From: Michal Hocko <mhocko@suse.com>
To: Peng Zhang <zhangpeng362@huawei.com>
Cc: linux-mm@kvack.org, shakeelb@google.com,
akpm@linux-foundation.org, wangkefeng.wang@huawei.com,
sunnanyong@huawei.com
Subject: Re: [PATCH 2/2] mm: oom: terminate the oom_evaluate_task() loop early
Date: Wed, 16 Aug 2023 08:48:20 +0200 [thread overview]
Message-ID: <ZNxxNOgtP0iXHzld@dhcp22.suse.cz> (raw)
In-Reply-To: <20230814063428.4111206-3-zhangpeng362@huawei.com>
On Mon 14-08-23 14:34:28, Peng Zhang wrote:
> From: ZhangPeng <zhangpeng362@huawei.com>
>
> If task is allocating a lot of memory and has been marked to be killed
> first, then it gets the highest score (LONG_MAX). Therefore, there is no
> need to continue to calculate the points of other tasks. Just terminate
> the oom_evaluate_task() loop early, when the task with the highest score
> is found. By doing this, we can get some performance gains in
> select_bad_process().
The point of evaluating all tasks is that there might be pre-existing
oom victim still being torn down. If you cut this short you might kill
more tasks than necessary.
So I do not think we want this patch.
> To implement it, the return value of oom_evaluate_task() is modified.
> When the task with the highest score is found (points == LONG_MAX),
> oom_evaluate_task() will return 1 and the loop will terminate early.
>
> Signed-off-by: ZhangPeng <zhangpeng362@huawei.com>
> ---
> mm/oom_kill.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/mm/oom_kill.c b/mm/oom_kill.c
> index 10f7826c4035..02d11ae043aa 100644
> --- a/mm/oom_kill.c
> +++ b/mm/oom_kill.c
> @@ -346,6 +346,8 @@ static int oom_evaluate_task(struct task_struct *task, void *arg)
> get_task_struct(task);
> oc->chosen = task;
> oc->chosen_points = points;
> + if (points == LONG_MAX)
> + return 1;
> next:
> return 0;
> abort:
> --
> 2.25.1
--
Michal Hocko
SUSE Labs
next prev parent reply other threads:[~2023-08-16 6:48 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-08-14 6:34 [PATCH 0/2] " Peng Zhang
2023-08-14 6:34 ` [PATCH 1/2] mm: oom: remove unnecessary goto in oom_evaluate_task() Peng Zhang
2023-08-16 6:46 ` Michal Hocko
2023-08-16 8:56 ` zhangpeng (AS)
2023-08-14 6:34 ` [PATCH 2/2] mm: oom: terminate the oom_evaluate_task() loop early Peng Zhang
2023-08-16 6:48 ` Michal Hocko [this message]
2023-08-16 8:57 ` zhangpeng (AS)
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=ZNxxNOgtP0iXHzld@dhcp22.suse.cz \
--to=mhocko@suse.com \
--cc=akpm@linux-foundation.org \
--cc=linux-mm@kvack.org \
--cc=shakeelb@google.com \
--cc=sunnanyong@huawei.com \
--cc=wangkefeng.wang@huawei.com \
--cc=zhangpeng362@huawei.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