1 .\" $NetBSD: bridge.4,v 1.5 2004/01/31 20:14:11 jdc Exp $
3 .\" Copyright 2001 Wasabi Systems, Inc.
4 .\" All rights reserved.
6 .\" Written by Jason R. Thorpe for Wasabi Systems, Inc.
8 .\" Redistribution and use in source and binary forms, with or without
9 .\" modification, are permitted provided that the following conditions
11 .\" 1. Redistributions of source code must retain the above copyright
12 .\" notice, this list of conditions and the following disclaimer.
13 .\" 2. Redistributions in binary form must reproduce the above copyright
14 .\" notice, this list of conditions and the following disclaimer in the
15 .\" documentation and/or other materials provided with the distribution.
16 .\" 3. All advertising materials mentioning features or use of this software
17 .\" must display the following acknowledgement:
18 .\" This product includes software developed for the NetBSD Project by
19 .\" Wasabi Systems, Inc.
20 .\" 4. The name of Wasabi Systems, Inc. may not be used to endorse
21 .\" or promote products derived from this software without specific prior
22 .\" written permission.
24 .\" THIS SOFTWARE IS PROVIDED BY WASABI SYSTEMS, INC. ``AS IS'' AND
25 .\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
26 .\" TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
27 .\" PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL WASABI SYSTEMS, INC
28 .\" BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
29 .\" CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
30 .\" SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
31 .\" INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
32 .\" CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
33 .\" ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
34 .\" POSSIBILITY OF SUCH DAMAGE.
43 .Nd network bridge device
45 To compile this driver into the kernel,
46 place the following line in your
47 kernel configuration file:
48 .Bd -ragged -offset indent
49 .Cd "device if_bridge"
52 Alternatively, to load the driver as a
53 module at boot time, place the following lines in
55 .Bd -literal -offset indent
62 driver creates a logical link between two or more IEEE 802 networks
64 .Dq "similar enough" )
66 For example, it is possible to bridge Ethernet and 802.11 networks together,
67 but it is not possible to bridge Ethernet and Token Ring together.
71 interface is created at runtime using interface cloning.
73 most easily done with the
83 interface randomly chooses a link (MAC) address in the range reserved for
84 locally administered addresses when it is created.
85 This address is guaranteed to be unique
89 interfaces on the local machine.
90 Thus you can theoretically have two bridges on the different machines with
91 the same link addresses.
92 The address can be changed by assigning the desired link address using
98 .Va net.link.bridge.inherit_mac
99 has non-zero value, newly created bridge will inherit MAC address
100 from its first member instead of choosing random link-level address.
101 This will provide more predictable bridge MAC without any
102 additional configuration, but currently this feature is known
103 to break some L2 protocols, for example PPPoE that is provided
108 Now this feature is considered as experimental and is turned off
111 A bridge can be used to provide several services, such as a simple
112 802.11-to-Ethernet bridge for wireless hosts, and traffic isolation.
114 A bridge works like a switch, forwarding traffic from one interface
116 Multicast and broadcast packets are always forwarded to all
117 interfaces that are part of the bridge.
118 For unicast traffic, the bridge learns which MAC addresses are associated
119 with which interfaces and will forward the traffic selectively.
121 All the bridged member interfaces need to be up in order to pass network traffic.
122 These can be enabled using
125 .Va ifconfig_ Ns Ao Ar interface Ac Ns Li ="up"
129 The MTU of the first member interface to be added is used as the bridge MTU.
130 All additional members are required to have exactly the same value.
132 The TXCSUM capability is disabled for any interface added to the bridge, and it
133 is restored when the interface is removed again.
137 where the packets are discarded after
139 processing, and are not processed or forwarded further.
140 This can be used to multiplex the input of two or more interfaces into a single
143 This is useful for reconstructing the traffic for network taps
144 that transmit the RX/TX signals out through two separate interfaces.
148 driver implements the Rapid Spanning Tree Protocol (RSTP or 802.1w) with
149 backwards compatibility with the legacy Spanning Tree Protocol (STP).
150 Spanning Tree is used to detect and remove loops in a network topology.
152 RSTP provides faster spanning tree convergence than legacy STP, the protocol
153 will exchange information with neighbouring switches to quickly transition to
154 forwarding without creating loops.
156 The code will default to RSTP mode but will downgrade any port connected to a
157 legacy STP network so is fully backward compatible.
158 A bridge can be forced to operate in STP mode without rapid state transitions
164 The bridge can log STP port changes to
167 .Va net.link.bridge.log_stp
171 Packet filtering can be used with any firewall package that hooks in via the
174 When filtering is enabled, bridged packets will pass through the filter
175 inbound on the originating interface, on the bridge interface and outbound on
176 the appropriate interfaces.
177 Either stage can be disabled.
178 The filtering behaviour can be controlled using
180 .Bl -tag -width ".Va net.link.bridge.pfil_onlyip"
181 .It Va net.link.bridge.pfil_onlyip
182 Controls the handling of non-IP packets which are not passed to
186 to only allow IP packets to pass (subject to firewall rules), set to
188 to unconditionally pass all non-IP Ethernet frames.
189 .It Va net.link.bridge.pfil_member
192 to enable filtering on the incoming and outgoing member interfaces, set
196 .It Va net.link.bridge.pfil_bridge
199 to enable filtering on the bridge interface, set
203 .It Va net.link.bridge.pfil_local_phys
206 to additionally filter on the physical interface for locally destined packets.
209 to disable this feature.
210 .It Va net.link.bridge.ipfw
213 to enable layer2 filtering with
218 This needs to be enabled for
227 will be disabled so that IPFW
228 is not run twice; these can be re-enabled if desired.
