[SRU][Vivid][PATCH] netfilter: bridge: forward IPv6 fragmented packets
Jay Vosburgh
jay.vosburgh at canonical.com
Wed Nov 18 00:48:48 UTC 2015
From: Bernhard Thaler <bernhard.thaler at wvnet.at>
BugLink: https://bugs.launchpad.net/nova/+bug/1463911
IPv6 fragmented packets are not forwarded on an ethernet bridge
with netfilter ip6_tables loaded. e.g. steps to reproduce
1) create a simple bridge like this
modprobe br_netfilter
brctl addbr br0
brctl addif br0 eth0
brctl addif br0 eth2
ifconfig eth0 up
ifconfig eth2 up
ifconfig br0 up
2) place a host with an IPv6 address on each side of the bridge
set IPv6 address on host A:
ip -6 addr add fd01:2345:6789:1::1/64 dev eth0
set IPv6 address on host B:
ip -6 addr add fd01:2345:6789:1::2/64 dev eth0
3) run a simple ping command on host A with packets > MTU
ping6 -s 4000 fd01:2345:6789:1::2
4) wait some time and run e.g. "ip6tables -t nat -nvL" on the bridge
IPv6 fragmented packets traverse the bridge cleanly until somebody runs.
"ip6tables -t nat -nvL". As soon as it is run (and netfilter modules are
loaded) IPv6 fragmented packets do not traverse the bridge any more (you
see no more responses in ping's output).
After applying this patch IPv6 fragmented packets traverse the bridge
cleanly in above scenario.
Signed-off-by: Bernhard Thaler <bernhard.thaler at wvnet.at>
[pablo at netfilter.org: small changes to br_nf_dev_queue_xmit]
Signed-off-by: Pablo Neira Ayuso <pablo at netfilter.org>
(backported from commit efb6de9b4ba0092b2c55f6a52d16294a8a698edd)
(backported from commit e70deecbf8e1562cac0b19f23848919e2f5d65aa)
(backported from commit d7b597421519d6f680eb8e152a0d8447466ee2d6)
Signed-off-by: Jay Vosburgh <jay.vosburgh at canonical.com>
---
include/linux/netfilter_ipv6.h | 1 +
include/linux/skbuff.h | 1 +
net/bridge/br_netfilter.c | 185 ++++++++++++++++++++++++++++++++---------
net/bridge/br_private.h | 6 +-
net/ipv6/netfilter.c | 1 +
5 files changed, 155 insertions(+), 39 deletions(-)
diff --git a/include/linux/netfilter_ipv6.h b/include/linux/netfilter_ipv6.h
index 64dad1cc..7c832ba 100644
--- a/include/linux/netfilter_ipv6.h
+++ b/include/linux/netfilter_ipv6.h
@@ -25,6 +25,7 @@ void ipv6_netfilter_fini(void);
struct nf_ipv6_ops {
int (*chk_addr)(struct net *net, const struct in6_addr *addr,
const struct net_device *dev, int strict);
+ int (*fragment)(struct sk_buff *skb, int (*output)(struct sk_buff *));
};
extern const struct nf_ipv6_ops __rcu *nf_ipv6_ops;
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 5fcc606..db71429 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -163,6 +163,7 @@ struct nf_conntrack {
struct nf_bridge_info {
atomic_t use;
unsigned int mask;
+ __u16 frag_max_size;
struct net_device *physindev;
struct net_device *physoutdev;
unsigned long data[32 / sizeof(unsigned long)];
diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c
index ba2bd0a..cb62adf 100644
--- a/net/bridge/br_netfilter.c
+++ b/net/bridge/br_netfilter.c
@@ -34,6 +34,7 @@
#include <net/ip.h>
#include <net/ipv6.h>
+#include <net/addrconf.h>
#include <net/route.h>
#include <net/netfilter/br_netfilter.h>
@@ -112,6 +113,71 @@ static inline __be16 pppoe_proto(const struct sk_buff *skb)
pppoe_proto(skb) == htons(PPP_IPV6) && \
brnf_filter_pppoe_tagged)
+/* largest possible L2 header, see br_nf_dev_queue_xmit() */
+#define NF_BRIDGE_MAX_MAC_HEADER_LENGTH (PPPOE_SES_HLEN + ETH_HLEN)
+
+#if IS_ENABLED(CONFIG_NF_DEFRAG_IPV4) || IS_ENABLED(CONFIG_NF_DEFRAG_IPV6)
+struct brnf_frag_data {
+ char mac[NF_BRIDGE_MAX_MAC_HEADER_LENGTH];
+ u8 encap_size;
+ u8 size;
+ u16 vlan_tci;
+ __be16 vlan_proto;
+};
+
+static DEFINE_PER_CPU(struct brnf_frag_data, brnf_frag_data_storage);
+#endif
+
+static int check_hbh_len(struct sk_buff *skb);
+
+static int br_validate_ipv6(struct sk_buff *skb)
+{
+ const struct ipv6hdr *hdr;
+ struct net_device *dev = skb->dev;
+ struct inet6_dev *idev = __in6_dev_get(skb->dev);
+ u32 pkt_len;
+ u8 ip6h_len = sizeof(struct ipv6hdr);
+
+ if (!pskb_may_pull(skb, ip6h_len))
+ goto inhdr_error;
+
+ if (skb->len < ip6h_len)
+ goto drop;
+
+ hdr = ipv6_hdr(skb);
+
+ if (hdr->version != 6)
+ goto inhdr_error;
+
+ pkt_len = ntohs(hdr->payload_len);
+
+ if (pkt_len || hdr->nexthdr != NEXTHDR_HOP) {
+ if (pkt_len + ip6h_len > skb->len) {
+ IP6_INC_STATS_BH(dev_net(dev), idev,
+ IPSTATS_MIB_INTRUNCATEDPKTS);
+ goto drop;
+ }
+ if (pskb_trim_rcsum(skb, pkt_len + ip6h_len)) {
+ IP6_INC_STATS_BH(dev_net(dev), idev,
+ IPSTATS_MIB_INDISCARDS);
+ goto drop;
+ }
+ }
+ if (hdr->nexthdr == NEXTHDR_HOP && check_hbh_len(skb))
+ goto drop;
+
+ memset(IP6CB(skb), 0, sizeof(struct inet6_skb_parm));
+ /* No IP options in IPv6 header; however it should be
+ * checked if some next headers need special treatment
+ */
+ return 0;
+
+inhdr_error:
+ IP6_INC_STATS_BH(dev_net(dev), idev, IPSTATS_MIB_INHDRERRORS);
+drop:
+ return -1;
+}
+
static inline struct rtable *bridge_parent_rtable(const struct net_device *dev)
{
struct net_bridge_port *port;
@@ -247,6 +313,8 @@ static int br_nf_pre_routing_finish_ipv6(struct sk_buff *skb)
struct nf_bridge_info *nf_bridge = skb->nf_bridge;
struct rtable *rt;
+ nf_bridge->frag_max_size = IP6CB(skb)->frag_max_size;
+
if (nf_bridge->mask & BRNF_PKT_TYPE) {
skb->pkt_type = PACKET_OTHERHOST;
nf_bridge->mask ^= BRNF_PKT_TYPE;
@@ -524,36 +592,15 @@ bad:
}
/* Replicate the checks that IPv6 does on packet reception and pass the packet
- * to ip6tables, which doesn't support NAT, so things are fairly simple. */
+ * to ip6tables.
+ */
static unsigned int br_nf_pre_routing_ipv6(const struct nf_hook_ops *ops,
struct sk_buff *skb,
const struct net_device *in,
const struct net_device *out,
int (*okfn)(struct sk_buff *))
{
- const struct ipv6hdr *hdr;
- u32 pkt_len;
-
- if (skb->len < sizeof(struct ipv6hdr))
- return NF_DROP;
-
- if (!pskb_may_pull(skb, sizeof(struct ipv6hdr)))
- return NF_DROP;
-
- hdr = ipv6_hdr(skb);
-
- if (hdr->version != 6)
- return NF_DROP;
-
- pkt_len = ntohs(hdr->payload_len);
-
- if (pkt_len || hdr->nexthdr != NEXTHDR_HOP) {
- if (pkt_len + sizeof(struct ipv6hdr) > skb->len)
- return NF_DROP;
- if (pskb_trim_rcsum(skb, pkt_len + sizeof(struct ipv6hdr)))
- return NF_DROP;
- }
- if (hdr->nexthdr == NEXTHDR_HOP && check_hbh_len(skb))
+ if (br_validate_ipv6(skb))
return NF_DROP;
nf_bridge_put(skb->nf_bridge);
@@ -658,6 +705,9 @@ static int br_nf_forward_finish(struct sk_buff *skb)
BR_INPUT_SKB_CB(skb)->frag_max_size = frag_max_size;
}
+ if (skb->protocol == htons(ETH_P_IPV6))
+ nf_bridge->frag_max_size = IP6CB(skb)->frag_max_size;
+
in = nf_bridge->physindev;
if (nf_bridge->mask & BRNF_PKT_TYPE) {
skb->pkt_type = PACKET_OTHERHOST;
@@ -728,6 +778,13 @@ static unsigned int br_nf_forward_ip(const struct nf_hook_ops *ops,
/* The physdev module checks on this */
nf_bridge->mask |= BRNF_BRIDGED;
+
+ if (pf == NFPROTO_IPV6) {
+ if (br_validate_ipv6(skb))
+ return NF_DROP;
+ IP6CB(skb)->frag_max_size = nf_bridge->frag_max_size;
+ }
+
nf_bridge->physoutdev = skb->dev;
if (pf == NFPROTO_IPV4)
skb->protocol = htons(ETH_P_IP);
@@ -776,35 +833,87 @@ static unsigned int br_nf_forward_arp(const struct nf_hook_ops *ops,
return NF_STOLEN;
}
-#if IS_ENABLED(CONFIG_NF_DEFRAG_IPV4)
+#if IS_ENABLED(CONFIG_NF_DEFRAG_IPV4) || IS_ENABLED(CONFIG_NF_DEFRAG_IPV6)
+static int br_nf_push_frag_xmit(struct sk_buff *skb)
+{
+ struct brnf_frag_data *data;
+ int err;
+
+ data = this_cpu_ptr(&brnf_frag_data_storage);
+ err = skb_cow_head(skb, data->size);
+
+ if (err) {
+ kfree_skb(skb);
+ return 0;
+ }
+
+ if (data->vlan_tci) {
+ skb->vlan_tci = data->vlan_tci;
+ skb->vlan_proto = data->vlan_proto;
+ }
+
+ skb_copy_to_linear_data_offset(skb, -data->size, data->mac, data->size);
+ __skb_push(skb, data->encap_size);
+
+ return br_dev_queue_push_xmit(skb);
+}
+#endif
+
static int br_nf_dev_queue_xmit(struct sk_buff *skb)
{
int ret;
int frag_max_size;
+ unsigned int mtu_reserved;
+
+ mtu_reserved = nf_bridge_mtu_reduction(skb);
+
+ if (skb_is_gso(skb) || skb->len + mtu_reserved <= skb->dev->mtu)
+ return br_dev_queue_push_xmit(skb);
/* This is wrong! We should preserve the original fragment
* boundaries by preserving frag_list rather than refragmenting.
*/
- if (skb->protocol == htons(ETH_P_IP) &&
- skb->len + nf_bridge_mtu_reduction(skb) > skb->dev->mtu &&
- !skb_is_gso(skb)) {
+#if IS_ENABLED(CONFIG_NF_CONNTRACK_IPV4)
+ if (skb->protocol == htons(ETH_P_IP)) {
frag_max_size = BR_INPUT_SKB_CB(skb)->frag_max_size;
if (br_parse_ip_options(skb))
/* Drop invalid packet */
return NF_DROP;
IPCB(skb)->frag_max_size = frag_max_size;
- ret = ip_fragment(skb, br_dev_queue_push_xmit);
- } else
- ret = br_dev_queue_push_xmit(skb);
+ return ip_fragment(skb, br_dev_queue_push_xmit);
+ }
+#endif
+#if IS_ENABLED(CONFIG_NF_DEFRAG_IPV6)
+ if (skb->protocol == htons(ETH_P_IPV6)) {
+ const struct nf_ipv6_ops *v6ops = nf_get_ipv6_ops();
+ struct brnf_frag_data *data;
- return ret;
-}
-#else
-static int br_nf_dev_queue_xmit(struct sk_buff *skb)
-{
- return br_dev_queue_push_xmit(skb);
-}
+ if (br_validate_ipv6(skb))
+ goto drop;
+
+ IP6CB(skb)->frag_max_size = skb->nf_bridge->frag_max_size;
+
+ nf_bridge_update_protocol(skb);
+
+ data = this_cpu_ptr(&brnf_frag_data_storage);
+ data->encap_size = nf_bridge_encap_header_len(skb);
+ data->size = ETH_HLEN + data->encap_size;
+
+ skb_copy_from_linear_data_offset(skb, -data->size, data->mac,
+ data->size);
+
+ if (v6ops)
+ return v6ops->fragment(skb, br_nf_push_frag_xmit);
+
+ kfree_skb(skb);
+ return -EMSGSIZE;
+ }
#endif
+ return br_dev_queue_push_xmit(skb);
+drop:
+ kfree_skb(skb);
+ return 0;
+}
/* PF_BRIDGE/POST_ROUTING ********************************************/
static unsigned int br_nf_post_routing(const struct nf_hook_ops *ops,
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index aea3d13..d9c880d 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -18,6 +18,7 @@
#include <linux/netpoll.h>
#include <linux/u64_stats_sync.h>
#include <net/route.h>
+#include <net/ip6_fib.h>
#include <linux/if_vlan.h>
#define BR_HASH_BITS 8
@@ -214,7 +215,10 @@ struct net_bridge
spinlock_t hash_lock;
struct hlist_head hash[BR_HASH_SIZE];
#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
- struct rtable fake_rtable;
+ union {
+ struct rtable fake_rtable;
+ struct rt6_info fake_rt6_info;
+ };
bool nf_call_iptables;
bool nf_call_ip6tables;
bool nf_call_arptables;
diff --git a/net/ipv6/netfilter.c b/net/ipv6/netfilter.c
index 398377a..3080815 100644
--- a/net/ipv6/netfilter.c
+++ b/net/ipv6/netfilter.c
@@ -191,6 +191,7 @@ static __sum16 nf_ip6_checksum_partial(struct sk_buff *skb, unsigned int hook,
static const struct nf_ipv6_ops ipv6ops = {
.chk_addr = ipv6_chk_addr,
+ .fragment = ip6_fragment,
};
static const struct nf_afinfo nf_ip6_afinfo = {
--
1.9.1
More information about the kernel-team
mailing list