mm: fix tracing in free_pcppages_bulk()

page->private gets re-used in __free_one_page() to store page order
(so trace_mm_page_pcpu_drain() may print order instead of migratetype)
thus migratetype value must be cached locally.

Fixes regression introduced in commit a701623 ("mm: fix migratetype
bug which slowed swapping").  This caused incorrect data to be attached
to the mm_page_pcpu_drain trace event.

[[email protected]: add comment]
Cc: Marek Szyprowski <[email protected]>
Cc: Michal Nazarewicz <[email protected]>
Acked-by: Minchan Kim <[email protected]>
Acked-by: Mel Gorman <[email protected]>
Cc: Hugh Dickins <[email protected]>
Signed-off-by: Bartlomiej Zolnierkiewicz <[email protected]>
Signed-off-by: Kyungmin Park <[email protected]>
Acked-by: KOSAKI Motohiro <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

Author: bzolnier

mm/page_alloc.c

@@ -668,12 +668,15 @@ static void free_pcppages_bulk(struct zone *zone, int count,
 			batch_free = to_free;
 
 		do {
+			int mt;	/* migratetype of the to-be-freed page */
+
 			page = list_entry(list->prev, struct page, lru);
 			/* must delete as __free_one_page list manipulates */
 			list_del(&page->lru);
+			mt = page_private(page);
 			/* MIGRATE_MOVABLE list may include MIGRATE_RESERVEs */
-			__free_one_page(page, zone, 0, page_private(page));
-			trace_mm_page_pcpu_drain(page, 0, page_private(page));
+			__free_one_page(page, zone, 0, mt);
+			trace_mm_page_pcpu_drain(page, 0, mt);
 		} while (--to_free && --batch_free && !list_empty(list));
 	}
 	__mod_zone_page_state(zone, NR_FREE_PAGES, count);