* Re: Trying to implement secondary loopback
From: Martitz, Thomas @ 2013-03-13 20:38 UTC (permalink / raw)
To: richard -rw- weinberger
Cc: netdev@vger.kernel.org, davem@davemloft.net, edumazet@google.com,
herbert@gondor.apana.org.au, ebiederm@xmission.com
In-Reply-To: <CAFLxGvxcyDVwC-iys6UdKMBB6u3UQx6s7-mwm5caoNu=N7rmeA@mail.gmail.com>
>
> ________________________________________
> Von: richard -rw- weinberger [richard.weinberger@gmail.com]
> Gesendet: Mittwoch, 13. März 2013 12:13
> An: Martitz, Thomas
> Cc: netdev@vger.kernel.org; davem@davemloft.net; edumazet@google.com; herbert@gondor.apana.org.au; ebiederm@xmission.com
> Betreff: Re: Trying to implement secondary loopback
>
> >
> > So my questions are:
> > * Is this on purpose/expected?
> > * Is there anyway around it so that I can have a custom loopback interface
> > without touching lo's functionality.
> > * Generally, what's the proper way (if any) implement a secondary loopback
> > interface?
>
> The only really question that matters is, why do you need a second
> loopback interface?
> Why can't you use any other pseudo interface?
>
What pseudo interface do you mean? I'm not aware of alternative pseudo interfaces.
I'm developing a NIC and a driver for it. Thus I'm implementing the interface. I'm trying to implement an interface that does exactly do what loopback does except that I want to extend the loopback path to PCIe. This is for testing & validation purposes.
I tried to implement an interface that's doesn't advertise as such. But the data transfer didn't go through my code. My test app is a simple client/server setup in a single process that sends/receive data to/on 127.0.0.1.
Best regards.
-----
visit us at
OFC 2013 / March 19-21 / Anaheim Convention Center, CA, USA / booth 11807
www.hhi.fraunhofer.de/events
^ permalink raw reply
* Re: Trying to implement secondary loopback
From: Eric W. Biederman @ 2013-03-13 20:46 UTC (permalink / raw)
To: Martitz, Thomas
Cc: richard -rw- weinberger, netdev@vger.kernel.org,
davem@davemloft.net, edumazet@google.com,
herbert@gondor.apana.org.au
In-Reply-To: <3D7E565A361FB844A410A5EFCDD7BA4002ECF5@MXSRV3.fe.hhi.de>
"Martitz, Thomas" <thomas.martitz@hhi.fraunhofer.de> writes:
>>
>> ________________________________________
>> Von: richard -rw- weinberger [richard.weinberger@gmail.com]
>> Gesendet: Mittwoch, 13. März 2013 12:13
>> An: Martitz, Thomas
>> Cc: netdev@vger.kernel.org; davem@davemloft.net; edumazet@google.com; herbert@gondor.apana.org.au; ebiederm@xmission.com
>> Betreff: Re: Trying to implement secondary loopback
>>
>> >
>> > So my questions are:
>> > * Is this on purpose/expected?
>> > * Is there anyway around it so that I can have a custom loopback interface
>> > without touching lo's functionality.
>> > * Generally, what's the proper way (if any) implement a secondary loopback
>> > interface?
>>
>> The only really question that matters is, why do you need a second
>> loopback interface?
>> Why can't you use any other pseudo interface?
>>
>
> What pseudo interface do you mean? I'm not aware of alternative pseudo interfaces.
>
> I'm developing a NIC and a driver for it. Thus I'm implementing the interface. I'm trying to implement an interface that does exactly do what loopback does except that I want to extend the loopback path to PCIe. This is for testing & validation purposes.
>
> I tried to implement an interface that's doesn't advertise as
> such. But the data transfer didn't go through my code. My test app is
> a simple client/server setup in a single process that sends/receive
> data to/on 127.0.0.1.
The easy solution is to setup two network namespaces and two nics.
And transmit data from one network namespace to another, through your
nics.
Eric
^ permalink raw reply
* AW: Trying to implement secondary loopback
From: Martitz, Thomas @ 2013-03-13 21:07 UTC (permalink / raw)
To: Eric W. Biederman
Cc: richard -rw- weinberger, netdev@vger.kernel.org,
davem@davemloft.net, edumazet@google.com,
herbert@gondor.apana.org.au
In-Reply-To: <87r4jjav00.fsf@xmission.com>
>
> ________________________________________
> Von: Eric W. Biederman [ebiederm@xmission.com]
> Gesendet: Mittwoch, 13. März 2013 21:46
> An: Martitz, Thomas
> Cc: richard -rw- weinberger; netdev@vger.kernel.org; davem@davemloft.net; edumazet@google.com; herbert@gondor.apana.org.au
> Betreff: Re: Trying to implement secondary loopback
>
> > I tried to implement an interface that's doesn't advertise as
> > such. But the data transfer didn't go through my code. My test app is
> > a simple client/server setup in a single process that sends/receive
> > data to/on 127.0.0.1.
>
> The easy solution is to setup two network namespaces and two nics.
> And transmit data from one network namespace to another, through your
> nics.
>
> Eric
>
That sounds a lot more involed than using loopback to send data back to the same process. And I guess just calling netif_rx() in the xmit() function wouldn't be sufficient either? Also I wouldn't know how to do that as I'm not yet very familiar with network namespaces (but that can clearly get fixed).
Best regards.
-----
visit us at
OFC 2013 / March 19-21 / Anaheim Convention Center, CA, USA / booth 11807
www.hhi.fraunhofer.de/events
^ permalink raw reply
* Re: AW: Trying to implement secondary loopback
From: Eric W. Biederman @ 2013-03-13 21:21 UTC (permalink / raw)
To: Martitz, Thomas
Cc: richard -rw- weinberger, netdev@vger.kernel.org,
davem@davemloft.net, edumazet@google.com,
herbert@gondor.apana.org.au
In-Reply-To: <3D7E565A361FB844A410A5EFCDD7BA4002EF11@MXSRV3.fe.hhi.de>
"Martitz, Thomas" <thomas.martitz@hhi.fraunhofer.de> writes:
>> Von: Eric W. Biederman [ebiederm@xmission.com]
>>
>> The easy solution is to setup two network namespaces and two nics.
>> And transmit data from one network namespace to another, through your
>> nics.
>
> That sounds a lot more involed than using loopback to send data back
> to the same process. And I guess just calling netif_rx() in the xmit()
> function wouldn't be sufficient either? Also I wouldn't know how to do
> that as I'm not yet very familiar with network namespaces (but that
> can clearly get fixed).
It may sound involved but it is just a few lines of code to set up.
ip netns add ns1
ip netns add ns2
ip link set nic1 netns ns1
ip link set nic2 netns ns2
int main(int argc, char **argv)
{
int netfd1, netfd2;
int sk1, sk2;
netfd1 = open(/var/run/netns/nic1);
netfd2 = open(/var/run/netns/nic2);
setns(netfd1, 0);
sk1 = socket(...);
setns(netfd2, 0);
sk2 = socket(...);
/* test test test */
}
And what is partidcularly interesting is that all of this works with
your drivers normal code paths without any kernel hacks.
Eric
^ permalink raw reply
* AW: AW: Trying to implement secondary loopback
From: Martitz, Thomas @ 2013-03-13 22:01 UTC (permalink / raw)
To: Eric W. Biederman
Cc: richard -rw- weinberger, netdev@vger.kernel.org,
davem@davemloft.net, edumazet@google.com,
herbert@gondor.apana.org.au
In-Reply-To: <878v5r9es7.fsf@xmission.com>
>
> ________________________________________
> Von: Eric W. Biederman [ebiederm@xmission.com]
> Gesendet: Mittwoch, 13. März 2013 22:21
> An: Martitz, Thomas
> Cc: richard -rw- weinberger; netdev@vger.kernel.org; davem@davemloft.net; edumazet@google.com; herbert@gondor.apana.org.au
> Betreff: Re: AW: Trying to implement secondary loopback
>
> "Martitz, Thomas" <thomas.martitz@hhi.fraunhofer.de> writes:
>
> >> Von: Eric W. Biederman [ebiederm@xmission.com]
> >>
> >> The easy solution is to setup two network namespaces and two nics.
> >> And transmit data from one network namespace to another, through your
> >> nics.
>
> >
> > That sounds a lot more involed than using loopback to send data back
> > to the same process. And I guess just calling netif_rx() in the xmit()
> > function wouldn't be sufficient either? Also I wouldn't know how to do
> > that as I'm not yet very familiar with network namespaces (but that
> > can clearly get fixed).
>
> It may sound involved but it is just a few lines of code to set up.
>
> ip netns add ns1
> ip netns add ns2
> ip link set nic1 netns ns1
> ip link set nic2 netns ns2
>
> int main(int argc, char **argv)
> {
> int netfd1, netfd2;
> int sk1, sk2;
> netfd1 = open(/var/run/netns/nic1);
> netfd2 = open(/var/run/netns/nic2);
>
> setns(netfd1, 0);
> sk1 = socket(...);
>
> setns(netfd2, 0);
> sk2 = socket(...);
>
> /* test test test */
> }
>
> And what is partidcularly interesting is that all of this works with
> your drivers normal code paths without any kernel hacks.
>
Thanks for the info, sounds very interesting. Now, how do I provide multiple interfaces within a single kernel module. Just call register_netdev() multiple times?
Best regards.
-----
visit us at
OFC 2013 / March 19-21 / Anaheim Convention Center, CA, USA / booth 11807
www.hhi.fraunhofer.de/events
^ permalink raw reply
* [PATCH net-next 0/6] Tunneling: Tunnel restructuring.
From: Pravin B Shelar @ 2013-03-13 22:48 UTC (permalink / raw)
To: davem; +Cc: netdev, jesse, Pravin B Shelar
Following patch series restructure GRE and IPIP tunneling code
to make it modular. It adds ip_tunnel module which acts as
generic tunneling layer which has common code.
These patches do not change any functionality.
Pravin B Shelar (6):
GRE: Use RTNL for atomic protocol registration.
GRE: Refactor GRE tunneling code.
IPIP: Use ip-tunneling code.
VXLAN: Fix vxlan stats handling.
VXLAN: Use IP Tunnels tunnel ENC encap API
tunnels: Use IP Tunnel stats APIs.
drivers/net/Kconfig | 1 +
drivers/net/ppp/pptp.c | 4 +
drivers/net/vxlan.c | 100 +---
include/net/gre.h | 51 ++
include/net/ip_tunnels.h | 169 ++++++
include/net/ipip.h | 88 ---
net/ipv4/Kconfig | 7 +
net/ipv4/Makefile | 1 +
net/ipv4/af_inet.c | 1 -
net/ipv4/gre.c | 23 +-
net/ipv4/ip_gre.c | 1492 ++++++++++------------------------------------
net/ipv4/ip_tunnel.c | 1036 ++++++++++++++++++++++++++++++++
net/ipv4/ip_vti.c | 42 +--
net/ipv4/ipip.c | 745 ++---------------------
net/ipv4/ipmr.c | 2 +-
net/ipv6/Kconfig | 2 +
net/ipv6/af_inet6.c | 1 -
net/ipv6/ip6_gre.c | 43 +--
net/ipv6/ip6_tunnel.c | 1 +
net/ipv6/sit.c | 39 +--
20 files changed, 1654 insertions(+), 2194 deletions(-)
create mode 100644 include/net/ip_tunnels.h
delete mode 100644 include/net/ipip.h
create mode 100644 net/ipv4/ip_tunnel.c
^ permalink raw reply
* [PATCH net-next 1/6] GRE: Use RTNL for atomic protocol registration.
From: Pravin B Shelar @ 2013-03-13 22:48 UTC (permalink / raw)
To: davem; +Cc: stephen, netdev, jesse, Pravin B Shelar
Currently gre demultiplexer uses its own spin-lock. We should
rather RTNL lock.
Suggested-by: Stephen Hemminger <--cc stephen@networkplumber.org>
Signed-off-by: Pravin B Shelar <pshelar@nicira.com>
---
drivers/net/ppp/pptp.c | 4 ++++
net/ipv4/gre.c | 18 +++++-------------
net/ipv4/ip_gre.c | 6 ++++++
3 files changed, 15 insertions(+), 13 deletions(-)
diff --git a/drivers/net/ppp/pptp.c b/drivers/net/ppp/pptp.c
index 162464f..672e7e3 100644
--- a/drivers/net/ppp/pptp.c
+++ b/drivers/net/ppp/pptp.c
@@ -673,7 +673,9 @@ static int __init pptp_init_module(void)
if (!callid_sock)
return -ENOMEM;
+ rtnl_lock();
err = gre_add_protocol(&gre_pptp_protocol, GREPROTO_PPTP);
+ rtnl_unlock();
if (err) {
pr_err("PPTP: can't add gre protocol\n");
goto out_mem_free;
@@ -707,7 +709,9 @@ static void __exit pptp_exit_module(void)
{
unregister_pppox_proto(PX_PROTO_PPTP);
proto_unregister(&pptp_sk_proto);
+ rtnl_lock();
gre_del_protocol(&gre_pptp_protocol, GREPROTO_PPTP);
+ rtnl_unlock();
vfree(callid_sock);
}
diff --git a/net/ipv4/gre.c b/net/ipv4/gre.c
index 7a4c710..fbaf33d 100644
--- a/net/ipv4/gre.c
+++ b/net/ipv4/gre.c
@@ -26,7 +26,6 @@
static const struct gre_protocol __rcu *gre_proto[GREPROTO_MAX] __read_mostly;
-static DEFINE_SPINLOCK(gre_proto_lock);
struct gre_base_hdr {
__be16 flags;
__be16 protocol;
@@ -35,19 +34,16 @@ struct gre_base_hdr {
int gre_add_protocol(const struct gre_protocol *proto, u8 version)
{
+ ASSERT_RTNL();
if (version >= GREPROTO_MAX)
goto err_out;
- spin_lock(&gre_proto_lock);
if (gre_proto[version])
- goto err_out_unlock;
+ goto err_out;
RCU_INIT_POINTER(gre_proto[version], proto);
- spin_unlock(&gre_proto_lock);
return 0;
-err_out_unlock:
- spin_unlock(&gre_proto_lock);
err_out:
return -1;
}
@@ -55,20 +51,16 @@ EXPORT_SYMBOL_GPL(gre_add_protocol);
int gre_del_protocol(const struct gre_protocol *proto, u8 version)
{
+ ASSERT_RTNL();
if (version >= GREPROTO_MAX)
goto err_out;
- spin_lock(&gre_proto_lock);
- if (rcu_dereference_protected(gre_proto[version],
- lockdep_is_held(&gre_proto_lock)) != proto)
- goto err_out_unlock;
+ if (rtnl_dereference(gre_proto[version]) != proto)
+ goto err_out;
RCU_INIT_POINTER(gre_proto[version], NULL);
- spin_unlock(&gre_proto_lock);
synchronize_rcu();
return 0;
-err_out_unlock:
- spin_unlock(&gre_proto_lock);
err_out:
return -1;
}
diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index a13a097..99cf2b4 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -1852,7 +1852,9 @@ static int __init ipgre_init(void)
if (err < 0)
return err;
+ rtnl_lock();
err = gre_add_protocol(&ipgre_protocol, GREPROTO_CISCO);
+ rtnl_unlock();
if (err < 0) {
pr_info("%s: can't add protocol\n", __func__);
goto add_proto_failed;
@@ -1872,7 +1874,9 @@ out:
tap_ops_failed:
rtnl_link_unregister(&ipgre_link_ops);
rtnl_link_failed:
+ rtnl_lock();
gre_del_protocol(&ipgre_protocol, GREPROTO_CISCO);
+ rtnl_unlock();
add_proto_failed:
unregister_pernet_device(&ipgre_net_ops);
goto out;
@@ -1882,8 +1886,10 @@ static void __exit ipgre_fini(void)
{
rtnl_link_unregister(&ipgre_tap_ops);
rtnl_link_unregister(&ipgre_link_ops);
+ rtnl_lock();
if (gre_del_protocol(&ipgre_protocol, GREPROTO_CISCO) < 0)
pr_info("%s: can't remove protocol\n", __func__);
+ rtnl_unlock();
unregister_pernet_device(&ipgre_net_ops);
}
--
1.7.1
^ permalink raw reply related
* [PATCH net-next 2/6] GRE: Refactor GRE tunneling code.
From: Pravin B Shelar @ 2013-03-13 22:48 UTC (permalink / raw)
To: davem; +Cc: netdev, jesse, Pravin B Shelar
Following patch refactors GRE code into ip tunneling code and GRE
specific code. Common tunneling code is moved to ip_tunnel module.
ip_tunnel module is written as generic library which can be used
by different tunneling implementations.
ip_tunnel module contains following components:
- packet xmit and rcv generic code. xmit flow looks like
(gre_xmit/ipip_xmit)->ip_tunnel_xmit->ip_local_out.
- hash table of all devices.
- lookup for tunnel devices.
- control plane operations like device create, destroy, ioctl, netlink
operations code.
- registration for tunneling modules, like gre, ipip etc.
- define single pcpu_tstats dev->tstats.
ipip.h header is renamed to ip_tunnel.h
Signed-off-by: Pravin B Shelar <pshelar@nicira.com>
---
drivers/net/vxlan.c | 2 +-
include/net/gre.h | 51 ++
include/net/ip_tunnels.h | 169 ++++++
include/net/ipip.h | 88 ---
net/ipv4/Kconfig | 5 +
net/ipv4/Makefile | 1 +
net/ipv4/af_inet.c | 1 -
net/ipv4/gre.c | 5 -
net/ipv4/ip_gre.c | 1490 ++++++++++------------------------------------
net/ipv4/ip_tunnel.c | 1036 ++++++++++++++++++++++++++++++++
net/ipv4/ip_vti.c | 2 +-
net/ipv4/ipip.c | 2 +-
net/ipv4/ipmr.c | 2 +-
net/ipv6/af_inet6.c | 1 -
net/ipv6/ip6_gre.c | 1 +
net/ipv6/ip6_tunnel.c | 1 +
net/ipv6/sit.c | 2 +-
17 files changed, 1569 insertions(+), 1290 deletions(-)
create mode 100644 include/net/ip_tunnels.h
delete mode 100644 include/net/ipip.h
create mode 100644 net/ipv4/ip_tunnel.c
diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index db0df07..338d44e 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -33,7 +33,7 @@
#include <net/arp.h>
#include <net/ndisc.h>
#include <net/ip.h>
-#include <net/ipip.h>
+#include <net/ip_tunnels.h>
#include <net/icmp.h>
#include <net/udp.h>
#include <net/rtnetlink.h>
diff --git a/include/net/gre.h b/include/net/gre.h
index 8266547..9f03a39 100644
--- a/include/net/gre.h
+++ b/include/net/gre.h
@@ -2,6 +2,7 @@
#define __LINUX_GRE_H
#include <linux/skbuff.h>
+#include <net/ip_tunnels.h>
#define GREPROTO_CISCO 0
#define GREPROTO_PPTP 1
@@ -12,7 +13,57 @@ struct gre_protocol {
void (*err_handler)(struct sk_buff *skb, u32 info);
};
+struct gre_base_hdr {
+ __be16 flags;
+ __be16 protocol;
+};
+#define GRE_HEADER_SECTION 4
+
int gre_add_protocol(const struct gre_protocol *proto, u8 version);
int gre_del_protocol(const struct gre_protocol *proto, u8 version);
+static inline __be16 gre_flags_to_tnl_flags(__be16 flags)
+{
+ __be16 tflags = 0;
+
+ if (flags & GRE_CSUM)
+ tflags |= TUNNEL_CSUM;
+ if (flags & GRE_ROUTING)
+ tflags |= TUNNEL_ROUTING;
+ if (flags & GRE_KEY)
+ tflags |= TUNNEL_KEY;
+ if (flags & GRE_SEQ)
+ tflags |= TUNNEL_SEQ;
+ if (flags & GRE_STRICT)
+ tflags |= TUNNEL_STRICT;
+ if (flags & GRE_REC)
+ tflags |= TUNNEL_REC;
+ if (flags & GRE_VERSION)
+ tflags |= TUNNEL_VERSION;
+
+ return tflags;
+}
+
+static inline __be16 tnl_flags_to_gre_flags(__be16 tflags)
+{
+ __be16 flags = 0;
+
+ if (tflags & TUNNEL_CSUM)
+ flags |= GRE_CSUM;
+ if (tflags & TUNNEL_ROUTING)
+ flags |= GRE_ROUTING;
+ if (tflags & TUNNEL_KEY)
+ flags |= GRE_KEY;
+ if (tflags & TUNNEL_SEQ)
+ flags |= GRE_SEQ;
+ if (tflags & TUNNEL_STRICT)
+ flags |= GRE_STRICT;
+ if (tflags & TUNNEL_REC)
+ flags |= GRE_REC;
+ if (tflags & TUNNEL_VERSION)
+ flags |= GRE_VERSION;
+
+ return flags;
+}
+
#endif
diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h
new file mode 100644
index 0000000..324d08f
--- /dev/null
+++ b/include/net/ip_tunnels.h
@@ -0,0 +1,169 @@
+#ifndef __NET_IP_TUNNEL_H
+#define __NET_IP_TUNNEL_H 1
+
+#include <linux/if_tunnel.h>
+#include <net/dsfield.h>
+#include <net/gro_cells.h>
+#include <net/inet_ecn.h>
+#include <net/ip.h>
+
+/* Keep error state on tunnel for 30 sec */
+#define IPTUNNEL_ERR_TIMEO (30*HZ)
+
+/* 6rd prefix/relay information */
+struct ip_tunnel_6rd_parm {
+ struct in6_addr prefix;
+ __be32 relay_prefix;
+ u16 prefixlen;
+ u16 relay_prefixlen;
+};
+
+struct ip_tunnel {
+ struct ip_tunnel __rcu *next;
+ struct hlist_node hash_node;
+ struct net_device *dev;
+
+ int err_count; /* Number of arrived ICMP errors */
+ unsigned long err_time; /* Time when the last ICMP error
+ * arrived */
+
+ /* These four fields used only by GRE */
+ __u32 i_seqno; /* The last seen seqno */
+ __u32 o_seqno; /* The last output seqno */
+ int hlen; /* Precalculated header length */
+ int mlink;
+
+ struct ip_tunnel_parm parms;
+
+ /* for SIT */
+#ifdef CONFIG_IPV6_SIT_6RD
+ struct ip_tunnel_6rd_parm ip6rd;
+#endif
+ struct ip_tunnel_prl_entry __rcu *prl; /* potential router list */
+ unsigned int prl_count; /* # of entries in PRL */
+ int ip_tnl_net_id;
+ struct gro_cells gro_cells;
+};
+
+#define TUNNEL_CSUM __cpu_to_be16(0x01)
+#define TUNNEL_ROUTING __cpu_to_be16(0x02)
+#define TUNNEL_KEY __cpu_to_be16(0x04)
+#define TUNNEL_SEQ __cpu_to_be16(0x08)
+#define TUNNEL_STRICT __cpu_to_be16(0x10)
+#define TUNNEL_REC __cpu_to_be16(0x20)
+#define TUNNEL_VERSION __cpu_to_be16(0x40)
+#define TUNNEL_NO_KEY __cpu_to_be16(0x80)
+
+struct tnl_ptk_info {
+ __be16 flags;
+ __be16 proto;
+ __be32 key;
+ __be32 seq;
+};
+
+struct ip_tunnel_prl_entry {
+ struct ip_tunnel_prl_entry __rcu *next;
+ __be32 addr;
+ u16 flags;
+ struct rcu_head rcu_head;
+};
+
+#define PACKET_RCVD 0
+#define PACKET_REJECT 1
+
+#define IP_TNL_HASH_BITS 10
+#define IP_TNL_HASH_SIZE (1 << IP_TNL_HASH_BITS)
+
+struct ip_tunnel_net {
+ struct hlist_head *tunnels;
+ struct net_device *fb_tunnel_dev;
+};
+
+int ip_tunnel_init(struct net_device *dev);
+void ip_tunnel_uninit(struct net_device *dev);
+void ip_tunnel_dellink(struct net_device *dev, struct list_head *head);
+int __net_init ip_tunnel_init_net(struct net *net, int ip_tnl_net_id,
+ struct rtnl_link_ops *ops, char *devname);
+
+void __net_exit ip_tunnel_delete_net(struct ip_tunnel_net *itn);
+
+int ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev,
+ const struct iphdr *tnl_params);
+int ip_tunnel_ioctl(struct net_device *dev, struct ip_tunnel_parm *p, int cmd);
+int ip_tunnel_change_mtu(struct net_device *dev, int new_mtu);
+
+struct rtnl_link_stats64 *ip_tunnel_get_stats64(struct net_device *dev,
+ struct rtnl_link_stats64 *tot);
+struct ip_tunnel *ip_tunnel_lookup(struct ip_tunnel_net *itn,
+ int link, __be16 flags,
+ __be32 remote, __be32 local,
+ __be32 key);
+
+int ip_tunnel_rcv(struct ip_tunnel *tunnel, struct sk_buff *skb,
+ const struct tnl_ptk_info *tpi, bool log_ecn_error);
+int ip_tunnel_changelink(struct net_device *dev, struct nlattr *tb[],
+ struct ip_tunnel_parm *p);
+int ip_tunnel_newlink(struct net_device *dev, struct nlattr *tb[],
+ struct ip_tunnel_parm *p);
+void ip_tunnel_setup(struct net_device *dev, int net_id);
+
+/* Extract dsfield from inner protocol */
+static inline u8 ip_tunnel_get_dsfield(const struct iphdr *iph,
+ const struct sk_buff *skb)
+{
+ if (skb->protocol == htons(ETH_P_IP))
+ return iph->tos;
+ else if (skb->protocol == htons(ETH_P_IPV6))
+ return ipv6_get_dsfield((const struct ipv6hdr *)iph);
+ else
+ return 0;
+}
+
+/* Propogate ECN bits out */
+static inline u8 ip_tunnel_ecn_encap(u8 tos,
+ const struct iphdr *iph,
+ const struct sk_buff *skb)
+{
+ u8 inner = ip_tunnel_get_dsfield(iph, skb);
+
+ return INET_ECN_encapsulate(tos, inner);
+}
+
+static inline void tunnel_ip_select_ident(struct sk_buff *skb,
+ const struct iphdr *old_iph,
+ struct dst_entry *dst)
+{
+ struct iphdr *iph = ip_hdr(skb);
+
+ if (iph->frag_off & htons(IP_DF))
+ iph->id = 0;
+ else {
+ /* Use inner packet iph-id if possible. */
+ if (skb->protocol == htons(ETH_P_IP) && old_iph->id)
+ iph->id = old_iph->id;
+ else
+ __ip_select_ident(iph, dst,
+ (skb_shinfo(skb)->gso_segs ?: 1) - 1);
+ }
+}
+
+static inline void iptunnel_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+ int err;
+ int pkt_len = skb->len - skb_transport_offset(skb);
+ struct pcpu_tstats *tstats = this_cpu_ptr(dev->tstats);
+
+ nf_reset(skb);
+
+ err = ip_local_out(skb);
+ if (likely(net_xmit_eval(err) == 0)) {
+ u64_stats_update_begin(&tstats->syncp);
+ tstats->tx_bytes += pkt_len;
+ tstats->tx_packets++;
+ u64_stats_update_end(&tstats->syncp);
+ } else {
+ dev->stats.tx_errors++;
+ dev->stats.tx_aborted_errors++;
+ }
+}
+#endif /* __NET_IP_TUNNEL_H */
diff --git a/include/net/ipip.h b/include/net/ipip.h
deleted file mode 100644
index 0c046e3..0000000
--- a/include/net/ipip.h
+++ /dev/null
@@ -1,88 +0,0 @@
-#ifndef __NET_IPIP_H
-#define __NET_IPIP_H 1
-
-#include <linux/if_tunnel.h>
-#include <net/gro_cells.h>
-#include <net/ip.h>
-
-/* Keep error state on tunnel for 30 sec */
-#define IPTUNNEL_ERR_TIMEO (30*HZ)
-
-/* 6rd prefix/relay information */
-struct ip_tunnel_6rd_parm {
- struct in6_addr prefix;
- __be32 relay_prefix;
- u16 prefixlen;
- u16 relay_prefixlen;
-};
-
-struct ip_tunnel {
- struct ip_tunnel __rcu *next;
- struct net_device *dev;
-
- int err_count; /* Number of arrived ICMP errors */
- unsigned long err_time; /* Time when the last ICMP error arrived */
-
- /* These four fields used only by GRE */
- __u32 i_seqno; /* The last seen seqno */
- __u32 o_seqno; /* The last output seqno */
- int hlen; /* Precalculated GRE header length */
- int mlink;
-
- struct ip_tunnel_parm parms;
-
- /* for SIT */
-#ifdef CONFIG_IPV6_SIT_6RD
- struct ip_tunnel_6rd_parm ip6rd;
-#endif
- struct ip_tunnel_prl_entry __rcu *prl; /* potential router list */
- unsigned int prl_count; /* # of entries in PRL */
-
- struct gro_cells gro_cells;
-};
-
-struct ip_tunnel_prl_entry {
- struct ip_tunnel_prl_entry __rcu *next;
- __be32 addr;
- u16 flags;
- struct rcu_head rcu_head;
-};
-
-static inline void iptunnel_xmit(struct sk_buff *skb, struct net_device *dev)
-{
- int err;
- int pkt_len = skb->len - skb_transport_offset(skb);
- struct pcpu_tstats *tstats = this_cpu_ptr(dev->tstats);
-
- nf_reset(skb);
-
- err = ip_local_out(skb);
- if (likely(net_xmit_eval(err) == 0)) {
- u64_stats_update_begin(&tstats->syncp);
- tstats->tx_bytes += pkt_len;
- tstats->tx_packets++;
- u64_stats_update_end(&tstats->syncp);
- } else {
- dev->stats.tx_errors++;
- dev->stats.tx_aborted_errors++;
- }
-}
-
-static inline void tunnel_ip_select_ident(struct sk_buff *skb,
- const struct iphdr *old_iph,
- struct dst_entry *dst)
-{
- struct iphdr *iph = ip_hdr(skb);
-
- if (iph->frag_off & htons(IP_DF))
- iph->id = 0;
- else {
- /* Use inner packet iph-id if possible. */
- if (skb->protocol == htons(ETH_P_IP) && old_iph->id)
- iph->id = old_iph->id;
- else
- __ip_select_ident(iph, dst,
- (skb_shinfo(skb)->gso_segs ?: 1) - 1);
- }
-}
-#endif
diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig
index 7944df7..2073226 100644
--- a/net/ipv4/Kconfig
+++ b/net/ipv4/Kconfig
@@ -186,9 +186,14 @@ config NET_IPGRE_DEMUX
This is helper module to demultiplex GRE packets on GRE version field criteria.
Required by ip_gre and pptp modules.
+config NET_IP_TUNNEL
+ tristate
+ default n
+
config NET_IPGRE
tristate "IP: GRE tunnels over IP"
depends on (IPV6 || IPV6=n) && NET_IPGRE_DEMUX
+ select NET_IP_TUNNEL
help
Tunneling means encapsulating data of one protocol type within
another protocol and sending it over a channel that understands the
diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile
index 15ca63e..089cb9f 100644
--- a/net/ipv4/Makefile
+++ b/net/ipv4/Makefile
@@ -13,6 +13,7 @@ obj-y := route.o inetpeer.o protocol.o \
fib_frontend.o fib_semantics.o fib_trie.o \
inet_fragment.o ping.o
+obj-$(CONFIG_NET_IP_TUNNEL) += ip_tunnel.o
obj-$(CONFIG_SYSCTL) += sysctl_net_ipv4.o
obj-$(CONFIG_PROC_FS) += proc.o
obj-$(CONFIG_IP_MULTIPLE_TABLES) += fib_rules.o
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 9e5882c..a1fca33 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -111,7 +111,6 @@
#include <net/sock.h>
#include <net/raw.h>
#include <net/icmp.h>
-#include <net/ipip.h>
#include <net/inet_common.h>
#include <net/xfrm.h>
#include <net/net_namespace.h>
diff --git a/net/ipv4/gre.c b/net/ipv4/gre.c
index fbaf33d..9d51d46 100644
--- a/net/ipv4/gre.c
+++ b/net/ipv4/gre.c
@@ -26,11 +26,6 @@
static const struct gre_protocol __rcu *gre_proto[GREPROTO_MAX] __read_mostly;
-struct gre_base_hdr {
- __be16 flags;
- __be16 protocol;
-};
-#define GRE_HEADER_SECTION 4
int gre_add_protocol(const struct gre_protocol *proto, u8 version)
{
diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index 99cf2b4..e40d7ba 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -37,7 +37,7 @@
#include <net/ip.h>
#include <net/icmp.h>
#include <net/protocol.h>
-#include <net/ipip.h>
+#include <net/ip_tunnels.h>
#include <net/arp.h>
#include <net/checksum.h>
#include <net/dsfield.h>
@@ -108,15 +108,6 @@
fatal route to network, even if it were you who configured
fatal static route: you are innocent. :-)
-
-
- 3. Really, ipv4/ipip.c, ipv4/ip_gre.c and ipv6/sit.c contain
- practically identical code. It would be good to glue them
- together, but it is not very evident, how to make them modular.
- sit is integral part of IPv6, ipip and gre are naturally modular.
- We could extract common parts (hash table, ioctl etc)
- to a separate module (ip_tunnel.c).
-
Alexey Kuznetsov.
*/
@@ -126,400 +117,129 @@ MODULE_PARM_DESC(log_ecn_error, "Log packets received with corrupted ECN");
static struct rtnl_link_ops ipgre_link_ops __read_mostly;
static int ipgre_tunnel_init(struct net_device *dev);
-static void ipgre_tunnel_setup(struct net_device *dev);
-static int ipgre_tunnel_bind_dev(struct net_device *dev);
-
-/* Fallback tunnel: no source, no destination, no key, no options */
-
-#define HASH_SIZE 16
static int ipgre_net_id __read_mostly;
-struct ipgre_net {
- struct ip_tunnel __rcu *tunnels[4][HASH_SIZE];
-
- struct net_device *fb_tunnel_dev;
-};
-
-/* Tunnel hash table */
-
-/*
- 4 hash tables:
-
- 3: (remote,local)
- 2: (remote,*)
- 1: (*,local)
- 0: (*,*)
-
- We require exact key match i.e. if a key is present in packet
- it will match only tunnel with the same key; if it is not present,
- it will match only keyless tunnel.
+static int ipgre_tap_net_id __read_mostly;
- All keysless packets, if not matched configured keyless tunnels
- will match fallback tunnel.
- */
+static __sum16 check_checksum(struct sk_buff *skb)
+{
+ __sum16 csum = 0;
-#define HASH(addr) (((__force u32)addr^((__force u32)addr>>4))&0xF)
+ switch (skb->ip_summed) {
+ case CHECKSUM_COMPLETE:
+ csum = csum_fold(skb->csum);
-#define tunnels_r_l tunnels[3]
-#define tunnels_r tunnels[2]
-#define tunnels_l tunnels[1]
-#define tunnels_wc tunnels[0]
+ if (!csum)
+ break;
+ /* Fall through. */
-static struct rtnl_link_stats64 *ipgre_get_stats64(struct net_device *dev,
- struct rtnl_link_stats64 *tot)
-{
- int i;
-
- for_each_possible_cpu(i) {
- const struct pcpu_tstats *tstats = per_cpu_ptr(dev->tstats, i);
- u64 rx_packets, rx_bytes, tx_packets, tx_bytes;
- unsigned int start;
-
- do {
- start = u64_stats_fetch_begin_bh(&tstats->syncp);
- rx_packets = tstats->rx_packets;
- tx_packets = tstats->tx_packets;
- rx_bytes = tstats->rx_bytes;
- tx_bytes = tstats->tx_bytes;
- } while (u64_stats_fetch_retry_bh(&tstats->syncp, start));
-
- tot->rx_packets += rx_packets;
- tot->tx_packets += tx_packets;
- tot->rx_bytes += rx_bytes;
- tot->tx_bytes += tx_bytes;
+ case CHECKSUM_NONE:
+ skb->csum = 0;
+ csum = __skb_checksum_complete(skb);
+ skb->ip_summed = CHECKSUM_COMPLETE;
+ break;
}
- tot->multicast = dev->stats.multicast;
- tot->rx_crc_errors = dev->stats.rx_crc_errors;
- tot->rx_fifo_errors = dev->stats.rx_fifo_errors;
- tot->rx_length_errors = dev->stats.rx_length_errors;
- tot->rx_frame_errors = dev->stats.rx_frame_errors;
- tot->rx_errors = dev->stats.rx_errors;
-
- tot->tx_fifo_errors = dev->stats.tx_fifo_errors;
- tot->tx_carrier_errors = dev->stats.tx_carrier_errors;
- tot->tx_dropped = dev->stats.tx_dropped;
- tot->tx_aborted_errors = dev->stats.tx_aborted_errors;
- tot->tx_errors = dev->stats.tx_errors;
-
- return tot;
+ return csum;
}
-/* Does key in tunnel parameters match packet */
-static bool ipgre_key_match(const struct ip_tunnel_parm *p,
- __be16 flags, __be32 key)
+static int ip_gre_calc_hlen(__be16 o_flags)
{
- if (p->i_flags & GRE_KEY) {
- if (flags & GRE_KEY)
- return key == p->i_key;
- else
- return false; /* key expected, none present */
- } else
- return !(flags & GRE_KEY);
-}
-
-/* Given src, dst and key, find appropriate for input tunnel. */
+ int addend = 4;
-static struct ip_tunnel *ipgre_tunnel_lookup(struct net_device *dev,
- __be32 remote, __be32 local,
- __be16 flags, __be32 key,
- __be16 gre_proto)
-{
- struct net *net = dev_net(dev);
- int link = dev->ifindex;
- unsigned int h0 = HASH(remote);
- unsigned int h1 = HASH(key);
- struct ip_tunnel *t, *cand = NULL;
- struct ipgre_net *ign = net_generic(net, ipgre_net_id);
- int dev_type = (gre_proto == htons(ETH_P_TEB)) ?
- ARPHRD_ETHER : ARPHRD_IPGRE;
- int score, cand_score = 4;
-
- for_each_ip_tunnel_rcu(t, ign->tunnels_r_l[h0 ^ h1]) {
- if (local != t->parms.iph.saddr ||
- remote != t->parms.iph.daddr ||
- !(t->dev->flags & IFF_UP))
- continue;
-
- if (!ipgre_key_match(&t->parms, flags, key))
- continue;
-
- if (t->dev->type != ARPHRD_IPGRE &&
- t->dev->type != dev_type)
- continue;
-
- score = 0;
- if (t->parms.link != link)
- score |= 1;
- if (t->dev->type != dev_type)
- score |= 2;
- if (score == 0)
- return t;
-
- if (score < cand_score) {
- cand = t;
- cand_score = score;
- }
- }
-
- for_each_ip_tunnel_rcu(t, ign->tunnels_r[h0 ^ h1]) {
- if (remote != t->parms.iph.daddr ||
- !(t->dev->flags & IFF_UP))
- continue;
-
- if (!ipgre_key_match(&t->parms, flags, key))
- continue;
-
- if (t->dev->type != ARPHRD_IPGRE &&
- t->dev->type != dev_type)
- continue;
-
- score = 0;
- if (t->parms.link != link)
- score |= 1;
- if (t->dev->type != dev_type)
- score |= 2;
- if (score == 0)
- return t;
-
- if (score < cand_score) {
- cand = t;
- cand_score = score;
- }
- }
-
- for_each_ip_tunnel_rcu(t, ign->tunnels_l[h1]) {
- if ((local != t->parms.iph.saddr &&
- (local != t->parms.iph.daddr ||
- !ipv4_is_multicast(local))) ||
- !(t->dev->flags & IFF_UP))
- continue;
-
- if (!ipgre_key_match(&t->parms, flags, key))
- continue;
-
- if (t->dev->type != ARPHRD_IPGRE &&
- t->dev->type != dev_type)
- continue;
-
- score = 0;
- if (t->parms.link != link)
- score |= 1;
- if (t->dev->type != dev_type)
- score |= 2;
- if (score == 0)
- return t;
-
- if (score < cand_score) {
- cand = t;
- cand_score = score;
- }
- }
-
- for_each_ip_tunnel_rcu(t, ign->tunnels_wc[h1]) {
- if (t->parms.i_key != key ||
- !(t->dev->flags & IFF_UP))
- continue;
-
- if (t->dev->type != ARPHRD_IPGRE &&
- t->dev->type != dev_type)
- continue;
-
- score = 0;
- if (t->parms.link != link)
- score |= 1;
- if (t->dev->type != dev_type)
- score |= 2;
- if (score == 0)
- return t;
-
- if (score < cand_score) {
- cand = t;
- cand_score = score;
- }
- }
-
- if (cand != NULL)
- return cand;
-
- dev = ign->fb_tunnel_dev;
- if (dev->flags & IFF_UP)
- return netdev_priv(dev);
-
- return NULL;
+ if (o_flags&TUNNEL_CSUM)
+ addend += 4;
+ if (o_flags&TUNNEL_KEY)
+ addend += 4;
+ if (o_flags&TUNNEL_SEQ)
+ addend += 4;
+ return addend;
}
-static struct ip_tunnel __rcu **__ipgre_bucket(struct ipgre_net *ign,
- struct ip_tunnel_parm *parms)
+static int parse_gre_header(struct sk_buff *skb, struct tnl_ptk_info *tpi,
+ bool *csum_err, int *hdr_len)
{
- __be32 remote = parms->iph.daddr;
- __be32 local = parms->iph.saddr;
- __be32 key = parms->i_key;
- unsigned int h = HASH(key);
- int prio = 0;
-
- if (local)
- prio |= 1;
- if (remote && !ipv4_is_multicast(remote)) {
- prio |= 2;
- h ^= HASH(remote);
- }
+ struct gre_base_hdr *greh = (struct gre_base_hdr *)skb->data;
+ __be32 *options = (__be32 *)(greh + 1);
- return &ign->tunnels[prio][h];
-}
+ if (unlikely(greh->flags & (GRE_VERSION | GRE_ROUTING)))
+ return -EINVAL;
-static inline struct ip_tunnel __rcu **ipgre_bucket(struct ipgre_net *ign,
- struct ip_tunnel *t)
-{
- return __ipgre_bucket(ign, &t->parms);
-}
+ tpi->flags = gre_flags_to_tnl_flags(greh->flags);
+ *hdr_len = ip_gre_calc_hlen(tpi->flags);
-static void ipgre_tunnel_link(struct ipgre_net *ign, struct ip_tunnel *t)
-{
- struct ip_tunnel __rcu **tp = ipgre_bucket(ign, t);
+ if (!pskb_may_pull(skb, *hdr_len))
+ return -EINVAL;
- rcu_assign_pointer(t->next, rtnl_dereference(*tp));
- rcu_assign_pointer(*tp, t);
-}
+ tpi->proto = greh->protocol;
-static void ipgre_tunnel_unlink(struct ipgre_net *ign, struct ip_tunnel *t)
-{
- struct ip_tunnel __rcu **tp;
- struct ip_tunnel *iter;
-
- for (tp = ipgre_bucket(ign, t);
- (iter = rtnl_dereference(*tp)) != NULL;
- tp = &iter->next) {
- if (t == iter) {
- rcu_assign_pointer(*tp, t->next);
- break;
+ if (greh->flags & GRE_CSUM) {
+ if (check_checksum(skb)) {
+ *csum_err = true;
+ return -EINVAL;
}
+ options++;
}
-}
-
-static struct ip_tunnel *ipgre_tunnel_find(struct net *net,
- struct ip_tunnel_parm *parms,
- int type)
-{
- __be32 remote = parms->iph.daddr;
- __be32 local = parms->iph.saddr;
- __be32 key = parms->i_key;
- int link = parms->link;
- struct ip_tunnel *t;
- struct ip_tunnel __rcu **tp;
- struct ipgre_net *ign = net_generic(net, ipgre_net_id);
-
- for (tp = __ipgre_bucket(ign, parms);
- (t = rtnl_dereference(*tp)) != NULL;
- tp = &t->next)
- if (local == t->parms.iph.saddr &&
- remote == t->parms.iph.daddr &&
- key == t->parms.i_key &&
- link == t->parms.link &&
- type == t->dev->type)
- break;
-
- return t;
-}
-
-static struct ip_tunnel *ipgre_tunnel_locate(struct net *net,
- struct ip_tunnel_parm *parms, int create)
-{
- struct ip_tunnel *t, *nt;
- struct net_device *dev;
- char name[IFNAMSIZ];
- struct ipgre_net *ign = net_generic(net, ipgre_net_id);
-
- t = ipgre_tunnel_find(net, parms, ARPHRD_IPGRE);
- if (t || !create)
- return t;
- if (parms->name[0])
- strlcpy(name, parms->name, IFNAMSIZ);
- else
- strcpy(name, "gre%d");
-
- dev = alloc_netdev(sizeof(*t), name, ipgre_tunnel_setup);
- if (!dev)
- return NULL;
-
- dev_net_set(dev, net);
-
- nt = netdev_priv(dev);
- nt->parms = *parms;
- dev->rtnl_link_ops = &ipgre_link_ops;
-
- dev->mtu = ipgre_tunnel_bind_dev(dev);
-
- if (register_netdevice(dev) < 0)
- goto failed_free;
-
- /* Can use a lockless transmit, unless we generate output sequences */
- if (!(nt->parms.o_flags & GRE_SEQ))
- dev->features |= NETIF_F_LLTX;
-
- dev_hold(dev);
- ipgre_tunnel_link(ign, nt);
- return nt;
+ if (greh->flags & GRE_KEY) {
+ tpi->key = *options;
+ options++;
+ } else
+ tpi->key = 0;
-failed_free:
- free_netdev(dev);
- return NULL;
-}
+ if (unlikely(greh->flags & GRE_SEQ)) {
+ tpi->seq = *options;
+ options++;
+ } else
+ tpi->seq = 0;
-static void ipgre_tunnel_uninit(struct net_device *dev)
-{
- struct net *net = dev_net(dev);
- struct ipgre_net *ign = net_generic(net, ipgre_net_id);
+ /* WCCP version 1 and 2 protocol decoding.
+ * - Change protocol to IP
+ * - When dealing with WCCPv2, Skip extra 4 bytes in GRE header
+ */
+ if (greh->flags == 0 && tpi->proto == htons(ETH_P_WCCP)) {
+ tpi->proto = htons(ETH_P_IP);
+ if ((*(u8 *)options & 0xF0) != 0x40) {
+ *hdr_len += 4;
+ if (!pskb_may_pull(skb, *hdr_len))
+ return -EINVAL;
+ }
+ }
- ipgre_tunnel_unlink(ign, netdev_priv(dev));
- dev_put(dev);
+ return 0;
}
-
static void ipgre_err(struct sk_buff *skb, u32 info)
{
-/* All the routers (except for Linux) return only
- 8 bytes of packet payload. It means, that precise relaying of
- ICMP in the real Internet is absolutely infeasible.
-
- Moreover, Cisco "wise men" put GRE key to the third word
- in GRE header. It makes impossible maintaining even soft state for keyed
- GRE tunnels with enabled checksum. Tell them "thank you".
+ /* All the routers (except for Linux) return only
+ 8 bytes of packet payload. It means, that precise relaying of
+ ICMP in the real Internet is absolutely infeasible.
- Well, I wonder, rfc1812 was written by Cisco employee,
- what the hell these idiots break standards established
- by themselves???
- */
+ Moreover, Cisco "wise men" put GRE key to the third word
+ in GRE header. It makes impossible maintaining even soft
+ state for keyed GRE tunnels with enabled checksum. Tell
+ them "thank you".
+ Well, I wonder, rfc1812 was written by Cisco employee,
+ what the hell these idiots break standards established
+ by themselves???
+ */
+ struct net *net = dev_net(skb->dev);
+ struct ip_tunnel_net *itn;
const struct iphdr *iph = (const struct iphdr *)skb->data;
- __be16 *p = (__be16 *)(skb->data+(iph->ihl<<2));
- int grehlen = (iph->ihl<<2) + 4;
const int type = icmp_hdr(skb)->type;
const int code = icmp_hdr(skb)->code;
struct ip_tunnel *t;
- __be16 flags;
- __be32 key = 0;
+ struct tnl_ptk_info tpi;
+ int hdr_len;
+ bool csum_err = false;
- flags = p[0];
- if (flags&(GRE_CSUM|GRE_KEY|GRE_SEQ|GRE_ROUTING|GRE_VERSION)) {
- if (flags&(GRE_VERSION|GRE_ROUTING))
+ if (parse_gre_header(skb, &tpi, &csum_err, &hdr_len)) {
+ if (!csum_err) /* ignore csum errors. */
return;
- if (flags&GRE_KEY) {
- grehlen += 4;
- if (flags&GRE_CSUM)
- grehlen += 4;
- }
}
- /* If only 8 bytes returned, keyed message will be dropped here */
- if (skb_headlen(skb) < grehlen)
- return;
-
- if (flags & GRE_KEY)
- key = *(((__be32 *)p) + (grehlen / 4) - 1);
-
switch (type) {
default:
case ICMP_PARAMETERPROB:
@@ -548,8 +268,13 @@ static void ipgre_err(struct sk_buff *skb, u32 info)
break;
}
- t = ipgre_tunnel_lookup(skb->dev, iph->daddr, iph->saddr,
- flags, key, p[1]);
+ if (tpi.proto == htons(ETH_P_TEB))
+ itn = net_generic(net, ipgre_tap_net_id);
+ else
+ itn = net_generic(net, ipgre_net_id);
+
+ t = ip_tunnel_lookup(itn, skb->dev->ifindex, tpi.flags,
+ iph->daddr, iph->saddr, tpi.key);
if (t == NULL)
return;
@@ -578,158 +303,33 @@ static void ipgre_err(struct sk_buff *skb, u32 info)
t->err_time = jiffies;
}
-static inline u8
-ipgre_ecn_encapsulate(u8 tos, const struct iphdr *old_iph, struct sk_buff *skb)
-{
- u8 inner = 0;
- if (skb->protocol == htons(ETH_P_IP))
- inner = old_iph->tos;
- else if (skb->protocol == htons(ETH_P_IPV6))
- inner = ipv6_get_dsfield((const struct ipv6hdr *)old_iph);
- return INET_ECN_encapsulate(tos, inner);
-}
-
static int ipgre_rcv(struct sk_buff *skb)
{
+ struct net *net = dev_net(skb->dev);
+ struct ip_tunnel_net *itn;
const struct iphdr *iph;
- u8 *h;
- __be16 flags;
- __sum16 csum = 0;
- __be32 key = 0;
- u32 seqno = 0;
struct ip_tunnel *tunnel;
- int offset = 4;
- __be16 gre_proto;
- int err;
+ struct tnl_ptk_info tpi;
+ int hdr_len;
+ bool csum_err = false;
- if (!pskb_may_pull(skb, 16))
+ if (parse_gre_header(skb, &tpi, &csum_err, &hdr_len) < 0)
goto drop;
- iph = ip_hdr(skb);
- h = skb->data;
- flags = *(__be16 *)h;
-
- if (flags&(GRE_CSUM|GRE_KEY|GRE_ROUTING|GRE_SEQ|GRE_VERSION)) {
- /* - Version must be 0.
- - We do not support routing headers.
- */
- if (flags&(GRE_VERSION|GRE_ROUTING))
- goto drop;
-
- if (flags&GRE_CSUM) {
- switch (skb->ip_summed) {
- case CHECKSUM_COMPLETE:
- csum = csum_fold(skb->csum);
- if (!csum)
- break;
- /* fall through */
- case CHECKSUM_NONE:
- skb->csum = 0;
- csum = __skb_checksum_complete(skb);
- skb->ip_summed = CHECKSUM_COMPLETE;
- }
- offset += 4;
- }
- if (flags&GRE_KEY) {
- key = *(__be32 *)(h + offset);
- offset += 4;
- }
- if (flags&GRE_SEQ) {
- seqno = ntohl(*(__be32 *)(h + offset));
- offset += 4;
- }
- }
+ if (tpi.proto == htons(ETH_P_TEB))
+ itn = net_generic(net, ipgre_tap_net_id);
+ else
+ itn = net_generic(net, ipgre_net_id);
- gre_proto = *(__be16 *)(h + 2);
+ iph = ip_hdr(skb);
+ tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, tpi.flags,
+ iph->saddr, iph->daddr, tpi.key);
- tunnel = ipgre_tunnel_lookup(skb->dev,
- iph->saddr, iph->daddr, flags, key,
- gre_proto);
if (tunnel) {
- struct pcpu_tstats *tstats;
-
- secpath_reset(skb);
-
- skb->protocol = gre_proto;
- /* WCCP version 1 and 2 protocol decoding.
- * - Change protocol to IP
- * - When dealing with WCCPv2, Skip extra 4 bytes in GRE header
- */
- if (flags == 0 && gre_proto == htons(ETH_P_WCCP)) {
- skb->protocol = htons(ETH_P_IP);
- if ((*(h + offset) & 0xF0) != 0x40)
- offset += 4;
- }
-
- skb->mac_header = skb->network_header;
- __pskb_pull(skb, offset);
- skb_postpull_rcsum(skb, skb_transport_header(skb), offset);
- skb->pkt_type = PACKET_HOST;
-#ifdef CONFIG_NET_IPGRE_BROADCAST
- if (ipv4_is_multicast(iph->daddr)) {
- /* Looped back packet, drop it! */
- if (rt_is_output_route(skb_rtable(skb)))
- goto drop;
- tunnel->dev->stats.multicast++;
- skb->pkt_type = PACKET_BROADCAST;
- }
-#endif
-
- if (((flags&GRE_CSUM) && csum) ||
- (!(flags&GRE_CSUM) && tunnel->parms.i_flags&GRE_CSUM)) {
- tunnel->dev->stats.rx_crc_errors++;
- tunnel->dev->stats.rx_errors++;
- goto drop;
- }
- if (tunnel->parms.i_flags&GRE_SEQ) {
- if (!(flags&GRE_SEQ) ||
- (tunnel->i_seqno && (s32)(seqno - tunnel->i_seqno) < 0)) {
- tunnel->dev->stats.rx_fifo_errors++;
- tunnel->dev->stats.rx_errors++;
- goto drop;
- }
- tunnel->i_seqno = seqno + 1;
- }
-
- /* Warning: All skb pointers will be invalidated! */
- if (tunnel->dev->type == ARPHRD_ETHER) {
- if (!pskb_may_pull(skb, ETH_HLEN)) {
- tunnel->dev->stats.rx_length_errors++;
- tunnel->dev->stats.rx_errors++;
- goto drop;
- }
-
- iph = ip_hdr(skb);
- skb->protocol = eth_type_trans(skb, tunnel->dev);
- skb_postpull_rcsum(skb, eth_hdr(skb), ETH_HLEN);
- }
-
- __skb_tunnel_rx(skb, tunnel->dev);
-
- skb_reset_network_header(skb);
- err = IP_ECN_decapsulate(iph, skb);
- if (unlikely(err)) {
- if (log_ecn_error)
- net_info_ratelimited("non-ECT from %pI4 with TOS=%#x\n",
- &iph->saddr, iph->tos);
- if (err > 1) {
- ++tunnel->dev->stats.rx_frame_errors;
- ++tunnel->dev->stats.rx_errors;
- goto drop;
- }
- }
-
- tstats = this_cpu_ptr(tunnel->dev->tstats);
- u64_stats_update_begin(&tstats->syncp);
- tstats->rx_packets++;
- tstats->rx_bytes += skb->len;
- u64_stats_update_end(&tstats->syncp);
-
- gro_cells_receive(&tunnel->gro_cells, skb);
+ ip_tunnel_rcv(tunnel, skb, &tpi, log_ecn_error);
return 0;
}
icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0);
-
drop:
kfree_skb(skb);
return 0;
@@ -746,7 +346,7 @@ static struct sk_buff *handle_offloads(struct ip_tunnel *tunnel, struct sk_buff
skb_shinfo(skb)->gso_type |= SKB_GSO_GRE;
return skb;
} else if (skb->ip_summed == CHECKSUM_PARTIAL &&
- tunnel->parms.o_flags&GRE_CSUM) {
+ tunnel->parms.o_flags&TUNNEL_CSUM) {
err = skb_checksum_help(skb);
if (unlikely(err))
goto error;
@@ -760,483 +360,145 @@ error:
return ERR_PTR(err);
}
-static netdev_tx_t ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
+static struct sk_buff *gre_build_header(struct sk_buff *skb,
+ const struct tnl_ptk_info *tpi,
+ int hdr_len)
{
- struct ip_tunnel *tunnel = netdev_priv(dev);
- const struct iphdr *old_iph;
- const struct iphdr *tiph;
- struct flowi4 fl4;
- u8 tos;
- __be16 df;
- struct rtable *rt; /* Route to the other host */
- struct net_device *tdev; /* Device to other host */
- struct iphdr *iph; /* Our new IP header */
- unsigned int max_headroom; /* The extra header space needed */
- int gre_hlen;
- __be32 dst;
- int mtu;
- u8 ttl;
- int err;
-
- skb = handle_offloads(tunnel, skb);
- if (IS_ERR(skb)) {
- dev->stats.tx_dropped++;
- return NETDEV_TX_OK;
- }
+ struct gre_base_hdr *greh;
- if (!skb->encapsulation) {
- skb_reset_inner_headers(skb);
- skb->encapsulation = 1;
- }
+ skb_push(skb, hdr_len);
- old_iph = ip_hdr(skb);
+ greh = (struct gre_base_hdr *)skb->data;
+ greh->flags = tnl_flags_to_gre_flags(tpi->flags);
+ greh->protocol = tpi->proto;
- if (dev->type == ARPHRD_ETHER)
- IPCB(skb)->flags = 0;
+ if (tpi->flags&(TUNNEL_KEY|TUNNEL_CSUM|TUNNEL_SEQ)) {
+ __be32 *ptr = (__be32 *)(((u8 *)greh) + hdr_len - 4);
- if (dev->header_ops && dev->type == ARPHRD_IPGRE) {
- gre_hlen = 0;
- if (skb->protocol == htons(ETH_P_IP))
- tiph = (const struct iphdr *)skb->data;
- else
- tiph = &tunnel->parms.iph;
- } else {
- gre_hlen = tunnel->hlen;
- tiph = &tunnel->parms.iph;
- }
-
- if ((dst = tiph->daddr) == 0) {
- /* NBMA tunnel */
-
- if (skb_dst(skb) == NULL) {
- dev->stats.tx_fifo_errors++;
- goto tx_error;
- }
-
- if (skb->protocol == htons(ETH_P_IP)) {
- rt = skb_rtable(skb);
- dst = rt_nexthop(rt, old_iph->daddr);
- }
-#if IS_ENABLED(CONFIG_IPV6)
- else if (skb->protocol == htons(ETH_P_IPV6)) {
- const struct in6_addr *addr6;
- struct neighbour *neigh;
- bool do_tx_error_icmp;
- int addr_type;
-
- neigh = dst_neigh_lookup(skb_dst(skb), &ipv6_hdr(skb)->daddr);
- if (neigh == NULL)
- goto tx_error;
-
- addr6 = (const struct in6_addr *)&neigh->primary_key;
- addr_type = ipv6_addr_type(addr6);
-
- if (addr_type == IPV6_ADDR_ANY) {
- addr6 = &ipv6_hdr(skb)->daddr;
- addr_type = ipv6_addr_type(addr6);
- }
-
- if ((addr_type & IPV6_ADDR_COMPATv4) == 0)
- do_tx_error_icmp = true;
- else {
- do_tx_error_icmp = false;
- dst = addr6->s6_addr32[3];
- }
- neigh_release(neigh);
- if (do_tx_error_icmp)
- goto tx_error_icmp;
- }
-#endif
- else
- goto tx_error;
- }
-
- ttl = tiph->ttl;
- tos = tiph->tos;
- if (tos & 0x1) {
- tos &= ~0x1;
- if (skb->protocol == htons(ETH_P_IP))
- tos = old_iph->tos;
- else if (skb->protocol == htons(ETH_P_IPV6))
- tos = ipv6_get_dsfield((const struct ipv6hdr *)old_iph);
- }
-
- rt = ip_route_output_gre(dev_net(dev), &fl4, dst, tiph->saddr,
- tunnel->parms.o_key, RT_TOS(tos),
- tunnel->parms.link);
- if (IS_ERR(rt)) {
- dev->stats.tx_carrier_errors++;
- goto tx_error;
- }
- tdev = rt->dst.dev;
-
- if (tdev == dev) {
- ip_rt_put(rt);
- dev->stats.collisions++;
- goto tx_error;
- }
-
- df = tiph->frag_off;
- if (df)
- mtu = dst_mtu(&rt->dst) - dev->hard_header_len - tunnel->hlen;
- else
- mtu = skb_dst(skb) ? dst_mtu(skb_dst(skb)) : dev->mtu;
-
- if (skb_dst(skb))
- skb_dst(skb)->ops->update_pmtu(skb_dst(skb), NULL, skb, mtu);
-
- if (skb->protocol == htons(ETH_P_IP)) {
- df |= (old_iph->frag_off&htons(IP_DF));
-
- if (!skb_is_gso(skb) &&
- (old_iph->frag_off&htons(IP_DF)) &&
- mtu < ntohs(old_iph->tot_len)) {
- icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED, htonl(mtu));
- ip_rt_put(rt);
- goto tx_error;
+ if (tpi->flags&TUNNEL_SEQ) {
+ *ptr = tpi->seq;
+ ptr--;
}
- }
-#if IS_ENABLED(CONFIG_IPV6)
- else if (skb->protocol == htons(ETH_P_IPV6)) {
- struct rt6_info *rt6 = (struct rt6_info *)skb_dst(skb);
-
- if (rt6 && mtu < dst_mtu(skb_dst(skb)) && mtu >= IPV6_MIN_MTU) {
- if ((tunnel->parms.iph.daddr &&
- !ipv4_is_multicast(tunnel->parms.iph.daddr)) ||
- rt6->rt6i_dst.plen == 128) {
- rt6->rt6i_flags |= RTF_MODIFIED;
- dst_metric_set(skb_dst(skb), RTAX_MTU, mtu);
- }
+ if (tpi->flags&TUNNEL_KEY) {
+ *ptr = tpi->key;
+ ptr--;
}
-
- if (!skb_is_gso(skb) &&
- mtu >= IPV6_MIN_MTU &&
- mtu < skb->len - tunnel->hlen + gre_hlen) {
- icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu);
- ip_rt_put(rt);
- goto tx_error;
+ if (tpi->flags&TUNNEL_CSUM &&
+ !(skb_shinfo(skb)->gso_type & SKB_GSO_GRE)) {
+ *(__sum16 *)ptr = 0;
+ *(__sum16 *)ptr = csum_fold(skb_checksum(skb, 0,
+ skb->len, 0));
}
}
-#endif
- if (tunnel->err_count > 0) {
- if (time_before(jiffies,
- tunnel->err_time + IPTUNNEL_ERR_TIMEO)) {
- tunnel->err_count--;
+ return skb;
+}
- dst_link_failure(skb);
- } else
- tunnel->err_count = 0;
- }
+static netdev_tx_t __gre_xmit(struct sk_buff *skb, struct net_device *dev,
+ const struct iphdr *tnl_params,
+ __be16 proto)
+{
+ struct ip_tunnel *tunnel = netdev_priv(dev);
+ struct tnl_ptk_info tpi;
- max_headroom = LL_RESERVED_SPACE(tdev) + gre_hlen + rt->dst.header_len;
-
- if (skb_headroom(skb) < max_headroom || skb_shared(skb)||
- (skb_cloned(skb) && !skb_clone_writable(skb, 0))) {
- struct sk_buff *new_skb = skb_realloc_headroom(skb, max_headroom);
- if (max_headroom > dev->needed_headroom)
- dev->needed_headroom = max_headroom;
- if (!new_skb) {
- ip_rt_put(rt);
- dev->stats.tx_dropped++;
- dev_kfree_skb(skb);
- return NETDEV_TX_OK;
- }
- if (skb->sk)
- skb_set_owner_w(new_skb, skb->sk);
- dev_kfree_skb(skb);
- skb = new_skb;
- old_iph = ip_hdr(skb);
- /* Warning : tiph value might point to freed memory */
+ if (!skb->encapsulation) {
+ skb_reset_inner_headers(skb);
+ skb->encapsulation = 1;
}
- skb_push(skb, gre_hlen);
- skb_reset_network_header(skb);
- skb_set_transport_header(skb, sizeof(*iph));
- memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt));
- IPCB(skb)->flags &= ~(IPSKB_XFRM_TUNNEL_SIZE | IPSKB_XFRM_TRANSFORMED |
- IPSKB_REROUTED);
- skb_dst_drop(skb);
- skb_dst_set(skb, &rt->dst);
-
- /*
- * Push down and install the IPIP header.
- */
-
- iph = ip_hdr(skb);
- iph->version = 4;
- iph->ihl = sizeof(struct iphdr) >> 2;
- iph->frag_off = df;
- iph->protocol = IPPROTO_GRE;
- iph->tos = ipgre_ecn_encapsulate(tos, old_iph, skb);
- iph->daddr = fl4.daddr;
- iph->saddr = fl4.saddr;
- iph->ttl = ttl;
+ tpi.flags = tunnel->parms.o_flags;
+ tpi.proto = proto;
+ tpi.key = tunnel->parms.o_key;
+ if (tunnel->parms.o_flags & TUNNEL_SEQ)
+ tunnel->o_seqno++;
+ tpi.seq = htonl(tunnel->o_seqno);
- tunnel_ip_select_ident(skb, old_iph, &rt->dst);
-
- if (ttl == 0) {
- if (skb->protocol == htons(ETH_P_IP))
- iph->ttl = old_iph->ttl;
-#if IS_ENABLED(CONFIG_IPV6)
- else if (skb->protocol == htons(ETH_P_IPV6))
- iph->ttl = ((const struct ipv6hdr *)old_iph)->hop_limit;
-#endif
- else
- iph->ttl = ip4_dst_hoplimit(&rt->dst);
- }
+ /* Need space for new headers (invalidates iph ptr) */
+ if (skb_cow_head(skb, dev->needed_headroom))
+ goto drop;
- ((__be16 *)(iph + 1))[0] = tunnel->parms.o_flags;
- ((__be16 *)(iph + 1))[1] = (dev->type == ARPHRD_ETHER) ?
- htons(ETH_P_TEB) : skb->protocol;
+ /* Push GRE header. */
+ skb = gre_build_header(skb, &tpi, tunnel->hlen);
+ if (!skb)
+ goto out;
- if (tunnel->parms.o_flags&(GRE_KEY|GRE_CSUM|GRE_SEQ)) {
- __be32 *ptr = (__be32 *)(((u8 *)iph) + tunnel->hlen - 4);
+ ip_tunnel_xmit(skb, dev, tnl_params);
- if (tunnel->parms.o_flags&GRE_SEQ) {
- ++tunnel->o_seqno;
- *ptr = htonl(tunnel->o_seqno);
- ptr--;
- }
- if (tunnel->parms.o_flags&GRE_KEY) {
- *ptr = tunnel->parms.o_key;
- ptr--;
- }
- /* Skip GRE checksum if skb is getting offloaded. */
- if (!(skb_shinfo(skb)->gso_type & SKB_GSO_GRE) &&
- (tunnel->parms.o_flags&GRE_CSUM)) {
- int offset = skb_transport_offset(skb);
-
- if (skb_has_shared_frag(skb)) {
- err = __skb_linearize(skb);
- if (err)
- goto tx_error;
- }
-
- *ptr = 0;
- *(__sum16 *)ptr = csum_fold(skb_checksum(skb, offset,
- skb->len - offset,
- 0));
- }
- }
-
- iptunnel_xmit(skb, dev);
return NETDEV_TX_OK;
-
-#if IS_ENABLED(CONFIG_IPV6)
-tx_error_icmp:
- dst_link_failure(skb);
-#endif
-tx_error:
- dev->stats.tx_errors++;
+drop:
dev_kfree_skb(skb);
+out:
+ dev->stats.tx_dropped++;
return NETDEV_TX_OK;
}
-static int ipgre_tunnel_bind_dev(struct net_device *dev)
+static netdev_tx_t ipgre_tap_xmit(struct sk_buff *skb, struct net_device *dev)
{
- struct net_device *tdev = NULL;
- struct ip_tunnel *tunnel;
- const struct iphdr *iph;
- int hlen = LL_MAX_HEADER;
- int mtu = ETH_DATA_LEN;
- int addend = sizeof(struct iphdr) + 4;
-
- tunnel = netdev_priv(dev);
- iph = &tunnel->parms.iph;
-
- /* Guess output device to choose reasonable mtu and needed_headroom */
-
- if (iph->daddr) {
- struct flowi4 fl4;
- struct rtable *rt;
-
- rt = ip_route_output_gre(dev_net(dev), &fl4,
- iph->daddr, iph->saddr,
- tunnel->parms.o_key,
- RT_TOS(iph->tos),
- tunnel->parms.link);
- if (!IS_ERR(rt)) {
- tdev = rt->dst.dev;
- ip_rt_put(rt);
- }
+ struct ip_tunnel *tunnel = netdev_priv(dev);
+ const struct iphdr *tnl_params = &tunnel->parms.iph;
- if (dev->type != ARPHRD_ETHER)
- dev->flags |= IFF_POINTOPOINT;
+ skb = handle_offloads(tunnel, skb);
+ if (IS_ERR(skb)) {
+ dev->stats.tx_dropped++;
+ return NETDEV_TX_OK;
}
- if (!tdev && tunnel->parms.link)
- tdev = __dev_get_by_index(dev_net(dev), tunnel->parms.link);
+ return __gre_xmit(skb, dev, tnl_params, htons(ETH_P_TEB));
+}
- if (tdev) {
- hlen = tdev->hard_header_len + tdev->needed_headroom;
- mtu = tdev->mtu;
- }
- dev->iflink = tunnel->parms.link;
-
- /* Precalculate GRE options length */
- if (tunnel->parms.o_flags&(GRE_CSUM|GRE_KEY|GRE_SEQ)) {
- if (tunnel->parms.o_flags&GRE_CSUM)
- addend += 4;
- if (tunnel->parms.o_flags&GRE_KEY)
- addend += 4;
- if (tunnel->parms.o_flags&GRE_SEQ)
- addend += 4;
- }
- dev->needed_headroom = addend + hlen;
- mtu -= dev->hard_header_len + addend;
+static netdev_tx_t ipgre_tunnel_xmit(struct sk_buff *skb,
+ struct net_device *dev)
+{
+ struct ip_tunnel *tunnel = netdev_priv(dev);
+ const struct iphdr *tnl_params;
- if (mtu < 68)
- mtu = 68;
+ skb = handle_offloads(tunnel, skb);
+ if (IS_ERR(skb)) {
+ dev->stats.tx_dropped++;
+ return NETDEV_TX_OK;
+ }
+ if (dev->header_ops) {
+ if (skb->protocol == htons(ETH_P_IP))
+ tnl_params = (const struct iphdr *)skb->data;
+ else
+ tnl_params = &tunnel->parms.iph;
- tunnel->hlen = addend;
- /* TCP offload with GRE SEQ is not supported. */
- if (!(tunnel->parms.o_flags & GRE_SEQ)) {
- dev->features |= NETIF_F_GSO_SOFTWARE;
- dev->hw_features |= NETIF_F_GSO_SOFTWARE;
+ /* Pull skb since ip_tunnel_xmit() needs skb->data pointing
+ * to gre header.
+ */
+ skb_pull(skb, tunnel->hlen + sizeof(struct iphdr));
+ } else {
+ tnl_params = &tunnel->parms.iph;
}
- return mtu;
+ return __gre_xmit(skb, dev, tnl_params, skb->protocol);
}
-static int
-ipgre_tunnel_ioctl (struct net_device *dev, struct ifreq *ifr, int cmd)
+static int ipgre_tunnel_ioctl(struct net_device *dev,
+ struct ifreq *ifr, int cmd)
{
int err = 0;
struct ip_tunnel_parm p;
- struct ip_tunnel *t;
- struct net *net = dev_net(dev);
- struct ipgre_net *ign = net_generic(net, ipgre_net_id);
-
- switch (cmd) {
- case SIOCGETTUNNEL:
- t = NULL;
- if (dev == ign->fb_tunnel_dev) {
- if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p))) {
- err = -EFAULT;
- break;
- }
- t = ipgre_tunnel_locate(net, &p, 0);
- }
- if (t == NULL)
- t = netdev_priv(dev);
- memcpy(&p, &t->parms, sizeof(p));
- if (copy_to_user(ifr->ifr_ifru.ifru_data, &p, sizeof(p)))
- err = -EFAULT;
- break;
-
- case SIOCADDTUNNEL:
- case SIOCCHGTUNNEL:
- err = -EPERM;
- if (!ns_capable(net->user_ns, CAP_NET_ADMIN))
- goto done;
-
- err = -EFAULT;
- if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p)))
- goto done;
-
- err = -EINVAL;
- if (p.iph.version != 4 || p.iph.protocol != IPPROTO_GRE ||
- p.iph.ihl != 5 || (p.iph.frag_off&htons(~IP_DF)) ||
- ((p.i_flags|p.o_flags)&(GRE_VERSION|GRE_ROUTING)))
- goto done;
- if (p.iph.ttl)
- p.iph.frag_off |= htons(IP_DF);
-
- if (!(p.i_flags&GRE_KEY))
- p.i_key = 0;
- if (!(p.o_flags&GRE_KEY))
- p.o_key = 0;
-
- t = ipgre_tunnel_locate(net, &p, cmd == SIOCADDTUNNEL);
-
- if (dev != ign->fb_tunnel_dev && cmd == SIOCCHGTUNNEL) {
- if (t != NULL) {
- if (t->dev != dev) {
- err = -EEXIST;
- break;
- }
- } else {
- unsigned int nflags = 0;
-
- t = netdev_priv(dev);
-
- if (ipv4_is_multicast(p.iph.daddr))
- nflags = IFF_BROADCAST;
- else if (p.iph.daddr)
- nflags = IFF_POINTOPOINT;
-
- if ((dev->flags^nflags)&(IFF_POINTOPOINT|IFF_BROADCAST)) {
- err = -EINVAL;
- break;
- }
- ipgre_tunnel_unlink(ign, t);
- synchronize_net();
- t->parms.iph.saddr = p.iph.saddr;
- t->parms.iph.daddr = p.iph.daddr;
- t->parms.i_key = p.i_key;
- t->parms.o_key = p.o_key;
- memcpy(dev->dev_addr, &p.iph.saddr, 4);
- memcpy(dev->broadcast, &p.iph.daddr, 4);
- ipgre_tunnel_link(ign, t);
- netdev_state_change(dev);
- }
- }
-
- if (t) {
- err = 0;
- if (cmd == SIOCCHGTUNNEL) {
- t->parms.iph.ttl = p.iph.ttl;
- t->parms.iph.tos = p.iph.tos;
- t->parms.iph.frag_off = p.iph.frag_off;
- if (t->parms.link != p.link) {
- t->parms.link = p.link;
- dev->mtu = ipgre_tunnel_bind_dev(dev);
- netdev_state_change(dev);
- }
- }
- if (copy_to_user(ifr->ifr_ifru.ifru_data, &t->parms, sizeof(p)))
- err = -EFAULT;
- } else
- err = (cmd == SIOCADDTUNNEL ? -ENOBUFS : -ENOENT);
- break;
-
- case SIOCDELTUNNEL:
- err = -EPERM;
- if (!ns_capable(net->user_ns, CAP_NET_ADMIN))
- goto done;
-
- if (dev == ign->fb_tunnel_dev) {
- err = -EFAULT;
- if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p)))
- goto done;
- err = -ENOENT;
- if ((t = ipgre_tunnel_locate(net, &p, 0)) == NULL)
- goto done;
- err = -EPERM;
- if (t == netdev_priv(ign->fb_tunnel_dev))
- goto done;
- dev = t->dev;
- }
- unregister_netdevice(dev);
- err = 0;
- break;
- default:
- err = -EINVAL;
+ if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p)))
+ return -EFAULT;
+ if (p.iph.version != 4 || p.iph.protocol != IPPROTO_GRE ||
+ p.iph.ihl != 5 || (p.iph.frag_off&htons(~IP_DF)) ||
+ ((p.i_flags|p.o_flags)&(GRE_VERSION|GRE_ROUTING))) {
+ return -EINVAL;
}
+ p.i_flags = gre_flags_to_tnl_flags(p.i_flags);
+ p.o_flags = gre_flags_to_tnl_flags(p.o_flags);
-done:
- return err;
-}
+ err = ip_tunnel_ioctl(dev, &p, cmd);
+ if (err)
+ return err;
-static int ipgre_tunnel_change_mtu(struct net_device *dev, int new_mtu)
-{
- struct ip_tunnel *tunnel = netdev_priv(dev);
- if (new_mtu < 68 ||
- new_mtu > 0xFFF8 - dev->hard_header_len - tunnel->hlen)
- return -EINVAL;
- dev->mtu = new_mtu;
+ p.i_flags = tnl_flags_to_gre_flags(p.i_flags);
+ p.o_flags = tnl_flags_to_gre_flags(p.o_flags);
+
+ if (copy_to_user(ifr->ifr_ifru.ifru_data, &p, sizeof(p)))
+ return -EFAULT;
return 0;
}
@@ -1274,17 +536,17 @@ static int ipgre_header(struct sk_buff *skb, struct net_device *dev,
const void *daddr, const void *saddr, unsigned int len)
{
struct ip_tunnel *t = netdev_priv(dev);
- struct iphdr *iph = (struct iphdr *)skb_push(skb, t->hlen);
- __be16 *p = (__be16 *)(iph+1);
+ struct iphdr *iph;
+ struct gre_base_hdr *greh;
- memcpy(iph, &t->parms.iph, sizeof(struct iphdr));
- p[0] = t->parms.o_flags;
- p[1] = htons(type);
+ iph = (struct iphdr *)skb_push(skb, t->hlen + sizeof(*iph));
+ greh = (struct gre_base_hdr *)(iph+1);
+ greh->flags = tnl_flags_to_gre_flags(t->parms.o_flags);
+ greh->protocol = htons(type);
- /*
- * Set the source hardware address.
