1.\" Copyright (c) 1990 The Regents of the University of California. 2.\" All rights reserved. 3.\" 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. 19.\" 20.\" This document is derived in part from the enet man page (enet.4) 21.\" distributed with 4.3BSD Unix. 22.\" 23.\" $FreeBSD: src/share/man/man4/bpf.4,v 1.21.2.11 2002/04/07 04:57:13 dd Exp $ 24.\" $DragonFly: src/share/man/man4/bpf.4,v 1.2 2003/06/17 04:36:58 dillon Exp $ 25.\" 26.Dd January 16, 1996 27.Dt BPF 4 28.Os 29.Sh NAME 30.Nm bpf 31.Nd Berkeley Packet Filter 32.Sh SYNOPSIS 33.Cd pseudo-device bpf 34.Sh DESCRIPTION 35The Berkeley Packet Filter 36provides a raw interface to data link layers in a protocol 37independent fashion. 38All packets on the network, even those destined for other hosts, 39are accessible through this mechanism. 40.Pp 41The packet filter appears as a character special device, 42.Pa /dev/bpf0 , 43.Pa /dev/bpf1 , 44etc. 45After opening the device, the file descriptor must be bound to a 46specific network interface with the 47.Dv BIOCSETIF 48ioctl. 49A given interface can be shared by multiple listeners, and the filter 50underlying each descriptor will see an identical packet stream. 51.Pp 52A separate device file is required for each minor device. 53If a file is in use, the open will fail and 54.Va errno 55will be set to 56.Er EBUSY . 57.Pp 58Associated with each open instance of a 59.Nm 60file is a user-settable packet filter. 61Whenever a packet is received by an interface, 62all file descriptors listening on that interface apply their filter. 63Each descriptor that accepts the packet receives its own copy. 64.Pp 65Reads from these files return the next group of packets 66that have matched the filter. 67To improve performance, the buffer passed to read must be 68the same size as the buffers used internally by 69.Nm . 70This size is returned by the 71.Dv BIOCGBLEN 72ioctl (see below), and 73can be set with 74.Dv BIOCSBLEN . 75Note that an individual packet larger than this size is necessarily 76truncated. 77.Pp 78The packet filter will support any link level protocol that has fixed length 79headers. Currently, only Ethernet, 80.Tn SLIP , 81and 82.Tn PPP 83drivers have been modified to interact with 84.Nm . 85.Pp 86Since packet data is in network byte order, applications should use the 87.Xr byteorder 3 88macros to extract multi-byte values. 89.Pp 90A packet can be sent out on the network by writing to a 91.Nm 92file descriptor. The writes are unbuffered, meaning only one 93packet can be processed per write. 94Currently, only writes to Ethernets and 95.Tn SLIP 96links are supported. 97.Sh IOCTLS 98The 99.Xr ioctl 2 100command codes below are defined in 101.Aq Pa net/bpf.h . 102All commands require 103these includes: 104.Bd -literal 105 #include <sys/types.h> 106 #include <sys/time.h> 107 #include <sys/ioctl.h> 108 #include <net/bpf.h> 109.Ed 110.Pp 111Additionally, 112.Dv BIOCGETIF 113and 114.Dv BIOCSETIF 115require 116.Aq Pa sys/socket.h 117and 118.Aq Pa net/if.h . 119.Pp 120In addition to 121.Dv FIONREAD 122and 123.Dv SIOCGIFADDR , 124the following commands may be applied to any open 125.Nm 126file. 127The (third) argument to 128.Xr ioctl 2 129should be a pointer to the type indicated. 130.Bl -tag -width BIOCGRTIMEOUT 131.It Dv BIOCGBLEN 132.Pq Li u_int 133Returns the required buffer length for reads on 134.Nm 135files. 136.It Dv BIOCSBLEN 137.Pq Li u_int 138Sets the buffer length for reads on 139.Nm 140files. The buffer must be set before the file is attached to an interface 141with 142.Dv BIOCSETIF . 143If the requested buffer size cannot be accommodated, the closest 144allowable size will be set and returned in the argument. 