xref: /386bsd/usr/src/usr.sbin/tcpdump/bpf/README (revision a2142627)
1Mon Jun 24 16:42:44 PDT 1991
2
3This directory contains the files necessary to install the
4Berkeley Packet Filter (BPF) in a BSD (or BSD-like) kernel.
5BPF is derived from the Stanford/CMU enet packet filter that was
6distributed with 4.3BSD release.  We have made no efforts to keep
7the two interfaces compatible.
8
9BPF has been tested on hp300's running BSD Tahoe, on Sparcstations
10running SunOS 4.1, and on Sun 3's running SunOS 3.5.  We have configured
11it into the BSD Lance ethernet driver, the Sun LANCE and Intel drivers,
12and our (soon to be released) SLIP driver (SunOS 3.5 and 4.1).  The modified
13BSD driver is included in this distribution, however, the Sun drivers cannot
14be made available.  There is a context diff in bpf/sunif/if_le.c-sunos4.1-diff
15for the SunOS 4.1 LANCE driver (if_le.c) [so you need SunOS source].
16
17Here's what you need to do:
18
19(1)	Add the following line to your config file.  The parameter
20	is an upper bound for two things: the number of simultaneuous open
21	files, and the number of hardware interfaces attached to BPF.
22
23pseudo-device   bpfilter 16
24
25	Add these lines to conf/files:
26
27net/bpf.c		optional bpfilter
28net/bpf_filter.c	optional bpfilter
29
30(2)	Copy these files into /sys/net:
31
32bpf/net/bpf.c
33bpf/net/bpf_filter.c
34bpf/net/bpf.h
35bpf/net/bpfcodes.h
36bpf/net/bpfdesc.h
37
38(3)	Install bpf.h and bpfcodes.h in /usr/include/net.
39
40(4)	Add an entry for BPF in `cdevsw'.  You need to add these
41	routines: bpfopen(), bpfclose(), bpfread(), bpfwrite(),
42	bpfioctl(), and bpfselect().
43
44	Create the special device files /dev/bpf0, /dev/bpf1, etc.
45	Make sure the major device number correpsonds to the entry in
46	cdevsw; the minor device number should be the same as the
47	trailing digit of the file name.
48
49	Access to the packet interface is controlled by the permissions
50	on the device files.  We recommend that access be restricted to
51	group `wheel'.  For example,
52
53		/etc/mknod /dev/bpf0 c {major dev} 0
54		/etc/mknod /dev/bpf1 c {major dev} 1
55		/etc/mknod /dev/bpf2 c {major dev} 2
56		...
57		chmod 440 /dev/bpf*
58		chgrp wheel /dev/bpf*
59
60	The highest allowable minor device number corresponds to the
61	number given in the "pseudo-device" config line (less one).
62
63(5)	Modify the link level device drivers to interact with BPF.
64	hpdev/if_le.c is an example driver for a LANCE Ethernet
65	interface on an hp300 series machine.  [If this is your
66	setup, go to (6).]
67
68	If you're starting from scratch, this is not too difficult.
69	All the BPF mods to hpdev/if_le.c are encapsulated with
70	`#ifdef NBPFILTER > 0', so they're easy to spot.  You need to:
71
72	a) Add includes for bpfilter.h and ../net/bpf.h.
73
74	b) Add a caddr_t to the softc.  This is the magic cookie
75	   that tells bpf_tap() who is talking to it.
76
77	c) Modify the attach routine to set up some device parameters
78	   [see hpdev/if_le.c:leattach()] and call bpfattach().
79
80	d) Make sure the driver can handle promiscuous operation,
81	   and that the routine ifpromisc() exists.  Ifpromisc()
82	   takes an ifp and a flag saying whether to enter or leave
83	   promiscuous operation.  It should reference count the
84	   calls and take actions only the last `off' or first `on'.
85	   The action it should take is setting/clearing the IFF_PROMISC
86	   bit, and calling the driver's SIOCIFFLAGS ioctl.  The
87	   driver should inspect the IFF_PROMISC bit and do the right
88	   thing.
89
90	d) Add calls to bpf_tap() at the following places:
91
92		i.  Right after the device interrupts and the packet is
93		    in contiguous interface memory.  This is before
94		    the packet has been copied in to mbufs.
95
96		ii. Right before the packet is transmitted.
97		    This is after the packet has been copied out of mbufs.
98
99		If the packet never exists in contiguous memory
100		(some interfaces can follow chains), you need to
101		call bpf_mtap instead.
102
103		To minimize the cost of the filter when there are no
104		listeners, bpf_tap() is only called when the magic
105		cookie in the driver's softc is nonzero.  (BPF will
106		set and clear it.)
107
108		Because BPF can force an interface into promiscuous mode,
109		you want to check that incoming packets are destined for
110		this host or are broadcast/multicast.  If neither is the
111		case, the packet should be tossed (after calling bpf_tap()).
112		This check only needs to be done when there are listeners.
113
114(6)	BPF calls the routine ifpromisc() to put an interface into
115	promiscuous mode.  SunOS 4.1 provides this routine; we have
116	provided our versions for SunOS 3.5 and BSD in the files
117	bpf/net/if-sunos3.c and bpf/net/if-bsd.c.  Merge this code
118	into net/if.c.	Additionally, the `if_pcount' integer field
119	must be added to the `struct ifnet' in net/if.h.
120
121(7)	That's it.  Run config, make depend, and make, and you're ready to go.
122
123
124All the code in these directories is subject to the standard Berkeley
125network software copyright:
126
127  Copyright (c) 1990 The Regents of the University of California.
128  All rights reserved.
129
130  Redistribution and use in source and binary forms, with or without
131  modification, are permitted provided that: (1) source code distributions
132  retain the above copyright notice and this paragraph in its entirety, (2)
133  distributions including binary code include the above copyright notice and
134  this paragraph in its entirety in the documentation or other materials
135  provided with the distribution, and (3) all advertising materials mentioning
136  features or use of this software display the following acknowledgement:
137  ``This product includes software developed by the University of California,
138  Lawrence Berkeley Laboratory and its contributors.'' Neither the name of
139  the University nor the names of its contributors may be used to endorse
140  or promote products derived from this software without specific prior
141  written permission.
142  THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED
143  WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF
144  MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.
145
146
147
148