229 .It Va net.link.bridge.ipfw_arp
232 to enable layer2 ARP filtering with
242 ARP and REVARP packets are forwarded without being filtered and others
243 that are not IP nor IPv6 packets are not forwarded when
246 IPFW can filter Ethernet types using
248 so all packets are passed to
249 the filter for processing.
251 The packets originating from the bridging host will be seen by
252 the filter on the interface that is looked up in the routing
255 The packets destined to the bridging host will be seen by the filter
256 on the interface with the MAC address equal to the packet's destination
258 There are situations when some of the bridge members are sharing
259 the same MAC address (for example the
261 interfaces: they are currently sharing the
262 MAC address of the parent physical interface).
263 It is not possible to distinguish between these interfaces using
264 their MAC address, excluding the case when the packet's destination
265 MAC address is equal to the MAC address of the interface on which
266 the packet was entered to the system.
267 In this case the filter will see the incoming packet on this
269 In all other cases the interface seen by the packet filter is chosen
270 from the list of bridge members with the same MAC address and the
271 result strongly depends on the member addition sequence and the
272 actual implementation of
274 It is not recommended to rely on the order chosen by the current
276 implementation: it can be changed in the future.
278 The previous paragraph is best illustrated with the following
283 the MAC address of the incoming packet's destination is
284 .Nm nn:nn:nn:nn:nn:nn ,
286 the interface on which packet entered the system is
291 .Nm xx:xx:xx:xx:xx:xx ,
293 there are possibly other bridge members with the same MAC address
294 .Nm xx:xx:xx:xx:xx:xx ,
296 the bridge has more than one interface that are sharing the
298 .Nm yy:yy:yy:yy:yy:yy ;
305 Then if the MAC address
306 .Nm nn:nn:nn:nn:nn:nn
308 .Nm xx:xx:xx:xx:xx:xx
309 then the filter will see the packet on the interface
311 no matter if there are any other bridge members carrying the same
313 But if the MAC address
314 .Nm nn:nn:nn:nn:nn:nn
316 .Nm yy:yy:yy:yy:yy:yy
317 then the interface that will be seen by the filter is one of the
319 It is not possible to predict the name of the actual interface
320 without the knowledge of the system state and the
322 implementation details.
324 This problem arises for any bridge members that are sharing the same
325 MAC address, not only to the
327 ones: they we taken just as the example of such situation.
328 So if one wants the filter the locally destined packets based on
329 their interface name, one should be aware of this implication.
330 The described situation will appear at least on the filtering bridges
331 that are doing IP-forwarding; in some of such cases it is better
332 to assign the IP address only to the
334 interface and not to the bridge members.
336 .Va net.link.bridge.pfil_local_phys
337 will let you do the additional filtering on the physical interface.
339 The following when placed in the file
341 will cause a bridge called
343 to be created, and will add the interfaces
347 to the bridge, and then enable packet forwarding.
348 Such a configuration could be used to implement a simple
349 802.11-to-Ethernet bridge (assuming the 802.11 interface is
351 .Bd -literal -offset indent
352 cloned_interfaces="bridge0"
353 ifconfig_bridge0="addm wlan0 addm fxp0 up"
356 For the bridge to forward packets all member interfaces and the bridge need
358 The above example would also require:
359 .Bd -literal -offset indent
360 create_args_wlan0="wlanmode hostap"
361 ifconfig_wlan0="up ssid my_ap mode 11g"
365 Consider a system with two 4-port Ethernet boards.
366 The following will cause a bridge consisting of all 8 ports with Rapid Spanning
367 Tree enabled to be created:
368 .Bd -literal -offset indent
369 ifconfig bridge0 create
371 addm fxp0 stp fxp0 \e
372 addm fxp1 stp fxp1 \e
373 addm fxp2 stp fxp2 \e
374 addm fxp3 stp fxp3 \e
375 addm fxp4 stp fxp4 \e
376 addm fxp5 stp fxp5 \e
377 addm fxp6 stp fxp6 \e
378 addm fxp7 stp fxp7 \e
382 The bridge can be used as a regular host interface at the same time as bridging
383 between its member ports.
384 In this example, the bridge connects em0 and em1, and will receive its IP
385 address through DHCP:
386 .Bd -literal -offset indent
387 cloned_interfaces="bridge0"
388 ifconfig_bridge0="addm em0 addm em1 DHCP"
393 The bridge can tunnel Ethernet across an IP internet using the EtherIP
395 This can be combined with
397 to provide an encrypted connection.
400 interface and set the local and remote IP addresses for the
401 tunnel, these are reversed on the remote bridge.
402 .Bd -literal -offset indent
404 ifconfig gif0 tunnel 1.2.3.4 5.6.7.8 up
405 ifconfig bridge0 create
406 ifconfig bridge0 addm fxp0 addm gif0 up
411 6.1, 6.2, 6.3, 7.0, 7.1, and 7.2 have a bug in the EtherIP protocol.
412 For more details and workaround, see
424 driver first appeared in
430 driver was originally written by
432 .Aq jason@thought.net
433 as part of an undergraduate independent study at the University of
434 North Carolina at Greensboro.
438 driver has been heavily modified from the original version by
440 .Aq thorpej@wasabisystems.com .
442 Rapid Spanning Tree Protocol (RSTP) support was added by
444 .Aq thompsa@FreeBSD.org .
448 driver currently supports only Ethernet and Ethernet-like (e.g., 802.11)
449 network devices, with exactly the same interface MTU size as the bridge device.