9.7.4 Opcodes

In the instruction descriptions below the following field descriptors are used:

%d

Destination general-purpose register whose role is to be destination of an operation.

%s

Source general-purpose register whose role is to be the source of an operation.

disp16

16-bit signed PC-relative offset, measured in number of 64-bit words, minus one.

disp32

32-bit signed PC-relative offset, measured in number of 64-bit words, minus one.

offset16

Signed 16-bit immediate.

imm32

Signed 32-bit immediate.

imm64

Signed 64-bit immediate.

9.7.4.1 Arithmetic instructions

The destination register in these instructions act like an accumulator.

add %d, (%s|imm32)

64-bit arithmetic addition.

sub %d, (%s|imm32)

64-bit arithmetic subtraction.

mul %d, (%s|imm32)

64-bit arithmetic multiplication.

div %d, (%s|imm32)

64-bit arithmetic integer division.

mod %d, (%s|imm32)

64-bit integer remainder.

and %d, (%s|imm32)

64-bit bit-wise “and” operation.

or %d, (%s|imm32)

64-bit bit-wise “or” operation.

xor %d, (%s|imm32)

64-bit bit-wise exclusive-or operation.

lsh %d, (%s|imm32)

64-bit left shift, by %s or imm32 bits.

rsh %d, (%s|imm32)

64-bit right logical shift, by %s or imm32 bits.

arsh %d, (%s|imm32)

64-bit right arithmetic shift, by %s or imm32 bits.

neg %d

64-bit arithmetic negation.

mov %d, (%s|imm32)

Move the 64-bit value of %s in %d, or load imm32 in %d.

9.7.4.2 32-bit arithmetic instructions

The destination register in these instructions act as an accumulator.

add32 %d, (%s|imm32)

32-bit arithmetic addition.

sub32 %d, (%s|imm32)

32-bit arithmetic subtraction.

mul32 %d, (%s|imm32)

32-bit arithmetic multiplication.

div32 %d, (%s|imm32)

32-bit arithmetic integer division.

mod32 %d, (%s|imm32)

32-bit integer remainder.

and32 %d, (%s|imm32)

32-bit bit-wise “and” operation.

or32 %d, (%s|imm32)

32-bit bit-wise “or” operation.

xor32 %d, (%s|imm32)

32-bit bit-wise exclusive-or operation.

lsh32 %d, (%s|imm32)

32-bit left shift, by %s or imm32 bits.

rsh32 %d, (%s|imm32)

32-bit right logical shift, by %s or imm32 bits.

arsh32 %d, (%s|imm32)

32-bit right arithmetic shift, by %s or imm32 bits.

neg32 %d

32-bit arithmetic negation.

mov32 %d, (%s|imm32)

Move the 32-bit value of %s in %d, or load imm32 in %d.

9.7.4.3 Endianness conversion instructions

endle %d, (8|16|32)

Convert the 8-bit, 16-bit or 32-bit value in %d to little-endian.

endbe %d, (8|16|32)

Convert the 8-bit, 16-bit or 32-bit value in %d to big-endian.

9.7.4.4 64-bit load and pseudo maps

lddw %d, imm64

Load the given signed 64-bit immediate, or pseudo map descriptor, to the destination register %d.

lddw %d, %map_fd(N)

Load the address of the given pseudo map fd N to the destination register %d.

9.7.4.5 Load instructions for socket filters

The following instructions are intended to be used in socket filters, and are therefore not general-purpose: they make assumptions on the contents of several registers. See the file Documentation/networking/filter.txt in the Linux kernel source tree for more information.

Absolute loads:

ldabsdw imm32

Absolute 64-bit load.

ldabsw imm32

Absolute 32-bit load.

ldabsh imm32

Absolute 16-bit load.

ldabsb imm32

Absolute 8-bit load.

Indirect loads:

ldinddw %s, imm32

Indirect 64-bit load.

ldindw %s, imm32

Indirect 32-bit load.

ldindh %s, imm32

Indirect 16-bit load.

ldindb %s, imm32

Indirect 8-bit load.

9.7.4.6 Generic load/store instructions

General-purpose load and store instructions are provided for several word sizes.

Load to register instructions:

ldxdw %d, [%s+offset16]

Generic 64-bit load.

ldxw %d, [%s+offset16]

Generic 32-bit load.

ldxh %d, [%s+offset16]

Generic 16-bit load.

ldxb %d, [%s+offset16]

Generic 8-bit load.

Store from register instructions:

stxdw [%d+offset16], %s

Generic 64-bit store.

stxw [%d+offset16], %s

Generic 32-bit store.

stxh [%d+offset16], %s

Generic 16-bit store.

stxb [%d+offset16], %s

Generic 8-bit store.

Store from immediates instructions:

stddw [%d+offset16], imm32

Store immediate as 64-bit.

stdw [%d+offset16], imm32

Store immediate as 32-bit.

stdh [%d+offset16], imm32

Store immediate as 16-bit.

stdb [%d+offset16], imm32

Store immediate as 8-bit.

9.7.4.7 Jump instructions

eBPF provides the following compare-and-jump instructions, which compare the values of the two given registers, or the values of a register and an immediate, and perform a branch in case the comparison holds true.

ja %d,(%s|imm32),disp16

Jump-always.

jeq %d,(%s|imm32),disp16

Jump if equal.

jgt %d,(%s|imm32),disp16

Jump if greater.

jge %d,(%s|imm32),disp16

Jump if greater or equal.

jlt %d,(%s|imm32),disp16

Jump if lesser.

jle %d,(%s|imm32),disp16

Jump if lesser or equal.

jset %d,(%s|imm32),disp16

Jump if signed equal.

jne %d,(%s|imm32),disp16

Jump if not equal.

jsgt %d,(%s|imm32),disp16

Jump if signed greater.

jsge %d,(%s|imm32),disp16

Jump if signed greater or equal.

jslt %d,(%s|imm32),disp16

Jump if signed lesser.

jsle %d,(%s|imm32),disp16

Jump if signed lesser or equal.

A call instruction is provided in order to perform calls to other eBPF functions, or to external kernel helpers:

call (disp32|imm32)

Jump and link to the offset disp32, or to the kernel helper function identified by imm32.

Finally:

exit

Terminate the eBPF program.

9.7.4.8 Atomic instructions

Atomic exchange-and-add instructions are provided in two flavors: one for swapping 64-bit quantities and another for 32-bit quantities.

xadddw [%d+offset16],%s

Exchange-and-add a 64-bit value at the specified location.

xaddw [%d+offset16],%s

Exchange-and-add a 32-bit value at the specified location.