/*
- * Copyright (c) 2010 Nicira Networks.
- * Distributed under the terms of the GNU GPL version 2.
+ * Copyright (c) 2007-2012 Nicira, Inc.
*
- * Significant portions of this file may be copied from parts of the Linux
- * kernel, by Linus Torvalds and others.
+ * 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
*/
-#include <linux/dcache.h>
#include <linux/etherdevice.h>
#include <linux/if.h>
+#include <linux/if_vlan.h>
+#include <linux/jhash.h>
+#include <linux/kconfig.h>
#include <linux/kernel.h>
#include <linux/list.h>
#include <linux/mutex.h>
#include <linux/percpu.h>
+#include <linux/rcupdate.h>
#include <linux/rtnetlink.h>
#include <linux/compat.h>
+#include <linux/version.h>
+#include <net/net_namespace.h>
+#include "datapath.h"
#include "vport.h"
-
-extern struct vport_ops netdev_vport_ops;
-extern struct vport_ops internal_vport_ops;
-extern struct vport_ops gre_vport_ops;
-
-static struct vport_ops *base_vport_ops_list[] = {
- &netdev_vport_ops,
- &internal_vport_ops,
- &gre_vport_ops,
+#include "vport-internal_dev.h"
+
+static void ovs_vport_record_error(struct vport *,
+ enum vport_err_type err_type);
+
+/* List of statically compiled vport implementations. Don't forget to also
+ * add yours to the list at the bottom of vport.h. */
+static const struct vport_ops *vport_ops_list[] = {
+ &ovs_netdev_vport_ops,
+ &ovs_internal_vport_ops,
+#if IS_ENABLED(CONFIG_NET_IPGRE_DEMUX)
+ &ovs_gre_vport_ops,
+ &ovs_gre64_vport_ops,
+#endif
+ &ovs_vxlan_vport_ops,
+ &ovs_lisp_vport_ops,
};
-static const struct vport_ops **vport_ops_list;
-static int n_vport_types;
-
+/* Protected by RCU read lock for reading, ovs_mutex for writing. */
static struct hlist_head *dev_table;
#define VPORT_HASH_BUCKETS 1024
-/* Both RTNL lock and vport_mutex need to be held when updating dev_table.
- *
- * If you use vport_locate and then perform some operations, you need to hold
- * one of these locks if you don't want the vport to be deleted out from under
- * you.
- *
- * If you get a reference to a vport through a dp_port, it is protected
- * by RCU and you need to hold rcu_read_lock instead when reading.
- *
- * If multiple locks are taken, the hierarchy is:
- * 1. RTNL
- * 2. DP
- * 3. vport
- */
-static DEFINE_MUTEX(vport_mutex);
-
-/**
- * vport_lock - acquire vport lock
- *
- * Acquire global vport lock. See above comment about locking requirements
- * and specific function definitions. May sleep.
- */
-void
-vport_lock(void)
-{
- mutex_lock(&vport_mutex);
-}
-
-/**
- * vport_unlock - release vport lock
- *
- * Release lock acquired with vport_lock.
- */
-void
-vport_unlock(void)
-{
- mutex_unlock(&vport_mutex);
-}
-
-#define ASSERT_VPORT() do { \
- if (unlikely(!mutex_is_locked(&vport_mutex))) { \
- printk(KERN_ERR "openvswitch: vport lock not held at %s (%d)\n", \
- __FILE__, __LINE__); \
- dump_stack(); \
- } \
-} while(0)
-
/**
- * vport_init - initialize vport subsystem
+ * ovs_vport_init - initialize vport subsystem
*
- * Called at module load time to initialize the vport subsystem and any
- * compiled in vport types.
+ * Called at module load time to initialize the vport subsystem.
*/
-int
-vport_init(void)
+int ovs_vport_init(void)
{
- int err;
- int i;
-
dev_table = kzalloc(VPORT_HASH_BUCKETS * sizeof(struct hlist_head),
GFP_KERNEL);
- if (!dev_table) {
- err = -ENOMEM;
- goto error;
- }
-
- vport_ops_list = kmalloc(ARRAY_SIZE(base_vport_ops_list) *
- sizeof(struct vport_ops *), GFP_KERNEL);
- if (!vport_ops_list) {
- err = -ENOMEM;
- goto error_dev_table;
- }
-
- for (i = 0; i < ARRAY_SIZE(base_vport_ops_list); i++) {
- struct vport_ops *new_ops = base_vport_ops_list[i];
-
- if (new_ops->get_stats && new_ops->flags & VPORT_F_GEN_STATS) {
- printk(KERN_INFO "openvswitch: both get_stats() and VPORT_F_GEN_STATS defined on vport %s, dropping VPORT_F_GEN_STATS\n", new_ops->type);
- new_ops->flags &= ~VPORT_F_GEN_STATS;
- }
-
- if (new_ops->init)
- err = new_ops->init();
- else
- err = 0;
-
- if (!err)
- vport_ops_list[n_vport_types++] = new_ops;
- else if (new_ops->flags & VPORT_F_REQUIRED) {
- vport_exit();
- goto error;
- }
- }
+ if (!dev_table)
+ return -ENOMEM;
return 0;
-
-error_dev_table:
- kfree(dev_table);
-error:
- return err;
-}
-
-static void
-vport_del_all(void)
-{
- int i;
-
- rtnl_lock();
- vport_lock();
-
- for (i = 0; i < VPORT_HASH_BUCKETS; i++) {
- struct hlist_head *bucket = &dev_table[i];
- struct vport *vport;
- struct hlist_node *node, *next;
-
- hlist_for_each_entry_safe(vport, node, next, bucket, hash_node)
- __vport_del(vport);
- }
-
- vport_unlock();
- rtnl_unlock();
}
/**
- * vport_exit - shutdown vport subsystem
+ * ovs_vport_exit - shutdown vport subsystem
*
- * Called at module exit time to shutdown the vport subsystem and any
- * initialized vport types.
+ * Called at module exit time to shutdown the vport subsystem.
