[IPV6] MROUTE: Support PIM-SM (SSM).
Based on ancient patch by Mickael Hoerdt
<hoerdt@clarinet.u-strasbg.fr>, which is available at
<http://www-r2.u-strasbg.fr/~hoerdt/dev/linux_ipv6_mforwarding/patch-linux-ipv6-mforwarding-0.1a>.
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
diff --git a/include/linux/mroute6.h b/include/linux/mroute6.h
index b921903..f6469fb 100644
--- a/include/linux/mroute6.h
+++ b/include/linux/mroute6.h
@@ -23,6 +23,8 @@
#define MRT6_ADD_MFC (MRT6_BASE+4) /* Add a multicast forwarding entry */
#define MRT6_DEL_MFC (MRT6_BASE+5) /* Delete a multicast forwarding entry */
#define MRT6_VERSION (MRT6_BASE+6) /* Get the kernel multicast version */
+#define MRT6_ASSERT (MRT6_BASE+7) /* Activate PIM assert mode */
+#define MRT6_PIM (MRT6_BASE+8) /* enable PIM code */
#define SIOCGETMIFCNT_IN6 SIOCPROTOPRIVATE /* IP protocol privates */
#define SIOCGETSGCNT_IN6 (SIOCPROTOPRIVATE+1)
@@ -217,6 +219,8 @@
struct mrt6msg {
#define MRT6MSG_NOCACHE 1
+#define MRT6MSG_WRONGMIF 2
+#define MRT6MSG_WHOLEPKT 3 /* used for use level encap */
__u8 im6_mbz; /* must be zero */
__u8 im6_msgtype; /* what type of message */
__u16 im6_mif; /* mif rec'd on */
diff --git a/net/ipv6/Kconfig b/net/ipv6/Kconfig
index 9a2ea81..82f987b 100644
--- a/net/ipv6/Kconfig
+++ b/net/ipv6/Kconfig
@@ -216,3 +216,10 @@
Experimental support for IPv6 multicast forwarding.
If unsure, say N.
+config IPV6_PIMSM_V2
+ bool "IPv6: PIM-SM version 2 support (EXPERIMENTAL)"
+ depends on IPV6_MROUTE
+ ---help---
+ Support for IPv6 PIM multicast routing protocol PIM-SMv2.
+ If unsure, say N.
+
diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c
index 1bdf3c1..2b70774 100644
--- a/net/ipv6/ip6mr.c
+++ b/net/ipv6/ip6mr.c
@@ -54,6 +54,7 @@
#include <net/ipv6.h>
#include <net/ip6_route.h>
#include <linux/mroute6.h>
+#include <linux/pim.h>
#include <net/addrconf.h>
#include <linux/netfilter_ipv6.h>
@@ -75,6 +76,13 @@
#define MIF_EXISTS(idx) (vif6_table[idx].dev != NULL)
+static int mroute_do_assert; /* Set in PIM assert */
+#ifdef CONFIG_IPV6_PIMSM_V2
+static int mroute_do_pim;
+#else
+#define mroute_do_pim 0
+#endif
+
static struct mfc6_cache *mfc6_cache_array[MFC_LINES]; /* Forwarding cache */
static struct mfc6_cache *mfc_unres_queue; /* Queue of unresolved entries */
@@ -97,6 +105,10 @@
static int ip6mr_cache_report(struct sk_buff *pkt, vifi_t vifi, int assert);
static int ip6mr_fill_mroute(struct sk_buff *skb, struct mfc6_cache *c, struct rtmsg *rtm);
+#ifdef CONFIG_IPV6_PIMSM_V2
+static struct inet6_protocol pim6_protocol;
+#endif
+
static struct timer_list ipmr_expire_timer;
@@ -339,6 +351,132 @@
};
#endif
+#ifdef CONFIG_IPV6_PIMSM_V2
+static int reg_vif_num = -1;
+
+static int pim6_rcv(struct sk_buff *skb)
+{
+ struct pimreghdr *pim;
+ struct ipv6hdr *encap;
+ struct net_device *reg_dev = NULL;
+
+ if (!pskb_may_pull(skb, sizeof(*pim) + sizeof(*encap)))
+ goto drop;
+
+ pim = (struct pimreghdr *)skb_transport_header(skb);
+ if (pim->type != ((PIM_VERSION << 4) | PIM_REGISTER) ||
+ (pim->flags & PIM_NULL_REGISTER) ||
+ (ip_compute_csum((void *)pim, sizeof(*pim)) != 0 &&
+ (u16)csum_fold(skb_checksum(skb, 0, skb->len, 0))))
+ goto drop;
+
+ /* check if the inner packet is destined to mcast group */
+ encap = (struct ipv6hdr *)(skb_transport_header(skb) +
+ sizeof(*pim));
+
+ if (!ipv6_addr_is_multicast(&encap->daddr) ||
+ encap->payload_len == 0 ||
+ ntohs(encap->payload_len) + sizeof(*pim) > skb->len)
+ goto drop;
+
+ read_lock(&mrt_lock);
+ if (reg_vif_num >= 0)
+ reg_dev = vif6_table[reg_vif_num].dev;
+ if (reg_dev)
+ dev_hold(reg_dev);
+ read_unlock(&mrt_lock);
+
+ if (reg_dev == NULL)
+ goto drop;
+
+ skb->mac_header = skb->network_header;
+ skb_pull(skb, (u8 *)encap - skb->data);
+ skb_reset_network_header(skb);
+ skb->dev = reg_dev;
+ skb->protocol = htons(ETH_P_IP);
+ skb->ip_summed = 0;
+ skb->pkt_type = PACKET_HOST;
+ dst_release(skb->dst);
+ ((struct net_device_stats *)netdev_priv(reg_dev))->rx_bytes += skb->len;
+ ((struct net_device_stats *)netdev_priv(reg_dev))->rx_packets++;
+ skb->dst = NULL;
+ nf_reset(skb);
+ netif_rx(skb);
+ dev_put(reg_dev);
+ return 0;
+ drop:
+ kfree_skb(skb);
+ return 0;
+}
+
+static struct inet6_protocol pim6_protocol = {
+ .handler = pim6_rcv,
+};
+
+/* Service routines creating virtual interfaces: PIMREG */
+
+static int reg_vif_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+ read_lock(&mrt_lock);
+ ((struct net_device_stats *)netdev_priv(dev))->tx_bytes += skb->len;
+ ((struct net_device_stats *)netdev_priv(dev))->tx_packets++;
+ ip6mr_cache_report(skb, reg_vif_num, MRT6MSG_WHOLEPKT);
+ read_unlock(&mrt_lock);
+ kfree_skb(skb);
+ return 0;
+}
+
+static struct net_device_stats *reg_vif_get_stats(struct net_device *dev)
+{
