On 15 October 2013 16:14, Tom Sutcliffe <tom.sutcliffe@xxxxxxxxxxx> wrote: > > On 15 Oct 2013, at 16:00, Peter Maydell <peter.maydell@xxxxxxxxxx> wrote: > >> On 15 October 2013 15:58, Tom Sutcliffe <tom.sutcliffe@xxxxxxxxxxx> wrote: >>> Thumbs up from me testing on Arndale. My only issue is that virt and vexpress-a15 add virtio-mmio devices in the opposite order to each other, for the same set of -device command line arguments. It would avoid future headaches if we could have these behave the same. My preference would be for the virt behaviour, as the -device order matches the order in which the guest Linux kernel adds them to /dev (for virtio-blk-devices at least). >> >> Oh yes, I'd forgotten you mentioned that. Did anybody ever >> track down *why* the kernel is reading the device tree >> backwards? > > Not me :) So apparently the kernel makes no guarantees at all about what order it might process the virtio-mmio transports in. This means that users mustn't rely on /dev/vda and /dev/vdb corresponding to particular virtio-blk devices on QEMU's command line -- you need to use UUIDs or something similar instead. I think this sucks, but that's the kernel for you. I'll probably change QEMU anyway, just because if there's no guarantee we might as well make qemu code do a simple forwards loop rather than a backwards one. -- PMM _______________________________________________ kvmarm mailing list kvmarm@xxxxxxxxxxxxxxxxxxxxx https://lists.cs.columbia.edu/cucslists/listinfo/kvmarm
