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

Pablo Neira Ayuso says: ==================== Netfilter fixes for net The following patchset contains Netfilter fixes for net: 1) Restrict range element expansion in ipset to avoid soft lockup, from Jozsef Kadlecsik. 2) Memleak in error path for nf_conntrack_bridge for IPv4 packets, from Yajun Deng. 3) Simplify conntrack garbage collection strategy to avoid frequent wake-ups, from Florian Westphal. 4) Fix NFNLA_HOOK_FUNCTION_NAME string, do not include module name. 5) Missing chain family netlink attribute in chain description in nfnetlink_hook. 6) Incorrect sequence number on nfnetlink_hook dumps. 7) Use netlink request family in reply message for consistency. 8) Remove offload_pickup sysctl, use conntrack for established state instead, from Florian Westphal. 9) Translate NFPROTO_INET/ingress to NFPROTO_NETDEV/ingress, since NFPROTO_INET is not exposed through nfnetlink_hook. * git://git.kernel.org/pub/scm/linux/kernel/git/pablo/nf: netfilter: nfnetlink_hook: translate inet ingress to netdev netfilter: conntrack: remove offload_pickup sysctl again netfilter: nfnetlink_hook: Use same family as request message netfilter: nfnetlink_hook: use the sequence number of the request message netfilter: nfnetlink_hook: missing chain family netfilter: nfnetlink_hook: strip off module name from hookfn netfilter: conntrack: collect all entries in one cycle netfilter: nf_conntrack_bridge: Fix memory leak when error netfilter: ipset: Limit the maximal range of consecutive elements to add/delete ==================== Link: https://lore.kernel.org/r/20210806151149.6356-1-pablo@netfilter.org Signed-off-by: Jakub Kicinski <kuba@kernel.org>
torvalds · Aug 6, 2021 · cc4e5ee · cc4e5ee
2 parents 704e624 + 269fc69
commit cc4e5ee
Show file tree

Hide file tree

Showing 21 changed files with 151 additions and 95 deletions.
diff --git a/Documentation/networking/nf_conntrack-sysctl.rst b/Documentation/networking/nf_conntrack-sysctl.rst
@@ -191,19 +191,9 @@ nf_flowtable_tcp_timeout - INTEGER (seconds)
         TCP connections may be offloaded from nf conntrack to nf flow table.
         Once aged, the connection is returned to nf conntrack with tcp pickup timeout.
 
-nf_flowtable_tcp_pickup - INTEGER (seconds)
-        default 120
-
-        TCP connection timeout after being aged from nf flow table offload.
-
 nf_flowtable_udp_timeout - INTEGER (seconds)
         default 30
 
         Control offload timeout for udp connections.
         UDP connections may be offloaded from nf conntrack to nf flow table.
         Once aged, the connection is returned to nf conntrack with udp pickup timeout.
-
-nf_flowtable_udp_pickup - INTEGER (seconds)
-        default 30
-
-        UDP connection timeout after being aged from nf flow table offload.
diff --git a/include/linux/netfilter/ipset/ip_set.h b/include/linux/netfilter/ipset/ip_set.h
@@ -196,6 +196,9 @@ struct ip_set_region {
 	u32 elements;		/* Number of elements vs timeout */
 };
 
+/* Max range where every element is added/deleted in one step */
+#define IPSET_MAX_RANGE		(1<<20)
+
 /* The max revision number supported by any set type + 1 */
 #define IPSET_REVISION_MAX	9
 

diff --git a/include/net/netns/conntrack.h b/include/net/netns/conntrack.h
@@ -30,7 +30,6 @@ struct nf_tcp_net {
 	u8 tcp_ignore_invalid_rst;
 #if IS_ENABLED(CONFIG_NF_FLOW_TABLE)
 	unsigned int offload_timeout;
-	unsigned int offload_pickup;
 #endif
 };
 
@@ -44,7 +43,6 @@ struct nf_udp_net {
 	unsigned int timeouts[UDP_CT_MAX];
 #if IS_ENABLED(CONFIG_NF_FLOW_TABLE)
 	unsigned int offload_timeout;
-	unsigned int offload_pickup;
 #endif
 };
 

diff --git a/include/uapi/linux/netfilter/nfnetlink_hook.h b/include/uapi/linux/netfilter/nfnetlink_hook.h
@@ -43,6 +43,15 @@ enum nfnl_hook_chain_info_attributes {
 };
 #define NFNLA_HOOK_INFO_MAX (__NFNLA_HOOK_INFO_MAX - 1)
 
