mir/ovs
2
0
Fork 0
mirror of https://github.com/openvswitch/ovs synced 2025-08-23 10:28:00 +00:00
Code Issues Releases Wiki Activity
ovs/lib/netdev-dummy.c

311 lines
8.0 KiB
C
Raw Normal View History

Add new "dummy" netdev and dpif implementations for use in unit tests.
2010-11-29 12:21:08 -08:00
/*
netdev: Clean up and refactor packet receive interface. The Open vSwitch tree only has one user of the ability for a netdev to receive packets from a network device. Thus, this commit simplifies the common-case use of the netdev interface by replacing the "ethertype" option from "struct netdev_options" by a new netdev_listen() call. The only user of netdev_listen() wants to receive all packets from a network device, so this commit also removes the ability to restrict the received packets to a particular protocol. (This ability was once used by the Open vSwitch integrated DHCP client, but that code has been removed.) This commit also simplifies and improves the implementation of the code in netdev-linux that started listening to a network device. Before, I had not figured out how to avoid receiving all packets on all devices before binding to a particular device, but I took a closer look at the kernel code and figured it out. I've tested that the userspace datapath (dpif-netdev), the only user of netdev_recv(), still works after this change.
2011-08-05 14:15:32 -07:00
* Copyright (c) 2010, 2011 Nicira Networks.
Add new "dummy" netdev and dpif implementations for use in unit tests.
2010-11-29 12:21:08 -08:00
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at:
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
#include <config.h>
#include "dummy.h"
#include <errno.h>
#include "list.h"
#include "netdev-provider.h"
#include "packets.h"
#include "shash.h"
#include "vlog.h"
VLOG_DEFINE_THIS_MODULE(netdev_dummy);
struct netdev_dev_dummy {
struct netdev_dev netdev_dev;
uint8_t hwaddr[ETH_ADDR_LEN];
int mtu;
struct netdev_stats stats;
enum netdev_flags flags;
netdev: New Function netdev_change_seq(). This new function will provide a much simpler replacement for netdev_monitor in the future.
2011-05-26 14:28:11 -07:00
unsigned int change_seq;
Add new "dummy" netdev and dpif implementations for use in unit tests.
2010-11-29 12:21:08 -08:00
};
struct netdev_dummy {
struct netdev netdev;
};
static int netdev_dummy_create(const struct netdev_class *, const char *,
netdev: Decouple creating and configuring network devices. Until now, each call to netdev_open() for a particular network device had to either specify a set of network device arguments that was either empty or (for devices that already existed) equal to the existing device's configuration. Unfortunately, the definition of "equality" in the latter case was mostly done in terms of strict equality of string-to-string maps, which caused problems in cases where, for example, one set of arguments specified the default value of an optional argument explicitly and the other omitted it. The netdev interface does have provisions for defining equality other ways, but this had only been done in one case that was especially problematic in practice. One way to solve this particular problem would be to carefully define equality in all the problematic cases. This commit takes another approach based on the realization that there is really no need to do any comparisons. Instead, it removes configuration at netdev_open() time entirely, because almost all of netdev_open()'s callers are not interested in creating and configuring a netdev. Most of them just want to open a configured device and use it. Therefore, this commit stops providing any configuration arguments to netdev_open() and the provider functions that it calls. Instead, a caller that does want to configure a device does so after it opens it, by calling netdev_set_config(). This change allows us to simplify the netdev interface a bit. There is no longer any need to implement argument comparisons. As a result, there is also no need for "struct netdev_dev" to keep track of configuration at all. Instead, the network devices that have configuration keep track of it in their own internal form. This new interface does mean that it becomes possible to accidentally create and try to use an unconfigured netdev that requires configuration. Bug #6677. Reported-by: Paul Ingram <paul@nicira.com>
2011-08-08 12:49:17 -07:00
struct netdev_dev **);
Add new "dummy" netdev and dpif implementations for use in unit tests.
