avo: A Go repository from mmcloughlin

Generate x86 Assembly with Go

avo makes high-performance Go assembly easier to write, review and maintain. The avo package presents a familiar assembly-like interface that simplifies development without sacrificing performance:

Use Go control structures for assembly generation; avo programs are Go programs
Register allocation: write functions with virtual registers and avo assigns physical registers for you
Automatically load arguments and store return values: ensure memory offsets are correct for complex structures
Generation of stub files to interface with your Go package

For more about avo:

Introductory talk "Better x86 Assembly Generation with Go" at dotGo 2019 (slides)
Longer tutorial at Gophercon 2019 showing a highly-optimized dot product (slides)
Watch Filippo Valsorda live code the rewrite of filippo.io/edwards25519 assembly with avo
Explore projects using avo
Discuss avo and general Go assembly topics in the #assembly channel of Gophers Slack

Note: APIs subject to change while avo is still in an experimental phase. You can use it to build real things but we suggest you pin a version with your package manager of choice.

Quick Start

Install avo with go get:

$ go get -u github.com/mmcloughlin/avo

avo assembly generators are pure Go programs. Here's a function that adds two uint64 values:

//go:build ignore

package main

import . "github.com/mmcloughlin/avo/build"

func main() {
	TEXT("Add", NOSPLIT, "func(x, y uint64) uint64")
	Doc("Add adds x and y.")
	x := Load(Param("x"), GP64())
	y := Load(Param("y"), GP64())
	ADDQ(x, y)
	Store(y, ReturnIndex(0))
	RET()
	Generate()
}

go run this code to see the assembly output. To integrate this into the rest of your Go package we recommend a go:generate line to produce the assembly and the corresponding Go stub file.

//go:generate go run asm.go -out add.s -stubs stub.go

After running go generate the add.s file will contain the Go assembly.

// Code generated by command: go run asm.go -out add.s -stubs stub.go. DO NOT EDIT.

#include "textflag.h"

// func Add(x uint64, y uint64) uint64
TEXT ·Add(SB), NOSPLIT, $0-24
	MOVQ x+0(FP), AX
	MOVQ y+8(FP), CX
	ADDQ AX, CX
	MOVQ CX, ret+16(FP)
	RET

The same call will produce the stub file stub.go which will enable the function to be called from your Go code.

// Code generated by command: go run asm.go -out add.s -stubs stub.go. DO NOT EDIT.

package add

// Add adds x and y.
func Add(x uint64, y uint64) uint64

See the examples/add directory for the complete working example.

Examples

See examples for the full suite of examples.

Slice Sum

Sum a slice of uint64s:

func main() {
	TEXT("Sum", NOSPLIT, "func(xs []uint64) uint64")
	Doc("Sum returns the sum of the elements in xs.")
	ptr := Load(Param("xs").Base(), GP64())
	n := Load(Param("xs").Len(), GP64())

	Comment("Initialize sum register to zero.")
	s := GP64()
	XORQ(s, s)

	Label("loop")
	Comment("Loop until zero bytes remain.")
	CMPQ(n, Imm(0))
	JE(LabelRef("done"))

	Comment("Load from pointer and add to running sum.")
	ADDQ(Mem{Base: ptr}, s)

	Comment("Advance pointer, decrement byte count.")
	ADDQ(Imm(8), ptr)
	DECQ(n)
	JMP(LabelRef("loop"))

	Label("done")
	Comment("Store sum to return value.")
	Store(s, ReturnIndex(0))
	RET()
	Generate()
}

The result from this code generator is:

// Code generated by command: go run asm.go -out sum.s -stubs stub.go. DO NOT EDIT.

#include "textflag.h"

// func Sum(xs []uint64) uint64
TEXT ·Sum(SB), NOSPLIT, $0-32
	MOVQ xs_base+0(FP), AX
	MOVQ xs_len+8(FP), CX

	// Initialize sum register to zero.
	XORQ DX, DX

loop:
	// Loop until zero bytes remain.
	CMPQ CX, $0x00
	JE   done

	// Load from pointer and add to running sum.
	ADDQ (AX), DX

	// Advance pointer, decrement byte count.
	ADDQ $0x08, AX
	DECQ CX
	JMP  loop

done:
	// Store sum to return value.
	MOVQ DX, ret+24(FP)
	RET

Full example at examples/sum.

Features

For demonstrations of avo features:

args: Loading function arguments.
returns: Building return values.
complex: Working with complex{64,128} types.
data: Defining DATA sections.
ext: Interacting with types from external packages.
pragma: Apply compiler directives to generated functions.

Real Examples

Implementations of full algorithms:

sha1: SHA-1 cryptographic hash.
fnv1a: FNV-1a hash function.
dot: Vector dot product.
md5x16: AVX-512 accelerated MD5.
geohash: Integer geohash encoding.
stadtx: StadtX hash port from dgryski/go-stadtx.

Adopters

Popular projects¹ using avo:

golang / go ⭐ 123.2k

The Go programming language

klauspost / compress ⭐ 4.7k

Optimized Go Compression Packages

golang / crypto ⭐ 3k

[mirror] Go supplementary cryptography libraries

klauspost / reedsolomon ⭐ 1.9k

Reed-Solomon Erasure Coding in Go

bytedance / gopkg ⭐ 1.7k

Universal Utilities for Go

cloudflare / circl ⭐ 1.3k

CIRCL: Cloudflare Interoperable Reusable Cryptographic Library

segmentio / asm ⭐ 869

Go library providing algorithms optimized to leverage the characteristics of modern CPUs

zeebo / xxh3 ⭐ 401

XXH3 algorithm in Go

zeebo / blake3 ⭐ 391

Pure Go implementation of BLAKE3 with AVX2 and SSE4.1 acceleration

lukechampine / blake3 ⭐ 354

An AVX-512 accelerated implementation of the BLAKE3 cryptographic hash function

See the full list of projects using avo.

Contributing

Contributions to avo are welcome:

Feedback from using avo in a real project is incredibly valuable. Consider porting an existing project to avo.
Submit bug reports to the issues page.
Pull requests accepted. Take a look at outstanding issues for ideas (especially the "good first issue" label).
Join us in the #assembly channel of Gophers Slack.

Credits

Inspired by the PeachPy and asmjit projects. Thanks to Damian Gryski for advice, and his extensive library of PeachPy Go projects.

License

avo is available under the BSD 3-Clause License.

Projects drawn from the avo third-party test suite. Popularity estimated from Github star count collected on Oct 1, 2024. ↩

mmcloughlin/avo