mm/memory-failure.c: fix wrong num_poisoned_pages in handling memory error on thp

num_poisoned_pages counts up the number of pages isolated by memory
errors.  But for thp, only one subpage is isolated because memory error
handler splits it, so it's wrong to add (1 << compound_trans_order).

Signed-off-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

diff --git a/mm/memory-failure.c b/mm/memory-failure.c
index 9cab165fd668e3caeeb2e2a72976432cdb938fff..9b1e5e7078a1f6bcb0bd8fb08d0a1b0593c908a0 100644 (file)
--- a/mm/memory-failure.c
+++ b/mm/memory-failure.c
@@ -1039,7 +1039,17 @@ int memory_failure(unsigned long pfn, int trapno, int flags)
                return 0;
        }
 
-       nr_pages = 1 << compound_trans_order(hpage);
+       /*
+        * Currently errors on hugetlbfs pages are contained in hugepage
+        * unit, so nr_pages should be 1 << compound_order. OTOH when
+        * errors are on transparent hugepages, they are supposed to be
+        * split and error containment is done in normal page unit.
+        * So nr_pages should be one in this case.
+        */
+       if (PageHuge(p))
+               nr_pages = 1 << compound_order(hpage);
+       else /* normal page or thp */
+               nr_pages = 1;
        atomic_long_add(nr_pages, &num_poisoned_pages);
 
        /*