[ovs-dev] [PATCH 05/13] Upstream VXLAN: Use ip_tunnel structure for vxlan preivate device.
Pravin B Shelar
pshelar at nicira.com
Thu Nov 22 15:56:48 UTC 2012
From: Pravin Shelar <pshelar at nicira.com>
Signed-off-by: Pravin B Shelar <pshelar at nicira.com>
---
drivers/net/vxlan.c | 116 +++++++++++++++++++++++++--------------------------
1 files changed, 57 insertions(+), 59 deletions(-)
diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index ce77b8b..3f94346 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -32,6 +32,7 @@
#include <net/arp.h>
#include <net/ndisc.h>
#include <net/ip.h>
+#include <net/ipip.h>
#include <net/icmp.h>
#include <net/udp.h>
#include <net/rtnetlink.h>
@@ -73,7 +74,7 @@ module_param(log_ecn_error, bool, 0644);
MODULE_PARM_DESC(log_ecn_error, "Log packets received with corrupted ECN");
/* per-net private data for this module */
-static unsigned int vxlan_net_id;
+static int vxlan_net_id;
struct vxlan_net {
struct socket *sock; /* UDP encap socket */
struct hlist_head vni_list[VNI_HASH_SIZE];
@@ -101,17 +102,12 @@ struct vxlan_stats {
/* Pseudo network device */
struct vxlan_dev {
+ struct ip_tunnel tunnel;
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 */
unsigned int link; /* link to multicast over */
__u16 port_min; /* source port range */
__u16 port_max;
- __u8 tos; /* TOS override */
- __u8 ttl;
u32 flags; /* VXLAN_F_* below */
unsigned long age_interval;
@@ -132,21 +128,21 @@ struct vxlan_dev {
/* salt for hash table */
static u32 vxlan_salt __read_mostly;
-static inline struct hlist_head *vni_head(struct net *net, u32 id)
+static inline struct hlist_head *vni_head(struct net *net, __be32 id)
{
struct vxlan_net *vn = net_generic(net, vxlan_net_id);
- return &vn->vni_list[hash_32(id, VNI_HASH_BITS)];
+ return &vn->vni_list[hash_32((__force u32)id, VNI_HASH_BITS)];
}
/* Look up VNI in a per net namespace table */
-static struct vxlan_dev *vxlan_find_vni(struct net *net, u32 id)
+static struct vxlan_dev *vxlan_find_vni(struct net *net, __be32 vni)
{
struct vxlan_dev *vxlan;
struct hlist_node *node;
- hlist_for_each_entry_rcu(vxlan, node, vni_head(net, id), hlist) {
- if (vxlan->vni == id)
+ hlist_for_each_entry_rcu(vxlan, node, vni_head(net, vni), hlist) {
+ if (vxlan->tunnel.parms.i_key == vni)
return vxlan;
}
@@ -180,7 +176,7 @@ static int vxlan_fdb_info(struct sk_buff *skb, struct vxlan_dev *vxlan,
} else
ndm->ndm_family = AF_BRIDGE;
ndm->ndm_state = fdb->state;
- ndm->ndm_ifindex = vxlan->dev->ifindex;
+ ndm->ndm_ifindex = vxlan->tunnel.dev->ifindex;
ndm->ndm_flags = NTF_SELF;
ndm->ndm_type = NDA_DST;
@@ -216,7 +212,7 @@ static inline size_t vxlan_nlmsg_size(void)
static void vxlan_fdb_notify(struct vxlan_dev *vxlan,
const struct vxlan_fdb *fdb, int type)
{
- struct net *net = dev_net(vxlan->dev);
+ struct net *net = dev_net(vxlan->tunnel.dev);
struct sk_buff *skb;
int err = -ENOBUFS;
@@ -311,7 +307,7 @@ static int vxlan_fdb_create(struct vxlan_dev *vxlan,
f = vxlan_find_mac(vxlan, mac);
if (f) {
if (flags & NLM_F_EXCL) {
- netdev_dbg(vxlan->dev,
+ netdev_dbg(vxlan->tunnel.dev,
"lost race to create %pM\n", mac);
return -EEXIST;
}
@@ -327,7 +323,7 @@ static int vxlan_fdb_create(struct vxlan_dev *vxlan,
if (vxlan->addrmax && vxlan->addrcnt >= vxlan->addrmax)
return -ENOSPC;
- netdev_dbg(vxlan->dev, "add %pM -> %pI4\n", mac, &ip);
+ netdev_dbg(vxlan->tunnel.dev, "add %pM -> %pI4\n", mac, &ip);
f = kmalloc(sizeof(*f), GFP_ATOMIC);
if (!f)
return -ENOMEM;
@@ -351,7 +347,7 @@ static int vxlan_fdb_create(struct vxlan_dev *vxlan,
static void vxlan_fdb_destroy(struct vxlan_dev *vxlan, struct vxlan_fdb *f)
{
- netdev_dbg(vxlan->dev,
+ netdev_dbg(vxlan->tunnel.dev,
"delete %pM\n", f->eth_addr);
--vxlan->addrcnt;
@@ -488,10 +484,10 @@ static bool vxlan_group_used(struct vxlan_net *vn,
if (vxlan == this)
continue;
- if (!netif_running(vxlan->dev))
+ if (!netif_running(vxlan->tunnel.dev))
continue;
- if (vxlan->gaddr == this->gaddr)
+ if (vxlan->tunnel.parms.iph.daddr == this->tunnel.parms.iph.daddr)
return true;
}
@@ -505,7 +501,7 @@ static int vxlan_join_group(struct net_device *dev)
struct vxlan_net *vn = net_generic(dev_net(dev), vxlan_net_id);
struct sock *sk = vn->sock->sk;
struct ip_mreqn mreq = {
- .imr_multiaddr.s_addr = vxlan->gaddr,
+ .imr_multiaddr.s_addr = vxlan->tunnel.parms.iph.daddr,
};
int err;
@@ -532,7 +528,7 @@ static int vxlan_leave_group(struct net_device *dev)
int err = 0;
struct sock *sk = vn->sock->sk;
struct ip_mreqn mreq = {
- .imr_multiaddr.s_addr = vxlan->gaddr,
+ .imr_multiaddr.s_addr = vxlan->tunnel.parms.iph.daddr,
};
/* Only leave group when last vxlan is done. */
@@ -556,7 +552,6 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
struct vxlanhdr *vxh;
struct vxlan_dev *vxlan;
struct vxlan_stats *stats;
- __u32 vni;
int err;
/* pop off outer UDP header */
@@ -578,16 +573,15 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
__skb_pull(skb, sizeof(struct vxlanhdr));
/* Is this VNI defined? */
- vni = ntohl(vxh->vx_vni) >> 8;
- vxlan = vxlan_find_vni(sock_net(sk), vni);
+ vxlan = vxlan_find_vni(sock_net(sk), vxh->vx_vni);
if (!vxlan) {
- netdev_dbg(skb->dev, "unknown vni %d\n", vni);
+ netdev_dbg(skb->dev, "unknown vni %d\n", ntohl(vxh->vx_vni) >> 8);
goto drop;
}
if (!pskb_may_pull(skb, ETH_HLEN)) {
- vxlan->dev->stats.rx_length_errors++;
- vxlan->dev->stats.rx_errors++;
+ vxlan->tunnel.dev->stats.rx_length_errors++;
+ vxlan->tunnel.dev->stats.rx_errors++;
goto drop;
}
@@ -595,17 +589,17 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
/* Re-examine inner Ethernet packet */
oip = ip_hdr(skb);
- skb->protocol = eth_type_trans(skb, vxlan->dev);
+ skb->protocol = eth_type_trans(skb, vxlan->tunnel.dev);
/* Ignore packet loops (and multicast echo) */
if (compare_ether_addr(eth_hdr(skb)->h_source,
- vxlan->dev->dev_addr) == 0)
+ vxlan->tunnel.dev->dev_addr) == 0)
goto drop;
if (vxlan->flags & VXLAN_F_LEARN)
vxlan_snoop(skb->dev, oip->saddr, eth_hdr(skb)->h_source);
- __skb_tunnel_rx(skb, vxlan->dev);
+ __skb_tunnel_rx(skb, vxlan->tunnel.dev);
skb_reset_network_header(skb);
skb->ip_summed = CHECKSUM_NONE;
