Loading...
1/*
2 * Copyright (c) 2007-2012 Nicira, Inc.
3 *
4 * This program is free software; you can redistribute it and/or
5 * modify it under the terms of version 2 of the GNU General Public
6 * License as published by the Free Software Foundation.
7 *
8 * This program is distributed in the hope that it will be useful, but
9 * WITHOUT ANY WARRANTY; without even the implied warranty of
10 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
11 * General Public License for more details.
12 *
13 * You should have received a copy of the GNU General Public License
14 * along with this program; if not, write to the Free Software
15 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
16 * 02110-1301, USA
17 */
18
19#include <linux/if_vlan.h>
20#include <linux/kernel.h>
21#include <linux/netdevice.h>
22#include <linux/etherdevice.h>
23#include <linux/ethtool.h>
24#include <linux/skbuff.h>
25
26#include <net/dst.h>
27#include <net/xfrm.h>
28#include <net/rtnetlink.h>
29
30#include "datapath.h"
31#include "vport-internal_dev.h"
32#include "vport-netdev.h"
33
34struct internal_dev {
35 struct vport *vport;
36};
37
38static struct vport_ops ovs_internal_vport_ops;
39
40static struct internal_dev *internal_dev_priv(struct net_device *netdev)
41{
42 return netdev_priv(netdev);
43}
44
45/* Called with rcu_read_lock_bh. */
46static int internal_dev_xmit(struct sk_buff *skb, struct net_device *netdev)
47{
48 int len, err;
49
50 len = skb->len;
51 rcu_read_lock();
52 err = ovs_vport_receive(internal_dev_priv(netdev)->vport, skb, NULL);
53 rcu_read_unlock();
54
55 if (likely(!err)) {
56 struct pcpu_sw_netstats *tstats = this_cpu_ptr(netdev->tstats);
57
58 u64_stats_update_begin(&tstats->syncp);
59 tstats->tx_bytes += len;
60 tstats->tx_packets++;
61 u64_stats_update_end(&tstats->syncp);
62 } else {
63 netdev->stats.tx_errors++;
64 }
65 return 0;
66}
67
68static int internal_dev_open(struct net_device *netdev)
69{
70 netif_start_queue(netdev);
71 return 0;
72}
73
74static int internal_dev_stop(struct net_device *netdev)
75{
76 netif_stop_queue(netdev);
77 return 0;
78}
79
80static void internal_dev_getinfo(struct net_device *netdev,
81 struct ethtool_drvinfo *info)
82{
83 strlcpy(info->driver, "openvswitch", sizeof(info->driver));
84}
85
86static const struct ethtool_ops internal_dev_ethtool_ops = {
87 .get_drvinfo = internal_dev_getinfo,
88 .get_link = ethtool_op_get_link,
89};
90
91static void internal_dev_destructor(struct net_device *dev)
92{
93 struct vport *vport = ovs_internal_dev_get_vport(dev);
94
95 ovs_vport_free(vport);
96}
97
98static void
99internal_get_stats(struct net_device *dev, struct rtnl_link_stats64 *stats)
100{
101 int i;
102
103 memset(stats, 0, sizeof(*stats));
104 stats->rx_errors = dev->stats.rx_errors;
105 stats->tx_errors = dev->stats.tx_errors;
106 stats->tx_dropped = dev->stats.tx_dropped;
107 stats->rx_dropped = dev->stats.rx_dropped;
108
109 for_each_possible_cpu(i) {
110 const struct pcpu_sw_netstats *percpu_stats;
111 struct pcpu_sw_netstats local_stats;
112 unsigned int start;
113
114 percpu_stats = per_cpu_ptr(dev->tstats, i);
115
116 do {
117 start = u64_stats_fetch_begin_irq(&percpu_stats->syncp);
118 local_stats = *percpu_stats;
119 } while (u64_stats_fetch_retry_irq(&percpu_stats->syncp, start));
120
121 stats->rx_bytes += local_stats.rx_bytes;
122 stats->rx_packets += local_stats.rx_packets;
123 stats->tx_bytes += local_stats.tx_bytes;
124 stats->tx_packets += local_stats.tx_packets;
125 }
126}
127
128static const struct net_device_ops internal_dev_netdev_ops = {
129 .ndo_open = internal_dev_open,
130 .ndo_stop = internal_dev_stop,
131 .ndo_start_xmit = internal_dev_xmit,
132 .ndo_set_mac_address = eth_mac_addr,
133 .ndo_get_stats64 = internal_get_stats,
134};
135
136static struct rtnl_link_ops internal_dev_link_ops __read_mostly = {
137 .kind = "openvswitch",
138};
139
140static void do_setup(struct net_device *netdev)
141{
142 ether_setup(netdev);
143
144 netdev->max_mtu = ETH_MAX_MTU;
145
146 netdev->netdev_ops = &internal_dev_netdev_ops;
147
148 netdev->priv_flags &= ~IFF_TX_SKB_SHARING;
149 netdev->priv_flags |= IFF_LIVE_ADDR_CHANGE | IFF_OPENVSWITCH |
150 IFF_NO_QUEUE;
151 netdev->needs_free_netdev = true;
152 netdev->priv_destructor = internal_dev_destructor;
153 netdev->ethtool_ops = &internal_dev_ethtool_ops;
154 netdev->rtnl_link_ops = &internal_dev_link_ops;
155
156 netdev->features = NETIF_F_LLTX | NETIF_F_SG | NETIF_F_FRAGLIST |
157 NETIF_F_HIGHDMA | NETIF_F_HW_CSUM |
158 NETIF_F_GSO_SOFTWARE | NETIF_F_GSO_ENCAP_ALL;
159
160 netdev->vlan_features = netdev->features;
161 netdev->hw_enc_features = netdev->features;
162 netdev->features |= NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_STAG_TX;
163 netdev->hw_features = netdev->features & ~NETIF_F_LLTX;
164
165 eth_hw_addr_random(netdev);
166}
167
168static struct vport *internal_dev_create(const struct vport_parms *parms)
169{
170 struct vport *vport;
171 struct internal_dev *internal_dev;
172 int err;
173
174 vport = ovs_vport_alloc(0, &ovs_internal_vport_ops, parms);
175 if (IS_ERR(vport)) {
176 err = PTR_ERR(vport);
177 goto error;
178 }
179
180 vport->dev = alloc_netdev(sizeof(struct internal_dev),
181 parms->name, NET_NAME_USER, do_setup);
182 if (!vport->dev) {
183 err = -ENOMEM;
184 goto error_free_vport;
185 }
186 vport->dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats);
187 if (!vport->dev->tstats) {
188 err = -ENOMEM;
189 goto error_free_netdev;
190 }
191
192 dev_net_set(vport->dev, ovs_dp_get_net(vport->dp));
193 internal_dev = internal_dev_priv(vport->dev);
194 internal_dev->vport = vport;
195
196 /* Restrict bridge port to current netns. */
197 if (vport->port_no == OVSP_LOCAL)
198 vport->dev->features |= NETIF_F_NETNS_LOCAL;
199
200 rtnl_lock();
201 err = register_netdevice(vport->dev);
202 if (err)
203 goto error_unlock;
204
205 dev_set_promiscuity(vport->dev, 1);
206 rtnl_unlock();
207 netif_start_queue(vport->dev);
208
209 return vport;
210
211error_unlock:
212 rtnl_unlock();
213 free_percpu(vport->dev->tstats);
214error_free_netdev:
215 free_netdev(vport->dev);
216error_free_vport:
217 ovs_vport_free(vport);
218error:
219 return ERR_PTR(err);
220}
221
222static void internal_dev_destroy(struct vport *vport)
223{
224 netif_stop_queue(vport->dev);
225 rtnl_lock();
226 dev_set_promiscuity(vport->dev, -1);
227
228 /* unregister_netdevice() waits for an RCU grace period. */
229 unregister_netdevice(vport->dev);
230 free_percpu(vport->dev->tstats);
231 rtnl_unlock();
232}
233
234static netdev_tx_t internal_dev_recv(struct sk_buff *skb)
235{
236 struct net_device *netdev = skb->dev;
237 struct pcpu_sw_netstats *stats;
238
239 if (unlikely(!(netdev->flags & IFF_UP))) {
240 kfree_skb(skb);
241 netdev->stats.rx_dropped++;
242 return NETDEV_TX_OK;
243 }
244
245 skb_dst_drop(skb);
246 nf_reset(skb);
247 secpath_reset(skb);
248
249 skb->pkt_type = PACKET_HOST;
250 skb->protocol = eth_type_trans(skb, netdev);
251 skb_postpull_rcsum(skb, eth_hdr(skb), ETH_HLEN);
252
253 stats = this_cpu_ptr(netdev->tstats);
254 u64_stats_update_begin(&stats->syncp);
255 stats->rx_packets++;
256 stats->rx_bytes += skb->len;
257 u64_stats_update_end(&stats->syncp);
258
259 netif_rx(skb);
260 return NETDEV_TX_OK;
261}
262
263static struct vport_ops ovs_internal_vport_ops = {
264 .type = OVS_VPORT_TYPE_INTERNAL,
265 .create = internal_dev_create,
266 .destroy = internal_dev_destroy,
267 .send = internal_dev_recv,
268};
269
270int ovs_is_internal_dev(const struct net_device *netdev)
271{
272 return netdev->netdev_ops == &internal_dev_netdev_ops;
273}
274
275struct vport *ovs_internal_dev_get_vport(struct net_device *netdev)
276{
277 if (!ovs_is_internal_dev(netdev))
278 return NULL;
279
280 return internal_dev_priv(netdev)->vport;
281}
282
283int ovs_internal_dev_rtnl_link_register(void)
284{
285 int err;
286
287 err = rtnl_link_register(&internal_dev_link_ops);
288 if (err < 0)
289 return err;
290
291 err = ovs_vport_ops_register(&ovs_internal_vport_ops);
292 if (err < 0)
293 rtnl_link_unregister(&internal_dev_link_ops);
294
295 return err;
296}
297
298void ovs_internal_dev_rtnl_link_unregister(void)
299{
300 ovs_vport_ops_unregister(&ovs_internal_vport_ops);
301 rtnl_link_unregister(&internal_dev_link_ops);
302}
1/*
2 * Copyright (c) 2007-2012 Nicira, Inc.
