2022-09-27 21:59:48 +03:00
.. contents ::
.. sectnum ::
========================================
eBPF Instruction Set Specification, v1.0
========================================
This document specifies version 1.0 of the eBPF instruction set.
2021-11-19 19:32:15 +03:00
2021-12-23 13:19:04 +03:00
Registers and calling convention
================================
eBPF has 10 general purpose registers and a read-only frame pointer register,
all of which are 64-bits wide.
The eBPF calling convention is defined as:
2022-09-27 21:59:48 +03:00
* R0: return value from function calls, and exit value for eBPF programs
* R1 - R5: arguments for function calls
* R6 - R9: callee saved registers that function calls will preserve
* R10: read-only frame pointer to access stack
2021-12-23 13:19:04 +03:00
R0 - R5 are scratch registers and eBPF programs needs to spill/fill them if
necessary across calls.
2021-11-19 19:32:15 +03:00
2022-01-03 21:35:51 +03:00
Instruction encoding
====================
2022-01-31 21:36:37 +03:00
eBPF has two instruction encodings:
2022-09-27 21:59:48 +03:00
* the basic instruction encoding, which uses 64 bits to encode an instruction
* the wide instruction encoding, which appends a second 64-bit immediate value
(imm64) after the basic instruction for a total of 128 bits.
2022-01-31 21:36:37 +03:00
The basic instruction encoding looks as follows:
2022-01-03 21:35:51 +03:00
2022-09-27 21:59:48 +03:00
============= ======= =============== ==================== ============
32 bits (MSB) 16 bits 4 bits 4 bits 8 bits (LSB)
============= ======= =============== ==================== ============
immediate offset source register destination register opcode
============= ======= =============== ==================== ============
2022-01-03 21:35:51 +03:00
Note that most instructions do not use all of the fields.
Unused fields shall be cleared to zero.
2021-12-23 13:19:05 +03:00
Instruction classes
2022-01-03 21:35:51 +03:00
-------------------
2021-11-19 19:32:15 +03:00
2021-12-23 13:19:05 +03:00
The three LSB bits of the 'opcode' field store the instruction class:
2021-11-19 19:32:15 +03:00
2022-09-27 21:59:48 +03:00
========= ===== =============================== ===================================
class value description reference
========= ===== =============================== ===================================
BPF_LD 0x00 non-standard load operations `Load and store instructions`_
BPF_LDX 0x01 load into register operations `Load and store instructions`_
BPF_ST 0x02 store from immediate operations `Load and store instructions`_
BPF_STX 0x03 store from register operations `Load and store instructions`_
BPF_ALU 0x04 32-bit arithmetic operations `Arithmetic and jump instructions`_
BPF_JMP 0x05 64-bit jump operations `Arithmetic and jump instructions`_
BPF_JMP32 0x06 32-bit jump operations `Arithmetic and jump instructions`_
BPF_ALU64 0x07 64-bit arithmetic operations `Arithmetic and jump instructions`_
========= ===== =============================== ===================================
2021-11-19 19:32:15 +03:00
2021-12-23 13:19:05 +03:00
Arithmetic and jump instructions
================================
2022-09-27 21:59:48 +03:00
For arithmetic and jump instructions (`` BPF_ALU `` , `` BPF_ALU64 `` , `` BPF_JMP `` and
`` BPF_JMP32 `` ), the 8-bit 'opcode' field is divided into three parts:
2021-11-19 19:32:15 +03:00
2022-09-27 21:59:48 +03:00
============== ====== =================
4 bits (MSB) 1 bit 3 bits (LSB)
============== ====== =================
operation code source instruction class
============== ====== =================
2021-11-19 19:32:15 +03:00
2021-12-23 13:19:05 +03:00
The 4th bit encodes the source operand:
2021-11-19 19:32:15 +03:00
2021-12-23 13:19:05 +03:00
====== ===== ========================================
source value description
====== ===== ========================================
BPF_K 0x00 use 32-bit immediate as source operand
BPF_X 0x08 use 'src_reg' register as source operand
====== ===== ========================================
2021-11-19 19:32:15 +03:00
2021-12-23 13:19:05 +03:00
The four MSB bits store the operation code.
2021-11-19 19:32:15 +03:00
2022-01-03 21:35:52 +03:00
Arithmetic instructions
-----------------------
2022-09-27 21:59:48 +03:00
`` BPF_ALU `` uses 32-bit wide operands while `` BPF_ALU64 `` uses 64-bit wide operands for
2022-01-03 21:35:52 +03:00
otherwise identical operations.
