Merge git://git.kernel.org/pub/scm/linux/kernel/git/pablo/nf-next

Pablo Neira Ayuso says:

====================
Netfilter updates for net-next

The following patchset contains Netfilter updates for net-next:

1) Use nfnetlink_unicast() instead of netlink_unicast() in nft_compat.

2) Remove call to nf_ct_l4proto_find() in flowtable offload timeout
   fixup.

3) CLUSTERIP registers ARP hook on demand, from Florian.

4) Use clusterip_net to store pernet warning, also from Florian.

5) Remove struct netns_xt, from Florian Westphal.

6) Enable ebtables hooks in initns on demand, from Florian.

7) Allow to filter conntrack netlink dump per status bits,
   from Florian Westphal.

8) Register x_tables hooks in initns on demand, from Florian.

9) Remove queue_handler from per-netns structure, again from Florian.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

7 files changed, 165 insertions, 72 deletions

diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c
index e81af33b233b..eb35c6151fb0 100644
--- a/net/netfilter/nf_conntrack_netlink.c
+++ b/net/netfilter/nf_conntrack_netlink.c
@@ -852,6 +852,11 @@ static int ctnetlink_done(struct netlink_callback *cb)
 	return 0;
 }
 
+struct ctnetlink_filter_u32 {
+	u32 val;
+	u32 mask;
+};
+
 struct ctnetlink_filter {
 	u8 family;
 
@@ -862,10 +867,8 @@ struct ctnetlink_filter {
 	struct nf_conntrack_tuple reply;
 	struct nf_conntrack_zone zone;
 
-	struct {
-		u_int32_t val;
-		u_int32_t mask;
-	} mark;
+	struct ctnetlink_filter_u32 mark;
+	struct ctnetlink_filter_u32 status;
 };
 
 static const struct nla_policy cta_filter_nla_policy[CTA_FILTER_MAX + 1] = {
@@ -907,6 +910,46 @@ static int ctnetlink_parse_tuple_filter(const struct nlattr * const cda[],
 					 struct nf_conntrack_zone *zone,
 					 u_int32_t flags);
 
+static int ctnetlink_filter_parse_mark(struct ctnetlink_filter_u32 *mark,
+				       const struct nlattr * const cda[])
+{
+#ifdef CONFIG_NF_CONNTRACK_MARK
+	if (cda[CTA_MARK]) {
+		mark->val = ntohl(nla_get_be32(cda[CTA_MARK]));
+
+		if (cda[CTA_MARK_MASK])
+			mark->mask = ntohl(nla_get_be32(cda[CTA_MARK_MASK]));
+		else
+			mark->mask = 0xffffffff;
+	} else if (cda[CTA_MARK_MASK]) {
+		return -EINVAL;
+	}
+#endif
+	return 0;
+}
+
+static int ctnetlink_filter_parse_status(struct ctnetlink_filter_u32 *status,
+					 const struct nlattr * const cda[])
+{
+	if (cda[CTA_STATUS]) {
+		status->val = ntohl(nla_get_be32(cda[CTA_STATUS]));
+		if (cda[CTA_STATUS_MASK])
+			status->mask = ntohl(nla_get_be32(cda[CTA_STATUS_MASK]));
+		else
+			status->mask = status->val;
+
+		/* status->val == 0? always true, else always false. */
+		if (status->mask == 0)
+			return -EINVAL;
+	} else if (cda[CTA_STATUS_MASK]) {
+		return -EINVAL;
+	}
+
+	/* CTA_STATUS is NLA_U32, if this fires UAPI needs to be extended */
+	BUILD_BUG_ON(__IPS_MAX_BIT >= 32);
+	return 0;
+}
+
 static struct ctnetlink_filter *
 ctnetlink_alloc_filter(const struct nlattr * const cda[], u8 family)
 {
@@ -924,18 +967,14 @@ ctnetlink_alloc_filter(const struct nlattr * const cda[], u8 family)
 
 	filter->family = family;
 