3 *
4 * This program is free software; you can redistribute it and/or
5 * modify it under the terms of version 2 of the GNU General Public
6 * License as published by the Free Software Foundation.
7 *
8 * This program is distributed in the hope that it will be useful, but
9 * WITHOUT ANY WARRANTY; without even the implied warranty of
10 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
11 * General Public License for more details.
12 *
13 * You should have received a copy of the GNU General Public License
14 * along with this program; if not, write to the Free Software
15 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
16 * 02110-1301, USA
17 */
18
19#include <linux/hardirq.h>
20#include <linux/if_vlan.h>
21#include <linux/kernel.h>
22#include <linux/netdevice.h>
23#include <linux/etherdevice.h>
24#include <linux/ethtool.h>
25#include <linux/skbuff.h>
26
27#include <net/dst.h>
28#include <net/xfrm.h>
29
30#include "datapath.h"
31#include "vport-internal_dev.h"
32#include "vport-netdev.h"
33
34struct internal_dev {
35 struct vport *vport;
36};
37
38static struct internal_dev *internal_dev_priv(struct net_device *netdev)
39{
40 return netdev_priv(netdev);
41}
42
43/* This function is only called by the kernel network layer.*/
44static struct rtnl_link_stats64 *internal_dev_get_stats(struct net_device *netdev,
45 struct rtnl_link_stats64 *stats)
46{
47 struct vport *vport = ovs_internal_dev_get_vport(netdev);
48 struct ovs_vport_stats vport_stats;
49
50 ovs_vport_get_stats(vport, &vport_stats);
51
52 /* The tx and rx stats need to be swapped because the
53 * switch and host OS have opposite perspectives. */
54 stats->rx_packets = vport_stats.tx_packets;
55 stats->tx_packets = vport_stats.rx_packets;
56 stats->rx_bytes = vport_stats.tx_bytes;
57 stats->tx_bytes = vport_stats.rx_bytes;
58 stats->rx_errors = vport_stats.tx_errors;
59 stats->tx_errors = vport_stats.rx_errors;
60 stats->rx_dropped = vport_stats.tx_dropped;
61 stats->tx_dropped = vport_stats.rx_dropped;
62
63 return stats;
64}
65
66/* Called with rcu_read_lock_bh. */
67static int internal_dev_xmit(struct sk_buff *skb, struct net_device *netdev)
68{
69 rcu_read_lock();
70 ovs_vport_receive(internal_dev_priv(netdev)->vport, skb, NULL);
71 rcu_read_unlock();
72 return 0;
73}
74
75static int internal_dev_open(struct net_device *netdev)
76{
77 netif_start_queue(netdev);
78 return 0;
79}
80
81static int internal_dev_stop(struct net_device *netdev)
82{
83 netif_stop_queue(netdev);
84 return 0;
85}
86
87static void internal_dev_getinfo(struct net_device *netdev,
88 struct ethtool_drvinfo *info)
89{
90 strlcpy(info->driver, "openvswitch", sizeof(info->driver));
91}
92
93static const struct ethtool_ops internal_dev_ethtool_ops = {
94 .get_drvinfo = internal_dev_getinfo,
95 .get_link = ethtool_op_get_link,
96};
97
98static int internal_dev_change_mtu(struct net_device *netdev, int new_mtu)
99{
100 if (new_mtu < 68)
101 return -EINVAL;
102
103 netdev->mtu = new_mtu;
104 return 0;
105}
106
107static void internal_dev_destructor(struct net_device *dev)
108{
109 struct vport *vport = ovs_internal_dev_get_vport(dev);
110
111 ovs_vport_free(vport);
112 free_netdev(dev);
113}
114
115static const struct net_device_ops internal_dev_netdev_ops = {
116 .ndo_open = internal_dev_open,
117 .ndo_stop = internal_dev_stop,
118 .ndo_start_xmit = internal_dev_xmit,
119 .ndo_set_mac_address = eth_mac_addr,
120 .ndo_change_mtu = internal_dev_change_mtu,
