1 .\" Copyright (c) 1990 The Regents of the University of California.
2 .\" All rights reserved.
4 .\" Redistribution and use in source and binary forms, with or without
5 .\" modification, are permitted provided that: (1) source code distributions
6 .\" retain the above copyright notice and this paragraph in its entirety, (2)
7 .\" distributions including binary code include the above copyright notice and
8 .\" this paragraph in its entirety in the documentation or other materials
9 .\" provided with the distribution, and (3) all advertising materials mentioning
10 .\" features or use of this software display the following acknowledgement:
11 .\" ``This product includes software developed by the University of California,
12 .\" Lawrence Berkeley Laboratory and its contributors.'' Neither the name of
13 .\" the University nor the names of its contributors may be used to endorse
14 .\" or promote products derived from this software without specific prior
15 .\" written permission.
16 .\" THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED
17 .\" WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF
18 .\" MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.
20 .\" This document is derived in part from the enet man page (enet.4)
21 .\" distributed with 4.3BSD Unix.
30 .Nd Berkeley Packet Filter
34 The Berkeley Packet Filter
35 provides a raw interface to data link layers in a protocol
37 All packets on the network, even those destined for other hosts,
38 are accessible through this mechanism.
40 The packet filter appears as a character special device,
44 After opening the device, the file descriptor must be bound to a
45 specific network interface with the
48 A given interface can be shared by multiple listeners, and the filter
49 underlying each descriptor will see an identical packet stream.
51 A separate device file is required for each minor device.
52 If a file is in use, the open will fail and
57 Associated with each open instance of a
59 file is a user-settable packet filter.
60 Whenever a packet is received by an interface,
61 all file descriptors listening on that interface apply their filter.
62 Each descriptor that accepts the packet receives its own copy.
64 Reads from these files return the next group of packets
65 that have matched the filter.
66 To improve performance, the buffer passed to read must be
67 the same size as the buffers used internally by
69 This size is returned by the
71 ioctl (see below), and
74 Note that an individual packet larger than this size is necessarily
77 The packet filter will support any link level protocol that has fixed length
79 Currently, only Ethernet,
83 drivers have been modified to interact with
86 Since packet data is in network byte order, applications should use the
88 macros to extract multi-byte values.
90 A packet can be sent out on the network by writing to a
93 The writes are unbuffered, meaning only one packet can be processed per write.
94 Currently, only writes to Ethernets and
100 command codes below are defined in
105 #include <sys/types.h>
106 #include <sys/time.h>
107 #include <sys/ioctl.h>
124 the following commands may be applied to any open
127 The (third) argument to
129 should be a pointer to the type indicated.
130 .Bl -tag -width BIOCGRTIMEOUT
133 Returns the required buffer length for reads on
138 Sets the buffer length for reads on
141 The buffer must be set before the file is attached to an interface
144 If the requested buffer size cannot be accommodated, the closest
145 allowable size will be set and returned in the argument.
146 A read call will result in
148 if it is passed a buffer that is not this size.
151 Returns the type of the data link layer underlying the attached interface.
153 is returned if no interface has been specified.
154 The device types, prefixed with
159 Forces the interface into promiscuous mode.
160 All packets, not just those destined for the local host, are processed.
161 Since more than one file can be listening on a given interface,
162 a listener that opened its interface non-promiscuously may receive
163 packets promiscuously.
164 This problem can be remedied with an appropriate filter.
166 Flushes the buffer of incoming packets,
167 and resets the statistics that are returned by BIOCGSTATS.
169 .Pq Li "struct ifreq"
170 Returns the name of the hardware interface that the file is listening on.
171 The name is returned in the ifr_name field of
175 All other fields are undefined.
177 .Pq Li "struct ifreq"
178 Sets the hardware interface associate with the file.
180 command must be performed before any packets can be read.
181 The device is indicated by name using the
186 Additionally, performs the actions of
190 .Pq Li "struct timeval"
191 Set or get the read timeout parameter.
193 specifies the length of time to wait before timing
194 out on a read request.
195 This parameter is initialized to zero by
197 indicating no timeout.
199 .Pq Li "struct bpf_stat"
200 Returns the following structure of packet statistics:
203 u_int bs_recv; /* number of packets received */
204 u_int bs_drop; /* number of packets dropped */
209 .Bl -hang -offset indent
211 the number of packets received by the descriptor since opened or reset
212 (including any buffered since the last read call);
215 the number of packets which were accepted by the filter but dropped by the
216 kernel because of buffer overflows
217 (i.e., the application's reads are not keeping up with the packet traffic).
223 based on the truth value of the argument.
224 When immediate mode is enabled, reads return immediately upon packet
226 Otherwise, a read will block until either the kernel buffer
227 becomes full or a timeout occurs.
228 This is useful for programs like
230 which must respond to messages in real time.
231 The default for a new file is off.
233 .Pq Li "struct bpf_program"
234 Sets the filter program used by the kernel to discard uninteresting
236 An array of instructions and its length is passed in using
237 the following structure:
241 struct bpf_insn *bf_insns;
245 The filter program is pointed to by the
247 field while its length in units of
248 .Sq Li struct bpf_insn
257 for an explanation of the filter language.
259 .Pq Li "struct bpf_version"
260 Returns the major and minor version numbers of the filter language currently
261 recognized by the kernel.
262 Before installing a filter, applications must check
263 that the current version is compatible with the running kernel.
264 Version numbers are compatible if the major numbers match and the application minor
265 is less than or equal to the kernel minor.
266 The kernel version number is returned in the following structure:
274 The current version numbers are given by
275 .Dv BPF_MAJOR_VERSION
277 .Dv BPF_MINOR_VERSION
280 An incompatible filter
281 may result in undefined behavior (most likely, an error returned by
283 or haphazard packet matching).
287 Set or get the status of the
290 Set to zero if the link level source address should be filled in automatically
291 by the interface output routine.
292 Set to one if the link level source
293 address will be written, as provided, to the wire.
294 This flag is initialized to zero by default.
298 Set or get the flag determining whether locally generated packets on the
299 interface should be returned by BPF.
300 Set to zero to see only incoming packets on the interface.
301 Set to one to see packets originating locally and remotely on the interface.
302 This flag is initialized to one by
306 The following structure is prepended to each packet returned by
310 struct timeval bh_tstamp; /* time stamp */
311 u_long bh_caplen; /* length of captured portion */
312 u_long bh_datalen; /* original length of packet */
313 u_short bh_hdrlen; /* length of bpf header (this struct
314 plus alignment padding */
318 The fields, whose values are stored in host order, and are:
320 .Bl -tag -compact -width bh_datalen
322 The time at which the packet was processed by the packet filter.
324 The length of the captured portion of the packet.
325 This is the minimum of
326 the truncation amount specified by the filter and the length of the packet.
328 The length of the packet off the wire.
329 This value is independent of the truncation amount specified by the filter.
333 header, which may not be equal to
334 .\" XXX - not really a function call
335 .Fn sizeof "struct bpf_hdr" .
340 field exists to account for
341 padding between the header and the link level protocol.
342 The purpose here is to guarantee proper alignment of the packet
343 data structures, which is required on alignment sensitive
344 architectures and improves performance on many other architectures.
345 The packet filter insures that the
347 and the network layer
348 header will be word aligned.
350 must be taken when accessing the link layer protocol fields on alignment
352 (This is not a problem on an Ethernet, since
353 the type field is a short falling on an even offset,
354 and the addresses are probably accessed in a bytewise fashion).
356 Additionally, individual packets are padded so that each starts
358 This requires that an application
359 has some knowledge of how to get from packet to packet.
366 It rounds up its argument to the nearest word aligned value (where a word is
372 points to the start of a packet, this expression
373 will advance it to the next packet:
374 .Dl p = (char *)p + BPF_WORDALIGN(p->bh_hdrlen + p->bh_caplen)
376 For the alignment mechanisms to work properly, the
379 must itself be word aligned.
383 will always return an aligned buffer.
385 A filter program is an array of instructions, with all branches forwardly
386 directed, terminated by a
389 Each instruction performs some action on the pseudo-machine state,
390 which consists of an accumulator, index register, scratch memory store,
391 and implicit program counter.
393 The following structure defines the instruction format:
405 field is used in different ways by different instructions,
410 fields are used as offsets
411 by the branch instructions.
412 The opcodes are encoded in a semi-hierarchical fashion.
413 There are eight classes of instructions:
423 Various other mode and
424 operator bits are or'd into the class to give the actual instructions.
425 The classes and modes are defined in
428 Below are the semantics for each defined
431 We use the convention that A is the accumulator, X is the index register,
432 P[] packet data, and M[] scratch memory store.
433 P[i:n] gives the data at byte offset
436 interpreted as a word (n=4),
437 unsigned halfword (n=2), or unsigned byte (n=1).
438 M[i] gives the i'th word in the scratch memory store, which is only
439 addressed in word units.
440 The memory store is indexed from 0 to
447 are the corresponding fields in the
448 instruction definition.
