[PATCH 0/6] ipset patches

[PATCH 0/6] ipset patches

[Jozsef Kadlecsik]

Hi Patrick and Pablo,

Here follows the patches which brings the kernel sync with ipset 6.9.1.
The first two are important bugfixes. Please consider applying them.

Best regards,
Jozsef

Jan Engelhardt (3):
  netfilter: ipset: avoid use of kernel-only types
  netfilter: ipset: expose userspace-relevant parts in ip_set.h
  netfilter: ipset: use NFPROTO_ constants

Joe Perches (1):
  netfilter: ipset: Remove unnecessary OOM logging messages

Jozsef Kadlecsik (2):
  netfilter: ipset: Autoload set type modules safely
  netfilter: ipset: Dumping error could lead to kernel BUG

 include/linux/netfilter/ipset/ip_set.h      |   31 ++++++++------
 include/linux/netfilter/xt_set.h            |    4 +-
 net/netfilter/ipset/ip_set_bitmap_ip.c      |    4 +-
 net/netfilter/ipset/ip_set_bitmap_ipmac.c   |    4 +-
 net/netfilter/ipset/ip_set_bitmap_port.c    |    4 +-
 net/netfilter/ipset/ip_set_core.c           |   57 +++++++++++++++++----------
 net/netfilter/ipset/ip_set_getport.c        |    4 +-
 net/netfilter/ipset/ip_set_hash_ip.c        |   18 ++++----
 net/netfilter/ipset/ip_set_hash_ipport.c    |   10 ++--
 net/netfilter/ipset/ip_set_hash_ipportip.c  |   10 ++--
 net/netfilter/ipset/ip_set_hash_ipportnet.c |   12 +++---
 net/netfilter/ipset/ip_set_hash_net.c       |   12 +++---
 net/netfilter/ipset/ip_set_hash_netiface.c  |   12 +++---
 net/netfilter/ipset/ip_set_hash_netport.c   |   12 +++---
 net/netfilter/ipset/ip_set_list_set.c       |    2 +-
 15 files changed, 108 insertions(+), 88 deletions(-)

[PATCH 1/6] netfilter: ipset: Autoload set type modules safely

[Jozsef Kadlecsik]

Jan Engelhardt noticed when userspace requests a set type unknown
to the kernel, it can lead to a loop due to the unsafe type module
loading. The issue is fixed in this patch.

Signed-off-by: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
---
 net/netfilter/ipset/ip_set_core.c |   36 ++++++++++++++++++++++++++----------
 1 files changed, 26 insertions(+), 10 deletions(-)

diff --git a/net/netfilter/ipset/ip_set_core.c b/net/netfilter/ipset/ip_set_core.c
index c012985..5370fd8 100644
--- a/net/netfilter/ipset/ip_set_core.c
+++ b/net/netfilter/ipset/ip_set_core.c
@@ -78,35 +78,42 @@ find_set_type(const char *name, u8 family, u8 revision)
 }
 
 /* Unlock, try to load a set type module and lock again */
-static int
-try_to_load_type(const char *name)
+static bool
+load_settype(const char *name)
 {
 	nfnl_unlock();
 	pr_debug("try to load ip_set_%s\n", name);
 	if (request_module("ip_set_%s", name) < 0) {
 		pr_warning("Can't find ip_set type %s\n", name);
 		nfnl_lock();
-		return -IPSET_ERR_FIND_TYPE;
+		return false;
 	}
 	nfnl_lock();
-	return -EAGAIN;
+	return true;
 }
 
 /* Find a set type and reference it */
+#define find_set_type_get(name, family, revision, found)	\
+	__find_set_type_get(name, family, revision, found, false)
+
 static int
-find_set_type_get(const char *name, u8 family, u8 revision,
-		  struct ip_set_type **found)
+__find_set_type_get(const char *name, u8 family, u8 revision,
+		    struct ip_set_type **found, bool retry)
 {
 	struct ip_set_type *type;
 	int err;
 
+	if (retry && !load_settype(name))
+		return -IPSET_ERR_FIND_TYPE;
+
 	rcu_read_lock();
 	*found = find_set_type(name, family, revision);
 	if (*found) {
 		err = !try_module_get((*found)->me) ? -EFAULT : 0;
 		goto unlock;
 	}
-	/* Make sure the type is loaded but we don't support the revision */
+	/* Make sure the type is already loaded
+	 * but we don't support the revision */
 	list_for_each_entry_rcu(type, &ip_set_type_list, list)
 		if (STREQ(type->name, name)) {
 			err = -IPSET_ERR_FIND_TYPE;
@@ -114,7 +121,8 @@ find_set_type_get(const char *name, u8 family, u8 revision,
 		}
 	rcu_read_unlock();
 
-	return try_to_load_type(name);
+	return retry ? -IPSET_ERR_FIND_TYPE :
+		__find_set_type_get(name, family, revision, found, true);
 
 unlock:
 	rcu_read_unlock();
@@ -125,12 +133,19 @@ unlock:
  * If we succeeded, the supported minimal and maximum revisions are
  * filled out.
  */
+#define find_set_type_minmax(name, family, min, max) \
+	__find_set_type_minmax(name, family, min, max, false)
+
 static int
-find_set_type_minmax(const char *name, u8 family, u8 *min, u8 *max)
+__find_set_type_minmax(const char *name, u8 family, u8 *min, u8 *max,
+		       bool retry)
 {
 	struct ip_set_type *type;
 	bool found = false;
 
+	if (retry && !load_settype(name))
+		return -IPSET_ERR_FIND_TYPE;
+
 	*min = 255; *max = 0;
 	rcu_read_lock();
 	list_for_each_entry_rcu(type, &ip_set_type_list, list)
@@ -146,7 +161,8 @@ find_set_type_minmax(const char *name, u8 family, u8 *min, u8 *max)
 	if (found)
 		return 0;
 
-	return try_to_load_type(name);
+	return retry ? -IPSET_ERR_FIND_TYPE :
+		__find_set_type_minmax(name, family, min, max, true);
 }
 
 #define family_name(f)	((f) == AF_INET ? "inet" : \
-- 
1.7.0.4

[PATCH 2/6] netfilter: ipset: Dumping error could lead to kernel BUG

[Jozsef Kadlecsik]

If there was a dumping error in the middle, the set-specific variable was
not zeroed out and thus the 'done' function of the dumping wrongly tried
to release the already released reference of the set. The already released
reference was caught by __ip_set_put and triggered a kernel BUG message.
The issue was reported by Jean-Philippe Menil.