-#ifdef CONFIG_NF_CONNTRACK_MARK
-	if (cda[CTA_MARK]) {
-		filter->mark.val = ntohl(nla_get_be32(cda[CTA_MARK]));
-		if (cda[CTA_MARK_MASK])
-			filter->mark.mask = ntohl(nla_get_be32(cda[CTA_MARK_MASK]));
-		else
-			filter->mark.mask = 0xffffffff;
-	} else if (cda[CTA_MARK_MASK]) {
-		err = -EINVAL;
+	err = ctnetlink_filter_parse_mark(&filter->mark, cda);
+	if (err)
 		goto err_filter;
-	}
-#endif
+
+	err = ctnetlink_filter_parse_status(&filter->status, cda);
+	if (err)
+		goto err_filter;
+
 	if (!cda[CTA_FILTER])
 		return filter;
 
@@ -989,7 +1028,7 @@ err_filter:
 
 static bool ctnetlink_needs_filter(u8 family, const struct nlattr * const *cda)
 {
-	return family || cda[CTA_MARK] || cda[CTA_FILTER];
+	return family || cda[CTA_MARK] || cda[CTA_FILTER] || cda[CTA_STATUS];
 }
 
 static int ctnetlink_start(struct netlink_callback *cb)
@@ -1082,6 +1121,7 @@ static int ctnetlink_filter_match(struct nf_conn *ct, void *data)
 {
 	struct ctnetlink_filter *filter = data;
 	struct nf_conntrack_tuple *tuple;
+	u32 status;
 
 	if (filter == NULL)
 		goto out;
@@ -1113,6 +1153,9 @@ static int ctnetlink_filter_match(struct nf_conn *ct, void *data)
 	if ((ct->mark & filter->mark.mask) != filter->mark.val)
 		goto ignore_entry;
 #endif
+	status = (u32)READ_ONCE(ct->status);
+	if ((status & filter->status.mask) != filter->status.val)
+		goto ignore_entry;
 
 out:
 	return 1;
@@ -1495,6 +1538,7 @@ static const struct nla_policy ct_nla_policy[CTA_MAX+1] = {
 	[CTA_LABELS_MASK]	= { .type = NLA_BINARY,
 				    .len = NF_CT_LABELS_MAX_SIZE },
 	[CTA_FILTER]		= { .type = NLA_NESTED },
+	[CTA_STATUS_MASK]	= { .type = NLA_U32 },
 };
 
 static int ctnetlink_flush_iterate(struct nf_conn *ct, void *data)
diff --git a/net/netfilter/nf_flow_table_core.c b/net/netfilter/nf_flow_table_core.c
index 8ecad71b3613..a106721a0bd7 100644
--- a/net/netfilter/nf_flow_table_core.c
+++ b/net/netfilter/nf_flow_table_core.c
@@ -180,15 +180,10 @@ static void flow_offload_fixup_tcp(struct ip_ct_tcp *tcp)
 
 static void flow_offload_fixup_ct_timeout(struct nf_conn *ct)
 {
-	const struct nf_conntrack_l4proto *l4proto;
 	struct net *net = nf_ct_net(ct);
 	int l4num = nf_ct_protonum(ct);
 	unsigned int timeout;
 
-	l4proto = nf_ct_l4proto_find(l4num);
-	if (!l4proto)
-		return;
-
 	if (l4num == IPPROTO_TCP) {
 		struct nf_tcp_net *tn = nf_tcp_pernet(net);
 
@@ -273,15 +268,10 @@ static const struct rhashtable_params nf_flow_offload_rhash_params = {
 
 unsigned long flow_offload_get_timeout(struct flow_offload *flow)
 {
-	const struct nf_conntrack_l4proto *l4proto;
 	unsigned long timeout = NF_FLOW_TIMEOUT;
 	struct net *net = nf_ct_net(flow->ct);
 	int l4num = nf_ct_protonum(flow->ct);
 
-	l4proto = nf_ct_l4proto_find(l4num);
-	if (!l4proto)
-		return timeout;
-
 	if (l4num == IPPROTO_TCP) {
 		struct nf_tcp_net *tn = nf_tcp_pernet(net);
 
diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c
index 7f2f69b609d8..6d12afabfe8a 100644
--- a/net/netfilter/nf_queue.c
+++ b/net/netfilter/nf_queue.c
@@ -21,6 +21,8 @@
 
