On Sat, 1 Dec 2018 at 09:28, <[email protected]> wrote: > Peter Maydell wrote: > >This kind of thing is why a PCI device is cleaner -- it just > >works on any machine with a PCI controller, it by default is > >something that the user can provide if they want to and not if > >they don't, and it's not a custom UI that management layers > >might need to special-case. The guest does need to do a bit > >of PCI probing and setup initially, but then it can just leave > >the MMIO BAR permanently mapped and write to that address > >when the guest panics.
> Usually, it doesn't matter if there is this pvpanic device or not, so I > haven't considered configurability from the beginning. It matters to people who don't want it, because it increases the attack surface presented to the guest. > I didn't expect you to strongly recommend the use of PCI devices. > Doing so is not just going to roll back some of the code in the kernel... > I'll think about your suggestions. My view on this hasn't changed since October: https://lists.nongnu.org/archive/html/qemu-devel/2018-10/msg03861.html if this really really needs to be an MMIO device then it can be done, but a PCI device is a much cleaner interface which is more widely usable than just the virt board. thanks -- PMM
