1) I will see what I can do about getting support from Marvell on this item.
2) By looking at the MDIO register range (0x2090300) in the memory browser, it does appear that the MDIO interface is still alive even after the System reset (my earlier statement was incorrect), judging by the transfers occurring in the USERACCESS1 register. I see the lower 16 bits getting populated with data read. However, my interrupt does not go off for some reason. I do see the raw interrupt bit and masked interrupt bits change, but the software is not diverted to the ISR(). This could be a configuration issue on my end. I will analyze this further.