]> git.kernelconcepts.de Git - karo-tx-linux.git/commitdiff
mm: collect LRU list heads into struct lruvec
authorJohannes Weiner <jweiner@redhat.com>
Thu, 8 Dec 2011 04:42:42 +0000 (15:42 +1100)
committerStephen Rothwell <sfr@canb.auug.org.au>
Tue, 13 Dec 2011 06:43:30 +0000 (17:43 +1100)
Having a unified structure with a LRU list set for both global zones and
per-memcg zones allows to keep that code simple which deals with LRU lists
and does not care about the container itself.

Once the per-memcg LRU lists directly link struct pages, the isolation
function and all other list manipulations are shared between the memcg
case and the global LRU case.

Signed-off-by: Johannes Weiner <jweiner@redhat.com>
Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Reviewed-by: Michal Hocko <mhocko@suse.cz>
Reviewed-by: Kirill A. Shutemov <kirill@shutemov.name>
Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Cc: Balbir Singh <bsingharora@gmail.com>
Cc: Ying Han <yinghan@google.com>
Cc: Greg Thelen <gthelen@google.com>
Cc: Michel Lespinasse <walken@google.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Minchan Kim <minchan.kim@gmail.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Hugh Dickins <hughd@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
include/linux/mm_inline.h
include/linux/mmzone.h
mm/memcontrol.c
mm/page_alloc.c
mm/swap.c
mm/vmscan.c

index 8f7d24712dc115790269442be6f8c245d4cf6145..e6a7ffe16d31fc77e87edb058c1c216a7580d013 100644 (file)
@@ -33,7 +33,7 @@ __add_page_to_lru_list(struct zone *zone, struct page *page, enum lru_list l,
 static inline void
 add_page_to_lru_list(struct zone *zone, struct page *page, enum lru_list l)
 {
-       __add_page_to_lru_list(zone, page, l, &zone->lru[l].list);
+       __add_page_to_lru_list(zone, page, l, &zone->lruvec.lists[l]);
 }
 
 static inline void
index a42915fd9e9969ee4e57d1511d525338642bfbb6..28cbda16743870650390b63be604bd639db7a47e 100644 (file)
@@ -159,6 +159,10 @@ static inline int is_unevictable_lru(enum lru_list l)
        return (l == LRU_UNEVICTABLE);
 }
 
+struct lruvec {
+       struct list_head lists[NR_LRU_LISTS];
+};
+
 /* Mask used at gathering information at once (see memcontrol.c) */
 #define LRU_ALL_FILE (BIT(LRU_INACTIVE_FILE) | BIT(LRU_ACTIVE_FILE))
 #define LRU_ALL_ANON (BIT(LRU_INACTIVE_ANON) | BIT(LRU_ACTIVE_ANON))
@@ -364,10 +368,8 @@ struct zone {
        ZONE_PADDING(_pad1_)
 
        /* Fields commonly accessed by the page reclaim scanner */
-       spinlock_t              lru_lock;       
-       struct zone_lru {
-               struct list_head list;
-       } lru[NR_LRU_LISTS];
+       spinlock_t              lru_lock;
+       struct lruvec           lruvec;
 
        struct zone_reclaim_stat reclaim_stat;
 
index bf84e2b7ef0f6a1470ecfc3d05f46c6e3b8862cf..22f25b0cffe523e7165d7f22a0606a33ad884b88 100644 (file)
@@ -132,10 +132,7 @@ struct mem_cgroup_reclaim_iter {
  * per-zone information in memory controller.
  */
 struct mem_cgroup_per_zone {
-       /*
-        * spin_lock to protect the per cgroup LRU
-        */
-       struct list_head        lists[NR_LRU_LISTS];
+       struct lruvec           lruvec;
        unsigned long           count[NR_LRU_LISTS];
 
        struct mem_cgroup_reclaim_iter reclaim_iter[DEF_PRIORITY + 1];
@@ -995,7 +992,7 @@ void mem_cgroup_rotate_reclaimable_page(struct page *page)
        /* Ensure pc->mem_cgroup is visible after reading PCG_USED. */
        smp_rmb();
        mz = page_cgroup_zoneinfo(pc->mem_cgroup, page);
-       list_move_tail(&pc->lru, &mz->lists[lru]);
+       list_move_tail(&pc->lru, &mz->lruvec.lists[lru]);
 }
 
 void mem_cgroup_rotate_lru_list(struct page *page, enum lru_list lru)
@@ -1013,7 +1010,7 @@ void mem_cgroup_rotate_lru_list(struct page *page, enum lru_list lru)
        /* Ensure pc->mem_cgroup is visible after reading PCG_USED. */
        smp_rmb();
        mz = page_cgroup_zoneinfo(pc->mem_cgroup, page);
-       list_move(&pc->lru, &mz->lists[lru]);
+       list_move(&pc->lru, &mz->lruvec.lists[lru]);
 }
 
 void mem_cgroup_add_lru_list(struct page *page, enum lru_list lru)
@@ -1043,7 +1040,7 @@ void mem_cgroup_add_lru_list(struct page *page, enum lru_list lru)
        /* huge page split is done under lru_lock. so, we have no races. */
        MEM_CGROUP_ZSTAT(mz, lru) += 1 << compound_order(page);
        SetPageCgroupAcctLRU(pc);
-       list_add(&pc->lru, &mz->lists[lru]);
+       list_add(&pc->lru, &mz->lruvec.lists[lru]);
 }
 
 /*
@@ -1241,7 +1238,7 @@ unsigned long mem_cgroup_isolate_pages(unsigned long nr_to_scan,
 
        BUG_ON(!mem_cont);
        mz = mem_cgroup_zoneinfo(mem_cont, nid, zid);
-       src = &mz->lists[lru];
+       src = &mz->lruvec.lists[lru];
 
        scan = 0;
        list_for_each_entry_safe_reverse(pc, tmp, src, lru) {
@@ -3628,7 +3625,7 @@ static int mem_cgroup_force_empty_list(struct mem_cgroup *memcg,
 
        zone = &NODE_DATA(node)->node_zones[zid];
        mz = mem_cgroup_zoneinfo(memcg, node, zid);
-       list = &mz->lists[lru];
+       list = &mz->lruvec.lists[lru];
 
        loop = MEM_CGROUP_ZSTAT(mz, lru);
        /* give some margin against EBUSY etc...*/
@@ -4724,7 +4721,7 @@ static int alloc_mem_cgroup_per_zone_info(struct mem_cgroup *memcg, int node)
        for (zone = 0; zone < MAX_NR_ZONES; zone++) {
                mz = &pn->zoneinfo[zone];
                for_each_lru(l)
-                       INIT_LIST_HEAD(&mz->lists[l]);
+                       INIT_LIST_HEAD(&mz->lruvec.lists[l]);
                mz->usage_in_excess = 0;
                mz->on_tree = false;
                mz->mem = memcg;
index af5467e3faa4d7cc5488f17f08d1e0a881e13ca4..7f28eb8ff607c6dd206784c9196cb1cafd78a076 100644 (file)
@@ -4317,7 +4317,7 @@ static void __paginginit free_area_init_core(struct pglist_data *pgdat,
 
                zone_pcp_init(zone);
                for_each_lru(l)
-                       INIT_LIST_HEAD(&zone->lru[l].list);
+                       INIT_LIST_HEAD(&zone->lruvec.lists[l]);
                zone->reclaim_stat.recent_rotated[0] = 0;
                zone->reclaim_stat.recent_rotated[1] = 0;
                zone->reclaim_stat.recent_scanned[0] = 0;
index 67a09a633a09baad606176f0899d94dbeb4d17d4..76ef79d3857c6252eb390ed08af0055f6171bb48 100644 (file)
--- a/mm/swap.c
+++ b/mm/swap.c
@@ -236,7 +236,7 @@ static void pagevec_move_tail_fn(struct page *page, void *arg)
 
