Internet-Draft eBPF ISA October 2023
Thaler Expires 21 April 2024 [Page]
Workgroup:
Network Working Group
Internet-Draft:
draft-ietf-bpf-isa-00
Published:
Intended Status:
Standards Track
Expires:
Author:
D. Thaler, Ed.
Microsoft

BPF Instruction Set Specification, v1.0

Abstract

This document specifies version 1.0 of the BPF instruction set.

Status of This Memo

This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79.

Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet-Drafts is at https://datatracker.ietf.org/drafts/current/.

Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress."

This Internet-Draft will expire on 21 April 2024.

▲

Table of Contents

1. Documentation conventions

For brevity and consistency, this document refers to families of types using a shorthand syntax and refers to several expository, mnemonic functions when describing the semantics of instructions. The range of valid values for those types and the semantics of those functions are defined in the following subsections.

1.1. Types

This document refers to integer types with the notation SN to specify a type's signedness (S) and bit width (N), respectively.

Table 1: Meaning of signedness notation.
S Meaning

u

unsigned

s

signed
Table 2: Meaning of bit-width notation.
N Bit width

8

8 bits

16

16 bits

32

32 bits

64

64 bits

128

128 bits

For example, u32 is a type whose valid values are all the 32-bit unsigned numbers and s16 is a types whose valid values are all the 16-bit signed numbers.

1.2. Functions

  • htobe16: Takes an unsigned 16-bit number in host-endian format and returns the equivalent number as an unsigned 16-bit number in big-endian format.
  • htobe32: Takes an unsigned 32-bit number in host-endian format and returns the equivalent number as an unsigned 32-bit number in big-endian format.
  • htobe64: Takes an unsigned 64-bit number in host-endian format and returns the equivalent number as an unsigned 64-bit number in big-endian format.
  • htole16: Takes an unsigned 16-bit number in host-endian format and returns the equivalent number as an unsigned 16-bit number in little-endian format.
  • htole32: Takes an unsigned 32-bit number in host-endian format and returns the equivalent number as an unsigned 32-bit number in little-endian format.
  • htole64: Takes an unsigned 64-bit number in host-endian format and returns the equivalent number as an unsigned 64-bit number in little-endian format.
  • bswap16: Takes an unsigned 16-bit number in either big- or little-endian format and returns the equivalent number with the same bit width but opposite endianness.
  • bswap32: Takes an unsigned 32-bit number in either big- or little-endian format and returns the equivalent number with the same bit width but opposite endianness.
  • bswap64: Takes an unsigned 64-bit number in either big- or little-endian format and returns the equivalent number with the same bit width but opposite endianness.

1.3. Definitions

Sign Extend

To sign extend an X -bit number, A, to a Y -bit number, B , means to

  1. Copy all X bits from A to the lower X bits of B.
  2. Set the value of the remaining Y - X bits of B to the value of the most-significant bit of A.

2. Instruction encoding

BPF has two instruction encodings:

The fields conforming an encoded basic instruction are stored in the following order:

  opcode:8 src_reg:4 dst_reg:4 offset:16 imm:32 // In little-endian BPF.
  opcode:8 dst_reg:4 src_reg:4 offset:16 imm:32 // In big-endian BPF.
imm

signed integer immediate value

offset

signed integer offset used with pointer arithmetic

src_reg

the source register number (0-10), except where otherwise specified (64-bit immediate instructions (Section 4.4) reuse this field for other purposes)

dst_reg

destination register number (0-10)

opcode

operation to perform

Note that the contents of multi-byte fields ('imm' and 'offset') are stored using big-endian byte ordering in big-endian BPF and little-endian byte ordering in little-endian BPF.

For example:

  opcode                  offset imm          assembly
         src_reg dst_reg
  07     0       1        00 00  44 33 22 11  r1 += 0x11223344 // little
         dst_reg src_reg
  07     1       0        00 00  11 22 33 44  r1 += 0x11223344 // big

Note that most instructions do not use all of the fields. Unused fields shall be cleared to zero.

As discussed below in 64-bit immediate instructions (Section 4.4), a 64-bit immediate instruction uses a 64-bit immediate value that is constructed as follows. The 64 bits following the basic instruction contain a pseudo instruction using the same format but with opcode, dst_reg, src_reg, and offset all set to zero, and imm containing the high 32 bits of the immediate value.

This is depicted in the following figure:

        basic_instruction
  .-----------------------------.
  |                             |
  code:8 regs:8 offset:16 imm:32 unused:32 imm:32
                                 |              |
                                 '--------------'
                                pseudo instruction

Thus the 64-bit immediate value is constructed as follows:

imm64 = (next_imm << 32) | imm

where 'next_imm' refers to the imm value of the pseudo instruction following the basic instruction. The unused bytes in the pseudo instruction are reserved and shall be cleared to zero.

