From: Jason Gunthorpe <jgg-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>
To: Alistair Popple <apopple-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>
Cc: Chris Li <chrisl-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>,
"T.J. Mercier"
<tjmercier-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>,
lsf-pc-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org,
linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org,
cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
Yosry Ahmed <yosryahmed-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>,
Tejun Heo <tj-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>,
Shakeel Butt <shakeelb-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>,
Muchun Song <muchun.song-fxUVXftIFDnyG1zEObXtfA@public.gmane.org>,
Johannes Weiner <hannes-druUgvl0LCNAfugRpC6u6w@public.gmane.org>,
Roman Gushchin
<roman.gushchin-fxUVXftIFDnyG1zEObXtfA@public.gmane.org>,
Kalesh Singh
<kaleshsingh-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>,
Yu Zhao <yuzhao-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>
Subject: Re: [LSF/MM/BPF TOPIC] Reducing zombie memcgs
Date: Fri, 12 May 2023 18:09:20 -0300 [thread overview]
Message-ID: <ZF6rACJzilA06oe+@nvidia.com> (raw)
In-Reply-To: <87jzxe9baj.fsf-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>
On Fri, May 12, 2023 at 06:45:13PM +1000, Alistair Popple wrote:
> However review comments suggested it needed to be added as part of
> memcg. As soon as we do that we have to address how we deal with shared
> memory. If we stick with the original RLIMIT proposal this discussion
> goes away, but based on feedback I think I need to at least investigate
> integrating it into memcg to get anything merged.
Personally I don't see how we can effectively solve the per-page
problem without also tracking all the owning memcgs for every
page. This means giving each struct page an array of memcgs
I suspect this will be too expensive to be realistically
implementable.
If it is done then we may not even need a pin controller on its own as
the main memcg should capture most of it. (althought it doesn't
distinguish between movable/swappable and non-swappable memory)
But this is all being done for the libvirt people, so it would be good
to involve them
Jason
WARNING: multiple messages have this Message-ID (diff)
From: Jason Gunthorpe <jgg@nvidia.com>
To: Alistair Popple <apopple@nvidia.com>
Cc: Chris Li <chrisl@kernel.org>,
"T.J. Mercier" <tjmercier@google.com>,
lsf-pc@lists.linux-foundation.org, linux-mm@kvack.org,
cgroups@vger.kernel.org, Yosry Ahmed <yosryahmed@google.com>,
Tejun Heo <tj@kernel.org>, Shakeel Butt <shakeelb@google.com>,
Muchun Song <muchun.song@linux.dev>,
Johannes Weiner <hannes@cmpxchg.org>,
Roman Gushchin <roman.gushchin@linux.dev>,
Kalesh Singh <kaleshsingh@google.com>,
Yu Zhao <yuzhao@google.com>
Subject: Re: [LSF/MM/BPF TOPIC] Reducing zombie memcgs
Date: Fri, 12 May 2023 18:09:20 -0300 [thread overview]
Message-ID: <ZF6rACJzilA06oe+@nvidia.com> (raw)
In-Reply-To: <87jzxe9baj.fsf@nvidia.com>
On Fri, May 12, 2023 at 06:45:13PM +1000, Alistair Popple wrote:
> However review comments suggested it needed to be added as part of
> memcg. As soon as we do that we have to address how we deal with shared
> memory. If we stick with the original RLIMIT proposal this discussion
> goes away, but based on feedback I think I need to at least investigate
> integrating it into memcg to get anything merged.
Personally I don't see how we can effectively solve the per-page
problem without also tracking all the owning memcgs for every
page. This means giving each struct page an array of memcgs
I suspect this will be too expensive to be realistically
implementable.
If it is done then we may not even need a pin controller on its own as
the main memcg should capture most of it. (althought it doesn't
distinguish between movable/swappable and non-swappable memory)
But this is all being done for the libvirt people, so it would be good
to involve them
Jason
next prev parent reply other threads:[~2023-05-12 21:09 UTC|newest]
Thread overview: 51+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-04-11 23:36 [LSF/MM/BPF TOPIC] Reducing zombie memcgs T.J. Mercier
2023-04-11 23:36 ` T.J. Mercier
[not found] ` <CABdmKX2M6koq4Q0Cmp_-=wbP0Qa190HdEGGaHfxNS05gAkUtPA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2023-04-11 23:48 ` Yosry Ahmed
2023-04-11 23:48 ` Yosry Ahmed
[not found] ` <CAJD7tkZw9uVPe5KH2xrihsv5nDmExJmkmsUPYP6Npvv6Q0NcVw-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2023-04-25 11:36 ` Yosry Ahmed
2023-04-25 11:36 ` Yosry Ahmed
[not found] ` <CAJD7tkb56gR0X5v3VHfmk3az3bOz=wF2jhEi+7Eek0J8XXBeWQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2023-04-25 18:42 ` Waiman Long
2023-04-25 18:42 ` Waiman Long
[not found] ` <27e15be8-d0eb-ed32-a0ec-5ec9b59f1f27-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2023-04-25 18:53 ` Yosry Ahmed
2023-04-25 18:53 ` Yosry Ahmed
[not found] ` <CAJD7tkb1W0bP3AU9KepOYPx-AD-fMKSfUhj_Cmth63RS9umMsg-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2023-04-26 20:15 ` Waiman Long
2023-04-26 20:15 ` Waiman Long
2023-05-01 16:38 ` Roman Gushchin
2023-05-01 16:38 ` Roman Gushchin
2023-05-02 7:18 ` Yosry Ahmed
2023-05-02 7:18 ` Yosry Ahmed
2023-05-02 20:02 ` Yosry Ahmed
2023-05-02 20:02 ` Yosry Ahmed
2023-05-03 22:15 ` Chris Li
2023-05-03 22:15 ` Chris Li
[not found] ` <ZFLdDyHoIdJSXJt+-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>
2023-05-04 11:58 ` Alistair Popple
2023-05-04 11:58 ` Alistair Popple
[not found] ` <874josz4rd.fsf-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>
2023-05-04 15:31 ` Chris Li
2023-05-04 15:31 ` Chris Li
[not found] ` <ZFPP71czDDxMPLQK-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>
2023-05-05 13:53 ` Alistair Popple
2023-05-05 13:53 ` Alistair Popple
[not found] ` <877ctm518f.fsf-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>
2023-05-06 22:49 ` Chris Li
2023-05-06 22:49 ` Chris Li
[not found] ` <ZFbZZPkSpsKMe8iR-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>
2023-05-08 8:17 ` Alistair Popple
2023-05-08 8:17 ` Alistair Popple
[not found] ` <87ttwnkzap.fsf-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>
2023-05-10 14:51 ` Chris Li
2023-05-10 14:51 ` Chris Li
[not found] ` <ZFuvhP5qGPivokc0-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>
2023-05-12 8:45 ` Alistair Popple
2023-05-12 8:45 ` Alistair Popple
[not found] ` <87jzxe9baj.fsf-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>
2023-05-12 21:09 ` Jason Gunthorpe [this message]
2023-05-12 21:09 ` Jason Gunthorpe
[not found] ` <ZF6rACJzilA06oe+-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>
2023-05-16 12:21 ` Alistair Popple
2023-05-16 12:21 ` Alistair Popple
[not found] ` <87y1lo8nwp.fsf-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>
2023-05-19 15:47 ` Jason Gunthorpe
2023-05-19 15:47 ` Jason Gunthorpe
2023-05-20 15:09 ` Chris Li
2023-05-20 15:09 ` Chris Li
2023-05-20 15:31 ` Chris Li
2023-05-20 15:31 ` Chris Li
[not found] ` <ZGjntWoAfgyT0doo-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>
2023-05-29 19:31 ` Jason Gunthorpe
2023-05-29 19:31 ` Jason Gunthorpe
2023-05-04 17:02 ` Shakeel Butt
2023-05-04 17:02 ` Shakeel Butt
[not found] ` <CALvZod4=+ANT6UR5h7Cp+0hKkVx6tPAaRa5iqBF=L2VBdMKERQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2023-05-04 17:36 ` Chris Li
2023-05-04 17:36 ` Chris Li
2023-05-12 3:08 ` Yosry Ahmed
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=ZF6rACJzilA06oe+@nvidia.com \
--to=jgg-ddmlm1+adcrqt0dzr+alfa@public.gmane.org \
--cc=apopple-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org \
--cc=cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=chrisl-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org \
--cc=hannes-druUgvl0LCNAfugRpC6u6w@public.gmane.org \
--cc=kaleshsingh-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org \
--cc=linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org \
--cc=lsf-pc-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org \
--cc=muchun.song-fxUVXftIFDnyG1zEObXtfA@public.gmane.org \
--cc=roman.gushchin-fxUVXftIFDnyG1zEObXtfA@public.gmane.org \
--cc=shakeelb-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org \
--cc=tj-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org \
--cc=tjmercier-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org \
--cc=yosryahmed-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org \
--cc=yuzhao-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 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.