tianon
commented
2 years ago I'm not sure what the difference is (and I'm not the best for debugging QEMU issues), but here's the command my Windows 10 container ends up running (which is doing extra work to get nested Hyper-V so I can run Windows Docker containers inside):
qemu-system-x86_64 -enable-kvm -smp cores=4 -m 4G -drive file=windows.qcow2,index=0,media=disk,discard=unmap,detect-zeroes=unmap,if=none,id=hda -device virtio-scsi-pci -device scsi-hd,drive=hda -netdev user,hostname=qemu-windows,hostfwd=tcp::22-:22,hostfwd=udp::22-:22,hostfwd=tcp::2375-:2375,hostfwd=udp::2375-:2375,id=net -device virtio-net-pci,netdev=net -vnc :0 -cpu Haswell,hv_relaxed,hv_spinlocks=0x1fff,hv_vapic,hv_time,+vmx -monitor stdio -fda virtio.vfd -usbdevice tablet
At a glance, it looks like you're missing KVM, which is definitely going to be at least a performance issue (if not related to your page fault).
This currently boots the windows screen then pagefaults and restarts.
The container is running the command
qemu-system-x86_64 -smp 3 -m 3072 -drive file=/tmp/hda.qcow2,index=0,media=disk,discard=unmap,detect-zeroes=unmap,if=none,id=hda -device virtio-scsi-pci -device scsi-hd,drive=hda -cdrom /tmp/debian.iso -boot order=d -netdev user,hostname=b999a51aa5d6,hostfwd=tcp::22-:22,hostfwd=udp::22-:22,hostfwd=tcp::2375-:2375,hostfwd=udp::2375-:2375,hostfwd=tcp::2376-:2376,hostfwd=udp::2376-:2376,hostfwd=tcp::3389-:3389,hostfwd=udp::3389-:3389,id=net -device virtio-net-pci,netdev=net -serial stdio -vnc :0