thp: reintroduce split_huge_page()

This patch adds implementation of split_huge_page() for new
refcountings.

Unlike previous implementation, new split_huge_page() can fail if
somebody holds GUP pin on the page.  It also means that pin on page
would prevent it from bening split under you.  It makes situation in
many places much cleaner.

The basic scheme of split_huge_page():

  - Check that sum of mapcounts of all subpage is equal to page_count()
    plus one (caller pin). Foll off with -EBUSY. This way we can avoid
    useless PMD-splits.

  - Freeze the page counters by splitting all PMD and setup migration
    PTEs.

  - Re-check sum of mapcounts against page_count(). Page's counts are
    stable now. -EBUSY if page is pinned.

  - Split compound page.

  - Unfreeze the page by removing migration entries.

Signed-off-by: Kirill A. Shutemov <[email protected]>
Tested-by: Sasha Levin <[email protected]>
Tested-by: Aneesh Kumar K.V <[email protected]>
Acked-by: Jerome Marchand <[email protected]>
Cc: Vlastimil Babka <[email protected]>
Cc: Andrea Arcangeli <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: Dave Hansen <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Rik van Riel <[email protected]>
Cc: Naoya Horiguchi <[email protected]>
Cc: Steve Capper <[email protected]>
Cc: Johannes Weiner <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Christoph Lameter <[email protected]>
Cc: David Rientjes <[email protected]>

Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

include/linux/huge_mm.h

@@ -90,8 +90,11 @@ extern bool is_vma_temporary_stack(struct vm_area_struct *vma);
 
 extern unsigned long transparent_hugepage_flags;
 
-#define split_huge_page_to_list(page, list) BUILD_BUG()
-#define split_huge_page(page) BUILD_BUG()
+int split_huge_page_to_list(struct page *page, struct list_head *list);
+static inline int split_huge_page(struct page *page)
+{
+	return split_huge_page_to_list(page, NULL);
+}
 
 void __split_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
 		unsigned long address);

include/linux/pagemap.h

@@ -394,10 +394,21 @@ static inline struct page *read_mapping_page(struct address_space *mapping,
  */
 static inline pgoff_t page_to_pgoff(struct page *page)
 {
+	pgoff_t pgoff;
+
 	if (unlikely(PageHeadHuge(page)))
 		return page->index << compound_order(page);
-	else
+
+	if (likely(!PageTransTail(page)))
 		return page->index << (PAGE_CACHE_SHIFT - PAGE_SHIFT);
+
+	/*
+	 *  We don't initialize ->index for tail pages: calculate based on
+	 *  head page
+	 */
+	pgoff = compound_head(page)->index << (PAGE_CACHE_SHIFT - PAGE_SHIFT);
+	pgoff += page - compound_head(page);
+	return pgoff;
 }
 
 /*