Skip to contentSkip to conversation

Netfilter fixes for net

PATCH In review Pablo Neira Ayuso 2 participants 7 messages Aug 19, 2019

Subsystems (4)
lore
Share & tools
lore

Pablo Neira Ayuso

author maintainer
#0 lore raw

cover Netfilter fixes for net

Hi,

The following patchset contains Netfilter fixes for net:

1) Remove IP MASQUERADING record in MAINTAINERS file, from Denis Efremov.

2) Counter arguments are swapped in ebtables, from Todd Seidelmann.

3) Missing netlink attribute validation in flow_offload extension.

4) Incorrect alignment in xt_nfacct that breaks 32-bits userspace / 64-bits kernels, from Juliana Rodrigueiro.

5) Missing include guard in nf_conntrack_h323_types.h, from Masahiro Yamada.

You can pull these changes from:

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

Thanks.

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

The following changes since commit cfef46d692efd852a0da6803f920cc756eea2855:

ravb: Fix use-after-free ravb_tstamp_skb (2019-08-18 14:19:14 -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 38a429c898ddd210cc35463b096389f97c3c5a73:

netfilter: add include guard to nf_conntrack_h323_types.h (2019-08-19 13:59:57 +0200)

----------------------------------------------------------------
Denis Efremov (1):
MAINTAINERS: Remove IP MASQUERADING record

Juliana Rodrigueiro (1):
netfilter: xt_nfacct: Fix alignment mismatch in xt_nfacct_match_info

Masahiro Yamada (1):
netfilter: add include guard to nf_conntrack_h323_types.h

Pablo Neira Ayuso (1):
netfilter: nft_flow_offload: missing netlink attribute policy

Todd Seidelmann (1):
netfilter: ebtables: Fix argument order to ADD_COUNTER

 MAINTAINERS                                       |  5 ----
 include/linux/netfilter/nf_conntrack_h323_types.h |  5 ++++
 include/uapi/linux/netfilter/xt_nfacct.h          |  5 ++++
 net/bridge/netfilter/ebtables.c                   |  8 ++---
 net/netfilter/nft_flow_offload.c                  |  6 ++++
 net/netfilter/xt_nfacct.c                         | 36 ++++++++++++++++-------
 6 files changed, 45 insertions(+), 20 deletions(-)
author maintainer
#1 lore raw

PATCH 1/5 MAINTAINERS: Remove IP MASQUERADING record

From: Denis Efremov <[email protected]>

This entry is in MAINTAINERS for historical purpose. It doesn't match current sources since the commit adf82accc5f5 ("netfilter: x_tables: merge ip and ipv6 masquerade modules") moved the module.
The net/netfilter/xt_MASQUERADE.c module is already under the netfilter section. Thus, there is no purpose to keep this separate entry in MAINTAINERS.

---
 MAINTAINERS | 5 -----
 1 file changed, 5 deletions(-)
MAINTAINERS +0 -5
index a416574780d6..6839cfd91dde 100644
@@ -8439,11 +8439,6 @@ S:	Maintained
 F:	fs/io_uring.c
 F:	include/uapi/linux/io_uring.h
 
-IP MASQUERADING
-M:	Juanjo Ciarlante <[email protected]>
-S:	Maintained
-F:	net/ipv4/netfilter/ipt_MASQUERADE.c
-
 IPMI SUBSYSTEM
 M:	Corey Minyard <[email protected]>
 L:	[email protected] (moderated for non-subscribers)
Signature · 1 line
2.11.0
author maintainer
#2 lore raw

PATCH 2/5 netfilter: ebtables: Fix argument order to ADD_COUNTER

In reply to Pablo Neira Ayuso

From: Todd Seidelmann <[email protected]>

The ordering of arguments to the x_tables ADD_COUNTER macro appears to be wrong in ebtables (cf. ip_tables.c, ip6_tables.c, and arp_tables.c).

This causes data corruption in the ebtables userspace tools because they get incorrect packet & byte counts from the kernel.

Fixes d72133e628803 ("netfilter: ebtables: use ADD_COUNTER macro")
---
 net/bridge/netfilter/ebtables.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
net/bridge/netfilter/ebtables.c +4 -4
index c8177a89f52c..4096d8a74a2b 100644
@@ -221,7 +221,7 @@ unsigned int ebt_do_table(struct sk_buff *skb,
 			return NF_DROP;
 		}
 
-		ADD_COUNTER(*(counter_base + i), 1, skb->len);
+		ADD_COUNTER(*(counter_base + i), skb->len, 1);
 
 		/* these should only watch: not modify, nor tell us
 		 * what to do with the packet
@@ -959,8 +959,8 @@ static void get_counters(const struct ebt_counter *oldcounters,
 			continue;
 		counter_base = COUNTER_BASE(oldcounters, nentries, cpu);
 		for (i = 0; i < nentries; i++)
-			ADD_COUNTER(counters[i], counter_base[i].pcnt,
-				    counter_base[i].bcnt);
+			ADD_COUNTER(counters[i], counter_base[i].bcnt,
+				    counter_base[i].pcnt);
 	}
 }
 
@@ -1280,7 +1280,7 @@ static int do_update_counters(struct net *net, const char *name,
 
 	/* we add to the counters of the first cpu */
 	for (i = 0; i < num_counters; i++)
-		ADD_COUNTER(t->private->counters[i], tmp[i].pcnt, tmp[i].bcnt);
+		ADD_COUNTER(t->private->counters[i], tmp[i].bcnt, tmp[i].pcnt);
 
 	write_unlock_bh(&t->lock);
 	ret = 0;
Signature · 1 line
2.11.0
author maintainer
#3 lore raw

PATCH 3/5 netfilter: nft_flow_offload: missing netlink attribute policy

In reply to Pablo Neira Ayuso

The netlink attribute policy for NFTA_FLOW_TABLE_NAME is missing.

Fixes a3c90f7a2323 ("netfilter: nf_tables: flow offload expression")
---
 net/netfilter/nft_flow_offload.c | 6 ++++++
 1 file changed, 6 insertions(+)
net/netfilter/nft_flow_offload.c +6 -0
index 060a4ed46d5e..01705ad74a9a 100644
@@ -149,6 +149,11 @@ static int nft_flow_offload_validate(const struct nft_ctx *ctx,
 	return nft_chain_validate_hooks(ctx->chain, hook_mask);
 }
 
+static const struct nla_policy nft_flow_offload_policy[NFTA_FLOW_MAX + 1] = {
+	[NFTA_FLOW_TABLE_NAME]	= { .type = NLA_STRING,
+				    .len = NFT_NAME_MAXLEN - 1 },
+};
+
 static int nft_flow_offload_init(const struct nft_ctx *ctx,
 				 const struct nft_expr *expr,
 				 const struct nlattr * const tb[])
@@ -207,6 +212,7 @@ static const struct nft_expr_ops nft_flow_offload_ops = {
 static struct nft_expr_type nft_flow_offload_type __read_mostly = {
 	.name		= "flow_offload",
 	.ops		= &nft_flow_offload_ops,
+	.policy		= nft_flow_offload_policy,
 	.maxattr	= NFTA_FLOW_MAX,
 	.owner		= THIS_MODULE,
 };
Signature · 1 line
2.11.0
author maintainer
#4 lore raw

PATCH 4/5 netfilter: xt_nfacct: Fix alignment mismatch in xt_nfacct_match_info

In reply to Pablo Neira Ayuso

From: Juliana Rodrigueiro <[email protected]>

When running a 64-bit kernel with a 32-bit iptables binary, the size of the xt_nfacct_match_info struct diverges.

kernel: sizeof(struct xt_nfacct_match_info) : 40
iptables: sizeof(struct xt_nfacct_match_info)) : 36

Trying to append nfacct related rules results in an unhelpful message. Although it is suggested to look for more information in dmesg, nothing can be found there.

    # iptables -A <chain> -m nfacct --nfacct-name <acct-object>
    iptables: Invalid argument. Run `dmesg' for more information.

This patch fixes the memory misalignment by enforcing 8-byte alignment within the struct's first revision. This solution is often used in many other uapi netfilter headers.

---
 include/uapi/linux/netfilter/xt_nfacct.h |  5 +++++
 net/netfilter/xt_nfacct.c                | 36 ++++++++++++++++++++++----------
 2 files changed, 30 insertions(+), 11 deletions(-)
include/uapi/linux/netfilter/xt_nfacct.h +5 -0
index 5c8a4d760ee3..b5123ab8d54a 100644
@@ -11,4 +11,9 @@ struct xt_nfacct_match_info {
 	struct nf_acct	*nfacct;
 };
 
+struct xt_nfacct_match_info_v1 {
+	char		name[NFACCT_NAME_MAX];
+	struct nf_acct	*nfacct __attribute__((aligned(8)));
+};
+
 #endif /* _XT_NFACCT_MATCH_H */
net/netfilter/xt_nfacct.c +25 -11
index d0ab1adf5bff..5aab6df74e0f 100644
@@ -54,25 +54,39 @@ nfacct_mt_destroy(const struct xt_mtdtor_param *par)
 	nfnl_acct_put(info->nfacct);
 }
 
-static struct xt_match nfacct_mt_reg __read_mostly = {
-	.name       = "nfacct",
-	.family     = NFPROTO_UNSPEC,
-	.checkentry = nfacct_mt_checkentry,
-	.match      = nfacct_mt,
-	.destroy    = nfacct_mt_destroy,
-	.matchsize  = sizeof(struct xt_nfacct_match_info),
-	.usersize   = offsetof(struct xt_nfacct_match_info, nfacct),
-	.me         = THIS_MODULE,
+static struct xt_match nfacct_mt_reg[] __read_mostly = {
+	{
+		.name       = "nfacct",
+		.revision   = 0,
+		.family     = NFPROTO_UNSPEC,
+		.checkentry = nfacct_mt_checkentry,
+		.match      = nfacct_mt,
+		.destroy    = nfacct_mt_destroy,
+		.matchsize  = sizeof(struct xt_nfacct_match_info),
+		.usersize   = offsetof(struct xt_nfacct_match_info, nfacct),
+		.me         = THIS_MODULE,
+	},
+	{
+		.name       = "nfacct",
+		.revision   = 1,
+		.family     = NFPROTO_UNSPEC,
+		.checkentry = nfacct_mt_checkentry,
+		.match      = nfacct_mt,
+		.destroy    = nfacct_mt_destroy,
+		.matchsize  = sizeof(struct xt_nfacct_match_info_v1),
+		.usersize   = offsetof(struct xt_nfacct_match_info_v1, nfacct),
+		.me         = THIS_MODULE,
+	},
 };
 
 static int __init nfacct_mt_init(void)
 {
-	return xt_register_match(&nfacct_mt_reg);
+	return xt_register_matches(nfacct_mt_reg, ARRAY_SIZE(nfacct_mt_reg));
 }
 
 static void __exit nfacct_mt_exit(void)
 {
-	xt_unregister_match(&nfacct_mt_reg);
+	xt_unregister_matches(nfacct_mt_reg, ARRAY_SIZE(nfacct_mt_reg));
 }
 
 module_init(nfacct_mt_init);
Signature · 1 line
2.11.0
author maintainer
#5 lore raw

PATCH 5/5 netfilter: add include guard to nf_conntrack_h323_types.h

In reply to Pablo Neira Ayuso

From: Masahiro Yamada <[email protected]>

Add a header include guard just in case.

---
 include/linux/netfilter/nf_conntrack_h323_types.h | 5 +++++
 1 file changed, 5 insertions(+)
include/linux/netfilter/nf_conntrack_h323_types.h +5 -0
index 7a6871ac8784..74c6f9241944 100644
@@ -4,6 +4,9 @@
  * Copyright (c) 2006 Jing Min Zhao <[email protected]>
  */
 
+#ifndef _NF_CONNTRACK_H323_TYPES_H
+#define _NF_CONNTRACK_H323_TYPES_H
+
 typedef struct TransportAddress_ipAddress {	/* SEQUENCE */
 	int options;		/* No use */
 	unsigned int ip;
@@ -931,3 +934,5 @@ typedef struct RasMessage {	/* CHOICE */
 		InfoRequestResponse infoRequestResponse;
 	};
 } RasMessage;
+
+#endif /* _NF_CONNTRACK_H323_TYPES_H */
Signature · 1 line
2.11.0
maintainer
#6 lore raw
In reply to Pablo Neira Ayuso

From: Pablo Neira Ayuso <[email protected]>
Date: Mon, 19 Aug 2019 20:49:06 +0200

Quoted 14 lines

The following patchset contains Netfilter fixes for net:

1) Remove IP MASQUERADING record in MAINTAINERS file, from Denis Efremov.

2) Counter arguments are swapped in ebtables, from Todd Seidelmann.

3) Missing netlink attribute validation in flow_offload extension.

4) Incorrect alignment in xt_nfacct that breaks 32-bits userspace / 64-bits kernels, from Juliana Rodrigueiro.

Quoted 6 lines

5) Missing include guard in nf_conntrack_h323_types.h, from Masahiro Yamada.

You can pull these changes from:

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

Pulled, thanks.