2.1. Instruction classes

The three LSB bits of the 'opcode' field store the instruction class:

Table 3
class value description reference

BPF_LD

0x00

non-standard load operations

Load and store instructions (Section 4)

BPF_LDX

0x01

load into register operations

Load and store instructions (Section 4)

BPF_ST

0x02

store from immediate operations

Load and store instructions (Section 4)

BPF_STX

0x03

store from register operations

Load and store instructions (Section 4)

BPF_ALU

0x04

32-bit arithmetic operations

Arithmetic and jump instructions (Section 3)

BPF_JMP

0x05

64-bit jump operations

Arithmetic and jump instructions (Section 3)

BPF_JMP32

0x06

32-bit jump operations

Arithmetic and jump instructions (Section 3)

BPF_ALU64

0x07

64-bit arithmetic operations

Arithmetic and jump instructions (Section 3)

3. Arithmetic and jump instructions

For arithmetic and jump instructions (BPF_ALU, BPF_ALU64, BPF_JMP and BPF_JMP32), the 8-bit 'opcode' field is divided into three parts:

Table 4
4 bits (MSB) 1 bit 3 bits (LSB)

code

source

instruction class
code

the operation code, whose meaning varies by instruction class

source

the source operand location, which unless otherwise specified is one of:

Table 5
source value description

BPF_K

0x00

use 32-bit 'imm' value as source operand

BPF_X

0x08

use 'src_reg' register value as source operand
instruction class

the instruction class (see Instruction classes (Section 2.1))

3.1. Arithmetic instructions

BPF_ALU uses 32-bit wide operands while BPF_ALU64 uses 64-bit wide operands for otherwise identical operations. The 'code' field encodes the operation as below, where 'src' and 'dst' refer to the values of the source and destination registers, respectively.

Table 6
code value offset description

BPF_ADD

0x00

0

dst += src

BPF_SUB

0x10

0

dst -= src

BPF_MUL

0x20

0

dst *= src

BPF_DIV

0x30

0

dst = (src != 0) ? (dst / src) : 0

BPF_SDIV

0x30

1

dst = (src != 0) ? (dst s/ src) : 0

BPF_OR

0x40

0

dst |= src

BPF_AND

0x50

0

dst &= src

BPF_LSH

0x60

0

dst <<= (src & mask)

BPF_RSH

0x70

0

dst >>= (src & mask)

BPF_NEG

0x80

0

dst = -dst

BPF_MOD

0x90

0

dst = (src != 0) ? (dst % src) : dst

BPF_SMOD

0x90

1

dst = (src != 0) ? (dst s% src) : dst

BPF_XOR

0xa0

0

dst ^= src

BPF_MOV

0xb0

0

dst = src

BPF_MOVSX

0xb0

8/16/32

dst = (s8,s16,s32)src

BPF_ARSH

0xc0

0

sign extending (Section 1.3) dst >>= (src & mask)

BPF_END

0xd0

0

byte swap operations (see Byte swap instructions (Section 3.2) below)

Underflow and overflow are allowed during arithmetic operations, meaning the 64-bit or 32-bit value will wrap. If BPF program execution would result in division by zero, the destination register is instead set to zero. If execution would result in modulo by zero, for BPF_ALU64 the value of the destination register is unchanged whereas for BPF_ALU the upper 32 bits of the destination register are zeroed.

BPF_ADD | BPF_X | BPF_ALU means:

  dst = (u32) ((u32) dst + (u32) src)

where '(u32)' indicates that the upper 32 bits are zeroed.

BPF_ADD | BPF_X | BPF_ALU64 means:

  dst = dst + src

BPF_XOR | BPF_K | BPF_ALU means:

  dst = (u32) dst ^ (u32) imm32

BPF_XOR | BPF_K | BPF_ALU64 means:

  dst = dst ^ imm32

Note that most instructions have instruction offset of 0. Only three instructions (BPF_SDIV, BPF_SMOD, BPF_MOVSX) have a non-zero offset.

The division and modulo operations support both unsigned and signed flavors.

For unsigned operations (BPF_DIV and BPF_MOD), for BPF_ALU, 'imm' is interpreted as a 32-bit unsigned value. For BPF_ALU64, 'imm' is first sign extended (Section 1.3) from 32 to 64 bits, and then interpreted as a 64-bit unsigned value.

For signed operations (BPF_SDIV and BPF_SMOD), for BPF_ALU, 'imm' is interpreted as a 32-bit signed value. For BPF_ALU64, 'imm' is first sign extended (Section 1.3) from 32 to 64 bits, and then interpreted as a 64-bit signed value.

Note that there are varying definitions of the signed modulo operation when the dividend or divisor are negative, where implementations often vary by language such that Python, Ruby, etc. differ from C, Go, Java, etc. This specification requires that signed modulo use truncated division (where -13 % 3 == -1) as implemented in C, Go, etc.:

a % n = a - n * trunc(a / n)

The BPF_MOVSX instruction does a move operation with sign extension. BPF_ALU | BPF_MOVSX sign extends (Section 1.3) 8-bit and 16-bit operands into 32 bit operands, and zeroes the remaining upper 32 bits. BPF_ALU64 | BPF_MOVSX sign extends (Section 1.3) 8-bit, 16-bit, and 32-bit operands into 64 bit operands.

Shift operations use a mask of 0x3F (63) for 64-bit operations and 0x1F (31) for 32-bit operations.

3.2. Byte swap instructions

The byte swap instructions use instruction classes of BPF_ALU and BPF_ALU64 and a 4-bit 'code' field of BPF_END.

The byte swap instructions operate on the destination register only and do not use a separate source register or immediate value.

For BPF_ALU, the 1-bit source operand field in the opcode is used to select what byte order the operation converts from or to. For BPF_ALU64, the 1-bit source operand field in the opcode is reserved and must be set to 0.

