The patch titled
memcg: add inactive_anon_is_low()
has been removed from the -mm tree. Its filename was
memcg-add-inactive_anon_is_low.patch
This patch was dropped because an updated version will be merged
The current -mm tree may be found at http://userweb.kernel.org/~akpm/mmotm/
------------------------------------------------------
Subject: memcg: add inactive_anon_is_low()
From: KOSAKI Motohiro <kosaki.motohiro@xxxxxxxxxxxxxx>
The inactive_anon_is_low() is key component of active/inactive anon
balancing on reclaim. However current inactive_anon_is_low() function
only consider global reclaim.
Therefore, we need following ugly scan_global_lru() condition.
if (lru == LRU_ACTIVE_ANON &&
(!scan_global_lru(sc) || inactive_anon_is_low(zone))) {
shrink_active_list(nr_to_scan, zone, sc, priority, file);
return 0;
It causes that memcg reclaim always deactivate pages when shrink_list() is
called. To make mem_cgroup_inactive_anon_is_low() improve active/inactive
anon balancing of memcgroup.
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@xxxxxxxxxxxxxx>
Cc: Cyrill Gorcunov <gorcunov@xxxxxxxxx>
Cc: "Pekka Enberg" <penberg@xxxxxxxxxxxxxx>
Acked-by: Rik van Riel <riel@xxxxxxxxxx>
Cc: Balbir Singh <balbir@xxxxxxxxxxxxxxxxxx>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@xxxxxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
include/linux/memcontrol.h | 9 ++++++
mm/memcontrol.c | 46 ++++++++++++++++++++++++++++++++++-
mm/vmscan.c | 36 +++++++++++++++++----------
3 files changed, 77 insertions(+), 14 deletions(-)
diff -puN include/linux/memcontrol.h~memcg-add-inactive_anon_is_low include/linux/memcontrol.h
--- a/include/linux/memcontrol.h~memcg-add-inactive_anon_is_low
+++ a/include/linux/memcontrol.h
@@ -90,6 +90,8 @@ extern void mem_cgroup_record_reclaim_pr
extern long mem_cgroup_calc_reclaim(struct mem_cgroup *mem, struct zone *zone,
int priority, enum lru_list lru);
+int mem_cgroup_inactive_anon_is_low(struct mem_cgroup *memcg,
+ struct zone *zone);
#ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP
extern int do_swap_account;
@@ -241,6 +243,13 @@ static inline bool mem_cgroup_oom_called
{
return false;
}
+
+static inline int
+mem_cgroup_inactive_anon_is_low(struct mem_cgroup *memcg, struct zone *zone)
+{
+ return 1;
+}
+
#endif /* CONFIG_CGROUP_MEM_CONT */
#endif /* _LINUX_MEMCONTROL_H */
diff -puN mm/memcontrol.c~memcg-add-inactive_anon_is_low mm/memcontrol.c
--- a/mm/memcontrol.c~memcg-add-inactive_anon_is_low
+++ a/mm/memcontrol.c
@@ -156,6 +156,9 @@ struct mem_cgroup {
unsigned long last_oom_jiffies;
int obsolete;
atomic_t refcnt;
+
+ unsigned int inactive_ratio;
+
/*
* statistics. This must be placed at the end of memcg.
*/
@@ -431,6 +434,20 @@ long mem_cgroup_calc_reclaim(struct mem_
return (nr_pages >> priority);
}
+int mem_cgroup_inactive_anon_is_low(struct mem_cgroup *memcg, struct zone *zone)
+{
+ unsigned long active;
+ unsigned long inactive;
+
+ inactive = mem_cgroup_get_all_zonestat(memcg, LRU_INACTIVE_ANON);
+ active = mem_cgroup_get_all_zonestat(memcg, LRU_ACTIVE_ANON);
+
+ if (inactive * memcg->inactive_ratio < active)
+ return 1;
+
+ return 0;
+}
+
unsigned long mem_cgroup_isolate_pages(unsigned long nr_to_scan,
struct list_head *dst,
unsigned long *scanned, int order,
@@ -1346,6 +1363,29 @@ int mem_cgroup_shrink_usage(struct mm_st
return 0;
}
+/*
+ * The inactive anon list should be small enough that the VM never has to
+ * do too much work, but large enough that each inactive page has a chance
+ * to be referenced again before it is swapped out.
+ *
+ * this calculation is straightforward porting from
+ * page_alloc.c::setup_per_zone_inactive_ratio().
+ * it describe more detail.
+ */
+static void mem_cgroup_set_inactive_ratio(struct mem_cgroup *memcg)
+{
+ unsigned int gb, ratio;
+
+ gb = res_counter_read_u64(&memcg->res, RES_LIMIT) >> 30;
+ if (gb)
+ ratio = int_sqrt(10 * gb);
+ else
+ ratio = 1;
+
+ memcg->inactive_ratio = ratio;
+
+}
+
static DEFINE_MUTEX(set_limit_mutex);
static int mem_cgroup_resize_limit(struct mem_cgroup *memcg,
@@ -1384,6 +1424,10 @@ static int mem_cgroup_resize_limit(struc
GFP_HIGHUSER_MOVABLE, false);
if (!progress) retry_count--;
}
+
+ if (!ret)
+ mem_cgroup_set_inactive_ratio(memcg);
+
return ret;
}
@@ -1968,7 +2012,7 @@ mem_cgroup_create(struct cgroup_subsys *
res_counter_init(&mem->res, NULL);
res_counter_init(&mem->memsw, NULL);
}
-
+ mem_cgroup_set_inactive_ratio(mem);
mem->last_scanned_child = NULL;
return &mem->css;
diff -puN mm/vmscan.c~memcg-add-inactive_anon_is_low mm/vmscan.c
--- a/mm/vmscan.c~memcg-add-inactive_anon_is_low
+++ a/mm/vmscan.c
@@ -1357,14 +1357,7 @@ static void shrink_active_list(unsigned
pagevec_release(&pvec);
}
-/**
- * inactive_anon_is_low - check if anonymous pages need to be deactivated
- * @zone: zone to check
- *
- * Returns true if the zone does not have enough inactive anon pages,
- * meaning some active anon pages need to be deactivated.
- */
-static int inactive_anon_is_low(struct zone *zone)
+static int inactive_anon_is_low_global(struct zone *zone)
{
unsigned long active, inactive;
@@ -1377,6 +1370,25 @@ static int inactive_anon_is_low(struct z
return 0;
}
+/**
+ * inactive_anon_is_low - check if anonymous pages need to be deactivated
+ * @zone: zone to check
+ * @sc: scan control of this context
+ *
+ * Returns true if the zone does not have enough inactive anon pages,
+ * meaning some active anon pages need to be deactivated.
+ */
+static int inactive_anon_is_low(struct zone *zone, struct scan_control *sc)
+{
+ int low;
+
+ if (scan_global_lru(sc))
+ low = inactive_anon_is_low_global(zone);
+ else
+ low = mem_cgroup_inactive_anon_is_low(sc->mem_cgroup, zone);
+ return low;
+}
+
static unsigned long shrink_list(enum lru_list lru, unsigned long nr_to_scan,
struct zone *zone, struct scan_control *sc, int priority)
{
@@ -1388,7 +1400,7 @@ static unsigned long shrink_list(enum lr
}
if (lru == LRU_ACTIVE_ANON &&
- (!scan_global_lru(sc) || inactive_anon_is_low(zone))) {
+ inactive_anon_is_low(zone, sc)) {
shrink_active_list(nr_to_scan, zone, sc, priority, file);
return 0;
}
@@ -1553,9 +1565,7 @@ static void shrink_zone(int priority, st
* Even if we did not try to evict anon pages at all, we want to
* rebalance the anon lru active/inactive ratio.
*/
- if (!scan_global_lru(sc) || inactive_anon_is_low(zone))
- shrink_active_list(SWAP_CLUSTER_MAX, zone, sc, priority, 0);
- else if (!scan_global_lru(sc))
+ if (inactive_anon_is_low(zone, sc))
shrink_active_list(SWAP_CLUSTER_MAX, zone, sc, priority, 0);
throttle_vm_writeout(sc->gfp_mask);
@@ -1851,7 +1861,7 @@ loop_again:
* Do some background aging of the anon list, to give
* pages a chance to be referenced before reclaiming.
*/
- if (inactive_anon_is_low(zone))
+ if (inactive_anon_is_low(zone, &sc))
shrink_active_list(SWAP_CLUSTER_MAX, zone,
&sc, priority, 0);
_
Patches currently in -mm which might be from kosaki.motohiro@xxxxxxxxxxxxxx are
vmscan-evict-streaming-io-first.patch
linux-next.patch
arch-ia64-kernel-iosapicc-fix-warnings.patch
mm-report-the-pagesize-backing-a-vma-in-proc-pid-smaps.patch
mm-report-the-mmu-pagesize-in-proc-pid-smaps.patch
oom-fix-zone_scan_mutex-name.patch
mm-get-rid-of-pagevec_release_nonlru.patch
cleanup-get-rid-of-ifdef-config_migration.patch
mm-more-likely-reclaim-madv_sequential-mappings.patch
mm-make-page_lock_anon_vma-static.patch
memcg-reclaim-shouldnt-change-zone-recent_rotated-statistics.patch
mm-make-init_section_page_cgroup-static.patch
mm-make-maddr-__iomem.patch
mm-make-mem_cgroup_resize_limit-static.patch
mm-make-scan_all_zones_unevictable_pages-static.patch
mm-make-scan_zone_unevictable_pages-static.patch
mm-make-setup_per_zone_inactive_ratio-static.patch
mm-make-vread-and-vwrite-declaration.patch
vmscan-bail-out-of-direct-reclaim-after-swap_cluster_max-pages.patch
mm-kill-zone_is_near_oom.patch
softirq-introduce-statistics-for-softirq.patch
proc-export-statistics-for-softirq-to-proc.patch
proc-update-document-for-proc-softirqs-and-proc-stat.patch
memcg-new-force_empty-to-free-pages-under-group-fix.patch
memcg-new-force_empty-to-free-pages-under-group-fix-fix.patch
memcg-add-inactive_anon_is_low.patch
memcg-add-mem_cgroup_zone_nr_pages.patch
memcg-add-zone_reclaim_stat.patch
memcg-remove-mem_cgroup_calc_reclaim.patch
memcg-show-inactive_ratio.patch
memcg-show-reclaim_stat.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
