An article in El Reg about this was previously discussed on HN [0], but this one has some more informative links and details. It's an interesting example of how even extremely simple, mostly legacy parts of core silicon (the Low Pin Count bus in this case) that shouldn't actually be strictly critical can sometimes in usage yield surprisingly serious failure modes. The impact this time may also be less generally obvious to the public vs issues like the 2011 series 6 chipset degradation, but result in more irritation for the HN crowd. Since the primary target usage is more infrastructure applications like routers/microservers/NAS just physically getting at everything could be aggravating depending on deployment.
0: https://news.ycombinator.com/item?id=13585048