *word = (unsigned long *)ext;
+
+	BUILD_BUG_ON(offsetof(struct nft_set_ext, genmask) != 0);
+	return test_bit(NFT_SET_ELEM_DEAD_BIT, word);
+}
+
 /**
  *	struct nft_trans - nf_tables object update in transaction
  *
@@ -1732,6 +1761,38 @@ struct nft_trans_flowtable {
 #define nft_trans_flowtable_flags(trans)	\
 	(((struct nft_trans_flowtable *)trans->data)->flags)
 
+#define NFT_TRANS_GC_BATCHCOUNT	256
+
+struct nft_trans_gc {
+	struct list_head	list;
+	struct net		*net;
+	struct nft_set		*set;
+	u32			seq;
+	u8			count;
+	void			*priv[NFT_TRANS_GC_BATCHCOUNT];
+	struct rcu_head		rcu;
+};
+
+struct nft_trans_gc *nft_trans_gc_alloc(struct nft_set *set,
+					unsigned int gc_seq, gfp_t gfp);
+void nft_trans_gc_destroy(struct nft_trans_gc *trans);
+
+struct nft_trans_gc *nft_trans_gc_queue_async(struct nft_trans_gc *gc,
+					      unsigned int gc_seq, gfp_t gfp);
+void nft_trans_gc_queue_async_done(struct nft_trans_gc *gc);
+
+struct nft_trans_gc *nft_trans_gc_queue_sync(struct nft_trans_gc *gc, gfp_t gfp);
+void nft_trans_gc_queue_sync_done(struct nft_trans_gc *trans);
+
+void nft_trans_gc_elem_add(struct nft_trans_gc *gc, void *priv);
+
+struct nft_trans_gc *nft_trans_gc_catchall(struct nft_trans_gc *gc,
+					   unsigned int gc_seq);
+
+void nft_setelem_data_deactivate(const struct net *net,
+				 const struct nft_set *set,
+				 struct nft_set_elem *elem);
+
 int __init nft_chain_filter_init(void);
 void nft_chain_filter_fini(void);
 
@@ -1758,6 +1819,7 @@ struct nftables_pernet {
 	struct mutex		commit_mutex;
 	u64			table_handle;
 	unsigned int		base_seq;
+	unsigned int		gc_seq;
 };
 
 extern unsigned int nf_tables_net_id;
diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index b4321869e5c6..c28bacb9479b 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -31,7 +31,9 @@ static LIST_HEAD(nf_tables_expressions);
 static LIST_HEAD(nf_tables_objects);
 static LIST_HEAD(nf_tables_flowtables);
 static LIST_HEAD(nf_tables_destroy_list);
+static LIST_HEAD(nf_tables_gc_list);
 static DEFINE_SPINLOCK(nf_tables_destroy_list_lock);
+static DEFINE_SPINLOCK(nf_tables_gc_list_lock);
 
 enum {
 	NFT_VALIDATE_SKIP	= 0,
@@ -120,6 +122,9 @@ static void nft_validate_state_update(struct nft_table *table, u8 new_validate_s
 static void nf_tables_trans_destroy_work(struct work_struct *w);
 static DECLARE_WORK(trans_destroy_work, nf_tables_trans_destroy_work);
 
+static void nft_trans_gc_work(struct work_struct *work);
+static DECLARE_WORK(trans_gc_work, nft_trans_gc_work);
+
 static void nft_ctx_init(struct nft_ctx *ctx,
 			 struct net *net,
 			 const struct sk_buff *skb,
@@ -582,10 +587,6 @@ static int nft_trans_set_add(const struct nft_ctx *ctx, int msg_type,
 	return __nft_trans_set_add(ctx, msg_type, set, NULL);
 }
 
-static void nft_setelem_data_deactivate(const struct net *net,
-					const struct nft_set *set,
-					struct nft_set_elem *elem);
-
 static int nft_mapelem_deactivate(const struct nft_ctx *ctx,
 				  struct nft_set *set,
 				  const struct nft_set_iter *iter,
@@ -5055,6 +5056,7 @@ static int nf_tables_newset(struct sk_buff *skb, const struct nfnl_info *info,
 
