Re: [Qemu-devel] [PATCH 0/5] VMWare PVSCSI paravirtual device implementation

[Paolo Bonzini]

Il 15/03/2012 12:47, Stefan Hajnoczi ha scritto:
> What is the V2V strategy?
> 
> Supporting these devices is fine if we have a way to convert guests to
> use virtio.  But if the plan is to keep the guests on VMware pv
> devices, then that will split the development effort on support and
> optimizing I/O devices.  All the performance work going into
> virtio-net (vhost-net), virtio-blk, and possibly virtio-scsi isn't
> easy to duplicate for VMware pv devices.  Also, we cannot extend
> VMware devices easily while retaining compatibility.

virtio-scsi and vmware-pvscsi should be equivalent in terms of
performance.  Of course vhost-scsi would be another story.

I think supporting these devices is good anyway as long as they are
expected to perform decently but not necessarily as well as virtio.
After all we do support both vmware-vga and qxl.

> I'm for supporting these devices, but we really need another step to
> move migrated guests onto virtio - otherwise users will be
> disappointed with KVM's support/performance.  Can you explain a bit
> how you want these devices to be used?

I'm curious about this too.

Paolo