Please read the whole thing I'm asking for a concept not the problem itself, but I have to show the problem to explain myself
Find the period of the small oscillations of a cylinder of radius r that rolls without slide inside a cylindrical surface of radius R.
I tried going with the conservation of mechanic energy $$E_m$$ which is equal to $$E_m=E_k+E_p+E_r$$ being $k, p, r$ kinetic, potential and rotational (energy). We know that $E_k=\frac{1}{2}mv^2=\frac{1}{2}m(\frac{d\theta}{dt}(R-r))^2$, being $\theta$ the angle of the small cilinder from the center of the big cilinder with respect to the vertical, and $E_p=-mg(R-r)\cos(\theta)$ but for $E_k=\frac{1}{2}I\omega_{cm}^2$ I had no idea on how to continue, since $\omega_{cm}$ here refers to the rotation of the small cylinder with respect to it's center of mass, not the rotation of the small cylinder with respect to the big cylinder so we can't do $\omega_{cm}=\frac{d\theta}{dt}$
Doubt. So I looked the solution and it said that the velocity of the small cylinder with respect to the big cylinder was equal to the velocity of the points inside the small cylinder with respect to the center of masses of the small cylinder, why is that? That makes no sense to me. The procedure they make afterwards is $v_{cm}=v_{O}\Rightarrow \omega_{cm}=\frac{R-r}{r}\omega_0=\frac{R-r}{r}\frac{d\theta}{dt}$ being $\omega_0$ the rotation of the small cylinder with respect to the center of the big cylinder.