memory-hotplug: fix NR_FREE_PAGES mismatch
authorWen Congyang <wency@cn.fujitsu.com>
Wed, 12 Dec 2012 00:00:52 +0000 (16:00 -0800)
committerLinus Torvalds <torvalds@linux-foundation.org>
Wed, 12 Dec 2012 01:22:23 +0000 (17:22 -0800)
NR_FREE_PAGES will be wrong after offlining pages.  We add/dec
NR_FREE_PAGES like this now:

1. move all pages in buddy system to MIGRATE_ISOLATE, and dec NR_FREE_PAGES

2. don't add NR_FREE_PAGES when it is freed and the migratetype is
   MIGRATE_ISOLATE

3. dec NR_FREE_PAGES when offlining isolated pages.

4. add NR_FREE_PAGES when undoing isolate pages.

When we come to step 3, all pages are in MIGRATE_ISOLATE list, and
NR_FREE_PAGES are right.  When we come to step4, all pages are not in
buddy system, so we don't change NR_FREE_PAGES in this step, but we change
NR_FREE_PAGES in step3.  So NR_FREE_PAGES is wrong after offlining pages.
So there is no need to change NR_FREE_PAGES in step3.

This patch also fixs a problem in step2: if the migratetype is
MIGRATE_ISOLATE, we should not add NR_FRR_PAGES when we remove pages from
pcppages.

Signed-off-by: Wen Congyang <wency@cn.fujitsu.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Jiang Liu <liuj97@gmail.com>
Cc: Len Brown <len.brown@intel.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Christoph Lameter <cl@linux.com>
Cc: Minchan Kim <minchan.kim@gmail.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Jianguo Wu <wujianguo106@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm/page_alloc.c

index 6f50cfe98a7b6dda0be18387f501c2458f362061..4dba04f06880f1a4a5a28ac08fb167f0899a925d 100644 (file)
@@ -667,11 +667,13 @@ static void free_pcppages_bulk(struct zone *zone, int count,
                        /* MIGRATE_MOVABLE list may include MIGRATE_RESERVEs */
                        __free_one_page(page, zone, 0, mt);
                        trace_mm_page_pcpu_drain(page, 0, mt);
-                       if (is_migrate_cma(mt))
-                               __mod_zone_page_state(zone, NR_FREE_CMA_PAGES, 1);
+                       if (likely(get_pageblock_migratetype(page) != MIGRATE_ISOLATE)) {
+                               __mod_zone_page_state(zone, NR_FREE_PAGES, 1);
+                               if (is_migrate_cma(mt))
+                                       __mod_zone_page_state(zone, NR_FREE_CMA_PAGES, 1);
+                       }
                } while (--to_free && --batch_free && !list_empty(list));
        }
-       __mod_zone_page_state(zone, NR_FREE_PAGES, count);
        spin_unlock(&zone->lock);
 }
 
@@ -6047,8 +6049,6 @@ __offline_isolated_pages(unsigned long start_pfn, unsigned long end_pfn)
                list_del(&page->lru);
                rmv_page_order(page);
                zone->free_area[order].nr_free--;
-               __mod_zone_page_state(zone, NR_FREE_PAGES,
-                                     - (1UL << order));
                for (i = 0; i < (1 << order); i++)
                        SetPageReserved((page+i));
                pfn += (1 << order);