more aggressively use lumpy reclaim

During an AIM7 run on a 16GB system, fork started failing around 32000
threads, despite the system having plenty of free swap and 15GB of
pageable memory.  This was on x86-64, so 8k stacks.

If a higher order allocation fails, we can either:
- keep evicting pages off the end of the LRUs and hope that
  we eventually create a contiguous region; this is somewhat
  unlikely if the system is under enough stress by new
  allocations
- after trying normal eviction for a bit, use lumpy reclaim

This patch switches the system to lumpy reclaim if the VM is having
trouble freeing enough pages, using the same threshold for detection as
used by pageout congestion wait.

Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

diff --git a/mm/vmscan.c b/mm/vmscan.c
index 9588973..a8347b6 100644 (file)
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -909,7 +909,8 @@ int isolate_lru_page(struct page *page)
  * of reclaimed pages
  */
 static unsigned long shrink_inactive_list(unsigned long max_scan,
-                       struct zone *zone, struct scan_control *sc, int file)
+                       struct zone *zone, struct scan_control *sc,
+                       int priority, int file)
 {
        LIST_HEAD(page_list);
        struct pagevec pvec;
@@ -927,8 +928,19 @@ static unsigned long shrink_inactive_list(unsigned long max_scan,
                unsigned long nr_freed;
                unsigned long nr_active;
                unsigned int count[NR_LRU_LISTS] = { 0, };
-               int mode = (sc->order > PAGE_ALLOC_COSTLY_ORDER) ?
-                                       ISOLATE_BOTH : ISOLATE_INACTIVE;
+               int mode = ISOLATE_INACTIVE;
+
+               /*
+                * If we need a large contiguous chunk of memory, or have
+                * trouble getting a small set of contiguous pages, we
+                * will reclaim both active and inactive pages.
+                *
+                * We use the same threshold as pageout congestion_wait below.
+                */
+               if (sc->order > PAGE_ALLOC_COSTLY_ORDER)
+                       mode = ISOLATE_BOTH;
+               else if (sc->order && priority < DEF_PRIORITY - 2)
+                       mode = ISOLATE_BOTH;
 
                nr_taken = sc->isolate_pages(sc->swap_cluster_max,
                             &page_list, &nr_scan, sc->order, mode,
@@ -1172,7 +1184,7 @@ static unsigned long shrink_list(enum lru_list lru, unsigned long nr_to_scan,
                shrink_active_list(nr_to_scan, zone, sc, priority, file);
                return 0;
        }
-       return shrink_inactive_list(nr_to_scan, zone, sc, file);
+       return shrink_inactive_list(nr_to_scan, zone, sc, priority, file);
 }
 
 /*