We have had a couple of CYPD3175 IC's fail in a prototype dedicated charging port design using the direct feedback system. The most faults we've had is that the chips stop responding to a plug insertion. We verify that the internal core is running because we can measure 1.8V and the CC line is pulled up to +3.3V before insertion and drops to 1.7V or so during insertion. These are prototype boards that were working until this sudden fault.
We've also had the controller IC suddenly start running very hot and fail after working with no problems for two weeks. We only noticed the failure because after a plug insertion the port turned on and off contiuously. We remove the cable and reinsterted it and the controller chip started to heat up and stop running.
We have studied the switch mode power supply for stability and the phase and gain margins are according to industry standards.