 #include "nf_internals.h"
 
+static const struct nf_queue_handler __rcu *nf_queue_handler;
+
 /*
  * Hook for nfnetlink_queue to register its queue handler.
  * We do this so that most of the NFQUEUE code can be modular.
@@ -29,20 +31,18 @@
  * receives, no matter what.
  */
 
-/* return EBUSY when somebody else is registered, return EEXIST if the
- * same handler is registered, return 0 in case of success. */
-void nf_register_queue_handler(struct net *net, const struct nf_queue_handler *qh)
+void nf_register_queue_handler(const struct nf_queue_handler *qh)
 {
 	/* should never happen, we only have one queueing backend in kernel */
-	WARN_ON(rcu_access_pointer(net->nf.queue_handler));
-	rcu_assign_pointer(net->nf.queue_handler, qh);
+	WARN_ON(rcu_access_pointer(nf_queue_handler));
+	rcu_assign_pointer(nf_queue_handler, qh);
 }
 EXPORT_SYMBOL(nf_register_queue_handler);
 
 /* The caller must flush their queue before this */
-void nf_unregister_queue_handler(struct net *net)
+void nf_unregister_queue_handler(void)
 {
-	RCU_INIT_POINTER(net->nf.queue_handler, NULL);
+	RCU_INIT_POINTER(nf_queue_handler, NULL);
 }
 EXPORT_SYMBOL(nf_unregister_queue_handler);
 
@@ -108,7 +108,7 @@ void nf_queue_nf_hook_drop(struct net *net)
 	const struct nf_queue_handler *qh;
 
 	rcu_read_lock();
-	qh = rcu_dereference(net->nf.queue_handler);
+	qh = rcu_dereference(nf_queue_handler);
 	if (qh)
 		qh->nf_hook_drop(net);
 	rcu_read_unlock();
@@ -149,12 +149,11 @@ static int __nf_queue(struct sk_buff *skb, const struct nf_hook_state *state,
 {
 	struct nf_queue_entry *entry = NULL;
 	const struct nf_queue_handler *qh;
-	struct net *net = state->net;
 	unsigned int route_key_size;
 	int status;
 
 	/* QUEUE == DROP if no one is waiting, to be safe. */
-	qh = rcu_dereference(net->nf.queue_handler);
+	qh = rcu_dereference(nf_queue_handler);
 	if (!qh)
 		return -ESRCH;
 
diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c
index f774de0fc24f..4c3fbaaeb103 100644
--- a/net/netfilter/nfnetlink_queue.c
+++ b/net/netfilter/nfnetlink_queue.c
@@ -951,6 +951,16 @@ static void nfqnl_nf_hook_drop(struct net *net)
 	struct nfnl_queue_net *q = nfnl_queue_pernet(net);
 	int i;
 
+	/* This function is also called on net namespace error unwind,
+	 * when pernet_ops->init() failed and ->exit() functions of the
+	 * previous pernet_ops gets called.
+	 *
+	 * This may result in a call to nfqnl_nf_hook_drop() before
+	 * struct nfnl_queue_net was allocated.
+	 */
+	if (!q)
+		return;
+
 	for (i = 0; i < INSTANCE_BUCKETS; i++) {
 		struct nfqnl_instance *inst;
 		struct hlist_head *head = &q->instance_table[i];
@@ -1502,7 +1512,6 @@ static int __net_init nfnl_queue_net_init(struct net *net)
 			&nfqnl_seq_ops, sizeof(struct iter_state)))
 		return -ENOMEM;
 #endif
-	nf_register_queue_handler(net, &nfqh);
 	return 0;
 }
 
@@ -1511,7 +1520,6 @@ static void __net_exit nfnl_queue_net_exit(struct net *net)
 	struct nfnl_queue_net *q = nfnl_queue_pernet(net);
 	unsigned int i;
 
