* [PATCH v2 01/21] mm: list_lru: fix list_lru_count_one() return value
2021-05-27 6:21 [PATCH v2 00/21] Optimize list lru memory consumption Muchun Song
@ 2021-05-27 6:21 ` Muchun Song
2021-05-27 6:21 ` [PATCH v2 02/21] mm: memcontrol: remove kmemcg_id reparenting Muchun Song
` (20 subsequent siblings)
21 siblings, 0 replies; 29+ messages in thread
From: Muchun Song @ 2021-05-27 6:21 UTC (permalink / raw)
To: willy, akpm, hannes, mhocko, vdavydov.dev, shakeelb, guro,
shy828301, alexs, richard.weiyang, david, trond.myklebust,
anna.schumaker
Cc: linux-fsdevel, linux-kernel, linux-mm, linux-nfs, zhengqi.arch,
duanxiongchun, fam.zheng, Muchun Song
Since commit 2788cf0c401c ("memcg: reparent list_lrus and free kmemcg_id
on css offline"), the ->nr_items can be negative during memory cgroup
reparenting. In this case, list_lru_count_one() returns an unusual and
huge value. It can surprise users. So let it return zero when ->nr_items
is negative.
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
---
mm/list_lru.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/mm/list_lru.c b/mm/list_lru.c
index cd58790d0fb3..4962d48d4410 100644
--- a/mm/list_lru.c
+++ b/mm/list_lru.c
@@ -176,13 +176,16 @@ unsigned long list_lru_count_one(struct list_lru *lru,
{
struct list_lru_node *nlru = &lru->node[nid];
struct list_lru_one *l;
- unsigned long count;
+ long count;
rcu_read_lock();
l = list_lru_from_memcg_idx(nlru, memcg_cache_id(memcg));
count = READ_ONCE(l->nr_items);
rcu_read_unlock();
+ if (unlikely(count < 0))
+ count = 0;
+
return count;
}
EXPORT_SYMBOL_GPL(list_lru_count_one);
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v2 02/21] mm: memcontrol: remove kmemcg_id reparenting
2021-05-27 6:21 [PATCH v2 00/21] Optimize list lru memory consumption Muchun Song
2021-05-27 6:21 ` [PATCH v2 01/21] mm: list_lru: fix list_lru_count_one() return value Muchun Song
@ 2021-05-27 6:21 ` Muchun Song
2021-05-27 6:21 ` [PATCH v2 03/21] mm: memcontrol: remove the kmem states Muchun Song
` (19 subsequent siblings)
21 siblings, 0 replies; 29+ messages in thread
From: Muchun Song @ 2021-05-27 6:21 UTC (permalink / raw)
To: willy, akpm, hannes, mhocko, vdavydov.dev, shakeelb, guro,
shy828301, alexs, richard.weiyang, david, trond.myklebust,
anna.schumaker
Cc: linux-fsdevel, linux-kernel, linux-mm, linux-nfs, zhengqi.arch,
duanxiongchun, fam.zheng, Muchun Song
Since slab objects and kmem pages are charged to object cgroup instead
of memory cgroup, memcg_reparent_objcgs() will reparent this cgroup and
all its descendants to the parent cgroup. This already makes further
list_lru_add()'s add elements to the parent's list. So we do not need
to change kmemcg_id of an offline cgroup to its parent's id. It is just
waste CPU cycles. Just remove those redundant code.
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
---
mm/memcontrol.c | 20 ++------------------
1 file changed, 2 insertions(+), 18 deletions(-)
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index db29b96f7311..9add859f69d7 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -3654,8 +3654,7 @@ static int memcg_online_kmem(struct mem_cgroup *memcg)
static void memcg_offline_kmem(struct mem_cgroup *memcg)
{
- struct cgroup_subsys_state *css;
- struct mem_cgroup *parent, *child;
+ struct mem_cgroup *parent;
int kmemcg_id;
if (memcg->kmem_state != KMEM_ONLINE)
@@ -3672,22 +3671,7 @@ static void memcg_offline_kmem(struct mem_cgroup *memcg)
kmemcg_id = memcg->kmemcg_id;
BUG_ON(kmemcg_id < 0);
- /*
- * Change kmemcg_id of this cgroup and all its descendants to the
- * parent's id, and then move all entries from this cgroup's list_lrus
- * to ones of the parent. After we have finished, all list_lrus
- * corresponding to this cgroup are guaranteed to remain empty. The
- * ordering is imposed by list_lru_node->lock taken by
- * memcg_drain_all_list_lrus().
- */
- rcu_read_lock(); /* can be called from css_free w/o cgroup_mutex */
- css_for_each_descendant_pre(css, &memcg->css) {
- child = mem_cgroup_from_css(css);
- BUG_ON(child->kmemcg_id != kmemcg_id);
- child->kmemcg_id = parent->kmemcg_id;
- }
- rcu_read_unlock();
-
+ /* memcg_reparent_objcgs() must be called before this. */
memcg_drain_all_list_lrus(kmemcg_id, parent);
memcg_free_cache_id(kmemcg_id);
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v2 03/21] mm: memcontrol: remove the kmem states
2021-05-27 6:21 [PATCH v2 00/21] Optimize list lru memory consumption Muchun Song
2021-05-27 6:21 ` [PATCH v2 01/21] mm: list_lru: fix list_lru_count_one() return value Muchun Song
2021-05-27 6:21 ` [PATCH v2 02/21] mm: memcontrol: remove kmemcg_id reparenting Muchun Song
@ 2021-05-27 6:21 ` Muchun Song
2021-05-27 6:21 ` [PATCH v2 04/21] mm: memcontrol: do it in mem_cgroup_css_online to make the kmem online Muchun Song
` (18 subsequent siblings)
21 siblings, 0 replies; 29+ messages in thread
From: Muchun Song @ 2021-05-27 6:21 UTC (permalink / raw)
To: willy, akpm, hannes, mhocko, vdavydov.dev, shakeelb, guro,
shy828301, alexs, richard.weiyang, david, trond.myklebust,
anna.schumaker
Cc: linux-fsdevel, linux-kernel, linux-mm, linux-nfs, zhengqi.arch,
duanxiongchun, fam.zheng, Muchun Song
Now the kmem states is only used to indicate whether the kmem is
offline. Because css_alloc() could fail, then we didn't make the
kmem offline. In this case, we need the kmem state to mark this
so that memcg_free_kmem() can make the kmem offline.
However, we can set ->kmemcg_id to -1 to indicate the kmem has
been offline. So we can remove the kmem states to simplify the
code.
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
---
include/linux/memcontrol.h | 7 -------
mm/memcontrol.c | 9 +++------
2 files changed, 3 insertions(+), 13 deletions(-)
diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index 0ce97eff79e2..6d0638e13fc1 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -182,12 +182,6 @@ struct mem_cgroup_thresholds {
struct mem_cgroup_threshold_ary *spare;
};
-enum memcg_kmem_state {
- KMEM_NONE,
- KMEM_ALLOCATED,
- KMEM_ONLINE,
-};
-
#if defined(CONFIG_SMP)
struct memcg_padding {
char x[0];
@@ -320,7 +314,6 @@ struct mem_cgroup {
#ifdef CONFIG_MEMCG_KMEM
int kmemcg_id;
- enum memcg_kmem_state kmem_state;
struct obj_cgroup __rcu *objcg;
struct list_head objcg_list; /* list of inherited objcgs */
#endif
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 9add859f69d7..23a9fc8dc143 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -3630,7 +3630,6 @@ static int memcg_online_kmem(struct mem_cgroup *memcg)
return 0;
BUG_ON(memcg->kmemcg_id >= 0);
- BUG_ON(memcg->kmem_state);
memcg_id = memcg_alloc_cache_id();
if (memcg_id < 0)
@@ -3647,7 +3646,6 @@ static int memcg_online_kmem(struct mem_cgroup *memcg)
static_branch_enable(&memcg_kmem_enabled_key);
memcg->kmemcg_id = memcg_id;
- memcg->kmem_state = KMEM_ONLINE;
return 0;
}
@@ -3657,11 +3655,9 @@ static void memcg_offline_kmem(struct mem_cgroup *memcg)
struct mem_cgroup *parent;
int kmemcg_id;
- if (memcg->kmem_state != KMEM_ONLINE)
+ if (cgroup_memory_nokmem)
return;
- memcg->kmem_state = KMEM_ALLOCATED;
-
parent = parent_mem_cgroup(memcg);
if (!parent)
parent = root_mem_cgroup;
@@ -3675,12 +3671,13 @@ static void memcg_offline_kmem(struct mem_cgroup *memcg)
memcg_drain_all_list_lrus(kmemcg_id, parent);
memcg_free_cache_id(kmemcg_id);
+ memcg->kmemcg_id = -1;
}
static void memcg_free_kmem(struct mem_cgroup *memcg)
{
/* css_alloc() failed, offlining didn't happen */
- if (unlikely(memcg->kmem_state == KMEM_ONLINE))
+ if (unlikely(memcg->kmemcg_id != -1))
memcg_offline_kmem(memcg);
}
#else
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v2 04/21] mm: memcontrol: do it in mem_cgroup_css_online to make the kmem online
2021-05-27 6:21 [PATCH v2 00/21] Optimize list lru memory consumption Muchun Song
` (2 preceding siblings ...)
2021-05-27 6:21 ` [PATCH v2 03/21] mm: memcontrol: remove the kmem states Muchun Song
@ 2021-05-27 6:21 ` Muchun Song
2021-05-27 6:21 ` [PATCH v2 05/21] mm: list_lru: remove lru node locking from memcg_update_list_lru_node Muchun Song
` (17 subsequent siblings)
21 siblings, 0 replies; 29+ messages in thread
From: Muchun Song @ 2021-05-27 6:21 UTC (permalink / raw)
To: willy, akpm, hannes, mhocko, vdavydov.dev, shakeelb, guro,
shy828301, alexs, richard.weiyang, david, trond.myklebust,
anna.schumaker
Cc: linux-fsdevel, linux-kernel, linux-mm, linux-nfs, zhengqi.arch,
duanxiongchun, fam.zheng, Muchun Song
If we do it in the mem_cgroup_css_online() to make the kmem online,
we do not need to set ->kmemcg_id when the kmem is offline. And we
also can remove memcg_free_kmem(). So just do that to simplify the
code.
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
---
mm/memcontrol.c | 42 +++++++++++++++---------------------------
1 file changed, 15 insertions(+), 27 deletions(-)
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 23a9fc8dc143..377ec9847179 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -3629,7 +3629,8 @@ static int memcg_online_kmem(struct mem_cgroup *memcg)
if (cgroup_memory_nokmem)
return 0;
- BUG_ON(memcg->kmemcg_id >= 0);
+ if (unlikely(mem_cgroup_is_root(memcg)))
+ return 0;
memcg_id = memcg_alloc_cache_id();
if (memcg_id < 0)
@@ -3658,6 +3659,9 @@ static void memcg_offline_kmem(struct mem_cgroup *memcg)
if (cgroup_memory_nokmem)
return;
+ if (unlikely(mem_cgroup_is_root(memcg)))
+ return;
+
parent = parent_mem_cgroup(memcg);
if (!parent)
parent = root_mem_cgroup;
@@ -3665,20 +3669,11 @@ static void memcg_offline_kmem(struct mem_cgroup *memcg)
memcg_reparent_objcgs(memcg, parent);
kmemcg_id = memcg->kmemcg_id;
- BUG_ON(kmemcg_id < 0);
/* memcg_reparent_objcgs() must be called before this. */
memcg_drain_all_list_lrus(kmemcg_id, parent);
memcg_free_cache_id(kmemcg_id);
- memcg->kmemcg_id = -1;
-}
-
-static void memcg_free_kmem(struct mem_cgroup *memcg)
-{
- /* css_alloc() failed, offlining didn't happen */
- if (unlikely(memcg->kmemcg_id != -1))
- memcg_offline_kmem(memcg);
}
#else
static int memcg_online_kmem(struct mem_cgroup *memcg)
@@ -3688,9 +3683,6 @@ static int memcg_online_kmem(struct mem_cgroup *memcg)
static void memcg_offline_kmem(struct mem_cgroup *memcg)
{
}
-static void memcg_free_kmem(struct mem_cgroup *memcg)
-{
-}
#endif /* CONFIG_MEMCG_KMEM */
static int memcg_update_kmem_max(struct mem_cgroup *memcg,
@@ -5219,7 +5211,6 @@ mem_cgroup_css_alloc(struct cgroup_subsys_state *parent_css)
{
struct mem_cgroup *parent = mem_cgroup_from_css(parent_css);
struct mem_cgroup *memcg, *old_memcg;
- long error = -ENOMEM;
old_memcg = set_active_memcg(parent);
memcg = mem_cgroup_alloc();
@@ -5249,38 +5240,36 @@ mem_cgroup_css_alloc(struct cgroup_subsys_state *parent_css)
}
/* The following stuff does not apply to the root */
- error = memcg_online_kmem(memcg);
- if (error)
- goto fail;
-
if (cgroup_subsys_on_dfl(memory_cgrp_subsys) && !cgroup_memory_nosocket)
static_branch_inc(&memcg_sockets_enabled_key);
return &memcg->css;
-fail:
- mem_cgroup_id_remove(memcg);
- mem_cgroup_free(memcg);
- return ERR_PTR(error);
}
static int mem_cgroup_css_online(struct cgroup_subsys_state *css)
{
struct mem_cgroup *memcg = mem_cgroup_from_css(css);
+ if (memcg_online_kmem(memcg))
+ goto remove_id;
+
/*
* A memcg must be visible for expand_shrinker_info()
* by the time the maps are allocated. So, we allocate maps
* here, when for_each_mem_cgroup() can't skip it.
*/
- if (alloc_shrinker_info(memcg)) {
- mem_cgroup_id_remove(memcg);
- return -ENOMEM;
- }
+ if (alloc_shrinker_info(memcg))
+ goto offline_kmem;
/* Online state pins memcg ID, memcg ID pins CSS */
refcount_set(&memcg->id.ref, 1);
css_get(css);
return 0;
+offline_kmem:
+ memcg_offline_kmem(memcg);
+remove_id:
+ mem_cgroup_id_remove(memcg);
+ return -ENOMEM;
}
static void mem_cgroup_css_offline(struct cgroup_subsys_state *css)
@@ -5338,7 +5327,6 @@ static void mem_cgroup_css_free(struct cgroup_subsys_state *css)
cancel_work_sync(&memcg->high_work);
mem_cgroup_remove_from_trees(memcg);
free_shrinker_info(memcg);
- memcg_free_kmem(memcg);
mem_cgroup_free(memcg);
}
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v2 05/21] mm: list_lru: remove lru node locking from memcg_update_list_lru_node
2021-05-27 6:21 [PATCH v2 00/21] Optimize list lru memory consumption Muchun Song
` (3 preceding siblings ...)
2021-05-27 6:21 ` [PATCH v2 04/21] mm: memcontrol: do it in mem_cgroup_css_online to make the kmem online Muchun Song
@ 2021-05-27 6:21 ` Muchun Song
2021-05-27 6:21 ` [PATCH v2 06/21] mm: list_lru: only add the memcg aware lrus to the list_lrus Muchun Song
` (16 subsequent siblings)
21 siblings, 0 replies; 29+ messages in thread
From: Muchun Song @ 2021-05-27 6:21 UTC (permalink / raw)
To: willy, akpm, hannes, mhocko, vdavydov.dev, shakeelb, guro,
shy828301, alexs, richard.weiyang, david, trond.myklebust,
anna.schumaker
Cc: linux-fsdevel, linux-kernel, linux-mm, linux-nfs, zhengqi.arch,
duanxiongchun, fam.zheng, Muchun Song
Since commit e5bc3af7734f ("rcu: Consolidate PREEMPT and !PREEMPT
synchronize_rcu()"), the critical section of spinlock can serve
as RCU read-side critical section. So we can remove the locking
from memcg_update_list_lru_node().
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
---
mm/list_lru.c | 11 +----------
1 file changed, 1 insertion(+), 10 deletions(-)
diff --git a/mm/list_lru.c b/mm/list_lru.c
index 4962d48d4410..e86d4d055d3c 100644
--- a/mm/list_lru.c
+++ b/mm/list_lru.c
@@ -403,18 +403,9 @@ static int memcg_update_list_lru_node(struct list_lru_node *nlru,
memcpy(&new->lru, &old->lru, old_size * sizeof(void *));
- /*
- * The locking below allows readers that hold nlru->lock avoid taking
- * rcu_read_lock (see list_lru_from_memcg_idx).
- *
- * Since list_lru_{add,del} may be called under an IRQ-safe lock,
- * we have to use IRQ-safe primitives here to avoid deadlock.
- */
- spin_lock_irq(&nlru->lock);
rcu_assign_pointer(nlru->memcg_lrus, new);
- spin_unlock_irq(&nlru->lock);
-
kvfree_rcu(old, rcu);
+
return 0;
}
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v2 06/21] mm: list_lru: only add the memcg aware lrus to the list_lrus
2021-05-27 6:21 [PATCH v2 00/21] Optimize list lru memory consumption Muchun Song
` (4 preceding siblings ...)
2021-05-27 6:21 ` [PATCH v2 05/21] mm: list_lru: remove lru node locking from memcg_update_list_lru_node Muchun Song
@ 2021-05-27 6:21 ` Muchun Song
2021-05-27 6:21 ` [PATCH v2 07/21] mm: list_lru: optimize the array of per memcg lists memory consumption Muchun Song
` (15 subsequent siblings)
21 siblings, 0 replies; 29+ messages in thread
From: Muchun Song @ 2021-05-27 6:21 UTC (permalink / raw)
To: willy, akpm, hannes, mhocko, vdavydov.dev, shakeelb, guro,
shy828301, alexs, richard.weiyang, david, trond.myklebust,
anna.schumaker
Cc: linux-fsdevel, linux-kernel, linux-mm, linux-nfs, zhengqi.arch,
duanxiongchun, fam.zheng, Muchun Song
We need to traverse every lru in the list_lrus in some routines, but skip
memcg non-aware lrus. However, we can only add the memcg aware lrus to
the list_lrus. This can be efficient to traverse the list.
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
---
mm/list_lru.c | 25 +++++++++++--------------
1 file changed, 11 insertions(+), 14 deletions(-)
diff --git a/mm/list_lru.c b/mm/list_lru.c
index e86d4d055d3c..bed699edabe5 100644
--- a/mm/list_lru.c
+++ b/mm/list_lru.c
@@ -18,8 +18,16 @@
static LIST_HEAD(list_lrus);
static DEFINE_MUTEX(list_lrus_mutex);
+static inline bool list_lru_memcg_aware(struct list_lru *lru)
+{
+ return lru->memcg_aware;
+}
+
static void list_lru_register(struct list_lru *lru)
{
+ if (!list_lru_memcg_aware(lru))
+ return;
+
mutex_lock(&list_lrus_mutex);
list_add(&lru->list, &list_lrus);
mutex_unlock(&list_lrus_mutex);
@@ -27,6 +35,9 @@ static void list_lru_register(struct list_lru *lru)
static void list_lru_unregister(struct list_lru *lru)
{
+ if (!list_lru_memcg_aware(lru))
+ return;
+
mutex_lock(&list_lrus_mutex);
list_del(&lru->list);
mutex_unlock(&list_lrus_mutex);
@@ -37,11 +48,6 @@ static int lru_shrinker_id(struct list_lru *lru)
return lru->shrinker_id;
}
-static inline bool list_lru_memcg_aware(struct list_lru *lru)
-{
- return lru->memcg_aware;
-}
-
static inline struct list_lru_one *
list_lru_from_memcg_idx(struct list_lru_node *nlru, int idx)
{
@@ -460,9 +466,6 @@ static int memcg_update_list_lru(struct list_lru *lru,
{
int i;
- if (!list_lru_memcg_aware(lru))
- return 0;
-
for_each_node(i) {
if (memcg_update_list_lru_node(&lru->node[i],
old_size, new_size))
@@ -485,9 +488,6 @@ static void memcg_cancel_update_list_lru(struct list_lru *lru,
{
int i;
- if (!list_lru_memcg_aware(lru))
- return;
-
for_each_node(i)
memcg_cancel_update_list_lru_node(&lru->node[i],
old_size, new_size);
@@ -546,9 +546,6 @@ static void memcg_drain_list_lru(struct list_lru *lru,
{
int i;
- if (!list_lru_memcg_aware(lru))
- return;
-
for_each_node(i)
memcg_drain_list_lru_node(lru, i, src_idx, dst_memcg);
}
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v2 07/21] mm: list_lru: optimize the array of per memcg lists memory consumption
2021-05-27 6:21 [PATCH v2 00/21] Optimize list lru memory consumption Muchun Song
` (5 preceding siblings ...)
2021-05-27 6:21 ` [PATCH v2 06/21] mm: list_lru: only add the memcg aware lrus to the list_lrus Muchun Song
@ 2021-05-27 6:21 ` Muchun Song
2021-05-27 6:21 ` [PATCH v2 08/21] mm: list_lru: remove memcg_aware field from struct list_lru Muchun Song
` (14 subsequent siblings)
21 siblings, 0 replies; 29+ messages in thread
From: Muchun Song @ 2021-05-27 6:21 UTC (permalink / raw)
To: willy, akpm, hannes, mhocko, vdavydov.dev, shakeelb, guro,
shy828301, alexs, richard.weiyang, david, trond.myklebust,
anna.schumaker
Cc: linux-fsdevel, linux-kernel, linux-mm, linux-nfs, zhengqi.arch,
duanxiongchun, fam.zheng, Muchun Song
The list_lru use an array to store the list_lru_one pointers, which is
per-memcg per-node. What if we run 10k containers in the system? The
size of the array of every list_lru can be 10k * number_of_node *
sizeof(void *). The array size becomes very big, the more numa node
in the system, the more memory we consume. We can convert the array
to per memcg instead of per-memcg per-node. It can save memory
especially when there are many numa nodes in the system. And also
simplify the code.
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
---
include/linux/list_lru.h | 17 +++--
mm/list_lru.c | 191 +++++++++++++++++------------------------------
2 files changed, 79 insertions(+), 129 deletions(-)
diff --git a/include/linux/list_lru.h b/include/linux/list_lru.h
index 9dcaa3e582c9..20a43904001d 100644
--- a/include/linux/list_lru.h
+++ b/include/linux/list_lru.h
@@ -31,10 +31,15 @@ struct list_lru_one {
long nr_items;
};
+struct list_lru_per_memcg {
+ /* array of per cgroup per node lists, indexed by node id */
+ struct list_lru_one nodes[0];
+};
+
struct list_lru_memcg {
- struct rcu_head rcu;
+ struct rcu_head rcu;
/* array of per cgroup lists, indexed by memcg_cache_id */
- struct list_lru_one *lru[];
+ struct list_lru_per_memcg *lrus[];
};
struct list_lru_node {
@@ -42,11 +47,7 @@ struct list_lru_node {
spinlock_t lock;
/* global list, used for the root cgroup in cgroup aware lrus */
struct list_lru_one lru;
-#ifdef CONFIG_MEMCG_KMEM
- /* for cgroup aware lrus points to per cgroup lists, otherwise NULL */
- struct list_lru_memcg __rcu *memcg_lrus;
-#endif
- long nr_items;
+ long nr_items;
} ____cacheline_aligned_in_smp;
struct list_lru {
@@ -55,6 +56,8 @@ struct list_lru {
struct list_head list;
int shrinker_id;
bool memcg_aware;
+ /* for cgroup aware lrus points to per cgroup lists, otherwise NULL */
+ struct list_lru_memcg __rcu *memcg_lrus;
#endif
};
diff --git a/mm/list_lru.c b/mm/list_lru.c
index bed699edabe5..e52f5a91fa0f 100644
--- a/mm/list_lru.c
+++ b/mm/list_lru.c
@@ -49,35 +49,38 @@ static int lru_shrinker_id(struct list_lru *lru)
}
static inline struct list_lru_one *
-list_lru_from_memcg_idx(struct list_lru_node *nlru, int idx)
+list_lru_from_memcg_idx(struct list_lru *lru, int nid, int idx)
{
struct list_lru_memcg *memcg_lrus;
+ struct list_lru_node *nlru = &lru->node[nid];
+
/*
* Either lock or RCU protects the array of per cgroup lists
- * from relocation (see memcg_update_list_lru_node).
+ * from relocation (see memcg_update_list_lru).
*/
- memcg_lrus = rcu_dereference_check(nlru->memcg_lrus,
+ memcg_lrus = rcu_dereference_check(lru->memcg_lrus,
lockdep_is_held(&nlru->lock));
if (memcg_lrus && idx >= 0)
- return memcg_lrus->lru[idx];
+ return &memcg_lrus->lrus[idx]->nodes[nid];
return &nlru->lru;
}
static inline struct list_lru_one *
-list_lru_from_kmem(struct list_lru_node *nlru, void *ptr,
+list_lru_from_kmem(struct list_lru *lru, int nid, void *ptr,
struct mem_cgroup **memcg_ptr)
{
+ struct list_lru_node *nlru = &lru->node[nid];
struct list_lru_one *l = &nlru->lru;
struct mem_cgroup *memcg = NULL;
- if (!nlru->memcg_lrus)
+ if (!lru->memcg_lrus)
goto out;
memcg = mem_cgroup_from_obj(ptr);
if (!memcg)
goto out;
- l = list_lru_from_memcg_idx(nlru, memcg_cache_id(memcg));
+ l = list_lru_from_memcg_idx(lru, nid, memcg_cache_id(memcg));
out:
if (memcg_ptr)
*memcg_ptr = memcg;
@@ -103,18 +106,18 @@ static inline bool list_lru_memcg_aware(struct list_lru *lru)
}
static inline struct list_lru_one *
-list_lru_from_memcg_idx(struct list_lru_node *nlru, int idx)
+list_lru_from_memcg_idx(struct list_lru *lru, int nid, int idx)
{
- return &nlru->lru;
+ return &lru->node[nid].lru;
}
static inline struct list_lru_one *
-list_lru_from_kmem(struct list_lru_node *nlru, void *ptr,
+list_lru_from_kmem(struct list_lru *lru, int nid, void *ptr,
struct mem_cgroup **memcg_ptr)
{
if (memcg_ptr)
*memcg_ptr = NULL;
- return &nlru->lru;
+ return &lru->node[nid].lru;
}
#endif /* CONFIG_MEMCG_KMEM */
@@ -127,7 +130,7 @@ bool list_lru_add(struct list_lru *lru, struct list_head *item)
spin_lock(&nlru->lock);
if (list_empty(item)) {
- l = list_lru_from_kmem(nlru, item, &memcg);
+ l = list_lru_from_kmem(lru, nid, item, &memcg);
list_add_tail(item, &l->list);
/* Set shrinker bit if the first element was added */
if (!l->nr_items++)
@@ -150,7 +153,7 @@ bool list_lru_del(struct list_lru *lru, struct list_head *item)
spin_lock(&nlru->lock);
if (!list_empty(item)) {
- l = list_lru_from_kmem(nlru, item, NULL);
+ l = list_lru_from_kmem(lru, nid, item, NULL);
list_del_init(item);
l->nr_items--;
nlru->nr_items--;
@@ -180,12 +183,11 @@ EXPORT_SYMBOL_GPL(list_lru_isolate_move);
unsigned long list_lru_count_one(struct list_lru *lru,
int nid, struct mem_cgroup *memcg)
{
- struct list_lru_node *nlru = &lru->node[nid];
struct list_lru_one *l;
long count;
rcu_read_lock();
- l = list_lru_from_memcg_idx(nlru, memcg_cache_id(memcg));
+ l = list_lru_from_memcg_idx(lru, nid, memcg_cache_id(memcg));
count = READ_ONCE(l->nr_items);
rcu_read_unlock();
@@ -206,16 +208,16 @@ unsigned long list_lru_count_node(struct list_lru *lru, int nid)
EXPORT_SYMBOL_GPL(list_lru_count_node);
static unsigned long
-__list_lru_walk_one(struct list_lru_node *nlru, int memcg_idx,
+__list_lru_walk_one(struct list_lru *lru, int nid, int memcg_idx,
list_lru_walk_cb isolate, void *cb_arg,
unsigned long *nr_to_walk)
{
-
+ struct list_lru_node *nlru = &lru->node[nid];
struct list_lru_one *l;
struct list_head *item, *n;
unsigned long isolated = 0;
- l = list_lru_from_memcg_idx(nlru, memcg_idx);
+ l = list_lru_from_memcg_idx(lru, nid, memcg_idx);
restart:
list_for_each_safe(item, n, &l->list) {
enum lru_status ret;
@@ -272,8 +274,8 @@ list_lru_walk_one(struct list_lru *lru, int nid, struct mem_cgroup *memcg,
unsigned long ret;
spin_lock(&nlru->lock);
- ret = __list_lru_walk_one(nlru, memcg_cache_id(memcg), isolate, cb_arg,
- nr_to_walk);
+ ret = __list_lru_walk_one(lru, nid, memcg_cache_id(memcg), isolate,
+ cb_arg, nr_to_walk);
spin_unlock(&nlru->lock);
return ret;
}
@@ -288,8 +290,8 @@ list_lru_walk_one_irq(struct list_lru *lru, int nid, struct mem_cgroup *memcg,
unsigned long ret;
spin_lock_irq(&nlru->lock);
- ret = __list_lru_walk_one(nlru, memcg_cache_id(memcg), isolate, cb_arg,
- nr_to_walk);
+ ret = __list_lru_walk_one(lru, nid, memcg_cache_id(memcg), isolate,
+ cb_arg, nr_to_walk);
spin_unlock_irq(&nlru->lock);
return ret;
}
@@ -308,7 +310,7 @@ unsigned long list_lru_walk_node(struct list_lru *lru, int nid,
struct list_lru_node *nlru = &lru->node[nid];
spin_lock(&nlru->lock);
- isolated += __list_lru_walk_one(nlru, memcg_idx,
+ isolated += __list_lru_walk_one(lru, nid, memcg_idx,
isolate, cb_arg,
nr_to_walk);
spin_unlock(&nlru->lock);
@@ -328,169 +330,114 @@ static void init_one_lru(struct list_lru_one *l)
}
#ifdef CONFIG_MEMCG_KMEM
-static void __memcg_destroy_list_lru_node(struct list_lru_memcg *memcg_lrus,
- int begin, int end)
+static void memcg_destroy_list_lru_range(struct list_lru_memcg *memcg_lrus,
+ int begin, int end)
{
int i;
for (i = begin; i < end; i++)
- kfree(memcg_lrus->lru[i]);
+ kfree(memcg_lrus->lrus[i]);
}
-static int __memcg_init_list_lru_node(struct list_lru_memcg *memcg_lrus,
- int begin, int end)
+static int memcg_init_list_lru_range(struct list_lru_memcg *memcg_lrus,
+ int begin, int end)
{
int i;
for (i = begin; i < end; i++) {
- struct list_lru_one *l;
+ int nid;
+ struct list_lru_per_memcg *lru;
- l = kmalloc(sizeof(struct list_lru_one), GFP_KERNEL);
- if (!l)
+ lru = kmalloc(struct_size(lru, nodes, nr_node_ids), GFP_KERNEL);
+ if (!lru)
goto fail;
- init_one_lru(l);
- memcg_lrus->lru[i] = l;
+ for_each_node(nid)
+ init_one_lru(&lru->nodes[nid]);
+ memcg_lrus->lrus[i] = lru;
}
return 0;
fail:
- __memcg_destroy_list_lru_node(memcg_lrus, begin, i);
+ memcg_destroy_list_lru_range(memcg_lrus, begin, i);
return -ENOMEM;
}
-static int memcg_init_list_lru_node(struct list_lru_node *nlru)
+static int memcg_init_list_lru(struct list_lru *lru, bool memcg_aware)
{
struct list_lru_memcg *memcg_lrus;
int size = memcg_nr_cache_ids;
+ lru->memcg_aware = memcg_aware;
+ if (!memcg_aware)
+ return 0;
+
memcg_lrus = kvmalloc(sizeof(*memcg_lrus) +
- size * sizeof(void *), GFP_KERNEL);
+ size * sizeof(memcg_lrus->lrus[0]), GFP_KERNEL);
if (!memcg_lrus)
return -ENOMEM;
- if (__memcg_init_list_lru_node(memcg_lrus, 0, size)) {
+ if (memcg_init_list_lru_range(memcg_lrus, 0, size)) {
kvfree(memcg_lrus);
return -ENOMEM;
}
- RCU_INIT_POINTER(nlru->memcg_lrus, memcg_lrus);
+ RCU_INIT_POINTER(lru->memcg_lrus, memcg_lrus);
return 0;
}
-static void memcg_destroy_list_lru_node(struct list_lru_node *nlru)
+static void memcg_destroy_list_lru(struct list_lru *lru)
{
struct list_lru_memcg *memcg_lrus;
+
+ if (!list_lru_memcg_aware(lru))
+ return;
+
/*
* This is called when shrinker has already been unregistered,
* and nobody can use it. So, there is no need to use kvfree_rcu().
*/
- memcg_lrus = rcu_dereference_protected(nlru->memcg_lrus, true);
- __memcg_destroy_list_lru_node(memcg_lrus, 0, memcg_nr_cache_ids);
+ memcg_lrus = rcu_dereference_protected(lru->memcg_lrus, true);
+ memcg_destroy_list_lru_range(memcg_lrus, 0, memcg_nr_cache_ids);
kvfree(memcg_lrus);
}
-static int memcg_update_list_lru_node(struct list_lru_node *nlru,
- int old_size, int new_size)
+static int memcg_update_list_lru(struct list_lru *lru, int old_size, int new_size)
{
struct list_lru_memcg *old, *new;
BUG_ON(old_size > new_size);
- old = rcu_dereference_protected(nlru->memcg_lrus,
+ old = rcu_dereference_protected(lru->memcg_lrus,
lockdep_is_held(&list_lrus_mutex));
- new = kvmalloc(sizeof(*new) + new_size * sizeof(void *), GFP_KERNEL);
+ new = kvmalloc(sizeof(*new) + new_size * sizeof(new->lrus[0]), GFP_KERNEL);
if (!new)
return -ENOMEM;
- if (__memcg_init_list_lru_node(new, old_size, new_size)) {
+ if (memcg_init_list_lru_range(new, old_size, new_size)) {
kvfree(new);
return -ENOMEM;
}
- memcpy(&new->lru, &old->lru, old_size * sizeof(void *));
+ memcpy(&new->lrus, &old->lrus, old_size * sizeof(new->lrus[0]));
- rcu_assign_pointer(nlru->memcg_lrus, new);
+ rcu_assign_pointer(lru->memcg_lrus, new);
kvfree_rcu(old, rcu);
return 0;
}
-static void memcg_cancel_update_list_lru_node(struct list_lru_node *nlru,
- int old_size, int new_size)
-{
- struct list_lru_memcg *memcg_lrus;
-
- memcg_lrus = rcu_dereference_protected(nlru->memcg_lrus,
- lockdep_is_held(&list_lrus_mutex));
- /* do not bother shrinking the array back to the old size, because we
- * cannot handle allocation failures here */
- __memcg_destroy_list_lru_node(memcg_lrus, old_size, new_size);
-}
-
-static int memcg_init_list_lru(struct list_lru *lru, bool memcg_aware)
-{
- int i;
-
- lru->memcg_aware = memcg_aware;
-
- if (!memcg_aware)
- return 0;
-
- for_each_node(i) {
- if (memcg_init_list_lru_node(&lru->node[i]))
- goto fail;
- }
- return 0;
-fail:
- for (i = i - 1; i >= 0; i--) {
- if (!lru->node[i].memcg_lrus)
- continue;
- memcg_destroy_list_lru_node(&lru->node[i]);
- }
- return -ENOMEM;
-}
-
-static void memcg_destroy_list_lru(struct list_lru *lru)
-{
- int i;
-
- if (!list_lru_memcg_aware(lru))
- return;
-
- for_each_node(i)
- memcg_destroy_list_lru_node(&lru->node[i]);
-}
-
-static int memcg_update_list_lru(struct list_lru *lru,
- int old_size, int new_size)
-{
- int i;
-
- for_each_node(i) {
- if (memcg_update_list_lru_node(&lru->node[i],
- old_size, new_size))
- goto fail;
- }
- return 0;
-fail:
- for (i = i - 1; i >= 0; i--) {
- if (!lru->node[i].memcg_lrus)
- continue;
-
- memcg_cancel_update_list_lru_node(&lru->node[i],
- old_size, new_size);
- }
- return -ENOMEM;
-}
-
static void memcg_cancel_update_list_lru(struct list_lru *lru,
int old_size, int new_size)
{
- int i;
+ struct list_lru_memcg *memcg_lrus;
- for_each_node(i)
- memcg_cancel_update_list_lru_node(&lru->node[i],
- old_size, new_size);
+ memcg_lrus = rcu_dereference_protected(lru->memcg_lrus,
+ lockdep_is_held(&list_lrus_mutex));
+ /*
+ * Do not bother shrinking the array back to the old size, because we
+ * cannot handle allocation failures here.
+ */
+ memcg_destroy_list_lru_range(memcg_lrus, old_size, new_size);
}
int memcg_update_all_list_lrus(int new_size)
@@ -527,8 +474,8 @@ static void memcg_drain_list_lru_node(struct list_lru *lru, int nid,
*/
spin_lock_irq(&nlru->lock);
- src = list_lru_from_memcg_idx(nlru, src_idx);
- dst = list_lru_from_memcg_idx(nlru, dst_idx);
+ src = list_lru_from_memcg_idx(lru, nid, src_idx);
+ dst = list_lru_from_memcg_idx(lru, nid, dst_idx);
list_splice_init(&src->list, &dst->list);
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v2 08/21] mm: list_lru: remove memcg_aware field from struct list_lru
2021-05-27 6:21 [PATCH v2 00/21] Optimize list lru memory consumption Muchun Song
` (6 preceding siblings ...)
2021-05-27 6:21 ` [PATCH v2 07/21] mm: list_lru: optimize the array of per memcg lists memory consumption Muchun Song
@ 2021-05-27 6:21 ` Muchun Song
2021-05-27 6:21 ` [PATCH v2 09/21] mm: introduce kmem_cache_alloc_lru Muchun Song
` (13 subsequent siblings)
21 siblings, 0 replies; 29+ messages in thread
From: Muchun Song @ 2021-05-27 6:21 UTC (permalink / raw)
To: willy, akpm, hannes, mhocko, vdavydov.dev, shakeelb, guro,
shy828301, alexs, richard.weiyang, david, trond.myklebust,
anna.schumaker
Cc: linux-fsdevel, linux-kernel, linux-mm, linux-nfs, zhengqi.arch,
duanxiongchun, fam.zheng, Muchun Song
We can use ->memcg_lrus to indicate if the list_lru is memcg aware.
So ->memcg_aware is redundant and just remove it.
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
---
include/linux/list_lru.h | 1 -
mm/list_lru.c | 9 +++++----
2 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/include/linux/list_lru.h b/include/linux/list_lru.h
index 20a43904001d..4a9e4aaecc69 100644
--- a/include/linux/list_lru.h
+++ b/include/linux/list_lru.h
@@ -55,7 +55,6 @@ struct list_lru {
#ifdef CONFIG_MEMCG_KMEM
struct list_head list;
int shrinker_id;
- bool memcg_aware;
/* for cgroup aware lrus points to per cgroup lists, otherwise NULL */
struct list_lru_memcg __rcu *memcg_lrus;
#endif
diff --git a/mm/list_lru.c b/mm/list_lru.c
index e52f5a91fa0f..8006c0fcc506 100644
--- a/mm/list_lru.c
+++ b/mm/list_lru.c
@@ -20,7 +20,7 @@ static DEFINE_MUTEX(list_lrus_mutex);
static inline bool list_lru_memcg_aware(struct list_lru *lru)
{
- return lru->memcg_aware;
+ return !!rcu_access_pointer(lru->memcg_lrus);
}
static void list_lru_register(struct list_lru *lru)
@@ -73,7 +73,7 @@ list_lru_from_kmem(struct list_lru *lru, int nid, void *ptr,
struct list_lru_one *l = &nlru->lru;
struct mem_cgroup *memcg = NULL;
- if (!lru->memcg_lrus)
+ if (!list_lru_memcg_aware(lru))
goto out;
memcg = mem_cgroup_from_obj(ptr);
@@ -367,9 +367,10 @@ static int memcg_init_list_lru(struct list_lru *lru, bool memcg_aware)
struct list_lru_memcg *memcg_lrus;
int size = memcg_nr_cache_ids;
- lru->memcg_aware = memcg_aware;
- if (!memcg_aware)
+ if (!memcg_aware) {
+ lru->memcg_lrus = NULL;
return 0;
+ }
memcg_lrus = kvmalloc(sizeof(*memcg_lrus) +
size * sizeof(memcg_lrus->lrus[0]), GFP_KERNEL);
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v2 09/21] mm: introduce kmem_cache_alloc_lru
2021-05-27 6:21 [PATCH v2 00/21] Optimize list lru memory consumption Muchun Song
` (7 preceding siblings ...)
2021-05-27 6:21 ` [PATCH v2 08/21] mm: list_lru: remove memcg_aware field from struct list_lru Muchun Song
@ 2021-05-27 6:21 ` Muchun Song
2021-05-27 6:21 ` [PATCH v2 10/21] fs: introduce alloc_inode_sb() to allocate filesystems specific inode Muchun Song
` (12 subsequent siblings)
21 siblings, 0 replies; 29+ messages in thread
From: Muchun Song @ 2021-05-27 6:21 UTC (permalink / raw)
To: willy, akpm, hannes, mhocko, vdavydov.dev, shakeelb, guro,
shy828301, alexs, richard.weiyang, david, trond.myklebust,
anna.schumaker
Cc: linux-fsdevel, linux-kernel, linux-mm, linux-nfs, zhengqi.arch,
duanxiongchun, fam.zheng, Muchun Song
We currently allocate scope for every memcg to be able to tracked on
every superblock instantiated in the system, regardless of whether
that superblock is even accessible to that memcg.
These huge memcg counts come from container hosts where memcgs are
confined to just a small subset of the total number of superblocks
that instantiated at any given point in time.
For these systems with huge container counts, list_lru does not need
the capability of tracking every memcg on every superblock. What it
comes down to is that adding the memcg to the list_lru at the first
insert. So introduce kmem_cache_alloc_lru to allocate objects and its
list_lru. In the later patch, we will convert all inode and dentry
allocation from kmem_cache_alloc to kmem_cache_alloc_lru.
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
---
include/linux/list_lru.h | 3 ++
include/linux/memcontrol.h | 14 ++++++
include/linux/slab.h | 4 ++
mm/list_lru.c | 115 +++++++++++++++++++++++++++++++++++++++++----
mm/memcontrol.c | 14 ------
mm/slab.c | 39 ++++++++++-----
mm/slab.h | 17 ++++++-
mm/slub.c | 42 +++++++++++------
8 files changed, 197 insertions(+), 51 deletions(-)
diff --git a/include/linux/list_lru.h b/include/linux/list_lru.h
index 4a9e4aaecc69..2083f4f2701f 100644
--- a/include/linux/list_lru.h
+++ b/include/linux/list_lru.h
@@ -55,11 +55,14 @@ struct list_lru {
#ifdef CONFIG_MEMCG_KMEM
struct list_head list;
int shrinker_id;
+ /* protects ->memcg_lrus->lrus[i] */
+ spinlock_t lock;
/* for cgroup aware lrus points to per cgroup lists, otherwise NULL */
struct list_lru_memcg __rcu *memcg_lrus;
#endif
};
+int list_lru_memcg_alloc(struct list_lru *lru, struct mem_cgroup *memcg, gfp_t gfp);
void list_lru_destroy(struct list_lru *lru);
int __list_lru_init(struct list_lru *lru, bool memcg_aware,
struct lock_class_key *key, struct shrinker *shrinker);
diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index 6d0638e13fc1..84a70c219e85 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -517,6 +517,20 @@ static inline struct mem_cgroup *page_memcg_check(struct page *page)
return (struct mem_cgroup *)(memcg_data & ~MEMCG_DATA_FLAGS_MASK);
}
+static inline struct mem_cgroup *get_mem_cgroup_from_objcg(struct obj_cgroup *objcg)
+{
+ struct mem_cgroup *memcg;
+
+ rcu_read_lock();
+retry:
+ memcg = obj_cgroup_memcg(objcg);
+ if (unlikely(!css_tryget(&memcg->css)))
+ goto retry;
+ rcu_read_unlock();
+
+ return memcg;
+}
+
#ifdef CONFIG_MEMCG_KMEM
/*
* PageMemcgKmem - check if the page has MemcgKmem flag set
diff --git a/include/linux/slab.h b/include/linux/slab.h
index 9d316aac0aba..346698208258 100644
--- a/include/linux/slab.h
+++ b/include/linux/slab.h
@@ -424,8 +424,12 @@ static __always_inline unsigned int __kmalloc_index(size_t size,
#define kmalloc_index(s) __kmalloc_index(s, true)
#endif /* !CONFIG_SLOB */
+struct list_lru;
+
void *__kmalloc(size_t size, gfp_t flags) __assume_kmalloc_alignment __malloc;
void *kmem_cache_alloc(struct kmem_cache *, gfp_t flags) __assume_slab_alignment __malloc;
+void *kmem_cache_alloc_lru(struct kmem_cache *s, struct list_lru *lru,
+ gfp_t gfpflags) __assume_slab_alignment __malloc;
void kmem_cache_free(struct kmem_cache *, void *);
/*
diff --git a/mm/list_lru.c b/mm/list_lru.c
index 8006c0fcc506..4ba1db6d4409 100644
--- a/mm/list_lru.c
+++ b/mm/list_lru.c
@@ -339,22 +339,30 @@ static void memcg_destroy_list_lru_range(struct list_lru_memcg *memcg_lrus,
kfree(memcg_lrus->lrus[i]);
}
+static struct list_lru_per_memcg *memcg_list_lru_alloc(gfp_t gfp)
+{
+ int nid;
+ struct list_lru_per_memcg *lru;
+
+ lru = kmalloc(struct_size(lru, nodes, nr_node_ids), gfp);
+ if (!lru)
+ return NULL;
+
+ for_each_node(nid)
+ init_one_lru(&lru->nodes[nid]);
+
+ return lru;
+}
+
static int memcg_init_list_lru_range(struct list_lru_memcg *memcg_lrus,
int begin, int end)
{
int i;
for (i = begin; i < end; i++) {
- int nid;
- struct list_lru_per_memcg *lru;
-
- lru = kmalloc(struct_size(lru, nodes, nr_node_ids), GFP_KERNEL);
- if (!lru)
+ memcg_lrus->lrus[i] = memcg_list_lru_alloc(GFP_KERNEL);
+ if (!memcg_lrus->lrus[i])
goto fail;
-
- for_each_node(nid)
- init_one_lru(&lru->nodes[nid]);
- memcg_lrus->lrus[i] = lru;
}
return 0;
fail:
@@ -372,6 +380,8 @@ static int memcg_init_list_lru(struct list_lru *lru, bool memcg_aware)
return 0;
}
+ spin_lock_init(&lru->lock);
+
memcg_lrus = kvmalloc(sizeof(*memcg_lrus) +
size * sizeof(memcg_lrus->lrus[0]), GFP_KERNEL);
if (!memcg_lrus)
@@ -419,9 +429,11 @@ static int memcg_update_list_lru(struct list_lru *lru, int old_size, int new_siz
return -ENOMEM;
}
+ spin_lock_irq(&lru->lock);
memcpy(&new->lrus, &old->lrus, old_size * sizeof(new->lrus[0]));
-
rcu_assign_pointer(lru->memcg_lrus, new);
+ spin_unlock_irq(&lru->lock);
+
kvfree_rcu(old, rcu);
return 0;
@@ -507,6 +519,89 @@ void memcg_drain_all_list_lrus(int src_idx, struct mem_cgroup *dst_memcg)
memcg_drain_list_lru(lru, src_idx, dst_memcg);
mutex_unlock(&list_lrus_mutex);
}
+
+static bool memcg_list_lru_skip_alloc(struct list_lru *lru,
+ struct mem_cgroup *memcg)
+{
+ struct list_lru_memcg *memcg_lrus;
+ int idx = memcg_cache_id(memcg);
+
+ if (unlikely(idx < 0))
+ return true;
+
+ rcu_read_lock();
+ memcg_lrus = rcu_dereference(lru->memcg_lrus);
+ if (memcg_lrus->lrus[idx]) {
+ rcu_read_unlock();
+ return true;
+ }
+ rcu_read_unlock();
+
+ return false;
+}
+
+int list_lru_memcg_alloc(struct list_lru *lru, struct mem_cgroup *memcg, gfp_t gfp)
+{
+ unsigned long flags;
+ struct list_lru_memcg *memcg_lrus;
+ int i;
+
+ struct list_lru_memcg {
+ struct list_lru_per_memcg *mlru;
+ struct mem_cgroup *memcg;
+ } *table;
+
+ if (!list_lru_memcg_aware(lru))
+ return 0;
+
+ if (memcg_list_lru_skip_alloc(lru, memcg))
+ return 0;
+
+ /*
+ * The allocated list_lru_per_memcg array is not accounted directly.
+ * Moreover, it should not come from DMA buffer and is not readily
+ * reclaimable. So those GFP bits should be masked off.
+ */
+ gfp &= ~(__GFP_DMA | __GFP_RECLAIMABLE | __GFP_ACCOUNT | __GFP_ZERO);
+ table = kmalloc_array(memcg->css.cgroup->level, sizeof(*table), gfp);
+ if (!table)
+ return -ENOMEM;
+
+ /*
+ * Because the list_lru can be reparented to the parent cgroup's
+ * list_lru, we should make sure that this cgroup and all its
+ * ancestors have allocated list_lru_per_memcg.
+ */
+ for (i = 0; memcg; memcg = parent_mem_cgroup(memcg), i++) {
+ if (memcg_list_lru_skip_alloc(lru, memcg))
+ break;
+
+ table[i].memcg = memcg;
+ table[i].mlru = memcg_list_lru_alloc(gfp);
+ if (!table[i].mlru) {
+ while (i--)
+ kfree(table[i].mlru);
+ kfree(table);
+ return -ENOMEM;
+ }
+ }
+
+ spin_lock_irqsave(&lru->lock, flags);
+ memcg_lrus = rcu_dereference_protected(lru->memcg_lrus, true);
+ while (i--) {
+ int index = memcg_cache_id(table[i].memcg);
+
+ if (memcg_lrus->lrus[index])
+ kfree(table[i].mlru);
+ else
+ memcg_lrus->lrus[index] = table[i].mlru;
+ }
+ spin_unlock_irqrestore(&lru->lock, flags);
+
+ kfree(table);
+
+ return 0;
+}
#else
static int memcg_init_list_lru(struct list_lru *lru, bool memcg_aware)
{
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 377ec9847179..09bafa82781f 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -2787,20 +2787,6 @@ static void commit_charge(struct page *page, struct mem_cgroup *memcg)
page->memcg_data = (unsigned long)memcg;
}
-static struct mem_cgroup *get_mem_cgroup_from_objcg(struct obj_cgroup *objcg)
-{
- struct mem_cgroup *memcg;
-
- rcu_read_lock();
-retry:
- memcg = obj_cgroup_memcg(objcg);
- if (unlikely(!css_tryget(&memcg->css)))
- goto retry;
- rcu_read_unlock();
-
- return memcg;
-}
-
#ifdef CONFIG_MEMCG_KMEM
/*
* The allocated objcg pointers array is not accounted directly.
diff --git a/mm/slab.c b/mm/slab.c
index d0f725637663..9a001aabc77b 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -3219,7 +3219,7 @@ slab_alloc_node(struct kmem_cache *cachep, gfp_t flags, int nodeid, size_t orig_
bool init = false;
flags &= gfp_allowed_mask;
- cachep = slab_pre_alloc_hook(cachep, &objcg, 1, flags);
+ cachep = slab_pre_alloc_hook(cachep, NULL, &objcg, 1, flags);
if (unlikely(!cachep))
return NULL;
@@ -3295,7 +3295,8 @@ __do_cache_alloc(struct kmem_cache *cachep, gfp_t flags)
#endif /* CONFIG_NUMA */
static __always_inline void *
-slab_alloc(struct kmem_cache *cachep, gfp_t flags, size_t orig_size, unsigned long caller)
+slab_alloc(struct kmem_cache *cachep, struct list_lru *lru, gfp_t flags,
+ size_t orig_size, unsigned long caller)
{
unsigned long save_flags;
void *objp;
@@ -3303,7 +3304,7 @@ slab_alloc(struct kmem_cache *cachep, gfp_t flags, size_t orig_size, unsigned lo
bool init = false;
flags &= gfp_allowed_mask;
- cachep = slab_pre_alloc_hook(cachep, &objcg, 1, flags);
+ cachep = slab_pre_alloc_hook(cachep, lru, &objcg, 1, flags);
if (unlikely(!cachep))
return NULL;
@@ -3492,6 +3493,18 @@ void ___cache_free(struct kmem_cache *cachep, void *objp,
__free_one(ac, objp);
}
+static __always_inline
+void *__kmem_cache_alloc_lru(struct kmem_cache *cachep, struct list_lru *lru,
+ gfp_t flags)
+{
+ void *ret = slab_alloc(cachep, lru, flags, cachep->object_size, _RET_IP_);
+
+ trace_kmem_cache_alloc(_RET_IP_, ret,
+ cachep->object_size, cachep->size, flags);
+
+ return ret;
+}
+
/**
* kmem_cache_alloc - Allocate an object
* @cachep: The cache to allocate from.
@@ -3504,15 +3517,17 @@ void ___cache_free(struct kmem_cache *cachep, void *objp,
*/
void *kmem_cache_alloc(struct kmem_cache *cachep, gfp_t flags)
{
- void *ret = slab_alloc(cachep, flags, cachep->object_size, _RET_IP_);
-
- trace_kmem_cache_alloc(_RET_IP_, ret,
- cachep->object_size, cachep->size, flags);
-
- return ret;
+ return __kmem_cache_alloc_lru(cachep, NULL, flags);
}
EXPORT_SYMBOL(kmem_cache_alloc);
+void *kmem_cache_alloc_lru(struct kmem_cache *cachep, struct list_lru *lru,
+ gfp_t flags)
+{
+ return __kmem_cache_alloc_lru(cachep, lru, flags);
+}
+EXPORT_SYMBOL(kmem_cache_alloc_lru);
+
static __always_inline void
cache_alloc_debugcheck_after_bulk(struct kmem_cache *s, gfp_t flags,
size_t size, void **p, unsigned long caller)
@@ -3529,7 +3544,7 @@ int kmem_cache_alloc_bulk(struct kmem_cache *s, gfp_t flags, size_t size,
size_t i;
struct obj_cgroup *objcg = NULL;
- s = slab_pre_alloc_hook(s, &objcg, size, flags);
+ s = slab_pre_alloc_hook(s, NULL, &objcg, size, flags);
if (!s)
return 0;
@@ -3570,7 +3585,7 @@ kmem_cache_alloc_trace(struct kmem_cache *cachep, gfp_t flags, size_t size)
{
void *ret;
- ret = slab_alloc(cachep, flags, size, _RET_IP_);
+ ret = slab_alloc(cachep, NULL, flags, size, _RET_IP_);
ret = kasan_kmalloc(cachep, ret, size, flags);
trace_kmalloc(_RET_IP_, ret,
@@ -3697,7 +3712,7 @@ static __always_inline void *__do_kmalloc(size_t size, gfp_t flags,
cachep = kmalloc_slab(size, flags);
if (unlikely(ZERO_OR_NULL_PTR(cachep)))
return cachep;
- ret = slab_alloc(cachep, flags, size, caller);
+ ret = slab_alloc(cachep, NULL, flags, size, caller);
ret = kasan_kmalloc(cachep, ret, size, flags);
trace_kmalloc(caller, ret,
diff --git a/mm/slab.h b/mm/slab.h
index b4cdf8687120..473b75777a4d 100644
--- a/mm/slab.h
+++ b/mm/slab.h
@@ -46,6 +46,7 @@ struct kmem_cache {
#include <linux/kmemleak.h>
#include <linux/random.h>
#include <linux/sched/mm.h>
+#include <linux/list_lru.h>
/*
* State of the slab allocator.
@@ -262,6 +263,7 @@ static inline size_t obj_full_size(struct kmem_cache *s)
* Returns false if the allocation should fail.
*/
static inline bool memcg_slab_pre_alloc_hook(struct kmem_cache *s,
+ struct list_lru *lru,
struct obj_cgroup **objcgp,
size_t objects, gfp_t flags)
{
@@ -277,6 +279,17 @@ static inline bool memcg_slab_pre_alloc_hook(struct kmem_cache *s,
if (!objcg)
return true;
+ if (lru) {
+ struct mem_cgroup *memcg = get_mem_cgroup_from_objcg(objcg);
+
+ if (list_lru_memcg_alloc(lru, memcg, flags)) {
+ css_put(&memcg->css);
+ obj_cgroup_put(objcg);
+ return false;
+ }
+ css_put(&memcg->css);
+ }
+
if (obj_cgroup_charge(objcg, flags, objects * obj_full_size(s))) {
obj_cgroup_put(objcg);
return false;
@@ -379,6 +392,7 @@ static inline void memcg_free_page_obj_cgroups(struct page *page)
}
static inline bool memcg_slab_pre_alloc_hook(struct kmem_cache *s,
+ struct list_lru *lru,
struct obj_cgroup **objcgp,
size_t objects, gfp_t flags)
{
@@ -477,6 +491,7 @@ static inline size_t slab_ksize(const struct kmem_cache *s)
}
static inline struct kmem_cache *slab_pre_alloc_hook(struct kmem_cache *s,
+ struct list_lru *lru,
struct obj_cgroup **objcgp,
size_t size, gfp_t flags)
{
@@ -487,7 +502,7 @@ static inline struct kmem_cache *slab_pre_alloc_hook(struct kmem_cache *s,
if (should_failslab(s, flags))
return NULL;
- if (!memcg_slab_pre_alloc_hook(s, objcgp, size, flags))
+ if (!memcg_slab_pre_alloc_hook(s, lru, objcgp, size, flags))
return NULL;
return s;
diff --git a/mm/slub.c b/mm/slub.c
index 6b896b8c36f0..d49c139d6dce 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -2862,7 +2862,7 @@ static __always_inline void maybe_wipe_obj_freeptr(struct kmem_cache *s,
*
* Otherwise we can simply pick the next object from the lockless free list.
*/
-static __always_inline void *slab_alloc_node(struct kmem_cache *s,
+static __always_inline void *slab_alloc_node(struct kmem_cache *s, struct list_lru *lru,
gfp_t gfpflags, int node, unsigned long addr, size_t orig_size)
{
void *object;
@@ -2872,7 +2872,7 @@ static __always_inline void *slab_alloc_node(struct kmem_cache *s,
struct obj_cgroup *objcg = NULL;
bool init = false;
- s = slab_pre_alloc_hook(s, &objcg, 1, gfpflags);
+ s = slab_pre_alloc_hook(s, lru, &objcg, 1, gfpflags);
if (!s)
return NULL;
@@ -2956,27 +2956,41 @@ static __always_inline void *slab_alloc_node(struct kmem_cache *s,
return object;
}
-static __always_inline void *slab_alloc(struct kmem_cache *s,
+static __always_inline void *slab_alloc(struct kmem_cache *s, struct list_lru *lru,
gfp_t gfpflags, unsigned long addr, size_t orig_size)
{
- return slab_alloc_node(s, gfpflags, NUMA_NO_NODE, addr, orig_size);
+ return slab_alloc_node(s, lru, gfpflags, NUMA_NO_NODE, addr, orig_size);
}
-void *kmem_cache_alloc(struct kmem_cache *s, gfp_t gfpflags)
+static __always_inline
+void *__kmem_cache_alloc_lru(struct kmem_cache *s, struct list_lru *lru,
+ gfp_t gfpflags)
{
- void *ret = slab_alloc(s, gfpflags, _RET_IP_, s->object_size);
+ void *ret = slab_alloc(s, lru, gfpflags, _RET_IP_, s->object_size);
trace_kmem_cache_alloc(_RET_IP_, ret, s->object_size,
s->size, gfpflags);
return ret;
}
+
+void *kmem_cache_alloc(struct kmem_cache *s, gfp_t gfpflags)
+{
+ return __kmem_cache_alloc_lru(s, NULL, gfpflags);
+}
EXPORT_SYMBOL(kmem_cache_alloc);
+void *kmem_cache_alloc_lru(struct kmem_cache *s, struct list_lru *lru,
+ gfp_t gfpflags)
+{
+ return __kmem_cache_alloc_lru(s, lru, gfpflags);
+}
+EXPORT_SYMBOL(kmem_cache_alloc_lru);
+
#ifdef CONFIG_TRACING
void *kmem_cache_alloc_trace(struct kmem_cache *s, gfp_t gfpflags, size_t size)
{
- void *ret = slab_alloc(s, gfpflags, _RET_IP_, size);
+ void *ret = slab_alloc(s, NULL, gfpflags, _RET_IP_, size);
trace_kmalloc(_RET_IP_, ret, size, s->size, gfpflags);
ret = kasan_kmalloc(s, ret, size, gfpflags);
return ret;
@@ -2987,7 +3001,7 @@ EXPORT_SYMBOL(kmem_cache_alloc_trace);
#ifdef CONFIG_NUMA
void *kmem_cache_alloc_node(struct kmem_cache *s, gfp_t gfpflags, int node)
{
- void *ret = slab_alloc_node(s, gfpflags, node, _RET_IP_, s->object_size);
+ void *ret = slab_alloc_node(s, NULL, gfpflags, node, _RET_IP_, s->object_size);
trace_kmem_cache_alloc_node(_RET_IP_, ret,
s->object_size, s->size, gfpflags, node);
@@ -3001,7 +3015,7 @@ void *kmem_cache_alloc_node_trace(struct kmem_cache *s,
gfp_t gfpflags,
int node, size_t size)
{
- void *ret = slab_alloc_node(s, gfpflags, node, _RET_IP_, size);
+ void *ret = slab_alloc_node(s, NULL, gfpflags, node, _RET_IP_, size);
trace_kmalloc_node(_RET_IP_, ret,
size, s->size, gfpflags, node);
@@ -3352,7 +3366,7 @@ int kmem_cache_alloc_bulk(struct kmem_cache *s, gfp_t flags, size_t size,
struct obj_cgroup *objcg = NULL;
/* memcg and kmem_cache debug support */
- s = slab_pre_alloc_hook(s, &objcg, size, flags);
+ s = slab_pre_alloc_hook(s, NULL, &objcg, size, flags);
if (unlikely(!s))
return false;
/*
@@ -4109,7 +4123,7 @@ void *__kmalloc(size_t size, gfp_t flags)
if (unlikely(ZERO_OR_NULL_PTR(s)))
return s;
- ret = slab_alloc(s, flags, _RET_IP_, size);
+ ret = slab_alloc(s, NULL, flags, _RET_IP_, size);
trace_kmalloc(_RET_IP_, ret, size, s->size, flags);
@@ -4157,7 +4171,7 @@ void *__kmalloc_node(size_t size, gfp_t flags, int node)
if (unlikely(ZERO_OR_NULL_PTR(s)))
return s;
- ret = slab_alloc_node(s, flags, node, _RET_IP_, size);
+ ret = slab_alloc_node(s, NULL, flags, node, _RET_IP_, size);
trace_kmalloc_node(_RET_IP_, ret, size, s->size, flags, node);
@@ -4619,7 +4633,7 @@ void *__kmalloc_track_caller(size_t size, gfp_t gfpflags, unsigned long caller)
if (unlikely(ZERO_OR_NULL_PTR(s)))
return s;
- ret = slab_alloc(s, gfpflags, caller, size);
+ ret = slab_alloc(s, NULL, gfpflags, caller, size);
/* Honor the call site pointer we received. */
trace_kmalloc(caller, ret, size, s->size, gfpflags);
@@ -4650,7 +4664,7 @@ void *__kmalloc_node_track_caller(size_t size, gfp_t gfpflags,
if (unlikely(ZERO_OR_NULL_PTR(s)))
return s;
- ret = slab_alloc_node(s, gfpflags, node, caller, size);
+ ret = slab_alloc_node(s, NULL, gfpflags, node, caller, size);
/* Honor the call site pointer we received. */
trace_kmalloc_node(caller, ret, size, s->size, gfpflags, node);
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v2 10/21] fs: introduce alloc_inode_sb() to allocate filesystems specific inode
2021-05-27 6:21 [PATCH v2 00/21] Optimize list lru memory consumption Muchun Song
` (8 preceding siblings ...)
2021-05-27 6:21 ` [PATCH v2 09/21] mm: introduce kmem_cache_alloc_lru Muchun Song
@ 2021-05-27 6:21 ` Muchun Song
2021-05-27 6:21 ` [PATCH v2 11/21] mm: dcache: use kmem_cache_alloc_lru() to allocate dentry Muchun Song
` (11 subsequent siblings)
21 siblings, 0 replies; 29+ messages in thread
From: Muchun Song @ 2021-05-27 6:21 UTC (permalink / raw)
To: willy, akpm, hannes, mhocko, vdavydov.dev, shakeelb, guro,
shy828301, alexs, richard.weiyang, david, trond.myklebust,
anna.schumaker
Cc: linux-fsdevel, linux-kernel, linux-mm, linux-nfs, zhengqi.arch,
duanxiongchun, fam.zheng, Muchun Song
The allocated inode cache will be added to its memcg list_lru, we should
allocate its list_lru_one structure in advance. Introduce alloc_inode_sb
which is used for allocating filesystems specific inodes to set up the
inode reclaim context correctly.
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
---
Documentation/filesystems/porting.rst | 5 +++++
drivers/dax/super.c | 2 +-
fs/9p/vfs_inode.c | 2 +-
fs/adfs/super.c | 2 +-
fs/affs/super.c | 2 +-
fs/afs/super.c | 2 +-
fs/befs/linuxvfs.c | 2 +-
fs/bfs/inode.c | 2 +-
fs/block_dev.c | 2 +-
fs/btrfs/inode.c | 2 +-
fs/ceph/inode.c | 2 +-
fs/cifs/cifsfs.c | 2 +-
fs/coda/inode.c | 2 +-
fs/ecryptfs/super.c | 2 +-
fs/efs/super.c | 2 +-
fs/erofs/super.c | 2 +-
fs/exfat/super.c | 2 +-
fs/ext2/super.c | 2 +-
fs/ext4/super.c | 2 +-
fs/f2fs/super.c | 2 +-
fs/fat/inode.c | 2 +-
fs/freevxfs/vxfs_super.c | 2 +-
fs/fuse/inode.c | 2 +-
fs/gfs2/super.c | 2 +-
fs/hfs/super.c | 2 +-
fs/hfsplus/super.c | 2 +-
fs/hostfs/hostfs_kern.c | 2 +-
fs/hpfs/super.c | 2 +-
fs/hugetlbfs/inode.c | 2 +-
fs/inode.c | 2 +-
fs/isofs/inode.c | 2 +-
fs/jffs2/super.c | 2 +-
fs/jfs/super.c | 2 +-
fs/minix/inode.c | 2 +-
fs/nfs/inode.c | 2 +-
fs/nilfs2/super.c | 2 +-
fs/ntfs/inode.c | 2 +-
fs/ocfs2/dlmfs/dlmfs.c | 2 +-
fs/ocfs2/super.c | 2 +-
fs/openpromfs/inode.c | 2 +-
fs/orangefs/super.c | 2 +-
fs/overlayfs/super.c | 2 +-
fs/proc/inode.c | 2 +-
fs/qnx4/inode.c | 2 +-
fs/qnx6/inode.c | 2 +-
fs/reiserfs/super.c | 2 +-
fs/romfs/super.c | 2 +-
fs/squashfs/super.c | 2 +-
fs/sysv/inode.c | 2 +-
fs/ubifs/super.c | 2 +-
fs/udf/super.c | 2 +-
fs/ufs/super.c | 2 +-
fs/vboxsf/super.c | 2 +-
fs/xfs/xfs_icache.c | 3 ++-
fs/zonefs/super.c | 2 +-
include/linux/fs.h | 11 +++++++++++
ipc/mqueue.c | 2 +-
mm/shmem.c | 2 +-
net/socket.c | 2 +-
net/sunrpc/rpc_pipe.c | 2 +-
60 files changed, 75 insertions(+), 58 deletions(-)
diff --git a/Documentation/filesystems/porting.rst b/Documentation/filesystems/porting.rst
index 0302035781be..7db1c462cfcc 100644
--- a/Documentation/filesystems/porting.rst
+++ b/Documentation/filesystems/porting.rst
@@ -45,6 +45,11 @@ typically between calling iget_locked() and unlocking the inode.
At some point that will become mandatory.
+**mandatory**
+
+The foo_inode_info should always be allocated through alloc_inode_sb() rather
+than kmem_cache_alloc().
+
---
**mandatory**
diff --git a/drivers/dax/super.c b/drivers/dax/super.c
index 5fa6ae9dbc8b..30474ba337c8 100644
--- a/drivers/dax/super.c
+++ b/drivers/dax/super.c
@@ -460,7 +460,7 @@ static struct inode *dax_alloc_inode(struct super_block *sb)
struct dax_device *dax_dev;
struct inode *inode;
- dax_dev = kmem_cache_alloc(dax_cache, GFP_KERNEL);
+ dax_dev = alloc_inode_sb(sb, dax_cache, GFP_KERNEL);
if (!dax_dev)
return NULL;
diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c
index 795706520b5e..5311f35accf5 100644
--- a/fs/9p/vfs_inode.c
+++ b/fs/9p/vfs_inode.c
@@ -223,7 +223,7 @@ v9fs_blank_wstat(struct p9_wstat *wstat)
struct inode *v9fs_alloc_inode(struct super_block *sb)
{
struct v9fs_inode *v9inode;
- v9inode = kmem_cache_alloc(v9fs_inode_cache, GFP_KERNEL);
+ v9inode = alloc_inode_sb(sb, v9fs_inode_cache, GFP_KERNEL);
if (!v9inode)
return NULL;
#ifdef CONFIG_9P_FSCACHE
diff --git a/fs/adfs/super.c b/fs/adfs/super.c
index bdbd26e571ed..e8bfc38239cd 100644
--- a/fs/adfs/super.c
+++ b/fs/adfs/super.c
@@ -220,7 +220,7 @@ static struct kmem_cache *adfs_inode_cachep;
static struct inode *adfs_alloc_inode(struct super_block *sb)
{
struct adfs_inode_info *ei;
- ei = kmem_cache_alloc(adfs_inode_cachep, GFP_KERNEL);
+ ei = alloc_inode_sb(sb, adfs_inode_cachep, GFP_KERNEL);
if (!ei)
return NULL;
return &ei->vfs_inode;
diff --git a/fs/affs/super.c b/fs/affs/super.c
index c6c2a513ec92..67458f63a1ab 100644
--- a/fs/affs/super.c
+++ b/fs/affs/super.c
@@ -100,7 +100,7 @@ static struct inode *affs_alloc_inode(struct super_block *sb)
{
struct affs_inode_info *i;
- i = kmem_cache_alloc(affs_inode_cachep, GFP_KERNEL);
+ i = alloc_inode_sb(sb, affs_inode_cachep, GFP_KERNEL);
if (!i)
return NULL;
diff --git a/fs/afs/super.c b/fs/afs/super.c
index e38bb1e7a4d2..0ecea5a94af9 100644
--- a/fs/afs/super.c
+++ b/fs/afs/super.c
@@ -677,7 +677,7 @@ static struct inode *afs_alloc_inode(struct super_block *sb)
{
struct afs_vnode *vnode;
- vnode = kmem_cache_alloc(afs_inode_cachep, GFP_KERNEL);
+ vnode = alloc_inode_sb(sb, afs_inode_cachep, GFP_KERNEL);
if (!vnode)
return NULL;
diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c
index c1ba13d19024..b4b3567ac655 100644
--- a/fs/befs/linuxvfs.c
+++ b/fs/befs/linuxvfs.c
@@ -277,7 +277,7 @@ befs_alloc_inode(struct super_block *sb)
{
struct befs_inode_info *bi;
- bi = kmem_cache_alloc(befs_inode_cachep, GFP_KERNEL);
+ bi = alloc_inode_sb(sb, befs_inode_cachep, GFP_KERNEL);
if (!bi)
return NULL;
return &bi->vfs_inode;
diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c
index fd691e4815c5..1926bec2c850 100644
--- a/fs/bfs/inode.c
+++ b/fs/bfs/inode.c
@@ -239,7 +239,7 @@ static struct kmem_cache *bfs_inode_cachep;
static struct inode *bfs_alloc_inode(struct super_block *sb)
{
struct bfs_inode_info *bi;
- bi = kmem_cache_alloc(bfs_inode_cachep, GFP_KERNEL);
+ bi = alloc_inode_sb(sb, bfs_inode_cachep, GFP_KERNEL);
if (!bi)
return NULL;
return &bi->vfs_inode;
diff --git a/fs/block_dev.c b/fs/block_dev.c
index b8abccd03e5d..14a0ef54ae50 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -796,7 +796,7 @@ static struct kmem_cache * bdev_cachep __read_mostly;
static struct inode *bdev_alloc_inode(struct super_block *sb)
{
- struct bdev_inode *ei = kmem_cache_alloc(bdev_cachep, GFP_KERNEL);
+ struct bdev_inode *ei = alloc_inode_sb(sb, bdev_cachep, GFP_KERNEL);
if (!ei)
return NULL;
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index b44c28455d4f..75decdbdbd45 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -8793,7 +8793,7 @@ struct inode *btrfs_alloc_inode(struct super_block *sb)
struct btrfs_inode *ei;
struct inode *inode;
- ei = kmem_cache_alloc(btrfs_inode_cachep, GFP_KERNEL);
+ ei = alloc_inode_sb(sb, btrfs_inode_cachep, GFP_KERNEL);
if (!ei)
return NULL;
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index e1c63adb196d..9fe9db45a6ab 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -447,7 +447,7 @@ struct inode *ceph_alloc_inode(struct super_block *sb)
struct ceph_inode_info *ci;
int i;
- ci = kmem_cache_alloc(ceph_inode_cachep, GFP_NOFS);
+ ci = alloc_inode_sb(sb, ceph_inode_cachep, GFP_NOFS);
if (!ci)
return NULL;
diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index 2ffcb29d5c8f..d68cb0ddb85a 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -363,7 +363,7 @@ static struct inode *
cifs_alloc_inode(struct super_block *sb)
{
struct cifsInodeInfo *cifs_inode;
- cifs_inode = kmem_cache_alloc(cifs_inode_cachep, GFP_KERNEL);
+ cifs_inode = alloc_inode_sb(sb, cifs_inode_cachep, GFP_KERNEL);
if (!cifs_inode)
return NULL;
cifs_inode->cifsAttrs = 0x20; /* default */
diff --git a/fs/coda/inode.c b/fs/coda/inode.c
index d9f1bd7153df..2185328b65c7 100644
--- a/fs/coda/inode.c
+++ b/fs/coda/inode.c
@@ -43,7 +43,7 @@ static struct kmem_cache * coda_inode_cachep;
static struct inode *coda_alloc_inode(struct super_block *sb)
{
struct coda_inode_info *ei;
- ei = kmem_cache_alloc(coda_inode_cachep, GFP_KERNEL);
+ ei = alloc_inode_sb(sb, coda_inode_cachep, GFP_KERNEL);
if (!ei)
return NULL;
memset(&ei->c_fid, 0, sizeof(struct CodaFid));
diff --git a/fs/ecryptfs/super.c b/fs/ecryptfs/super.c
index 39116af0390f..0b1c878317ab 100644
--- a/fs/ecryptfs/super.c
+++ b/fs/ecryptfs/super.c
@@ -38,7 +38,7 @@ static struct inode *ecryptfs_alloc_inode(struct super_block *sb)
struct ecryptfs_inode_info *inode_info;
struct inode *inode = NULL;
- inode_info = kmem_cache_alloc(ecryptfs_inode_info_cache, GFP_KERNEL);
+ inode_info = alloc_inode_sb(sb, ecryptfs_inode_info_cache, GFP_KERNEL);
if (unlikely(!inode_info))
goto out;
if (ecryptfs_init_crypt_stat(&inode_info->crypt_stat)) {
diff --git a/fs/efs/super.c b/fs/efs/super.c
index 62b155b9366b..b287f47c165b 100644
--- a/fs/efs/super.c
+++ b/fs/efs/super.c
@@ -69,7 +69,7 @@ static struct kmem_cache * efs_inode_cachep;
static struct inode *efs_alloc_inode(struct super_block *sb)
{
struct efs_inode_info *ei;
- ei = kmem_cache_alloc(efs_inode_cachep, GFP_KERNEL);
+ ei = alloc_inode_sb(sb, efs_inode_cachep, GFP_KERNEL);
if (!ei)
return NULL;
return &ei->vfs_inode;
diff --git a/fs/erofs/super.c b/fs/erofs/super.c
index bbf3bbd908e0..03f2dd684f8d 100644
--- a/fs/erofs/super.c
+++ b/fs/erofs/super.c
@@ -83,7 +83,7 @@ static void erofs_inode_init_once(void *ptr)
static struct inode *erofs_alloc_inode(struct super_block *sb)
{
struct erofs_inode *vi =
- kmem_cache_alloc(erofs_inode_cachep, GFP_KERNEL);
+ alloc_inode_sb(sb, erofs_inode_cachep, GFP_KERNEL);
if (!vi)
return NULL;
diff --git a/fs/exfat/super.c b/fs/exfat/super.c
index d38d17a77e76..c88e0f218df0 100644
--- a/fs/exfat/super.c
+++ b/fs/exfat/super.c
@@ -182,7 +182,7 @@ static struct inode *exfat_alloc_inode(struct super_block *sb)
{
struct exfat_inode_info *ei;
- ei = kmem_cache_alloc(exfat_inode_cachep, GFP_NOFS);
+ ei = alloc_inode_sb(sb, exfat_inode_cachep, GFP_NOFS);
if (!ei)
return NULL;
diff --git a/fs/ext2/super.c b/fs/ext2/super.c
index 21e09fbaa46f..7eba9d6fbdc6 100644
--- a/fs/ext2/super.c
+++ b/fs/ext2/super.c
@@ -180,7 +180,7 @@ static struct kmem_cache * ext2_inode_cachep;
static struct inode *ext2_alloc_inode(struct super_block *sb)
{
struct ext2_inode_info *ei;
- ei = kmem_cache_alloc(ext2_inode_cachep, GFP_KERNEL);
+ ei = alloc_inode_sb(sb, ext2_inode_cachep, GFP_KERNEL);
if (!ei)
return NULL;
ei->i_block_alloc_info = NULL;
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 7dc94f3e18e6..b2c486d224b4 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -1275,7 +1275,7 @@ static struct inode *ext4_alloc_inode(struct super_block *sb)
{
struct ext4_inode_info *ei;
- ei = kmem_cache_alloc(ext4_inode_cachep, GFP_NOFS);
+ ei = alloc_inode_sb(sb, ext4_inode_cachep, GFP_NOFS);
if (!ei)
return NULL;
diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
index 7d325bfaf65a..444134950ae4 100644
--- a/fs/f2fs/super.c
+++ b/fs/f2fs/super.c
@@ -1169,7 +1169,7 @@ static struct inode *f2fs_alloc_inode(struct super_block *sb)
{
struct f2fs_inode_info *fi;
- fi = kmem_cache_alloc(f2fs_inode_cachep, GFP_F2FS_ZERO);
+ fi = alloc_inode_sb(sb, f2fs_inode_cachep, GFP_F2FS_ZERO);
if (!fi)
return NULL;
diff --git a/fs/fat/inode.c b/fs/fat/inode.c
index bab9b202b496..adc8a525553a 100644
--- a/fs/fat/inode.c
+++ b/fs/fat/inode.c
@@ -744,7 +744,7 @@ static struct kmem_cache *fat_inode_cachep;
static struct inode *fat_alloc_inode(struct super_block *sb)
{
struct msdos_inode_info *ei;
- ei = kmem_cache_alloc(fat_inode_cachep, GFP_NOFS);
+ ei = alloc_inode_sb(sb, fat_inode_cachep, GFP_NOFS);
if (!ei)
return NULL;
diff --git a/fs/freevxfs/vxfs_super.c b/fs/freevxfs/vxfs_super.c
index 578a5062706e..22eed5a73ac2 100644
--- a/fs/freevxfs/vxfs_super.c
+++ b/fs/freevxfs/vxfs_super.c
@@ -124,7 +124,7 @@ static struct inode *vxfs_alloc_inode(struct super_block *sb)
{
struct vxfs_inode_info *vi;
- vi = kmem_cache_alloc(vxfs_inode_cachep, GFP_KERNEL);
+ vi = alloc_inode_sb(sb, vxfs_inode_cachep, GFP_KERNEL);
if (!vi)
return NULL;
inode_init_once(&vi->vfs_inode);
diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c
index 393e36b74dc4..3fe7d1484ae6 100644
--- a/fs/fuse/inode.c
+++ b/fs/fuse/inode.c
@@ -73,7 +73,7 @@ static struct inode *fuse_alloc_inode(struct super_block *sb)
{
struct fuse_inode *fi;
- fi = kmem_cache_alloc(fuse_inode_cachep, GFP_KERNEL);
+ fi = alloc_inode_sb(sb, fuse_inode_cachep, GFP_KERNEL);
if (!fi)
return NULL;
diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c
index 4d4ceb0b6903..e9e8ad6f0c66 100644
--- a/fs/gfs2/super.c
+++ b/fs/gfs2/super.c
@@ -1496,7 +1496,7 @@ static struct inode *gfs2_alloc_inode(struct super_block *sb)
{
struct gfs2_inode *ip;
- ip = kmem_cache_alloc(gfs2_inode_cachep, GFP_KERNEL);
+ ip = alloc_inode_sb(sb, gfs2_inode_cachep, GFP_KERNEL);
if (!ip)
return NULL;
ip->i_flags = 0;
diff --git a/fs/hfs/super.c b/fs/hfs/super.c
index 44d07c9e3a7f..e40334c93183 100644
--- a/fs/hfs/super.c
+++ b/fs/hfs/super.c
@@ -162,7 +162,7 @@ static struct inode *hfs_alloc_inode(struct super_block *sb)
{
struct hfs_inode_info *i;
- i = kmem_cache_alloc(hfs_inode_cachep, GFP_KERNEL);
+ i = alloc_inode_sb(sb, hfs_inode_cachep, GFP_KERNEL);
return i ? &i->vfs_inode : NULL;
}
diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c
index b9e3db3f855f..8479add998b5 100644
--- a/fs/hfsplus/super.c
+++ b/fs/hfsplus/super.c
@@ -624,7 +624,7 @@ static struct inode *hfsplus_alloc_inode(struct super_block *sb)
{
struct hfsplus_inode_info *i;
- i = kmem_cache_alloc(hfsplus_inode_cachep, GFP_KERNEL);
+ i = alloc_inode_sb(sb, hfsplus_inode_cachep, GFP_KERNEL);
return i ? &i->vfs_inode : NULL;
}
diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c
index 7d0c3dbb2898..96dc97ae48e4 100644
--- a/fs/hostfs/hostfs_kern.c
+++ b/fs/hostfs/hostfs_kern.c
@@ -222,7 +222,7 @@ static struct inode *hostfs_alloc_inode(struct super_block *sb)
{
struct hostfs_inode_info *hi;
- hi = kmem_cache_alloc(hostfs_inode_cache, GFP_KERNEL_ACCOUNT);
+ hi = alloc_inode_sb(sb, hostfs_inode_cache, GFP_KERNEL_ACCOUNT);
if (hi == NULL)
return NULL;
hi->fd = -1;
diff --git a/fs/hpfs/super.c b/fs/hpfs/super.c
index a7dbfc892022..1cb89595b875 100644
--- a/fs/hpfs/super.c
+++ b/fs/hpfs/super.c
@@ -232,7 +232,7 @@ static struct kmem_cache * hpfs_inode_cachep;
static struct inode *hpfs_alloc_inode(struct super_block *sb)
{
struct hpfs_inode_info *ei;
- ei = kmem_cache_alloc(hpfs_inode_cachep, GFP_NOFS);
+ ei = alloc_inode_sb(sb, hpfs_inode_cachep, GFP_NOFS);
if (!ei)
return NULL;
return &ei->vfs_inode;
diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
index 9d9e0097c1d3..7a3095f786ec 100644
--- a/fs/hugetlbfs/inode.c
+++ b/fs/hugetlbfs/inode.c
@@ -1108,7 +1108,7 @@ static struct inode *hugetlbfs_alloc_inode(struct super_block *sb)
if (unlikely(!hugetlbfs_dec_free_inodes(sbinfo)))
return NULL;
- p = kmem_cache_alloc(hugetlbfs_inode_cachep, GFP_KERNEL);
+ p = alloc_inode_sb(sb, hugetlbfs_inode_cachep, GFP_KERNEL);
if (unlikely(!p)) {
hugetlbfs_inc_free_inodes(sbinfo);
return NULL;
diff --git a/fs/inode.c b/fs/inode.c
index c93500d84264..60ab51416a16 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -232,7 +232,7 @@ static struct inode *alloc_inode(struct super_block *sb)
if (ops->alloc_inode)
inode = ops->alloc_inode(sb);
else
- inode = kmem_cache_alloc(inode_cachep, GFP_KERNEL);
+ inode = alloc_inode_sb(sb, inode_cachep, GFP_KERNEL);
if (!inode)
return NULL;
diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c
index 21edc423b79f..ef6c29501702 100644
--- a/fs/isofs/inode.c
+++ b/fs/isofs/inode.c
@@ -70,7 +70,7 @@ static struct kmem_cache *isofs_inode_cachep;
static struct inode *isofs_alloc_inode(struct super_block *sb)
{
struct iso_inode_info *ei;
- ei = kmem_cache_alloc(isofs_inode_cachep, GFP_KERNEL);
+ ei = alloc_inode_sb(sb, isofs_inode_cachep, GFP_KERNEL);
if (!ei)
return NULL;
return &ei->vfs_inode;
diff --git a/fs/jffs2/super.c b/fs/jffs2/super.c
index 81ca58c10b72..7ea37f49f1e1 100644
--- a/fs/jffs2/super.c
+++ b/fs/jffs2/super.c
@@ -39,7 +39,7 @@ static struct inode *jffs2_alloc_inode(struct super_block *sb)
{
struct jffs2_inode_info *f;
- f = kmem_cache_alloc(jffs2_inode_cachep, GFP_KERNEL);
+ f = alloc_inode_sb(sb, jffs2_inode_cachep, GFP_KERNEL);
if (!f)
return NULL;
return &f->vfs_inode;
diff --git a/fs/jfs/super.c b/fs/jfs/super.c
index 1f0ffabbde56..65255541d44e 100644
--- a/fs/jfs/super.c
+++ b/fs/jfs/super.c
@@ -102,7 +102,7 @@ static struct inode *jfs_alloc_inode(struct super_block *sb)
{
struct jfs_inode_info *jfs_inode;
- jfs_inode = kmem_cache_alloc(jfs_inode_cachep, GFP_NOFS);
+ jfs_inode = alloc_inode_sb(sb, jfs_inode_cachep, GFP_NOFS);
if (!jfs_inode)
return NULL;
#ifdef CONFIG_QUOTA
diff --git a/fs/minix/inode.c b/fs/minix/inode.c
index a532a99bbe81..f30b1daac386 100644
--- a/fs/minix/inode.c
+++ b/fs/minix/inode.c
@@ -63,7 +63,7 @@ static struct kmem_cache * minix_inode_cachep;
static struct inode *minix_alloc_inode(struct super_block *sb)
{
struct minix_inode_info *ei;
- ei = kmem_cache_alloc(minix_inode_cachep, GFP_KERNEL);
+ ei = alloc_inode_sb(sb, minix_inode_cachep, GFP_KERNEL);
if (!ei)
return NULL;
return &ei->vfs_inode;
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 529c4099f482..fa3d8e8b695f 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -2240,7 +2240,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
struct inode *nfs_alloc_inode(struct super_block *sb)
{
struct nfs_inode *nfsi;
- nfsi = kmem_cache_alloc(nfs_inode_cachep, GFP_KERNEL);
+ nfsi = alloc_inode_sb(sb, nfs_inode_cachep, GFP_KERNEL);
if (!nfsi)
return NULL;
nfsi->flags = 0UL;
diff --git a/fs/nilfs2/super.c b/fs/nilfs2/super.c
index 4abd928b0bc8..e7d7ad7a1e50 100644
--- a/fs/nilfs2/super.c
+++ b/fs/nilfs2/super.c
@@ -151,7 +151,7 @@ struct inode *nilfs_alloc_inode(struct super_block *sb)
{
struct nilfs_inode_info *ii;
- ii = kmem_cache_alloc(nilfs_inode_cachep, GFP_NOFS);
+ ii = alloc_inode_sb(sb, nilfs_inode_cachep, GFP_NOFS);
if (!ii)
return NULL;
ii->i_bh = NULL;
diff --git a/fs/ntfs/inode.c b/fs/ntfs/inode.c
index f5c058b3192c..1e053bb58ead 100644
--- a/fs/ntfs/inode.c
+++ b/fs/ntfs/inode.c
@@ -310,7 +310,7 @@ struct inode *ntfs_alloc_big_inode(struct super_block *sb)
ntfs_inode *ni;
ntfs_debug("Entering.");
- ni = kmem_cache_alloc(ntfs_big_inode_cache, GFP_NOFS);
+ ni = alloc_inode_sb(sb, ntfs_big_inode_cache, GFP_NOFS);
if (likely(ni != NULL)) {
ni->state = 0;
return VFS_I(ni);
diff --git a/fs/ocfs2/dlmfs/dlmfs.c b/fs/ocfs2/dlmfs/dlmfs.c
index fa0a14f199eb..e360543ad7e7 100644
--- a/fs/ocfs2/dlmfs/dlmfs.c
+++ b/fs/ocfs2/dlmfs/dlmfs.c
@@ -280,7 +280,7 @@ static struct inode *dlmfs_alloc_inode(struct super_block *sb)
{
struct dlmfs_inode_private *ip;
- ip = kmem_cache_alloc(dlmfs_inode_cache, GFP_NOFS);
+ ip = alloc_inode_sb(sb, dlmfs_inode_cache, GFP_NOFS);
if (!ip)
return NULL;
diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c
index c86bd4e60e20..cf044448130f 100644
--- a/fs/ocfs2/super.c
+++ b/fs/ocfs2/super.c
@@ -549,7 +549,7 @@ static struct inode *ocfs2_alloc_inode(struct super_block *sb)
{
struct ocfs2_inode_info *oi;
- oi = kmem_cache_alloc(ocfs2_inode_cachep, GFP_NOFS);
+ oi = alloc_inode_sb(sb, ocfs2_inode_cachep, GFP_NOFS);
if (!oi)
return NULL;
diff --git a/fs/openpromfs/inode.c b/fs/openpromfs/inode.c
index f825176ff4ed..f0b7f4d51a17 100644
--- a/fs/openpromfs/inode.c
+++ b/fs/openpromfs/inode.c
@@ -335,7 +335,7 @@ static struct inode *openprom_alloc_inode(struct super_block *sb)
{
struct op_inode_info *oi;
- oi = kmem_cache_alloc(op_inode_cachep, GFP_KERNEL);
+ oi = alloc_inode_sb(sb, op_inode_cachep, GFP_KERNEL);
if (!oi)
return NULL;
diff --git a/fs/orangefs/super.c b/fs/orangefs/super.c
index ee5efdc35cc1..5731ce285c98 100644
--- a/fs/orangefs/super.c
+++ b/fs/orangefs/super.c
@@ -106,7 +106,7 @@ static struct inode *orangefs_alloc_inode(struct super_block *sb)
{
struct orangefs_inode_s *orangefs_inode;
- orangefs_inode = kmem_cache_alloc(orangefs_inode_cache, GFP_KERNEL);
+ orangefs_inode = alloc_inode_sb(sb, orangefs_inode_cache, GFP_KERNEL);
if (!orangefs_inode)
return NULL;
diff --git a/fs/overlayfs/super.c b/fs/overlayfs/super.c
index b01d4147520d..6d8d16bf4b57 100644
--- a/fs/overlayfs/super.c
+++ b/fs/overlayfs/super.c
@@ -174,7 +174,7 @@ static struct kmem_cache *ovl_inode_cachep;
static struct inode *ovl_alloc_inode(struct super_block *sb)
{
- struct ovl_inode *oi = kmem_cache_alloc(ovl_inode_cachep, GFP_KERNEL);
+ struct ovl_inode *oi = alloc_inode_sb(sb, ovl_inode_cachep, GFP_KERNEL);
if (!oi)
return NULL;
diff --git a/fs/proc/inode.c b/fs/proc/inode.c
index 599eb724ff2d..cc0a406d3a19 100644
--- a/fs/proc/inode.c
+++ b/fs/proc/inode.c
@@ -66,7 +66,7 @@ static struct inode *proc_alloc_inode(struct super_block *sb)
{
struct proc_inode *ei;
- ei = kmem_cache_alloc(proc_inode_cachep, GFP_KERNEL);
+ ei = alloc_inode_sb(sb, proc_inode_cachep, GFP_KERNEL);
if (!ei)
return NULL;
ei->pid = NULL;
diff --git a/fs/qnx4/inode.c b/fs/qnx4/inode.c
index 3fb7fc819b4f..a635bb6615e9 100644
--- a/fs/qnx4/inode.c
+++ b/fs/qnx4/inode.c
@@ -338,7 +338,7 @@ static struct kmem_cache *qnx4_inode_cachep;
static struct inode *qnx4_alloc_inode(struct super_block *sb)
{
struct qnx4_inode_info *ei;
- ei = kmem_cache_alloc(qnx4_inode_cachep, GFP_KERNEL);
+ ei = alloc_inode_sb(sb, qnx4_inode_cachep, GFP_KERNEL);
if (!ei)
return NULL;
return &ei->vfs_inode;
diff --git a/fs/qnx6/inode.c b/fs/qnx6/inode.c
index 61191f7bdf62..9d8e7e9788a1 100644
--- a/fs/qnx6/inode.c
+++ b/fs/qnx6/inode.c
@@ -597,7 +597,7 @@ static struct kmem_cache *qnx6_inode_cachep;
static struct inode *qnx6_alloc_inode(struct super_block *sb)
{
struct qnx6_inode_info *ei;
- ei = kmem_cache_alloc(qnx6_inode_cachep, GFP_KERNEL);
+ ei = alloc_inode_sb(sb, qnx6_inode_cachep, GFP_KERNEL);
if (!ei)
return NULL;
return &ei->vfs_inode;
diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c
index 3ffafc73acf0..d9e8b0368ce9 100644
--- a/fs/reiserfs/super.c
+++ b/fs/reiserfs/super.c
@@ -639,7 +639,7 @@ static struct kmem_cache *reiserfs_inode_cachep;
static struct inode *reiserfs_alloc_inode(struct super_block *sb)
{
struct reiserfs_inode_info *ei;
- ei = kmem_cache_alloc(reiserfs_inode_cachep, GFP_KERNEL);
+ ei = alloc_inode_sb(sb, reiserfs_inode_cachep, GFP_KERNEL);
if (!ei)
return NULL;
atomic_set(&ei->openers, 0);
diff --git a/fs/romfs/super.c b/fs/romfs/super.c
index 259f684d9236..9e6bbb4219de 100644
--- a/fs/romfs/super.c
+++ b/fs/romfs/super.c
@@ -375,7 +375,7 @@ static struct inode *romfs_alloc_inode(struct super_block *sb)
{
struct romfs_inode_info *inode;
- inode = kmem_cache_alloc(romfs_inode_cachep, GFP_KERNEL);
+ inode = alloc_inode_sb(sb, romfs_inode_cachep, GFP_KERNEL);
return inode ? &inode->vfs_inode : NULL;
}
diff --git a/fs/squashfs/super.c b/fs/squashfs/super.c
index 88cc94be1076..16b785cb78c4 100644
--- a/fs/squashfs/super.c
+++ b/fs/squashfs/super.c
@@ -466,7 +466,7 @@ static void __exit exit_squashfs_fs(void)
static struct inode *squashfs_alloc_inode(struct super_block *sb)
{
struct squashfs_inode_info *ei =
- kmem_cache_alloc(squashfs_inode_cachep, GFP_KERNEL);
+ alloc_inode_sb(sb, squashfs_inode_cachep, GFP_KERNEL);
return ei ? &ei->vfs_inode : NULL;
}
diff --git a/fs/sysv/inode.c b/fs/sysv/inode.c
index be47263b8605..9e8d4a6fb2f3 100644
--- a/fs/sysv/inode.c
+++ b/fs/sysv/inode.c
@@ -306,7 +306,7 @@ static struct inode *sysv_alloc_inode(struct super_block *sb)
{
struct sysv_inode_info *si;
- si = kmem_cache_alloc(sysv_inode_cachep, GFP_KERNEL);
+ si = alloc_inode_sb(sb, sysv_inode_cachep, GFP_KERNEL);
if (!si)
return NULL;
return &si->vfs_inode;
diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c
index 7b572e1414ba..6851040c598e 100644
--- a/fs/ubifs/super.c
+++ b/fs/ubifs/super.c
@@ -268,7 +268,7 @@ static struct inode *ubifs_alloc_inode(struct super_block *sb)
{
struct ubifs_inode *ui;
- ui = kmem_cache_alloc(ubifs_inode_slab, GFP_NOFS);
+ ui = alloc_inode_sb(sb, ubifs_inode_slab, GFP_NOFS);
if (!ui)
return NULL;
diff --git a/fs/udf/super.c b/fs/udf/super.c
index 2f83c1204e20..bdeaa5d88a9e 100644
--- a/fs/udf/super.c
+++ b/fs/udf/super.c
@@ -141,7 +141,7 @@ static struct kmem_cache *udf_inode_cachep;
static struct inode *udf_alloc_inode(struct super_block *sb)
{
struct udf_inode_info *ei;
- ei = kmem_cache_alloc(udf_inode_cachep, GFP_KERNEL);
+ ei = alloc_inode_sb(sb, udf_inode_cachep, GFP_KERNEL);
if (!ei)
return NULL;
diff --git a/fs/ufs/super.c b/fs/ufs/super.c
index 74028b5a7b0a..ba83ad6e9d39 100644
--- a/fs/ufs/super.c
+++ b/fs/ufs/super.c
@@ -1443,7 +1443,7 @@ static struct inode *ufs_alloc_inode(struct super_block *sb)
{
struct ufs_inode_info *ei;
- ei = kmem_cache_alloc(ufs_inode_cachep, GFP_NOFS);
+ ei = alloc_inode_sb(sb, ufs_inode_cachep, GFP_NOFS);
if (!ei)
return NULL;
diff --git a/fs/vboxsf/super.c b/fs/vboxsf/super.c
index 4f5e59f06284..050ef855158b 100644
--- a/fs/vboxsf/super.c
+++ b/fs/vboxsf/super.c
@@ -244,7 +244,7 @@ static struct inode *vboxsf_alloc_inode(struct super_block *sb)
{
struct vboxsf_inode *sf_i;
- sf_i = kmem_cache_alloc(vboxsf_inode_cachep, GFP_NOFS);
+ sf_i = alloc_inode_sb(sb, vboxsf_inode_cachep, GFP_NOFS);
if (!sf_i)
return NULL;
diff --git a/fs/xfs/xfs_icache.c b/fs/xfs/xfs_icache.c
index 3c81daca0e9a..8c8b059e1e26 100644
--- a/fs/xfs/xfs_icache.c
+++ b/fs/xfs/xfs_icache.c
@@ -40,7 +40,8 @@ xfs_inode_alloc(
* XXX: If this didn't occur in transactions, we could drop GFP_NOFAIL
* and return NULL here on ENOMEM.
*/
- ip = kmem_cache_alloc(xfs_inode_zone, GFP_KERNEL | __GFP_NOFAIL);
+ ip = alloc_inode_sb(mp->m_super, xfs_inode_zone,
+ GFP_KERNEL | __GFP_NOFAIL);
if (inode_init_always(mp->m_super, VFS_I(ip))) {
kmem_cache_free(xfs_inode_zone, ip);
diff --git a/fs/zonefs/super.c b/fs/zonefs/super.c
index cd145d318b17..335b58251b52 100644
--- a/fs/zonefs/super.c
+++ b/fs/zonefs/super.c
@@ -1152,7 +1152,7 @@ static struct inode *zonefs_alloc_inode(struct super_block *sb)
{
struct zonefs_inode_info *zi;
- zi = kmem_cache_alloc(zonefs_inode_cachep, GFP_KERNEL);
+ zi = alloc_inode_sb(sb, zonefs_inode_cachep, GFP_KERNEL);
if (!zi)
return NULL;
diff --git a/include/linux/fs.h b/include/linux/fs.h
index c3c88fdb9b2a..cb016debb46e 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -41,6 +41,7 @@
#include <linux/stddef.h>
#include <linux/mount.h>
#include <linux/cred.h>
+#include <linux/slab.h>
#include <asm/byteorder.h>
#include <uapi/linux/fs.h>
@@ -3200,6 +3201,16 @@ extern void free_inode_nonrcu(struct inode *inode);
extern int should_remove_suid(struct dentry *);
extern int file_remove_privs(struct file *);
+/*
+ * This must be used for allocating filesystems specific inodes to set
+ * up the inode reclaim context correctly.
+ */
+static inline void *
+alloc_inode_sb(struct super_block *sb, struct kmem_cache *cache, gfp_t gfp)
+{
+ return kmem_cache_alloc_lru(cache, &sb->s_inode_lru, gfp);
+}
+
extern void __insert_inode_hash(struct inode *, unsigned long hashval);
static inline void insert_inode_hash(struct inode *inode)
{
diff --git a/ipc/mqueue.c b/ipc/mqueue.c
index 4e4e61111500..310178dc5480 100644
--- a/ipc/mqueue.c
+++ b/ipc/mqueue.c
@@ -483,7 +483,7 @@ static struct inode *mqueue_alloc_inode(struct super_block *sb)
{
struct mqueue_inode_info *ei;
- ei = kmem_cache_alloc(mqueue_inode_cachep, GFP_KERNEL);
+ ei = alloc_inode_sb(sb, mqueue_inode_cachep, GFP_KERNEL);
if (!ei)
return NULL;
return &ei->vfs_inode;
diff --git a/mm/shmem.c b/mm/shmem.c
index ccce139d4e5c..67c8db781727 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -3736,7 +3736,7 @@ static struct kmem_cache *shmem_inode_cachep;
static struct inode *shmem_alloc_inode(struct super_block *sb)
{
struct shmem_inode_info *info;
- info = kmem_cache_alloc(shmem_inode_cachep, GFP_KERNEL);
+ info = alloc_inode_sb(sb, shmem_inode_cachep, GFP_KERNEL);
if (!info)
return NULL;
return &info->vfs_inode;
diff --git a/net/socket.c b/net/socket.c
index 27e3e7d53f8e..d59844b02a2d 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -250,7 +250,7 @@ static struct inode *sock_alloc_inode(struct super_block *sb)
{
struct socket_alloc *ei;
- ei = kmem_cache_alloc(sock_inode_cachep, GFP_KERNEL);
+ ei = alloc_inode_sb(sb, sock_inode_cachep, GFP_KERNEL);
if (!ei)
return NULL;
init_waitqueue_head(&ei->socket.wq.wait);
diff --git a/net/sunrpc/rpc_pipe.c b/net/sunrpc/rpc_pipe.c
index 09c000d490a1..30707f39694e 100644
--- a/net/sunrpc/rpc_pipe.c
+++ b/net/sunrpc/rpc_pipe.c
@@ -197,7 +197,7 @@ static struct inode *
rpc_alloc_inode(struct super_block *sb)
{
struct rpc_inode *rpci;
- rpci = kmem_cache_alloc(rpc_inode_cachep, GFP_KERNEL);
+ rpci = alloc_inode_sb(sb, rpc_inode_cachep, GFP_KERNEL);
if (!rpci)
return NULL;
return &rpci->vfs_inode;
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v2 11/21] mm: dcache: use kmem_cache_alloc_lru() to allocate dentry
2021-05-27 6:21 [PATCH v2 00/21] Optimize list lru memory consumption Muchun Song
` (9 preceding siblings ...)
2021-05-27 6:21 ` [PATCH v2 10/21] fs: introduce alloc_inode_sb() to allocate filesystems specific inode Muchun Song
@ 2021-05-27 6:21 ` Muchun Song
2021-05-27 6:21 ` [PATCH v2 12/21] xarray: use kmem_cache_alloc_lru to allocate xa_node Muchun Song
` (10 subsequent siblings)
21 siblings, 0 replies; 29+ messages in thread
From: Muchun Song @ 2021-05-27 6:21 UTC (permalink / raw)
To: willy, akpm, hannes, mhocko, vdavydov.dev, shakeelb, guro,
shy828301, alexs, richard.weiyang, david, trond.myklebust,
anna.schumaker
Cc: linux-fsdevel, linux-kernel, linux-mm, linux-nfs, zhengqi.arch,
duanxiongchun, fam.zheng, Muchun Song
Like inode cache, the dentry will also be added to its memcg list_lru.
So replace kmem_cache_alloc() with kmem_cache_alloc_lru() to allocate
dentry.
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
---
fs/dcache.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/fs/dcache.c b/fs/dcache.c
index cf871a81f4fd..36d4806d7284 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -1741,7 +1741,8 @@ static struct dentry *__d_alloc(struct super_block *sb, const struct qstr *name)
char *dname;
int err;
- dentry = kmem_cache_alloc(dentry_cache, GFP_KERNEL);
+ dentry = kmem_cache_alloc_lru(dentry_cache, &sb->s_dentry_lru,
+ GFP_KERNEL);
if (!dentry)
return NULL;
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v2 12/21] xarray: use kmem_cache_alloc_lru to allocate xa_node
2021-05-27 6:21 [PATCH v2 00/21] Optimize list lru memory consumption Muchun Song
` (10 preceding siblings ...)
2021-05-27 6:21 ` [PATCH v2 11/21] mm: dcache: use kmem_cache_alloc_lru() to allocate dentry Muchun Song
@ 2021-05-27 6:21 ` Muchun Song
2021-05-27 6:21 ` [PATCH v2 13/21] mm: workingset: use xas_set_lru() to pass shadow_nodes Muchun Song
` (9 subsequent siblings)
21 siblings, 0 replies; 29+ messages in thread
From: Muchun Song @ 2021-05-27 6:21 UTC (permalink / raw)
To: willy, akpm, hannes, mhocko, vdavydov.dev, shakeelb, guro,
shy828301, alexs, richard.weiyang, david, trond.myklebust,
anna.schumaker
Cc: linux-fsdevel, linux-kernel, linux-mm, linux-nfs, zhengqi.arch,
duanxiongchun, fam.zheng, Muchun Song
The workingset will add the xa_node to the shadow_nodes list. So the
allocation of xa_node should be done by kmem_cache_alloc_lru(). The
workingset can use xas_set_lru() to pass the list_lru.
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
---
include/linux/xarray.h | 9 ++++++++-
lib/xarray.c | 10 +++++-----
2 files changed, 13 insertions(+), 6 deletions(-)
diff --git a/include/linux/xarray.h b/include/linux/xarray.h
index a91e3d90df8a..31f3e5ef3c7b 100644
--- a/include/linux/xarray.h
+++ b/include/linux/xarray.h
@@ -1317,6 +1317,7 @@ struct xa_state {
struct xa_node *xa_node;
struct xa_node *xa_alloc;
xa_update_node_t xa_update;
+ struct list_lru *xa_lru;
};
/*
@@ -1336,7 +1337,8 @@ struct xa_state {
.xa_pad = 0, \
.xa_node = XAS_RESTART, \
.xa_alloc = NULL, \
- .xa_update = NULL \
+ .xa_update = NULL, \
+ .xa_lru = NULL, \
}
/**
@@ -1613,6 +1615,11 @@ static inline void xas_set_update(struct xa_state *xas, xa_update_node_t update)
xas->xa_update = update;
}
+static inline void xas_set_lru(struct xa_state *xas, struct list_lru *lru)
+{
+ xas->xa_lru = lru;
+}
+
/**
* xas_next_entry() - Advance iterator to next present entry.
* @xas: XArray operation state.
diff --git a/lib/xarray.c b/lib/xarray.c
index f5d8f54907b4..e9b818abc823 100644
--- a/lib/xarray.c
+++ b/lib/xarray.c
@@ -302,7 +302,7 @@ bool xas_nomem(struct xa_state *xas, gfp_t gfp)
}
if (xas->xa->xa_flags & XA_FLAGS_ACCOUNT)
gfp |= __GFP_ACCOUNT;
- xas->xa_alloc = kmem_cache_alloc(radix_tree_node_cachep, gfp);
+ xas->xa_alloc = kmem_cache_alloc_lru(radix_tree_node_cachep, xas->xa_lru, gfp);
if (!xas->xa_alloc)
return false;
xas->xa_alloc->parent = NULL;
@@ -334,10 +334,10 @@ static bool __xas_nomem(struct xa_state *xas, gfp_t gfp)
gfp |= __GFP_ACCOUNT;
if (gfpflags_allow_blocking(gfp)) {
xas_unlock_type(xas, lock_type);
- xas->xa_alloc = kmem_cache_alloc(radix_tree_node_cachep, gfp);
+ xas->xa_alloc = kmem_cache_alloc_lru(radix_tree_node_cachep, xas->xa_lru, gfp);
xas_lock_type(xas, lock_type);
} else {
- xas->xa_alloc = kmem_cache_alloc(radix_tree_node_cachep, gfp);
+ xas->xa_alloc = kmem_cache_alloc_lru(radix_tree_node_cachep, xas->xa_lru, gfp);
}
if (!xas->xa_alloc)
return false;
@@ -371,7 +371,7 @@ static void *xas_alloc(struct xa_state *xas, unsigned int shift)
if (xas->xa->xa_flags & XA_FLAGS_ACCOUNT)
gfp |= __GFP_ACCOUNT;
- node = kmem_cache_alloc(radix_tree_node_cachep, gfp);
+ node = kmem_cache_alloc_lru(radix_tree_node_cachep, xas->xa_lru, gfp);
if (!node) {
xas_set_err(xas, -ENOMEM);
return NULL;
@@ -1014,7 +1014,7 @@ void xas_split_alloc(struct xa_state *xas, void *entry, unsigned int order,
void *sibling = NULL;
struct xa_node *node;
- node = kmem_cache_alloc(radix_tree_node_cachep, gfp);
+ node = kmem_cache_alloc_lru(radix_tree_node_cachep, xas->xa_lru, gfp);
if (!node)
goto nomem;
node->array = xas->xa;
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v2 13/21] mm: workingset: use xas_set_lru() to pass shadow_nodes
2021-05-27 6:21 [PATCH v2 00/21] Optimize list lru memory consumption Muchun Song
` (11 preceding siblings ...)
2021-05-27 6:21 ` [PATCH v2 12/21] xarray: use kmem_cache_alloc_lru to allocate xa_node Muchun Song
@ 2021-05-27 6:21 ` Muchun Song
2021-05-27 6:21 ` [PATCH v2 14/21] nfs42: use a specific kmem_cache to allocate nfs4_xattr_entry Muchun Song
` (8 subsequent siblings)
21 siblings, 0 replies; 29+ messages in thread
From: Muchun Song @ 2021-05-27 6:21 UTC (permalink / raw)
To: willy, akpm, hannes, mhocko, vdavydov.dev, shakeelb, guro,
shy828301, alexs, richard.weiyang, david, trond.myklebust,
anna.schumaker
Cc: linux-fsdevel, linux-kernel, linux-mm, linux-nfs, zhengqi.arch,
duanxiongchun, fam.zheng, Muchun Song
The workingset will add the xa_node to shadow_nodes, so we should use
xas_set_lru() to pass the list_lru which we want to insert xa_node
into to set up the xa_node reclaim context correctly.
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
---
include/linux/swap.h | 5 ++++-
mm/workingset.c | 2 +-
2 files changed, 5 insertions(+), 2 deletions(-)
diff --git a/include/linux/swap.h b/include/linux/swap.h
index 46d51d058d05..a2e1363b1509 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -323,9 +323,12 @@ void workingset_activation(struct page *page);
/* Only track the nodes of mappings with shadow entries */
void workingset_update_node(struct xa_node *node);
+extern struct list_lru shadow_nodes;
#define mapping_set_update(xas, mapping) do { \
- if (!dax_mapping(mapping) && !shmem_mapping(mapping)) \
+ if (!dax_mapping(mapping) && !shmem_mapping(mapping)) { \
xas_set_update(xas, workingset_update_node); \
+ xas_set_lru(xas, &shadow_nodes); \
+ } \
} while (0)
/* linux/mm/page_alloc.c */
diff --git a/mm/workingset.c b/mm/workingset.c
index 4f7a306ce75a..e50827fc3994 100644
--- a/mm/workingset.c
+++ b/mm/workingset.c
@@ -426,7 +426,7 @@ void workingset_activation(struct page *page)
* point where they would still be useful.
*/
-static struct list_lru shadow_nodes;
+struct list_lru shadow_nodes;
void workingset_update_node(struct xa_node *node)
{
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v2 14/21] nfs42: use a specific kmem_cache to allocate nfs4_xattr_entry
2021-05-27 6:21 [PATCH v2 00/21] Optimize list lru memory consumption Muchun Song
` (12 preceding siblings ...)
2021-05-27 6:21 ` [PATCH v2 13/21] mm: workingset: use xas_set_lru() to pass shadow_nodes Muchun Song
@ 2021-05-27 6:21 ` Muchun Song
2021-05-27 6:21 ` [PATCH v2 15/21] mm: list_lru: allocate list_lru_one only when needed Muchun Song
` (7 subsequent siblings)
21 siblings, 0 replies; 29+ messages in thread
From: Muchun Song @ 2021-05-27 6:21 UTC (permalink / raw)
To: willy, akpm, hannes, mhocko, vdavydov.dev, shakeelb, guro,
shy828301, alexs, richard.weiyang, david, trond.myklebust,
anna.schumaker
Cc: linux-fsdevel, linux-kernel, linux-mm, linux-nfs, zhengqi.arch,
duanxiongchun, fam.zheng, Muchun Song
If we want to add the allocated objects to its list_lru, we should use
kmem_cache_alloc_lru() to allocate objects. So intruduce
nfs4_xattr_entry_cachep which is used to allocate nfs4_xattr_entry.
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
---
fs/nfs/nfs42xattr.c | 95 ++++++++++++++++++++++++++---------------------------
1 file changed, 47 insertions(+), 48 deletions(-)
diff --git a/fs/nfs/nfs42xattr.c b/fs/nfs/nfs42xattr.c
index 1c4d2a05b401..5b7af9080db0 100644
--- a/fs/nfs/nfs42xattr.c
+++ b/fs/nfs/nfs42xattr.c
@@ -81,7 +81,7 @@ struct nfs4_xattr_entry {
struct hlist_node hnode;
struct list_head lru;
struct list_head dispose;
- char *xattr_name;
+ const char *xattr_name;
void *xattr_value;
size_t xattr_size;
struct nfs4_xattr_bucket *bucket;
@@ -98,6 +98,7 @@ static struct list_lru nfs4_xattr_entry_lru;
static struct list_lru nfs4_xattr_large_entry_lru;
static struct kmem_cache *nfs4_xattr_cache_cachep;
+static struct kmem_cache *nfs4_xattr_entry_cachep;
/*
* Hashing helper functions.
@@ -177,49 +178,27 @@ nfs4_xattr_alloc_entry(const char *name, const void *value,
{
struct nfs4_xattr_entry *entry;
void *valp;
- char *namep;
- size_t alloclen, slen;
- char *buf;
- uint32_t flags;
+ const char *namep;
+ uint32_t flags = len > PAGE_SIZE ? NFS4_XATTR_ENTRY_EXTVAL : 0;
+ gfp_t gfp = GFP_KERNEL_ACCOUNT | GFP_NOFS;
+ struct list_lru *lru;
BUILD_BUG_ON(sizeof(struct nfs4_xattr_entry) +
XATTR_NAME_MAX + 1 > PAGE_SIZE);
- alloclen = sizeof(struct nfs4_xattr_entry);
- if (name != NULL) {
- slen = strlen(name) + 1;
- alloclen += slen;
- } else
- slen = 0;
-
- if (alloclen + len <= PAGE_SIZE) {
- alloclen += len;
- flags = 0;
- } else {
- flags = NFS4_XATTR_ENTRY_EXTVAL;
- }
-
- buf = kmalloc(alloclen, GFP_KERNEL_ACCOUNT | GFP_NOFS);
- if (buf == NULL)
+ lru = flags & NFS4_XATTR_ENTRY_EXTVAL ? &nfs4_xattr_large_entry_lru :
+ &nfs4_xattr_entry_lru;
+ entry = kmem_cache_alloc_lru(nfs4_xattr_entry_cachep, lru, gfp);
+ if (!entry)
return NULL;
- entry = (struct nfs4_xattr_entry *)buf;
-
- if (name != NULL) {
- namep = buf + sizeof(struct nfs4_xattr_entry);
- memcpy(namep, name, slen);
- } else {
- namep = NULL;
- }
-
-
- if (flags & NFS4_XATTR_ENTRY_EXTVAL) {
- valp = kvmalloc(len, GFP_KERNEL_ACCOUNT | GFP_NOFS);
- if (valp == NULL) {
- kfree(buf);
- return NULL;
- }
- } else if (len != 0) {
- valp = buf + sizeof(struct nfs4_xattr_entry) + slen;
+ namep = kstrdup_const(name, gfp);
+ if (!namep && name)
+ goto free_buf;
+
+ if (len != 0) {
+ valp = kvmalloc(len, gfp);
+ if (!valp)
+ goto free_name;
} else
valp = NULL;
@@ -232,23 +211,23 @@ nfs4_xattr_alloc_entry(const char *name, const void *value,
entry->flags = flags;
entry->xattr_value = valp;
- kref_init(&entry->ref);
entry->xattr_name = namep;
entry->xattr_size = len;
- entry->bucket = NULL;
- INIT_LIST_HEAD(&entry->lru);
- INIT_LIST_HEAD(&entry->dispose);
- INIT_HLIST_NODE(&entry->hnode);
return entry;
+free_name:
+ kfree_const(namep);
+free_buf:
+ kmem_cache_free(nfs4_xattr_entry_cachep, entry);
+ return NULL;
}
static void
nfs4_xattr_free_entry(struct nfs4_xattr_entry *entry)
{
- if (entry->flags & NFS4_XATTR_ENTRY_EXTVAL)
- kvfree(entry->xattr_value);
- kfree(entry);
+ kvfree(entry->xattr_value);
+ kfree_const(entry->xattr_name);
+ kmem_cache_free(nfs4_xattr_entry_cachep, entry);
}
static void
@@ -289,7 +268,7 @@ nfs4_xattr_alloc_cache(void)
{
struct nfs4_xattr_cache *cache;
- cache = kmem_cache_alloc(nfs4_xattr_cache_cachep,
+ cache = kmem_cache_alloc_lru(nfs4_xattr_cache_cachep, &nfs4_xattr_cache_lru,
GFP_KERNEL_ACCOUNT | GFP_NOFS);
if (cache == NULL)
return NULL;
@@ -992,6 +971,17 @@ static void nfs4_xattr_cache_init_once(void *p)
INIT_LIST_HEAD(&cache->dispose);
}
+static void nfs4_xattr_entry_init_once(void *p)
+{
+ struct nfs4_xattr_entry *entry = p;
+
+ kref_init(&entry->ref);
+ entry->bucket = NULL;
+ INIT_LIST_HEAD(&entry->lru);
+ INIT_LIST_HEAD(&entry->dispose);
+ INIT_HLIST_NODE(&entry->hnode);
+}
+
int __init nfs4_xattr_cache_init(void)
{
int ret = 0;
@@ -1003,6 +993,13 @@ int __init nfs4_xattr_cache_init(void)
if (nfs4_xattr_cache_cachep == NULL)
return -ENOMEM;
+ nfs4_xattr_entry_cachep = kmem_cache_create("nfs4_xattr_entry",
+ sizeof(struct nfs4_xattr_entry), 0,
+ (SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD | SLAB_ACCOUNT),
+ nfs4_xattr_entry_init_once);
+ if (!nfs4_xattr_entry_cachep)
+ goto out5;
+
ret = list_lru_init_memcg(&nfs4_xattr_large_entry_lru,
&nfs4_xattr_large_entry_shrinker);
if (ret)
@@ -1040,6 +1037,8 @@ int __init nfs4_xattr_cache_init(void)
out3:
list_lru_destroy(&nfs4_xattr_large_entry_lru);
out4:
+ kmem_cache_destroy(nfs4_xattr_entry_cachep);
+out5:
kmem_cache_destroy(nfs4_xattr_cache_cachep);
return ret;
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v2 15/21] mm: list_lru: allocate list_lru_one only when needed
2021-05-27 6:21 [PATCH v2 00/21] Optimize list lru memory consumption Muchun Song
` (13 preceding siblings ...)
2021-05-27 6:21 ` [PATCH v2 14/21] nfs42: use a specific kmem_cache to allocate nfs4_xattr_entry Muchun Song
@ 2021-05-27 6:21 ` Muchun Song
2021-05-27 6:21 ` [PATCH v2 16/21] mm: list_lru: rename memcg_drain_all_list_lrus to memcg_reparent_list_lrus Muchun Song
` (6 subsequent siblings)
21 siblings, 0 replies; 29+ messages in thread
From: Muchun Song @ 2021-05-27 6:21 UTC (permalink / raw)
To: willy, akpm, hannes, mhocko, vdavydov.dev, shakeelb, guro,
shy828301, alexs, richard.weiyang, david, trond.myklebust,
anna.schumaker
Cc: linux-fsdevel, linux-kernel, linux-mm, linux-nfs, zhengqi.arch,
duanxiongchun, fam.zheng, Muchun Song
In our server, we found a suspected memory leak problem. The kmalloc-32
consumes more than 6GB of memory. Other kmem_caches consume less than
2GB memory.
After our in-depth analysis, the memory consumption of kmalloc-32 slab
cache is the cause of list_lru_one allocation.
crash> p memcg_nr_cache_ids
memcg_nr_cache_ids = $2 = 24574
memcg_nr_cache_ids is very large and memory consumption of each list_lru
can be calculated with the following formula.
num_numa_node * memcg_nr_cache_ids * 32 (kmalloc-32)
There are 4 numa nodes in our system, so each list_lru consumes ~3MB.
crash> list super_blocks | wc -l
952
Every mount will register 2 list lrus, one is for inode, another is for
dentry. There are 952 super_blocks. So the total memory is 952 * 2 * 3
MB (~5.6GB). But the number of memory cgroup is less than 500. So I
guess more than 12286 containers have been deployed on this machine (I
do not know why there are so many containers, it may be a user's bug or
the user really want to do that). And memcg_nr_cache_ids has not been
reduced to a suitable value. This can waste a lot of memory.
Now the infrastructure for dynamic list_lru_one allocation is ready, so
remove statically allocated memory code to save memory.
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
---
include/linux/list_lru.h | 8 +--
mm/list_lru.c | 125 +++++++++++++++++++++++++++--------------------
mm/memcontrol.c | 7 ++-
3 files changed, 81 insertions(+), 59 deletions(-)
diff --git a/include/linux/list_lru.h b/include/linux/list_lru.h
index 2083f4f2701f..00c15dc1ede3 100644
--- a/include/linux/list_lru.h
+++ b/include/linux/list_lru.h
@@ -32,14 +32,15 @@ struct list_lru_one {
};
struct list_lru_per_memcg {
+ struct rcu_head rcu;
/* array of per cgroup per node lists, indexed by node id */
- struct list_lru_one nodes[0];
+ struct list_lru_one nodes[];
};
struct list_lru_memcg {
struct rcu_head rcu;
/* array of per cgroup lists, indexed by memcg_cache_id */
- struct list_lru_per_memcg *lrus[];
+ struct list_lru_per_memcg __rcu *lrus[];
};
struct list_lru_node {
@@ -75,7 +76,8 @@ int __list_lru_init(struct list_lru *lru, bool memcg_aware,
__list_lru_init((lru), true, NULL, shrinker)
int memcg_update_all_list_lrus(int num_memcgs);
-void memcg_drain_all_list_lrus(int src_idx, struct mem_cgroup *dst_memcg);
+void memcg_drain_all_list_lrus(struct mem_cgroup *src_memcg,
+ struct mem_cgroup *dst_memcg);
/**
* list_lru_add: add an element to the lru list's tail
diff --git a/mm/list_lru.c b/mm/list_lru.c
index 4ba1db6d4409..43eadb0834e1 100644
--- a/mm/list_lru.c
+++ b/mm/list_lru.c
@@ -60,8 +60,12 @@ list_lru_from_memcg_idx(struct list_lru *lru, int nid, int idx)
*/
memcg_lrus = rcu_dereference_check(lru->memcg_lrus,
lockdep_is_held(&nlru->lock));
- if (memcg_lrus && idx >= 0)
- return &memcg_lrus->lrus[idx]->nodes[nid];
+ if (memcg_lrus && idx >= 0) {
+ struct list_lru_per_memcg *mlru;
+
+ mlru = rcu_dereference_check(memcg_lrus->lrus[idx], true);
+ return mlru ? &mlru->nodes[nid] : NULL;
+ }
return &nlru->lru;
}
@@ -184,11 +188,12 @@ unsigned long list_lru_count_one(struct list_lru *lru,
int nid, struct mem_cgroup *memcg)
{
struct list_lru_one *l;
- long count;
+ long count = 0;
rcu_read_lock();
l = list_lru_from_memcg_idx(lru, nid, memcg_cache_id(memcg));
- count = READ_ONCE(l->nr_items);
+ if (l)
+ count = READ_ONCE(l->nr_items);
rcu_read_unlock();
if (unlikely(count < 0))
@@ -217,8 +222,11 @@ __list_lru_walk_one(struct list_lru *lru, int nid, int memcg_idx,
struct list_head *item, *n;
unsigned long isolated = 0;
- l = list_lru_from_memcg_idx(lru, nid, memcg_idx);
restart:
+ l = list_lru_from_memcg_idx(lru, nid, memcg_idx);
+ if (!l)
+ goto out;
+
list_for_each_safe(item, n, &l->list) {
enum lru_status ret;
@@ -262,6 +270,7 @@ __list_lru_walk_one(struct list_lru *lru, int nid, int memcg_idx,
BUG();
}
}
+out:
return isolated;
}
@@ -354,20 +363,26 @@ static struct list_lru_per_memcg *memcg_list_lru_alloc(gfp_t gfp)
return lru;
}
-static int memcg_init_list_lru_range(struct list_lru_memcg *memcg_lrus,
- int begin, int end)
+static void memcg_list_lru_free(struct list_lru *lru, int src_idx)
{
- int i;
+ struct list_lru_memcg *memcg_lrus;
+ struct list_lru_per_memcg *mlru;
- for (i = begin; i < end; i++) {
- memcg_lrus->lrus[i] = memcg_list_lru_alloc(GFP_KERNEL);
- if (!memcg_lrus->lrus[i])
- goto fail;
- }
- return 0;
-fail:
- memcg_destroy_list_lru_range(memcg_lrus, begin, i);
- return -ENOMEM;
+ spin_lock_irq(&lru->lock);
+ memcg_lrus = rcu_dereference_protected(lru->memcg_lrus, true);
+ mlru = rcu_dereference_protected(memcg_lrus->lrus[src_idx], true);
+ if (mlru)
+ rcu_assign_pointer(memcg_lrus->lrus[src_idx], NULL);
+ spin_unlock_irq(&lru->lock);
+
+ /*
+ * The __list_lru_walk_one() can walk the list of this node.
+ * We need kvfree_rcu() here. And the walking of the list
+ * is under lru->node[nid]->lock, which can serve as a RCU
+ * read-side critical section.
+ */
+ if (mlru)
+ kvfree_rcu(mlru, rcu);
}
static int memcg_init_list_lru(struct list_lru *lru, bool memcg_aware)
@@ -382,15 +397,11 @@ static int memcg_init_list_lru(struct list_lru *lru, bool memcg_aware)
spin_lock_init(&lru->lock);
- memcg_lrus = kvmalloc(sizeof(*memcg_lrus) +
+ memcg_lrus = kvzalloc(sizeof(*memcg_lrus) +
size * sizeof(memcg_lrus->lrus[0]), GFP_KERNEL);
if (!memcg_lrus)
return -ENOMEM;
- if (memcg_init_list_lru_range(memcg_lrus, 0, size)) {
- kvfree(memcg_lrus);
- return -ENOMEM;
- }
RCU_INIT_POINTER(lru->memcg_lrus, memcg_lrus);
return 0;
@@ -424,13 +435,9 @@ static int memcg_update_list_lru(struct list_lru *lru, int old_size, int new_siz
if (!new)
return -ENOMEM;
- if (memcg_init_list_lru_range(new, old_size, new_size)) {
- kvfree(new);
- return -ENOMEM;
- }
-
spin_lock_irq(&lru->lock);
memcpy(&new->lrus, &old->lrus, old_size * sizeof(new->lrus[0]));
+ memset(&new->lrus[old_size], 0, (new_size - old_size) * sizeof(new->lrus[0]));
rcu_assign_pointer(lru->memcg_lrus, new);
spin_unlock_irq(&lru->lock);
@@ -439,20 +446,6 @@ static int memcg_update_list_lru(struct list_lru *lru, int old_size, int new_siz
return 0;
}
-static void memcg_cancel_update_list_lru(struct list_lru *lru,
- int old_size, int new_size)
-{
- struct list_lru_memcg *memcg_lrus;
-
- memcg_lrus = rcu_dereference_protected(lru->memcg_lrus,
- lockdep_is_held(&list_lrus_mutex));
- /*
- * Do not bother shrinking the array back to the old size, because we
- * cannot handle allocation failures here.
- */
- memcg_destroy_list_lru_range(memcg_lrus, old_size, new_size);
-}
-
int memcg_update_all_list_lrus(int new_size)
{
int ret = 0;
@@ -463,15 +456,10 @@ int memcg_update_all_list_lrus(int new_size)
list_for_each_entry(lru, &list_lrus, list) {
ret = memcg_update_list_lru(lru, old_size, new_size);
if (ret)
- goto fail;
+ break;
}
-out:
mutex_unlock(&list_lrus_mutex);
return ret;
-fail:
- list_for_each_entry_continue_reverse(lru, &list_lrus, list)
- memcg_cancel_update_list_lru(lru, old_size, new_size);
- goto out;
}
static void memcg_drain_list_lru_node(struct list_lru *lru, int nid,
@@ -488,6 +476,8 @@ static void memcg_drain_list_lru_node(struct list_lru *lru, int nid,
spin_lock_irq(&nlru->lock);
src = list_lru_from_memcg_idx(lru, nid, src_idx);
+ if (!src)
+ goto out;
dst = list_lru_from_memcg_idx(lru, nid, dst_idx);
list_splice_init(&src->list, &dst->list);
@@ -497,7 +487,7 @@ static void memcg_drain_list_lru_node(struct list_lru *lru, int nid,
set_shrinker_bit(dst_memcg, nid, lru_shrinker_id(lru));
src->nr_items = 0;
}
-
+out:
spin_unlock_irq(&nlru->lock);
}
@@ -508,11 +498,37 @@ static void memcg_drain_list_lru(struct list_lru *lru,
for_each_node(i)
memcg_drain_list_lru_node(lru, i, src_idx, dst_memcg);
+
+ memcg_list_lru_free(lru, src_idx);
}
-void memcg_drain_all_list_lrus(int src_idx, struct mem_cgroup *dst_memcg)
+void memcg_drain_all_list_lrus(struct mem_cgroup *src_memcg,
+ struct mem_cgroup *dst_memcg)
{
+ struct cgroup_subsys_state *css;
struct list_lru *lru;
+ int src_idx = src_memcg->kmemcg_id;
+
+ /*
+ * Change kmemcg_id of this cgroup and all its descendants to the
+ * parent's id, and then move all entries from this cgroup's list_lrus
+ * to ones of the parent.
+ *
+ * After we have finished, all list_lrus corresponding to this cgroup
+ * are guaranteed to remain empty. So we can safely free this cgroup's
+ * list lrus which is freed in memcg_list_lru_free().
+ * Changing ->kmemcg_id to the parent can prevent list_lru_memcg_alloc()
+ * from allocating list lrus for this cgroup after calling
+ * memcg_list_lru_free().
+ */
+ rcu_read_lock();
+ css_for_each_descendant_pre(css, &src_memcg->css) {
+ struct mem_cgroup *memcg;
+
+ memcg = mem_cgroup_from_css(css);
+ memcg->kmemcg_id = dst_memcg->kmemcg_id;
+ }
+ rcu_read_unlock();
mutex_lock(&list_lrus_mutex);
list_for_each_entry(lru, &list_lrus, list)
@@ -531,7 +547,7 @@ static bool memcg_list_lru_skip_alloc(struct list_lru *lru,
rcu_read_lock();
memcg_lrus = rcu_dereference(lru->memcg_lrus);
- if (memcg_lrus->lrus[idx]) {
+ if (rcu_access_pointer(memcg_lrus->lrus[idx])) {
rcu_read_unlock();
return true;
}
@@ -546,7 +562,7 @@ int list_lru_memcg_alloc(struct list_lru *lru, struct mem_cgroup *memcg, gfp_t g
struct list_lru_memcg *memcg_lrus;
int i;
- struct list_lru_memcg {
+ struct list_lru_memcg_table {
struct list_lru_per_memcg *mlru;
struct mem_cgroup *memcg;
} *table;
@@ -590,11 +606,12 @@ int list_lru_memcg_alloc(struct list_lru *lru, struct mem_cgroup *memcg, gfp_t g
memcg_lrus = rcu_dereference_protected(lru->memcg_lrus, true);
while (i--) {
int index = memcg_cache_id(table[i].memcg);
+ struct list_lru_per_memcg *mlru = table[i].mlru;
- if (memcg_lrus->lrus[index])
- kfree(table[i].mlru);
+ if (index < 0 || rcu_dereference_protected(memcg_lrus->lrus[index], true))
+ kfree(mlru);
else
- memcg_lrus->lrus[index] = table[i].mlru;
+ rcu_assign_pointer(memcg_lrus->lrus[index], mlru);
}
spin_unlock_irqrestore(&lru->lock, flags);
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 09bafa82781f..1994513a33d4 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -3654,10 +3654,13 @@ static void memcg_offline_kmem(struct mem_cgroup *memcg)
memcg_reparent_objcgs(memcg, parent);
+ /*
+ * memcg_drain_all_list_lrus() can change memcg->kmemcg_id.
+ * Cache it to @kmemcg_id.
+ */
kmemcg_id = memcg->kmemcg_id;
- /* memcg_reparent_objcgs() must be called before this. */
- memcg_drain_all_list_lrus(kmemcg_id, parent);
+ memcg_drain_all_list_lrus(memcg, parent);
memcg_free_cache_id(kmemcg_id);
}
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v2 16/21] mm: list_lru: rename memcg_drain_all_list_lrus to memcg_reparent_list_lrus
2021-05-27 6:21 [PATCH v2 00/21] Optimize list lru memory consumption Muchun Song
` (14 preceding siblings ...)
2021-05-27 6:21 ` [PATCH v2 15/21] mm: list_lru: allocate list_lru_one only when needed Muchun Song
@ 2021-05-27 6:21 ` Muchun Song
2021-05-27 6:21 ` [PATCH v2 17/21] mm: list_lru: replace linear array with xarray Muchun Song
` (5 subsequent siblings)
21 siblings, 0 replies; 29+ messages in thread
From: Muchun Song @ 2021-05-27 6:21 UTC (permalink / raw)
To: willy, akpm, hannes, mhocko, vdavydov.dev, shakeelb, guro,
shy828301, alexs, richard.weiyang, david, trond.myklebust,
anna.schumaker
Cc: linux-fsdevel, linux-kernel, linux-mm, linux-nfs, zhengqi.arch,
duanxiongchun, fam.zheng, Muchun Song
The purpose of the memcg_drain_all_list_lrus() is list_lrus reparenting.
It is very similar to memcg_reparent_objcgs(). Rename it to
memcg_reparent_list_lrus() so that the name can more consistent with
memcg_reparent_objcgs().
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
---
include/linux/list_lru.h | 3 +--
mm/list_lru.c | 25 ++++++++++++-------------
mm/memcontrol.c | 4 ++--
3 files changed, 15 insertions(+), 17 deletions(-)
diff --git a/include/linux/list_lru.h b/include/linux/list_lru.h
index 00c15dc1ede3..60970ea48c29 100644
--- a/include/linux/list_lru.h
+++ b/include/linux/list_lru.h
@@ -76,8 +76,7 @@ int __list_lru_init(struct list_lru *lru, bool memcg_aware,
__list_lru_init((lru), true, NULL, shrinker)
int memcg_update_all_list_lrus(int num_memcgs);
-void memcg_drain_all_list_lrus(struct mem_cgroup *src_memcg,
- struct mem_cgroup *dst_memcg);
+void memcg_reparent_list_lrus(struct mem_cgroup *memcg, struct mem_cgroup *parent);
/**
* list_lru_add: add an element to the lru list's tail
diff --git a/mm/list_lru.c b/mm/list_lru.c
index 43eadb0834e1..dec116a12629 100644
--- a/mm/list_lru.c
+++ b/mm/list_lru.c
@@ -462,8 +462,8 @@ int memcg_update_all_list_lrus(int new_size)
return ret;
}
-static void memcg_drain_list_lru_node(struct list_lru *lru, int nid,
- int src_idx, struct mem_cgroup *dst_memcg)
+static void memcg_reparent_list_lru_node(struct list_lru *lru, int nid,
+ int src_idx, struct mem_cgroup *dst_memcg)
{
struct list_lru_node *nlru = &lru->node[nid];
int dst_idx = dst_memcg->kmemcg_id;
@@ -491,23 +491,22 @@ static void memcg_drain_list_lru_node(struct list_lru *lru, int nid,
spin_unlock_irq(&nlru->lock);
}
-static void memcg_drain_list_lru(struct list_lru *lru,
- int src_idx, struct mem_cgroup *dst_memcg)
+static void memcg_reparent_list_lru(struct list_lru *lru,
+ int src_idx, struct mem_cgroup *dst_memcg)
{
int i;
for_each_node(i)
- memcg_drain_list_lru_node(lru, i, src_idx, dst_memcg);
+ memcg_reparent_list_lru_node(lru, i, src_idx, dst_memcg);
memcg_list_lru_free(lru, src_idx);
}
-void memcg_drain_all_list_lrus(struct mem_cgroup *src_memcg,
- struct mem_cgroup *dst_memcg)
+void memcg_reparent_list_lrus(struct mem_cgroup *memcg, struct mem_cgroup *parent)
{
struct cgroup_subsys_state *css;
struct list_lru *lru;
- int src_idx = src_memcg->kmemcg_id;
+ int src_idx = memcg->kmemcg_id;
/*
* Change kmemcg_id of this cgroup and all its descendants to the
@@ -522,17 +521,17 @@ void memcg_drain_all_list_lrus(struct mem_cgroup *src_memcg,
* memcg_list_lru_free().
*/
rcu_read_lock();
- css_for_each_descendant_pre(css, &src_memcg->css) {
- struct mem_cgroup *memcg;
+ css_for_each_descendant_pre(css, &memcg->css) {
+ struct mem_cgroup *child;
- memcg = mem_cgroup_from_css(css);
- memcg->kmemcg_id = dst_memcg->kmemcg_id;
+ child = mem_cgroup_from_css(css);
+ child->kmemcg_id = parent->kmemcg_id;
}
rcu_read_unlock();
mutex_lock(&list_lrus_mutex);
list_for_each_entry(lru, &list_lrus, list)
- memcg_drain_list_lru(lru, src_idx, dst_memcg);
+ memcg_reparent_list_lru(lru, src_idx, parent);
mutex_unlock(&list_lrus_mutex);
}
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 1994513a33d4..205bf710486c 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -3655,12 +3655,12 @@ static void memcg_offline_kmem(struct mem_cgroup *memcg)
memcg_reparent_objcgs(memcg, parent);
/*
- * memcg_drain_all_list_lrus() can change memcg->kmemcg_id.
+ * memcg_reparent_list_lrus() can change memcg->kmemcg_id.
* Cache it to @kmemcg_id.
*/
kmemcg_id = memcg->kmemcg_id;
- memcg_drain_all_list_lrus(memcg, parent);
+ memcg_reparent_list_lrus(memcg, parent);
memcg_free_cache_id(kmemcg_id);
}
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v2 17/21] mm: list_lru: replace linear array with xarray
2021-05-27 6:21 [PATCH v2 00/21] Optimize list lru memory consumption Muchun Song
` (15 preceding siblings ...)
2021-05-27 6:21 ` [PATCH v2 16/21] mm: list_lru: rename memcg_drain_all_list_lrus to memcg_reparent_list_lrus Muchun Song
@ 2021-05-27 6:21 ` Muchun Song
2021-05-27 12:07 ` Matthew Wilcox
2021-05-27 6:21 ` [PATCH v2 18/21] mm: memcontrol: reuse memory cgroup ID for kmem ID Muchun Song
` (4 subsequent siblings)
21 siblings, 1 reply; 29+ messages in thread
From: Muchun Song @ 2021-05-27 6:21 UTC (permalink / raw)
To: willy, akpm, hannes, mhocko, vdavydov.dev, shakeelb, guro,
shy828301, alexs, richard.weiyang, david, trond.myklebust,
anna.schumaker
Cc: linux-fsdevel, linux-kernel, linux-mm, linux-nfs, zhengqi.arch,
duanxiongchun, fam.zheng, Muchun Song
If we run 10k containers in the system, the size of the
list_lru_memcg->lrus can be ~96KB per list_lru. When we decrease the
number containers, the size of the array will not be shrinked. It is
not scalable. The xarray is a good choice for this case. We can save
a lot of memory when there are tens of thousands continers in the
system. If we use xarray, we also can remove the logic code of
resizing array, which can simplify the code.
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
---
include/linux/list_lru.h | 12 +---
include/linux/memcontrol.h | 23 ------
mm/list_lru.c | 173 +++++++++++++++------------------------------
mm/memcontrol.c | 77 ++------------------
4 files changed, 63 insertions(+), 222 deletions(-)
diff --git a/include/linux/list_lru.h b/include/linux/list_lru.h
index 60970ea48c29..d7c9bd29e836 100644
--- a/include/linux/list_lru.h
+++ b/include/linux/list_lru.h
@@ -11,6 +11,7 @@
#include <linux/list.h>
#include <linux/nodemask.h>
#include <linux/shrinker.h>
+#include <linux/xarray.h>
struct mem_cgroup;
@@ -37,12 +38,6 @@ struct list_lru_per_memcg {
struct list_lru_one nodes[];
};
-struct list_lru_memcg {
- struct rcu_head rcu;
- /* array of per cgroup lists, indexed by memcg_cache_id */
- struct list_lru_per_memcg __rcu *lrus[];
-};
-
struct list_lru_node {
/* protects all lists on the node, including per cgroup */
spinlock_t lock;
@@ -56,10 +51,8 @@ struct list_lru {
#ifdef CONFIG_MEMCG_KMEM
struct list_head list;
int shrinker_id;
- /* protects ->memcg_lrus->lrus[i] */
- spinlock_t lock;
/* for cgroup aware lrus points to per cgroup lists, otherwise NULL */
- struct list_lru_memcg __rcu *memcg_lrus;
+ struct xarray *xa;
#endif
};
@@ -75,7 +68,6 @@ int __list_lru_init(struct list_lru *lru, bool memcg_aware,
#define list_lru_init_memcg(lru, shrinker) \
__list_lru_init((lru), true, NULL, shrinker)
-int memcg_update_all_list_lrus(int num_memcgs);
void memcg_reparent_list_lrus(struct mem_cgroup *memcg, struct mem_cgroup *parent);
/**
diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index 84a70c219e85..e129c7067f63 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -1631,18 +1631,6 @@ void obj_cgroup_uncharge(struct obj_cgroup *objcg, size_t size);
extern struct static_key_false memcg_kmem_enabled_key;
-extern int memcg_nr_cache_ids;
-void memcg_get_cache_ids(void);
-void memcg_put_cache_ids(void);
-
-/*
- * Helper macro to loop through all memcg-specific caches. Callers must still
- * check if the cache is valid (it is either valid or NULL).
- * the slab_mutex must be held when looping through those caches
- */
-#define for_each_memcg_cache_index(_idx) \
- for ((_idx) = 0; (_idx) < memcg_nr_cache_ids; (_idx)++)
-
static inline bool memcg_kmem_enabled(void)
{
return static_branch_likely(&memcg_kmem_enabled_key);
@@ -1695,9 +1683,6 @@ static inline void __memcg_kmem_uncharge_page(struct page *page, int order)
{
}
-#define for_each_memcg_cache_index(_idx) \
- for (; NULL; )
-
static inline bool memcg_kmem_enabled(void)
{
return false;
@@ -1708,14 +1693,6 @@ static inline int memcg_cache_id(struct mem_cgroup *memcg)
return -1;
}
-static inline void memcg_get_cache_ids(void)
-{
-}
-
-static inline void memcg_put_cache_ids(void)
-{
-}
-
static inline struct mem_cgroup *mem_cgroup_from_obj(void *p)
{
return NULL;
diff --git a/mm/list_lru.c b/mm/list_lru.c
index dec116a12629..37052864bf78 100644
--- a/mm/list_lru.c
+++ b/mm/list_lru.c
@@ -20,7 +20,7 @@ static DEFINE_MUTEX(list_lrus_mutex);
static inline bool list_lru_memcg_aware(struct list_lru *lru)
{
- return !!rcu_access_pointer(lru->memcg_lrus);
+ return !!lru->xa;
}
static void list_lru_register(struct list_lru *lru)
@@ -51,22 +51,12 @@ static int lru_shrinker_id(struct list_lru *lru)
static inline struct list_lru_one *
list_lru_from_memcg_idx(struct list_lru *lru, int nid, int idx)
{
- struct list_lru_memcg *memcg_lrus;
- struct list_lru_node *nlru = &lru->node[nid];
+ if (list_lru_memcg_aware(lru) && idx >= 0) {
+ struct list_lru_per_memcg *mlru = xa_load(lru->xa, idx);
- /*
- * Either lock or RCU protects the array of per cgroup lists
- * from relocation (see memcg_update_list_lru).
- */
- memcg_lrus = rcu_dereference_check(lru->memcg_lrus,
- lockdep_is_held(&nlru->lock));
- if (memcg_lrus && idx >= 0) {
- struct list_lru_per_memcg *mlru;
-
- mlru = rcu_dereference_check(memcg_lrus->lrus[idx], true);
return mlru ? &mlru->nodes[nid] : NULL;
}
- return &nlru->lru;
+ return &lru->node[nid].lru;
}
static inline struct list_lru_one *
@@ -310,16 +300,18 @@ unsigned long list_lru_walk_node(struct list_lru *lru, int nid,
unsigned long *nr_to_walk)
{
long isolated = 0;
- int memcg_idx;
isolated += list_lru_walk_one(lru, nid, NULL, isolate, cb_arg,
nr_to_walk);
if (*nr_to_walk > 0 && list_lru_memcg_aware(lru)) {
- for_each_memcg_cache_index(memcg_idx) {
+ struct list_lru_per_memcg *mlru;
+ unsigned long index;
+
+ xa_for_each(lru->xa, index, mlru) {
struct list_lru_node *nlru = &lru->node[nid];
spin_lock(&nlru->lock);
- isolated += __list_lru_walk_one(lru, nid, memcg_idx,
+ isolated += __list_lru_walk_one(lru, nid, index,
isolate, cb_arg,
nr_to_walk);
spin_unlock(&nlru->lock);
@@ -339,15 +331,6 @@ static void init_one_lru(struct list_lru_one *l)
}
#ifdef CONFIG_MEMCG_KMEM
-static void memcg_destroy_list_lru_range(struct list_lru_memcg *memcg_lrus,
- int begin, int end)
-{
- int i;
-
- for (i = begin; i < end; i++)
- kfree(memcg_lrus->lrus[i]);
-}
-
static struct list_lru_per_memcg *memcg_list_lru_alloc(gfp_t gfp)
{
int nid;
@@ -365,15 +348,7 @@ static struct list_lru_per_memcg *memcg_list_lru_alloc(gfp_t gfp)
static void memcg_list_lru_free(struct list_lru *lru, int src_idx)
{
- struct list_lru_memcg *memcg_lrus;
- struct list_lru_per_memcg *mlru;
-
- spin_lock_irq(&lru->lock);
- memcg_lrus = rcu_dereference_protected(lru->memcg_lrus, true);
- mlru = rcu_dereference_protected(memcg_lrus->lrus[src_idx], true);
- if (mlru)
- rcu_assign_pointer(memcg_lrus->lrus[src_idx], NULL);
- spin_unlock_irq(&lru->lock);
+ struct list_lru_per_memcg *mlru = xa_erase_irq(lru->xa, src_idx);
/*
* The __list_lru_walk_one() can walk the list of this node.
@@ -387,79 +362,35 @@ static void memcg_list_lru_free(struct list_lru *lru, int src_idx)
static int memcg_init_list_lru(struct list_lru *lru, bool memcg_aware)
{
- struct list_lru_memcg *memcg_lrus;
- int size = memcg_nr_cache_ids;
-
if (!memcg_aware) {
- lru->memcg_lrus = NULL;
+ lru->xa = NULL;
return 0;
}
- spin_lock_init(&lru->lock);
-
- memcg_lrus = kvzalloc(sizeof(*memcg_lrus) +
- size * sizeof(memcg_lrus->lrus[0]), GFP_KERNEL);
- if (!memcg_lrus)
+ lru->xa = kmalloc(sizeof(*lru->xa), GFP_KERNEL);
+ if (!lru->xa)
return -ENOMEM;
-
- RCU_INIT_POINTER(lru->memcg_lrus, memcg_lrus);
+ xa_init_flags(lru->xa, XA_FLAGS_LOCK_IRQ);
return 0;
}
static void memcg_destroy_list_lru(struct list_lru *lru)
{
- struct list_lru_memcg *memcg_lrus;
+ XA_STATE(xas, lru->xa, 0);
+ struct list_lru_per_memcg *mlru;
if (!list_lru_memcg_aware(lru))
return;
- /*
- * This is called when shrinker has already been unregistered,
- * and nobody can use it. So, there is no need to use kvfree_rcu().
- */
- memcg_lrus = rcu_dereference_protected(lru->memcg_lrus, true);
- memcg_destroy_list_lru_range(memcg_lrus, 0, memcg_nr_cache_ids);
- kvfree(memcg_lrus);
-}
-
-static int memcg_update_list_lru(struct list_lru *lru, int old_size, int new_size)
-{
- struct list_lru_memcg *old, *new;
-
- BUG_ON(old_size > new_size);
-
- old = rcu_dereference_protected(lru->memcg_lrus,
- lockdep_is_held(&list_lrus_mutex));
- new = kvmalloc(sizeof(*new) + new_size * sizeof(new->lrus[0]), GFP_KERNEL);
- if (!new)
- return -ENOMEM;
-
- spin_lock_irq(&lru->lock);
- memcpy(&new->lrus, &old->lrus, old_size * sizeof(new->lrus[0]));
- memset(&new->lrus[old_size], 0, (new_size - old_size) * sizeof(new->lrus[0]));
- rcu_assign_pointer(lru->memcg_lrus, new);
- spin_unlock_irq(&lru->lock);
-
- kvfree_rcu(old, rcu);
-
- return 0;
-}
-
-int memcg_update_all_list_lrus(int new_size)
-{
- int ret = 0;
- struct list_lru *lru;
- int old_size = memcg_nr_cache_ids;
-
- mutex_lock(&list_lrus_mutex);
- list_for_each_entry(lru, &list_lrus, list) {
- ret = memcg_update_list_lru(lru, old_size, new_size);
- if (ret)
- break;
+ xas_lock_irq(&xas);
+ xas_for_each(&xas, mlru, ULONG_MAX) {
+ kfree(mlru);
+ xas_store(&xas, NULL);
}
- mutex_unlock(&list_lrus_mutex);
- return ret;
+ xas_unlock_irq(&xas);
+
+ kfree(lru->xa);
}
static void memcg_reparent_list_lru_node(struct list_lru *lru, int nid,
@@ -538,28 +469,18 @@ void memcg_reparent_list_lrus(struct mem_cgroup *memcg, struct mem_cgroup *paren
static bool memcg_list_lru_skip_alloc(struct list_lru *lru,
struct mem_cgroup *memcg)
{
- struct list_lru_memcg *memcg_lrus;
int idx = memcg_cache_id(memcg);
- if (unlikely(idx < 0))
+ if (unlikely(idx < 0) || xa_load(lru->xa, idx))
return true;
-
- rcu_read_lock();
- memcg_lrus = rcu_dereference(lru->memcg_lrus);
- if (rcu_access_pointer(memcg_lrus->lrus[idx])) {
- rcu_read_unlock();
- return true;
- }
- rcu_read_unlock();
-
return false;
}
int list_lru_memcg_alloc(struct list_lru *lru, struct mem_cgroup *memcg, gfp_t gfp)
{
+ XA_STATE(xas, lru->xa, 0);
unsigned long flags;
- struct list_lru_memcg *memcg_lrus;
- int i;
+ int i, ret = 0;
struct list_lru_memcg_table {
struct list_lru_per_memcg *mlru;
@@ -601,22 +522,45 @@ int list_lru_memcg_alloc(struct list_lru *lru, struct mem_cgroup *memcg, gfp_t g
}
}
- spin_lock_irqsave(&lru->lock, flags);
- memcg_lrus = rcu_dereference_protected(lru->memcg_lrus, true);
+ xas_lock_irqsave(&xas, flags);
while (i--) {
int index = memcg_cache_id(table[i].memcg);
struct list_lru_per_memcg *mlru = table[i].mlru;
- if (index < 0 || rcu_dereference_protected(memcg_lrus->lrus[index], true))
+ xas_set(&xas, index);
+retry:
+ if (unlikely(index < 0 || ret || xas_load(&xas))) {
kfree(mlru);
- else
- rcu_assign_pointer(memcg_lrus->lrus[index], mlru);
+ } else {
+ ret = xa_err(xas_store(&xas, mlru));
+ if (ret == -ENOMEM) {
+ xas_unlock_irqrestore(&xas, flags);
+ if (xas_nomem(&xas, gfp))
+ ret = 0;
+ xas_lock_irqsave(&xas, flags);
+ /*
+ * The xas lock has been released, this memcg
+ * can be reparented before us. So reload
+ * memcg id. More details see the comments
+ * in memcg_reparent_list_lrus().
+ */
+ index = memcg_cache_id(table[i].memcg);
+ if (index < 0)
+ ret = 0;
+ else if (!ret && index != xas.xa_index)
+ xas_set(&xas, index);
+ goto retry;
+ }
+ }
}
- spin_unlock_irqrestore(&lru->lock, flags);
+ /* xas_nomem() is used to free memory instead of memory allocation. */
+ if (xas.xa_alloc)
+ xas_nomem(&xas, gfp);
+ xas_unlock_irqrestore(&xas, flags);
kfree(table);
- return 0;
+ return ret;
}
#else
static int memcg_init_list_lru(struct list_lru *lru, bool memcg_aware)
@@ -641,7 +585,6 @@ int __list_lru_init(struct list_lru *lru, bool memcg_aware,
else
lru->shrinker_id = -1;
#endif
- memcg_get_cache_ids();
lru->node = kcalloc(nr_node_ids, sizeof(*lru->node), GFP_KERNEL);
if (!lru->node)
@@ -664,7 +607,6 @@ int __list_lru_init(struct list_lru *lru, bool memcg_aware,
list_lru_register(lru);
out:
- memcg_put_cache_ids();
return err;
}
EXPORT_SYMBOL_GPL(__list_lru_init);
@@ -675,8 +617,6 @@ void list_lru_destroy(struct list_lru *lru)
if (!lru->node)
return;
- memcg_get_cache_ids();
-
list_lru_unregister(lru);
memcg_destroy_list_lru(lru);
@@ -686,6 +626,5 @@ void list_lru_destroy(struct list_lru *lru)
#ifdef CONFIG_MEMCG_KMEM
lru->shrinker_id = -1;
#endif
- memcg_put_cache_ids();
}
EXPORT_SYMBOL_GPL(list_lru_destroy);
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 205bf710486c..82b147597138 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -345,42 +345,17 @@ static void memcg_reparent_objcgs(struct mem_cgroup *memcg,
* This will be used as a shrinker list's index.
* The main reason for not using cgroup id for this:
* this works better in sparse environments, where we have a lot of memcgs,
- * but only a few kmem-limited. Or also, if we have, for instance, 200
- * memcgs, and none but the 200th is kmem-limited, we'd have to have a
- * 200 entry array for that.
- *
- * The current size of the caches array is stored in memcg_nr_cache_ids. It
- * will double each time we have to increase it.
+ * but only a few kmem-limited.
*/
static DEFINE_IDA(memcg_cache_ida);
-int memcg_nr_cache_ids;
-
-/* Protects memcg_nr_cache_ids */
-static DECLARE_RWSEM(memcg_cache_ids_sem);
-
-void memcg_get_cache_ids(void)
-{
- down_read(&memcg_cache_ids_sem);
-}
-
-void memcg_put_cache_ids(void)
-{
- up_read(&memcg_cache_ids_sem);
-}
/*
- * MIN_SIZE is different than 1, because we would like to avoid going through
- * the alloc/free process all the time. In a small machine, 4 kmem-limited
- * cgroups is a reasonable guess. In the future, it could be a parameter or
- * tunable, but that is strictly not necessary.
- *
* MAX_SIZE should be as large as the number of cgrp_ids. Ideally, we could get
* this constant directly from cgroup, but it is understandable that this is
* better kept as an internal representation in cgroup.c. In any case, the
* cgrp_id space is not getting any smaller, and we don't have to necessarily
* increase ours as well if it increases.
*/
-#define MEMCG_CACHES_MIN_SIZE 4
#define MEMCG_CACHES_MAX_SIZE MEM_CGROUP_ID_MAX
/*
@@ -2903,49 +2878,6 @@ __always_inline struct obj_cgroup *get_obj_cgroup_from_current(void)
return objcg;
}
-static int memcg_alloc_cache_id(void)
-{
- int id, size;
- int err;
-
- id = ida_simple_get(&memcg_cache_ida,
- 0, MEMCG_CACHES_MAX_SIZE, GFP_KERNEL);
- if (id < 0)
- return id;
-
- if (id < memcg_nr_cache_ids)
- return id;
-
- /*
- * There's no space for the new id in memcg_caches arrays,
- * so we have to grow them.
- */
- down_write(&memcg_cache_ids_sem);
-
- size = 2 * (id + 1);
- if (size < MEMCG_CACHES_MIN_SIZE)
- size = MEMCG_CACHES_MIN_SIZE;
- else if (size > MEMCG_CACHES_MAX_SIZE)
- size = MEMCG_CACHES_MAX_SIZE;
-
- err = memcg_update_all_list_lrus(size);
- if (!err)
- memcg_nr_cache_ids = size;
-
- up_write(&memcg_cache_ids_sem);
-
- if (err) {
- ida_simple_remove(&memcg_cache_ida, id);
- return err;
- }
- return id;
-}
-
-static void memcg_free_cache_id(int id)
-{
- ida_simple_remove(&memcg_cache_ida, id);
-}
-
/*
* obj_cgroup_uncharge_pages: uncharge a number of kernel pages from a objcg
* @objcg: object cgroup to uncharge
@@ -3618,13 +3550,14 @@ static int memcg_online_kmem(struct mem_cgroup *memcg)
if (unlikely(mem_cgroup_is_root(memcg)))
return 0;
- memcg_id = memcg_alloc_cache_id();
+ memcg_id = ida_simple_get(&memcg_cache_ida, 0, MEMCG_CACHES_MAX_SIZE,
+ GFP_KERNEL);
if (memcg_id < 0)
return memcg_id;
objcg = obj_cgroup_alloc();
if (!objcg) {
- memcg_free_cache_id(memcg_id);
+ ida_simple_remove(&memcg_cache_ida, memcg_id);
return -ENOMEM;
}
objcg->memcg = memcg;
@@ -3662,7 +3595,7 @@ static void memcg_offline_kmem(struct mem_cgroup *memcg)
memcg_reparent_list_lrus(memcg, parent);
- memcg_free_cache_id(kmemcg_id);
+ ida_simple_remove(&memcg_cache_ida, kmemcg_id);
}
#else
static int memcg_online_kmem(struct mem_cgroup *memcg)
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* Re: [PATCH v2 17/21] mm: list_lru: replace linear array with xarray
2021-05-27 6:21 ` [PATCH v2 17/21] mm: list_lru: replace linear array with xarray Muchun Song
@ 2021-05-27 12:07 ` Matthew Wilcox
2021-05-28 3:43 ` Muchun Song
0 siblings, 1 reply; 29+ messages in thread
From: Matthew Wilcox @ 2021-05-27 12:07 UTC (permalink / raw)
To: Muchun Song
Cc: akpm, hannes, mhocko, vdavydov.dev, shakeelb, guro, shy828301,
alexs, richard.weiyang, david, trond.myklebust, anna.schumaker,
linux-fsdevel, linux-kernel, linux-mm, linux-nfs, zhengqi.arch,
duanxiongchun, fam.zheng
On Thu, May 27, 2021 at 02:21:44PM +0800, Muchun Song wrote:
> If we run 10k containers in the system, the size of the
> list_lru_memcg->lrus can be ~96KB per list_lru. When we decrease the
> number containers, the size of the array will not be shrinked. It is
> not scalable. The xarray is a good choice for this case. We can save
> a lot of memory when there are tens of thousands continers in the
> system. If we use xarray, we also can remove the logic code of
> resizing array, which can simplify the code.
I am all for this, in concept. Some thoughts below ...
> @@ -56,10 +51,8 @@ struct list_lru {
> #ifdef CONFIG_MEMCG_KMEM
> struct list_head list;
> int shrinker_id;
> - /* protects ->memcg_lrus->lrus[i] */
> - spinlock_t lock;
> /* for cgroup aware lrus points to per cgroup lists, otherwise NULL */
> - struct list_lru_memcg __rcu *memcg_lrus;
> + struct xarray *xa;
> #endif
Normally, we embed an xarray in its containing structure instead of
allocating it. It's only a pointer, int and spinlock, so generally
16 bytes, as opposed to the 8 bytes for the pointer and a 16 byte
allocation. There is a minor wrinkle in that currently 'NULL' is
used to indicate "is not cgroup aware". Maybe there's another way
to indicate that?
> @@ -51,22 +51,12 @@ static int lru_shrinker_id(struct list_lru *lru)
> static inline struct list_lru_one *
> list_lru_from_memcg_idx(struct list_lru *lru, int nid, int idx)
> {
> - struct list_lru_memcg *memcg_lrus;
> - struct list_lru_node *nlru = &lru->node[nid];
> + if (list_lru_memcg_aware(lru) && idx >= 0) {
> + struct list_lru_per_memcg *mlru = xa_load(lru->xa, idx);
>
> - /*
> - * Either lock or RCU protects the array of per cgroup lists
> - * from relocation (see memcg_update_list_lru).
> - */
> - memcg_lrus = rcu_dereference_check(lru->memcg_lrus,
> - lockdep_is_held(&nlru->lock));
> - if (memcg_lrus && idx >= 0) {
> - struct list_lru_per_memcg *mlru;
> -
> - mlru = rcu_dereference_check(memcg_lrus->lrus[idx], true);
> return mlru ? &mlru->nodes[nid] : NULL;
> }
> - return &nlru->lru;
> + return &lru->node[nid].lru;
> }
... perhaps we move the xarray out from under the #ifdef and use index 0
for non-memcg-aware lrus? The XArray is specially optimised for arrays
which only have one entry at 0.
> int list_lru_memcg_alloc(struct list_lru *lru, struct mem_cgroup *memcg, gfp_t gfp)
> {
> + XA_STATE(xas, lru->xa, 0);
> unsigned long flags;
> - struct list_lru_memcg *memcg_lrus;
> - int i;
> + int i, ret = 0;
>
> struct list_lru_memcg_table {
> struct list_lru_per_memcg *mlru;
> @@ -601,22 +522,45 @@ int list_lru_memcg_alloc(struct list_lru *lru, struct mem_cgroup *memcg, gfp_t g
> }
> }
>
> - spin_lock_irqsave(&lru->lock, flags);
> - memcg_lrus = rcu_dereference_protected(lru->memcg_lrus, true);
> + xas_lock_irqsave(&xas, flags);
> while (i--) {
> int index = memcg_cache_id(table[i].memcg);
> struct list_lru_per_memcg *mlru = table[i].mlru;
>
> - if (index < 0 || rcu_dereference_protected(memcg_lrus->lrus[index], true))
> + xas_set(&xas, index);
> +retry:
> + if (unlikely(index < 0 || ret || xas_load(&xas))) {
> kfree(mlru);
> - else
> - rcu_assign_pointer(memcg_lrus->lrus[index], mlru);
> + } else {
> + ret = xa_err(xas_store(&xas, mlru));
This is mixing advanced and normal XArray concepts ... sorry to have
confused you. I think what you meant to do here was:
xas_store(&xas, mlru);
ret = xas_error(&xas);
Or you can avoid introducing 'ret' at all, and keep your errors in the
xa_state. You're kind of mirroring the xa_state errors into 'ret'
anyway, so that seems easier to understand?
> - memcg_id = memcg_alloc_cache_id();
> + memcg_id = ida_simple_get(&memcg_cache_ida, 0, MEMCG_CACHES_MAX_SIZE,
> + GFP_KERNEL);
memcg_id = ida_alloc_max(&memcg_cache_ida,
MEMCG_CACHES_MAX_SIZE - 1, GFP_KERNEL);
... although i think there's actually a fencepost error, and this really
should be MEMCG_CACHES_MAX_SIZE.
> objcg = obj_cgroup_alloc();
> if (!objcg) {
> - memcg_free_cache_id(memcg_id);
> + ida_simple_remove(&memcg_cache_ida, memcg_id);
ida_free(&memcg_cache_ida, memcg_id);
^ permalink raw reply [flat|nested] 29+ messages in thread* Re: [External] Re: [PATCH v2 17/21] mm: list_lru: replace linear array with xarray
2021-05-27 12:07 ` Matthew Wilcox
@ 2021-05-28 3:43 ` Muchun Song
0 siblings, 0 replies; 29+ messages in thread
From: Muchun Song @ 2021-05-28 3:43 UTC (permalink / raw)
To: Matthew Wilcox
Cc: Andrew Morton, Johannes Weiner, Michal Hocko, Vladimir Davydov,
Shakeel Butt, Roman Gushchin, Yang Shi, Alex Shi, Wei Yang,
Dave Chinner, trond.myklebust, anna.schumaker, linux-fsdevel,
LKML, Linux Memory Management List, linux-nfs, zhengqi.arch,
Xiongchun duan, fam.zheng
On Thu, May 27, 2021 at 8:08 PM Matthew Wilcox <willy@infradead.org> wrote:
>
> On Thu, May 27, 2021 at 02:21:44PM +0800, Muchun Song wrote:
> > If we run 10k containers in the system, the size of the
> > list_lru_memcg->lrus can be ~96KB per list_lru. When we decrease the
> > number containers, the size of the array will not be shrinked. It is
> > not scalable. The xarray is a good choice for this case. We can save
> > a lot of memory when there are tens of thousands continers in the
> > system. If we use xarray, we also can remove the logic code of
> > resizing array, which can simplify the code.
>
> I am all for this, in concept. Some thoughts below ...
>
> > @@ -56,10 +51,8 @@ struct list_lru {
> > #ifdef CONFIG_MEMCG_KMEM
> > struct list_head list;
> > int shrinker_id;
> > - /* protects ->memcg_lrus->lrus[i] */
> > - spinlock_t lock;
> > /* for cgroup aware lrus points to per cgroup lists, otherwise NULL */
> > - struct list_lru_memcg __rcu *memcg_lrus;
> > + struct xarray *xa;
> > #endif
>
> Normally, we embed an xarray in its containing structure instead of
> allocating it. It's only a pointer, int and spinlock, so generally
> 16 bytes, as opposed to the 8 bytes for the pointer and a 16 byte
> allocation. There is a minor wrinkle in that currently 'NULL' is
> used to indicate "is not cgroup aware". Maybe there's another way
> to indicate that?
Sure. I can drop patch 8 in this series. In that case, we can use
->memcg_aware to indicate that.
>
> > @@ -51,22 +51,12 @@ static int lru_shrinker_id(struct list_lru *lru)
> > static inline struct list_lru_one *
> > list_lru_from_memcg_idx(struct list_lru *lru, int nid, int idx)
> > {
> > - struct list_lru_memcg *memcg_lrus;
> > - struct list_lru_node *nlru = &lru->node[nid];
> > + if (list_lru_memcg_aware(lru) && idx >= 0) {
> > + struct list_lru_per_memcg *mlru = xa_load(lru->xa, idx);
> >
> > - /*
> > - * Either lock or RCU protects the array of per cgroup lists
> > - * from relocation (see memcg_update_list_lru).
> > - */
> > - memcg_lrus = rcu_dereference_check(lru->memcg_lrus,
> > - lockdep_is_held(&nlru->lock));
> > - if (memcg_lrus && idx >= 0) {
> > - struct list_lru_per_memcg *mlru;
> > -
> > - mlru = rcu_dereference_check(memcg_lrus->lrus[idx], true);
> > return mlru ? &mlru->nodes[nid] : NULL;
> > }
> > - return &nlru->lru;
> > + return &lru->node[nid].lru;
> > }
>
> ... perhaps we move the xarray out from under the #ifdef and use index 0
> for non-memcg-aware lrus? The XArray is specially optimised for arrays
> which only have one entry at 0.
Sounds like a good idea. I can do a try.
>
> > int list_lru_memcg_alloc(struct list_lru *lru, struct mem_cgroup *memcg, gfp_t gfp)
> > {
> > + XA_STATE(xas, lru->xa, 0);
> > unsigned long flags;
> > - struct list_lru_memcg *memcg_lrus;
> > - int i;
> > + int i, ret = 0;
> >
> > struct list_lru_memcg_table {
> > struct list_lru_per_memcg *mlru;
> > @@ -601,22 +522,45 @@ int list_lru_memcg_alloc(struct list_lru *lru, struct mem_cgroup *memcg, gfp_t g
> > }
> > }
> >
> > - spin_lock_irqsave(&lru->lock, flags);
> > - memcg_lrus = rcu_dereference_protected(lru->memcg_lrus, true);
> > + xas_lock_irqsave(&xas, flags);
> > while (i--) {
> > int index = memcg_cache_id(table[i].memcg);
> > struct list_lru_per_memcg *mlru = table[i].mlru;
> >
> > - if (index < 0 || rcu_dereference_protected(memcg_lrus->lrus[index], true))
> > + xas_set(&xas, index);
> > +retry:
> > + if (unlikely(index < 0 || ret || xas_load(&xas))) {
> > kfree(mlru);
> > - else
> > - rcu_assign_pointer(memcg_lrus->lrus[index], mlru);
> > + } else {
> > + ret = xa_err(xas_store(&xas, mlru));
>
> This is mixing advanced and normal XArray concepts ... sorry to have
> confused you. I think what you meant to do here was:
>
> xas_store(&xas, mlru);
> ret = xas_error(&xas);
Sure. Thanks for pointing it out. It's my bad usage.
>
> Or you can avoid introducing 'ret' at all, and keep your errors in the
> xa_state. You're kind of mirroring the xa_state errors into 'ret'
> anyway, so that seems easier to understand?
Make sense. I will do this in the next version. Thanks for your
all suggestions.
>
> > - memcg_id = memcg_alloc_cache_id();
> > + memcg_id = ida_simple_get(&memcg_cache_ida, 0, MEMCG_CACHES_MAX_SIZE,
> > + GFP_KERNEL);
>
> memcg_id = ida_alloc_max(&memcg_cache_ida,
> MEMCG_CACHES_MAX_SIZE - 1, GFP_KERNEL);
>
> ... although i think there's actually a fencepost error, and this really
> should be MEMCG_CACHES_MAX_SIZE.
Totally agree. I have fixed this issue in patch 19.
>
> > objcg = obj_cgroup_alloc();
> > if (!objcg) {
> > - memcg_free_cache_id(memcg_id);
> > + ida_simple_remove(&memcg_cache_ida, memcg_id);
>
> ida_free(&memcg_cache_ida, memcg_id);
I Will update to this new API.
>
^ permalink raw reply [flat|nested] 29+ messages in thread* Re: [External] Re: [PATCH v2 17/21] mm: list_lru: replace linear array with xarray
@ 2021-05-28 3:43 ` Muchun Song
0 siblings, 0 replies; 29+ messages in thread
From: Muchun Song @ 2021-05-28 3:43 UTC (permalink / raw)
To: Matthew Wilcox
Cc: Andrew Morton, Johannes Weiner, Michal Hocko, Vladimir Davydov,
Shakeel Butt, Roman Gushchin, Yang Shi, Alex Shi, Wei Yang,
Dave Chinner, trond.myklebust, anna.schumaker, linux-fsdevel,
LKML, Linux Memory Management List, linux-nfs, zhengqi.arch,
Xiongchun duan, fam.zheng
On Thu, May 27, 2021 at 8:08 PM Matthew Wilcox <willy@infradead.org> wrote:
>
> On Thu, May 27, 2021 at 02:21:44PM +0800, Muchun Song wrote:
> > If we run 10k containers in the system, the size of the
> > list_lru_memcg->lrus can be ~96KB per list_lru. When we decrease the
> > number containers, the size of the array will not be shrinked. It is
> > not scalable. The xarray is a good choice for this case. We can save
> > a lot of memory when there are tens of thousands continers in the
> > system. If we use xarray, we also can remove the logic code of
> > resizing array, which can simplify the code.
>
> I am all for this, in concept. Some thoughts below ...
>
> > @@ -56,10 +51,8 @@ struct list_lru {
> > #ifdef CONFIG_MEMCG_KMEM
> > struct list_head list;
> > int shrinker_id;
> > - /* protects ->memcg_lrus->lrus[i] */
> > - spinlock_t lock;
> > /* for cgroup aware lrus points to per cgroup lists, otherwise NULL */
> > - struct list_lru_memcg __rcu *memcg_lrus;
> > + struct xarray *xa;
> > #endif
>
> Normally, we embed an xarray in its containing structure instead of
> allocating it. It's only a pointer, int and spinlock, so generally
> 16 bytes, as opposed to the 8 bytes for the pointer and a 16 byte
> allocation. There is a minor wrinkle in that currently 'NULL' is
> used to indicate "is not cgroup aware". Maybe there's another way
> to indicate that?
Sure. I can drop patch 8 in this series. In that case, we can use
->memcg_aware to indicate that.
>
> > @@ -51,22 +51,12 @@ static int lru_shrinker_id(struct list_lru *lru)
> > static inline struct list_lru_one *
> > list_lru_from_memcg_idx(struct list_lru *lru, int nid, int idx)
> > {
> > - struct list_lru_memcg *memcg_lrus;
> > - struct list_lru_node *nlru = &lru->node[nid];
> > + if (list_lru_memcg_aware(lru) && idx >= 0) {
> > + struct list_lru_per_memcg *mlru = xa_load(lru->xa, idx);
> >
> > - /*
> > - * Either lock or RCU protects the array of per cgroup lists
> > - * from relocation (see memcg_update_list_lru).
> > - */
> > - memcg_lrus = rcu_dereference_check(lru->memcg_lrus,
> > - lockdep_is_held(&nlru->lock));
> > - if (memcg_lrus && idx >= 0) {
> > - struct list_lru_per_memcg *mlru;
> > -
> > - mlru = rcu_dereference_check(memcg_lrus->lrus[idx], true);
> > return mlru ? &mlru->nodes[nid] : NULL;
> > }
> > - return &nlru->lru;
> > + return &lru->node[nid].lru;
> > }
>
> ... perhaps we move the xarray out from under the #ifdef and use index 0
> for non-memcg-aware lrus? The XArray is specially optimised for arrays
> which only have one entry at 0.
Sounds like a good idea. I can do a try.
>
> > int list_lru_memcg_alloc(struct list_lru *lru, struct mem_cgroup *memcg, gfp_t gfp)
> > {
> > + XA_STATE(xas, lru->xa, 0);
> > unsigned long flags;
> > - struct list_lru_memcg *memcg_lrus;
> > - int i;
> > + int i, ret = 0;
> >
> > struct list_lru_memcg_table {
> > struct list_lru_per_memcg *mlru;
> > @@ -601,22 +522,45 @@ int list_lru_memcg_alloc(struct list_lru *lru, struct mem_cgroup *memcg, gfp_t g
> > }
> > }
> >
> > - spin_lock_irqsave(&lru->lock, flags);
> > - memcg_lrus = rcu_dereference_protected(lru->memcg_lrus, true);
> > + xas_lock_irqsave(&xas, flags);
> > while (i--) {
> > int index = memcg_cache_id(table[i].memcg);
> > struct list_lru_per_memcg *mlru = table[i].mlru;
> >
> > - if (index < 0 || rcu_dereference_protected(memcg_lrus->lrus[index], true))
> > + xas_set(&xas, index);
> > +retry:
> > + if (unlikely(index < 0 || ret || xas_load(&xas))) {
> > kfree(mlru);
> > - else
> > - rcu_assign_pointer(memcg_lrus->lrus[index], mlru);
> > + } else {
> > + ret = xa_err(xas_store(&xas, mlru));
>
> This is mixing advanced and normal XArray concepts ... sorry to have
> confused you. I think what you meant to do here was:
>
> xas_store(&xas, mlru);
> ret = xas_error(&xas);
Sure. Thanks for pointing it out. It's my bad usage.
>
> Or you can avoid introducing 'ret' at all, and keep your errors in the
> xa_state. You're kind of mirroring the xa_state errors into 'ret'
> anyway, so that seems easier to understand?
Make sense. I will do this in the next version. Thanks for your
all suggestions.
>
> > - memcg_id = memcg_alloc_cache_id();
> > + memcg_id = ida_simple_get(&memcg_cache_ida, 0, MEMCG_CACHES_MAX_SIZE,
> > + GFP_KERNEL);
>
> memcg_id = ida_alloc_max(&memcg_cache_ida,
> MEMCG_CACHES_MAX_SIZE - 1, GFP_KERNEL);
>
> ... although i think there's actually a fencepost error, and this really
> should be MEMCG_CACHES_MAX_SIZE.
Totally agree. I have fixed this issue in patch 19.
>
> > objcg = obj_cgroup_alloc();
> > if (!objcg) {
> > - memcg_free_cache_id(memcg_id);
> > + ida_simple_remove(&memcg_cache_ida, memcg_id);
>
> ida_free(&memcg_cache_ida, memcg_id);
I Will update to this new API.
>
^ permalink raw reply [flat|nested] 29+ messages in thread* Re: [External] Re: [PATCH v2 17/21] mm: list_lru: replace linear array with xarray
2021-05-28 3:43 ` Muchun Song
@ 2021-05-28 8:04 ` Muchun Song
-1 siblings, 0 replies; 29+ messages in thread
From: Muchun Song @ 2021-05-28 8:04 UTC (permalink / raw)
To: Matthew Wilcox
Cc: Andrew Morton, Johannes Weiner, Michal Hocko, Vladimir Davydov,
Shakeel Butt, Roman Gushchin, Yang Shi, Alex Shi, Wei Yang,
Dave Chinner, trond.myklebust, anna.schumaker, linux-fsdevel,
LKML, Linux Memory Management List, linux-nfs, zhengqi.arch,
Xiongchun duan, fam.zheng
On Fri, May 28, 2021 at 11:43 AM Muchun Song <songmuchun@bytedance.com> wrote:
>
> On Thu, May 27, 2021 at 8:08 PM Matthew Wilcox <willy@infradead.org> wrote:
> >
> > On Thu, May 27, 2021 at 02:21:44PM +0800, Muchun Song wrote:
> > > If we run 10k containers in the system, the size of the
> > > list_lru_memcg->lrus can be ~96KB per list_lru. When we decrease the
> > > number containers, the size of the array will not be shrinked. It is
> > > not scalable. The xarray is a good choice for this case. We can save
> > > a lot of memory when there are tens of thousands continers in the
> > > system. If we use xarray, we also can remove the logic code of
> > > resizing array, which can simplify the code.
> >
> > I am all for this, in concept. Some thoughts below ...
> >
> > > @@ -56,10 +51,8 @@ struct list_lru {
> > > #ifdef CONFIG_MEMCG_KMEM
> > > struct list_head list;
> > > int shrinker_id;
> > > - /* protects ->memcg_lrus->lrus[i] */
> > > - spinlock_t lock;
> > > /* for cgroup aware lrus points to per cgroup lists, otherwise NULL */
> > > - struct list_lru_memcg __rcu *memcg_lrus;
> > > + struct xarray *xa;
> > > #endif
> >
> > Normally, we embed an xarray in its containing structure instead of
> > allocating it. It's only a pointer, int and spinlock, so generally
> > 16 bytes, as opposed to the 8 bytes for the pointer and a 16 byte
> > allocation. There is a minor wrinkle in that currently 'NULL' is
> > used to indicate "is not cgroup aware". Maybe there's another way
> > to indicate that?
>
> Sure. I can drop patch 8 in this series. In that case, we can use
> ->memcg_aware to indicate that.
>
>
> >
> > > @@ -51,22 +51,12 @@ static int lru_shrinker_id(struct list_lru *lru)
> > > static inline struct list_lru_one *
> > > list_lru_from_memcg_idx(struct list_lru *lru, int nid, int idx)
> > > {
> > > - struct list_lru_memcg *memcg_lrus;
> > > - struct list_lru_node *nlru = &lru->node[nid];
> > > + if (list_lru_memcg_aware(lru) && idx >= 0) {
> > > + struct list_lru_per_memcg *mlru = xa_load(lru->xa, idx);
> > >
> > > - /*
> > > - * Either lock or RCU protects the array of per cgroup lists
> > > - * from relocation (see memcg_update_list_lru).
> > > - */
> > > - memcg_lrus = rcu_dereference_check(lru->memcg_lrus,
> > > - lockdep_is_held(&nlru->lock));
> > > - if (memcg_lrus && idx >= 0) {
> > > - struct list_lru_per_memcg *mlru;
> > > -
> > > - mlru = rcu_dereference_check(memcg_lrus->lrus[idx], true);
> > > return mlru ? &mlru->nodes[nid] : NULL;
> > > }
> > > - return &nlru->lru;
> > > + return &lru->node[nid].lru;
> > > }
> >
> > ... perhaps we move the xarray out from under the #ifdef and use index 0
> > for non-memcg-aware lrus? The XArray is specially optimised for arrays
> > which only have one entry at 0.
>
> Sounds like a good idea. I can do a try.
I have thought more about this. If we do this, we need to allocate a
list_lru_per_memcg structure for the root memcg. Since the structure
of list_lru_node already aligns with cache line size. From this point
of view, this just wastes memory. We do not gain anything. Right?
Another approach is introducing a new structure of list_lru_nodes,
which is described as following.
struct list_lru_nodes {
struct list_lru_node nodes[0];
};
Then we insert struct list_lru_nodes to the XArray (index == 0).
There will be two different types in the XArray. If index == 0,
the xa_load() returns list_lru_nodes pointer, otherwise, it returns
list_lru_per_memcg pointer. So list_lru_from_memcg_idx() still
need to handle different cases.
It looks like both approaches do not have any obvious
advantages.
What do you think about this, Mattew?
>
> >
> > > int list_lru_memcg_alloc(struct list_lru *lru, struct mem_cgroup *memcg, gfp_t gfp)
> > > {
> > > + XA_STATE(xas, lru->xa, 0);
> > > unsigned long flags;
> > > - struct list_lru_memcg *memcg_lrus;
> > > - int i;
> > > + int i, ret = 0;
> > >
> > > struct list_lru_memcg_table {
> > > struct list_lru_per_memcg *mlru;
> > > @@ -601,22 +522,45 @@ int list_lru_memcg_alloc(struct list_lru *lru, struct mem_cgroup *memcg, gfp_t g
> > > }
> > > }
> > >
> > > - spin_lock_irqsave(&lru->lock, flags);
> > > - memcg_lrus = rcu_dereference_protected(lru->memcg_lrus, true);
> > > + xas_lock_irqsave(&xas, flags);
> > > while (i--) {
> > > int index = memcg_cache_id(table[i].memcg);
> > > struct list_lru_per_memcg *mlru = table[i].mlru;
> > >
> > > - if (index < 0 || rcu_dereference_protected(memcg_lrus->lrus[index], true))
> > > + xas_set(&xas, index);
> > > +retry:
> > > + if (unlikely(index < 0 || ret || xas_load(&xas))) {
> > > kfree(mlru);
> > > - else
> > > - rcu_assign_pointer(memcg_lrus->lrus[index], mlru);
> > > + } else {
> > > + ret = xa_err(xas_store(&xas, mlru));
> >
> > This is mixing advanced and normal XArray concepts ... sorry to have
> > confused you. I think what you meant to do here was:
> >
> > xas_store(&xas, mlru);
> > ret = xas_error(&xas);
>
> Sure. Thanks for pointing it out. It's my bad usage.
>
> >
> > Or you can avoid introducing 'ret' at all, and keep your errors in the
> > xa_state. You're kind of mirroring the xa_state errors into 'ret'
> > anyway, so that seems easier to understand?
>
> Make sense. I will do this in the next version. Thanks for your
> all suggestions.
>
> >
> > > - memcg_id = memcg_alloc_cache_id();
> > > + memcg_id = ida_simple_get(&memcg_cache_ida, 0, MEMCG_CACHES_MAX_SIZE,
> > > + GFP_KERNEL);
> >
> > memcg_id = ida_alloc_max(&memcg_cache_ida,
> > MEMCG_CACHES_MAX_SIZE - 1, GFP_KERNEL);
> >
> > ... although i think there's actually a fencepost error, and this really
> > should be MEMCG_CACHES_MAX_SIZE.
>
> Totally agree. I have fixed this issue in patch 19.
>
> >
> > > objcg = obj_cgroup_alloc();
> > > if (!objcg) {
> > > - memcg_free_cache_id(memcg_id);
> > > + ida_simple_remove(&memcg_cache_ida, memcg_id);
> >
> > ida_free(&memcg_cache_ida, memcg_id);
>
> I Will update to this new API.
>
> >
^ permalink raw reply [flat|nested] 29+ messages in thread* Re: [External] Re: [PATCH v2 17/21] mm: list_lru: replace linear array with xarray
@ 2021-05-28 8:04 ` Muchun Song
0 siblings, 0 replies; 29+ messages in thread
From: Muchun Song @ 2021-05-28 8:04 UTC (permalink / raw)
To: Matthew Wilcox
Cc: Andrew Morton, Johannes Weiner, Michal Hocko, Vladimir Davydov,
Shakeel Butt, Roman Gushchin, Yang Shi, Alex Shi, Wei Yang,
Dave Chinner, trond.myklebust, anna.schumaker, linux-fsdevel,
LKML, Linux Memory Management List, linux-nfs, zhengqi.arch,
Xiongchun duan, fam.zheng
On Fri, May 28, 2021 at 11:43 AM Muchun Song <songmuchun@bytedance.com> wrote:
>
> On Thu, May 27, 2021 at 8:08 PM Matthew Wilcox <willy@infradead.org> wrote:
> >
> > On Thu, May 27, 2021 at 02:21:44PM +0800, Muchun Song wrote:
> > > If we run 10k containers in the system, the size of the
> > > list_lru_memcg->lrus can be ~96KB per list_lru. When we decrease the
> > > number containers, the size of the array will not be shrinked. It is
> > > not scalable. The xarray is a good choice for this case. We can save
> > > a lot of memory when there are tens of thousands continers in the
> > > system. If we use xarray, we also can remove the logic code of
> > > resizing array, which can simplify the code.
> >
> > I am all for this, in concept. Some thoughts below ...
> >
> > > @@ -56,10 +51,8 @@ struct list_lru {
> > > #ifdef CONFIG_MEMCG_KMEM
> > > struct list_head list;
> > > int shrinker_id;
> > > - /* protects ->memcg_lrus->lrus[i] */
> > > - spinlock_t lock;
> > > /* for cgroup aware lrus points to per cgroup lists, otherwise NULL */
> > > - struct list_lru_memcg __rcu *memcg_lrus;
> > > + struct xarray *xa;
> > > #endif
> >
> > Normally, we embed an xarray in its containing structure instead of
> > allocating it. It's only a pointer, int and spinlock, so generally
> > 16 bytes, as opposed to the 8 bytes for the pointer and a 16 byte
> > allocation. There is a minor wrinkle in that currently 'NULL' is
> > used to indicate "is not cgroup aware". Maybe there's another way
> > to indicate that?
>
> Sure. I can drop patch 8 in this series. In that case, we can use
> ->memcg_aware to indicate that.
>
>
> >
> > > @@ -51,22 +51,12 @@ static int lru_shrinker_id(struct list_lru *lru)
> > > static inline struct list_lru_one *
> > > list_lru_from_memcg_idx(struct list_lru *lru, int nid, int idx)
> > > {
> > > - struct list_lru_memcg *memcg_lrus;
> > > - struct list_lru_node *nlru = &lru->node[nid];
> > > + if (list_lru_memcg_aware(lru) && idx >= 0) {
> > > + struct list_lru_per_memcg *mlru = xa_load(lru->xa, idx);
> > >
> > > - /*
> > > - * Either lock or RCU protects the array of per cgroup lists
> > > - * from relocation (see memcg_update_list_lru).
> > > - */
> > > - memcg_lrus = rcu_dereference_check(lru->memcg_lrus,
> > > - lockdep_is_held(&nlru->lock));
> > > - if (memcg_lrus && idx >= 0) {
> > > - struct list_lru_per_memcg *mlru;
> > > -
> > > - mlru = rcu_dereference_check(memcg_lrus->lrus[idx], true);
> > > return mlru ? &mlru->nodes[nid] : NULL;
> > > }
> > > - return &nlru->lru;
> > > + return &lru->node[nid].lru;
> > > }
> >
> > ... perhaps we move the xarray out from under the #ifdef and use index 0
> > for non-memcg-aware lrus? The XArray is specially optimised for arrays
> > which only have one entry at 0.
>
> Sounds like a good idea. I can do a try.
I have thought more about this. If we do this, we need to allocate a
list_lru_per_memcg structure for the root memcg. Since the structure
of list_lru_node already aligns with cache line size. From this point
of view, this just wastes memory. We do not gain anything. Right?
Another approach is introducing a new structure of list_lru_nodes,
which is described as following.
struct list_lru_nodes {
struct list_lru_node nodes[0];
};
Then we insert struct list_lru_nodes to the XArray (index == 0).
There will be two different types in the XArray. If index == 0,
the xa_load() returns list_lru_nodes pointer, otherwise, it returns
list_lru_per_memcg pointer. So list_lru_from_memcg_idx() still
need to handle different cases.
It looks like both approaches do not have any obvious
advantages.
What do you think about this, Mattew?
>
> >
> > > int list_lru_memcg_alloc(struct list_lru *lru, struct mem_cgroup *memcg, gfp_t gfp)
> > > {
> > > + XA_STATE(xas, lru->xa, 0);
> > > unsigned long flags;
> > > - struct list_lru_memcg *memcg_lrus;
> > > - int i;
> > > + int i, ret = 0;
> > >
> > > struct list_lru_memcg_table {
> > > struct list_lru_per_memcg *mlru;
> > > @@ -601,22 +522,45 @@ int list_lru_memcg_alloc(struct list_lru *lru, struct mem_cgroup *memcg, gfp_t g
> > > }
> > > }
> > >
> > > - spin_lock_irqsave(&lru->lock, flags);
> > > - memcg_lrus = rcu_dereference_protected(lru->memcg_lrus, true);
> > > + xas_lock_irqsave(&xas, flags);
> > > while (i--) {
> > > int index = memcg_cache_id(table[i].memcg);
> > > struct list_lru_per_memcg *mlru = table[i].mlru;
> > >
> > > - if (index < 0 || rcu_dereference_protected(memcg_lrus->lrus[index], true))
> > > + xas_set(&xas, index);
> > > +retry:
> > > + if (unlikely(index < 0 || ret || xas_load(&xas))) {
> > > kfree(mlru);
> > > - else
> > > - rcu_assign_pointer(memcg_lrus->lrus[index], mlru);
> > > + } else {
> > > + ret = xa_err(xas_store(&xas, mlru));
> >
> > This is mixing advanced and normal XArray concepts ... sorry to have
> > confused you. I think what you meant to do here was:
> >
> > xas_store(&xas, mlru);
> > ret = xas_error(&xas);
>
> Sure. Thanks for pointing it out. It's my bad usage.
>
> >
> > Or you can avoid introducing 'ret' at all, and keep your errors in the
> > xa_state. You're kind of mirroring the xa_state errors into 'ret'
> > anyway, so that seems easier to understand?
>
> Make sense. I will do this in the next version. Thanks for your
> all suggestions.
>
> >
> > > - memcg_id = memcg_alloc_cache_id();
> > > + memcg_id = ida_simple_get(&memcg_cache_ida, 0, MEMCG_CACHES_MAX_SIZE,
> > > + GFP_KERNEL);
> >
> > memcg_id = ida_alloc_max(&memcg_cache_ida,
> > MEMCG_CACHES_MAX_SIZE - 1, GFP_KERNEL);
> >
> > ... although i think there's actually a fencepost error, and this really
> > should be MEMCG_CACHES_MAX_SIZE.
>
> Totally agree. I have fixed this issue in patch 19.
>
> >
> > > objcg = obj_cgroup_alloc();
> > > if (!objcg) {
> > > - memcg_free_cache_id(memcg_id);
> > > + ida_simple_remove(&memcg_cache_ida, memcg_id);
> >
> > ida_free(&memcg_cache_ida, memcg_id);
>
> I Will update to this new API.
>
> >
^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH v2 18/21] mm: memcontrol: reuse memory cgroup ID for kmem ID
2021-05-27 6:21 [PATCH v2 00/21] Optimize list lru memory consumption Muchun Song
` (16 preceding siblings ...)
2021-05-27 6:21 ` [PATCH v2 17/21] mm: list_lru: replace linear array with xarray Muchun Song
@ 2021-05-27 6:21 ` Muchun Song
2021-05-27 6:21 ` [PATCH v2 19/21] mm: memcontrol: fix cannot alloc the maximum memcg ID Muchun Song
` (3 subsequent siblings)
21 siblings, 0 replies; 29+ messages in thread
From: Muchun Song @ 2021-05-27 6:21 UTC (permalink / raw)
To: willy, akpm, hannes, mhocko, vdavydov.dev, shakeelb, guro,
shy828301, alexs, richard.weiyang, david, trond.myklebust,
anna.schumaker
Cc: linux-fsdevel, linux-kernel, linux-mm, linux-nfs, zhengqi.arch,
duanxiongchun, fam.zheng, Muchun Song
There are two idrs being used by memory cgroup, one is for kmem ID,
another is for memory cgroup ID. The maximum ID of both is 64Ki.
Both of them can limit the total number of memory cgroups. Actually,
we can reuse memory cgroup ID for kmem ID to simplify the code.
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
---
mm/memcontrol.c | 40 +++-------------------------------------
1 file changed, 3 insertions(+), 37 deletions(-)
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 82b147597138..ae3ad1001824 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -342,23 +342,6 @@ static void memcg_reparent_objcgs(struct mem_cgroup *memcg,
}
/*
- * This will be used as a shrinker list's index.
- * The main reason for not using cgroup id for this:
- * this works better in sparse environments, where we have a lot of memcgs,
- * but only a few kmem-limited.
- */
-static DEFINE_IDA(memcg_cache_ida);
-
-/*
- * MAX_SIZE should be as large as the number of cgrp_ids. Ideally, we could get
- * this constant directly from cgroup, but it is understandable that this is
- * better kept as an internal representation in cgroup.c. In any case, the
- * cgrp_id space is not getting any smaller, and we don't have to necessarily
- * increase ours as well if it increases.
- */
-#define MEMCG_CACHES_MAX_SIZE MEM_CGROUP_ID_MAX
-
-/*
* A lot of the calls to the cache allocation functions are expected to be
* inlined by the compiler. Since the calls to memcg_slab_pre_alloc_hook() are
* conditional to this static branch, we'll have to allow modules that does
@@ -3542,7 +3525,6 @@ static u64 mem_cgroup_read_u64(struct cgroup_subsys_state *css,
static int memcg_online_kmem(struct mem_cgroup *memcg)
{
struct obj_cgroup *objcg;
- int memcg_id;
if (cgroup_memory_nokmem)
return 0;
@@ -3550,22 +3532,16 @@ static int memcg_online_kmem(struct mem_cgroup *memcg)
if (unlikely(mem_cgroup_is_root(memcg)))
return 0;
- memcg_id = ida_simple_get(&memcg_cache_ida, 0, MEMCG_CACHES_MAX_SIZE,
- GFP_KERNEL);
- if (memcg_id < 0)
- return memcg_id;
-
objcg = obj_cgroup_alloc();
- if (!objcg) {
- ida_simple_remove(&memcg_cache_ida, memcg_id);
+ if (!objcg)
return -ENOMEM;
- }
+
objcg->memcg = memcg;
rcu_assign_pointer(memcg->objcg, objcg);
static_branch_enable(&memcg_kmem_enabled_key);
- memcg->kmemcg_id = memcg_id;
+ memcg->kmemcg_id = memcg->id.id;
return 0;
}
@@ -3573,7 +3549,6 @@ static int memcg_online_kmem(struct mem_cgroup *memcg)
static void memcg_offline_kmem(struct mem_cgroup *memcg)
{
struct mem_cgroup *parent;
- int kmemcg_id;
if (cgroup_memory_nokmem)
return;
@@ -3586,16 +3561,7 @@ static void memcg_offline_kmem(struct mem_cgroup *memcg)
parent = root_mem_cgroup;
memcg_reparent_objcgs(memcg, parent);
-
- /*
- * memcg_reparent_list_lrus() can change memcg->kmemcg_id.
- * Cache it to @kmemcg_id.
- */
- kmemcg_id = memcg->kmemcg_id;
-
memcg_reparent_list_lrus(memcg, parent);
-
- ida_simple_remove(&memcg_cache_ida, kmemcg_id);
}
#else
static int memcg_online_kmem(struct mem_cgroup *memcg)
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v2 19/21] mm: memcontrol: fix cannot alloc the maximum memcg ID
2021-05-27 6:21 [PATCH v2 00/21] Optimize list lru memory consumption Muchun Song
` (17 preceding siblings ...)
2021-05-27 6:21 ` [PATCH v2 18/21] mm: memcontrol: reuse memory cgroup ID for kmem ID Muchun Song
@ 2021-05-27 6:21 ` Muchun Song
2021-05-27 6:21 ` [PATCH v2 20/21] mm: list_lru: rename list_lru_per_memcg to list_lru_memcg Muchun Song
` (2 subsequent siblings)
21 siblings, 0 replies; 29+ messages in thread
From: Muchun Song @ 2021-05-27 6:21 UTC (permalink / raw)
To: willy, akpm, hannes, mhocko, vdavydov.dev, shakeelb, guro,
shy828301, alexs, richard.weiyang, david, trond.myklebust,
anna.schumaker
Cc: linux-fsdevel, linux-kernel, linux-mm, linux-nfs, zhengqi.arch,
duanxiongchun, fam.zheng, Muchun Song
The idr_alloc() does not include @max ID. So in the current implementation,
the maximum memcg ID is 65534 instead of 65535. It seems a bug. So fix this.
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
---
mm/memcontrol.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index ae3ad1001824..a1c8ec858593 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -5044,7 +5044,7 @@ static struct mem_cgroup *mem_cgroup_alloc(void)
return ERR_PTR(error);
memcg->id.id = idr_alloc(&mem_cgroup_idr, NULL,
- 1, MEM_CGROUP_ID_MAX,
+ 1, MEM_CGROUP_ID_MAX + 1,
GFP_KERNEL);
if (memcg->id.id < 0) {
error = memcg->id.id;
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v2 20/21] mm: list_lru: rename list_lru_per_memcg to list_lru_memcg
2021-05-27 6:21 [PATCH v2 00/21] Optimize list lru memory consumption Muchun Song
` (18 preceding siblings ...)
2021-05-27 6:21 ` [PATCH v2 19/21] mm: memcontrol: fix cannot alloc the maximum memcg ID Muchun Song
@ 2021-05-27 6:21 ` Muchun Song
2021-05-27 6:21 ` [PATCH v2 21/21] mm: memcontrol: rename memcg_cache_id to memcg_kmem_id Muchun Song
2021-06-16 10:53 ` Muchun Song
21 siblings, 0 replies; 29+ messages in thread
From: Muchun Song @ 2021-05-27 6:21 UTC (permalink / raw)
To: willy, akpm, hannes, mhocko, vdavydov.dev, shakeelb, guro,
shy828301, alexs, richard.weiyang, david, trond.myklebust,
anna.schumaker
Cc: linux-fsdevel, linux-kernel, linux-mm, linux-nfs, zhengqi.arch,
duanxiongchun, fam.zheng, Muchun Song
Before now, the name of list_lru_memcg was occupied. Since previous
patch, the name is free. So rename list_lru_per_memcg to list_lru_memcg,
it is more brief.
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
---
include/linux/list_lru.h | 2 +-
mm/list_lru.c | 20 ++++++++++----------
2 files changed, 11 insertions(+), 11 deletions(-)
diff --git a/include/linux/list_lru.h b/include/linux/list_lru.h
index d7c9bd29e836..7497719ec71c 100644
--- a/include/linux/list_lru.h
+++ b/include/linux/list_lru.h
@@ -32,7 +32,7 @@ struct list_lru_one {
long nr_items;
};
-struct list_lru_per_memcg {
+struct list_lru_memcg {
struct rcu_head rcu;
/* array of per cgroup per node lists, indexed by node id */
struct list_lru_one nodes[];
diff --git a/mm/list_lru.c b/mm/list_lru.c
index 37052864bf78..77efdd0c8b24 100644
--- a/mm/list_lru.c
+++ b/mm/list_lru.c
@@ -52,7 +52,7 @@ static inline struct list_lru_one *
list_lru_from_memcg_idx(struct list_lru *lru, int nid, int idx)
{
if (list_lru_memcg_aware(lru) && idx >= 0) {
- struct list_lru_per_memcg *mlru = xa_load(lru->xa, idx);
+ struct list_lru_memcg *mlru = xa_load(lru->xa, idx);
return mlru ? &mlru->nodes[nid] : NULL;
}
@@ -304,7 +304,7 @@ unsigned long list_lru_walk_node(struct list_lru *lru, int nid,
isolated += list_lru_walk_one(lru, nid, NULL, isolate, cb_arg,
nr_to_walk);
if (*nr_to_walk > 0 && list_lru_memcg_aware(lru)) {
- struct list_lru_per_memcg *mlru;
+ struct list_lru_memcg *mlru;
unsigned long index;
xa_for_each(lru->xa, index, mlru) {
@@ -331,10 +331,10 @@ static void init_one_lru(struct list_lru_one *l)
}
#ifdef CONFIG_MEMCG_KMEM
-static struct list_lru_per_memcg *memcg_list_lru_alloc(gfp_t gfp)
+static struct list_lru_memcg *memcg_list_lru_alloc(gfp_t gfp)
{
int nid;
- struct list_lru_per_memcg *lru;
+ struct list_lru_memcg *lru;
lru = kmalloc(struct_size(lru, nodes, nr_node_ids), gfp);
if (!lru)
@@ -348,7 +348,7 @@ static struct list_lru_per_memcg *memcg_list_lru_alloc(gfp_t gfp)
static void memcg_list_lru_free(struct list_lru *lru, int src_idx)
{
- struct list_lru_per_memcg *mlru = xa_erase_irq(lru->xa, src_idx);
+ struct list_lru_memcg *mlru = xa_erase_irq(lru->xa, src_idx);
/*
* The __list_lru_walk_one() can walk the list of this node.
@@ -378,7 +378,7 @@ static int memcg_init_list_lru(struct list_lru *lru, bool memcg_aware)
static void memcg_destroy_list_lru(struct list_lru *lru)
{
XA_STATE(xas, lru->xa, 0);
- struct list_lru_per_memcg *mlru;
+ struct list_lru_memcg *mlru;
if (!list_lru_memcg_aware(lru))
return;
@@ -483,7 +483,7 @@ int list_lru_memcg_alloc(struct list_lru *lru, struct mem_cgroup *memcg, gfp_t g
int i, ret = 0;
struct list_lru_memcg_table {
- struct list_lru_per_memcg *mlru;
+ struct list_lru_memcg *mlru;
struct mem_cgroup *memcg;
} *table;
@@ -494,7 +494,7 @@ int list_lru_memcg_alloc(struct list_lru *lru, struct mem_cgroup *memcg, gfp_t g
return 0;
/*
- * The allocated list_lru_per_memcg array is not accounted directly.
+ * The allocated list_lru_memcg array is not accounted directly.
* Moreover, it should not come from DMA buffer and is not readily
* reclaimable. So those GFP bits should be masked off.
*/
@@ -506,7 +506,7 @@ int list_lru_memcg_alloc(struct list_lru *lru, struct mem_cgroup *memcg, gfp_t g
/*
* Because the list_lru can be reparented to the parent cgroup's
* list_lru, we should make sure that this cgroup and all its
- * ancestors have allocated list_lru_per_memcg.
+ * ancestors have allocated list_lru_memcg.
*/
for (i = 0; memcg; memcg = parent_mem_cgroup(memcg), i++) {
if (memcg_list_lru_skip_alloc(lru, memcg))
@@ -525,7 +525,7 @@ int list_lru_memcg_alloc(struct list_lru *lru, struct mem_cgroup *memcg, gfp_t g
xas_lock_irqsave(&xas, flags);
while (i--) {
int index = memcg_cache_id(table[i].memcg);
- struct list_lru_per_memcg *mlru = table[i].mlru;
+ struct list_lru_memcg *mlru = table[i].mlru;
xas_set(&xas, index);
retry:
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* [PATCH v2 21/21] mm: memcontrol: rename memcg_cache_id to memcg_kmem_id
2021-05-27 6:21 [PATCH v2 00/21] Optimize list lru memory consumption Muchun Song
` (19 preceding siblings ...)
2021-05-27 6:21 ` [PATCH v2 20/21] mm: list_lru: rename list_lru_per_memcg to list_lru_memcg Muchun Song
@ 2021-05-27 6:21 ` Muchun Song
2021-06-16 10:53 ` Muchun Song
21 siblings, 0 replies; 29+ messages in thread
From: Muchun Song @ 2021-05-27 6:21 UTC (permalink / raw)
To: willy, akpm, hannes, mhocko, vdavydov.dev, shakeelb, guro,
shy828301, alexs, richard.weiyang, david, trond.myklebust,
anna.schumaker
Cc: linux-fsdevel, linux-kernel, linux-mm, linux-nfs, zhengqi.arch,
duanxiongchun, fam.zheng, Muchun Song
The memcg_cache_id is introduced by commit 2633d7a02823 ("slab/slub:
consider a memcg parameter in kmem_create_cache"). It is used to index
in the kmem_cache->memcg_params->memcg_caches array. Since
kmem_cache->memcg_params.memcg_caches has been removed by commit
9855609bde03 ("mm: memcg/slab: use a single set of kmem_caches for
all accounted allocations"). So the name does not need to reflect cache
related. Just rename it to memcg_kmem_id. And it can reflect kmem
related.
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
---
include/linux/memcontrol.h | 4 ++--
mm/list_lru.c | 14 +++++++-------
2 files changed, 9 insertions(+), 9 deletions(-)
diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index e129c7067f63..637d5854fdda 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -1654,7 +1654,7 @@ static inline void memcg_kmem_uncharge_page(struct page *page, int order)
* A helper for accessing memcg's kmem_id, used for getting
* corresponding LRU lists.
*/
-static inline int memcg_cache_id(struct mem_cgroup *memcg)
+static inline int memcg_kmem_id(struct mem_cgroup *memcg)
{
return memcg ? memcg->kmemcg_id : -1;
}
@@ -1688,7 +1688,7 @@ static inline bool memcg_kmem_enabled(void)
return false;
}
-static inline int memcg_cache_id(struct mem_cgroup *memcg)
+static inline int memcg_kmem_id(struct mem_cgroup *memcg)
{
return -1;
}
diff --git a/mm/list_lru.c b/mm/list_lru.c
index 77efdd0c8b24..1ee2b28ded7d 100644
--- a/mm/list_lru.c
+++ b/mm/list_lru.c
@@ -74,7 +74,7 @@ list_lru_from_kmem(struct list_lru *lru, int nid, void *ptr,
if (!memcg)
goto out;
- l = list_lru_from_memcg_idx(lru, nid, memcg_cache_id(memcg));
+ l = list_lru_from_memcg_idx(lru, nid, memcg_kmem_id(memcg));
out:
if (memcg_ptr)
*memcg_ptr = memcg;
@@ -181,7 +181,7 @@ unsigned long list_lru_count_one(struct list_lru *lru,
long count = 0;
rcu_read_lock();
- l = list_lru_from_memcg_idx(lru, nid, memcg_cache_id(memcg));
+ l = list_lru_from_memcg_idx(lru, nid, memcg_kmem_id(memcg));
if (l)
count = READ_ONCE(l->nr_items);
rcu_read_unlock();
@@ -273,7 +273,7 @@ list_lru_walk_one(struct list_lru *lru, int nid, struct mem_cgroup *memcg,
unsigned long ret;
spin_lock(&nlru->lock);
- ret = __list_lru_walk_one(lru, nid, memcg_cache_id(memcg), isolate,
+ ret = __list_lru_walk_one(lru, nid, memcg_kmem_id(memcg), isolate,
cb_arg, nr_to_walk);
spin_unlock(&nlru->lock);
return ret;
@@ -289,7 +289,7 @@ list_lru_walk_one_irq(struct list_lru *lru, int nid, struct mem_cgroup *memcg,
unsigned long ret;
spin_lock_irq(&nlru->lock);
- ret = __list_lru_walk_one(lru, nid, memcg_cache_id(memcg), isolate,
+ ret = __list_lru_walk_one(lru, nid, memcg_kmem_id(memcg), isolate,
cb_arg, nr_to_walk);
spin_unlock_irq(&nlru->lock);
return ret;
@@ -469,7 +469,7 @@ void memcg_reparent_list_lrus(struct mem_cgroup *memcg, struct mem_cgroup *paren
static bool memcg_list_lru_skip_alloc(struct list_lru *lru,
struct mem_cgroup *memcg)
{
- int idx = memcg_cache_id(memcg);
+ int idx = memcg_kmem_id(memcg);
if (unlikely(idx < 0) || xa_load(lru->xa, idx))
return true;
@@ -524,7 +524,7 @@ int list_lru_memcg_alloc(struct list_lru *lru, struct mem_cgroup *memcg, gfp_t g
xas_lock_irqsave(&xas, flags);
while (i--) {
- int index = memcg_cache_id(table[i].memcg);
+ int index = memcg_kmem_id(table[i].memcg);
struct list_lru_memcg *mlru = table[i].mlru;
xas_set(&xas, index);
@@ -544,7 +544,7 @@ int list_lru_memcg_alloc(struct list_lru *lru, struct mem_cgroup *memcg, gfp_t g
* memcg id. More details see the comments
* in memcg_reparent_list_lrus().
*/
- index = memcg_cache_id(table[i].memcg);
+ index = memcg_kmem_id(table[i].memcg);
if (index < 0)
ret = 0;
else if (!ret && index != xas.xa_index)
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread* Re: [PATCH v2 00/21] Optimize list lru memory consumption
2021-05-27 6:21 [PATCH v2 00/21] Optimize list lru memory consumption Muchun Song
@ 2021-06-16 10:53 ` Muchun Song
2021-05-27 6:21 ` [PATCH v2 02/21] mm: memcontrol: remove kmemcg_id reparenting Muchun Song
` (20 subsequent siblings)
21 siblings, 0 replies; 29+ messages in thread
From: Muchun Song @ 2021-06-16 10:53 UTC (permalink / raw)
To: Matthew Wilcox, Andrew Morton, Johannes Weiner, Michal Hocko,
Vladimir Davydov, Shakeel Butt, Roman Gushchin, Yang Shi,
Alex Shi, Wei Yang, Dave Chinner, trond.myklebust,
anna.schumaker
Cc: linux-fsdevel, LKML, Linux Memory Management List, linux-nfs,
zhengqi.arch, Xiongchun duan, fam.zheng
I'm very sorry to bother everyone. Ping guys. Any comments on this series?
On Thu, May 27, 2021 at 2:24 PM Muchun Song <songmuchun@bytedance.com> wrote:
>
> In our server, we found a suspected memory leak problem. The kmalloc-32
> consumes more than 6GB of memory. Other kmem_caches consume less than 2GB
> memory.
>
> After our in-depth analysis, the memory consumption of kmalloc-32 slab
> cache is the cause of list_lru_one allocation.
>
> crash> p memcg_nr_cache_ids
> memcg_nr_cache_ids = $2 = 24574
>
> memcg_nr_cache_ids is very large and memory consumption of each list_lru
> can be calculated with the following formula.
>
> num_numa_node * memcg_nr_cache_ids * 32 (kmalloc-32)
>
> There are 4 numa nodes in our system, so each list_lru consumes ~3MB.
>
> crash> list super_blocks | wc -l
> 952
>
> Every mount will register 2 list lrus, one is for inode, another is for
> dentry. There are 952 super_blocks. So the total memory is 952 * 2 * 3
> MB (~5.6GB). But now the number of memory cgroups is less than 500. So I
> guess more than 12286 memory cgroups have been created on this machine (I
> do not know why there are so many cgroups, it may be a user's bug or
> the user really want to do that). Because memcg_nr_cache_ids has not been
> reduced to a suitable value. This can waste a lot of memory. If we want
> to reduce memcg_nr_cache_ids, we have to reboot the server. This is not
> what we want.
>
> In order to reduce memcg_nr_cache_ids, I had posted a patchset [1] to do
> this. But this did not fundamentally solve the problem.
>
> We currently allocate scope for every memcg to be able to tracked on every
> superblock instantiated in the system, regardless of whether that superblock
> is even accessible to that memcg.
>
> These huge memcg counts come from container hosts where memcgs are confined
> to just a small subset of the total number of superblocks that instantiated
> at any given point in time.
>
> For these systems with huge container counts, list_lru does not need the
> capability of tracking every memcg on every superblock.
>
> What it comes down to is that the list_lru is only needed for a given memcg
> if that memcg is instatiating and freeing objects on a given list_lru.
>
> As Dave said, "Which makes me think we should be moving more towards 'add the
> memcg to the list_lru at the first insert' model rather than 'instantiate
> all at memcg init time just in case'."
>
> This patchset aims to optimize the list lru memory consumption from different
> aspects.
>
> Patch 1-6 are code simplification.
> Patch 7 converts the array from per-memcg per-node to per-memcg
> Patch 8 is code simplification.
> Patch 9-15 let list_lru allocation dynamically.
> Patch 16 is code cleanup.
> Patch 17 use xarray to optimize per memcg pointer array size.
> Patch 18-21 is code simplification.
>
> I had done a easy test to show the optimization. I create 10k memory cgroups
> and mount 10k filesystems in the systems. We use free command to show how many
> memory does the systems comsumes after this operation (There are 2 numa nodes
> in the system).
>
> +-----------------------+------------------------+
> | condition | memory consumption |
> +-----------------------+------------------------+
> | without this patchset | 24464 MB |
> +-----------------------+------------------------+
> | after patch 7 | 21957 MB | <--------+
> +-----------------------+------------------------+ |
> | after patch 15 | 6895 MB | |
> +-----------------------+------------------------+ |
> | after patch 17 | 4367 MB | |
> +-----------------------+------------------------+ |
> |
> The more the number of nodes, the more obvious the effect---+
>
> BTW, there was a recent discussion [2] on the same issue.
>
> [1] https://lore.kernel.org/linux-fsdevel/20210428094949.43579-1-songmuchun@bytedance.com/
> [2] https://lore.kernel.org/linux-fsdevel/20210405054848.GA1077931@in.ibm.com/
>
> Changelog in v2:
> 1. Update Documentation/filesystems/porting.rst suggested by Dave.
> 2. Add a comment above alloc_inode_sb() suggested by Dave.
> 2. Rework some patch's commit log.
> 3. Add patch 18-21.
>
> Thanks Dave.
>
> Muchun Song (21):
> mm: list_lru: fix list_lru_count_one() return value
> mm: memcontrol: remove kmemcg_id reparenting
> mm: memcontrol: remove the kmem states
> mm: memcontrol: do it in mem_cgroup_css_online to make the kmem online
> mm: list_lru: remove lru node locking from memcg_update_list_lru_node
> mm: list_lru: only add the memcg aware lrus to the list_lrus
> mm: list_lru: optimize the array of per memcg lists memory consumption
> mm: list_lru: remove memcg_aware field from struct list_lru
> mm: introduce kmem_cache_alloc_lru
> fs: introduce alloc_inode_sb() to allocate filesystems specific inode
> mm: dcache: use kmem_cache_alloc_lru() to allocate dentry
> xarray: use kmem_cache_alloc_lru to allocate xa_node
> mm: workingset: use xas_set_lru() to pass shadow_nodes
> nfs42: use a specific kmem_cache to allocate nfs4_xattr_entry
> mm: list_lru: allocate list_lru_one only when needed
> mm: list_lru: rename memcg_drain_all_list_lrus to
> memcg_reparent_list_lrus
> mm: list_lru: replace linear array with xarray
> mm: memcontrol: reuse memory cgroup ID for kmem ID
> mm: memcontrol: fix cannot alloc the maximum memcg ID
> mm: list_lru: rename list_lru_per_memcg to list_lru_memcg
> mm: memcontrol: rename memcg_cache_id to memcg_kmem_id
>
> Documentation/filesystems/porting.rst | 5 +
> drivers/dax/super.c | 2 +-
> fs/9p/vfs_inode.c | 2 +-
> fs/adfs/super.c | 2 +-
> fs/affs/super.c | 2 +-
> fs/afs/super.c | 2 +-
> fs/befs/linuxvfs.c | 2 +-
> fs/bfs/inode.c | 2 +-
> fs/block_dev.c | 2 +-
> fs/btrfs/inode.c | 2 +-
> fs/ceph/inode.c | 2 +-
> fs/cifs/cifsfs.c | 2 +-
> fs/coda/inode.c | 2 +-
> fs/dcache.c | 3 +-
> fs/ecryptfs/super.c | 2 +-
> fs/efs/super.c | 2 +-
> fs/erofs/super.c | 2 +-
> fs/exfat/super.c | 2 +-
> fs/ext2/super.c | 2 +-
> fs/ext4/super.c | 2 +-
> fs/f2fs/super.c | 2 +-
> fs/fat/inode.c | 2 +-
> fs/freevxfs/vxfs_super.c | 2 +-
> fs/fuse/inode.c | 2 +-
> fs/gfs2/super.c | 2 +-
> fs/hfs/super.c | 2 +-
> fs/hfsplus/super.c | 2 +-
> fs/hostfs/hostfs_kern.c | 2 +-
> fs/hpfs/super.c | 2 +-
> fs/hugetlbfs/inode.c | 2 +-
> fs/inode.c | 2 +-
> fs/isofs/inode.c | 2 +-
> fs/jffs2/super.c | 2 +-
> fs/jfs/super.c | 2 +-
> fs/minix/inode.c | 2 +-
> fs/nfs/inode.c | 2 +-
> fs/nfs/nfs42xattr.c | 95 ++++----
> fs/nilfs2/super.c | 2 +-
> fs/ntfs/inode.c | 2 +-
> fs/ocfs2/dlmfs/dlmfs.c | 2 +-
> fs/ocfs2/super.c | 2 +-
> fs/openpromfs/inode.c | 2 +-
> fs/orangefs/super.c | 2 +-
> fs/overlayfs/super.c | 2 +-
> fs/proc/inode.c | 2 +-
> fs/qnx4/inode.c | 2 +-
> fs/qnx6/inode.c | 2 +-
> fs/reiserfs/super.c | 2 +-
> fs/romfs/super.c | 2 +-
> fs/squashfs/super.c | 2 +-
> fs/sysv/inode.c | 2 +-
> fs/ubifs/super.c | 2 +-
> fs/udf/super.c | 2 +-
> fs/ufs/super.c | 2 +-
> fs/vboxsf/super.c | 2 +-
> fs/xfs/xfs_icache.c | 3 +-
> fs/zonefs/super.c | 2 +-
> include/linux/fs.h | 11 +
> include/linux/list_lru.h | 18 +-
> include/linux/memcontrol.h | 48 ++--
> include/linux/slab.h | 4 +
> include/linux/swap.h | 5 +-
> include/linux/xarray.h | 9 +-
> ipc/mqueue.c | 2 +-
> lib/xarray.c | 10 +-
> mm/list_lru.c | 447 +++++++++++++++++-----------------
> mm/memcontrol.c | 185 ++------------
> mm/shmem.c | 2 +-
> mm/slab.c | 39 ++-
> mm/slab.h | 17 +-
> mm/slub.c | 42 ++--
> mm/workingset.c | 2 +-
> net/socket.c | 2 +-
> net/sunrpc/rpc_pipe.c | 2 +-
> 74 files changed, 480 insertions(+), 577 deletions(-)
>
> --
> 2.11.0
>
^ permalink raw reply [flat|nested] 29+ messages in thread* Re: [PATCH v2 00/21] Optimize list lru memory consumption
@ 2021-06-16 10:53 ` Muchun Song
0 siblings, 0 replies; 29+ messages in thread
From: Muchun Song @ 2021-06-16 10:53 UTC (permalink / raw)
To: Matthew Wilcox, Andrew Morton, Johannes Weiner, Michal Hocko,
Vladimir Davydov, Shakeel Butt, Roman Gushchin, Yang Shi,
Alex Shi, Wei Yang, Dave Chinner, trond.myklebust,
anna.schumaker
Cc: linux-fsdevel, LKML, Linux Memory Management List, linux-nfs,
zhengqi.arch, Xiongchun duan, fam.zheng
I'm very sorry to bother everyone. Ping guys. Any comments on this series?
On Thu, May 27, 2021 at 2:24 PM Muchun Song <songmuchun@bytedance.com> wrote:
>
> In our server, we found a suspected memory leak problem. The kmalloc-32
> consumes more than 6GB of memory. Other kmem_caches consume less than 2GB
> memory.
>
> After our in-depth analysis, the memory consumption of kmalloc-32 slab
> cache is the cause of list_lru_one allocation.
>
> crash> p memcg_nr_cache_ids
> memcg_nr_cache_ids = $2 = 24574
>
> memcg_nr_cache_ids is very large and memory consumption of each list_lru
> can be calculated with the following formula.
>
> num_numa_node * memcg_nr_cache_ids * 32 (kmalloc-32)
>
> There are 4 numa nodes in our system, so each list_lru consumes ~3MB.
>
> crash> list super_blocks | wc -l
> 952
>
> Every mount will register 2 list lrus, one is for inode, another is for
> dentry. There are 952 super_blocks. So the total memory is 952 * 2 * 3
> MB (~5.6GB). But now the number of memory cgroups is less than 500. So I
> guess more than 12286 memory cgroups have been created on this machine (I
> do not know why there are so many cgroups, it may be a user's bug or
> the user really want to do that). Because memcg_nr_cache_ids has not been
> reduced to a suitable value. This can waste a lot of memory. If we want
> to reduce memcg_nr_cache_ids, we have to reboot the server. This is not
> what we want.
>
> In order to reduce memcg_nr_cache_ids, I had posted a patchset [1] to do
> this. But this did not fundamentally solve the problem.
>
> We currently allocate scope for every memcg to be able to tracked on every
> superblock instantiated in the system, regardless of whether that superblock
> is even accessible to that memcg.
>
> These huge memcg counts come from container hosts where memcgs are confined
> to just a small subset of the total number of superblocks that instantiated
> at any given point in time.
>
> For these systems with huge container counts, list_lru does not need the
> capability of tracking every memcg on every superblock.
>
> What it comes down to is that the list_lru is only needed for a given memcg
> if that memcg is instatiating and freeing objects on a given list_lru.
>
> As Dave said, "Which makes me think we should be moving more towards 'add the
> memcg to the list_lru at the first insert' model rather than 'instantiate
> all at memcg init time just in case'."
>
> This patchset aims to optimize the list lru memory consumption from different
> aspects.
>
> Patch 1-6 are code simplification.
> Patch 7 converts the array from per-memcg per-node to per-memcg
> Patch 8 is code simplification.
> Patch 9-15 let list_lru allocation dynamically.
> Patch 16 is code cleanup.
> Patch 17 use xarray to optimize per memcg pointer array size.
> Patch 18-21 is code simplification.
>
> I had done a easy test to show the optimization. I create 10k memory cgroups
> and mount 10k filesystems in the systems. We use free command to show how many
> memory does the systems comsumes after this operation (There are 2 numa nodes
> in the system).
>
> +-----------------------+------------------------+
> | condition | memory consumption |
> +-----------------------+------------------------+
> | without this patchset | 24464 MB |
> +-----------------------+------------------------+
> | after patch 7 | 21957 MB | <--------+
> +-----------------------+------------------------+ |
> | after patch 15 | 6895 MB | |
> +-----------------------+------------------------+ |
> | after patch 17 | 4367 MB | |
> +-----------------------+------------------------+ |
> |
> The more the number of nodes, the more obvious the effect---+
>
> BTW, there was a recent discussion [2] on the same issue.
>
> [1] https://lore.kernel.org/linux-fsdevel/20210428094949.43579-1-songmuchun@bytedance.com/
> [2] https://lore.kernel.org/linux-fsdevel/20210405054848.GA1077931@in.ibm.com/
>
> Changelog in v2:
> 1. Update Documentation/filesystems/porting.rst suggested by Dave.
> 2. Add a comment above alloc_inode_sb() suggested by Dave.
> 2. Rework some patch's commit log.
> 3. Add patch 18-21.
>
> Thanks Dave.
>
> Muchun Song (21):
> mm: list_lru: fix list_lru_count_one() return value
> mm: memcontrol: remove kmemcg_id reparenting
> mm: memcontrol: remove the kmem states
> mm: memcontrol: do it in mem_cgroup_css_online to make the kmem online
> mm: list_lru: remove lru node locking from memcg_update_list_lru_node
> mm: list_lru: only add the memcg aware lrus to the list_lrus
> mm: list_lru: optimize the array of per memcg lists memory consumption
> mm: list_lru: remove memcg_aware field from struct list_lru
> mm: introduce kmem_cache_alloc_lru
> fs: introduce alloc_inode_sb() to allocate filesystems specific inode
> mm: dcache: use kmem_cache_alloc_lru() to allocate dentry
> xarray: use kmem_cache_alloc_lru to allocate xa_node
> mm: workingset: use xas_set_lru() to pass shadow_nodes
> nfs42: use a specific kmem_cache to allocate nfs4_xattr_entry
> mm: list_lru: allocate list_lru_one only when needed
> mm: list_lru: rename memcg_drain_all_list_lrus to
> memcg_reparent_list_lrus
> mm: list_lru: replace linear array with xarray
> mm: memcontrol: reuse memory cgroup ID for kmem ID
> mm: memcontrol: fix cannot alloc the maximum memcg ID
> mm: list_lru: rename list_lru_per_memcg to list_lru_memcg
> mm: memcontrol: rename memcg_cache_id to memcg_kmem_id
>
> Documentation/filesystems/porting.rst | 5 +
> drivers/dax/super.c | 2 +-
> fs/9p/vfs_inode.c | 2 +-
> fs/adfs/super.c | 2 +-
> fs/affs/super.c | 2 +-
> fs/afs/super.c | 2 +-
> fs/befs/linuxvfs.c | 2 +-
> fs/bfs/inode.c | 2 +-
> fs/block_dev.c | 2 +-
> fs/btrfs/inode.c | 2 +-
> fs/ceph/inode.c | 2 +-
> fs/cifs/cifsfs.c | 2 +-
> fs/coda/inode.c | 2 +-
> fs/dcache.c | 3 +-
> fs/ecryptfs/super.c | 2 +-
> fs/efs/super.c | 2 +-
> fs/erofs/super.c | 2 +-
> fs/exfat/super.c | 2 +-
> fs/ext2/super.c | 2 +-
> fs/ext4/super.c | 2 +-
> fs/f2fs/super.c | 2 +-
> fs/fat/inode.c | 2 +-
> fs/freevxfs/vxfs_super.c | 2 +-
> fs/fuse/inode.c | 2 +-
> fs/gfs2/super.c | 2 +-
> fs/hfs/super.c | 2 +-
> fs/hfsplus/super.c | 2 +-
> fs/hostfs/hostfs_kern.c | 2 +-
> fs/hpfs/super.c | 2 +-
> fs/hugetlbfs/inode.c | 2 +-
> fs/inode.c | 2 +-
> fs/isofs/inode.c | 2 +-
> fs/jffs2/super.c | 2 +-
> fs/jfs/super.c | 2 +-
> fs/minix/inode.c | 2 +-
> fs/nfs/inode.c | 2 +-
> fs/nfs/nfs42xattr.c | 95 ++++----
> fs/nilfs2/super.c | 2 +-
> fs/ntfs/inode.c | 2 +-
> fs/ocfs2/dlmfs/dlmfs.c | 2 +-
> fs/ocfs2/super.c | 2 +-
> fs/openpromfs/inode.c | 2 +-
> fs/orangefs/super.c | 2 +-
> fs/overlayfs/super.c | 2 +-
> fs/proc/inode.c | 2 +-
> fs/qnx4/inode.c | 2 +-
> fs/qnx6/inode.c | 2 +-
> fs/reiserfs/super.c | 2 +-
> fs/romfs/super.c | 2 +-
> fs/squashfs/super.c | 2 +-
> fs/sysv/inode.c | 2 +-
> fs/ubifs/super.c | 2 +-
> fs/udf/super.c | 2 +-
> fs/ufs/super.c | 2 +-
> fs/vboxsf/super.c | 2 +-
> fs/xfs/xfs_icache.c | 3 +-
> fs/zonefs/super.c | 2 +-
> include/linux/fs.h | 11 +
> include/linux/list_lru.h | 18 +-
> include/linux/memcontrol.h | 48 ++--
> include/linux/slab.h | 4 +
> include/linux/swap.h | 5 +-
> include/linux/xarray.h | 9 +-
> ipc/mqueue.c | 2 +-
> lib/xarray.c | 10 +-
> mm/list_lru.c | 447 +++++++++++++++++-----------------
> mm/memcontrol.c | 185 ++------------
> mm/shmem.c | 2 +-
> mm/slab.c | 39 ++-
> mm/slab.h | 17 +-
> mm/slub.c | 42 ++--
> mm/workingset.c | 2 +-
> net/socket.c | 2 +-
> net/sunrpc/rpc_pipe.c | 2 +-
> 74 files changed, 480 insertions(+), 577 deletions(-)
>
> --
> 2.11.0
>
^ permalink raw reply [flat|nested] 29+ messages in thread