Signed-off-by: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
---
 net/netfilter/ipset/ip_set_core.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/net/netfilter/ipset/ip_set_core.c b/net/netfilter/ipset/ip_set_core.c
index 5370fd8..7fa4a89 100644
--- a/net/netfilter/ipset/ip_set_core.c
+++ b/net/netfilter/ipset/ip_set_core.c
@@ -1143,6 +1143,7 @@ release_refcount:
 	if (ret || !cb->args[2]) {
 		pr_debug("release set %s\n", ip_set_list[index]->name);
 		ip_set_put_byindex(index);
+		cb->args[2] = 0;
 	}
 out:
 	if (nlh) {
-- 
1.7.0.4

[PATCH 3/6] netfilter: ipset: Remove unnecessary OOM logging messages

[Jozsef Kadlecsik]

From: Joe Perches <joe@perches.com>

Removing unnecessary messages saves code and text.

Site specific OOM messages are duplications of a generic MM
out of memory message and aren't really useful, so just
delete them.

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
---
 net/netfilter/ipset/ip_set_core.c |    4 +---
 1 files changed, 1 insertions(+), 3 deletions(-)

diff --git a/net/netfilter/ipset/ip_set_core.c b/net/netfilter/ipset/ip_set_core.c
index 7fa4a89..d5c4b5b 100644
--- a/net/netfilter/ipset/ip_set_core.c
+++ b/net/netfilter/ipset/ip_set_core.c
@@ -1717,10 +1717,8 @@ ip_set_init(void)
 
 	ip_set_list = kzalloc(sizeof(struct ip_set *) * ip_set_max,
 			      GFP_KERNEL);
-	if (!ip_set_list) {
-		pr_err("ip_set: Unable to create ip_set_list\n");
+	if (!ip_set_list)
 		return -ENOMEM;
-	}
 
 	ret = nfnetlink_subsys_register(&ip_set_netlink_subsys);
 	if (ret != 0) {
-- 
1.7.0.4

[PATCH 4/6] netfilter: ipset: avoid use of kernel-only types

[Jozsef Kadlecsik]

From: Jan Engelhardt <jengelh@medozas.de>

When using the xt_set.h header in userspace, one will get these gcc
reports:

ipset/ip_set.h:184:1: error: unknown type name "u16"
In file included from libxt_SET.c:21:0:
netfilter/xt_set.h:61:2: error: unknown type name "u32"
netfilter/xt_set.h:62:2: error: unknown type name "u32"

Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
Signed-off-by: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
---
 include/linux/netfilter/xt_set.h |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/include/linux/netfilter/xt_set.h b/include/linux/netfilter/xt_set.h
index c0405ac..e3a9978 100644
--- a/include/linux/netfilter/xt_set.h
+++ b/include/linux/netfilter/xt_set.h
@@ -58,8 +58,8 @@ struct xt_set_info_target_v1 {
 struct xt_set_info_target_v2 {
 	struct xt_set_info add_set;
 	struct xt_set_info del_set;
-	u32 flags;
-	u32 timeout;
+	__u32 flags;
+	__u32 timeout;
 };
 
 #endif /*_XT_SET_H*/
-- 
1.7.0.4

[PATCH 5/6] netfilter: ipset: expose userspace-relevant parts in ip_set.h

[Jozsef Kadlecsik]

From: Jan Engelhardt <jengelh@medozas.de>

iptables's libxt_SET.c depends on these.

Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
Signed-off-by: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
---
 include/linux/netfilter/ipset/ip_set.h |   26 ++++++++++++++------------
 1 files changed, 14 insertions(+), 12 deletions(-)

diff --git a/include/linux/netfilter/ipset/ip_set.h b/include/linux/netfilter/ipset/ip_set.h
index 3540c6e..c853158 100644
--- a/include/linux/netfilter/ipset/ip_set.h
+++ b/include/linux/netfilter/ipset/ip_set.h
@@ -11,6 +11,8 @@
  * published by the Free Software Foundation.
  */
 
+#include <linux/types.h>
+
 /* The protocol version */
 #define IPSET_PROTOCOL		6
 
@@ -168,19 +170,10 @@ enum ipset_adt {
 	IPSET_CADT_MAX,
 };
 
-#ifdef __KERNEL__
-#include <linux/ip.h>
-#include <linux/ipv6.h>
-#include <linux/netlink.h>
-#include <linux/netfilter.h>
-#include <linux/netfilter/x_tables.h>
-#include <linux/vmalloc.h>
-#include <net/netlink.h>
-
 /* Sets are identified by an index in kernel space. Tweak with ip_set_id_t
  * and IPSET_INVALID_ID if you want to increase the max number of sets.
  */
-typedef u16 ip_set_id_t;
+typedef __u16 ip_set_id_t;
 
 #define IPSET_INVALID_ID		65535
 
@@ -203,6 +196,15 @@ enum ip_set_kopt {
 	IPSET_DIM_THREE_SRC = (1 << IPSET_DIM_THREE),
 };
 
+#ifdef __KERNEL__
+#include <linux/ip.h>
+#include <linux/ipv6.h>
+#include <linux/netlink.h>
+#include <linux/netfilter.h>
+#include <linux/netfilter/x_tables.h>
+#include <linux/vmalloc.h>
+#include <net/netlink.h>
+
 /* Set features */
 enum ip_set_feature {
 	IPSET_TYPE_IP_FLAG = 0,
@@ -450,6 +452,8 @@ bitmap_bytes(u32 a, u32 b)
 	return 4 * ((((b - a + 8) / 8) + 3) / 4);
 }
 
+#endif /* __KERNEL__ */
+
 /* Interface to iptables/ip6tables */
 
 #define SO_IP_SET		83
@@ -475,6 +479,4 @@ struct ip_set_req_version {
 	unsigned version;
 };
 
-#endif	/* __KERNEL__ */
-
 #endif /*_IP_SET_H */
-- 
1.7.0.4

[PATCH 6/6] netfilter: ipset: use NFPROTO_ constants

[Jozsef Kadlecsik]

From: Jan Engelhardt <jengelh@medozas.de>

ipset is actually using NFPROTO values rather than AF (xt_set passes
that along).