+enum nfnl_hook_chain_desc_attributes {
+	NFNLA_CHAIN_UNSPEC,
+	NFNLA_CHAIN_TABLE,
+	NFNLA_CHAIN_FAMILY,
+	NFNLA_CHAIN_NAME,
+	__NFNLA_CHAIN_MAX,
+};
+#define NFNLA_CHAIN_MAX (__NFNLA_CHAIN_MAX - 1)
+
 /**
  * enum nfnl_hook_chaintype - chain type
  *

diff --git a/net/bridge/netfilter/nf_conntrack_bridge.c b/net/bridge/netfilter/nf_conntrack_bridge.c
@@ -88,6 +88,12 @@ static int nf_br_ip_fragment(struct net *net, struct sock *sk,
 
 			skb = ip_fraglist_next(&iter);
 		}
+
+		if (!err)
+			return 0;
+
+		kfree_skb_list(iter.frag);
+
 		return err;
 	}
 slow_path:

diff --git a/net/netfilter/ipset/ip_set_hash_ip.c b/net/netfilter/ipset/ip_set_hash_ip.c
@@ -132,8 +132,11 @@ hash_ip4_uadt(struct ip_set *set, struct nlattr *tb[],
 		ret = ip_set_get_hostipaddr4(tb[IPSET_ATTR_IP_TO], &ip_to);
 		if (ret)
 			return ret;
-		if (ip > ip_to)
+		if (ip > ip_to) {
+			if (ip_to == 0)
+				return -IPSET_ERR_HASH_ELEM;
 			swap(ip, ip_to);
+		}
 	} else if (tb[IPSET_ATTR_CIDR]) {
 		u8 cidr = nla_get_u8(tb[IPSET_ATTR_CIDR]);
 
@@ -144,6 +147,10 @@ hash_ip4_uadt(struct ip_set *set, struct nlattr *tb[],
 
 	hosts = h->netmask == 32 ? 1 : 2 << (32 - h->netmask - 1);
 
+	/* 64bit division is not allowed on 32bit */
+	if (((u64)ip_to - ip + 1) >> (32 - h->netmask) > IPSET_MAX_RANGE)
+		return -ERANGE;
+
 	if (retried) {
 		ip = ntohl(h->next.ip);
 		e.ip = htonl(ip);

diff --git a/net/netfilter/ipset/ip_set_hash_ipmark.c b/net/netfilter/ipset/ip_set_hash_ipmark.c
@@ -121,6 +121,8 @@ hash_ipmark4_uadt(struct ip_set *set, struct nlattr *tb[],
 
 	e.mark = ntohl(nla_get_be32(tb[IPSET_ATTR_MARK]));
 	e.mark &= h->markmask;
+	if (e.mark == 0 && e.ip == 0)
+		return -IPSET_ERR_HASH_ELEM;
 
 	if (adt == IPSET_TEST ||
 	    !(tb[IPSET_ATTR_IP_TO] || tb[IPSET_ATTR_CIDR])) {
@@ -133,8 +135,11 @@ hash_ipmark4_uadt(struct ip_set *set, struct nlattr *tb[],
 		ret = ip_set_get_hostipaddr4(tb[IPSET_ATTR_IP_TO], &ip_to);
 		if (ret)
 			return ret;
-		if (ip > ip_to)
+		if (ip > ip_to) {
+			if (e.mark == 0 && ip_to == 0)
+				return -IPSET_ERR_HASH_ELEM;
 			swap(ip, ip_to);
+		}
 	} else if (tb[IPSET_ATTR_CIDR]) {
 		u8 cidr = nla_get_u8(tb[IPSET_ATTR_CIDR]);
 
@@ -143,6 +148,9 @@ hash_ipmark4_uadt(struct ip_set *set, struct nlattr *tb[],
 		ip_set_mask_from_to(ip, ip_to, cidr);
 	}
 
+	if (((u64)ip_to - ip + 1) > IPSET_MAX_RANGE)
+		return -ERANGE;
+
 	if (retried)
 		ip = ntohl(h->next.ip);
 	for (; ip <= ip_to; ip++) {

diff --git a/net/netfilter/ipset/ip_set_hash_ipport.c b/net/netfilter/ipset/ip_set_hash_ipport.c
@@ -173,6 +173,9 @@ hash_ipport4_uadt(struct ip_set *set, struct nlattr *tb[],
 			swap(port, port_to);
 	}
 