*/
-void
-vport_exit(void)
+void ovs_vport_exit(void)
{
- int i;
-
- vport_del_all();
-
- for (i = 0; i < n_vport_types; i++) {
- if (vport_ops_list[i]->exit)
- vport_ops_list[i]->exit();
- }
-
- kfree(vport_ops_list);
kfree(dev_table);
}
-/**
- * vport_add - add vport device (for userspace callers)
- *
- * @uvport_config: New port configuration.
- *
- * Creates a new vport with the specified configuration (which is dependent
- * on device type). This function is for userspace callers and assumes no
- * locks are held.
- */
-static int
-do_vport_add(struct odp_vport_add *vport_config)
-{
- struct vport *vport;
- int err = 0;
-
- vport_config->port_type[VPORT_TYPE_SIZE - 1] = '\0';
- vport_config->devname[IFNAMSIZ - 1] = '\0';
-
- rtnl_lock();
-
- vport = vport_locate(vport_config->devname);
- if (vport) {
- err = -EEXIST;
- goto out;
- }
-
- vport_lock();
- vport = __vport_add(vport_config->devname, vport_config->port_type,
- vport_config->config);
- vport_unlock();
-
- if (IS_ERR(vport))
- err = PTR_ERR(vport);
-
-out:
- rtnl_unlock();
- return err;
-}
-
-int
-vport_add(const struct odp_vport_add __user *uvport_config)
-{
- struct odp_vport_add vport_config;
-
- if (copy_from_user(&vport_config, uvport_config, sizeof(struct odp_vport_add)))
- return -EFAULT;
-
- return do_vport_add(&vport_config);
-}
-
-#ifdef CONFIG_COMPAT
-int
-compat_vport_add(struct compat_odp_vport_add *ucompat)
-{
- struct compat_odp_vport_add compat;
- struct odp_vport_add vport_config;
-
- if (copy_from_user(&compat, ucompat, sizeof(struct compat_odp_vport_add)))
- return -EFAULT;
-
- memcpy(vport_config.port_type, compat.port_type, VPORT_TYPE_SIZE);
- memcpy(vport_config.devname, compat.devname, IFNAMSIZ);
- vport_config.config = compat_ptr(compat.config);
-
- return do_vport_add(&vport_config);
-}
-#endif
-
-/**
- * vport_mod - modify existing vport device (for userspace callers)
- *
- * @uvport_config: New configuration for vport
- *
- * Modifies an existing device with the specified configuration (which is
- * dependent on device type). This function is for userspace callers and
- * assumes no locks are held.
- */
-static int
-do_vport_mod(struct odp_vport_mod *vport_config)
-{
- struct vport *vport;
- int err;
-
- vport_config->devname[IFNAMSIZ - 1] = '\0';
-
- rtnl_lock();
-
- vport = vport_locate(vport_config->devname);
- if (!vport) {
- err = -ENODEV;
- goto out;
- }
-
- vport_lock();
- err = __vport_mod(vport, vport_config->config);
- vport_unlock();
-
-out:
- rtnl_unlock();
- return err;
-}
-
-int
-vport_mod(const struct odp_vport_mod __user *uvport_config)
-{
- struct odp_vport_mod vport_config;
-
- if (copy_from_user(&vport_config, uvport_config, sizeof(struct odp_vport_mod)))
- return -EFAULT;
-
- return do_vport_mod(&vport_config);
-}
-
-#ifdef CONFIG_COMPAT
-int
-compat_vport_mod(struct compat_odp_vport_mod *ucompat)
-{
- struct compat_odp_vport_mod compat;
- struct odp_vport_mod vport_config;
-
- if (copy_from_user(&compat, ucompat, sizeof(struct compat_odp_vport_mod)))
- return -EFAULT;
-
- memcpy(vport_config.devname, compat.devname, IFNAMSIZ);
- vport_config.config = compat_ptr(compat.config);
-
- return do_vport_mod(&vport_config);
-}
-#endif
-
-/**
- * vport_del - delete existing vport device (for userspace callers)
- *
- * @udevname: Name of device to delete
- *
- * Deletes the specified device. Detaches the device from a datapath first
- * if it is attached. Deleting the device will fail if it does not exist or it
- * is the datapath local port. It is also possible to fail for less obvious
- * reasons, such as lack of memory. This function is for userspace callers and
- * assumes no locks are held.
- */
-int
-vport_del(const char __user *udevname)
-{
- char devname[IFNAMSIZ];
- struct vport *vport;
- struct dp_port *dp_port;
- int err = 0;
- int retval;
-
- retval = strncpy_from_user(devname, udevname, IFNAMSIZ);
- if (retval < 0)
- return -EFAULT;
- else if (retval >= IFNAMSIZ)
- return -ENAMETOOLONG;
-
- rtnl_lock();
-
- vport = vport_locate(devname);
- if (!vport) {
- err = -ENODEV;
- goto out;
- }
-
- dp_port = vport_get_dp_port(vport);
- if (dp_port) {
- struct datapath *dp = dp_port->dp;
-
- mutex_lock(&dp->mutex);
-
- if (!strcmp(dp_name(dp), devname)) {
- err = -EINVAL;
- goto dp_port_out;
- }
-
- err = dp_detach_port(dp_port, 0);
-
-dp_port_out:
- mutex_unlock(&dp->mutex);
-
- if (err)
- goto out;
- }
-
- vport_lock();
- err = __vport_del(vport);
- vport_unlock();
-
-out:
- rtnl_unlock();
- return err;
-}
-
-/**
- * vport_stats_get - retrieve device stats (for userspace callers)
- *
- * @ustats_req: Stats request parameters.
- *
- * Retrieves transmit, receive, and error stats for the given device. This
- * function is for userspace callers and assumes no locks are held.
- */
-int
-vport_stats_get(struct odp_vport_stats_req __user *ustats_req)
+static struct hlist_head *hash_bucket(struct net *net, const char *name)
{
- struct odp_vport_stats_req stats_req;
- struct vport *vport;
- int err;
-
- if (copy_from_user(&stats_req, ustats_req, sizeof(struct odp_vport_stats_req)))
- return -EFAULT;
-
- stats_req.devname[IFNAMSIZ - 1] = '\0';
-
- vport_lock();
-
- vport = vport_locate(stats_req.devname);
- if (!vport) {
- err = -ENODEV;
- goto out;
- }
-
- if (vport->ops->get_stats) {
- rcu_read_lock();
- err = vport->ops->get_stats(vport, &stats_req.stats);
- rcu_read_unlock();
-
- } else if (vport->ops->flags & VPORT_F_GEN_STATS) {
- int i;
-
- memset(&stats_req.stats, 0, sizeof(struct odp_vport_stats));
-
- for_each_possible_cpu(i) {
- const struct vport_percpu_stats *percpu_stats;
-
- percpu_stats = per_cpu_ptr(vport->percpu_stats, i);
- stats_req.stats.rx_bytes += percpu_stats->rx_bytes;
- stats_req.stats.rx_packets += percpu_stats->rx_packets;
- stats_req.stats.tx_bytes += percpu_stats->tx_bytes;
- stats_req.stats.tx_packets += percpu_stats->tx_packets;
- }
-
- spin_lock_bh(&vport->err_stats.lock);
-
- stats_req.stats.rx_dropped = vport->err_stats.rx_dropped;
- stats_req.stats.rx_errors = vport->err_stats.rx_errors
- + vport->err_stats.rx_frame_err
- + vport->err_stats.rx_over_err
- + vport->err_stats.rx_crc_err;
- stats_req.stats.rx_frame_err = vport->err_stats.rx_frame_err;
- stats_req.stats.rx_over_err = vport->err_stats.rx_over_err;
- stats_req.stats.rx_crc_err = vport->err_stats.rx_crc_err;
- stats_req.stats.tx_dropped = vport->err_stats.tx_dropped;
- stats_req.stats.tx_errors = vport->err_stats.tx_errors;
- stats_req.stats.collisions = vport->err_stats.collisions;
-
- spin_unlock_bh(&vport->err_stats.lock);
-
- err = 0;
- } else
- err = -EOPNOTSUPP;
-
-out:
- vport_unlock();
-
- if (!err)
- if (copy_to_user(ustats_req, &stats_req, sizeof(struct odp_vport_stats_req)))
- err = -EFAULT;
-
- return err;
-}
-
-/**
- * vport_ether_get - retrieve device Ethernet address (for userspace callers)
- *
- * @uvport_ether: Ethernet address request parameters.
- *
- * Retrieves the Ethernet address of the given device. This function is for
- * userspace callers and assumes no locks are held.
- */
-int
-vport_ether_get(struct odp_vport_ether __user *uvport_ether)
-{
- struct odp_vport_ether vport_ether;
- struct vport *vport;
- int err = 0;
-
- if (copy_from_user(&vport_ether, uvport_ether, sizeof(struct odp_vport_ether)))
- return -EFAULT;
-
- vport_ether.devname[IFNAMSIZ - 1] = '\0';
-
- vport_lock();
-
- vport = vport_locate(vport_ether.devname);
- if (!vport) {
- err = -ENODEV;
- goto out;
- }
-
- rcu_read_lock();
- memcpy(vport_ether.ether_addr, vport_get_addr(vport), ETH_ALEN);
- rcu_read_unlock();
-
-out:
- vport_unlock();
-
- if (!err)
- if (copy_to_user(uvport_ether, &vport_ether, sizeof(struct odp_vport_ether)))
- err = -EFAULT;
-
- return err;
-}
-
-/**
- * vport_ether_set - set device Ethernet address (for userspace callers)
- *
- * @uvport_ether: Ethernet address request parameters.
- *
- * Sets the Ethernet address of the given device. Some devices may not support
- * setting the Ethernet address, in which case the result will always be
- * -EOPNOTSUPP. This function is for userspace callers and assumes no locks
- * are held.
- */
-int
-vport_ether_set(struct odp_vport_ether __user *uvport_ether)
-{
- struct odp_vport_ether vport_ether;
- struct vport *vport;
- int err;
-
- if (copy_from_user(&vport_ether, uvport_ether, sizeof(struct odp_vport_ether)))
- return -EFAULT;
-
- vport_ether.devname[IFNAMSIZ - 1] = '\0';
-
- rtnl_lock();
- vport_lock();
-
- vport = vport_locate(vport_ether.devname);
- if (!vport) {
- err = -ENODEV;
- goto out;
- }
-
- err = vport_set_addr(vport, vport_ether.ether_addr);
-
-out:
- vport_unlock();
- rtnl_unlock();
- return err;
-}
-
-/**
- * vport_mut_get - retrieve device MTU (for userspace callers)
- *
- * @uvport_mtu: MTU request parameters.
- *
- * Retrieves the MTU of the given device. This function is for userspace
- * callers and assumes no locks are held.
- */
-int
-vport_mtu_get(struct odp_vport_mtu __user *uvport_mtu)
-{
- struct odp_vport_mtu vport_mtu;
- struct vport *vport;
- int err = 0;
-
- if (copy_from_user(&vport_mtu, uvport_mtu, sizeof(struct odp_vport_mtu)))
- return -EFAULT;
-
- vport_mtu.devname[IFNAMSIZ - 1] = '\0';
-
- vport_lock();
-
- vport = vport_locate(vport_mtu.devname);
- if (!vport) {
- err = -ENODEV;
- goto out;
- }
-
- vport_mtu.mtu = vport_get_mtu(vport);
-
-out:
- vport_unlock();
-
- if (!err)
- if (copy_to_user(uvport_mtu, &vport_mtu, sizeof(struct odp_vport_mtu)))
- err = -EFAULT;
-
- return err;
-}
-
-/**
- * vport_mtu_set - set device MTU (for userspace callers)
- *
- * @uvport_mtu: MTU request parameters.
- *
- * Sets the MTU of the given device. Some devices may not support setting the
- * MTU, in which case the result will always be -EOPNOTSUPP. This function is
- * for userspace callers and assumes no locks are held.
- */
-int
-vport_mtu_set(struct odp_vport_mtu __user *uvport_mtu)
-{
- struct odp_vport_mtu vport_mtu;
- struct vport *vport;
- int err;
-
- if (copy_from_user(&vport_mtu, uvport_mtu, sizeof(struct odp_vport_mtu)))
- return -EFAULT;
-
- vport_mtu.devname[IFNAMSIZ - 1] = '\0';
-
- rtnl_lock();
- vport_lock();
-
- vport = vport_locate(vport_mtu.devname);
- if (!vport) {
- err = -ENODEV;
- goto out;
- }
-
- err = vport_set_mtu(vport, vport_mtu.mtu);
-
-out:
- vport_unlock();
- rtnl_unlock();
- return err;
-}
-
-static struct hlist_head *
-hash_bucket(const char *name)
-{
- unsigned int hash = full_name_hash(name, strlen(name));
+ unsigned int hash = jhash(name, strlen(name), (unsigned long) net);
return &dev_table[hash & (VPORT_HASH_BUCKETS - 1)];
}
/**
- * vport_locate - find a port that has already been created
+ * ovs_vport_locate - find a port that has already been created
*
* @name: name of port to find
*
- * Either RTNL or vport lock must be acquired before calling this function
- * and held while using the found port. See the locking comments at the
- * top of the file.
+ * Must be called with ovs or RCU read lock.
*/
-struct vport *
-vport_locate(const char *name)
+struct vport *ovs_vport_locate(struct net *net, const char *name)
{
- struct hlist_head *bucket = hash_bucket(name);
+ struct hlist_head *bucket = hash_bucket(net, name);
struct vport *vport;
- struct hlist_node *node;
-
- if (unlikely(!mutex_is_locked(&vport_mutex) && !rtnl_is_locked())) {
- printk(KERN_ERR "openvswitch: neither RTNL nor vport lock held in vport_locate\n");
- dump_stack();
- }
- rcu_read_lock();
-
- hlist_for_each_entry(vport, node, bucket, hash_node)
- if (!strcmp(name, vport_get_name(vport)))
- goto out;
-
- vport = NULL;
-
-out:
- rcu_read_unlock();
- return vport;
-}
-
-static void
-register_vport(struct vport *vport)
-{
- hlist_add_head(&vport->hash_node, hash_bucket(vport_get_name(vport)));
-}
+ hlist_for_each_entry_rcu(vport, bucket, hash_node)
+ if (!strcmp(name, vport->ops->get_name(vport)) &&
+ net_eq(ovs_dp_get_net(vport->dp), net))
+ return vport;
-static void
-unregister_vport(struct vport *vport)
-{
- hlist_del(&vport->hash_node);
+ return NULL;
}
/**
- * vport_alloc - allocate and initialize new vport
+ * ovs_vport_alloc - allocate and initialize new vport
*
* @priv_size: Size of private data area to allocate.
* @ops: vport device ops
* Allocate and initialize a new vport defined by @ops. The vport will contain
* a private data area of size @priv_size that can be accessed using
* vport_priv(). vports that are no longer needed should be released with
- * vport_free().
+ * ovs_vport_free().
*/
-struct vport *
-vport_alloc(int priv_size, const struct vport_ops *ops)
+struct vport *ovs_vport_alloc(int priv_size, const struct vport_ops *ops,
+ const struct vport_parms *parms)
{
struct vport *vport;
size_t alloc_size;
+ int i;
alloc_size = sizeof(struct vport);
if (priv_size) {
if (!vport)
return ERR_PTR(-ENOMEM);
+ vport->dp = parms->dp;
+ vport->port_no = parms->port_no;
vport->ops = ops;
+ INIT_HLIST_NODE(&vport->dp_hash_node);
- if (vport->ops->flags & VPORT_F_GEN_STATS) {
- vport->percpu_stats = alloc_percpu(struct vport_percpu_stats);
- if (!vport->percpu_stats)
- return ERR_PTR(-ENOMEM);
+ if (ovs_vport_set_upcall_portids(vport, parms->upcall_portids))
+ return ERR_PTR(-EINVAL);
- spin_lock_init(&vport->err_stats.lock);
+ vport->percpu_stats = alloc_percpu(struct pcpu_sw_netstats);
+ if (!vport->percpu_stats) {
+ kfree(vport);
+ return ERR_PTR(-ENOMEM);
+ }
+
+ for_each_possible_cpu(i) {
+ struct pcpu_sw_netstats *vport_stats;
+ vport_stats = per_cpu_ptr(vport->percpu_stats, i);
+ u64_stats_init(&vport_stats->syncp);
}
+ spin_lock_init(&vport->stats_lock);
+
return vport;
}
/**
- * vport_free - uninitialize and free vport
+ * ovs_vport_free - uninitialize and free vport
*
* @vport: vport to free
*
- * Frees a vport allocated with vport_alloc() when it is no longer needed.
+ * Frees a vport allocated with ovs_vport_alloc() when it is no longer needed.
+ *
+ * The caller must ensure that an RCU grace period has passed since the last
+ * time @vport was in a datapath.
*/
-void
-vport_free(struct vport *vport)
+void ovs_vport_free(struct vport *vport)
{
- if (vport->ops->flags & VPORT_F_GEN_STATS)
- free_percpu(vport->percpu_stats);
-
+ kfree((struct vport_portids __force *)vport->upcall_portids);
+ free_percpu(vport->percpu_stats);
kfree(vport);
}
/**
- * __vport_add - add vport device (for kernel callers)
+ * ovs_vport_add - add vport device (for kernel callers)
*
- * @name: Name of new device.
- * @type: Type of new device (to be matched against types in registered vport
- * ops).
- * @config: Device type specific configuration. Userspace pointer.
+ * @parms: Information about new vport.
*
- * Creates a new vport with the specified configuration (which is dependent
- * on device type). Both RTNL and vport locks must be held.
+ * Creates a new vport with the specified configuration (which is dependent on
+ * device type). ovs_mutex must be held.
*/
-struct vport *
-__vport_add(const char *name, const char *type, const void __user *config)
+struct vport *ovs_vport_add(const struct vport_parms *parms)
{
struct vport *vport;
int err = 0;
int i;
- ASSERT_RTNL();
- ASSERT_VPORT();
+ for (i = 0; i < ARRAY_SIZE(vport_ops_list); i++) {
+ if (vport_ops_list[i]->type == parms->type) {
+ struct hlist_head *bucket;
- for (i = 0; i < n_vport_types; i++) {
- if (!strcmp(vport_ops_list[i]->type, type)) {
- vport = vport_ops_list[i]->create(name, config);
+ vport = vport_ops_list[i]->create(parms);
if (IS_ERR(vport)) {
err = PTR_ERR(vport);
goto out;
}
- register_vport(vport);
+ bucket = hash_bucket(ovs_dp_get_net(vport->dp),
+ vport->ops->get_name(vport));
+ hlist_add_head_rcu(&vport->hash_node, bucket);
return vport;
}
}
}
/**
- * __vport_mod - modify existing vport device (for kernel callers)
+ * ovs_vport_set_options - modify existing vport device (for kernel callers)
*
* @vport: vport to modify.
- * @config: Device type specific configuration. Userspace pointer.
+ * @options: New configuration.
*
* Modifies an existing device with the specified configuration (which is
- * dependent on device type). Both RTNL and vport locks must be held.
+ * dependent on device type). ovs_mutex must be held.
*/
-int
-__vport_mod(struct vport *vport, const void __user *config)
+int ovs_vport_set_options(struct vport *vport, struct nlattr *options)
{
- ASSERT_RTNL();
- ASSERT_VPORT();
-
- if (vport->ops->modify)
- return vport->ops->modify(vport, config);
- else
+ if (!vport->ops->set_options)
return -EOPNOTSUPP;
+ return vport->ops->set_options(vport, options);
}
/**
- * __vport_del - delete existing vport device (for kernel callers)
+ * ovs_vport_del - delete existing vport device
*
* @vport: vport to delete.
*
- * Deletes the specified device. The device must not be currently attached to
- * a datapath. It is possible to fail for reasons such as lack of memory.
- * Both RTNL and vport locks must be held.
+ * Detaches @vport from its datapath and destroys it. It is possible to fail
+ * for reasons such as lack of memory. ovs_mutex must be held.
*/
-int
-__vport_del(struct vport *vport)
+void ovs_vport_del(struct vport *vport)
{
- ASSERT_RTNL();
- ASSERT_VPORT();
- BUG_ON(vport_get_dp_port(vport));
-
- unregister_vport(vport);
+ ASSERT_OVSL();
- return vport->ops->destroy(vport);
+ hlist_del_rcu(&vport->hash_node);
+ vport->ops->destroy(vport);
}
/**
- * vport_attach - attach a vport to a datapath
+ * ovs_vport_set_stats - sets offset device stats
*
- * @vport: vport to attach.
- * @dp_port: Datapath port to attach the vport to.
+ * @vport: vport on which to set stats
+ * @stats: stats to set
*
- * Attaches a vport to a specific datapath so that packets may be exchanged.
- * Both ports must be currently unattached. @dp_port must be successfully
- * attached to a vport before it is connected to a datapath and must not be
- * modified while connected. RTNL lock and the appropriate DP mutex must be held.
+ * Provides a set of transmit, receive, and error stats to be added as an
+ * offset to the collected data when stats are retrieved. Some devices may not
+ * support setting the stats, in which case the result will always be
+ * -EOPNOTSUPP.
+ *
+ * Must be called with ovs_mutex.
*/
-int
-vport_attach(struct vport *vport, struct dp_port *dp_port)
+void ovs_vport_set_stats(struct vport *vport, struct ovs_vport_stats *stats)
{
- ASSERT_RTNL();
-
- if (dp_port->vport)
- return -EBUSY;
-
- if (vport_get_dp_port(vport))
- return -EBUSY;
-
- if (vport->ops->attach) {
- int err;
-
- err = vport->ops->attach(vport);
- if (err)
- return err;
- }
-
- dp_port->vport = vport;
- rcu_assign_pointer(vport->dp_port, dp_port);
-
- return 0;
+ spin_lock_bh(&vport->stats_lock);
+ vport->offset_stats = *stats;
+ spin_unlock_bh(&vport->stats_lock);
}
/**
- * vport_detach - detach a vport from a datapath
+ * ovs_vport_get_stats - retrieve device stats
*
- * @vport: vport to detach.
+ * @vport: vport from which to retrieve the stats
+ * @stats: location to store stats
*
- * Detaches a vport from a datapath. May fail for a variety of reasons,
- * including lack of memory. RTNL lock and the appropriate DP mutex must be held.
+ * Retrieves transmit, receive, and error stats for the given device.
+ *
+ * Must be called with ovs_mutex or rcu_read_lock.
*/
-int
-vport_detach(struct vport *vport)
+void ovs_vport_get_stats(struct vport *vport, struct ovs_vport_stats *stats)
{
- struct dp_port *dp_port;
+ int i;
- ASSERT_RTNL();
+ /* We potentially have 3 sources of stats that need to be
+ * combined: those we have collected (split into err_stats and
+ * percpu_stats), offset_stats from set_stats(), and device
+ * error stats from netdev->get_stats() (for errors that happen
+ * downstream and therefore aren't reported through our
+ * vport_record_error() function).
+ * Stats from first two sources are merged and reported by ovs over
+ * OVS_VPORT_ATTR_STATS.
+ * netdev-stats can be directly read over netlink-ioctl.
+ */
- dp_port = vport_get_dp_port(vport);
- if (!dp_port)
- return -EINVAL;
+ spin_lock_bh(&vport->stats_lock);
- dp_port->vport = NULL;
- rcu_assign_pointer(vport->dp_port, NULL);
+ *stats = vport->offset_stats;
- if (vport->ops->detach)
- return vport->ops->detach(vport);
- else
- return 0;
-}
+ stats->rx_errors += vport->err_stats.rx_errors;
+ stats->tx_errors += vport->err_stats.tx_errors;
+ stats->tx_dropped += vport->err_stats.tx_dropped;
+ stats->rx_dropped += vport->err_stats.rx_dropped;
-/**
- * vport_set_mtu - set device MTU (for kernel callers)
- *
- * @vport: vport on which to set MTU.
