From: Kamezawa Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>,
David Rientjes <rientjes@google.com>,
linux-mm@kvack.org, mgorman@suse.de, dhillf@gmail.com,
aarcange@redhat.com, mhocko@suse.cz, hannes@cmpxchg.org,
linux-kernel@vger.kernel.org, cgroups@vger.kernel.org,
Ying Han <yinghan@google.com>
Subject: Re: [PATCH -V6 07/14] memcg: Add HugeTLB extension
Date: Mon, 11 Jun 2012 12:55:05 +0900 [thread overview]
Message-ID: <4FD56C19.4060307@jp.fujitsu.com> (raw)
In-Reply-To: <20120608160612.dea6d1ce.akpm@linux-foundation.org>
(2012/06/09 8:06), Andrew Morton wrote:
> On Wed, 30 May 2012 20:13:31 +0530
> "Aneesh Kumar K.V"<aneesh.kumar@linux.vnet.ibm.com> wrote:
>
>>>>
>>>> - code: seperating hugetlb bits out from memcg bits to avoid growing
>>>> mm/memcontrol.c beyond its current 5650 lines, and
>>>>
>>>
>>> I can definitely look at spliting mm/memcontrol.c
>>>
>>>
>>>> - performance: not incurring any overhead of enabling memcg for per-
>>>> page tracking that is unnecessary if users only want to limit hugetlb
>>>> pages.
>>>>
>>
>> Since Andrew didn't sent the patchset to Linus because of this
>> discussion, I looked at reworking the patchset as a seperate
>> controller. The patchset I sent here
>>
>> http://thread.gmane.org/gmane.linux.kernel.mm/79230
>>
>> have seen minimal testing. I also folded the fixup patches
>> Andrew had in -mm to original patchset.
>>
>> Let me know if the changes looks good.
>
> This is starting to be a problem. I'm still sitting on the old version
> of this patchset and it will start to get in the way of other work.
>
> We now have this new version of the patchset which implements a
> separate controller but it is unclear to me which way we want to go.
>
> Can the memcg developers please drop everything else and make a
> decision here?
Following is a summary in my point of view.
I think there are several topics.
- overheads.
(A) IMHO, runtime overhead will be negligible because...
- if hugetlb is used, anonymous memory accouning doesn't add much overheads
because they're not used.
- when it comes to file-cache accounting, I/O dominates performance rather
than memcg..
- but you may see some overheads with 100+ cpu system...I'm not sure.
(B) memory space overhead will not be negligible.
- now, memcg uses 16bytes per page....4GB/1TB.
This may be an obvious overhead to the system if working set size are
quite big and the apps want to use huge size memory.
(C) what hugetlbfs is.
- hugetlb is statically allocated. So, they're not usual memory.
Then, hugetlb cgroup is better.
- IMHO, hugetlb is memory. And I thought memory.limit_in_bytes should
take it into account....
(D) code duplication
- memory cgroup and hugetlb cgroup will have similar hooks,codes,UIs.
- we need some #ifdef if we have consolidated memory/hugetlb cgroup.
(E) user experience
- with independent hugetlb cgroup, users can disable memory cgroup.
- with consolidated memcg+hugetlb cgroup, we'll be able to limit
usual page + hugetlb usage by a limit.
Now, I think...
1. I need to agree that overhead is _not_ negligible.
2. THP should be the way rather than hugetlb for my main target platform.
(shmem/tmpfs should support THP. we need study.)
user-experience should be fixed by THP+tmpfs+memcg.
3. It seems Aneesh decided to have independent hugetlb cgroup.
So, now, I admit to have independent hugetlb cgroup.
Other opinions ?
Thanks,
-Kame
next prev parent reply other threads:[~2012-06-11 3:57 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-04-16 10:44 [PATCH -V6 00/14] memcg: Add memcg extension to control HugeTLB allocation Aneesh Kumar K.V
2012-04-16 10:44 ` [PATCH -V6 01/14] hugetlb: rename max_hstate to hugetlb_max_hstate Aneesh Kumar K.V
2012-05-24 21:11 ` David Rientjes
2012-04-16 10:44 ` [PATCH -V6 02/14] hugetlbfs: don't use ERR_PTR with VM_FAULT* values Aneesh Kumar K.V
2012-05-24 21:17 ` David Rientjes
2012-04-16 10:44 ` [PATCH -V6 03/14] hugetlbfs: Add an inline helper for finding hstate index Aneesh Kumar K.V
2012-05-24 21:22 ` David Rientjes
2012-05-27 20:07 ` Aneesh Kumar K.V
2012-04-16 10:44 ` [PATCH -V6 04/14] hugetlb: Use mmu_gather instead of a temporary linked list for accumulating pages Aneesh Kumar K.V
2012-04-23 23:44 ` Andrew Morton
2012-04-16 10:44 ` [PATCH -V6 05/14] hugetlb: Avoid taking i_mmap_mutex in unmap_single_vma for hugetlb Aneesh Kumar K.V
2012-04-16 10:44 ` [PATCH -V6 06/14] hugetlb: Simplify migrate_huge_page Aneesh Kumar K.V
2012-05-24 21:35 ` David Rientjes
2012-05-27 20:13 ` Aneesh Kumar K.V
2012-04-16 10:44 ` [PATCH -V6 07/14] memcg: Add HugeTLB extension Aneesh Kumar K.V
2012-05-02 0:20 ` Paul Gortmaker
2012-05-03 4:37 ` Aneesh Kumar K.V
2012-05-24 21:52 ` David Rientjes
2012-05-24 22:57 ` Andrew Morton
2012-05-24 23:20 ` David Rientjes
2012-05-27 20:28 ` Aneesh Kumar K.V
2012-05-30 14:43 ` Aneesh Kumar K.V
2012-06-08 23:06 ` Andrew Morton
2012-06-09 14:16 ` Aneesh Kumar K.V
2012-06-10 1:55 ` David Rientjes
2012-06-10 15:04 ` Aneesh Kumar K.V
2012-06-11 3:55 ` Kamezawa Hiroyuki [this message]
2012-06-11 9:23 ` David Rientjes
2012-06-15 22:31 ` Aditya Kali
2012-06-16 20:26 ` David Rientjes
2012-06-11 9:32 ` Michal Hocko
2012-04-16 10:44 ` [PATCH -V6 08/14] hugetlb: add charge/uncharge calls for HugeTLB alloc/free Aneesh Kumar K.V
2012-04-16 10:44 ` [PATCH -V6 09/14] memcg: track resource index in cftype private Aneesh Kumar K.V
2012-04-16 10:44 ` [PATCH -V6 10/14] hugetlbfs: Add memcg control files for hugetlbfs Aneesh Kumar K.V
2012-04-16 23:13 ` Andrew Morton
2012-04-18 6:15 ` [PATCH] memcg: Use scnprintf instead of sprintf Aneesh Kumar K.V
2012-04-18 22:36 ` Andrew Morton
2012-04-19 8:26 ` Andreas Schwab
2012-04-18 6:16 ` [PATCH -V6 10/14] hugetlbfs: Add memcg control files for hugetlbfs Aneesh Kumar K.V
2012-04-16 10:44 ` [PATCH -V6 11/14] hugetlbfs: Add a list for tracking in-use HugeTLB pages Aneesh Kumar K.V
2012-04-16 10:44 ` [PATCH -V6 12/14] memcg: move HugeTLB resource count to parent cgroup on memcg removal Aneesh Kumar K.V
2012-04-23 22:45 ` Andrew Morton
2012-04-16 10:44 ` [PATCH -V6 13/14] hugetlb: migrate memcg info from oldpage to new page during migration Aneesh Kumar K.V
2012-04-16 10:44 ` [PATCH -V6 14/14] memcg: Add memory controller documentation for hugetlb management Aneesh Kumar K.V
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=4FD56C19.4060307@jp.fujitsu.com \
--to=kamezawa.hiroyu@jp.fujitsu.com \
--cc=aarcange@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=aneesh.kumar@linux.vnet.ibm.com \
--cc=cgroups@vger.kernel.org \
--cc=dhillf@gmail.com \
--cc=hannes@cmpxchg.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mgorman@suse.de \
--cc=mhocko@suse.cz \
--cc=rientjes@google.com \
--cc=yinghan@google.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 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).