2011-10-26 04:26:31 +02:00
|
|
|
/*
|
2012-05-04 03:55:23 +02:00
|
|
|
* Copyright (c) 2007-2012 Nicira, Inc.
|
2011-10-26 04:26:31 +02:00
|
|
|
*
|
|
|
|
* 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/init.h>
|
|
|
|
#include <linux/module.h>
|
|
|
|
#include <linux/if_arp.h>
|
|
|
|
#include <linux/if_vlan.h>
|
|
|
|
#include <linux/in.h>
|
|
|
|
#include <linux/ip.h>
|
|
|
|
#include <linux/jhash.h>
|
|
|
|
#include <linux/delay.h>
|
|
|
|
#include <linux/time.h>
|
|
|
|
#include <linux/etherdevice.h>
|
|
|
|
#include <linux/genetlink.h>
|
|
|
|
#include <linux/kernel.h>
|
|
|
|
#include <linux/kthread.h>
|
|
|
|
#include <linux/mutex.h>
|
|
|
|
#include <linux/percpu.h>
|
|
|
|
#include <linux/rcupdate.h>
|
|
|
|
#include <linux/tcp.h>
|
|
|
|
#include <linux/udp.h>
|
|
|
|
#include <linux/ethtool.h>
|
|
|
|
#include <linux/wait.h>
|
|
|
|
#include <asm/div64.h>
|
|
|
|
#include <linux/highmem.h>
|
|
|
|
#include <linux/netfilter_bridge.h>
|
|
|
|
#include <linux/netfilter_ipv4.h>
|
|
|
|
#include <linux/inetdevice.h>
|
|
|
|
#include <linux/list.h>
|
2013-04-15 22:23:03 +02:00
|
|
|
#include <linux/lockdep.h>
|
2011-10-26 04:26:31 +02:00
|
|
|
#include <linux/openvswitch.h>
|
|
|
|
#include <linux/rculist.h>
|
|
|
|
#include <linux/dmi.h>
|
|
|
|
#include <linux/workqueue.h>
|
|
|
|
#include <net/genetlink.h>
|
2012-02-23 04:58:59 +01:00
|
|
|
#include <net/net_namespace.h>
|
|
|
|
#include <net/netns/generic.h>
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
#include "datapath.h"
|
|
|
|
#include "flow.h"
|
|
|
|
#include "vport-internal_dev.h"
|
2013-04-29 15:06:41 +02:00
|
|
|
#include "vport-netdev.h"
|
2011-10-26 04:26:31 +02:00
|
|
|
|
2012-02-23 04:58:59 +01:00
|
|
|
|
|
|
|
#define REHASH_FLOW_INTERVAL (10 * 60 * HZ)
|
|
|
|
static void rehash_flow_table(struct work_struct *work);
|
|
|
|
static DECLARE_DELAYED_WORK(rehash_flow_wq, rehash_flow_table);
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
int ovs_net_id __read_mostly;
|
|
|
|
|
2013-03-29 14:46:50 +01:00
|
|
|
static void ovs_notify(struct sk_buff *skb, struct genl_info *info,
|
|
|
|
struct genl_multicast_group *grp)
|
|
|
|
{
|
|
|
|
genl_notify(skb, genl_info_net(info), info->snd_portid,
|
|
|
|
grp->id, info->nlhdr, GFP_KERNEL);
|
|
|
|
}
|
|
|
|
|
2011-10-26 04:26:31 +02:00
|
|
|
/**
|
|
|
|
* DOC: Locking:
|
|
|
|
*
|
2013-04-15 22:23:03 +02:00
|
|
|
* All writes e.g. Writes to device state (add/remove datapath, port, set
|
|
|
|
* operations on vports, etc.), Writes to other state (flow table
|
|
|
|
* modifications, set miscellaneous datapath parameters, etc.) are protected
|
|
|
|
* by ovs_lock.
|
2011-10-26 04:26:31 +02:00
|
|
|
*
|
|
|
|
* Reads are protected by RCU.
|
|
|
|
*
|
|
|
|
* There are a few special cases (mostly stats) that have their own
|
|
|
|
* synchronization but they nest under all of above and don't interact with
|
|
|
|
* each other.
|
2013-04-15 22:23:03 +02:00
|
|
|
*
|
|
|
|
* The RTNL lock nests inside ovs_mutex.
|
2011-10-26 04:26:31 +02:00
|
|
|
*/
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
static DEFINE_MUTEX(ovs_mutex);
|
|
|
|
|
|
|
|
void ovs_lock(void)
|
|
|
|
{
|
|
|
|
mutex_lock(&ovs_mutex);
|
|
|
|
}
|
|
|
|
|
|
|
|
void ovs_unlock(void)
|
|
|
|
{
|
|
|
|
mutex_unlock(&ovs_mutex);
|
|
|
|
}
|
|
|
|
|
|
|
|
#ifdef CONFIG_LOCKDEP
|
|
|
|
int lockdep_ovsl_is_held(void)
|
|
|
|
{
|
|
|
|
if (debug_locks)
|
|
|
|
return lockdep_is_held(&ovs_mutex);
|
|
|
|
else
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
2011-10-26 04:26:31 +02:00
|
|
|
static struct vport *new_vport(const struct vport_parms *);
|
2012-02-23 04:58:59 +01:00
|
|
|
static int queue_gso_packets(struct net *, int dp_ifindex, struct sk_buff *,
|
2011-10-26 04:26:31 +02:00
|
|
|
const struct dp_upcall_info *);
|
2012-02-23 04:58:59 +01:00
|
|
|
static int queue_userspace_packet(struct net *, int dp_ifindex,
|
|
|
|
struct sk_buff *,
|
2011-10-26 04:26:31 +02:00
|
|
|
const struct dp_upcall_info *);
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
/* Must be called with rcu_read_lock or ovs_mutex. */
|
2012-02-23 04:58:59 +01:00
|
|
|
static struct datapath *get_dp(struct net *net, int dp_ifindex)
|
2011-10-26 04:26:31 +02:00
|
|
|
{
|
|
|
|
struct datapath *dp = NULL;
|
|
|
|
struct net_device *dev;
|
|
|
|
|
|
|
|
rcu_read_lock();
|
2012-02-23 04:58:59 +01:00
|
|
|
dev = dev_get_by_index_rcu(net, dp_ifindex);
|
2011-10-26 04:26:31 +02:00
|
|
|
if (dev) {
|
|
|
|
struct vport *vport = ovs_internal_dev_get_vport(dev);
|
|
|
|
if (vport)
|
|
|
|
dp = vport->dp;
|
|
|
|
}
|
|
|
|
rcu_read_unlock();
|
|
|
|
|
|
|
|
return dp;
|
|
|
|
}
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
/* Must be called with rcu_read_lock or ovs_mutex. */
|
2011-10-26 04:26:31 +02:00
|
|
|
const char *ovs_dp_name(const struct datapath *dp)
|
|
|
|
{
|
2013-04-15 22:23:03 +02:00
|
|
|
struct vport *vport = ovs_vport_ovsl_rcu(dp, OVSP_LOCAL);
|
2011-10-26 04:26:31 +02:00
|
|
|
return vport->ops->get_name(vport);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int get_dpifindex(struct datapath *dp)
|
|
|
|
{
|
|
|
|
struct vport *local;
|
|
|
|
int ifindex;
|
|
|
|
|
|
|
|
rcu_read_lock();
|
|
|
|
|
2012-08-23 21:40:54 +02:00
|
|
|
local = ovs_vport_rcu(dp, OVSP_LOCAL);
|
2011-10-26 04:26:31 +02:00
|
|
|
if (local)
|
2013-04-29 15:06:41 +02:00
|
|
|
ifindex = netdev_vport_priv(local)->dev->ifindex;
|
2011-10-26 04:26:31 +02:00
|
|
|
else
|
|
|
|
ifindex = 0;
|
|
|
|
|
|
|
|
rcu_read_unlock();
|
|
|
|
|
|
|
|
return ifindex;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void destroy_dp_rcu(struct rcu_head *rcu)
|
|
|
|
{
|
|
|
|
struct datapath *dp = container_of(rcu, struct datapath, rcu);
|
|
|
|
|
|
|
|
ovs_flow_tbl_destroy((__force struct flow_table *)dp->table);
|
|
|
|
free_percpu(dp->stats_percpu);
|
2012-02-23 04:58:59 +01:00
|
|
|
release_net(ovs_dp_get_net(dp));
|
2012-08-23 21:40:54 +02:00
|
|
|
kfree(dp->ports);
|
2011-10-26 04:26:31 +02:00
|
|
|
kfree(dp);
|
|
|
|
}
|
|
|
|
|
2012-08-23 21:40:54 +02:00
|
|
|
static struct hlist_head *vport_hash_bucket(const struct datapath *dp,
|
|
|
|
u16 port_no)
|
|
|
|
{
|
|
|
|
return &dp->ports[port_no & (DP_VPORT_HASH_BUCKETS - 1)];
|
|
|
|
}
|
|
|
|
|
|
|
|
struct vport *ovs_lookup_vport(const struct datapath *dp, u16 port_no)
|
|
|
|
{
|
|
|
|
struct vport *vport;
|
|
|
|
struct hlist_head *head;
|
|
|
|
|
|
|
|
head = vport_hash_bucket(dp, port_no);
|
hlist: drop the node parameter from iterators
I'm not sure why, but the hlist for each entry iterators were conceived
list_for_each_entry(pos, head, member)
The hlist ones were greedy and wanted an extra parameter:
hlist_for_each_entry(tpos, pos, head, member)
Why did they need an extra pos parameter? I'm not quite sure. Not only
they don't really need it, it also prevents the iterator from looking
exactly like the list iterator, which is unfortunate.
Besides the semantic patch, there was some manual work required:
- Fix up the actual hlist iterators in linux/list.h
- Fix up the declaration of other iterators based on the hlist ones.
- A very small amount of places were using the 'node' parameter, this
was modified to use 'obj->member' instead.
- Coccinelle didn't handle the hlist_for_each_entry_safe iterator
properly, so those had to be fixed up manually.
The semantic patch which is mostly the work of Peter Senna Tschudin is here:
@@
iterator name hlist_for_each_entry, hlist_for_each_entry_continue, hlist_for_each_entry_from, hlist_for_each_entry_rcu, hlist_for_each_entry_rcu_bh, hlist_for_each_entry_continue_rcu_bh, for_each_busy_worker, ax25_uid_for_each, ax25_for_each, inet_bind_bucket_for_each, sctp_for_each_hentry, sk_for_each, sk_for_each_rcu, sk_for_each_from, sk_for_each_safe, sk_for_each_bound, hlist_for_each_entry_safe, hlist_for_each_entry_continue_rcu, nr_neigh_for_each, nr_neigh_for_each_safe, nr_node_for_each, nr_node_for_each_safe, for_each_gfn_indirect_valid_sp, for_each_gfn_sp, for_each_host;
type T;
expression a,c,d,e;
identifier b;
statement S;
@@
-T b;
<+... when != b
(
hlist_for_each_entry(a,
- b,
c, d) S
|
hlist_for_each_entry_continue(a,
- b,
c) S
|
hlist_for_each_entry_from(a,
- b,
c) S
|
hlist_for_each_entry_rcu(a,
- b,
c, d) S
|
hlist_for_each_entry_rcu_bh(a,
- b,
c, d) S
|
hlist_for_each_entry_continue_rcu_bh(a,
- b,
c) S
|
for_each_busy_worker(a, c,
- b,
d) S
|
ax25_uid_for_each(a,
- b,
c) S
|
ax25_for_each(a,
- b,
c) S
|
inet_bind_bucket_for_each(a,
- b,
c) S
|
sctp_for_each_hentry(a,
- b,
c) S
|
sk_for_each(a,
- b,
c) S
|
sk_for_each_rcu(a,
- b,
c) S
|
sk_for_each_from
-(a, b)
+(a)
S
+ sk_for_each_from(a) S
|
sk_for_each_safe(a,
- b,
c, d) S
|
sk_for_each_bound(a,
- b,
c) S
|
hlist_for_each_entry_safe(a,
- b,
c, d, e) S
|
hlist_for_each_entry_continue_rcu(a,
- b,
c) S
|
nr_neigh_for_each(a,
- b,
c) S
|
nr_neigh_for_each_safe(a,
- b,
c, d) S
|
nr_node_for_each(a,
- b,
c) S
|
nr_node_for_each_safe(a,
- b,
c, d) S
|
- for_each_gfn_sp(a, c, d, b) S
+ for_each_gfn_sp(a, c, d) S
|
- for_each_gfn_indirect_valid_sp(a, c, d, b) S
+ for_each_gfn_indirect_valid_sp(a, c, d) S
|
for_each_host(a,
- b,
c) S
|
for_each_host_safe(a,
- b,
c, d) S
|
for_each_mesh_entry(a,
- b,
c, d) S
)
...+>
[akpm@linux-foundation.org: drop bogus change from net/ipv4/raw.c]
[akpm@linux-foundation.org: drop bogus hunk from net/ipv6/raw.c]
[akpm@linux-foundation.org: checkpatch fixes]
[akpm@linux-foundation.org: fix warnings]
[akpm@linux-foudnation.org: redo intrusive kvm changes]
Tested-by: Peter Senna Tschudin <peter.senna@gmail.com>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-28 02:06:00 +01:00
|
|
|
hlist_for_each_entry_rcu(vport, head, dp_hash_node) {
|
2012-08-23 21:40:54 +02:00
|
|
|
if (vport->port_no == port_no)
|
|
|
|
return vport;
|
|
|
|
}
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
/* Called with ovs_mutex. */
|
2011-10-26 04:26:31 +02:00
|
|
|
static struct vport *new_vport(const struct vport_parms *parms)
|
|
|
|
{
|
|
|
|
struct vport *vport;
|
|
|
|
|
|
|
|
vport = ovs_vport_add(parms);
|
|
|
|
if (!IS_ERR(vport)) {
|
|
|
|
struct datapath *dp = parms->dp;
|
2012-08-23 21:40:54 +02:00
|
|
|
struct hlist_head *head = vport_hash_bucket(dp, vport->port_no);
|
2011-10-26 04:26:31 +02:00
|
|
|
|
2012-08-23 21:40:54 +02:00
|
|
|
hlist_add_head_rcu(&vport->dp_hash_node, head);
|
2011-10-26 04:26:31 +02:00
|
|
|
}
|
|
|
|
return vport;
|
|
|
|
}
|
|
|
|
|
|
|
|
void ovs_dp_detach_port(struct vport *p)
|
|
|
|
{
|
2013-04-15 22:23:03 +02:00
|
|
|
ASSERT_OVSL();
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
/* First drop references to device. */
|
2012-08-23 21:40:54 +02:00
|
|
|
hlist_del_rcu(&p->dp_hash_node);
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
/* Then destroy it. */
|
|
|
|
ovs_vport_del(p);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Must be called with rcu_read_lock. */
|
|
|
|
void ovs_dp_process_received_packet(struct vport *p, struct sk_buff *skb)
|
|
|
|
{
|
|
|
|
struct datapath *dp = p->dp;
|
|
|
|
struct sw_flow *flow;
|
|
|
|
struct dp_stats_percpu *stats;
|
|
|
|
struct sw_flow_key key;
|
|
|
|
u64 *stats_counter;
|
|
|
|
int error;
|
|
|
|
int key_len;
|
|
|
|
|
2012-11-13 02:52:25 +01:00
|
|
|
stats = this_cpu_ptr(dp->stats_percpu);
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
/* Extract flow from 'skb' into 'key'. */
|
|
|
|
error = ovs_flow_extract(skb, p->port_no, &key, &key_len);
|
|
|
|
if (unlikely(error)) {
|
|
|
|
kfree_skb(skb);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Look up flow. */
|
|
|
|
flow = ovs_flow_tbl_lookup(rcu_dereference(dp->table), &key, key_len);
|
|
|
|
if (unlikely(!flow)) {
|
|
|
|
struct dp_upcall_info upcall;
|
|
|
|
|
|
|
|
upcall.cmd = OVS_PACKET_CMD_MISS;
|
|
|
|
upcall.key = &key;
|
|
|
|
upcall.userdata = NULL;
|
2012-09-07 22:12:54 +02:00
|
|
|
upcall.portid = p->upcall_portid;
|
2011-10-26 04:26:31 +02:00
|
|
|
ovs_dp_upcall(dp, skb, &upcall);
|
|
|
|
consume_skb(skb);
|
|
|
|
stats_counter = &stats->n_missed;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
OVS_CB(skb)->flow = flow;
|
|
|
|
|
|
|
|
stats_counter = &stats->n_hit;
|
|
|
|
ovs_flow_used(OVS_CB(skb)->flow, skb);
|
|
|
|
ovs_execute_actions(dp, skb);
|
|
|
|
|
|
|
|
out:
|
|
|
|
/* Update datapath statistics. */
|
|
|
|
u64_stats_update_begin(&stats->sync);
|
|
|
|
(*stats_counter)++;
|
|
|
|
u64_stats_update_end(&stats->sync);
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct genl_family dp_packet_genl_family = {
|
|
|
|
.id = GENL_ID_GENERATE,
|
|
|
|
.hdrsize = sizeof(struct ovs_header),
|
|
|
|
.name = OVS_PACKET_FAMILY,
|
|
|
|
.version = OVS_PACKET_VERSION,
|
2012-02-23 04:58:59 +01:00
|
|
|
.maxattr = OVS_PACKET_ATTR_MAX,
|
2013-04-23 09:48:48 +02:00
|
|
|
.netnsok = true,
|
|
|
|
.parallel_ops = true,
|
2011-10-26 04:26:31 +02:00
|
|
|
};
|
|
|
|
|
|
|
|
int ovs_dp_upcall(struct datapath *dp, struct sk_buff *skb,
|
2012-02-23 04:58:59 +01:00
|
|
|
const struct dp_upcall_info *upcall_info)
|
2011-10-26 04:26:31 +02:00
|
|
|
{
|
|
|
|
struct dp_stats_percpu *stats;
|
|
|
|
int dp_ifindex;
|
|
|
|
int err;
|
|
|
|
|
2012-09-07 22:12:54 +02:00
|
|
|
if (upcall_info->portid == 0) {
|
2011-10-26 04:26:31 +02:00
|
|
|
err = -ENOTCONN;
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
dp_ifindex = get_dpifindex(dp);
|
|
|
|
if (!dp_ifindex) {
|
|
|
|
err = -ENODEV;
|
|
|
|
goto err;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!skb_is_gso(skb))
|
2012-02-23 04:58:59 +01:00
|
|
|
err = queue_userspace_packet(ovs_dp_get_net(dp), dp_ifindex, skb, upcall_info);
|
2011-10-26 04:26:31 +02:00
|
|
|
else
|
2012-02-23 04:58:59 +01:00
|
|
|
err = queue_gso_packets(ovs_dp_get_net(dp), dp_ifindex, skb, upcall_info);
|
2011-10-26 04:26:31 +02:00
|
|
|
if (err)
|
|
|
|
goto err;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
err:
|
2012-11-13 02:52:25 +01:00
|
|
|
stats = this_cpu_ptr(dp->stats_percpu);
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
u64_stats_update_begin(&stats->sync);
|
|
|
|
stats->n_lost++;
|
|
|
|
u64_stats_update_end(&stats->sync);
|
|
|
|
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2012-02-23 04:58:59 +01:00
|
|
|
static int queue_gso_packets(struct net *net, int dp_ifindex,
|
|
|
|
struct sk_buff *skb,
|
2011-10-26 04:26:31 +02:00
|
|
|
const struct dp_upcall_info *upcall_info)
|
|
|
|
{
|
2012-07-20 23:47:54 +02:00
|
|
|
unsigned short gso_type = skb_shinfo(skb)->gso_type;
|
2011-10-26 04:26:31 +02:00
|
|
|
struct dp_upcall_info later_info;
|
|
|
|
struct sw_flow_key later_key;
|
|
|
|
struct sk_buff *segs, *nskb;
|
|
|
|
int err;
|
|
|
|
|
2013-02-05 17:36:38 +01:00
|
|
|
segs = __skb_gso_segment(skb, NETIF_F_SG | NETIF_F_HW_CSUM, false);
|
2012-07-20 23:46:29 +02:00
|
|
|
if (IS_ERR(segs))
|
|
|
|
return PTR_ERR(segs);
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
/* Queue all of the segments. */
|
|
|
|
skb = segs;
|
|
|
|
do {
|
2012-02-23 04:58:59 +01:00
|
|
|
err = queue_userspace_packet(net, dp_ifindex, skb, upcall_info);
|
2011-10-26 04:26:31 +02:00
|
|
|
if (err)
|
|
|
|
break;
|
|
|
|
|
2012-07-20 23:47:54 +02:00
|
|
|
if (skb == segs && gso_type & SKB_GSO_UDP) {
|
2011-10-26 04:26:31 +02:00
|
|
|
/* The initial flow key extracted by ovs_flow_extract()
|
|
|
|
* in this case is for a first fragment, so we need to
|
|
|
|
* properly mark later fragments.