- */
+ memcpy(iph, &t->parms.iph, sizeof(struct iphdr));
+ /* Set the source hardware address. */
if (saddr)
memcpy(&iph->saddr, saddr, 4);
if (daddr)
@@ -1292,7 +554,7 @@ static int ipgre_header(struct sk_buff *skb, struct net_device *dev,
if (iph->daddr)
return t->hlen;
- return -t->hlen;
+ return -(t->hlen + sizeof(*iph));
}
static int ipgre_header_parse(const struct sk_buff *skb, unsigned char *haddr)
@@ -1351,26 +613,17 @@ static int ipgre_close(struct net_device *dev)
static const struct net_device_ops ipgre_netdev_ops = {
.ndo_init = ipgre_tunnel_init,
- .ndo_uninit = ipgre_tunnel_uninit,
+ .ndo_uninit = ip_tunnel_uninit,
#ifdef CONFIG_NET_IPGRE_BROADCAST
.ndo_open = ipgre_open,
.ndo_stop = ipgre_close,
#endif
.ndo_start_xmit = ipgre_tunnel_xmit,
.ndo_do_ioctl = ipgre_tunnel_ioctl,
- .ndo_change_mtu = ipgre_tunnel_change_mtu,
- .ndo_get_stats64 = ipgre_get_stats64,
+ .ndo_change_mtu = ip_tunnel_change_mtu,
+ .ndo_get_stats64 = ip_tunnel_get_stats64,
};
-static void ipgre_dev_free(struct net_device *dev)
-{
- struct ip_tunnel *tunnel = netdev_priv(dev);
-
- gro_cells_destroy(&tunnel->gro_cells);
- free_percpu(dev->tstats);
- free_netdev(dev);
-}
-
#define GRE_FEATURES (NETIF_F_SG | \
NETIF_F_FRAGLIST | \
NETIF_F_HIGHDMA | \
@@ -1379,35 +632,49 @@ static void ipgre_dev_free(struct net_device *dev)
static void ipgre_tunnel_setup(struct net_device *dev)
{
dev->netdev_ops = &ipgre_netdev_ops;
- dev->destructor = ipgre_dev_free;
+ ip_tunnel_setup(dev, ipgre_net_id);
+}
- dev->type = ARPHRD_IPGRE;
- dev->needed_headroom = LL_MAX_HEADER + sizeof(struct iphdr) + 4;
+static void __gre_tunnel_init(struct net_device *dev)
+{
+ struct ip_tunnel *tunnel;
+
+ tunnel = netdev_priv(dev);
+ tunnel->hlen = ip_gre_calc_hlen(tunnel->parms.o_flags);
+ tunnel->parms.iph.protocol = IPPROTO_GRE;
+
+ dev->needed_headroom = LL_MAX_HEADER + sizeof(struct iphdr) + 4;
dev->mtu = ETH_DATA_LEN - sizeof(struct iphdr) - 4;
- dev->flags = IFF_NOARP;
dev->iflink = 0;
- dev->addr_len = 4;
- dev->features |= NETIF_F_NETNS_LOCAL;
- dev->priv_flags &= ~IFF_XMIT_DST_RELEASE;
- dev->features |= GRE_FEATURES;
+ dev->features |= NETIF_F_NETNS_LOCAL | GRE_FEATURES;
dev->hw_features |= GRE_FEATURES;
+
+ if (!(tunnel->parms.o_flags & TUNNEL_SEQ)) {
+ /* TCP offload with GRE SEQ is not supported. */
+ dev->features |= NETIF_F_GSO_SOFTWARE;
+ dev->hw_features |= NETIF_F_GSO_SOFTWARE;
+ /* Can use a lockless transmit, unless we generate
+ * output sequences
+ */
+ dev->features |= NETIF_F_LLTX;
+ }
}
static int ipgre_tunnel_init(struct net_device *dev)
{
- struct ip_tunnel *tunnel;
- struct iphdr *iph;
- int err;
+ struct ip_tunnel *tunnel = netdev_priv(dev);
+ struct iphdr *iph = &tunnel->parms.iph;
- tunnel = netdev_priv(dev);
- iph = &tunnel->parms.iph;
+ __gre_tunnel_init(dev);
- tunnel->dev = dev;
- strcpy(tunnel->parms.name, dev->name);
+ memcpy(dev->dev_addr, &iph->saddr, 4);
+ memcpy(dev->broadcast, &iph->daddr, 4);
- memcpy(dev->dev_addr, &tunnel->parms.iph.saddr, 4);
- memcpy(dev->broadcast, &tunnel->parms.iph.daddr, 4);
+ dev->type = ARPHRD_IPGRE;
+ dev->flags = IFF_NOARP;
+ dev->priv_flags &= ~IFF_XMIT_DST_RELEASE;
+ dev->addr_len = 4;
if (iph->daddr) {
#ifdef CONFIG_NET_IPGRE_BROADCAST
@@ -1421,106 +688,30 @@ static int ipgre_tunnel_init(struct net_device *dev)
} else
dev->header_ops = &ipgre_header_ops;
- dev->tstats = alloc_percpu(struct pcpu_tstats);
- if (!dev->tstats)
- return -ENOMEM;
-
- err = gro_cells_init(&tunnel->gro_cells, dev);
- if (err) {
- free_percpu(dev->tstats);
- return err;
- }
-
- return 0;
-}
-
-static void ipgre_fb_tunnel_init(struct net_device *dev)
-{
- struct ip_tunnel *tunnel = netdev_priv(dev);
- struct iphdr *iph = &tunnel->parms.iph;
-
- tunnel->dev = dev;
- strcpy(tunnel->parms.name, dev->name);
-
- iph->version = 4;
- iph->protocol = IPPROTO_GRE;
- iph->ihl = 5;
- tunnel->hlen = sizeof(struct iphdr) + 4;
-
- dev_hold(dev);
+ return ip_tunnel_init(dev);
}
-
static const struct gre_protocol ipgre_protocol = {
.handler = ipgre_rcv,
.err_handler = ipgre_err,
};
-static void ipgre_destroy_tunnels(struct ipgre_net *ign, struct list_head *head)
-{
- int prio;
-
- for (prio = 0; prio < 4; prio++) {
- int h;
- for (h = 0; h < HASH_SIZE; h++) {
- struct ip_tunnel *t;
-
- t = rtnl_dereference(ign->tunnels[prio][h]);
-
- while (t != NULL) {
- unregister_netdevice_queue(t->dev, head);
- t = rtnl_dereference(t->next);
- }
- }
- }
-}
-
static int __net_init ipgre_init_net(struct net *net)
{
- struct ipgre_net *ign = net_generic(net, ipgre_net_id);
- int err;
-
- ign->fb_tunnel_dev = alloc_netdev(sizeof(struct ip_tunnel), "gre0",
- ipgre_tunnel_setup);
- if (!ign->fb_tunnel_dev) {
- err = -ENOMEM;
- goto err_alloc_dev;
- }
- dev_net_set(ign->fb_tunnel_dev, net);
-
- ipgre_fb_tunnel_init(ign->fb_tunnel_dev);
- ign->fb_tunnel_dev->rtnl_link_ops = &ipgre_link_ops;
-
- if ((err = register_netdev(ign->fb_tunnel_dev)))
- goto err_reg_dev;
-
- rcu_assign_pointer(ign->tunnels_wc[0],
- netdev_priv(ign->fb_tunnel_dev));
- return 0;
-
-err_reg_dev:
- ipgre_dev_free(ign->fb_tunnel_dev);
-err_alloc_dev:
- return err;
+ return ip_tunnel_init_net(net, ipgre_net_id, &ipgre_link_ops, NULL);
}
static void __net_exit ipgre_exit_net(struct net *net)
{
- struct ipgre_net *ign;
- LIST_HEAD(list);
-
- ign = net_generic(net, ipgre_net_id);
- rtnl_lock();
- ipgre_destroy_tunnels(ign, &list);
- unregister_netdevice_many(&list);
- rtnl_unlock();
+ struct ip_tunnel_net *itn = net_generic(net, ipgre_net_id);
+ ip_tunnel_delete_net(itn);
}
static struct pernet_operations ipgre_net_ops = {
.init = ipgre_init_net,
.exit = ipgre_exit_net,
.id = &ipgre_net_id,
- .size = sizeof(struct ipgre_net),
+ .size = sizeof(struct ip_tunnel_net),
};
static int ipgre_tunnel_validate(struct nlattr *tb[], struct nlattr *data[])
@@ -1565,8 +756,8 @@ out:
return ipgre_tunnel_validate(tb, data);
}
-static void ipgre_netlink_parms(struct nlattr *data[],
- struct ip_tunnel_parm *parms)
+static void ipgre_netlink_parms(struct nlattr *data[], struct nlattr *tb[],
+ struct ip_tunnel_parm *parms)
{
memset(parms, 0, sizeof(*parms));
@@ -1579,10 +770,10 @@ static void ipgre_netlink_parms(struct nlattr *data[],
parms->link = nla_get_u32(data[IFLA_GRE_LINK]);
if (data[IFLA_GRE_IFLAGS])
- parms->i_flags = nla_get_be16(data[IFLA_GRE_IFLAGS]);
+ parms->i_flags = gre_flags_to_tnl_flags(nla_get_be16(data[IFLA_GRE_IFLAGS]));
if (data[IFLA_GRE_OFLAGS])
- parms->o_flags = nla_get_be16(data[IFLA_GRE_OFLAGS]);
+ parms->o_flags = gre_flags_to_tnl_flags(nla_get_be16(data[IFLA_GRE_OFLAGS]));
if (data[IFLA_GRE_IKEY])
parms->i_key = nla_get_be32(data[IFLA_GRE_IKEY]);
@@ -1608,146 +799,44 @@ static void ipgre_netlink_parms(struct nlattr *data[],
static int ipgre_tap_init(struct net_device *dev)
{
- struct ip_tunnel *tunnel;
-
- tunnel = netdev_priv(dev);
-
- tunnel->dev = dev;
- strcpy(tunnel->parms.name, dev->name);
-
- ipgre_tunnel_bind_dev(dev);
+ __gre_tunnel_init(dev);
- dev->tstats = alloc_percpu(struct pcpu_tstats);
- if (!dev->tstats)
- return -ENOMEM;
-
- return 0;
+ return ip_tunnel_init(dev);
}
static const struct net_device_ops ipgre_tap_netdev_ops = {
.ndo_init = ipgre_tap_init,
- .ndo_uninit = ipgre_tunnel_uninit,
- .ndo_start_xmit = ipgre_tunnel_xmit,
+ .ndo_uninit = ip_tunnel_uninit,
+ .ndo_start_xmit = ipgre_tap_xmit,
.ndo_set_mac_address = eth_mac_addr,
.ndo_validate_addr = eth_validate_addr,
- .ndo_change_mtu = ipgre_tunnel_change_mtu,
- .ndo_get_stats64 = ipgre_get_stats64,
+ .ndo_change_mtu = ip_tunnel_change_mtu,
+ .ndo_get_stats64 = ip_tunnel_get_stats64,
};
static void ipgre_tap_setup(struct net_device *dev)
{
-
ether_setup(dev);
-
dev->netdev_ops = &ipgre_tap_netdev_ops;
- dev->destructor = ipgre_dev_free;
-
- dev->iflink = 0;
- dev->features |= NETIF_F_NETNS_LOCAL;
-
- dev->features |= GRE_FEATURES;
- dev->hw_features |= GRE_FEATURES;
+ ip_tunnel_setup(dev, ipgre_tap_net_id);
}
-static int ipgre_newlink(struct net *src_net, struct net_device *dev, struct nlattr *tb[],
- struct nlattr *data[])
+static int ipgre_newlink(struct net *src_net, struct net_device *dev,
+ struct nlattr *tb[], struct nlattr *data[])
{
- struct ip_tunnel *nt;
- struct net *net = dev_net(dev);
- struct ipgre_net *ign = net_generic(net, ipgre_net_id);
- int mtu;
- int err;
-
- nt = netdev_priv(dev);
- ipgre_netlink_parms(data, &nt->parms);
-
- if (ipgre_tunnel_find(net, &nt->parms, dev->type))
- return -EEXIST;
-
- if (dev->type == ARPHRD_ETHER && !tb[IFLA_ADDRESS])
- eth_hw_addr_random(dev);
-
- mtu = ipgre_tunnel_bind_dev(dev);
- if (!tb[IFLA_MTU])
- dev->mtu = mtu;
-
- /* Can use a lockless transmit, unless we generate output sequences */
- if (!(nt->parms.o_flags & GRE_SEQ))
- dev->features |= NETIF_F_LLTX;
-
- err = register_netdevice(dev);
- if (err)
- goto out;
-
- dev_hold(dev);
- ipgre_tunnel_link(ign, nt);
+ struct ip_tunnel_parm p;
-out:
- return err;
+ ipgre_netlink_parms(data, tb, &p);
+ return ip_tunnel_newlink(dev, tb, &p);
}
static int ipgre_changelink(struct net_device *dev, struct nlattr *tb[],
struct nlattr *data[])
{
- struct ip_tunnel *t, *nt;
- struct net *net = dev_net(dev);
- struct ipgre_net *ign = net_generic(net, ipgre_net_id);
struct ip_tunnel_parm p;
- int mtu;
-
- if (dev == ign->fb_tunnel_dev)
- return -EINVAL;
-
- nt = netdev_priv(dev);
- ipgre_netlink_parms(data, &p);
-
- t = ipgre_tunnel_locate(net, &p, 0);
-
- if (t) {
- if (t->dev != dev)
- return -EEXIST;
- } else {
- t = nt;
-
- if (dev->type != ARPHRD_ETHER) {
- unsigned int nflags = 0;
-
- if (ipv4_is_multicast(p.iph.daddr))
- nflags = IFF_BROADCAST;
- else if (p.iph.daddr)
- nflags = IFF_POINTOPOINT;
-
- if ((dev->flags ^ nflags) &
- (IFF_POINTOPOINT | IFF_BROADCAST))
- return -EINVAL;
- }
- ipgre_tunnel_unlink(ign, t);
- t->parms.iph.saddr = p.iph.saddr;
- t->parms.iph.daddr = p.iph.daddr;
- t->parms.i_key = p.i_key;
- if (dev->type != ARPHRD_ETHER) {
- memcpy(dev->dev_addr, &p.iph.saddr, 4);
- memcpy(dev->broadcast, &p.iph.daddr, 4);
- }
- ipgre_tunnel_link(ign, t);
- netdev_state_change(dev);
- }
-
- t->parms.o_key = p.o_key;
- t->parms.iph.ttl = p.iph.ttl;
- t->parms.iph.tos = p.iph.tos;
- t->parms.iph.frag_off = p.iph.frag_off;
-
- if (t->parms.link != p.link) {
- t->parms.link = p.link;
- mtu = ipgre_tunnel_bind_dev(dev);
- if (!tb[IFLA_MTU])
- dev->mtu = mtu;
- netdev_state_change(dev);
- }
-
- return 0;
+ ipgre_netlink_parms(data, tb, &p);
+ return ip_tunnel_changelink(dev, tb, &p);
}
static size_t ipgre_get_size(const struct net_device *dev)
@@ -1782,8 +871,8 @@ static int ipgre_fill_info(struct sk_buff *skb, const struct net_device *dev)
struct ip_tunnel_parm *p = &t->parms;
if (nla_put_u32(skb, IFLA_GRE_LINK, p->link) ||
- nla_put_be16(skb, IFLA_GRE_IFLAGS, p->i_flags) ||
- nla_put_be16(skb, IFLA_GRE_OFLAGS, p->o_flags) ||
+ nla_put_be16(skb, IFLA_GRE_IFLAGS, tnl_flags_to_gre_flags(p->i_flags)) ||
+ nla_put_be16(skb, IFLA_GRE_OFLAGS, tnl_flags_to_gre_flags(p->o_flags)) ||
nla_put_be32(skb, IFLA_GRE_IKEY, p->i_key) ||
nla_put_be32(skb, IFLA_GRE_OKEY, p->o_key) ||
nla_put_be32(skb, IFLA_GRE_LOCAL, p->iph.saddr) ||
@@ -1821,6 +910,7 @@ static struct rtnl_link_ops ipgre_link_ops __read_mostly = {
.validate = ipgre_tunnel_validate,
.newlink = ipgre_newlink,
.changelink = ipgre_changelink,
+ .dellink = ip_tunnel_dellink,
.get_size = ipgre_get_size,
.fill_info = ipgre_fill_info,
};
@@ -1834,13 +924,28 @@ static struct rtnl_link_ops ipgre_tap_ops __read_mostly = {
.validate = ipgre_tap_validate,
.newlink = ipgre_newlink,
.changelink = ipgre_changelink,
+ .dellink = ip_tunnel_dellink,
.get_size = ipgre_get_size,
.fill_info = ipgre_fill_info,
};
-/*
- * And now the modules code and kernel interface.
- */
+static int __net_init ipgre_tap_init_net(struct net *net)
+{
+ return ip_tunnel_init_net(net, ipgre_tap_net_id, &ipgre_tap_ops, NULL);
+}
+
+static void __net_exit ipgre_tap_exit_net(struct net *net)
+{
+ struct ip_tunnel_net *itn = net_generic(net, ipgre_tap_net_id);
+ ip_tunnel_delete_net(itn);
+}
+
+static struct pernet_operations ipgre_tap_net_ops = {
+ .init = ipgre_tap_init_net,
+ .exit = ipgre_tap_exit_net,
+ .id = &ipgre_tap_net_id,
+ .size = sizeof(struct ip_tunnel_net),
+};
static int __init ipgre_init(void)
{
@@ -1852,6 +957,10 @@ static int __init ipgre_init(void)
if (err < 0)
return err;
+ err = register_pernet_device(&ipgre_tap_net_ops);
+ if (err < 0)
+ goto pnet_tap_faied;
+
rtnl_lock();
err = gre_add_protocol(&ipgre_protocol, GREPROTO_CISCO);
rtnl_unlock();
@@ -1868,18 +977,17 @@ static int __init ipgre_init(void)
if (err < 0)
goto tap_ops_failed;
-out:
- return err;
+ return 0;
tap_ops_failed:
rtnl_link_unregister(&ipgre_link_ops);
rtnl_link_failed:
- rtnl_lock();
gre_del_protocol(&ipgre_protocol, GREPROTO_CISCO);
- rtnl_unlock();
add_proto_failed:
+ unregister_pernet_device(&ipgre_tap_net_ops);
+pnet_tap_faied:
unregister_pernet_device(&ipgre_net_ops);
- goto out;
+ return err;
}
static void __exit ipgre_fini(void)
@@ -1890,6 +998,7 @@ static void __exit ipgre_fini(void)
if (gre_del_protocol(&ipgre_protocol, GREPROTO_CISCO) < 0)
pr_info("%s: can't remove protocol\n", __func__);
rtnl_unlock();
+ unregister_pernet_device(&ipgre_tap_net_ops);
unregister_pernet_device(&ipgre_net_ops);
}
@@ -1899,3 +1008,4 @@ MODULE_LICENSE("GPL");
MODULE_ALIAS_RTNL_LINK("gre");
MODULE_ALIAS_RTNL_LINK("gretap");
MODULE_ALIAS_NETDEV("gre0");
+MODULE_ALIAS_NETDEV("gretap0");
diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c
new file mode 100644
index 0000000..88b4432
--- /dev/null
+++ b/net/ipv4/ip_tunnel.c
@@ -0,0 +1,1036 @@
+/*
+ * Copyright (c) 2013 Nicira, Inc.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/capability.h>
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/uaccess.h>
+#include <linux/skbuff.h>
+#include <linux/netdevice.h>
+#include <linux/in.h>
+#include <linux/tcp.h>
+#include <linux/udp.h>
+#include <linux/if_arp.h>
+#include <linux/mroute.h>
+#include <linux/init.h>
+#include <linux/in6.h>
+#include <linux/inetdevice.h>
+#include <linux/igmp.h>
+#include <linux/netfilter_ipv4.h>
+#include <linux/etherdevice.h>
+#include <linux/if_ether.h>
+#include <linux/if_vlan.h>
+#include <linux/rculist.h>
+
+#include <net/sock.h>
+#include <net/ip.h>
+#include <net/icmp.h>
+#include <net/protocol.h>
+#include <net/ip_tunnels.h>
+#include <net/arp.h>
+#include <net/checksum.h>
+#include <net/dsfield.h>
+#include <net/inet_ecn.h>
+#include <net/xfrm.h>
+#include <net/net_namespace.h>
+#include <net/netns/generic.h>
+#include <net/rtnetlink.h>
+
+#if IS_ENABLED(CONFIG_IPV6)
+#include <net/ipv6.h>
+#include <net/ip6_fib.h>
+#include <net/ip6_route.h>
+#endif
+
+static unsigned int ip_tunnel_hash(struct ip_tunnel_net *itn,
+ __be32 key, __be32 remote)
+{
+ return hash_32((__force u32)key ^ (__force u32)remote,
+ IP_TNL_HASH_BITS);
+}
+
+/* Often modified stats are per cpu, other are shared (netdev->stats) */
+struct rtnl_link_stats64 *ip_tunnel_get_stats64(struct net_device *dev,
+ struct rtnl_link_stats64 *tot)
+{
+ int i;
+
+ for_each_possible_cpu(i) {
+ const struct pcpu_tstats *tstats = per_cpu_ptr(dev->tstats, i);
+ u64 rx_packets, rx_bytes, tx_packets, tx_bytes;
+ unsigned int start;
+
+ do {
+ start = u64_stats_fetch_begin_bh(&tstats->syncp);
+ rx_packets = tstats->rx_packets;
+ tx_packets = tstats->tx_packets;
+ rx_bytes = tstats->rx_bytes;
+ tx_bytes = tstats->tx_bytes;
+ } while (u64_stats_fetch_retry_bh(&tstats->syncp, start));
+
+ tot->rx_packets += rx_packets;
+ tot->tx_packets += tx_packets;
+ tot->rx_bytes += rx_bytes;
+ tot->tx_bytes += tx_bytes;
+ }
+
+ tot->multicast = dev->stats.multicast;
+
+ tot->rx_crc_errors = dev->stats.rx_crc_errors;
+ tot->rx_fifo_errors = dev->stats.rx_fifo_errors;
+ tot->rx_length_errors = dev->stats.rx_length_errors;
+ tot->rx_frame_errors = dev->stats.rx_frame_errors;
+ tot->rx_errors = dev->stats.rx_errors;
+
+ tot->tx_fifo_errors = dev->stats.tx_fifo_errors;
+ tot->tx_carrier_errors = dev->stats.tx_carrier_errors;
+ tot->tx_dropped = dev->stats.tx_dropped;
+ tot->tx_aborted_errors = dev->stats.tx_aborted_errors;
+ tot->tx_errors = dev->stats.tx_errors;
+
+ tot->collisions = dev->stats.collisions;
+
+ return tot;
+}
+EXPORT_SYMBOL(ip_tunnel_get_stats64);
+
+static bool ip_tunnel_key_match(const struct ip_tunnel_parm *p,
+ __be16 flags, __be32 key)
+{
+ if (p->i_flags & TUNNEL_KEY) {
+ if (flags & TUNNEL_KEY)
+ return key == p->i_key;
+ else
+ /* key expected, none present */
+ return false;
+ } else
+ return !(flags & TUNNEL_KEY);
+}
+
+/* Fallback tunnel: no source, no destination, no key, no options
+
+ Tunnel hash table:
+ We require exact key match i.e. if a key is present in packet
+ it will match only tunnel with the same key; if it is not present,
+ it will match only keyless tunnel.
+
+ All keysless packets, if not matched configured keyless tunnels
+ will match fallback tunnel.
+ Given src, dst and key, find appropriate for input tunnel.
+*/
+struct ip_tunnel *ip_tunnel_lookup(struct ip_tunnel_net *itn,
+ int link, __be16 flags,
+ __be32 remote, __be32 local,
+ __be32 key)
+{
+ unsigned int hash;
+ struct ip_tunnel *t, *cand = NULL;
+ struct hlist_head *head;
+
+ hash = ip_tunnel_hash(itn, key, remote);
+ head = &itn->tunnels[hash];
+
+ hlist_for_each_entry_rcu(t, head, hash_node) {
+ if (local != t->parms.iph.saddr ||
+ remote != t->parms.iph.daddr ||
+ !(t->dev->flags & IFF_UP))
+ continue;
+
+ if (!ip_tunnel_key_match(&t->parms, flags, key))
+ continue;
+
+ if (t->parms.link == link)
+ return t;
+ else
+ cand = t;
+ }
+
+ hlist_for_each_entry_rcu(t, head, hash_node) {
+ if (remote != t->parms.iph.daddr ||
+ !(t->dev->flags & IFF_UP))
+ continue;
+
+ if (!ip_tunnel_key_match(&t->parms, flags, key))
+ continue;
+
+ if (t->parms.link == link)
+ return t;
+ else if (!cand)
+ cand = t;
+ }
+
+ hash = ip_tunnel_hash(itn, key, 0);
+ head = &itn->tunnels[hash];
+
+ hlist_for_each_entry_rcu(t, head, hash_node) {
+ if ((local != t->parms.iph.saddr &&
+ (local != t->parms.iph.daddr ||
+ !ipv4_is_multicast(local))) ||
+ !(t->dev->flags & IFF_UP))
+ continue;
+
+ if (!ip_tunnel_key_match(&t->parms, flags, key))
+ continue;
+
+ if (t->parms.link == link)
+ return t;
+ else if (!cand)
+ cand = t;
+ }
+
+ if (flags & TUNNEL_NO_KEY)
+ goto skip_key_lookup;
+
+ hlist_for_each_entry_rcu(t, head, hash_node) {
+ if (t->parms.i_key != key ||
+ !(t->dev->flags & IFF_UP))
+ continue;
+
+ if (t->parms.link == link)
+ return t;
+ else if (!cand)
+ cand = t;
+ }
+
+skip_key_lookup:
+ if (cand)
+ return cand;
+
+ if (itn->fb_tunnel_dev && itn->fb_tunnel_dev->flags & IFF_UP)
+ return netdev_priv(itn->fb_tunnel_dev);
+
+
+ return NULL;
+}
+EXPORT_SYMBOL(ip_tunnel_lookup);
+
+static struct hlist_head *ip_bucket(struct ip_tunnel_net *itn,
+ struct ip_tunnel_parm *parms)
+{
+ unsigned int h;
+ __be32 remote;
+
+ if (parms->iph.daddr && !ipv4_is_multicast(parms->iph.daddr))
+ remote = parms->iph.daddr;
+ else
+ remote = 0;
+
+ h = ip_tunnel_hash(itn, parms->i_key, remote);
+ return &itn->tunnels[h];
+}
+
+static void ip_tunnel_add(struct ip_tunnel_net *itn, struct ip_tunnel *t)
+{
+ struct hlist_head *head = ip_bucket(itn, &t->parms);
+
+ hlist_add_head_rcu(&t->hash_node, head);
+}
+
+static void ip_tunnel_del(struct ip_tunnel_net *itn, struct ip_tunnel *t)
+{
+ hlist_del_init_rcu(&t->hash_node);
+}
+
+static struct ip_tunnel *ip_tunnel_find(struct ip_tunnel_net *itn,
+ struct ip_tunnel_parm *parms,
+ int type)
+{
+ __be32 remote = parms->iph.daddr;
+ __be32 local = parms->iph.saddr;
+ __be32 key = parms->i_key;
+ int link = parms->link;
+ struct ip_tunnel *t = NULL;
+ struct hlist_head *head = ip_bucket(itn, parms);
+
+ hlist_for_each_entry_rcu(t, head, hash_node) {
+ if (local == t->parms.iph.saddr &&
+ remote == t->parms.iph.daddr &&
+ key == t->parms.i_key &&
+ link == t->parms.link &&
+ type == t->dev->type)
+ break;
+ }
+ return t;
+}
+
+static struct net_device *__ip_tunnel_create(struct net *net,
+ const struct rtnl_link_ops *ops,
+ struct ip_tunnel_parm *parms)
+{
+ int err;
+ struct ip_tunnel *tunnel;
+ struct net_device *dev;
+ char name[IFNAMSIZ];
+
+ if (parms->name[0])
+ strlcpy(name, parms->name, IFNAMSIZ);
+ else {
+ if (strlen(ops->kind) + 3 >= IFNAMSIZ) {
+ err = -E2BIG;
+ goto failed;
+ }
+ strlcpy(name, ops->kind, IFNAMSIZ);
+ strncat(name, "%d", 2);
+ }
+
+ ASSERT_RTNL();
+ dev = alloc_netdev(ops->priv_size, name, ops->setup);
+ if (!dev) {
+ err = -ENOMEM;
+ goto failed;
+ }
+ dev_net_set(dev, net);
+
+ dev->rtnl_link_ops = ops;
+
+ tunnel = netdev_priv(dev);
+ tunnel->parms = *parms;
+
+ err = register_netdevice(dev);
+ if (err)
+ goto failed_free;
+
+ return dev;
+
+failed_free:
+ free_netdev(dev);
+failed:
+ return ERR_PTR(err);
+}
+
+static inline struct rtable *ip_route_output_tunnel(struct net *net,
+ struct flowi4 *fl4,
+ int proto,
+ __be32 daddr, __be32 saddr,
+ __be32 key, __u8 tos, int oif)
+{
+ memset(fl4, 0, sizeof(*fl4));
+ fl4->flowi4_oif = oif;
+ fl4->daddr = daddr;
+ fl4->saddr = saddr;
+ fl4->flowi4_tos = tos;
+ fl4->flowi4_proto = proto;
+ fl4->fl4_gre_key = key;
+ return ip_route_output_key(net, fl4);
+}
+
+static int ip_tunnel_bind_dev(struct net_device *dev)
+{
+ struct net_device *tdev = NULL;
+ struct ip_tunnel *tunnel = netdev_priv(dev);
+ const struct iphdr *iph;
+ int hlen = LL_MAX_HEADER;
+ int mtu = ETH_DATA_LEN;
+ int t_hlen = tunnel->hlen + sizeof(struct iphdr);
+
+ iph = &tunnel->parms.iph;
+
+ /* Guess output device to choose reasonable mtu and needed_headroom */
+ if (iph->daddr) {
+ struct flowi4 fl4;
+ struct rtable *rt;
+
+ rt = ip_route_output_tunnel(dev_net(dev), &fl4,
+ tunnel->parms.iph.protocol,
+ iph->daddr, iph->saddr,
+ tunnel->parms.o_key,
+ RT_TOS(iph->tos),
+ tunnel->parms.link);
+ if (!IS_ERR(rt)) {
+ tdev = rt->dst.dev;
+ ip_rt_put(rt);
+ }
+ if (dev->type != ARPHRD_ETHER)
+ dev->flags |= IFF_POINTOPOINT;
+ }
+
+ if (!tdev && tunnel->parms.link)
+ tdev = __dev_get_by_index(dev_net(dev), tunnel->parms.link);
+
+ if (tdev) {
+ hlen = tdev->hard_header_len + tdev->needed_headroom;
+ mtu = tdev->mtu;
+ }
+ dev->iflink = tunnel->parms.link;
+
+ dev->needed_headroom = t_hlen + hlen;
+ mtu -= (dev->hard_header_len + t_hlen);
+
+ if (mtu < 68)
+ mtu = 68;
+
+ return mtu;
+}
+
+static struct ip_tunnel *ip_tunnel_create(struct net *net,
+ struct ip_tunnel_net *itn,
+ struct ip_tunnel_parm *parms)
+{
+ struct ip_tunnel *nt, *fbt;
+ struct net_device *dev;
+
+ BUG_ON(!itn->fb_tunnel_dev);
+ fbt = netdev_priv(itn->fb_tunnel_dev);
+ dev = __ip_tunnel_create(net, itn->fb_tunnel_dev->rtnl_link_ops, parms);
+ if (IS_ERR(dev))
+ return NULL;
+
+ dev->mtu = ip_tunnel_bind_dev(dev);
+
+ nt = netdev_priv(dev);
+ ip_tunnel_add(itn, nt);
+ return nt;
+}
+
+int ip_tunnel_rcv(struct ip_tunnel *tunnel, struct sk_buff *skb,
+ const struct tnl_ptk_info *tpi, bool log_ecn_error)
+{
+ struct pcpu_tstats *tstats;
+ const struct iphdr *iph = ip_hdr(skb);
+ int err;
+
+ secpath_reset(skb);
+
+ skb->protocol = tpi->proto;
+
+ skb->mac_header = skb->network_header;
+ __pskb_pull(skb, tunnel->hlen);
+ skb_postpull_rcsum(skb, skb_transport_header(skb), tunnel->hlen);
+#ifdef CONFIG_NET_IPGRE_BROADCAST
+ if (ipv4_is_multicast(iph->daddr)) {
+ /* Looped back packet, drop it! */
+ if (rt_is_output_route(skb_rtable(skb)))
+ goto drop;
+ tunnel->dev->stats.multicast++;
+ skb->pkt_type = PACKET_BROADCAST;
+ }
+#endif
+
+ if ((!(tpi->flags&TUNNEL_CSUM) && (tunnel->parms.i_flags&TUNNEL_CSUM)) ||
+ ((tpi->flags&TUNNEL_CSUM) && !(tunnel->parms.i_flags&TUNNEL_CSUM))) {
+ tunnel->dev->stats.rx_crc_errors++;
+ tunnel->dev->stats.rx_errors++;
+ goto drop;
+ }
+
+ if (tunnel->parms.i_flags&TUNNEL_SEQ) {
+ if (!(tpi->flags&TUNNEL_SEQ) ||
+ (tunnel->i_seqno && (s32)(ntohl(tpi->seq) - tunnel->i_seqno) < 0)) {
+ tunnel->dev->stats.rx_fifo_errors++;
+ tunnel->dev->stats.rx_errors++;
+ goto drop;
+ }
+ tunnel->i_seqno = ntohl(tpi->seq) + 1;
+ }
+
+ /* Warning: All skb pointers will be invalidated! */
+ if (tunnel->dev->type == ARPHRD_ETHER) {
+ if (!pskb_may_pull(skb, ETH_HLEN)) {
+ tunnel->dev->stats.rx_length_errors++;
+ tunnel->dev->stats.rx_errors++;
+ goto drop;
+ }
+
+ iph = ip_hdr(skb);
+ skb->protocol = eth_type_trans(skb, tunnel->dev);
+ skb_postpull_rcsum(skb, eth_hdr(skb), ETH_HLEN);
+ }
+
+ skb->pkt_type = PACKET_HOST;
+ __skb_tunnel_rx(skb, tunnel->dev);
+
+ skb_reset_network_header(skb);
+ err = IP_ECN_decapsulate(iph, skb);
+ if (unlikely(err)) {
+ if (log_ecn_error)
+ net_info_ratelimited("non-ECT from %pI4 with TOS=%#x\n",
+ &iph->saddr, iph->tos);
+ if (err > 1) {
+ ++tunnel->dev->stats.rx_frame_errors;
+ ++tunnel->dev->stats.rx_errors;
+ goto drop;
+ }
+ }
+
+ tstats = this_cpu_ptr(tunnel->dev->tstats);
+ u64_stats_update_begin(&tstats->syncp);
+ tstats->rx_packets++;
+ tstats->rx_bytes += skb->len;
+ u64_stats_update_end(&tstats->syncp);
+
+ gro_cells_receive(&tunnel->gro_cells, skb);
+
+ return 0;
+
+drop:
+ kfree_skb(skb);
+ return 0;
+}
+EXPORT_SYMBOL(ip_tunnel_rcv);
+
+int ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev,
+ const struct iphdr *tnl_params)
+{
+ struct ip_tunnel *tunnel = netdev_priv(dev);
+ const struct iphdr *inner_iph;
+ struct iphdr *iph;
+ struct flowi4 fl4;
+ u8 tos, ttl;
+ __be16 df;
+ struct rtable *rt; /* Route to the other host */
+ struct net_device *tdev; /* Device to other host */
+ unsigned int max_headroom; /* The extra header space needed */
+ __be32 dst;
+ int mtu;
+
+ inner_iph = (const struct iphdr *)skb_inner_network_header(skb);
+
+ dst = tnl_params->daddr;
+ if (dst == 0) {
+ /* NBMA tunnel */
+
+ if (skb_dst(skb) == NULL) {
+ dev->stats.tx_fifo_errors++;
+ goto tx_error;
+ }
+
+ if (skb->protocol == htons(ETH_P_IP)) {
+ rt = skb_rtable(skb);
+ dst = rt_nexthop(rt, inner_iph->daddr);
+ }
+#if IS_ENABLED(CONFIG_IPV6)
+ else if (skb->protocol == htons(ETH_P_IPV6)) {
+ const struct in6_addr *addr6;
+ struct neighbour *neigh;
+ bool do_tx_error_icmp;
+ int addr_type;
+
+ neigh = dst_neigh_lookup(skb_dst(skb),
+ &ipv6_hdr(skb)->daddr);
+ if (neigh == NULL)
+ goto tx_error;
+
+ addr6 = (const struct in6_addr *)&neigh->primary_key;
+ addr_type = ipv6_addr_type(addr6);
+
+ if (addr_type == IPV6_ADDR_ANY) {
+ addr6 = &ipv6_hdr(skb)->daddr;
+ addr_type = ipv6_addr_type(addr6);
+ }
+
+ if ((addr_type & IPV6_ADDR_COMPATv4) == 0)
+ do_tx_error_icmp = true;
+ else {
+ do_tx_error_icmp = false;
+ dst = addr6->s6_addr32[3];
+ }
+ neigh_release(neigh);
+ if (do_tx_error_icmp)
+ goto tx_error_icmp;
+ }
+#endif
+ else
+ goto tx_error;
+ }
+
+ tos = tnl_params->tos;
+ if (tos & 0x1) {
+ tos &= ~0x1;
+ if (skb->protocol == htons(ETH_P_IP))
+ tos = inner_iph->tos;
+ else if (skb->protocol == htons(ETH_P_IPV6))
+ tos = ipv6_get_dsfield((const struct ipv6hdr *)inner_iph);
+ }
+
+ rt = ip_route_output_tunnel(dev_net(dev), &fl4,
+ tunnel->parms.iph.protocol,
+ dst, tnl_params->saddr,
+ tunnel->parms.o_key,
+ RT_TOS(tos),
+ tunnel->parms.link);
+ if (IS_ERR(rt)) {
+ dev->stats.tx_carrier_errors++;
+ goto tx_error;
+ }
+ tdev = rt->dst.dev;
+
+ if (tdev == dev) {
+ ip_rt_put(rt);
+ dev->stats.collisions++;
+ goto tx_error;
+ }
+
+ df = tnl_params->frag_off;
+
+ if (df)
+ mtu = dst_mtu(&rt->dst) - dev->hard_header_len
+ - sizeof(struct iphdr);
+ else
+ mtu = skb_dst(skb) ? dst_mtu(skb_dst(skb)) : dev->mtu;
+
+ if (skb_dst(skb))
+ skb_dst(skb)->ops->update_pmtu(skb_dst(skb), NULL, skb, mtu);
+
+ if (skb->protocol == htons(ETH_P_IP)) {
+ df |= (inner_iph->frag_off&htons(IP_DF));
+
+ if (!skb_is_gso(skb) &&
+ (inner_iph->frag_off&htons(IP_DF)) &&
+ mtu < ntohs(inner_iph->tot_len)) {
+ icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED, htonl(mtu));
+ ip_rt_put(rt);
+ goto tx_error;
+ }
+ }
+#if IS_ENABLED(CONFIG_IPV6)
+ else if (skb->protocol == htons(ETH_P_IPV6)) {
+ struct rt6_info *rt6 = (struct rt6_info *)skb_dst(skb);
+
+ if (rt6 && mtu < dst_mtu(skb_dst(skb)) &&
+ mtu >= IPV6_MIN_MTU) {
+ if ((tunnel->parms.iph.daddr &&
+ !ipv4_is_multicast(tunnel->parms.iph.daddr)) ||
+ rt6->rt6i_dst.plen == 128) {
+ rt6->rt6i_flags |= RTF_MODIFIED;
+ dst_metric_set(skb_dst(skb), RTAX_MTU, mtu);
+ }
+ }
+
+ if (!skb_is_gso(skb) && mtu >= IPV6_MIN_MTU &&
+ mtu < skb->len) {
+ icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu);
+ ip_rt_put(rt);
+ goto tx_error;
+ }
+ }
+#endif
+
+ if (tunnel->err_count > 0) {
+ if (time_before(jiffies,
+ tunnel->err_time + IPTUNNEL_ERR_TIMEO)) {
+ tunnel->err_count--;
+
+ dst_link_failure(skb);
+ } else
+ tunnel->err_count = 0;
+ }
+
+ ttl = tnl_params->ttl;
+ if (ttl == 0) {
+ if (skb->protocol == htons(ETH_P_IP))
+ ttl = inner_iph->ttl;
+#if IS_ENABLED(CONFIG_IPV6)
+ else if (skb->protocol == htons(ETH_P_IPV6))
+ ttl = ((const struct ipv6hdr *)inner_iph)->hop_limit;
+#endif
+ else
+ ttl = ip4_dst_hoplimit(&rt->dst);
+ }
+
+ max_headroom = LL_RESERVED_SPACE(tdev) + sizeof(struct iphdr)
+ + rt->dst.header_len;
+ if (max_headroom > dev->needed_headroom) {
+ dev->needed_headroom = max_headroom;
+ if (skb_cow_head(skb, dev->needed_headroom)) {
+ dev->stats.tx_dropped++;
+ dev_kfree_skb(skb);
+ return NETDEV_TX_OK;
+ }
+ }
+
+ skb_dst_drop(skb);
+ skb_dst_set(skb, &rt->dst);
+ memset(IPCB(skb), 0, sizeof(*IPCB(skb)));
+
+ /* Push down and install the IP header. */
+ skb_push(skb, sizeof(struct iphdr));
+ skb_reset_network_header(skb);
+
+ iph = ip_hdr(skb);
+ inner_iph = (const struct iphdr *)skb_inner_network_header(skb);
+
+ iph->version = 4;
+ iph->ihl = sizeof(struct iphdr) >> 2;
+ iph->frag_off = df;
+ iph->protocol = tnl_params->protocol;
+ iph->tos = ip_tunnel_ecn_encap(tos, inner_iph, skb);
+ iph->daddr = fl4.daddr;
+ iph->saddr = fl4.saddr;
+ iph->ttl = ttl;
+ tunnel_ip_select_ident(skb, inner_iph, &rt->dst);
+
+ iptunnel_xmit(skb, dev);
+ return 0;
+
+#if IS_ENABLED(CONFIG_IPV6)
+tx_error_icmp:
+ dst_link_failure(skb);
+#endif
+tx_error:
+ dev->stats.tx_errors++;
+ dev_kfree_skb(skb);
+ return 0;
+}
+EXPORT_SYMBOL(ip_tunnel_xmit);
+
+static void ip_tunnel_update(struct ip_tunnel_net *itn,
+ struct ip_tunnel *t,
+ struct net_device *dev,
+ struct ip_tunnel_parm *p,
+ bool set_mtu)
+{
+ ip_tunnel_del(itn, t);
+ t->parms.iph.saddr = p->iph.saddr;
+ t->parms.iph.daddr = p->iph.daddr;
+ t->parms.i_key = p->i_key;
+ t->parms.o_key = p->o_key;
+ if (dev->type != ARPHRD_ETHER) {
+ memcpy(dev->dev_addr, &p->iph.saddr, 4);
+ memcpy(dev->broadcast, &p->iph.daddr, 4);
+ }
+ ip_tunnel_add(itn, t);
+
+ t->parms.iph.ttl = p->iph.ttl;
+ t->parms.iph.tos = p->iph.tos;
+ t->parms.iph.frag_off = p->iph.frag_off;
+
+ if (t->parms.link != p->link) {
+ int mtu;
+
+ t->parms.link = p->link;
+ mtu = ip_tunnel_bind_dev(dev);
+ if (set_mtu)
+ dev->mtu = mtu;
+ }
+ netdev_state_change(dev);
+}
+
+int ip_tunnel_ioctl(struct net_device *dev, struct ip_tunnel_parm *p, int cmd)
+{
+ int err = 0;
+ struct ip_tunnel *t;
+ struct net *net = dev_net(dev);
+ struct ip_tunnel *tunnel = netdev_priv(dev);
+ struct ip_tunnel_net *itn = net_generic(net, tunnel->ip_tnl_net_id);
+
+ BUG_ON(!itn->fb_tunnel_dev);
+ switch (cmd) {
+ case SIOCGETTUNNEL:
+ t = NULL;
+ if (dev == itn->fb_tunnel_dev)
+ t = ip_tunnel_find(itn, p, itn->fb_tunnel_dev->type);
+ if (t == NULL)
+ t = netdev_priv(dev);
+ memcpy(p, &t->parms, sizeof(*p));
+ break;
+
+ case SIOCADDTUNNEL:
+ case SIOCCHGTUNNEL:
+ err = -EPERM;
+ if (!ns_capable(net->user_ns, CAP_NET_ADMIN))
+ goto done;
+ if (p->iph.ttl)
+ p->iph.frag_off |= htons(IP_DF);
+ if (!(p->i_flags&TUNNEL_KEY))
+ p->i_key = 0;
+ if (!(p->o_flags&TUNNEL_KEY))
+ p->o_key = 0;
+
+ t = ip_tunnel_find(itn, p, itn->fb_tunnel_dev->type);
+
+ if (!t && (cmd == SIOCADDTUNNEL))
+ t = ip_tunnel_create(net, itn, p);
+
+ if (dev != itn->fb_tunnel_dev && cmd == SIOCCHGTUNNEL) {
+ if (t != NULL) {
+ if (t->dev != dev) {
+ err = -EEXIST;
+ break;
+ }
+ } else {
+ unsigned int nflags = 0;
+
+ if (ipv4_is_multicast(p->iph.daddr))
+ nflags = IFF_BROADCAST;
+ else if (p->iph.daddr)
+ nflags = IFF_POINTOPOINT;
+
+ if ((dev->flags^nflags)&(IFF_POINTOPOINT|IFF_BROADCAST)) {
+ err = -EINVAL;
+ break;
+ }
+
+ t = netdev_priv(dev);
+ }
+ }
+
+ if (t) {
+ err = 0;
+ ip_tunnel_update(itn, t, dev, p, true);
+ } else
+ err = (cmd == SIOCADDTUNNEL ? -ENOBUFS : -ENOENT);
+ break;
+
+ case SIOCDELTUNNEL:
+ err = -EPERM;
+ if (!ns_capable(net->user_ns, CAP_NET_ADMIN))
+ goto done;
+
+ if (dev == itn->fb_tunnel_dev) {
+ err = -ENOENT;
+ t = ip_tunnel_find(itn, p, itn->fb_tunnel_dev->type);
+ if (t == NULL)
+ goto done;
+ err = -EPERM;
+ if (t == netdev_priv(itn->fb_tunnel_dev))
+ goto done;
+ dev = t->dev;
+ }
+ unregister_netdevice(dev);
+ err = 0;
+ break;
+
+ default:
+ err = -EINVAL;
+ }
+
+done:
+ return err;
+}
+EXPORT_SYMBOL(ip_tunnel_ioctl);
+
+int ip_tunnel_change_mtu(struct net_device *dev, int new_mtu)
+{
+ struct ip_tunnel *tunnel = netdev_priv(dev);
+ int t_hlen = tunnel->hlen + sizeof(struct iphdr);
+
+ if (new_mtu < 68 ||
+ new_mtu > 0xFFF8 - dev->hard_header_len - t_hlen)
+ return -EINVAL;
+ dev->mtu = new_mtu;
+ return 0;
+}
+EXPORT_SYMBOL(ip_tunnel_change_mtu);
+
+static void ip_tunnel_dev_free(struct net_device *dev)
+{
+ struct ip_tunnel *tunnel = netdev_priv(dev);
+
+ gro_cells_destroy(&tunnel->gro_cells);
+ free_percpu(dev->tstats);
+ free_netdev(dev);
+}
+
+void ip_tunnel_dellink(struct net_device *dev, struct list_head *head)
+{
+ struct net *net = dev_net(dev);
+ struct ip_tunnel *tunnel = netdev_priv(dev);
+ struct ip_tunnel_net *itn;
+
+ itn = net_generic(net, tunnel->ip_tnl_net_id);
+
+ if (itn->fb_tunnel_dev != dev) {
+ ip_tunnel_del(itn, netdev_priv(dev));
+ unregister_netdevice_queue(dev, head);
+ }
+}
+EXPORT_SYMBOL(ip_tunnel_dellink);
+
+int __net_init ip_tunnel_init_net(struct net *net, int ip_tnl_net_id,
+ struct rtnl_link_ops *ops, char *devname)
+{
+ struct ip_tunnel_net *itn = net_generic(net, ip_tnl_net_id);
+ struct ip_tunnel_parm parms;
+
+ itn->tunnels = kzalloc(IP_TNL_HASH_SIZE * sizeof(struct hlist_head), GFP_KERNEL);
+ if (!itn->tunnels)
+ return -ENOMEM;
+
+ if (!ops) {
+ itn->fb_tunnel_dev = NULL;
+ return 0;
+ }
+ memset(&parms, 0, sizeof(parms));
+ if (devname)
+ strlcpy(parms.name, devname, IFNAMSIZ);
+
+ rtnl_lock();
+ itn->fb_tunnel_dev = __ip_tunnel_create(net, ops, &parms);
+ rtnl_unlock();
+ if (IS_ERR(itn->fb_tunnel_dev)) {
+ kfree(itn->tunnels);
+ return PTR_ERR(itn->fb_tunnel_dev);
+ }
+
+ return 0;
+}
+EXPORT_SYMBOL(ip_tunnel_init_net);
+
+static void ip_tunnel_destroy(struct ip_tunnel_net *itn, struct list_head *head)
+{
+ int h;
+
+ for (h = 0; h < IP_TNL_HASH_SIZE; h++) {
+ struct ip_tunnel *t;
+ struct hlist_node *n;
+ struct hlist_head *thead = &itn->tunnels[h];
+
+ hlist_for_each_entry_safe(t, n, thead, hash_node)
+ unregister_netdevice_queue(t->dev, head);
+ }
+ if (itn->fb_tunnel_dev)
+ unregister_netdevice_queue(itn->fb_tunnel_dev, head);
+}
+
+void __net_exit ip_tunnel_delete_net(struct ip_tunnel_net *itn)
+{
+ LIST_HEAD(list);
+
+ rtnl_lock();
+ ip_tunnel_destroy(itn, &list);
+ unregister_netdevice_many(&list);
+ rtnl_unlock();
+ kfree(itn->tunnels);
+}
+EXPORT_SYMBOL(ip_tunnel_delete_net);
+
+int ip_tunnel_newlink(struct net_device *dev, struct nlattr *tb[],
+ struct ip_tunnel_parm *p)
+{
+ struct ip_tunnel *nt;
+ struct net *net = dev_net(dev);
+ struct ip_tunnel_net *itn;
+ int mtu;
+ int err;
+
+ nt = netdev_priv(dev);
+ itn = net_generic(net, nt->ip_tnl_net_id);
+
+ if (ip_tunnel_find(itn, p, dev->type))
+ return -EEXIST;
+
+ nt->parms = *p;
+ err = register_netdevice(dev);
+ if (err)
+ goto out;
+
+ if (dev->type == ARPHRD_ETHER && !tb[IFLA_ADDRESS])
+ eth_hw_addr_random(dev);
+
+ mtu = ip_tunnel_bind_dev(dev);
+ if (!tb[IFLA_MTU])
+ dev->mtu = mtu;
+
+ ip_tunnel_add(itn, nt);
+
+out:
+ return err;
+}
+EXPORT_SYMBOL(ip_tunnel_newlink);
+
+int ip_tunnel_changelink(struct net_device *dev, struct nlattr *tb[],
+ struct ip_tunnel_parm *p)
+{
+ struct ip_tunnel *t, *nt;
+ struct net *net = dev_net(dev);
+ struct ip_tunnel *tunnel = netdev_priv(dev);
+ struct ip_tunnel_net *itn = net_generic(net, tunnel->ip_tnl_net_id);
+
+ if (dev == itn->fb_tunnel_dev)
+ return -EINVAL;
+
+ nt = netdev_priv(dev);
+
+ t = ip_tunnel_find(itn, p, dev->type);
+
+ if (t) {
+ if (t->dev != dev)
+ return -EEXIST;
+ } else {
+ t = nt;
+
+ if (dev->type != ARPHRD_ETHER) {
+ unsigned int nflags = 0;
+
+ if (ipv4_is_multicast(p->iph.daddr))
+ nflags = IFF_BROADCAST;
+ else if (p->iph.daddr)
+ nflags = IFF_POINTOPOINT;
+
+ if ((dev->flags ^ nflags) &
+ (IFF_POINTOPOINT | IFF_BROADCAST))
+ return -EINVAL;
+ }
+ }
+
+ ip_tunnel_update(itn, t, dev, p, !tb[IFLA_MTU]);
+ return 0;
+}
+EXPORT_SYMBOL(ip_tunnel_changelink);
+
+int ip_tunnel_init(struct net_device *dev)
+{
+ struct ip_tunnel *tunnel = netdev_priv(dev);
+ struct iphdr *iph = &tunnel->parms.iph;
+ int err;
+
+ dev->destructor = ip_tunnel_dev_free;
+ dev->tstats = alloc_percpu(struct pcpu_tstats);
+ if (!dev->tstats)
+ return -ENOMEM;
+
+ err = gro_cells_init(&tunnel->gro_cells, dev);
+ if (err) {
+ free_percpu(dev->tstats);
+ return err;
+ }
+
+ tunnel->dev = dev;
+ strcpy(tunnel->parms.name, dev->name);
+ iph->version = 4;
+ iph->ihl = 5;
+
+ return 0;
+}
+EXPORT_SYMBOL(ip_tunnel_init);
+
+void ip_tunnel_uninit(struct net_device *dev)
+{
+ struct net *net = dev_net(dev);
+ struct ip_tunnel *tunnel = netdev_priv(dev);
+ struct ip_tunnel_net *itn;
+
+ itn = net_generic(net, tunnel->ip_tnl_net_id);
+ /* fb_tunnel_dev will be unregisted in net-exit call. */
+ if (itn->fb_tunnel_dev != dev)
+ ip_tunnel_del(itn, netdev_priv(dev));
+}
+EXPORT_SYMBOL(ip_tunnel_uninit);
+
+void ip_tunnel_setup(struct net_device *dev, int net_id)
+{
+ struct ip_tunnel *tunnel = netdev_priv(dev);
+ tunnel->ip_tnl_net_id = net_id;
+}
+EXPORT_SYMBOL(ip_tunnel_setup);
+
+MODULE_LICENSE("GPL");
diff --git a/net/ipv4/ip_vti.c b/net/ipv4/ip_vti.c
index c3a4233..6a628fb 100644
--- a/net/ipv4/ip_vti.c
+++ b/net/ipv4/ip_vti.c
@@ -38,7 +38,7 @@
#include <net/sock.h>
#include <net/ip.h>
#include <net/icmp.h>
-#include <net/ipip.h>
+#include <net/ip_tunnels.h>
#include <net/inet_ecn.h>
#include <net/xfrm.h>
#include <net/net_namespace.h>
diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c
index 34e006f..a557d6a 100644
--- a/net/ipv4/ipip.c
+++ b/net/ipv4/ipip.c
@@ -111,7 +111,7 @@
#include <net/sock.h>
#include <net/ip.h>
#include <net/icmp.h>
-#include <net/ipip.h>
+#include <net/ip_tunnels.h>
#include <net/inet_ecn.h>
#include <net/xfrm.h>
#include <net/net_namespace.h>
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index 5f95b3a..fd61fe1 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -61,7 +61,7 @@
#include <linux/netfilter_ipv4.h>
#include <linux/compat.h>
#include <linux/export.h>
-#include <net/ipip.h>
+#include <net/ip_tunnels.h>
#include <net/checksum.h>
#include <net/netlink.h>
#include <net/fib_rules.h>
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index f56277f..ab5c7ad 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -49,7 +49,6 @@
#include <net/udp.h>
#include <net/udplite.h>
#include <net/tcp.h>
-#include <net/ipip.h>
#include <net/protocol.h>
#include <net/inet_common.h>
#include <net/route.h>
diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c
index 6a6ba73..af4a5e2 100644
--- a/net/ipv6/ip6_gre.c
+++ b/net/ipv6/ip6_gre.c
@@ -38,6 +38,7 @@
#include <net/sock.h>
#include <net/ip.h>
+#include <net/ip_tunnel.h>
#include <net/icmp.h>
#include <net/protocol.h>
#include <net/addrconf.h>
diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c
index bef3fed..1e55866 100644
--- a/net/ipv6/ip6_tunnel.c
+++ b/net/ipv6/ip6_tunnel.c
@@ -47,6 +47,7 @@
#include <net/icmp.h>
#include <net/ip.h>
+#include <net/ip_tunnels.h>
#include <net/ipv6.h>
#include <net/ip6_route.h>
#include <net/addrconf.h>
diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c
index 898e671..ee4fc57 100644
--- a/net/ipv6/sit.c
+++ b/net/ipv6/sit.c
@@ -49,7 +49,7 @@
#include <net/ip.h>
#include <net/udp.h>
#include <net/icmp.h>
-#include <net/ipip.h>
+#include <net/ip_tunnels.h>
#include <net/inet_ecn.h>
#include <net/xfrm.h>
#include <net/dsfield.h>
--
1.7.1
^ permalink raw reply related
* [PATCH net-next 3/6] IPIP: Use ip-tunneling code.
From: Pravin B Shelar @ 2013-03-13 22:48 UTC (permalink / raw)
To: davem; +Cc: netdev, jesse, Pravin B Shelar
Reuse common ip-tunneling code which is re-factored from GRE
module.
Signed-off-by: Pravin B Shelar <pshelar@nicira.com>
---
drivers/net/Kconfig | 1 +
net/ipv4/Kconfig | 1 +
net/ipv4/ipip.c | 743 ++++-----------------------------------------------
3 files changed, 57 insertions(+), 688 deletions(-)
diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index 87f1d39..3835321 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -151,6 +151,7 @@ config MACVTAP
config VXLAN
tristate "Virtual eXtensible Local Area Network (VXLAN)"
depends on INET
+ select NET_IP_TUNNEL
---help---
This allows one to create vxlan virtual interfaces that provide
Layer 2 Networks over Layer 3 Networks. VXLAN is often used
diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig
index 2073226..053b53e 100644
--- a/net/ipv4/Kconfig
+++ b/net/ipv4/Kconfig
@@ -166,6 +166,7 @@ config IP_PNP_RARP
config NET_IPIP
tristate "IP: tunneling"
select INET_TUNNEL
+ select NET_IP_TUNNEL
---help---
Tunneling means encapsulating data of one protocol type within
another protocol and sending it over a channel that understands the
diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c
index a557d6a..ea9eb03 100644
--- a/net/ipv4/ipip.c
+++ b/net/ipv4/ipip.c
@@ -117,221 +117,15 @@
#include <net/net_namespace.h>
#include <net/netns/generic.h>
-#define HASH_SIZE 16
-#define HASH(addr) (((__force u32)addr^((__force u32)addr>>4))&0xF)
-
static bool log_ecn_error = true;
module_param(log_ecn_error, bool, 0644);
MODULE_PARM_DESC(log_ecn_error, "Log packets received with corrupted ECN");
static int ipip_net_id __read_mostly;
-struct ipip_net {
- struct ip_tunnel __rcu *tunnels_r_l[HASH_SIZE];
- struct ip_tunnel __rcu *tunnels_r[HASH_SIZE];
- struct ip_tunnel __rcu *tunnels_l[HASH_SIZE];
- struct ip_tunnel __rcu *tunnels_wc[1];
- struct ip_tunnel __rcu **tunnels[4];
-
- struct net_device *fb_tunnel_dev;
-};
static int ipip_tunnel_init(struct net_device *dev);
-static void ipip_tunnel_setup(struct net_device *dev);
-static void ipip_dev_free(struct net_device *dev);
static struct rtnl_link_ops ipip_link_ops __read_mostly;
-static struct rtnl_link_stats64 *ipip_get_stats64(struct net_device *dev,
- struct rtnl_link_stats64 *tot)
-{
- int i;
-
- for_each_possible_cpu(i) {
- const struct pcpu_tstats *tstats = per_cpu_ptr(dev->tstats, i);
- u64 rx_packets, rx_bytes, tx_packets, tx_bytes;
- unsigned int start;
-
- do {
- start = u64_stats_fetch_begin_bh(&tstats->syncp);
- rx_packets = tstats->rx_packets;
- tx_packets = tstats->tx_packets;
- rx_bytes = tstats->rx_bytes;
- tx_bytes = tstats->tx_bytes;
- } while (u64_stats_fetch_retry_bh(&tstats->syncp, start));
-
- tot->rx_packets += rx_packets;
- tot->tx_packets += tx_packets;
- tot->rx_bytes += rx_bytes;
- tot->tx_bytes += tx_bytes;
- }
-
- tot->tx_fifo_errors = dev->stats.tx_fifo_errors;
- tot->tx_carrier_errors = dev->stats.tx_carrier_errors;
- tot->tx_dropped = dev->stats.tx_dropped;
- tot->tx_aborted_errors = dev->stats.tx_aborted_errors;
- tot->tx_errors = dev->stats.tx_errors;
- tot->collisions = dev->stats.collisions;
-
- return tot;
-}
-
-static struct ip_tunnel *ipip_tunnel_lookup(struct net *net,
- __be32 remote, __be32 local)
-{
- unsigned int h0 = HASH(remote);
- unsigned int h1 = HASH(local);
- struct ip_tunnel *t;
- struct ipip_net *ipn = net_generic(net, ipip_net_id);
-
- for_each_ip_tunnel_rcu(t, ipn->tunnels_r_l[h0 ^ h1])
- if (local == t->parms.iph.saddr &&
- remote == t->parms.iph.daddr && (t->dev->flags&IFF_UP))
- return t;
-
- for_each_ip_tunnel_rcu(t, ipn->tunnels_r[h0])
- if (remote == t->parms.iph.daddr && (t->dev->flags&IFF_UP))
- return t;
-
- for_each_ip_tunnel_rcu(t, ipn->tunnels_l[h1])
- if (local == t->parms.iph.saddr && (t->dev->flags&IFF_UP))
- return t;
-
- t = rcu_dereference(ipn->tunnels_wc[0]);
- if (t && (t->dev->flags&IFF_UP))
- return t;
- return NULL;
-}
-
-static struct ip_tunnel __rcu **__ipip_bucket(struct ipip_net *ipn,
- struct ip_tunnel_parm *parms)
-{
- __be32 remote = parms->iph.daddr;
- __be32 local = parms->iph.saddr;
- unsigned int h = 0;
- int prio = 0;
-
- if (remote) {
- prio |= 2;
- h ^= HASH(remote);
- }
- if (local) {
- prio |= 1;
- h ^= HASH(local);
- }
- return &ipn->tunnels[prio][h];
-}
-
-static inline struct ip_tunnel __rcu **ipip_bucket(struct ipip_net *ipn,
- struct ip_tunnel *t)
-{
- return __ipip_bucket(ipn, &t->parms);
-}
-
-static void ipip_tunnel_unlink(struct ipip_net *ipn, struct ip_tunnel *t)
-{
- struct ip_tunnel __rcu **tp;
- struct ip_tunnel *iter;
-
- for (tp = ipip_bucket(ipn, t);
- (iter = rtnl_dereference(*tp)) != NULL;
- tp = &iter->next) {
- if (t == iter) {
- rcu_assign_pointer(*tp, t->next);
- break;
- }
- }
-}
-
-static void ipip_tunnel_link(struct ipip_net *ipn, struct ip_tunnel *t)
-{
- struct ip_tunnel __rcu **tp = ipip_bucket(ipn, t);
-
- rcu_assign_pointer(t->next, rtnl_dereference(*tp));
- rcu_assign_pointer(*tp, t);
-}
-
-static int ipip_tunnel_create(struct net_device *dev)
-{
- struct ip_tunnel *t = netdev_priv(dev);
- struct net *net = dev_net(dev);
- struct ipip_net *ipn = net_generic(net, ipip_net_id);
- int err;
-
- err = ipip_tunnel_init(dev);
- if (err < 0)
- goto out;
-
- err = register_netdevice(dev);
- if (err < 0)
- goto out;
-
- strcpy(t->parms.name, dev->name);
- dev->rtnl_link_ops = &ipip_link_ops;
-
- dev_hold(dev);
- ipip_tunnel_link(ipn, t);
- return 0;
-
-out:
- return err;
-}
-
-static struct ip_tunnel *ipip_tunnel_locate(struct net *net,
- struct ip_tunnel_parm *parms, int create)
-{
- __be32 remote = parms->iph.daddr;
- __be32 local = parms->iph.saddr;
- struct ip_tunnel *t, *nt;
- struct ip_tunnel __rcu **tp;
- struct net_device *dev;
- char name[IFNAMSIZ];
- struct ipip_net *ipn = net_generic(net, ipip_net_id);
-
- for (tp = __ipip_bucket(ipn, parms);
- (t = rtnl_dereference(*tp)) != NULL;
- tp = &t->next) {
- if (local == t->parms.iph.saddr && remote == t->parms.iph.daddr)
- return t;
- }
- if (!create)
- return NULL;
-
- if (parms->name[0])
- strlcpy(name, parms->name, IFNAMSIZ);
- else
- strcpy(name, "tunl%d");
-
- dev = alloc_netdev(sizeof(*t), name, ipip_tunnel_setup);
- if (dev == NULL)
- return NULL;
-
- dev_net_set(dev, net);
-
- nt = netdev_priv(dev);
- nt->parms = *parms;
-
- if (ipip_tunnel_create(dev) < 0)
- goto failed_free;
-
- return nt;
-
-failed_free:
- ipip_dev_free(dev);
- return NULL;
-}
-
-/* called with RTNL */
-static void ipip_tunnel_uninit(struct net_device *dev)
-{
- struct net *net = dev_net(dev);
- struct ipip_net *ipn = net_generic(net, ipip_net_id);
-
- if (dev == ipn->fb_tunnel_dev)
- RCU_INIT_POINTER(ipn->tunnels_wc[0], NULL);
- else
- ipip_tunnel_unlink(ipn, netdev_priv(dev));
- dev_put(dev);
-}
-
static int ipip_err(struct sk_buff *skb, u32 info)
{
@@ -339,41 +133,17 @@ static int ipip_err(struct sk_buff *skb, u32 info)
8 bytes of packet payload. It means, that precise relaying of
ICMP in the real Internet is absolutely infeasible.
*/
+ struct net *net = dev_net(skb->dev);
+ struct ip_tunnel_net *itn = net_generic(net, ipip_net_id);
const struct iphdr *iph = (const struct iphdr *)skb->data;
- const int type = icmp_hdr(skb)->type;
- const int code = icmp_hdr(skb)->code;
struct ip_tunnel *t;
int err;
-
- switch (type) {
- default:
- case ICMP_PARAMETERPROB:
- return 0;
-
- case ICMP_DEST_UNREACH:
- switch (code) {
- case ICMP_SR_FAILED:
- case ICMP_PORT_UNREACH:
- /* Impossible event. */
- return 0;
- default:
- /* All others are translated to HOST_UNREACH.
- rfc2003 contains "deep thoughts" about NET_UNREACH,
- I believe they are just ether pollution. --ANK
- */
- break;
- }
- break;
- case ICMP_TIME_EXCEEDED:
- if (code != ICMP_EXC_TTL)
- return 0;
- break;
- case ICMP_REDIRECT:
- break;
- }
+ const int type = icmp_hdr(skb)->type;
+ const int code = icmp_hdr(skb)->code;
err = -ENOENT;
- t = ipip_tunnel_lookup(dev_net(skb->dev), iph->daddr, iph->saddr);
+ t = ip_tunnel_lookup(itn, skb->dev->ifindex, TUNNEL_NO_KEY,
+ iph->daddr, iph->saddr, 0);
if (t == NULL)
goto out;
@@ -403,53 +173,29 @@ static int ipip_err(struct sk_buff *skb, u32 info)
else
t->err_count = 1;
t->err_time = jiffies;
-out:
+out:
return err;
}
+static const struct tnl_ptk_info tpi = {
+ /* no tunnel info required for ipip. */
+ .proto = htons(ETH_P_IP),
+};
+
static int ipip_rcv(struct sk_buff *skb)
{
+ struct net *net = dev_net(skb->dev);
+ struct ip_tunnel_net *itn = net_generic(net, ipip_net_id);
struct ip_tunnel *tunnel;
const struct iphdr *iph = ip_hdr(skb);
- int err;
-
- tunnel = ipip_tunnel_lookup(dev_net(skb->dev), iph->saddr, iph->daddr);
- if (tunnel != NULL) {
- struct pcpu_tstats *tstats;
+ tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, TUNNEL_NO_KEY,
+ iph->saddr, iph->daddr, 0);
+ if (tunnel) {
if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb))
goto drop;
-
- secpath_reset(skb);
-
- skb->mac_header = skb->network_header;
- skb_reset_network_header(skb);
- skb->protocol = htons(ETH_P_IP);
- skb->pkt_type = PACKET_HOST;
-
- __skb_tunnel_rx(skb, tunnel->dev);
-
- err = IP_ECN_decapsulate(iph, skb);
- if (unlikely(err)) {
- if (log_ecn_error)
- net_info_ratelimited("non-ECT from %pI4 with TOS=%#x\n",
- &iph->saddr, iph->tos);
- if (err > 1) {
- ++tunnel->dev->stats.rx_frame_errors;
- ++tunnel->dev->stats.rx_errors;
- goto drop;
- }
- }
-
- tstats = this_cpu_ptr(tunnel->dev->tstats);
- u64_stats_update_begin(&tstats->syncp);
- tstats->rx_packets++;
- tstats->rx_bytes += skb->len;
- u64_stats_update_end(&tstats->syncp);
-
- netif_rx(skb);
- return 0;
+ return ip_tunnel_rcv(tunnel, skb, &tpi, log_ecn_error);
}
return -1;
@@ -463,333 +209,65 @@ drop:
* This function assumes it is being called from dev_queue_xmit()
* and that skb is filled properly by that function.
*/
-
static netdev_tx_t ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
{
struct ip_tunnel *tunnel = netdev_priv(dev);
const struct iphdr *tiph = &tunnel->parms.iph;
- u8 tos = tunnel->parms.iph.tos;
- __be16 df = tiph->frag_off;
- struct rtable *rt; /* Route to the other host */
- struct net_device *tdev; /* Device to other host */
- const struct iphdr *old_iph;
- struct iphdr *iph; /* Our new IP header */
- unsigned int max_headroom; /* The extra header space needed */
- __be32 dst = tiph->daddr;
- struct flowi4 fl4;
- int mtu;
if (skb->protocol != htons(ETH_P_IP))
goto tx_error;
- old_iph = ip_hdr(skb);
-
- if (tos & 1)
- tos = old_iph->tos;
-
- if (!dst) {
- /* NBMA tunnel */
- if ((rt = skb_rtable(skb)) == NULL) {
- dev->stats.tx_fifo_errors++;
- goto tx_error;
- }
- dst = rt_nexthop(rt, old_iph->daddr);
- }
-
- rt = ip_route_output_ports(dev_net(dev), &fl4, NULL,
- dst, tiph->saddr,
- 0, 0,
- IPPROTO_IPIP, RT_TOS(tos),
- tunnel->parms.link);
- if (IS_ERR(rt)) {
- dev->stats.tx_carrier_errors++;
- goto tx_error_icmp;
- }
- tdev = rt->dst.dev;
-
- if (tdev == dev) {
- ip_rt_put(rt);
- dev->stats.collisions++;
- goto tx_error;
- }
-
- df |= old_iph->frag_off & htons(IP_DF);
-
- if (df) {
- mtu = dst_mtu(&rt->dst) - sizeof(struct iphdr);
-
- if (mtu < 68) {
- dev->stats.collisions++;
- ip_rt_put(rt);
- goto tx_error;
- }
-
- if (skb_dst(skb))
- skb_dst(skb)->ops->update_pmtu(skb_dst(skb), NULL, skb, mtu);
-
- if ((old_iph->frag_off & htons(IP_DF)) &&
- mtu < ntohs(old_iph->tot_len)) {
- icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED,
- htonl(mtu));
- ip_rt_put(rt);
- goto tx_error;
- }
- }
-
- if (tunnel->err_count > 0) {
- if (time_before(jiffies,
- tunnel->err_time + IPTUNNEL_ERR_TIMEO)) {
- tunnel->err_count--;
- dst_link_failure(skb);
- } else
- tunnel->err_count = 0;
- }
-
- /*
- * Okay, now see if we can stuff it in the buffer as-is.
- */
- max_headroom = (LL_RESERVED_SPACE(tdev)+sizeof(struct iphdr));
-
- if (skb_headroom(skb) < max_headroom || skb_shared(skb) ||
- (skb_cloned(skb) && !skb_clone_writable(skb, 0))) {
- struct sk_buff *new_skb = skb_realloc_headroom(skb, max_headroom);
- if (!new_skb) {
- ip_rt_put(rt);
- dev->stats.tx_dropped++;
- dev_kfree_skb(skb);
- return NETDEV_TX_OK;
- }
- if (skb->sk)
- skb_set_owner_w(new_skb, skb->sk);
- dev_kfree_skb(skb);
- skb = new_skb;
- old_iph = ip_hdr(skb);
- }
if (!skb->encapsulation) {
skb_reset_inner_headers(skb);
skb->encapsulation = 1;
}
- if (skb->ip_summed != CHECKSUM_PARTIAL)
- skb->ip_summed = CHECKSUM_NONE;
-
- skb->transport_header = skb->network_header;
- skb_push(skb, sizeof(struct iphdr));
- skb_reset_network_header(skb);
- memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt));
- IPCB(skb)->flags &= ~(IPSKB_XFRM_TUNNEL_SIZE | IPSKB_XFRM_TRANSFORMED |
- IPSKB_REROUTED);
- skb_dst_drop(skb);
- skb_dst_set(skb, &rt->dst);
-
- /*
- * Push down and install the IPIP header.
- */
-
- iph = ip_hdr(skb);
- iph->version = 4;
- iph->ihl = sizeof(struct iphdr)>>2;
- iph->frag_off = df;
- iph->protocol = IPPROTO_IPIP;
- iph->tos = INET_ECN_encapsulate(tos, old_iph->tos);
- iph->daddr = fl4.daddr;
- iph->saddr = fl4.saddr;
- tunnel_ip_select_ident(skb, old_iph, &rt->dst);
-
- if ((iph->ttl = tiph->ttl) == 0)
- iph->ttl = old_iph->ttl;
-
- iptunnel_xmit(skb, dev);
+ ip_tunnel_xmit(skb, dev, tiph);
return NETDEV_TX_OK;
-tx_error_icmp:
- dst_link_failure(skb);
tx_error:
dev->stats.tx_errors++;
dev_kfree_skb(skb);
return NETDEV_TX_OK;
}
-static void ipip_tunnel_bind_dev(struct net_device *dev)
-{
- struct net_device *tdev = NULL;
- struct ip_tunnel *tunnel;
- const struct iphdr *iph;
-
- tunnel = netdev_priv(dev);
- iph = &tunnel->parms.iph;
-
- if (iph->daddr) {
- struct rtable *rt;
- struct flowi4 fl4;
-
- rt = ip_route_output_ports(dev_net(dev), &fl4, NULL,
- iph->daddr, iph->saddr,
- 0, 0,
- IPPROTO_IPIP,
- RT_TOS(iph->tos),
- tunnel->parms.link);
- if (!IS_ERR(rt)) {
- tdev = rt->dst.dev;
- ip_rt_put(rt);
- }
- dev->flags |= IFF_POINTOPOINT;
- }
-
- if (!tdev && tunnel->parms.link)
- tdev = __dev_get_by_index(dev_net(dev), tunnel->parms.link);
-
- if (tdev) {
- dev->hard_header_len = tdev->hard_header_len + sizeof(struct iphdr);
- dev->mtu = tdev->mtu - sizeof(struct iphdr);
- }
- dev->iflink = tunnel->parms.link;
-}
-
-static void ipip_tunnel_update(struct ip_tunnel *t, struct ip_tunnel_parm *p)
-{
- struct net *net = dev_net(t->dev);
- struct ipip_net *ipn = net_generic(net, ipip_net_id);
-
- ipip_tunnel_unlink(ipn, t);
- synchronize_net();
- t->parms.iph.saddr = p->iph.saddr;
- t->parms.iph.daddr = p->iph.daddr;
- memcpy(t->dev->dev_addr, &p->iph.saddr, 4);
- memcpy(t->dev->broadcast, &p->iph.daddr, 4);
- ipip_tunnel_link(ipn, t);
- t->parms.iph.ttl = p->iph.ttl;
- t->parms.iph.tos = p->iph.tos;
- t->parms.iph.frag_off = p->iph.frag_off;
- if (t->parms.link != p->link) {
- t->parms.link = p->link;
- ipip_tunnel_bind_dev(t->dev);
- }
- netdev_state_change(t->dev);
-}
-
static int
-ipip_tunnel_ioctl (struct net_device *dev, struct ifreq *ifr, int cmd)
+ipip_tunnel_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
{
int err = 0;
struct ip_tunnel_parm p;
- struct ip_tunnel *t;
- struct net *net = dev_net(dev);
- struct ipip_net *ipn = net_generic(net, ipip_net_id);
-
- switch (cmd) {
- case SIOCGETTUNNEL:
- t = NULL;
- if (dev == ipn->fb_tunnel_dev) {
- if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p))) {
- err = -EFAULT;
- break;
- }
- t = ipip_tunnel_locate(net, &p, 0);
- }
- if (t == NULL)
- t = netdev_priv(dev);
- memcpy(&p, &t->parms, sizeof(p));
- if (copy_to_user(ifr->ifr_ifru.ifru_data, &p, sizeof(p)))
- err = -EFAULT;
- break;
-
- case SIOCADDTUNNEL:
- case SIOCCHGTUNNEL:
- err = -EPERM;
- if (!ns_capable(net->user_ns, CAP_NET_ADMIN))
- goto done;
-
- err = -EFAULT;
- if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p)))
- goto done;
-
- err = -EINVAL;
- if (p.iph.version != 4 || p.iph.protocol != IPPROTO_IPIP ||
- p.iph.ihl != 5 || (p.iph.frag_off&htons(~IP_DF)))
- goto done;
- if (p.iph.ttl)
- p.iph.frag_off |= htons(IP_DF);
-
- t = ipip_tunnel_locate(net, &p, cmd == SIOCADDTUNNEL);
-
- if (dev != ipn->fb_tunnel_dev && cmd == SIOCCHGTUNNEL) {
- if (t != NULL) {
- if (t->dev != dev) {
- err = -EEXIST;
- break;
- }
- } else {
- if (((dev->flags&IFF_POINTOPOINT) && !p.iph.daddr) ||
- (!(dev->flags&IFF_POINTOPOINT) && p.iph.daddr)) {
- err = -EINVAL;
- break;
- }
- t = netdev_priv(dev);
- }
-
- ipip_tunnel_update(t, &p);
- }
-
- if (t) {
- err = 0;
- if (copy_to_user(ifr->ifr_ifru.ifru_data, &t->parms, sizeof(p)))
- err = -EFAULT;
- } else
- err = (cmd == SIOCADDTUNNEL ? -ENOBUFS : -ENOENT);
- break;
-
- case SIOCDELTUNNEL:
- err = -EPERM;
- if (!ns_capable(net->user_ns, CAP_NET_ADMIN))
- goto done;
-
- if (dev == ipn->fb_tunnel_dev) {
- err = -EFAULT;
- if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p)))
- goto done;
- err = -ENOENT;
- if ((t = ipip_tunnel_locate(net, &p, 0)) == NULL)
- goto done;
- err = -EPERM;
- if (t->dev == ipn->fb_tunnel_dev)
- goto done;
- dev = t->dev;
- }
- unregister_netdevice(dev);
- err = 0;
- break;
-
- default:
- err = -EINVAL;
- }
-done:
- return err;
-}
+ if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p)))
+ return -EFAULT;
-static int ipip_tunnel_change_mtu(struct net_device *dev, int new_mtu)
-{
- if (new_mtu < 68 || new_mtu > 0xFFF8 - sizeof(struct iphdr))
+ if (p.iph.version != 4 || p.iph.protocol != IPPROTO_IPIP ||
+ p.iph.ihl != 5 || (p.iph.frag_off&htons(~IP_DF)))
+ return -EINVAL;
+ if (p.i_key || p.o_key || p.i_flags || p.o_flags)
return -EINVAL;
- dev->mtu = new_mtu;
+ if (p.iph.ttl)
+ p.iph.frag_off |= htons(IP_DF);
+
+
+ err = ip_tunnel_ioctl(dev, &p, cmd);
+ if (err)
+ return err;
+
+ if (copy_to_user(ifr->ifr_ifru.ifru_data, &p, sizeof(p)))
+ return -EFAULT;
+
return 0;
}
static const struct net_device_ops ipip_netdev_ops = {
- .ndo_uninit = ipip_tunnel_uninit,
+ .ndo_init = ipip_tunnel_init,
+ .ndo_uninit = ip_tunnel_uninit,
.ndo_start_xmit = ipip_tunnel_xmit,
.ndo_do_ioctl = ipip_tunnel_ioctl,
- .ndo_change_mtu = ipip_tunnel_change_mtu,
- .ndo_get_stats64 = ipip_get_stats64,
+ .ndo_change_mtu = ip_tunnel_change_mtu,
+ .ndo_get_stats64 = ip_tunnel_get_stats64,
};
-static void ipip_dev_free(struct net_device *dev)
-{
- free_percpu(dev->tstats);
- free_netdev(dev);
-}
-
#define IPIP_FEATURES (NETIF_F_SG | \
NETIF_F_FRAGLIST | \
NETIF_F_HIGHDMA | \
@@ -798,11 +276,8 @@ static void ipip_dev_free(struct net_device *dev)
static void ipip_tunnel_setup(struct net_device *dev)
{
dev->netdev_ops = &ipip_netdev_ops;
- dev->destructor = ipip_dev_free;
dev->type = ARPHRD_TUNNEL;
- dev->hard_header_len = LL_MAX_HEADER + sizeof(struct iphdr);
- dev->mtu = ETH_DATA_LEN - sizeof(struct iphdr);
dev->flags = IFF_NOARP;
dev->iflink = 0;
dev->addr_len = 4;
@@ -812,46 +287,19 @@ static void ipip_tunnel_setup(struct net_device *dev)
dev->features |= IPIP_FEATURES;
dev->hw_features |= IPIP_FEATURES;
+ ip_tunnel_setup(dev, ipip_net_id);
}
static int ipip_tunnel_init(struct net_device *dev)
{
struct ip_tunnel *tunnel = netdev_priv(dev);
- tunnel->dev = dev;
-
memcpy(dev->dev_addr, &tunnel->parms.iph.saddr, 4);
memcpy(dev->broadcast, &tunnel->parms.iph.daddr, 4);
- ipip_tunnel_bind_dev(dev);
-
- dev->tstats = alloc_percpu(struct pcpu_tstats);
- if (!dev->tstats)
- return -ENOMEM;
-
- return 0;
-}
-
-static int __net_init ipip_fb_tunnel_init(struct net_device *dev)
-{
- struct ip_tunnel *tunnel = netdev_priv(dev);
- struct iphdr *iph = &tunnel->parms.iph;
- struct ipip_net *ipn = net_generic(dev_net(dev), ipip_net_id);
-
- tunnel->dev = dev;
- strcpy(tunnel->parms.name, dev->name);
-
- iph->version = 4;
- iph->protocol = IPPROTO_IPIP;
- iph->ihl = 5;
-
- dev->tstats = alloc_percpu(struct pcpu_tstats);
- if (!dev->tstats)
- return -ENOMEM;
-
- dev_hold(dev);
- rcu_assign_pointer(ipn->tunnels_wc[0], tunnel);
- return 0;
+ tunnel->hlen = 0;
+ tunnel->parms.iph.protocol = IPPROTO_IPIP;
+ return ip_tunnel_init(dev);
}
static void ipip_netlink_parms(struct nlattr *data[],
@@ -891,28 +339,16 @@ static void ipip_netlink_parms(struct nlattr *data[],
static int ipip_newlink(struct net *src_net, struct net_device *dev,
struct nlattr *tb[], struct nlattr *data[])
{
- struct net *net = dev_net(dev);
- struct ip_tunnel *nt;
-
- nt = netdev_priv(dev);
- ipip_netlink_parms(data, &nt->parms);
-
- if (ipip_tunnel_locate(net, &nt->parms, 0))
- return -EEXIST;
+ struct ip_tunnel_parm p;
- return ipip_tunnel_create(dev);
+ ipip_netlink_parms(data, &p);
+ return ip_tunnel_newlink(dev, tb, &p);
}
static int ipip_changelink(struct net_device *dev, struct nlattr *tb[],
struct nlattr *data[])
{
- struct ip_tunnel *t;
struct ip_tunnel_parm p;
- struct net *net = dev_net(dev);
- struct ipip_net *ipn = net_generic(net, ipip_net_id);
-
- if (dev == ipn->fb_tunnel_dev)
- return -EINVAL;
ipip_netlink_parms(data, &p);
@@ -920,16 +356,7 @@ static int ipip_changelink(struct net_device *dev, struct nlattr *tb[],
(!(dev->flags & IFF_POINTOPOINT) && p.iph.daddr))
return -EINVAL;
- t = ipip_tunnel_locate(net, &p, 0);
-
- if (t) {
- if (t->dev != dev)
- return -EEXIST;
- } else
- t = netdev_priv(dev);
-
- ipip_tunnel_update(t, &p);
- return 0;
+ return ip_tunnel_changelink(dev, tb, &p);
}
static size_t ipip_get_size(const struct net_device *dev)
@@ -986,6 +413,7 @@ static struct rtnl_link_ops ipip_link_ops __read_mostly = {
.setup = ipip_tunnel_setup,
.newlink = ipip_newlink,
.changelink = ipip_changelink,
+ .dellink = ip_tunnel_dellink,
.get_size = ipip_get_size,
.fill_info = ipip_fill_info,
};
@@ -996,90 +424,29 @@ static struct xfrm_tunnel ipip_handler __read_mostly = {
.priority = 1,
};
-static const char banner[] __initconst =
- KERN_INFO "IPv4 over IPv4 tunneling driver\n";
-
-static void ipip_destroy_tunnels(struct ipip_net *ipn, struct list_head *head)
-{
- int prio;
-
- for (prio = 1; prio < 4; prio++) {
- int h;
- for (h = 0; h < HASH_SIZE; h++) {
- struct ip_tunnel *t;
-
- t = rtnl_dereference(ipn->tunnels[prio][h]);
- while (t != NULL) {
- unregister_netdevice_queue(t->dev, head);
- t = rtnl_dereference(t->next);
- }
- }
- }
-}
-
static int __net_init ipip_init_net(struct net *net)
{
- struct ipip_net *ipn = net_generic(net, ipip_net_id);
- struct ip_tunnel *t;
- int err;
-
- ipn->tunnels[0] = ipn->tunnels_wc;
- ipn->tunnels[1] = ipn->tunnels_l;
- ipn->tunnels[2] = ipn->tunnels_r;
- ipn->tunnels[3] = ipn->tunnels_r_l;
-
- ipn->fb_tunnel_dev = alloc_netdev(sizeof(struct ip_tunnel),
- "tunl0",
- ipip_tunnel_setup);
- if (!ipn->fb_tunnel_dev) {
- err = -ENOMEM;
- goto err_alloc_dev;
- }
- dev_net_set(ipn->fb_tunnel_dev, net);
-
- err = ipip_fb_tunnel_init(ipn->fb_tunnel_dev);
- if (err)
- goto err_reg_dev;
-
- if ((err = register_netdev(ipn->fb_tunnel_dev)))
- goto err_reg_dev;
-
- t = netdev_priv(ipn->fb_tunnel_dev);
-
- strcpy(t->parms.name, ipn->fb_tunnel_dev->name);
- return 0;
-
-err_reg_dev:
- ipip_dev_free(ipn->fb_tunnel_dev);
-err_alloc_dev:
- /* nothing */
- return err;
+ return ip_tunnel_init_net(net, ipip_net_id, &ipip_link_ops, "tunl0");
}
static void __net_exit ipip_exit_net(struct net *net)
{
- struct ipip_net *ipn = net_generic(net, ipip_net_id);
- LIST_HEAD(list);
-
- rtnl_lock();
- ipip_destroy_tunnels(ipn, &list);
- unregister_netdevice_queue(ipn->fb_tunnel_dev, &list);
- unregister_netdevice_many(&list);
- rtnl_unlock();
+ struct ip_tunnel_net *itn = net_generic(net, ipip_net_id);
+ ip_tunnel_delete_net(itn);
}
static struct pernet_operations ipip_net_ops = {
.init = ipip_init_net,
.exit = ipip_exit_net,
.id = &ipip_net_id,
- .size = sizeof(struct ipip_net),
+ .size = sizeof(struct ip_tunnel_net),
};
static int __init ipip_init(void)
{
int err;
- printk(banner);
+ pr_info("ipip: IPv4 over IPv4 tunneling driver\n");
err = register_pernet_device(&ipip_net_ops);
if (err < 0)
--
1.7.1
^ permalink raw reply related
* [PATCH net-next 4/6] VXLAN: Fix vxlan stats handling.