121 .ndo_get_stats64 = internal_dev_get_stats,
122};
123
124static void do_setup(struct net_device *netdev)
125{
126 ether_setup(netdev);
127
128 netdev->netdev_ops = &internal_dev_netdev_ops;
129
130 netdev->priv_flags &= ~IFF_TX_SKB_SHARING;
131 netdev->priv_flags |= IFF_LIVE_ADDR_CHANGE;
132 netdev->destructor = internal_dev_destructor;
133 SET_ETHTOOL_OPS(netdev, &internal_dev_ethtool_ops);
134 netdev->tx_queue_len = 0;
135
136 netdev->features = NETIF_F_LLTX | NETIF_F_SG | NETIF_F_FRAGLIST |
137 NETIF_F_HIGHDMA | NETIF_F_HW_CSUM | NETIF_F_GSO_SOFTWARE;
138
139 netdev->vlan_features = netdev->features;
140 netdev->features |= NETIF_F_HW_VLAN_CTAG_TX;
141 netdev->hw_features = netdev->features & ~NETIF_F_LLTX;
142 eth_hw_addr_random(netdev);
143}
144
145static struct vport *internal_dev_create(const struct vport_parms *parms)
146{
147 struct vport *vport;
148 struct netdev_vport *netdev_vport;
149 struct internal_dev *internal_dev;
150 int err;
151
152 vport = ovs_vport_alloc(sizeof(struct netdev_vport),
153 &ovs_internal_vport_ops, parms);
154 if (IS_ERR(vport)) {
155 err = PTR_ERR(vport);
156 goto error;
157 }
158
159 netdev_vport = netdev_vport_priv(vport);
160
161 netdev_vport->dev = alloc_netdev(sizeof(struct internal_dev),
162 parms->name, do_setup);
163 if (!netdev_vport->dev) {
164 err = -ENOMEM;
165 goto error_free_vport;
166 }
167
168 dev_net_set(netdev_vport->dev, ovs_dp_get_net(vport->dp));
169 internal_dev = internal_dev_priv(netdev_vport->dev);
170 internal_dev->vport = vport;
171
172 /* Restrict bridge port to current netns. */
173 if (vport->port_no == OVSP_LOCAL)
174 netdev_vport->dev->features |= NETIF_F_NETNS_LOCAL;
175
176 rtnl_lock();
177 err = register_netdevice(netdev_vport->dev);
178 if (err)
179 goto error_free_netdev;
180
181 dev_set_promiscuity(netdev_vport->dev, 1);
182 rtnl_unlock();
183 netif_start_queue(netdev_vport->dev);
184
185 return vport;
186
187error_free_netdev:
188 rtnl_unlock();
189 free_netdev(netdev_vport->dev);
190error_free_vport:
191 ovs_vport_free(vport);
192error:
193 return ERR_PTR(err);
194}
195
196static void internal_dev_destroy(struct vport *vport)
197{
198 struct netdev_vport *netdev_vport = netdev_vport_priv(vport);
199
200 netif_stop_queue(netdev_vport->dev);
201 rtnl_lock();
202 dev_set_promiscuity(netdev_vport->dev, -1);
203
204 /* unregister_netdevice() waits for an RCU grace period. */
205 unregister_netdevice(netdev_vport->dev);
206
207 rtnl_unlock();
208}
209
210static int internal_dev_recv(struct vport *vport, struct sk_buff *skb)
211{
212 struct net_device *netdev = netdev_vport_priv(vport)->dev;
213 int len;
214
215 len = skb->len;
216
217 skb_dst_drop(skb);
218 nf_reset(skb);
219 secpath_reset(skb);
220
221 skb->dev = netdev;
222 skb->pkt_type = PACKET_HOST;
223 skb->protocol = eth_type_trans(skb, netdev);
224 skb_postpull_rcsum(skb, eth_hdr(skb), ETH_HLEN);
225
226 netif_rx(skb);
227
228 return len;
229}
230
231const struct vport_ops ovs_internal_vport_ops = {
232 .type = OVS_VPORT_TYPE_INTERNAL,
233 .create = internal_dev_create,
234 .destroy = internal_dev_destroy,
235 .get_name = ovs_netdev_get_name,
236 .send = internal_dev_recv,
237};
238
239int ovs_is_internal_dev(const struct net_device *netdev)
240{
241 return netdev->netdev_ops == &internal_dev_netdev_ops;
242}
243
244struct vport *ovs_internal_dev_get_vport(struct net_device *netdev)
245{
246 if (!ovs_is_internal_dev(netdev))
247 return NULL;
248
249 return internal_dev_priv(netdev)->vport;
250}