Commit ef8f2327 authored by Mel Gorman's avatar Mel Gorman Committed by Linus Torvalds

mm, memcg: move memcg limit enforcement from zones to nodes

Memcg needs adjustment after moving LRUs to the node.  Limits are
tracked per memcg but the soft-limit excess is tracked per zone.  As
global page reclaim is based on the node, it is easy to imagine a
situation where a zone soft limit is exceeded even though the memcg
limit is fine.

This patch moves the soft limit tree the node.  Technically, all the
variable names should also change but people are already familiar by the
meaning of "mz" even if "mn" would be a more appropriate name now.

Link: http://lkml.kernel.org/r/1467970510-21195-15-git-send-email-mgorman@techsingularity.netSigned-off-by: default avatarMel Gorman <mgorman@techsingularity.net>
Acked-by: default avatarMichal Hocko <mhocko@suse.com>
Cc: Hillf Danton <hillf.zj@alibaba-inc.com>
Acked-by: default avatarJohannes Weiner <hannes@cmpxchg.org>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Rik van Riel <riel@surriel.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: default avatarAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
parent a9dd0a83
...@@ -60,7 +60,7 @@ enum mem_cgroup_stat_index { ...@@ -60,7 +60,7 @@ enum mem_cgroup_stat_index {
}; };
struct mem_cgroup_reclaim_cookie { struct mem_cgroup_reclaim_cookie {
struct zone *zone; pg_data_t *pgdat;
int priority; int priority;
unsigned int generation; unsigned int generation;
}; };
...@@ -118,7 +118,7 @@ struct mem_cgroup_reclaim_iter { ...@@ -118,7 +118,7 @@ struct mem_cgroup_reclaim_iter {
/* /*
* per-zone information in memory controller. * per-zone information in memory controller.
*/ */
struct mem_cgroup_per_zone { struct mem_cgroup_per_node {
struct lruvec lruvec; struct lruvec lruvec;
unsigned long lru_size[NR_LRU_LISTS]; unsigned long lru_size[NR_LRU_LISTS];
...@@ -132,10 +132,6 @@ struct mem_cgroup_per_zone { ...@@ -132,10 +132,6 @@ struct mem_cgroup_per_zone {
/* use container_of */ /* use container_of */
}; };
struct mem_cgroup_per_node {
struct mem_cgroup_per_zone zoneinfo[MAX_NR_ZONES];
};
struct mem_cgroup_threshold { struct mem_cgroup_threshold {
struct eventfd_ctx *eventfd; struct eventfd_ctx *eventfd;
unsigned long threshold; unsigned long threshold;
...@@ -314,19 +310,15 @@ void mem_cgroup_uncharge_list(struct list_head *page_list); ...@@ -314,19 +310,15 @@ void mem_cgroup_uncharge_list(struct list_head *page_list);
void mem_cgroup_migrate(struct page *oldpage, struct page *newpage); void mem_cgroup_migrate(struct page *oldpage, struct page *newpage);
static inline struct mem_cgroup_per_zone * static struct mem_cgroup_per_node *
mem_cgroup_zone_zoneinfo(struct mem_cgroup *memcg, struct zone *zone) mem_cgroup_nodeinfo(struct mem_cgroup *memcg, int nid)
{ {
int nid = zone_to_nid(zone); return memcg->nodeinfo[nid];
int zid = zone_idx(zone);
return &memcg->nodeinfo[nid]->zoneinfo[zid];
} }
/** /**
* mem_cgroup_lruvec - get the lru list vector for a node or a memcg zone * mem_cgroup_lruvec - get the lru list vector for a node or a memcg zone
* @node: node of the wanted lruvec * @node: node of the wanted lruvec
* @zone: zone of the wanted lruvec
* @memcg: memcg of the wanted lruvec * @memcg: memcg of the wanted lruvec
* *
* Returns the lru list vector holding pages for a given @node or a given * Returns the lru list vector holding pages for a given @node or a given
...@@ -334,9 +326,9 @@ mem_cgroup_zone_zoneinfo(struct mem_cgroup *memcg, struct zone *zone) ...@@ -334,9 +326,9 @@ mem_cgroup_zone_zoneinfo(struct mem_cgroup *memcg, struct zone *zone)
* is disabled. * is disabled.
*/ */
static inline struct lruvec *mem_cgroup_lruvec(struct pglist_data *pgdat, static inline struct lruvec *mem_cgroup_lruvec(struct pglist_data *pgdat,
struct zone *zone, struct mem_cgroup *memcg) struct mem_cgroup *memcg)
{ {
struct mem_cgroup_per_zone *mz; struct mem_cgroup_per_node *mz;
struct lruvec *lruvec; struct lruvec *lruvec;
if (mem_cgroup_disabled()) { if (mem_cgroup_disabled()) {
...@@ -344,7 +336,7 @@ static inline struct lruvec *mem_cgroup_lruvec(struct pglist_data *pgdat, ...@@ -344,7 +336,7 @@ static inline struct lruvec *mem_cgroup_lruvec(struct pglist_data *pgdat,
goto out; goto out;
} }
mz = mem_cgroup_zone_zoneinfo(memcg, zone); mz = mem_cgroup_nodeinfo(memcg, pgdat->node_id);
lruvec = &mz->lruvec; lruvec = &mz->lruvec;
out: out:
/* /*
...@@ -352,8 +344,8 @@ static inline struct lruvec *mem_cgroup_lruvec(struct pglist_data *pgdat, ...@@ -352,8 +344,8 @@ static inline struct lruvec *mem_cgroup_lruvec(struct pglist_data *pgdat,
* we have to be prepared to initialize lruvec->pgdat here; * we have to be prepared to initialize lruvec->pgdat here;
* and if offlined then reonlined, we need to reinitialize it. * and if offlined then reonlined, we need to reinitialize it.