        if (PageLRU(page) && !PageActive(page) && !PageUnevictable(page)) {
                enum lru_list lru = page_lru_base_type(page);
-               list_move_tail(&page->lru, &zone->lru[lru].list);
+               list_move_tail(&page->lru, &zone->lruvec.lists[lru]);
                mem_cgroup_rotate_reclaimable_page(page);
                (*pgmoved)++;
        }
@@ -480,7 +480,7 @@ static void lru_deactivate_fn(struct page *page, void *arg)
                 * The page's writeback ends up during pagevec
                 * We moves tha page into tail of inactive.
                 */
-               list_move_tail(&page->lru, &zone->lru[lru].list);
+               list_move_tail(&page->lru, &zone->lruvec.lists[lru]);
                mem_cgroup_rotate_reclaimable_page(page);
                __count_vm_event(PGROTATED);
        }
@@ -654,7 +654,6 @@ void lru_add_page_tail(struct zone* zone,
        int active;
        enum lru_list lru;
        const int file = 0;
-       struct list_head *head;
 
        VM_BUG_ON(!PageHead(page));
        VM_BUG_ON(PageCompound(page_tail));
@@ -674,10 +673,10 @@ void lru_add_page_tail(struct zone* zone,
                }
                update_page_reclaim_stat(zone, page_tail, file, active);
                if (likely(PageLRU(page)))
-                       head = page->lru.prev;
+                       __add_page_to_lru_list(zone, page_tail, lru,
+                                              page->lru.prev);
                else
-                       head = &zone->lru[lru].list;
-               __add_page_to_lru_list(zone, page_tail, lru, head);
+                       add_page_to_lru_list(zone, page_tail, lru);
        } else {
                SetPageUnevictable(page_tail);
                add_page_to_lru_list(zone, page_tail, LRU_UNEVICTABLE);
index 313cc54e5d0796f2f90fa1ce46fe77c73c3ba18d..4740785191641df240dd323b3f6b543c68b87a59 100644 (file)
@@ -1250,8 +1250,8 @@ static unsigned long isolate_pages_global(unsigned long nr,
                lru += LRU_ACTIVE;
        if (file)
                lru += LRU_FILE;
-       return isolate_lru_pages(nr, &z->lru[lru].list, dst, scanned, order,
-                                                               mode, file);
+       return isolate_lru_pages(nr, &z->lruvec.lists[lru], dst,
+                                scanned, order, mode, file);
 }
 
 /*
@@ -1630,7 +1630,7 @@ static void move_active_pages_to_lru(struct zone *zone,
                VM_BUG_ON(PageLRU(page));
                SetPageLRU(page);
 
-               list_move(&page->lru, &zone->lru[lru].list);
+               list_move(&page->lru, &zone->lruvec.lists[lru]);
                mem_cgroup_add_lru_list(page, lru);
                pgmoved += hpage_nr_pages(page);
 
@@ -3447,7 +3447,7 @@ retry:
                enum lru_list l = page_lru_base_type(page);
 
                __dec_zone_state(zone, NR_UNEVICTABLE);
-               list_move(&page->lru, &zone->lru[l].list);
+               list_move(&page->lru, &zone->lruvec.lists[l]);
                mem_cgroup_move_lists(page, LRU_UNEVICTABLE, l);
                __inc_zone_state(zone, NR_INACTIVE_ANON + l);
                __count_vm_event(UNEVICTABLE_PGRESCUED);
@@ -3456,7 +3456,7 @@ retry:
                 * rotate unevictable list
                 */
                SetPageUnevictable(page);
-               list_move(&page->lru, &zone->lru[LRU_UNEVICTABLE].list);
+               list_move(&page->lru, &zone->lruvec.lists[LRU_UNEVICTABLE]);
                mem_cgroup_rotate_lru_list(page, LRU_UNEVICTABLE);
                if (page_evictable(page, NULL))
                        goto retry;