Table 7
class source value description

BPF_ALU

BPF_TO_LE

0x00

convert between host byte order and little endian

BPF_ALU

BPF_TO_BE

0x08

convert between host byte order and big endian

BPF_ALU64

Reserved

0x00

do byte swap unconditionally

The 'imm' field encodes the width of the swap operations. The following widths are supported: 16, 32 and 64.

Examples:

BPF_ALU | BPF_TO_LE | BPF_END with imm = 16/32/64 means:

  dst = htole16(dst)
  dst = htole32(dst)
  dst = htole64(dst)

BPF_ALU | BPF_TO_BE | BPF_END with imm = 16/32/64 means:

  dst = htobe16(dst)
  dst = htobe32(dst)
  dst = htobe64(dst)

BPF_ALU64 | BPF_TO_LE | BPF_END with imm = 16/32/64 means:

  dst = bswap16(dst)
  dst = bswap32(dst)
  dst = bswap64(dst)

3.3. Jump instructions

BPF_JMP32 uses 32-bit wide operands while BPF_JMP uses 64-bit wide operands for otherwise identical operations. The 'code' field encodes the operation as below:

Table 8
code value src description notes

BPF_JA

0x0

0x0

PC += offset

BPF_JMP class

BPF_JA

0x0

0x0

PC += imm

BPF_JMP32 class

BPF_JEQ

0x1

any

PC += offset if dst == src

BPF_JGT

0x2

any

PC += offset if dst > src

unsigned

BPF_JGE

0x3

any

PC += offset if dst >= src

unsigned

BPF_JSET

0x4

any

PC += offset if dst & src

BPF_JNE

0x5

any

PC += offset if dst != src

BPF_JSGT

0x6

any

PC += offset if dst > src

signed

BPF_JSGE

0x7

any

PC += offset if dst >= src

signed

BPF_CALL

0x8

0x0

call helper function by address

see Helper functions (Section 3.3.1)

BPF_CALL

0x8

0x1

call PC += imm

see Program-local functions (Section 3.3.2)

BPF_CALL

0x8

0x2

call helper function by BTF ID

see Helper functions (Section 3.3.1)

BPF_EXIT

0x9

0x0

return

BPF_JMP only

BPF_JLT

0xa

any

PC += offset if dst < src

unsigned

BPF_JLE

0xb

any

PC += offset if dst <= src

unsigned

BPF_JSLT

0xc

any

PC += offset if dst < src

signed

BPF_JSLE

0xd

any

PC += offset if dst <= src

signed

The BPF program needs to store the return value into register R0 before doing a BPF_EXIT.

Example:

BPF_JSGE | BPF_X | BPF_JMP32 (0x7e) means:

  if (s32)dst s>= (s32)src goto +offset

where 's>=' indicates a signed '>=' comparison.

BPF_JA | BPF_K | BPF_JMP32 (0x06) means:

  gotol +imm

where 'imm' means the branch offset comes from insn 'imm' field.

Note that there are two flavors of BPF_JA instructions. The BPF_JMP class permits a 16-bit jump offset specified by the 'offset' field, whereas the BPF_JMP32 class permits a 32-bit jump offset specified by the 'imm' field. A > 16-bit conditional jump may be converted to a < 16-bit conditional jump plus a 32-bit unconditional jump.

3.3.1. Helper functions

Helper functions are a concept whereby BPF programs can call into a set of function calls exposed by the underlying platform.

Historically, each helper function was identified by an address encoded in the imm field. The available helper functions may differ for each program type, but address values are unique across all program types.

Platforms that support the BPF Type Format (BTF) support identifying a helper function by a BTF ID encoded in the imm field, where the BTF ID identifies the helper name and type.

3.3.2. Program-local functions

Program-local functions are functions exposed by the same BPF program as the caller, and are referenced by offset from the call instruction, similar to BPF_JA. The offset is encoded in the imm field of the call instruction. A BPF_EXIT within the program-local function will return to the caller.

4. Load and store instructions

For load and store instructions (BPF_LD, BPF_LDX, BPF_ST, and BPF_STX), the 8-bit 'opcode' field is divided as:

Table 9
3 bits (MSB) 2 bits 3 bits (LSB)

mode

size

instruction class

The mode modifier is one of:

Table 10
mode modifier value description reference

BPF_IMM

0x00

64-bit immediate instructions

64-bit immediate instructions (Section 4.4)

BPF_ABS

0x20

legacy BPF packet access (absolute)

Legacy BPF Packet access instructions (Section 4.5)

BPF_IND

0x40

legacy BPF packet access (indirect)

Legacy BPF Packet access instructions (Section 4.5)

BPF_MEM

0x60

regular load and store operations

Regular load and store operations (Section 4.1)

BPF_MEMSX

0x80

sign-extension load operations

Sign-extension load operations (Section 4.2)

BPF_ATOMIC

0xc0

atomic operations

Atomic operations (Section 4.3)

The size modifier is one of:

Table 11
size modifier value description

BPF_W

0x00

word (4 bytes)

BPF_H

0x08

half word (2 bytes)

BPF_B

0x10

byte

BPF_DW

0x18

double word (8 bytes)

4.1. Regular load and store operations

The BPF_MEM mode modifier is used to encode regular load and store instructions that transfer data between a register and memory.