2010-11-29 12:21:08 -08:00
static void netdev_dummy_poll_notify(const struct netdev *);
static bool
is_dummy_class(const struct netdev_class *class)
{
return class->create == netdev_dummy_create;
}
static struct netdev_dev_dummy *
netdev_dev_dummy_cast(const struct netdev_dev *netdev_dev)
{
assert(is_dummy_class(netdev_dev_get_class(netdev_dev)));
return CONTAINER_OF(netdev_dev, struct netdev_dev_dummy, netdev_dev);
}
static struct netdev_dummy *
netdev_dummy_cast(const struct netdev *netdev)
{
struct netdev_dev *netdev_dev = netdev_get_dev(netdev);
assert(is_dummy_class(netdev_dev_get_class(netdev_dev)));
return CONTAINER_OF(netdev, struct netdev_dummy, netdev);
}
static int
netdev_dummy_create(const struct netdev_class *class, const char *name,
struct netdev_dev **netdev_devp)
{
static unsigned int n = 0xaa550000;
struct netdev_dev_dummy *netdev_dev;
netdev_dev = xzalloc(sizeof *netdev_dev);
netdev: Decouple creating and configuring network devices. Until now, each call to netdev_open() for a particular network device had to either specify a set of network device arguments that was either empty or (for devices that already existed) equal to the existing device's configuration. Unfortunately, the definition of "equality" in the latter case was mostly done in terms of strict equality of string-to-string maps, which caused problems in cases where, for example, one set of arguments specified the default value of an optional argument explicitly and the other omitted it. The netdev interface does have provisions for defining equality other ways, but this had only been done in one case that was especially problematic in practice. One way to solve this particular problem would be to carefully define equality in all the problematic cases. This commit takes another approach based on the realization that there is really no need to do any comparisons. Instead, it removes configuration at netdev_open() time entirely, because almost all of netdev_open()'s callers are not interested in creating and configuring a netdev. Most of them just want to open a configured device and use it. Therefore, this commit stops providing any configuration arguments to netdev_open() and the provider functions that it calls. Instead, a caller that does want to configure a device does so after it opens it, by calling netdev_set_config(). This change allows us to simplify the netdev interface a bit. There is no longer any need to implement argument comparisons. As a result, there is also no need for "struct netdev_dev" to keep track of configuration at all. Instead, the network devices that have configuration keep track of it in their own internal form. This new interface does mean that it becomes possible to accidentally create and try to use an unconfigured netdev that requires configuration. Bug #6677. Reported-by: Paul Ingram <paul@nicira.com>
2011-08-08 12:49:17 -07:00
netdev_dev_init(&netdev_dev->netdev_dev, name, class);
Add new "dummy" netdev and dpif implementations for use in unit tests.
2010-11-29 12:21:08 -08:00
netdev_dev->hwaddr[0] = 0xaa;
netdev_dev->hwaddr[1] = 0x55;
netdev_dev->hwaddr[2] = n >> 24;
netdev_dev->hwaddr[3] = n >> 16;
netdev_dev->hwaddr[4] = n >> 8;
netdev_dev->hwaddr[5] = n;
netdev_dev->mtu = 1500;
netdev_dev->flags = 0;
netdev: New Function netdev_change_seq(). This new function will provide a much simpler replacement for netdev_monitor in the future.
2011-05-26 14:28:11 -07:00
netdev_dev->change_seq = 1;
Add new "dummy" netdev and dpif implementations for use in unit tests.
2010-11-29 12:21:08 -08:00
n++;
*netdev_devp = &netdev_dev->netdev_dev;
return 0;
}
static void
netdev_dummy_destroy(struct netdev_dev *netdev_dev_)
{
struct netdev_dev_dummy *netdev_dev = netdev_dev_dummy_cast(netdev_dev_);
free(netdev_dev);
}
static int
netdev: Clean up and refactor packet receive interface. The Open vSwitch tree only has one user of the ability for a netdev to receive packets from a network device. Thus, this commit simplifies the common-case use of the netdev interface by replacing the "ethertype" option from "struct netdev_options" by a new netdev_listen() call. The only user of netdev_listen() wants to receive all packets from a network device, so this commit also removes the ability to restrict the received packets to a particular protocol. (This ability was once used by the Open vSwitch integrated DHCP client, but that code has been removed.) This commit also simplifies and improves the implementation of the code in netdev-linux that started listening to a network device. Before, I had not figured out how to avoid receiving all packets on all devices before binding to a particular device, but I took a closer look at the kernel code and figured it out. I've tested that the userspace datapath (dpif-netdev), the only user of netdev_recv(), still works after this change.
2011-08-05 14:15:32 -07:00
netdev_dummy_open(struct netdev_dev *netdev_dev_, struct netdev **netdevp)
Add new "dummy" netdev and dpif implementations for use in unit tests.
