On Thu, Sep 17, 2020 at 01:35:56PM -0700, Linus Torvalds wrote: > For that to happen, we'd need to have the vma flag so that we wouldn't > have any worry about non-pinners, but as you suggested, I think even > just a mm-wide counter - or flag - to deal with the fast-bup case is > likely perfectly sufficient. Would mm_struct.pinned_vm suffice? Though now I'm not 100% certain whether all pin_user_pages*() callers are accounting correctly upon pinned_vm. My gut feeling is that bc3e53f682d93df67 does not convert all the old locked_vm users with page pinnings (e.g., mm_iommu_do_alloc, vaddr_get_pfn, etc.; didn't try harder). -- Peter Xu
