[lkml]   [2008]   [Mar]   [4]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
    Patch in this message
    Subject[patch 20/20] cull non-reclaimable anon pages from the LRU at fault time
    V2 -> V3:
    + rebase to 23-mm1 atop RvR's split lru series.

    V1 -> V2:
    + no changes

    Optional part of "noreclaim infrastructure"

    In the fault paths that install new anonymous pages, check whether
    the page is reclaimable or not using lru_cache_add_active_or_noreclaim().
    If the page is reclaimable, just add it to the active lru list [via
    the pagevec cache], else add it to the noreclaim list.

    This "proactive" culling in the fault path mimics the handling of
    mlocked pages in Nick Piggin's series to keep mlocked pages off
    the lru lists.


    1) This patch is optional--e.g., if one is concerned about the
    additional test in the fault path. We can defer the moving of
    nonreclaimable pages until when vmscan [shrink_*_list()]
    encounters them. Vmscan will only need to handle such pages

    2) I moved the call to page_add_new_anon_rmap() to before the test
    for page_reclaimable() and thus before the calls to
    lru_cache_add_{active|noreclaim}(), so that page_reclaimable()
    could recognize the page as anon.
    TBD: I think this reordering is OK, but the previous order may
    have existed to close some obscure race?

    3) The 'vma' argument to page_reclaimable() is require to notice that
    we're faulting a page into an mlock()ed vma w/o having to scan the
    page's rmap in the fault path. Culling mlock()ed anon pages is
    currently the only reason for this patch.

    Signed-off-by: Lee Schermerhorn <>
    Signed-off-by: Rik van Riel <>

    Index: linux-2.6.25-rc3-mm1/mm/memory.c
    --- linux-2.6.25-rc3-mm1.orig/mm/memory.c 2008-03-04 15:30:20.000000000 -0500
    +++ linux-2.6.25-rc3-mm1/mm/memory.c 2008-03-04 16:22:58.000000000 -0500
    @@ -1678,7 +1678,7 @@ gotten:
    set_pte_at(mm, address, page_table, entry);
    update_mmu_cache(vma, address, entry);
    - lru_cache_add_active_anon(new_page);
    + lru_cache_add_active_or_noreclaim(new_page, vma);
    page_add_new_anon_rmap(new_page, vma, address);

    /* Free the old page.. */
    @@ -2147,7 +2147,7 @@ static int do_anonymous_page(struct mm_s
    goto release;
    inc_mm_counter(mm, anon_rss);
    - lru_cache_add_active_anon(page);
    + lru_cache_add_active_or_noreclaim(page, vma);
    page_add_new_anon_rmap(page, vma, address);
    set_pte_at(mm, address, page_table, entry);

    @@ -2289,10 +2289,10 @@ static int __do_fault(struct mm_struct *
    entry = maybe_mkwrite(pte_mkdirty(entry), vma);
    set_pte_at(mm, address, page_table, entry);
    if (anon) {
    - inc_mm_counter(mm, anon_rss);
    + inc_mm_counter(mm, anon_rss);
    - lru_cache_add_active_anon(page);
    - page_add_new_anon_rmap(page, vma, address);
    + lru_cache_add_active_or_noreclaim(page, vma);
    + page_add_new_anon_rmap(page, vma, address);
    } else {
    inc_mm_counter(mm, file_rss);
    Index: linux-2.6.25-rc3-mm1/mm/swap_state.c
    --- linux-2.6.25-rc3-mm1.orig/mm/swap_state.c 2008-03-04 15:46:42.000000000 -0500
    +++ linux-2.6.25-rc3-mm1/mm/swap_state.c 2008-03-04 16:22:58.000000000 -0500
    @@ -300,7 +300,10 @@ struct page *read_swap_cache_async(swp_e
    * Initiate read into locked page and return.
    - lru_cache_add_anon(new_page);
    + if (!page_reclaimable(new_page, vma))
    + lru_cache_add_noreclaim(new_page);
    + else
    + lru_cache_add_anon(new_page);
    swap_readpage(NULL, new_page);
    return new_page;
    All Rights Reversed

     \ /
      Last update: 2008-03-05 00:15    [W:0.046 / U:2.640 seconds]
    ©2003-2016 Jasper Spaans. hosted at Digital OceanAdvertise on this site