From: "Huang, Ying" <ying.huang@intel.com>
To: Zi Yan <ziy@nvidia.com>
Cc: <linux-mm@kvack.org>, <linux-kernel@vger.kernel.org>,
Andrew Morton <akpm@linux-foundation.org>,
Yang Shi <shy828301@gmail.com>,
Baolin Wang <baolin.wang@linux.alibaba.com>,
Oscar Salvador <osalvador@suse.de>,
"Matthew Wilcox" <willy@infradead.org>
Subject: Re: [RFC 0/6] migrate_pages(): batch TLB flushing
Date: Thu, 22 Sep 2022 09:45:11 +0800 [thread overview]
Message-ID: <878rmckwrc.fsf@yhuang6-desk2.ccr.corp.intel.com> (raw)
In-Reply-To: <FE41BDA8-F7EC-4FBC-9647-A5A835CDECB0@nvidia.com> (Zi Yan's message of "Wed, 21 Sep 2022 11:47:38 -0400")
Zi Yan <ziy@nvidia.com> writes:
> On 21 Sep 2022, at 2:06, Huang Ying wrote:
>
>> From: "Huang, Ying" <ying.huang@intel.com>
>>
>> Now, migrate_pages() migrate pages one by one, like the fake code as
>> follows,
>>
>> for each page
>> unmap
>> flush TLB
>> copy
>> restore map
>>
>> If multiple pages are passed to migrate_pages(), there are
>> opportunities to batch the TLB flushing and copying. That is, we can
>> change the code to something as follows,
>>
>> for each page
>> unmap
>> for each page
>> flush TLB
>> for each page
>> copy
>> for each page
>> restore map
>>
>> The total number of TLB flushing IPI can be reduced considerably. And
>> we may use some hardware accelerator such as DSA to accelerate the
>> page copying.
>>
>> So in this patch, we refactor the migrate_pages() implementation and
>> implement the TLB flushing batching. Base on this, hardware
>> accelerated page copying can be implemented.
>>
>> If too many pages are passed to migrate_pages(), in the naive batched
>> implementation, we may unmap too many pages at the same time. The
>> possibility for a task to wait for the migrated pages to be mapped
>> again increases. So the latency may be hurt. To deal with this
>> issue, the max number of pages be unmapped in batch is restricted to
>> no more than HPAGE_PMD_NR. That is, the influence is at the same
>> level of THP migration.
>>
>> We use the following test to measure the performance impact of the
>> patchset,
>>
>> On a 2-socket Intel server,
>>
>> - Run pmbench memory accessing benchmark
>>
>> - Run `migratepages` to migrate pages of pmbench between node 0 and
>> node 1 back and forth.
>>
>> With the patch, the TLB flushing IPI reduces 99.1% during the test and
>> the number of pages migrated successfully per second increases 291.7%.
>
> Thank you for the patchset. Batching page migration will definitely
> improve its throughput from my past experiments[1] and starting with
> TLB flushing is a good first step.
Thanks for the pointer, the patch description provides valuable information
for me already!
> BTW, what is the rationality behind the increased page migration
> success rate per second?
From perf profiling data, in the base kernel,
migrate_pages.migrate_to_node.do_migrate_pages.kernel_migrate_pages.__x64_sys_migrate_pages: 2.87
ptep_clear_flush.try_to_migrate_one.rmap_walk_anon.try_to_migrate.__unmap_and_move: 2.39
Because pmbench run in the system too, the CPU cycles of migrate_pages()
is about 2.87%. While the CPU cycles for TLB flushing is 2.39%. That
is, 2.39/2.87 = 83.3% CPU cycles of migrate_pages() are used for TLB
flushing.
After batching the TLB flushing, the perf profiling data becomes,
migrate_pages.migrate_to_node.do_migrate_pages.kernel_migrate_pages.__x64_sys_migrate_pages: 2.77
move_to_new_folio.migrate_pages_batch.migrate_pages.migrate_to_node.do_migrate_pages: 1.68
copy_page.folio_copy.migrate_folio.move_to_new_folio.migrate_pages_batch: 1.21
1.21/2.77 = 43.7% CPU cycles of migrate_pages() are used for page
copying now.
try_to_migrate_one: 0.23
The CPU cycles of unmapping and TLB flushing becomes 0.23/2.77 = 8.3% of
migrate_pages().
All in all, after the optimization, we do much less TLB flushing, which
consumes a lot of CPU cycles before the optimization. So the throughput
of migrate_pages() increases greatly.