450 refers to the length of the packet.
452 .Bl -tag -width BPF_STXx
454 These instructions copy a value into the accumulator.
455 The type of the source operand is specified by an
457 and can be a constant
459 packet data at a fixed offset
461 packet data at a variable offset
465 or a word in the scratch memory store
471 the data size must be specified as a word
477 The semantics of all the recognized
482 BPF_LD+BPF_W+BPF_ABS A <- P[k:4]
483 BPF_LD+BPF_H+BPF_ABS A <- P[k:2]
484 BPF_LD+BPF_B+BPF_ABS A <- P[k:1]
485 BPF_LD+BPF_W+BPF_IND A <- P[X+k:4]
486 BPF_LD+BPF_H+BPF_IND A <- P[X+k:2]
487 BPF_LD+BPF_B+BPF_IND A <- P[X+k:1]
488 BPF_LD+BPF_W+BPF_LEN A <- len
489 BPF_LD+BPF_IMM A <- k
490 BPF_LD+BPF_MEM A <- M[k]
493 These instructions load a value into the index register.
495 the addressing modes are more restrictive than those of the accumulator loads,
498 a hack for efficiently loading the IP header length.
501 BPF_LDX+BPF_W+BPF_IMM X <- k
502 BPF_LDX+BPF_W+BPF_MEM X <- M[k]
503 BPF_LDX+BPF_W+BPF_LEN X <- len
504 BPF_LDX+BPF_B+BPF_MSH X <- 4*(P[k:1]&0xf)
507 This instruction stores the accumulator into the scratch memory.
508 We do not need an addressing mode since there is only one possibility
515 This instruction stores the index register in the scratch memory store.
521 The alu instructions perform operations between the accumulator and
522 index register or constant, and store the result back in the accumulator.
523 For binary operations, a source mode is required
529 BPF_ALU+BPF_ADD+BPF_K A <- A + k
530 BPF_ALU+BPF_SUB+BPF_K A <- A - k
531 BPF_ALU+BPF_MUL+BPF_K A <- A * k
532 BPF_ALU+BPF_DIV+BPF_K A <- A / k
533 BPF_ALU+BPF_AND+BPF_K A <- A & k
534 BPF_ALU+BPF_OR+BPF_K A <- A | k
535 BPF_ALU+BPF_LSH+BPF_K A <- A << k
536 BPF_ALU+BPF_RSH+BPF_K A <- A >> k
537 BPF_ALU+BPF_ADD+BPF_X A <- A + X
538 BPF_ALU+BPF_SUB+BPF_X A <- A - X
539 BPF_ALU+BPF_MUL+BPF_X A <- A * X
540 BPF_ALU+BPF_DIV+BPF_X A <- A / X
541 BPF_ALU+BPF_AND+BPF_X A <- A & X
542 BPF_ALU+BPF_OR+BPF_X A <- A | X
543 BPF_ALU+BPF_LSH+BPF_X A <- A << X
544 BPF_ALU+BPF_RSH+BPF_X A <- A >> X
545 BPF_ALU+BPF_NEG A <- -A
548 The jump instructions alter flow of control.
550 compare the accumulator against a constant
552 or the index register
554 If the result is true (or non-zero),
555 the true branch is taken, otherwise the false branch is taken.
556 Jump offsets are encoded in 8 bits so the longest jump is 256 instructions.
557 However, the jump always
559 opcode uses the 32 bit
561 field as the offset, allowing arbitrarily distant destinations.
562 All conditionals use unsigned comparison conventions.
565 BPF_JMP+BPF_JA pc += k
566 BPF_JMP+BPF_JGT+BPF_K pc += (A > k) ? jt : jf
567 BPF_JMP+BPF_JGE+BPF_K pc += (A >= k) ? jt : jf
568 BPF_JMP+BPF_JEQ+BPF_K pc += (A == k) ? jt : jf
569 BPF_JMP+BPF_JSET+BPF_K pc += (A & k) ? jt : jf
570 BPF_JMP+BPF_JGT+BPF_X pc += (A > X) ? jt : jf
571 BPF_JMP+BPF_JGE+BPF_X pc += (A >= X) ? jt : jf
572 BPF_JMP+BPF_JEQ+BPF_X pc += (A == X) ? jt : jf
573 BPF_JMP+BPF_JSET+BPF_X pc += (A & X) ? jt : jf
576 The return instructions terminate the filter program and specify the amount
577 of packet to accept (i.e., they return the truncation amount).
578 A return value of zero indicates that the packet should be ignored.
579 The return value is either a constant
585 BPF_RET+BPF_A accept A bytes
586 BPF_RET+BPF_K accept k bytes
589 The miscellaneous category was created for anything that does not
590 fit into the above classes, and for any new instructions that might need to
592 Currently, these are the register transfer instructions
593 that copy the index register to the accumulator or vice versa.
596 BPF_MISC+BPF_TAX X <- A
597 BPF_MISC+BPF_TXA A <- X
603 interface provides the following macros to facilitate
605 .Fn BPF_STMT opcode operand
607 .Fn BPF_JUMP opcode operand true_offset false_offset .
609 .Bl -tag -compact -width /dev/bpfXXX
610 .It Pa /dev/bpf Ns Sy n
611 the packet filter device
614 The following filter is taken from the Reverse ARP Daemon.
615 It accepts only Reverse ARP requests.
617 struct bpf_insn insns[] = {
618 BPF_STMT(BPF_LD+BPF_H+BPF_ABS, 12),
619 BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, ETHERTYPE_REVARP, 0, 3),
620 BPF_STMT(BPF_LD+BPF_H+BPF_ABS, 20),
621 BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, REVARP_REQUEST, 0, 1),
622 BPF_STMT(BPF_RET+BPF_K, sizeof(struct ether_arp) +
623 sizeof(struct ether_header)),
624 BPF_STMT(BPF_RET+BPF_K, 0),
628 This filter accepts only IP packets between host 128.3.112.15 and
631 struct bpf_insn insns[] = {
632 BPF_STMT(BPF_LD+BPF_H+BPF_ABS, 12),
633 BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, ETHERTYPE_IP, 0, 8),
634 BPF_STMT(BPF_LD+BPF_W+BPF_ABS, 26),
635 BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 0x8003700f, 0, 2),
636 BPF_STMT(BPF_LD+BPF_W+BPF_ABS, 30),
637 BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 0x80037023, 3, 4),
638 BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 0x80037023, 0, 3),
639 BPF_STMT(BPF_LD+BPF_W+BPF_ABS, 30),
640 BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 0x8003700f, 0, 1),
641 BPF_STMT(BPF_RET+BPF_K, (u_int)-1),
642 BPF_STMT(BPF_RET+BPF_K, 0),
646 Finally, this filter returns only TCP finger packets.
647 We must parse the IP header to reach the TCP header.
651 checks that the IP fragment offset is 0 so we are sure
652 that we have a TCP header.
654 struct bpf_insn insns[] = {
655 BPF_STMT(BPF_LD+BPF_H+BPF_ABS, 12),
656 BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, ETHERTYPE_IP, 0, 10),
657 BPF_STMT(BPF_LD+BPF_B+BPF_ABS, 23),
658 BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, IPPROTO_TCP, 0, 8),
659 BPF_STMT(BPF_LD+BPF_H+BPF_ABS, 20),
660 BPF_JUMP(BPF_JMP+BPF_JSET+BPF_K, 0x1fff, 6, 0),
661 BPF_STMT(BPF_LDX+BPF_B+BPF_MSH, 14),
662 BPF_STMT(BPF_LD+BPF_H+BPF_IND, 14),
663 BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 79, 2, 0),
664 BPF_STMT(BPF_LD+BPF_H+BPF_IND, 16),
665 BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 79, 0, 1),
666 BPF_STMT(BPF_RET+BPF_K, (u_int)-1),
667 BPF_STMT(BPF_RET+BPF_K, 0),
679 .%T "An efficient, extensible, and portable network monitor"
682 The Enet packet filter was created in 1980 by Mike Accetta and
683 Rick Rashid at Carnegie-Mellon University.
685 Stanford, ported the code to
687 and continued its development from
689 Since then, it has evolved into the Ultrix Packet Filter at
701 of Lawrence Berkeley Laboratory, implemented BPF in
703 Much of the design is due to
706 The read buffer must be of a fixed size (returned by the
710 A file that does not request promiscuous mode may receive promiscuously
711 received packets as a side effect of another file requesting this
712 mode on the same hardware interface.
713 This could be fixed in the kernel with additional processing overhead.
714 However, we favor the model where
715 all files must assume that the interface is promiscuous, and if
716 so desired, must utilize a filter to reject foreign packets.
718 Data link protocols with variable length headers are not currently supported.
722 flag has been observed to work incorrectly on some interface
723 types, including those with hardware loopback rather than software loopback,
724 and point-to-point interfaces.
725 It appears to function correctly on a
726 broad range of Ethernet-style interfaces.