@@ -615,8 +609,8 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
net_info_ratelimited("non-ECT from %pI4 with TOS=%#x\n",
&oip->saddr, oip->tos);
if (err > 1) {
- ++vxlan->dev->stats.rx_frame_errors;
- ++vxlan->dev->stats.rx_errors;
+ ++vxlan->tunnel.dev->stats.rx_frame_errors;
+ ++vxlan->tunnel.dev->stats.rx_errors;
goto drop;
}
}
@@ -845,7 +839,7 @@ static netdev_tx_t vxlan_xmit(struct sk_buff *skb, struct net_device *dev)
f = vxlan_find_mac(vxlan, eth->h_dest);
if (f == NULL) {
did_rsc = false;
- dst = vxlan->gaddr;
+ dst = vxlan->tunnel.parms.iph.daddr;
if (!dst && (vxlan->flags & VXLAN_F_L2MISS) &&
!is_multicast_ether_addr(eth->h_dest))
vxlan_fdb_miss(vxlan, eth->h_dest);
@@ -882,21 +876,21 @@ static netdev_tx_t vxlan_xmit(struct sk_buff *skb, struct net_device *dev)
old_iph = ip_hdr(skb);
- ttl = vxlan->ttl;
+ ttl = vxlan->tunnel.parms.iph.ttl;
if (!ttl && IN_MULTICAST(ntohl(dst)))
ttl = 1;
- tos = vxlan->tos;
+ tos = vxlan->tunnel.parms.iph.tos;
if (tos == 1)
tos = vxlan_get_dsfield(old_iph, skb);
src_port = vxlan_src_port(vxlan, skb);
memset(&fl4, 0, sizeof(fl4));
- fl4.flowi4_oif = vxlan->link;
+ fl4.flowi4_oif = vxlan->tunnel.parms.link;
fl4.flowi4_tos = RT_TOS(tos);
fl4.daddr = dst;
- fl4.saddr = vxlan->saddr;
+ fl4.saddr = vxlan->tunnel.parms.iph.saddr;
rt = ip_route_output_key(dev_net(dev), &fl4);
if (IS_ERR(rt)) {
@@ -920,7 +914,7 @@ static netdev_tx_t vxlan_xmit(struct sk_buff *skb, struct net_device *dev)
vxh = (struct vxlanhdr *) __skb_push(skb, sizeof(*vxh));
vxh->vx_flags = htonl(VXLAN_FLAGS);
- vxh->vx_vni = htonl(vxlan->vni << 8);
+ vxh->vx_vni = vxlan->tunnel.parms.i_key;
__skb_push(skb, sizeof(*uh));
skb_reset_transport_header(skb);
@@ -982,7 +976,7 @@ static void vxlan_cleanup(unsigned long arg)
unsigned long next_timer = jiffies + FDB_AGE_INTERVAL;
unsigned int h;
- if (!netif_running(vxlan->dev))
+ if (!netif_running(vxlan->tunnel.dev))
return;
spin_lock_bh(&vxlan->hash_lock);
@@ -998,7 +992,7 @@ static void vxlan_cleanup(unsigned long arg)
timeout = f->used + vxlan->age_interval * HZ;
if (time_before_eq(timeout, jiffies)) {
- netdev_dbg(vxlan->dev,
+ netdev_dbg(vxlan->tunnel.dev,
"garbage collect %pM\n",
f->eth_addr);
f->state = NUD_STALE;
@@ -1030,7 +1024,7 @@ static int vxlan_open(struct net_device *dev)
struct vxlan_dev *vxlan = netdev_priv(dev);
int err;
- if (vxlan->gaddr) {
+ if (vxlan->tunnel.parms.iph.daddr) {
err = vxlan_join_group(dev);
if (err)
return err;
@@ -1064,7 +1058,7 @@ static int vxlan_stop(struct net_device *dev)
{
struct vxlan_dev *vxlan = netdev_priv(dev);
- if (vxlan->gaddr)
+ if (vxlan->tunnel.parms.iph.daddr)
vxlan_leave_group(dev);
del_timer_sync(&vxlan->age_timer);
@@ -1180,7 +1174,7 @@ static void vxlan_setup(struct net_device *dev)
vxlan->port_min = low;
vxlan->port_max = high;
- vxlan->dev = dev;
+ vxlan->tunnel.dev = dev;
for (h = 0; h < FDB_HASH_SIZE; ++h)
INIT_HLIST_HEAD(&vxlan->fdb_head[h]);
@@ -1259,25 +1253,26 @@ static int vxlan_newlink(struct net *net, struct net_device *dev,
return -EINVAL;