145A read call will result in 146.Er EIO 147if it is passed a buffer that is not this size. 148.It Dv BIOCGDLT 149.Pq Li u_int 150Returns the type of the data link layer underlying the attached interface. 151.Er EINVAL 152is returned if no interface has been specified. 153The device types, prefixed with 154.Dq Li DLT_ , 155are defined in 156.Aq Pa net/bpf.h . 157.It Dv BIOCPROMISC 158Forces the interface into promiscuous mode. 159All packets, not just those destined for the local host, are processed. 160Since more than one file can be listening on a given interface, 161a listener that opened its interface non-promiscuously may receive 162packets promiscuously. This problem can be remedied with an 163appropriate filter. 164.It Dv BIOCFLUSH 165Flushes the buffer of incoming packets, 166and resets the statistics that are returned by BIOCGSTATS. 167.It Dv BIOCGETIF 168.Pq Li "struct ifreq" 169Returns the name of the hardware interface that the file is listening on. 170The name is returned in the ifr_name field of 171the 172.Li ifreq 173structure. 174All other fields are undefined. 175.It Dv BIOCSETIF 176.Pq Li "struct ifreq" 177Sets the hardware interface associate with the file. This 178command must be performed before any packets can be read. 179The device is indicated by name using the 180.Li ifr_name 181field of the 182.Li ifreq 183structure. 184Additionally, performs the actions of 185.Dv BIOCFLUSH . 186.It Dv BIOCSRTIMEOUT 187.It Dv BIOCGRTIMEOUT 188.Pq Li "struct timeval" 189Set or get the read timeout parameter. 190The argument 191specifies the length of time to wait before timing 192out on a read request. 193This parameter is initialized to zero by 194.Xr open 2 , 195indicating no timeout. 196.It Dv BIOCGSTATS 197.Pq Li "struct bpf_stat" 198Returns the following structure of packet statistics: 199.Bd -literal 200struct bpf_stat { 201 u_int bs_recv; /* number of packets received */ 202 u_int bs_drop; /* number of packets dropped */ 203}; 204.Ed 205.Pp 206The fields are: 207.Bl -hang -offset indent 208.It Li bs_recv 209the number of packets received by the descriptor since opened or reset 210(including any buffered since the last read call); 211and 212.It Li bs_drop 213the number of packets which were accepted by the filter but dropped by the 214kernel because of buffer overflows 215(i.e., the application's reads aren't keeping up with the packet traffic). 216.El 217.It Dv BIOCIMMEDIATE 218.Pq Li u_int 219Enable or disable 220.Dq immediate mode , 221based on the truth value of the argument. 222When immediate mode is enabled, reads return immediately upon packet 223reception. Otherwise, a read will block until either the kernel buffer 224becomes full or a timeout occurs. 225This is useful for programs like 226.Xr rarpd 8 227which must respond to messages in real time. 228The default for a new file is off. 229.It Dv BIOCSETF 230.Pq Li "struct bpf_program" 231Sets the filter program used by the kernel to discard uninteresting 232packets. An array of instructions and its length is passed in using 233the following structure: 234.Bd -literal 235struct bpf_program { 236 int bf_len; 237 struct bpf_insn *bf_insns; 238}; 239.Ed 240.Pp 241The filter program is pointed to by the 242.Li bf_insns 243field while its length in units of 244.Sq Li struct bpf_insn 245is given by the 246.Li bf_len 247field. 248Also, the actions of 249.Dv BIOCFLUSH 250are performed. 251See section 252.Sx "FILTER MACHINE" 253for an explanation of the filter language. 254.It Dv BIOCVERSION 255.Pq Li "struct bpf_version" 256Returns the major and minor version numbers of the filter language currently 257recognized by the kernel. Before installing a filter, applications must check 258that the current version is compatible with the running kernel. Version 259numbers are compatible if the major numbers match and the application minor 260is less than or equal to the kernel minor. The kernel version number is 261returned in the following structure: 262.Bd -literal 263struct bpf_version { 264 u_short bv_major; 265 u_short bv_minor; 266}; 267.Ed 268.Pp 269The current version numbers are given by 270.Dv BPF_MAJOR_VERSION 271and 272.Dv BPF_MINOR_VERSION 273from 274.Aq Pa net/bpf.h . 275An incompatible filter 276may result in undefined behavior (most likely, an error returned by 277.Fn ioctl 278or haphazard packet matching). 279.It Dv BIOCSHDRCMPLT 280.It Dv BIOCGHDRCMPLT 281.Pq Li u_int 282Set or get the status of the 283.Dq header complete 284flag. 285Set to zero if the link level source address should be filled in automatically 286by the interface output routine. Set to one if the link level source 287address will be written, as provided, to the wire. This flag is initialized 288to zero by default. 289.It Dv BIOCSSEESENT 290.It Dv BIOCGSEESENT 291.Pq Li u_int 292Set or get the flag determining whether locally generated packets on the 293interface should be returned by BPF. Set to zero to see only incoming 294packets on the interface. Set to one to see packets originating 295locally and remotely on the interface. This flag is initialized to one by 296default. 297.El 298.Sh BPF HEADER 299The following structure is prepended to each packet returned by 300.Xr read 2 : 301.Bd -literal 302struct bpf_hdr { 303 struct timeval bh_tstamp; /* time stamp */ 304 u_long bh_caplen; /* length of captured portion */ 305 u_long bh_datalen; /* original length of packet */ 306 u_short bh_hdrlen; /* length of bpf header (this struct 307 plus alignment padding */ 308}; 309.Ed 310.Pp 311The fields, whose values are stored in host order, and are: 312.Pp 313.Bl -tag -compact -width bh_datalen 314.It Li bh_tstamp 315The time at which the packet was processed by the packet filter. 316.It Li bh_caplen 317The length of the captured portion of the packet. This is the minimum of 318the truncation amount specified by the filter and the length of the packet. 319.It Li bh_datalen 320The length of the packet off the wire. 321This value is independent of the truncation amount specified by the filter. 322.It Li bh_hdrlen 323The length of the 324.Nm 325header, which may not be equal to 326.\" XXX - not really a function call 327.Fn sizeof "struct bpf_hdr" . 328.El 329.Pp 330The 331.Li bh_hdrlen 332field exists to account for 333padding between the header and the link level protocol. 334The purpose here is to guarantee proper alignment of the packet 335data structures, which is required on alignment sensitive 336architectures and improves performance on many other architectures. 337The packet filter insures that the 338.Li bpf_hdr 339and the network layer 340header will be word aligned. Suitable precautions 341must be taken when accessing the link layer protocol fields on alignment 342restricted machines. (This isn't a problem on an Ethernet, since 343the type field is a short falling on an even offset, 344and the addresses are probably accessed in a bytewise fashion). 345.Pp 346Additionally, individual packets are padded so that each starts 347on a word boundary. This requires that an application 348has some knowledge of how to get from packet to packet. 349The macro 350.Dv BPF_WORDALIGN 351is defined in 352.Aq Pa net/bpf.h 353to facilitate 354this process. It rounds up its argument 355to the nearest word aligned value (where a word is 356.Dv BPF_ALIGNMENT 357bytes wide). 358.Pp 359For example, if 360.Sq Li p 361points to the start of a packet, this expression 362will advance it to the next packet: 363.Dl p = (char *)p + BPF_WORDALIGN(p->bh_hdrlen + p->bh_caplen) 364.Pp 365For the alignment mechanisms to work properly, the 366buffer passed to 367.Xr read 2 368must itself be word aligned. 369The 370.Xr malloc 3 371function 372will always return an aligned buffer. 373.Sh FILTER MACHINE 374A filter program is an array of instructions, with all branches forwardly 375directed, terminated by a 376.Em return 377instruction. 378Each instruction performs some action on the pseudo-machine state, 379which consists of an accumulator, index register, scratch memory store, 380and implicit program counter. 381.Pp 382The following structure defines the instruction format: 383.Bd -literal 384struct bpf_insn { 385 u_short code; 386 u_char jt; 387 u_char jf; 388 u_long k; 389}; 390.Ed 391.Pp 392The 393.Li k 394field is used in different ways by different instructions, 395and the 396.Li jt 397and 398.Li jf 399fields are used as offsets 400by the branch instructions. 401The opcodes are encoded in a semi-hierarchical fashion. 402There are eight classes of instructions: 403.Dv BPF_LD , 404.Dv BPF_LDX , 405.Dv BPF_ST , 406.Dv BPF_STX , 407.Dv BPF_ALU , 408.Dv BPF_JMP , 409.Dv BPF_RET , 410and 411.Dv BPF_MISC . 412Various other mode and 413operator bits are or'd into the class to give the actual instructions. 414The classes and modes are defined in 415.Aq Pa net/bpf.h . 416.Pp 417Below are the semantics for each defined 418.Nm 419instruction. 420We use the convention that A is the accumulator, X is the index register, 421P[] packet data, and M[] scratch memory store. 422P[i:n] gives the data at byte offset 423.Dq i 424in the packet, 425interpreted as a word (n=4), 426unsigned halfword (n=2), or unsigned byte (n=1). 427M[i] gives the i'th word in the scratch memory store, which is only 428addressed in word units. The memory store is indexed from 0 to 429.Dv BPF_MEMWORDS 430- 1. 431.Li k , 432.Li jt , 433and 434.Li jf 435are the corresponding fields in the 436instruction definition. 437.Dq len 438refers to the length of the packet. 439.Pp 440.Bl -tag -width BPF_STXx 441.It Dv BPF_LD 442These instructions copy a value into the accumulator. The type of the 443source operand is specified by an 444.Dq addressing mode 445and can be a constant 446.Pq Dv BPF_IMM , 447packet data at a fixed offset 448.Pq Dv BPF_ABS , 449packet data at a variable offset 450.Pq Dv BPF_IND , 451the packet length 452.Pq Dv BPF_LEN , 453or a word in the scratch memory store 454.Pq Dv BPF_MEM . 455For 456.Dv BPF_IND 457and 458.Dv BPF_ABS , 459the data size must be specified as a word 460.Pq Dv BPF_W , 461halfword 462.Pq Dv BPF_H , 463or byte 464.Pq Dv BPF_B . 465The semantics of all the recognized 466.Dv BPF_LD 467instructions follow. 468.Pp 469.Bl -tag -width "BPF_LD+BPF_W+BPF_IND" -compact 470.It Li BPF_LD+BPF_W+BPF_ABS 471A <- P[k:4] 472.It Li BPF_LD+BPF_H+BPF_ABS 473A <- P[k:2] 474.It Li BPF_LD+BPF_B+BPF_ABS 475A <- P[k:1] 476.It Li BPF_LD+BPF_W+BPF_IND 477A <- P[X+k:4] 478.It Li BPF_LD+BPF_H+BPF_IND 479A <- P[X+k:2] 480.It Li BPF_LD+BPF_B+BPF_IND 481A <- P[X+k:1] 482.It Li BPF_LD+BPF_W+BPF_LEN 483A <- len 484.It Li BPF_LD+BPF_IMM 485A <- k 486.It Li BPF_LD+BPF_MEM 487A <- M[k] 488.El 489.It Dv BPF_LDX 490These instructions load a value into the index register. Note that 491the addressing modes are more restrictive than those of the accumulator loads, 492but they include 493.Dv BPF_MSH , 494a hack for efficiently loading the IP header length. 495.Pp 496.Bl -tag -width "BPF_LDX+BPF_W+BPF_MEM" -compact 497.It Li BPF_LDX+BPF_W+BPF_IMM 498X <- k 499.It Li BPF_LDX+BPF_W+BPF_MEM 500X <- M[k] 501.It Li BPF_LDX+BPF_W+BPF_LEN 502X <- len 503.It Li BPF_LDX+BPF_B+BPF_MSH 504X <- 4*(P[k:1]&0xf) 505.El 506.It Dv BPF_ST 507This instruction stores the accumulator into the scratch memory. 