use to determine whether it equals 0 or not.

[cdpzyafk]

r.Cmp(zeroInt) != 0 => r.Sign() != 0

func BenchmarkCmpZeroInt(b *testing.B) {
positiveVal := big.NewInt(314)
negtiveVal := big.NewInt(-314)
b.Run("cmpzero", func(b *testing.B) {
for i := 0; i < b.N; i++ {
zeroInt.Cmp(zeroInt)
}
})

b.Run("cmppositive", func(b *testing.B) {
	for i := 0; i < b.N; i++ {
		positiveVal.Cmp(zeroInt)
	}
})

b.Run("cmpnegtive", func(b *testing.B) {
	for i := 0; i < b.N; i++ {
		negtiveVal.Cmp(zeroInt)
	}
})

b.Run("signzero", func(b *testing.B) {
	for i := 0; i < b.N; i++ {
		zeroInt.Sign()
	}
})

b.Run("signpositive", func(b *testing.B) {
	for i := 0; i < b.N; i++ {
		positiveVal.Sign()
	}
})

b.Run("signnegtive", func(b *testing.B) {
	for i := 0; i < b.N; i++ {
		negtiveVal.Sign()
	}
})

}

goos: darwin
goarch: amd64
pkg: github.com/shopspring/decimal
cpu: Intel(R) Core(TM) i5-1038NG7 CPU @ 2.00GHz
BenchmarkCmpZeroInt
BenchmarkCmpZeroInt/cmpzero
BenchmarkCmpZeroInt/cmpzero-8 729437467 1.714 ns/op 0 B/op 0 allocs/op
BenchmarkCmpZeroInt/cmppositive
BenchmarkCmpZeroInt/cmppositive-8 240192255 5.657 ns/op 0 B/op 0 allocs/op
BenchmarkCmpZeroInt/cmpnegtive
BenchmarkCmpZeroInt/cmpnegtive-8 463588371 2.386 ns/op 0 B/op 0 allocs/op
BenchmarkCmpZeroInt/signzero
BenchmarkCmpZeroInt/signzero-8 1000000000 0.4847 ns/op 0 B/op 0 allocs/op
BenchmarkCmpZeroInt/signpositive
BenchmarkCmpZeroInt/signpositive-8 1000000000 0.5291 ns/op 0 B/op 0 allocs/op
BenchmarkCmpZeroInt/signnegtive
BenchmarkCmpZeroInt/signnegtive-8 1000000000 0.5308 ns/op 0 B/op 0 allocs/op

[mwoss]

If I correctly understand your PR, bigInt.Sign() == 0 is way more performant than bigInt.Cmp(zeroInt) when validating if given big.Int is a zero value. Could you just reformat your PR description and title so it's a bit more descriptive, just for future reference.

[mwoss]

Also deeply apologize for late review! :<