*/ */
if (unlikely(lruvec->pgdat != zone->zone_pgdat)) if (unlikely(lruvec->pgdat != pgdat))
lruvec->pgdat = zone->zone_pgdat; lruvec->pgdat = pgdat;
return lruvec; return lruvec;
} }
...@@ -446,9 +438,9 @@ unsigned long mem_cgroup_node_nr_lru_pages(struct mem_cgroup *memcg, ...@@ -446,9 +438,9 @@ unsigned long mem_cgroup_node_nr_lru_pages(struct mem_cgroup *memcg,
static inline static inline
unsigned long mem_cgroup_get_lru_size(struct lruvec *lruvec, enum lru_list lru) unsigned long mem_cgroup_get_lru_size(struct lruvec *lruvec, enum lru_list lru)
{ {
struct mem_cgroup_per_zone *mz; struct mem_cgroup_per_node *mz;
mz = container_of(lruvec, struct mem_cgroup_per_zone, lruvec); mz = container_of(lruvec, struct mem_cgroup_per_node, lruvec);
return mz->lru_size[lru]; return mz->lru_size[lru];
} }
...@@ -519,7 +511,7 @@ static inline void mem_cgroup_dec_page_stat(struct page *page, ...@@ -519,7 +511,7 @@ static inline void mem_cgroup_dec_page_stat(struct page *page,
mem_cgroup_update_page_stat(page, idx, -1); mem_cgroup_update_page_stat(page, idx, -1);
} }
unsigned long mem_cgroup_soft_limit_reclaim(struct zone *zone, int order, unsigned long mem_cgroup_soft_limit_reclaim(pg_data_t *pgdat, int order,
gfp_t gfp_mask, gfp_t gfp_mask,
unsigned long *total_scanned); unsigned long *total_scanned);
...@@ -611,7 +603,7 @@ static inline void mem_cgroup_migrate(struct page *old, struct page *new) ...@@ -611,7 +603,7 @@ static inline void mem_cgroup_migrate(struct page *old, struct page *new)
} }
static inline struct lruvec *mem_cgroup_lruvec(struct pglist_data *pgdat, static inline struct lruvec *mem_cgroup_lruvec(struct pglist_data *pgdat,
struct zone *zone, struct mem_cgroup *memcg) struct mem_cgroup *memcg)
{ {
return node_lruvec(pgdat); return node_lruvec(pgdat);
} }
...@@ -723,7 +715,7 @@ static inline void mem_cgroup_dec_page_stat(struct page *page, ...@@ -723,7 +715,7 @@ static inline void mem_cgroup_dec_page_stat(struct page *page,
} }
static inline static inline
unsigned long mem_cgroup_soft_limit_reclaim(struct zone *zone, int order, unsigned long mem_cgroup_soft_limit_reclaim(pg_data_t *pgdat, int order,
gfp_t gfp_mask, gfp_t gfp_mask,
unsigned long *total_scanned) unsigned long *total_scanned)
{ {
......
...@@ -318,7 +318,7 @@ extern unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *memcg, ...@@ -318,7 +318,7 @@ extern unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *memcg,
bool may_swap); bool may_swap);
extern unsigned long mem_cgroup_shrink_node(struct mem_cgroup *mem, extern unsigned long mem_cgroup_shrink_node(struct mem_cgroup *mem,
gfp_t gfp_mask, bool noswap, gfp_t gfp_mask, bool noswap,
struct zone *zone, pg_data_t *pgdat,
unsigned long *nr_scanned); unsigned long *nr_scanned);
extern unsigned long shrink_all_memory(unsigned long nr_pages); extern unsigned long shrink_all_memory(unsigned long nr_pages);
extern int vm_swappiness; extern int vm_swappiness;
......
...@@ -132,15 +132,11 @@ static const char * const mem_cgroup_lru_names[] = { ...@@ -132,15 +132,11 @@ static const char * const mem_cgroup_lru_names[] = {
* their hierarchy representation * their hierarchy representation
*/ */
struct mem_cgroup_tree_per_zone { struct mem_cgroup_tree_per_node {
struct rb_root rb_root; struct rb_root rb_root;
spinlock_t lock; spinlock_t lock;
}; };
struct mem_cgroup_tree_per_node {
struct mem_cgroup_tree_per_zone rb_tree_per_zone[MAX_NR_ZONES];
};
struct mem_cgroup_tree { struct mem_cgroup_tree {
struct mem_cgroup_tree_per_node *rb_tree_per_node[MAX_NUMNODES]; struct mem_cgroup_tree_per_node *rb_tree_per_node[MAX_NUMNODES];
}; };
...@@ -374,37 +370,35 @@ ino_t page_cgroup_ino(struct page *page) ...@@ -374,37 +370,35 @@ ino_t page_cgroup_ino(struct page *page)
return ino; return ino;
} }
static struct mem_cgroup_per_zone * static struct mem_cgroup_per_node *
mem_cgroup_page_zoneinfo(struct mem_cgroup *memcg, struct page *page) mem_cgroup_page_nodeinfo(struct mem_cgroup *memcg, struct page *page)
{ {
int nid = page_to_nid(page); int nid = page_to_nid(page);
int zid = page_zonenum(page);
return &memcg->nodeinfo[nid]->zoneinfo[zid]; return memcg->nodeinfo[nid];
} }
static struct mem_cgroup_tree_per_zone * static struct mem_cgroup_tree_per_node *
soft_limit_tree_node_zone(int nid, int zid) soft_limit_tree_node(int nid)
{ {
return &soft_limit_tree.rb_tree_per_node[nid]->rb_tree_per_zone[zid]; return soft_limit_tree.rb_tree_per_node[nid];
} }
static struct mem_cgroup_tree_per_zone * static struct mem_cgroup_tree_per_node *
soft_limit_tree_from_page(struct page *page) soft_limit_tree_from_page(struct page *page)
{ {
int nid = page_to_nid(page); int nid = page_to_nid(page);
int zid = page_zonenum(page);
return &soft_limit_tree.rb_tree_per_node[nid]->rb_tree_per_zone[zid]; return soft_limit_tree.rb_tree_per_node[nid];
} }
static void __mem_cgroup_insert_exceeded(struct mem_cgroup_per_zone *mz, static void __mem_cgroup_insert_exceeded(struct mem_cgroup_per_node *mz,
struct mem_cgroup_tree_per_zone *mctz, struct mem_cgroup_tree_per_node *mctz,
unsigned long new_usage_in_excess) unsigned long new_usage_in_excess)
{ {
struct rb_node **p = &mctz->rb_root.rb_node; struct rb_node **p = &mctz->rb_root.rb_node;
struct rb_node *parent = NULL; struct rb_node *parent = NULL;
struct mem_cgroup_per_zone *mz_node; struct mem_cgroup_per_node *mz_node;
if (mz->on_tree) if (mz->on_tree)
return; return;
...@@ -414,7 +408,7 @@ static void __mem_cgroup_insert_exceeded(struct mem_cgroup_per_zone *mz, ...@@ -414,7 +408,7 @@ static void __mem_cgroup_insert_exceeded(struct mem_cgroup_per_zone *mz,
return; return;
while (*p) { while (*p) {
parent = *p; parent = *p;
mz_node = rb_entry(parent, struct mem_cgroup_per_zone, mz_node = rb_entry(parent, struct mem_cgroup_per_node,
tree_node); tree_node);
if (mz->usage_in_excess < mz_node->usage_in_excess) if (mz->usage_in_excess < mz_node->usage_in_excess)
p = &(*p)->rb_left; p = &(*p)->rb_left;
...@@ -430,8 +424,8 @@ static void __mem_cgroup_insert_exceeded(struct mem_cgroup_per_zone *mz, ...@@ -430,8 +424,8 @@ static void __mem_cgroup_insert_exceeded(struct mem_cgroup_per_zone *mz,
mz->on_tree = true; mz->on_tree = true;
} }
static void __mem_cgroup_remove_exceeded(struct mem_cgroup_per_zone *mz, static void __mem_cgroup_remove_exceeded(struct mem_cgroup_per_node *mz,
struct mem_cgroup_tree_per_zone *mctz) struct mem_cgroup_tree_per_node *mctz)
{ {
if (!mz->on_tree) if (!mz->on_tree)
return; return;
...@@ -439,8 +433,8 @@ static void __mem_cgroup_remove_exceeded(struct mem_cgroup_per_zone *mz, ...@@ -439,8 +433,8 @@ static void __mem_cgroup_remove_exceeded(struct mem_cgroup_per_zone *mz,
mz->on_tree = false; mz->on_tree = false;
} }
static void mem_cgroup_remove_exceeded(struct mem_cgroup_per_zone *mz, static void mem_cgroup_remove_exceeded(struct mem_cgroup_per_node *mz,
struct mem_cgroup_tree_per_zone *mctz) struct mem_cgroup_tree_per_node *mctz)
{ {
unsigned long flags; unsigned long flags;
...@@ -464,8 +458,8 @@ static unsigned long soft_limit_excess(struct mem_cgroup *memcg) ...@@ -464,8 +458,8 @@ static unsigned long soft_limit_excess(struct mem_cgroup *memcg)
static void mem_cgroup_update_tree(struct mem_cgroup *memcg, struct page *page) static void mem_cgroup_update_tree(struct mem_cgroup *memcg, struct page *page)
{ {
unsigned long excess; unsigned long excess;
struct mem_cgroup_per_zone *mz; struct mem_cgroup_per_node *mz;
struct mem_cgroup_tree_per_zone *mctz; struct mem_cgroup_tree_per_node *mctz;
mctz = soft_limit_tree_from_page(page); mctz = soft_limit_tree_from_page(page);
/* /*
...@@ -473,7 +467,7 @@ static void mem_cgroup_update_tree(struct mem_cgroup *memcg, struct page *page) ...@@ -473,7 +467,7 @@ static void mem_cgroup_update_tree(struct mem_cgroup *memcg, struct page *page)
* because their event counter is not touched. * because their event counter is not touched.
