aboutsummaryrefslogtreecommitdiffstats
path: root/include/linux/mmzone.h
diff options
context:
space:
mode:
authorJiang Liu <liuj97@gmail.com>2012-12-12 16:52:12 -0500
committerLinus Torvalds <torvalds@linux-foundation.org>2012-12-12 20:38:34 -0500
commit9feedc9d831e18ae6d0d15aa562e5e46ba53647b (patch)
treecb26ff54b0f02c4905772288b27f99b8b384ad6d /include/linux/mmzone.h
parentc2d23f919bafcbc2259f5257d9a7d729802f0e3a (diff)
mm: introduce new field "managed_pages" to struct zone
Currently a zone's present_pages is calcuated as below, which is inaccurate and may cause trouble to memory hotplug. spanned_pages - absent_pages - memmap_pages - dma_reserve. During fixing bugs caused by inaccurate zone->present_pages, we found zone->present_pages has been abused. The field zone->present_pages may have different meanings in different contexts: 1) pages existing in a zone. 2) pages managed by the buddy system. For more discussions about the issue, please refer to: http://lkml.org/lkml/2012/11/5/866 https://patchwork.kernel.org/patch/1346751/ This patchset tries to introduce a new field named "managed_pages" to struct zone, which counts "pages managed by the buddy system". And revert zone->present_pages to count "physical pages existing in a zone", which also keep in consistence with pgdat->node_present_pages. We will set an initial value for zone->managed_pages in function free_area_init_core() and will adjust it later if the initial value is inaccurate. For DMA/normal zones, the initial value is set to: (spanned_pages - absent_pages - memmap_pages - dma_reserve) Later zone->managed_pages will be adjusted to the accurate value when the bootmem allocator frees all free pages to the buddy system in function free_all_bootmem_node() and free_all_bootmem(). The bootmem allocator doesn't touch highmem pages, so highmem zones' managed_pages is set to the accurate value "spanned_pages - absent_pages" in function free_area_init_core() and won't be updated anymore. This patch also adds a new field "managed_pages" to /proc/zoneinfo and sysrq showmem. [akpm@linux-foundation.org: small comment tweaks] Signed-off-by: Jiang Liu <jiang.liu@huawei.com> Cc: Wen Congyang <wency@cn.fujitsu.com> Cc: David Rientjes <rientjes@google.com> Cc: Maciej Rutecki <maciej.rutecki@gmail.com> Tested-by: Chris Clayton <chris2553@googlemail.com> Cc: "Rafael J . Wysocki" <rjw@sisk.pl> Cc: Mel Gorman <mgorman@suse.de> Cc: Minchan Kim <minchan@kernel.org> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Michal Hocko <mhocko@suse.cz> Cc: Jianguo Wu <wujianguo@huawei.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Diffstat (limited to 'include/linux/mmzone.h')
-rw-r--r--include/linux/mmzone.h41
1 files changed, 34 insertions, 7 deletions
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 0c0b1d608a69..cd55dad56aac 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -460,17 +460,44 @@ struct zone {
460 unsigned long zone_start_pfn; 460 unsigned long zone_start_pfn;
461 461
462 /* 462 /*
463 * zone_start_pfn, spanned_pages and present_pages are all 463 * spanned_pages is the total pages spanned by the zone, including
464 * protected by span_seqlock. It is a seqlock because it has 464 * holes, which is calculated as:
465 * to be read outside of zone->lock, and it is done in the main 465 * spanned_pages = zone_end_pfn - zone_start_pfn;
466 * allocator path. But, it is written quite infrequently.
467 * 466 *
468 * The lock is declared along with zone->lock because it is 467 * present_pages is physical pages existing within the zone, which
468 * is calculated as:
469 * present_pages = spanned_pages - absent_pages(pags in holes);
470 *
471 * managed_pages is present pages managed by the buddy system, which
472 * is calculated as (reserved_pages includes pages allocated by the
473 * bootmem allocator):
474 * managed_pages = present_pages - reserved_pages;
475 *
476 * So present_pages may be used by memory hotplug or memory power
477 * management logic to figure out unmanaged pages by checking
478 * (present_pages - managed_pages). And managed_pages should be used
479 * by page allocator and vm scanner to calculate all kinds of watermarks
480 * and thresholds.
481 *
482 * Locking rules:
483 *
484 * zone_start_pfn and spanned_pages are protected by span_seqlock.
485 * It is a seqlock because it has to be read outside of zone->lock,
486 * and it is done in the main allocator path. But, it is written
487 * quite infrequently.
488 *
489 * The span_seq lock is declared along with zone->lock because it is
469 * frequently read in proximity to zone->lock. It's good to 490 * frequently read in proximity to zone->lock. It's good to
470 * give them a chance of being in the same cacheline. 491 * give them a chance of being in the same cacheline.
492 *
493 * Write access to present_pages and managed_pages at runtime should
494 * be protected by lock_memory_hotplug()/unlock_memory_hotplug().
495 * Any reader who can't tolerant drift of present_pages and
496 * managed_pages should hold memory hotplug lock to get a stable value.
471 */ 497 */
472 unsigned long spanned_pages; /* total size, including holes */ 498 unsigned long spanned_pages;
473 unsigned long present_pages; /* amount of memory (excluding holes) */ 499 unsigned long present_pages;
500 unsigned long managed_pages;
474 501
475 /* 502 /*
476 * rarely used fields: 503 * rarely used fields: