On Mon, Mar 16, 2020 at 09:10:53AM -0300, Jason Gunthorpe wrote: > On Mon, Mar 16, 2020 at 10:13:47AM +0100, Christoph Hellwig wrote: > > On Wed, Mar 11, 2020 at 03:35:06PM -0300, Jason Gunthorpe wrote: > > > From: Jason Gunthorpe <jgg@xxxxxxxxxxxx> > > > > > > Currently if a special PTE is encountered hmm_range_fault() immediately > > > returns EFAULT and sets the HMM_PFN_SPECIAL error output (which nothing > > > uses). > > > > > > EFAULT should only be returned after testing with hmm_pte_need_fault(). > > > > > > Also pte_devmap() and pte_special() are exclusive, and there is no need to > > > check IS_ENABLED, pte_special() is stubbed out to return false on > > > unsupported architectures. > > > > I think the right fix is to just kill HMM_PFN_SPECIAL and treat any > > fault on special ptes that aren't the zero page as an error. > > I have another series that is doing that - this change is to make the > next series make sense and not introduce new control logic too. Ok. I had some cleanups like this based of older trees, but if you are active in this area I think I'll let you handle it. > Even when this is switched to ERROR it still needs to have the > hmm_range_fault() logic this patch introduces. True.