|
|
|
|
*/
|
|
|
|
later_key = *upcall_info->key;
|
|
|
|
later_key.ip.frag = OVS_FRAG_TYPE_LATER;
|
|
|
|
|
|
|
|
later_info = *upcall_info;
|
|
|
|
later_info.key = &later_key;
|
|
|
|
upcall_info = &later_info;
|
|
|
|
}
|
|
|
|
} while ((skb = skb->next));
|
|
|
|
|
|
|
|
/* Free all of the segments. */
|
|
|
|
skb = segs;
|
|
|
|
do {
|
|
|
|
nskb = skb->next;
|
|
|
|
if (err)
|
|
|
|
kfree_skb(skb);
|
|
|
|
else
|
|
|
|
consume_skb(skb);
|
|
|
|
} while ((skb = nskb));
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2013-03-29 14:46:49 +01:00
|
|
|
static size_t key_attr_size(void)
|
|
|
|
{
|
|
|
|
return nla_total_size(4) /* OVS_KEY_ATTR_PRIORITY */
|
2013-06-18 02:50:18 +02:00
|
|
|
+ nla_total_size(0) /* OVS_KEY_ATTR_TUNNEL */
|
|
|
|
+ nla_total_size(8) /* OVS_TUNNEL_KEY_ATTR_ID */
|
|
|
|
+ nla_total_size(4) /* OVS_TUNNEL_KEY_ATTR_IPV4_SRC */
|
|
|
|
+ nla_total_size(4) /* OVS_TUNNEL_KEY_ATTR_IPV4_DST */
|
|
|
|
+ nla_total_size(1) /* OVS_TUNNEL_KEY_ATTR_TOS */
|
|
|
|
+ nla_total_size(1) /* OVS_TUNNEL_KEY_ATTR_TTL */
|
|
|
|
+ nla_total_size(0) /* OVS_TUNNEL_KEY_ATTR_DONT_FRAGMENT */
|
|
|
|
+ nla_total_size(0) /* OVS_TUNNEL_KEY_ATTR_CSUM */
|
2013-03-29 14:46:49 +01:00
|
|
|
+ nla_total_size(4) /* OVS_KEY_ATTR_IN_PORT */
|
|
|
|
+ nla_total_size(4) /* OVS_KEY_ATTR_SKB_MARK */
|
|
|
|
+ nla_total_size(12) /* OVS_KEY_ATTR_ETHERNET */
|
|
|
|
+ nla_total_size(2) /* OVS_KEY_ATTR_ETHERTYPE */
|
|
|
|
+ nla_total_size(4) /* OVS_KEY_ATTR_8021Q */
|
|
|
|
+ nla_total_size(0) /* OVS_KEY_ATTR_ENCAP */
|
|
|
|
+ nla_total_size(2) /* OVS_KEY_ATTR_ETHERTYPE */
|
|
|
|
+ nla_total_size(40) /* OVS_KEY_ATTR_IPV6 */
|
|
|
|
+ nla_total_size(2) /* OVS_KEY_ATTR_ICMPV6 */
|
|
|
|
+ nla_total_size(28); /* OVS_KEY_ATTR_ND */
|
|
|
|
}
|
|
|
|
|
|
|
|
static size_t upcall_msg_size(const struct sk_buff *skb,
|
|
|
|
const struct nlattr *userdata)
|
|
|
|
{
|
|
|
|
size_t size = NLMSG_ALIGN(sizeof(struct ovs_header))
|
|
|
|
+ nla_total_size(skb->len) /* OVS_PACKET_ATTR_PACKET */
|
|
|
|
+ nla_total_size(key_attr_size()); /* OVS_PACKET_ATTR_KEY */
|
|
|
|
|
|
|
|
/* OVS_PACKET_ATTR_USERDATA */
|
|
|
|
if (userdata)
|
|
|
|
size += NLA_ALIGN(userdata->nla_len);
|
|
|
|
|
|
|
|
return size;
|
|
|
|
}
|
|
|
|
|
2012-02-23 04:58:59 +01:00
|
|
|
static int queue_userspace_packet(struct net *net, int dp_ifindex,
|
|
|
|
struct sk_buff *skb,
|
2011-10-26 04:26:31 +02:00
|
|
|
const struct dp_upcall_info *upcall_info)
|
|
|
|
{
|
|
|
|
struct ovs_header *upcall;
|
|
|
|
struct sk_buff *nskb = NULL;
|
|
|
|
struct sk_buff *user_skb; /* to be queued to userspace */
|
|
|
|
struct nlattr *nla;
|
|
|
|
int err;
|
|
|
|
|
|
|
|
if (vlan_tx_tag_present(skb)) {
|
|
|
|
nskb = skb_clone(skb, GFP_ATOMIC);
|
|
|
|
if (!nskb)
|
|
|
|
return -ENOMEM;
|
|
|
|
|
2013-04-19 04:04:30 +02:00
|
|
|
nskb = __vlan_put_tag(nskb, nskb->vlan_proto, vlan_tx_tag_get(nskb));
|
2012-05-13 10:44:18 +02:00
|
|
|
if (!nskb)
|
2011-10-26 04:26:31 +02:00
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
nskb->vlan_tci = 0;
|
|
|
|
skb = nskb;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (nla_attr_size(skb->len) > USHRT_MAX) {
|
|
|
|
err = -EFBIG;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
2013-03-29 14:46:49 +01:00
|
|
|
user_skb = genlmsg_new(upcall_msg_size(skb, upcall_info->userdata), GFP_ATOMIC);
|
2011-10-26 04:26:31 +02:00
|
|
|
if (!user_skb) {
|
|
|
|
err = -ENOMEM;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
upcall = genlmsg_put(user_skb, 0, 0, &dp_packet_genl_family,
|
|
|
|
0, upcall_info->cmd);
|
|
|
|
upcall->dp_ifindex = dp_ifindex;
|
|
|
|
|
|
|
|
nla = nla_nest_start(user_skb, OVS_PACKET_ATTR_KEY);
|
|
|
|
ovs_flow_to_nlattrs(upcall_info->key, user_skb);
|
|
|
|
nla_nest_end(user_skb, nla);
|
|
|
|
|
|
|
|
if (upcall_info->userdata)
|
2013-02-16 02:29:22 +01:00
|
|
|
__nla_put(user_skb, OVS_PACKET_ATTR_USERDATA,
|
|
|
|
nla_len(upcall_info->userdata),
|
|
|
|
nla_data(upcall_info->userdata));
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
nla = __nla_reserve(user_skb, OVS_PACKET_ATTR_PACKET, skb->len);
|
|
|
|
|
|
|
|
skb_copy_and_csum_dev(skb, nla_data(nla));
|
|
|
|
|
2013-02-15 20:07:43 +01:00
|
|
|
genlmsg_end(user_skb, upcall);
|
2012-09-07 22:12:54 +02:00
|
|
|
err = genlmsg_unicast(net, user_skb, upcall_info->portid);
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
out:
|
|
|
|
kfree_skb(nskb);
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
/* Called with ovs_mutex. */
|
2012-02-23 04:58:59 +01:00
|
|
|
static int flush_flows(struct datapath *dp)
|
2011-10-26 04:26:31 +02:00
|
|
|
{
|
|
|
|
struct flow_table *old_table;
|
|
|
|
struct flow_table *new_table;
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
old_table = ovsl_dereference(dp->table);
|
2011-10-26 04:26:31 +02:00
|
|
|
new_table = ovs_flow_tbl_alloc(TBL_MIN_BUCKETS);
|
|
|
|
if (!new_table)
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
rcu_assign_pointer(dp->table, new_table);
|
|
|
|
|
|
|
|
ovs_flow_tbl_deferred_destroy(old_table);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2013-06-18 02:50:12 +02:00
|
|
|
static struct nlattr *reserve_sfa_size(struct sw_flow_actions **sfa, int attr_len)
|
|
|
|
{
|
|
|
|
|
|
|
|
struct sw_flow_actions *acts;
|
|
|
|
int new_acts_size;
|
|
|
|
int req_size = NLA_ALIGN(attr_len);
|
|
|
|
int next_offset = offsetof(struct sw_flow_actions, actions) +
|
|
|
|
(*sfa)->actions_len;
|
|
|
|
|
|
|
|
if (req_size <= (ksize(*sfa) - next_offset))
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
new_acts_size = ksize(*sfa) * 2;
|
|
|
|
|
|
|
|
if (new_acts_size > MAX_ACTIONS_BUFSIZE) {
|
|
|
|
if ((MAX_ACTIONS_BUFSIZE - next_offset) < req_size)
|
|
|
|
return ERR_PTR(-EMSGSIZE);
|
|
|
|
new_acts_size = MAX_ACTIONS_BUFSIZE;
|
|
|
|
}
|
|
|
|
|
|
|
|
acts = ovs_flow_actions_alloc(new_acts_size);
|
|
|
|
if (IS_ERR(acts))
|
|
|
|
return (void *)acts;
|
|
|
|
|
|
|
|
memcpy(acts->actions, (*sfa)->actions, (*sfa)->actions_len);
|
|
|
|
acts->actions_len = (*sfa)->actions_len;
|
|
|
|
kfree(*sfa);
|
|
|
|
*sfa = acts;
|
|
|
|
|
|
|
|
out:
|
|
|
|
(*sfa)->actions_len += req_size;
|
|
|
|
return (struct nlattr *) ((unsigned char *)(*sfa) + next_offset);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int add_action(struct sw_flow_actions **sfa, int attrtype, void *data, int len)
|
|
|
|
{
|
|
|
|
struct nlattr *a;
|
|
|
|
|
|
|
|
a = reserve_sfa_size(sfa, nla_attr_size(len));
|
|
|
|
if (IS_ERR(a))
|
|
|
|
return PTR_ERR(a);
|
|
|
|
|
|
|
|
a->nla_type = attrtype;
|
|
|
|
a->nla_len = nla_attr_size(len);
|
|
|
|
|
|
|
|
if (data)
|
|
|
|
memcpy(nla_data(a), data, len);
|
|
|
|
memset((unsigned char *) a + a->nla_len, 0, nla_padlen(len));
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline int add_nested_action_start(struct sw_flow_actions **sfa, int attrtype)
|
|
|
|
{
|
|
|
|
int used = (*sfa)->actions_len;
|
|
|
|
int err;
|
|
|
|
|
|
|
|
err = add_action(sfa, attrtype, NULL, 0);
|
|
|
|
if (err)
|
|
|
|
return err;
|
|
|
|
|
|
|
|
return used;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void add_nested_action_end(struct sw_flow_actions *sfa, int st_offset)
|
|
|
|
{
|
|
|
|
struct nlattr *a = (struct nlattr *) ((unsigned char *)sfa->actions + st_offset);
|
|
|
|
|
|
|
|
a->nla_len = sfa->actions_len - st_offset;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int validate_and_copy_actions(const struct nlattr *attr,
|
|
|
|
const struct sw_flow_key *key, int depth,
|
|
|
|
struct sw_flow_actions **sfa);
|
2011-10-26 04:26:31 +02:00
|
|
|
|
2013-06-18 02:50:12 +02:00
|
|
|
static int validate_and_copy_sample(const struct nlattr *attr,
|
|
|
|
const struct sw_flow_key *key, int depth,
|
|
|
|
struct sw_flow_actions **sfa)
|
2011-10-26 04:26:31 +02:00
|
|
|
{
|
|
|
|
const struct nlattr *attrs[OVS_SAMPLE_ATTR_MAX + 1];
|
|
|
|
const struct nlattr *probability, *actions;
|
|
|
|
const struct nlattr *a;
|
2013-06-18 02:50:12 +02:00
|
|
|
int rem, start, err, st_acts;
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
memset(attrs, 0, sizeof(attrs));
|
|
|
|
nla_for_each_nested(a, attr, rem) {
|
|
|
|
int type = nla_type(a);
|
|
|
|
if (!type || type > OVS_SAMPLE_ATTR_MAX || attrs[type])
|
|
|
|
return -EINVAL;
|
|
|
|
attrs[type] = a;
|
|
|
|
}
|
|
|
|
if (rem)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
probability = attrs[OVS_SAMPLE_ATTR_PROBABILITY];
|
|
|
|
if (!probability || nla_len(probability) != sizeof(u32))
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
actions = attrs[OVS_SAMPLE_ATTR_ACTIONS];
|
|
|
|
if (!actions || (nla_len(actions) && nla_len(actions) < NLA_HDRLEN))
|
|
|
|
return -EINVAL;
|
2013-06-18 02:50:12 +02:00
|
|
|
|
|
|
|
/* validation done, copy sample action. */
|
|
|
|
start = add_nested_action_start(sfa, OVS_ACTION_ATTR_SAMPLE);
|
|
|
|
if (start < 0)
|
|
|
|
return start;
|
|
|
|
err = add_action(sfa, OVS_SAMPLE_ATTR_PROBABILITY, nla_data(probability), sizeof(u32));
|
|
|
|
if (err)
|
|
|
|
return err;
|
|
|
|
st_acts = add_nested_action_start(sfa, OVS_SAMPLE_ATTR_ACTIONS);
|
|
|
|
if (st_acts < 0)
|
|
|
|
return st_acts;
|
|
|
|
|
|
|
|
err = validate_and_copy_actions(actions, key, depth + 1, sfa);
|
|
|
|
if (err)
|
|
|
|
return err;
|
|
|
|
|
|
|
|
add_nested_action_end(*sfa, st_acts);
|
|
|
|
add_nested_action_end(*sfa, start);
|
|
|
|
|
|
|
|
return 0;
|
2011-10-26 04:26:31 +02:00
|
|
|
}
|
|
|
|
|
2012-05-08 02:21:53 +02:00
|
|
|
static int validate_tp_port(const struct sw_flow_key *flow_key)
|
|
|
|
{
|
|
|
|
if (flow_key->eth.type == htons(ETH_P_IP)) {
|
2012-08-07 00:49:47 +02:00
|
|
|
if (flow_key->ipv4.tp.src || flow_key->ipv4.tp.dst)
|
2012-05-08 02:21:53 +02:00
|
|
|
return 0;
|
|
|
|
} else if (flow_key->eth.type == htons(ETH_P_IPV6)) {
|
2012-08-07 00:49:47 +02:00
|
|
|
if (flow_key->ipv6.tp.src || flow_key->ipv6.tp.dst)
|
2012-05-08 02:21:53 +02:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
return -EINVAL;
|
|
|
|
}
|
|
|
|
|
2013-06-18 02:50:18 +02:00
|
|
|
static int validate_and_copy_set_tun(const struct nlattr *attr,
|
|
|
|
struct sw_flow_actions **sfa)
|
|
|
|
{
|
|
|
|
struct ovs_key_ipv4_tunnel tun_key;
|
|
|
|
int err, start;
|
|
|
|
|
|
|
|
err = ovs_ipv4_tun_from_nlattr(nla_data(attr), &tun_key);
|
|
|
|
if (err)
|
|
|
|
return err;
|
|
|
|
|
|
|
|
start = add_nested_action_start(sfa, OVS_ACTION_ATTR_SET);
|
|
|
|
if (start < 0)
|
|
|
|
return start;
|
|
|
|
|
|
|
|
err = add_action(sfa, OVS_KEY_ATTR_IPV4_TUNNEL, &tun_key, sizeof(tun_key));
|
|
|
|
add_nested_action_end(*sfa, start);
|
|
|
|
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2011-10-26 04:26:31 +02:00
|
|
|
static int validate_set(const struct nlattr *a,
|
2013-06-18 02:50:18 +02:00
|
|
|
const struct sw_flow_key *flow_key,
|
|
|
|
struct sw_flow_actions **sfa,
|
|
|
|
bool *set_tun)
|
2011-10-26 04:26:31 +02:00
|
|
|
{
|
|
|
|
const struct nlattr *ovs_key = nla_data(a);
|
|
|
|
int key_type = nla_type(ovs_key);
|
|
|
|
|
|
|
|
/* There can be only one key in a action */
|
|
|
|
if (nla_total_size(nla_len(ovs_key)) != nla_len(a))
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
if (key_type > OVS_KEY_ATTR_MAX ||
|
2013-06-18 02:50:18 +02:00
|
|
|
(ovs_key_lens[key_type] != nla_len(ovs_key) &&
|
|
|
|
ovs_key_lens[key_type] != -1))
|
2011-10-26 04:26:31 +02:00
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
switch (key_type) {
|
|
|
|
const struct ovs_key_ipv4 *ipv4_key;
|
2012-11-14 00:44:14 +01:00
|
|
|
const struct ovs_key_ipv6 *ipv6_key;
|
2013-06-18 02:50:18 +02:00