 	INIT_LIST_HEAD(&set->bindings);
 	INIT_LIST_HEAD(&set->catchall_list);
+	refcount_set(&set->refs, 1);
 	set->table = table;
 	write_pnet(&set->net, net);
 	set->ops = ops;
@@ -5122,6 +5124,14 @@ static void nft_set_catchall_destroy(const struct nft_ctx *ctx,
 	}
 }
 
+static void nft_set_put(struct nft_set *set)
+{
+	if (refcount_dec_and_test(&set->refs)) {
+		kfree(set->name);
+		kvfree(set);
+	}
+}
+
 static void nft_set_destroy(const struct nft_ctx *ctx, struct nft_set *set)
 {
 	int i;
@@ -5134,8 +5144,7 @@ static void nft_set_destroy(const struct nft_ctx *ctx, struct nft_set *set)
 
 	set->ops->destroy(ctx, set);
 	nft_set_catchall_destroy(ctx, set);
-	kfree(set->name);
-	kvfree(set);
+	nft_set_put(set);
 }
 
 static int nf_tables_delset(struct sk_buff *skb, const struct nfnl_info *info,
@@ -6278,7 +6287,8 @@ struct nft_set_ext *nft_set_catchall_lookup(const struct net *net,
 	list_for_each_entry_rcu(catchall, &set->catchall_list, list) {
 		ext = nft_set_elem_ext(set, catchall->elem);
 		if (nft_set_elem_active(ext, genmask) &&
-		    !nft_set_elem_expired(ext))
+		    !nft_set_elem_expired(ext) &&
+		    !nft_set_elem_is_dead(ext))
 			return ext;
 	}
 
@@ -6933,9 +6943,9 @@ static void nft_setelem_data_activate(const struct net *net,
 		nft_use_inc_restore(&(*nft_set_ext_obj(ext))->use);
 }
 
-static void nft_setelem_data_deactivate(const struct net *net,
-					const struct nft_set *set,
-					struct nft_set_elem *elem)
+void nft_setelem_data_deactivate(const struct net *net,
+				 const struct nft_set *set,
+				 struct nft_set_elem *elem)
 {
 	const struct nft_set_ext *ext = nft_set_elem_ext(set, elem->priv);
 
@@ -9418,6 +9428,207 @@ void nft_chain_del(struct nft_chain *chain)
 	list_del_rcu(&chain->list);
 }
 