*/ */
for (; memcg; memcg = parent_mem_cgroup(memcg)) { for (; memcg; memcg = parent_mem_cgroup(memcg)) {
mz = mem_cgroup_page_zoneinfo(memcg, page); mz = mem_cgroup_page_nodeinfo(memcg, page);
excess = soft_limit_excess(memcg); excess = soft_limit_excess(memcg);
/* /*
* We have to update the tree if mz is on RB-tree or * We have to update the tree if mz is on RB-tree or
...@@ -498,24 +492,22 @@ static void mem_cgroup_update_tree(struct mem_cgroup *memcg, struct page *page) ...@@ -498,24 +492,22 @@ static void mem_cgroup_update_tree(struct mem_cgroup *memcg, struct page *page)
static void mem_cgroup_remove_from_trees(struct mem_cgroup *memcg) static void mem_cgroup_remove_from_trees(struct mem_cgroup *memcg)
{ {
struct mem_cgroup_tree_per_zone *mctz; struct mem_cgroup_tree_per_node *mctz;
struct mem_cgroup_per_zone *mz; struct mem_cgroup_per_node *mz;
int nid, zid; int nid;
for_each_node(nid) { for_each_node(nid) {
for (zid = 0; zid < MAX_NR_ZONES; zid++) { mz = mem_cgroup_nodeinfo(memcg, nid);
mz = &memcg->nodeinfo[nid]->zoneinfo[zid]; mctz = soft_limit_tree_node(nid);
mctz = soft_limit_tree_node_zone(nid, zid);
mem_cgroup_remove_exceeded(mz, mctz); mem_cgroup_remove_exceeded(mz, mctz);
} }
}
} }
static struct mem_cgroup_per_zone * static struct mem_cgroup_per_node *
__mem_cgroup_largest_soft_limit_node(struct mem_cgroup_tree_per_zone *mctz) __mem_cgroup_largest_soft_limit_node(struct mem_cgroup_tree_per_node *mctz)
{ {
struct rb_node *rightmost = NULL; struct rb_node *rightmost = NULL;
struct mem_cgroup_per_zone *mz; struct mem_cgroup_per_node *mz;
retry: retry:
mz = NULL; mz = NULL;
...@@ -523,7 +515,7 @@ __mem_cgroup_largest_soft_limit_node(struct mem_cgroup_tree_per_zone *mctz) ...@@ -523,7 +515,7 @@ __mem_cgroup_largest_soft_limit_node(struct mem_cgroup_tree_per_zone *mctz)
if (!rightmost) if (!rightmost)
goto done; /* Nothing to reclaim from */ goto done; /* Nothing to reclaim from */
mz = rb_entry(rightmost, struct mem_cgroup_per_zone, tree_node); mz = rb_entry(rightmost, struct mem_cgroup_per_node, tree_node);
/* /*
* Remove the node now but someone else can add it back, * Remove the node now but someone else can add it back,
* we will to add it back at the end of reclaim to its correct * we will to add it back at the end of reclaim to its correct
...@@ -537,10 +529,10 @@ __mem_cgroup_largest_soft_limit_node(struct mem_cgroup_tree_per_zone *mctz) ...@@ -537,10 +529,10 @@ __mem_cgroup_largest_soft_limit_node(struct mem_cgroup_tree_per_zone *mctz)
return mz; return mz;
} }
static struct mem_cgroup_per_zone * static struct mem_cgroup_per_node *
mem_cgroup_largest_soft_limit_node(struct mem_cgroup_tree_per_zone *mctz) mem_cgroup_largest_soft_limit_node(struct mem_cgroup_tree_per_node *mctz)
{ {
struct mem_cgroup_per_zone *mz; struct mem_cgroup_per_node *mz;
spin_lock_irq(&mctz->lock); spin_lock_irq(&mctz->lock);
mz = __mem_cgroup_largest_soft_limit_node(mctz); mz = __mem_cgroup_largest_soft_limit_node(mctz);
...@@ -634,21 +626,17 @@ unsigned long mem_cgroup_node_nr_lru_pages(struct mem_cgroup *memcg, ...@@ -634,21 +626,17 @@ unsigned long mem_cgroup_node_nr_lru_pages(struct mem_cgroup *memcg,
int nid, unsigned int lru_mask) int nid, unsigned int lru_mask)
{ {
unsigned long nr = 0; unsigned long nr = 0;
int zid; struct mem_cgroup_per_node *mz;
enum lru_list lru;
VM_BUG_ON((unsigned)nid >= nr_node_ids); VM_BUG_ON((unsigned)nid >= nr_node_ids);
for (zid = 0; zid < MAX_NR_ZONES; zid++) {
struct mem_cgroup_per_zone *mz;
enum lru_list lru;
for_each_lru(lru) { for_each_lru(lru) {
if (!(BIT(lru) & lru_mask)) if (!(BIT(lru) & lru_mask))
continue; continue;
mz = &memcg->nodeinfo[nid]->zoneinfo[zid]; mz = mem_cgroup_nodeinfo(memcg, nid);
nr += mz->lru_size[lru]; nr += mz->lru_size[lru];
} }
}
return nr; return nr;
} }
...@@ -800,9 +788,9 @@ struct mem_cgroup *mem_cgroup_iter(struct mem_cgroup *root, ...@@ -800,9 +788,9 @@ struct mem_cgroup *mem_cgroup_iter(struct mem_cgroup *root,
rcu_read_lock(); rcu_read_lock();
if (reclaim) { if (reclaim) {
struct mem_cgroup_per_zone *mz; struct mem_cgroup_per_node *mz;
mz = mem_cgroup_zone_zoneinfo(root, reclaim->zone); mz = mem_cgroup_nodeinfo(root, reclaim->pgdat->node_id);
iter = &mz->iter[reclaim->priority]; iter = &mz->iter[reclaim->priority];
if (prev && reclaim->generation != iter->generation) if (prev && reclaim->generation != iter->generation)
...@@ -901,14 +889,13 @@ static void invalidate_reclaim_iterators(struct mem_cgroup *dead_memcg) ...@@ -901,14 +889,13 @@ static void invalidate_reclaim_iterators(struct mem_cgroup *dead_memcg)
{ {
struct mem_cgroup *memcg = dead_memcg; struct mem_cgroup *memcg = dead_memcg;
struct mem_cgroup_reclaim_iter *iter; struct mem_cgroup_reclaim_iter *iter;
struct mem_cgroup_per_zone *mz; struct mem_cgroup_per_node *mz;
int nid, zid; int nid;
int i; int i;
while ((memcg = parent_mem_cgroup(memcg))) { while ((memcg = parent_mem_cgroup(memcg))) {
for_each_node(nid) { for_each_node(nid) {
for (zid = 0; zid < MAX_NR_ZONES; zid++) { mz = mem_cgroup_nodeinfo(memcg, nid);
mz = &memcg->nodeinfo[nid]->zoneinfo[zid];
for (i = 0; i <= DEF_PRIORITY; i++) { for (i = 0; i <= DEF_PRIORITY; i++) {
iter = &mz->iter[i]; iter = &mz->iter[i];
cmpxchg(&iter->position, cmpxchg(&iter->position,
...@@ -916,7 +903,6 @@ static void invalidate_reclaim_iterators(struct mem_cgroup *dead_memcg) ...@@ -916,7 +903,6 @@ static void invalidate_reclaim_iterators(struct mem_cgroup *dead_memcg)
} }
} }
} }
}
} }
/* /*
...@@ -945,7 +931,7 @@ static void invalidate_reclaim_iterators(struct mem_cgroup *dead_memcg) ...@@ -945,7 +931,7 @@ static void invalidate_reclaim_iterators(struct mem_cgroup *dead_memcg)
*/ */
struct lruvec *mem_cgroup_page_lruvec(struct page *page, struct pglist_data *pgdat) struct lruvec *mem_cgroup_page_lruvec(struct page *page, struct pglist_data *pgdat)
{ {
struct mem_cgroup_per_zone *mz; struct mem_cgroup_per_node *mz;
struct mem_cgroup *memcg; struct mem_cgroup *memcg;
struct lruvec *lruvec; struct lruvec *lruvec;
...@@ -962,7 +948,7 @@ struct lruvec *mem_cgroup_page_lruvec(struct page *page, struct pglist_data *pgd ...@@ -962,7 +948,7 @@ struct lruvec *mem_cgroup_page_lruvec(struct page *page, struct pglist_data *pgd
if (!memcg) if (!memcg)
memcg = root_mem_cgroup; memcg = root_mem_cgroup;
mz = mem_cgroup_page_zoneinfo(memcg, page); mz = mem_cgroup_page_nodeinfo(memcg, page);
lruvec = &mz->lruvec; lruvec = &mz->lruvec;
out: out:
/* /*
...@@ -989,7 +975,7 @@ struct lruvec *mem_cgroup_page_lruvec(struct page *page, struct pglist_data *pgd ...@@ -989,7 +975,7 @@ struct lruvec *mem_cgroup_page_lruvec(struct page *page, struct pglist_data *pgd
void mem_cgroup_update_lru_size(struct lruvec *lruvec, enum lru_list lru, void mem_cgroup_update_lru_size(struct lruvec *lruvec, enum lru_list lru,
enum zone_type zid, int nr_pages) enum zone_type zid, int nr_pages)
{ {
struct mem_cgroup_per_zone *mz; struct mem_cgroup_per_node *mz;
unsigned long *lru_size; unsigned long *lru_size;
long size; long size;
bool empty; bool empty;
...@@ -999,7 +985,7 @@ void mem_cgroup_update_lru_size(struct lruvec *lruvec, enum lru_list lru, ...@@ -999,7 +985,7 @@ void mem_cgroup_update_lru_size(struct lruvec *lruvec, enum lru_list lru,
if (mem_cgroup_disabled()) if (mem_cgroup_disabled())
return; return;
mz = container_of(lruvec, struct mem_cgroup_per_zone, lruvec); mz = container_of(lruvec, struct mem_cgroup_per_node, lruvec);
lru_size = mz->lru_size + lru; lru_size = mz->lru_size + lru;
empty = list_empty(lruvec->lists + lru); empty = list_empty(lruvec->lists + lru);
...@@ -1392,7 +1378,7 @@ int mem_cgroup_select_victim_node(struct mem_cgroup *memcg) ...@@ -1392,7 +1378,7 @@ int mem_cgroup_select_victim_node(struct mem_cgroup *memcg)
#endif #endif
static int mem_cgroup_soft_reclaim(struct mem_cgroup *root_memcg, static int mem_cgroup_soft_reclaim(struct mem_cgroup *root_memcg,
struct zone *zone, pg_data_t *pgdat,
gfp_t gfp_mask, gfp_t gfp_mask,
unsigned long *total_scanned) unsigned long *total_scanned)
{ {
...@@ -1402,7 +1388,7 @@ static int mem_cgroup_soft_reclaim(struct mem_cgroup *root_memcg, ...@@ -1402,7 +1388,7 @@ static int mem_cgroup_soft_reclaim(struct mem_cgroup *root_memcg,
unsigned long excess; unsigned long excess;
unsigned long nr_scanned; unsigned long nr_scanned;
struct mem_cgroup_reclaim_cookie reclaim = { struct mem_cgroup_reclaim_cookie reclaim = {
.zone = zone, .pgdat = pgdat,
.priority = 0, .priority = 0,
}; };
...@@ -1433,7 +1419,7 @@ static int mem_cgroup_soft_reclaim(struct mem_cgroup *root_memcg, ...@@ -1433,7 +1419,7 @@ static int mem_cgroup_soft_reclaim(struct mem_cgroup *root_memcg,
continue; continue;
} }
total += mem_cgroup_shrink_node(victim, gfp_mask, false, total += mem_cgroup_shrink_node(victim, gfp_mask, false,
zone, &nr_scanned); pgdat, &nr_scanned);
*total_scanned += nr_scanned; *total_scanned += nr_scanned;
if (!soft_limit_excess(root_memcg)) if (!soft_limit_excess(root_memcg))
break; break;
...@@ -2560,22 +2546,22 @@ static int mem_cgroup_resize_memsw_limit(struct mem_cgroup *memcg, ...@@ -2560,22 +2546,22 @@ static int mem_cgroup_resize_memsw_limit(struct mem_cgroup *memcg,
return ret; return ret;
} }
unsigned long mem_cgroup_soft_limit_reclaim(struct zone *zone, int order, unsigned long mem_cgroup_soft_limit_reclaim(pg_data_t *pgdat, int order,
gfp_t gfp_mask, gfp_t gfp_mask,
unsigned long *total_scanned) unsigned long *total_scanned)
{ {
unsigned long nr_reclaimed = 0; unsigned long nr_reclaimed = 0;
struct mem_cgroup_per_zone *mz, *next_mz = NULL; struct mem_cgroup_per_node *mz, *next_mz = NULL;
unsigned long reclaimed; unsigned long reclaimed;
int loop = 0; int loop = 0;
struct mem_cgroup_tree_per_zone *mctz; struct mem_cgroup_tree_per_node *mctz;
unsigned long excess; unsigned long excess;
unsigned long nr_scanned; unsigned long nr_scanned;
if (order > 0) if (order > 0)
return 0; return 0;
mctz = soft_limit_tree_node_zone(zone_to_nid(zone), zone_idx(zone)); mctz = soft_limit_tree_node(pgdat->node_id);
/* /*
* This loop can run a while, specially if mem_cgroup's continuously * This loop can run a while, specially if mem_cgroup's continuously
* keep exceeding their soft limit and putting the system under * keep exceeding their soft limit and putting the system under
...@@ -2590,7 +2576,7 @@ unsigned long mem_cgroup_soft_limit_reclaim(struct zone *zone, int order, ...@@ -2590,7 +2576,7 @@ unsigned long mem_cgroup_soft_limit_reclaim(struct zone *zone, int order,
break; break;
nr_scanned = 0; nr_scanned = 0;
reclaimed = mem_cgroup_soft_reclaim(mz->memcg, zone, reclaimed = mem_cgroup_soft_reclaim(mz->memcg, pgdat,
gfp_mask, &nr_scanned); gfp_mask, &nr_scanned);
nr_reclaimed += reclaimed; nr_reclaimed += reclaimed;
*total_scanned += nr_scanned; *total_scanned += nr_scanned;
...@@ -3211,15 +3197,14 @@ static int memcg_stat_show(struct seq_file *m, void *v) ...@@ -3211,15 +3197,14 @@ static int memcg_stat_show(struct seq_file *m, void *v)
#ifdef CONFIG_DEBUG_VM #ifdef CONFIG_DEBUG_VM
{ {
int nid, zid; pg_data_t *pgdat;
struct mem_cgroup_per_zone *mz; struct mem_cgroup_per_node *mz;
struct zone_reclaim_stat *rstat; struct zone_reclaim_stat *rstat;
unsigned long recent_rotated[2] = {0, 0}; unsigned long recent_rotated[2] = {0, 0};
unsigned long recent_scanned[2] = {0, 0}; unsigned long recent_scanned[2] = {0, 0};
for_each_online_node(nid) for_each_online_pgdat(pgdat) {
for (zid = 0; zid < MAX_NR_ZONES; zid++) { mz = mem_cgroup_nodeinfo(memcg, pgdat->node_id);
mz = &memcg->nodeinfo[nid]->zoneinfo[zid];
rstat = &mz->lruvec.reclaim_stat; rstat = &mz->lruvec.reclaim_stat;
recent_rotated[0] += rstat->recent_rotated[0]; recent_rotated[0] += rstat->recent_rotated[0];
...@@ -4106,11 +4091,10 @@ struct mem_cgroup *mem_cgroup_from_id(unsigned short id) ...@@ -4106,11 +4091,10 @@ struct mem_cgroup *mem_cgroup_from_id(unsigned short id)
return idr_find(&mem_cgroup_idr, id); return idr_find(&mem_cgroup_idr, id);
} }
static int alloc_mem_cgroup_per_zone_info(struct mem_cgroup *memcg, int node) static int alloc_mem_cgroup_per_node_info(struct mem_cgroup *memcg, int node)
{ {
struct mem_cgroup_per_node *pn; struct mem_cgroup_per_node *pn;
struct mem_cgroup_per_zone *mz; int tmp = node;
int zone, tmp = node;
/* /*
* This routine is called against possible nodes. * This routine is called against possible nodes.
* But it's BUG to call kmalloc() against offline node. * But it's BUG to call kmalloc() against offline node.
...@@ -4125,18 +4109,16 @@ static int alloc_mem_cgroup_per_zone_info(struct mem_cgroup *memcg, int node) ...@@ -4125,18 +4109,16 @@ static int alloc_mem_cgroup_per_zone_info(struct mem_cgroup *memcg, int node)
if (!pn) if (!pn)
return 1; return 1;
for (zone = 0; zone < MAX_NR_ZONES; zone++) { lruvec_init(&pn->lruvec);
mz = &pn->zoneinfo[zone]; pn->usage_in_excess = 0;
lruvec_init(&mz->lruvec); pn->on_tree = false;
mz->usage_in_excess = 0; pn->memcg = memcg;
mz->on_tree = false;
mz->memcg = memcg;
}
memcg->nodeinfo[node] = pn; memcg->nodeinfo[node] = pn;
return 0; return 0;
} }
static void free_mem_cgroup_per_zone_info(struct mem_cgroup *memcg, int node) static void free_mem_cgroup_per_node_info(struct mem_cgroup *memcg, int node)
{ {
kfree(memcg->nodeinfo[node]); kfree(memcg->nodeinfo[node]);
} }
...@@ -4147,7 +4129,7 @@ static void mem_cgroup_free(struct mem_cgroup *memcg) ...@@ -4147,7 +4129,7 @@ static void mem_cgroup_free(struct mem_cgroup *memcg)
memcg_wb_domain_exit(memcg); memcg_wb_domain_exit(memcg);
for_each_node(node) for_each_node(node)
free_mem_cgroup_per_zone_info(memcg, node); free_mem_cgroup_per_node_info(memcg, node);
free_percpu(memcg->stat); free_percpu(memcg->stat);
kfree(memcg); kfree(memcg);
} }
...@@ -4176,7 +4158,7 @@ static struct mem_cgroup *mem_cgroup_alloc(void) ...@@ -4176,7 +4158,7 @@ static struct mem_cgroup *mem_cgroup_alloc(void)
goto fail; goto fail;
for_each_node(node) for_each_node(node)
if (alloc_mem_cgroup_per_zone_info(memcg, node)) if (alloc_mem_cgroup_per_node_info(memcg, node))
goto fail; goto fail;
if (memcg_wb_domain_init(memcg, GFP_KERNEL)) if (memcg_wb_domain_init(memcg, GFP_KERNEL))
...@@ -5779,18 +5761,12 @@ static int __init mem_cgroup_init(void) ...@@ -5779,18 +5761,12 @@ static int __init mem_cgroup_init(void)
for_each_node(node) { for_each_node(node) {
struct mem_cgroup_tree_per_node *rtpn; struct mem_cgroup_tree_per_node *rtpn;
int zone;
rtpn = kzalloc_node(sizeof(*rtpn), GFP_KERNEL, rtpn = kzalloc_node(sizeof(*rtpn), GFP_KERNEL,
node_online(node) ? node : NUMA_NO_NODE); node_online(node) ? node : NUMA_NO_NODE);
for (zone = 0; zone < MAX_NR_ZONES; zone++) { rtpn->rb_root = RB_ROOT;
struct mem_cgroup_tree_per_zone *rtpz; spin_lock_init(&rtpn->lock);
rtpz = &rtpn->rb_tree_per_zone[zone];
rtpz->rb_root = RB_ROOT;
spin_lock_init(&rtpz->lock);
}
soft_limit_tree.rb_tree_per_node[node] = rtpn; soft_limit_tree.rb_tree_per_node[node] = rtpn;
} }
......
...@@ -2229,8 +2229,7 @@ static inline void init_tlb_ubc(void) ...@@ -2229,8 +2229,7 @@ static inline void init_tlb_ubc(void)
static void shrink_node_memcg(struct pglist_data *pgdat, struct mem_cgroup *memcg, static void shrink_node_memcg(struct pglist_data *pgdat, struct mem_cgroup *memcg,
struct scan_control *sc, unsigned long *lru_pages) struct scan_control *sc, unsigned long *lru_pages)
{ {
struct zone *zone = &pgdat->node_zones[sc->reclaim_idx]; struct lruvec *lruvec = mem_cgroup_lruvec(pgdat, memcg);
struct lruvec *lruvec = mem_cgroup_lruvec(pgdat, zone, memcg);
unsigned long nr[NR_LRU_LISTS]; unsigned long nr[NR_LRU_LISTS];
unsigned long targets[NR_LRU_LISTS]; unsigned long targets[NR_LRU_LISTS];
unsigned long nr_to_scan; unsigned long nr_to_scan;
...@@ -2439,7 +2438,7 @@ static bool shrink_node(pg_data_t *pgdat, struct scan_control *sc, ...@@ -2439,7 +2438,7 @@ static bool shrink_node(pg_data_t *pgdat, struct scan_control *sc,
do { do {
struct mem_cgroup *root = sc->target_mem_cgroup; struct mem_cgroup *root = sc->target_mem_cgroup;
struct mem_cgroup_reclaim_cookie reclaim = { struct mem_cgroup_reclaim_cookie reclaim = {
.zone = &pgdat->node_zones[classzone_idx], .pgdat = pgdat,
.priority = sc->priority, .priority = sc->priority,
}; };
unsigned long node_lru_pages = 0; unsigned long node_lru_pages = 0;
...@@ -2647,7 +2646,7 @@ static void shrink_zones(struct zonelist *zonelist, struct scan_control *sc) ...@@ -2647,7 +2646,7 @@ static void shrink_zones(struct zonelist *zonelist, struct scan_control *sc)
* and balancing, not for a memcg's limit. * and balancing, not for a memcg's limit.
*/ */
nr_soft_scanned = 0; nr_soft_scanned = 0;
nr_soft_reclaimed = mem_cgroup_soft_limit_reclaim(zone, nr_soft_reclaimed = mem_cgroup_soft_limit_reclaim(zone->zone_pgdat,
sc->order, sc->gfp_mask, sc->order, sc->gfp_mask,
&nr_soft_scanned); &nr_soft_scanned);
sc->nr_reclaimed += nr_soft_reclaimed; sc->nr_reclaimed += nr_soft_reclaimed;
...@@ -2917,7 +2916,7 @@ unsigned long try_to_free_pages(struct zonelist *zonelist, int order, ...@@ -2917,7 +2916,7 @@ unsigned long try_to_free_pages(struct zonelist *zonelist, int order,
unsigned long mem_cgroup_shrink_node(struct mem_cgroup *memcg, unsigned long mem_cgroup_shrink_node(struct mem_cgroup *memcg,
gfp_t gfp_mask, bool noswap, gfp_t gfp_mask, bool noswap,
struct zone *zone, pg_data_t *pgdat,
unsigned long *nr_scanned) unsigned long *nr_scanned)
{ {
struct scan_control sc = { struct scan_control sc = {
...@@ -2944,7 +2943,7 @@ unsigned long mem_cgroup_shrink_node(struct mem_cgroup *memcg, ...@@ -2944,7 +2943,7 @@ unsigned long mem_cgroup_shrink_node(struct mem_cgroup *memcg,
* will pick up pages from other mem cgroup's as well. We hack * will pick up pages from other mem cgroup's as well. We hack
* the priority and make it zero. * the priority and make it zero.