|
|
|
int err;
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
case OVS_KEY_ATTR_PRIORITY:
|
2012-11-26 20:24:11 +01:00
|
|
|
case OVS_KEY_ATTR_SKB_MARK:
|
2011-10-26 04:26:31 +02:00
|
|
|
case OVS_KEY_ATTR_ETHERNET:
|
|
|
|
break;
|
|
|
|
|
2013-06-18 02:50:18 +02:00
|
|
|
case OVS_KEY_ATTR_TUNNEL:
|
|
|
|
*set_tun = true;
|
|
|
|
err = validate_and_copy_set_tun(a, sfa);
|
|
|
|
if (err)
|
|
|
|
return err;
|
|
|
|
break;
|
|
|
|
|
2011-10-26 04:26:31 +02:00
|
|
|
case OVS_KEY_ATTR_IPV4:
|
|
|
|
if (flow_key->eth.type != htons(ETH_P_IP))
|
|
|
|
return -EINVAL;
|
|
|
|
|
2012-08-07 00:49:47 +02:00
|
|
|
if (!flow_key->ip.proto)
|
2011-10-26 04:26:31 +02:00
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
ipv4_key = nla_data(ovs_key);
|
|
|
|
if (ipv4_key->ipv4_proto != flow_key->ip.proto)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
if (ipv4_key->ipv4_frag != flow_key->ip.frag)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
break;
|
|
|
|
|
2012-11-14 00:44:14 +01:00
|
|
|
case OVS_KEY_ATTR_IPV6:
|
|
|
|
if (flow_key->eth.type != htons(ETH_P_IPV6))
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
if (!flow_key->ip.proto)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
ipv6_key = nla_data(ovs_key);
|
|
|
|
if (ipv6_key->ipv6_proto != flow_key->ip.proto)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
if (ipv6_key->ipv6_frag != flow_key->ip.frag)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
if (ntohl(ipv6_key->ipv6_label) & 0xFFF00000)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
break;
|
|
|
|
|
2011-10-26 04:26:31 +02:00
|
|
|
case OVS_KEY_ATTR_TCP:
|
|
|
|
if (flow_key->ip.proto != IPPROTO_TCP)
|
|
|
|
return -EINVAL;
|
|
|
|
|
2012-05-08 02:21:53 +02:00
|
|
|
return validate_tp_port(flow_key);
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
case OVS_KEY_ATTR_UDP:
|
|
|
|
if (flow_key->ip.proto != IPPROTO_UDP)
|
|
|
|
return -EINVAL;
|
|
|
|
|
2012-05-08 02:21:53 +02:00
|
|
|
return validate_tp_port(flow_key);
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
default:
|
|
|
|
return -EINVAL;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int validate_userspace(const struct nlattr *attr)
|
|
|
|
{
|
|
|
|
static const struct nla_policy userspace_policy[OVS_USERSPACE_ATTR_MAX + 1] = {
|
|
|
|
[OVS_USERSPACE_ATTR_PID] = {.type = NLA_U32 },
|
2013-02-16 02:29:22 +01:00
|
|
|
[OVS_USERSPACE_ATTR_USERDATA] = {.type = NLA_UNSPEC },
|
2011-10-26 04:26:31 +02:00
|
|
|
};
|
|
|
|
struct nlattr *a[OVS_USERSPACE_ATTR_MAX + 1];
|
|
|
|
int error;
|
|
|
|
|
|
|
|
error = nla_parse_nested(a, OVS_USERSPACE_ATTR_MAX,
|
|
|
|
attr, userspace_policy);
|
|
|
|
if (error)
|
|
|
|
return error;
|
|
|
|
|
|
|
|
if (!a[OVS_USERSPACE_ATTR_PID] ||
|
|
|
|
!nla_get_u32(a[OVS_USERSPACE_ATTR_PID]))
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2013-06-18 02:50:12 +02:00
|
|
|
static int copy_action(const struct nlattr *from,
|
|
|
|
struct sw_flow_actions **sfa)
|
|
|
|
{
|
|
|
|
int totlen = NLA_ALIGN(from->nla_len);
|
|
|
|
struct nlattr *to;
|
|
|
|
|
|
|
|
to = reserve_sfa_size(sfa, from->nla_len);
|
|
|
|
if (IS_ERR(to))
|
|
|
|
return PTR_ERR(to);
|
|
|
|
|
|
|
|
memcpy(to, from, totlen);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int validate_and_copy_actions(const struct nlattr *attr,
|
|
|
|
const struct sw_flow_key *key,
|
|
|
|
int depth,
|
|
|
|
struct sw_flow_actions **sfa)
|
2011-10-26 04:26:31 +02:00
|
|
|
{
|
|
|
|
const struct nlattr *a;
|
|
|
|
int rem, err;
|
|
|
|
|
|
|
|
if (depth >= SAMPLE_ACTION_DEPTH)
|
|
|
|
return -EOVERFLOW;
|
|
|
|
|
|
|
|
nla_for_each_nested(a, attr, rem) {
|
|
|
|
/* Expected argument lengths, (u32)-1 for variable length. */
|
|
|
|
static const u32 action_lens[OVS_ACTION_ATTR_MAX + 1] = {
|
|
|
|
[OVS_ACTION_ATTR_OUTPUT] = sizeof(u32),
|
|
|
|
[OVS_ACTION_ATTR_USERSPACE] = (u32)-1,
|
|
|
|
[OVS_ACTION_ATTR_PUSH_VLAN] = sizeof(struct ovs_action_push_vlan),
|
|
|
|
[OVS_ACTION_ATTR_POP_VLAN] = 0,
|
|
|
|
[OVS_ACTION_ATTR_SET] = (u32)-1,
|
|
|
|
[OVS_ACTION_ATTR_SAMPLE] = (u32)-1
|
|
|
|
};
|
|
|
|
const struct ovs_action_push_vlan *vlan;
|
|
|
|
int type = nla_type(a);
|
2013-06-18 02:50:12 +02:00
|
|
|
bool skip_copy;
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
if (type > OVS_ACTION_ATTR_MAX ||
|
|
|
|
(action_lens[type] != nla_len(a) &&
|
|
|
|
action_lens[type] != (u32)-1))
|
|
|
|
return -EINVAL;
|
|
|
|
|
2013-06-18 02:50:12 +02:00
|
|
|
skip_copy = false;
|
2011-10-26 04:26:31 +02:00
|
|
|
switch (type) {
|
|
|
|
case OVS_ACTION_ATTR_UNSPEC:
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
case OVS_ACTION_ATTR_USERSPACE:
|
|
|
|
err = validate_userspace(a);
|
|
|
|
if (err)
|
|
|
|
return err;
|
|
|
|
break;
|
|
|
|
|
|
|
|
case OVS_ACTION_ATTR_OUTPUT:
|
|
|
|
if (nla_get_u32(a) >= DP_MAX_PORTS)
|
|
|
|
return -EINVAL;
|
|
|
|
break;
|
|
|
|
|
|
|
|
|
|
|
|
case OVS_ACTION_ATTR_POP_VLAN:
|
|
|
|
break;
|
|
|
|
|
|
|
|
case OVS_ACTION_ATTR_PUSH_VLAN:
|
|
|
|
vlan = nla_data(a);
|
|
|
|
if (vlan->vlan_tpid != htons(ETH_P_8021Q))
|
|
|
|
return -EINVAL;
|
|
|
|
if (!(vlan->vlan_tci & htons(VLAN_TAG_PRESENT)))
|
|
|
|
return -EINVAL;
|
|
|
|
break;
|
|
|
|
|
|
|
|
case OVS_ACTION_ATTR_SET:
|
2013-06-18 02:50:18 +02:00
|
|
|
err = validate_set(a, key, sfa, &skip_copy);
|
2011-10-26 04:26:31 +02:00
|
|
|
if (err)
|
|
|
|
return err;
|
|
|
|
break;
|
|
|
|
|
|
|
|
case OVS_ACTION_ATTR_SAMPLE:
|
2013-06-18 02:50:12 +02:00
|
|
|
err = validate_and_copy_sample(a, key, depth, sfa);
|
2011-10-26 04:26:31 +02:00
|
|
|
if (err)
|
|
|
|
return err;
|
2013-06-18 02:50:12 +02:00
|
|
|
skip_copy = true;
|
2011-10-26 04:26:31 +02:00
|
|
|
break;
|
|
|
|
|
|
|
|
default:
|
|
|
|
return -EINVAL;
|
|
|
|
}
|
2013-06-18 02:50:12 +02:00
|
|
|
if (!skip_copy) {
|
|
|
|
err = copy_action(a, sfa);
|
|
|
|
if (err)
|
|
|
|
return err;
|
|
|
|
}
|
2011-10-26 04:26:31 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
if (rem > 0)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void clear_stats(struct sw_flow *flow)
|
|
|
|
{
|
|
|
|
flow->used = 0;
|
|
|
|
flow->tcp_flags = 0;
|
|
|
|
flow->packet_count = 0;
|
|
|
|
flow->byte_count = 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int ovs_packet_cmd_execute(struct sk_buff *skb, struct genl_info *info)
|
|
|
|
{
|
|
|
|
struct ovs_header *ovs_header = info->userhdr;
|
|
|
|
struct nlattr **a = info->attrs;
|
|
|
|
struct sw_flow_actions *acts;
|
|
|
|
struct sk_buff *packet;
|
|
|
|
struct sw_flow *flow;
|
|
|
|
struct datapath *dp;
|
|
|
|
struct ethhdr *eth;
|
|
|
|
int len;
|
|
|
|
int err;
|
|
|
|
int key_len;
|
|
|
|
|
|
|
|
err = -EINVAL;
|
|
|
|
if (!a[OVS_PACKET_ATTR_PACKET] || !a[OVS_PACKET_ATTR_KEY] ||
|
2013-03-29 14:46:47 +01:00
|
|
|
!a[OVS_PACKET_ATTR_ACTIONS])
|
2011-10-26 04:26:31 +02:00
|
|
|
goto err;
|
|
|
|
|
|
|
|
len = nla_len(a[OVS_PACKET_ATTR_PACKET]);
|
|
|
|
packet = __dev_alloc_skb(NET_IP_ALIGN + len, GFP_KERNEL);
|
|
|
|
err = -ENOMEM;
|
|
|
|
if (!packet)
|
|
|
|
goto err;
|
|
|
|
skb_reserve(packet, NET_IP_ALIGN);
|
|
|
|
|
2013-03-29 14:46:48 +01:00
|
|
|
nla_memcpy(__skb_put(packet, len), a[OVS_PACKET_ATTR_PACKET], len);
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
skb_reset_mac_header(packet);
|
|
|
|
eth = eth_hdr(packet);
|
|
|
|
|
|
|
|
/* Normally, setting the skb 'protocol' field would be handled by a
|
|
|
|
* call to eth_type_trans(), but it assumes there's a sending
|
|
|
|
* device, which we may not have. */
|
2013-03-28 05:38:25 +01:00
|
|
|
if (ntohs(eth->h_proto) >= ETH_P_802_3_MIN)
|
2011-10-26 04:26:31 +02:00
|
|
|
packet->protocol = eth->h_proto;
|
|
|
|
else
|
|
|
|
packet->protocol = htons(ETH_P_802_2);
|
|
|
|
|
|
|
|
/* Build an sw_flow for sending this packet. */
|
|
|
|
flow = ovs_flow_alloc();
|
|
|
|
err = PTR_ERR(flow);
|
|
|
|
if (IS_ERR(flow))
|
|
|
|
goto err_kfree_skb;
|
|
|
|
|
|
|
|
err = ovs_flow_extract(packet, -1, &flow->key, &key_len);
|
|
|
|
if (err)
|
|
|
|
goto err_flow_free;
|
|
|
|
|
2013-06-18 02:50:28 +02:00
|
|
|
err = ovs_flow_metadata_from_nlattrs(flow, key_len, a[OVS_PACKET_ATTR_KEY]);
|
2011-10-26 04:26:31 +02:00
|
|
|
if (err)
|
|
|
|
goto err_flow_free;
|
2013-06-18 02:50:12 +02:00
|
|
|
acts = ovs_flow_actions_alloc(nla_len(a[OVS_PACKET_ATTR_ACTIONS]));
|
2011-10-26 04:26:31 +02:00
|
|
|
err = PTR_ERR(acts);
|
|
|
|
if (IS_ERR(acts))
|
|
|
|
goto err_flow_free;
|
2013-06-18 02:50:12 +02:00
|
|
|
|
|
|
|
err = validate_and_copy_actions(a[OVS_PACKET_ATTR_ACTIONS], &flow->key, 0, &acts);
|
2011-10-26 04:26:31 +02:00
|
|
|
rcu_assign_pointer(flow->sf_acts, acts);
|
2013-06-18 02:50:12 +02:00
|
|
|
if (err)
|
|
|
|
goto err_flow_free;
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
OVS_CB(packet)->flow = flow;
|
|
|
|
packet->priority = flow->key.phy.priority;
|
2012-11-26 20:24:11 +01:00
|
|
|
packet->mark = flow->key.phy.skb_mark;
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
rcu_read_lock();
|
2012-02-23 04:58:59 +01:00
|
|
|
dp = get_dp(sock_net(skb->sk), ovs_header->dp_ifindex);
|
2011-10-26 04:26:31 +02:00
|
|
|
err = -ENODEV;
|
|
|
|
if (!dp)
|
|
|
|
goto err_unlock;
|
|
|
|
|
|
|
|
local_bh_disable();
|
|
|
|
err = ovs_execute_actions(dp, packet);
|
|
|
|
local_bh_enable();
|
|
|
|
rcu_read_unlock();
|
|
|
|
|
|
|
|
ovs_flow_free(flow);
|
|
|
|
return err;
|
|
|
|
|
|
|
|
err_unlock:
|
|
|
|
rcu_read_unlock();
|
|
|
|
err_flow_free:
|
|
|
|
ovs_flow_free(flow);
|
|
|
|
err_kfree_skb:
|
|
|
|
kfree_skb(packet);
|
|
|
|
err:
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
|
|
|
static const struct nla_policy packet_policy[OVS_PACKET_ATTR_MAX + 1] = {
|
2013-03-29 14:46:47 +01:00
|
|
|
[OVS_PACKET_ATTR_PACKET] = { .len = ETH_HLEN },
|
2011-10-26 04:26:31 +02:00
|
|
|
[OVS_PACKET_ATTR_KEY] = { .type = NLA_NESTED },
|
|
|
|
[OVS_PACKET_ATTR_ACTIONS] = { .type = NLA_NESTED },
|
|
|
|
};
|
|
|
|
|
|
|
|
static struct genl_ops dp_packet_genl_ops[] = {
|
|
|
|
{ .cmd = OVS_PACKET_CMD_EXECUTE,
|
|
|
|
.flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
|
|
|
|
.policy = packet_policy,
|
|
|
|
.doit = ovs_packet_cmd_execute
|
|
|
|
}
|
|
|
|
};
|
|
|
|
|
|
|
|
static void get_dp_stats(struct datapath *dp, struct ovs_dp_stats *stats)
|
|
|
|
{
|
|
|
|
int i;
|
2013-04-15 22:23:03 +02:00
|
|
|
struct flow_table *table = ovsl_dereference(dp->table);
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
stats->n_flows = ovs_flow_tbl_count(table);
|
|
|
|
|
|
|
|
stats->n_hit = stats->n_missed = stats->n_lost = 0;
|
|
|
|
for_each_possible_cpu(i) {
|
|
|
|
const struct dp_stats_percpu *percpu_stats;
|
|
|
|
struct dp_stats_percpu local_stats;
|
|
|
|
unsigned int start;
|
|
|
|
|
|
|
|
percpu_stats = per_cpu_ptr(dp->stats_percpu, i);
|
|
|
|
|
|
|
|
do {
|
|
|
|
start = u64_stats_fetch_begin_bh(&percpu_stats->sync);
|
|
|
|
local_stats = *percpu_stats;
|
|
|
|
} while (u64_stats_fetch_retry_bh(&percpu_stats->sync, start));
|
|
|
|
|
|
|
|
stats->n_hit += local_stats.n_hit;
|
|
|
|
stats->n_missed += local_stats.n_missed;
|
|
|
|
stats->n_lost += local_stats.n_lost;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static const struct nla_policy flow_policy[OVS_FLOW_ATTR_MAX + 1] = {
|
|
|
|
[OVS_FLOW_ATTR_KEY] = { .type = NLA_NESTED },
|
|
|
|
[OVS_FLOW_ATTR_ACTIONS] = { .type = NLA_NESTED },
|
|
|
|
[OVS_FLOW_ATTR_CLEAR] = { .type = NLA_FLAG },
|
|
|
|
};
|
|
|
|
|
|
|
|
static struct genl_family dp_flow_genl_family = {
|
|
|
|
.id = GENL_ID_GENERATE,
|
|
|
|
.hdrsize = sizeof(struct ovs_header),