-	nf_unregister_queue_handler(net);
 #ifdef CONFIG_PROC_FS
 	remove_proc_entry("nfnetlink_queue", net->nf.proc_netfilter);
 #endif
@@ -1555,6 +1563,8 @@ static int __init nfnetlink_queue_init(void)
 		goto cleanup_netlink_subsys;
 	}
 
+	nf_register_queue_handler(&nfqh);
+
 	return status;
 
 cleanup_netlink_subsys:
@@ -1568,6 +1578,7 @@ out:
 
 static void __exit nfnetlink_queue_fini(void)
 {
+	nf_unregister_queue_handler();
 	unregister_netdevice_notifier(&nfqnl_dev_notifier);
 	nfnetlink_subsys_unregister(&nfqnl_subsys);
 	netlink_unregister_notifier(&nfqnl_rtnl_notifier);
diff --git a/net/netfilter/nft_compat.c b/net/netfilter/nft_compat.c
index 639c337c885b..272bcdb1392d 100644
--- a/net/netfilter/nft_compat.c
+++ b/net/netfilter/nft_compat.c
@@ -683,14 +683,12 @@ static int nfnl_compat_get_rcu(struct sk_buff *skb,
 		goto out_put;
 	}
 
-	ret = netlink_unicast(info->sk, skb2, NETLINK_CB(skb).portid,
-			      MSG_DONTWAIT);
-	if (ret > 0)
-		ret = 0;
+	ret = nfnetlink_unicast(skb2, info->net, NETLINK_CB(skb).portid);
 out_put:
 	rcu_read_lock();
 	module_put(THIS_MODULE);
-	return ret == -EAGAIN ? -ENOBUFS : ret;
+
+	return ret;
 }
 
 static const struct nla_policy nfnl_compat_policy_get[NFTA_COMPAT_MAX+1] = {
diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c
index 84e58ee501a4..25524e393349 100644
--- a/net/netfilter/x_tables.c
+++ b/net/netfilter/x_tables.c
@@ -39,6 +39,20 @@ MODULE_DESCRIPTION("{ip,ip6,arp,eb}_tables backend module");
 #define XT_PCPU_BLOCK_SIZE 4096
 #define XT_MAX_TABLE_SIZE	(512 * 1024 * 1024)
 
+struct xt_template {
+	struct list_head list;
+
+	/* called when table is needed in the given netns */
+	int (*table_init)(struct net *net);
+
+	struct module *me;
+
+	/* A unique name... */
+	char name[XT_TABLE_MAXNAMELEN];
+};
+
+static struct list_head xt_templates[NFPROTO_NUMPROTO];
+
 struct xt_pernet {
 	struct list_head tables[NFPROTO_NUMPROTO];
 };
@@ -1221,48 +1235,43 @@ struct xt_table *xt_find_table_lock(struct net *net, u_int8_t af,
 				    const char *name)
 {
 	struct xt_pernet *xt_net = net_generic(net, xt_pernet_id);
-	struct xt_table *t, *found = NULL;
+	struct module *owner = NULL;
+	struct xt_template *tmpl;
+	struct xt_table *t;
 
 	mutex_lock(&xt[af].mutex);
 	list_for_each_entry(t, &xt_net->tables[af], list)
 		if (strcmp(t->name, name) == 0 && try_module_get(t->me))
 			return t;
 
-	if (net == &init_net)
-		goto out;
-
-	/* Table doesn't exist in this netns, re-try init */
-	xt_net = net_generic(&init_net, xt_pernet_id);
-	list_for_each_entry(t, &xt_net->tables[af], list) {
+	/* Table doesn't exist in this netns, check larval list */
+	list_for_each_entry(tmpl, &xt_templates[af], list) {
 		int err;
 
-		if (strcmp(t->name, name))
+		if (strcmp(tmpl->name, name))
 			continue;
-		if (!try_module_get(t->me))
+		if (!try_module_get(tmpl->me))
 			goto out;
+
+		owner = tmpl->me;
+
 		mutex_unlock(&xt[af].mutex);
-		err = t->table_init(net);
+		err = tmpl->table_init(net);
 		if (err < 0) {
-			module_put(t->me);
+			module_put(owner);
 			return ERR_PTR(err);
 		}
 
