From: Balbir Singh <bsingharora@gmail.com>
To: Michal Hocko <mhocko@kernel.org>
Cc: linux-mm@kvack.org, linuxppc-dev@lists.ozlabs.org,
Tejun Heo <tj@kernel.org>,
Andrew Morton <akpm@linux-foundation.org>,
Johannes Weiner <hannes@cmpxchg.org>,
Vladimir Davydov <vdavydov.dev@gmail.com>
Subject: Re: [mm v2 0/3] Support memory cgroup hotplug
Date: Wed, 23 Nov 2016 18:50:42 +1100 [thread overview]
Message-ID: <342ebcca-b54c-4bc6-906b-653042caae06@gmail.com> (raw)
In-Reply-To: <20161123072543.GD2864@dhcp22.suse.cz>
On 23/11/16 18:25, Michal Hocko wrote:
> On Wed 23-11-16 15:36:51, Balbir Singh wrote:
>> In the absence of hotplug we use extra memory proportional to
>> (possible_nodes - online_nodes) * number_of_cgroups. PPC64 has a patch
>> to disable large consumption with large number of cgroups. This patch
>> adds hotplug support to memory cgroups and reverts the commit that
>> limited possible nodes to online nodes.
>
> Balbir,
> I have asked this in the previous version but there still seems to be a
> lack of information of _why_ do we want this, _how_ much do we save on
> the memory overhead on most systems and _why_ the additional complexity
> is really worth it. Please make sure to add all this in the cover
> letter.
>
The data is in the patch referred to in patch 3. The order of waste was
200MB for 400 cgroup directories enough for us to restrict possible_map
to online_map. These patches allow us to have a larger possible map and
allow onlining nodes not in the online_map, which is currently a restriction
on ppc64.
A typical system that I use has about 100-150 directories, depending on the
number of users/docker instances/configuration/virtual machines. These numbers
will only grow as we pack more of these instances on them.
>From a complexity view point, the patches are quite straight forward.
> I still didn't get to look into those patches because I am swamped with
> other things but to be honest I do not really see a strong justification
> to make it high priority for me.
>
I am OK if you need more time to review them, but I've been pushing them
to fix the cases I've mentioned above.
Balbir Singh.
WARNING: multiple messages have this Message-ID (diff)
From: Balbir Singh <bsingharora@gmail.com>
To: Michal Hocko <mhocko@kernel.org>
Cc: linux-mm@kvack.org, linuxppc-dev@lists.ozlabs.org,
Tejun Heo <tj@kernel.org>,
Andrew Morton <akpm@linux-foundation.org>,
Johannes Weiner <hannes@cmpxchg.org>,
Vladimir Davydov <vdavydov.dev@gmail.com>
Subject: Re: [mm v2 0/3] Support memory cgroup hotplug
Date: Wed, 23 Nov 2016 18:50:42 +1100 [thread overview]
Message-ID: <342ebcca-b54c-4bc6-906b-653042caae06@gmail.com> (raw)
In-Reply-To: <20161123072543.GD2864@dhcp22.suse.cz>
On 23/11/16 18:25, Michal Hocko wrote:
> On Wed 23-11-16 15:36:51, Balbir Singh wrote:
>> In the absence of hotplug we use extra memory proportional to
>> (possible_nodes - online_nodes) * number_of_cgroups. PPC64 has a patch
>> to disable large consumption with large number of cgroups. This patch
>> adds hotplug support to memory cgroups and reverts the commit that
>> limited possible nodes to online nodes.
>
> Balbir,
> I have asked this in the previous version but there still seems to be a
> lack of information of _why_ do we want this, _how_ much do we save on
> the memory overhead on most systems and _why_ the additional complexity
> is really worth it. Please make sure to add all this in the cover
> letter.
>
The data is in the patch referred to in patch 3. The order of waste was
200MB for 400 cgroup directories enough for us to restrict possible_map
to online_map. These patches allow us to have a larger possible map and
allow onlining nodes not in the online_map, which is currently a restriction
on ppc64.
A typical system that I use has about 100-150 directories, depending on the
number of users/docker instances/configuration/virtual machines. These numbers
will only grow as we pack more of these instances on them.
>From a complexity view point, the patches are quite straight forward.
> I still didn't get to look into those patches because I am swamped with
> other things but to be honest I do not really see a strong justification
> to make it high priority for me.
>
I am OK if you need more time to review them, but I've been pushing them
to fix the cases I've mentioned above.
Balbir Singh.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2016-11-23 7:50 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-11-23 4:36 [mm v2 0/3] Support memory cgroup hotplug Balbir Singh
2016-11-23 4:36 ` Balbir Singh
2016-11-23 4:36 ` [mm v2 1/3] mm: Add basic infrastructure for memcg hotplug support Balbir Singh
2016-11-23 4:36 ` Balbir Singh
2016-11-23 4:36 ` [mm v2 2/3] mm: Move operations to hotplug callbacks Balbir Singh
2016-11-23 4:36 ` Balbir Singh
2016-11-23 4:36 ` [mm v2 3/3] powerpc/mm: fix node_possible_map limitations Balbir Singh
2016-11-23 4:36 ` Balbir Singh
2016-11-23 7:25 ` [mm v2 0/3] Support memory cgroup hotplug Michal Hocko
2016-11-23 7:25 ` Michal Hocko
2016-11-23 7:50 ` Balbir Singh [this message]
2016-11-23 7:50 ` Balbir Singh
2016-11-23 8:07 ` Michal Hocko
2016-11-23 8:07 ` Michal Hocko
2016-11-23 8:37 ` Balbir Singh
2016-11-23 8:37 ` Balbir Singh
2016-11-23 9:28 ` Michal Hocko
2016-11-23 9:28 ` Michal Hocko
2016-11-23 13:05 ` Balbir Singh
2016-11-23 13:05 ` Balbir Singh
2016-11-23 13:22 ` Michal Hocko
2016-11-23 13:22 ` Michal Hocko
2016-11-28 21:10 ` Tejun Heo
2016-11-28 21:10 ` Tejun Heo
2016-11-29 0:09 ` Balbir Singh
2016-11-29 0:09 ` Balbir Singh
2016-11-29 0:42 ` Tejun Heo
2016-11-29 0:42 ` Tejun Heo
2016-11-29 4:57 ` Balbir Singh
2016-11-29 4:57 ` Balbir Singh
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=342ebcca-b54c-4bc6-906b-653042caae06@gmail.com \
--to=bsingharora@gmail.com \
--cc=akpm@linux-foundation.org \
--cc=hannes@cmpxchg.org \
--cc=linux-mm@kvack.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=mhocko@kernel.org \
--cc=tj@kernel.org \
--cc=vdavydov.dev@gmail.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 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.