+	if (((u64)ip_to - ip + 1)*(port_to - port + 1) > IPSET_MAX_RANGE)
+		return -ERANGE;
+
 	if (retried)
 		ip = ntohl(h->next.ip);
 	for (; ip <= ip_to; ip++) {

diff --git a/net/netfilter/ipset/ip_set_hash_ipportip.c b/net/netfilter/ipset/ip_set_hash_ipportip.c
@@ -180,6 +180,9 @@ hash_ipportip4_uadt(struct ip_set *set, struct nlattr *tb[],
 			swap(port, port_to);
 	}
 
+	if (((u64)ip_to - ip + 1)*(port_to - port + 1) > IPSET_MAX_RANGE)
+		return -ERANGE;
+
 	if (retried)
 		ip = ntohl(h->next.ip);
 	for (; ip <= ip_to; ip++) {

diff --git a/net/netfilter/ipset/ip_set_hash_ipportnet.c b/net/netfilter/ipset/ip_set_hash_ipportnet.c
@@ -253,6 +253,9 @@ hash_ipportnet4_uadt(struct ip_set *set, struct nlattr *tb[],
 			swap(port, port_to);
 	}
 
+	if (((u64)ip_to - ip + 1)*(port_to - port + 1) > IPSET_MAX_RANGE)
+		return -ERANGE;
+
 	ip2_to = ip2_from;
 	if (tb[IPSET_ATTR_IP2_TO]) {
 		ret = ip_set_get_hostipaddr4(tb[IPSET_ATTR_IP2_TO], &ip2_to);

diff --git a/net/netfilter/ipset/ip_set_hash_net.c b/net/netfilter/ipset/ip_set_hash_net.c
@@ -140,7 +140,7 @@ hash_net4_uadt(struct ip_set *set, struct nlattr *tb[],
 	ipset_adtfn adtfn = set->variant->adt[adt];
 	struct hash_net4_elem e = { .cidr = HOST_MASK };
 	struct ip_set_ext ext = IP_SET_INIT_UEXT(set);
-	u32 ip = 0, ip_to = 0;
+	u32 ip = 0, ip_to = 0, ipn, n = 0;
 	int ret;
 
 	if (tb[IPSET_ATTR_LINENO])
@@ -188,6 +188,15 @@ hash_net4_uadt(struct ip_set *set, struct nlattr *tb[],
 		if (ip + UINT_MAX == ip_to)
 			return -IPSET_ERR_HASH_RANGE;
 	}
+	ipn = ip;
+	do {
+		ipn = ip_set_range_to_cidr(ipn, ip_to, &e.cidr);
+		n++;
+	} while (ipn++ < ip_to);
+
+	if (n > IPSET_MAX_RANGE)
+		return -ERANGE;
+
 	if (retried)
 		ip = ntohl(h->next.ip);
 	do {

diff --git a/net/netfilter/ipset/ip_set_hash_netiface.c b/net/netfilter/ipset/ip_set_hash_netiface.c
@@ -202,7 +202,7 @@ hash_netiface4_uadt(struct ip_set *set, struct nlattr *tb[],
 	ipset_adtfn adtfn = set->variant->adt[adt];
 	struct hash_netiface4_elem e = { .cidr = HOST_MASK, .elem = 1 };
 	struct ip_set_ext ext = IP_SET_INIT_UEXT(set);
-	u32 ip = 0, ip_to = 0;
+	u32 ip = 0, ip_to = 0, ipn, n = 0;
 	int ret;
 
 	if (tb[IPSET_ATTR_LINENO])
@@ -256,6 +256,14 @@ hash_netiface4_uadt(struct ip_set *set, struct nlattr *tb[],
 	} else {
 		ip_set_mask_from_to(ip, ip_to, e.cidr);
 	}
+	ipn = ip;
+	do {
+		ipn = ip_set_range_to_cidr(ipn, ip_to, &e.cidr);
+		n++;
+	} while (ipn++ < ip_to);
+
+	if (n > IPSET_MAX_RANGE)
+		return -ERANGE;
 
 	if (retried)
 		ip = ntohl(h->next.ip);

