From: Qi Zheng <qi.zheng@linux.dev>
To: hannes@cmpxchg.org, hughd@google.com, mhocko@suse.com,
roman.gushchin@linux.dev, shakeel.butt@linux.dev,
muchun.song@linux.dev, david@redhat.com,
lorenzo.stoakes@oracle.com, ziy@nvidia.com, harry.yoo@oracle.com,
imran.f.khan@oracle.com, kamalesh.babulal@oracle.com,
axelrasmussen@google.com, yuanchu@google.com, weixugc@google.com,
akpm@linux-foundation.org
Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org,
cgroups@vger.kernel.org, Muchun Song <songmuchun@bytedance.com>,
Qi Zheng <zhengqi.arch@bytedance.com>
Subject: [PATCH v1 06/26] mm: memcontrol: return root object cgroup for root memory cgroup
Date: Tue, 28 Oct 2025 21:58:19 +0800 [thread overview]
Message-ID: <5e9743f291e7ca7b8f052775e993090ed66cfa80.1761658310.git.zhengqi.arch@bytedance.com> (raw)
In-Reply-To: <cover.1761658310.git.zhengqi.arch@bytedance.com>
From: Muchun Song <songmuchun@bytedance.com>
Memory cgroup functions such as get_mem_cgroup_from_folio() and
get_mem_cgroup_from_mm() return a valid memory cgroup pointer,
even for the root memory cgroup. In contrast, the situation for
object cgroups has been different.
Previously, the root object cgroup couldn't be returned because
it didn't exist. Now that a valid root object cgroup exists, for
the sake of consistency, it's necessary to align the behavior of
object-cgroup-related operations with that of memory cgroup APIs.
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
Signed-off-by: Qi Zheng <zhengqi.arch@bytedance.com>
---
include/linux/memcontrol.h | 29 +++++++++++++++++-------
mm/memcontrol.c | 45 ++++++++++++++++++++------------------
mm/percpu.c | 2 +-
3 files changed, 46 insertions(+), 30 deletions(-)
diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index 6185d8399a54e..9fdbd4970021d 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -332,6 +332,7 @@ struct mem_cgroup {
#define MEMCG_CHARGE_BATCH 64U
extern struct mem_cgroup *root_mem_cgroup;
+extern struct obj_cgroup *root_obj_cgroup;
enum page_memcg_data_flags {
/* page->memcg_data is a pointer to an slabobj_ext vector */
@@ -549,6 +550,11 @@ static inline bool mem_cgroup_is_root(struct mem_cgroup *memcg)
return (memcg == root_mem_cgroup);
}
+static inline bool obj_cgroup_is_root(const struct obj_cgroup *objcg)
+{
+ return objcg == root_obj_cgroup;
+}
+
static inline bool mem_cgroup_disabled(void)
{
return !cgroup_subsys_enabled(memory_cgrp_subsys);
@@ -773,23 +779,26 @@ struct mem_cgroup *mem_cgroup_from_css(struct cgroup_subsys_state *css){
static inline bool obj_cgroup_tryget(struct obj_cgroup *objcg)
{
+ if (obj_cgroup_is_root(objcg))
+ return true;
return percpu_ref_tryget(&objcg->refcnt);
}
-static inline void obj_cgroup_get(struct obj_cgroup *objcg)
+static inline void obj_cgroup_get_many(struct obj_cgroup *objcg,
+ unsigned long nr)
{
- percpu_ref_get(&objcg->refcnt);
+ if (!obj_cgroup_is_root(objcg))
+ percpu_ref_get_many(&objcg->refcnt, nr);
}
-static inline void obj_cgroup_get_many(struct obj_cgroup *objcg,
- unsigned long nr)
+static inline void obj_cgroup_get(struct obj_cgroup *objcg)
{
- percpu_ref_get_many(&objcg->refcnt, nr);
+ obj_cgroup_get_many(objcg, 1);
}
static inline void obj_cgroup_put(struct obj_cgroup *objcg)
{
- if (objcg)
+ if (objcg && !obj_cgroup_is_root(objcg))
percpu_ref_put(&objcg->refcnt);
}
@@ -1094,6 +1103,11 @@ static inline bool mem_cgroup_is_root(struct mem_cgroup *memcg)
return true;
}
+static inline bool obj_cgroup_is_root(const struct obj_cgroup *objcg)
+{
+ return true;
+}
+
static inline bool mem_cgroup_disabled(void)
{
return true;
@@ -1710,8 +1724,7 @@ static inline struct obj_cgroup *get_obj_cgroup_from_current(void)
{
struct obj_cgroup *objcg = current_obj_cgroup();
- if (objcg)
- obj_cgroup_get(objcg);
+ obj_cgroup_get(objcg);
return objcg;
}
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 2afd7f99ca101..d484b632c790f 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -83,6 +83,8 @@ EXPORT_SYMBOL(memory_cgrp_subsys);
struct mem_cgroup *root_mem_cgroup __read_mostly;
EXPORT_SYMBOL(root_mem_cgroup);
+struct obj_cgroup *root_obj_cgroup __read_mostly;
+
/* Active memory cgroup to use from an interrupt context */
DEFINE_PER_CPU(struct mem_cgroup *, int_active_memcg);
EXPORT_PER_CPU_SYMBOL_GPL(int_active_memcg);
@@ -2642,15 +2644,14 @@ struct mem_cgroup *mem_cgroup_from_slab_obj(void *p)
static struct obj_cgroup *__get_obj_cgroup_from_memcg(struct mem_cgroup *memcg)
{
- struct obj_cgroup *objcg = NULL;
+ for (; memcg; memcg = parent_mem_cgroup(memcg)) {
+ struct obj_cgroup *objcg = rcu_dereference(memcg->objcg);
- for (; !mem_cgroup_is_root(memcg); memcg = parent_mem_cgroup(memcg)) {
- objcg = rcu_dereference(memcg->objcg);
if (likely(objcg && obj_cgroup_tryget(objcg)))
- break;
- objcg = NULL;
+ return objcg;
}
- return objcg;
+
+ return NULL;
}
static struct obj_cgroup *current_objcg_update(void)
@@ -2724,18 +2725,17 @@ __always_inline struct obj_cgroup *current_obj_cgroup(void)
* Objcg reference is kept by the task, so it's safe
* to use the objcg by the current task.
*/
- return objcg;
+ return objcg ? : root_obj_cgroup;
}
memcg = this_cpu_read(int_active_memcg);
if (unlikely(memcg))
goto from_memcg;
- return NULL;
+ return root_obj_cgroup;
from_memcg:
- objcg = NULL;
- for (; !mem_cgroup_is_root(memcg); memcg = parent_mem_cgroup(memcg)) {
+ for (; memcg; memcg = parent_mem_cgroup(memcg)) {
/*
* Memcg pointer is protected by scope (see set_active_memcg())
* and is pinning the corresponding objcg, so objcg can't go
@@ -2744,10 +2744,10 @@ __always_inline struct obj_cgroup *current_obj_cgroup(void)
*/
objcg = rcu_dereference_check(memcg->objcg, 1);
if (likely(objcg))
- break;
+ return objcg;
}
- return objcg;
+ return root_obj_cgroup;
}
struct obj_cgroup *get_obj_cgroup_from_folio(struct folio *folio)
@@ -2761,14 +2761,8 @@ struct obj_cgroup *get_obj_cgroup_from_folio(struct folio *folio)
objcg = __folio_objcg(folio);
obj_cgroup_get(objcg);
} else {
- struct mem_cgroup *memcg;
-
rcu_read_lock();
- memcg = __folio_memcg(folio);
- if (memcg)
- objcg = __get_obj_cgroup_from_memcg(memcg);
- else
- objcg = NULL;
+ objcg = __get_obj_cgroup_from_memcg(__folio_memcg(folio));
rcu_read_unlock();
}
return objcg;
@@ -2871,7 +2865,7 @@ int __memcg_kmem_charge_page(struct page *page, gfp_t gfp, int order)
int ret = 0;
objcg = current_obj_cgroup();
- if (objcg) {
+ if (!obj_cgroup_is_root(objcg)) {
ret = obj_cgroup_charge_pages(objcg, gfp, 1 << order);
if (!ret) {
obj_cgroup_get(objcg);
@@ -3172,7 +3166,7 @@ bool __memcg_slab_post_alloc_hook(struct kmem_cache *s, struct list_lru *lru,
* obj_cgroup_get() is used to get a permanent reference.
*/
objcg = current_obj_cgroup();
- if (!objcg)
+ if (obj_cgroup_is_root(objcg))
return true;
/*
@@ -3859,6 +3853,9 @@ static int mem_cgroup_css_online(struct cgroup_subsys_state *css)
if (!objcg)
goto free_shrinker;
+ if (unlikely(mem_cgroup_is_root(memcg)))
+ root_obj_cgroup = objcg;
+
objcg->memcg = memcg;
rcu_assign_pointer(memcg->objcg, objcg);
obj_cgroup_get(objcg);
@@ -5479,6 +5476,9 @@ void obj_cgroup_charge_zswap(struct obj_cgroup *objcg, size_t size)
if (!cgroup_subsys_on_dfl(memory_cgrp_subsys))
return;
+ if (obj_cgroup_is_root(objcg))
+ return;
+
VM_WARN_ON_ONCE(!(current->flags & PF_MEMALLOC));
/* PF_MEMALLOC context, charging must succeed */
@@ -5506,6 +5506,9 @@ void obj_cgroup_uncharge_zswap(struct obj_cgroup *objcg, size_t size)
if (!cgroup_subsys_on_dfl(memory_cgrp_subsys))
return;
+ if (obj_cgroup_is_root(objcg))
+ return;
+
obj_cgroup_uncharge(objcg, size);
rcu_read_lock();
diff --git a/mm/percpu.c b/mm/percpu.c
index 81462ce5866e1..78bdffe1fcb57 100644
--- a/mm/percpu.c
+++ b/mm/percpu.c
@@ -1616,7 +1616,7 @@ static bool pcpu_memcg_pre_alloc_hook(size_t size, gfp_t gfp,
return true;
objcg = current_obj_cgroup();
- if (!objcg)
+ if (obj_cgroup_is_root(objcg))
return true;
if (obj_cgroup_charge(objcg, gfp, pcpu_obj_full_size(size)))
--
2.20.1
next prev parent reply other threads:[~2025-10-28 14:02 UTC|newest]
Thread overview: 107+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-10-28 13:58 [PATCH v1 00/26] Eliminate Dying Memory Cgroup Qi Zheng
2025-10-28 13:58 ` [PATCH v1 01/26] mm: memcontrol: remove dead code of checking parent memory cgroup Qi Zheng
2025-11-07 1:40 ` Harry Yoo
2025-11-20 9:07 ` Chen Ridong
2025-10-28 13:58 ` [PATCH v1 02/26] mm: workingset: use folio_lruvec() in workingset_refault() Qi Zheng
2025-11-07 1:55 ` Harry Yoo
2025-10-28 13:58 ` [PATCH v1 03/26] mm: rename unlock_page_lruvec_irq and its variants Qi Zheng
2025-11-07 2:03 ` Harry Yoo
2025-11-20 12:27 ` Chen Ridong
2025-10-28 13:58 ` [PATCH v1 04/26] mm: vmscan: refactor move_folios_to_lru() Qi Zheng
2025-11-07 5:11 ` Harry Yoo
2025-11-07 6:41 ` Qi Zheng
2025-11-07 13:20 ` Harry Yoo
2025-11-08 6:32 ` Shakeel Butt
2025-11-10 2:13 ` Harry Yoo
2025-11-10 4:30 ` Qi Zheng
2025-11-10 5:43 ` Harry Yoo
2025-11-10 6:11 ` Qi Zheng
2025-11-10 16:47 ` Shakeel Butt
2025-11-11 0:42 ` Harry Yoo
2025-11-11 3:04 ` Qi Zheng
2025-11-11 3:16 ` Harry Yoo
2025-11-11 3:23 ` Qi Zheng
2025-11-11 8:49 ` Sebastian Andrzej Siewior
2025-11-11 16:44 ` Shakeel Butt
2025-11-12 7:49 ` Sebastian Andrzej Siewior
2025-11-12 8:46 ` Harry Yoo
2025-11-12 8:54 ` Sebastian Andrzej Siewior
2025-11-12 15:45 ` Steven Rostedt
2025-11-11 3:17 ` Shakeel Butt
2025-11-11 3:24 ` Qi Zheng
2025-11-07 7:18 ` Sebastian Andrzej Siewior
2025-10-28 13:58 ` [PATCH v1 05/26] mm: memcontrol: allocate object cgroup for non-kmem case Qi Zheng
2025-11-17 8:02 ` Harry Yoo
2025-11-21 3:58 ` Chen Ridong
2025-11-21 8:17 ` Qi Zheng
2025-10-28 13:58 ` Qi Zheng [this message]
2025-11-17 9:17 ` [PATCH v1 06/26] mm: memcontrol: return root object cgroup for root memory cgroup Harry Yoo
2025-11-17 9:41 ` Harry Yoo
2025-11-18 11:31 ` Qi Zheng
2025-11-18 11:28 ` Qi Zheng
2025-11-18 12:11 ` Qi Zheng
2025-11-19 7:24 ` Harry Yoo
2025-11-19 7:42 ` Qi Zheng
2025-11-18 12:12 ` Harry Yoo
2025-11-19 6:40 ` Qi Zheng
2025-10-28 13:58 ` [PATCH v1 07/26] mm: memcontrol: prevent memory cgroup release in get_mem_cgroup_from_folio() Qi Zheng
2025-11-19 8:06 ` Harry Yoo
2025-11-20 13:32 ` Qi Zheng
2025-10-28 13:58 ` [PATCH v1 08/26] buffer: prevent memory cgroup release in folio_alloc_buffers() Qi Zheng
2025-11-19 8:10 ` Harry Yoo
2025-10-28 13:58 ` [PATCH v1 09/26] writeback: prevent memory cgroup release in writeback module Qi Zheng
2025-11-19 9:18 ` Harry Yoo
2025-10-28 13:58 ` [PATCH v1 10/26] mm: memcontrol: prevent memory cgroup release in count_memcg_folio_events() Qi Zheng
2025-11-19 9:21 ` Harry Yoo
2025-10-28 13:58 ` [PATCH v1 11/26] mm: page_io: prevent memory cgroup release in page_io module Qi Zheng
2025-11-19 9:26 ` Harry Yoo
2025-11-20 13:34 ` Qi Zheng
2025-10-28 13:58 ` [PATCH v1 12/26] mm: migrate: prevent memory cgroup release in folio_migrate_mapping() Qi Zheng
2025-11-19 10:00 ` Harry Yoo
2025-10-28 13:58 ` [PATCH v1 13/26] mm: mglru: prevent memory cgroup release in mglru Qi Zheng
2025-11-19 10:13 ` Harry Yoo
2025-11-20 13:39 ` Qi Zheng
2025-10-28 13:58 ` [PATCH v1 14/26] mm: memcontrol: prevent memory cgroup release in mem_cgroup_swap_full() Qi Zheng
2025-11-20 7:51 ` Harry Yoo
2025-10-28 13:58 ` [PATCH v1 15/26] mm: workingset: prevent memory cgroup release in lru_gen_eviction() Qi Zheng
2025-11-20 8:26 ` Harry Yoo
2025-10-28 13:58 ` [PATCH v1 16/26] mm: thp: prevent memory cgroup release in folio_split_queue_lock{_irqsave}() Qi Zheng
2025-11-20 8:53 ` Harry Yoo
2025-10-28 13:58 ` [PATCH v1 17/26] mm: workingset: prevent lruvec release in workingset_refault() Qi Zheng
2025-11-20 9:40 ` Harry Yoo
2025-10-28 13:58 ` [PATCH v1 18/26] mm: zswap: prevent lruvec release in zswap_folio_swapin() Qi Zheng
2025-11-20 9:42 ` Harry Yoo
2025-10-28 13:58 ` [PATCH v1 19/26] mm: swap: prevent lruvec release in swap module Qi Zheng
2025-11-20 9:52 ` Harry Yoo
2025-11-20 13:41 ` Qi Zheng
2025-10-28 13:58 ` [PATCH v1 20/26] mm: workingset: prevent lruvec release in workingset_activation() Qi Zheng
2025-11-20 9:54 ` Harry Yoo
2025-10-28 13:58 ` [PATCH v1 21/26] mm: memcontrol: prepare for reparenting LRU pages for lruvec lock Qi Zheng
2025-11-04 6:49 ` kernel test robot
2025-11-04 8:59 ` Qi Zheng
2025-11-21 3:15 ` Harry Yoo
2025-11-21 8:01 ` Qi Zheng
2025-10-28 13:58 ` [PATCH v1 22/26] mm: vmscan: prepare for reparenting traditional LRU folios Qi Zheng
2025-11-21 10:11 ` Harry Yoo
2025-10-28 13:58 ` [PATCH v1 23/26] mm: vmscan: prepare for reparenting MGLRU folios Qi Zheng
2025-11-25 9:55 ` Harry Yoo
2025-11-26 2:44 ` Qi Zheng
2025-11-26 13:48 ` Harry Yoo
2025-11-27 3:48 ` Qi Zheng
2025-12-01 15:40 ` Qi Zheng
2025-12-01 21:50 ` Yuanchu Xie
2025-12-02 3:04 ` Qi Zheng
2025-10-28 13:58 ` [PATCH v1 24/26] mm: memcontrol: refactor memcg_reparent_objcgs() Qi Zheng
2025-10-28 13:58 ` [PATCH v1 25/26] mm: memcontrol: eliminate the problem of dying memory cgroup for LRU folios Qi Zheng
2025-11-14 17:56 ` Michal Koutný
2025-11-20 11:56 ` Chen Ridong
2025-11-20 13:45 ` Qi Zheng
2025-10-28 13:58 ` [PATCH v1 26/26] mm: lru: add VM_WARN_ON_ONCE_FOLIO to lru maintenance helpers Qi Zheng
2025-10-28 20:58 ` [syzbot ci] Re: Eliminate Dying Memory Cgroup syzbot ci
2025-10-29 0:22 ` Harry Yoo
2025-10-29 0:25 ` syzbot ci
2025-10-29 3:12 ` Qi Zheng
2025-10-29 7:53 ` [PATCH v1 00/26] " Michal Hocko
2025-10-29 8:05 ` Qi Zheng
2025-10-31 10:35 ` Michal Hocko
2025-11-03 3:33 ` Qi Zheng
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=5e9743f291e7ca7b8f052775e993090ed66cfa80.1761658310.git.zhengqi.arch@bytedance.com \
--to=qi.zheng@linux.dev \
--cc=akpm@linux-foundation.org \
--cc=axelrasmussen@google.com \
--cc=cgroups@vger.kernel.org \
--cc=david@redhat.com \
--cc=hannes@cmpxchg.org \
--cc=harry.yoo@oracle.com \
--cc=hughd@google.com \
--cc=imran.f.khan@oracle.com \
--cc=kamalesh.babulal@oracle.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=lorenzo.stoakes@oracle.com \
--cc=mhocko@suse.com \
--cc=muchun.song@linux.dev \
--cc=roman.gushchin@linux.dev \
--cc=shakeel.butt@linux.dev \
--cc=songmuchun@bytedance.com \
--cc=weixugc@google.com \
--cc=yuanchu@google.com \
--cc=zhengqi.arch@bytedance.com \
--cc=ziy@nvidia.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox