From: Anshuman Khandual <khandual@linux.vnet.ibm.com>
To: Zi Yan <zi.yan@cs.rutgers.edu>,
Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: "linux-mm@kvack.org" <linux-mm@kvack.org>,
"Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>,
Hugh Dickins <hughd@google.com>,
Andrew Morton <akpm@linux-foundation.org>,
Dave Hansen <dave.hansen@intel.com>,
Andrea Arcangeli <aarcange@redhat.com>,
Mel Gorman <mgorman@techsingularity.net>,
Michal Hocko <mhocko@kernel.org>,
Vlastimil Babka <vbabka@suse.cz>,
Pavel Emelyanov <xemul@parallels.com>,
Balbir Singh <bsingharora@gmail.com>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
Naoya Horiguchi <nao.horiguchi@gmail.com>
Subject: Re: [PATCH v2 00/12] mm: page migration enhancement for thp
Date: Fri, 11 Nov 2016 09:18:44 +0530 [thread overview]
Message-ID: <58253F9C.6040307@linux.vnet.ibm.com> (raw)
In-Reply-To: <D34FA575-7C5D-4E9D-B337-A925F1A89C66@cs.rutgers.edu>
On 11/10/2016 07:31 PM, Zi Yan wrote:
> On 9 Nov 2016, at 18:52, Naoya Horiguchi wrote:
>
>> Hi Anshuman,
>>
>> On Wed, Nov 09, 2016 at 04:03:04PM +0530, Anshuman Khandual wrote:
>>> On 11/08/2016 05:01 AM, Naoya Horiguchi wrote:
>>>> Hi everyone,
>>>>
>>>> I've updated thp migration patches for v4.9-rc2-mmotm-2016-10-27-18-27
>>>> with feedbacks for ver.1.
>>>>
>>>> General description (no change since ver.1)
>>>> ===========================================
>>>>
>>>> This patchset enhances page migration functionality to handle thp migration
>>>> for various page migration's callers:
>>>> - mbind(2)
>>>> - move_pages(2)
>>>> - migrate_pages(2)
>>>> - cgroup/cpuset migration
>>>> - memory hotremove
>>>> - soft offline
>>>>
>>>> The main benefit is that we can avoid unnecessary thp splits, which helps us
>>>> avoid performance decrease when your applications handles NUMA optimization on
>>>> their own.
>>>>
>>>> The implementation is similar to that of normal page migration, the key point
>>>> is that we modify a pmd to a pmd migration entry in swap-entry like format.
>>>
>>> Will it be better to have new THP_MIGRATE_SUCCESS and THP_MIGRATE_FAIL
>>> VM events to capture how many times the migration worked without first
>>> splitting the huge page and how many time it did not work ?
>>
>> Thank you for the suggestion.
>> I think that's helpful, so will try it in next version.
>>
>>> Also do you
>>> have a test case which demonstrates this THP migration and kind of shows
>>> its better than the present split and move method ?
>>
>> I don't have test cases which compare thp migration and split-then-migration
>> with some numbers. Maybe measuring/comparing the overhead of migration is
>> a good start point, although I think the real benefit of thp migration comes
>> from workload "after migration" by avoiding thp split.
>
> Migrating 4KB pages has much lower (~1/3) throughput than 2MB pages.
I assume the 2MB throughput you mentioned is with this THP migration
feature enabled.
>
> What I get is that on average it takes 1987.38 us to migrate 512 4KB pages and
> 658.54 us to migrate 1 2MB page.
>
> I did the test in a two-socket Intel Xeon E5-2640v4 box. I used migrate_pages()
> system call to migrate pages. MADV_NOHUGEPAGE and MADV_HUGEPAGE are used to
> make 4KB and 2MB pages and each page’s flags are checked to make sure the page
> size is 4KB or 2MB THP.
>
> There is no split page. But the page migration time already tells the story.
Right. Just wondering if we can add a test case which measures just
this migration time improvement by avoiding the split not the TLB
based improvement which the workload will receive as an addition.