|
|
|
|
.name = OVS_FLOW_FAMILY,
|
|
|
|
.version = OVS_FLOW_VERSION,
|
2012-02-23 04:58:59 +01:00
|
|
|
.maxattr = OVS_FLOW_ATTR_MAX,
|
2013-04-23 09:48:48 +02:00
|
|
|
.netnsok = true,
|
|
|
|
.parallel_ops = true,
|
2011-10-26 04:26:31 +02:00
|
|
|
};
|
|
|
|
|
|
|
|
static struct genl_multicast_group ovs_dp_flow_multicast_group = {
|
|
|
|
.name = OVS_FLOW_MCGROUP
|
|
|
|
};
|
|
|
|
|
2013-06-18 02:50:12 +02:00
|
|
|
static int actions_to_attr(const struct nlattr *attr, int len, struct sk_buff *skb);
|
|
|
|
static int sample_action_to_attr(const struct nlattr *attr, struct sk_buff *skb)
|
|
|
|
{
|
|
|
|
const struct nlattr *a;
|
|
|
|
struct nlattr *start;
|
|
|
|
int err = 0, rem;
|
|
|
|
|
|
|
|
start = nla_nest_start(skb, OVS_ACTION_ATTR_SAMPLE);
|
|
|
|
if (!start)
|
|
|
|
return -EMSGSIZE;
|
|
|
|
|
|
|
|
nla_for_each_nested(a, attr, rem) {
|
|
|
|
int type = nla_type(a);
|
|
|
|
struct nlattr *st_sample;
|
|
|
|
|
|
|
|
switch (type) {
|
|
|
|
case OVS_SAMPLE_ATTR_PROBABILITY:
|
|
|
|
if (nla_put(skb, OVS_SAMPLE_ATTR_PROBABILITY, sizeof(u32), nla_data(a)))
|
|
|
|
return -EMSGSIZE;
|
|
|
|
break;
|
|
|
|
case OVS_SAMPLE_ATTR_ACTIONS:
|
|
|
|
st_sample = nla_nest_start(skb, OVS_SAMPLE_ATTR_ACTIONS);
|
|
|
|
if (!st_sample)
|
|
|
|
return -EMSGSIZE;
|
|
|
|
err = actions_to_attr(nla_data(a), nla_len(a), skb);
|
|
|
|
if (err)
|
|
|
|
return err;
|
|
|
|
nla_nest_end(skb, st_sample);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
nla_nest_end(skb, start);
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2013-06-18 02:50:18 +02:00
|
|
|
static int set_action_to_attr(const struct nlattr *a, struct sk_buff *skb)
|
|
|
|
{
|
|
|
|
const struct nlattr *ovs_key = nla_data(a);
|
|
|
|
int key_type = nla_type(ovs_key);
|
|
|
|
struct nlattr *start;
|
|
|
|
int err;
|
|
|
|
|
|
|
|
switch (key_type) {
|
|
|
|
case OVS_KEY_ATTR_IPV4_TUNNEL:
|
|
|
|
start = nla_nest_start(skb, OVS_ACTION_ATTR_SET);
|
|
|
|
if (!start)
|
|
|
|
return -EMSGSIZE;
|
|
|
|
|
|
|
|
err = ovs_ipv4_tun_to_nlattr(skb, nla_data(ovs_key));
|
|
|
|
if (err)
|
|
|
|
return err;
|
|
|
|
nla_nest_end(skb, start);
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
if (nla_put(skb, OVS_ACTION_ATTR_SET, nla_len(a), ovs_key))
|
|
|
|
return -EMSGSIZE;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2013-06-18 02:50:12 +02:00
|
|
|
static int actions_to_attr(const struct nlattr *attr, int len, struct sk_buff *skb)
|
|
|
|
{
|
|
|
|
const struct nlattr *a;
|
|
|
|
int rem, err;
|
|
|
|
|
|
|
|
nla_for_each_attr(a, attr, len, rem) {
|
|
|
|
int type = nla_type(a);
|
|
|
|
|
|
|
|
switch (type) {
|
2013-06-18 02:50:18 +02:00
|
|
|
case OVS_ACTION_ATTR_SET:
|
|
|
|
err = set_action_to_attr(a, skb);
|
|
|
|
if (err)
|
|
|
|
return err;
|
|
|
|
break;
|
|
|
|
|
2013-06-18 02:50:12 +02:00
|
|
|
case OVS_ACTION_ATTR_SAMPLE:
|
|
|
|
err = sample_action_to_attr(a, skb);
|
|
|
|
if (err)
|
|
|
|
return err;
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
if (nla_put(skb, type, nla_len(a), nla_data(a)))
|
|
|
|
return -EMSGSIZE;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2013-03-29 14:46:49 +01:00
|
|
|
static size_t ovs_flow_cmd_msg_size(const struct sw_flow_actions *acts)
|
|
|
|
{
|
|
|
|
return NLMSG_ALIGN(sizeof(struct ovs_header))
|
|
|
|
+ nla_total_size(key_attr_size()) /* OVS_FLOW_ATTR_KEY */
|
|
|
|
+ nla_total_size(sizeof(struct ovs_flow_stats)) /* OVS_FLOW_ATTR_STATS */
|
|
|
|
+ nla_total_size(1) /* OVS_FLOW_ATTR_TCP_FLAGS */
|
|
|
|
+ nla_total_size(8) /* OVS_FLOW_ATTR_USED */
|
|
|
|
+ nla_total_size(acts->actions_len); /* OVS_FLOW_ATTR_ACTIONS */
|
|
|
|
}
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
/* Called with ovs_mutex. */
|
2011-10-26 04:26:31 +02:00
|
|
|
static int ovs_flow_cmd_fill_info(struct sw_flow *flow, struct datapath *dp,
|
2012-09-07 22:12:54 +02:00
|
|
|
struct sk_buff *skb, u32 portid,
|
2011-10-26 04:26:31 +02:00
|
|
|
u32 seq, u32 flags, u8 cmd)
|
|
|
|
{
|
|
|
|
const int skb_orig_len = skb->len;
|
|
|
|
const struct sw_flow_actions *sf_acts;
|
2013-06-18 02:50:12 +02:00
|
|
|
struct nlattr *start;
|
2011-10-26 04:26:31 +02:00
|
|
|
struct ovs_flow_stats stats;
|
|
|
|
struct ovs_header *ovs_header;
|
|
|
|
struct nlattr *nla;
|
|
|
|
unsigned long used;
|
|
|
|
u8 tcp_flags;
|
|
|
|
int err;
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
sf_acts = ovsl_dereference(flow->sf_acts);
|
2011-10-26 04:26:31 +02:00
|
|
|
|
2012-09-07 22:12:54 +02:00
|
|
|
ovs_header = genlmsg_put(skb, portid, seq, &dp_flow_genl_family, flags, cmd);
|
2011-10-26 04:26:31 +02:00
|
|
|
if (!ovs_header)
|
|
|
|
return -EMSGSIZE;
|
|
|
|
|
|
|
|
ovs_header->dp_ifindex = get_dpifindex(dp);
|
|
|
|
|
|
|
|
nla = nla_nest_start(skb, OVS_FLOW_ATTR_KEY);
|
|
|
|
if (!nla)
|
|
|
|
goto nla_put_failure;
|
|
|
|
err = ovs_flow_to_nlattrs(&flow->key, skb);
|
|
|
|
if (err)
|
|
|
|
goto error;
|
|
|
|
nla_nest_end(skb, nla);
|
|
|
|
|
|
|
|
spin_lock_bh(&flow->lock);
|
|
|
|
used = flow->used;
|
|
|
|
stats.n_packets = flow->packet_count;
|
|
|
|
stats.n_bytes = flow->byte_count;
|
|
|
|
tcp_flags = flow->tcp_flags;
|
|
|
|
spin_unlock_bh(&flow->lock);
|
|
|
|
|
2012-03-30 05:20:48 +02:00
|
|
|
if (used &&
|
|
|
|
nla_put_u64(skb, OVS_FLOW_ATTR_USED, ovs_flow_used_time(used)))
|
|
|
|
goto nla_put_failure;
|
2011-10-26 04:26:31 +02:00
|
|
|
|
2012-03-30 05:20:48 +02:00
|
|
|
if (stats.n_packets &&
|
|
|
|
nla_put(skb, OVS_FLOW_ATTR_STATS,
|
|
|
|
sizeof(struct ovs_flow_stats), &stats))
|
|
|
|
goto nla_put_failure;
|
2011-10-26 04:26:31 +02:00
|
|
|
|
2012-03-30 05:20:48 +02:00
|
|
|
if (tcp_flags &&
|
|
|
|
nla_put_u8(skb, OVS_FLOW_ATTR_TCP_FLAGS, tcp_flags))
|
|
|
|
goto nla_put_failure;
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
/* If OVS_FLOW_ATTR_ACTIONS doesn't fit, skip dumping the actions if
|
|
|
|
* this is the first flow to be dumped into 'skb'. This is unusual for
|
|
|
|
* Netlink but individual action lists can be longer than
|
|
|
|
* NLMSG_GOODSIZE and thus entirely undumpable if we didn't do this.
|
|
|
|
* The userspace caller can always fetch the actions separately if it
|
|
|
|
* really wants them. (Most userspace callers in fact don't care.)
|
|
|
|
*
|
|
|
|
* This can only fail for dump operations because the skb is always
|
|
|
|
* properly sized for single flows.
|
|
|
|
*/
|
2013-06-18 02:50:12 +02:00
|
|
|
start = nla_nest_start(skb, OVS_FLOW_ATTR_ACTIONS);
|
|
|
|
if (start) {
|
|
|
|
err = actions_to_attr(sf_acts->actions, sf_acts->actions_len, skb);
|
|
|
|
if (!err)
|
|
|
|
nla_nest_end(skb, start);
|
|
|
|
else {
|
|
|
|
if (skb_orig_len)
|
|
|
|
goto error;
|
|
|
|
|
|
|
|
nla_nest_cancel(skb, start);
|
|
|
|
}
|
|
|
|
} else if (skb_orig_len)
|
|
|
|
goto nla_put_failure;
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
return genlmsg_end(skb, ovs_header);
|
|
|
|
|
|
|
|
nla_put_failure:
|
|
|
|
err = -EMSGSIZE;
|
|
|
|
error:
|
|
|
|
genlmsg_cancel(skb, ovs_header);
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct sk_buff *ovs_flow_cmd_alloc_info(struct sw_flow *flow)
|
|
|
|
{
|
|
|
|
const struct sw_flow_actions *sf_acts;
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
sf_acts = ovsl_dereference(flow->sf_acts);
|
2011-10-26 04:26:31 +02:00
|
|
|
|
2013-03-29 14:46:49 +01:00
|
|
|
return genlmsg_new(ovs_flow_cmd_msg_size(sf_acts), GFP_KERNEL);
|
2011-10-26 04:26:31 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
static struct sk_buff *ovs_flow_cmd_build_info(struct sw_flow *flow,
|
|
|
|
struct datapath *dp,
|
2012-09-07 22:12:54 +02:00
|
|
|
u32 portid, u32 seq, u8 cmd)
|
2011-10-26 04:26:31 +02:00
|
|
|
{
|
|
|
|
struct sk_buff *skb;
|
|
|
|
int retval;
|
|
|
|
|
|
|
|
skb = ovs_flow_cmd_alloc_info(flow);
|
|
|
|
if (!skb)
|
|
|
|
return ERR_PTR(-ENOMEM);
|
|
|
|
|
2012-09-07 22:12:54 +02:00
|
|
|
retval = ovs_flow_cmd_fill_info(flow, dp, skb, portid, seq, 0, cmd);
|
2011-10-26 04:26:31 +02:00
|
|
|
BUG_ON(retval < 0);
|
|
|
|
return skb;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info)
|
|
|
|
{
|
|
|
|
struct nlattr **a = info->attrs;
|
|
|
|
struct ovs_header *ovs_header = info->userhdr;
|
|
|
|
struct sw_flow_key key;
|
|
|
|
struct sw_flow *flow;
|
|
|
|
struct sk_buff *reply;
|
|
|
|
struct datapath *dp;
|
|
|
|
struct flow_table *table;
|
2013-06-18 02:50:12 +02:00
|
|
|
struct sw_flow_actions *acts = NULL;
|
2011-10-26 04:26:31 +02:00
|
|
|
int error;
|
|
|
|
int key_len;
|
|
|
|
|
|
|
|
/* Extract key. */
|
|
|
|
error = -EINVAL;
|
|
|
|
if (!a[OVS_FLOW_ATTR_KEY])
|
|
|
|
goto error;
|
|
|
|
error = ovs_flow_from_nlattrs(&key, &key_len, a[OVS_FLOW_ATTR_KEY]);
|
|
|
|
if (error)
|
|
|
|
goto error;
|
|
|
|
|
|
|
|
/* Validate actions. */
|
|
|
|
if (a[OVS_FLOW_ATTR_ACTIONS]) {
|
2013-06-18 02:50:12 +02:00
|
|
|
acts = ovs_flow_actions_alloc(nla_len(a[OVS_FLOW_ATTR_ACTIONS]));
|
|
|
|
error = PTR_ERR(acts);
|
|
|
|
if (IS_ERR(acts))
|
2011-10-26 04:26:31 +02:00
|
|
|
goto error;
|
2013-06-18 02:50:12 +02:00
|
|
|
|
|
|
|
error = validate_and_copy_actions(a[OVS_FLOW_ATTR_ACTIONS], &key, 0, &acts);
|
|
|
|
if (error)
|
|
|
|
goto err_kfree;
|
2011-10-26 04:26:31 +02:00
|
|
|
} else if (info->genlhdr->cmd == OVS_FLOW_CMD_NEW) {
|
|
|
|
error = -EINVAL;
|
|
|
|
goto error;
|
|
|
|
}
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_lock();
|
2012-02-23 04:58:59 +01:00
|
|
|
dp = get_dp(sock_net(skb->sk), ovs_header->dp_ifindex);
|
2011-10-26 04:26:31 +02:00
|
|
|
error = -ENODEV;
|
|
|
|
if (!dp)
|
2013-04-15 22:23:03 +02:00
|
|
|
goto err_unlock_ovs;
|
2011-10-26 04:26:31 +02:00
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
table = ovsl_dereference(dp->table);
|
2011-10-26 04:26:31 +02:00
|
|
|
flow = ovs_flow_tbl_lookup(table, &key, key_len);
|
|
|
|
if (!flow) {
|
|
|
|
/* Bail out if we're not allowed to create a new flow. */
|
|
|
|
error = -ENOENT;
|
|
|
|
if (info->genlhdr->cmd == OVS_FLOW_CMD_SET)
|
2013-04-15 22:23:03 +02:00
|
|
|
goto err_unlock_ovs;
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
/* Expand table, if necessary, to make room. */
|
|
|
|
if (ovs_flow_tbl_need_to_expand(table)) {
|
|
|
|
struct flow_table *new_table;
|
|
|
|
|
|
|
|
new_table = ovs_flow_tbl_expand(table);
|
|
|
|
if (!IS_ERR(new_table)) {
|
|
|
|
rcu_assign_pointer(dp->table, new_table);
|
|
|
|
ovs_flow_tbl_deferred_destroy(table);
|
2013-04-15 22:23:03 +02:00
|
|
|
table = ovsl_dereference(dp->table);
|
2011-10-26 04:26:31 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Allocate flow. */
|
|
|
|
flow = ovs_flow_alloc();
|
|
|
|
if (IS_ERR(flow)) {
|
|
|
|
error = PTR_ERR(flow);
|
2013-04-15 22:23:03 +02:00
|
|
|
goto err_unlock_ovs;
|
2011-10-26 04:26:31 +02:00
|
|
|
}
|
|
|
|
clear_stats(flow);
|
|
|
|
|
|
|
|
rcu_assign_pointer(flow->sf_acts, acts);
|
|
|
|
|
|
|
|
/* Put flow in bucket. */
|
2013-06-18 02:50:28 +02:00
|
|
|
ovs_flow_tbl_insert(table, flow, &key, key_len);
|
2011-10-26 04:26:31 +02:00
|
|
|
|
2012-09-07 22:12:54 +02:00
|
|
|
reply = ovs_flow_cmd_build_info(flow, dp, info->snd_portid,
|
2011-10-26 04:26:31 +02:00
|
|
|
info->snd_seq,
|
|
|
|
OVS_FLOW_CMD_NEW);
|
|
|
|
} else {
|
|
|
|
/* We found a matching flow. */
|
|
|
|
struct sw_flow_actions *old_acts;
|
|
|
|
|
|
|
|
/* Bail out if we're not allowed to modify an existing flow.
|
|
|
|
* We accept NLM_F_CREATE in place of the intended NLM_F_EXCL
|
|
|
|
* because Generic Netlink treats the latter as a dump
|
|
|
|
* request. We also accept NLM_F_EXCL in case that bug ever
|
|
|
|
* gets fixed.
|
|
|
|
*/
|
|
|
|
error = -EEXIST;
|
|
|
|
if (info->genlhdr->cmd == OVS_FLOW_CMD_NEW &&
|
|
|
|
info->nlhdr->nlmsg_flags & (NLM_F_CREATE | NLM_F_EXCL))
|
2013-04-15 22:23:03 +02:00
|
|
|
goto err_unlock_ovs;
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
/* Update actions. */
|
2013-04-15 22:23:03 +02:00
|
|
|
old_acts = ovsl_dereference(flow->sf_acts);
|
2013-06-18 02:50:12 +02:00
|
|
|
rcu_assign_pointer(flow->sf_acts, acts);
|
|
|
|
ovs_flow_deferred_free_acts(old_acts);
|
2011-10-26 04:26:31 +02:00
|
|
|
|
2012-09-07 22:12:54 +02:00
|
|
|
reply = ovs_flow_cmd_build_info(flow, dp, info->snd_portid,
|
2011-10-26 04:26:31 +02:00
|
|
|
info->snd_seq, OVS_FLOW_CMD_NEW);
|
|
|
|
|
|
|
|
/* Clear stats. */
|
|
|
|
if (a[OVS_FLOW_ATTR_CLEAR]) {
|
|
|
|
spin_lock_bh(&flow->lock);
|
|
|
|
clear_stats(flow);
|
|
|
|
spin_unlock_bh(&flow->lock);
|
|
|
|
}
|
|
|
|
}
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_unlock();
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
if (!IS_ERR(reply))
|
2013-03-29 14:46:50 +01:00
|
|
|
ovs_notify(reply, info, &ovs_dp_flow_multicast_group);
|
2011-10-26 04:26:31 +02:00
|
|
|
else
|
2012-02-23 04:58:59 +01:00
|
|
|
netlink_set_err(sock_net(skb->sk)->genl_sock, 0,
|
2011-10-26 04:26:31 +02:00
|
|
|
ovs_dp_flow_multicast_group.id, PTR_ERR(reply));
|
|
|
|
return 0;
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
err_unlock_ovs:
|
|
|
|
ovs_unlock();
|
2013-06-18 02:50:12 +02:00
|
|
|
err_kfree:
|
|
|
|
kfree(acts);
|
2011-10-26 04:26:31 +02:00
|
|
|
error:
|
|
|
|
return error;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int ovs_flow_cmd_get(struct sk_buff *skb, struct genl_info *info)
|
|
|
|
{
|
|
|
|
struct nlattr **a = info->attrs;
|
|
|
|
struct ovs_header *ovs_header = info->userhdr;
|
|
|
|
struct sw_flow_key key;
|
|
|
|
struct sk_buff *reply;
|
|
|
|
struct sw_flow *flow;
|
|
|
|
struct datapath *dp;
|
|
|
|
struct flow_table *table;
|
|
|
|
int err;
|
|
|
|
int key_len;
|
|
|
|
|
|
|
|
if (!a[OVS_FLOW_ATTR_KEY])
|
|
|
|
return -EINVAL;
|
|
|
|
err = ovs_flow_from_nlattrs(&key, &key_len, a[OVS_FLOW_ATTR_KEY]);
|
|
|
|
if (err)
|
|
|
|
return err;
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_lock();
|
2012-02-23 04:58:59 +01:00
|
|
|
dp = get_dp(sock_net(skb->sk), ovs_header->dp_ifindex);
|
2013-04-15 22:23:03 +02:00
|
|
|
if (!dp) {
|
|
|
|
err = -ENODEV;
|
|
|
|
goto unlock;
|
|
|
|
}
|
2011-10-26 04:26:31 +02:00
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
table = ovsl_dereference(dp->table);
|
2011-10-26 04:26:31 +02:00
|
|
|
flow = ovs_flow_tbl_lookup(table, &key, key_len);
|
2013-04-15 22:23:03 +02:00
|
|
|
if (!flow) {
|
|
|
|
err = -ENOENT;
|
|
|
|
goto unlock;
|
|
|
|
}
|
2011-10-26 04:26:31 +02:00
|
|
|
|
2012-09-07 22:12:54 +02:00
|
|
|
reply = ovs_flow_cmd_build_info(flow, dp, info->snd_portid,
|
2011-10-26 04:26:31 +02:00
|
|
|
info->snd_seq, OVS_FLOW_CMD_NEW);
|
2013-04-15 22:23:03 +02:00
|
|
|
if (IS_ERR(reply)) {
|
|
|
|
err = PTR_ERR(reply);
|
|
|
|
goto unlock;
|
|
|
|
}
|
2011-10-26 04:26:31 +02:00
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_unlock();
|
2011-10-26 04:26:31 +02:00
|
|
|
return genlmsg_reply(reply, info);
|
2013-04-15 22:23:03 +02:00
|
|
|
unlock:
|
|
|
|
ovs_unlock();
|
|
|
|
return err;
|
2011-10-26 04:26:31 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
static int ovs_flow_cmd_del(struct sk_buff *skb, struct genl_info *info)
|
|
|
|
{
|
|
|
|
struct nlattr **a = info->attrs;
|
|
|
|
struct ovs_header *ovs_header = info->userhdr;
|
|
|
|
struct sw_flow_key key;
|
|
|
|
struct sk_buff *reply;
|
|
|
|
struct sw_flow *flow;
|
|
|
|
struct datapath *dp;
|
|
|
|
struct flow_table *table;
|
|
|
|
int err;
|
|
|
|
int key_len;
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_lock();
|
2012-02-23 04:58:59 +01:00
|
|
|
dp = get_dp(sock_net(skb->sk), ovs_header->dp_ifindex);
|
2013-04-15 22:23:03 +02:00
|
|
|
if (!dp) {
|
|
|
|
err = -ENODEV;
|
|
|
|
goto unlock;
|
|
|
|
}
|
2012-02-23 04:58:59 +01:00
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
if (!a[OVS_FLOW_ATTR_KEY]) {
|
|
|
|
err = flush_flows(dp);
|
|
|
|
goto unlock;
|
|
|
|
}
|
2011-10-26 04:26:31 +02:00
|
|
|
err = ovs_flow_from_nlattrs(&key, &key_len, a[OVS_FLOW_ATTR_KEY]);
|
|
|
|
if (err)
|
2013-04-15 22:23:03 +02:00
|
|
|
goto unlock;
|
2011-10-26 04:26:31 +02:00
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
table = ovsl_dereference(dp->table);
|
2011-10-26 04:26:31 +02:00
|
|
|
flow = ovs_flow_tbl_lookup(table, &key, key_len);
|
2013-04-15 22:23:03 +02:00
|
|
|
if (!flow) {
|
|
|
|
err = -ENOENT;
|
|
|
|
goto unlock;
|
|
|
|
}
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
reply = ovs_flow_cmd_alloc_info(flow);
|
2013-04-15 22:23:03 +02:00
|
|
|
if (!reply) {
|
|
|
|
err = -ENOMEM;
|
|
|
|
goto unlock;
|
|
|
|
}
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
ovs_flow_tbl_remove(table, flow);
|
|
|
|
|
2012-09-07 22:12:54 +02:00
|
|
|
err = ovs_flow_cmd_fill_info(flow, dp, reply, info->snd_portid,
|
2011-10-26 04:26:31 +02:00
|
|
|
info->snd_seq, 0, OVS_FLOW_CMD_DEL);
|
|
|
|
BUG_ON(err < 0);
|
|
|
|
|
|
|
|
ovs_flow_deferred_free(flow);
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_unlock();
|
2011-10-26 04:26:31 +02:00
|
|
|
|
2013-03-29 14:46:50 +01:00
|
|
|
ovs_notify(reply, info, &ovs_dp_flow_multicast_group);
|
2011-10-26 04:26:31 +02:00
|
|
|
return 0;
|
2013-04-15 22:23:03 +02:00
|
|
|
unlock:
|
|
|
|
ovs_unlock();
|
|
|
|
return err;
|
2011-10-26 04:26:31 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
static int ovs_flow_cmd_dump(struct sk_buff *skb, struct netlink_callback *cb)
|
|
|
|
{
|
|
|
|
struct ovs_header *ovs_header = genlmsg_data(nlmsg_data(cb->nlh));
|
|
|
|
struct datapath *dp;
|
|
|
|
struct flow_table *table;
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_lock();
|
2012-02-23 04:58:59 +01:00
|
|
|
dp = get_dp(sock_net(skb->sk), ovs_header->dp_ifindex);
|
2013-04-15 22:23:03 +02:00
|
|
|
if (!dp) {
|
|
|
|
ovs_unlock();
|
2011-10-26 04:26:31 +02:00
|
|
|
return -ENODEV;
|
2013-04-15 22:23:03 +02:00
|
|
|
}
|
2011-10-26 04:26:31 +02:00
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
table = ovsl_dereference(dp->table);
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
for (;;) {
|
|
|
|
struct sw_flow *flow;
|
|
|
|
u32 bucket, obj;
|
|
|
|
|
|
|
|
bucket = cb->args[0];
|
|
|
|
obj = cb->args[1];
|
|
|
|
flow = ovs_flow_tbl_next(table, &bucket, &obj);
|
|
|
|
if (!flow)
|
|
|
|
break;
|
|
|
|
|
|
|
|
if (ovs_flow_cmd_fill_info(flow, dp, skb,
|
2012-09-07 22:12:54 +02:00
|
|
|
NETLINK_CB(cb->skb).portid,
|
2011-10-26 04:26:31 +02:00
|
|
|
cb->nlh->nlmsg_seq, NLM_F_MULTI,
|
|
|
|
OVS_FLOW_CMD_NEW) < 0)
|
|
|
|
break;
|
|
|
|
|
|
|
|
cb->args[0] = bucket;
|
|
|
|
cb->args[1] = obj;
|
|
|
|
}
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_unlock();
|
2011-10-26 04:26:31 +02:00
|
|
|
return skb->len;
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct genl_ops dp_flow_genl_ops[] = {
|
|
|
|
{ .cmd = OVS_FLOW_CMD_NEW,
|
|
|
|
.flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
|
|
|
|
.policy = flow_policy,
|
|
|
|
.doit = ovs_flow_cmd_new_or_set
|
|
|
|
},
|
|
|
|
{ .cmd = OVS_FLOW_CMD_DEL,
|
|
|
|
.flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
|
|
|
|
.policy = flow_policy,
|
|
|
|
.doit = ovs_flow_cmd_del
|
|
|
|
},
|
|
|
|
{ .cmd = OVS_FLOW_CMD_GET,
|
|
|
|
.flags = 0, /* OK for unprivileged users. */
|
|
|
|
.policy = flow_policy,
|
|
|
|
.doit = ovs_flow_cmd_get,
|
|
|
|
.dumpit = ovs_flow_cmd_dump
|
|
|
|
},
|
|
|
|
{ .cmd = OVS_FLOW_CMD_SET,
|
|
|
|
.flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
|
|
|
|
.policy = flow_policy,
|
|
|
|
.doit = ovs_flow_cmd_new_or_set,
|
|
|
|
},
|
|
|
|
};
|
|
|
|
|
|
|
|
static const struct nla_policy datapath_policy[OVS_DP_ATTR_MAX + 1] = {
|
|
|
|
[OVS_DP_ATTR_NAME] = { .type = NLA_NUL_STRING, .len = IFNAMSIZ - 1 },
|
|
|
|
[OVS_DP_ATTR_UPCALL_PID] = { .type = NLA_U32 },
|
|
|
|
};
|
|
|
|
|
|
|
|
static struct genl_family dp_datapath_genl_family = {
|
|
|
|
.id = GENL_ID_GENERATE,
|
|
|
|
.hdrsize = sizeof(struct ovs_header),
|
|
|
|
.name = OVS_DATAPATH_FAMILY,
|
|
|
|
.version = OVS_DATAPATH_VERSION,
|
2012-02-23 04:58:59 +01:00
|
|
|
.maxattr = OVS_DP_ATTR_MAX,
|
2013-04-23 09:48:48 +02:00
|
|
|
.netnsok = true,
|
|
|
|
.parallel_ops = true,
|
2011-10-26 04:26:31 +02:00
|
|
|
};
|
|
|
|
|
|
|
|
static struct genl_multicast_group ovs_dp_datapath_multicast_group = {
|
|
|
|
.name = OVS_DATAPATH_MCGROUP
|
|
|
|
};
|
|
|
|
|
2013-03-29 14:46:49 +01:00
|
|
|
static size_t ovs_dp_cmd_msg_size(void)
|
|
|
|
{
|
|
|
|
size_t msgsize = NLMSG_ALIGN(sizeof(struct ovs_header));
|
|
|
|
|
|
|
|
msgsize += nla_total_size(IFNAMSIZ);
|
|
|
|
msgsize += nla_total_size(sizeof(struct ovs_dp_stats));
|
|
|
|
|
|
|
|
return msgsize;
|
|
|
|
}
|
|
|
|
|
2011-10-26 04:26:31 +02:00
|
|
|
static int ovs_dp_cmd_fill_info(struct datapath *dp, struct sk_buff *skb,
|
2012-09-07 22:12:54 +02:00
|
|
|
u32 portid, u32 seq, u32 flags, u8 cmd)
|
2011-10-26 04:26:31 +02:00
|
|
|
{
|
|
|
|
struct ovs_header *ovs_header;
|
|
|
|
struct ovs_dp_stats dp_stats;
|
|
|
|
int err;
|
|
|
|
|
2012-09-07 22:12:54 +02:00
|
|
|
ovs_header = genlmsg_put(skb, portid, seq, &dp_datapath_genl_family,
|
2011-10-26 04:26:31 +02:00
|
|
|
flags, cmd);
|
|
|
|
if (!ovs_header)
|
|
|
|
goto error;
|
|
|
|
|
|
|
|
ovs_header->dp_ifindex = get_dpifindex(dp);
|
|
|
|
|
|
|
|
rcu_read_lock();
|
|
|
|
err = nla_put_string(skb, OVS_DP_ATTR_NAME, ovs_dp_name(dp));
|
|
|
|
rcu_read_unlock();
|
|
|
|
if (err)
|
|
|
|
goto nla_put_failure;
|
|
|
|
|
|
|
|
get_dp_stats(dp, &dp_stats);
|
2012-03-30 05:20:48 +02:00
|
|
|
if (nla_put(skb, OVS_DP_ATTR_STATS, sizeof(struct ovs_dp_stats), &dp_stats))
|
|
|
|
goto nla_put_failure;
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
return genlmsg_end(skb, ovs_header);
|
|
|
|
|
|
|
|
nla_put_failure:
|
|
|
|
genlmsg_cancel(skb, ovs_header);
|
|
|
|
error:
|
|
|
|
return -EMSGSIZE;
|
|
|
|
}
|
|
|
|
|
2012-09-07 22:12:54 +02:00
|
|
|
static struct sk_buff *ovs_dp_cmd_build_info(struct datapath *dp, u32 portid,
|
2011-10-26 04:26:31 +02:00
|
|
|
u32 seq, u8 cmd)
|
|
|
|
{
|
|
|
|
struct sk_buff *skb;
|
|
|
|
int retval;
|
|
|
|
|
2013-03-29 14:46:49 +01:00
|
|
|
skb = genlmsg_new(ovs_dp_cmd_msg_size(), GFP_KERNEL);
|
2011-10-26 04:26:31 +02:00
|
|
|
if (!skb)
|
|
|
|
return ERR_PTR(-ENOMEM);
|
|
|
|
|
2012-09-07 22:12:54 +02:00
|
|
|
retval = ovs_dp_cmd_fill_info(dp, skb, portid, seq, 0, cmd);
|
2011-10-26 04:26:31 +02:00
|
|
|
if (retval < 0) {
|
|
|
|
kfree_skb(skb);
|
|
|
|
return ERR_PTR(retval);
|
|
|
|
}
|
|
|
|
return skb;
|
|
|
|
}
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
/* Called with ovs_mutex. */
|
2012-02-23 04:58:59 +01:00
|
|
|
static struct datapath *lookup_datapath(struct net *net,
|
|
|
|
struct ovs_header *ovs_header,
|
2011-10-26 04:26:31 +02:00
|
|
|
struct nlattr *a[OVS_DP_ATTR_MAX + 1])
|
|
|
|
{
|
|
|
|
struct datapath *dp;
|
|
|
|
|
|
|
|
if (!a[OVS_DP_ATTR_NAME])
|
2012-02-23 04:58:59 +01:00
|
|
|
dp = get_dp(net, ovs_header->dp_ifindex);
|
2011-10-26 04:26:31 +02:00
|
|
|
else {
|
|
|
|
struct vport *vport;
|
|
|
|
|
|
|
|
rcu_read_lock();
|
2012-02-23 04:58:59 +01:00
|
|
|
vport = ovs_vport_locate(net, nla_data(a[OVS_DP_ATTR_NAME]));
|
2011-10-26 04:26:31 +02:00
|
|
|
dp = vport && vport->port_no == OVSP_LOCAL ? vport->dp : NULL;
|
|
|
|
rcu_read_unlock();
|
|
|
|
}
|
|
|
|
return dp ? dp : ERR_PTR(-ENODEV);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info)
|
|
|
|
{
|
|
|
|
struct nlattr **a = info->attrs;
|
|
|
|
struct vport_parms parms;
|
|
|
|
struct sk_buff *reply;
|
|
|
|
struct datapath *dp;
|
|
|
|
struct vport *vport;
|
2012-02-23 04:58:59 +01:00
|
|
|
struct ovs_net *ovs_net;
|
2012-08-23 21:40:54 +02:00
|
|
|
int err, i;
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
err = -EINVAL;
|
|
|
|
if (!a[OVS_DP_ATTR_NAME] || !a[OVS_DP_ATTR_UPCALL_PID])
|
|
|
|
goto err;
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_lock();
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
err = -ENOMEM;
|
|
|
|
dp = kzalloc(sizeof(*dp), GFP_KERNEL);
|
|
|
|
if (dp == NULL)
|
2013-04-15 22:23:03 +02:00
|
|
|
goto err_unlock_ovs;
|
2012-02-23 04:58:59 +01:00
|
|
|
|
|
|
|
ovs_dp_set_net(dp, hold_net(sock_net(skb->sk)));
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
/* Allocate table. */
|
|
|
|
err = -ENOMEM;
|
|
|
|
rcu_assign_pointer(dp->table, ovs_flow_tbl_alloc(TBL_MIN_BUCKETS));
|
|
|
|
if (!dp->table)
|
|
|
|
goto err_free_dp;
|
|
|
|
|
|
|
|
dp->stats_percpu = alloc_percpu(struct dp_stats_percpu);
|
|
|
|
if (!dp->stats_percpu) {
|
|
|
|
err = -ENOMEM;
|
|
|
|
goto err_destroy_table;
|
|
|
|
}
|
|
|
|
|
2012-08-23 21:40:54 +02:00
|
|
|
dp->ports = kmalloc(DP_VPORT_HASH_BUCKETS * sizeof(struct hlist_head),
|
|
|
|
GFP_KERNEL);
|
|
|
|
if (!dp->ports) {
|
|
|
|
err = -ENOMEM;
|
|
|
|
goto err_destroy_percpu;
|
|
|
|
}
|
|
|
|
|
|
|
|
for (i = 0; i < DP_VPORT_HASH_BUCKETS; i++)
|
|
|
|
INIT_HLIST_HEAD(&dp->ports[i]);
|
|
|
|
|
2011-10-26 04:26:31 +02:00
|
|
|
/* Set up our datapath device. */
|
|
|
|
parms.name = nla_data(a[OVS_DP_ATTR_NAME]);
|
|
|
|
parms.type = OVS_VPORT_TYPE_INTERNAL;
|
|
|
|
parms.options = NULL;
|
|
|
|
parms.dp = dp;
|
|
|
|
parms.port_no = OVSP_LOCAL;
|
2012-09-07 22:12:54 +02:00
|
|
|
parms.upcall_portid = nla_get_u32(a[OVS_DP_ATTR_UPCALL_PID]);
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
vport = new_vport(&parms);
|
|
|
|
if (IS_ERR(vport)) {
|
|
|
|
err = PTR_ERR(vport);
|
|
|
|
if (err == -EBUSY)
|
|
|
|
err = -EEXIST;
|
|
|
|
|
2012-08-23 21:40:54 +02:00
|
|
|
goto err_destroy_ports_array;
|
2011-10-26 04:26:31 +02:00
|
|
|
}
|
|
|
|
|
2012-09-07 22:12:54 +02:00
|
|
|
reply = ovs_dp_cmd_build_info(dp, info->snd_portid,
|
2011-10-26 04:26:31 +02:00
|
|
|
info->snd_seq, OVS_DP_CMD_NEW);
|
|
|
|
err = PTR_ERR(reply);
|
|
|
|
if (IS_ERR(reply))
|
|
|
|
goto err_destroy_local_port;
|
|
|
|
|
2012-02-23 04:58:59 +01:00
|
|
|
ovs_net = net_generic(ovs_dp_get_net(dp), ovs_net_id);
|
|
|
|
list_add_tail(&dp->list_node, &ovs_net->dps);
|
2013-04-15 22:23:03 +02:00
|
|
|
|
|
|
|
ovs_unlock();
|
2011-10-26 04:26:31 +02:00
|
|
|
|
2013-03-29 14:46:50 +01:00
|
|
|
ovs_notify(reply, info, &ovs_dp_datapath_multicast_group);
|
2011-10-26 04:26:31 +02:00
|
|
|
return 0;
|
|
|
|
|
|
|
|
err_destroy_local_port:
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_dp_detach_port(ovs_vport_ovsl(dp, OVSP_LOCAL));
|
2012-08-23 21:40:54 +02:00
|
|
|
err_destroy_ports_array:
|
|
|
|
kfree(dp->ports);
|
2011-10-26 04:26:31 +02:00
|
|
|
err_destroy_percpu:
|
|
|
|
free_percpu(dp->stats_percpu);
|
|
|
|
err_destroy_table:
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_flow_tbl_destroy(ovsl_dereference(dp->table));
|
2011-10-26 04:26:31 +02:00
|
|
|
err_free_dp:
|
2012-02-23 04:58:59 +01:00
|
|
|
release_net(ovs_dp_get_net(dp));
|
2011-10-26 04:26:31 +02:00
|
|
|
kfree(dp);
|
2013-04-15 22:23:03 +02:00
|
|
|
err_unlock_ovs:
|
|
|
|
ovs_unlock();
|
2011-10-26 04:26:31 +02:00
|
|
|
err:
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
/* Called with ovs_mutex. */
|
2012-02-23 04:58:59 +01:00
|
|
|
static void __dp_destroy(struct datapath *dp)
|
2011-10-26 04:26:31 +02:00
|
|
|
{
|
2012-08-23 21:40:54 +02:00
|
|
|
int i;
|
2011-10-26 04:26:31 +02:00
|
|
|
|
2012-08-23 21:40:54 +02:00
|
|
|
for (i = 0; i < DP_VPORT_HASH_BUCKETS; i++) {
|
|
|
|
struct vport *vport;
|
hlist: drop the node parameter from iterators
I'm not sure why, but the hlist for each entry iterators were conceived
list_for_each_entry(pos, head, member)
The hlist ones were greedy and wanted an extra parameter:
hlist_for_each_entry(tpos, pos, head, member)
Why did they need an extra pos parameter? I'm not quite sure. Not only
they don't really need it, it also prevents the iterator from looking
exactly like the list iterator, which is unfortunate.
