mem-hwpoison: fix page refcount around isolate_lru_page()

Drop first page reference only after calling isolate_lru_page() to keep
page stable reference while isolating.
Signed-off-by: NKonstantin Khlebnikov <khlebnikov@openvz.org>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Rik van Riel <riel@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

mm/memory-failure.c

@@ -1440,16 +1440,12 @@ int soft_offline_page(struct page *page, int flags)
 	 */
 	ret = invalidate_inode_page(page);
 	unlock_page(page);
-
 	/*
-	 * Drop count because page migration doesn't like raised
-	 * counts. The page could get re-allocated, but if it becomes
-	 * LRU the isolation will just fail.
 	 * RED-PEN would be better to keep it isolated here, but we
 	 * would need to fix isolation locking first.
 	 */
-	put_page(page);
 	if (ret == 1) {
+		put_page(page);
 		ret = 0;
 		pr_info("soft_offline: %#lx: invalidated\n", pfn);
 		goto done;
@@ -1461,6 +1457,11 @@ int soft_offline_page(struct page *page, int flags)
 	 * handles a large number of cases for us.
 	 */
 	ret = isolate_lru_page(page);
+	/*
+	 * Drop page reference which is came from get_any_page()
+	 * successful isolate_lru_page() already took another one.
+	 */
+	put_page(page);
 	if (!ret) {
 		LIST_HEAD(pagelist);