xref: /freebsd/share/man/man4/tap.4 (revision 38f0b757fd84d17d0fc24739a7cda160c4516d81)
1.\" $FreeBSD$
2.\" Based on PR#2411
3.\"
4.Dd January 26, 2012
5.Dt TAP 4
6.Os
7.Sh NAME
8.Nm tap
9.Nd Ethernet tunnel software network interface
10.Sh SYNOPSIS
11.Cd device tap
12.Sh DESCRIPTION
13The
14.Nm
15interface is a software loopback mechanism that can be loosely
16described as the network interface analog of the
17.Xr pty 4 ,
18that is,
19.Nm
20does for network interfaces what the
21.Nm pty
22driver does for terminals.
23.Pp
24The
25.Nm
26driver, like the
27.Nm pty
28driver, provides two interfaces: an interface like the usual facility
29it is simulating
30(an Ethernet network interface in the case of
31.Nm ,
32or a terminal for
33.Nm pty ) ,
34and a character-special device
35.Dq control
36interface.
37.Pp
38The network interfaces are named
39.Dq Li tap0 ,
40.Dq Li tap1 ,
41etc., one for each control device that has been opened.
42These Ethernet network interfaces persist until
43.Pa if_tap.ko
44module is unloaded, or until removed with "ifconfig destroy" (see below).
45.Pp
46.Nm
47devices are created using interface cloning.
48This is done using the
49.Dq ifconfig tap Ns Sy N No create
50command.
51This is the preferred method of creating
52.Nm
53devices.
54The same method allows removal of interfaces.
55For this, use the
56.Dq ifconfig tap Ns Sy N No destroy
57command.
58.Pp
59If the
60.Xr sysctl 8
61variable
62.Va net.link.tap.devfs_cloning
63is non-zero, the
64.Nm
65interface
66permits opens on the special control device
67.Pa /dev/tap .
68When this device is opened,
69.Nm
70will return a handle for the lowest unused
71.Nm
72device (use
73.Xr devname 3
74to determine which).
75.Pp
76.Bf Em
77Disabling the legacy devfs cloning functionality may break existing
78applications which use
79.Nm ,
80such as
81.Tn VMware
82and
83.Xr ssh 1 .
84It therefore defaults to being enabled until further notice.
85.Ef
86.Pp
87Control devices (once successfully opened) persist until
88.Pa if_tap.ko
89is unloaded or the interface is destroyed.
90.Pp
91Each interface supports the usual Ethernet network interface
92.Xr ioctl 2 Ns s
93and thus can be used with
94.Xr ifconfig 8
95like any other Ethernet interface.
96When the system chooses to transmit
97an Ethernet frame on the network interface, the frame can be read from
98the control device
99(it appears as
100.Dq input
101there);
102writing an Ethernet frame to the control device generates an input frame on
103the network interface, as if the
104(non-existent)
105hardware had just received it.
106.Pp
107The Ethernet tunnel device, normally
108.Pa /dev/tap Ns Sy N ,
109is exclusive-open
110(it cannot be opened if it is already open)
111and is restricted to the super-user, unless the
112.Xr sysctl 8
113variable
114.Va net.link.tap.user_open
115is non-zero.
116If the
117.Xr sysctl 8
118variable
119.Va net.link.tap.up_on_open
120is non-zero, the tunnel device will be marked
121.Dq up
122when the control device is opened.
123A
124.Fn read
125call will return an error
126.Pq Er EHOSTDOWN
127if the interface is not
128.Dq ready .
129Once the interface is ready,
130.Fn read
131will return an Ethernet frame if one is available; if not, it will
132either block until one is or return
133.Er EWOULDBLOCK ,
134depending on whether non-blocking I/O has been enabled.
135If the frame
136is longer than is allowed for in the buffer passed to
137.Fn read ,
138the extra data will be silently dropped.
139.Pp
140A
141.Xr write 2
142call passes an Ethernet frame in to be
143.Dq received
144on the pseudo-interface.
145Each
146.Fn write
147call supplies exactly one frame; the frame length is taken from the
148amount of data provided to
149.Fn write .
150Writes will not block; if the frame cannot be accepted
151for a transient reason
152(e.g., no buffer space available),
153it is silently dropped; if the reason is not transient
154(e.g., frame too large),
155an error is returned.
156The following
157.Xr ioctl 2
158calls are supported
159(defined in
160.In net/if_tap.h ) :
161.Bl -tag -width VMIO_SIOCSETMACADDR
162.It Dv TAPSIFINFO
163Set network interface information (line speed, MTU and type).
164The argument should be a pointer to a
165.Va struct tapinfo .
166.It Dv TAPGIFINFO
167Retrieve network interface information (line speed, MTU and type).
168The argument should be a pointer to a
169.Va struct tapinfo .
170.It Dv TAPSDEBUG
171The argument should be a pointer to an
172.Va int ;
173this sets the internal debugging variable to that value.
174What, if
175anything, this variable controls is not documented here; see the source
176code.
177.It Dv TAPGDEBUG
178The argument should be a pointer to an
179.Va int ;
180this stores the internal debugging variable's value into it.
181.It Dv TAPGIFNAME
182Retrieve network interface name.
183The argument should be a pointer to a
184.Va struct ifreq .
185The interface name will be returned in the
186.Va ifr_name
187field.
188.It Dv FIONBIO
189Turn non-blocking I/O for reads off or on, according as the argument
190.Va int Ns 's
191value is or is not zero
192(Writes are always nonblocking).
193.It Dv FIOASYNC
194Turn asynchronous I/O for reads
195(i.e., generation of
196.Dv SIGIO
197when data is available to be read)
198off or on, according as the argument
199.Va int Ns 's
200value is or is not zero.
201.It Dv FIONREAD
202If any frames are queued to be read, store the size of the first one into the argument
203.Va int ;
204otherwise, store zero.
205.It Dv TIOCSPGRP
206Set the process group to receive
207.Dv SIGIO
208signals, when asynchronous I/O is enabled, to the argument
209.Va int
210value.
211.It Dv TIOCGPGRP
212Retrieve the process group value for
213.Dv SIGIO
214signals into the argument
215.Va int
216value.
217.It Dv SIOCGIFADDR
218Retrieve the Media Access Control
219.Pq Dv MAC
220address of the
221.Dq remote
222side.
223This command is used by the VMware port and expected to be executed on
224descriptor, associated with control device
225(usually
226.Pa /dev/vmnet Ns Sy N
227or
228.Pa /dev/tap Ns Sy N ) .
229The
230.Va buffer ,
231which is passed as the argument, is expected to have enough space to store
232the
233.Dv MAC
234address.
235At the open time both
236.Dq local
237and
238.Dq remote
239.Dv MAC
240addresses are the same, so this command could be used to retrieve the
241.Dq local
242.Dv MAC
243address.
244.It Dv SIOCSIFADDR
245Set the Media Access Control
246.Pq Dv MAC
247address of the
248.Dq remote
249side.
250This command is used by VMware port and expected to be executed on
251a descriptor, associated with control device
252(usually
253.Pa /dev/vmnet Ns Sy N ) .
254.El
255.Pp
256The control device also supports
257.Xr select 2
258for read; selecting for write is pointless, and always succeeds, since
259writes are always non-blocking.
260.Pp
261On the last close of the data device, the interface is
262brought down
263(as if with
264.Dq ifconfig tap Ns Sy N No down )
265unless the device is a
266.Em VMnet
267device.
268All queued frames are thrown away.
269If the interface is up when the data
270device is not open, output frames are thrown away rather than
271letting them pile up.
272.Pp
273The
274.Nm
275device can also be used with the VMware port as a replacement
276for the old
277.Em VMnet
278device driver.
279The driver uses the minor number
280to select between
281.Nm
282and
283.Nm vmnet
284devices.
285.Em VMnet
286minor numbers begin at
287.Va 0x800000
288+
289.Va N ;
290where
291.Va N
292is a
293.Em VMnet
294unit number.
295In this case the control device is expected to be
296.Pa /dev/vmnet Ns Sy N ,
297and the network interface will be
298.Sy vmnet Ns Ar N .
299Additionally,
300.Em VMnet
301devices do not
302.Xr ifconfig 8
303themselves down when the
304control device is closed.
305Everything else is the same.
306.Pp
307In addition to the above mentioned
308.Xr ioctl 2
309calls, there is an additional one for the VMware port.
310.Bl -tag -width VMIO_SIOCSETMACADDR
311.It Dv VMIO_SIOCSIFFLAGS
312VMware
313.Dv SIOCSIFFLAGS .
314.El
315.Sh SEE ALSO
316.Xr inet 4 ,
317.Xr intro 4
318