Besides the semantic patch, there was some manual work required:
- Fix up the actual hlist iterators in linux/list.h
- Fix up the declaration of other iterators based on the hlist ones.
- A very small amount of places were using the 'node' parameter, this
was modified to use 'obj->member' instead.
- Coccinelle didn't handle the hlist_for_each_entry_safe iterator
properly, so those had to be fixed up manually.
The semantic patch which is mostly the work of Peter Senna Tschudin is here:
@@
iterator name hlist_for_each_entry, hlist_for_each_entry_continue, hlist_for_each_entry_from, hlist_for_each_entry_rcu, hlist_for_each_entry_rcu_bh, hlist_for_each_entry_continue_rcu_bh, for_each_busy_worker, ax25_uid_for_each, ax25_for_each, inet_bind_bucket_for_each, sctp_for_each_hentry, sk_for_each, sk_for_each_rcu, sk_for_each_from, sk_for_each_safe, sk_for_each_bound, hlist_for_each_entry_safe, hlist_for_each_entry_continue_rcu, nr_neigh_for_each, nr_neigh_for_each_safe, nr_node_for_each, nr_node_for_each_safe, for_each_gfn_indirect_valid_sp, for_each_gfn_sp, for_each_host;
type T;
expression a,c,d,e;
identifier b;
statement S;
@@
-T b;
<+... when != b
(
hlist_for_each_entry(a,
- b,
c, d) S
|
hlist_for_each_entry_continue(a,
- b,
c) S
|
hlist_for_each_entry_from(a,
- b,
c) S
|
hlist_for_each_entry_rcu(a,
- b,
c, d) S
|
hlist_for_each_entry_rcu_bh(a,
- b,
c, d) S
|
hlist_for_each_entry_continue_rcu_bh(a,
- b,
c) S
|
for_each_busy_worker(a, c,
- b,
d) S
|
ax25_uid_for_each(a,
- b,
c) S
|
ax25_for_each(a,
- b,
c) S
|
inet_bind_bucket_for_each(a,
- b,
c) S
|
sctp_for_each_hentry(a,
- b,
c) S
|
sk_for_each(a,
- b,
c) S
|
sk_for_each_rcu(a,
- b,
c) S
|
sk_for_each_from
-(a, b)
+(a)
S
+ sk_for_each_from(a) S
|
sk_for_each_safe(a,
- b,
c, d) S
|
sk_for_each_bound(a,
- b,
c) S
|
hlist_for_each_entry_safe(a,
- b,
c, d, e) S
|
hlist_for_each_entry_continue_rcu(a,
- b,
c) S
|
nr_neigh_for_each(a,
- b,
c) S
|
nr_neigh_for_each_safe(a,
- b,
c, d) S
|
nr_node_for_each(a,
- b,
c) S
|
nr_node_for_each_safe(a,
- b,
c, d) S
|
- for_each_gfn_sp(a, c, d, b) S
+ for_each_gfn_sp(a, c, d) S
|
- for_each_gfn_indirect_valid_sp(a, c, d, b) S
+ for_each_gfn_indirect_valid_sp(a, c, d) S
|
for_each_host(a,
- b,
c) S
|
for_each_host_safe(a,
- b,
c, d) S
|
for_each_mesh_entry(a,
- b,
c, d) S
)
...+>
[akpm@linux-foundation.org: drop bogus change from net/ipv4/raw.c]
[akpm@linux-foundation.org: drop bogus hunk from net/ipv6/raw.c]
[akpm@linux-foundation.org: checkpatch fixes]
[akpm@linux-foundation.org: fix warnings]
[akpm@linux-foudnation.org: redo intrusive kvm changes]
Tested-by: Peter Senna Tschudin <peter.senna@gmail.com>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-28 02:06:00 +01:00
|
|
|
struct hlist_node *n;
|
2012-08-23 21:40:54 +02:00
|
|
|
|
hlist: drop the node parameter from iterators
I'm not sure why, but the hlist for each entry iterators were conceived
list_for_each_entry(pos, head, member)
The hlist ones were greedy and wanted an extra parameter:
hlist_for_each_entry(tpos, pos, head, member)
Why did they need an extra pos parameter? I'm not quite sure. Not only
they don't really need it, it also prevents the iterator from looking
exactly like the list iterator, which is unfortunate.
Besides the semantic patch, there was some manual work required:
- Fix up the actual hlist iterators in linux/list.h
- Fix up the declaration of other iterators based on the hlist ones.
- A very small amount of places were using the 'node' parameter, this
was modified to use 'obj->member' instead.
- Coccinelle didn't handle the hlist_for_each_entry_safe iterator
properly, so those had to be fixed up manually.
The semantic patch which is mostly the work of Peter Senna Tschudin is here:
@@
iterator name hlist_for_each_entry, hlist_for_each_entry_continue, hlist_for_each_entry_from, hlist_for_each_entry_rcu, hlist_for_each_entry_rcu_bh, hlist_for_each_entry_continue_rcu_bh, for_each_busy_worker, ax25_uid_for_each, ax25_for_each, inet_bind_bucket_for_each, sctp_for_each_hentry, sk_for_each, sk_for_each_rcu, sk_for_each_from, sk_for_each_safe, sk_for_each_bound, hlist_for_each_entry_safe, hlist_for_each_entry_continue_rcu, nr_neigh_for_each, nr_neigh_for_each_safe, nr_node_for_each, nr_node_for_each_safe, for_each_gfn_indirect_valid_sp, for_each_gfn_sp, for_each_host;
type T;
expression a,c,d,e;
identifier b;
statement S;
@@
-T b;
<+... when != b
(
hlist_for_each_entry(a,
- b,
c, d) S
|
hlist_for_each_entry_continue(a,
- b,
c) S
|
hlist_for_each_entry_from(a,
- b,
c) S
|
hlist_for_each_entry_rcu(a,
- b,
c, d) S
|
hlist_for_each_entry_rcu_bh(a,
- b,
c, d) S
|
hlist_for_each_entry_continue_rcu_bh(a,
- b,
c) S
|
for_each_busy_worker(a, c,
- b,
d) S
|
ax25_uid_for_each(a,
- b,
c) S
|
ax25_for_each(a,
- b,
c) S
|
inet_bind_bucket_for_each(a,
- b,
c) S
|
sctp_for_each_hentry(a,
- b,
c) S
|
sk_for_each(a,
- b,
c) S
|
sk_for_each_rcu(a,
- b,
c) S
|
sk_for_each_from
-(a, b)
+(a)
S
+ sk_for_each_from(a) S
|
sk_for_each_safe(a,
- b,
c, d) S
|
sk_for_each_bound(a,
- b,
c) S
|
hlist_for_each_entry_safe(a,
- b,
c, d, e) S
|
hlist_for_each_entry_continue_rcu(a,
- b,
c) S
|
nr_neigh_for_each(a,
- b,
c) S
|
nr_neigh_for_each_safe(a,
- b,
c, d) S
|
nr_node_for_each(a,
- b,
c) S
|
nr_node_for_each_safe(a,
- b,
c, d) S
|
- for_each_gfn_sp(a, c, d, b) S
+ for_each_gfn_sp(a, c, d) S
|
- for_each_gfn_indirect_valid_sp(a, c, d, b) S
+ for_each_gfn_indirect_valid_sp(a, c, d) S
|
for_each_host(a,
- b,
c) S
|
for_each_host_safe(a,
- b,
c, d) S
|
for_each_mesh_entry(a,
- b,
c, d) S
)
...+>
[akpm@linux-foundation.org: drop bogus change from net/ipv4/raw.c]
[akpm@linux-foundation.org: drop bogus hunk from net/ipv6/raw.c]
[akpm@linux-foundation.org: checkpatch fixes]
[akpm@linux-foundation.org: fix warnings]
[akpm@linux-foudnation.org: redo intrusive kvm changes]
Tested-by: Peter Senna Tschudin <peter.senna@gmail.com>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-28 02:06:00 +01:00
|
|
|
hlist_for_each_entry_safe(vport, n, &dp->ports[i], dp_hash_node)
|
2012-08-23 21:40:54 +02:00
|
|
|
if (vport->port_no != OVSP_LOCAL)
|
|
|
|
ovs_dp_detach_port(vport);
|
|
|
|
}
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
list_del(&dp->list_node);
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
/* OVSP_LOCAL is datapath internal port. We need to make sure that
|
|
|
|
* all port in datapath are destroyed first before freeing datapath.
