Fisher information for $\sigma^2$

Let $x_1,x_2,\dots,x_n$ be a random sample from $N(\mu,\sigma^2)$, the parameter $\mu$ being known. If the unknown $(\theta=\sigma^2)$ parameter is $\sigma^2$ with the parametric space $\Theta=\{\theta\mid0<\theta<\infty\}$, then $\mathbb{E}_\theta\Bigl(\dfrac{\partial^2\log f_\theta(x)}{\partial\theta^2}\Bigr)$ is equal to: