#include <linux/delay.h>
#include <linux/kthread.h>
#include <linux/freezer.h>
+#include <linux/memcontrol.h>
#include <asm/tlbflush.h>
#include <asm/div64.h>
int all_unreclaimable;
int order;
+
+ /*
+ * Pages that have (or should have) IO pending. If we run into
+ * a lot of these, we're better off waiting a little for IO to
+ * finish rather than scanning more pages in the VM.
+ */
+ int nr_io_pages;
+
+ /* Which cgroup do we reclaim from */
+ struct mem_cgroup *mem_cgroup;
+
+ /* Pluggable isolate pages callback */
+ unsigned long (*isolate_pages)(unsigned long nr, struct list_head *dst,
+ unsigned long *scanned, int order, int mode,
+ struct zone *z, struct mem_cgroup *mem_cont,
+ int active);
};
#define lru_to_page(_head) (list_entry((_head)->prev, struct page, lru))
static LIST_HEAD(shrinker_list);
static DECLARE_RWSEM(shrinker_rwsem);
+#ifdef CONFIG_CGROUP_MEM_CONT
+#define scan_global_lru(sc) (!(sc)->mem_cgroup)
+#else
+#define scan_global_lru(sc) (1)
+#endif
+
/*
* Add a shrinker callback to be called from the vm
*/
* percentages of the lru and ageable caches. This should balance the seeks
* generated by these structures.
*
- * If the vm encounted mapped pages on the LRU it increase the pressure on
+ * If the vm encountered mapped pages on the LRU it increase the pressure on
* slab to avoid swapping.
*
* We do weird things to avoid (scanned*seeks*entries) overflowing 32 bits.
*/
if (sync_writeback == PAGEOUT_IO_SYNC && may_enter_fs)
wait_on_page_writeback(page);
- else
+ else {
+ sc->nr_io_pages++;
goto keep_locked;
+ }
}
- referenced = page_referenced(page, 1);
+ referenced = page_referenced(page, 1, sc->mem_cgroup);
/* In active use or really unfreeable? Activate it. */
if (sc->order <= PAGE_ALLOC_COSTLY_ORDER &&
referenced && page_mapping_inuse(page))
if (PageDirty(page)) {
if (sc->order <= PAGE_ALLOC_COSTLY_ORDER && referenced)
goto keep_locked;
- if (!may_enter_fs)
+ if (!may_enter_fs) {
+ sc->nr_io_pages++;
goto keep_locked;
+ }
if (!sc->may_writepage)
goto keep_locked;
case PAGE_ACTIVATE:
goto activate_locked;
case PAGE_SUCCESS:
- if (PageWriteback(page) || PageDirty(page))
+ if (PageWriteback(page) || PageDirty(page)) {
+ sc->nr_io_pages++;
goto keep;
+ }
/*
* A synchronous write - probably a ramdisk. Go
* ahead and try to reclaim the page.
*
* returns 0 on success, -ve errno on failure.
*/
-static int __isolate_lru_page(struct page *page, int mode)
+int __isolate_lru_page(struct page *page, int mode)
{
int ret = -EINVAL;
return nr_taken;
}
+static unsigned long isolate_pages_global(unsigned long nr,
+ struct list_head *dst,
+ unsigned long *scanned, int order,
+ int mode, struct zone *z,
+ struct mem_cgroup *mem_cont,
+ int active)
+{
+ if (active)
+ return isolate_lru_pages(nr, &z->active_list, dst,
+ scanned, order, mode);
+ else
+ return isolate_lru_pages(nr, &z->inactive_list, dst,
+ scanned, order, mode);
+}
+
/*
* clear_active_flags() is a helper for shrink_active_list(), clearing
* any active bits from the pages in the list.
unsigned long nr_freed;
unsigned long nr_active;
- nr_taken = isolate_lru_pages(sc->swap_cluster_max,
- &zone->inactive_list,
+ nr_taken = sc->isolate_pages(sc->swap_cluster_max,
&page_list, &nr_scan, sc->order,
(sc->order > PAGE_ALLOC_COSTLY_ORDER)?
- ISOLATE_BOTH : ISOLATE_INACTIVE);
+ ISOLATE_BOTH : ISOLATE_INACTIVE,
+ zone, sc->mem_cgroup, 0);
nr_active = clear_active_flags(&page_list);
__count_vm_events(PGDEACTIVATE, nr_active);
lru_add_drain();
spin_lock_irq(&zone->lru_lock);
- pgmoved = isolate_lru_pages(nr_pages, &zone->active_list,
- &l_hold, &pgscanned, sc->order, ISOLATE_ACTIVE);
+ pgmoved = sc->isolate_pages(nr_pages, &l_hold, &pgscanned, sc->order,
+ ISOLATE_ACTIVE, zone,
+ sc->mem_cgroup, 1);
zone->pages_scanned += pgscanned;
__mod_zone_page_state(zone, NR_ACTIVE, -pgmoved);
spin_unlock_irq(&zone->lru_lock);
if (page_mapped(page)) {
if (!reclaim_mapped ||
(total_swap_pages == 0 && PageAnon(page)) ||
- page_referenced(page, 0)) {
+ page_referenced(page, 0, sc->mem_cgroup)) {
list_add(&page->lru, &l_active);
continue;
}
ClearPageActive(page);
list_move(&page->lru, &zone->inactive_list);
+ mem_cgroup_move_lists(page_get_page_cgroup(page), false);
pgmoved++;
if (!pagevec_add(&pvec, page)) {
__mod_zone_page_state(zone, NR_INACTIVE, pgmoved);
SetPageLRU(page);
VM_BUG_ON(!PageActive(page));
list_move(&page->lru, &zone->active_list);
+ mem_cgroup_move_lists(page_get_page_cgroup(page), true);
pgmoved++;
if (!pagevec_add(&pvec, page)) {
__mod_zone_page_state(zone, NR_ACTIVE, pgmoved);
* holds filesystem locks which prevent writeout this might not work, and the
* allocation attempt will fail.
*/
-unsigned long try_to_free_pages(struct zone **zones, int order, gfp_t gfp_mask)
+static unsigned long do_try_to_free_pages(struct zone **zones, gfp_t gfp_mask,
+ struct scan_control *sc)
{
int priority;
int ret = 0;
struct reclaim_state *reclaim_state = current->reclaim_state;
unsigned long lru_pages = 0;
int i;
- struct scan_control sc = {
- .gfp_mask = gfp_mask,
- .may_writepage = !laptop_mode,
- .swap_cluster_max = SWAP_CLUSTER_MAX,
- .may_swap = 1,
- .swappiness = vm_swappiness,
- .order = order,
- };
count_vm_event(ALLOCSTALL);
}
for (priority = DEF_PRIORITY; priority >= 0; priority--) {
- sc.nr_scanned = 0;
+ sc->nr_scanned = 0;
+ sc->nr_io_pages = 0;
if (!priority)
disable_swap_token();
- nr_reclaimed += shrink_zones(priority, zones, &sc);
- shrink_slab(sc.nr_scanned, gfp_mask, lru_pages);
- if (reclaim_state) {
- nr_reclaimed += reclaim_state->reclaimed_slab;
- reclaim_state->reclaimed_slab = 0;
+ nr_reclaimed += shrink_zones(priority, zones, sc);
+ /*
+ * Don't shrink slabs when reclaiming memory from
+ * over limit cgroups
+ */
+ if (scan_global_lru(sc)) {
+ shrink_slab(sc->nr_scanned, gfp_mask, lru_pages);
+ if (reclaim_state) {
+ nr_reclaimed += reclaim_state->reclaimed_slab;
+ reclaim_state->reclaimed_slab = 0;
+ }
}
- total_scanned += sc.nr_scanned;
- if (nr_reclaimed >= sc.swap_cluster_max) {
+ total_scanned += sc->nr_scanned;
+ if (nr_reclaimed >= sc->swap_cluster_max) {
ret = 1;
goto out;
}
* that's undesirable in laptop mode, where we *want* lumpy
* writeout. So in laptop mode, write out the whole world.
*/
- if (total_scanned > sc.swap_cluster_max +
- sc.swap_cluster_max / 2) {
+ if (total_scanned > sc->swap_cluster_max +
+ sc->swap_cluster_max / 2) {
wakeup_pdflush(laptop_mode ? 0 : total_scanned);
- sc.may_writepage = 1;
+ sc->may_writepage = 1;
}
/* Take a nap, wait for some writeback to complete */
- if (sc.nr_scanned && priority < DEF_PRIORITY - 2)
+ if (sc->nr_scanned && priority < DEF_PRIORITY - 2 &&
+ sc->nr_io_pages > sc->swap_cluster_max)
congestion_wait(WRITE, HZ/10);
}
/* top priority shrink_caches still had more to do? don't OOM, then */
- if (!sc.all_unreclaimable)
+ if (!sc->all_unreclaimable && scan_global_lru(sc))
ret = 1;
out:
/*
*/
if (priority < 0)
priority = 0;
- for (i = 0; zones[i] != 0; i++) {
+ for (i = 0; zones[i] != NULL; i++) {
struct zone *zone = zones[i];
if (!cpuset_zone_allowed_hardwall(zone, GFP_KERNEL))
return ret;
}
+unsigned long try_to_free_pages(struct zone **zones, int order, gfp_t gfp_mask)
+{
+ struct scan_control sc = {
+ .gfp_mask = gfp_mask,
+ .may_writepage = !laptop_mode,
+ .swap_cluster_max = SWAP_CLUSTER_MAX,
+ .may_swap = 1,
+ .swappiness = vm_swappiness,
+ .order = order,
+ .mem_cgroup = NULL,
+ .isolate_pages = isolate_pages_global,
+ };
+
+ return do_try_to_free_pages(zones, gfp_mask, &sc);
+}
+
+#ifdef CONFIG_CGROUP_MEM_CONT
+
+unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *mem_cont,
+ gfp_t gfp_mask)
+{
+ struct scan_control sc = {
+ .gfp_mask = gfp_mask,
+ .may_writepage = !laptop_mode,
+ .may_swap = 1,
+ .swap_cluster_max = SWAP_CLUSTER_MAX,
+ .swappiness = vm_swappiness,
+ .order = 0,
+ .mem_cgroup = mem_cont,
+ .isolate_pages = mem_cgroup_isolate_pages,
+ };
+ struct zone **zones;
+ int target_zone = gfp_zone(GFP_HIGHUSER_MOVABLE);
+
+ zones = NODE_DATA(numa_node_id())->node_zonelists[target_zone].zones;
+ if (do_try_to_free_pages(zones, sc.gfp_mask, &sc))
+ return 1;
+ return 0;
+}
+#endif
+
/*
* For kswapd, balance_pgdat() will work across all this node's zones until
* they are all at pages_high.
.swap_cluster_max = SWAP_CLUSTER_MAX,
.swappiness = vm_swappiness,
.order = order,
+ .mem_cgroup = NULL,
+ .isolate_pages = isolate_pages_global,
};
/*
* temp_priority is used to remember the scanning priority at which
if (!priority)
disable_swap_token();
+ sc.nr_io_pages = 0;
all_zones_ok = 1;
/*
* OK, kswapd is getting into trouble. Take a nap, then take
* another pass across the zones.
*/
- if (total_scanned && priority < DEF_PRIORITY - 2)
+ if (total_scanned && priority < DEF_PRIORITY - 2 &&
+ sc.nr_io_pages > sc.swap_cluster_max)
congestion_wait(WRITE, HZ/10);
/*
.swap_cluster_max = nr_pages,
.may_writepage = 1,
.swappiness = vm_swappiness,
+ .isolate_pages = isolate_pages_global,
};
current->reclaim_state = &reclaim_state;
SWAP_CLUSTER_MAX),
.gfp_mask = gfp_mask,
.swappiness = vm_swappiness,
+ .isolate_pages = isolate_pages_global,
};
unsigned long slab_reclaimable;