mm: fix possible PMD dirty bit lost in set_pmd_migration_entry()

In set_pmd_migration_entry(), pmdp_invalidate() is used to change PMD
atomically.  But the PMD is read before that with an ordinary memory
reading.  If the THP (transparent huge page) is written between the PMD
reading and pmdp_invalidate(), the PMD dirty bit may be lost, and cause
data corruption.  The race window is quite small, but still possible in
theory, so need to be fixed.

The race is fixed via using the return value of pmdp_invalidate() to get
the original content of PMD, which is a read/modify/write atomic
operation.  So no THP writing can occur in between.

The race has been introduced when the THP migration support is added in
the commit 616b837 ("mm: thp: enable thp migration in generic path").
But this fix depends on the commit d52605d ("mm: do not lose dirty
and accessed bits in pmdp_invalidate()").  So it's easy to be backported
after v4.16.  But the race window is really small, so it may be fine not
to backport the fix at all.

Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: "Huang, Ying" <[email protected]>
Reviewed-by: Zi Yan <[email protected]>
Reviewed-by: William Kucharski <[email protected]>
Acked-by: Kirill A. Shutemov <[email protected]>
Cc: <[email protected]>
Cc: Vlastimil Babka <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Andrea Arcangeli <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Linus Torvalds <[email protected]>

Author: yhuang-intel

mm/huge_memory.c

@@ -3043,8 +3043,7 @@ void set_pmd_migration_entry(struct page_vma_mapped_walk *pvmw,
 		return;
 
 	flush_cache_range(vma, address, address + HPAGE_PMD_SIZE);
-	pmdval = *pvmw->pmd;
-	pmdp_invalidate(vma, address, pvmw->pmd);
+	pmdval = pmdp_invalidate(vma, address, pvmw->pmd);
 	if (pmd_dirty(pmdval))
 		set_page_dirty(page);
 	entry = make_migration_entry(page, pmd_write(pmdval));