linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Kaiyang Zhao <kaiyang2@cs.cmu.edu>
To: David Rientjes <rientjes@google.com>
Cc: linux-mm@kvack.org, cgroups@vger.kernel.org,
	roman.gushchin@linux.dev, shakeel.butt@linux.dev,
	muchun.song@linux.dev, akpm@linux-foundation.org,
	mhocko@kernel.org, nehagholkar@meta.com, abhishekd@meta.com,
	hannes@cmpxchg.org
Subject: Re: [PATCH] mm,memcg: provide per-cgroup counters for NUMA balancing operations
Date: Tue, 13 Aug 2024 18:21:20 +0000	[thread overview]
Message-ID: <ZrukILyQhMAKWwTe@localhost.localhost> (raw)
In-Reply-To: <284406af-56b8-4b66-750f-10f9d38cfac7@google.com>

On Mon, Aug 12, 2024 at 10:09:55PM -0700, David Rientjes wrote:
> Does this include top-tier specific memory limits as well?

Yes, we plan to have top-tier specific memory protection (like memory.low)
and memory limit (like memory.high). Exactly what the interface will
look like will probably depends on the community feedback. Maybe it's
just adding a memory_top_tier.low/high knob, maybe something else.

> And is your primary motivation the promotion path through NUMA Balancing 
> or are you also looking at demotion to develop a comprehensive policy for 
> memory placement using these limits?

Both promotion and demotion/reclaim path. If we want top-tier memory
protection and limit, moderating both promotion and demotion is I think
a must for converging on a stable distribution of memory usage on top-tier
and slower tier.

Kaiyang


  reply	other threads:[~2024-08-13 18:21 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-08-09 21:21 kaiyang2
2024-08-10  0:28 ` Andrew Morton
2024-08-11 20:16 ` David Rientjes
2024-08-12 22:49   ` Kaiyang Zhao
2024-08-13  5:09     ` David Rientjes
2024-08-13 18:21       ` Kaiyang Zhao [this message]
2024-08-14 20:48     ` Andrew Morton

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=ZrukILyQhMAKWwTe@localhost.localhost \
    --to=kaiyang2@cs.cmu.edu \
    --cc=abhishekd@meta.com \
    --cc=akpm@linux-foundation.org \
    --cc=cgroups@vger.kernel.org \
    --cc=hannes@cmpxchg.org \
    --cc=linux-mm@kvack.org \
    --cc=mhocko@kernel.org \
    --cc=muchun.song@linux.dev \
    --cc=nehagholkar@meta.com \
    --cc=rientjes@google.com \
    --cc=roman.gushchin@linux.dev \
    --cc=shakeel.butt@linux.dev \
    /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