mm/ksm.c: fix inconsistent accounting of zero pages

When using KSM with use_zero_pages, we replace anonymous pages
containing only zeroes with actual zero pages, which are not anonymous.
We need to do proper accounting of the mm counters, otherwise we will
get wrong values in /proc and a BUG message in dmesg when tearing down
the mm.

Link: http://lkml.kernel.org/r/1522931274-15552-1-git-send-email-imbrenda@linux.vnet.ibm.com
Fixes: e86c59b ("mm/ksm: improve deduplication of zero pages with colouring")
Signed-off-by: Claudio Imbrenda <imbrenda@linux.vnet.ibm.com>
Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: Gerald Schaefer <gerald.schaefer@de.ibm.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/ksm.c

@@ -1131,6 +1131,13 @@ static int replace_page(struct vm_area_struct *vma, struct page *page,
 	} else {
 		newpte = pte_mkspecial(pfn_pte(page_to_pfn(kpage),
 					       vma->vm_page_prot));
+		/*
+		 * We're replacing an anonymous page with a zero page, which is
+		 * not anonymous. We need to do proper accounting otherwise we
+		 * will get wrong values in /proc, and a BUG message in dmesg
+		 * when tearing down the mm.
+		 */
+		dec_mm_counter(mm, MM_ANONPAGES);
 	}
 
 	flush_cache_page(vma, addr, pte_pfn(*ptep));