void un_m_add2f(f32 *v, f32 *a, f32 *b) {
    v[0] = a[0] + b[0];
    v[1] = a[1] + b[1];
}

void un_m_add3f(f32 *v, f32 *a, f32 *b) {
    v[0] = a[0] + b[0];
    v[1] = a[1] + b[1];
    v[2] = a[2] + b[2];
}

void un_m_add4f(f32 *v, f32 *a, f32 *b) {
    v[0] = a[0] + b[0];
    v[1] = a[1] + b[1];
    v[2] = a[2] + b[2];
    v[3] = a[3] + b[3];
}

void un_m_sub2f(f32 *v, f32 *a, f32 *b) {
    v[0] = a[0] - b[0];
    v[1] = a[1] - b[1];
}

void un_m_sub3f(f32 *v, f32 *a, f32 *b) {
    v[0] = a[0] - b[0];
    v[1] = a[1] - b[1];
    v[2] = a[2] - b[2];
}

void un_m_sub4f(f32 *v, f32 *a, f32 *b) {
    v[0] = a[0] - b[0];
    v[1] = a[1] - b[1];
    v[2] = a[2] - b[2];
    v[3] = a[3] - b[3];
}

void un_m_add_scalar2f(f32 *v, f32 *a, f32 scalar) {
    v[0] = a[0] + scalar;
    v[1] = a[1] + scalar;
}

void un_m_add_scalar3f(f32 *v, f32 *a, f32 scalar) {
    v[0] = a[0] + scalar;
    v[1] = a[1] + scalar;
    v[2] = a[2] + scalar;
}

void un_m_add_scalar4f(f32 *v, f32 *a, f32 scalar) {
    v[0] = a[0] + scalar;
    v[1] = a[1] + scalar;
    v[2] = a[2] + scalar;
    v[3] = a[3] + scalar;
}

void un_m_sub_scalar2f(f32 *v, f32 *a, f32 scalar) {
    v[0] = a[0] - scalar;
    v[1] = a[1] - scalar;
}

void un_m_sub_scalar3f(f32 *v, f32 *a, f32 scalar) {
    v[0] = a[0] - scalar;
    v[1] = a[1] - scalar;
    v[2] = a[2] - scalar;
}

void un_m_sub_scalar4f(f32 *v, f32 *a, f32 scalar) {
    v[0] = a[0] - scalar;
    v[1] = a[1] - scalar;
    v[2] = a[2] - scalar;
    v[3] = a[3] - scalar;
}

void un_m_mul_scalar2f(f32 *v, f32 *a, f32 scalar) {
    v[0] = a[0] * scalar;
    v[1] = a[1] * scalar;
}

void un_m_mul_scalar3f(f32 *v, f32 *a, f32 scalar) {
    v[0] = a[0] * scalar;
    v[1] = a[1] * scalar;
    v[2] = a[2] * scalar;
}

void un_m_mul_scalar4f(f32 *v, f32 *a, f32 scalar) {
    v[0] = a[0] * scalar;
    v[1] = a[1] * scalar;
    v[2] = a[2] * scalar;
    v[3] = a[3] * scalar;
}

void un_m_div_scalar2f(f32 *v, f32 *a, f32 scalar) {
    v[0] = a[0] / scalar;
    v[1] = a[1] / scalar;
}

void un_m_div_scalar3f(f32 *v, f32 *a, f32 scalar) {
    v[0] = a[0] / scalar;
    v[1] = a[1] / scalar;
    v[2] = a[2] / scalar;
}

void un_m_div_scalar4f(f32 *v, f32 *a, f32 scalar) {
    v[0] = a[0] / scalar;
    v[1] = a[1] / scalar;
    v[2] = a[2] / scalar;
    v[3] = a[3] / scalar;
}

void un_m_dot2f(f32 *v, f32 *a, f32 *b) {
    *v = a[0] * b[0] + a[1] * b[1];
}

void un_m_dot3f(f32 *v, f32 *a, f32 *b) {
    *v = a[0] * b[0] + a[1] * b[1] + a[2] * b[2];
}

void un_m_dot4f(f32 *v, f32 *a, f32 *b) {
    *v = a[0] * b[0] + a[1] * b[1] + a[2] * b[2] + a[3] * b[3];
}

void un_m_hadamard2f(f32 *v, f32 *a, f32 *b) {
    v[0] = a[0] * b[0];
    v[1] = a[1] * b[1];
}

void un_m_hadamard3f(f32 *v, f32 *a, f32 *b) {
    v[0] = a[0] * b[0];
    v[1] = a[1] * b[1];
    v[2] = a[2] * b[2];
}

void un_m_hadamard4f(f32 *v, f32 *a, f32 *b) {
    v[0] = a[0] * b[0];
    v[1] = a[1] * b[1];
    v[2] = a[2] * b[2];
    v[3] = a[3] * b[3];
}

void un_m_cross2f(f32 *v, f32 *a) {
    v[0] = a[1];
    v[1] =-a[0];
}

void un_m_cross3f(f32 *v, f32 *a, f32 *b) {
	v[0] = a[1] * b[2] - a[2] * b[1];
	v[1] = a[2] * b[0] - a[0] * b[2];
	v[2] = a[0] * b[1] - a[1] * b[0];
}

void un_m_magnitude2f(f32 *v, f32 *a) {
    *v = sqrtf(a[0] * a[0] + a[1] * a[1]);
}

void un_m_magnitude3f(f32 *v, f32 *a) {
    *v = sqrtf(a[0] * a[0] + a[1] * a[1] + a[2] * a[2]);
}

void un_m_magnitude4f(f32 *v, f32 *a) {
    *v = sqrtf(a[0] * a[0] + a[1] * a[1] + a[2] * a[2] + a[3] * a[3]);
}

void un_m_magnitude_sqr2f(f32 *v, f32 *a) {
    *v = a[0] * a[0] + a[1] * a[1];
}

void un_m_magnitude_sqr3f(f32 *v, f32 *a) {
    *v = a[0] * a[0] + a[1] * a[1] + a[2] * a[2];
}

void un_m_magnitude_sqr4f(f32 *v, f32 *a) {
    *v = a[0] * a[0] + a[1] * a[1] + a[2] * a[2] + a[3] * a[3];
}

void un_m_distance2f(f32 *v, f32 *a, f32 *b) {
    f32 f[2];
    un_m_sub2f(f, b, a);
    *v = sqrtf(f[0] * f[0] + f[1] * f[1]);
}

void un_m_distance3f(f32 *v, f32 *a, f32 *b) {
    f32 f[3];
    un_m_sub3f(f, b, a);
    *v = sqrtf(f[0] * f[0] + f[1] * f[1] + f[2] * f[2]);
}

void un_m_distance4f(f32 *v, f32 *a, f32 *b) {
    f32 f[4];
    un_m_sub4f(f, b, a);
    *v = sqrtf(f[0] * f[0] + f[1] * f[1] + f[2] * f[2] + f[3] * f[3]);
}

void un_m_distance_sqr2f(f32 *v, f32 *a, f32 *b) {
    f32 f[2];
    un_m_sub2f(f, b, a);
    *v = f[0] * f[0] + f[1] * f[1];
}

void un_m_distance_sqr3f(f32 *v, f32 *a, f32 *b) {
    f32 f[3];
    un_m_sub3f(f, b, a);
    *v = f[0] * f[0] + f[1] * f[1] + f[2] * f[2];
}

void un_m_distance_sqr4f(f32 *v, f32 *a, f32 *b) {
    f32 f[4];
    un_m_sub4f(f, b, a);
    *v = f[0] * f[0] + f[1] * f[1] + f[2] * f[2] + f[3] * f[3];
}

void un_m_normalize2f(f32 *v, f32 *a) {
    f32 f;
    un_m_magnitude2f(&f, a);
    v[0] = a[0] / f;
    v[1] = a[1] / f;
}

void un_m_normalize3f(f32 *v, f32 *a) {
    f32 f;
    un_m_magnitude3f(&f, a);
    v[0] = a[0] / f;
    v[1] = a[1] / f;
    v[2] = a[2] / f;
}

void un_m_normalize4f(f32 *v, f32 *a) {
    f32 f;
    un_m_magnitude4f(&f, a);
    v[0] = a[0] / f;
    v[1] = a[1] / f;
    v[2] = a[2] / f;
    v[3] = a[3] / f;
}

void un_m_project2f(f32 *v, f32 *a, f32 *onto) {
    f32 dot, magn, scale;

    un_m_dot2f(&dot, a, onto);
    un_m_magnitude_sqr2f(&magn, onto);

    if (magn < EPSILON) {
        un_memory_set((void*)v, 0, sizeof(*v) * 2);
        return;
    }

    scale = dot / magn;
    v[0] = onto[0] * scale;
    v[1] = onto[1] * scale;
}

void un_m_project3f(f32 *v, f32 *a, f32 *onto) {
    f32 dot, magn, scale;

    un_m_dot3f(&dot, a, onto);
    un_m_magnitude_sqr3f(&magn, onto);

    if (magn < EPSILON) {
        un_memory_set((void*)v, 0, sizeof(*v) * 3);
        return;
    }

    scale = dot / magn;
    v[0] = onto[0] * scale;
    v[1] = onto[1] * scale;
    v[2] = onto[2] * scale;
}

void un_m_project4f(f32 *v, f32 *a, f32 *onto) {
    f32 dot, magn, scale;

    un_m_dot4f(&dot, a, onto);
    un_m_magnitude_sqr4f(&magn, onto);

    if (magn < EPSILON) {
        un_memory_set((void*)v, 0, sizeof(*v) * 4);
        return;
    }

    scale = dot / magn;
    v[0] = onto[0] * scale;
    v[1] = onto[1] * scale;
    v[2] = onto[2] * scale;
    v[3] = onto[3] * scale;
}

void un_m_reflect2f(f32 *v, f32 *a, f32 *normal) {
    f32 dot;

    un_m_dot2f(&dot, a, normal);
    un_m_mul_scalar2f(v, normal, 2.0f * dot);
    un_m_sub2f(v, v, a);
}

void un_m_reflect3f(f32 *v, f32 *a, f32 *normal) {
    f32 dot;

    un_m_dot3f(&dot, a, normal);
    un_m_mul_scalar3f(v, normal, 2.0f * dot);
    un_m_sub3f(v, v, a);
}

void un_m_reflect4f(f32 *v, f32 *a, f32 *normal) {
    f32 dot;

    un_m_dot4f(&dot, a, normal);
    un_m_mul_scalar4f(v, normal, 2.0f * dot);
    un_m_sub4f(v, v, a);
}