BPF_MEM | <size> | BPF_STX means:

  *(size *) (dst + offset) = src

BPF_MEM | <size> | BPF_ST means:

  *(size *) (dst + offset) = imm32

BPF_MEM | <size> | BPF_LDX means:

  dst = *(unsigned size *) (src + offset)

Where size is one of: BPF_B, BPF_H, BPF_W, or BPF_DW and 'unsigned size' is one of u8, u16, u32 or u64.

4.2. Sign-extension load operations

The BPF_MEMSX mode modifier is used to encode sign-extension (Section 1.3) load instructions that transfer data between a register and memory.

BPF_MEMSX | <size> | BPF_LDX means:

  dst = *(signed size *) (src + offset)

Where size is one of: BPF_B, BPF_H or BPF_W, and 'signed size' is one of s8, s16 or s32.

4.3. Atomic operations

Atomic operations are operations that operate on memory and can not be interrupted or corrupted by other access to the same memory region by other BPF programs or means outside of this specification.

All atomic operations supported by BPF are encoded as store operations that use the BPF_ATOMIC mode modifier as follows:

  • BPF_ATOMIC | BPF_W | BPF_STX for 32-bit operations
  • BPF_ATOMIC | BPF_DW | BPF_STX for 64-bit operations
  • 8-bit and 16-bit wide atomic operations are not supported.

The 'imm' field is used to encode the actual atomic operation. Simple atomic operation use a subset of the values defined to encode arithmetic operations in the 'imm' field to encode the atomic operation:

Table 12
imm value description

BPF_ADD

0x00

atomic add

BPF_OR

0x40

atomic or

BPF_AND

0x50

atomic and

BPF_XOR

0xa0

atomic xor

BPF_ATOMIC | BPF_W | BPF_STX with 'imm' = BPF_ADD means:

  *(u32 *)(dst + offset) += src

BPF_ATOMIC | BPF_DW | BPF_STX with 'imm' = BPF ADD means:

  *(u64 *)(dst + offset) += src

In addition to the simple atomic operations, there also is a modifier and two complex atomic operations:

Table 13
imm value description

BPF_FETCH

0x01

modifier: return old value

BPF_XCHG

0xe0 | BPF_FETCH

atomic exchange

BPF_CMPXCHG

0xf0 | BPF_FETCH

atomic compare and exchange

The BPF_FETCH modifier is optional for simple atomic operations, and always set for the complex atomic operations. If the BPF_FETCH flag is set, then the operation also overwrites src with the value that was in memory before it was modified.

The BPF_XCHG operation atomically exchanges src with the value addressed by dst + offset.

The BPF_CMPXCHG operation atomically compares the value addressed by dst + offset with R0. If they match, the value addressed by dst + offset is replaced with src. In either case, the value that was at dst + offset before the operation is zero-extended and loaded back to R0.

4.4. 64-bit immediate instructions

Instructions with the BPF_IMM 'mode' modifier use the wide instruction encoding defined in Instruction encoding (Section 2), and use the 'src' field of the basic instruction to hold an opcode subtype.

The following table defines a set of BPF_IMM | BPF_DW | BPF_LD instructions with opcode subtypes in the 'src' field, using new terms such as "map" defined further below:

Table 14
opcode construction opcode src pseudocode imm type dst type

BPF_IMM | BPF_DW | BPF_LD

0x18

0x0

dst = imm64

integer

integer

BPF_IMM | BPF_DW | BPF_LD

0x18

0x1

dst = map_by_fd(imm)

map fd

map

BPF_IMM | BPF_DW | BPF_LD

0x18

0x2

dst = map_val(map_by_fd(imm)) + next_imm

map fd

data pointer

BPF_IMM | BPF_DW | BPF_LD

0x18

0x3

dst = var_addr(imm)

variable id

data pointer

BPF_IMM | BPF_DW | BPF_LD

0x18

0x4

dst = code_addr(imm)

integer

code pointer

BPF_IMM | BPF_DW | BPF_LD

0x18

0x5

dst = map_by_idx(imm)

map index

map

BPF_IMM | BPF_DW | BPF_LD

0x18

0x6

dst = map_val(map_by_idx(imm)) + next_imm

map index

data pointer

where

  • map_by_fd(imm) means to convert a 32-bit file descriptor into an address of a map (see Maps (Section 4.4.1))
  • map_by_idx(imm) means to convert a 32-bit index into an address of a map
  • map_val(map) gets the address of the first value in a given map
  • var_addr(imm) gets the address of a platform variable (see Platform Variables (Section 4.4.2)) with a given id
  • code_addr(imm) gets the address of the instruction at a specified relative offset in number of (64-bit) instructions
  • the 'imm type' can be used by disassemblers for display
  • the 'dst type' can be used for verification and JIT compilation purposes

4.4.1. Maps

Maps are shared memory regions accessible by BPF programs on some platforms. A map can have various semantics as defined in a separate document, and may or may not have a single contiguous memory region, but the 'map_val(map)' is currently only defined for maps that do have a single contiguous memory region.

Each map can have a file descriptor (fd) if supported by the platform, where 'map_by_fd(imm)' means to get the map with the specified file descriptor. Each BPF program can also be defined to use a set of maps associated with the program at load time, and 'map_by_idx(imm)' means to get the map with the given index in the set associated with the BPF program containing the instruction.

4.4.2. Platform Variables

Platform variables are memory regions, identified by integer ids, exposed by the runtime and accessible by BPF programs on some platforms. The 'var_addr(imm)' operation means to get the address of the memory region identified by the given id.