|
2011-10-26 04:26:31 +02:00
|
|
|
*/
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_dp_detach_port(ovs_vport_ovsl(dp, OVSP_LOCAL));
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
call_rcu(&dp->rcu, destroy_dp_rcu);
|
2012-02-23 04:58:59 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
static int ovs_dp_cmd_del(struct sk_buff *skb, struct genl_info *info)
|
|
|
|
{
|
|
|
|
struct sk_buff *reply;
|
|
|
|
struct datapath *dp;
|
|
|
|
int err;
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_lock();
|
2012-02-23 04:58:59 +01:00
|
|
|
dp = lookup_datapath(sock_net(skb->sk), info->userhdr, info->attrs);
|
|
|
|
err = PTR_ERR(dp);
|
|
|
|
if (IS_ERR(dp))
|
2013-04-15 22:23:03 +02:00
|
|
|
goto unlock;
|
2012-02-23 04:58:59 +01:00
|
|
|
|
2012-09-07 22:12:54 +02:00
|
|
|
reply = ovs_dp_cmd_build_info(dp, info->snd_portid,
|
2012-02-23 04:58:59 +01:00
|
|
|
info->snd_seq, OVS_DP_CMD_DEL);
|
|
|
|
err = PTR_ERR(reply);
|
|
|
|
if (IS_ERR(reply))
|
2013-04-15 22:23:03 +02:00
|
|
|
goto unlock;
|
2012-02-23 04:58:59 +01:00
|
|
|
|
|
|
|
__dp_destroy(dp);
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_unlock();
|
2011-10-26 04:26:31 +02:00
|
|
|
|
2013-03-29 14:46:50 +01:00
|
|
|
ovs_notify(reply, info, &ovs_dp_datapath_multicast_group);
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
return 0;
|
2013-04-15 22:23:03 +02:00
|
|
|
unlock:
|
|
|
|
ovs_unlock();
|
|
|
|
return err;
|
2011-10-26 04:26:31 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
static int ovs_dp_cmd_set(struct sk_buff *skb, struct genl_info *info)
|
|
|
|
{
|
|
|
|
struct sk_buff *reply;
|
|
|
|
struct datapath *dp;
|
|
|
|
int err;
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_lock();
|
2012-02-23 04:58:59 +01:00
|
|
|
dp = lookup_datapath(sock_net(skb->sk), info->userhdr, info->attrs);
|
2013-04-15 22:23:03 +02:00
|
|
|
err = PTR_ERR(dp);
|
2011-10-26 04:26:31 +02:00
|
|
|
if (IS_ERR(dp))
|
2013-04-15 22:23:03 +02:00
|
|
|
goto unlock;
|
2011-10-26 04:26:31 +02:00
|
|
|
|
2012-09-07 22:12:54 +02:00
|
|
|
reply = ovs_dp_cmd_build_info(dp, info->snd_portid,
|
2011-10-26 04:26:31 +02:00
|
|
|
info->snd_seq, OVS_DP_CMD_NEW);
|
|
|
|
if (IS_ERR(reply)) {
|
|
|
|
err = PTR_ERR(reply);
|
2012-02-23 04:58:59 +01:00
|
|
|
netlink_set_err(sock_net(skb->sk)->genl_sock, 0,
|
2011-10-26 04:26:31 +02:00
|
|
|
ovs_dp_datapath_multicast_group.id, err);
|
2013-04-15 22:23:03 +02:00
|
|
|
err = 0;
|
|
|
|
goto unlock;
|
2011-10-26 04:26:31 +02:00
|
|
|
}
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_unlock();
|
2013-03-29 14:46:50 +01:00
|
|
|
ovs_notify(reply, info, &ovs_dp_datapath_multicast_group);
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
return 0;
|
2013-04-15 22:23:03 +02:00
|
|
|
unlock:
|
|
|
|
ovs_unlock();
|
|
|
|
return err;
|
2011-10-26 04:26:31 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
static int ovs_dp_cmd_get(struct sk_buff *skb, struct genl_info *info)
|
|
|
|
{
|
|
|
|
struct sk_buff *reply;
|
|
|
|
struct datapath *dp;
|
2013-04-15 22:23:03 +02:00
|
|
|
int err;
|
2011-10-26 04:26:31 +02:00
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_lock();
|
2012-02-23 04:58:59 +01:00
|
|
|
dp = lookup_datapath(sock_net(skb->sk), info->userhdr, info->attrs);
|
2013-04-15 22:23:03 +02:00
|
|
|
if (IS_ERR(dp)) {
|
|
|
|
err = PTR_ERR(dp);
|
|
|
|
goto unlock;
|
|
|
|
}
|
2011-10-26 04:26:31 +02:00
|
|
|
|
2012-09-07 22:12:54 +02:00
|
|
|
reply = ovs_dp_cmd_build_info(dp, info->snd_portid,
|
2011-10-26 04:26:31 +02:00
|
|
|
info->snd_seq, OVS_DP_CMD_NEW);
|
2013-04-15 22:23:03 +02:00
|
|
|
if (IS_ERR(reply)) {
|
|
|
|
err = PTR_ERR(reply);
|
|
|
|
goto unlock;
|
|
|
|
}
|
2011-10-26 04:26:31 +02:00
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_unlock();
|
2011-10-26 04:26:31 +02:00
|
|
|
return genlmsg_reply(reply, info);
|
2013-04-15 22:23:03 +02:00
|
|
|
|
|
|
|
unlock:
|
|
|
|
ovs_unlock();
|
|
|
|
return err;
|
2011-10-26 04:26:31 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
static int ovs_dp_cmd_dump(struct sk_buff *skb, struct netlink_callback *cb)
|
|
|
|
{
|
2012-02-23 04:58:59 +01:00
|
|
|
struct ovs_net *ovs_net = net_generic(sock_net(skb->sk), ovs_net_id);
|
2011-10-26 04:26:31 +02:00
|
|
|
struct datapath *dp;
|
|
|
|
int skip = cb->args[0];
|
|
|
|
int i = 0;
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_lock();
|
2012-02-23 04:58:59 +01:00
|
|
|
list_for_each_entry(dp, &ovs_net->dps, list_node) {
|
2012-01-17 14:33:39 +01:00
|
|
|
if (i >= skip &&
|
2012-09-07 22:12:54 +02:00
|
|
|
ovs_dp_cmd_fill_info(dp, skb, NETLINK_CB(cb->skb).portid,
|
2011-10-26 04:26:31 +02:00
|
|
|
cb->nlh->nlmsg_seq, NLM_F_MULTI,
|
|
|
|
OVS_DP_CMD_NEW) < 0)
|
|
|
|
break;
|
|
|
|
i++;
|
|
|
|
}
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_unlock();
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
cb->args[0] = i;
|
|
|
|
|
|
|
|
return skb->len;
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct genl_ops dp_datapath_genl_ops[] = {
|
|
|
|
{ .cmd = OVS_DP_CMD_NEW,
|
|
|
|
.flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
|
|
|
|
.policy = datapath_policy,
|
|
|
|
.doit = ovs_dp_cmd_new
|
|
|
|
},
|
|
|
|
{ .cmd = OVS_DP_CMD_DEL,
|
|
|
|
.flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
|
|
|
|
.policy = datapath_policy,
|
|
|
|
.doit = ovs_dp_cmd_del
|
|
|
|
},
|
|
|
|
{ .cmd = OVS_DP_CMD_GET,
|
|
|
|
.flags = 0, /* OK for unprivileged users. */
|
|
|
|
.policy = datapath_policy,
|
|
|
|
.doit = ovs_dp_cmd_get,
|
|
|
|
.dumpit = ovs_dp_cmd_dump
|
|
|
|
},
|
|
|
|
{ .cmd = OVS_DP_CMD_SET,
|
|
|
|
.flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
|
|
|
|
.policy = datapath_policy,
|
|
|
|
.doit = ovs_dp_cmd_set,
|
|
|
|
},
|
|
|
|
};
|
|
|
|
|
|
|
|
static const struct nla_policy vport_policy[OVS_VPORT_ATTR_MAX + 1] = {
|
|
|
|
[OVS_VPORT_ATTR_NAME] = { .type = NLA_NUL_STRING, .len = IFNAMSIZ - 1 },
|
|
|
|
[OVS_VPORT_ATTR_STATS] = { .len = sizeof(struct ovs_vport_stats) },
|
|
|
|
[OVS_VPORT_ATTR_PORT_NO] = { .type = NLA_U32 },
|
|
|
|
[OVS_VPORT_ATTR_TYPE] = { .type = NLA_U32 },
|
|
|
|
[OVS_VPORT_ATTR_UPCALL_PID] = { .type = NLA_U32 },
|
|
|
|
[OVS_VPORT_ATTR_OPTIONS] = { .type = NLA_NESTED },
|
|
|
|
};
|
|
|
|
|
|
|
|
static struct genl_family dp_vport_genl_family = {
|
|
|
|
.id = GENL_ID_GENERATE,
|
|
|
|
.hdrsize = sizeof(struct ovs_header),
|
|
|
|
.name = OVS_VPORT_FAMILY,
|
|
|
|
.version = OVS_VPORT_VERSION,
|
2012-02-23 04:58:59 +01:00
|
|
|
.maxattr = OVS_VPORT_ATTR_MAX,
|
2013-04-23 09:48:48 +02:00
|
|
|
.netnsok = true,
|
|
|
|
.parallel_ops = true,
|
2011-10-26 04:26:31 +02:00
|
|
|
};
|
|
|
|
|
|
|
|
struct genl_multicast_group ovs_dp_vport_multicast_group = {
|
|
|
|
.name = OVS_VPORT_MCGROUP
|
|
|
|
};
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
/* Called with ovs_mutex or RCU read lock. */
|
2011-10-26 04:26:31 +02:00
|
|
|
static int ovs_vport_cmd_fill_info(struct vport *vport, struct sk_buff *skb,
|
2012-09-07 22:12:54 +02:00
|
|
|
u32 portid, u32 seq, u32 flags, u8 cmd)
|
2011-10-26 04:26:31 +02:00
|
|
|
{
|
|
|
|
struct ovs_header *ovs_header;
|
|
|
|
struct ovs_vport_stats vport_stats;
|
|
|
|
int err;
|
|
|
|
|
2012-09-07 22:12:54 +02:00
|
|
|
ovs_header = genlmsg_put(skb, portid, seq, &dp_vport_genl_family,
|
2011-10-26 04:26:31 +02:00
|
|
|
flags, cmd);
|
|
|
|
if (!ovs_header)
|
|
|
|
return -EMSGSIZE;
|
|
|
|
|
|
|
|
ovs_header->dp_ifindex = get_dpifindex(vport->dp);
|
|
|
|
|
2012-03-30 05:20:48 +02:00
|
|
|
if (nla_put_u32(skb, OVS_VPORT_ATTR_PORT_NO, vport->port_no) ||
|
|
|
|
nla_put_u32(skb, OVS_VPORT_ATTR_TYPE, vport->ops->type) ||
|
|
|
|
nla_put_string(skb, OVS_VPORT_ATTR_NAME, vport->ops->get_name(vport)) ||
|
2012-09-07 22:12:54 +02:00
|
|
|
nla_put_u32(skb, OVS_VPORT_ATTR_UPCALL_PID, vport->upcall_portid))
|
2012-03-30 05:20:48 +02:00
|
|
|
goto nla_put_failure;
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
ovs_vport_get_stats(vport, &vport_stats);
|
2012-03-30 05:20:48 +02:00
|
|
|
if (nla_put(skb, OVS_VPORT_ATTR_STATS, sizeof(struct ovs_vport_stats),
|
|
|
|
&vport_stats))
|
|
|
|
goto nla_put_failure;
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
err = ovs_vport_get_options(vport, skb);
|
|
|
|
if (err == -EMSGSIZE)
|
|
|
|
goto error;
|
|
|
|
|
|
|
|
return genlmsg_end(skb, ovs_header);
|
|
|
|
|
|
|
|
nla_put_failure:
|
|
|
|
err = -EMSGSIZE;
|
|
|
|
error:
|
|
|
|
genlmsg_cancel(skb, ovs_header);
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
/* Called with ovs_mutex or RCU read lock. */
|
2012-09-07 22:12:54 +02:00
|
|
|
struct sk_buff *ovs_vport_cmd_build_info(struct vport *vport, u32 portid,
|
2011-10-26 04:26:31 +02:00
|
|
|
u32 seq, u8 cmd)
|
|
|
|
{
|
|
|
|
struct sk_buff *skb;
|
|
|
|
int retval;
|
|
|
|
|
|
|
|
skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_ATOMIC);
|
|
|
|
if (!skb)
|
|
|
|
return ERR_PTR(-ENOMEM);
|
|
|
|
|
2012-09-07 22:12:54 +02:00
|
|
|
retval = ovs_vport_cmd_fill_info(vport, skb, portid, seq, 0, cmd);
|
2013-03-26 23:48:38 +01:00
|
|
|
BUG_ON(retval < 0);
|
|
|
|
|
2011-10-26 04:26:31 +02:00
|
|
|
return skb;
|
|
|
|
}
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
/* Called with ovs_mutex or RCU read lock. */
|
2012-02-23 04:58:59 +01:00
|
|
|
static struct vport *lookup_vport(struct net *net,
|
|
|
|
struct ovs_header *ovs_header,
|
2011-10-26 04:26:31 +02:00
|
|
|
struct nlattr *a[OVS_VPORT_ATTR_MAX + 1])
|
|
|
|
{
|
|
|
|
struct datapath *dp;
|
|
|
|
struct vport *vport;
|
|
|
|
|
|
|
|
if (a[OVS_VPORT_ATTR_NAME]) {
|
2012-02-23 04:58:59 +01:00
|
|
|
vport = ovs_vport_locate(net, nla_data(a[OVS_VPORT_ATTR_NAME]));
|
2011-10-26 04:26:31 +02:00
|
|
|
if (!vport)
|
|
|
|
return ERR_PTR(-ENODEV);
|
2012-03-07 00:04:04 +01:00
|
|
|
if (ovs_header->dp_ifindex &&
|
|
|
|
ovs_header->dp_ifindex != get_dpifindex(vport->dp))
|
|
|
|
return ERR_PTR(-ENODEV);
|
2011-10-26 04:26:31 +02:00
|
|
|
return vport;
|
|
|
|
} else if (a[OVS_VPORT_ATTR_PORT_NO]) {
|
|
|
|
u32 port_no = nla_get_u32(a[OVS_VPORT_ATTR_PORT_NO]);
|
|
|
|
|
|
|
|
if (port_no >= DP_MAX_PORTS)
|
|
|
|
return ERR_PTR(-EFBIG);
|
|
|
|
|
2012-02-23 04:58:59 +01:00
|
|
|
dp = get_dp(net, ovs_header->dp_ifindex);
|
2011-10-26 04:26:31 +02:00
|
|
|
if (!dp)
|
|
|
|
return ERR_PTR(-ENODEV);
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
vport = ovs_vport_ovsl_rcu(dp, port_no);
|
2011-10-26 04:26:31 +02:00
|
|
|
if (!vport)
|
2013-01-09 23:27:35 +01:00
|
|
|
return ERR_PTR(-ENODEV);
|
2011-10-26 04:26:31 +02:00
|
|
|
return vport;
|
|
|
|
} else
|
|
|
|
return ERR_PTR(-EINVAL);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int ovs_vport_cmd_new(struct sk_buff *skb, struct genl_info *info)
|
|
|
|
{
|
|
|
|
struct nlattr **a = info->attrs;
|
|
|
|
struct ovs_header *ovs_header = info->userhdr;
|
|
|
|
struct vport_parms parms;
|
|
|
|
struct sk_buff *reply;
|
|
|
|
struct vport *vport;
|
|
|
|
struct datapath *dp;
|
|
|
|
u32 port_no;
|
|
|
|
int err;
|
|
|
|
|
|
|
|
err = -EINVAL;
|
|
|
|
if (!a[OVS_VPORT_ATTR_NAME] || !a[OVS_VPORT_ATTR_TYPE] ||
|
|
|
|
!a[OVS_VPORT_ATTR_UPCALL_PID])
|
|
|
|
goto exit;
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_lock();
|
2012-02-23 04:58:59 +01:00
|
|
|
dp = get_dp(sock_net(skb->sk), ovs_header->dp_ifindex);
|
2011-10-26 04:26:31 +02:00
|
|
|
err = -ENODEV;
|
|
|
|
if (!dp)
|
|
|
|
goto exit_unlock;
|
|
|
|
|
|
|
|
if (a[OVS_VPORT_ATTR_PORT_NO]) {
|
|
|
|
port_no = nla_get_u32(a[OVS_VPORT_ATTR_PORT_NO]);
|
|
|
|
|
|
|
|
err = -EFBIG;
|
|
|
|
if (port_no >= DP_MAX_PORTS)
|
|
|
|
goto exit_unlock;
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
vport = ovs_vport_ovsl(dp, port_no);
|
2011-10-26 04:26:31 +02:00
|
|
|
err = -EBUSY;
|
|
|
|
if (vport)
|
|
|
|
goto exit_unlock;
|
|
|
|
} else {
|
|
|
|
for (port_no = 1; ; port_no++) {
|
|
|
|
if (port_no >= DP_MAX_PORTS) {
|
|
|
|
err = -EFBIG;
|
|
|
|
goto exit_unlock;
|
|
|
|
}
|
2013-04-15 22:23:03 +02:00
|
|
|
vport = ovs_vport_ovsl(dp, port_no);
|
2011-10-26 04:26:31 +02:00
|
|
|
if (!vport)
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
parms.name = nla_data(a[OVS_VPORT_ATTR_NAME]);
|
|
|
|
parms.type = nla_get_u32(a[OVS_VPORT_ATTR_TYPE]);
|
|
|
|
parms.options = a[OVS_VPORT_ATTR_OPTIONS];
|
|
|
|
parms.dp = dp;
|
|
|
|
parms.port_no = port_no;
|
2012-09-07 22:12:54 +02:00
|
|
|
parms.upcall_portid = nla_get_u32(a[OVS_VPORT_ATTR_UPCALL_PID]);
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
vport = new_vport(&parms);
|
|
|
|
err = PTR_ERR(vport);
|
|
|
|
if (IS_ERR(vport))
|
|
|
|
goto exit_unlock;
|
|
|
|
|
2013-02-08 22:18:01 +01:00
|
|
|
err = 0;
|
2012-09-07 22:12:54 +02:00
|
|
|
reply = ovs_vport_cmd_build_info(vport, info->snd_portid, info->snd_seq,
|
2011-10-26 04:26:31 +02:00
|
|
|
OVS_VPORT_CMD_NEW);
|
|
|
|
if (IS_ERR(reply)) {
|
|
|
|
err = PTR_ERR(reply);
|
|
|
|
ovs_dp_detach_port(vport);
|
|
|
|
goto exit_unlock;
|
|
|
|
}
|
2013-03-29 14:46:50 +01:00
|
|
|
|
|
|
|
ovs_notify(reply, info, &ovs_dp_vport_multicast_group);
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
exit_unlock:
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_unlock();
|
2011-10-26 04:26:31 +02:00
|
|
|
exit:
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int ovs_vport_cmd_set(struct sk_buff *skb, struct genl_info *info)
|
|
|
|
{
|
|
|
|
struct nlattr **a = info->attrs;
|
|
|
|
struct sk_buff *reply;
|
|
|
|
struct vport *vport;
|
|
|
|
int err;
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_lock();
|
2012-02-23 04:58:59 +01:00
|
|
|
vport = lookup_vport(sock_net(skb->sk), info->userhdr, a);
|
2011-10-26 04:26:31 +02:00
|
|
|
err = PTR_ERR(vport);
|
|
|
|
if (IS_ERR(vport))
|
|
|
|
goto exit_unlock;
|
|
|
|
|
|
|
|
if (a[OVS_VPORT_ATTR_TYPE] &&
|
2013-05-13 17:15:26 +02:00
|
|
|
nla_get_u32(a[OVS_VPORT_ATTR_TYPE]) != vport->ops->type) {
|
2011-10-26 04:26:31 +02:00
|
|
|
err = -EINVAL;
|
2013-05-13 17:15:26 +02:00
|
|
|
goto exit_unlock;
|
|
|
|
}
|
2011-10-26 04:26:31 +02:00
|
|
|
|
2013-03-26 23:48:38 +01:00
|
|
|
reply = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
|
|
|
|
if (!reply) {
|
|
|
|
err = -ENOMEM;
|
|
|
|
goto exit_unlock;
|
|
|
|
}
|
|
|
|
|
2013-05-13 17:15:26 +02:00
|
|
|
if (a[OVS_VPORT_ATTR_OPTIONS]) {
|
2011-10-26 04:26:31 +02:00
|
|
|
err = ovs_vport_set_options(vport, a[OVS_VPORT_ATTR_OPTIONS]);
|
2013-05-13 17:15:26 +02:00
|
|
|
if (err)
|
|
|
|
goto exit_free;
|
|
|
|
}
|
2013-03-26 23:48:38 +01:00
|
|
|
|
2012-04-09 21:12:12 +02:00
|
|
|
if (a[OVS_VPORT_ATTR_UPCALL_PID])
|
2012-09-07 22:12:54 +02:00
|
|
|
vport->upcall_portid = nla_get_u32(a[OVS_VPORT_ATTR_UPCALL_PID]);
|
2011-10-26 04:26:31 +02:00
|
|
|
|
2013-03-26 23:48:38 +01:00
|
|
|
err = ovs_vport_cmd_fill_info(vport, reply, info->snd_portid,
|
|
|
|
info->snd_seq, 0, OVS_VPORT_CMD_NEW);
|
|
|
|
BUG_ON(err < 0);
|
2011-10-26 04:26:31 +02:00
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_unlock();
|
2013-03-29 14:46:50 +01:00
|
|
|
ovs_notify(reply, info, &ovs_dp_vport_multicast_group);
|
2013-04-15 22:23:03 +02:00
|
|
|
return 0;
|
2011-10-26 04:26:31 +02:00
|
|
|
|
2013-03-26 23:48:38 +01:00
|
|
|
exit_free:
|
|
|
|
kfree_skb(reply);
|
2011-10-26 04:26:31 +02:00
|
|
|
exit_unlock:
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_unlock();
|
2011-10-26 04:26:31 +02:00
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int ovs_vport_cmd_del(struct sk_buff *skb, struct genl_info *info)
|
|
|
|
{
|
|
|
|
struct nlattr **a = info->attrs;
|
|
|
|
struct sk_buff *reply;
|
|
|
|
struct vport *vport;
|
|
|
|
int err;
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_lock();
|
2012-02-23 04:58:59 +01:00
|
|
|
vport = lookup_vport(sock_net(skb->sk), info->userhdr, a);
|
2011-10-26 04:26:31 +02:00
|
|
|
err = PTR_ERR(vport);
|
|
|
|
if (IS_ERR(vport))
|
|
|
|
goto exit_unlock;
|
|
|
|
|
|
|
|
if (vport->port_no == OVSP_LOCAL) {
|
|
|
|
err = -EINVAL;
|
|
|
|
goto exit_unlock;
|
|
|
|
}
|
|
|
|
|
2013-06-18 02:50:12 +02:00
|
|
|
reply = ovs_vport_cmd_build_info(vport, info->snd_portid,
|
|
|
|
info->snd_seq, OVS_VPORT_CMD_DEL);
|
2011-10-26 04:26:31 +02:00
|
|
|
err = PTR_ERR(reply);
|
|
|
|
if (IS_ERR(reply))
|
|
|
|
goto exit_unlock;
|
|
|
|
|
2013-02-08 18:30:23 +01:00
|
|
|
err = 0;
|
2011-10-26 04:26:31 +02:00
|
|
|
ovs_dp_detach_port(vport);
|
|
|
|
|
2013-03-29 14:46:50 +01:00
|
|
|
ovs_notify(reply, info, &ovs_dp_vport_multicast_group);
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
exit_unlock:
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_unlock();
|
2011-10-26 04:26:31 +02:00
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int ovs_vport_cmd_get(struct sk_buff *skb, struct genl_info *info)
|
|
|
|
{
|
|
|
|
struct nlattr **a = info->attrs;
|
|
|
|
struct ovs_header *ovs_header = info->userhdr;
|
|
|
|
struct sk_buff *reply;
|
|
|
|
struct vport *vport;
|
|
|
|
int err;
|
|
|
|
|
|
|
|
rcu_read_lock();
|
2012-02-23 04:58:59 +01:00
|
|
|
vport = lookup_vport(sock_net(skb->sk), ovs_header, a);
|
2011-10-26 04:26:31 +02:00
|
|
|
err = PTR_ERR(vport);
|
|
|
|
if (IS_ERR(vport))
|
|
|
|
goto exit_unlock;
|
|
|
|
|
2013-06-18 02:50:12 +02:00
|
|
|
reply = ovs_vport_cmd_build_info(vport, info->snd_portid,
|
|
|
|
info->snd_seq, OVS_VPORT_CMD_NEW);
|
2011-10-26 04:26:31 +02:00
|
|
|
err = PTR_ERR(reply);
|
|
|
|
if (IS_ERR(reply))
|
|
|
|
goto exit_unlock;
|
|
|
|
|
|
|
|
rcu_read_unlock();
|
|
|
|
|
|
|
|
return genlmsg_reply(reply, info);
|
|
|
|
|
|
|
|
exit_unlock:
|
|
|
|
rcu_read_unlock();
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int ovs_vport_cmd_dump(struct sk_buff *skb, struct netlink_callback *cb)
|
|
|
|
{
|
|
|
|
struct ovs_header *ovs_header = genlmsg_data(nlmsg_data(cb->nlh));
|
|
|
|
struct datapath *dp;
|
2012-08-23 21:40:54 +02:00
|
|
|
int bucket = cb->args[0], skip = cb->args[1];
|
|
|
|
int i, j = 0;
|
2011-10-26 04:26:31 +02:00
|
|
|
|
2012-02-23 04:58:59 +01:00
|
|
|
dp = get_dp(sock_net(skb->sk), ovs_header->dp_ifindex);
|
2011-10-26 04:26:31 +02:00
|
|
|
if (!dp)
|
|
|
|
return -ENODEV;
|
|
|
|
|
|
|
|
rcu_read_lock();
|
2012-08-23 21:40:54 +02:00
|
|
|
for (i = bucket; i < DP_VPORT_HASH_BUCKETS; i++) {
|
2011-10-26 04:26:31 +02:00
|
|
|
struct vport *vport;
|
2012-08-23 21:40:54 +02:00
|
|
|
|
|
|
|
j = 0;
|
hlist: drop the node parameter from iterators
I'm not sure why, but the hlist for each entry iterators were conceived
list_for_each_entry(pos, head, member)
The hlist ones were greedy and wanted an extra parameter:
hlist_for_each_entry(tpos, pos, head, member)
Why did they need an extra pos parameter? I'm not quite sure. Not only
they don't really need it, it also prevents the iterator from looking
exactly like the list iterator, which is unfortunate.