I will add these data in the next version of patch.
Best Regards,
Huang, Ying
>>
>> This patchset is based on v6.0-rc5 and the following patchset,
>>
>> [PATCH -V3 0/8] migrate_pages(): fix several bugs in error path
>> https://lore.kernel.org/lkml/20220817081408.513338-1-ying.huang@intel.com/
>>
>> The migrate_pages() related code is converting to folio now. So this
>> patchset cannot apply recent akpm/mm-unstable branch. This patchset
>> is used to check the basic idea. If it is OK, I will rebase the
>> patchset on top of folio changes.
>>
>> Best Regards,
>> Huang, Ying
>
>
> [1] https://lwn.net/Articles/784925/
>
> --
> Best Regards,
> Yan, Zi
next prev parent reply other threads:[~2022-09-22 1:45 UTC|newest]
Thread overview: 50+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-09-21 6:06 [RFC 0/6] migrate_pages(): batch TLB flushing Huang Ying
2022-09-21 6:06 ` [RFC 1/6] mm/migrate_pages: separate huge page and normal pages migration Huang Ying
2022-09-21 15:55 ` Zi Yan
2022-09-22 1:14 ` Huang, Ying
2022-09-22 6:03 ` Baolin Wang
2022-09-22 6:22 ` Huang, Ying
2022-09-21 6:06 ` [RFC 2/6] mm/migrate_pages: split unmap_and_move() to _unmap() and _move() Huang Ying
2022-09-21 16:08 ` Zi Yan
2022-09-22 1:15 ` Huang, Ying
2022-09-22 6:36 ` Baolin Wang
2022-09-26 9:28 ` Alistair Popple
2022-09-26 18:06 ` Yang Shi
2022-09-27 0:02 ` Alistair Popple
2022-09-27 1:51 ` Huang, Ying
2022-09-27 20:34 ` John Hubbard
2022-09-27 20:57 ` Yang Shi
2022-09-28 0:59 ` Alistair Popple
2022-09-28 1:41 ` Huang, Ying
2022-09-28 1:44 ` John Hubbard
2022-09-28 1:49 ` Yang Shi
2022-09-28 1:56 ` John Hubbard
2022-09-28 2:14 ` Yang Shi
2022-09-28 2:57 ` John Hubbard
2022-09-28 3:25 ` Yang Shi
2022-09-28 3:39 ` Yang Shi
2022-09-27 20:56 ` Yang Shi
2022-09-27 20:54 ` Yang Shi
2022-09-21 6:06 ` [RFC 3/6] mm/migrate_pages: restrict number of pages to migrate in batch Huang Ying
2022-09-21 16:10 ` Zi Yan
2022-09-21 16:15 ` Zi Yan
2022-09-22 1:15 ` Huang, Ying
2022-09-21 6:06 ` [RFC 4/6] mm/migrate_pages: batch _unmap and _move Huang Ying
2022-09-21 6:06 ` [RFC 5/6] mm/migrate_pages: share more code between " Huang Ying
2022-09-21 6:06 ` [RFC 6/6] mm/migrate_pages: batch flushing TLB Huang Ying
2022-09-21 15:47 ` [RFC 0/6] migrate_pages(): batch TLB flushing Zi Yan
2022-09-22 1:45 ` Huang, Ying [this message]
2022-09-22 3:47 ` haoxin
2022-09-22 4:36 ` Huang, Ying
2022-09-22 12:50 ` Bharata B Rao
2022-09-23 7:52 ` Huang, Ying
2022-09-27 10:46 ` Bharata B Rao
2022-09-28 1:46 ` Huang, Ying
2022-09-26 9:11 ` Alistair Popple
2022-09-27 11:21 ` haoxin
2022-09-28 2:01 ` Huang, Ying
2022-09-28 3:33 ` haoxin
2022-09-28 4:53 ` Huang, Ying
2022-11-01 14:49 ` Hesham Almatary
2022-11-02 3:14 ` Huang, Ying
2022-11-02 14:13 ` Hesham Almatary
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=878rmckwrc.fsf@yhuang6-desk2.ccr.corp.intel.com \
--to=ying.huang@intel.com \
--cc=akpm@linux-foundation.org \
--cc=baolin.wang@linux.alibaba.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=osalvador@suse.de \
--cc=shy828301@gmail.com \
--cc=willy@infradead.org \
--cc=ziy@nvidia.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).