func TestCarryPropagate(t *testing.T) {
	asmLikeGeneric := func(a [5]uint64) bool {
		t1 := &Element{a[0], a[1], a[2], a[3], a[4]}
		t2 := &Element{a[0], a[1], a[2], a[3], a[4]}

		t1.carryPropagate()
		t2.carryPropagateGeneric()

		if *t1 != *t2 {
			t.Logf("got: %#v,\nexpected: %#v", t1, t2)
		}

		return *t1 == *t2 && isInBounds(t2)
	}

	if err := quick.Check(asmLikeGeneric, quickCheckConfig(1024)); err != nil {

	// assembly. Probably because the body of this function is so simple that
	// the compiler can figure out better optimizations by inlining the carry
	// propagation.
	return v.carryPropagateGeneric()
}

// Subtract sets v = a - b, and returns v.
func (v *Element) Subtract(a, b *Element) *Element {
	// We first add 2 * p, to guarantee the subtraction won't underflow, and