2022-09-27 21:59:48 +03:00
The 'code' field encodes the operation as below:
======== ===== ==========================================================
code value description
======== ===== ==========================================================
BPF_ADD 0x00 dst += src
BPF_SUB 0x10 dst -= src
BPF_MUL 0x20 dst \*= src
BPF_DIV 0x30 dst /= src
BPF_OR 0x40 dst \|= src
BPF_AND 0x50 dst &= src
BPF_LSH 0x60 dst <<= src
BPF_RSH 0x70 dst >>= src
BPF_NEG 0x80 dst = ~src
BPF_MOD 0x90 dst %= src
BPF_XOR 0xa0 dst ^= src
BPF_MOV 0xb0 dst = src
BPF_ARSH 0xc0 sign extending shift right
BPF_END 0xd0 byte swap operations (see `Byte swap instructions`_ below)
======== ===== ==========================================================
`` BPF_ADD | BPF_X | BPF_ALU `` means::
2022-01-03 21:35:52 +03:00
dst_reg = (u32) dst_reg + (u32) src_reg;
2022-09-27 21:59:48 +03:00
`` BPF_ADD | BPF_X | BPF_ALU64 `` means::
2022-01-03 21:35:52 +03:00
dst_reg = dst_reg + src_reg
2022-09-27 21:59:48 +03:00
`` BPF_XOR | BPF_K | BPF_ALU `` means::
2022-01-03 21:35:52 +03:00
src_reg = (u32) src_reg ^ (u32) imm32
2022-09-27 21:59:48 +03:00
`` BPF_XOR | BPF_K | BPF_ALU64 `` means::
2022-01-03 21:35:52 +03:00
src_reg = src_reg ^ imm32
2022-01-31 21:36:34 +03:00
Byte swap instructions
2022-09-27 21:59:48 +03:00
~~~~~~~~~~~~~~~~~~~~~~
2022-01-31 21:36:34 +03:00
2022-06-09 11:39:37 +03:00
The byte swap instructions use an instruction class of `` BPF_ALU `` and a 4-bit
2022-09-27 21:59:48 +03:00
'code' field of `` BPF_END `` .
2022-01-31 21:36:34 +03:00
2022-04-28 12:55:52 +03:00
The byte swap instructions operate on the destination register
2022-01-31 21:36:34 +03:00
only and do not use a separate source register or immediate value.
2022-08-29 09:52:39 +03:00
The 1-bit source operand field in the opcode is used to select what byte
2022-01-31 21:36:34 +03:00
order the operation convert from or to:
2022-09-27 21:59:48 +03:00
========= ===== =================================================
source value description
========= ===== =================================================
BPF_TO_LE 0x00 convert between host byte order and little endian
BPF_TO_BE 0x08 convert between host byte order and big endian
========= ===== =================================================
2022-01-31 21:36:34 +03:00
2022-09-27 21:59:48 +03:00
The 'imm' field encodes the width of the swap operations. The following widths
2022-01-31 21:36:34 +03:00
are supported: 16, 32 and 64.
Examples:
`` BPF_ALU | BPF_TO_LE | BPF_END `` with imm = 16 means::
dst_reg = htole16(dst_reg)
`` BPF_ALU | BPF_TO_BE | BPF_END `` with imm = 64 means::
dst_reg = htobe64(dst_reg)
2022-01-03 21:35:52 +03:00
Jump instructions
-----------------
2022-09-27 21:59:48 +03:00
`` BPF_JMP32 `` uses 32-bit wide operands while `` BPF_JMP `` uses 64-bit wide operands for
2022-01-03 21:35:52 +03:00
otherwise identical operations.
2022-09-27 21:59:48 +03:00
The 'code' field encodes the operation as below:
======== ===== ========================= ============
code value description notes
======== ===== ========================= ============
BPF_JA 0x00 PC += off BPF_JMP only
BPF_JEQ 0x10 PC += off if dst == src
BPF_JGT 0x20 PC += off if dst > src unsigned
BPF_JGE 0x30 PC += off if dst >= src unsigned
BPF_JSET 0x40 PC += off if dst & src
BPF_JNE 0x50 PC += off if dst != src
BPF_JSGT 0x60 PC += off if dst > src signed
BPF_JSGE 0x70 PC += off if dst >= src signed
BPF_CALL 0x80 function call
BPF_EXIT 0x90 function / program return BPF_JMP only
BPF_JLT 0xa0 PC += off if dst < src unsigned
BPF_JLE 0xb0 PC += off if dst <= src unsigned
BPF_JSLT 0xc0 PC += off if dst < src signed
BPF_JSLE 0xd0 PC += off if dst <= src signed
======== ===== ========================= ============
2021-12-23 13:19:04 +03:00
2022-01-03 21:35:52 +03:00
The eBPF program needs to store the return value into register R0 before doing a
BPF_EXIT.