-		found = t;
-
 		mutex_lock(&xt[af].mutex);
 		break;
 	}
 
-	if (!found)
-		goto out;
-
-	xt_net = net_generic(net, xt_pernet_id);
 	/* and once again: */
 	list_for_each_entry(t, &xt_net->tables[af], list)
 		if (strcmp(t->name, name) == 0)
 			return t;
 
-	module_put(found->me);
+	module_put(owner);
  out:
 	mutex_unlock(&xt[af].mutex);
 	return ERR_PTR(-ENOENT);
@@ -1749,6 +1758,58 @@ xt_hook_ops_alloc(const struct xt_table *table, nf_hookfn *fn)
 }
 EXPORT_SYMBOL_GPL(xt_hook_ops_alloc);
 
+int xt_register_template(const struct xt_table *table,
+			 int (*table_init)(struct net *net))
+{
+	int ret = -EEXIST, af = table->af;
+	struct xt_template *t;
+
+	mutex_lock(&xt[af].mutex);
+
+	list_for_each_entry(t, &xt_templates[af], list) {
+		if (WARN_ON_ONCE(strcmp(table->name, t->name) == 0))
+			goto out_unlock;
+	}
+
+	ret = -ENOMEM;
+	t = kzalloc(sizeof(*t), GFP_KERNEL);
+	if (!t)
+		goto out_unlock;
+
+	BUILD_BUG_ON(sizeof(t->name) != sizeof(table->name));
+
+	strscpy(t->name, table->name, sizeof(t->name));
+	t->table_init = table_init;
+	t->me = table->me;
+	list_add(&t->list, &xt_templates[af]);
+	ret = 0;
+out_unlock:
+	mutex_unlock(&xt[af].mutex);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(xt_register_template);
+
+void xt_unregister_template(const struct xt_table *table)
+{
+	struct xt_template *t;
+	int af = table->af;
+
+	mutex_lock(&xt[af].mutex);
+	list_for_each_entry(t, &xt_templates[af], list) {
+		if (strcmp(table->name, t->name))
+			continue;
+
+		list_del(&t->list);
+		mutex_unlock(&xt[af].mutex);
+		kfree(t);
+		return;
+	}
+
+	mutex_unlock(&xt[af].mutex);
+	WARN_ON_ONCE(1);
+}
+EXPORT_SYMBOL_GPL(xt_unregister_template);
+
 int xt_proto_init(struct net *net, u_int8_t af)
 {
 #ifdef CONFIG_PROC_FS
@@ -1937,6 +1998,7 @@ static int __init xt_init(void)
 #endif
 		INIT_LIST_HEAD(&xt[i].target);
 		INIT_LIST_HEAD(&xt[i].match);
+		INIT_LIST_HEAD(&xt_templates[i]);
 	}
 	rv = register_pernet_subsys(&xt_net_ops);
 	if (rv < 0)
diff --git a/net/netfilter/xt_CT.c b/net/netfilter/xt_CT.c
index 12404d221026..0a913ce07425 100644
--- a/net/netfilter/xt_CT.c
+++ b/net/netfilter/xt_CT.c
@@ -351,21 +351,10 @@ notrack_tg(struct sk_buff *skb, const struct xt_action_param *par)
 	return XT_CONTINUE;
 }
 
-static int notrack_chk(const struct xt_tgchk_param *par)
-{
-	if (!par->net->xt.notrack_deprecated_warning) {
-		pr_info("netfilter: NOTRACK target is deprecated, "
-			"use CT instead or upgrade iptables\n");
-		par->net->xt.notrack_deprecated_warning = true;
-	}
-	return 0;
-}
-
 static struct xt_target notrack_tg_reg __read_mostly = {
 	.name		= "NOTRACK",
 	.revision	= 0,
 	.family		= NFPROTO_UNSPEC,
-	.checkentry	= notrack_chk,
 	.target		= notrack_tg,
 	.table		= "raw",
 	.me		= THIS_MODULE,