The patch titled Subject: mm: consider all swapped back pages in used-once logic has been removed from the -mm tree. Its filename was mm-consider-all-swapped-back-pages-in-used-once-logic.patch This patch was dropped because it was merged into mainline or a subsystem tree The current -mm tree may be found at http://userweb.kernel.org/~akpm/mmotm/ ------------------------------------------------------ From: Michal Hocko <mhocko@xxxxxxx> Subject: mm: consider all swapped back pages in used-once logic Commit 64574746 ("vmscan: detect mapped file pages used only once") made mapped pages have another round in inactive list because they might be just short lived and so we could consider them again next time. This heuristic helps to reduce pressure on the active list with a streaming IO worklods. This patch fixes a regression introduced by this commit for heavy shmem based workloads because unlike Anon pages, which are excluded from this heuristic because they are usually long lived, shmem pages are handled as a regular page cache. This doesn't work quite well, unfortunately, if the workload is mostly backed by shmem (in memory database sitting on 80% of memory) with a streaming IO in the background (backup - up to 20% of memory). Anon inactive list is full of (dirty) shmem pages when watermarks are hit. Shmem pages are kept in the inactive list (they are referenced) in the first round and it is hard to reclaim anything else so we reach lower scanning priorities very quickly which leads to an excessive swap out. Let's fix this by excluding all swap backed pages (they tend to be long lived wrt. the regular page cache anyway) from used-once heuristic and rather activate them if they are referenced. The customer's workload is shmem backed database (80% of RAM) and they are measuring transactions/s with an IO in the background (20%). Transactions touch more or less random rows in the table. The transaction rate fell by a factor of 3 (in the worst case) because of commit 64574746. This patch restores the previous numbers. Signed-off-by: Michal Hocko <mhocko@xxxxxxx> Acked-by: Johannes Weiner <hannes@xxxxxxxxxxx> Cc: Mel Gorman <mel@xxxxxxxxx> Cc: Minchan Kim <minchan@xxxxxxxxxx> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@xxxxxxxxxxxxxx> Reviewed-by: Rik van Riel <riel@xxxxxxxxxx> Cc: <stable@xxxxxxxxxxxxxxx> [2.6.34+] Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- mm/vmscan.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff -puN mm/vmscan.c~mm-consider-all-swapped-back-pages-in-used-once-logic mm/vmscan.c --- a/mm/vmscan.c~mm-consider-all-swapped-back-pages-in-used-once-logic +++ a/mm/vmscan.c @@ -657,7 +657,7 @@ static enum page_references page_check_r return PAGEREF_RECLAIM; if (referenced_ptes) { - if (PageAnon(page)) + if (PageSwapBacked(page)) return PAGEREF_ACTIVATE; /* * All mapped pages start out with page table _ Patches currently in -mm which might be from mhocko@xxxxxxx are origin.patch hugetlb-rename-max_hstate-to-hugetlb_max_hstate.patch hugetlbfs-dont-use-err_ptr-with-vm_fault-values.patch hugetlbfs-add-an-inline-helper-for-finding-hstate-index.patch hugetlbfs-add-an-inline-helper-for-finding-hstate-index-fix.patch hugetlb-use-mmu_gather-instead-of-a-temporary-linked-list-for-accumulating-pages.patch hugetlb-use-mmu_gather-instead-of-a-temporary-linked-list-for-accumulating-pages-fix.patch hugetlb-use-mmu_gather-instead-of-a-temporary-linked-list-for-accumulating-pages-fix-fix.patch hugetlb-avoid-taking-i_mmap_mutex-in-unmap_single_vma-for-hugetlb.patch hugetlb-simplify-migrate_huge_page.patch hugetlb-simplify-migrate_huge_page-fix.patch memcg-add-hugetlb-extension.patch memcg-add-hugetlb-extension-fix.patch memcg-add-hugetlb-extension-fix-fix.patch hugetlb-add-charge-uncharge-calls-for-hugetlb-alloc-free.patch memcg-track-resource-index-in-cftype-private.patch hugetlbfs-add-memcg-control-files-for-hugetlbfs.patch hugetlbfs-add-memcg-control-files-for-hugetlbfs-use-scnprintf-instead-of-sprintf.patch hugetlbfs-add-memcg-control-files-for-hugetlbfs-use-scnprintf-instead-of-sprintf-fix.patch hugetlbfs-add-memcg-control-files-for-hugetlbfs-add-failcnt-support-for-hugetlb-extension.patch hugetlbfs-add-a-list-for-tracking-in-use-hugetlb-pages.patch memcg-move-hugetlb-resource-count-to-parent-cgroup-on-memcg-removal.patch memcg-move-hugetlb-resource-count-to-parent-cgroup-on-memcg-removal-fix.patch memcg-move-hugetlb-resource-count-to-parent-cgroup-on-memcg-removal-fix-fix.patch hugetlb-migrate-memcg-info-from-oldpage-to-new-page-during-migration.patch memcg-add-memory-controller-documentation-for-hugetlb-management.patch memcg-fix-error-code-in-hugetlb_force_memcg_empty.patch memcg-use-res_counter_uncharge_until-in-mem_cgroup_move_hugetlb_parent.patch memcg-move-charges-to-root-cgroup-if-use_hierarchy=0-in-mem_cgroup_move_hugetlb_parent.patch -- To unsubscribe from this list: send the line "unsubscribe mm-commits" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html