xref: /freebsd/share/man/man4/ng_bpf.4 (revision dd41de95a84d979615a2ef11df6850622bf6184e)
1.\" Copyright (c) 1999 Whistle Communications, Inc.
2.\" All rights reserved.
3.\"
4.\" Subject to the following obligations and disclaimer of warranty, use and
5.\" redistribution of this software, in source or object code forms, with or
6.\" without modifications are expressly permitted by Whistle Communications;
7.\" provided, however, that:
8.\" 1. Any and all reproductions of the source or object code must include the
9.\"    copyright notice above and the following disclaimer of warranties; and
10.\" 2. No rights are granted, in any manner or form, to use Whistle
11.\"    Communications, Inc. trademarks, including the mark "WHISTLE
12.\"    COMMUNICATIONS" on advertising, endorsements, or otherwise except as
13.\"    such appears in the above copyright notice or in the software.
14.\"
15.\" THIS SOFTWARE IS BEING PROVIDED BY WHISTLE COMMUNICATIONS "AS IS", AND
16.\" TO THE MAXIMUM EXTENT PERMITTED BY LAW, WHISTLE COMMUNICATIONS MAKES NO
17.\" REPRESENTATIONS OR WARRANTIES, EXPRESS OR IMPLIED, REGARDING THIS SOFTWARE,
18.\" INCLUDING WITHOUT LIMITATION, ANY AND ALL IMPLIED WARRANTIES OF
19.\" MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, OR NON-INFRINGEMENT.
20.\" WHISTLE COMMUNICATIONS DOES NOT WARRANT, GUARANTEE, OR MAKE ANY
21.\" REPRESENTATIONS REGARDING THE USE OF, OR THE RESULTS OF THE USE OF THIS
22.\" SOFTWARE IN TERMS OF ITS CORRECTNESS, ACCURACY, RELIABILITY OR OTHERWISE.
23.\" IN NO EVENT SHALL WHISTLE COMMUNICATIONS BE LIABLE FOR ANY DAMAGES
24.\" RESULTING FROM OR ARISING OUT OF ANY USE OF THIS SOFTWARE, INCLUDING
25.\" WITHOUT LIMITATION, ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY,
26.\" PUNITIVE, OR CONSEQUENTIAL DAMAGES, PROCUREMENT OF SUBSTITUTE GOODS OR
27.\" SERVICES, LOSS OF USE, DATA OR PROFITS, HOWEVER CAUSED AND UNDER ANY
28.\" THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
29.\" (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
30.\" THIS SOFTWARE, EVEN IF WHISTLE COMMUNICATIONS IS ADVISED OF THE POSSIBILITY
31.\" OF SUCH DAMAGE.
32.\"
33.\" Author: Archie Cobbs <archie@FreeBSD.org>
34.\"
35.\" $FreeBSD$
36.\" $Whistle: ng_bpf.8,v 1.2 1999/12/03 01:57:12 archie Exp $
37.\"
38.Dd September 20, 2020
39.Dt NG_BPF 4
40.Os
41.Sh NAME
42.Nm ng_bpf
43.Nd Berkeley packet filter netgraph node type
44.Sh SYNOPSIS
45.In sys/types.h
46.In net/bpf.h
47.In netgraph.h
48.In netgraph/ng_bpf.h
49.Sh DESCRIPTION
50The
51.Nm bpf
52node type allows Berkeley Packet Filter (see
53.Xr bpf 4 )
54filters to be applied to data travelling through a Netgraph network.
55Each node allows an arbitrary number of connections to arbitrarily
56named hooks.
57With each hook is associated a
58.Xr bpf 4
59filter program which is applied to incoming data only, a destination hook
60for matching packets, a destination hook for non-matching packets,
61and various statistics counters.
62.Pp
63A
64.Xr bpf 4
65program returns an unsigned integer, which is normally interpreted as
66the length of the prefix of the packet to return.
67In the context of this
68node type, returning zero is considered a non-match, in which case the
69entire packet is delivered out the non-match destination hook.
70Returning a value greater than zero causes the packet to be truncated
71to that length and delivered out the match destination hook.
72Either or both destination hooks may be the empty string, or may
73not exist, in which case the packet is dropped.
74.Pp
75New hooks are initially configured to drop all packets.
76A new filter program may be installed using the
77.Dv NGM_BPF_SET_PROGRAM
78control message.
79.Sh HOOKS
80This node type supports any number of hooks having arbitrary names.
81.Sh CONTROL MESSAGES
82This node type supports the generic control messages, plus the following:
83.Bl -tag -width foo
84.It Dv NGM_BPF_SET_PROGRAM Pq Ic setprogram
85This command sets the filter program that will be applied to incoming
86data on a hook.
87The following structure must be supplied as an argument:
88.Bd -literal -offset 4n
89struct ng_bpf_hookprog {
90  char            thisHook[NG_HOOKSIZ];     /* name of hook */
91  char            ifMatch[NG_HOOKSIZ];      /* match dest hook */
92  char            ifNotMatch[NG_HOOKSIZ];   /* !match dest hook */
93  int32_t         bpf_prog_len;             /* #insns in program */
94  struct bpf_insn bpf_prog[];               /* bpf program */
95};
96.Ed
97.Pp
98The hook to be updated is specified in
99.Dv thisHook .
100The BPF program is the sequence of instructions in the
101.Dv bpf_prog
102array; there must be
103.Dv bpf_prog_len
104of them.
105Matching and non-matching incoming packets are delivered out the hooks named
106.Dv ifMatch
107and
108.Dv ifNotMatch ,
109respectively.
110The program must be a valid
111.Xr bpf 4
112program or else
113.Er EINVAL
114is returned.
115.It Dv NGM_BPF_GET_PROGRAM Pq Ic getprogram
116This command takes an ASCII
117string argument, the hook name, and returns the
118corresponding
119.Dv "struct ng_bpf_hookprog"
120as shown above.
121.It Dv NGM_BPF_GET_STATS Pq Ic getstats
122This command takes an ASCII
123string argument, the hook name, and returns the
124statistics associated with the hook as a
125.Dv "struct ng_bpf_hookstat" .
126.It Dv NGM_BPF_CLR_STATS Pq Ic clrstats
127This command takes an ASCII
128string argument, the hook name, and clears the
129statistics associated with the hook.
130.It Dv NGM_BPF_GETCLR_STATS Pq Ic getclrstats
131This command is identical to
132.Dv NGM_BPF_GET_STATS ,
133except that the statistics are also atomically cleared.
134.El
135.Sh SHUTDOWN
136This node shuts down upon receipt of a
137.Dv NGM_SHUTDOWN
138control message, or when all hooks have been disconnected.
139.Sh EXAMPLES
140It is possible to configure a node from the command line, using
141.Xr tcpdump 1
142to generate raw BPF instructions which are then transformed
143into the ASCII form of a
144.Dv NGM_BPF_SET_PROGRAM
145control message, as demonstrated here:
146.Bd -literal -offset 4n
147#!/bin/sh
148
149PATTERN="tcp dst port 80"
150NODEPATH="my_node:"
151INHOOK="hook1"
152MATCHHOOK="hook2"
153NOTMATCHHOOK="hook3"
154
155BPFPROG=$( tcpdump -s 8192 -p -ddd ${PATTERN} | \\
156           ( read len ; \\
157             echo -n "bpf_prog_len=$len " ; \\
158             echo -n "bpf_prog=[" ; \\
159             while read code jt jf k ; do \\
160                 echo -n " { code=$code jt=$jt jf=$jf k=$k }" ; \\
161             done ; \\
162             echo " ]" ) )
163
164ngctl msg ${NODEPATH} setprogram { thisHook=\\"${INHOOK}\\" \\
165  ifMatch=\\"${MATCHHOOK}\\" \\
166  ifNotMatch=\\"${NOTMATCHHOOK}\\" \\
167  ${BPFPROG} }
168.Ed
169.Pp
170Based on the previous example, it is possible to prevent a jail (or a VM)
171from spoofing by allowing only traffic that has the expected ethernet and
172IP addresses:
173.Bd -literal -offset 4n
174#!/bin/sh
175
176NODEPATH="my_node:"
177JAIL_MAC="0a:00:de:ad:be:ef"
178JAIL_IP="128.66.1.42"
179JAIL_HOOK="jail"
180HOST_HOOK="host"
181DEBUG_HOOK="nomatch"
182
183bpf_prog() {
184    local PATTERN=$1
185
186    tcpdump -s 8192 -p -ddd ${PATTERN} | (
187        read len
188        echo -n "bpf_prog_len=$len "
189        echo -n "bpf_prog=["
190        while read code jt jf k ; do
191            echo -n " { code=$code jt=$jt jf=$jf k=$k }"
192        done
193        echo " ]"
194    )
195}
196
197# Prevent jail from spoofing (filter packets coming from jail)
198ngctl msg ${NODEPATH} setprogram {                        \\
199    thisHook=\\"${JAIL_HOOK}\\"                             \\
200    ifMatch=\\"${HOST_HOOK}\\"                              \\
201    ifNotMatch=\\"${DEBUG_HOOK}\\"                          \\
202    $(bpf_prog "ether src ${JAIL_MAC} && src ${JAIL_IP}") \\
203}
204
205# Prevent jail from receiving spoofed packets (filter packets
206# coming from host)
207ngctl msg ${NODEPATH} setprogram {                        \\
208    thisHook=\\"${HOST_HOOK}\\"                             \\
209    ifMatch=\\"${JAIL_HOOK}\\"                              \\
210    ifNotMatch=\\"${DEBUG_HOOK}\\"                          \\
211    $(bpf_prog "ether dst ${JAIL_MAC} && dst ${JAIL_IP}") \\
212}
213.Ed
214.Sh SEE ALSO
215.Xr bpf 4 ,
216.Xr netgraph 4 ,
217.Xr ngctl 8
218.Sh HISTORY
219The
220.Nm
221node type was implemented in
222.Fx 4.0 .
223.Sh AUTHORS
224.An Archie Cobbs Aq Mt archie@FreeBSD.org
225.Sh BUGS
226When built as a loadable kernel module, this module includes the file
227.Pa net/bpf_filter.c .
228Although loading the module should fail if
229.Pa net/bpf_filter.c
230already exists in the kernel, currently it does not, and the duplicate
231copies of the file do not interfere.
232However, this may change in the future.
233