Besides the semantic patch, there was some manual work required:
- Fix up the actual hlist iterators in linux/list.h
- Fix up the declaration of other iterators based on the hlist ones.
- A very small amount of places were using the 'node' parameter, this
was modified to use 'obj->member' instead.
- Coccinelle didn't handle the hlist_for_each_entry_safe iterator
properly, so those had to be fixed up manually.
The semantic patch which is mostly the work of Peter Senna Tschudin is here:
@@
iterator name hlist_for_each_entry, hlist_for_each_entry_continue, hlist_for_each_entry_from, hlist_for_each_entry_rcu, hlist_for_each_entry_rcu_bh, hlist_for_each_entry_continue_rcu_bh, for_each_busy_worker, ax25_uid_for_each, ax25_for_each, inet_bind_bucket_for_each, sctp_for_each_hentry, sk_for_each, sk_for_each_rcu, sk_for_each_from, sk_for_each_safe, sk_for_each_bound, hlist_for_each_entry_safe, hlist_for_each_entry_continue_rcu, nr_neigh_for_each, nr_neigh_for_each_safe, nr_node_for_each, nr_node_for_each_safe, for_each_gfn_indirect_valid_sp, for_each_gfn_sp, for_each_host;
type T;
expression a,c,d,e;
identifier b;
statement S;
@@
-T b;
<+... when != b
(
hlist_for_each_entry(a,
- b,
c, d) S
|
hlist_for_each_entry_continue(a,
- b,
c) S
|
hlist_for_each_entry_from(a,
- b,
c) S
|
hlist_for_each_entry_rcu(a,
- b,
c, d) S
|
hlist_for_each_entry_rcu_bh(a,
- b,
c, d) S
|
hlist_for_each_entry_continue_rcu_bh(a,
- b,
c) S
|
for_each_busy_worker(a, c,
- b,
d) S
|
ax25_uid_for_each(a,
- b,
c) S
|
ax25_for_each(a,
- b,
c) S
|
inet_bind_bucket_for_each(a,
- b,
c) S
|
sctp_for_each_hentry(a,
- b,
c) S
|
sk_for_each(a,
- b,
c) S
|
sk_for_each_rcu(a,
- b,
c) S
|
sk_for_each_from
-(a, b)
+(a)
S
+ sk_for_each_from(a) S
|
sk_for_each_safe(a,
- b,
c, d) S
|
sk_for_each_bound(a,
- b,
c) S
|
hlist_for_each_entry_safe(a,
- b,
c, d, e) S
|
hlist_for_each_entry_continue_rcu(a,
- b,
c) S
|
nr_neigh_for_each(a,
- b,
c) S
|
nr_neigh_for_each_safe(a,
- b,
c, d) S
|
nr_node_for_each(a,
- b,
c) S
|
nr_node_for_each_safe(a,
- b,
c, d) S
|
- for_each_gfn_sp(a, c, d, b) S
+ for_each_gfn_sp(a, c, d) S
|
- for_each_gfn_indirect_valid_sp(a, c, d, b) S
+ for_each_gfn_indirect_valid_sp(a, c, d) S
|
for_each_host(a,
- b,
c) S
|
for_each_host_safe(a,
- b,
c, d) S
|
for_each_mesh_entry(a,
- b,
c, d) S
)
...+>
[akpm@linux-foundation.org: drop bogus change from net/ipv4/raw.c]
[akpm@linux-foundation.org: drop bogus hunk from net/ipv6/raw.c]
[akpm@linux-foundation.org: checkpatch fixes]
[akpm@linux-foundation.org: fix warnings]
[akpm@linux-foudnation.org: redo intrusive kvm changes]
Tested-by: Peter Senna Tschudin <peter.senna@gmail.com>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-28 02:06:00 +01:00
|
|
|
hlist_for_each_entry_rcu(vport, &dp->ports[i], dp_hash_node) {
|
2012-08-23 21:40:54 +02:00
|
|
|
if (j >= skip &&
|
|
|
|
ovs_vport_cmd_fill_info(vport, skb,
|
2012-09-07 22:12:54 +02:00
|
|
|
NETLINK_CB(cb->skb).portid,
|
2012-08-23 21:40:54 +02:00
|
|
|
cb->nlh->nlmsg_seq,
|
|
|
|
NLM_F_MULTI,
|
|
|
|
OVS_VPORT_CMD_NEW) < 0)
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
j++;
|
|
|
|
}
|
|
|
|
skip = 0;
|
2011-10-26 04:26:31 +02:00
|
|
|
}
|
2012-08-23 21:40:54 +02:00
|
|
|
out:
|
2011-10-26 04:26:31 +02:00
|
|
|
rcu_read_unlock();
|
|
|
|
|
2012-08-23 21:40:54 +02:00
|
|
|
cb->args[0] = i;
|
|
|
|
cb->args[1] = j;
|
2011-10-26 04:26:31 +02:00
|
|
|
|
2012-08-23 21:40:54 +02:00
|
|
|
return skb->len;
|
2011-10-26 04:26:31 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
static struct genl_ops dp_vport_genl_ops[] = {
|
|
|
|
{ .cmd = OVS_VPORT_CMD_NEW,
|
|
|
|
.flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
|
|
|
|
.policy = vport_policy,
|
|
|
|
.doit = ovs_vport_cmd_new
|
|
|
|
},
|
|
|
|
{ .cmd = OVS_VPORT_CMD_DEL,
|
|
|
|
.flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
|
|
|
|
.policy = vport_policy,
|
|
|
|
.doit = ovs_vport_cmd_del
|
|
|
|
},
|
|
|
|
{ .cmd = OVS_VPORT_CMD_GET,
|
|
|
|
.flags = 0, /* OK for unprivileged users. */
|
|
|
|
.policy = vport_policy,
|
|
|
|
.doit = ovs_vport_cmd_get,
|
|
|
|
.dumpit = ovs_vport_cmd_dump
|
|
|
|
},
|
|
|
|
{ .cmd = OVS_VPORT_CMD_SET,
|
|
|
|
.flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
|
|
|
|
.policy = vport_policy,
|
|
|
|
.doit = ovs_vport_cmd_set,
|
|
|
|
},
|
|
|
|
};
|
|
|
|
|
|
|
|
struct genl_family_and_ops {
|
|
|
|
struct genl_family *family;
|
|
|
|
struct genl_ops *ops;
|
|
|
|
int n_ops;
|
|
|
|
struct genl_multicast_group *group;
|
|
|
|
};
|
|
|
|
|
|
|
|
static const struct genl_family_and_ops dp_genl_families[] = {
|
|
|
|
{ &dp_datapath_genl_family,
|
|
|
|
dp_datapath_genl_ops, ARRAY_SIZE(dp_datapath_genl_ops),
|
|
|
|
&ovs_dp_datapath_multicast_group },
|
|
|
|
{ &dp_vport_genl_family,
|
|
|
|
dp_vport_genl_ops, ARRAY_SIZE(dp_vport_genl_ops),
|
|
|
|
&ovs_dp_vport_multicast_group },
|
|
|
|
{ &dp_flow_genl_family,
|
|
|
|
dp_flow_genl_ops, ARRAY_SIZE(dp_flow_genl_ops),
|
|
|
|
&ovs_dp_flow_multicast_group },
|
|
|
|
{ &dp_packet_genl_family,
|
|
|
|
dp_packet_genl_ops, ARRAY_SIZE(dp_packet_genl_ops),
|
|
|
|
NULL },
|
|
|
|
};
|
|
|
|
|
|
|
|
static void dp_unregister_genl(int n_families)
|
|
|
|
{
|
|
|
|
int i;
|
|
|
|
|
|
|
|
for (i = 0; i < n_families; i++)
|
|
|
|
genl_unregister_family(dp_genl_families[i].family);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int dp_register_genl(void)
|
|
|
|
{
|
|
|
|
int n_registered;
|
|
|
|
int err;
|
|
|
|
int i;
|
|
|
|
|
|
|
|
n_registered = 0;
|
|
|
|
for (i = 0; i < ARRAY_SIZE(dp_genl_families); i++) {
|
|
|
|
const struct genl_family_and_ops *f = &dp_genl_families[i];
|
|
|
|
|
|
|
|
err = genl_register_family_with_ops(f->family, f->ops,
|
|
|
|
f->n_ops);
|
|
|
|
if (err)
|
|
|
|
goto error;
|
|
|
|
n_registered++;
|
|
|
|
|
|
|
|
if (f->group) {
|
|
|
|
err = genl_register_mc_group(f->family, f->group);
|
|
|
|
if (err)
|
|
|
|
goto error;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
error:
|
|
|
|
dp_unregister_genl(n_registered);
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2012-02-23 04:58:59 +01:00
|
|
|
static void rehash_flow_table(struct work_struct *work)
|
|
|
|
{
|
|
|
|
struct datapath *dp;
|
|
|
|
struct net *net;
|
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_lock();
|
2012-02-23 04:58:59 +01:00
|
|
|
rtnl_lock();
|
|
|
|
for_each_net(net) {
|
|
|
|
struct ovs_net *ovs_net = net_generic(net, ovs_net_id);
|
|
|
|
|
|
|
|
list_for_each_entry(dp, &ovs_net->dps, list_node) {
|
2013-04-15 22:23:03 +02:00
|
|
|
struct flow_table *old_table = ovsl_dereference(dp->table);
|
2012-02-23 04:58:59 +01:00
|
|
|
struct flow_table *new_table;
|
|
|
|
|
|
|
|
new_table = ovs_flow_tbl_rehash(old_table);
|
|
|
|
if (!IS_ERR(new_table)) {
|
|
|
|
rcu_assign_pointer(dp->table, new_table);
|
|
|
|
ovs_flow_tbl_deferred_destroy(old_table);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
rtnl_unlock();
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_unlock();
|
2012-02-23 04:58:59 +01:00
|
|
|
schedule_delayed_work(&rehash_flow_wq, REHASH_FLOW_INTERVAL);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int __net_init ovs_init_net(struct net *net)
|
|
|
|
{
|
|
|
|
struct ovs_net *ovs_net = net_generic(net, ovs_net_id);
|
|
|
|
|
|
|
|
INIT_LIST_HEAD(&ovs_net->dps);
|
2013-04-15 22:23:03 +02:00
|
|
|
INIT_WORK(&ovs_net->dp_notify_work, ovs_dp_notify_wq);
|
2012-02-23 04:58:59 +01:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void __net_exit ovs_exit_net(struct net *net)
|
|
|
|
{
|
|
|
|
struct datapath *dp, *dp_next;
|
2013-04-15 22:23:03 +02:00
|
|
|
struct ovs_net *ovs_net = net_generic(net, ovs_net_id);
|
2012-02-23 04:58:59 +01:00
|
|
|
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_lock();
|
2012-02-23 04:58:59 +01:00
|
|
|
list_for_each_entry_safe(dp, dp_next, &ovs_net->dps, list_node)
|
|
|
|
__dp_destroy(dp);
|
2013-04-15 22:23:03 +02:00
|
|
|
ovs_unlock();
|
|
|
|
|
|
|
|
cancel_work_sync(&ovs_net->dp_notify_work);
|
2012-02-23 04:58:59 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
static struct pernet_operations ovs_net_ops = {
|
|
|
|
.init = ovs_init_net,
|
|
|
|
.exit = ovs_exit_net,
|
|
|
|
.id = &ovs_net_id,
|
|
|
|
.size = sizeof(struct ovs_net),
|
|
|
|
};
|
|
|
|
|
2011-10-26 04:26:31 +02:00
|
|
|
static int __init dp_init(void)
|
|
|
|
{
|
|
|
|
int err;
|
|
|
|
|
2013-01-09 08:19:55 +01:00
|
|
|
BUILD_BUG_ON(sizeof(struct ovs_skb_cb) > FIELD_SIZEOF(struct sk_buff, cb));
|
2011-10-26 04:26:31 +02:00
|
|
|
|
|
|
|
pr_info("Open vSwitch switching datapath\n");
|
|
|
|
|
|
|
|
err = ovs_flow_init();
|
|
|
|
if (err)
|
|
|
|
goto error;
|
|
|
|
|
|
|
|
err = ovs_vport_init();
|
|
|
|
if (err)
|
|
|
|
goto error_flow_exit;
|
|
|
|
|
2012-02-23 04:58:59 +01:00
|
|
|
err = register_pernet_device(&ovs_net_ops);
|
2011-10-26 04:26:31 +02:00
|
|
|
if (err)
|
|
|
|
goto error_vport_exit;
|
|
|
|
|
2012-02-23 04:58:59 +01:00
|
|
|
err = register_netdevice_notifier(&ovs_dp_device_notifier);
|
|
|
|
if (err)
|
|
|
|
goto error_netns_exit;
|
|
|
|
|
2011-10-26 04:26:31 +02:00
|
|
|
err = dp_register_genl();
|
|
|
|
if (err < 0)
|
|
|
|
goto error_unreg_notifier;
|
|
|
|
|
|
|
|
schedule_delayed_work(&rehash_flow_wq, REHASH_FLOW_INTERVAL);
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
error_unreg_notifier:
|
|
|
|
unregister_netdevice_notifier(&ovs_dp_device_notifier);
|
2012-02-23 04:58:59 +01:00
|
|
|
error_netns_exit:
|
|
|
|
unregister_pernet_device(&ovs_net_ops);
|
2011-10-26 04:26:31 +02:00
|
|
|
error_vport_exit:
|
|
|
|
ovs_vport_exit();
|
|
|
|
error_flow_exit:
|
|
|
|
ovs_flow_exit();
|
|
|
|
error:
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void dp_cleanup(void)
|
|
|
|
{
|
|
|
|
cancel_delayed_work_sync(&rehash_flow_wq);
|
|
|
|
dp_unregister_genl(ARRAY_SIZE(dp_genl_families));
|
|
|
|
unregister_netdevice_notifier(&ovs_dp_device_notifier);
|
2012-02-23 04:58:59 +01:00
|
|
|
unregister_pernet_device(&ovs_net_ops);
|
|
|
|
rcu_barrier();
|
2011-10-26 04:26:31 +02:00
|
|
|
ovs_vport_exit();
|
|
|
|
ovs_flow_exit();
|
|
|
|
}
|
|
|
|
|
|
|
|
module_init(dp_init);
|
|
|
|
module_exit(dp_cleanup);
|
|
|
|
|
|
|
|
MODULE_DESCRIPTION("Open vSwitch switching datapath");
|
|
|
|
MODULE_LICENSE("GPL");
|