diff --git a/net/netfilter/ipset/ip_set_hash_netnet.c b/net/netfilter/ipset/ip_set_hash_netnet.c
@@ -168,7 +168,8 @@ hash_netnet4_uadt(struct ip_set *set, struct nlattr *tb[],
 	struct hash_netnet4_elem e = { };
 	struct ip_set_ext ext = IP_SET_INIT_UEXT(set);
 	u32 ip = 0, ip_to = 0;
-	u32 ip2 = 0, ip2_from = 0, ip2_to = 0;
+	u32 ip2 = 0, ip2_from = 0, ip2_to = 0, ipn;
+	u64 n = 0, m = 0;
 	int ret;
 
 	if (tb[IPSET_ATTR_LINENO])
@@ -244,6 +245,19 @@ hash_netnet4_uadt(struct ip_set *set, struct nlattr *tb[],
 	} else {
 		ip_set_mask_from_to(ip2_from, ip2_to, e.cidr[1]);
 	}
+	ipn = ip;
+	do {
+		ipn = ip_set_range_to_cidr(ipn, ip_to, &e.cidr[0]);
+		n++;
+	} while (ipn++ < ip_to);
+	ipn = ip2_from;
+	do {
+		ipn = ip_set_range_to_cidr(ipn, ip2_to, &e.cidr[1]);
+		m++;
+	} while (ipn++ < ip2_to);
+
+	if (n*m > IPSET_MAX_RANGE)
+		return -ERANGE;
 
 	if (retried) {
 		ip = ntohl(h->next.ip[0]);

diff --git a/net/netfilter/ipset/ip_set_hash_netport.c b/net/netfilter/ipset/ip_set_hash_netport.c
@@ -158,7 +158,8 @@ hash_netport4_uadt(struct ip_set *set, struct nlattr *tb[],
 	ipset_adtfn adtfn = set->variant->adt[adt];
 	struct hash_netport4_elem e = { .cidr = HOST_MASK - 1 };
 	struct ip_set_ext ext = IP_SET_INIT_UEXT(set);
-	u32 port, port_to, p = 0, ip = 0, ip_to = 0;
+	u32 port, port_to, p = 0, ip = 0, ip_to = 0, ipn;
+	u64 n = 0;
 	bool with_ports = false;
 	u8 cidr;
 	int ret;
@@ -235,6 +236,14 @@ hash_netport4_uadt(struct ip_set *set, struct nlattr *tb[],
 	} else {
 		ip_set_mask_from_to(ip, ip_to, e.cidr + 1);
 	}
+	ipn = ip;
+	do {
+		ipn = ip_set_range_to_cidr(ipn, ip_to, &cidr);
+		n++;
+	} while (ipn++ < ip_to);
+
+	if (n*(port_to - port + 1) > IPSET_MAX_RANGE)
+		return -ERANGE;
 
 	if (retried) {
 		ip = ntohl(h->next.ip);

diff --git a/net/netfilter/ipset/ip_set_hash_netportnet.c b/net/netfilter/ipset/ip_set_hash_netportnet.c
@@ -182,7 +182,8 @@ hash_netportnet4_uadt(struct ip_set *set, struct nlattr *tb[],
 	struct hash_netportnet4_elem e = { };
 	struct ip_set_ext ext = IP_SET_INIT_UEXT(set);
 	u32 ip = 0, ip_to = 0, p = 0, port, port_to;
-	u32 ip2_from = 0, ip2_to = 0, ip2;
+	u32 ip2_from = 0, ip2_to = 0, ip2, ipn;
+	u64 n = 0, m = 0;
 	bool with_ports = false;
 	int ret;
 
@@ -284,6 +285,19 @@ hash_netportnet4_uadt(struct ip_set *set, struct nlattr *tb[],
 	} else {
 		ip_set_mask_from_to(ip2_from, ip2_to, e.cidr[1]);
 	}
+	ipn = ip;
+	do {
+		ipn = ip_set_range_to_cidr(ipn, ip_to, &e.cidr[0]);
+		n++;
+	} while (ipn++ < ip_to);
+	ipn = ip2_from;
+	do {
+		ipn = ip_set_range_to_cidr(ipn, ip2_to, &e.cidr[1]);
+		m++;
+	} while (ipn++ < ip2_to);
+
+	if (n*m*(port_to - port + 1) > IPSET_MAX_RANGE)
+		return -ERANGE;
 
 	if (retried) {
 		ip = ntohl(h->next.ip[0]);