The machining efficiency and accuracy are two important indexes to evaluate industrial robots. Traditional methods eliminate the tracking error by designing advanced controller parameters following scheduling the feedrate profile. With the step-by-step way, it is a challenge to improve the machining efficiency and suppress tracking errors simultaneously. In this paper, a novel time-optimal and tracking error controlled motion planning method is developed to achieve high motion velocity and limit tracking errors. Firstly, based on the kinematic and dynamic models, the feedback control model of the robot joints are framed with considered of the spatial posture of robots. Secondly, a time-optimal feedrate planning model is developed with Linear Programming(LP) strategy, which involves the kinematic and dynamic constraints of the tool tip and robot joints. During this process, a sliding window strategy is designed to decrease the calculation scale and improve the computation efficiency for long toolpaths. And then, a tracking error controlled motion planning model is proposed to optimize displacement commands based on the tracking error controlled constraints. The simulation results verified the advantages of the developed algorithm in terms of the machining efficiency and motion accuracy.