From: JoonSoo Kim <js1304@gmail.com>
To: Glauber Costa <glommer@parallels.com>
Cc: Andrew Morton <akpm@linux-foundation.org>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org,
kamezawa.hiroyu@jp.fujitsu.com,
Johannes Weiner <hannes@cmpxchg.org>, Tejun Heo <tj@kernel.org>,
Michal Hocko <mhocko@suse.cz>, Christoph Lameter <cl@linux.com>,
Pekka Enberg <penberg@kernel.org>,
David Rientjes <rientjes@google.com>,
Greg Thelen <gthelen@google.com>
Subject: Re: [PATCH v6 00/29] kmem controller for memcg.
Date: Sat, 3 Nov 2012 04:25:59 +0900 [thread overview]
Message-ID: <CAAmzW4O74e3J9M3Q86Y0wXX6Pfp8GDpv6jAB5ebJPHfAxAeL0Q@mail.gmail.com> (raw)
In-Reply-To: <50937918.7080302@parallels.com>
Hello, Glauber.
2012/11/2 Glauber Costa <glommer@parallels.com>:
> On 11/02/2012 04:04 AM, Andrew Morton wrote:
>> On Thu, 1 Nov 2012 16:07:16 +0400
>> Glauber Costa <glommer@parallels.com> wrote:
>>
>>> Hi,
>>>
>>> This work introduces the kernel memory controller for memcg. Unlike previous
>>> submissions, this includes the whole controller, comprised of slab and stack
>>> memory.
>>
>> I'm in the middle of (re)reading all this. Meanwhile I'll push it all
>> out to http://ozlabs.org/~akpm/mmots/ for the crazier testers.
>>
>> One thing:
>>
>>> Numbers can be found at https://lkml.org/lkml/2012/9/13/239
>>
>> You claim in the above that the fork worload is 'slab intensive". Or
>> at least, you seem to - it's a bit fuzzy.
>>
>> But how slab intensive is it, really?
>>
>> What is extremely slab intensive is networking. The networking guys
>> are very sensitive to slab performance. If this hasn't already been
>> done, could you please determine what impact this has upon networking?
>> I expect Eric Dumazet, Dave Miller and Tom Herbert could suggest
>> testing approaches.
>>
>
> I can test it, but unfortunately I am unlikely to get to prepare a good
> environment before Barcelona.
>
> I know, however, that Greg Thelen was testing netperf in his setup.
> Greg, do you have any publishable numbers you could share?
Below is my humble opinion.
I am worrying about data cache footprint which is possibly caused by
this patchset, especially slab implementation.
If there are several memcg cgroups, each cgroup has it's own kmem_caches.
When each group do slab-intensive job hard, data cache may be overflowed easily,
and cache miss rate will be high, therefore this would decrease system
performance highly.
Is there any result about this?
Thanks.
--
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:[~2012-11-02 19:26 UTC|newest]
Thread overview: 75+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-11-01 12:07 [PATCH v6 00/29] kmem controller for memcg Glauber Costa
2012-11-01 12:07 ` [PATCH v6 01/29] memcg: Make it possible to use the stock for more than one page Glauber Costa
2012-11-01 12:07 ` [PATCH v6 02/29] memcg: Reclaim when more than one page needed Glauber Costa
2012-11-01 12:07 ` [PATCH v6 03/29] memcg: change defines to an enum Glauber Costa
2012-11-01 12:07 ` [PATCH v6 04/29] kmem accounting basic infrastructure Glauber Costa
2012-11-01 12:07 ` [PATCH v6 05/29] Add a __GFP_KMEMCG flag Glauber Costa
2012-11-01 19:58 ` Christoph Lameter
2012-11-01 12:07 ` [PATCH v6 06/29] memcg: kmem controller infrastructure Glauber Costa
2012-11-01 20:03 ` Christoph Lameter
2012-11-01 12:07 ` [PATCH v6 07/29] mm: Allocate kernel pages to the right memcg Glauber Costa
2012-11-01 12:07 ` [PATCH v6 08/29] res_counter: return amount of charges after res_counter_uncharge Glauber Costa
2012-11-01 12:07 ` [PATCH v6 09/29] memcg: kmem accounting lifecycle management Glauber Costa
2012-11-01 12:07 ` [PATCH v6 10/29] memcg: use static branches when code not in use Glauber Costa
2012-11-01 12:07 ` [PATCH v6 11/29] memcg: allow a memcg with kmem charges to be destructed Glauber Costa
2012-11-02 0:05 ` Andrew Morton
2012-11-02 7:50 ` Glauber Costa
2012-11-06 10:54 ` Michal Hocko
2012-11-01 12:07 ` [PATCH v6 12/29] execute the whole memcg freeing in free_worker Glauber Costa
2012-11-01 12:07 ` [PATCH v6 13/29] protect architectures where THREAD_SIZE >= PAGE_SIZE against fork bombs Glauber Costa
2012-11-01 12:07 ` [PATCH v6 14/29] Add documentation about the kmem controller Glauber Costa
2012-11-01 12:07 ` [PATCH v6 15/29] slab/slub: struct memcg_params Glauber Costa
2012-11-01 12:07 ` [PATCH v6 16/29] slab: annotate on-slab caches nodelist locks Glauber Costa
2012-11-01 12:07 ` [PATCH v6 17/29] consider a memcg parameter in kmem_create_cache Glauber Costa
2012-11-01 12:07 ` [PATCH v6 18/29] Allocate memory for memcg caches whenever a new memcg appears Glauber Costa
2012-11-06 0:23 ` Andrew Morton
2012-11-07 7:05 ` Glauber Costa
2012-11-07 7:10 ` Andrew Morton
2012-11-01 12:07 ` [PATCH v6 19/29] memcg: infrastructure to match an allocation to the right cache Glauber Costa
2012-11-06 0:28 ` Andrew Morton
2012-11-06 8:03 ` Michal Hocko
2012-11-08 11:05 ` Michal Hocko
2012-11-08 14:33 ` Michal Hocko
2012-11-07 7:04 ` Glauber Costa
2012-11-07 7:13 ` Andrew Morton
2012-11-01 12:07 ` [PATCH v6 20/29] memcg: skip memcg kmem allocations in specified code regions Glauber Costa
2012-11-06 0:33 ` Andrew Morton
2012-11-01 12:07 ` [PATCH v6 21/29] sl[au]b: always get the cache from its page in kmem_cache_free Glauber Costa
2012-11-01 12:07 ` [PATCH v6 22/29] sl[au]b: Allocate objects from memcg cache Glauber Costa
2012-11-01 12:07 ` [PATCH v6 23/29] memcg: destroy memcg caches Glauber Costa
2012-11-02 0:05 ` Andrew Morton
2012-11-02 7:46 ` Glauber Costa
2012-11-02 20:19 ` Michal Hocko
2012-11-06 0:40 ` Andrew Morton
2012-11-01 12:07 ` [PATCH v6 24/29] memcg/sl[au]b Track all the memcg children of a kmem_cache Glauber Costa
2012-11-01 12:07 ` [PATCH v6 25/29] memcg/sl[au]b: shrink dead caches Glauber Costa
2012-11-06 0:48 ` Andrew Morton
2012-11-07 7:13 ` Glauber Costa
2012-11-07 7:16 ` Andrew Morton
2012-11-07 9:22 ` Glauber Costa
2012-11-07 22:46 ` Andrew Morton
2012-11-08 7:13 ` Glauber Costa
2012-11-08 17:15 ` Christoph Lameter
2012-11-08 19:21 ` Andrew Morton
2012-11-08 22:31 ` Glauber Costa
2012-11-08 22:40 ` Andrew Morton
2012-11-09 20:06 ` Christoph Lameter
2012-11-09 20:04 ` Christoph Lameter
2012-11-01 12:07 ` [PATCH v6 26/29] Aggregate memcg cache values in slabinfo Glauber Costa
2012-11-06 0:57 ` Andrew Morton
2012-11-01 12:07 ` [PATCH v6 27/29] slab: propagate tunables values Glauber Costa
2012-11-01 12:07 ` [PATCH v6 28/29] slub: slub-specific propagation changes Glauber Costa
2012-11-06 19:25 ` Andrew Morton
2012-11-07 15:53 ` Sasha Levin
2012-11-08 6:51 ` Glauber Costa
2012-11-09 3:37 ` Sasha Levin
2012-11-14 12:06 ` Glauber Costa
2012-11-01 12:07 ` [PATCH v6 29/29] Add slab-specific documentation about the kmem controller Glauber Costa
2012-11-02 0:04 ` [PATCH v6 00/29] kmem controller for memcg Andrew Morton
2012-11-02 7:41 ` Glauber Costa
2012-11-02 19:25 ` JoonSoo Kim [this message]
2012-11-02 23:06 ` Tejun Heo
2012-11-05 8:14 ` Glauber Costa
2012-11-05 8:18 ` Glauber Costa
2012-11-03 3:36 ` Greg Thelen
2012-11-02 8:30 ` Pekka Enberg
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=CAAmzW4O74e3J9M3Q86Y0wXX6Pfp8GDpv6jAB5ebJPHfAxAeL0Q@mail.gmail.com \
--to=js1304@gmail.com \
--cc=akpm@linux-foundation.org \
--cc=cl@linux.com \
--cc=glommer@parallels.com \
--cc=gthelen@google.com \
--cc=hannes@cmpxchg.org \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@suse.cz \
--cc=penberg@kernel.org \
--cc=rientjes@google.com \
--cc=tj@kernel.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;
as well as URLs for NNTP newsgroup(s).