- * @mtu: New MTU.
- *
- * Sets the MTU of the given device. Some devices may not support setting the
- * MTU, in which case the result will always be -EOPNOTSUPP. RTNL lock must
- * be held.
- */
-int
-vport_set_mtu(struct vport *vport, int mtu)
-{
- ASSERT_RTNL();
+ spin_unlock_bh(&vport->stats_lock);
- if (mtu < 68)
- return -EINVAL;
+ for_each_possible_cpu(i) {
+ const struct pcpu_sw_netstats *percpu_stats;
+ struct pcpu_sw_netstats local_stats;
+ unsigned int start;
- if (vport->ops->set_mtu)
- return vport->ops->set_mtu(vport, mtu);
- else
- return -EOPNOTSUPP;
-}
+ percpu_stats = per_cpu_ptr(vport->percpu_stats, i);
-/**
- * vport_set_addr - set device Ethernet address (for kernel callers)
- *
- * @vport: vport on which to set Ethernet address.
- * @addr: New address.
- *
- * Sets the Ethernet address of the given device. Some devices may not support
- * setting the Ethernet address, in which case the result will always be
- * -EOPNOTSUPP. RTNL lock must be held.
- */
-int
-vport_set_addr(struct vport *vport, const unsigned char *addr)
-{
- ASSERT_RTNL();
+ do {
+ start = u64_stats_fetch_begin_bh(&percpu_stats->syncp);
+ local_stats = *percpu_stats;
+ } while (u64_stats_fetch_retry_bh(&percpu_stats->syncp, start));
- if (!is_valid_ether_addr(addr))
- return -EADDRNOTAVAIL;
-
- if (vport->ops->set_addr)
- return vport->ops->set_addr(vport, addr);
- else
- return -EOPNOTSUPP;
+ stats->rx_bytes += local_stats.rx_bytes;
+ stats->rx_packets += local_stats.rx_packets;
+ stats->tx_bytes += local_stats.tx_bytes;
+ stats->tx_packets += local_stats.tx_packets;
+ }
}
/**
- * vport_get_name - retrieve device name
+ * ovs_vport_get_options - retrieve device options
*
- * @vport: vport from which to retrieve the name.
+ * @vport: vport from which to retrieve the options.
+ * @skb: sk_buff where options should be appended.
*
- * Retrieves the name of the given device. Either RTNL lock or rcu_read_lock
- * must be held for the entire duration that the name is in use.
- */
-const char *
-vport_get_name(const struct vport *vport)
-{
- return vport->ops->get_name(vport);
-}
-
-/**
- * vport_get_type - retrieve device type
+ * Retrieves the configuration of the given device, appending an
+ * %OVS_VPORT_ATTR_OPTIONS attribute that in turn contains nested
+ * vport-specific attributes to @skb.
*
- * @vport: vport from which to retrieve the type.
+ * Returns 0 if successful, -EMSGSIZE if @skb has insufficient room, or another
+ * negative error code if a real error occurred. If an error occurs, @skb is
+ * left unmodified.
*
- * Retrieves the type of the given device. Either RTNL lock or rcu_read_lock
- * must be held for the entire duration that the type is in use.
+ * Must be called with ovs_mutex or rcu_read_lock.
*/
-const char *
-vport_get_type(const struct vport *vport)
+int ovs_vport_get_options(const struct vport *vport, struct sk_buff *skb)
{
- return vport->ops->type;
-}
+ struct nlattr *nla;
+ int err;
-/**
- * vport_get_addr - retrieve device Ethernet address (for kernel callers)
- *
- * @vport: vport from which to retrieve the Ethernet address.
- *
- * Retrieves the Ethernet address of the given device. Either RTNL lock or
- * rcu_read_lock must be held for the entire duration that the Ethernet address
- * is in use.
- */
-const unsigned char *
-vport_get_addr(const struct vport *vport)
-{
- return vport->ops->get_addr(vport);
-}
+ if (!vport->ops->get_options)
+ return 0;
-/**
- * vport_get_dp_port - retrieve attached datapath port
- *
- * @vport: vport from which to retrieve the datapath port.
- *
- * Retrieves the attached datapath port or null if not attached. Either RTNL
- * lock or rcu_read_lock must be held for the entire duration that the datapath
- * port is being accessed.
- */
-struct dp_port *
-vport_get_dp_port(const struct vport *vport)
-{
- return rcu_dereference(vport->dp_port);
-}
+ nla = nla_nest_start(skb, OVS_VPORT_ATTR_OPTIONS);
+ if (!nla)
+ return -EMSGSIZE;
-/**
- * vport_get_kobj - retrieve associated kobj
- *
- * @vport: vport from which to retrieve the associated kobj
- *
- * Retrieves the associated kobj or null if no kobj. The returned kobj is
- * valid for as long as the vport exists.
- */
-struct kobject *
-vport_get_kobj(const struct vport *vport)
-{
- if (vport->ops->get_kobj)
- return vport->ops->get_kobj(vport);
- else
- return NULL;
-}
+ err = vport->ops->get_options(vport, skb);
+ if (err) {
+ nla_nest_cancel(skb, nla);
+ return err;
+ }
-/**
- * vport_get_flags - retrieve device flags
- *
- * @vport: vport from which to retrieve the flags
- *
- * Retrieves the flags of the given device. Either RTNL lock or rcu_read_lock
- * must be held.
- */
-unsigned
-vport_get_flags(const struct vport *vport)
-{
- return vport->ops->get_dev_flags(vport);
+ nla_nest_end(skb, nla);
+ return 0;
}
-/**
- * vport_get_flags - check whether device is running
- *
- * @vport: vport on which to check status.
- *
- * Checks whether the given device is running. Either RTNL lock or
- * rcu_read_lock must be held.
- */
-int
-vport_is_running(const struct vport *vport)
+static void vport_portids_destroy_rcu_cb(struct rcu_head *rcu)
{
- return vport->ops->is_running(vport);
+ struct vport_portids *ids = container_of(rcu, struct vport_portids,
+ rcu);
+
+ kfree(ids);
}
/**
- * vport_get_flags - retrieve device operating state
+ * ovs_vport_set_upcall_portids - set upcall portids of @vport.
*
- * @vport: vport from which to check status
+ * @vport: vport to modify.
+ * @ids: new configuration, an array of port ids.
*
- * Retrieves the RFC2863 operstate of the given device. Either RTNL lock or
- * rcu_read_lock must be held.
- */
-unsigned char
-vport_get_operstate(const struct vport *vport)
-{
- return vport->ops->get_operstate(vport);
-}
-
-/**
- * vport_get_ifindex - retrieve device system interface index
+ * Sets the vport's upcall_portids to @ids.
*
- * @vport: vport from which to retrieve index
+ * Returns 0 if successful, -EINVAL if @ids is zero length or cannot be parsed
+ * as an array of U32.
*
- * Retrieves the system interface index of the given device. Not all devices
- * will have system indexes, in which case the index of the datapath local
- * port is returned. Returns a negative index on error. Either RTNL lock or
- * rcu_read_lock must be held.
+ * Must be called with ovs_mutex.
*/
-int
-vport_get_ifindex(const struct vport *vport)
+int ovs_vport_set_upcall_portids(struct vport *vport, struct nlattr *ids)
{
- const struct dp_port *dp_port;
+ struct vport_portids *old, *vport_portids;
+
+ if (!nla_len(ids) || nla_len(ids) % sizeof(u32))
+ return -EINVAL;
+
+ old = ovsl_dereference(vport->upcall_portids);
+
+ vport_portids = kmalloc(sizeof *vport_portids + nla_len(ids),
+ GFP_KERNEL);
+ vport_portids->n_ids = nla_len(ids) / sizeof(u32);
+ vport_portids->rn_ids = reciprocal_value(vport_portids->n_ids);
+ nla_memcpy(vport_portids->ids, ids, nla_len(ids));
- if (vport->ops->get_ifindex)
- return vport->ops->get_ifindex(vport);
+ rcu_assign_pointer(vport->upcall_portids, vport_portids);
- /* If we don't actually have an ifindex, use the local port's.
- * Userspace doesn't check it anyways. */
- dp_port = vport_get_dp_port(vport);
- if (!dp_port)
- return -EAGAIN;
+ if (old)
+ call_rcu(&old->rcu, vport_portids_destroy_rcu_cb);
- return vport_get_ifindex(dp_port->dp->ports[ODPP_LOCAL]->vport);
+ return 0;
}
/**
- * vport_get_iflink - retrieve device system link index
+ * ovs_vport_get_upcall_portids - get the upcall_portids of @vport.
+ *
+ * @vport: vport from which to retrieve the portids.
+ * @skb: sk_buff where portids should be appended.
*
- * @vport: vport from which to retrieve index
+ * Retrieves the configuration of the given vport, appending the
+ * %OVS_VPORT_ATTR_UPCALL_PID attribute which is the array of upcall
+ * portids to @skb.
*
- * Retrieves the system link index of the given device. The link is the index
- * of the interface on which the packet will actually be sent. In most cases
- * this is the same as the ifindex but may be different for tunnel devices.
- * Returns a negative index on error. Either RTNL lock or rcu_read_lock must
- * be held.
+ * Returns 0 if successful, -EMSGSIZE if @skb has insufficient room.
+ * If an error occurs, @skb is left unmodified. Must be called with
+ * ovs_mutex or rcu_read_lock.
*/
-int
-vport_get_iflink(const struct vport *vport)
+int ovs_vport_get_upcall_portids(const struct vport *vport,
+ struct sk_buff *skb)
{
- if (vport->ops->get_iflink)
- return vport->ops->get_iflink(vport);
+ struct vport_portids *ids;
+
+ ids = rcu_dereference_ovsl(vport->upcall_portids);
- /* If we don't have an iflink, use the ifindex. In most cases they
- * are the same. */
- return vport_get_ifindex(vport);
+ if (vport->dp->user_features & OVS_DP_F_VPORT_PIDS)
+ return nla_put(skb, OVS_VPORT_ATTR_UPCALL_PID,
+ ids->n_ids * sizeof(u32), (void *) ids->ids);
+ else
+ return nla_put_u32(skb, OVS_VPORT_ATTR_UPCALL_PID, ids->ids[0]);
}
/**
- * vport_get_mtu - retrieve device MTU (for kernel callers)
+ * ovs_vport_find_upcall_portid - find the upcall portid to send upcall.
+ *
+ * @vport: vport from which the missed packet is received.
+ * @skb: skb that the missed packet was received.
*
- * @vport: vport from which to retrieve MTU
+ * Uses the skb_get_hash() to select the upcall portid to send the
+ * upcall.
*
- * Retrieves the MTU of the given device. Either RTNL lock or rcu_read_lock
- * must be held.
+ * Returns the portid of the target socket. Must be called with rcu_read_lock.
*/
-int
-vport_get_mtu(const struct vport *vport)
+u32 ovs_vport_find_upcall_portid(const struct vport *p, struct sk_buff *skb)
{
- return vport->ops->get_mtu(vport);
+ struct vport_portids *ids;
+ u32 hash;
+
+ ids = rcu_dereference(p->upcall_portids);
+
+ if (ids->n_ids == 1 && ids->ids[0] == 0)
+ return 0;
+
+ hash = skb_get_hash(skb);
+ return ids->ids[hash - ids->n_ids * reciprocal_divide(hash, ids->rn_ids)];
}
/**
- * vport_receive - pass up received packet to the datapath for processing
+ * ovs_vport_receive - pass up received packet to the datapath for processing
*
* @vport: vport that received the packet
* @skb: skb that was received
+ * @tun_key: tunnel (if any) that carried packet
*
* Must be called with rcu_read_lock. The packet cannot be shared and
* skb->data should point to the Ethernet header. The caller must have already
* called compute_ip_summed() to initialize the checksumming fields.
