On 03.04.2013 11:20, Christophe Fergeau wrote: > The current handling of bus types has some issues: > - it assumes that if the design uses a disk controller hanging off > a PCI bus, then it can use virtio, which is not true for > Windows for example unless an additional driver is installed > - it checks for "ide", "sata", "virtio" bus names, but they are not > used in libosinfo, and "sata is not mentioned in libosinfo.rng > - if the bus type could not determined, falling back to an IDE > bus should be a safe guess > > This commit changes the code to guessing the best disk controller > to use, and then derives the bus type from it when needed. > One limitation of this approach is that we are currently limited to > virtio or IDE as libosinfo is not expressive enough for us to tell > if a given disk controller is IDE/SATA/SCSI/... > One way of making this distinction possible would be to attach the > PCI subclass to libosinfo device descriptions as this contains the > information we need. > --- > libvirt-designer/libvirt-designer-domain.c | 185 ++++++++++++++--------------- > 1 file changed, 91 insertions(+), 94 deletions(-) For some reason this patch cannot be applied cleanly: Applying: Rework disk bus type handling error: patch failed: libvirt-designer/libvirt-designer-domain.c:992 error: libvirt-designer/libvirt-designer-domain.c: patch does not apply Patch failed at 0001 Rework disk bus type handling Michal -- libvir-list mailing list libvir-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/libvir-list