From: Kinsey Ho <kinseyho@google.com>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org,
Yosry Ahmed <yosryahmed@google.com>,
Roman Gushchin <roman.gushchin@linux.dev>,
Kinsey Ho <kinseyho@google.com>
Subject: [PATCH mm-unstable v1 3/4] mm: restart if multiple traversals raced
Date: Wed, 24 Jul 2024 19:02:13 +0000 [thread overview]
Message-ID: <20240724190214.1108049-4-kinseyho@google.com> (raw)
In-Reply-To: <20240724190214.1108049-1-kinseyho@google.com>
Currently, if multiple reclaimers raced on the same position, the
reclaimers which detect the race will still reclaim from the same memcg.
Instead, the reclaimers which detect the race should move on to the next
memcg in the hierarchy.
So, in the case where multiple traversals race, jump back to the start
of the mem_cgroup_iter() function to find the next memcg in the
hierarchy to reclaim from.
Signed-off-by: Kinsey Ho <kinseyho@google.com>
---
include/linux/memcontrol.h | 4 ++--
mm/memcontrol.c | 14 ++++++++++----
2 files changed, 12 insertions(+), 6 deletions(-)
diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index 4cbab85e2e56..2b354abe6d48 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -57,7 +57,7 @@ enum memcg_memory_event {
struct mem_cgroup_reclaim_cookie {
pg_data_t *pgdat;
- unsigned int generation;
+ int generation;
};
#ifdef CONFIG_MEMCG
@@ -77,7 +77,7 @@ struct lruvec_stats;
struct mem_cgroup_reclaim_iter {
struct mem_cgroup __rcu *position;
/* scan generation, increased every round-trip */
- unsigned int generation;
+ atomic_t generation;
};
/*
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index f672bc47c6b5..4314a2b8848d 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -1005,18 +1005,20 @@ struct mem_cgroup *mem_cgroup_iter(struct mem_cgroup *root,
rcu_read_lock();
restart:
if (reclaim) {
+ int gen;
struct mem_cgroup_per_node *mz;
mz = root->nodeinfo[reclaim->pgdat->node_id];
iter = &mz->iter;
+ gen = atomic_read(&iter->generation);
/*
* On start, join the current reclaim iteration cycle.
* Exit when a concurrent walker completes it.
*/
if (!prev)
- reclaim->generation = iter->generation;
- else if (reclaim->generation != iter->generation)
+ reclaim->generation = gen;
+ else if (reclaim->generation != gen)
goto out_unlock;
pos = rcu_dereference(iter->position);
@@ -1050,10 +1052,14 @@ struct mem_cgroup *mem_cgroup_iter(struct mem_cgroup *root,
* thread, so check that the value hasn't changed since we read
* it to avoid reclaiming from the same cgroup twice.
*/
- (void)cmpxchg(&iter->position, pos, memcg);
+ if (cmpxchg(&iter->position, pos, memcg) != pos) {
+ if (css && css != &root->css)
+ css_put(css);
+ goto restart;
+ }
if (!memcg) {
- iter->generation++;
+ atomic_inc(&iter->generation);
/*
* Reclaimers share the hierarchy walk, and a
--
2.45.2.1089.g2a221341d9-goog
next prev parent reply other threads:[~2024-07-24 19:02 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-07-24 19:02 [PATCH mm-unstable v1 0/4] Improve mem_cgroup_iter() Kinsey Ho
2024-07-24 19:02 ` [PATCH mm-unstable v1 1/4] mm: don't hold css->refcnt during traversal Kinsey Ho
2024-07-25 16:33 ` kernel test robot
2024-07-25 20:43 ` Johannes Weiner
2024-07-25 21:09 ` Roman Gushchin
2024-07-25 22:33 ` Yosry Ahmed
2024-08-01 21:46 ` Kinsey Ho
2024-08-01 22:32 ` Kinsey Ho
2024-08-02 1:28 ` Johannes Weiner
2024-07-24 19:02 ` [PATCH mm-unstable v1 2/4] mm: increment gen # before restarting traversal Kinsey Ho
2024-07-25 20:53 ` Johannes Weiner
2024-07-24 19:02 ` Kinsey Ho [this message]
2024-07-24 19:02 ` [PATCH mm-unstable v1 4/4] mm: clean up mem_cgroup_iter() Kinsey Ho
2024-07-25 18:15 ` kernel test robot
2024-07-25 0:25 ` [PATCH mm-unstable v1 0/4] Improve mem_cgroup_iter() Roman Gushchin
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=20240724190214.1108049-4-kinseyho@google.com \
--to=kinseyho@google.com \
--cc=akpm@linux-foundation.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=roman.gushchin@linux.dev \
--cc=yosryahmed@google.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