On Thu, Apr 15, 2021 at 01:33:24PM +0800, Lu Baolu wrote: > Hi Jason, > > On 4/14/21 7:26 PM, Jason Gunthorpe wrote: > > On Wed, Apr 14, 2021 at 02:22:09PM +0800, Lu Baolu wrote: > > > > > I still worry about supervisor pasid allocation. > > > > > > If we use iommu_sva_alloc_pasid() to allocate a supervisor pasid, which > > > mm should the pasid be set? I've ever thought about passing &init_mm to > > > iommu_sva_alloc_pasid(). But if you add "mm != current->mm", this seems > > > not to work. Or do you prefer a separated interface for supervisor pasid > > > allocation/free? > > > > Without a mm_struct it is not SVA, so don't use SVA APIs for whatever > > a 'supervisor pasid' is > > The supervisor PASID has its mm_struct. The only difference is that the > device will set priv=1 in its DMA transactions with the PASID.
Soemthing like init_mm should not be used with 'SVA' Jason
