ipip.c

/*
 *	Linux NET3:	IP/IP protocol decoder.
 *
 *	Authors:
 *		Sam Lantinga (slouken@cs.ucdavis.edu)  02/01/95
 *
 *	Fixes:
 *		Alan Cox	:	Merged and made usable non modular (its so tiny its silly as
 *					a module taking up 2 pages).
 *		Alan Cox	: 	Fixed bug with 1.3.18 and IPIP not working (now needs to set skb->h.iph)
 *					to keep ip_forward happy.
 *		Alan Cox	:	More fixes for 1.3.21, and firewall fix. Maybe this will work soon 8).
 *		Kai Schulte	:	Fixed #defines for IP_FIREWALL->FIREWALL
 *              David Woodhouse :       Perform some basic ICMP handling.
 *                                      IPIP Routing without decapsulation.
 *              Carlos Picoto   :       GRE over IP support
 *		Alexey Kuznetsov:	Reworked. Really, now it is truncated version of ipv4/ip_gre.c.
 *					I do not want to merge them together.
 *
 *	This program is free software; you can redistribute it and/or
 *	modify it under the terms of the GNU General Public License
 *	as published by the Free Software Foundation; either version
 *	2 of the License, or (at your option) any later version.
 *
 */

/* tunnel.c: an IP tunnel driver

	The purpose of this driver is to provide an IP tunnel through
	which you can tunnel network traffic transparently across subnets.

	This was written by looking at Nick Holloway's dummy driver
	Thanks for the great code!

		-Sam Lantinga	(slouken@cs.ucdavis.edu)  02/01/95

	Minor tweaks:
		Cleaned up the code a little and added some pre-1.3.0 tweaks.
		dev->hard_header/hard_header_len changed to use no headers.
		Comments/bracketing tweaked.
		Made the tunnels use dev->name not tunnel: when error reporting.
		Added tx_dropped stat

		-Alan Cox	(alan@lxorguk.ukuu.org.uk) 21 March 95

	Reworked:
		Changed to tunnel to destination gateway in addition to the
			tunnel's pointopoint address
		Almost completely rewritten
		Note:  There is currently no firewall or ICMP handling done.

		-Sam Lantinga	(slouken@cs.ucdavis.edu) 02/13/96

*/

/* Things I wish I had known when writing the tunnel driver:

	When the tunnel_xmit() function is called, the skb contains the
	packet to be sent (plus a great deal of extra info), and dev
	contains the tunnel device that _we_ are.

	When we are passed a packet, we are expected to fill in the
	source address with our source IP address.

	What is the proper way to allocate, copy and free a buffer?
	After you allocate it, it is a "0 length" chunk of memory
	starting at zero.  If you want to add headers to the buffer
	later, you'll have to call "skb_reserve(skb, amount)" with
	the amount of memory you want reserved.  Then, you call
	"skb_put(skb, amount)" with the amount of space you want in
	the buffer.  skb_put() returns a pointer to the top (#0) of
	that buffer.  skb->len is set to the amount of space you have
	"allocated" with skb_put().  You can then write up to skb->len
	bytes to that buffer.  If you need more, you can call skb_put()
	again with the additional amount of space you need.  You can
	find out how much more space you can allocate by calling
	"skb_tailroom(skb)".
	Now, to add header space, call "skb_push(skb, header_len)".
	This creates space at the beginning of the buffer and returns
	a pointer to this new space.  If later you need to strip a
	header from a buffer, call "skb_pull(skb, header_len)".
	skb_headroom() will return how much space is left at the top
	of the buffer (before the main data).  Remember, this headroom
	space must be reserved before the skb_put() function is called.
	*/

/*
   This version of net/ipv4/ipip.c is cloned of net/ipv4/ip_gre.c

   For comments look at net/ipv4/ip_gre.c --ANK
 */


#include <linux/capability.h>
#include <linux/module.h>
#include <linux/types.h>
#include <linux/kernel.h>
#include <linux/slab.h>
#include <asm/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/netfilter_ipv4.h>
#include <linux/if_ether.h>

#include <net/sock.h>
#include <net/ip.h>
#include <net/icmp.h>
#include <net/ipip.h>
#include <net/inet_ecn.h>
#include <net/xfrm.h>
#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;