From: Johannes Weiner <hannes@cmpxchg.org>
To: Yosry Ahmed <yosryahmed@google.com>
Cc: "Tejun Heo" <tj@kernel.org>, "Josef Bacik" <josef@toxicpanda.com>,
"Jens Axboe" <axboe@kernel.dk>,
"Zefan Li" <lizefan.x@bytedance.com>,
"Michal Hocko" <mhocko@kernel.org>,
"Roman Gushchin" <roman.gushchin@linux.dev>,
"Shakeel Butt" <shakeelb@google.com>,
"Muchun Song" <muchun.song@linux.dev>,
"Andrew Morton" <akpm@linux-foundation.org>,
"Michal Koutný" <mkoutny@suse.com>,
"Vasily Averin" <vasily.averin@linux.dev>,
cgroups@vger.kernel.org, linux-block@vger.kernel.org,
linux-kernel@vger.kernel.org, linux-mm@kvack.org,
bpf@vger.kernel.org
Subject: Re: [PATCH v1 5/9] memcg: replace stats_flush_lock with an atomic
Date: Tue, 28 Mar 2023 13:53:22 -0400 [thread overview]
Message-ID: <ZCMpklJZqwWHro0u@cmpxchg.org> (raw)
In-Reply-To: <20230328061638.203420-6-yosryahmed@google.com>
On Tue, Mar 28, 2023 at 06:16:34AM +0000, Yosry Ahmed wrote:
> As Johannes notes in [1], stats_flush_lock is currently used to:
> (a) Protect updated to stats_flush_threshold.
> (b) Protect updates to flush_next_time.
> (c) Serializes calls to cgroup_rstat_flush() based on those ratelimits.
>
> However:
>
> 1. stats_flush_threshold is already an atomic
>
> 2. flush_next_time is not atomic. The writer is locked, but the reader
> is lockless. If the reader races with a flush, you could see this:
>
> if (time_after(jiffies, flush_next_time))
> spin_trylock()
> flush_next_time = now + delay
> flush()
> spin_unlock()
> spin_trylock()
> flush_next_time = now + delay
> flush()
> spin_unlock()
>
> which means we already can get flushes at a higher frequency than
> FLUSH_TIME during races. But it isn't really a problem.
>
> The reader could also see garbled partial updates, so it needs at
> least READ_ONCE and WRITE_ONCE protection.
>
> 3. Serializing cgroup_rstat_flush() calls against the ratelimit
> factors is currently broken because of the race in 2. But the race
> is actually harmless, all we might get is the occasional earlier
> flush. If there is no delta, the flush won't do much. And if there
> is, the flush is justified.
>
> So the lock can be removed all together. However, the lock also served
> the purpose of preventing a thundering herd problem for concurrent
> flushers, see [2]. Use an atomic instead to serve the purpose of
> unifying concurrent flushers.
>
> [1]https://lore.kernel.org/lkml/20230323172732.GE739026@cmpxchg.org/
> [2]https://lore.kernel.org/lkml/20210716212137.1391164-2-shakeelb@google.com/
>
> Signed-off-by: Yosry Ahmed <yosryahmed@google.com>
With Shakeel's suggestion:
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
next prev parent reply other threads:[~2023-03-28 17:53 UTC|newest]
Thread overview: 41+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-03-28 6:16 [PATCH v1 0/9] memcg: make rstat flushing irq and sleep friendly Yosry Ahmed
2023-03-28 6:16 ` [PATCH v1 1/9] cgroup: rename cgroup_rstat_flush_"irqsafe" to "atomic" Yosry Ahmed
2023-03-28 13:24 ` Shakeel Butt
2023-03-28 17:42 ` Johannes Weiner
2023-03-28 6:16 ` [PATCH v1 2/9] memcg: rename mem_cgroup_flush_stats_"delayed" to "ratelimited" Yosry Ahmed
2023-03-28 13:25 ` Shakeel Butt
2023-03-28 17:42 ` Johannes Weiner
2023-03-28 6:16 ` [PATCH v1 3/9] memcg: do not flush stats in irq context Yosry Ahmed
2023-03-28 13:26 ` Shakeel Butt
2023-03-28 17:43 ` Johannes Weiner
2023-03-28 6:16 ` [PATCH v1 4/9] cgroup: rstat: add WARN_ON_ONCE() if flushing outside task context Yosry Ahmed
2023-03-28 14:59 ` Shakeel Butt
2023-03-28 17:49 ` Johannes Weiner
2023-03-28 18:59 ` Yosry Ahmed
2023-03-28 22:18 ` Yosry Ahmed
2023-03-28 6:16 ` [PATCH v1 5/9] memcg: replace stats_flush_lock with an atomic Yosry Ahmed
2023-03-28 14:15 ` Shakeel Butt
2023-03-28 18:52 ` Yosry Ahmed
2023-03-28 19:28 ` Shakeel Butt
2023-03-28 19:34 ` Yosry Ahmed
2023-03-28 19:42 ` Yosry Ahmed
2023-03-28 17:53 ` Johannes Weiner [this message]
2023-03-28 6:16 ` [PATCH v1 6/9] memcg: sleep during flushing stats in safe contexts Yosry Ahmed
2023-03-28 15:09 ` Shakeel Butt
2023-03-28 18:35 ` Johannes Weiner
2023-03-28 18:45 ` Yosry Ahmed
2023-03-28 19:06 ` Johannes Weiner
2023-03-28 19:26 ` Yosry Ahmed
2023-03-28 6:16 ` [PATCH v1 7/9] workingset: memcg: sleep when flushing stats in workingset_refault() Yosry Ahmed
2023-03-28 15:18 ` Shakeel Butt
2023-03-28 18:47 ` Johannes Weiner
2023-03-28 19:25 ` Yosry Ahmed
2023-03-28 18:43 ` Johannes Weiner
2023-03-28 6:16 ` [PATCH v1 8/9] vmscan: memcg: sleep when flushing stats during reclaim Yosry Ahmed
2023-03-28 15:19 ` Shakeel Butt
2023-03-28 19:01 ` Yosry Ahmed
2023-03-28 19:29 ` Shakeel Butt
2023-03-28 18:49 ` Johannes Weiner
2023-03-28 6:16 ` [PATCH v1 9/9] memcg: do not modify rstat tree for zero updates Yosry Ahmed
2023-03-28 15:20 ` Shakeel Butt
2023-03-28 18:50 ` Johannes Weiner
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=ZCMpklJZqwWHro0u@cmpxchg.org \
--to=hannes@cmpxchg.org \
--cc=akpm@linux-foundation.org \
--cc=axboe@kernel.dk \
--cc=bpf@vger.kernel.org \
--cc=cgroups@vger.kernel.org \
--cc=josef@toxicpanda.com \
--cc=linux-block@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=lizefan.x@bytedance.com \
--cc=mhocko@kernel.org \
--cc=mkoutny@suse.com \
--cc=muchun.song@linux.dev \
--cc=roman.gushchin@linux.dev \
--cc=shakeelb@google.com \
--cc=tj@kernel.org \
--cc=vasily.averin@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;
as well as URLs for NNTP newsgroup(s).