Ah. Gotcha. For what it's worth, it's giving me: root at qemuarm:~# qemu-system-arm -enable-kvm -M vexpress-a9 -cpu cortex-a9 -drive if=none,file=/arm-wheezy.img,id=myimg -global virtio-blk-mmio.drive=myimg -kern el /zImage-guest -nographic -append 'root=/dev/mmcblk0 console=ttyAMA0 rootwait earlyprintk debug' Uncompressing Linux... done, booting the kernel. Error: unrecognized/unsupported processor variant (0x00000000) Which is?.unexpected?especially given that it boots without mmio? But it's too early to complain about that yet. I'll take a crack at it first . Alex On Mar 13, 2012, at 1:45 PM, Peter Maydell wrote: > 2012/3/13 Alexander Golec <thejfasi at gmail.com>: >> I've merged in Peter Maydell's virtio-mmio code onto the head, and I'm >> getting the following exchange: >> >> root at qemuarm:~# qemu-system-arm -enable-kvm -M vexpress-a15 -cpu cortex-a9 >> -drive file=/arm-wheezy.img,if=virtio -kernel /zImage-guest -nographic >> -append 'root=/dev/mmcblk0 console=ttyAMA0 rootwait earlyprintk debug' >> qemu-system-arm: drive property not set >> qemu-system-arm: Initialization of device virtio-blk-mmio failed > > Yeah, you have to manually point the virtio-mmio at the drive with > options like > "-drive if=none,file=whatever.img,id=myimg -global virtio-blk-mmio.drive=myimg" > > I mentioned this in the cover letter for my original RFC patchset > but I forgot to point you at that, sorry. Here it is: > http://lists.nongnu.org/archive/html/qemu-devel/2011-11/msg01870.html > > (You don't want '-drive if=virtio' because that will create a PCI > virtio device.) > > -- PMM -------------- next part -------------- An HTML attachment was scrubbed... URL: https://lists.cs.columbia.edu/pipermail/android-virt/attachments/20120313/0e9cc223/attachment.html
