All of lore.kernel.org
 help / color / mirror / Atom feed
From: Yue Zhao <findns94@gmail.com>
To: akpm@linux-foundation.org
Cc: roman.gushchin@linux.dev, hannes@cmpxchg.org, mhocko@kernel.org,
	shakeelb@google.com, muchun.song@linux.dev, willy@infradead.org,
	linux-mm@kvack.org, cgroups@vger.kernel.org,
	linux-kernel@vger.kernel.org, tangyeechou@gmail.com,
	Yue Zhao <findns94@gmail.com>
Subject: [PATCH v2, 4/4] mm, memcg: Prevent memory.soft_limit_in_bytes load/store tearing
Date: Mon,  6 Mar 2023 23:41:38 +0800	[thread overview]
Message-ID: <20230306154138.3775-5-findns94@gmail.com> (raw)
In-Reply-To: <20230306154138.3775-1-findns94@gmail.com>

The knob for cgroup v1 memory controller: memory.soft_limit_in_bytes
is not protected by any locking so it can be modified while it is used.
This is not an actual problem because races are unlikely.
But it is better to use READ_ONCE/WRITE_ONCE to prevent compiler from
doing anything funky.

The access of memcg->soft_limit is lockless,
so it can be concurrently set at the same time as we are
trying to read it.

Signed-off-by: Yue Zhao <findns94@gmail.com>
---
 mm/memcontrol.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 26605b2f51b1..20566f59bbcb 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -3728,7 +3728,7 @@ static u64 mem_cgroup_read_u64(struct cgroup_subsys_state *css,
 	case RES_FAILCNT:
 		return counter->failcnt;
 	case RES_SOFT_LIMIT:
-		return (u64)memcg->soft_limit * PAGE_SIZE;
+		return (u64)READ_ONCE(memcg->soft_limit) * PAGE_SIZE;
 	default:
 		BUG();
 	}
@@ -3870,7 +3870,7 @@ static ssize_t mem_cgroup_write(struct kernfs_open_file *of,
 		if (IS_ENABLED(CONFIG_PREEMPT_RT)) {
 			ret = -EOPNOTSUPP;
 		} else {
-			memcg->soft_limit = nr_pages;
+			WRITE_ONCE(memcg->soft_limit, nr_pages);
 			ret = 0;
 		}
 		break;
-- 
2.17.1


WARNING: multiple messages have this Message-ID (diff)
From: Yue Zhao <findns94-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
To: akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org
Cc: roman.gushchin-fxUVXftIFDnyG1zEObXtfA@public.gmane.org,
	hannes-druUgvl0LCNAfugRpC6u6w@public.gmane.org,
	mhocko-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org,
	shakeelb-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org,
	muchun.song-fxUVXftIFDnyG1zEObXtfA@public.gmane.org,
	willy-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org,
	linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org,
	cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	tangyeechou-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org,
	Yue Zhao <findns94-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Subject: [PATCH v2, 4/4] mm, memcg: Prevent memory.soft_limit_in_bytes load/store tearing
Date: Mon,  6 Mar 2023 23:41:38 +0800	[thread overview]
Message-ID: <20230306154138.3775-5-findns94@gmail.com> (raw)
In-Reply-To: <20230306154138.3775-1-findns94-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>

The knob for cgroup v1 memory controller: memory.soft_limit_in_bytes
is not protected by any locking so it can be modified while it is used.
This is not an actual problem because races are unlikely.
But it is better to use READ_ONCE/WRITE_ONCE to prevent compiler from
doing anything funky.

The access of memcg->soft_limit is lockless,
so it can be concurrently set at the same time as we are
trying to read it.

Signed-off-by: Yue Zhao <findns94-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
---
 mm/memcontrol.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 26605b2f51b1..20566f59bbcb 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -3728,7 +3728,7 @@ static u64 mem_cgroup_read_u64(struct cgroup_subsys_state *css,
 	case RES_FAILCNT:
 		return counter->failcnt;
 	case RES_SOFT_LIMIT:
-		return (u64)memcg->soft_limit * PAGE_SIZE;
+		return (u64)READ_ONCE(memcg->soft_limit) * PAGE_SIZE;
 	default:
 		BUG();
 	}
@@ -3870,7 +3870,7 @@ static ssize_t mem_cgroup_write(struct kernfs_open_file *of,
 		if (IS_ENABLED(CONFIG_PREEMPT_RT)) {
 			ret = -EOPNOTSUPP;
 		} else {
-			memcg->soft_limit = nr_pages;
+			WRITE_ONCE(memcg->soft_limit, nr_pages);
 			ret = 0;
 		}
 		break;
-- 
2.17.1


  parent reply	other threads:[~2023-03-06 15:43 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-03-06 15:41 [PATCH v2, 0/4] mm, memcg: cgroup v1 and v2 tunable load/store tearing fixes Yue Zhao
2023-03-06 15:41 ` Yue Zhao
2023-03-06 15:41 ` [PATCH v2, 1/4] mm, memcg: Prevent memory.oom.group load/store tearing Yue Zhao
2023-03-06 15:41   ` Yue Zhao
2023-03-06 17:47   ` Michal Hocko
2023-03-06 17:47     ` Michal Hocko
2023-03-06 15:41 ` [PATCH v2, 2/4] mm, memcg: Prevent memory.swappiness " Yue Zhao
2023-03-06 17:50   ` Michal Hocko
2023-03-06 17:50     ` Michal Hocko
2023-03-06 15:41 ` [PATCH v2, 3/4] mm, memcg: Prevent memory.oom_control " Yue Zhao
2023-03-06 15:41   ` Yue Zhao
2023-03-06 17:53   ` Michal Hocko
2023-03-06 17:53     ` Michal Hocko
2023-03-08 14:51     ` Martin Zhao
2023-03-08 14:51       ` Martin Zhao
2023-03-06 15:41 ` Yue Zhao [this message]
2023-03-06 15:41   ` [PATCH v2, 4/4] mm, memcg: Prevent memory.soft_limit_in_bytes " Yue Zhao
2023-03-06 17:55   ` Michal Hocko
2023-03-06 17:55     ` Michal Hocko
2023-03-06 16:25 ` [PATCH v2, 0/4] mm, memcg: cgroup v1 and v2 tunable load/store tearing fixes Shakeel Butt
2023-03-06 16:25   ` Shakeel Butt
2023-03-06 17:51 ` Roman Gushchin
2023-03-06 17:51   ` 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=20230306154138.3775-5-findns94@gmail.com \
    --to=findns94@gmail.com \
    --cc=akpm@linux-foundation.org \
    --cc=cgroups@vger.kernel.org \
    --cc=hannes@cmpxchg.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mhocko@kernel.org \
    --cc=muchun.song@linux.dev \
    --cc=roman.gushchin@linux.dev \
    --cc=shakeelb@google.com \
    --cc=tangyeechou@gmail.com \
    --cc=willy@infradead.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.