Benchmark tips#1: https://github.com/golang/go/issues/14813

popcnt.go

package main

// Hamming weight algorithm
// https://en.wikipedia.org/wiki/Hamming_weight

const m1 = 0x5555555555555555
const m2 = 0x3333333333333333
const m4 = 0x0f0f0f0f0f0f0f0f
const h01 = 0x0101010101010101

func main() {
	popcnt(112)
}

func popcnt(x uint64) int {
	x -= (x >> 1) & m1
	x = (x & m2) + ((x >> 2) & m2)
	x = (x + (x >> 4)) & m4
	return int((x * h01) >> 56)
}

popcnt_test.go

package main

import "testing"

func BenchmarkBadPopcnt(b *testing.B) {
	for i := 0; i < b.N; i++ {
		popcnt(uint64(i))
	}
}

var result int

func BenchmarkPopcnt(b *testing.B) {
	var n int
	for i := 0; i < b.N; i++ {
		n = popcnt(uint64(i))
	}
	result = n
}

PS> go test -bench . --benchmem
goos: windows
goarch: amd64
pkg: github.com/toversus/ssa-optimize
BenchmarkBadPopcnt-8    2000000000               0.25 ns/op            0 B/op          0 allocs/op
BenchmarkPopcnt-8       1000000000               2.01 ns/op            0 B/op          0 allocs/op
PASS