Confusion about Eigenvalue Decomposition of 2D Gaussian Covariance

[pyh3552]

forward.cu Line229,230,231

float mid = 0.5f * (cov.x + cov.z);
float lambda1 = mid + sqrt(max(0.1f, mid * mid - det));
float lambda2 = mid - sqrt(max(0.1f, mid * mid - det));

I think the code here is solving the eigenvalues of a symmetric matrix:
$$\det\left([[a-\lambda,b], [b, c-\lambda]]\right)=(a-\lambda)(c-\lambda)-b^{2}=0$$
$$\lambda^{2}-(a+c)\lambda+(ac-b^{2})=0$$
$$\lambda=0.5 * \Bigg[(a+c)\pm \sqrt{(a+c)^{2}-4*(ac-b^{2})}\Bigg]$$

But the code seems to have omitted $0.5*$ and $4*$