2010-11-29 12:21:08 -08:00
{
struct netdev_dummy *netdev;
netdev = xmalloc(sizeof *netdev);
netdev_init(&netdev->netdev, netdev_dev_);
*netdevp = &netdev->netdev;
return 0;
}
static void
netdev_dummy_close(struct netdev *netdev_)
{
struct netdev_dummy *netdev = netdev_dummy_cast(netdev_);
free(netdev);
}
netdev: Clean up and refactor packet receive interface. The Open vSwitch tree only has one user of the ability for a netdev to receive packets from a network device. Thus, this commit simplifies the common-case use of the netdev interface by replacing the "ethertype" option from "struct netdev_options" by a new netdev_listen() call. The only user of netdev_listen() wants to receive all packets from a network device, so this commit also removes the ability to restrict the received packets to a particular protocol. (This ability was once used by the Open vSwitch integrated DHCP client, but that code has been removed.) This commit also simplifies and improves the implementation of the code in netdev-linux that started listening to a network device. Before, I had not figured out how to avoid receiving all packets on all devices before binding to a particular device, but I took a closer look at the kernel code and figured it out. I've tested that the userspace datapath (dpif-netdev), the only user of netdev_recv(), still works after this change.
2011-08-05 14:15:32 -07:00
static int
netdev_dummy_listen(struct netdev *netdev_ OVS_UNUSED)
{
/* It's OK to listen on a dummy device. It just never receives any
* packets. */
return 0;
}
static int
netdev_dummy_recv(struct netdev *netdev_ OVS_UNUSED,
void *buffer OVS_UNUSED, size_t size OVS_UNUSED)
{
/* A dummy device never receives any packets. */
return -EAGAIN;
}
Add new "dummy" netdev and dpif implementations for use in unit tests.
2010-11-29 12:21:08 -08:00
static int
netdev_dummy_set_etheraddr(struct netdev *netdev,
const uint8_t mac[ETH_ADDR_LEN])
{
struct netdev_dev_dummy *dev =
netdev_dev_dummy_cast(netdev_get_dev(netdev));
if (!eth_addr_equals(dev->hwaddr, mac)) {
memcpy(dev->hwaddr, mac, ETH_ADDR_LEN);
netdev_dummy_poll_notify(netdev);
}
return 0;
}
static int
netdev_dummy_get_etheraddr(const struct netdev *netdev,
uint8_t mac[ETH_ADDR_LEN])
{
const struct netdev_dev_dummy *dev =
netdev_dev_dummy_cast(netdev_get_dev(netdev));
memcpy(mac, dev->hwaddr, ETH_ADDR_LEN);
return 0;
}
static int
netdev_dummy_get_mtu(const struct netdev *netdev, int *mtup)
{
const struct netdev_dev_dummy *dev =
netdev_dev_dummy_cast(netdev_get_dev(netdev));
*mtup = dev->mtu;
return 0;
}
datapath: Strip down vport interface : OVS_VPORT_ATTR_MTU There is no need to have vport attribute MTU (OVS_VPORT_ATTR_MTU) as linux net-dev-ioctl can be used to get/set MTU for linux device. Following patch removes OVS_VPORT_ATTR_MTU from datapath protocol. This patch also adds netdev_set_mtu interface. So that MTU adjustments can be done from OVS userspace. get_mtu() interface is also changed, now get_mtu() returns EOPNOTSUPP rather than returning 0 and setting *pmtu to INT_MAX in case there is no MTU attribute for given device. Signed-off-by: Pravin B Shelar <pshelar@nicira.com> Acked-by: Jesse Gross <jesse@nicira.com>
2011-09-12 17:12:52 -07:00
static int
netdev_dummy_set_mtu(const struct netdev *netdev, int mtu)
{
struct netdev_dev_dummy *dev =
netdev_dev_dummy_cast(netdev_get_dev(netdev));
dev->mtu = mtu;
return 0;
}
Add new "dummy" netdev and dpif implementations for use in unit tests.