next prev parent reply other threads:[~2016-11-11 3:49 UTC|newest]
Thread overview: 71+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-11-07 23:31 [PATCH v2 00/12] mm: page migration enhancement for thp Naoya Horiguchi
2016-11-07 23:31 ` [PATCH v2 01/12] mm: x86: move _PAGE_SWP_SOFT_DIRTY from bit 7 to bit 6 Naoya Horiguchi
2016-11-10 23:29 ` Dave Hansen
2016-11-11 1:08 ` Naoya Horiguchi
2016-11-11 11:12 ` Kirill A. Shutemov
2016-11-07 23:31 ` [PATCH v2 02/12] mm: mempolicy: add queue_pages_node_check() Naoya Horiguchi
2016-11-07 23:31 ` [PATCH v2 03/12] mm: thp: introduce separate TTU flag for thp freezing Naoya Horiguchi
2016-11-10 8:31 ` Anshuman Khandual
2016-11-10 9:09 ` Naoya Horiguchi
2016-11-11 3:18 ` Anshuman Khandual
2016-11-11 11:16 ` Kirill A. Shutemov
2016-11-07 23:31 ` [PATCH v2 04/12] mm: thp: introduce CONFIG_ARCH_ENABLE_THP_MIGRATION Naoya Horiguchi
2016-11-11 11:18 ` Kirill A. Shutemov
2016-11-15 2:05 ` Naoya Horiguchi
2016-11-28 14:21 ` Michal Hocko
2016-11-29 7:50 ` Naoya Horiguchi
2016-11-29 8:45 ` Michal Hocko
2016-11-07 23:31 ` [PATCH v2 05/12] mm: thp: add core routines for thp/pmd migration Naoya Horiguchi
2016-11-08 3:05 ` kbuild test robot
2016-11-08 6:46 ` Naoya Horiguchi
2016-11-08 8:13 ` Anshuman Khandual
2016-11-08 8:16 ` Naoya Horiguchi
2016-11-10 8:29 ` Anshuman Khandual
2016-11-10 9:43 ` Naoya Horiguchi
2016-11-14 11:45 ` Kirill A. Shutemov
2016-11-15 4:57 ` Naoya Horiguchi
2016-11-28 14:31 ` Michal Hocko
2016-11-29 7:57 ` Naoya Horiguchi
2016-11-07 23:31 ` [PATCH v2 06/12] mm: thp: enable thp migration in generic path Naoya Horiguchi
2016-11-28 14:33 ` Michal Hocko
2016-11-29 8:16 ` Naoya Horiguchi
2016-11-07 23:31 ` [PATCH v2 07/12] mm: thp: check pmd migration entry in common path Naoya Horiguchi
2016-11-08 0:23 ` kbuild test robot
2016-11-08 1:22 ` Naoya Horiguchi
2016-11-10 8:36 ` Anshuman Khandual
2016-11-10 9:12 ` Naoya Horiguchi
2016-11-10 9:08 ` Hillf Danton
2016-11-10 9:21 ` Naoya Horiguchi
2016-11-10 9:28 ` Hillf Danton
2016-11-10 9:34 ` Naoya Horiguchi
2016-11-17 23:56 ` Kirill A. Shutemov
2016-11-29 6:46 ` Naoya Horiguchi
2016-11-28 14:35 ` Michal Hocko
2016-11-07 23:31 ` [PATCH v2 08/12] mm: soft-dirty: keep soft-dirty bits over thp migration Naoya Horiguchi
2016-11-10 8:38 ` Anshuman Khandual
2016-11-10 9:13 ` Naoya Horiguchi
2016-11-07 23:31 ` [PATCH v2 09/12] mm: hwpoison: soft offline supports " Naoya Horiguchi
2016-11-10 10:31 ` Balbir Singh
2016-11-10 23:58 ` Naoya Horiguchi
2016-11-14 23:22 ` Balbir Singh
2016-11-07 23:31 ` [PATCH v2 10/12] mm: mempolicy: mbind and migrate_pages support " Naoya Horiguchi
2016-11-25 12:27 ` Anshuman Khandual
2016-11-29 7:07 ` Naoya Horiguchi
2016-11-07 23:31 ` [PATCH v2 11/12] mm: migrate: move_pages() supports " Naoya Horiguchi
2016-11-18 0:01 ` Kirill A. Shutemov
2016-11-07 23:31 ` [PATCH v2 12/12] mm: memory_hotplug: memory hotremove " Naoya Horiguchi
2016-11-08 0:29 ` kbuild test robot
2016-11-08 0:30 ` kbuild test robot
2016-11-08 1:36 ` Naoya Horiguchi
2016-11-08 1:41 ` [PATCH] mm: fix unused variable warning Naoya Horiguchi
2016-11-09 2:32 ` [PATCH v2 00/12] mm: page migration enhancement for thp Balbir Singh
2016-11-09 4:59 ` Naoya Horiguchi
2016-11-09 21:28 ` Balbir Singh
2016-11-09 10:33 ` Anshuman Khandual
2016-11-09 23:52 ` Naoya Horiguchi
2016-11-10 14:01 ` Zi Yan
2016-11-11 3:48 ` Anshuman Khandual [this message]
2017-01-30 10:47 ` Anshuman Khandual
2017-01-31 19:54 ` Zi Yan
2017-02-03 3:12 ` Hillf Danton
2017-02-03 3:24 ` Zi Yan
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=58253F9C.6040307@linux.vnet.ibm.com \
--to=khandual@linux.vnet.ibm.com \
--cc=aarcange@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=bsingharora@gmail.com \
--cc=dave.hansen@intel.com \
--cc=hughd@google.com \
--cc=kirill.shutemov@linux.intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mgorman@techsingularity.net \
--cc=mhocko@kernel.org \
--cc=n-horiguchi@ah.jp.nec.com \
--cc=nao.horiguchi@gmail.com \
--cc=vbabka@suse.cz \
--cc=xemul@parallels.com \
--cc=zi.yan@cs.rutgers.edu \
/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).