+ return (struct net_device_stats *)netdev_priv(dev);
+}
+
+static void reg_vif_setup(struct net_device *dev)
+{
+ dev->type = ARPHRD_PIMREG;
+ dev->mtu = 1500 - sizeof(struct ipv6hdr) - 8;
+ dev->flags = IFF_NOARP;
+ dev->hard_start_xmit = reg_vif_xmit;
+ dev->get_stats = reg_vif_get_stats;
+ dev->destructor = free_netdev;
+}
+
+static struct net_device *ip6mr_reg_vif(void)
+{
+ struct net_device *dev;
+ struct inet6_dev *in_dev;
+
+ dev = alloc_netdev(sizeof(struct net_device_stats), "pim6reg",
+ reg_vif_setup);
+
+ if (dev == NULL)
+ return NULL;
+
+ if (register_netdevice(dev)) {
+ free_netdev(dev);
+ return NULL;
+ }
+ dev->iflink = 0;
+
+ in_dev = ipv6_find_idev(dev);
+ if (!in_dev)
+ goto failure;
+
+ if (dev_open(dev))
+ goto failure;
+
+ return dev;
+
+failure:
+ /* allow the register to be completed before unregistering. */
+ rtnl_unlock();
+ rtnl_lock();
+
+ unregister_netdevice(dev);
+ return NULL;
+}
+#endif
+
/*
* Delete a VIF entry
*/
@@ -361,6 +499,11 @@
return -EADDRNOTAVAIL;
}
+#ifdef CONFIG_IPV6_PIMSM_V2
+ if (vifi == reg_vif_num)
+ reg_vif_num = -1;
+#endif
+
if (vifi + 1 == maxvif) {
int tmp;
for (tmp = vifi - 1; tmp >= 0; tmp--) {
@@ -480,6 +623,19 @@
return -EADDRINUSE;
switch (vifc->mif6c_flags) {
+#ifdef CONFIG_IPV6_PIMSM_V2
+ case MIFF_REGISTER:
+ /*
+ * Special Purpose VIF in PIM
+ * All the packets will be sent to the daemon
+ */
+ if (reg_vif_num >= 0)
+ return -EADDRINUSE;
+ dev = ip6mr_reg_vif();
+ if (!dev)
+ return -ENOBUFS;
+ break;
+#endif
case 0:
dev = dev_get_by_index(&init_net, vifc->mif6c_pifi);
if (!dev)
@@ -512,6 +668,10 @@
write_lock_bh(&mrt_lock);
dev_hold(dev);
v->dev = dev;
+#ifdef CONFIG_IPV6_PIMSM_V2
+ if (v->flags & MIFF_REGISTER)
+ reg_vif_num = vifi;
+#endif
if (vifi + 1 > maxvif)
maxvif = vifi + 1;
write_unlock_bh(&mrt_lock);
@@ -599,7 +759,13 @@
struct mrt6msg *msg;
int ret;
- skb = alloc_skb(sizeof(struct ipv6hdr) + sizeof(*msg), GFP_ATOMIC);
+#ifdef CONFIG_IPV6_PIMSM_V2
+ if (assert == MRT6MSG_WHOLEPKT)
+ skb = skb_realloc_headroom(pkt, -skb_network_offset(pkt)
+ +sizeof(*msg));
+ else
+#endif
+ skb = alloc_skb(sizeof(struct ipv6hdr) + sizeof(*msg), GFP_ATOMIC);
if (!skb)
return -ENOBUFS;
@@ -609,6 +775,29 @@
skb->ip_summed = CHECKSUM_UNNECESSARY;
+#ifdef CONFIG_IPV6_PIMSM_V2
+ if (assert == MRT6MSG_WHOLEPKT) {
+ /* Ugly, but we have no choice with this interface.
+ Duplicate old header, fix length etc.
