[PATCH 0/7] Netfilter fixes for net

Hi David,

The following patchset contains Netfilter fixes for your net tree,
they are:

1) Allow nf_tables reject expression from input, forward and output hooks,
   since only there the routing information is available, otherwise we crash.

2) Fix unsafe list iteration when flushing timeout and accouting objects.

3) Fix refcount leak on timeout policy parsing failure.

4) Unlink timeout object for unconfirmed conntracks too

5) Missing validation of pkttype mangling from bridge family.

6) Fix refcount leak on ebtables on second lookup for the specific
   bridge match extension, this patch from Sabrina Dubroca.

7) Remove unnecessary ip_hdr() in nf_tables_netdev family.

Patches from 1-5 and 7 from Liping Zhang.

You can pull these changes from:

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

Thanks!

----------------------------------------------------------------

The following changes since commit 51af96b53469f3b8cfcfe0504d0ff87239175b78:

  mlxsw: router: Enable neighbors to be created on stacked devices (2016-08-24 09:39:04 -0700)

are available in the git repository at:

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

for you to fetch changes up to c73c2484901139c28383b58eabcbf4d613e91518:

  netfilter: nf_tables_netdev: remove redundant ip_hdr assignment (2016-08-30 11:41:04 +0200)

----------------------------------------------------------------
Liping Zhang (6):
      netfilter: nft_reject: restrict to INPUT/FORWARD/OUTPUT
      netfilter: nfnetlink: use list_for_each_entry_safe to delete all objects
      netfilter: cttimeout: put back l4proto when replacing timeout policy
      netfilter: cttimeout: unlink timeout objs in the unconfirmed ct lists
      netfilter: nft_meta: improve the validity check of pkttype set expr
      netfilter: nf_tables_netdev: remove redundant ip_hdr assignment

Sabrina Dubroca (1):
      netfilter: ebtables: put module reference when an incorrect extension is found

 include/net/netfilter/nft_meta.h       |  4 +++
 include/net/netfilter/nft_reject.h     |  4 +++
 net/bridge/netfilter/ebtables.c        |  2 ++
 net/bridge/netfilter/nft_meta_bridge.c |  1 +
 net/ipv4/netfilter/nft_reject_ipv4.c   |  1 +
 net/ipv6/netfilter/nft_reject_ipv6.c   |  1 +
 net/netfilter/nf_tables_netdev.c       |  1 -
 net/netfilter/nfnetlink_acct.c         |  6 ++---
 net/netfilter/nfnetlink_cttimeout.c    | 49 +++++++++++++++++++---------------
 net/netfilter/nft_meta.c               | 17 +++++++++---
 net/netfilter/nft_reject.c             | 16 +++++++++++
 net/netfilter/nft_reject_inet.c        |  7 ++++-
 12 files changed, 79 insertions(+), 30 deletions(-)

From: Liping Zhang <redacted>

After I add the nft rule "nft add rule filter prerouting reject
with tcp reset", kernel panic happened on my system:
  NULL pointer dereference at ...
  IP: [<ffffffff81b9db2f>] nf_send_reset+0xaf/0x400
  Call Trace:
  [<ffffffff81b9da80>] ? nf_reject_ip_tcphdr_get+0x160/0x160
  [<ffffffffa0928061>] nft_reject_ipv4_eval+0x61/0xb0 [nft_reject_ipv4]
  [<ffffffffa08e836a>] nft_do_chain+0x1fa/0x890 [nf_tables]
  [<ffffffffa08e8170>] ? __nft_trace_packet+0x170/0x170 [nf_tables]
  [<ffffffffa06e0900>] ? nf_ct_invert_tuple+0xb0/0xc0 [nf_conntrack]
  [<ffffffffa07224d4>] ? nf_nat_setup_info+0x5d4/0x650 [nf_nat]
  [...]

Because in the PREROUTING chain, routing information is not exist,
then we will dereference the NULL pointer and oops happen.

So we restrict reject expression to INPUT, FORWARD and OUTPUT chain.
This is consistent with iptables REJECT target.

Signed-off-by: Liping Zhang <redacted>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nft_reject.h   |  4 ++++
 net/ipv4/netfilter/nft_reject_ipv4.c |  1 +
 net/ipv6/netfilter/nft_reject_ipv6.c |  1 +
 net/netfilter/nft_reject.c           | 16 ++++++++++++++++
 net/netfilter/nft_reject_inet.c      |  7 ++++++-
 5 files changed, 28 insertions(+), 1 deletion(-)

diff --git a/include/net/netfilter/nft_reject.h b/include/net/netfilter/nft_reject.h
index 60fa153..02e28c5 100644
--- a/include/net/netfilter/nft_reject.h
+++ b/include/net/netfilter/nft_reject.h

@@ -8,6 +8,10 @@ struct nft_reject {
 
 extern const struct nla_policy nft_reject_policy[];
 
+int nft_reject_validate(const struct nft_ctx *ctx,
+			const struct nft_expr *expr,
+			const struct nft_data **data);
+
 int nft_reject_init(const struct nft_ctx *ctx,
 		    const struct nft_expr *expr,
 		    const struct nlattr * const tb[]);

diff --git a/net/ipv4/netfilter/nft_reject_ipv4.c b/net/ipv4/netfilter/nft_reject_ipv4.c
index c24f41c..2c2553b 100644
--- a/net/ipv4/netfilter/nft_reject_ipv4.c
+++ b/net/ipv4/netfilter/nft_reject_ipv4.c

@@ -46,6 +46,7 @@ static const struct nft_expr_ops nft_reject_ipv4_ops = {
 	.eval		= nft_reject_ipv4_eval,
 	.init		= nft_reject_init,
 	.dump		= nft_reject_dump,
+	.validate	= nft_reject_validate,
 };
 
 static struct nft_expr_type nft_reject_ipv4_type __read_mostly = {

diff --git a/net/ipv6/netfilter/nft_reject_ipv6.c b/net/ipv6/netfilter/nft_reject_ipv6.c
index 533cd57..92bda99 100644
--- a/net/ipv6/netfilter/nft_reject_ipv6.c
+++ b/net/ipv6/netfilter/nft_reject_ipv6.c

@@ -47,6 +47,7 @@ static const struct nft_expr_ops nft_reject_ipv6_ops = {
 	.eval		= nft_reject_ipv6_eval,
 	.init		= nft_reject_init,
 	.dump		= nft_reject_dump,
+	.validate	= nft_reject_validate,
 };
 
 static struct nft_expr_type nft_reject_ipv6_type __read_mostly = {

diff --git a/net/netfilter/nft_reject.c b/net/netfilter/nft_reject.c
index 0522fc9..c64de3f7 100644
--- a/net/netfilter/nft_reject.c
+++ b/net/netfilter/nft_reject.c

@@ -26,11 +26,27 @@ const struct nla_policy nft_reject_policy[NFTA_REJECT_MAX + 1] = {
 };
 EXPORT_SYMBOL_GPL(nft_reject_policy);
 
+int nft_reject_validate(const struct nft_ctx *ctx,
+			const struct nft_expr *expr,
+			const struct nft_data **data)
+{
+	return nft_chain_validate_hooks(ctx->chain,
+					(1 << NF_INET_LOCAL_IN) |
+					(1 << NF_INET_FORWARD) |
+					(1 << NF_INET_LOCAL_OUT));
+}
+EXPORT_SYMBOL_GPL(nft_reject_validate);
+
 int nft_reject_init(const struct nft_ctx *ctx,
 		    const struct nft_expr *expr,
 		    const struct nlattr * const tb[])
 {
 	struct nft_reject *priv = nft_expr_priv(expr);
+	int err;
+
+	err = nft_reject_validate(ctx, expr, NULL);
+	if (err < 0)
+		return err;
 
 	if (tb[NFTA_REJECT_TYPE] == NULL)
 		return -EINVAL;

diff --git a/net/netfilter/nft_reject_inet.c b/net/netfilter/nft_reject_inet.c
index 759ca52..e79d9ca 100644
--- a/net/netfilter/nft_reject_inet.c
+++ b/net/netfilter/nft_reject_inet.c

@@ -66,7 +66,11 @@ static int nft_reject_inet_init(const struct nft_ctx *ctx,
 				const struct nlattr * const tb[])
 {
 	struct nft_reject *priv = nft_expr_priv(expr);
-	int icmp_code;
+	int icmp_code, err;
+
+	err = nft_reject_validate(ctx, expr, NULL);
+	if (err < 0)
+		return err;
 
 	if (tb[NFTA_REJECT_TYPE] == NULL)
 		return -EINVAL;

@@ -124,6 +128,7 @@ static const struct nft_expr_ops nft_reject_inet_ops = {
 	.eval		= nft_reject_inet_eval,
 	.init		= nft_reject_inet_init,
 	.dump		= nft_reject_inet_dump,
+	.validate	= nft_reject_validate,
 };
 
 static struct nft_expr_type nft_reject_inet_type __read_mostly = {

-- 
2.1.4

From: Liping Zhang <redacted>

cttimeout and acct objects are deleted from the list while traversing
it, so use list_for_each_entry is unsafe here.

Signed-off-by: Liping Zhang <redacted>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nfnetlink_acct.c      | 6 +++---
 net/netfilter/nfnetlink_cttimeout.c | 5 +++--
 2 files changed, 6 insertions(+), 5 deletions(-)

diff --git a/net/netfilter/nfnetlink_acct.c b/net/netfilter/nfnetlink_acct.c
index 70eb2f6a..d44d89b 100644
--- a/net/netfilter/nfnetlink_acct.c
+++ b/net/netfilter/nfnetlink_acct.c

@@ -343,12 +343,12 @@ static int nfnl_acct_del(struct net *net, struct sock *nfnl,
 			 struct sk_buff *skb, const struct nlmsghdr *nlh,
 			 const struct nlattr * const tb[])
 {
-	char *acct_name;
-	struct nf_acct *cur;
+	struct nf_acct *cur, *tmp;
 	int ret = -ENOENT;
+	char *acct_name;
 
 	if (!tb[NFACCT_NAME]) {
-		list_for_each_entry(cur, &net->nfnl_acct_list, head)
+		list_for_each_entry_safe(cur, tmp, &net->nfnl_acct_list, head)
 			nfnl_acct_try_del(cur);
 
 		return 0;

diff --git a/net/netfilter/nfnetlink_cttimeout.c b/net/netfilter/nfnetlink_cttimeout.c
index 68216cd..f74fee1 100644
--- a/net/netfilter/nfnetlink_cttimeout.c
+++ b/net/netfilter/nfnetlink_cttimeout.c

@@ -350,12 +350,13 @@ static int cttimeout_del_timeout(struct net *net, struct sock *ctnl,
 				 const struct nlmsghdr *nlh,
 				 const struct nlattr * const cda[])
 {
-	struct ctnl_timeout *cur;
+	struct ctnl_timeout *cur, *tmp;
 	int ret = -ENOENT;
 	char *name;
 
 	if (!cda[CTA_TIMEOUT_NAME]) {
-		list_for_each_entry(cur, &net->nfct_timeout_list, head)
+		list_for_each_entry_safe(cur, tmp, &net->nfct_timeout_list,
+					 head)
 			ctnl_timeout_try_del(net, cur);
 
 		return 0;

-- 
2.1.4

From: Liping Zhang <redacted>

We forget to call nf_ct_l4proto_put when replacing the existing
timeout policy. Acctually, there's no need to get ct l4proto
before doing replace, so we can move it to a later position.

Signed-off-by: Liping Zhang <redacted>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nfnetlink_cttimeout.c | 33 +++++++++++++++------------------
 1 file changed, 15 insertions(+), 18 deletions(-)

diff --git a/net/netfilter/nfnetlink_cttimeout.c b/net/netfilter/nfnetlink_cttimeout.c
index f74fee1..6844c7a 100644
--- a/net/netfilter/nfnetlink_cttimeout.c
+++ b/net/netfilter/nfnetlink_cttimeout.c

@@ -98,31 +98,28 @@ static int cttimeout_new_timeout(struct net *net, struct sock *ctnl,
 		break;
 	}
 
-	l4proto = nf_ct_l4proto_find_get(l3num, l4num);
-
-	/* This protocol is not supportted, skip. */
-	if (l4proto->l4proto != l4num) {
-		ret = -EOPNOTSUPP;
-		goto err_proto_put;
-	}
-
 	if (matching) {
 		if (nlh->nlmsg_flags & NLM_F_REPLACE) {
 			/* You cannot replace one timeout policy by another of
 			 * different kind, sorry.
 			 */
 			if (matching->l3num != l3num ||
-			    matching->l4proto->l4proto != l4num) {
-				ret = -EINVAL;
-				goto err_proto_put;
-			}
-
-			ret = ctnl_timeout_parse_policy(&matching->data,
-							l4proto, net,
-							cda[CTA_TIMEOUT_DATA]);
-			return ret;
+			    matching->l4proto->l4proto != l4num)
+				return -EINVAL;
+
+			return ctnl_timeout_parse_policy(&matching->data,
+							 matching->l4proto, net,
+							 cda[CTA_TIMEOUT_DATA]);
 		}
-		ret = -EBUSY;
+
+		return -EBUSY;
+	}
+
+	l4proto = nf_ct_l4proto_find_get(l3num, l4num);
+
+	/* This protocol is not supportted, skip. */
+	if (l4proto->l4proto != l4num) {
+		ret = -EOPNOTSUPP;
 		goto err_proto_put;
 	}
 

-- 
2.1.4

From: Liping Zhang <redacted>

"meta pkttype set" is only supported on prerouting chain with bridge
family and ingress chain with netdev family.

But the validate check is incomplete, and the user can add the nft
rules on input chain with bridge family, for example:
  # nft add table bridge filter
  # nft add chain bridge filter input {type filter hook input \
    priority 0 \;}
  # nft add chain bridge filter test
  # nft add rule bridge filter test meta pkttype set unicast
  # nft add rule bridge filter input jump test

This patch fixes the problem.

Signed-off-by: Liping Zhang <redacted>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nft_meta.h       |  4 ++++
 net/bridge/netfilter/nft_meta_bridge.c |  1 +
 net/netfilter/nft_meta.c               | 17 +++++++++++++----
 3 files changed, 18 insertions(+), 4 deletions(-)

diff --git a/include/net/netfilter/nft_meta.h b/include/net/netfilter/nft_meta.h
index d27588c..1139cde 100644
--- a/include/net/netfilter/nft_meta.h
+++ b/include/net/netfilter/nft_meta.h

@@ -36,4 +36,8 @@ void nft_meta_set_eval(const struct nft_expr *expr,
 void nft_meta_set_destroy(const struct nft_ctx *ctx,
 			  const struct nft_expr *expr);
 
+int nft_meta_set_validate(const struct nft_ctx *ctx,
+			  const struct nft_expr *expr,
+			  const struct nft_data **data);
+
 #endif

diff --git a/net/bridge/netfilter/nft_meta_bridge.c b/net/bridge/netfilter/nft_meta_bridge.c
index 4b901d9..ad47a92 100644
--- a/net/bridge/netfilter/nft_meta_bridge.c
+++ b/net/bridge/netfilter/nft_meta_bridge.c

@@ -86,6 +86,7 @@ static const struct nft_expr_ops nft_meta_bridge_set_ops = {
 	.init		= nft_meta_set_init,
 	.destroy	= nft_meta_set_destroy,
 	.dump		= nft_meta_set_dump,
+	.validate	= nft_meta_set_validate,
 };
 
 static const struct nft_expr_ops *

diff --git a/net/netfilter/nft_meta.c b/net/netfilter/nft_meta.c
index 2863f34..8a6bc76 100644
--- a/net/netfilter/nft_meta.c
+++ b/net/netfilter/nft_meta.c

@@ -291,10 +291,16 @@ int nft_meta_get_init(const struct nft_ctx *ctx,
 }
 EXPORT_SYMBOL_GPL(nft_meta_get_init);
 
-static int nft_meta_set_init_pkttype(const struct nft_ctx *ctx)
+int nft_meta_set_validate(const struct nft_ctx *ctx,
+			  const struct nft_expr *expr,
+			  const struct nft_data **data)
 {
+	struct nft_meta *priv = nft_expr_priv(expr);
 	unsigned int hooks;
 
+	if (priv->key != NFT_META_PKTTYPE)
+		return 0;
+
 	switch (ctx->afi->family) {
 	case NFPROTO_BRIDGE:
 		hooks = 1 << NF_BR_PRE_ROUTING;

@@ -308,6 +314,7 @@ static int nft_meta_set_init_pkttype(const struct nft_ctx *ctx)
 
 	return nft_chain_validate_hooks(ctx->chain, hooks);
 }
+EXPORT_SYMBOL_GPL(nft_meta_set_validate);
 
 int nft_meta_set_init(const struct nft_ctx *ctx,
 		      const struct nft_expr *expr,

@@ -327,15 +334,16 @@ int nft_meta_set_init(const struct nft_ctx *ctx,
 		len = sizeof(u8);
 		break;
 	case NFT_META_PKTTYPE:
-		err = nft_meta_set_init_pkttype(ctx);
-		if (err)
-			return err;
 		len = sizeof(u8);
 		break;
 	default:
 		return -EOPNOTSUPP;
 	}
 
+	err = nft_meta_set_validate(ctx, expr, NULL);
+	if (err < 0)
+		return err;
+
 	priv->sreg = nft_parse_register(tb[NFTA_META_SREG]);
 	err = nft_validate_register_load(priv->sreg, len);
 	if (err < 0)

@@ -407,6 +415,7 @@ static const struct nft_expr_ops nft_meta_set_ops = {
 	.init		= nft_meta_set_init,
 	.destroy	= nft_meta_set_destroy,
 	.dump		= nft_meta_set_dump,
+	.validate	= nft_meta_set_validate,
 };
 
 static const struct nft_expr_ops *

-- 
2.1.4

From: Sabrina Dubroca <sd@queasysnail.net>

commit bcf493428840 ("netfilter: ebtables: Fix extension lookup with
identical name") added a second lookup in case the extension that was
found during the first lookup matched another extension with the same
name, but didn't release the reference on the incorrect module.

Fixes: bcf493428840 ("netfilter: ebtables: Fix extension lookup with identical name")
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Acked-by: Phil Sutter <phil@nwl.cc>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/bridge/netfilter/ebtables.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/net/bridge/netfilter/ebtables.c b/net/bridge/netfilter/ebtables.c
index cceac5b..0833c25 100644
--- a/net/bridge/netfilter/ebtables.c
+++ b/net/bridge/netfilter/ebtables.c

@@ -368,6 +368,8 @@ ebt_check_match(struct ebt_entry_match *m, struct xt_mtchk_param *par,
 
 	match = xt_find_match(NFPROTO_BRIDGE, m->u.name, 0);
 	if (IS_ERR(match) || match->family != NFPROTO_BRIDGE) {
+		if (!IS_ERR(match))
+			module_put(match->me);
 		request_module("ebt_%s", m->u.name);
 		match = xt_find_match(NFPROTO_BRIDGE, m->u.name, 0);
 	}

-- 
2.1.4

From: Liping Zhang <redacted>

KASAN reported this bug:
  BUG: KASAN: use-after-free in icmp_packet+0x25/0x50 [nf_conntrack_ipv4] at
  addr ffff880002db08c8
  Read of size 4 by task lt-nf-queue/19041
  Call Trace:
  <IRQ>  [<ffffffff815eeebb>] dump_stack+0x63/0x88
  [<ffffffff813386f8>] kasan_report_error+0x528/0x560
  [<ffffffff81338cc8>] kasan_report+0x58/0x60
  [<ffffffffa07393f5>] ? icmp_packet+0x25/0x50 [nf_conntrack_ipv4]
  [<ffffffff81337551>] __asan_load4+0x61/0x80
  [<ffffffffa07393f5>] icmp_packet+0x25/0x50 [nf_conntrack_ipv4]
  [<ffffffffa06ecaa0>] nf_conntrack_in+0x550/0x980 [nf_conntrack]
  [<ffffffffa06ec550>] ? __nf_conntrack_confirm+0xb10/0xb10 [nf_conntrack]
  [ ... ]

The main reason is that we missed to unlink the timeout objects in the
unconfirmed ct lists, so we will access the timeout objects that have
already been freed.

Signed-off-by: Liping Zhang <redacted>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nfnetlink_cttimeout.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/net/netfilter/nfnetlink_cttimeout.c b/net/netfilter/nfnetlink_cttimeout.c
index 6844c7a..139e086 100644
--- a/net/netfilter/nfnetlink_cttimeout.c
+++ b/net/netfilter/nfnetlink_cttimeout.c

@@ -302,7 +302,16 @@ static void ctnl_untimeout(struct net *net, struct ctnl_timeout *timeout)
 	const struct hlist_nulls_node *nn;
 	unsigned int last_hsize;
 	spinlock_t *lock;
-	int i;
+	int i, cpu;
+
+	for_each_possible_cpu(cpu) {
+		struct ct_pcpu *pcpu = per_cpu_ptr(net->ct.pcpu_lists, cpu);
+
+		spin_lock_bh(&pcpu->lock);
+		hlist_nulls_for_each_entry(h, nn, &pcpu->unconfirmed, hnnode)
+			untimeout(h, timeout);
+		spin_unlock_bh(&pcpu->lock);
+	}
 
 	local_bh_disable();
 restart:

-- 
2.1.4

From: Liping Zhang <redacted>

We have already use skb_header_pointer to get the ip header pointer,
so there's no need to use ip_hdr again. Moreover, in NETDEV INGRESS
hook, ip header maybe not linear, so use ip_hdr is not appropriate,
remove it.

Signed-off-by: Liping Zhang <redacted>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nf_tables_netdev.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/net/netfilter/nf_tables_netdev.c b/net/netfilter/nf_tables_netdev.c
index 5eefe4a..75d696f 100644
--- a/net/netfilter/nf_tables_netdev.c
+++ b/net/netfilter/nf_tables_netdev.c

@@ -30,7 +30,6 @@ nft_netdev_set_pktinfo_ipv4(struct nft_pktinfo *pkt,
 	if (!iph)
 		return;
 
-	iph = ip_hdr(skb);
 	if (iph->ihl < 5 || iph->version != 4)
 		return;
 

-- 
2.1.4

From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Tue, 30 Aug 2016 13:26:16 +0200

The following patchset contains Netfilter fixes for your net tree,
they are:

 ...

You can pull these changes from:

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

Pulled, thanks a lot Pablo.