.\" -*- nroff -*- .\" .\" $NetBSD: bpf.4,v 1.65 2022/11/30 06:03:41 ozaki-r Exp $ .\" .\" Copyright (c) 1990, 1991, 1992, 1993, 1994 .\" The Regents of the University of California. All rights reserved. .\" .\" Redistribution and use in source and binary forms, with or without .\" modification, are permitted provided that: (1) source code distributions .\" retain the above copyright notice and this paragraph in its entirety, (2) .\" distributions including binary code include the above copyright notice and .\" this paragraph in its entirety in the documentation or other materials .\" provided with the distribution, and (3) all advertising materials mentioning .\" features or use of this software display the following acknowledgement: .\" ``This product includes software developed by the University of California, .\" Lawrence Berkeley Laboratory and its contributors.'' Neither the name of .\" the University nor the names of its contributors may be used to endorse .\" or promote products derived from this software without specific prior .\" written permission. .\" THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED .\" WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF .\" MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. .\" .\" This document is derived in part from the enet man page (enet.4) .\" distributed with 4.3BSD Unix. .\" .Dd November 30, 2022 .Dt BPF 4 .Os .Sh NAME .Nm bpf .Nd Berkeley Packet Filter raw network interface .Sh SYNOPSIS .Cd "pseudo-device bpfilter" .Sh DESCRIPTION The Berkeley Packet Filter provides a raw interface to data link layers in a protocol independent fashion. All packets on the network, even those destined for other hosts, are accessible through this mechanism. .Pp The packet filter appears as a character special device, .Pa /dev/bpf . After opening the device, the file descriptor must be bound to a specific network interface with the .Dv BIOCSETIF ioctl. A given interface can be shared by multiple listeners, and the filter underlying each descriptor will see an identical packet stream. .Pp Associated with each open instance of a .Nm file is a user-settable packet filter. Whenever a packet is received by an interface, all file descriptors listening on that interface apply their filter. Each descriptor that accepts the packet receives its own copy. .Pp Reads from these files return the next group of packets that have matched the filter. To improve performance, the buffer passed to read must be the same size as the buffers used internally by .Nm . This size is returned by the .Dv BIOCGBLEN ioctl (see below), and can be set with .Dv BIOCSBLEN . Note that an individual packet larger than this size is necessarily truncated. .Pp Since packet data is in network byte order, applications should use the .Xr byteorder 3 macros to extract multi-byte values. .Pp A packet can be sent out on the network by writing to a .Nm file descriptor. The writes are unbuffered, meaning only one packet can be processed per write. Currently, only writes to Ethernet-based (including Wi-Fi), SLIP and loopback links are supported. .Sh IOCTLS The .Xr ioctl 2 command codes below are defined in .In net/bpf.h . All commands require these includes: .Bd -literal -offset indent #include #include #include #include .Ed .Pp Additionally, .Dv BIOCGETIF and .Dv BIOCSETIF require .Pa . .Pp The (third) argument to the .Xr ioctl 2 should be a pointer to the type indicated. .Bl -tag -width indent -offset indent .It Dv BIOCGBLEN ( u_int ) Returns the required buffer length for reads on .Nm files. .It Dv BIOCSBLEN ( u_int ) Sets the buffer length for reads on .Nm files. The buffer must be set before the file is attached to an interface with .Dv BIOCSETIF . If the requested buffer size cannot be accommodated, the closest allowable