+ And all this only to mangle msg->im6_msgtype and
+ to set msg->im6_mbz to "mbz" :-)
+ */
+ skb_push(skb, -skb_network_offset(pkt));
+
+ skb_push(skb, sizeof(*msg));
+ skb_reset_transport_header(skb);
+ msg = (struct mrt6msg *)skb_transport_header(skb);
+ msg->im6_mbz = 0;
+ msg->im6_msgtype = MRT6MSG_WHOLEPKT;
+ msg->im6_mif = reg_vif_num;
+ msg->im6_pad = 0;
+ ipv6_addr_copy(&msg->im6_src, &ipv6_hdr(pkt)->saddr);
+ ipv6_addr_copy(&msg->im6_dst, &ipv6_hdr(pkt)->daddr);
+
+ skb->ip_summed = CHECKSUM_UNNECESSARY;
+ } else
+#endif
+ {
/*
* Copy the IP header
*/
@@ -635,6 +824,7 @@
skb->ip_summed = CHECKSUM_UNNECESSARY;
skb_pull(skb, sizeof(struct ipv6hdr));
+ }
if (mroute6_socket == NULL) {
kfree_skb(skb);
@@ -1034,6 +1224,44 @@
return ret;
/*
+ * Control PIM assert (to activate pim will activate assert)
+ */
+ case MRT6_ASSERT:
+ {
+ int v;
+ if (get_user(v, (int __user *)optval))
+ return -EFAULT;
+ mroute_do_assert = !!v;
+ return 0;
+ }
+
+#ifdef CONFIG_IPV6_PIMSM_V2
+ case MRT6_PIM:
+ {
+ int v, ret;
+ if (get_user(v, (int __user *)optval))
+ return -EFAULT;
+ v = !!v;
+ rtnl_lock();
+ ret = 0;
+ if (v != mroute_do_pim) {
+ mroute_do_pim = v;
+ mroute_do_assert = v;
+ if (mroute_do_pim)
+ ret = inet6_add_protocol(&pim6_protocol,
+ IPPROTO_PIM);
+ else
+ ret = inet6_del_protocol(&pim6_protocol,
+ IPPROTO_PIM);
+ if (ret < 0)
+ ret = -EAGAIN;
+ }
+ rtnl_unlock();
+ return ret;
+ }
+
+#endif
+ /*
* Spurious command, or MRT_VERSION which you cannot
* set.
*/
@@ -1056,6 +1284,14 @@
case MRT6_VERSION:
val = 0x0305;
break;
+#ifdef CONFIG_IPV6_PIMSM_V2
+ case MRT6_PIM:
+ val = mroute_do_pim;
+ break;
+#endif
+ case MRT6_ASSERT:
+ val = mroute_do_assert;
+ break;
default:
return -ENOPROTOOPT;
}
@@ -1151,6 +1387,18 @@
if (vif->dev == NULL)
goto out_free;
+#ifdef CONFIG_IPV6_PIMSM_V2
+ if (vif->flags & MIFF_REGISTER) {
+ vif->pkt_out++;
+ vif->bytes_out += skb->len;
+ ((struct net_device_stats *)netdev_priv(vif->dev))->tx_bytes += skb->len;
+ ((struct net_device_stats *)netdev_priv(vif->dev))->tx_packets++;
+ ip6mr_cache_report(skb, vifi, MRT6MSG_WHOLEPKT);
+ kfree_skb(skb);
+ return 0;
+ }
+#endif
+
ipv6h = ipv6_hdr(skb);
fl = (struct flowi) {
@@ -1220,6 +1468,30 @@
cache->mfc_un.res.pkt++;
cache->mfc_un.res.bytes += skb->len;
+ /*
+ * Wrong interface: drop packet and (maybe) send PIM assert.
+ */
+ if (vif6_table[vif].dev != skb->dev) {
+ int true_vifi;
+
+ cache->mfc_un.res.wrong_if++;
+ true_vifi = ip6mr_find_vif(skb->dev);
+
+ if (true_vifi >= 0 && mroute_do_assert &&
+ /* pimsm uses asserts, when switching from RPT to SPT,
+ so that we cannot check that packet arrived on an oif.
+ It is bad, but otherwise we would need to move pretty
+ large chunk of pimd to kernel. Ough... --ANK
+ */
+ (mroute_do_pim || cache->mfc_un.res.ttls[true_vifi] < 255) &&
+ time_after(jiffies,
+ cache->mfc_un.res.last_assert + MFC_ASSERT_THRESH)) {
+ cache->mfc_un.res.last_assert = jiffies;
+ ip6mr_cache_report(skb, true_vifi, MRT6MSG_WRONGMIF);
+ }
+ goto dont_forward;
+ }
+
vif6_table[vif].pkt_in++;
vif6_table[vif].bytes_in += skb->len;
@@ -1241,6 +1513,7 @@
return 0;
}
+dont_forward:
kfree_skb(skb);
return 0;
}