2021-11-19 19:32:15 +03:00
2021-12-23 13:19:05 +03:00
Load and store instructions
===========================
2022-09-27 21:59:48 +03:00
For load and store instructions (`` BPF_LD `` , `` BPF_LDX `` , `` BPF_ST `` , and `` BPF_STX `` ), the
2021-12-23 13:19:05 +03:00
8-bit 'opcode' field is divided as:
2022-09-27 21:59:48 +03:00
============ ====== =================
3 bits (MSB) 2 bits 3 bits (LSB)
============ ====== =================
mode size instruction class
============ ====== =================
The mode modifier is one of:
============= ===== ==================================== =============
mode modifier value description reference
============= ===== ==================================== =============
BPF_IMM 0x00 64-bit immediate instructions `64-bit immediate instructions`_
BPF_ABS 0x20 legacy BPF packet access (absolute) `Legacy BPF Packet access instructions`_
BPF_IND 0x40 legacy BPF packet access (indirect) `Legacy BPF Packet access instructions`_
BPF_MEM 0x60 regular load and store operations `Regular load and store operations`_
BPF_ATOMIC 0xc0 atomic operations `Atomic operations`_
============= ===== ==================================== =============
2021-12-23 13:19:05 +03:00
The size modifier is one of:
2021-11-19 19:32:15 +03:00
2021-12-23 13:19:05 +03:00
============= ===== =====================
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)
============= ===== =====================
2021-11-19 19:32:15 +03:00
2022-01-31 21:36:35 +03:00
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::
2021-11-19 19:32:15 +03:00
2021-12-23 13:19:05 +03:00
*(size * ) (dst_reg + off) = src_reg
2021-11-19 19:32:15 +03:00
2022-01-31 21:36:35 +03:00
`` BPF_MEM | <size> | BPF_ST `` means::
2021-11-19 19:32:15 +03:00
2021-12-23 13:19:05 +03:00
*(size * ) (dst_reg + off) = imm32
2022-01-31 21:36:35 +03:00
`` BPF_MEM | <size> | BPF_LDX `` means::
2021-12-23 13:19:05 +03:00
dst_reg = *(size * ) (src_reg + off)
2022-01-31 21:36:35 +03:00
Where size is one of: `` BPF_B `` , `` BPF_H `` , `` BPF_W `` , or `` BPF_DW `` .
2021-12-23 13:19:05 +03:00
Atomic operations
-----------------
2021-11-19 19:32:15 +03:00
2022-01-31 21:36:38 +03:00
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 eBPF programs or means outside of this specification.
2021-11-19 19:32:15 +03:00
2022-01-31 21:36:38 +03:00
All atomic operations supported by eBPF are encoded as store operations
that use the `` BPF_ATOMIC `` mode modifier as follows:
2021-11-19 19:32:15 +03:00
2022-09-27 21:59:48 +03:00
* `` 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.
2021-11-19 19:32:15 +03:00
2022-09-27 21:59:48 +03:00
The 'imm' field is used to encode the actual atomic operation.
2022-01-31 21:36:38 +03:00
Simple atomic operation use a subset of the values defined to encode
2022-09-27 21:59:48 +03:00
arithmetic operations in the 'imm' field to encode the atomic operation:
2021-11-19 19:32:15 +03:00
2022-09-27 21:59:48 +03:00
======== ===== ===========
imm value description
======== ===== ===========
BPF_ADD 0x00 atomic add
BPF_OR 0x40 atomic or
BPF_AND 0x50 atomic and
BPF_XOR 0xa0 atomic xor
======== ===== ===========
2021-11-19 19:32:15 +03:00
2022-09-27 21:59:48 +03:00
`` BPF_ATOMIC | BPF_W | BPF_STX `` with 'imm' = BPF_ADD means::
2021-11-19 19:32:15 +03:00
2022-01-31 21:36:38 +03:00
*(u32 * )(dst_reg + off16) += src_reg
2021-11-19 19:32:15 +03:00
2022-09-27 21:59:48 +03:00
`` BPF_ATOMIC | BPF_DW | BPF_STX `` with 'imm' = BPF ADD means::
2021-11-19 19:32:15 +03:00
2022-01-31 21:36:38 +03:00
*(u64 * )(dst_reg + off16) += src_reg
2021-11-19 19:32:15 +03:00
2022-01-31 21:36:38 +03:00
In addition to the simple atomic operations, there also is a modifier and
two complex atomic operations:
2022-09-27 21:59:48 +03:00
=========== ================ ===========================
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
=========== ================ ===========================
2022-01-31 21:36:38 +03:00
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_reg `` with the value that
was in memory before it was modified.
The `` BPF_XCHG `` operation atomically exchanges `` src_reg `` with the value
addressed by `` dst_reg + off `` .
The `` BPF_CMPXCHG `` operation atomically compares the value addressed by
`` dst_reg + off `` with `` R0 `` . If they match, the value addressed by
`` dst_reg + off `` is replaced with `` src_reg `` . In either case, the
value that was at `` dst_reg + off `` before the operation is zero-extended
and loaded back to `` R0 `` .
2021-11-19 19:32:15 +03:00
2022-01-31 21:36:37 +03:00
64-bit immediate instructions
-----------------------------
2022-09-27 21:59:48 +03:00
Instructions with the `` BPF_IMM `` 'mode' modifier use the wide instruction
2022-01-31 21:36:37 +03:00
encoding for an extra imm64 value.
There is currently only one such instruction.
`` BPF_LD | BPF_DW | BPF_IMM `` means::
dst_reg = imm64
2021-12-23 13:19:05 +03:00
2021-12-23 13:19:06 +03:00
2022-01-31 21:36:36 +03:00
Legacy BPF Packet access instructions
-------------------------------------
2021-12-23 13:19:06 +03:00
2022-09-27 21:59:44 +03:00
eBPF 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.