*/ */
shrink_node_memcg(zone->zone_pgdat, memcg, &sc, &lru_pages); shrink_node_memcg(pgdat, memcg, &sc, &lru_pages);
trace_mm_vmscan_memcg_softlimit_reclaim_end(sc.nr_reclaimed); trace_mm_vmscan_memcg_softlimit_reclaim_end(sc.nr_reclaimed);
...@@ -2994,7 +2993,7 @@ unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *memcg, ...@@ -2994,7 +2993,7 @@ unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *memcg,
#endif #endif
static void age_active_anon(struct pglist_data *pgdat, static void age_active_anon(struct pglist_data *pgdat,
struct zone *zone, struct scan_control *sc) struct scan_control *sc)
{ {
struct mem_cgroup *memcg; struct mem_cgroup *memcg;
...@@ -3003,7 +3002,7 @@ static void age_active_anon(struct pglist_data *pgdat, ...@@ -3003,7 +3002,7 @@ static void age_active_anon(struct pglist_data *pgdat,
memcg = mem_cgroup_iter(NULL, NULL, NULL); memcg = mem_cgroup_iter(NULL, NULL, NULL);
do { do {
struct lruvec *lruvec = mem_cgroup_lruvec(pgdat, zone, memcg); struct lruvec *lruvec = mem_cgroup_lruvec(pgdat, memcg);
if (inactive_list_is_low(lruvec, false)) if (inactive_list_is_low(lruvec, false))
shrink_active_list(SWAP_CLUSTER_MAX, lruvec, shrink_active_list(SWAP_CLUSTER_MAX, lruvec,
...@@ -3193,7 +3192,7 @@ static int balance_pgdat(pg_data_t *pgdat, int order, int classzone_idx) ...@@ -3193,7 +3192,7 @@ static int balance_pgdat(pg_data_t *pgdat, int order, int classzone_idx)
* pages are rotated regardless of classzone as this is * pages are rotated regardless of classzone as this is
* about consistent aging. * about consistent aging.
*/ */
age_active_anon(pgdat, &pgdat->node_zones[MAX_NR_ZONES - 1], &sc); age_active_anon(pgdat, &sc);
/* /*
* If we're getting trouble reclaiming, start doing writepage * If we're getting trouble reclaiming, start doing writepage
...@@ -3205,7 +3204,7 @@ static int balance_pgdat(pg_data_t *pgdat, int order, int classzone_idx) ...@@ -3205,7 +3204,7 @@ static int balance_pgdat(pg_data_t *pgdat, int order, int classzone_idx)
/* Call soft limit reclaim before calling shrink_node. */ /* Call soft limit reclaim before calling shrink_node. */
sc.nr_scanned = 0; sc.nr_scanned = 0;
nr_soft_scanned = 0; nr_soft_scanned = 0;
nr_soft_reclaimed = mem_cgroup_soft_limit_reclaim(zone, sc.order, nr_soft_reclaimed = mem_cgroup_soft_limit_reclaim(pgdat, sc.order,
sc.gfp_mask, &nr_soft_scanned); sc.gfp_mask, &nr_soft_scanned);
sc.nr_reclaimed += nr_soft_reclaimed; sc.nr_reclaimed += nr_soft_reclaimed;
......
...@@ -218,7 +218,7 @@ void *workingset_eviction(struct address_space *mapping, struct page *page) ...@@ -218,7 +218,7 @@ void *workingset_eviction(struct address_space *mapping, struct page *page)
VM_BUG_ON_PAGE(page_count(page), page); VM_BUG_ON_PAGE(page_count(page), page);
VM_BUG_ON_PAGE(!PageLocked(page), page); VM_BUG_ON_PAGE(!PageLocked(page), page);
lruvec = mem_cgroup_lruvec(zone->zone_pgdat, zone, memcg); lruvec = mem_cgroup_lruvec(zone->zone_pgdat, memcg);
eviction = atomic_long_inc_return(&lruvec->inactive_age); eviction = atomic_long_inc_return(&lruvec->inactive_age);
return pack_shadow(memcgid, zone, eviction); return pack_shadow(memcgid, zone, eviction);
} }
...@@ -267,7 +267,7 @@ bool workingset_refault(void *shadow) ...@@ -267,7 +267,7 @@ bool workingset_refault(void *shadow)
rcu_read_unlock(); rcu_read_unlock();
return false; return false;
} }
lruvec = mem_cgroup_lruvec(zone->zone_pgdat, zone, memcg); lruvec = mem_cgroup_lruvec(zone->zone_pgdat, memcg);
refault = atomic_long_read(&lruvec->inactive_age); refault = atomic_long_read(&lruvec->inactive_age);
active_file = lruvec_lru_size(lruvec, LRU_ACTIVE_FILE); active_file = lruvec_lru_size(lruvec, LRU_ACTIVE_FILE);
rcu_read_unlock(); rcu_read_unlock();
...@@ -319,7 +319,7 @@ void workingset_activation(struct page *page) ...@@ -319,7 +319,7 @@ void workingset_activation(struct page *page)
memcg = page_memcg_rcu(page); memcg = page_memcg_rcu(page);
if (!mem_cgroup_disabled() && !memcg) if (!mem_cgroup_disabled() && !memcg)
goto out; goto out;
lruvec = mem_cgroup_lruvec(page_pgdat(page), page_zone(page), memcg); lruvec = mem_cgroup_lruvec(page_pgdat(page), memcg);
atomic_long_inc(&lruvec->inactive_age); atomic_long_inc(&lruvec->inactive_age);
out: out:
rcu_read_unlock(); rcu_read_unlock();
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment