On Thu, Mar 04, 2021 at 07:20:22AM +0000, Liu, Yi L wrote: > > > However, IOMMU is a system device which has little value to be exposed > > to > > > the userspace. Not to mention the device-IOMMU affinity/topology. VFIO > > > nicely abstracts IOMMU from the userspace, why do we want to reverse > > that? > > > > The other patch was talking about a /dev/ioasid - why can't this stuff > > be run over that? > > The stuff in this patch are actually iommu domain operations, which are > finally supported by iommu domain ops. While /dev/ioasid in another patch > is created for IOASID allocation/free to fit the PASID allocation requirement > from both vSVA and vDPA. It has no idea about iommu domain and neither the > device information. Without such info, /dev/ioasid is unable to run this > stuff. Why can't it know? My point was that VFIO should interact with /dev/ioasid to exchange the information both sides need and once exchanged the control over IOMMU should be done through /dev/ioasid, not inside VFIO. Jason