Signed-off-by: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
---
 include/linux/netfilter/ipset/ip_set.h      |    5 ++++-
 net/netfilter/ipset/ip_set_bitmap_ip.c      |    4 ++--
 net/netfilter/ipset/ip_set_bitmap_ipmac.c   |    4 ++--
 net/netfilter/ipset/ip_set_bitmap_port.c    |    4 ++--
 net/netfilter/ipset/ip_set_core.c           |   16 ++++++++--------
 net/netfilter/ipset/ip_set_getport.c        |    4 ++--
 net/netfilter/ipset/ip_set_hash_ip.c        |   18 +++++++++---------
 net/netfilter/ipset/ip_set_hash_ipport.c    |   10 +++++-----
 net/netfilter/ipset/ip_set_hash_ipportip.c  |   10 +++++-----
 net/netfilter/ipset/ip_set_hash_ipportnet.c |   12 ++++++------
 net/netfilter/ipset/ip_set_hash_net.c       |   12 ++++++------
 net/netfilter/ipset/ip_set_hash_netiface.c  |   12 ++++++------
 net/netfilter/ipset/ip_set_hash_netport.c   |   12 ++++++------
 net/netfilter/ipset/ip_set_list_set.c       |    2 +-
 14 files changed, 64 insertions(+), 61 deletions(-)

diff --git a/include/linux/netfilter/ipset/ip_set.h b/include/linux/netfilter/ipset/ip_set.h
index c853158..e921766 100644
--- a/include/linux/netfilter/ipset/ip_set.h
+++ b/include/linux/netfilter/ipset/ip_set.h
@@ -290,7 +290,10 @@ struct ip_set_type {
 	u8 features;
 	/* Set type dimension */
 	u8 dimension;
-	/* Supported family: may be AF_UNSPEC for both AF_INET/AF_INET6 */
+	/*
+	 * Supported family: may be NFPROTO_UNSPEC for both
+	 * NFPROTO_IPV4/NFPROTO_IPV6.
+	 */
 	u8 family;
 	/* Type revisions */
 	u8 revision_min, revision_max;
diff --git a/net/netfilter/ipset/ip_set_bitmap_ip.c b/net/netfilter/ipset/ip_set_bitmap_ip.c
index e3e7399..a72a4df 100644
--- a/net/netfilter/ipset/ip_set_bitmap_ip.c
+++ b/net/netfilter/ipset/ip_set_bitmap_ip.c
@@ -442,7 +442,7 @@ init_map_ip(struct ip_set *set, struct bitmap_ip *map,
 	map->timeout = IPSET_NO_TIMEOUT;
 
 	set->data = map;
-	set->family = AF_INET;
+	set->family = NFPROTO_IPV4;
 
 	return true;
 }
@@ -550,7 +550,7 @@ static struct ip_set_type bitmap_ip_type __read_mostly = {
 	.protocol	= IPSET_PROTOCOL,
 	.features	= IPSET_TYPE_IP,
 	.dimension	= IPSET_DIM_ONE,
-	.family		= AF_INET,
+	.family		= NFPROTO_IPV4,
 	.revision_min	= 0,
 	.revision_max	= 0,
 	.create		= bitmap_ip_create,
diff --git a/net/netfilter/ipset/ip_set_bitmap_ipmac.c b/net/netfilter/ipset/ip_set_bitmap_ipmac.c
index 56096f5..81324c1 100644
--- a/net/netfilter/ipset/ip_set_bitmap_ipmac.c
+++ b/net/netfilter/ipset/ip_set_bitmap_ipmac.c
@@ -543,7 +543,7 @@ init_map_ipmac(struct ip_set *set, struct bitmap_ipmac *map,
 	map->timeout = IPSET_NO_TIMEOUT;
 
 	set->data = map;
-	set->family = AF_INET;
+	set->family = NFPROTO_IPV4;
 
 	return true;
 }
@@ -623,7 +623,7 @@ static struct ip_set_type bitmap_ipmac_type = {
 	.protocol	= IPSET_PROTOCOL,
 	.features	= IPSET_TYPE_IP | IPSET_TYPE_MAC,
 	.dimension	= IPSET_DIM_TWO,
-	.family		= AF_INET,
+	.family		= NFPROTO_IPV4,
 	.revision_min	= 0,
 	.revision_max	= 0,
 	.create		= bitmap_ipmac_create,
diff --git a/net/netfilter/ipset/ip_set_bitmap_port.c b/net/netfilter/ipset/ip_set_bitmap_port.c
index 29ba93b..382ec28 100644
--- a/net/netfilter/ipset/ip_set_bitmap_port.c
+++ b/net/netfilter/ipset/ip_set_bitmap_port.c
@@ -422,7 +422,7 @@ init_map_port(struct ip_set *set, struct bitmap_port *map,
 	map->timeout = IPSET_NO_TIMEOUT;
 
 	set->data = map;
-	set->family = AF_UNSPEC;
+	set->family = NFPROTO_UNSPEC;
 
