mm: re-architect the VM_UNPAGED logic
This replaces the (in my opinion horrible) VM_UNMAPPED logic with very
explicit support for a "remapped page range" aka VM_PFNMAP. It allows a
VM area to contain an arbitrary range of page table entries that the VM
never touches, and never considers to be normal pages.
Any user of "remap_pfn_range()" automatically gets this new
functionality, and doesn't even have to mark the pages reserved or
indeed mark them any other way. It just works. As a side effect, doing
mmap() on /dev/mem works for arbitrary ranges.
Sparc update from David in the next commit.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
diff --git a/mm/rmap.c b/mm/rmap.c
index 2e034a0..6389cda 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -226,8 +226,6 @@
/*
* At what user virtual address is page expected in vma? checking that the
* page matches the vma: currently only used on anon pages, by unuse_vma;
- * and by extraordinary checks on anon pages in VM_UNPAGED vmas, taking
- * care that an mmap of /dev/mem might window free and foreign pages.
*/
unsigned long page_address_in_vma(struct page *page, struct vm_area_struct *vma)
{
@@ -614,7 +612,6 @@
struct page *page;
unsigned long address;
unsigned long end;
- unsigned long pfn;
address = (vma->vm_start + cursor) & CLUSTER_MASK;
end = address + CLUSTER_SIZE;
@@ -643,15 +640,8 @@
for (; address < end; pte++, address += PAGE_SIZE) {
if (!pte_present(*pte))
continue;
-
- pfn = pte_pfn(*pte);
- if (unlikely(!pfn_valid(pfn))) {
- print_bad_pte(vma, *pte, address);
- continue;
- }
-
- page = pfn_to_page(pfn);
- BUG_ON(PageAnon(page));
+ page = vm_normal_page(vma, address, *pte);
+ BUG_ON(!page || PageAnon(page));
if (ptep_clear_flush_young(vma, address, pte))
continue;