+static void nft_trans_gc_setelem_remove(struct nft_ctx *ctx,
+					struct nft_trans_gc *trans)
+{
+	void **priv = trans->priv;
+	unsigned int i;
+
+	for (i = 0; i < trans->count; i++) {
+		struct nft_set_elem elem = {
+			.priv = priv[i],
+		};
+
+		nft_setelem_data_deactivate(ctx->net, trans->set, &elem);
+		nft_setelem_remove(ctx->net, trans->set, &elem);
+	}
+}
+
+void nft_trans_gc_destroy(struct nft_trans_gc *trans)
+{
+	nft_set_put(trans->set);
+	put_net(trans->net);
+	kfree(trans);
+}
+
+static void nft_trans_gc_trans_free(struct rcu_head *rcu)
+{
+	struct nft_set_elem elem = {};
+	struct nft_trans_gc *trans;
+	struct nft_ctx ctx = {};
+	unsigned int i;
+
+	trans = container_of(rcu, struct nft_trans_gc, rcu);
+	ctx.net	= read_pnet(&trans->set->net);
+
+	for (i = 0; i < trans->count; i++) {
+		elem.priv = trans->priv[i];
+		if (!nft_setelem_is_catchall(trans->set, &elem))
+			atomic_dec(&trans->set->nelems);
+
+		nf_tables_set_elem_destroy(&ctx, trans->set, elem.priv);
+	}
+
+	nft_trans_gc_destroy(trans);
+}
+
+static bool nft_trans_gc_work_done(struct nft_trans_gc *trans)
+{
+	struct nftables_pernet *nft_net;
+	struct nft_ctx ctx = {};
+
+	nft_net = nft_pernet(trans->net);
+
+	mutex_lock(&nft_net->commit_mutex);
+
+	/* Check for race with transaction, otherwise this batch refers to
+	 * stale objects that might not be there anymore. Skip transaction if
+	 * set has been destroyed from control plane transaction in case gc
+	 * worker loses race.
+	 */
+	if (READ_ONCE(nft_net->gc_seq) != trans->seq || trans->set->dead) {
+		mutex_unlock(&nft_net->commit_mutex);
+		return false;
+	}
+
+	ctx.net = trans->net;
+	ctx.table = trans->set->table;
+
+	nft_trans_gc_setelem_remove(&ctx, trans);
+	mutex_unlock(&nft_net->commit_mutex);
+
+	return true;
+}
+
+static void nft_trans_gc_work(struct work_struct *work)
+{
+	struct nft_trans_gc *trans, *next;
+	LIST_HEAD(trans_gc_list);
+
+	spin_lock(&nf_tables_destroy_list_lock);
+	list_splice_init(&nf_tables_gc_list, &trans_gc_list);
+	spin_unlock(&nf_tables_destroy_list_lock);
+
+	list_for_each_entry_safe(trans, next, &trans_gc_list, list) {
+		list_del(&trans->list);
+		if (!nft_trans_gc_work_done(trans)) {
+			nft_trans_gc_destroy(trans);
+			continue;
+		}
+		call_rcu(&trans->rcu, nft_trans_gc_trans_free);
+	}
+}
+
+struct nft_trans_gc *nft_trans_gc_alloc(struct nft_set *set,
+					unsigned int gc_seq, gfp_t gfp)
+{
+	struct net *net = read_pnet(&set->net);
+	struct nft_trans_gc *trans;
+
+	trans = kzalloc(sizeof(*trans), gfp);
+	if (!trans)
+		return NULL;
+
+	refcount_inc(&set->refs);
+	trans->set = set;
+	trans->net = get_net(net);
+	trans->seq = gc_seq;
+
+	return trans;
+}
+
+void nft_trans_gc_elem_add(struct nft_trans_gc *trans, void *priv)
+{
+	trans->priv[trans->count++] = priv;
+}
+
+static void nft_trans_gc_queue_work(struct nft_trans_gc *trans)
+{
+	spin_lock(&nf_tables_gc_list_lock);
+	list_add_tail(&trans->list, &nf_tables_gc_list);
+	spin_unlock(&nf_tables_gc_list_lock);
+
+	schedule_work(&trans_gc_work);
+}
+
+static int nft_trans_gc_space(struct nft_trans_gc *trans)
+{
+	return NFT_TRANS_GC_BATCHCOUNT - trans->count;
+}
+
+struct nft_trans_gc *nft_trans_gc_queue_async(struct nft_trans_gc *gc,
+					      unsigned int gc_seq, gfp_t gfp)
+{
+	if (nft_trans_gc_space(gc))
+		return gc;
+
+	nft_trans_gc_queue_work(gc);
+
+	return nft_trans_gc_alloc(gc->set, gc_seq, gfp);
+}
+
+void nft_trans_gc_queue_async_done(struct nft_trans_gc *trans)
+{
+	if (trans->count == 0) {
+		nft_trans_gc_destroy(trans);
+		return;
+	}
+
+	nft_trans_gc_queue_work(trans);
+}
+
+struct nft_trans_gc *nft_trans_gc_queue_sync(struct nft_trans_gc *gc, gfp_t gfp)
+{
+	if (WARN_ON_ONCE(!lockdep_commit_lock_is_held(gc->net)))
+		return NULL;
+
+	if (nft_trans_gc_space(gc))
+		return gc;
+
+	call_rcu(&gc->rcu, nft_trans_gc_trans_free);
+
+	return nft_trans_gc_alloc(gc->set, 0, gfp);
+}
+
+void nft_trans_gc_queue_sync_done(struct nft_trans_gc *trans)
+{
+	WARN_ON_ONCE(!lockdep_commit_lock_is_held(trans->net));
+
+	if (trans->count == 0) {
+		nft_trans_gc_destroy(trans);
+		return;
+	}
+
+	call_rcu(&trans->rcu, nft_trans_gc_trans_free);
+}
+
+struct nft_trans_gc *nft_trans_gc_catchall(struct nft_trans_gc *gc,
+					   unsigned int gc_seq)
+{
+	struct nft_set_elem_catchall *catchall;
+	const struct nft_set *set = gc->set;
+	struct nft_set_ext *ext;
+
+	list_for_each_entry_rcu(catchall, &set->catchall_list, list) {
+		ext = nft_set_elem_ext(set, catchall->elem);
+
+		if (!nft_set_elem_expired(ext))
+			continue;
+		if (nft_set_elem_is_dead(ext))
+			goto dead_elem;
+
+		nft_set_elem_dead(ext);
+dead_elem:
+		gc = nft_trans_gc_queue_async(gc, gc_seq, GFP_ATOMIC);
+		if (!gc)
+			return NULL;
+
+		nft_trans_gc_elem_add(gc, catchall->elem);
+	}
+
+	return gc;
+}
+
 static void nf_tables_module_autoload_cleanup(struct net *net)
 {
 	struct nftables_pernet *nft_net = nft_pernet(net);
@@ -9580,11 +9791,11 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 {
 	struct nftables_pernet *nft_net = nft_pernet(net);
 	struct nft_trans *trans, *next;
+	unsigned int base_seq, gc_seq;
 	LIST_HEAD(set_update_list);
 	struct nft_trans_elem *te;
 	struct nft_chain *chain;
 	struct nft_table *table;
-	unsigned int base_seq;
 	LIST_HEAD(adl);
 	int err;
 
