On 12/23/2011 01:42 AM, Paolo Bonzini wrote: > On 12/23/2011 12:15 AM, Daniel P. Berrange wrote: >> The 'type' attribute refers to how the host emulator deals with the >> disk. >> >> The 'device' attribute refers to what type of device hardware is exposed >> to the guest. >> >> What we're doing here is controlling whether the host emulator allows >> SG_IO. The guest visible device hardware has not changed at all. Thus >> using the 'type' attribute is the correct approach. > > You could also say that the host is still accessing the device in the > same way (via /dev/sdX). The guest visible device changes because in > one case it allows SG_IO, in the other it doesn't, even if the host > device is the same. I also tend to agree that 1) this is a guest-visible change in behavior for the same host resource, and 2) Paolo's proposals for virtio-scsi LUN pass-through present device='lun'. https://www.redhat.com/archives/libvir-list/2011-December/msg01013.html Or do we mix the two concepts? That is, type='lun' is used to control SG_IO enabling, while device='lun' is used to control LUN pass-through, and you can even have the combination type='lun' device='lun'? -- Eric Blake eblake@xxxxxxxxxx +1-919-301-3266 Libvirt virtualization library http://libvirt.org
Attachment:
signature.asc
Description: OpenPGP digital signature
-- libvir-list mailing list libvir-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/libvir-list