2010-11-29 12:21:08 -08:00
static int
netdev_dummy_get_stats(const struct netdev *netdev, struct netdev_stats *stats)
{
const struct netdev_dev_dummy *dev =
netdev_dev_dummy_cast(netdev_get_dev(netdev));
*stats = dev->stats;
return 0;
}
static int
netdev_dummy_set_stats(struct netdev *netdev, const struct netdev_stats *stats)
{
struct netdev_dev_dummy *dev =
netdev_dev_dummy_cast(netdev_get_dev(netdev));
dev->stats = *stats;
return 0;
}
static int
netdev_dummy_update_flags(struct netdev *netdev,
enum netdev_flags off, enum netdev_flags on,
enum netdev_flags *old_flagsp)
{
struct netdev_dev_dummy *dev =
netdev_dev_dummy_cast(netdev_get_dev(netdev));
if ((off | on) & ~(NETDEV_UP | NETDEV_PROMISC)) {
return EINVAL;
}
*old_flagsp = dev->flags;
dev->flags |= on;
dev->flags &= ~off;
if (*old_flagsp != dev->flags) {
netdev_dummy_poll_notify(netdev);
}
return 0;
}
netdev: New Function netdev_change_seq(). This new function will provide a much simpler replacement for netdev_monitor in the future.
2011-05-26 14:28:11 -07:00
static unsigned int
netdev_dummy_change_seq(const struct netdev *netdev)
{
return netdev_dev_dummy_cast(netdev_get_dev(netdev))->change_seq;
}
Add new "dummy" netdev and dpif implementations for use in unit tests.
2010-11-29 12:21:08 -08:00
/* Helper functions. */
static void
netdev_dummy_poll_notify(const struct netdev *netdev)
{
netdev: New Function netdev_change_seq(). This new function will provide a much simpler replacement for netdev_monitor in the future.
2011-05-26 14:28:11 -07:00
struct netdev_dev_dummy *dev =
netdev_dev_dummy_cast(netdev_get_dev(netdev));
Add new "dummy" netdev and dpif implementations for use in unit tests.
2010-11-29 12:21:08 -08:00
netdev: New Function netdev_change_seq(). This new function will provide a much simpler replacement for netdev_monitor in the future.
2011-05-26 14:28:11 -07:00
dev->change_seq++;
if (!dev->change_seq) {
dev->change_seq++;
}
Add new "dummy" netdev and dpif implementations for use in unit tests.
2010-11-29 12:21:08 -08:00
}
static const struct netdev_class dummy_class = {
"dummy",
NULL, /* init */
NULL, /* run */
NULL, /* wait */
netdev_dummy_create,
netdev_dummy_destroy,
netdev: Decouple creating and configuring network devices. Until now, each call to netdev_open() for a particular network device had to either specify a set of network device arguments that was either empty or (for devices that already existed) equal to the existing device's configuration. Unfortunately, the definition of "equality" in the latter case was mostly done in terms of strict equality of string-to-string maps, which caused problems in cases where, for example, one set of arguments specified the default value of an optional argument explicitly and the other omitted it. The netdev interface does have provisions for defining equality other ways, but this had only been done in one case that was especially problematic in practice. One way to solve this particular problem would be to carefully define equality in all the problematic cases. This commit takes another approach based on the realization that there is really no need to do any comparisons. Instead, it removes configuration at netdev_open() time entirely, because almost all of netdev_open()'s callers are not interested in creating and configuring a netdev. Most of them just want to open a configured device and use it. Therefore, this commit stops providing any configuration arguments to netdev_open() and the provider functions that it calls. Instead, a caller that does want to configure a device does so after it opens it, by calling netdev_set_config(). This change allows us to simplify the netdev interface a bit. There is no longer any need to implement argument comparisons. As a result, there is also no need for "struct netdev_dev" to keep track of configuration at all. Instead, the network devices that have configuration keep track of it in their own internal form. This new interface does mean that it becomes possible to accidentally create and try to use an unconfigured netdev that requires configuration. Bug #6677. Reported-by: Paul Ingram <paul@nicira.com>
2011-08-08 12:49:17 -07:00
NULL, /* get_config */
netdev: Add methods to do netdev-specific argument comparisons. When doing a netdev_open(), a check is first done to make sure the arguments are equivalent for any open devices with the same name. In most cases, a simple shash comparison is sufficient. However, IPsec key configuration is handled by an external program, so it is not pushed down into the kernel module. Thus, when the "unparse_config" method is called on an existing IPsec-based vport, a simple comparison with the returned data will not match the original configuration. This commit adds code to allow netdev-specific argument comparisons and has "ipsec_gre" make use of them. Bug #5575
2011-06-13 19:26:47 -07:00
NULL, /* set_config */
Add new "dummy" netdev and dpif implementations for use in unit tests.
2010-11-29 12:21:08 -08:00
netdev_dummy_open,
netdev_dummy_close,
netdev: Clean up and refactor packet receive interface. The Open vSwitch tree only has one user of the ability for a netdev to receive packets from a network device. Thus, this commit simplifies the common-case use of the netdev interface by replacing the "ethertype" option from "struct netdev_options" by a new netdev_listen() call. The only user of netdev_listen() wants to receive all packets from a network device, so this commit also removes the ability to restrict the received packets to a particular protocol. (This ability was once used by the Open vSwitch integrated DHCP client, but that code has been removed.) This commit also simplifies and improves the implementation of the code in netdev-linux that started listening to a network device. Before, I had not figured out how to avoid receiving all packets on all devices before binding to a particular device, but I took a closer look at the kernel code and figured it out. I've tested that the userspace datapath (dpif-netdev), the only user of netdev_recv(), still works after this change.
2011-08-05 14:15:32 -07:00
netdev_dummy_listen, /* listen */
netdev_dummy_recv, /* recv */
Add new "dummy" netdev and dpif implementations for use in unit tests.
2010-11-29 12:21:08 -08:00
NULL, /* recv_wait */
NULL, /* drain */
NULL, /* send */
NULL, /* send_wait */
netdev_dummy_set_etheraddr,
netdev_dummy_get_etheraddr,
netdev_dummy_get_mtu,
datapath: Strip down vport interface : OVS_VPORT_ATTR_MTU There is no need to have vport attribute MTU (OVS_VPORT_ATTR_MTU) as linux net-dev-ioctl can be used to get/set MTU for linux device. Following patch removes OVS_VPORT_ATTR_MTU from datapath protocol. This patch also adds netdev_set_mtu interface. So that MTU adjustments can be done from OVS userspace. get_mtu() interface is also changed, now get_mtu() returns EOPNOTSUPP rather than returning 0 and setting *pmtu to INT_MAX in case there is no MTU attribute for given device. Signed-off-by: Pravin B Shelar <pshelar@nicira.com> Acked-by: Jesse Gross <jesse@nicira.com>
2011-09-12 17:12:52 -07:00
netdev_dummy_set_mtu,
Add new "dummy" netdev and dpif implementations for use in unit tests.
2010-11-29 12:21:08 -08:00
NULL, /* get_ifindex */
NULL, /* get_carrier */
vswitchd: New column "link_resets". An interface's 'link_resets' column represents the number of times Open vSwitch has observed its link_state change.
2011-10-14 12:49:57 -07:00
NULL, /* get_carrier_resets */
vswitchd: Add miimon support. This commit allows users to check link status in bonded ports using MII instead of carrier.
2011-01-07 16:22:34 -08:00
NULL, /* get_miimon */
Add new "dummy" netdev and dpif implementations for use in unit tests.
2010-11-29 12:21:08 -08:00
netdev_dummy_get_stats,
netdev_dummy_set_stats,
NULL, /* get_features */
NULL, /* set_advertisements */
NULL, /* set_policing */
NULL, /* get_qos_types */
NULL, /* get_qos_capabilities */
NULL, /* get_qos */
NULL, /* set_qos */
NULL, /* get_queue */
NULL, /* set_queue */
NULL, /* delete_queue */
NULL, /* get_queue_stats */
NULL, /* dump_queues */
NULL, /* dump_queue_stats */
NULL, /* get_in4 */
NULL, /* set_in4 */
NULL, /* get_in6 */
NULL, /* add_router */
NULL, /* get_next_hop */
bridge: Move tunnel_egress_iface to status column. This commit removes the tunnel_egress_iface column from the interface table and moves it's data to the status column. In the process it reverts the database to version 1.0.0.
2011-01-05 11:51:15 -08:00
NULL, /* get_status */
Add new "dummy" netdev and dpif implementations for use in unit tests.
2010-11-29 12:21:08 -08:00
NULL, /* arp_lookup */
netdev_dummy_update_flags,
netdev: New Function netdev_change_seq(). This new function will provide a much simpler replacement for netdev_monitor in the future.
2011-05-26 14:28:11 -07:00
netdev_dummy_change_seq
Add new "dummy" netdev and dpif implementations for use in unit tests.
2010-11-29 12:21:08 -08:00
};
void
netdev_dummy_register(void)
{
netdev_register_provider(&dummy_class);
}
Reference in New Issue Copy Permalink