4.5. Legacy BPF Packet access instructions

BPF previously introduced special instructions for access to packet data that were carried over from classic BPF. However, these instructions are deprecated and should no longer be used.

5. IANA Considerations

This document proposes a new IANA registry for BPF instructions, as follows:

6. Acknowledgements

This draft was generated from instruction-set.rst in the Linux kernel repository, to which a number of other individuals have authored contributions over time, including Akhil Raj, Alexei Starovoitov, Brendan Jackman, Christoph Hellwig, Daniel Borkmann, Ilya Leoshkevich, Jiong Wang, Jose E. Marchesi, Kosuke Fujimoto, Shahab Vahedi, Tiezhu Yang, Will Hawkins, and Zheng Yejian, with review and suggestions by many others including Alan Jowett, Andrii Nakryiko, David Vernet, Jim Harris, Quentin Monnet, Song Liu, Shung-Hsi Yu, Stanislav Fomichev, and Yonghong Song.

7. Appendix

Initial values for the BPF Instruction registry are given below. The descriptions in this table are informative. In case of any discrepancy, the reference is authoritative.

Table 15
opcode src imm offset description reference

0x00

0x0

any

0

(additional immediate value)

64-bit immediate instructions (Section 4.4)

0x04

0x0

any

0

dst = (u32)((u32)dst + (u32)imm)

Arithmetic instructions (Section 3.1)

0x05

0x0

0x00

any

goto +offset

Jump instructions (Section 3.3)

0x06

0x0

any

0

goto +imm

Jump instructions (Section 3.3)

0x07

0x0

any

0

dst += imm

Arithmetic instructions (Section 3.1)

0x0c

any

0x00

0

dst = (u32)((u32)dst + (u32)src)

Arithmetic instructions (Section 3.1)

0x0f

any

0x00

0

dst += src

Arithmetic instructions (Section 3.1)

0x14

0x0

any

0

dst = (u32)((u32)dst - (u32)imm)

Arithmetic instructions (Section 3.1)

0x15

0x0

any

any

if dst == imm goto +offset

Jump instructions (Section 3.3)

0x16

0x0

any

any

if (u32)dst == imm goto +offset

Jump instructions (Section 3.3)

0x17

0x0

any

0

dst -= imm

Arithmetic instructions (Section 3.1)

0x18

0x0

any

0

dst = imm64

64-bit immediate instructions (Section 4.4)

0x18

0x1

any

0

dst = map_by_fd(imm)

64-bit immediate instructions (Section 4.4)

0x18

0x2

any

0

dst = mva(map_by_fd(imm)) + next_imm

64-bit immediate instructions (Section 4.4)

0x18

0x3

any

0

dst = variable_addr(imm)

64-bit immediate instructions (Section 4.4)

0x18

0x4

any

0

dst = code_addr(imm)

64-bit immediate instructions (Section 4.4)

0x18

0x5

any

0

dst = map_by_idx(imm)

64-bit immediate instructions (Section 4.4)

0x18

0x6

any

0

dst = mva(map_by_idx(imm)) + next_imm

64-bit immediate instructions (Section 4.4)

0x1c

any

0x00

0

dst = (u32)((u32)dst - (u32)src)

Arithmetic instructions (Section 3.1)

0x1d

any

0x00

any

if dst == src goto +offset

Jump instructions (Section 3.3)

0x1e

any

0x00

any

if (u32)dst == (u32)src goto +offset

Jump instructions (Section 3.3)

0x1f

any

0x00

0

dst -= src

Arithmetic instructions (Section 3.1)

0x20

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x21

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x22

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x23

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x24

0x0

any

0

dst = (u32)(dst * imm)

Arithmetic instructions (Section 3.1)

0x25

0x0

any

any

if dst > imm goto +offset

Jump instructions (Section 3.3)

0x26

0x0

any

any

if (u32)dst > imm goto +offset

Jump instructions (Section 3.3)

0x27

0x0

any

0

dst *= imm

Arithmetic instructions (Section 3.1)

0x28

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x29

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x2a

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x2b

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x2c

any

0x00

0

dst = (u32)(dst * src)

Arithmetic instructions (Section 3.1)

0x2d

any

0x00

any

if dst > src goto +offset

Jump instructions (Section 3.3)

0x2e

any

0x00

any

if (u32)dst > (u32)src goto +offset

Jump instructions (Section 3.3)

0x2f

any

0x00

0

dst *= src

Arithmetic instructions (Section 3.1)

0x30

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x31

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x32

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x33

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x34

0x0

any

0

dst = (u32)((imm != 0) ? ((u32)dst / (u32)imm) : 0)

Arithmetic instructions (Section 3.1)

0x34

0x0

any

1

dst = (u32)((imm != 0) ? ((s32)dst s/ imm) : 0)

Arithmetic instructions (Section 3.1)

0x35

0x0

any

any

if dst >= imm goto +offset

Jump instructions (Section 3.3)

0x36

0x0

any

any

if (u32)dst >= imm goto +offset

Jump instructions (Section 3.3)

0x37

0x0

any

0

dst = (imm != 0) ? (dst / (u32)imm) : 0

Arithmetic instructions (Section 3.1)

0x37

0x0

any

1

dst = (imm != 0) ? (dst s/ imm) : 0

Arithmetic instructions (Section 3.1)

