From: Rik van Riel 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 --- mm/vmscan.c | 20 ++++++++++++++++---- 1 file changed, 16 insertions(+), 4 deletions(-) Index: linux-2.6.26-rc5-mm2/mm/vmscan.c =================================================================== --- linux-2.6.26-rc5-mm2.orig/mm/vmscan.c 2008-06-10 13:40:19.000000000 -0400 +++ linux-2.6.26-rc5-mm2/mm/vmscan.c 2008-06-10 13:41:25.000000000 -0400 @@ -875,7 +875,8 @@ 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; @@ -893,8 +894,19 @@ static unsigned long shrink_inactive_lis 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, @@ -1175,7 +1187,7 @@ static unsigned long shrink_list(enum lr 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); } /* -- All Rights Reversed -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/