In this letter, we consider the use of a continuous-time dynamic feedback controller for linear model predictive control (MPC). The controller incorporates both integral-action and anti-windup conditioning and ensures optimality through online emulation of the solution trajectory of an underlying MPC problem. We establish closed-loop stability when the controller is interconnected with an open-loop stable linear-time invariant system using the passivity argument and the invariance principle. We include a computational example to illustrate the effectiveness and the constraint handling capability of the proposed controller.