0x38

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x39

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x3a

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x3b

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x3c

any

any

0

dst = (u32)((src != 0) ? ((u32)dst / (u32)src) : 0)

Arithmetic instructions (Section 3.1)

0x3c

any

any

1

dst = (u32)((src != 0) ? ((s32)dst s/(s32)src) : 0)

Arithmetic instructions (Section 3.1)

0x3d

any

0x00

any

if dst >= src goto +offset

Jump instructions (Section 3.3)

0x3e

any

0x00

any

if (u32)dst >= (u32)src goto +offset

Jump instructions (Section 3.3)

0x3f

any

0x00

0

dst = (src != 0) ? (dst / src) : 0

Arithmetic instructions (Section 3.1)

0x3f

any

0x00

1

dst = (src != 0) ? (dst s/ src) : 0

Arithmetic instructions (Section 3.1)

0x40

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x41

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x42

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x43

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x44

0x0

any

0

dst = (u32)(dst | imm)

Arithmetic instructions (Section 3.1)

0x45

0x0

any

any

if dst & imm goto +offset

Jump instructions (Section 3.3)

0x46

0x0

any

any

if (u32)dst & imm goto +offset

Jump instructions (Section 3.3)

0x47

0x0

any

0

dst |= imm

Arithmetic instructions (Section 3.1)

0x48

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x49

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x4a

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x4b

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x4c

any

0x00

0

dst = (u32)(dst | src)

Arithmetic instructions (Section 3.1)

0x4d

any

0x00

any

if dst & src goto +offset

Jump instructions (Section 3.3)

0x4e

any

0x00

any

if (u32)dst & (u32)src goto +offset

Jump instructions (Section 3.3)

0x4f

any

0x00

0

dst |= src

Arithmetic instructions (Section 3.1)

0x50

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x51

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x52

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x53

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x54

0x0

any

0

dst = (u32)(dst & imm)

Arithmetic instructions (Section 3.1)

0x55

0x0

any

any

if dst != imm goto +offset

Jump instructions (Section 3.3)

0x56

0x0

any

any

if (u32)dst != imm goto +offset

Jump instructions (Section 3.3)

0x57

0x0

any

0

dst &= imm

Arithmetic instructions (Section 3.1)

0x58

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x59

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x5a

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x5b

any

any

any

(deprecated, implementation-specific)

Legacy BPF Packet access instructions (Section 4.5)

0x5c

any

0x00

0

dst = (u32)(dst & src)

Arithmetic instructions (Section 3.1)

0x5d

any

0x00

any

if dst != src goto +offset

Jump instructions (Section 3.3)

0x5e

any

0x00

any

if (u32)dst != (u32)src goto +offset

Jump instructions (Section 3.3)

0x5f

any

0x00

0

dst &= src

Arithmetic instructions (Section 3.1)

0x61

any

0x00

any

dst = *(u32 *)(src + offset)

Load and store instructions (Section 4)

0x62

0x0

any

any

*(u32 *)(dst + offset) = imm

Load and store instructions (Section 4)

0x63

any

0x00

any

*(u32 *)(dst + offset) = src

Load and store instructions (Section 4)

0x64

0x0

any

0

dst = (u32)(dst << imm)

Arithmetic instructions (Section 3.1)

0x65

0x0

any

any

if dst s> imm goto +offset

Jump instructions (Section 3.3)

0x66

0x0

any

any

if (s32)dst s> (s32)imm goto +offset

Jump instructions (Section 3.3)

0x67

0x0

any

0

dst <<= imm

Arithmetic instructions (Section 3.1)

0x69

any

0x00

any

dst = *(u16 *)(src + offset)

Load and store instructions (Section 4)

0x6a

0x0

any

any

*(u16 *)(dst + offset) = imm

Load and store instructions (Section 4)

0x6b

any

0x00

any

*(u16 *)(dst + offset) = src

Load and store instructions (Section 4)

0x6c

any

0x00

0

dst = (u32)(dst << src)

Arithmetic instructions (Section 3.1)

0x6d

any

0x00

any

if dst s> src goto +offset

Jump instructions (Section 3.3)

0x6e

any

0x00

any

if (s32)dst s> (s32)src goto +offset

Jump instructions (Section 3.3)

0x6f

any

0x00

0

dst <<= src

Arithmetic instructions (Section 3.1)

0x71

any

0x00

any

dst = *(u8 *)(src + offset)

Load and store instructions (Section 4)

0x72

0x0

any

any

*(u8 *)(dst + offset) = imm

Load and store instructions (Section 4)

0x73

any

0x00

any

*(u8 *)(dst + offset) = src

Load and store instructions (Section 4)

0x74

0x0

any

0

dst = (u32)(dst >> imm)

Arithmetic instructions (Section 3.1)

0x75

0x0

any

any

if dst s>= imm goto +offset

Jump instructions (Section 3.3)

0x76

0x0

any

any

if (s32)dst s>= (s32)imm goto +offset

Jump instructions (Section 3.3)

0x77

0x0

any

0

dst >>= imm

Arithmetic instructions (Section 3.1)

0x79

any

0x00

any

dst = *(u64 *)(src + offset)

Load and store instructions (Section 4)

0x7a

0x0

any

any

*(u64 *)(dst + offset) = imm

Load and store instructions (Section 4)

0x7b

any

0x00

any

*(u64 *)(dst + offset) = src

Load and store instructions (Section 4)

0x7c

any

0x00

0

dst = (u32)(dst >> src)