From: Pravin B Shelar @ 2013-03-13 22:48 UTC (permalink / raw)
To: davem; +Cc: stephen, xiyou.wangcong, netdev, jesse, Pravin B Shelar
Fixes bug in VXLAN code where is iptunnel_xmit() called with NULL
dev->tstats.
This bug was introduced in commit 6aed0c8bf7d2f389b (tunnel: use
iptunnel_xmit() again).
Following patch fixes bug by setting dev->tstats. It uses ip_tunnel
module code to share stats function.
CC: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: Pravin B Shelar <pshelar@nicira.com>
---
drivers/net/vxlan.c | 72 +++++----------------------------------------------
1 files changed, 7 insertions(+), 65 deletions(-)
diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index 338d44e..d844168 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -92,20 +92,10 @@ struct vxlan_fdb {
u8 eth_addr[ETH_ALEN];
};
-/* Per-cpu network traffic stats */
-struct vxlan_stats {
- u64 rx_packets;
- u64 rx_bytes;
- u64 tx_packets;
- u64 tx_bytes;
- struct u64_stats_sync syncp;
-};
-
/* Pseudo network device */
struct vxlan_dev {
struct hlist_node hlist;
struct net_device *dev;
- struct vxlan_stats __percpu *stats;
__u32 vni; /* virtual network id */
__be32 gaddr; /* multicast group */
__be32 saddr; /* source address */
@@ -556,7 +546,7 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
struct iphdr *oip;
struct vxlanhdr *vxh;
struct vxlan_dev *vxlan;
- struct vxlan_stats *stats;
+ struct pcpu_tstats *stats;
__u32 vni;
int err;
@@ -632,7 +622,7 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
}
}
- stats = this_cpu_ptr(vxlan->stats);
+ stats = this_cpu_ptr(vxlan->dev->tstats);
u64_stats_update_begin(&stats->syncp);
stats->rx_packets++;
stats->rx_bytes += skb->len;
@@ -884,8 +874,7 @@ static netdev_tx_t vxlan_xmit(struct sk_buff *skb, struct net_device *dev)
/* short-circuited back to local bridge */
if (netif_rx(skb) == NET_RX_SUCCESS) {
- struct vxlan_stats *stats =
- this_cpu_ptr(vxlan->stats);
+ struct pcpu_tstats *stats = this_cpu_ptr(dev->tstats);
u64_stats_update_begin(&stats->syncp);
stats->tx_packets++;
@@ -1035,10 +1024,8 @@ static void vxlan_cleanup(unsigned long arg)
/* Setup stats when device is created */
static int vxlan_init(struct net_device *dev)
{
- struct vxlan_dev *vxlan = netdev_priv(dev);
-
- vxlan->stats = alloc_percpu(struct vxlan_stats);
- if (!vxlan->stats)
+ dev->tstats = alloc_percpu(struct pcpu_tstats);
+ if (!dev->tstats)
return -ENOMEM;
return 0;
@@ -1094,49 +1081,6 @@ static int vxlan_stop(struct net_device *dev)
return 0;
}
-/* Merge per-cpu statistics */
-static struct rtnl_link_stats64 *vxlan_stats64(struct net_device *dev,
- struct rtnl_link_stats64 *stats)
-{
- struct vxlan_dev *vxlan = netdev_priv(dev);
- struct vxlan_stats tmp, sum = { 0 };
- unsigned int cpu;
-
- for_each_possible_cpu(cpu) {
- unsigned int start;
- const struct vxlan_stats *stats
- = per_cpu_ptr(vxlan->stats, cpu);
-
- do {
- start = u64_stats_fetch_begin_bh(&stats->syncp);
- memcpy(&tmp, stats, sizeof(tmp));
- } while (u64_stats_fetch_retry_bh(&stats->syncp, start));
-
- sum.tx_bytes += tmp.tx_bytes;
- sum.tx_packets += tmp.tx_packets;
- sum.rx_bytes += tmp.rx_bytes;
- sum.rx_packets += tmp.rx_packets;
- }
-
- stats->tx_bytes = sum.tx_bytes;
- stats->tx_packets = sum.tx_packets;
- stats->rx_bytes = sum.rx_bytes;
- stats->rx_packets = sum.rx_packets;
-
- stats->multicast = dev->stats.multicast;
- stats->rx_length_errors = dev->stats.rx_length_errors;
- stats->rx_frame_errors = dev->stats.rx_frame_errors;
- stats->rx_errors = dev->stats.rx_errors;
-
- stats->tx_dropped = dev->stats.tx_dropped;
- stats->tx_carrier_errors = dev->stats.tx_carrier_errors;
- stats->tx_aborted_errors = dev->stats.tx_aborted_errors;
- stats->collisions = dev->stats.collisions;
- stats->tx_errors = dev->stats.tx_errors;
-
- return stats;
-}
-
/* Stub, nothing needs to be done. */
static void vxlan_set_multicast_list(struct net_device *dev)
{
@@ -1147,7 +1091,7 @@ static const struct net_device_ops vxlan_netdev_ops = {
.ndo_open = vxlan_open,
.ndo_stop = vxlan_stop,
.ndo_start_xmit = vxlan_xmit,
- .ndo_get_stats64 = vxlan_stats64,
+ .ndo_get_stats64 = ip_tunnel_get_stats64,
.ndo_set_rx_mode = vxlan_set_multicast_list,
.ndo_change_mtu = eth_change_mtu,
.ndo_validate_addr = eth_validate_addr,
@@ -1164,9 +1108,7 @@ static struct device_type vxlan_type = {
static void vxlan_free(struct net_device *dev)
{
- struct vxlan_dev *vxlan = netdev_priv(dev);
-
- free_percpu(vxlan->stats);
+ free_percpu(dev->tstats);
free_netdev(dev);
}
--
1.7.1
^ permalink raw reply related
* [PATCH net-next 5/6] VXLAN: Use IP Tunnels tunnel ENC encap API
From: Pravin B Shelar @ 2013-03-13 22:48 UTC (permalink / raw)
To: davem; +Cc: stephen, netdev, jesse, Pravin B Shelar
Use common ecn_encap functions from ip_tunnel module.
Signed-off-by: Pravin B Shelar <pshelar@nicira.com>
---
drivers/net/vxlan.c | 26 ++------------------------
1 files changed, 2 insertions(+), 24 deletions(-)
diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index d844168..74e37eb 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -753,28 +753,6 @@ static bool route_shortcircuit(struct net_device *dev, struct sk_buff *skb)
return false;
}
-/* Extract dsfield from inner protocol */
-static inline u8 vxlan_get_dsfield(const struct iphdr *iph,
- const struct sk_buff *skb)
-{
- if (skb->protocol == htons(ETH_P_IP))
- return iph->tos;
- else if (skb->protocol == htons(ETH_P_IPV6))
- return ipv6_get_dsfield((const struct ipv6hdr *)iph);
- else
- return 0;
-}
-
-/* Propogate ECN bits out */
-static inline u8 vxlan_ecn_encap(u8 tos,
- const struct iphdr *iph,
- const struct sk_buff *skb)
-{
- u8 inner = vxlan_get_dsfield(iph, skb);
-
- return INET_ECN_encapsulate(tos, inner);
-}
-
static void vxlan_sock_free(struct sk_buff *skb)
{
sock_put(skb->sk);
@@ -906,7 +884,7 @@ static netdev_tx_t vxlan_xmit(struct sk_buff *skb, struct net_device *dev)
tos = vxlan->tos;
if (tos == 1)
- tos = vxlan_get_dsfield(old_iph, skb);
+ tos = ip_tunnel_get_dsfield(old_iph, skb);
src_port = vxlan_src_port(vxlan, skb);
@@ -957,7 +935,7 @@ static netdev_tx_t vxlan_xmit(struct sk_buff *skb, struct net_device *dev)
iph->ihl = sizeof(struct iphdr) >> 2;
iph->frag_off = df;
iph->protocol = IPPROTO_UDP;
- iph->tos = vxlan_ecn_encap(tos, old_iph, skb);
+ iph->tos = ip_tunnel_ecn_encap(tos, old_iph, skb);
iph->daddr = dst;
iph->saddr = fl4.saddr;
iph->ttl = ttl ? : ip4_dst_hoplimit(&rt->dst);
--
1.7.1
^ permalink raw reply related
* [PATCH net-next 6/6] tunnels: Use IP Tunnel stats APIs.
From: Pravin B Shelar @ 2013-03-13 22:48 UTC (permalink / raw)
To: davem; +Cc: netdev, jesse, Pravin B Shelar
Use common function to calculate rtnl_link_stats64 stats.
Signed-off-by: Pravin B Shelar <pshelar@nicira.com>
---
net/ipv4/Kconfig | 1 +
net/ipv4/ip_vti.c | 40 +---------------------------------------
net/ipv6/Kconfig | 2 ++
net/ipv6/ip6_gre.c | 42 +-----------------------------------------
net/ipv6/sit.c | 37 +------------------------------------
5 files changed, 6 insertions(+), 116 deletions(-)
diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig
index 053b53e..8603ca8 100644
--- a/net/ipv4/Kconfig
+++ b/net/ipv4/Kconfig
@@ -319,6 +319,7 @@ config SYN_COOKIES
config NET_IPVTI
tristate "Virtual (secure) IP: tunneling"
select INET_TUNNEL
+ select NET_IP_TUNNEL
depends on INET_XFRM_MODE_TUNNEL
---help---
Tunneling means encapsulating data of one protocol type within
diff --git a/net/ipv4/ip_vti.c b/net/ipv4/ip_vti.c
index 6a628fb..9d2bdb2 100644
--- a/net/ipv4/ip_vti.c
+++ b/net/ipv4/ip_vti.c
@@ -82,44 +82,6 @@ static int vti_tunnel_bind_dev(struct net_device *dev);
} while (0)
-static struct rtnl_link_stats64 *vti_get_stats64(struct net_device *dev,
- struct rtnl_link_stats64 *tot)
-{
- int i;
-
- for_each_possible_cpu(i) {
- const struct pcpu_tstats *tstats = per_cpu_ptr(dev->tstats, i);
- u64 rx_packets, rx_bytes, tx_packets, tx_bytes;
- unsigned int start;
-
- do {
- start = u64_stats_fetch_begin_bh(&tstats->syncp);
- rx_packets = tstats->rx_packets;
- tx_packets = tstats->tx_packets;
- rx_bytes = tstats->rx_bytes;
- tx_bytes = tstats->tx_bytes;
- } while (u64_stats_fetch_retry_bh(&tstats->syncp, start));
-
- tot->rx_packets += rx_packets;
- tot->tx_packets += tx_packets;
- tot->rx_bytes += rx_bytes;
- tot->tx_bytes += tx_bytes;
- }
-
- tot->multicast = dev->stats.multicast;
- tot->rx_crc_errors = dev->stats.rx_crc_errors;
- tot->rx_fifo_errors = dev->stats.rx_fifo_errors;
- tot->rx_length_errors = dev->stats.rx_length_errors;
- tot->rx_errors = dev->stats.rx_errors;
- tot->tx_fifo_errors = dev->stats.tx_fifo_errors;
- tot->tx_carrier_errors = dev->stats.tx_carrier_errors;
- tot->tx_dropped = dev->stats.tx_dropped;
- tot->tx_aborted_errors = dev->stats.tx_aborted_errors;
- tot->tx_errors = dev->stats.tx_errors;
-
- return tot;
-}
-
static struct ip_tunnel *vti_tunnel_lookup(struct net *net,
__be32 remote, __be32 local)
{
@@ -597,7 +559,7 @@ static const struct net_device_ops vti_netdev_ops = {
.ndo_start_xmit = vti_tunnel_xmit,
.ndo_do_ioctl = vti_tunnel_ioctl,
.ndo_change_mtu = vti_tunnel_change_mtu,
- .ndo_get_stats64 = vti_get_stats64,
+ .ndo_get_stats64 = ip_tunnel_get_stats64,
};
static void vti_dev_free(struct net_device *dev)
diff --git a/net/ipv6/Kconfig b/net/ipv6/Kconfig
index ed0b9e2..11b13ea 100644
--- a/net/ipv6/Kconfig
+++ b/net/ipv6/Kconfig
@@ -156,6 +156,7 @@ config INET6_XFRM_MODE_ROUTEOPTIMIZATION
config IPV6_SIT
tristate "IPv6: IPv6-in-IPv4 tunnel (SIT driver)"
select INET_TUNNEL
+ select NET_IP_TUNNEL
select IPV6_NDISC_NODETYPE
default y
---help---
@@ -201,6 +202,7 @@ config IPV6_TUNNEL
config IPV6_GRE
tristate "IPv6: GRE tunnel"
select IPV6_TUNNEL
+ select NET_IP_TUNNEL
---help---
Tunneling means encapsulating data of one protocol type within
another protocol and sending it over a channel that understands the
diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c
index af4a5e2..499a545 100644
--- a/net/ipv6/ip6_gre.c
+++ b/net/ipv6/ip6_gre.c
@@ -111,46 +111,6 @@ static u32 HASH_ADDR(const struct in6_addr *addr)
#define tunnels_l tunnels[1]
#define tunnels_wc tunnels[0]
-static struct rtnl_link_stats64 *ip6gre_get_stats64(struct net_device *dev,
- struct rtnl_link_stats64 *tot)
-{
- int i;
-
- for_each_possible_cpu(i) {
- const struct pcpu_tstats *tstats = per_cpu_ptr(dev->tstats, i);
- u64 rx_packets, rx_bytes, tx_packets, tx_bytes;
- unsigned int start;
-
- do {
- start = u64_stats_fetch_begin_bh(&tstats->syncp);
- rx_packets = tstats->rx_packets;
- tx_packets = tstats->tx_packets;
- rx_bytes = tstats->rx_bytes;
- tx_bytes = tstats->tx_bytes;
- } while (u64_stats_fetch_retry_bh(&tstats->syncp, start));
-
- tot->rx_packets += rx_packets;
- tot->tx_packets += tx_packets;
- tot->rx_bytes += rx_bytes;
- tot->tx_bytes += tx_bytes;
- }
-
- tot->multicast = dev->stats.multicast;
- tot->rx_crc_errors = dev->stats.rx_crc_errors;
- tot->rx_fifo_errors = dev->stats.rx_fifo_errors;
- tot->rx_length_errors = dev->stats.rx_length_errors;
- tot->rx_frame_errors = dev->stats.rx_frame_errors;
- tot->rx_errors = dev->stats.rx_errors;
-
- tot->tx_fifo_errors = dev->stats.tx_fifo_errors;
- tot->tx_carrier_errors = dev->stats.tx_carrier_errors;
- tot->tx_dropped = dev->stats.tx_dropped;
- tot->tx_aborted_errors = dev->stats.tx_aborted_errors;
- tot->tx_errors = dev->stats.tx_errors;
-
- return tot;
-}
-
/* Given src, dst and key, find appropriate for input tunnel. */
static struct ip6_tnl *ip6gre_tunnel_lookup(struct net_device *dev,
@@ -1257,7 +1217,7 @@ static const struct net_device_ops ip6gre_netdev_ops = {
.ndo_start_xmit = ip6gre_tunnel_xmit,
.ndo_do_ioctl = ip6gre_tunnel_ioctl,
.ndo_change_mtu = ip6gre_tunnel_change_mtu,
- .ndo_get_stats64 = ip6gre_get_stats64,
+ .ndo_get_stats64 = ip_tunnel_get_stats64,
};
static void ip6gre_dev_free(struct net_device *dev)
diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c
index ee4fc57..35f28d0 100644
--- a/net/ipv6/sit.c
+++ b/net/ipv6/sit.c
@@ -87,41 +87,6 @@ struct sit_net {
struct net_device *fb_tunnel_dev;
};
-static struct rtnl_link_stats64 *ipip6_get_stats64(struct net_device *dev,
- struct rtnl_link_stats64 *tot)
-{
- int i;
-
- for_each_possible_cpu(i) {
- const struct pcpu_tstats *tstats = per_cpu_ptr(dev->tstats, i);
- u64 rx_packets, rx_bytes, tx_packets, tx_bytes;
- unsigned int start;
-
- do {
- start = u64_stats_fetch_begin_bh(&tstats->syncp);
- rx_packets = tstats->rx_packets;
- tx_packets = tstats->tx_packets;
- rx_bytes = tstats->rx_bytes;
- tx_bytes = tstats->tx_bytes;
- } while (u64_stats_fetch_retry_bh(&tstats->syncp, start));
-
- tot->rx_packets += rx_packets;
- tot->tx_packets += tx_packets;
- tot->rx_bytes += rx_bytes;
- tot->tx_bytes += tx_bytes;
- }
-
- tot->rx_errors = dev->stats.rx_errors;
- tot->rx_frame_errors = dev->stats.rx_frame_errors;
- tot->tx_fifo_errors = dev->stats.tx_fifo_errors;
- tot->tx_carrier_errors = dev->stats.tx_carrier_errors;
- tot->tx_dropped = dev->stats.tx_dropped;
- tot->tx_aborted_errors = dev->stats.tx_aborted_errors;
- tot->tx_errors = dev->stats.tx_errors;
-
- return tot;
-}
-
/*
* Must be invoked with rcu_read_lock
*/
@@ -1202,7 +1167,7 @@ static const struct net_device_ops ipip6_netdev_ops = {
.ndo_start_xmit = ipip6_tunnel_xmit,
.ndo_do_ioctl = ipip6_tunnel_ioctl,
.ndo_change_mtu = ipip6_tunnel_change_mtu,
- .ndo_get_stats64= ipip6_get_stats64,
+ .ndo_get_stats64 = ip_tunnel_get_stats64,
};
static void ipip6_dev_free(struct net_device *dev)
--
1.7.1
^ permalink raw reply related
* Re: [PATCH net-next 05/22] cxgb4: Add T5 write combining support
From: Casey Leedom @ 2013-03-13 22:54 UTC (permalink / raw)
To: David Laight
Cc: Vipul Pandya, Steve Wise, David Miller, netdev, linux-rdma,
linux-scsi, roland, JBottomley, Dimitrios Michailidis,
Naresh Kumar Inna, Divy Le Ray, Santosh Rastapur, Arvind Bhushan,
Abhishek Agrawal
In-Reply-To: <AE90C24D6B3A694183C094C60CF0A2F6026B718B@saturn3.aculab.com>
On 03/13/13 08:43, David Laight wrote:
> From my recollection of the x86 architecture, the memory barriers are
> hardly ever needed, certainly not in the places where, for example a
> ppc needs them. I'd actually suspect that the normal wmb() for x86
> should be a nop. About the only place where any on the fence
> instructions are needed are in relation to write combining accesses.
> In particular I don't believe they are ever needed to synchronise
> uncached accesses with each other, or with cached accesses (which are
> snooped). David
The question at hand is how should we indicate that we're finished
with a Write Combined set of stores in an architecturally neutral
manner? Is wmb() a good approach? Vipul has noted that the iWarp code
uses a new "wc_wmb()" for this purpose which seems to be the same
_currently_ as wmb() for all current platforms. I presume that the
iWarp folks pick a new name to offer themselves some cover in the future.
And yes, the code sequence that was accidentally included in Vipul's
previous patches should never have been submitted for inclusion in
kernel.org. It got missed in our internal reviews and I apologize for that.
Casey
^ permalink raw reply
* Re: [PATCH net-next 2/4] bridge: Allow an ability to designate an uplink port
From: Stephen Hemminger @ 2013-03-13 22:54 UTC (permalink / raw)
To: Michael S. Tsirkin; +Cc: Vlad Yasevich, netdev, bridge
In-Reply-To: <20130313203333.GC9821@redhat.com>
On Wed, 13 Mar 2013 22:33:33 +0200
"Michael S. Tsirkin" <mst@redhat.com> wrote:
> On Tue, Mar 12, 2013 at 09:45:24PM -0400, Vlad Yasevich wrote:
> > Allow a ports to be designated as uplink. Multiple ports
> > may be designated as uplinks and they will be kept in a
> > list.
> >
> > Signed-off-by: Vlad Yasevich <vyasevic@redhat.com>
>
> It looks like if you make two links uplink, bridging
> between them won't work at all?
Maybe better to mark which ports are terminal ports rather than marking
uplink ports?
^ permalink raw reply
* Re: [PATCH net-next 1/6] GRE: Use RTNL for atomic protocol registration.
From: Stephen Hemminger @ 2013-03-13 22:57 UTC (permalink / raw)
To: Pravin B Shelar; +Cc: davem, netdev, jesse
In-Reply-To: <1363214907-3257-1-git-send-email-pshelar@nicira.com>
On Wed, 13 Mar 2013 15:48:27 -0700
Pravin B Shelar <pshelar@nicira.com> wrote:
> Currently gre demultiplexer uses its own spin-lock. We should
> rather RTNL lock.
>
> Suggested-by: Stephen Hemminger <--cc stephen@networkplumber.org>
> Signed-off-by: Pravin B Shelar <pshelar@nicira.com>
Contradicting myself :-)
This does have added safety but some people eventually want to get rid
of the big network lock. That said, it makes sense that you would need a
supersized lock when playing with multiple devices.
^ permalink raw reply
* pull request: batman-adv 2013-03-13
From: Antonio Quartulli @ 2013-03-13 22:56 UTC (permalink / raw)
To: davem; +Cc: netdev, b.a.t.m.a.n
Hello David,
here you have our first patchset intended for net-next/linux-3.10.
These patches include a new big component implementing a first version of
Network Coding for mesh networks. This code is the result of Martin Hundebøll
and Jeppe Ledet-Pedersen's Master Thesis[1]. The target of Network Coding is to
increase throughput by fusing multiple packets in one wifi transmission (thus
saving airtime).
This new component also introduced a new Kbuild option which can be used to
decide whether to include this feature or not. However, a switch to dynamically
turn off the behaviour at runtime has been provided too. DebugFS has also been
extended with some knobs to be used in case of debugging only.
The rest is clean up work.
Please pull or let me know if there is any problem!
Thank you,
Antonio
[1] http://downloads.open-mesh.org/batman/papers/batman-adv_network_coding.pdf
The following changes since commit 7f02d1601cf05ce79fde78cb9b9bc2e375f87126:
qlcnic: Bump up the version to 5.1.37 (2013-03-13 05:35:05 -0400)
are available in the git repository at:
git://git.open-mesh.org/linux-merge.git tags/batman-adv-for-davem
for you to fetch changes up to 2df5278b0267c799f3e877e8eeddbb6e93cda0bb:
batman-adv: network coding - receive coded packets and decode them (2013-03-13 22:53:51 +0100)
----------------------------------------------------------------
Included changes:
- introduction of the new Network Coding component. This new mechanism aims to
increase throughput by fusing multiple packets in one transmission.
- minor cleanups
----------------------------------------------------------------
Antonio Quartulli (1):
batman-adv: don't use !! in bool conversion
Marek Lindner (1):
batman-adv: replace redundant primary_if_get calls
Martin Hundebøll (7):
batman-adv: Return reason for failure in batadv_check_unicast_packet()
batman-adv: network coding - add the initial infrastructure code
batman-adv: network coding - detect coding nodes and remove these after timeout
batman-adv: network coding - buffer unicast packets before forward
batman-adv: network coding - code and transmit packets if possible
batman-adv: network coding - save overheard and tx packets for decoding
batman-adv: network coding - receive coded packets and decode them
Documentation/ABI/testing/sysfs-class-net-mesh | 8 +
net/batman-adv/Kconfig | 14 +
net/batman-adv/Makefile | 1 +
net/batman-adv/bat_iv_ogm.c | 5 +
net/batman-adv/debugfs.c | 18 +
net/batman-adv/distributed-arp-table.c | 22 +-
net/batman-adv/main.c | 6 +
net/batman-adv/main.h | 12 +-
net/batman-adv/network-coding.c | 1821 ++++++++++++++++++++++++
net/batman-adv/network-coding.h | 123 ++
net/batman-adv/originator.c | 6 +
net/batman-adv/packet.h | 33 +
net/batman-adv/routing.c | 49 +-
net/batman-adv/send.c | 5 +
net/batman-adv/soft-interface.c | 14 +
net/batman-adv/sysfs.c | 16 +-
net/batman-adv/translation-table.c | 29 +-
net/batman-adv/types.h | 136 ++
net/batman-adv/unicast.c | 6 +-
19 files changed, 2259 insertions(+), 65 deletions(-)
create mode 100644 net/batman-adv/network-coding.c
create mode 100644 net/batman-adv/network-coding.h
^ permalink raw reply
* [PATCH 1/9] batman-adv: replace redundant primary_if_get calls
From: Antonio Quartulli @ 2013-03-13 22:56 UTC (permalink / raw)
To: davem; +Cc: netdev, b.a.t.m.a.n, Marek Lindner, Antonio Quartulli
In-Reply-To: <1363215399-18875-1-git-send-email-ordex@autistici.org>
From: Marek Lindner <lindner_marek@yahoo.de>
The batadv_priv struct carries a pointer to its own interface
struct. Therefore, it is not necessary to retrieve the soft_iface
via the primary interface.
Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
net/batman-adv/distributed-arp-table.c | 22 ++++------------------
net/batman-adv/sysfs.c | 10 +---------
net/batman-adv/translation-table.c | 25 ++++++-------------------
3 files changed, 11 insertions(+), 46 deletions(-)
diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c
index d54188a..8e15d96 100644
--- a/net/batman-adv/distributed-arp-table.c
+++ b/net/batman-adv/distributed-arp-table.c
@@ -816,7 +816,6 @@ bool batadv_dat_snoop_outgoing_arp_request(struct batadv_priv *bat_priv,
bool ret = false;
struct batadv_dat_entry *dat_entry = NULL;
struct sk_buff *skb_new;
- struct batadv_hard_iface *primary_if = NULL;
if (!atomic_read(&bat_priv->distributed_arp_table))
goto out;
@@ -838,22 +837,18 @@ bool batadv_dat_snoop_outgoing_arp_request(struct batadv_priv *bat_priv,
dat_entry = batadv_dat_entry_hash_find(bat_priv, ip_dst);
if (dat_entry) {
- primary_if = batadv_primary_if_get_selected(bat_priv);
- if (!primary_if)
- goto out;
-
skb_new = arp_create(ARPOP_REPLY, ETH_P_ARP, ip_src,
- primary_if->soft_iface, ip_dst, hw_src,
+ bat_priv->soft_iface, ip_dst, hw_src,
dat_entry->mac_addr, hw_src);
if (!skb_new)
goto out;
skb_reset_mac_header(skb_new);
skb_new->protocol = eth_type_trans(skb_new,
- primary_if->soft_iface);
+ bat_priv->soft_iface);
bat_priv->stats.rx_packets++;
bat_priv->stats.rx_bytes += skb->len + ETH_HLEN;
- primary_if->soft_iface->last_rx = jiffies;
+ bat_priv->soft_iface->last_rx = jiffies;
netif_rx(skb_new);
batadv_dbg(BATADV_DBG_DAT, bat_priv, "ARP request replied locally\n");
@@ -866,8 +861,6 @@ bool batadv_dat_snoop_outgoing_arp_request(struct batadv_priv *bat_priv,
out:
if (dat_entry)
batadv_dat_entry_free_ref(dat_entry);
- if (primary_if)
- batadv_hardif_free_ref(primary_if);
return ret;
}
@@ -887,7 +880,6 @@ bool batadv_dat_snoop_incoming_arp_request(struct batadv_priv *bat_priv,
__be32 ip_src, ip_dst;
uint8_t *hw_src;
struct sk_buff *skb_new;
- struct batadv_hard_iface *primary_if = NULL;
struct batadv_dat_entry *dat_entry = NULL;
bool ret = false;
int err;
@@ -912,12 +904,8 @@ bool batadv_dat_snoop_incoming_arp_request(struct batadv_priv *bat_priv,
if (!dat_entry)
goto out;
- primary_if = batadv_primary_if_get_selected(bat_priv);
- if (!primary_if)
- goto out;
-
skb_new = arp_create(ARPOP_REPLY, ETH_P_ARP, ip_src,
- primary_if->soft_iface, ip_dst, hw_src,
+ bat_priv->soft_iface, ip_dst, hw_src,
dat_entry->mac_addr, hw_src);
if (!skb_new)
@@ -941,8 +929,6 @@ bool batadv_dat_snoop_incoming_arp_request(struct batadv_priv *bat_priv,
out:
if (dat_entry)
batadv_dat_entry_free_ref(dat_entry);
- if (primary_if)
- batadv_hardif_free_ref(primary_if);
if (ret)
kfree_skb(skb);
return ret;
diff --git a/net/batman-adv/sysfs.c b/net/batman-adv/sysfs.c
index afbba31..6a44fed 100644
--- a/net/batman-adv/sysfs.c
+++ b/net/batman-adv/sysfs.c
@@ -688,15 +688,10 @@ int batadv_throw_uevent(struct batadv_priv *bat_priv, enum batadv_uev_type type,
enum batadv_uev_action action, const char *data)
{
int ret = -ENOMEM;
- struct batadv_hard_iface *primary_if;
struct kobject *bat_kobj;
char *uevent_env[4] = { NULL, NULL, NULL, NULL };
- primary_if = batadv_primary_if_get_selected(bat_priv);
- if (!primary_if)
- goto out;
-
- bat_kobj = &primary_if->soft_iface->dev.kobj;
+ bat_kobj = &bat_priv->soft_iface->dev.kobj;
uevent_env[0] = kmalloc(strlen(BATADV_UEV_TYPE_VAR) +
strlen(batadv_uev_type_str[type]) + 1,
@@ -732,9 +727,6 @@ out:
kfree(uevent_env[1]);
kfree(uevent_env[2]);
- if (primary_if)
- batadv_hardif_free_ref(primary_if);
-
if (ret)
batadv_dbg(BATADV_DBG_BATMAN, bat_priv,
"Impossible to send uevent for (%s,%s,%s) event (err: %d)\n",
diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c
index 98a66a0..7e9e264 100644
--- a/net/batman-adv/translation-table.c
+++ b/net/batman-adv/translation-table.c
@@ -385,25 +385,19 @@ static void batadv_tt_prepare_packet_buff(struct batadv_priv *bat_priv,
int *packet_buff_len,
int min_packet_len)
{
- struct batadv_hard_iface *primary_if;
int req_len;
- primary_if = batadv_primary_if_get_selected(bat_priv);
-
req_len = min_packet_len;
req_len += batadv_tt_len(atomic_read(&bat_priv->tt.local_changes));
/* if we have too many changes for one packet don't send any
* and wait for the tt table request which will be fragmented
*/
- if ((!primary_if) || (req_len > primary_if->soft_iface->mtu))
+ if (req_len > bat_priv->soft_iface->mtu)
req_len = min_packet_len;
batadv_tt_realloc_packet_buff(packet_buff, packet_buff_len,
min_packet_len, req_len);
-
- if (primary_if)
- batadv_hardif_free_ref(primary_if);
}
static int batadv_tt_changes_fill_buff(struct batadv_priv *bat_priv,
@@ -1580,7 +1574,7 @@ static int batadv_tt_global_valid(const void *entry_ptr,
static struct sk_buff *
batadv_tt_response_fill_table(uint16_t tt_len, uint8_t ttvn,
struct batadv_hashtable *hash,
- struct batadv_hard_iface *primary_if,
+ struct batadv_priv *bat_priv,
int (*valid_cb)(const void *, const void *),
void *cb_data)
{
@@ -1594,8 +1588,8 @@ batadv_tt_response_fill_table(uint16_t tt_len, uint8_t ttvn,
uint32_t i;
size_t len;
- if (tt_query_size + tt_len > primary_if->soft_iface->mtu) {
- tt_len = primary_if->soft_iface->mtu - tt_query_size;
+ if (tt_query_size + tt_len > bat_priv->soft_iface->mtu) {
+ tt_len = bat_priv->soft_iface->mtu - tt_query_size;
tt_len -= tt_len % sizeof(struct batadv_tt_change);
}
tt_tot = tt_len / sizeof(struct batadv_tt_change);
@@ -1715,7 +1709,6 @@ batadv_send_other_tt_response(struct batadv_priv *bat_priv,
{
struct batadv_orig_node *req_dst_orig_node;
struct batadv_orig_node *res_dst_orig_node = NULL;
- struct batadv_hard_iface *primary_if = NULL;
uint8_t orig_ttvn, req_ttvn, ttvn;
int ret = false;
unsigned char *tt_buff;
@@ -1740,10 +1733,6 @@ batadv_send_other_tt_response(struct batadv_priv *bat_priv,
if (!res_dst_orig_node)
goto out;
- primary_if = batadv_primary_if_get_selected(bat_priv);
- if (!primary_if)
- goto out;
-
orig_ttvn = (uint8_t)atomic_read(&req_dst_orig_node->last_ttvn);
req_ttvn = tt_request->ttvn;
@@ -1791,7 +1780,7 @@ batadv_send_other_tt_response(struct batadv_priv *bat_priv,
skb = batadv_tt_response_fill_table(tt_len, ttvn,
bat_priv->tt.global_hash,
- primary_if,
+ bat_priv,
batadv_tt_global_valid,
req_dst_orig_node);
if (!skb)
@@ -1828,8 +1817,6 @@ out:
batadv_orig_node_free_ref(res_dst_orig_node);
if (req_dst_orig_node)
batadv_orig_node_free_ref(req_dst_orig_node);
- if (primary_if)
- batadv_hardif_free_ref(primary_if);
if (!ret)
kfree_skb(skb);
return ret;
@@ -1907,7 +1894,7 @@ batadv_send_my_tt_response(struct batadv_priv *bat_priv,
skb = batadv_tt_response_fill_table(tt_len, ttvn,
bat_priv->tt.local_hash,
- primary_if,
+ bat_priv,
batadv_tt_local_valid_entry,
NULL);
if (!skb)
--
1.8.1.5
^ permalink raw reply related
* [PATCH 2/9] batman-adv: Return reason for failure in batadv_check_unicast_packet()
From: Antonio Quartulli @ 2013-03-13 22:56 UTC (permalink / raw)
To: davem
Cc: netdev, b.a.t.m.a.n, Martin Hundebøll, Marek Lindner,
Antonio Quartulli
In-Reply-To: <1363215399-18875-1-git-send-email-ordex@autistici.org>
From: Martin Hundebøll <martin@hundeboll.net>
batadv_check_unicast_packet() is changed to return a value based on the
reason to drop the packet, which will be useful information for
future users of batadv_check_unicast_packet().
Signed-off-by: Martin Hundebøll <martin@hundeboll.net>
Acked-by: Antonio Quartulli <ordex@autistici.org>
Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
net/batman-adv/routing.c | 18 ++++++++++++++----
1 file changed, 14 insertions(+), 4 deletions(-)
diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c
index 5ee21ce..322c97a 100644
--- a/net/batman-adv/routing.c
+++ b/net/batman-adv/routing.c
@@ -548,27 +548,37 @@ batadv_find_ifalter_router(struct batadv_orig_node *primary_orig,
return router;
}
+/**
+ * batadv_check_unicast_packet - Check for malformed unicast packets
+ * @skb: packet to check
+ * @hdr_size: size of header to pull
+ *
+ * Check for short header and bad addresses in given packet. Returns negative
+ * value when check fails and 0 otherwise. The negative value depends on the
+ * reason: -ENODATA for bad header, -EBADR for broadcast destination or source,
+ * and -EREMOTE for non-local (other host) destination.
+ */
static int batadv_check_unicast_packet(struct sk_buff *skb, int hdr_size)
{
struct ethhdr *ethhdr;
/* drop packet if it has not necessary minimum size */
if (unlikely(!pskb_may_pull(skb, hdr_size)))
- return -1;
+ return -ENODATA;
ethhdr = (struct ethhdr *)skb_mac_header(skb);
/* packet with unicast indication but broadcast recipient */
if (is_broadcast_ether_addr(ethhdr->h_dest))
- return -1;
+ return -EBADR;
/* packet with broadcast sender address */
if (is_broadcast_ether_addr(ethhdr->h_source))
- return -1;
+ return -EBADR;
/* not for me */
if (!batadv_is_my_mac(ethhdr->h_dest))
- return -1;
+ return -EREMOTE;
return 0;
}
--
1.8.1.5
^ permalink raw reply related
* [PATCH 3/9] batman-adv: don't use !! in bool conversion
From: Antonio Quartulli @ 2013-03-13 22:56 UTC (permalink / raw)
To: davem; +Cc: netdev, b.a.t.m.a.n, Antonio Quartulli, Marek Lindner
In-Reply-To: <1363215399-18875-1-git-send-email-ordex@autistici.org>
In C standard any expression different from 0 will be converted to
'true' when casting to bool (whatever is the length of the value).
Therefore all the "!!" conversions can be removed.
Signed-off-by: Antonio Quartulli <ordex@autistici.org>
Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
---
net/batman-adv/translation-table.c | 4 ++--
net/batman-adv/unicast.c | 6 +++---
2 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c
index 7e9e264..9322320 100644
--- a/net/batman-adv/translation-table.c
+++ b/net/batman-adv/translation-table.c
@@ -902,7 +902,7 @@ out_remove:
/* remove address from local hash if present */
local_flags = batadv_tt_local_remove(bat_priv, tt_addr,
"global tt received",
- !!(flags & BATADV_TT_CLIENT_ROAM));
+ flags & BATADV_TT_CLIENT_ROAM);
tt_global_entry->common.flags |= local_flags & BATADV_TT_CLIENT_WIFI;
if (!(flags & BATADV_TT_CLIENT_ROAM))
@@ -2515,7 +2515,7 @@ bool batadv_tt_global_client_is_roaming(struct batadv_priv *bat_priv,
if (!tt_global_entry)
goto out;
- ret = !!(tt_global_entry->common.flags & BATADV_TT_CLIENT_ROAM);
+ ret = tt_global_entry->common.flags & BATADV_TT_CLIENT_ROAM;
batadv_tt_global_entry_free_ref(tt_global_entry);
out:
return ret;
diff --git a/net/batman-adv/unicast.c b/net/batman-adv/unicast.c
index 50e079f..0bb3b59 100644
--- a/net/batman-adv/unicast.c
+++ b/net/batman-adv/unicast.c
@@ -122,7 +122,7 @@ batadv_frag_search_packet(struct list_head *head,
{
struct batadv_frag_packet_list_entry *tfp;
struct batadv_unicast_frag_packet *tmp_up = NULL;
- int is_head_tmp, is_head;
+ bool is_head_tmp, is_head;
uint16_t search_seqno;
if (up->flags & BATADV_UNI_FRAG_HEAD)
@@ -130,7 +130,7 @@ batadv_frag_search_packet(struct list_head *head,
else
search_seqno = ntohs(up->seqno)-1;
- is_head = !!(up->flags & BATADV_UNI_FRAG_HEAD);
+ is_head = up->flags & BATADV_UNI_FRAG_HEAD;
list_for_each_entry(tfp, head, list) {
if (!tfp->skb)
@@ -142,7 +142,7 @@ batadv_frag_search_packet(struct list_head *head,
tmp_up = (struct batadv_unicast_frag_packet *)tfp->skb->data;
if (tfp->seqno == search_seqno) {
- is_head_tmp = !!(tmp_up->flags & BATADV_UNI_FRAG_HEAD);
+ is_head_tmp = tmp_up->flags & BATADV_UNI_FRAG_HEAD;
if (is_head_tmp != is_head)
return tfp;
else
--
1.8.1.5
^ permalink raw reply related
* [PATCH 4/9] batman-adv: network coding - add the initial infrastructure code
From: Antonio Quartulli @ 2013-03-13 22:56 UTC (permalink / raw)
To: davem
Cc: netdev, b.a.t.m.a.n, Martin Hundebøll, Marek Lindner,
Antonio Quartulli
In-Reply-To: <1363215399-18875-1-git-send-email-ordex@autistici.org>
From: Martin Hundebøll <martin@hundeboll.net>
Network coding exploits the 802.11 shared medium to allow multiple
packets to be sent in a single transmission. In brief, a relay can XOR
two packets, and send the coded packet to two destinations. The
receivers can decode one of the original packets by XOR'ing the coded
packet with the other original packet. This will lead to increased
throughput in topologies where two packets cross one relay.
In a simple topology with three nodes, it takes four transmissions
without network coding to get one packet from Node A to Node B and one
from Node B to Node A:
1. Node A ---- p1 ---> Node R Node B
2. Node A Node R <--- p2 ---- Node B
3. Node A <--- p2 ---- Node R Node B
4. Node A Node R ---- p1 ---> Node B
With network coding, the relay only needs one transmission, which saves
us one slot of valuable airtime:
1. Node A ---- p1 ---> Node R Node B
2. Node A Node R <--- p2 ---- Node B
3. Node A <- p1 x p2 - Node R - p1 x p2 -> Node B
The same principle holds for a topology including five nodes. Here the
packets from Node A and Node B are overheard by Node C and Node D,
respectively. This allows Node R to send a network coded packet to save
one transmission:
Node A Node B
| \ / |
| p1 p2 |
| \ / |
p1 > Node R < p2
| |
| / \ |
| p1 x p2 p1 x p2 |
v / \ v
/ \
Node C < > Node D
More information is available on the open-mesh.org wiki[1].
This patch adds the initial code to support network coding in
batman-adv. It sets up a worker thread to do house keeping and adds a
sysfs file to enable/disable network coding. The feature is disabled by
default, as it requires a wifi-driver with working promiscuous mode, and
also because it adds a small delay at each hop.
[1] http://www.open-mesh.org/projects/batman-adv/wiki/Catwoman
Signed-off-by: Martin Hundebøll <martin@hundeboll.net>
Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
Documentation/ABI/testing/sysfs-class-net-mesh | 8 +++
net/batman-adv/Kconfig | 14 +++++
net/batman-adv/Makefile | 1 +
net/batman-adv/main.c | 6 ++
net/batman-adv/main.h | 4 +-
net/batman-adv/network-coding.c | 81 ++++++++++++++++++++++++++
net/batman-adv/network-coding.h | 48 +++++++++++++++
net/batman-adv/soft-interface.c | 3 +
net/batman-adv/sysfs.c | 6 ++
net/batman-adv/types.h | 14 +++++
10 files changed, 184 insertions(+), 1 deletion(-)
create mode 100644 net/batman-adv/network-coding.c
create mode 100644 net/batman-adv/network-coding.h
diff --git a/Documentation/ABI/testing/sysfs-class-net-mesh b/Documentation/ABI/testing/sysfs-class-net-mesh
index bc41da6..bdcd8b4 100644
--- a/Documentation/ABI/testing/sysfs-class-net-mesh
+++ b/Documentation/ABI/testing/sysfs-class-net-mesh
@@ -67,6 +67,14 @@ Description:
Defines the penalty which will be applied to an
originator message's tq-field on every hop.
+What: /sys/class/net/<mesh_iface>/mesh/network_coding
+Date: Nov 2012
+Contact: Martin Hundeboll <martin@hundeboll.net>
+Description:
+ Controls whether Network Coding (using some magic
+ to send fewer wifi packets but still the same
+ content) is enabled or not.
+
What: /sys/class/net/<mesh_iface>/mesh/orig_interval
Date: May 2010
Contact: Marek Lindner <lindner_marek@yahoo.de>
diff --git a/net/batman-adv/Kconfig b/net/batman-adv/Kconfig
index 8d8afb1..fa780b7 100644
--- a/net/batman-adv/Kconfig
+++ b/net/batman-adv/Kconfig
@@ -36,6 +36,20 @@ config BATMAN_ADV_DAT
mesh networks. If you think that your network does not need
this option you can safely remove it and save some space.
+config BATMAN_ADV_NC
+ bool "Network Coding"
+ depends on BATMAN_ADV
+ default n
+ help
+ This option enables network coding, a mechanism that aims to
+ increase the overall network throughput by fusing multiple
+ packets in one transmission.
+ Note that interfaces controlled by batman-adv must be manually
+ configured to have promiscuous mode enabled in order to make
+ network coding work.
+ If you think that your network does not need this feature you
+ can safely disable it and save some space.
+
config BATMAN_ADV_DEBUG
bool "B.A.T.M.A.N. debugging"
depends on BATMAN_ADV
diff --git a/net/batman-adv/Makefile b/net/batman-adv/Makefile
index e45e3b4..4b8f192 100644
--- a/net/batman-adv/Makefile
+++ b/net/batman-adv/Makefile
@@ -30,6 +30,7 @@ batman-adv-y += hard-interface.o
batman-adv-y += hash.o
batman-adv-y += icmp_socket.o
batman-adv-y += main.o
+batman-adv-$(CONFIG_BATMAN_ADV_NC) += network-coding.o
batman-adv-y += originator.o
batman-adv-y += ring_buffer.o
batman-adv-y += routing.o
diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c
index 0488d70..0495a7d 100644
--- a/net/batman-adv/main.c
+++ b/net/batman-adv/main.c
@@ -35,6 +35,7 @@
#include "vis.h"
#include "hash.h"
#include "bat_algo.h"
+#include "network-coding.h"
/* List manipulations on hardif_list have to be rtnl_lock()'ed,
@@ -135,6 +136,10 @@ int batadv_mesh_init(struct net_device *soft_iface)
if (ret < 0)
goto err;
+ ret = batadv_nc_init(bat_priv);
+ if (ret < 0)
+ goto err;
+
atomic_set(&bat_priv->gw.reselect, 0);
atomic_set(&bat_priv->mesh_state, BATADV_MESH_ACTIVE);
@@ -157,6 +162,7 @@ void batadv_mesh_free(struct net_device *soft_iface)
batadv_gw_node_purge(bat_priv);
batadv_originator_free(bat_priv);
+ batadv_nc_free(bat_priv);
batadv_tt_free(bat_priv);
diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index ced08b9..59ba2ff 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -185,6 +185,7 @@ __be32 batadv_skb_crc32(struct sk_buff *skb, u8 *payload_ptr);
* @BATADV_DBG_TT: translation table messages
* @BATADV_DBG_BLA: bridge loop avoidance messages
* @BATADV_DBG_DAT: ARP snooping and DAT related messages
+ * @BATADV_DBG_NC: network coding related messages
* @BATADV_DBG_ALL: the union of all the above log levels
*/
enum batadv_dbg_level {
@@ -193,7 +194,8 @@ enum batadv_dbg_level {
BATADV_DBG_TT = BIT(2),
BATADV_DBG_BLA = BIT(3),
BATADV_DBG_DAT = BIT(4),
- BATADV_DBG_ALL = 31,
+ BATADV_DBG_NC = BIT(5),
+ BATADV_DBG_ALL = 63,
};
#ifdef CONFIG_BATMAN_ADV_DEBUG
diff --git a/net/batman-adv/network-coding.c b/net/batman-adv/network-coding.c
new file mode 100644
index 0000000..9c2d54b
--- /dev/null
+++ b/net/batman-adv/network-coding.c
@@ -0,0 +1,81 @@
+/* Copyright (C) 2012-2013 B.A.T.M.A.N. contributors:
+ *
+ * Martin Hundebøll, Jeppe Ledet-Pedersen
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA
+ */
+
+#include "main.h"
+#include "network-coding.h"
+
+static void batadv_nc_worker(struct work_struct *work);
+
+/**
+ * batadv_nc_start_timer - initialise the nc periodic worker
+ * @bat_priv: the bat priv with all the soft interface information
+ */
+static void batadv_nc_start_timer(struct batadv_priv *bat_priv)
+{
+ queue_delayed_work(batadv_event_workqueue, &bat_priv->nc.work,
+ msecs_to_jiffies(10));
+}
+
+/**
+ * batadv_nc_init - initialise coding hash table and start house keeping
+ * @bat_priv: the bat priv with all the soft interface information
+ */
+int batadv_nc_init(struct batadv_priv *bat_priv)
+{
+ INIT_DELAYED_WORK(&bat_priv->nc.work, batadv_nc_worker);
+ batadv_nc_start_timer(bat_priv);
+
+ return 0;
+}
+
+/**
+ * batadv_nc_init_bat_priv - initialise the nc specific bat_priv variables
+ * @bat_priv: the bat priv with all the soft interface information
+ */
+void batadv_nc_init_bat_priv(struct batadv_priv *bat_priv)
+{
+ atomic_set(&bat_priv->network_coding, 1);
+}
+
+/**
+ * batadv_nc_worker - periodic task for house keeping related to network coding
+ * @work: kernel work struct
+ */
+static void batadv_nc_worker(struct work_struct *work)
+{
+ struct delayed_work *delayed_work;
+ struct batadv_priv_nc *priv_nc;
+ struct batadv_priv *bat_priv;
+
+ delayed_work = container_of(work, struct delayed_work, work);
+ priv_nc = container_of(delayed_work, struct batadv_priv_nc, work);
+ bat_priv = container_of(priv_nc, struct batadv_priv, nc);
+
+ /* Schedule a new check */
+ batadv_nc_start_timer(bat_priv);
+}
+
+/**
+ * batadv_nc_free - clean up network coding memory
+ * @bat_priv: the bat priv with all the soft interface information
+ */
+void batadv_nc_free(struct batadv_priv *bat_priv)
+{
+ cancel_delayed_work_sync(&bat_priv->nc.work);
+}
diff --git a/net/batman-adv/network-coding.h b/net/batman-adv/network-coding.h
new file mode 100644
index 0000000..7483cba
--- /dev/null
+++ b/net/batman-adv/network-coding.h
@@ -0,0 +1,48 @@
+/* Copyright (C) 2012-2013 B.A.T.M.A.N. contributors:
+ *
+ * Martin Hundebøll, Jeppe Ledet-Pedersen
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA
+ */
+
+#ifndef _NET_BATMAN_ADV_NETWORK_CODING_H_
+#define _NET_BATMAN_ADV_NETWORK_CODING_H_
+
+#ifdef CONFIG_BATMAN_ADV_NC
+
+int batadv_nc_init(struct batadv_priv *bat_priv);
+void batadv_nc_free(struct batadv_priv *bat_priv);
+void batadv_nc_init_bat_priv(struct batadv_priv *bat_priv);
+
+#else /* ifdef CONFIG_BATMAN_ADV_NC */
+
+static inline int batadv_nc_init(struct batadv_priv *bat_priv)
+{
+ return 0;
+}
+
+static inline void batadv_nc_free(struct batadv_priv *bat_priv)
+{
+ return;
+}
+
+static inline void batadv_nc_init_bat_priv(struct batadv_priv *bat_priv)
+{
+ return;
+}
+
+#endif /* ifdef CONFIG_BATMAN_ADV_NC */
+
+#endif /* _NET_BATMAN_ADV_NETWORK_CODING_H_ */
diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c
index 2711e87..7188e07 100644
--- a/net/batman-adv/soft-interface.c
+++ b/net/batman-adv/soft-interface.c
@@ -37,6 +37,7 @@
#include <linux/if_ether.h>
#include "unicast.h"
#include "bridge_loop_avoidance.h"
+#include "network-coding.h"
static int batadv_get_settings(struct net_device *dev, struct ethtool_cmd *cmd);
@@ -544,6 +545,8 @@ struct net_device *batadv_softif_create(const char *name)
if (ret < 0)
goto unreg_soft_iface;
+ batadv_nc_init_bat_priv(bat_priv);
+
ret = batadv_sysfs_add_meshif(soft_iface);
if (ret < 0)
goto unreg_soft_iface;
diff --git a/net/batman-adv/sysfs.c b/net/batman-adv/sysfs.c
index 6a44fed..ce39f62 100644
--- a/net/batman-adv/sysfs.c
+++ b/net/batman-adv/sysfs.c
@@ -442,6 +442,9 @@ static BATADV_ATTR(gw_bandwidth, S_IRUGO | S_IWUSR, batadv_show_gw_bwidth,
#ifdef CONFIG_BATMAN_ADV_DEBUG
BATADV_ATTR_SIF_UINT(log_level, S_IRUGO | S_IWUSR, 0, BATADV_DBG_ALL, NULL);
#endif
+#ifdef CONFIG_BATMAN_ADV_NC
+BATADV_ATTR_SIF_BOOL(network_coding, S_IRUGO | S_IWUSR, NULL);
+#endif
static struct batadv_attribute *batadv_mesh_attrs[] = {
&batadv_attr_aggregated_ogms,
@@ -464,6 +467,9 @@ static struct batadv_attribute *batadv_mesh_attrs[] = {
#ifdef CONFIG_BATMAN_ADV_DEBUG
&batadv_attr_log_level,
#endif
+#ifdef CONFIG_BATMAN_ADV_NC
+ &batadv_attr_network_coding,
+#endif
NULL,
};
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 4cd87a0..83bfe7c 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -428,6 +428,14 @@ struct batadv_priv_dat {
#endif
/**
+ * struct batadv_priv_nc - per mesh interface network coding private data
+ * @work: work queue callback item for cleanup
+ */
+struct batadv_priv_nc {
+ struct delayed_work work;
+};
+
+/**
* struct batadv_priv - per mesh interface data
* @mesh_state: current status of the mesh (inactive/active/deactivating)
* @soft_iface: net device which holds this struct as private data
@@ -470,6 +478,8 @@ struct batadv_priv_dat {
* @tt: translation table data
* @vis: vis data
* @dat: distributed arp table data
+ * @network_coding: bool indicating whether network coding is enabled
+ * @batadv_priv_nc: network coding data
*/
struct batadv_priv {
atomic_t mesh_state;
@@ -522,6 +532,10 @@ struct batadv_priv {
#ifdef CONFIG_BATMAN_ADV_DAT
struct batadv_priv_dat dat;
#endif
+#ifdef CONFIG_BATMAN_ADV_NC
+ atomic_t network_coding;
+ struct batadv_priv_nc nc;
+#endif /* CONFIG_BATMAN_ADV_NC */
};
/**
--
1.8.1.5
^ permalink raw reply related
* [PATCH 5/9] batman-adv: network coding - detect coding nodes and remove these after timeout
From: Antonio Quartulli @ 2013-03-13 22:56 UTC (permalink / raw)
To: davem
Cc: netdev, b.a.t.m.a.n, Martin Hundebøll, Marek Lindner,
Antonio Quartulli
In-Reply-To: <1363215399-18875-1-git-send-email-ordex@autistici.org>
From: Martin Hundebøll <martin@hundeboll.net>
To use network coding efficiently, a relay must know when neighbor nodes
are likely to have enough information to be able to decode a network
coded packet. This is detected by using OGMs from batman-adv to discover
when one neighbor is in range of another neighbor. The relay check the
TLL to detect when an OGM is forwarded from one neighbor by another
neighbor, and thereby knows that the two neighbors are in range and thus
overhear packets sent by each other.
This information is saved in the orig_node struct to be used when
searching for coding opportunities. Two lists are added to the
orig_node struct: One for neighbors that can hear the orig_node
(outgoing nc_nodes) and one for neighbors that the orig_node can hear
(incoming nc_nodes).
Information about nc_nodes is kept for 10 seconds and is available
through debugfs in batman_adv/nc_nodes to use when debugging network
coding.
Signed-off-by: Martin Hundebøll <martin@hundeboll.net>
Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
net/batman-adv/bat_iv_ogm.c | 5 +
net/batman-adv/debugfs.c | 18 ++
net/batman-adv/main.h | 2 +
net/batman-adv/network-coding.c | 414 ++++++++++++++++++++++++++++++++++++++++
net/batman-adv/network-coding.h | 47 +++++
net/batman-adv/originator.c | 6 +
net/batman-adv/types.h | 32 ++++
7 files changed, 524 insertions(+)
diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c
index a5bb0a7..071f288 100644
--- a/net/batman-adv/bat_iv_ogm.c
+++ b/net/batman-adv/bat_iv_ogm.c
@@ -27,6 +27,7 @@
#include "hard-interface.h"
#include "send.h"
#include "bat_algo.h"
+#include "network-coding.h"
static struct batadv_neigh_node *
batadv_iv_ogm_neigh_new(struct batadv_hard_iface *hard_iface,
@@ -1185,6 +1186,10 @@ static void batadv_iv_ogm_process(const struct ethhdr *ethhdr,
if (!orig_neigh_node)
goto out;
+ /* Update nc_nodes of the originator */
+ batadv_nc_update_nc_node(bat_priv, orig_node, orig_neigh_node,
+ batadv_ogm_packet, is_single_hop_neigh);
+
orig_neigh_router = batadv_orig_node_get_router(orig_neigh_node);
/* drop packet if sender is not a direct neighbor and if we
diff --git a/net/batman-adv/debugfs.c b/net/batman-adv/debugfs.c
index 6ae8651..f186a55 100644
--- a/net/batman-adv/debugfs.c
+++ b/net/batman-adv/debugfs.c
@@ -32,6 +32,7 @@
#include "icmp_socket.h"
#include "bridge_loop_avoidance.h"
#include "distributed-arp-table.h"
+#include "network-coding.h"
static struct dentry *batadv_debugfs;
@@ -310,6 +311,14 @@ struct batadv_debuginfo {
const struct file_operations fops;
};
+#ifdef CONFIG_BATMAN_ADV_NC
+static int batadv_nc_nodes_open(struct inode *inode, struct file *file)
+{
+ struct net_device *net_dev = (struct net_device *)inode->i_private;
+ return single_open(file, batadv_nc_nodes_seq_print_text, net_dev);
+}
+#endif
+
#define BATADV_DEBUGINFO(_name, _mode, _open) \
struct batadv_debuginfo batadv_debuginfo_##_name = { \
.attr = { .name = __stringify(_name), \
@@ -348,6 +357,9 @@ static BATADV_DEBUGINFO(dat_cache, S_IRUGO, batadv_dat_cache_open);
static BATADV_DEBUGINFO(transtable_local, S_IRUGO,
batadv_transtable_local_open);
static BATADV_DEBUGINFO(vis_data, S_IRUGO, batadv_vis_data_open);
+#ifdef CONFIG_BATMAN_ADV_NC
+static BATADV_DEBUGINFO(nc_nodes, S_IRUGO, batadv_nc_nodes_open);
+#endif
static struct batadv_debuginfo *batadv_mesh_debuginfos[] = {
&batadv_debuginfo_originators,
@@ -362,6 +374,9 @@ static struct batadv_debuginfo *batadv_mesh_debuginfos[] = {
#endif
&batadv_debuginfo_transtable_local,
&batadv_debuginfo_vis_data,
+#ifdef CONFIG_BATMAN_ADV_NC
+ &batadv_debuginfo_nc_nodes,
+#endif
NULL,
};
@@ -431,6 +446,9 @@ int batadv_debugfs_add_meshif(struct net_device *dev)
}
}
+ if (batadv_nc_init_debugfs(bat_priv) < 0)
+ goto rem_attr;
+
return 0;
rem_attr:
debugfs_remove_recursive(bat_priv->debug_dir);
diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index 59ba2ff..8a10619 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -105,6 +105,8 @@
#define BATADV_RESET_PROTECTION_MS 30000
#define BATADV_EXPECTED_SEQNO_RANGE 65536
+#define BATADV_NC_NODE_TIMEOUT 10000 /* Milliseconds */
+
enum batadv_mesh_state {
BATADV_MESH_INACTIVE,
BATADV_MESH_ACTIVE,
diff --git a/net/batman-adv/network-coding.c b/net/batman-adv/network-coding.c
index 9c2d54b..ff4985d 100644
--- a/net/batman-adv/network-coding.c
+++ b/net/batman-adv/network-coding.c
@@ -17,8 +17,12 @@
* 02110-1301, USA
*/
+#include <linux/debugfs.h>
+
#include "main.h"
#include "network-coding.h"
+#include "originator.h"
+#include "hard-interface.h"
static void batadv_nc_worker(struct work_struct *work);
@@ -51,6 +55,151 @@ int batadv_nc_init(struct batadv_priv *bat_priv)
void batadv_nc_init_bat_priv(struct batadv_priv *bat_priv)
{
atomic_set(&bat_priv->network_coding, 1);
+ bat_priv->nc.min_tq = 200;
+}
+
+/**
+ * batadv_nc_init_orig - initialise the nc fields of an orig_node
+ * @orig_node: the orig_node which is going to be initialised
+ */
+void batadv_nc_init_orig(struct batadv_orig_node *orig_node)
+{
+ INIT_LIST_HEAD(&orig_node->in_coding_list);
+ INIT_LIST_HEAD(&orig_node->out_coding_list);
+ spin_lock_init(&orig_node->in_coding_list_lock);
+ spin_lock_init(&orig_node->out_coding_list_lock);
+}
+
+/**
+ * batadv_nc_node_free_rcu - rcu callback to free an nc node and remove
+ * its refcount on the orig_node
+ * @rcu: rcu pointer of the nc node
+ */
+static void batadv_nc_node_free_rcu(struct rcu_head *rcu)
+{
+ struct batadv_nc_node *nc_node;
+
+ nc_node = container_of(rcu, struct batadv_nc_node, rcu);
+ batadv_orig_node_free_ref(nc_node->orig_node);
+ kfree(nc_node);
+}
+
+/**
+ * batadv_nc_node_free_ref - decrements the nc node refcounter and possibly
+ * frees it
+ * @nc_node: the nc node to free
+ */
+static void batadv_nc_node_free_ref(struct batadv_nc_node *nc_node)
+{
+ if (atomic_dec_and_test(&nc_node->refcount))
+ call_rcu(&nc_node->rcu, batadv_nc_node_free_rcu);
+}
+
+/**
+ * batadv_nc_to_purge_nc_node - checks whether an nc node has to be purged
+ * @bat_priv: the bat priv with all the soft interface information
+ * @nc_node: the nc node to check
+ *
+ * Returns true if the entry has to be purged now, false otherwise
+ */
+static bool batadv_nc_to_purge_nc_node(struct batadv_priv *bat_priv,
+ struct batadv_nc_node *nc_node)
+{
+ if (atomic_read(&bat_priv->mesh_state) != BATADV_MESH_ACTIVE)
+ return true;
+
+ return batadv_has_timed_out(nc_node->last_seen, BATADV_NC_NODE_TIMEOUT);
+}
+
+/**
+ * batadv_nc_purge_orig_nc_nodes - go through list of nc nodes and purge stale
+ * entries
+ * @bat_priv: the bat priv with all the soft interface information
+ * @list: list of nc nodes
+ * @lock: nc node list lock
+ * @to_purge: function in charge to decide whether an entry has to be purged or
+ * not. This function takes the nc node as argument and has to return
+ * a boolean value: true if the entry has to be deleted, false
+ * otherwise
+ */
+static void
+batadv_nc_purge_orig_nc_nodes(struct batadv_priv *bat_priv,
+ struct list_head *list,
+ spinlock_t *lock,
+ bool (*to_purge)(struct batadv_priv *,
+ struct batadv_nc_node *))
+{
+ struct batadv_nc_node *nc_node, *nc_node_tmp;
+
+ /* For each nc_node in list */
+ spin_lock_bh(lock);
+ list_for_each_entry_safe(nc_node, nc_node_tmp, list, list) {
+ /* if an helper function has been passed as parameter,
+ * ask it if the entry has to be purged or not
+ */
+ if (to_purge && !to_purge(bat_priv, nc_node))
+ continue;
+
+ batadv_dbg(BATADV_DBG_NC, bat_priv,
+ "Removing nc_node %pM -> %pM\n",
+ nc_node->addr, nc_node->orig_node->orig);
+ list_del_rcu(&nc_node->list);
+ batadv_nc_node_free_ref(nc_node);
+ }
+ spin_unlock_bh(lock);
+}
+
+/**
+ * batadv_nc_purge_orig - purges all nc node data attached of the given
+ * originator
+ * @bat_priv: the bat priv with all the soft interface information
+ * @orig_node: orig_node with the nc node entries to be purged
+ * @to_purge: function in charge to decide whether an entry has to be purged or
+ * not. This function takes the nc node as argument and has to return
+ * a boolean value: true is the entry has to be deleted, false
+ * otherwise
+ */
+void batadv_nc_purge_orig(struct batadv_priv *bat_priv,
+ struct batadv_orig_node *orig_node,
+ bool (*to_purge)(struct batadv_priv *,
+ struct batadv_nc_node *))
+{
+ /* Check ingoing nc_node's of this orig_node */
+ batadv_nc_purge_orig_nc_nodes(bat_priv, &orig_node->in_coding_list,
+ &orig_node->in_coding_list_lock,
+ to_purge);
+
+ /* Check outgoing nc_node's of this orig_node */
+ batadv_nc_purge_orig_nc_nodes(bat_priv, &orig_node->out_coding_list,
+ &orig_node->out_coding_list_lock,
+ to_purge);
+}
+
+/**
+ * batadv_nc_purge_orig_hash - traverse entire originator hash to check if they
+ * have timed out nc nodes
+ * @bat_priv: the bat priv with all the soft interface information
+ */
+static void batadv_nc_purge_orig_hash(struct batadv_priv *bat_priv)
+{
+ struct batadv_hashtable *hash = bat_priv->orig_hash;
+ struct hlist_head *head;
+ struct batadv_orig_node *orig_node;
+ uint32_t i;
+
+ if (!hash)
+ return;
+
+ /* For each orig_node */
+ for (i = 0; i < hash->size; i++) {
+ head = &hash->table[i];
+
+ rcu_read_lock();
+ hlist_for_each_entry_rcu(orig_node, head, hash_entry)
+ batadv_nc_purge_orig(bat_priv, orig_node,
+ batadv_nc_to_purge_nc_node);
+ rcu_read_unlock();
+ }
}
/**
@@ -67,11 +216,197 @@ static void batadv_nc_worker(struct work_struct *work)
priv_nc = container_of(delayed_work, struct batadv_priv_nc, work);
bat_priv = container_of(priv_nc, struct batadv_priv, nc);
+ batadv_nc_purge_orig_hash(bat_priv);
+
/* Schedule a new check */
batadv_nc_start_timer(bat_priv);
}
/**
+ * batadv_can_nc_with_orig - checks whether the given orig node is suitable for
+ * coding or not
+ * @bat_priv: the bat priv with all the soft interface information
+ * @orig_node: neighboring orig node which may be used as nc candidate
+ * @ogm_packet: incoming ogm packet also used for the checks
+ *
+ * Returns true if:
+ * 1) The OGM must have the most recent sequence number.
+ * 2) The TTL must be decremented by one and only one.
+ * 3) The OGM must be received from the first hop from orig_node.
+ * 4) The TQ value of the OGM must be above bat_priv->nc.min_tq.
+ */
+static bool batadv_can_nc_with_orig(struct batadv_priv *bat_priv,
+ struct batadv_orig_node *orig_node,
+ struct batadv_ogm_packet *ogm_packet)
+{
+ if (orig_node->last_real_seqno != ogm_packet->seqno)
+ return false;
+ if (orig_node->last_ttl != ogm_packet->header.ttl + 1)
+ return false;
+ if (!batadv_compare_eth(ogm_packet->orig, ogm_packet->prev_sender))
+ return false;
+ if (ogm_packet->tq < bat_priv->nc.min_tq)
+ return false;
+
+ return true;
+}
+
+/**
+ * batadv_nc_find_nc_node - search for an existing nc node and return it
+ * @orig_node: orig node originating the ogm packet
+ * @orig_neigh_node: neighboring orig node from which we received the ogm packet
+ * (can be equal to orig_node)
+ * @in_coding: traverse incoming or outgoing network coding list
+ *
+ * Returns the nc_node if found, NULL otherwise.
+ */
+static struct batadv_nc_node
+*batadv_nc_find_nc_node(struct batadv_orig_node *orig_node,
+ struct batadv_orig_node *orig_neigh_node,
+ bool in_coding)
+{
+ struct batadv_nc_node *nc_node, *nc_node_out = NULL;
+ struct list_head *list;
+
+ if (in_coding)
+ list = &orig_neigh_node->in_coding_list;
+ else
+ list = &orig_neigh_node->out_coding_list;
+
+ /* Traverse list of nc_nodes to orig_node */
+ rcu_read_lock();
+ list_for_each_entry_rcu(nc_node, list, list) {
+ if (!batadv_compare_eth(nc_node->addr, orig_node->orig))
+ continue;
+
+ if (!atomic_inc_not_zero(&nc_node->refcount))
+ continue;
+
+ /* Found a match */
+ nc_node_out = nc_node;
+ break;
+ }
+ rcu_read_unlock();
+
+ return nc_node_out;
+}
+
+/**
+ * batadv_nc_get_nc_node - retrieves an nc node or creates the entry if it was
+ * not found
+ * @bat_priv: the bat priv with all the soft interface information
+ * @orig_node: orig node originating the ogm packet
+ * @orig_neigh_node: neighboring orig node from which we received the ogm packet
+ * (can be equal to orig_node)
+ * @in_coding: traverse incoming or outgoing network coding list
+ *
+ * Returns the nc_node if found or created, NULL in case of an error.
+ */
+static struct batadv_nc_node
+*batadv_nc_get_nc_node(struct batadv_priv *bat_priv,
+ struct batadv_orig_node *orig_node,
+ struct batadv_orig_node *orig_neigh_node,
+ bool in_coding)
+{
+ struct batadv_nc_node *nc_node;
+ spinlock_t *lock; /* Used to lock list selected by "int in_coding" */
+ struct list_head *list;
+
+ /* Check if nc_node is already added */
+ nc_node = batadv_nc_find_nc_node(orig_node, orig_neigh_node, in_coding);
+
+ /* Node found */
+ if (nc_node)
+ return nc_node;
+
+ nc_node = kzalloc(sizeof(*nc_node), GFP_ATOMIC);
+ if (!nc_node)
+ return NULL;
+
+ if (!atomic_inc_not_zero(&orig_neigh_node->refcount))
+ goto free;
+
+ /* Initialize nc_node */
+ INIT_LIST_HEAD(&nc_node->list);
+ memcpy(nc_node->addr, orig_node->orig, ETH_ALEN);
+ nc_node->orig_node = orig_neigh_node;
+ atomic_set(&nc_node->refcount, 2);
+
+ /* Select ingoing or outgoing coding node */
+ if (in_coding) {
+ lock = &orig_neigh_node->in_coding_list_lock;
+ list = &orig_neigh_node->in_coding_list;
+ } else {
+ lock = &orig_neigh_node->out_coding_list_lock;
+ list = &orig_neigh_node->out_coding_list;
+ }
+
+ batadv_dbg(BATADV_DBG_NC, bat_priv, "Adding nc_node %pM -> %pM\n",
+ nc_node->addr, nc_node->orig_node->orig);
+
+ /* Add nc_node to orig_node */
+ spin_lock_bh(lock);
+ list_add_tail_rcu(&nc_node->list, list);
+ spin_unlock_bh(lock);
+
+ return nc_node;
+
+free:
+ kfree(nc_node);
+ return NULL;
+}
+
+/**
+ * batadv_nc_update_nc_node - updates stored incoming and outgoing nc node structs
+ * (best called on incoming OGMs)
+ * @bat_priv: the bat priv with all the soft interface information
+ * @orig_node: orig node originating the ogm packet
+ * @orig_neigh_node: neighboring orig node from which we received the ogm packet
+ * (can be equal to orig_node)
+ * @ogm_packet: incoming ogm packet
+ * @is_single_hop_neigh: orig_node is a single hop neighbor
+ */
+void batadv_nc_update_nc_node(struct batadv_priv *bat_priv,
+ struct batadv_orig_node *orig_node,
+ struct batadv_orig_node *orig_neigh_node,
+ struct batadv_ogm_packet *ogm_packet,
+ int is_single_hop_neigh)
+{
+ struct batadv_nc_node *in_nc_node = NULL, *out_nc_node = NULL;
+
+ /* Check if network coding is enabled */
+ if (!atomic_read(&bat_priv->network_coding))
+ goto out;
+
+ /* accept ogms from 'good' neighbors and single hop neighbors */
+ if (!batadv_can_nc_with_orig(bat_priv, orig_node, ogm_packet) &&
+ !is_single_hop_neigh)
+ goto out;
+
+ /* Add orig_node as in_nc_node on hop */
+ in_nc_node = batadv_nc_get_nc_node(bat_priv, orig_node,
+ orig_neigh_node, true);
+ if (!in_nc_node)
+ goto out;
+
+ in_nc_node->last_seen = jiffies;
+
+ /* Add hop as out_nc_node on orig_node */
+ out_nc_node = batadv_nc_get_nc_node(bat_priv, orig_neigh_node,
+ orig_node, false);
+ if (!out_nc_node)
+ goto out;
+
+ out_nc_node->last_seen = jiffies;
+
+out:
+ if (in_nc_node)
+ batadv_nc_node_free_ref(in_nc_node);
+ if (out_nc_node)
+ batadv_nc_node_free_ref(out_nc_node);
+}
+
+/**
* batadv_nc_free - clean up network coding memory
* @bat_priv: the bat priv with all the soft interface information
*/
@@ -79,3 +414,82 @@ void batadv_nc_free(struct batadv_priv *bat_priv)
{
cancel_delayed_work_sync(&bat_priv->nc.work);
}
+
+/**
+ * batadv_nc_nodes_seq_print_text - print the nc node information
+ * @seq: seq file to print on
+ * @offset: not used
+ */
+int batadv_nc_nodes_seq_print_text(struct seq_file *seq, void *offset)
+{
+ struct net_device *net_dev = (struct net_device *)seq->private;
+ struct batadv_priv *bat_priv = netdev_priv(net_dev);
+ struct batadv_hashtable *hash = bat_priv->orig_hash;
+ struct batadv_hard_iface *primary_if;
+ struct hlist_head *head;
+ struct batadv_orig_node *orig_node;
+ struct batadv_nc_node *nc_node;
+ int i;
+
+ primary_if = batadv_seq_print_text_primary_if_get(seq);
+ if (!primary_if)
+ goto out;
+
+ /* Traverse list of originators */
+ for (i = 0; i < hash->size; i++) {
+ head = &hash->table[i];
+
+ /* For each orig_node in this bin */
+ rcu_read_lock();
+ hlist_for_each_entry_rcu(orig_node, head, hash_entry) {
+ seq_printf(seq, "Node: %pM\n", orig_node->orig);
+
+ seq_printf(seq, " Ingoing: ");
+ /* For each in_nc_node to this orig_node */
+ list_for_each_entry_rcu(nc_node,
+ &orig_node->in_coding_list,
+ list)
+ seq_printf(seq, "%pM ",
+ nc_node->addr);
+ seq_printf(seq, "\n");
+
+ seq_printf(seq, " Outgoing: ");
+ /* For out_nc_node to this orig_node */
+ list_for_each_entry_rcu(nc_node,
+ &orig_node->out_coding_list,
+ list)
+ seq_printf(seq, "%pM ",
+ nc_node->addr);
+ seq_printf(seq, "\n\n");
+ }
+ rcu_read_unlock();
+ }
+
+out:
+ if (primary_if)
+ batadv_hardif_free_ref(primary_if);
+ return 0;
+}
+
+/**
+ * batadv_nc_init_debugfs - create nc folder and related files in debugfs
+ * @bat_priv: the bat priv with all the soft interface information
+ */
+int batadv_nc_init_debugfs(struct batadv_priv *bat_priv)
+{
+ struct dentry *nc_dir, *file;
+
+ nc_dir = debugfs_create_dir("nc", bat_priv->debug_dir);
+ if (!nc_dir)
+ goto out;
+
+ file = debugfs_create_u8("min_tq", S_IRUGO | S_IWUSR, nc_dir,
+ &bat_priv->nc.min_tq);
+ if (!file)
+ goto out;
+
+ return 0;
+
+out:
+ return -ENOMEM;
+}
diff --git a/net/batman-adv/network-coding.h b/net/batman-adv/network-coding.h
index 7483cba..4c56cd9 100644
--- a/net/batman-adv/network-coding.h
+++ b/net/batman-adv/network-coding.h
@@ -24,7 +24,19 @@
int batadv_nc_init(struct batadv_priv *bat_priv);
void batadv_nc_free(struct batadv_priv *bat_priv);
+void batadv_nc_update_nc_node(struct batadv_priv *bat_priv,
+ struct batadv_orig_node *orig_node,
+ struct batadv_orig_node *orig_neigh_node,
+ struct batadv_ogm_packet *ogm_packet,
+ int is_single_hop_neigh);
+void batadv_nc_purge_orig(struct batadv_priv *bat_priv,
+ struct batadv_orig_node *orig_node,
+ bool (*to_purge)(struct batadv_priv *,
+ struct batadv_nc_node *));
void batadv_nc_init_bat_priv(struct batadv_priv *bat_priv);
+void batadv_nc_init_orig(struct batadv_orig_node *orig_node);
+int batadv_nc_nodes_seq_print_text(struct seq_file *seq, void *offset);
+int batadv_nc_init_debugfs(struct batadv_priv *bat_priv);
#else /* ifdef CONFIG_BATMAN_ADV_NC */
@@ -38,11 +50,46 @@ static inline void batadv_nc_free(struct batadv_priv *bat_priv)
return;
}
+static inline void
+batadv_nc_update_nc_node(struct batadv_priv *bat_priv,
+ struct batadv_orig_node *orig_node,
+ struct batadv_orig_node *orig_neigh_node,
+ struct batadv_ogm_packet *ogm_packet,
+ int is_single_hop_neigh)
+{
+ return;
+}
+
+static inline void
+batadv_nc_purge_orig(struct batadv_priv *bat_priv,
+ struct batadv_orig_node *orig_node,
+ bool (*to_purge)(struct batadv_priv *,
+ struct batadv_nc_node *))
+{
+ return;
+}
+
static inline void batadv_nc_init_bat_priv(struct batadv_priv *bat_priv)
{
return;
}
+static inline void batadv_nc_init_orig(struct batadv_orig_node *orig_node)
+{
+ return;
+}
+
+static inline int batadv_nc_nodes_seq_print_text(struct seq_file *seq,
+ void *offset)
+{
+ return 0;
+}
+
+static inline int batadv_nc_init_debugfs(struct batadv_priv *bat_priv)
+{
+ return 0;
+}
+
#endif /* ifdef CONFIG_BATMAN_ADV_NC */
#endif /* _NET_BATMAN_ADV_NETWORK_CODING_H_ */
diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index 96fb80b..585e684 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -28,6 +28,7 @@
#include "unicast.h"
#include "soft-interface.h"
#include "bridge_loop_avoidance.h"
+#include "network-coding.h"
/* hash class keys */
static struct lock_class_key batadv_orig_hash_lock_class_key;
@@ -142,6 +143,9 @@ static void batadv_orig_node_free_rcu(struct rcu_head *rcu)
spin_unlock_bh(&orig_node->neigh_list_lock);
+ /* Free nc_nodes */
+ batadv_nc_purge_orig(orig_node->bat_priv, orig_node, NULL);
+
batadv_frag_list_free(&orig_node->frag_list);
batadv_tt_global_del_orig(orig_node->bat_priv, orig_node,
"originator timed out");
@@ -219,6 +223,8 @@ struct batadv_orig_node *batadv_get_orig_node(struct batadv_priv *bat_priv,
spin_lock_init(&orig_node->neigh_list_lock);
spin_lock_init(&orig_node->tt_buff_lock);
+ batadv_nc_init_orig(orig_node);
+
/* extra reference for return */
atomic_set(&orig_node->refcount, 2);
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 83bfe7c..1544ab4 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -128,6 +128,10 @@ struct batadv_hard_iface {
* @bond_list: list of bonding candidates
* @refcount: number of contexts the object is used
* @rcu: struct used for freeing in an RCU-safe manner
+ * @in_coding_list: list of nodes this orig can hear
+ * @out_coding_list: list of nodes that can hear this orig
+ * @in_coding_list_lock: protects in_coding_list
+ * @out_coding_list_lock: protects out_coding_list
*/
struct batadv_orig_node {
uint8_t orig[ETH_ALEN];
@@ -171,6 +175,12 @@ struct batadv_orig_node {
struct list_head bond_list;
atomic_t refcount;
struct rcu_head rcu;
+#ifdef CONFIG_BATMAN_ADV_NC
+ struct list_head in_coding_list;
+ struct list_head out_coding_list;
+ spinlock_t in_coding_list_lock; /* Protects in_coding_list */
+ spinlock_t out_coding_list_lock; /* Protects out_coding_list */
+#endif
};
/**
@@ -430,9 +440,13 @@ struct batadv_priv_dat {
/**
* struct batadv_priv_nc - per mesh interface network coding private data
* @work: work queue callback item for cleanup
+ * @debug_dir: dentry for nc subdir in batman-adv directory in debugfs
+ * @min_tq: only consider neighbors for encoding if neigh_tq > min_tq
*/
struct batadv_priv_nc {
struct delayed_work work;
+ struct dentry *debug_dir;
+ u8 min_tq;
};
/**
@@ -716,6 +730,24 @@ struct batadv_tt_roam_node {
};
/**
+ * struct batadv_nc_node - network coding node
+ * @list: next and prev pointer for the list handling
+ * @addr: the node's mac address
+ * @refcount: number of contexts the object is used by
+ * @rcu: struct used for freeing in an RCU-safe manner
+ * @orig_node: pointer to corresponding orig node struct
+ * @last_seen: timestamp of last ogm received from this node
+ */
+struct batadv_nc_node {
+ struct list_head list;
+ uint8_t addr[ETH_ALEN];
+ atomic_t refcount;
+ struct rcu_head rcu;
+ struct batadv_orig_node *orig_node;
+ unsigned long last_seen;
+};
+
+/**
* struct batadv_forw_packet - structure for bcast packets to be sent/forwarded
* @list: list node for batadv_socket_client::queue_list
* @send_time: execution time for delayed_work (packet sending)
--
1.8.1.5
^ permalink raw reply related
* [PATCH 6/9] batman-adv: network coding - buffer unicast packets before forward
From: Antonio Quartulli @ 2013-03-13 22:56 UTC (permalink / raw)
To: davem
Cc: netdev, b.a.t.m.a.n, Martin Hundebøll, Marek Lindner,
Antonio Quartulli
In-Reply-To: <1363215399-18875-1-git-send-email-ordex@autistici.org>
From: Martin Hundebøll <martin@hundeboll.net>
Two be able to network code two packets, one packet must be buffered
until the next is available. This is done in a "coding buffer", which is
essentially a hash table with lists of packets. Each entry in the hash
table corresponds to a specific src-dst pair, which has a linked list of
packets that are buffered.
This patch adds skbs to the buffer just before forwarding them. The
buffer is traversed every 10 ms, where timed skbs are removed from the
buffer and transmitted. To allow experiments with the network coding
scheme, the timeout is tunable through a file in debugfs.
Signed-off-by: Martin Hundebøll <martin@hundeboll.net>
Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
net/batman-adv/network-coding.c | 487 ++++++++++++++++++++++++++++++++++++++++
net/batman-adv/network-coding.h | 10 +
net/batman-adv/routing.c | 18 +-
net/batman-adv/types.h | 49 ++++
4 files changed, 557 insertions(+), 7 deletions(-)
diff --git a/net/batman-adv/network-coding.c b/net/batman-adv/network-coding.c
index ff4985d..2c38c54 100644
--- a/net/batman-adv/network-coding.c
+++ b/net/batman-adv/network-coding.c
@@ -20,10 +20,14 @@
#include <linux/debugfs.h>
#include "main.h"
+#include "hash.h"
#include "network-coding.h"
+#include "send.h"
#include "originator.h"
#include "hard-interface.h"
+static struct lock_class_key batadv_nc_coding_hash_lock_class_key;
+
static void batadv_nc_worker(struct work_struct *work);
/**
@@ -42,10 +46,25 @@ static void batadv_nc_start_timer(struct batadv_priv *bat_priv)
*/
int batadv_nc_init(struct batadv_priv *bat_priv)
{
+ bat_priv->nc.timestamp_fwd_flush = jiffies;
+
+ if (bat_priv->nc.coding_hash)
+ return 0;
+
+ bat_priv->nc.coding_hash = batadv_hash_new(128);
+ if (!bat_priv->nc.coding_hash)
+ goto err;
+
+ batadv_hash_set_lock_class(bat_priv->nc.coding_hash,
+ &batadv_nc_coding_hash_lock_class_key);
+
INIT_DELAYED_WORK(&bat_priv->nc.work, batadv_nc_worker);
batadv_nc_start_timer(bat_priv);
return 0;
+
+err:
+ return -ENOMEM;
}
/**
@@ -56,6 +75,7 @@ void batadv_nc_init_bat_priv(struct batadv_priv *bat_priv)
{
atomic_set(&bat_priv->network_coding, 1);
bat_priv->nc.min_tq = 200;
+ bat_priv->nc.max_fwd_delay = 10;
}
/**
@@ -96,6 +116,30 @@ static void batadv_nc_node_free_ref(struct batadv_nc_node *nc_node)
}
/**
+ * batadv_nc_path_free_ref - decrements the nc path refcounter and possibly
+ * frees it
+ * @nc_path: the nc node to free
+ */
+static void batadv_nc_path_free_ref(struct batadv_nc_path *nc_path)
+{
+ if (atomic_dec_and_test(&nc_path->refcount))
+ kfree_rcu(nc_path, rcu);
+}
+
+/**
+ * batadv_nc_packet_free - frees nc packet
+ * @nc_packet: the nc packet to free
+ */
+static void batadv_nc_packet_free(struct batadv_nc_packet *nc_packet)
+{
+ if (nc_packet->skb)
+ kfree_skb(nc_packet->skb);
+
+ batadv_nc_path_free_ref(nc_packet->nc_path);
+ kfree(nc_packet);
+}
+
+/**
* batadv_nc_to_purge_nc_node - checks whether an nc node has to be purged
* @bat_priv: the bat priv with all the soft interface information
* @nc_node: the nc node to check
@@ -112,6 +156,26 @@ static bool batadv_nc_to_purge_nc_node(struct batadv_priv *bat_priv,
}
/**
+ * batadv_nc_to_purge_nc_path_coding - checks whether an nc path has timed out
+ * @bat_priv: the bat priv with all the soft interface information
+ * @nc_path: the nc path to check
+ *
+ * Returns true if the entry has to be purged now, false otherwise
+ */
+static bool batadv_nc_to_purge_nc_path_coding(struct batadv_priv *bat_priv,
+ struct batadv_nc_path *nc_path)
+{
+ if (atomic_read(&bat_priv->mesh_state) != BATADV_MESH_ACTIVE)
+ return true;
+
+ /* purge the path when no packets has been added for 10 times the
+ * max_fwd_delay time
+ */
+ return batadv_has_timed_out(nc_path->last_valid,
+ bat_priv->nc.max_fwd_delay * 10);
+}
+
+/**
* batadv_nc_purge_orig_nc_nodes - go through list of nc nodes and purge stale
* entries
* @bat_priv: the bat priv with all the soft interface information
@@ -203,6 +267,262 @@ static void batadv_nc_purge_orig_hash(struct batadv_priv *bat_priv)
}
/**
+ * batadv_nc_purge_paths - traverse all nc paths part of the hash and remove
+ * unused ones
+ * @bat_priv: the bat priv with all the soft interface information
+ * @hash: hash table containing the nc paths to check
+ * @to_purge: function in charge to decide whether an entry has to be purged or
+ * not. This function takes the nc node as argument and has to return
+ * a boolean value: true is the entry has to be deleted, false
+ * otherwise
+ */
+static void batadv_nc_purge_paths(struct batadv_priv *bat_priv,
+ struct batadv_hashtable *hash,
+ bool (*to_purge)(struct batadv_priv *,
+ struct batadv_nc_path *))
+{
+ struct hlist_head *head;
+ struct hlist_node *node_tmp;
+ struct batadv_nc_path *nc_path;
+ spinlock_t *lock; /* Protects lists in hash */
+ uint32_t i;
+
+ for (i = 0; i < hash->size; i++) {
+ head = &hash->table[i];
+ lock = &hash->list_locks[i];
+
+ /* For each nc_path in this bin */
+ spin_lock_bh(lock);
+ hlist_for_each_entry_safe(nc_path, node_tmp, head, hash_entry) {
+ /* if an helper function has been passed as parameter,
+ * ask it if the entry has to be purged or not
+ */
+ if (to_purge && !to_purge(bat_priv, nc_path))
+ continue;
+
+ /* purging an non-empty nc_path should never happen, but
+ * is observed under high CPU load. Delay the purging
+ * until next iteration to allow the packet_list to be
+ * emptied first.
+ */
+ if (!unlikely(list_empty(&nc_path->packet_list))) {
+ net_ratelimited_function(printk,
+ KERN_WARNING
+ "Skipping free of non-empty nc_path (%pM -> %pM)!\n",
+ nc_path->prev_hop,
+ nc_path->next_hop);
+ continue;
+ }
+
+ /* nc_path is unused, so remove it */
+ batadv_dbg(BATADV_DBG_NC, bat_priv,
+ "Remove nc_path %pM -> %pM\n",
+ nc_path->prev_hop, nc_path->next_hop);
+ hlist_del_rcu(&nc_path->hash_entry);
+ batadv_nc_path_free_ref(nc_path);
+ }
+ spin_unlock_bh(lock);
+ }
+}
+
+/**
+ * batadv_nc_hash_key_gen - computes the nc_path hash key
+ * @key: buffer to hold the final hash key
+ * @src: source ethernet mac address going into the hash key
+ * @dst: destination ethernet mac address going into the hash key
+ */
+static void batadv_nc_hash_key_gen(struct batadv_nc_path *key, const char *src,
+ const char *dst)
+{
+ memcpy(key->prev_hop, src, sizeof(key->prev_hop));
+ memcpy(key->next_hop, dst, sizeof(key->next_hop));
+}
+
+/**
+ * batadv_nc_hash_choose - compute the hash value for an nc path
+ * @data: data to hash
+ * @size: size of the hash table
+ *
+ * Returns the selected index in the hash table for the given data.
+ */
+static uint32_t batadv_nc_hash_choose(const void *data, uint32_t size)
+{
+ const struct batadv_nc_path *nc_path = data;
+ uint32_t hash = 0;
+
+ hash = batadv_hash_bytes(hash, &nc_path->prev_hop,
+ sizeof(nc_path->prev_hop));
+ hash = batadv_hash_bytes(hash, &nc_path->next_hop,
+ sizeof(nc_path->next_hop));
+
+ hash += (hash << 3);
+ hash ^= (hash >> 11);
+ hash += (hash << 15);
+
+ return hash % size;
+}
+
+/**
+ * batadv_nc_hash_compare - comparing function used in the network coding hash
+ * tables
+ * @node: node in the local table
+ * @data2: second object to compare the node to
+ *
+ * Returns 1 if the two entry are the same, 0 otherwise
+ */
+static int batadv_nc_hash_compare(const struct hlist_node *node,
+ const void *data2)
+{
+ const struct batadv_nc_path *nc_path1, *nc_path2;
+
+ nc_path1 = container_of(node, struct batadv_nc_path, hash_entry);
+ nc_path2 = data2;
+
+ /* Return 1 if the two keys are identical */
+ if (memcmp(nc_path1->prev_hop, nc_path2->prev_hop,
+ sizeof(nc_path1->prev_hop)) != 0)
+ return 0;
+
+ if (memcmp(nc_path1->next_hop, nc_path2->next_hop,
+ sizeof(nc_path1->next_hop)) != 0)
+ return 0;
+
+ return 1;
+}
+
+/**
+ * batadv_nc_hash_find - search for an existing nc path and return it
+ * @hash: hash table containing the nc path
+ * @data: search key
+ *
+ * Returns the nc_path if found, NULL otherwise.
+ */
+static struct batadv_nc_path *
+batadv_nc_hash_find(struct batadv_hashtable *hash,
+ void *data)
+{
+ struct hlist_head *head;
+ struct batadv_nc_path *nc_path, *nc_path_tmp = NULL;
+ int index;
+
+ if (!hash)
+ return NULL;
+
+ index = batadv_nc_hash_choose(data, hash->size);
+ head = &hash->table[index];
+
+ rcu_read_lock();
+ hlist_for_each_entry_rcu(nc_path, head, hash_entry) {
+ if (!batadv_nc_hash_compare(&nc_path->hash_entry, data))
+ continue;
+
+ if (!atomic_inc_not_zero(&nc_path->refcount))
+ continue;
+
+ nc_path_tmp = nc_path;
+ break;
+ }
+ rcu_read_unlock();
+
+ return nc_path_tmp;
+}
+
+/**
+ * batadv_nc_send_packet - send non-coded packet and free nc_packet struct
+ * @nc_packet: the nc packet to send
+ */
+static void batadv_nc_send_packet(struct batadv_nc_packet *nc_packet)
+{
+ batadv_send_skb_packet(nc_packet->skb,
+ nc_packet->neigh_node->if_incoming,
+ nc_packet->nc_path->next_hop);
+ nc_packet->skb = NULL;
+ batadv_nc_packet_free(nc_packet);
+}
+
+/**
+ * batadv_nc_fwd_flush - Checks the timestamp of the given nc packet.
+ * @bat_priv: the bat priv with all the soft interface information
+ * @nc_path: the nc path the packet belongs to
+ * @nc_packet: the nc packet to be checked
+ *
+ * Checks whether the given nc packet has hit its forward timeout. If so, the
+ * packet is no longer delayed, immediately sent and the entry deleted from the
+ * queue. Has to be called with the appropriate locks.
+ *
+ * Returns false as soon as the entry in the fifo queue has not been timed out
+ * yet and true otherwise.
+ */
+static bool batadv_nc_fwd_flush(struct batadv_priv *bat_priv,
+ struct batadv_nc_path *nc_path,
+ struct batadv_nc_packet *nc_packet)
+{
+ unsigned long timeout = bat_priv->nc.max_fwd_delay;
+
+ /* Packets are added to tail, so the remaining packets did not time
+ * out and we can stop processing the current queue
+ */
+ if (atomic_read(&bat_priv->mesh_state) == BATADV_MESH_ACTIVE &&
+ !batadv_has_timed_out(nc_packet->timestamp, timeout))
+ return false;
+
+ /* Send packet */
+ batadv_inc_counter(bat_priv, BATADV_CNT_FORWARD);
+ batadv_add_counter(bat_priv, BATADV_CNT_FORWARD_BYTES,
+ nc_packet->skb->len + ETH_HLEN);
+ list_del(&nc_packet->list);
+ batadv_nc_send_packet(nc_packet);
+
+ return true;
+}
+
+/**
+ * batadv_nc_process_nc_paths - traverse given nc packet pool and free timed out
+ * nc packets
+ * @bat_priv: the bat priv with all the soft interface information
+ * @hash: to be processed hash table
+ * @process_fn: Function called to process given nc packet. Should return true
+ * to encourage this function to proceed with the next packet.
+ * Otherwise the rest of the current queue is skipped.
+ */
+static void
+batadv_nc_process_nc_paths(struct batadv_priv *bat_priv,
+ struct batadv_hashtable *hash,
+ bool (*process_fn)(struct batadv_priv *,
+ struct batadv_nc_path *,
+ struct batadv_nc_packet *))
+{
+ struct hlist_head *head;
+ struct batadv_nc_packet *nc_packet, *nc_packet_tmp;
+ struct batadv_nc_path *nc_path;
+ bool ret;
+ int i;
+
+ if (!hash)
+ return;
+
+ /* Loop hash table bins */
+ for (i = 0; i < hash->size; i++) {
+ head = &hash->table[i];
+
+ /* Loop coding paths */
+ rcu_read_lock();
+ hlist_for_each_entry_rcu(nc_path, head, hash_entry) {
+ /* Loop packets */
+ spin_lock_bh(&nc_path->packet_list_lock);
+ list_for_each_entry_safe(nc_packet, nc_packet_tmp,
+ &nc_path->packet_list, list) {
+ ret = process_fn(bat_priv, nc_path, nc_packet);
+ if (!ret)
+ break;
+ }
+ spin_unlock_bh(&nc_path->packet_list_lock);
+ }
+ rcu_read_unlock();
+ }
+}
+
+/**
* batadv_nc_worker - periodic task for house keeping related to network coding
* @work: kernel work struct
*/
@@ -211,12 +531,23 @@ static void batadv_nc_worker(struct work_struct *work)
struct delayed_work *delayed_work;
struct batadv_priv_nc *priv_nc;
struct batadv_priv *bat_priv;
+ unsigned long timeout;
delayed_work = container_of(work, struct delayed_work, work);
priv_nc = container_of(delayed_work, struct batadv_priv_nc, work);
bat_priv = container_of(priv_nc, struct batadv_priv, nc);
batadv_nc_purge_orig_hash(bat_priv);
+ batadv_nc_purge_paths(bat_priv, bat_priv->nc.coding_hash,
+ batadv_nc_to_purge_nc_path_coding);
+
+ timeout = bat_priv->nc.max_fwd_delay;
+
+ if (batadv_has_timed_out(bat_priv->nc.timestamp_fwd_flush, timeout)) {
+ batadv_nc_process_nc_paths(bat_priv, bat_priv->nc.coding_hash,
+ batadv_nc_fwd_flush);
+ bat_priv->nc.timestamp_fwd_flush = jiffies;
+ }
/* Schedule a new check */
batadv_nc_start_timer(bat_priv);
@@ -407,12 +738,163 @@ out:
}
/**
+ * batadv_nc_get_path - get existing nc_path or allocate a new one
+ * @bat_priv: the bat priv with all the soft interface information
+ * @hash: hash table containing the nc path
+ * @src: ethernet source address - first half of the nc path search key
+ * @dst: ethernet destination address - second half of the nc path search key
+ *
+ * Returns pointer to nc_path if the path was found or created, returns NULL
+ * on error.
+ */
+static struct batadv_nc_path *batadv_nc_get_path(struct batadv_priv *bat_priv,
+ struct batadv_hashtable *hash,
+ uint8_t *src,
+ uint8_t *dst)
+{
+ int hash_added;
+ struct batadv_nc_path *nc_path, nc_path_key;
+
+ batadv_nc_hash_key_gen(&nc_path_key, src, dst);
+
+ /* Search for existing nc_path */
+ nc_path = batadv_nc_hash_find(hash, (void *)&nc_path_key);
+
+ if (nc_path) {
+ /* Set timestamp to delay removal of nc_path */
+ nc_path->last_valid = jiffies;
+ return nc_path;
+ }
+
+ /* No existing nc_path was found; create a new */
+ nc_path = kzalloc(sizeof(*nc_path), GFP_ATOMIC);
+
+ if (!nc_path)
+ return NULL;
+
+ /* Initialize nc_path */
+ INIT_LIST_HEAD(&nc_path->packet_list);
+ spin_lock_init(&nc_path->packet_list_lock);
+ atomic_set(&nc_path->refcount, 2);
+ nc_path->last_valid = jiffies;
+ memcpy(nc_path->next_hop, dst, ETH_ALEN);
+ memcpy(nc_path->prev_hop, src, ETH_ALEN);
+
+ batadv_dbg(BATADV_DBG_NC, bat_priv, "Adding nc_path %pM -> %pM\n",
+ nc_path->prev_hop,
+ nc_path->next_hop);
+
+ /* Add nc_path to hash table */
+ hash_added = batadv_hash_add(hash, batadv_nc_hash_compare,
+ batadv_nc_hash_choose, &nc_path_key,
+ &nc_path->hash_entry);
+
+ if (hash_added < 0) {
+ kfree(nc_path);
+ return NULL;
+ }
+
+ return nc_path;
+}
+
+/**
+ * batadv_nc_skb_add_to_path - buffer skb for later encoding / decoding
+ * @skb: skb to add to path
+ * @nc_path: path to add skb to
+ * @neigh_node: next hop to forward packet to
+ * @packet_id: checksum to identify packet
+ *
+ * Returns true if the packet was buffered or false in case of an error.
+ */
+static bool batadv_nc_skb_add_to_path(struct sk_buff *skb,
+ struct batadv_nc_path *nc_path,
+ struct batadv_neigh_node *neigh_node,
+ __be32 packet_id)
+{
+ struct batadv_nc_packet *nc_packet;
+
+ nc_packet = kzalloc(sizeof(*nc_packet), GFP_ATOMIC);
+ if (!nc_packet)
+ return false;
+
+ /* Initialize nc_packet */
+ nc_packet->timestamp = jiffies;
+ nc_packet->packet_id = packet_id;
+ nc_packet->skb = skb;
+ nc_packet->neigh_node = neigh_node;
+ nc_packet->nc_path = nc_path;
+
+ /* Add coding packet to list */
+ spin_lock_bh(&nc_path->packet_list_lock);
+ list_add_tail(&nc_packet->list, &nc_path->packet_list);
+ spin_unlock_bh(&nc_path->packet_list_lock);
+
+ return true;
+}
+
+/**
+ * batadv_nc_skb_forward - try to code a packet or add it to the coding packet
+ * buffer
+ * @skb: data skb to forward
+ * @neigh_node: next hop to forward packet to
+ * @ethhdr: pointer to the ethernet header inside the skb
+ *
+ * Returns true if the skb was consumed (encoded packet sent) or false otherwise
+ */
+bool batadv_nc_skb_forward(struct sk_buff *skb,
+ struct batadv_neigh_node *neigh_node,
+ struct ethhdr *ethhdr)
+{
+ const struct net_device *netdev = neigh_node->if_incoming->soft_iface;
+ struct batadv_priv *bat_priv = netdev_priv(netdev);
+ struct batadv_unicast_packet *packet;
+ struct batadv_nc_path *nc_path;
+ __be32 packet_id;
+ u8 *payload;
+
+ /* Check if network coding is enabled */
+ if (!atomic_read(&bat_priv->network_coding))
+ goto out;
+
+ /* We only handle unicast packets */
+ payload = skb_network_header(skb);
+ packet = (struct batadv_unicast_packet *)payload;
+ if (packet->header.packet_type != BATADV_UNICAST)
+ goto out;
+
+ /* Find or create a nc_path for this src-dst pair */
+ nc_path = batadv_nc_get_path(bat_priv,
+ bat_priv->nc.coding_hash,
+ ethhdr->h_source,
+ neigh_node->addr);
+
+ if (!nc_path)
+ goto out;
+
+ /* Add skb to nc_path */
+ packet_id = batadv_skb_crc32(skb, payload + sizeof(*packet));
+ if (!batadv_nc_skb_add_to_path(skb, nc_path, neigh_node, packet_id))
+ goto free_nc_path;
+
+ /* Packet is consumed */
+ return true;
+
+free_nc_path:
+ batadv_nc_path_free_ref(nc_path);
+out:
+ /* Packet is not consumed */
+ return false;
+}
+
+/**
* batadv_nc_free - clean up network coding memory
* @bat_priv: the bat priv with all the soft interface information
*/
void batadv_nc_free(struct batadv_priv *bat_priv)
{
cancel_delayed_work_sync(&bat_priv->nc.work);
+ batadv_nc_purge_paths(bat_priv, bat_priv->nc.coding_hash, NULL);
+ batadv_hash_destroy(bat_priv->nc.coding_hash);
}
/**
@@ -488,6 +970,11 @@ int batadv_nc_init_debugfs(struct batadv_priv *bat_priv)
if (!file)
goto out;
+ file = debugfs_create_u32("max_fwd_delay", S_IRUGO | S_IWUSR, nc_dir,
+ &bat_priv->nc.max_fwd_delay);
+ if (!file)
+ goto out;
+
return 0;
out:
diff --git a/net/batman-adv/network-coding.h b/net/batman-adv/network-coding.h
index 4c56cd9..c32602a 100644
--- a/net/batman-adv/network-coding.h
+++ b/net/batman-adv/network-coding.h
@@ -35,6 +35,9 @@ void batadv_nc_purge_orig(struct batadv_priv *bat_priv,
struct batadv_nc_node *));
void batadv_nc_init_bat_priv(struct batadv_priv *bat_priv);
void batadv_nc_init_orig(struct batadv_orig_node *orig_node);
+bool batadv_nc_skb_forward(struct sk_buff *skb,
+ struct batadv_neigh_node *neigh_node,
+ struct ethhdr *ethhdr);
int batadv_nc_nodes_seq_print_text(struct seq_file *seq, void *offset);
int batadv_nc_init_debugfs(struct batadv_priv *bat_priv);
@@ -79,6 +82,13 @@ static inline void batadv_nc_init_orig(struct batadv_orig_node *orig_node)
return;
}
+static inline bool batadv_nc_skb_forward(struct sk_buff *skb,
+ struct batadv_neigh_node *neigh_node,
+ struct ethhdr *ethhdr)
+{
+ return false;
+}
+
static inline int batadv_nc_nodes_seq_print_text(struct seq_file *seq,
void *offset)
{
diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c
index 322c97a..44fda7c 100644
--- a/net/batman-adv/routing.c
+++ b/net/batman-adv/routing.c
@@ -29,6 +29,7 @@
#include "unicast.h"
#include "bridge_loop_avoidance.h"
#include "distributed-arp-table.h"
+#include "network-coding.h"
static int batadv_route_unicast_packet(struct sk_buff *skb,
struct batadv_hard_iface *recv_if);
@@ -860,14 +861,17 @@ static int batadv_route_unicast_packet(struct sk_buff *skb,
/* decrement ttl */
unicast_packet->header.ttl--;
- /* Update stats counter */
- batadv_inc_counter(bat_priv, BATADV_CNT_FORWARD);
- batadv_add_counter(bat_priv, BATADV_CNT_FORWARD_BYTES,
- skb->len + ETH_HLEN);
+ /* network code packet if possible */
+ if (batadv_nc_skb_forward(skb, neigh_node, ethhdr)) {
+ ret = NET_RX_SUCCESS;
+ } else if (batadv_send_skb_to_orig(skb, orig_node, recv_if)) {
+ ret = NET_RX_SUCCESS;
- /* route it */
- if (batadv_send_skb_to_orig(skb, orig_node, recv_if))
- ret = NET_RX_SUCCESS;
+ /* Update stats counter */
+ batadv_inc_counter(bat_priv, BATADV_CNT_FORWARD);
+ batadv_add_counter(bat_priv, BATADV_CNT_FORWARD_BYTES,
+ skb->len + ETH_HLEN);
+ }
out:
if (neigh_node)
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 1544ab4..43caf7f 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -442,11 +442,19 @@ struct batadv_priv_dat {
* @work: work queue callback item for cleanup
* @debug_dir: dentry for nc subdir in batman-adv directory in debugfs
* @min_tq: only consider neighbors for encoding if neigh_tq > min_tq
+ * @max_fwd_delay: maximum packet forward delay to allow coding of packets
+ * @timestamp_fwd_flush: timestamp of last forward packet queue flush
+ * @coding_hash: Hash table used to buffer skbs while waiting for another
+ * incoming skb to code it with. Skbs are added to the buffer just before being
+ * forwarded in routing.c
*/
struct batadv_priv_nc {
struct delayed_work work;
struct dentry *debug_dir;
u8 min_tq;
+ u32 max_fwd_delay;
+ unsigned long timestamp_fwd_flush;
+ struct batadv_hashtable *coding_hash;
};
/**
@@ -748,6 +756,47 @@ struct batadv_nc_node {
};
/**
+ * struct batadv_nc_path - network coding path
+ * @hash_entry: next and prev pointer for the list handling
+ * @rcu: struct used for freeing in an RCU-safe manner
+ * @refcount: number of contexts the object is used by
+ * @packet_list: list of buffered packets for this path
+ * @packet_list_lock: access lock for packet list
+ * @next_hop: next hop (destination) of path
+ * @prev_hop: previous hop (source) of path
+ * @last_valid: timestamp for last validation of path
+ */
+struct batadv_nc_path {
+ struct hlist_node hash_entry;
+ struct rcu_head rcu;
+ atomic_t refcount;
+ struct list_head packet_list;
+ spinlock_t packet_list_lock; /* Protects packet_list */
+ uint8_t next_hop[ETH_ALEN];
+ uint8_t prev_hop[ETH_ALEN];
+ unsigned long last_valid;
+};
+
+/**
+ * struct batadv_nc_packet - network coding packet used when coding and
+ * decoding packets
+ * @list: next and prev pointer for the list handling
+ * @packet_id: crc32 checksum of skb data
+ * @timestamp: field containing the info when the packet was added to path
+ * @neigh_node: pointer to original next hop neighbor of skb
+ * @skb: skb which can be encoded or used for decoding
+ * @nc_path: pointer to path this nc packet is attached to
+ */
+struct batadv_nc_packet {
+ struct list_head list;
+ __be32 packet_id;
+ unsigned long timestamp;
+ struct batadv_neigh_node *neigh_node;
+ struct sk_buff *skb;
+ struct batadv_nc_path *nc_path;
+};
+
+/**
* struct batadv_forw_packet - structure for bcast packets to be sent/forwarded
* @list: list node for batadv_socket_client::queue_list
* @send_time: execution time for delayed_work (packet sending)
--
1.8.1.5
^ permalink raw reply related
* [PATCH 7/9] batman-adv: network coding - code and transmit packets if possible
From: Antonio Quartulli @ 2013-03-13 22:56 UTC (permalink / raw)
To: davem
Cc: netdev, b.a.t.m.a.n, Martin Hundebøll, Marek Lindner,
Antonio Quartulli
In-Reply-To: <1363215399-18875-1-git-send-email-ordex@autistici.org>
From: Martin Hundebøll <martin@hundeboll.net>
Before adding forward-skbs to the coding buffer, the buffer is searched
for a potential coding opportunity. If one is found, the two packets are
network coded and transmitted right away. If not, the forward-skb is
added to the buffer.
Network coded packets are transmitted with information about the two
receivers and the two coded packets. The first receiver is given by the
MAC header, while the second is given in the payload/bat-header. The
second receiver uses promiscuous mode to receive the packet and check
the second destination.
Signed-off-by: Martin Hundebøll <martin@hundeboll.net>
Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
net/batman-adv/main.h | 6 +
net/batman-adv/network-coding.c | 401 ++++++++++++++++++++++++++++++++++++++++
net/batman-adv/packet.h | 33 ++++
net/batman-adv/soft-interface.c | 6 +
net/batman-adv/types.h | 20 ++
5 files changed, 466 insertions(+)
diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index 8a10619..0afd4ee 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -302,4 +302,10 @@ static inline uint64_t batadv_sum_counter(struct batadv_priv *bat_priv,
return sum;
}
+/* Define a macro to reach the control buffer of the skb. The members of the
+ * control buffer are defined in struct batadv_skb_cb in types.h.
+ * The macro is inspired by the similar macro TCP_SKB_CB() in tcp.h.
+ */
+#define BATADV_SKB_CB(__skb) ((struct batadv_skb_cb *)&((__skb)->cb[0]))
+
#endif /* _NET_BATMAN_ADV_MAIN_H_ */
diff --git a/net/batman-adv/network-coding.c b/net/batman-adv/network-coding.c
index 2c38c54..fce2846 100644
--- a/net/batman-adv/network-coding.c
+++ b/net/batman-adv/network-coding.c
@@ -798,6 +798,403 @@ static struct batadv_nc_path *batadv_nc_get_path(struct batadv_priv *bat_priv,
}
/**
+ * batadv_nc_random_weight_tq - scale the receivers TQ-value to avoid unfair
+ * selection of a receiver with slightly lower TQ than the other
+ * @tq: to be weighted tq value
+ */
+static uint8_t batadv_nc_random_weight_tq(uint8_t tq)
+{
+ uint8_t rand_val, rand_tq;
+
+ get_random_bytes(&rand_val, sizeof(rand_val));
+
+ /* randomize the estimated packet loss (max TQ - estimated TQ) */
+ rand_tq = rand_val * (BATADV_TQ_MAX_VALUE - tq);
+
+ /* normalize the randomized packet loss */
+ rand_tq /= BATADV_TQ_MAX_VALUE;
+
+ /* convert to (randomized) estimated tq again */
+ return BATADV_TQ_MAX_VALUE - rand_tq;
+}
+
+/**
+ * batadv_nc_memxor - XOR destination with source
+ * @dst: byte array to XOR into
+ * @src: byte array to XOR from
+ * @len: length of destination array
+ */
+static void batadv_nc_memxor(char *dst, const char *src, unsigned int len)
+{
+ unsigned int i;
+
+ for (i = 0; i < len; ++i)
+ dst[i] ^= src[i];
+}
+
+/**
+ * batadv_nc_code_packets - code a received unicast_packet with an nc packet
+ * into a coded_packet and send it
+ * @bat_priv: the bat priv with all the soft interface information
+ * @skb: data skb to forward
+ * @ethhdr: pointer to the ethernet header inside the skb
+ * @nc_packet: structure containing the packet to the skb can be coded with
+ * @neigh_node: next hop to forward packet to
+ *
+ * Returns true if both packets are consumed, false otherwise.
+ */
+static bool batadv_nc_code_packets(struct batadv_priv *bat_priv,
+ struct sk_buff *skb,
+ struct ethhdr *ethhdr,
+ struct batadv_nc_packet *nc_packet,
+ struct batadv_neigh_node *neigh_node)
+{
+ uint8_t tq_weighted_neigh, tq_weighted_coding;
+ struct sk_buff *skb_dest, *skb_src;
+ struct batadv_unicast_packet *packet1;
+ struct batadv_unicast_packet *packet2;
+ struct batadv_coded_packet *coded_packet;
+ struct batadv_neigh_node *neigh_tmp, *router_neigh;
+ struct batadv_neigh_node *router_coding = NULL;
+ uint8_t *first_source, *first_dest, *second_source, *second_dest;
+ __be32 packet_id1, packet_id2;
+ size_t count;
+ bool res = false;
+ int coding_len;
+ int unicast_size = sizeof(*packet1);
+ int coded_size = sizeof(*coded_packet);
+ int header_add = coded_size - unicast_size;
+
+ router_neigh = batadv_orig_node_get_router(neigh_node->orig_node);
+ if (!router_neigh)
+ goto out;
+
+ neigh_tmp = nc_packet->neigh_node;
+ router_coding = batadv_orig_node_get_router(neigh_tmp->orig_node);
+ if (!router_coding)
+ goto out;
+
+ tq_weighted_neigh = batadv_nc_random_weight_tq(router_neigh->tq_avg);
+ tq_weighted_coding = batadv_nc_random_weight_tq(router_coding->tq_avg);
+
+ /* Select one destination for the MAC-header dst-field based on
+ * weighted TQ-values.
+ */
+ if (tq_weighted_neigh >= tq_weighted_coding) {
+ /* Destination from nc_packet is selected for MAC-header */
+ first_dest = nc_packet->nc_path->next_hop;
+ first_source = nc_packet->nc_path->prev_hop;
+ second_dest = neigh_node->addr;
+ second_source = ethhdr->h_source;
+ packet1 = (struct batadv_unicast_packet *)nc_packet->skb->data;
+ packet2 = (struct batadv_unicast_packet *)skb->data;
+ packet_id1 = nc_packet->packet_id;
+ packet_id2 = batadv_skb_crc32(skb,
+ skb->data + sizeof(*packet2));
+ } else {
+ /* Destination for skb is selected for MAC-header */
+ first_dest = neigh_node->addr;
+ first_source = ethhdr->h_source;
+ second_dest = nc_packet->nc_path->next_hop;
+ second_source = nc_packet->nc_path->prev_hop;
+ packet1 = (struct batadv_unicast_packet *)skb->data;
+ packet2 = (struct batadv_unicast_packet *)nc_packet->skb->data;
+ packet_id1 = batadv_skb_crc32(skb,
+ skb->data + sizeof(*packet1));
+ packet_id2 = nc_packet->packet_id;
+ }
+
+ /* Instead of zero padding the smallest data buffer, we
+ * code into the largest.
+ */
+ if (skb->len <= nc_packet->skb->len) {
+ skb_dest = nc_packet->skb;
+ skb_src = skb;
+ } else {
+ skb_dest = skb;
+ skb_src = nc_packet->skb;
+ }
+
+ /* coding_len is used when decoding the packet shorter packet */
+ coding_len = skb_src->len - unicast_size;
+
+ if (skb_linearize(skb_dest) < 0 || skb_linearize(skb_src) < 0)
+ goto out;
+
+ skb_push(skb_dest, header_add);
+
+ coded_packet = (struct batadv_coded_packet *)skb_dest->data;
+ skb_reset_mac_header(skb_dest);
+
+ coded_packet->header.packet_type = BATADV_CODED;
+ coded_packet->header.version = BATADV_COMPAT_VERSION;
+ coded_packet->header.ttl = packet1->header.ttl;
+
+ /* Info about first unicast packet */
+ memcpy(coded_packet->first_source, first_source, ETH_ALEN);
+ memcpy(coded_packet->first_orig_dest, packet1->dest, ETH_ALEN);
+ coded_packet->first_crc = packet_id1;
+ coded_packet->first_ttvn = packet1->ttvn;
+
+ /* Info about second unicast packet */
+ memcpy(coded_packet->second_dest, second_dest, ETH_ALEN);
+ memcpy(coded_packet->second_source, second_source, ETH_ALEN);
+ memcpy(coded_packet->second_orig_dest, packet2->dest, ETH_ALEN);
+ coded_packet->second_crc = packet_id2;
+ coded_packet->second_ttl = packet2->header.ttl;
+ coded_packet->second_ttvn = packet2->ttvn;
+ coded_packet->coded_len = htons(coding_len);
+
+ /* This is where the magic happens: Code skb_src into skb_dest */
+ batadv_nc_memxor(skb_dest->data + coded_size,
+ skb_src->data + unicast_size, coding_len);
+
+ /* Update counters accordingly */
+ if (BATADV_SKB_CB(skb_src)->decoded &&
+ BATADV_SKB_CB(skb_dest)->decoded) {
+ /* Both packets are recoded */
+ count = skb_src->len + ETH_HLEN;
+ count += skb_dest->len + ETH_HLEN;
+ batadv_add_counter(bat_priv, BATADV_CNT_NC_RECODE, 2);
+ batadv_add_counter(bat_priv, BATADV_CNT_NC_RECODE_BYTES, count);
+ } else if (!BATADV_SKB_CB(skb_src)->decoded &&
+ !BATADV_SKB_CB(skb_dest)->decoded) {
+ /* Both packets are newly coded */
+ count = skb_src->len + ETH_HLEN;
+ count += skb_dest->len + ETH_HLEN;
+ batadv_add_counter(bat_priv, BATADV_CNT_NC_CODE, 2);
+ batadv_add_counter(bat_priv, BATADV_CNT_NC_CODE_BYTES, count);
+ } else if (BATADV_SKB_CB(skb_src)->decoded &&
+ !BATADV_SKB_CB(skb_dest)->decoded) {
+ /* skb_src recoded and skb_dest is newly coded */
+ batadv_inc_counter(bat_priv, BATADV_CNT_NC_RECODE);
+ batadv_add_counter(bat_priv, BATADV_CNT_NC_RECODE_BYTES,
+ skb_src->len + ETH_HLEN);
+ batadv_inc_counter(bat_priv, BATADV_CNT_NC_CODE);
+ batadv_add_counter(bat_priv, BATADV_CNT_NC_CODE_BYTES,
+ skb_dest->len + ETH_HLEN);
+ } else if (!BATADV_SKB_CB(skb_src)->decoded &&
+ BATADV_SKB_CB(skb_dest)->decoded) {
+ /* skb_src is newly coded and skb_dest is recoded */
+ batadv_inc_counter(bat_priv, BATADV_CNT_NC_CODE);
+ batadv_add_counter(bat_priv, BATADV_CNT_NC_CODE_BYTES,
+ skb_src->len + ETH_HLEN);
+ batadv_inc_counter(bat_priv, BATADV_CNT_NC_RECODE);
+ batadv_add_counter(bat_priv, BATADV_CNT_NC_RECODE_BYTES,
+ skb_dest->len + ETH_HLEN);
+ }
+
+ /* skb_src is now coded into skb_dest, so free it */
+ kfree_skb(skb_src);
+
+ /* avoid duplicate free of skb from nc_packet */
+ nc_packet->skb = NULL;
+ batadv_nc_packet_free(nc_packet);
+
+ /* Send the coded packet and return true */
+ batadv_send_skb_packet(skb_dest, neigh_node->if_incoming, first_dest);
+ res = true;
+out:
+ if (router_neigh)
+ batadv_neigh_node_free_ref(router_neigh);
+ if (router_coding)
+ batadv_neigh_node_free_ref(router_coding);
+ return res;
+}
+
+/**
+ * batadv_nc_skb_coding_possible - true if a decoded skb is available at dst.
+ * @skb: data skb to forward
+ * @dst: destination mac address of the other skb to code with
+ * @src: source mac address of skb
+ *
+ * Whenever we network code a packet we have to check whether we received it in
+ * a network coded form. If so, we may not be able to use it for coding because
+ * some neighbors may also have received (overheard) the packet in the network
+ * coded form without being able to decode it. It is hard to know which of the
+ * neighboring nodes was able to decode the packet, therefore we can only
+ * re-code the packet if the source of the previous encoded packet is involved.
+ * Since the source encoded the packet we can be certain it has all necessary
+ * decode information.
+ *
+ * Returns true if coding of a decoded packet is allowed.
+ */
+static bool batadv_nc_skb_coding_possible(struct sk_buff *skb,
+ uint8_t *dst, uint8_t *src)
+{
+ if (BATADV_SKB_CB(skb)->decoded && !batadv_compare_eth(dst, src))
+ return false;
+ else
+ return true;
+}
+
+/**
+ * batadv_nc_path_search - Find the coding path matching in_nc_node and
+ * out_nc_node to retrieve a buffered packet that can be used for coding.
+ * @bat_priv: the bat priv with all the soft interface information
+ * @in_nc_node: pointer to skb next hop's neighbor nc node
+ * @out_nc_node: pointer to skb source's neighbor nc node
+ * @skb: data skb to forward
+ * @eth_dst: next hop mac address of skb
+ *
+ * Returns true if coding of a decoded skb is allowed.
+ */
+static struct batadv_nc_packet *
+batadv_nc_path_search(struct batadv_priv *bat_priv,
+ struct batadv_nc_node *in_nc_node,
+ struct batadv_nc_node *out_nc_node,
+ struct sk_buff *skb,
+ uint8_t *eth_dst)
+{
+ struct batadv_nc_path *nc_path, nc_path_key;
+ struct batadv_nc_packet *nc_packet_out = NULL;
+ struct batadv_nc_packet *nc_packet, *nc_packet_tmp;
+ struct batadv_hashtable *hash = bat_priv->nc.coding_hash;
+ int idx;
+
+ if (!hash)
+ return NULL;
+
+ /* Create almost path key */
+ batadv_nc_hash_key_gen(&nc_path_key, in_nc_node->addr,
+ out_nc_node->addr);
+ idx = batadv_nc_hash_choose(&nc_path_key, hash->size);
+
+ /* Check for coding opportunities in this nc_path */
+ rcu_read_lock();
+ hlist_for_each_entry_rcu(nc_path, &hash->table[idx], hash_entry) {
+ if (!batadv_compare_eth(nc_path->prev_hop, in_nc_node->addr))
+ continue;
+
+ if (!batadv_compare_eth(nc_path->next_hop, out_nc_node->addr))
+ continue;
+
+ spin_lock_bh(&nc_path->packet_list_lock);
+ if (list_empty(&nc_path->packet_list)) {
+ spin_unlock_bh(&nc_path->packet_list_lock);
+ continue;
+ }
+
+ list_for_each_entry_safe(nc_packet, nc_packet_tmp,
+ &nc_path->packet_list, list) {
+ if (!batadv_nc_skb_coding_possible(nc_packet->skb,
+ eth_dst,
+ in_nc_node->addr))
+ continue;
+
+ /* Coding opportunity is found! */
+ list_del(&nc_packet->list);
+ nc_packet_out = nc_packet;
+ break;
+ }
+
+ spin_unlock_bh(&nc_path->packet_list_lock);
+ break;
+ }
+ rcu_read_unlock();
+
+ return nc_packet_out;
+}
+
+/**
+ * batadv_nc_skb_src_search - Loops through the list of neighoring nodes of the
+ * skb's sender (may be equal to the originator).
+ * @bat_priv: the bat priv with all the soft interface information
+ * @skb: data skb to forward
+ * @eth_dst: next hop mac address of skb
+ * @eth_src: source mac address of skb
+ * @in_nc_node: pointer to skb next hop's neighbor nc node
+ *
+ * Returns an nc packet if a suitable coding packet was found, NULL otherwise.
+ */
+static struct batadv_nc_packet *
+batadv_nc_skb_src_search(struct batadv_priv *bat_priv,
+ struct sk_buff *skb,
+ uint8_t *eth_dst,
+ uint8_t *eth_src,
+ struct batadv_nc_node *in_nc_node)
+{
+ struct batadv_orig_node *orig_node;
+ struct batadv_nc_node *out_nc_node;
+ struct batadv_nc_packet *nc_packet = NULL;
+
+ orig_node = batadv_orig_hash_find(bat_priv, eth_src);
+ if (!orig_node)
+ return NULL;
+
+ rcu_read_lock();
+ list_for_each_entry_rcu(out_nc_node,
+ &orig_node->out_coding_list, list) {
+ /* Check if the skb is decoded and if recoding is possible */
+ if (!batadv_nc_skb_coding_possible(skb,
+ out_nc_node->addr, eth_src))
+ continue;
+
+ /* Search for an opportunity in this nc_path */
+ nc_packet = batadv_nc_path_search(bat_priv, in_nc_node,
+ out_nc_node, skb, eth_dst);
+ if (nc_packet)
+ break;
+ }
+ rcu_read_unlock();
+
+ batadv_orig_node_free_ref(orig_node);
+ return nc_packet;
+}
+
+/**
+ * batadv_nc_skb_dst_search - Loops through list of neighboring nodes to dst.
+ * @skb: data skb to forward
+ * @neigh_node: next hop to forward packet to
+ * @ethhdr: pointer to the ethernet header inside the skb
+ *
+ * Loops through list of neighboring nodes the next hop has a good connection to
+ * (receives OGMs with a sufficient quality). We need to find a neighbor of our
+ * next hop that potentially sent a packet which our next hop also received
+ * (overheard) and has stored for later decoding.
+ *
+ * Returns true if the skb was consumed (encoded packet sent) or false otherwise
+ */
+static bool batadv_nc_skb_dst_search(struct sk_buff *skb,
+ struct batadv_neigh_node *neigh_node,
+ struct ethhdr *ethhdr)
+{
+ struct net_device *netdev = neigh_node->if_incoming->soft_iface;
+ struct batadv_priv *bat_priv = netdev_priv(netdev);
+ struct batadv_orig_node *orig_node = neigh_node->orig_node;
+ struct batadv_nc_node *nc_node;
+ struct batadv_nc_packet *nc_packet = NULL;
+
+ rcu_read_lock();
+ list_for_each_entry_rcu(nc_node, &orig_node->in_coding_list, list) {
+ /* Search for coding opportunity with this in_nc_node */
+ nc_packet = batadv_nc_skb_src_search(bat_priv, skb,
+ neigh_node->addr,
+ ethhdr->h_source, nc_node);
+
+ /* Opportunity was found, so stop searching */
+ if (nc_packet)
+ break;
+ }
+ rcu_read_unlock();
+
+ if (!nc_packet)
+ return false;
+
+ /* Code and send packets */
+ if (batadv_nc_code_packets(bat_priv, skb, ethhdr, nc_packet,
+ neigh_node))
+ return true;
+
+ /* out of mem ? Coding failed - we have to free the buffered packet
+ * to avoid memleaks. The skb passed as argument will be dealt with
+ * by the calling function.
+ */
+ batadv_nc_send_packet(nc_packet);
+ return false;
+}
+
+/**
* batadv_nc_skb_add_to_path - buffer skb for later encoding / decoding
* @skb: skb to add to path
* @nc_path: path to add skb to
@@ -862,6 +1259,10 @@ bool batadv_nc_skb_forward(struct sk_buff *skb,
if (packet->header.packet_type != BATADV_UNICAST)
goto out;
+ /* Try to find a coding opportunity and send the skb if one is found */
+ if (batadv_nc_skb_dst_search(skb, neigh_node, ethhdr))
+ return true;
+
/* Find or create a nc_path for this src-dst pair */
nc_path = batadv_nc_get_path(bat_priv,
bat_priv->nc.coding_hash,
diff --git a/net/batman-adv/packet.h b/net/batman-adv/packet.h
index ed0aa89..a079958 100644
--- a/net/batman-adv/packet.h
+++ b/net/batman-adv/packet.h
@@ -30,6 +30,7 @@ enum batadv_packettype {
BATADV_TT_QUERY = 0x07,
BATADV_ROAM_ADV = 0x08,
BATADV_UNICAST_4ADDR = 0x09,
+ BATADV_CODED = 0x0a,
};
/**
@@ -278,4 +279,36 @@ struct batadv_tt_change {
uint8_t addr[ETH_ALEN];
} __packed;
+/**
+ * struct batadv_coded_packet - network coded packet
+ * @header: common batman packet header and ttl of first included packet
+ * @reserved: Align following fields to 2-byte boundaries
+ * @first_source: original source of first included packet
+ * @first_orig_dest: original destinal of first included packet
+ * @first_crc: checksum of first included packet
+ * @first_ttvn: tt-version number of first included packet
+ * @second_ttl: ttl of second packet
+ * @second_dest: second receiver of this coded packet
+ * @second_source: original source of second included packet
+ * @second_orig_dest: original destination of second included packet
+ * @second_crc: checksum of second included packet
+ * @second_ttvn: tt version number of second included packet
+ * @coded_len: length of network coded part of the payload
+ */
+struct batadv_coded_packet {
+ struct batadv_header header;
+ uint8_t first_ttvn;
+ /* uint8_t first_dest[ETH_ALEN]; - saved in mac header destination */
+ uint8_t first_source[ETH_ALEN];
+ uint8_t first_orig_dest[ETH_ALEN];
+ __be32 first_crc;
+ uint8_t second_ttl;
+ uint8_t second_ttvn;
+ uint8_t second_dest[ETH_ALEN];
+ uint8_t second_source[ETH_ALEN];
+ uint8_t second_orig_dest[ETH_ALEN];
+ __be32 second_crc;
+ uint16_t coded_len;
+};
+
#endif /* _NET_BATMAN_ADV_PACKET_H_ */
diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c
index 7188e07..7e463c3 100644
--- a/net/batman-adv/soft-interface.c
+++ b/net/batman-adv/soft-interface.c
@@ -665,6 +665,12 @@ static const struct {
{ "dat_put_rx" },
{ "dat_cached_reply_tx" },
#endif
+#ifdef CONFIG_BATMAN_ADV_NC
+ { "nc_code" },
+ { "nc_code_bytes" },
+ { "nc_recode" },
+ { "nc_recode_bytes" },
+#endif
};
static void batadv_get_strings(struct net_device *dev, uint32_t stringset,
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 43caf7f..42d7438 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -275,6 +275,10 @@ struct batadv_bcast_duplist_entry {
* @BATADV_CNT_DAT_PUT_RX: received dht PUT traffic packet counter
* @BATADV_CNT_DAT_CACHED_REPLY_TX: transmitted dat cache reply traffic packet
* counter
+ * @BATADV_CNT_NC_CODE: transmitted nc-combined traffic packet counter
+ * @BATADV_CNT_NC_CODE_BYTES: transmitted nc-combined traffic bytes counter
+ * @BATADV_CNT_NC_RECODE: transmitted nc-recombined traffic packet counter
+ * @BATADV_CNT_NC_RECODE_BYTES: transmitted nc-recombined traffic bytes counter
* @BATADV_CNT_NUM: number of traffic counters
*/
enum batadv_counters {
@@ -302,6 +306,12 @@ enum batadv_counters {
BATADV_CNT_DAT_PUT_RX,
BATADV_CNT_DAT_CACHED_REPLY_TX,
#endif
+#ifdef CONFIG_BATMAN_ADV_NC
+ BATADV_CNT_NC_CODE,
+ BATADV_CNT_NC_CODE_BYTES,
+ BATADV_CNT_NC_RECODE,
+ BATADV_CNT_NC_RECODE_BYTES,
+#endif
BATADV_CNT_NUM,
};
@@ -797,6 +807,16 @@ struct batadv_nc_packet {
};
/**
+ * batadv_skb_cb - control buffer structure used to store private data relevant
+ * to batman-adv in the skb->cb buffer in skbs.
+ * @decoded: Marks a skb as decoded, which is checked when searching for coding
+ * opportunities in network-coding.c
+ */
+struct batadv_skb_cb {
+ bool decoded;
+};
+
+/**
* struct batadv_forw_packet - structure for bcast packets to be sent/forwarded
* @list: list node for batadv_socket_client::queue_list
* @send_time: execution time for delayed_work (packet sending)
--
1.8.1.5
^ permalink raw reply related
* [PATCH 8/9] batman-adv: network coding - save overheard and tx packets for decoding
From: Antonio Quartulli @ 2013-03-13 22:56 UTC (permalink / raw)
To: davem
Cc: netdev, b.a.t.m.a.n, Martin Hundebøll, Marek Lindner,
Antonio Quartulli
In-Reply-To: <1363215399-18875-1-git-send-email-ordex@autistici.org>
From: Martin Hundebøll <martin@hundeboll.net>
To be able to decode a network coded packet, a node must already know
one of the two coded packets. This is done by buffering skbs before
transmission and buffering packets sniffed with promiscuous mode from
other hosts.
Packets are kept in a buffer similar to the one with forward-skbs: A
hash table, where each entry, which corresponds to a src-dst pair, has a
linked list packets.
Signed-off-by: Martin Hundebøll <martin@hundeboll.net>
Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
net/batman-adv/network-coding.c | 208 +++++++++++++++++++++++++++++++++++++++-
net/batman-adv/network-coding.h | 18 ++++
net/batman-adv/routing.c | 13 ++-
net/batman-adv/send.c | 5 +
net/batman-adv/soft-interface.c | 1 +
net/batman-adv/types.h | 11 +++
6 files changed, 253 insertions(+), 3 deletions(-)
diff --git a/net/batman-adv/network-coding.c b/net/batman-adv/network-coding.c
index fce2846..3d2ed2f 100644
--- a/net/batman-adv/network-coding.c
+++ b/net/batman-adv/network-coding.c
@@ -27,6 +27,7 @@
#include "hard-interface.h"
static struct lock_class_key batadv_nc_coding_hash_lock_class_key;
+static struct lock_class_key batadv_nc_decoding_hash_lock_class_key;
static void batadv_nc_worker(struct work_struct *work);
@@ -47,8 +48,9 @@ static void batadv_nc_start_timer(struct batadv_priv *bat_priv)
int batadv_nc_init(struct batadv_priv *bat_priv)
{
bat_priv->nc.timestamp_fwd_flush = jiffies;
+ bat_priv->nc.timestamp_sniffed_purge = jiffies;
- if (bat_priv->nc.coding_hash)
+ if (bat_priv->nc.coding_hash || bat_priv->nc.decoding_hash)
return 0;
bat_priv->nc.coding_hash = batadv_hash_new(128);
@@ -58,6 +60,13 @@ int batadv_nc_init(struct batadv_priv *bat_priv)
batadv_hash_set_lock_class(bat_priv->nc.coding_hash,
&batadv_nc_coding_hash_lock_class_key);
+ bat_priv->nc.decoding_hash = batadv_hash_new(128);
+ if (!bat_priv->nc.decoding_hash)
+ goto err;
+
+ batadv_hash_set_lock_class(bat_priv->nc.coding_hash,
+ &batadv_nc_decoding_hash_lock_class_key);
+
INIT_DELAYED_WORK(&bat_priv->nc.work, batadv_nc_worker);
batadv_nc_start_timer(bat_priv);
@@ -76,6 +85,7 @@ void batadv_nc_init_bat_priv(struct batadv_priv *bat_priv)
atomic_set(&bat_priv->network_coding, 1);
bat_priv->nc.min_tq = 200;
bat_priv->nc.max_fwd_delay = 10;
+ bat_priv->nc.max_buffer_time = 200;
}
/**
@@ -176,6 +186,26 @@ static bool batadv_nc_to_purge_nc_path_coding(struct batadv_priv *bat_priv,
}
/**
+ * batadv_nc_to_purge_nc_path_decoding - checks whether an nc path has timed out
+ * @bat_priv: the bat priv with all the soft interface information
+ * @nc_path: the nc path to check
+ *
+ * Returns true if the entry has to be purged now, false otherwise
+ */
+static bool batadv_nc_to_purge_nc_path_decoding(struct batadv_priv *bat_priv,
+ struct batadv_nc_path *nc_path)
+{
+ if (atomic_read(&bat_priv->mesh_state) != BATADV_MESH_ACTIVE)
+ return true;
+
+ /* purge the path when no packets has been added for 10 times the
+ * max_buffer time
+ */
+ return batadv_has_timed_out(nc_path->last_valid,
+ bat_priv->nc.max_buffer_time*10);
+}
+
+/**
* batadv_nc_purge_orig_nc_nodes - go through list of nc nodes and purge stale
* entries
* @bat_priv: the bat priv with all the soft interface information
@@ -441,6 +471,43 @@ static void batadv_nc_send_packet(struct batadv_nc_packet *nc_packet)
}
/**
+ * batadv_nc_sniffed_purge - Checks timestamp of given sniffed nc_packet.
+ * @bat_priv: the bat priv with all the soft interface information
+ * @nc_path: the nc path the packet belongs to
+ * @nc_packet: the nc packet to be checked
+ *
+ * Checks whether the given sniffed (overheard) nc_packet has hit its buffering
+ * timeout. If so, the packet is no longer kept and the entry deleted from the
+ * queue. Has to be called with the appropriate locks.
+ *
+ * Returns false as soon as the entry in the fifo queue has not been timed out
+ * yet and true otherwise.
+ */
+static bool batadv_nc_sniffed_purge(struct batadv_priv *bat_priv,
+ struct batadv_nc_path *nc_path,
+ struct batadv_nc_packet *nc_packet)
+{
+ unsigned long timeout = bat_priv->nc.max_buffer_time;
+ bool res = false;
+
+ /* Packets are added to tail, so the remaining packets did not time
+ * out and we can stop processing the current queue
+ */
+ if (atomic_read(&bat_priv->mesh_state) == BATADV_MESH_ACTIVE &&
+ !batadv_has_timed_out(nc_packet->timestamp, timeout))
+ goto out;
+
+ /* purge nc packet */
+ list_del(&nc_packet->list);
+ batadv_nc_packet_free(nc_packet);
+
+ res = true;
+
+out:
+ return res;
+}
+
+/**
* batadv_nc_fwd_flush - Checks the timestamp of the given nc packet.
* @bat_priv: the bat priv with all the soft interface information
* @nc_path: the nc path the packet belongs to
@@ -540,6 +607,8 @@ static void batadv_nc_worker(struct work_struct *work)
batadv_nc_purge_orig_hash(bat_priv);
batadv_nc_purge_paths(bat_priv, bat_priv->nc.coding_hash,
batadv_nc_to_purge_nc_path_coding);
+ batadv_nc_purge_paths(bat_priv, bat_priv->nc.decoding_hash,
+ batadv_nc_to_purge_nc_path_decoding);
timeout = bat_priv->nc.max_fwd_delay;
@@ -549,6 +618,13 @@ static void batadv_nc_worker(struct work_struct *work)
bat_priv->nc.timestamp_fwd_flush = jiffies;
}
+ if (batadv_has_timed_out(bat_priv->nc.timestamp_sniffed_purge,
+ bat_priv->nc.max_buffer_time)) {
+ batadv_nc_process_nc_paths(bat_priv, bat_priv->nc.decoding_hash,
+ batadv_nc_sniffed_purge);
+ bat_priv->nc.timestamp_sniffed_purge = jiffies;
+ }
+
/* Schedule a new check */
batadv_nc_start_timer(bat_priv);
}
@@ -1143,6 +1219,41 @@ batadv_nc_skb_src_search(struct batadv_priv *bat_priv,
}
/**
+ * batadv_nc_skb_store_before_coding - set the ethernet src and dst of the
+ * unicast skb before it is stored for use in later decoding
+ * @bat_priv: the bat priv with all the soft interface information
+ * @skb: data skb to store
+ * @eth_dst_new: new destination mac address of skb
+ */
+static void batadv_nc_skb_store_before_coding(struct batadv_priv *bat_priv,
+ struct sk_buff *skb,
+ uint8_t *eth_dst_new)
+{
+ struct ethhdr *ethhdr;
+
+ /* Copy skb header to change the mac header */
+ skb = pskb_copy(skb, GFP_ATOMIC);
+ if (!skb)
+ return;
+
+ /* Set the mac header as if we actually sent the packet uncoded */
+ ethhdr = (struct ethhdr *)skb_mac_header(skb);
+ memcpy(ethhdr->h_source, ethhdr->h_dest, ETH_ALEN);
+ memcpy(ethhdr->h_dest, eth_dst_new, ETH_ALEN);
+
+ /* Set data pointer to MAC header to mimic packets from our tx path */
+ skb_push(skb, ETH_HLEN);
+
+ /* Add the packet to the decoding packet pool */
+ batadv_nc_skb_store_for_decoding(bat_priv, skb);
+
+ /* batadv_nc_skb_store_for_decoding() clones the skb, so we must free
+ * our ref
+ */
+ kfree_skb(skb);
+}
+
+/**
* batadv_nc_skb_dst_search - Loops through list of neighboring nodes to dst.
* @skb: data skb to forward
* @neigh_node: next hop to forward packet to
@@ -1181,6 +1292,12 @@ static bool batadv_nc_skb_dst_search(struct sk_buff *skb,
if (!nc_packet)
return false;
+ /* Save packets for later decoding */
+ batadv_nc_skb_store_before_coding(bat_priv, skb,
+ neigh_node->addr);
+ batadv_nc_skb_store_before_coding(bat_priv, nc_packet->skb,
+ nc_packet->neigh_node->addr);
+
/* Code and send packets */
if (batadv_nc_code_packets(bat_priv, skb, ethhdr, nc_packet,
neigh_node))
@@ -1288,14 +1405,98 @@ out:
}
/**
+ * batadv_nc_skb_store_for_decoding - save a clone of the skb which can be used
+ * when decoding coded packets
+ * @bat_priv: the bat priv with all the soft interface information
+ * @skb: data skb to store
+ */
+void batadv_nc_skb_store_for_decoding(struct batadv_priv *bat_priv,
+ struct sk_buff *skb)
+{
+ struct batadv_unicast_packet *packet;
+ struct batadv_nc_path *nc_path;
+ struct ethhdr *ethhdr = (struct ethhdr *)skb_mac_header(skb);
+ __be32 packet_id;
+ u8 *payload;
+
+ /* Check if network coding is enabled */
+ if (!atomic_read(&bat_priv->network_coding))
+ goto out;
+
+ /* Check for supported packet type */
+ payload = skb_network_header(skb);
+ packet = (struct batadv_unicast_packet *)payload;
+ if (packet->header.packet_type != BATADV_UNICAST)
+ goto out;
+
+ /* Find existing nc_path or create a new */
+ nc_path = batadv_nc_get_path(bat_priv,
+ bat_priv->nc.decoding_hash,
+ ethhdr->h_source,
+ ethhdr->h_dest);
+
+ if (!nc_path)
+ goto out;
+
+ /* Clone skb and adjust skb->data to point at batman header */
+ skb = skb_clone(skb, GFP_ATOMIC);
+ if (unlikely(!skb))
+ goto free_nc_path;
+
+ if (unlikely(!pskb_may_pull(skb, ETH_HLEN)))
+ goto free_skb;
+
+ if (unlikely(!skb_pull_rcsum(skb, ETH_HLEN)))
+ goto free_skb;
+
+ /* Add skb to nc_path */
+ packet_id = batadv_skb_crc32(skb, payload + sizeof(*packet));
+ if (!batadv_nc_skb_add_to_path(skb, nc_path, NULL, packet_id))
+ goto free_skb;
+
+ batadv_inc_counter(bat_priv, BATADV_CNT_NC_BUFFER);
+ return;
+
+free_skb:
+ kfree_skb(skb);
+free_nc_path:
+ batadv_nc_path_free_ref(nc_path);
+out:
+ return;
+}
+
+/**
+ * batadv_nc_skb_store_sniffed_unicast - check if a received unicast packet
+ * should be saved in the decoding buffer and, if so, store it there
+ * @bat_priv: the bat priv with all the soft interface information
+ * @skb: unicast skb to store
+ */
+void batadv_nc_skb_store_sniffed_unicast(struct batadv_priv *bat_priv,
+ struct sk_buff *skb)
+{
+ struct ethhdr *ethhdr = (struct ethhdr *)skb_mac_header(skb);
+
+ if (batadv_is_my_mac(ethhdr->h_dest))
+ return;
+
+ /* Set data pointer to MAC header to mimic packets from our tx path */
+ skb_push(skb, ETH_HLEN);
+
+ batadv_nc_skb_store_for_decoding(bat_priv, skb);
+}
+
+/**
* batadv_nc_free - clean up network coding memory
* @bat_priv: the bat priv with all the soft interface information
*/
void batadv_nc_free(struct batadv_priv *bat_priv)
{
cancel_delayed_work_sync(&bat_priv->nc.work);
+
batadv_nc_purge_paths(bat_priv, bat_priv->nc.coding_hash, NULL);
batadv_hash_destroy(bat_priv->nc.coding_hash);
+ batadv_nc_purge_paths(bat_priv, bat_priv->nc.decoding_hash, NULL);
+ batadv_hash_destroy(bat_priv->nc.decoding_hash);
}
/**
@@ -1376,6 +1577,11 @@ int batadv_nc_init_debugfs(struct batadv_priv *bat_priv)
if (!file)
goto out;
+ file = debugfs_create_u32("max_buffer_time", S_IRUGO | S_IWUSR, nc_dir,
+ &bat_priv->nc.max_buffer_time);
+ if (!file)
+ goto out;
+
return 0;
out:
diff --git a/net/batman-adv/network-coding.h b/net/batman-adv/network-coding.h
index c32602a..4fa6d0c 100644
--- a/net/batman-adv/network-coding.h
+++ b/net/batman-adv/network-coding.h
@@ -38,6 +38,10 @@ void batadv_nc_init_orig(struct batadv_orig_node *orig_node);
bool batadv_nc_skb_forward(struct sk_buff *skb,
struct batadv_neigh_node *neigh_node,
struct ethhdr *ethhdr);
+void batadv_nc_skb_store_for_decoding(struct batadv_priv *bat_priv,
+ struct sk_buff *skb);
+void batadv_nc_skb_store_sniffed_unicast(struct batadv_priv *bat_priv,
+ struct sk_buff *skb);
int batadv_nc_nodes_seq_print_text(struct seq_file *seq, void *offset);
int batadv_nc_init_debugfs(struct batadv_priv *bat_priv);
@@ -89,6 +93,20 @@ static inline bool batadv_nc_skb_forward(struct sk_buff *skb,
return false;
}
+static inline void
+batadv_nc_skb_store_for_decoding(struct batadv_priv *bat_priv,
+ struct sk_buff *skb)
+{
+ return;
+}
+
+static inline void
+batadv_nc_skb_store_sniffed_unicast(struct batadv_priv *bat_priv,
+ struct sk_buff *skb)
+{
+ return;
+}
+
static inline int batadv_nc_nodes_seq_print_text(struct seq_file *seq,
void *offset)
{
diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c
index 44fda7c..8f88967 100644
--- a/net/batman-adv/routing.c
+++ b/net/batman-adv/routing.c
@@ -1047,7 +1047,7 @@ int batadv_recv_unicast_packet(struct sk_buff *skb,
struct batadv_unicast_4addr_packet *unicast_4addr_packet;
uint8_t *orig_addr;
struct batadv_orig_node *orig_node = NULL;
- int hdr_size = sizeof(*unicast_packet);
+ int check, hdr_size = sizeof(*unicast_packet);
bool is4addr;
unicast_packet = (struct batadv_unicast_packet *)skb->data;
@@ -1058,7 +1058,16 @@ int batadv_recv_unicast_packet(struct sk_buff *skb,
if (is4addr)
hdr_size = sizeof(*unicast_4addr_packet);
- if (batadv_check_unicast_packet(skb, hdr_size) < 0)
+ /* function returns -EREMOTE for promiscuous packets */
+ check = batadv_check_unicast_packet(skb, hdr_size);
+
+ /* Even though the packet is not for us, we might save it to use for
+ * decoding a later received coded packet
+ */
+ if (check == -EREMOTE)
+ batadv_nc_skb_store_sniffed_unicast(bat_priv, skb);
+
+ if (check < 0)
return NET_RX_DROP;
if (!batadv_check_unicast_ttvn(bat_priv, skb))
diff --git a/net/batman-adv/send.c b/net/batman-adv/send.c
index a67cffd..263cfd1 100644
--- a/net/batman-adv/send.c
+++ b/net/batman-adv/send.c
@@ -27,6 +27,7 @@
#include "vis.h"
#include "gateway_common.h"
#include "originator.h"
+#include "network-coding.h"
#include <linux/if_ether.h>
@@ -39,6 +40,7 @@ int batadv_send_skb_packet(struct sk_buff *skb,
struct batadv_hard_iface *hard_iface,
const uint8_t *dst_addr)
{
+ struct batadv_priv *bat_priv = netdev_priv(hard_iface->soft_iface);
struct ethhdr *ethhdr;
if (hard_iface->if_status != BATADV_IF_ACTIVE)
@@ -70,6 +72,9 @@ int batadv_send_skb_packet(struct sk_buff *skb,
skb->dev = hard_iface->net_dev;
+ /* Save a clone of the skb to use when decoding coded packets */
+ batadv_nc_skb_store_for_decoding(bat_priv, skb);
+
/* dev_queue_xmit() returns a negative result on error. However on
* congestion and traffic shaping, it drops and returns NET_XMIT_DROP
* (which is > 0). This will not be treated as an error.
diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c
index 7e463c3..75204ec 100644
--- a/net/batman-adv/soft-interface.c
+++ b/net/batman-adv/soft-interface.c
@@ -670,6 +670,7 @@ static const struct {
{ "nc_code_bytes" },
{ "nc_recode" },
{ "nc_recode_bytes" },
+ { "nc_buffer" },
#endif
};
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 42d7438..5f3640d 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -279,6 +279,7 @@ struct batadv_bcast_duplist_entry {
* @BATADV_CNT_NC_CODE_BYTES: transmitted nc-combined traffic bytes counter
* @BATADV_CNT_NC_RECODE: transmitted nc-recombined traffic packet counter
* @BATADV_CNT_NC_RECODE_BYTES: transmitted nc-recombined traffic bytes counter
+ * @BATADV_CNT_NC_BUFFER: counter for packets buffered for later nc decoding
* @BATADV_CNT_NUM: number of traffic counters
*/
enum batadv_counters {
@@ -311,6 +312,7 @@ enum batadv_counters {
BATADV_CNT_NC_CODE_BYTES,
BATADV_CNT_NC_RECODE,
BATADV_CNT_NC_RECODE_BYTES,
+ BATADV_CNT_NC_BUFFER,
#endif
BATADV_CNT_NUM,
};
@@ -453,18 +455,27 @@ struct batadv_priv_dat {
* @debug_dir: dentry for nc subdir in batman-adv directory in debugfs
* @min_tq: only consider neighbors for encoding if neigh_tq > min_tq
* @max_fwd_delay: maximum packet forward delay to allow coding of packets
+ * @max_buffer_time: buffer time for sniffed packets used to decoding
* @timestamp_fwd_flush: timestamp of last forward packet queue flush
+ * @timestamp_sniffed_purge: timestamp of last sniffed packet queue purge
* @coding_hash: Hash table used to buffer skbs while waiting for another
* incoming skb to code it with. Skbs are added to the buffer just before being
* forwarded in routing.c
+ * @decoding_hash: Hash table used to buffer skbs that might be needed to decode
+ * a received coded skb. The buffer is used for 1) skbs arriving on the
+ * soft-interface; 2) skbs overheard on the hard-interface; and 3) skbs
+ * forwarded by batman-adv.
*/
struct batadv_priv_nc {
struct delayed_work work;
struct dentry *debug_dir;
u8 min_tq;
u32 max_fwd_delay;
+ u32 max_buffer_time;
unsigned long timestamp_fwd_flush;
+ unsigned long timestamp_sniffed_purge;
struct batadv_hashtable *coding_hash;
+ struct batadv_hashtable *decoding_hash;
};
/**
--
1.8.1.5
^ permalink raw reply related
* [PATCH 9/9] batman-adv: network coding - receive coded packets and decode them
From: Antonio Quartulli @ 2013-03-13 22:56 UTC (permalink / raw)
To: davem
Cc: netdev, b.a.t.m.a.n, Martin Hundebøll, Marek Lindner,
Antonio Quartulli
In-Reply-To: <1363215399-18875-1-git-send-email-ordex@autistici.org>
From: Martin Hundebøll <martin@hundeboll.net>
When receiving a network coded packet, the decoding buffer is searched
for a packet to use for decoding. The source, destination, and crc32 from
the coded packet is used to identify the wanted packet. The decoded
packet is passed to the usual unicast receiver function, as had it never
been network coded.
Signed-off-by: Martin Hundebøll <martin@hundeboll.net>
Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
Signed-off-by: Antonio Quartulli <ordex@autistici.org>
---
net/batman-adv/network-coding.c | 232 ++++++++++++++++++++++++++++++++++++++++
net/batman-adv/soft-interface.c | 4 +
net/batman-adv/types.h | 10 ++
3 files changed, 246 insertions(+)
diff --git a/net/batman-adv/network-coding.c b/net/batman-adv/network-coding.c
index 3d2ed2f..5728079 100644
--- a/net/batman-adv/network-coding.c
+++ b/net/batman-adv/network-coding.c
@@ -25,11 +25,14 @@
#include "send.h"
#include "originator.h"
#include "hard-interface.h"
+#include "routing.h"
static struct lock_class_key batadv_nc_coding_hash_lock_class_key;
static struct lock_class_key batadv_nc_decoding_hash_lock_class_key;
static void batadv_nc_worker(struct work_struct *work);
+static int batadv_nc_recv_coded_packet(struct sk_buff *skb,
+ struct batadv_hard_iface *recv_if);
/**
* batadv_nc_start_timer - initialise the nc periodic worker
@@ -67,6 +70,11 @@ int batadv_nc_init(struct batadv_priv *bat_priv)
batadv_hash_set_lock_class(bat_priv->nc.coding_hash,
&batadv_nc_decoding_hash_lock_class_key);
+ /* Register our packet type */
+ if (batadv_recv_handler_register(BATADV_CODED,
+ batadv_nc_recv_coded_packet) < 0)
+ goto err;
+
INIT_DELAYED_WORK(&bat_priv->nc.work, batadv_nc_worker);
batadv_nc_start_timer(bat_priv);
@@ -1486,11 +1494,235 @@ void batadv_nc_skb_store_sniffed_unicast(struct batadv_priv *bat_priv,
}
/**
+ * batadv_nc_skb_decode_packet - decode given skb using the decode data stored
+ * in nc_packet
+ * @skb: unicast skb to decode
+ * @nc_packet: decode data needed to decode the skb
+ *
+ * Returns pointer to decoded unicast packet if the packet was decoded or NULL
+ * in case of an error.
+ */
+static struct batadv_unicast_packet *
+batadv_nc_skb_decode_packet(struct sk_buff *skb,
+ struct batadv_nc_packet *nc_packet)
+{
+ const int h_size = sizeof(struct batadv_unicast_packet);
+ const int h_diff = sizeof(struct batadv_coded_packet) - h_size;
+ struct batadv_unicast_packet *unicast_packet;
+ struct batadv_coded_packet coded_packet_tmp;
+ struct ethhdr *ethhdr, ethhdr_tmp;
+ uint8_t *orig_dest, ttl, ttvn;
+ unsigned int coding_len;
+
+ /* Save headers temporarily */
+ memcpy(&coded_packet_tmp, skb->data, sizeof(coded_packet_tmp));
+ memcpy(ðhdr_tmp, skb_mac_header(skb), sizeof(ethhdr_tmp));
+
+ if (skb_cow(skb, 0) < 0)
+ return NULL;
+
+ if (unlikely(!skb_pull_rcsum(skb, h_diff)))
+ return NULL;
+
+ /* Data points to batman header, so set mac header 14 bytes before
+ * and network to data
+ */
+ skb_set_mac_header(skb, -ETH_HLEN);
+ skb_reset_network_header(skb);
+
+ /* Reconstruct original mac header */
+ ethhdr = (struct ethhdr *)skb_mac_header(skb);
+ memcpy(ethhdr, ðhdr_tmp, sizeof(*ethhdr));
+
+ /* Select the correct unicast header information based on the location
+ * of our mac address in the coded_packet header
+ */
+ if (batadv_is_my_mac(coded_packet_tmp.second_dest)) {
+ /* If we are the second destination the packet was overheard,
+ * so the Ethernet address must be copied to h_dest and
+ * pkt_type changed from PACKET_OTHERHOST to PACKET_HOST
+ */
+ memcpy(ethhdr->h_dest, coded_packet_tmp.second_dest, ETH_ALEN);
+ skb->pkt_type = PACKET_HOST;
+
+ orig_dest = coded_packet_tmp.second_orig_dest;
+ ttl = coded_packet_tmp.second_ttl;
+ ttvn = coded_packet_tmp.second_ttvn;
+ } else {
+ orig_dest = coded_packet_tmp.first_orig_dest;
+ ttl = coded_packet_tmp.header.ttl;
+ ttvn = coded_packet_tmp.first_ttvn;
+ }
+
+ coding_len = ntohs(coded_packet_tmp.coded_len);
+
+ if (coding_len > skb->len)
+ return NULL;
+
+ /* Here the magic is reversed:
+ * extract the missing packet from the received coded packet
+ */
+ batadv_nc_memxor(skb->data + h_size,
+ nc_packet->skb->data + h_size,
+ coding_len);
+
+ /* Resize decoded skb if decoded with larger packet */
+ if (nc_packet->skb->len > coding_len + h_size)
+ pskb_trim_rcsum(skb, coding_len + h_size);
+
+ /* Create decoded unicast packet */
+ unicast_packet = (struct batadv_unicast_packet *)skb->data;
+ unicast_packet->header.packet_type = BATADV_UNICAST;
+ unicast_packet->header.version = BATADV_COMPAT_VERSION;
+ unicast_packet->header.ttl = ttl;
+ memcpy(unicast_packet->dest, orig_dest, ETH_ALEN);
+ unicast_packet->ttvn = ttvn;
+
+ batadv_nc_packet_free(nc_packet);
+ return unicast_packet;
+}
+
+/**
+ * batadv_nc_find_decoding_packet - search through buffered decoding data to
+ * find the data needed to decode the coded packet
+ * @bat_priv: the bat priv with all the soft interface information
+ * @ethhdr: pointer to the ethernet header inside the coded packet
+ * @coded: coded packet we try to find decode data for
+ *
+ * Returns pointer to nc packet if the needed data was found or NULL otherwise.
+ */
+static struct batadv_nc_packet *
+batadv_nc_find_decoding_packet(struct batadv_priv *bat_priv,
+ struct ethhdr *ethhdr,
+ struct batadv_coded_packet *coded)
+{
+ struct batadv_hashtable *hash = bat_priv->nc.decoding_hash;
+ struct batadv_nc_packet *tmp_nc_packet, *nc_packet = NULL;
+ struct batadv_nc_path *nc_path, nc_path_key;
+ uint8_t *dest, *source;
+ __be32 packet_id;
+ int index;
+
+ if (!hash)
+ return NULL;
+
+ /* Select the correct packet id based on the location of our mac-addr */
+ dest = ethhdr->h_source;
+ if (!batadv_is_my_mac(coded->second_dest)) {
+ source = coded->second_source;
+ packet_id = coded->second_crc;
+ } else {
+ source = coded->first_source;
+ packet_id = coded->first_crc;
+ }
+
+ batadv_nc_hash_key_gen(&nc_path_key, source, dest);
+ index = batadv_nc_hash_choose(&nc_path_key, hash->size);
+
+ /* Search for matching coding path */
+ rcu_read_lock();
+ hlist_for_each_entry_rcu(nc_path, &hash->table[index], hash_entry) {
+ /* Find matching nc_packet */
+ spin_lock_bh(&nc_path->packet_list_lock);
+ list_for_each_entry(tmp_nc_packet,
+ &nc_path->packet_list, list) {
+ if (packet_id == tmp_nc_packet->packet_id) {
+ list_del(&tmp_nc_packet->list);
+
+ nc_packet = tmp_nc_packet;
+ break;
+ }
+ }
+ spin_unlock_bh(&nc_path->packet_list_lock);
+
+ if (nc_packet)
+ break;
+ }
+ rcu_read_unlock();
+
+ if (!nc_packet)
+ batadv_dbg(BATADV_DBG_NC, bat_priv,
+ "No decoding packet found for %u\n", packet_id);
+
+ return nc_packet;
+}
+
+/**
+ * batadv_nc_recv_coded_packet - try to decode coded packet and enqueue the
+ * resulting unicast packet
+ * @skb: incoming coded packet
+ * @recv_if: pointer to interface this packet was received on
+ */
+static int batadv_nc_recv_coded_packet(struct sk_buff *skb,
+ struct batadv_hard_iface *recv_if)
+{
+ struct batadv_priv *bat_priv = netdev_priv(recv_if->soft_iface);
+ struct batadv_unicast_packet *unicast_packet;
+ struct batadv_coded_packet *coded_packet;
+ struct batadv_nc_packet *nc_packet;
+ struct ethhdr *ethhdr;
+ int hdr_size = sizeof(*coded_packet);
+
+ /* Check if network coding is enabled */
+ if (!atomic_read(&bat_priv->network_coding))
+ return NET_RX_DROP;
+
+ /* Make sure we can access (and remove) header */
+ if (unlikely(!pskb_may_pull(skb, hdr_size)))
+ return NET_RX_DROP;
+
+ coded_packet = (struct batadv_coded_packet *)skb->data;
+ ethhdr = (struct ethhdr *)skb_mac_header(skb);
+
+ /* Verify frame is destined for us */
+ if (!batadv_is_my_mac(ethhdr->h_dest) &&
+ !batadv_is_my_mac(coded_packet->second_dest))
+ return NET_RX_DROP;
+
+ /* Update stat counter */
+ if (batadv_is_my_mac(coded_packet->second_dest))
+ batadv_inc_counter(bat_priv, BATADV_CNT_NC_SNIFFED);
+
+ nc_packet = batadv_nc_find_decoding_packet(bat_priv, ethhdr,
+ coded_packet);
+ if (!nc_packet) {
+ batadv_inc_counter(bat_priv, BATADV_CNT_NC_DECODE_FAILED);
+ return NET_RX_DROP;
+ }
+
+ /* Make skb's linear, because decoding accesses the entire buffer */
+ if (skb_linearize(skb) < 0)
+ goto free_nc_packet;
+
+ if (skb_linearize(nc_packet->skb) < 0)
+ goto free_nc_packet;
+
+ /* Decode the packet */
+ unicast_packet = batadv_nc_skb_decode_packet(skb, nc_packet);
+ if (!unicast_packet) {
+ batadv_inc_counter(bat_priv, BATADV_CNT_NC_DECODE_FAILED);
+ goto free_nc_packet;
+ }
+
+ /* Mark packet as decoded to do correct recoding when forwarding */
+ BATADV_SKB_CB(skb)->decoded = true;
+ batadv_inc_counter(bat_priv, BATADV_CNT_NC_DECODE);
+ batadv_add_counter(bat_priv, BATADV_CNT_NC_DECODE_BYTES,
+ skb->len + ETH_HLEN);
+ return batadv_recv_unicast_packet(skb, recv_if);
+
+free_nc_packet:
+ batadv_nc_packet_free(nc_packet);
+ return NET_RX_DROP;
+}
+
+/**
* batadv_nc_free - clean up network coding memory
* @bat_priv: the bat priv with all the soft interface information
*/
void batadv_nc_free(struct batadv_priv *bat_priv)
{
+ batadv_recv_handler_unregister(BATADV_CODED);
cancel_delayed_work_sync(&bat_priv->nc.work);
batadv_nc_purge_paths(bat_priv, bat_priv->nc.coding_hash, NULL);
diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c
index 75204ec..f93ae42 100644
--- a/net/batman-adv/soft-interface.c
+++ b/net/batman-adv/soft-interface.c
@@ -671,6 +671,10 @@ static const struct {
{ "nc_recode" },
{ "nc_recode_bytes" },
{ "nc_buffer" },
+ { "nc_decode" },
+ { "nc_decode_bytes" },
+ { "nc_decode_failed" },
+ { "nc_sniffed" },
#endif
};
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 5f3640d..aba8364 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -280,6 +280,12 @@ struct batadv_bcast_duplist_entry {
* @BATADV_CNT_NC_RECODE: transmitted nc-recombined traffic packet counter
* @BATADV_CNT_NC_RECODE_BYTES: transmitted nc-recombined traffic bytes counter
* @BATADV_CNT_NC_BUFFER: counter for packets buffered for later nc decoding
+ * @BATADV_CNT_NC_DECODE: received and nc-decoded traffic packet counter
+ * @BATADV_CNT_NC_DECODE_BYTES: received and nc-decoded traffic bytes counter
+ * @BATADV_CNT_NC_DECODE_FAILED: received and decode-failed traffic packet
+ * counter
+ * @BATADV_CNT_NC_SNIFFED: counter for nc-decoded packets received in promisc
+ * mode.
* @BATADV_CNT_NUM: number of traffic counters
*/
enum batadv_counters {
@@ -313,6 +319,10 @@ enum batadv_counters {
BATADV_CNT_NC_RECODE,
BATADV_CNT_NC_RECODE_BYTES,
BATADV_CNT_NC_BUFFER,
+ BATADV_CNT_NC_DECODE,
+ BATADV_CNT_NC_DECODE_BYTES,
+ BATADV_CNT_NC_DECODE_FAILED,
+ BATADV_CNT_NC_SNIFFED,
#endif
BATADV_CNT_NUM,
};
--
1.8.1.5
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox