From: Johannes Weiner <hannes-druUgvl0LCNAfugRpC6u6w@public.gmane.org>
To: Mina Almasry <almasrymina-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>
Cc: Huang Ying <ying.huang-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>,
Yang Shi
<yang.shi-KPsoFbNs7GizrGE5bRqYAgC/G2K4zDHf@public.gmane.org>,
Yosry Ahmed <yosryahmed-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>,
Tim Chen <tim.c.chen-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>,
weixugc-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org,
shakeelb-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org,
gthelen-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org,
fvdl-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org,
Michal Hocko <mhocko-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>,
Roman Gushchin
<roman.gushchin-fxUVXftIFDnyG1zEObXtfA@public.gmane.org>,
Muchun Song <songmuchun-EC8Uxl6Npydl57MIdRCFDg@public.gmane.org>,
Andrew Morton
<akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org>,
linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org
Subject: Re: [RFC PATCH V1] mm: Disable demotion from proactive reclaim
Date: Wed, 23 Nov 2022 13:00:35 -0500 [thread overview]
Message-ID: <Y35fw2JSAeAddONg@cmpxchg.org> (raw)
In-Reply-To: <20221122203850.2765015-1-almasrymina-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>
Hello Mina,
On Tue, Nov 22, 2022 at 12:38:45PM -0800, Mina Almasry wrote:
> Since commit 3f1509c57b1b ("Revert "mm/vmscan: never demote for memcg
> reclaim""), the proactive reclaim interface memory.reclaim does both
> reclaim and demotion. This is likely fine for us for latency critical
> jobs where we would want to disable proactive reclaim entirely, and is
> also fine for latency tolerant jobs where we would like to both
> proactively reclaim and demote.
>
> However, for some latency tiers in the middle we would like to demote but
> not reclaim. This is because reclaim and demotion incur different latency
> costs to the jobs in the cgroup. Demoted memory would still be addressable
> by the userspace at a higher latency, but reclaimed memory would need to
> incur a pagefault.
>
> To address this, I propose having reclaim-only and demotion-only
> mechanisms in the kernel. There are a couple possible
> interfaces to carry this out I considered:
>
> 1. Disable demotion in the memory.reclaim interface and add a new
> demotion interface (memory.demote).
> 2. Extend memory.reclaim with a "demote=<int>" flag to configure the demotion
> behavior in the kernel like so:
> - demote=0 would disable demotion from this call.
> - demote=1 would allow the kernel to demote if it desires.
> - demote=2 would only demote if possible but not attempt any
> other form of reclaim.
Unfortunately, our proactive reclaim stack currently relies on
memory.reclaim doing both. It may not stay like that, but I'm a bit
wary of changing user-visible semantics post-facto.
In patch 2, you're adding a node interface to memory.demote. Can you
add this to memory.reclaim instead? This would allow you to control
demotion and reclaim independently as you please: if you call it on a
node with demotion targets, it will demote; if you call it on a node
without one, it'll reclaim. And current users will remain unaffected.
next prev parent reply other threads:[~2022-11-23 18:00 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-11-22 20:38 [RFC PATCH V1] mm: Disable demotion from proactive reclaim Mina Almasry
[not found] ` <20221122203850.2765015-1-almasrymina-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>
2022-11-22 20:38 ` [RFC PATCH v1] mm: Add memory.demote for proactive demotion only Mina Almasry
2022-11-22 20:38 ` [RFC PATCH v1 3/4] mm: Fix demotion-only scanning anon pages Mina Almasry
[not found] ` <20221122203850.2765015-3-almasrymina-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>
