On Thu, Nov 08, 2018 at 02:09:17PM -0600, Bjorn Helgaas wrote: > > I'm having second thoughts about this. One thing I'm uncomfortable > with is that sprinkling pci_dev_is_disconnected() around feels ad hoc > instead of systematic, in the sense that I don't know how we convince > ourselves that this (and only this) is the correct place to put it.
You know how the kernel provides ZERO_PAGE, wouldn't it be cool if we also had a ONES_PAGE and could remap all virtual addresses from a memory mapped device to that page on an ungraceful disconnect? I do not know how to accomplish that, so might just be crazy talk... But if it is possible, that would be a pretty nifty way to solve this problem.
