From: Muchun Song <songmuchun@bytedance.com>
To: guro@fb.com, hannes@cmpxchg.org, mhocko@kernel.org,
akpm@linux-foundation.org, shakeelb@google.com,
vdavydov.dev@gmail.com
Cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org,
duanxiongchun@bytedance.com,
Muchun Song <songmuchun@bytedance.com>
Subject: [RFC PATCH 10/15] mm: thp: make deferred split queue lock safe when the LRU pages reparented
Date: Tue, 30 Mar 2021 18:15:26 +0800 [thread overview]
Message-ID: <20210330101531.82752-11-songmuchun@bytedance.com> (raw)
In-Reply-To: <20210330101531.82752-1-songmuchun@bytedance.com>
Similar to lruvec lock, we use the same approach to make the lock safe
when the LRU pages reparented.
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
---
mm/huge_memory.c | 26 ++++++++++++++++++++++++++
1 file changed, 26 insertions(+)
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 186dc11e8992..434cc7283a64 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -496,6 +496,8 @@ static struct deferred_split *lock_split_queue(struct page *page)
struct deferred_split *queue;
struct mem_cgroup *memcg;
+ rcu_read_lock();
+retry:
memcg = page_memcg(compound_head(page));
if (memcg)
queue = &memcg->deferred_split_queue;
@@ -503,6 +505,17 @@ static struct deferred_split *lock_split_queue(struct page *page)
queue = &NODE_DATA(page_to_nid(page))->deferred_split_queue;
spin_lock(&queue->split_queue_lock);
+ if (unlikely(memcg != page_memcg(page))) {
+ spin_unlock(&queue->split_queue_lock);
+ goto retry;
+ }
+
+ /*
+ * Preemption is disabled in the internal of spin_lock, which can serve
+ * as RCU read-side critical sections.
+ */
+ rcu_read_unlock();
+
return queue;
}
@@ -512,6 +525,8 @@ static struct deferred_split *lock_split_queue_irqsave(struct page *page,
struct deferred_split *queue;
struct mem_cgroup *memcg;
+ rcu_read_lock();
+retry:
memcg = page_memcg(compound_head(page));
if (memcg)
queue = &memcg->deferred_split_queue;
@@ -519,6 +534,17 @@ static struct deferred_split *lock_split_queue_irqsave(struct page *page,
queue = &NODE_DATA(page_to_nid(page))->deferred_split_queue;
spin_lock_irqsave(&queue->split_queue_lock, *flags);
+ if (unlikely(memcg != page_memcg(page))) {
+ spin_unlock_irqrestore(&queue->split_queue_lock, *flags);
+ goto retry;
+ }
+
+ /*
+ * Preemption is disabled in the internal of spin_lock, which can serve
+ * as RCU read-side critical sections.
+ */
+ rcu_read_unlock();
+
return queue;
}
#else
--
2.11.0
next prev parent reply other threads:[~2021-03-30 10:22 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-30 10:15 [RFC PATCH 00/15] Use obj_cgroup APIs to charge the LRU pages Muchun Song
2021-03-30 10:15 ` [RFC PATCH 01/15] mm: memcontrol: fix page charging in page replacement Muchun Song
2021-04-02 15:07 ` Johannes Weiner
2021-03-30 10:15 ` [RFC PATCH 02/15] mm: memcontrol: bail out early when !mm in get_mem_cgroup_from_mm Muchun Song
2021-04-02 15:08 ` Johannes Weiner
2021-03-30 10:15 ` [RFC PATCH 03/15] mm: memcontrol: remove the pgdata parameter of mem_cgroup_page_lruvec Muchun Song
2021-04-02 15:22 ` Johannes Weiner
2021-03-30 10:15 ` [RFC PATCH 04/15] mm: memcontrol: use lruvec_memcg in lruvec_holds_page_lru_lock Muchun Song
2021-04-02 16:21 ` Johannes Weiner
2021-04-03 12:37 ` [External] " Muchun Song
2021-03-30 10:15 ` [RFC PATCH 05/15] mm: memcontrol: simplify the logic of objcg pinning memcg Muchun Song
2021-03-30 10:15 ` [RFC PATCH 06/15] mm: memcontrol: move the objcg infrastructure out of CONFIG_MEMCG_KMEM Muchun Song
2021-03-30 10:15 ` [RFC PATCH 07/15] mm: memcontrol: introduce compact_lock_page_lruvec_irqsave Muchun Song
2021-03-30 10:15 ` [RFC PATCH 08/15] mm: memcontrol: make lruvec lock safe when the LRU pages reparented Muchun Song
2021-03-30 10:15 ` [RFC PATCH 09/15] mm: thp: introduce lock/unlock_split_queue{_irqsave}() Muchun Song
2021-03-30 10:15 ` Muchun Song [this message]
2021-03-30 10:15 ` [RFC PATCH 11/15] mm: memcontrol: make all the callers of page_memcg() safe Muchun Song
2021-03-30 10:15 ` [RFC PATCH 12/15] mm: memcontrol: introduce memcg_reparent_ops Muchun Song
2021-03-30 10:15 ` [RFC PATCH 13/15] mm: memcontrol: use obj_cgroup APIs to charge the LRU pages Muchun Song
2021-03-30 10:15 ` [RFC PATCH 14/15] mm: memcontrol: rename {un}lock_page_memcg() to {un}lock_page_objcg() Muchun Song
2021-03-30 10:15 ` [RFC PATCH 15/15] mm: lru: add VM_BUG_ON_PAGE to lru maintenance function Muchun Song
2021-03-30 18:34 ` [RFC PATCH 00/15] Use obj_cgroup APIs to charge the LRU pages Shakeel Butt
2021-03-30 18:58 ` Roman Gushchin
2021-03-30 21:30 ` Johannes Weiner
2021-03-30 22:05 ` Roman Gushchin
2021-03-31 15:17 ` Johannes Weiner
2021-04-01 16:07 ` [External] " Muchun Song
2021-04-01 17:15 ` Shakeel Butt
2021-04-02 3:14 ` Muchun Song
2021-04-02 17:30 ` Johannes Weiner
2021-04-01 21:34 ` Roman Gushchin
2021-04-01 22:55 ` Yang Shi
2021-04-02 4:03 ` [External] " Muchun Song
2021-03-30 21:10 ` Johannes Weiner
2021-03-31 0:28 ` Shakeel Butt
2021-03-31 3:28 ` Balbir Singh
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=20210330101531.82752-11-songmuchun@bytedance.com \
--to=songmuchun@bytedance.com \
--cc=akpm@linux-foundation.org \
--cc=duanxiongchun@bytedance.com \
--cc=guro@fb.com \
--cc=hannes@cmpxchg.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@kernel.org \
--cc=shakeelb@google.com \
--cc=vdavydov.dev@gmail.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