From: Roman Gushchin <guro@fb.com>
To: Michal Hocko <mhocko@suse.com>
Cc: Miaohe Lin <linmiaohe@huawei.com>,
"hannes@cmpxchg.org" <hannes@cmpxchg.org>,
"vdavydov.dev@gmail.com" <vdavydov.dev@gmail.com>,
"akpm@linux-foundation.org" <akpm@linux-foundation.org>,
"shakeelb@google.com" <shakeelb@google.com>,
"willy@infradead.org" <willy@infradead.org>,
"alexs@kernel.org" <alexs@kernel.org>,
"richard.weiyang@gmail.com" <richard.weiyang@gmail.com>,
"songmuchun@bytedance.com" <songmuchun@bytedance.com>,
"linux-mm@kvack.org" <linux-mm@kvack.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"cgroups@vger.kernel.org" <cgroups@vger.kernel.org>
Subject: Re: [PATCH 2/5] mm, memcg: narrow the scope of percpu_charge_mutex
Date: Tue, 3 Aug 2021 07:13:35 +0000 [thread overview]
Message-ID: <898130C3-FDF2-41C2-81DD-D33721C977C6@fb.com> (raw)
In-Reply-To: <YQjsHTW+46bG9XsV@dhcp22.suse.cz>
I’d go with atomic_dec().
Sent from my iPhone
> On Aug 3, 2021, at 00:11, Michal Hocko <mhocko@suse.com> wrote:
>
> On Tue 03-08-21 14:29:13, Miaohe Lin wrote:
> [...]
>> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
>> index 616d1a72ece3..6210b1124929 100644
>> --- a/mm/memcontrol.c
>> +++ b/mm/memcontrol.c
>> @@ -2208,11 +2208,11 @@ static void refill_stock(struct mem_cgroup *memcg, unsigned int nr_pages)
>> */
>> static void drain_all_stock(struct mem_cgroup *root_memcg)
>> {
>> - static DEFINE_MUTEX(percpu_charge_mutex);
>> int cpu, curcpu;
>> + static atomic_t drainer = ATOMIC_INIT(0);
>>
>> /* If someone's already draining, avoid adding running more workers. */
>> - if (!mutex_trylock(&percpu_charge_mutex))
>> + if (atomic_cmpxchg(&drainer, 0, 1) != 0)
>> return;
>> /*
>> * Notify other cpus that system-wide "drain" is running
>> @@ -2244,7 +2244,7 @@ static void drain_all_stock(struct mem_cgroup *root_memcg)
>> }
>> }
>> put_cpu();
>> - mutex_unlock(&percpu_charge_mutex);
>> + atomic_set(&drainer, 0);
>
> atomic_set doesn't imply memory barrier IIRC. Is this safe?
>
> --
> Michal Hocko
> SUSE Labs
WARNING: multiple messages have this Message-ID (diff)
From: Roman Gushchin <guro-b10kYP2dOMg@public.gmane.org>
To: Michal Hocko <mhocko-IBi9RG/b67k@public.gmane.org>
Cc: Miaohe Lin <linmiaohe-hv44wF8Li93QT0dZR+AlfA@public.gmane.org>,
"hannes-druUgvl0LCNAfugRpC6u6w@public.gmane.org"
<hannes-druUgvl0LCNAfugRpC6u6w@public.gmane.org>,
"vdavydov.dev-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org"
<vdavydov.dev-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>,
"akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org"
<akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org>,
"shakeelb-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org"
<shakeelb-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>,
"willy-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org"
<willy-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>,
"alexs-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org"
<alexs-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>,
"richard.weiyang-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org"
<richard.weiyang-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>,
"songmuchun-EC8Uxl6Npydl57MIdRCFDg@public.gmane.org"
<songmuchun-EC8Uxl6Npydl57MIdRCFDg@public.gmane.org>,
"linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org"
<linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org>,
"linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org"
<linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
"cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org"
<cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>
Subject: Re: [PATCH 2/5] mm, memcg: narrow the scope of percpu_charge_mutex
Date: Tue, 3 Aug 2021 07:13:35 +0000 [thread overview]
Message-ID: <898130C3-FDF2-41C2-81DD-D33721C977C6@fb.com> (raw)
In-Reply-To: <YQjsHTW+46bG9XsV-2MMpYkNvuYDjFM9bn6wA6Q@public.gmane.org>
I’d go with atomic_dec().
Sent from my iPhone
> On Aug 3, 2021, at 00:11, Michal Hocko <mhocko@suse.com> wrote:
>
> On Tue 03-08-21 14:29:13, Miaohe Lin wrote:
> [...]
>> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
>> index 616d1a72ece3..6210b1124929 100644
>> --- a/mm/memcontrol.c
>> +++ b/mm/memcontrol.c
>> @@ -2208,11 +2208,11 @@ static void refill_stock(struct mem_cgroup *memcg, unsigned int nr_pages)
>> */
>> static void drain_all_stock(struct mem_cgroup *root_memcg)
>> {
>> - static DEFINE_MUTEX(percpu_charge_mutex);
>> int cpu, curcpu;
>> + static atomic_t drainer = ATOMIC_INIT(0);
>>
>> /* If someone's already draining, avoid adding running more workers. */
>> - if (!mutex_trylock(&percpu_charge_mutex))
>> + if (atomic_cmpxchg(&drainer, 0, 1) != 0)
>> return;
>> /*
>> * Notify other cpus that system-wide "drain" is running
>> @@ -2244,7 +2244,7 @@ static void drain_all_stock(struct mem_cgroup *root_memcg)
>> }
>> }
>> put_cpu();
>> - mutex_unlock(&percpu_charge_mutex);
>> + atomic_set(&drainer, 0);
>
> atomic_set doesn't imply memory barrier IIRC. Is this safe?
>
> --
> Michal Hocko
> SUSE Labs
next prev parent reply other threads:[~2021-08-03 7:14 UTC|newest]
Thread overview: 97+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-07-29 12:57 [PATCH 0/5] Cleanups and fixup for memcontrol Miaohe Lin
2021-07-29 12:57 ` Miaohe Lin
2021-07-29 12:57 ` [PATCH 1/5] mm, memcg: remove unused functions Miaohe Lin
2021-07-29 12:57 ` Miaohe Lin
2021-07-29 14:07 ` Shakeel Butt
2021-07-29 14:07 ` Shakeel Butt
2021-07-29 14:07 ` Shakeel Butt
2021-07-30 2:39 ` Muchun Song
2021-07-30 2:39 ` Muchun Song
2021-07-30 2:39 ` Muchun Song
2021-07-30 2:57 ` Roman Gushchin
2021-07-30 2:57 ` Roman Gushchin
2021-07-30 6:45 ` Michal Hocko
2021-07-30 6:45 ` Michal Hocko
2021-07-29 12:57 ` [PATCH 2/5] mm, memcg: narrow the scope of percpu_charge_mutex Miaohe Lin
2021-07-29 12:57 ` Miaohe Lin
2021-07-30 2:42 ` Muchun Song
2021-07-30 2:42 ` Muchun Song
2021-07-30 2:42 ` Muchun Song
2021-07-30 3:06 ` Roman Gushchin
2021-07-30 3:06 ` Roman Gushchin
2021-07-30 6:50 ` Michal Hocko
2021-07-30 6:50 ` Michal Hocko
2021-07-31 2:29 ` Miaohe Lin
2021-07-31 2:29 ` Miaohe Lin
2021-08-02 6:49 ` Michal Hocko
2021-08-02 6:49 ` Michal Hocko
2021-08-02 9:54 ` Miaohe Lin
2021-08-02 9:54 ` Miaohe Lin
2021-08-03 3:40 ` Roman Gushchin
2021-08-03 3:40 ` Roman Gushchin
2021-08-03 6:29 ` Miaohe Lin
2021-08-03 6:29 ` Miaohe Lin
2021-08-03 7:11 ` Michal Hocko
2021-08-03 7:11 ` Michal Hocko
2021-08-03 7:13 ` Roman Gushchin [this message]
2021-08-03 7:13 ` Roman Gushchin
2021-08-03 7:27 ` Michal Hocko
2021-08-03 7:27 ` Michal Hocko
2021-08-03 9:33 ` Muchun Song
2021-08-03 9:33 ` Muchun Song
2021-08-03 9:33 ` Muchun Song
2021-08-03 10:50 ` Miaohe Lin
2021-08-03 10:50 ` Miaohe Lin
2021-08-03 14:15 ` Johannes Weiner
2021-08-03 14:15 ` Johannes Weiner
2021-08-04 8:20 ` Michal Hocko
2021-08-04 8:20 ` Michal Hocko
2021-08-05 1:44 ` Miaohe Lin
2021-08-05 1:44 ` Miaohe Lin
2021-07-30 6:46 ` Michal Hocko
2021-07-29 12:57 ` [PATCH 3/5] mm, memcg: save some atomic ops when flush is already true Miaohe Lin
2021-07-29 12:57 ` Miaohe Lin
2021-07-29 14:40 ` Shakeel Butt
2021-07-29 14:40 ` Shakeel Butt
2021-07-29 14:40 ` Shakeel Butt
2021-07-30 2:37 ` Muchun Song
2021-07-30 2:37 ` Muchun Song
2021-07-30 2:37 ` Muchun Song
2021-07-30 3:07 ` Roman Gushchin
2021-07-30 3:07 ` Roman Gushchin
2021-07-30 6:51 ` Michal Hocko
2021-07-30 6:51 ` Michal Hocko
2021-07-29 12:57 ` [PATCH 4/5] mm, memcg: avoid possible NULL pointer dereferencing in mem_cgroup_init() Miaohe Lin
2021-07-29 12:57 ` Miaohe Lin
2021-07-29 13:52 ` Matthew Wilcox
2021-07-29 13:52 ` Matthew Wilcox
2021-07-30 1:50 ` Miaohe Lin
2021-07-30 1:50 ` Miaohe Lin
2021-07-30 3:12 ` Roman Gushchin
2021-07-30 3:12 ` Roman Gushchin
2021-07-30 6:29 ` Miaohe Lin
2021-07-30 6:29 ` Miaohe Lin
2021-07-30 6:44 ` Michal Hocko
2021-07-30 6:44 ` Michal Hocko
2021-07-31 2:05 ` Miaohe Lin
2021-07-31 2:05 ` Miaohe Lin
2021-08-02 6:43 ` Michal Hocko
2021-08-02 6:43 ` Michal Hocko
2021-08-02 10:00 ` Miaohe Lin
2021-08-02 10:00 ` Miaohe Lin
2021-08-02 10:42 ` Michal Hocko
2021-08-02 10:42 ` Michal Hocko
2021-08-02 11:18 ` Miaohe Lin
2021-08-02 11:18 ` Miaohe Lin
2021-07-29 12:57 ` [PATCH 5/5] mm, memcg: always call __mod_node_page_state() with preempt disabled Miaohe Lin
2021-07-29 12:57 ` Miaohe Lin
2021-07-29 14:39 ` Shakeel Butt
2021-07-29 14:39 ` Shakeel Butt
2021-07-29 14:39 ` Shakeel Butt
2021-07-30 1:52 ` Miaohe Lin
2021-07-30 1:52 ` Miaohe Lin
2021-07-30 2:33 ` [External] " Muchun Song
2021-07-30 2:33 ` Muchun Song
2021-07-30 2:33 ` Muchun Song
2021-07-30 2:46 ` Miaohe Lin
2021-07-30 2:46 ` Miaohe Lin
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=898130C3-FDF2-41C2-81DD-D33721C977C6@fb.com \
--to=guro@fb.com \
--cc=akpm@linux-foundation.org \
--cc=alexs@kernel.org \
--cc=cgroups@vger.kernel.org \
--cc=hannes@cmpxchg.org \
--cc=linmiaohe@huawei.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@suse.com \
--cc=richard.weiyang@gmail.com \
--cc=shakeelb@google.com \
--cc=songmuchun@bytedance.com \
--cc=vdavydov.dev@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.