*/
-void
-vport_receive(struct vport *vport, struct sk_buff *skb)
+void ovs_vport_receive(struct vport *vport, struct sk_buff *skb,
+ struct ovs_key_ipv4_tunnel *tun_key)
{
- struct dp_port *dp_port = vport_get_dp_port(vport);
-
- if (!dp_port) {
- vport_record_error(vport, VPORT_E_RX_DROPPED);
- kfree_skb(skb);
-
- return;
- }
-
- if (vport->ops->flags & VPORT_F_GEN_STATS) {
- struct vport_percpu_stats *stats;
+ struct pcpu_sw_netstats *stats;
- local_bh_disable();
+ stats = this_cpu_ptr(vport->percpu_stats);
+ u64_stats_update_begin(&stats->syncp);
+ stats->rx_packets++;
+ stats->rx_bytes += skb->len;
+ u64_stats_update_end(&stats->syncp);
- stats = per_cpu_ptr(vport->percpu_stats, smp_processor_id());
- stats->rx_packets++;
- stats->rx_bytes += skb->len;
-
- local_bh_enable();
- }
-
- if (!(vport->ops->flags & VPORT_F_TUN_ID))
- OVS_CB(skb)->tun_id = 0;
-
- dp_process_received_packet(dp_port, skb);
+ OVS_CB(skb)->tun_key = tun_key;
+ ovs_dp_process_received_packet(vport, skb);
}
/**
- * vport_send - send a packet on a device
+ * ovs_vport_send - send a packet on a device
*
* @vport: vport on which to send the packet
* @skb: skb to send
*
- * Sends the given packet and returns the length of data sent. Either RTNL
+ * Sends the given packet and returns the length of data sent. Either ovs
* lock or rcu_read_lock must be held.
*/
-int
-vport_send(struct vport *vport, struct sk_buff *skb)
+int ovs_vport_send(struct vport *vport, struct sk_buff *skb)
{
- int sent;
-
- sent = vport->ops->send(vport, skb);
+ int sent = vport->ops->send(vport, skb);
- if (vport->ops->flags & VPORT_F_GEN_STATS && sent > 0) {
- struct vport_percpu_stats *stats;
+ if (likely(sent > 0)) {
+ struct pcpu_sw_netstats *stats;
- local_bh_disable();
+ stats = this_cpu_ptr(vport->percpu_stats);
- stats = per_cpu_ptr(vport->percpu_stats, smp_processor_id());
+ u64_stats_update_begin(&stats->syncp);
stats->tx_packets++;
stats->tx_bytes += sent;
-
- local_bh_enable();
- }
+ u64_stats_update_end(&stats->syncp);
+ } else if (sent < 0) {
+ ovs_vport_record_error(vport, VPORT_E_TX_ERROR);
+ kfree_skb(skb);
+ } else
+ ovs_vport_record_error(vport, VPORT_E_TX_DROPPED);
return sent;
}
/**
- * vport_record_error - indicate device error to generic stats layer
+ * ovs_vport_record_error - indicate device error to generic stats layer
*
* @vport: vport that encountered the error
* @err_type: one of enum vport_err_type types to indicate the error type
*
* If using the vport generic stats layer indicate that an error of the given
- * type has occured.
+ * type has occurred.
*/
-void
-vport_record_error(struct vport *vport, enum vport_err_type err_type)
+static void ovs_vport_record_error(struct vport *vport,
+ enum vport_err_type err_type)
{
- if (vport->ops->flags & VPORT_F_GEN_STATS) {
-
- spin_lock_bh(&vport->err_stats.lock);
-
- switch (err_type) {
- case VPORT_E_RX_DROPPED:
- vport->err_stats.rx_dropped++;
- break;
-
- case VPORT_E_RX_ERROR:
- vport->err_stats.rx_errors++;
- break;
+ spin_lock(&vport->stats_lock);
- case VPORT_E_RX_FRAME:
- vport->err_stats.rx_frame_err++;
- break;
+ switch (err_type) {
+ case VPORT_E_RX_DROPPED:
+ vport->err_stats.rx_dropped++;
+ break;
- case VPORT_E_RX_OVER:
- vport->err_stats.rx_over_err++;
- break;
+ case VPORT_E_RX_ERROR:
+ vport->err_stats.rx_errors++;
+ break;
- case VPORT_E_RX_CRC:
- vport->err_stats.rx_crc_err++;
- break;
+ case VPORT_E_TX_DROPPED:
+ vport->err_stats.tx_dropped++;
+ break;
- case VPORT_E_TX_DROPPED:
- vport->err_stats.tx_dropped++;
- break;
+ case VPORT_E_TX_ERROR:
+ vport->err_stats.tx_errors++;
+ break;
+ }
- case VPORT_E_TX_ERROR:
- vport->err_stats.tx_errors++;
- break;
+ spin_unlock(&vport->stats_lock);
+}
- case VPORT_E_COLLISION:
- vport->err_stats.collisions++;
- break;
- };
+static void free_vport_rcu(struct rcu_head *rcu)
+{
+ struct vport *vport = container_of(rcu, struct vport, rcu);
- spin_unlock_bh(&vport->err_stats.lock);
- }
+ ovs_vport_free(vport);
}
-/**
- * vport_gen_ether_addr - generate an Ethernet address
- *
- * @addr: location to store generated address
- *
- * Generates a random Ethernet address for use when creating a device that
- * has no natural address.
- */
-void
-vport_gen_ether_addr(u8 *addr)
+void ovs_vport_deferred_free(struct vport *vport)
{
- random_ether_addr(addr);
-
- /* Set the OUI to the Nicira one. */
- addr[0] = 0x00;
- addr[1] = 0x23;
- addr[2] = 0x20;
+ if (!vport)
+ return;
- /* Set the top bit to indicate random address. */
- addr[3] |= 0x80;
+ call_rcu(&vport->rcu, free_vport_rcu);
}
git://git.onelab.eu / sliver-openvswitch.git / blobdiff