508We do not need an addressing mode since there is only one possibility 509for the destination. 510.Pp 511.Bl -tag -width "BPF_ST" -compact 512.It Li BPF_ST 513M[k] <- A 514.El 515.It Dv BPF_STX 516This instruction stores the index register in the scratch memory store. 517.Pp 518.Bl -tag -width "BPF_STX" -compact 519.It Li BPF_STX 520M[k] <- X 521.El 522.It Dv BPF_ALU 523The alu instructions perform operations between the accumulator and 524index register or constant, and store the result back in the accumulator. 525For binary operations, a source mode is required 526.Dv ( BPF_K 527or 528.Dv BPF_X ) . 529.Pp 530.Bl -tag -width "BPF_ALU+BPF_MUL+BPF_K" -compact 531.It Li BPF_ALU+BPF_ADD+BPF_K 532A <- A + k 533.It Li BPF_ALU+BPF_SUB+BPF_K 534A <- A - k 535.It Li BPF_ALU+BPF_MUL+BPF_K 536A <- A * k 537.It Li BPF_ALU+BPF_DIV+BPF_K 538A <- A / k 539.It Li BPF_ALU+BPF_AND+BPF_K 540A <- A & k 541.It Li BPF_ALU+BPF_OR+BPF_K 542A <- A | k 543.It Li BPF_ALU+BPF_LSH+BPF_K 544A <- A << k 545.It Li BPF_ALU+BPF_RSH+BPF_K 546A <- A >> k 547.It Li BPF_ALU+BPF_ADD+BPF_X 548A <- A + X 549.It Li BPF_ALU+BPF_SUB+BPF_X 550A <- A - X 551.It Li BPF_ALU+BPF_MUL+BPF_X 552A <- A * X 553.It Li BPF_ALU+BPF_DIV+BPF_X 554A <- A / X 555.It Li BPF_ALU+BPF_AND+BPF_X 556A <- A & X 557.It Li BPF_ALU+BPF_OR+BPF_X 558A <- A | X 559.It Li BPF_ALU+BPF_LSH+BPF_X 560A <- A << X 561.It Li BPF_ALU+BPF_RSH+BPF_X 562A <- A >> X 563.It Li BPF_ALU+BPF_NEG 564A <- -A 565.El 566.It Dv BPF_JMP 567The jump instructions alter flow of control. Conditional jumps 568compare the accumulator against a constant 569.Pq Dv BPF_K 570or the index register 571.Pq Dv BPF_X . 572If the result is true (or non-zero), 573the true branch is taken, otherwise the false branch is taken. 574Jump offsets are encoded in 8 bits so the longest jump is 256 instructions. 575However, the jump always 576.Pq Dv BPF_JA 577opcode uses the 32 bit 578.Li k 579field as the offset, allowing arbitrarily distant destinations. 580All conditionals use unsigned comparison conventions. 581.Pp 582.Bl -tag -width "BPF_JMP+BPF_KSET+BPF_X" -compact 583.It Li BPF_JMP+BPF_JA 584pc += k 585.It Li BPF_JMP+BPF_JGT+BPF_K 586pc += (A > k) ? jt : jf 587.It Li BPF_JMP+BPF_JGE+BPF_K 588pc += (A >= k) ? jt : jf 589.It Li BPF_JMP+BPF_JEQ+BPF_K 590pc += (A == k) ? jt : jf 591.It Li BPF_JMP+BPF_JSET+BPF_K 592pc += (A & k) ? jt : jf 593.It Li BPF_JMP+BPF_JGT+BPF_X 594pc += (A > X) ? jt : jf 595.It Li BPF_JMP+BPF_JGE+BPF_X 596pc += (A >= X) ? jt : jf 597.It Li BPF_JMP+BPF_JEQ+BPF_X 598pc += (A == X) ? jt : jf 599.It Li BPF_JMP+BPF_JSET+BPF_X 600pc += (A & X) ? jt : jf 601.El 602.It Dv BPF_RET 603The return instructions terminate the filter program and specify the amount 604of packet to accept (i.e., they return the truncation amount). A return 605value of zero indicates that the packet should be ignored. 606The return value is either a constant 607.Pq Dv BPF_K 608or the accumulator 609.Pq Dv BPF_A . 610.Pp 611.Bl -tag -width "BPF_RET+BPF_K" -compact 612.It Li BPF_RET+BPF_A 613accept A bytes 614.It Li BPF_RET+BPF_K 615accept k bytes 616.El 617.It Dv BPF_MISC 618The miscellaneous category was created for anything that doesn't 619fit into the above classes, and for any new instructions that might need to 620be added. Currently, these are the register transfer instructions 621that copy the index register to the accumulator or vice versa. 622.Pp 623.Bl -tag -width "BPF_MISC+BPF_TAX" -compact 624.It Li BPF_MISC+BPF_TAX 625X <- A 626.It Li BPF_MISC+BPF_TXA 627A <- X 628.El 629.El 630.Pp 631The 632.Nm 633interface provides the following macros to facilitate 634array initializers: 635.Fn BPF_STMT opcode operand 636and 637.Fn BPF_JUMP opcode operand true_offset false_offset . 638.Sh EXAMPLES 639The following filter is taken from the Reverse ARP Daemon. It accepts 640only Reverse ARP requests. 641.Bd -literal 642struct bpf_insn insns[] = { 643 BPF_STMT(BPF_LD+BPF_H+BPF_ABS, 12), 644 BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, ETHERTYPE_REVARP, 0, 3), 645 BPF_STMT(BPF_LD+BPF_H+BPF_ABS, 20), 646 BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, REVARP_REQUEST, 0, 1), 647 BPF_STMT(BPF_RET+BPF_K, sizeof(struct ether_arp) + 648 sizeof(struct ether_header)), 649 BPF_STMT(BPF_RET+BPF_K, 0), 650}; 651.Ed 652.Pp 653This filter accepts only IP packets between host 128.3.112.15 and 654128.3.112.35. 655.Bd -literal 656struct bpf_insn insns[] = { 657 BPF_STMT(BPF_LD+BPF_H+BPF_ABS, 12), 658 BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, ETHERTYPE_IP, 0, 8), 659 BPF_STMT(BPF_LD+BPF_W+BPF_ABS, 26), 660 BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 0x8003700f, 0, 2), 661 BPF_STMT(BPF_LD+BPF_W+BPF_ABS, 30), 662 BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 0x80037023, 3, 4), 663 BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 0x80037023, 0, 3), 664 BPF_STMT(BPF_LD+BPF_W+BPF_ABS, 30), 665 BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 0x8003700f, 0, 1), 666 BPF_STMT(BPF_RET+BPF_K, (u_int)-1), 667 BPF_STMT(BPF_RET+BPF_K, 0), 668}; 669.Ed 670.Pp 671Finally, this filter returns only TCP finger packets. We must parse 672the IP header to reach the TCP header. The 673.Dv BPF_JSET 674instruction 675checks that the IP fragment offset is 0 so we are sure 676that we have a TCP header. 677.Bd -literal 678struct bpf_insn insns[] = { 679 BPF_STMT(BPF_LD+BPF_H+BPF_ABS, 12), 680 BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, ETHERTYPE_IP, 0, 10), 681 BPF_STMT(BPF_LD+BPF_B+BPF_ABS, 23), 682 BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, IPPROTO_TCP, 0, 8), 683 BPF_STMT(BPF_LD+BPF_H+BPF_ABS, 20), 684 BPF_JUMP(BPF_JMP+BPF_JSET+BPF_K, 0x1fff, 6, 0), 685 BPF_STMT(BPF_LDX+BPF_B+BPF_MSH, 14), 686 BPF_STMT(BPF_LD+BPF_H+BPF_IND, 14), 687 BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 79, 2, 0), 688 BPF_STMT(BPF_LD+BPF_H+BPF_IND, 16), 689 BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 79, 0, 1), 690 BPF_STMT(BPF_RET+BPF_K, (u_int)-1), 691 BPF_STMT(BPF_RET+BPF_K, 0), 692}; 693.Ed 694.Sh SEE ALSO 695.Xr tcpdump 1 , 696.Xr ioctl 2 , 697.Xr byteorder 3 , 698.Xr ng_bpf 4 699.Rs 700.%A McCanne, S. 701.%A Jacobson V. 702.%T "An efficient, extensible, and portable network monitor" 703.Re 704.Sh FILES 705.Bl -tag -compact -width /dev/bpfXXX 706.It Pa /dev/bpf Ns Sy n 707the packet filter device 708.El 709.Sh BUGS 710The read buffer must be of a fixed size (returned by the 711.Dv BIOCGBLEN 712ioctl). 713.Pp 714A file that does not request promiscuous mode may receive promiscuously 715received packets as a side effect of another file requesting this 716mode on the same hardware interface. This could be fixed in the kernel 717with additional processing overhead. However, we favor the model where 718all files must assume that the interface is promiscuous, and if 719so desired, must utilize a filter to reject foreign packets. 720.Pp 721Data link protocols with variable length headers are not currently supported. 722.Sh HISTORY 723The Enet packet filter was created in 1980 by Mike Accetta and 724Rick Rashid at Carnegie-Mellon University. Jeffrey Mogul, at 725Stanford, ported the code to 726.Bx 727and continued its development from 7281983 on. Since then, it has evolved into the Ultrix Packet Filter 729at 730.Tn DEC , 731a 732.Tn STREAMS 733.Tn NIT 734module under 735.Tn SunOS 4.1 , 736and 737.Tn BPF . 738.Sh AUTHORS 739.An -nosplit 740.An Steven McCanne , 741of Lawrence Berkeley Laboratory, implemented BPF in 742Summer 1990. Much of the design is due to 743.An Van Jacobson . 744