size will be set and returned in the argument. A read call will result in .Er EINVAL if it is passed a buffer that is not this size. .It Dv BIOCGDLT ( u_int ) Returns the type of the data link layer underlying the attached interface. .Er EINVAL is returned if no interface has been specified. The device types, prefixed with .Dq DLT_ , are defined in .In net/bpf.h . .It Dv BIOCGDLTLIST ( struct bpf_dltlist ) Returns an array of the available types of the data link layer underlying the attached interface: .Bd -literal -offset indent struct bpf_dltlist { u_int bfl_len; u_int *bfl_list; }; .Ed .Pp The available types are returned in the array pointed to by the .Va bfl_list field while their length in u_int is supplied to the .Va bfl_len field. .Er ENOMEM is returned if there is not enough buffer space and .Er EFAULT is returned if a bad address is encountered. The .Va bfl_len field is modified on return to indicate the actual length in u_int of the array returned. If .Va bfl_list is .Dv NULL , the .Va bfl_len field is set to indicate the required length of an array in u_int. .It Dv BIOCSDLT ( u_int ) Changes the type of the data link layer underlying the attached interface. .Er EINVAL is returned if no interface has been specified or the specified type is not available for the interface. .It Dv BIOCPROMISC Forces the interface into promiscuous mode. All packets, not just those destined for the local host, are processed. Since more than one file can be listening on a given interface, a listener that opened its interface non-promiscuously may receive packets promiscuously. This problem can be remedied with an appropriate filter. .Pp The interface remains in promiscuous mode until all files listening promiscuously are closed. .It Dv BIOCFLUSH Flushes the buffer of incoming packets, and resets the statistics that are returned by .Dv BIOCGSTATS . .It Dv BIOCGETIF ( struct ifreq ) Returns the name of the hardware interface that the file is listening on. The name is returned in the ifr_name field of .Fa ifr . All other fields are undefined. .It Dv BIOCSETIF ( struct ifreq ) Sets the hardware interface associated with the file. This command must be performed before any packets can be read. The device is indicated by name using the .Dv ifr_name field of the .Fa ifreq . Additionally, performs the actions of .Dv BIOCFLUSH . .It Dv BIOCSRTIMEOUT , BIOCGRTIMEOUT ( struct timeval ) Sets or gets the read timeout parameter. The .Fa timeval specifies the length of time to wait before timing out on a read request. This parameter is initialized to zero by .Xr open 2 , indicating no timeout. .It Dv BIOCGSTATS ( struct bpf_stat ) Returns the following structure of packet statistics: .Bd -literal -offset indent struct bpf_stat { uint64_t bs_recv; uint64_t bs_drop; uint64_t bs_capt; uint64_t bs_padding[13]; }; .Ed .Pp The fields are: .Bl -tag -width bs_recv -offset indent .It Va bs_recv the number of packets received by the descriptor since opened or reset (including any buffered since the last read call); .It Va bs_drop the number of packets which were accepted by the filter but dropped by the kernel because of buffer overflows (i.e., the application's reads aren't keeping up with the packet traffic); and .It Va bs_capt the number of packets accepted by the filter. .El .It Dv BIOCIMMEDIATE ( u_int ) Enables or disables .Dq immediate mode , based on the truth value of the argument. When immediate mode is enabled, reads return immediately upon packet reception. Otherwise, a read will block until either the kernel buffer becomes full or a timeout occurs. This is useful for programs like .Xr rarpd 8 , which must respond to messages in real time. The default for a new file is off. .Dv BIOCLOCK Set the locked flag on the bpf descriptor. This prevents the execution of ioctl commands which could change the underlying operating parameters of the device. .It Dv BIOCSETF ( struct bpf_program ) Sets the filter program used by the kernel to discard uninteresting packets. An array of instructions and its length are passed in using the following structure: .Bd -literal -offset indent struct bpf_program { u_int bf_len; struct bpf_insn *bf_insns; }; .Ed .Pp The filter program is pointed to by the .Va bf_insns field while its length in units of .Sq struct bpf_insn is given by the .Va bf_len field. Also, the actions of .Dv BIOCFLUSH are performed. .Pp See section .Sy FILTER MACHINE for an explanation of the filter language. .It Dv BIOCSETWF ( struct bpf_program ) Sets the write filter program used by the kernel to control what type of packets can be written to the interface. See the .Dv BIOCSETF command for more information on the bpf filter program. .It Dv BIOCVERSION ( struct bpf_version ) Returns the major and minor version numbers of the filter language currently recognized by the kernel. Before installing a filter, applications must check that the current version is compatible with the running kernel. Version numbers are compatible if the major numbers match and the application minor is less than or equal to the kernel minor. The kernel version number is returned in the following structure: .Bd -literal -offset indent struct bpf_version { u_short bv_major; u_short bv_minor; }; .Ed .Pp The current version numbers are given by .Dv BPF_MAJOR_VERSION and .Dv BPF_MINOR_VERSION from .In net/bpf.h . An incompatible filter may result in undefined behavior (most likely, an error returned by .Xr ioctl 2 or haphazard packet matching). .It Dv BIOCSRSIG , BIOCGRSIG ( u_int ) Sets or gets the receive signal. This signal will be sent to the process or process group specified by .Dv FIOSETOWN . It defaults to .Dv SIGIO . .It Dv BIOCGHDRCMPLT , BIOCSHDRCMPLT ( u_int ) Sets or gets the status of the .Dq header complete flag. Set to zero if the link level source address should be filled in automatically by the interface output routine. Set to one if the link level source address will be written, as provided, to the wire. This flag is initialized to zero by default. .It Dv BIOCGSEESENT , BIOCSSEESENT ( u_int ) These commands are obsolete but left for compatibility. Use .Dv BIOCSDIRECTION and .Dv BIOCGDIRECTION instead. Set or get the flag determining whether locally generated packets on the interface should be returned by BPF. Set to zero to see only incoming packets on the interface. Set to one to see packets originating locally and remotely on the interface. This flag is initialized to one by default. .It Dv BIOCSDIRECTION .It Dv BIOCGDIRECTION .Pq Li u_int Set or get the setting determining whether incoming, outgoing, or all packets on the interface should be returned by BPF. Set to .Dv BPF_D_IN to see only incoming packets on the interface. Set to .Dv BPF_D_INOUT to see packets originating locally and remotely on the interface. Set to .Dv BPF_D_OUT to see only outgoing packets on the interface. This setting is initialized to .Dv BPF_D_INOUT by default. .It Dv BIOCFEEDBACK , BIOCSFEEDBACK , BIOCGFEEDBACK ( u_int ) Set (or get) .Dq packet feedback mode . This allows injected packets to be fed back as input to the interface when output via the interface is successful. The first name is meant for .Fx compatibility, the two others follow the Get/Set convention. .\"When .\".Dv BPF_D_INOUT .\"direction is set, injected Injected outgoing packets are not returned by BPF to avoid duplication. This flag is initialized to zero by default. .El .Sh STANDARD IOCTLS .Nm now supports several standard .Xr ioctl 2 Ns 's which allow the user to do async and/or non-blocking I/O to an open .Nm bpf file descriptor. .Bl -tag -width indent -offset indent .It Dv FIONREAD ( int ) Returns the number of bytes that are immediately available for reading. .It Dv FIONBIO ( int ) Set or clear non-blocking I/O. If arg is non-zero, then doing a .Xr read 2 when no data is available will return -1 and .Va errno will be set to .Er EAGAIN . If arg is zero, non-blocking I/O is disabled. Note: setting this overrides the timeout set by .Dv BIOCSRTIMEOUT . .It Dv FIOASYNC ( int ) Enable or disable async I/O. When enabled (arg is non-zero), the process or process group specified by .Dv FIOSETOWN will start receiving SIGIO's when packets arrive. Note that you must do an .Dv FIOSETOWN in order for this to take effect, as the system will not default this for you. The signal may be changed via .Dv BIOCSRSIG . .It Dv FIOSETOWN , FIOGETOWN ( int ) Set or get the process or process group (if negative) that should receive SIGIO when packets are available. The signal may be changed using .Dv BIOCSRSIG (see above). .El .Sh BPF HEADER The following structure is prepended to each packet returned by .Xr read 2 : .Bd -literal -offset indent struct bpf_hdr { struct bpf_timeval bh_tstamp; uint32_t bh_caplen; uint32_t bh_datalen; uint16_t bh_hdrlen; }; .Ed .Pp The fields, whose values are stored in host order, are: .Bl -tag -width bh_datalen -offset indent .It Va bh_tstamp The time at which the packet was processed by the packet filter. This structure differs from the standard .Vt struct timeval in that both members are of type .Vt long . .It Va bh_caplen The length of the captured portion of the packet. This is the minimum of the truncation amount specified by the filter and the length of the packet. .It Va bh_datalen The length of the packet off the wire. This value is independent of the truncation amount specified by the filter. .It Va bh_hdrlen The length of the BPF header, which may not be equal to .Em sizeof(struct bpf_hdr) . .El .Pp The .Va bh_hdrlen field exists to account for padding between the header and the link level protocol. The purpose here is to guarantee proper alignment of the packet data structures, which is required on alignment sensitive architectures and improves performance on many other architectures. The packet filter ensures that the .Va bpf_hdr and the .Em network layer header will be word aligned. Suitable precautions must be taken when accessing the link layer protocol fields on alignment restricted machines. (This isn't a problem on an Ethernet, since the type field is a short falling on an even offset, and the addresses are probably accessed in a bytewise fashion). .Pp Additionally, individual packets are padded so that each starts on a word boundary. This requires that an application has some knowledge of how to get from packet to packet. The macro .Dv BPF_WORDALIGN is defined in .In net/bpf.h to facilitate this process. It rounds up its argument to the nearest word aligned value (where a word is .Dv BPF_ALIGNMENT bytes wide). .Pp For example, if .Sq Va p points to the start of a packet, this expression will advance it to the next packet: .Pp .Dl p = (char *)p + BPF_WORDALIGN(p->bh_hdrlen + p->bh_caplen) .Pp For the alignment mechanisms to work properly, the buffer passed to .Xr read 2 must itself be word aligned. .Xr malloc 3 will always return an aligned buffer. .Sh FILTER MACHINE A filter program is an array of instructions, with all branches forwardly directed, terminated by a .Sy return instruction. Each instruction performs some action on the pseudo-machine state, which consists of an accumulator, index register, scratch memory store, and implicit program counter. .Pp The following structure defines the instruction format: .Bd -literal -offset indent struct bpf_insn { uint16_t code; u_char jt; u_char jf; uint32_t k; }; .Ed .Pp The .Va k field is used in different ways by different instructions, and the .Va jt and .Va jf fields are used as offsets by the branch instructions. The opcodes are encoded in a semi-hierarchical fashion. There are eight classes of instructions: BPF_LD, BPF_LDX, BPF_ST, BPF_STX, BPF_ALU, BPF_JMP, BPF_RET, and BPF_MISC. Various other mode and operator bits are or'd into the class to give the actual instructions. The classes and modes are defined in .In net/bpf.h . .Pp Below are the semantics for each defined BPF instruction. We use the convention that A is the accumulator, X is the index register, P[] packet data, and M[] scratch memory store. P[i:n] gives the data at byte offset .Dq i in the packet, interpreted as a word (n=4), unsigned halfword (n=2), or unsigned byte (n=1). M[i] gives the i'th word in the scratch memory store, which is only addressed in word units. The memory store is indexed from 0 to BPF_MEMWORDS-1. .Va k , .Va jt , and .Va jf are the corresponding fields in the instruction definition. .Dq len refers to the length of the packet. .Bl -tag -width indent -offset indent .It Sy BPF_LD These instructions copy a value into the accumulator. The type of the source operand is specified by an .Dq addressing mode and can be a constant .Sy ( BPF_IMM ) , packet data at a fixed offset .Sy ( BPF_ABS ) , packet data at a variable offset .Sy ( BPF_IND ) , the packet length .Sy ( BPF_LEN ) , or a word in the scratch memory store .Sy ( BPF_MEM ) . For .Sy BPF_IND and .Sy BPF_ABS , the data size must be specified as a word .Sy ( BPF_W ) , halfword .Sy ( BPF_H ) , or byte .Sy ( BPF_B ) . Arithmetic overflow when calculating a variable offset terminates the filter program and the packet is ignored. The semantics of all the recognized BPF_LD instructions follow. .Bl -column "BPF_LD_BPF_W_BPF_ABS" "A <- P[k:4]" -offset indent .It Sy BPF_LD+BPF_W+BPF_ABS Ta A <- P[k:4] .It Sy BPF_LD+BPF_H+BPF_ABS Ta A <- P[k:2] .It Sy BPF_LD+BPF_B+BPF_ABS Ta A <- P[k:1] .It Sy BPF_LD+BPF_W+BPF_IND Ta A <- P[X+k:4] .It Sy BPF_LD+BPF_H+BPF_IND Ta A <- P[X+k:2] .It Sy BPF_LD+BPF_B+BPF_IND Ta A <- P[X+k:1] .It Sy BPF_LD+BPF_W+BPF_LEN Ta A <- len .It Sy BPF_LD+BPF_IMM Ta A <- k .It Sy BPF_LD+BPF_MEM Ta A <- M[k] .El .It Sy BPF_LDX These instructions load a value into the index register. Note that the addressing modes are more restricted than those of the accumulator loads, but they include .Sy BPF_MSH , a hack for efficiently loading the IP header length. .Bl -column "BPF_LDX_BPF_W_BPF_MEM" "X <- k" -offset indent .It Sy BPF_LDX+BPF_W+BPF_IMM Ta X <- k .It Sy BPF_LDX+BPF_W+BPF_MEM Ta X <- M[k] .It Sy BPF_LDX+BPF_W+BPF_LEN Ta X <- len .It Sy BPF_LDX+BPF_B+BPF_MSH Ta X <- 4*(P[k:1]&0xf) .El .It Sy BPF_ST This instruction stores the accumulator into the scratch memory. We do not need an addressing mode since there is only one possibility for the destination. .Bl -column "BPF_ST" "M[k] <- A" -offset indent .It Sy BPF_ST Ta M[k] <- A .El .It Sy BPF_STX This instruction stores the index register in the scratch memory store. .Bl -column "BPF_STX" "M[k] <- X" -offset indent .It Sy BPF_STX Ta M[k] <- X .El .It Sy BPF_ALU The alu instructions perform operations between the accumulator and index register or constant, and store the result back in the accumulator. For binary operations, a source mode is required .Sy ( BPF_K or .Sy BPF_X ) . .Bl -column "BPF_ALU_BPF_ADD_BPF_K" "A <- A + k" -offset indent .It Sy BPF_ALU+BPF_ADD+BPF_K Ta A <- A + k .It Sy BPF_ALU+BPF_SUB+BPF_K Ta A <- A - k .It Sy BPF_ALU+BPF_MUL+BPF_K Ta A <- A * k .It Sy BPF_ALU+BPF_DIV+BPF_K Ta A <- A / k .It Sy BPF_ALU+BPF_AND+BPF_K Ta A <- A & k .It Sy BPF_ALU+BPF_OR+BPF_K Ta A <- A | k .It Sy BPF_ALU+BPF_LSH+BPF_K Ta A <- A << k .It Sy BPF_ALU+BPF_RSH+BPF_K Ta A <- A >> k .It Sy BPF_ALU+BPF_ADD+BPF_X Ta A <- A + X .It Sy BPF_ALU+BPF_SUB+BPF_X Ta A <- A - X .It Sy BPF_ALU+BPF_MUL+BPF_X Ta A <- A * X .It Sy BPF_ALU+BPF_DIV+BPF_X Ta A <- A / X .It Sy BPF_ALU+BPF_AND+BPF_X Ta A <- A & X .It Sy BPF_ALU+BPF_OR+BPF_X Ta A <- A | X .It Sy BPF_ALU+BPF_LSH+BPF_X Ta A <- A << X .It Sy BPF_ALU+BPF_RSH+BPF_X Ta A <- A >> X .It Sy BPF_ALU+BPF_NEG Ta A <- -A .El .It Sy BPF_JMP The jump instructions alter flow of control. Conditional jumps compare the accumulator against a constant .Sy ( BPF_K ) or the index register .Sy ( BPF_X ) . If the result is true (or non-zero), the true branch is taken, otherwise the false branch is taken. Jump offsets are encoded in 8 bits so the longest jump is 256 instructions. However, the jump always .Sy ( BPF_JA ) opcode uses the 32 bit .Va k field as the offset, allowing arbitrarily distant destinations. All conditionals use unsigned comparison conventions. .Bl -column "BPF_JMP+BPF_JSET+BPF_K" "pc += (A \*[Ge] k) ? jt : jf" -offset indent .It Sy BPF_JMP+BPF_JA Ta pc += k .It Sy BPF_JMP+BPF_JGT+BPF_K Ta "pc += (A > k) ? jt : jf" .It Sy BPF_JMP+BPF_JGE+BPF_K Ta "pc += (A \*[Ge] k) ? jt : jf" .It Sy BPF_JMP+BPF_JEQ+BPF_K Ta "pc += (A == k) ? jt : jf" .It Sy BPF_JMP+BPF_JSET+BPF_K Ta "pc += (A & k) ? jt : jf" .It Sy BPF_JMP+BPF_JGT+BPF_X Ta "pc += (A > X) ? jt : jf" .It Sy BPF_JMP+BPF_JGE+BPF_X Ta "pc += (A \*[Ge] X) ? jt : jf" .It Sy BPF_JMP+BPF_JEQ+BPF_X Ta "pc += (A == X) ? jt : jf" .It Sy BPF_JMP+BPF_JSET+BPF_X Ta "pc += (A & X) ? jt : jf" .El .It Sy BPF_RET The return instructions terminate the filter program and specify the amount of packet to accept (i.e., they return the truncation amount). A return value of zero indicates that the packet should be ignored. The return value is either a constant .Sy ( BPF_K ) or the accumulator .Sy ( BPF_A ) . .Bl -column "BPF_RET+BPF_A" "accept A bytes" -offset indent .It Sy BPF_RET+BPF_A Ta accept A bytes .It Sy BPF_RET+BPF_K Ta accept k bytes .El .It Sy BPF_MISC The miscellaneous category was created for anything that doesn't fit into the above classes, and for any new instructions that might need to be added. Currently, these are the register transfer instructions that copy the index register to the accumulator or vice versa. .Bl -column "BPF_MISC+BPF_TAX" "X <- A" -offset indent .It Sy BPF_MISC+BPF_TAX Ta X <- A .It Sy BPF_MISC+BPF_TXA Ta A <- X .El .Pp Also, two instructions to call a "coprocessor" if initialized by the kernel component. There is no coprocessor by default. .Bl -column "BPF_MISC+BPF_COPX" "A <- funcs[X](...)" -offset indent .It Sy BPF_MISC+BPF_COP Ta A <- funcs[k](..) .It Sy BPF_MISC+BPF_COPX Ta A <- funcs[X](..) .El .Pp If the coprocessor is not set or the function index is out of range, these instructions will abort the program and return zero. .El .Pp The BPF interface provides the following macros to facilitate array initializers: .Bd -unfilled -offset indent .Fn BPF_STMT opcode operand .Fn BPF_JUMP opcode operand true_offset false_offset .Ed .Sh SYSCTLS The following sysctls are available when .Nm is enabled: .Bl -tag -width "XnetXbpfXmaxbufsizeXX" .It Li net.bpf.maxbufsize Sets the maximum buffer size available for .Nm peers. .It Li net.bpf.stats Shows .Nm statistics. They can be retrieved with the .Xr netstat 1 utility. .It Li net.bpf.peers Shows the current .Nm peers. This is only available to the super user and can also be retrieved with the .Xr netstat 1 utility. .El .Pp On architectures with .Xr bpfjit 4 support, the additional sysctl is available: .Bl -tag -width "XnetXbpfXjitXX" .It Li net.bpf.jit Toggle .Sy Just-In-Time compilation of new filter programs. In order to enable Just-In-Time compilation, the bpfjit kernel module must be loaded. Changing a value of this sysctl doesn't affect existing filter programs. .El .Sh FILES .Pa /dev/bpf .Sh EXAMPLES The following filter is taken from the Reverse ARP Daemon. It accepts only Reverse ARP requests. .Bd -literal -offset indent struct bpf_insn insns[] = { BPF_STMT(BPF_LD+BPF_H+BPF_ABS, 12), BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, ETHERTYPE_REVARP, 0, 3), BPF_STMT(BPF_LD+BPF_H+BPF_ABS, 20), BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, REVARP_REQUEST, 0, 1), BPF_STMT(BPF_RET+BPF_K, sizeof(struct ether_arp) + sizeof(struct ether_header)), BPF_STMT(BPF_RET+BPF_K, 0), }; .Ed .Pp This filter accepts only IP packets between host 128.3.112.15 and 128.3.112.35. .Bd -literal -offset indent struct bpf_insn insns[] = { BPF_STMT(BPF_LD+BPF_H+BPF_ABS, 12), BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, ETHERTYPE_IP, 0, 8), BPF_STMT(BPF_LD+BPF_W+BPF_ABS, 26), BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 0x8003700f, 0, 2), BPF_STMT(BPF_LD+BPF_W+BPF_ABS, 30), BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 0x80037023, 3, 4), BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 0x80037023, 0, 3), BPF_STMT(BPF_LD+BPF_W+BPF_ABS, 30), BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 0x8003700f, 0, 1), BPF_STMT(BPF_RET+BPF_K, (u_int)-1), BPF_STMT(BPF_RET+BPF_K, 0), }; .Ed .Pp Finally, this filter returns only TCP finger packets. We must parse the IP header to reach the TCP header. The .Sy BPF_JSET instruction checks that the IP fragment offset is 0 so we are sure that we have a TCP header. .Bd -literal -offset indent struct bpf_insn insns[] = { BPF_STMT(BPF_LD+BPF_H+BPF_ABS, 12), BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, ETHERTYPE_IP, 0, 10), BPF_STMT(BPF_LD+BPF_B+BPF_ABS, 23), BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, IPPROTO_TCP, 0, 8), BPF_STMT(BPF_LD+BPF_H+BPF_ABS, 20), BPF_JUMP(BPF_JMP+BPF_JSET+BPF_K, 0x1fff, 6, 0), BPF_STMT(BPF_LDX+BPF_B+BPF_MSH, 14), BPF_STMT(BPF_LD+BPF_H+BPF_IND, 14), BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 79, 2, 0), BPF_STMT(BPF_LD+BPF_H+BPF_IND, 16), BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 79, 0, 1), BPF_STMT(BPF_RET+BPF_K, (u_int)-1), BPF_STMT(BPF_RET+BPF_K, 0), }; .Ed .Sh SEE ALSO .Xr ioctl 2 , .Xr read 2 , .Xr select 2 , .Xr signal 3 , .Xr bpfjit 4 , .Xr tcpdump 8 .Rs .%T "The BSD Packet Filter: A New Architecture for User-level Packet Capture" .%A S. McCanne .%A V. Jacobson .%J Proceedings of the 1993 Winter USENIX .%C Technical Conference, San Diego, CA .Re .Sh HISTORY The Enet packet filter was created in 1980 by Mike Accetta and Rick Rashid at Carnegie-Mellon University. Jeffrey Mogul, at Stanford, ported the code to BSD and continued its development from 1983 on. Since then, it has evolved into the ULTRIX Packet Filter at DEC, a STREAMS NIT module under SunOS 4.1, and BPF. .Sh AUTHORS .An -nosplit .An Steven McCanne , of Lawrence Berkeley Laboratory, implemented BPF in Summer 1990. The design was in collaboration with .An Van Jacobson , also of Lawrence Berkeley Laboratory. .Sh BUGS The read buffer must be of a fixed size (returned by the .Dv BIOCGBLEN ioctl). .Pp A file that does not request promiscuous mode may receive promiscuously received packets as a side effect of another file requesting this mode on the same hardware interface. This could be fixed in the kernel with additional processing overhead. However, we favor the model where all files must assume that the interface is promiscuous, and if so desired, must use a filter to reject foreign packets. .Pp Under SunOS, if a BPF application reads more than 2^31 bytes of data, read will fail in .Er EINVAL . You can either fix the bug in SunOS, or lseek to 0 when read fails for this reason. .Pp .Dq Immediate mode and the .Dq read timeout are misguided features. This functionality can be emulated with non-blocking mode and .Xr select 2 .