vni = nla_get_u32(data[IFLA_VXLAN_ID]);
- if (vxlan_find_vni(net, vni)) {
+ vxlan->tunnel.parms.i_key = htonl(vni << 8);
+
+ if (vxlan_find_vni(net, vxlan->tunnel.parms.i_key)) {
pr_info("duplicate VNI %u\n", vni);
return -EEXIST;
}
- vxlan->vni = vni;
if (data[IFLA_VXLAN_GROUP])
- vxlan->gaddr = nla_get_be32(data[IFLA_VXLAN_GROUP]);
+ vxlan->tunnel.parms.iph.daddr = nla_get_be32(data[IFLA_VXLAN_GROUP]);
if (data[IFLA_VXLAN_LOCAL])
- vxlan->saddr = nla_get_be32(data[IFLA_VXLAN_LOCAL]);
+ vxlan->tunnel.parms.iph.saddr = nla_get_be32(data[IFLA_VXLAN_LOCAL]);
if (data[IFLA_VXLAN_LINK] &&
- (vxlan->link = nla_get_u32(data[IFLA_VXLAN_LINK]))) {
+ (vxlan->tunnel.parms.link = nla_get_u32(data[IFLA_VXLAN_LINK]))) {
struct net_device *lowerdev
- = __dev_get_by_index(net, vxlan->link);
+ = __dev_get_by_index(net, vxlan->tunnel.parms.link);
if (!lowerdev) {
- pr_info("ifindex %d does not exist\n", vxlan->link);
+ pr_info("ifindex %d does not exist\n", vxlan->tunnel.parms.link);
return -ENODEV;
}
@@ -1290,10 +1285,10 @@ static int vxlan_newlink(struct net *net, struct net_device *dev,
}
if (data[IFLA_VXLAN_TOS])
- vxlan->tos = nla_get_u8(data[IFLA_VXLAN_TOS]);
+ vxlan->tunnel.parms.iph.tos = nla_get_u8(data[IFLA_VXLAN_TOS]);
if (data[IFLA_VXLAN_TTL])
- vxlan->ttl = nla_get_u8(data[IFLA_VXLAN_TTL]);
+ vxlan->tunnel.parms.iph.ttl = nla_get_u8(data[IFLA_VXLAN_TTL]);
if (!data[IFLA_VXLAN_LEARNING] || nla_get_u8(data[IFLA_VXLAN_LEARNING]))
vxlan->flags |= VXLAN_F_LEARN;
@@ -1327,7 +1322,7 @@ static int vxlan_newlink(struct net *net, struct net_device *dev,
err = register_netdevice(dev);
if (!err)
- hlist_add_head_rcu(&vxlan->hlist, vni_head(net, vxlan->vni));
+ hlist_add_head_rcu(&vxlan->hlist, vni_head(net, vxlan->tunnel.parms.i_key));
return err;
}
@@ -1369,20 +1364,23 @@ static int vxlan_fill_info(struct sk_buff *skb, const struct net_device *dev)
.high = htons(vxlan->port_max),
};
- if (nla_put_u32(skb, IFLA_VXLAN_ID, vxlan->vni))
+ if (nla_put_u32(skb, IFLA_VXLAN_ID, ntohl(vxlan->tunnel.parms.i_key) >> 8))
goto nla_put_failure;
- if (vxlan->gaddr && nla_put_be32(skb, IFLA_VXLAN_GROUP, vxlan->gaddr))
+ if (vxlan->tunnel.parms.iph.daddr &&
+ nla_put_be32(skb, IFLA_VXLAN_GROUP, vxlan->tunnel.parms.iph.daddr))
goto nla_put_failure;
- if (vxlan->link && nla_put_u32(skb, IFLA_VXLAN_LINK, vxlan->link))
+ if (vxlan->tunnel.parms.link &&
+ nla_put_u32(skb, IFLA_VXLAN_LINK, vxlan->tunnel.parms.link))
goto nla_put_failure;
- if (vxlan->saddr && nla_put_be32(skb, IFLA_VXLAN_LOCAL, vxlan->saddr))
+ if (vxlan->tunnel.parms.iph.saddr &&
+ nla_put_be32(skb, IFLA_VXLAN_LOCAL, vxlan->tunnel.parms.iph.saddr))
goto nla_put_failure;
- if (nla_put_u8(skb, IFLA_VXLAN_TTL, vxlan->ttl) ||
- nla_put_u8(skb, IFLA_VXLAN_TOS, vxlan->tos) ||
+ if (nla_put_u8(skb, IFLA_VXLAN_TTL, vxlan->tunnel.parms.iph.ttl) ||
+ nla_put_u8(skb, IFLA_VXLAN_TOS, vxlan->tunnel.parms.iph.tos) ||
nla_put_u8(skb, IFLA_VXLAN_LEARNING,
!!(vxlan->flags & VXLAN_F_LEARN)) ||
nla_put_u8(skb, IFLA_VXLAN_PROXY,
--
1.7.1
More information about the dev
mailing list