 	return true;
 }
@@ -483,7 +483,7 @@ static struct ip_set_type bitmap_port_type = {
 	.protocol	= IPSET_PROTOCOL,
 	.features	= IPSET_TYPE_PORT,
 	.dimension	= IPSET_DIM_ONE,
-	.family		= AF_UNSPEC,
+	.family		= NFPROTO_UNSPEC,
 	.revision_min	= 0,
 	.revision_max	= 0,
 	.create		= bitmap_port_create,
diff --git a/net/netfilter/ipset/ip_set_core.c b/net/netfilter/ipset/ip_set_core.c
index d5c4b5b..65fc2e1 100644
--- a/net/netfilter/ipset/ip_set_core.c
+++ b/net/netfilter/ipset/ip_set_core.c
@@ -70,7 +70,7 @@ find_set_type(const char *name, u8 family, u8 revision)
 
 	list_for_each_entry_rcu(type, &ip_set_type_list, list)
 		if (STREQ(type->name, name) &&
-		    (type->family == family || type->family == AF_UNSPEC) &&
+		    (type->family == family || type->family == NFPROTO_UNSPEC) &&
 		    revision >= type->revision_min &&
 		    revision <= type->revision_max)
 			return type;
@@ -150,7 +150,7 @@ __find_set_type_minmax(const char *name, u8 family, u8 *min, u8 *max,
 	rcu_read_lock();
 	list_for_each_entry_rcu(type, &ip_set_type_list, list)
 		if (STREQ(type->name, name) &&
-		    (type->family == family || type->family == AF_UNSPEC)) {
+		    (type->family == family || type->family == NFPROTO_UNSPEC)) {
 			found = true;
 			if (type->revision_min < *min)
 				*min = type->revision_min;
@@ -165,8 +165,8 @@ __find_set_type_minmax(const char *name, u8 family, u8 *min, u8 *max,
 		__find_set_type_minmax(name, family, min, max, true);
 }
 
-#define family_name(f)	((f) == AF_INET ? "inet" : \
-			 (f) == AF_INET6 ? "inet6" : "any")
+#define family_name(f)	((f) == NFPROTO_IPV4 ? "inet" : \
+			 (f) == NFPROTO_IPV6 ? "inet6" : "any")
 
 /* Register a set type structure. The type is identified by
  * the unique triple of name, family and revision.
@@ -355,7 +355,7 @@ ip_set_test(ip_set_id_t index, const struct sk_buff *skb,
 	pr_debug("set %s, index %u\n", set->name, index);
 
 	if (opt->dim < set->type->dimension ||
-	    !(opt->family == set->family || set->family == AF_UNSPEC))
+	    !(opt->family == set->family || set->family == NFPROTO_UNSPEC))
 		return 0;
 
 	read_lock_bh(&set->lock);
@@ -388,7 +388,7 @@ ip_set_add(ip_set_id_t index, const struct sk_buff *skb,
 	pr_debug("set %s, index %u\n", set->name, index);
 
 	if (opt->dim < set->type->dimension ||
-	    !(opt->family == set->family || set->family == AF_UNSPEC))
+	    !(opt->family == set->family || set->family == NFPROTO_UNSPEC))
 		return 0;
 
 	write_lock_bh(&set->lock);
@@ -411,7 +411,7 @@ ip_set_del(ip_set_id_t index, const struct sk_buff *skb,
 	pr_debug("set %s, index %u\n", set->name, index);
 
 	if (opt->dim < set->type->dimension ||
-	    !(opt->family == set->family || set->family == AF_UNSPEC))
+	    !(opt->family == set->family || set->family == NFPROTO_UNSPEC))
 		return 0;
 
 	write_lock_bh(&set->lock);
@@ -576,7 +576,7 @@ start_msg(struct sk_buff *skb, u32 pid, u32 seq, unsigned int flags,
 		return NULL;
 
 	nfmsg = nlmsg_data(nlh);
-	nfmsg->nfgen_family = AF_INET;
+	nfmsg->nfgen_family = NFPROTO_IPV4;
 	nfmsg->version = NFNETLINK_V0;
 	nfmsg->res_id = 0;
 
diff --git a/net/netfilter/ipset/ip_set_getport.c b/net/netfilter/ipset/ip_set_getport.c
index 757143b..58ca4e1 100644
--- a/net/netfilter/ipset/ip_set_getport.c
+++ b/net/netfilter/ipset/ip_set_getport.c
@@ -133,10 +133,10 @@ ip_set_get_ip_port(const struct sk_buff *skb, u8 pf, bool src, __be16 *port)
 	u8 proto;
 
 	switch (pf) {
-	case AF_INET:
+	case NFPROTO_IPV4:
 		ret = ip_set_get_ip4_port(skb, src, port, &proto);
 		break;
-	case AF_INET6:
+	case NFPROTO_IPV6:
 		ret = ip_set_get_ip6_port(skb, src, port, &proto);
 		break;
 	default:
diff --git a/net/netfilter/ipset/ip_set_hash_ip.c b/net/netfilter/ipset/ip_set_hash_ip.c
index f2d576e..14a8628 100644
--- a/net/netfilter/ipset/ip_set_hash_ip.c
+++ b/net/netfilter/ipset/ip_set_hash_ip.c
@@ -366,11 +366,11 @@ hash_ip_create(struct ip_set *set, struct nlattr *tb[], u32 flags)
 	u8 netmask, hbits;
 	struct ip_set_hash *h;
 
-	if (!(set->family == AF_INET || set->family == AF_INET6))
+	if (!(set->family == NFPROTO_IPV4 || set->family == NFPROTO_IPV6))
 		return -IPSET_ERR_INVALID_FAMILY;
-	netmask = set->family == AF_INET ? 32 : 128;
+	netmask = set->family == NFPROTO_IPV4 ? 32 : 128;
 	pr_debug("Create set %s with family %s\n",
-		 set->name, set->family == AF_INET ? "inet" : "inet6");
+		 set->name, set->family == NFPROTO_IPV4 ? "inet" : "inet6");
 
 	if (unlikely(!ip_set_optattr_netorder(tb, IPSET_ATTR_HASHSIZE) ||
 		     !ip_set_optattr_netorder(tb, IPSET_ATTR_MAXELEM) ||
@@ -389,8 +389,8 @@ hash_ip_create(struct ip_set *set, struct nlattr *tb[], u32 flags)
 	if (tb[IPSET_ATTR_NETMASK]) {
 		netmask = nla_get_u8(tb[IPSET_ATTR_NETMASK]);
 
-		if ((set->family == AF_INET && netmask > 32) ||
-		    (set->family == AF_INET6 && netmask > 128) ||
+		if ((set->family == NFPROTO_IPV4 && netmask > 32) ||
+		    (set->family == NFPROTO_IPV6 && netmask > 128) ||
 		    netmask == 0)
 			return -IPSET_ERR_INVALID_NETMASK;
 	}
@@ -419,15 +419,15 @@ hash_ip_create(struct ip_set *set, struct nlattr *tb[], u32 flags)
 	if (tb[IPSET_ATTR_TIMEOUT]) {
 		h->timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]);
 
-		set->variant = set->family == AF_INET
+		set->variant = set->family == NFPROTO_IPV4
 			? &hash_ip4_tvariant : &hash_ip6_tvariant;
 
-		if (set->family == AF_INET)
+		if (set->family == NFPROTO_IPV4)
 			hash_ip4_gc_init(set);
 		else
 			hash_ip6_gc_init(set);
 	} else {
-		set->variant = set->family == AF_INET
+		set->variant = set->family == NFPROTO_IPV4
 			? &hash_ip4_variant : &hash_ip6_variant;
 	}
 
@@ -443,7 +443,7 @@ static struct ip_set_type hash_ip_type __read_mostly = {
 	.protocol	= IPSET_PROTOCOL,
 	.features	= IPSET_TYPE_IP,
 	.dimension	= IPSET_DIM_ONE,
-	.family		= AF_UNSPEC,
+	.family		= NFPROTO_UNSPEC,
 	.revision_min	= 0,
 	.revision_max	= 0,
 	.create		= hash_ip_create,
diff --git a/net/netfilter/ipset/ip_set_hash_ipport.c b/net/netfilter/ipset/ip_set_hash_ipport.c
index 6ee10f5..30a6273 100644
--- a/net/netfilter/ipset/ip_set_hash_ipport.c
+++ b/net/netfilter/ipset/ip_set_hash_ipport.c
@@ -450,7 +450,7 @@ hash_ipport_create(struct ip_set *set, struct nlattr *tb[], u32 flags)
 	u32 hashsize = IPSET_DEFAULT_HASHSIZE, maxelem = IPSET_DEFAULT_MAXELEM;
 	u8 hbits;
 
-	if (!(set->family == AF_INET || set->family == AF_INET6))
+	if (!(set->family == NFPROTO_IPV4 || set->family == NFPROTO_IPV6))
 		return -IPSET_ERR_INVALID_FAMILY;
 
 	if (unlikely(!ip_set_optattr_netorder(tb, IPSET_ATTR_HASHSIZE) ||
@@ -490,15 +490,15 @@ hash_ipport_create(struct ip_set *set, struct nlattr *tb[], u32 flags)
 	if (tb[IPSET_ATTR_TIMEOUT]) {
 		h->timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]);
 
-		set->variant = set->family == AF_INET
+		set->variant = set->family == NFPROTO_IPV4
 			? &hash_ipport4_tvariant : &hash_ipport6_tvariant;
 
-		if (set->family == AF_INET)
+		if (set->family == NFPROTO_IPV4)
 			hash_ipport4_gc_init(set);
 		else
 			hash_ipport6_gc_init(set);
 	} else {
-		set->variant = set->family == AF_INET
+		set->variant = set->family == NFPROTO_IPV4
 			? &hash_ipport4_variant : &hash_ipport6_variant;
 	}
 
@@ -514,7 +514,7 @@ static struct ip_set_type hash_ipport_type __read_mostly = {
 	.protocol	= IPSET_PROTOCOL,
 	.features	= IPSET_TYPE_IP | IPSET_TYPE_PORT,
 	.dimension	= IPSET_DIM_TWO,
-	.family		= AF_UNSPEC,
+	.family		= NFPROTO_UNSPEC,
 	.revision_min	= 0,
 	.revision_max	= 1,	/* SCTP and UDPLITE support added */
 	.create		= hash_ipport_create,
diff --git a/net/netfilter/ipset/ip_set_hash_ipportip.c b/net/netfilter/ipset/ip_set_hash_ipportip.c
index fb90e34..55de642 100644
--- a/net/netfilter/ipset/ip_set_hash_ipportip.c
+++ b/net/netfilter/ipset/ip_set_hash_ipportip.c
@@ -468,7 +468,7 @@ hash_ipportip_create(struct ip_set *set, struct nlattr *tb[], u32 flags)
 	u32 hashsize = IPSET_DEFAULT_HASHSIZE, maxelem = IPSET_DEFAULT_MAXELEM;
 	u8 hbits;
 
-	if (!(set->family == AF_INET || set->family == AF_INET6))
+	if (!(set->family == NFPROTO_IPV4 || set->family == NFPROTO_IPV6))
 		return -IPSET_ERR_INVALID_FAMILY;
 
 	if (unlikely(!ip_set_optattr_netorder(tb, IPSET_ATTR_HASHSIZE) ||
@@ -508,15 +508,15 @@ hash_ipportip_create(struct ip_set *set, struct nlattr *tb[], u32 flags)
 	if (tb[IPSET_ATTR_TIMEOUT]) {
 		h->timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]);
 
-		set->variant = set->family == AF_INET
+		set->variant = set->family == NFPROTO_IPV4
 			? &hash_ipportip4_tvariant : &hash_ipportip6_tvariant;
 
-		if (set->family == AF_INET)
+		if (set->family == NFPROTO_IPV4)
 			hash_ipportip4_gc_init(set);
 		else
 			hash_ipportip6_gc_init(set);
 	} else {
-		set->variant = set->family == AF_INET
+		set->variant = set->family == NFPROTO_IPV4
 			? &hash_ipportip4_variant : &hash_ipportip6_variant;
 	}
 
@@ -532,7 +532,7 @@ static struct ip_set_type hash_ipportip_type __read_mostly = {
 	.protocol	= IPSET_PROTOCOL,
 	.features	= IPSET_TYPE_IP | IPSET_TYPE_PORT | IPSET_TYPE_IP2,
 	.dimension	= IPSET_DIM_THREE,
-	.family		= AF_UNSPEC,
+	.family		= NFPROTO_UNSPEC,
 	.revision_min	= 0,
 	.revision_max	= 1,	/* SCTP and UDPLITE support added */
 	.create		= hash_ipportip_create,
diff --git a/net/netfilter/ipset/ip_set_hash_ipportnet.c b/net/netfilter/ipset/ip_set_hash_ipportnet.c
index deb3e3d..6ee4f72 100644
--- a/net/netfilter/ipset/ip_set_hash_ipportnet.c
+++ b/net/netfilter/ipset/ip_set_hash_ipportnet.c
@@ -554,7 +554,7 @@ hash_ipportnet_create(struct ip_set *set, struct nlattr *tb[], u32 flags)
 	u32 hashsize = IPSET_DEFAULT_HASHSIZE, maxelem = IPSET_DEFAULT_MAXELEM;
 	u8 hbits;
 
-	if (!(set->family == AF_INET || set->family == AF_INET6))
+	if (!(set->family == NFPROTO_IPV4 || set->family == NFPROTO_IPV6))
 		return -IPSET_ERR_INVALID_FAMILY;
 
 	if (unlikely(!ip_set_optattr_netorder(tb, IPSET_ATTR_HASHSIZE) ||
@@ -573,7 +573,7 @@ hash_ipportnet_create(struct ip_set *set, struct nlattr *tb[], u32 flags)
 
 	h = kzalloc(sizeof(*h)
 		    + sizeof(struct ip_set_hash_nets)
-		      * (set->family == AF_INET ? 32 : 128), GFP_KERNEL);
+		      * (set->family == NFPROTO_IPV4 ? 32 : 128), GFP_KERNEL);
 	if (!h)
 		return -ENOMEM;
 
@@ -596,16 +596,16 @@ hash_ipportnet_create(struct ip_set *set, struct nlattr *tb[], u32 flags)
 	if (tb[IPSET_ATTR_TIMEOUT]) {
 		h->timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]);
 
-		set->variant = set->family == AF_INET
+		set->variant = set->family == NFPROTO_IPV4
 			? &hash_ipportnet4_tvariant
 			: &hash_ipportnet6_tvariant;
 
-		if (set->family == AF_INET)
+		if (set->family == NFPROTO_IPV4)
 			hash_ipportnet4_gc_init(set);
 		else
 			hash_ipportnet6_gc_init(set);
 	} else {
-		set->variant = set->family == AF_INET
+		set->variant = set->family == NFPROTO_IPV4
 			? &hash_ipportnet4_variant : &hash_ipportnet6_variant;
 	}
 
@@ -621,7 +621,7 @@ static struct ip_set_type hash_ipportnet_type __read_mostly = {
 	.protocol	= IPSET_PROTOCOL,
 	.features	= IPSET_TYPE_IP | IPSET_TYPE_PORT | IPSET_TYPE_IP2,
 	.dimension	= IPSET_DIM_THREE,
-	.family		= AF_UNSPEC,
+	.family		= NFPROTO_UNSPEC,
 	.revision_min	= 0,
 	/*		  1	   SCTP and UDPLITE support added */
 	.revision_max	= 2,	/* Range as input support for IPv4 added */
diff --git a/net/netfilter/ipset/ip_set_hash_net.c b/net/netfilter/ipset/ip_set_hash_net.c
index 60d0165..48e35ba 100644
--- a/net/netfilter/ipset/ip_set_hash_net.c
+++ b/net/netfilter/ipset/ip_set_hash_net.c
@@ -406,7 +406,7 @@ hash_net_create(struct ip_set *set, struct nlattr *tb[], u32 flags)
 	struct ip_set_hash *h;
 	u8 hbits;
 
-	if (!(set->family == AF_INET || set->family == AF_INET6))
+	if (!(set->family == NFPROTO_IPV4 || set->family == NFPROTO_IPV6))
 		return -IPSET_ERR_INVALID_FAMILY;
 
 	if (unlikely(!ip_set_optattr_netorder(tb, IPSET_ATTR_HASHSIZE) ||
@@ -425,7 +425,7 @@ hash_net_create(struct ip_set *set, struct nlattr *tb[], u32 flags)
 
 	h = kzalloc(sizeof(*h)
 		    + sizeof(struct ip_set_hash_nets)
-		      * (set->family == AF_INET ? 32 : 128), GFP_KERNEL);
+		      * (set->family == NFPROTO_IPV4 ? 32 : 128), GFP_KERNEL);
 	if (!h)
 		return -ENOMEM;
 
@@ -448,15 +448,15 @@ hash_net_create(struct ip_set *set, struct nlattr *tb[], u32 flags)
 	if (tb[IPSET_ATTR_TIMEOUT]) {
 		h->timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]);
 
-		set->variant = set->family == AF_INET
+		set->variant = set->family == NFPROTO_IPV4
 			? &hash_net4_tvariant : &hash_net6_tvariant;
 
-		if (set->family == AF_INET)
+		if (set->family == NFPROTO_IPV4)
 			hash_net4_gc_init(set);
 		else
 			hash_net6_gc_init(set);
 	} else {
-		set->variant = set->family == AF_INET
+		set->variant = set->family == NFPROTO_IPV4
 			? &hash_net4_variant : &hash_net6_variant;
 	}
 
@@ -472,7 +472,7 @@ static struct ip_set_type hash_net_type __read_mostly = {
 	.protocol	= IPSET_PROTOCOL,
 	.features	= IPSET_TYPE_IP,
 	.dimension	= IPSET_DIM_ONE,
-	.family		= AF_UNSPEC,
+	.family		= NFPROTO_UNSPEC,
 	.revision_min	= 0,
 	.revision_max	= 1,	/* Range as input support for IPv4 added */
 	.create		= hash_net_create,
diff --git a/net/netfilter/ipset/ip_set_hash_netiface.c b/net/netfilter/ipset/ip_set_hash_netiface.c
index e13095d..a9fb4af 100644
--- a/net/netfilter/ipset/ip_set_hash_netiface.c
+++ b/net/netfilter/ipset/ip_set_hash_netiface.c
@@ -678,7 +678,7 @@ hash_netiface_create(struct ip_set *set, struct nlattr *tb[], u32 flags)
 	u32 hashsize = IPSET_DEFAULT_HASHSIZE, maxelem = IPSET_DEFAULT_MAXELEM;
 	u8 hbits;
 
-	if (!(set->family == AF_INET || set->family == AF_INET6))
+	if (!(set->family == NFPROTO_IPV4 || set->family == NFPROTO_IPV6))
 		return -IPSET_ERR_INVALID_FAMILY;
 
 	if (unlikely(!ip_set_optattr_netorder(tb, IPSET_ATTR_HASHSIZE) ||
@@ -697,7 +697,7 @@ hash_netiface_create(struct ip_set *set, struct nlattr *tb[], u32 flags)
 
 	h = kzalloc(sizeof(*h)
 		    + sizeof(struct ip_set_hash_nets)
-		      * (set->family == AF_INET ? 32 : 128), GFP_KERNEL);
+		      * (set->family == NFPROTO_IPV4 ? 32 : 128), GFP_KERNEL);
 	if (!h)
 		return -ENOMEM;
 
@@ -722,15 +722,15 @@ hash_netiface_create(struct ip_set *set, struct nlattr *tb[], u32 flags)
 	if (tb[IPSET_ATTR_TIMEOUT]) {
 		h->timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]);
 
-		set->variant = set->family == AF_INET
+		set->variant = set->family == NFPROTO_IPV4
 			? &hash_netiface4_tvariant : &hash_netiface6_tvariant;
 
-		if (set->family == AF_INET)
+		if (set->family == NFPROTO_IPV4)
 			hash_netiface4_gc_init(set);
 		else
 			hash_netiface6_gc_init(set);
 	} else {
-		set->variant = set->family == AF_INET
+		set->variant = set->family == NFPROTO_IPV4
 			? &hash_netiface4_variant : &hash_netiface6_variant;
 	}
 
@@ -746,7 +746,7 @@ static struct ip_set_type hash_netiface_type __read_mostly = {
 	.protocol	= IPSET_PROTOCOL,
 	.features	= IPSET_TYPE_IP | IPSET_TYPE_IFACE,
 	.dimension	= IPSET_DIM_TWO,
-	.family		= AF_UNSPEC,
+	.family		= NFPROTO_UNSPEC,
 	.revision_min	= 0,
 	.create		= hash_netiface_create,
 	.create_policy	= {
diff --git a/net/netfilter/ipset/ip_set_hash_netport.c b/net/netfilter/ipset/ip_set_hash_netport.c
index 8f9de72..1fcc102 100644
--- a/net/netfilter/ipset/ip_set_hash_netport.c
+++ b/net/netfilter/ipset/ip_set_hash_netport.c
@@ -507,7 +507,7 @@ hash_netport_create(struct ip_set *set, struct nlattr *tb[], u32 flags)
 	u32 hashsize = IPSET_DEFAULT_HASHSIZE, maxelem = IPSET_DEFAULT_MAXELEM;
 	u8 hbits;
 
-	if (!(set->family == AF_INET || set->family == AF_INET6))
+	if (!(set->family == NFPROTO_IPV4 || set->family == NFPROTO_IPV6))
 		return -IPSET_ERR_INVALID_FAMILY;
 
 	if (unlikely(!ip_set_optattr_netorder(tb, IPSET_ATTR_HASHSIZE) ||
@@ -526,7 +526,7 @@ hash_netport_create(struct ip_set *set, struct nlattr *tb[], u32 flags)
 
 	h = kzalloc(sizeof(*h)
 		    + sizeof(struct ip_set_hash_nets)
-		      * (set->family == AF_INET ? 32 : 128), GFP_KERNEL);
+		      * (set->family == NFPROTO_IPV4 ? 32 : 128), GFP_KERNEL);
 	if (!h)
 		return -ENOMEM;
 
@@ -549,15 +549,15 @@ hash_netport_create(struct ip_set *set, struct nlattr *tb[], u32 flags)
 	if (tb[IPSET_ATTR_TIMEOUT]) {
 		h->timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]);
 
-		set->variant = set->family == AF_INET
+		set->variant = set->family == NFPROTO_IPV4
 			? &hash_netport4_tvariant : &hash_netport6_tvariant;
 
-		if (set->family == AF_INET)
+		if (set->family == NFPROTO_IPV4)
 			hash_netport4_gc_init(set);
 		else
 			hash_netport6_gc_init(set);
 	} else {
-		set->variant = set->family == AF_INET
+		set->variant = set->family == NFPROTO_IPV4
 			? &hash_netport4_variant : &hash_netport6_variant;
 	}
 
@@ -573,7 +573,7 @@ static struct ip_set_type hash_netport_type __read_mostly = {
 	.protocol	= IPSET_PROTOCOL,
 	.features	= IPSET_TYPE_IP | IPSET_TYPE_PORT,
 	.dimension	= IPSET_DIM_TWO,
-	.family		= AF_UNSPEC,
+	.family		= NFPROTO_UNSPEC,
 	.revision_min	= 0,
 	/*		  1	   SCTP and UDPLITE support added */
 	.revision_max	= 2,	/* Range as input support for IPv4 added */
diff --git a/net/netfilter/ipset/ip_set_list_set.c b/net/netfilter/ipset/ip_set_list_set.c
index 4d10819..7e095f9 100644
--- a/net/netfilter/ipset/ip_set_list_set.c
+++ b/net/netfilter/ipset/ip_set_list_set.c
@@ -575,7 +575,7 @@ static struct ip_set_type list_set_type __read_mostly = {
 	.protocol	= IPSET_PROTOCOL,
 	.features	= IPSET_TYPE_NAME | IPSET_DUMP_LAST,
 	.dimension	= IPSET_DIM_ONE,
-	.family		= AF_UNSPEC,
+	.family		= NFPROTO_UNSPEC,
 	.revision_min	= 0,
 	.revision_max	= 0,
 	.create		= list_set_create,
-- 
1.7.0.4

Re: [PATCH 0/6] ipset patches

[Pablo Neira Ayuso]

On Tue, Sep 06, 2011 at 09:59:01PM +0200, Jozsef Kadlecsik wrote:
> Hi Patrick and Pablo,
> 
> Here follows the patches which brings the kernel sync with ipset 6.9.1.
> The first two are important bugfixes. Please consider applying them.

AFAIK, all trees are frozen in master.kernel.org due to the security
issue. It seems we'll have to wait until we can apply these patches.

Re: [PATCH 0/6] ipset patches

[Jozsef Kadlecsik]

On Wed, 7 Sep 2011, Pablo Neira Ayuso wrote:

> On Tue, Sep 06, 2011 at 09:59:01PM +0200, Jozsef Kadlecsik wrote:
> > Hi Patrick and Pablo,
> > 
> > Here follows the patches which brings the kernel sync with ipset 6.9.1.
> > The first two are important bugfixes. Please consider applying them.
> 
> AFAIK, all trees are frozen in master.kernel.org due to the security
> issue. It seems we'll have to wait until we can apply these patches.

Yeah, I know. But with the new ipset release it seemed best to send the 
patches anyway and wait for the cleaning up and verifications at 
kernel.org.

Best regards,
Jozsef
-
E-mail  : kadlec@blackhole.kfki.hu, kadlec@mail.kfki.hu
PGP key : http://www.kfki.hu/~kadlec/pgp_public_key.txt
Address : KFKI Research Institute for Particle and Nuclear Physics
          H-1525 Budapest 114, POB. 49, Hungary

Re: [PATCH 0/6] ipset patches

[Patrick McHardy]

On 07.09.2011 14:04, Jozsef Kadlecsik wrote:
> On Wed, 7 Sep 2011, Pablo Neira Ayuso wrote:
> 
>> On Tue, Sep 06, 2011 at 09:59:01PM +0200, Jozsef Kadlecsik wrote:
>>> Hi Patrick and Pablo,
>>>
>>> Here follows the patches which brings the kernel sync with ipset 6.9.1.
>>> The first two are important bugfixes. Please consider applying them.
>>
>> AFAIK, all trees are frozen in master.kernel.org due to the security
>> issue. It seems we'll have to wait until we can apply these patches.
> 
> Yeah, I know. But with the new ipset release it seemed best to send the 
> patches anyway and wait for the cleaning up and verifications at 
> kernel.org.

I think for now I'll apply all patches to my local tree and will either
push them out to kernel.org in case it is available again during the
next days or temporarily move the tree to somewhere else.

Re: [PATCH 1/6] netfilter: ipset: Autoload set type modules safely

[Patrick McHardy]

On 06.09.2011 21:59, Jozsef Kadlecsik wrote:
> Jan Engelhardt noticed when userspace requests a set type unknown
> to the kernel, it can lead to a loop due to the unsafe type module
> loading. The issue is fixed in this patch.

Applied, thanks.

Re: [PATCH 2/6] netfilter: ipset: Dumping error could lead to kernel BUG

[Patrick McHardy]

On 06.09.2011 21:59, Jozsef Kadlecsik wrote:
> If there was a dumping error in the middle, the set-specific variable was
> not zeroed out and thus the 'done' function of the dumping wrongly tried
> to release the already released reference of the set. The already released
> reference was caught by __ip_set_put and triggered a kernel BUG message.
> The issue was reported by Jean-Philippe Menil.

Applied, thanks.

Re: [PATCH 3/6] netfilter: ipset: Remove unnecessary OOM logging messages

[Patrick McHardy]

On 06.09.2011 21:59, Jozsef Kadlecsik wrote:
> From: Joe Perches <joe@perches.com>
> 
> Removing unnecessary messages saves code and text.
> 
> Site specific OOM messages are duplications of a generic MM
> out of memory message and aren't really useful, so just
> delete them.

This one is already in my tree.

Re: [PATCH 4/6] netfilter: ipset: avoid use of kernel-only types

[Patrick McHardy]

On 06.09.2011 21:59, Jozsef Kadlecsik wrote:
> When using the xt_set.h header in userspace, one will get these gcc
> reports:
> 
> ipset/ip_set.h:184:1: error: unknown type name "u16"
> In file included from libxt_SET.c:21:0:
> netfilter/xt_set.h:61:2: error: unknown type name "u32"
> netfilter/xt_set.h:62:2: error: unknown type name "u32"

Applied, thanks.

Re: [PATCH 5/6] netfilter: ipset: expose userspace-relevant parts in ip_set.h

[Patrick McHardy]

On 06.09.2011 21:59, Jozsef Kadlecsik wrote:
> iptables's libxt_SET.c depends on these.

Applied, thanks.

Re: [PATCH 6/6] netfilter: ipset: use NFPROTO_ constants

[Patrick McHardy]

On 06.09.2011 21:59, Jozsef Kadlecsik wrote:
> ipset is actually using NFPROTO values rather than AF (xt_set passes
> that along).
> 

Also applied, thanks.