On 16 November 2011 14:33, Paolo Bonzini <pbonz...@redhat.com> wrote: > On 11/14/2011 03:55 PM, Peter Maydell wrote: >> >> This set of patches implements the QEMU end of the MMIO virtio transport >> (as specified by Appendix X of the latest virtio spec from here >> http://ozlabs.org/~rusty/virtio-spec/virtio.pdf >> and implemented by patches which I think are going into Linux 3.2). > > How does this compare against hw/syborg_virtio.c?
Pawel may have more detail, but to me the significant difference is that virtio-mmio is an implementation of a specification extension agreed with the virtio spec maintainers, whereas syborg doesn't seem to be mentioned in the virtio spec anywhere, so I am unsure what it is intended to be implementing. (There are some technical differences too, like virtio-mmio allowing the guest to specify queue sizes and alignments; these mostly came out of the process of agreeing the spec extension.) -- PMM