@@ -9661,6 +9872,10 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 
 	WRITE_ONCE(nft_net->base_seq, base_seq);
 
+	/* Bump gc counter, it becomes odd, this is the busy mark. */
+	gc_seq = READ_ONCE(nft_net->gc_seq);
+	WRITE_ONCE(nft_net->gc_seq, ++gc_seq);
+
 	/* step 3. Start new generation, rules_gen_X now in use. */
 	net->nft.gencursor = nft_gencursor_next(net);
 
@@ -9768,6 +9983,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 			break;
 		case NFT_MSG_DELSET:
 		case NFT_MSG_DESTROYSET:
+			nft_trans_set(trans)->dead = 1;
 			list_del_rcu(&nft_trans_set(trans)->list);
 			nf_tables_set_notify(&trans->ctx, nft_trans_set(trans),
 					     trans->msg_type, GFP_KERNEL);
@@ -9870,6 +10086,8 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 	nft_commit_notify(net, NETLINK_CB(skb).portid);
 	nf_tables_gen_notify(net, skb, NFT_MSG_NEWGEN);
 	nf_tables_commit_audit_log(&adl, nft_net->base_seq);
+
+	WRITE_ONCE(nft_net->gc_seq, ++gc_seq);
 	nf_tables_commit_release(net);
 
 	return 0;
@@ -10919,6 +11137,7 @@ static int __net_init nf_tables_init_net(struct net *net)
 	INIT_LIST_HEAD(&nft_net->notify_list);
 	mutex_init(&nft_net->commit_mutex);
 	nft_net->base_seq = 1;
+	nft_net->gc_seq = 0;
 
 	return 0;
 }
@@ -10947,10 +11166,16 @@ static void __net_exit nf_tables_exit_net(struct net *net)
 	WARN_ON_ONCE(!list_empty(&nft_net->notify_list));
 }
 
+static void nf_tables_exit_batch(struct list_head *net_exit_list)
+{
+	flush_work(&trans_gc_work);
+}
+
 static struct pernet_operations nf_tables_net_ops = {
 	.init		= nf_tables_init_net,
 	.pre_exit	= nf_tables_pre_exit_net,
 	.exit		= nf_tables_exit_net,
+	.exit_batch	= nf_tables_exit_batch,
 	.id		= &nf_tables_net_id,
 	.size		= sizeof(struct nftables_pernet),
 };
@@ -11022,6 +11247,7 @@ static void __exit nf_tables_module_exit(void)
 	nft_chain_filter_fini();
 	nft_chain_route_fini();
 	unregister_pernet_subsys(&nf_tables_net_ops);
+	cancel_work_sync(&trans_gc_work);
 	cancel_work_sync(&trans_destroy_work);
 	rcu_barrier();
 	rhltable_destroy(&nft_objname_ht);