2022-11-24 5:27 ` Huang, Ying
2022-11-22 20:38 ` [RFC PATCH v1 4/4] mm: Add nodes= arg to memory.demote Mina Almasry
2022-11-23 18:00 ` Johannes Weiner [this message]
[not found] ` <Y35fw2JSAeAddONg-druUgvl0LCNAfugRpC6u6w@public.gmane.org>
2022-11-23 21:20 ` [RFC PATCH V1] mm: Disable demotion from proactive reclaim Mina Almasry
[not found] ` <CAHS8izN+xqM67XLT4y5qyYnGQMUWRQCJrdvf2gjTHd8nZ_=0sw-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2022-11-23 21:35 ` Yosry Ahmed
[not found] ` <CAJD7tkZNW=u1TD-Fd_3RuzRNtaFjxihbGm0836QHkdp0Nn-vyQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2022-11-23 22:30 ` Johannes Weiner
2022-11-23 23:47 ` Yosry Ahmed
2022-11-23 21:58 ` Johannes Weiner
[not found] ` <Y36XchdgTCsMP4jT-druUgvl0LCNAfugRpC6u6w@public.gmane.org>
2022-11-23 22:37 ` Mina Almasry
2022-11-24 5:51 ` Huang, Ying
[not found] ` <874juonbmv.fsf-fFUE1NP8JkzwuUmzmnQr+vooFf0ArEBIu+b9c/7xato@public.gmane.org>
2022-11-28 22:24 ` Yang Shi
[not found] ` <CAHbLzkrmxyzH4R7a9sJQavrUyKCEiNYeA543+sdJLsgRPrwBwQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2022-11-29 0:53 ` Huang, Ying
[not found] ` <87a64ad1iz.fsf-fFUE1NP8JkzwuUmzmnQr+vooFf0ArEBIu+b9c/7xato@public.gmane.org>
2022-11-29 17:27 ` Yang Shi
[not found] ` <CAHbLzkpVZf-3K0Ys8HG8x6D_XpPChB-H2XMYar7UwnNDeMiw8w-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2022-11-30 5:31 ` Huang, Ying
[not found] ` <87ilixatyw.fsf-fFUE1NP8JkzwuUmzmnQr+vooFf0ArEBIu+b9c/7xato@public.gmane.org>
2022-11-30 18:49 ` Yang Shi
[not found] ` <CAHbLzkr_njh2xtAf6RME_Fe0TgTKdC4mcsUe24orqVScjibUrA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2022-12-01 1:51 ` Huang, Ying
[not found] ` <87h6yfao37.fsf-fFUE1NP8JkzwuUmzmnQr+vooFf0ArEBIu+b9c/7xato@public.gmane.org>
2022-12-01 22:45 ` Yang Shi
[not found] ` <CAHbLzkr9k8fvBGVskN1sMJiLX_JkWW7OrrscUrA0xASh+rYN7Q-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2022-12-02 1:57 ` Huang, Ying
2022-11-29 18:08 ` Johannes Weiner
[not found] ` <Y4ZKgXdUXZ3ycMp4-druUgvl0LCNAfugRpC6u6w@public.gmane.org>
2022-11-30 3:55 ` Huang, Ying
[not found] ` <87wn7dayfz.fsf-fFUE1NP8JkzwuUmzmnQr+vooFf0ArEBIu+b9c/7xato@public.gmane.org>
2022-12-01 20:40 ` Mina Almasry
[not found] ` <CAHS8izNc=NeMcNudOB0r4aYbZURpU+Y9xpek_KpO=55R+7ec5g-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2022-12-02 2:01 ` Huang, Ying
[not found] ` <87tu2e36nw.fsf-fFUE1NP8JkzwuUmzmnQr+vooFf0ArEBIu+b9c/7xato@public.gmane.org>
2022-12-02 2:06 ` Mina Almasry
2022-11-30 2:14 ` Mina Almasry
[not found] ` <CAHS8izMhoHVCGxXGt8qRtf-fPpAR8=pTy7Rc3j2=Wf8vJz-C+g-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2022-11-30 5:39 ` Huang, Ying
[not found] ` <87edtlatmg.fsf-fFUE1NP8JkzwuUmzmnQr+vooFf0ArEBIu+b9c/7xato@public.gmane.org>
2022-11-30 6:06 ` Mina Almasry
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=Y35fw2JSAeAddONg@cmpxchg.org \
--to=hannes-druugvl0lcnafugrpc6u6w@public.gmane.org \
--cc=akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org \
--cc=almasrymina-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org \
--cc=cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=fvdl-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org \
--cc=gthelen-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org \
--cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org \
--cc=mhocko-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org \
--cc=roman.gushchin-fxUVXftIFDnyG1zEObXtfA@public.gmane.org \
--cc=shakeelb-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org \
--cc=songmuchun-EC8Uxl6Npydl57MIdRCFDg@public.gmane.org \
--cc=tim.c.chen-VuQAYsv1563Yd54FQh9/CA@public.gmane.org \
--cc=weixugc-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org \
--cc=yang.shi-KPsoFbNs7GizrGE5bRqYAgC/G2K4zDHf@public.gmane.org \
--cc=ying.huang-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org \
--cc=yosryahmed-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox