> From: Jason Gunthorpe <jgg@xxxxxxxxxx> > Sent: Wednesday, November 16, 2022 8:33 AM > > > This is the comment: > > > > > > /* > > > * This is part of the VFIO compatibility support for VFIO_TYPE1_IOMMU. > > > That > > > * mode permits splitting a mapped area up, and then one of the splits is > > > * unmapped. Doing this normally would cause us to violate our invariant > of > > > * pairing map/unmap. Thus, to support old VFIO compatibility disable > > > support > > > * for batching consecutive PFNs. All PFNs mapped into the iommu are > done > > > in > > > * PAGE_SIZE units, not larger or smaller. > > > */ > > > static int batch_iommu_map_small(struct iommu_domain *domain, > > > unsigned long iova, phys_addr_t paddr, > > > size_t size, int prot) > > > > > > > I meant a comment in iopt_calculate_iova_alignment(). > > How about "see batch_iommu_map_small()" ? > yes, that works. btw is there a better name to reflect strict PAGE_SIZE only? disable_large_pages literally implies no >PAGE_SIZE alignment. This is the confusion easily raised why it also refers to no subpage alignment.
