On 05/04/13 14:53, Paolo Bonzini wrote:
Il 04/04/2013 21:40, Osier Yang ha scritto:
>> Depending on the pool type, this should be treated as file or block.
>> You do this correctly in patch 8, but I think it is not complete. For
>> example, will your patches allow SCSI passthrough of a volume inside an
>> iSCSI pool?
> No, it will be another patch set. On top of Han cheng's patches.
scsi-generic support for iSCSI pool would need his patches, but
scsi-block shouldn't.
Oh, right, for a "lun" device, it's only valid for block type disk
currently, so it can't work for a volume type disk (and the
volume is of block type), I'm making the patch.
My impression is that you already have "half baked" support for this
feature in patches 7/8, and you should either complete it, or leave it
out of this series.
Paolo
> I'm waiting for those patches go in first, or I will rebase/update
> it if Han doesn't have enough time to do it, and do the work
> after that.