comparing var big.NewInt(0) vs allocating big.NewInt(0) everytime

cmp_test.go

package main

import (
	"math/big"
	"testing"
)

var bns = []*big.Int{
	big.NewInt(0),
	big.NewInt(10),
	big.NewInt(10000),
}

func BenchmarkRawAllocCmp(b *testing.B) {
	var recv bool
	for i := 0; i < b.N; i++ {
		for _, bn := range bns {
			recv = cmpRawAlloc(bn)
		}
	}
	b.ReportAllocs()

	if recv {
		recv = false
	}
}

func BenchmarkVarCmp(b *testing.B) {
	var recv bool
	for i := 0; i < b.N; i++ {
		for _, bn := range bns {
			recv = cmpVar(bn)
		}
	}
	b.ReportAllocs()

	if recv {
		recv = false
	}
}

func cmpRawAlloc(bn *big.Int) bool {
	return bn.Cmp(big.NewInt(0)) == 0
}

var bigZero = big.NewInt(0)

func cmpVar(bn *big.Int) bool {
	return bn.Cmp(bigZero) == 0
}

results.txt

$ go test -v -run=^$ -bench=.
goos: darwin
goarch: amd64
BenchmarkRawAllocCmp-4   	10000000	       168 ns/op	      96 B/op	       3 allocs/op
BenchmarkVarCmp-4        	50000000	        29.1 ns/op	       0 B/op	       0 allocs/op
PASS