Arithmetic instructions (Section 3.1)

0x7d

any

0x00

any

if dst s>= src goto +offset

Jump instructions (Section 3.3)

0x7e

any

0x00

any

if (s32)dst s>= (s32)src goto +offset

Jump instructions (Section 3.3)

0x7f

any

0x00

0

dst >>= src

Arithmetic instructions (Section 3.1)

0x84

0x0

0x00

0

dst = (u32)-dst

Arithmetic instructions (Section 3.1)

0x85

0x0

any

0

call helper function by address

Helper functions (Section 3.3.1)

0x85

0x1

any

0

call PC += imm

Program-local functions (Section 3.3.2)

0x85

0x2

any

0

call helper function by BTF ID

Helper functions (Section 3.3.1)

0x87

0x0

0x00

0

dst = -dst

Arithmetic instructions (Section 3.1)

0x94

0x0

any

0

dst = (u32)((imm != 0)?((u32)dst % (u32)imm) : dst)

Arithmetic instructions (Section 3.1)

0x94

0x0

any

1

dst = (u32)((imm != 0) ? ((s32)dst s% imm) : dst)

Arithmetic instructions (Section 3.1)

0x95

0x0

0x00

0

return

Jump instructions (Section 3.3)

0x97

0x0

any

0

dst = (imm != 0) ? (dst % (u32)imm) : dst

Arithmetic instructions (Section 3.1)

0x97

0x0

any

1

dst = (imm != 0) ? (dst s% imm) : dst

Arithmetic instructions (Section 3.1)

0x9c

any

0x00

0

dst = (u32)((src != 0)?((u32)dst % (u32)src) : dst)

Arithmetic instructions (Section 3.1)

0x9c

any

0x00

1

dst = (u32)((src != 0)?((s32)dst s% (s32)src) :dst)

Arithmetic instructions (Section 3.1)

0x9f

any

0x00

0

dst = (src != 0) ? (dst % src) : dst

Arithmetic instructions (Section 3.1)

0x9f

any

0x00

1

dst = (src != 0) ? (dst s% src) : dst

Arithmetic instructions (Section 3.1)

0xa4

0x0

any

0

dst = (u32)(dst ^ imm)

Arithmetic instructions (Section 3.1)

0xa5

0x0

any

any

if dst < imm goto +offset

Jump instructions (Section 3.3)

0xa6

0x0

any

any

if (u32)dst < imm goto +offset

Jump instructions (Section 3.3)

0xa7

0x0

any

0

dst ^= imm

Arithmetic instructions (Section 3.1)

0xac

any

0x00

0

dst = (u32)(dst ^ src)

Arithmetic instructions (Section 3.1)

0xad

any

0x00

any

if dst < src goto +offset

Jump instructions (Section 3.3)

0xae

any

0x00

any

if (u32)dst < (u32)src goto +offset

Jump instructions (Section 3.3)

0xaf

any

0x00

0

dst ^= src

Arithmetic instructions (Section 3.1)

0xb4

0x0

any

0

dst = (u32) imm

Arithmetic instructions (Section 3.1)

0xb4

0x0

any

8

dst = (u32) (s32) (s8) imm

Arithmetic instructions (Section 3.1)

0xb4

0x0

any

16

dst = (u32) (s32) (s16) imm

Arithmetic instructions (Section 3.1)

0xb5

0x0

any

any

if dst <= imm goto +offset

Jump instructions (Section 3.3)

0xb6

0x0

any

any

if (u32)dst <= imm goto +offset

Jump instructions (Section 3.3)

0xb7

0x0

any

0

dst = imm

Arithmetic instructions (Section 3.1)

0xb7

0x0

any

8

dst = (s64) (s8) imm

Arithmetic instructions (Section 3.1)

0xb7

0x0

any

16

dst = (s64) (s16) imm

Arithmetic instructions (Section 3.1)

0xb7

0x0

any

32

dst = (s64) (s32) imm

Arithmetic instructions (Section 3.1)

0xbc

any

0x00

0

dst = (u32) src

Arithmetic instructions (Section 3.1)

0xbc

any

0x00

8

dst = (u32) (s32) (s8) src

Arithmetic instructions (Section 3.1)

0xbc

any

0x00

16

dst = (u32) (s32) (s16) src

Arithmetic instructions (Section 3.1)

0xbd

any

0x00

any

if dst <= src goto +offset

Jump instructions (Section 3.3)

0xbe

any

0x00

any

if (u32)dst <= (u32)src goto +offset

Jump instructions (Section 3.3)

0xbf

any

0x00

0

dst = src

Arithmetic instructions (Section 3.1)

0xbf

any

0x00

8

dst = (s64) (s8) src

Arithmetic instructions (Section 3.1)

0xbf

any

0x00

16

dst = (s64) (s16) src

Arithmetic instructions (Section 3.1)

0xbf

any

0x00

32

dst = (s64) (s32) src

Arithmetic instructions (Section 3.1)

0xc3

any

0x00

any

lock *(u32 *)(dst + offset) += src

Atomic operations (Section 4.3)

0xc3

any

0x01

any

lock:

     *(u32 *)(dst + offset) += src
     src = *(u32 *)(dst + offset)

Atomic operations (Section 4.3)

0xc3

any

0x40

any

*(u32 *)(dst + offset) |= src

Atomic operations (Section 4.3)

0xc3

any

0x41

any

lock:

     *(u32 *)(dst + offset) |= src
     src = *(u32 *)(dst + offset)

Atomic operations (Section 4.3)

0xc3

any

0x50

any

*(u32 *)(dst + offset) &= src

Atomic operations (Section 4.3)

0xc3

any

0x51

any

lock:

     *(u32 *)(dst + offset) &= src
     src = *(u32 *)(dst + offset)

Atomic operations (Section 4.3)

0xc3

any

0xa0

any

*(u32 *)(dst + offset) ^= src

Atomic operations (Section 4.3)

0xc3

any

0xa1

any

lock:

     *(u32 *)(dst + offset) ^= src
     src = *(u32 *)(dst + offset)

Atomic operations (Section 4.3)

0xc3

any

0xe1

any

lock:

     temp = *(u32 *)(dst + offset)
     *(u32 *)(dst + offset) = src
     src = temp

Atomic operations (Section 4.3)

0xc3

any

0xf1

any

lock:

     temp = *(u32 *)(dst + offset)
     if *(u32)(dst + offset) == R0
        *(u32)(dst + offset) = src
     R0 = temp

Atomic operations (Section 4.3)

0xc4

0x0

any

0

dst = (u32)(dst s>> imm)

Arithmetic instructions (Section 3.1)

0xc5

0x0

any

any

if dst s< imm goto +offset

Jump instructions (Section 3.3)

0xc6

0x0

any

any

if (s32)dst s< (s32)imm goto +offset

Jump instructions (Section 3.3)

0xc7

0x0

any

0

dst s>>= imm

Arithmetic instructions (Section 3.1)

0xcc

any

0x00

0

dst = (u32)(dst s>> src)

Arithmetic instructions (Section 3.1)

0xcd

any

0x00

any

if dst s< src goto +offset

Jump instructions (Section 3.3)

0xce

any

0x00

any

if (s32)dst s< (s32)src goto +offset

Jump instructions (Section 3.3)

0xcf

any

0x00

0

dst s>>= src

Arithmetic instructions (Section 3.1)

0xd4

0x0

0x10

0

dst = htole16(dst)

Byte swap instructions (Section 3.2)

0xd4

0x0

0x20

0

dst = htole32(dst)

Byte swap instructions (Section 3.2)

0xd4

0x0

0x40

0

dst = htole64(dst)

Byte swap instructions (Section 3.2)

0xd5

0x0

any

any

if dst s<= imm goto +offset

Jump instructions (Section 3.3)

0xd6

0x0

any

any

if (s32)dst s<= (s32)imm goto +offset

Jump instructions (Section 3.3)

0xd7

0x0

0x10

0

dst = bswap16(dst)

Byte swap instructions (Section 3.2)

0xd7

0x0

0x20

0

dst = bswap32(dst)

Byte swap instructions (Section 3.2)

0xd7

0x0

0x40

0

dst = bswap64(dst)

Byte swap instructions (Section 3.2)

0xdb

any

0x00

any

lock *(u64 *)(dst + offset) += src

Atomic operations (Section 4.3)

0xdb

any

0x01

any

lock:

     *(u64 *)(dst + offset) += src
     src = *(u64 *)(dst + offset)

Atomic operations (Section 4.3)

0xdb

any

0x40

any

*(u64 *)(dst + offset) |= src

Atomic operations (Section 4.3)

0xdb

any

0x41

any

lock:

     *(u64 *)(dst + offset) |= src
     src = *(u64 *)(dst + offset)

Atomic operations (Section 4.3)

0xdb

any

0x50

any

*(u64 *)(dst + offset) &= src

Atomic operations (Section 4.3)

0xdb

any

0x51

any

lock:

     *(u64 *)(dst + offset) &= src
     src = *(u64 *)(dst + offset)

Atomic operations (Section 4.3)

0xdb

any

0xa0

any

*(u64 *)(dst + offset) ^= src

Atomic operations (Section 4.3)

0xdb

any

0xa1

any

lock:

     *(u64 *)(dst + offset) ^= src
     src = *(u64 *)(dst + offset)

Atomic operations (Section 4.3)

0xdb

any

0xe1

any

lock:

     temp = *(u64 *)(dst + offset)
     *(u64 *)(dst + offset) = src
     src = temp

Atomic operations (Section 4.3)

0xdb

any

0xf1

any

lock:

     temp = *(u64 *)(dst + offset)
     if *(u64)(dst + offset) == R0
        *(u64)(dst + offset) = src
     R0 = temp

Atomic operations (Section 4.3)

0xdc

0x0

0x10

0

dst = htobe16(dst)

Byte swap instructions (Section 3.2)

0xdc

0x0

0x20

0

dst = htobe32(dst)

Byte swap instructions (Section 3.2)

0xdc

0x0

0x40

0

dst = htobe64(dst)

Byte swap instructions (Section 3.2)

0xdd

any

0x00

any

if dst s<= src goto +offset

Jump instructions (Section 3.3)

0xde

any

0x00

any

if (s32)dst s<= (s32)src goto +offset

Jump instructions (Section 3.3)

8. Normative References

[RFC8126]
"Guidelines for Writing an IANA Considerations Section in RFCs", RFC 8126, <https://www.rfc-editor.org/rfc/rfc8126>.

Author's Address

Dave Thaler (editor)
Microsoft
Redmond, WA 98052
United States of America