Commit 510ded33 authored by Tejun Heo's avatar Tejun Heo Committed by Linus Torvalds

slab: implement slab_root_caches list

With kmem cgroup support enabled, kmem_caches can be created and
destroyed frequently and a great number of near empty kmem_caches can
accumulate if there are a lot of transient cgroups and the system is not
under memory pressure.  When memory reclaim starts under such
conditions, it can lead to consecutive deactivation and destruction of
many kmem_caches, easily hundreds of thousands on moderately large
systems, exposing scalability issues in the current slab management
code.  This is one of the patches to address the issue.

slab_caches currently lists all caches including root and memcg ones.
This is the only data structure which lists the root caches and
iterating root caches can only be done by walking the list while
skipping over memcg caches.  As there can be a huge number of memcg
caches, this can become very expensive.

This also can make /proc/slabinfo behave very badly.  seq_file processes
reads in 4k chunks and seeks to the previous Nth position on slab_caches
list to resume after each chunk.  With a lot of memcg cache churns on
the list, reading /proc/slabinfo can become very slow and its content
often ends up with duplicate and/or missing entries.

This patch adds a new list slab_root_caches which lists only the root
caches.  When memcg is not enabled, it becomes just an alias of
slab_caches.  memcg specific list operations are collected into
memcg_[un]link_cache().

Link: http://lkml.kernel.org/r/20170117235411.9408-7-tj@kernel.orgSigned-off-by: default avatarTejun Heo <tj@kernel.org>
Reported-by: default avatarJay Vana <jsvana@fb.com>
Acked-by: default avatarVladimir Davydov <vdavydov@tarantool.org>
Cc: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: default avatarAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
parent bc2791f8
...@@ -556,6 +556,8 @@ struct memcg_cache_array { ...@@ -556,6 +556,8 @@ struct memcg_cache_array {
* used to index child cachces during allocation and cleared * used to index child cachces during allocation and cleared
* early during shutdown. * early during shutdown.
* *
* @root_caches_node: List node for slab_root_caches list.
*
* @children: List of all child caches. While the child caches are also * @children: List of all child caches. While the child caches are also
* reachable through @memcg_caches, a child cache remains on * reachable through @memcg_caches, a child cache remains on
* this list until it is actually destroyed. * this list until it is actually destroyed.
...@@ -573,6 +575,7 @@ struct memcg_cache_params { ...@@ -573,6 +575,7 @@ struct memcg_cache_params {
union { union {
struct { struct {
struct memcg_cache_array __rcu *memcg_caches; struct memcg_cache_array __rcu *memcg_caches;
struct list_head __root_caches_node;
struct list_head children; struct list_head children;
}; };
struct { struct {
......
...@@ -201,6 +201,11 @@ void __kmem_cache_free_bulk(struct kmem_cache *, size_t, void **); ...@@ -201,6 +201,11 @@ void __kmem_cache_free_bulk(struct kmem_cache *, size_t, void **);
int __kmem_cache_alloc_bulk(struct kmem_cache *, gfp_t, size_t, void **); int __kmem_cache_alloc_bulk(struct kmem_cache *, gfp_t, size_t, void **);
#if defined(CONFIG_MEMCG) && !defined(CONFIG_SLOB) #if defined(CONFIG_MEMCG) && !defined(CONFIG_SLOB)
/* List of all root caches. */
extern struct list_head slab_root_caches;
#define root_caches_node memcg_params.__root_caches_node
/* /*
* Iterate over all memcg caches of the given root cache. The caller must hold * Iterate over all memcg caches of the given root cache. The caller must hold
* slab_mutex. * slab_mutex.
...@@ -300,9 +305,14 @@ static __always_inline void memcg_uncharge_slab(struct page *page, int order, ...@@ -300,9 +305,14 @@ static __always_inline void memcg_uncharge_slab(struct page *page, int order,
} }
extern void slab_init_memcg_params(struct kmem_cache *); extern void slab_init_memcg_params(struct kmem_cache *);
extern void memcg_link_cache(struct kmem_cache *s);
#else /* CONFIG_MEMCG && !CONFIG_SLOB */ #else /* CONFIG_MEMCG && !CONFIG_SLOB */
/* If !memcg, all caches are root. */
#define slab_root_caches slab_caches
#define root_caches_node list
#define for_each_memcg_cache(iter, root) \ #define for_each_memcg_cache(iter, root) \
for ((void)(iter), (void)(root); 0; ) for ((void)(iter), (void)(root); 0; )
...@@ -347,6 +357,11 @@ static inline void memcg_uncharge_slab(struct page *page, int order, ...@@ -347,6 +357,11 @@ static inline void memcg_uncharge_slab(struct page *page, int order,
static inline void slab_init_memcg_params(struct kmem_cache *s) static inline void slab_init_memcg_params(struct kmem_cache *s)
{ {
} }
static inline void memcg_link_cache(struct kmem_cache *s)
{
}
#endif /* CONFIG_MEMCG && !CONFIG_SLOB */ #endif /* CONFIG_MEMCG && !CONFIG_SLOB */
static inline struct kmem_cache *cache_from_obj(struct kmem_cache *s, void *x) static inline struct kmem_cache *cache_from_obj(struct kmem_cache *s, void *x)
......
...@@ -138,6 +138,9 @@ int __kmem_cache_alloc_bulk(struct kmem_cache *s, gfp_t flags, size_t nr, ...@@ -138,6 +138,9 @@ int __kmem_cache_alloc_bulk(struct kmem_cache *s, gfp_t flags, size_t nr,
} }
#if defined(CONFIG_MEMCG) && !defined(CONFIG_SLOB) #if defined(CONFIG_MEMCG) && !defined(CONFIG_SLOB)
LIST_HEAD(slab_root_caches);
void slab_init_memcg_params(struct kmem_cache *s) void slab_init_memcg_params(struct kmem_cache *s)
{ {
s->memcg_params.root_cache = NULL; s->memcg_params.root_cache = NULL;
...@@ -183,9 +186,6 @@ static int update_memcg_params(struct kmem_cache *s, int new_array_size) ...@@ -183,9 +186,6 @@ static int update_memcg_params(struct kmem_cache *s, int new_array_size)
{ {
struct memcg_cache_array *old, *new; struct memcg_cache_array *old, *new;
if (!is_root_cache(s))
return 0;
new = kzalloc(sizeof(struct memcg_cache_array) + new = kzalloc(sizeof(struct memcg_cache_array) +
new_array_size * sizeof(void *), GFP_KERNEL); new_array_size * sizeof(void *), GFP_KERNEL);
if (!new) if (!new)
...@@ -209,7 +209,7 @@ int memcg_update_all_caches(int num_memcgs) ...@@ -209,7 +209,7 @@ int memcg_update_all_caches(int num_memcgs)
int ret = 0; int ret = 0;
mutex_lock(&slab_mutex); mutex_lock(&slab_mutex);
list_for_each_entry(s, &slab_caches, list) { list_for_each_entry(s, &slab_root_caches, root_caches_node) {
ret = update_memcg_params(s, num_memcgs); ret = update_memcg_params(s, num_memcgs);
/* /*
* Instead of freeing the memory, we'll just leave the caches * Instead of freeing the memory, we'll just leave the caches
...@@ -222,10 +222,26 @@ int memcg_update_all_caches(int num_memcgs) ...@@ -222,10 +222,26 @@ int memcg_update_all_caches(int num_memcgs)
return ret; return ret;
} }
static void unlink_memcg_cache(struct kmem_cache *s) void memcg_link_cache(struct kmem_cache *s)
{
if (is_root_cache(s)) {
list_add(&s->root_caches_node, &slab_root_caches);
} else {
list_add(&s->memcg_params.children_node,
&s->memcg_params.root_cache->memcg_params.children);
list_add(&s->memcg_params.kmem_caches_node,
&s->memcg_params.memcg->kmem_caches);
}
}
static void memcg_unlink_cache(struct kmem_cache *s)
{ {
if (is_root_cache(s)) {
list_del(&s->root_caches_node);
} else {
list_del(&s->memcg_params.children_node); list_del(&s->memcg_params.children_node);
list_del(&s->memcg_params.kmem_caches_node); list_del(&s->memcg_params.kmem_caches_node);
}
} }
#else #else
static inline int init_memcg_params(struct kmem_cache *s, static inline int init_memcg_params(struct kmem_cache *s,
...@@ -238,7 +254,7 @@ static inline void destroy_memcg_params(struct kmem_cache *s) ...@@ -238,7 +254,7 @@ static inline void destroy_memcg_params(struct kmem_cache *s)
{ {
} }
static inline void unlink_memcg_cache(struct kmem_cache *s) static inline void memcg_unlink_cache(struct kmem_cache *s)
{ {
} }
#endif /* CONFIG_MEMCG && !CONFIG_SLOB */ #endif /* CONFIG_MEMCG && !CONFIG_SLOB */
...@@ -285,7 +301,7 @@ struct kmem_cache *find_mergeable(size_t size, size_t align, ...@@ -285,7 +301,7 @@ struct kmem_cache *find_mergeable(size_t size, size_t align,
if (flags & SLAB_NEVER_MERGE) if (flags & SLAB_NEVER_MERGE)
return NULL; return NULL;
list_for_each_entry_reverse(s, &slab_caches, list) { list_for_each_entry_reverse(s, &slab_root_caches, root_caches_node) {
if (slab_unmergeable(s)) if (slab_unmergeable(s))
continue; continue;
...@@ -369,6 +385,7 @@ static struct kmem_cache *create_cache(const char *name, ...@@ -369,6 +385,7 @@ static struct kmem_cache *create_cache(const char *name,
s->refcount = 1; s->refcount = 1;
list_add(&s->list, &slab_caches); list_add(&s->list, &slab_caches);
memcg_link_cache(s);
out: out:
if (err) if (err)
return ERR_PTR(err); return ERR_PTR(err);
...@@ -514,9 +531,8 @@ static int shutdown_cache(struct kmem_cache *s) ...@@ -514,9 +531,8 @@ static int shutdown_cache(struct kmem_cache *s)
if (__kmem_cache_shutdown(s) != 0) if (__kmem_cache_shutdown(s) != 0)
return -EBUSY; return -EBUSY;
memcg_unlink_cache(s);
list_del(&s->list); list_del(&s->list);
if (!is_root_cache(s))
unlink_memcg_cache(s);
if (s->flags & SLAB_DESTROY_BY_RCU) { if (s->flags & SLAB_DESTROY_BY_RCU) {
list_add_tail(&s->list, &slab_caches_to_rcu_destroy); list_add_tail(&s->list, &slab_caches_to_rcu_destroy);
...@@ -596,10 +612,6 @@ void memcg_create_kmem_cache(struct mem_cgroup *memcg, ...@@ -596,10 +612,6 @@ void memcg_create_kmem_cache(struct mem_cgroup *memcg,
goto out_unlock; goto out_unlock;
} }
list_add(&s->memcg_params.children_node,
&root_cache->memcg_params.children);
list_add(&s->memcg_params.kmem_caches_node, &memcg->kmem_caches);
/* /*
* Since readers won't lock (see cache_from_memcg_idx()), we need a * Since readers won't lock (see cache_from_memcg_idx()), we need a
* barrier here to ensure nobody will see the kmem_cache partially * barrier here to ensure nobody will see the kmem_cache partially
...@@ -627,10 +639,7 @@ void memcg_deactivate_kmem_caches(struct mem_cgroup *memcg) ...@@ -627,10 +639,7 @@ void memcg_deactivate_kmem_caches(struct mem_cgroup *memcg)
get_online_mems(); get_online_mems();
mutex_lock(&slab_mutex); mutex_lock(&slab_mutex);
list_for_each_entry(s, &slab_caches, list) { list_for_each_entry(s, &slab_root_caches, root_caches_node) {
if (!is_root_cache(s))
continue;
arr = rcu_dereference_protected(s->memcg_params.memcg_caches, arr = rcu_dereference_protected(s->memcg_params.memcg_caches,
lockdep_is_held(&slab_mutex)); lockdep_is_held(&slab_mutex));
c = arr->entries[idx]; c = arr->entries[idx];
...@@ -829,6 +838,7 @@ struct kmem_cache *__init create_kmalloc_cache(const char *name, size_t size, ...@@ -829,6 +838,7 @@ struct kmem_cache *__init create_kmalloc_cache(const char *name, size_t size,
create_boot_cache(s, name, size, flags); create_boot_cache(s, name, size, flags);
list_add(&s->list, &slab_caches); list_add(&s->list, &slab_caches);
memcg_link_cache(s);
s->refcount = 1; s->refcount = 1;
return s; return s;
} }
...@@ -1136,12 +1146,12 @@ static void print_slabinfo_header(struct seq_file *m) ...@@ -1136,12 +1146,12 @@ static void print_slabinfo_header(struct seq_file *m)
void *slab_start(struct seq_file *m, loff_t *pos) void *slab_start(struct seq_file *m, loff_t *pos)
{ {
mutex_lock(&slab_mutex); mutex_lock(&slab_mutex);
return seq_list_start(&slab_caches, *pos); return seq_list_start(&slab_root_caches, *pos);
} }
void *slab_next(struct seq_file *m, void *p, loff_t *pos) void *slab_next(struct seq_file *m, void *p, loff_t *pos)
{ {
return seq_list_next(p, &slab_caches, pos); return seq_list_next(p, &slab_root_caches, pos);
} }
void slab_stop(struct seq_file *m, void *p) void slab_stop(struct seq_file *m, void *p)
...@@ -1193,11 +1203,10 @@ static void cache_show(struct kmem_cache *s, struct seq_file *m) ...@@ -1193,11 +1203,10 @@ static void cache_show(struct kmem_cache *s, struct seq_file *m)
static int slab_show(struct seq_file *m, void *p) static int slab_show(struct seq_file *m, void *p)
{ {
struct kmem_cache *s = list_entry(p, struct kmem_cache, list); struct kmem_cache *s = list_entry(p, struct kmem_cache, root_caches_node);
if (p == slab_caches.next) if (p == slab_root_caches.next)
print_slabinfo_header(m); print_slabinfo_header(m);
if (is_root_cache(s))
cache_show(s, m); cache_show(s, m);
return 0; return 0;
} }
......
...@@ -4127,6 +4127,7 @@ static struct kmem_cache * __init bootstrap(struct kmem_cache *static_cache) ...@@ -4127,6 +4127,7 @@ static struct kmem_cache * __init bootstrap(struct kmem_cache *static_cache)
} }
slab_init_memcg_params(s); slab_init_memcg_params(s);
list_add(&s->list, &slab_caches); list_add(&s->list, &slab_caches);
memcg_link_cache(s);
return s; return s;
} }
......
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