I think to measure the mos capacitance as a function of DC voltage, e.g. the control voltage is supported in most simulators by tracing the internal variables. For oscillators the problem is that if the capacitance curve is nonlinear also the amplitude of the ocillator determinate the frequency. That is most crictical for PMOS caps. At the threshold voltage the cap change by a factor 2x over 100-200mV. If the amplitude varies also the frequency. That is a serious practical problem.