From: Matthew Wilcox <willy@infradead.org>
To: Baolin Wang <baolin.wang@linux.alibaba.com>
Cc: akpm@linux-foundation.org, hughd@google.com, david@redhat.com,
wangkefeng.wang@huawei.com, ying.huang@intel.com,
21cnbao@gmail.com, ryan.roberts@arm.com, shy828301@gmail.com,
ziy@nvidia.com, ioworker0@gmail.com, da.gomez@samsung.com,
p.raghav@samsung.com, linux-mm@kvack.org,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH v5 0/6] add mTHP support for anonymous shmem
Date: Thu, 4 Jul 2024 19:43:26 +0100 [thread overview]
Message-ID: <ZobtTmzj0AmNXcav@casper.infradead.org> (raw)
In-Reply-To: <cover.1718090413.git.baolin.wang@linux.alibaba.com>
On Tue, Jun 11, 2024 at 06:11:04PM +0800, Baolin Wang wrote:
> Anonymous pages have already been supported for multi-size (mTHP) allocation
> through commit 19eaf44954df, that can allow THP to be configured through the
> sysfs interface located at '/sys/kernel/mm/transparent_hugepage/hugepage-XXkb/enabled'.
>
> However, the anonymous shmem will ignore the anonymous mTHP rule configured
> through the sysfs interface, and can only use the PMD-mapped THP, that is not
> reasonable. Many implement anonymous page sharing through mmap(MAP_SHARED |
> MAP_ANONYMOUS), especially in database usage scenarios, therefore, users expect
> to apply an unified mTHP strategy for anonymous pages, also including the
> anonymous shared pages, in order to enjoy the benefits of mTHP. For example,
> lower latency than PMD-mapped THP, smaller memory bloat than PMD-mapped THP,
> contiguous PTEs on ARM architecture to reduce TLB miss etc.
OK, this makes sense.
> As discussed in the bi-weekly MM meeting[1], the mTHP controls should control
> all of shmem, not only anonymous shmem, but support will be added iteratively.
> Therefore, this patch set starts with support for anonymous shmem.
But then this doesn't. You say first that users want the same controls
to control all anonymous memory, then you introduce a completely
separate set of controls for shared anonymous memory.
shmem has two uses:
- MAP_ANONYMOUS | MAP_SHARED (this patch set)
- tmpfs
For the second use case we don't want controls *at all*, we want the
same heiristics used for all other filesystems to apply to tmpfs.
There's no reason to have separate controls for choosing folio size
in shmem.
> The primary strategy is similar to supporting anonymous mTHP. Introduce
> a new interface '/mm/transparent_hugepage/hugepage-XXkb/shmem_enabled',
> which can have almost the same values as the top-level
> '/sys/kernel/mm/transparent_hugepage/shmem_enabled', with adding a new
> additional "inherit" option and dropping the testing options 'force' and
> 'deny'. By default all sizes will be set to "never" except PMD size, which
> is set to "inherit". This ensures backward compatibility with the anonymous
> shmem enabled of the top level, meanwhile also allows independent control of
> anonymous shmem enabled for each mTHP.
next prev parent reply other threads:[~2024-07-04 18:43 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-06-11 10:11 [PATCH v5 0/6] add mTHP support for anonymous shmem Baolin Wang
2024-06-11 10:11 ` [PATCH v5 1/6] mm: memory: extend finish_fault() to support large folio Baolin Wang
2024-06-11 14:38 ` Zi Yan
2024-06-12 9:28 ` Baolin Wang
2024-06-12 13:40 ` Kefeng Wang
2024-06-13 0:51 ` Baolin Wang
2024-06-11 10:11 ` [PATCH v5 2/6] mm: shmem: add THP validation for PMD-mapped THP related statistics Baolin Wang
2024-06-11 10:11 ` [PATCH v5 3/6] mm: shmem: add multi-size THP sysfs interface for anonymous shmem Baolin Wang
2024-06-11 10:11 ` [PATCH v5 4/6] mm: shmem: add mTHP support " Baolin Wang
2024-07-03 17:25 ` Ryan Roberts
2024-07-04 11:15 ` Baolin Wang
2024-07-04 13:58 ` Ryan Roberts
2024-07-04 14:01 ` David Hildenbrand
2024-07-04 15:05 ` Bang Li
2024-07-04 15:54 ` Ryan Roberts
2024-07-05 2:56 ` Baolin Wang
2024-07-05 8:55 ` Ryan Roberts
2024-07-04 14:46 ` Bang Li
2024-07-05 3:01 ` Baolin Wang
2024-07-05 8:42 ` Ryan Roberts
2024-07-05 8:57 ` Baolin Wang
2024-07-05 9:05 ` Ryan Roberts
2024-06-11 10:11 ` [PATCH v5 5/6] mm: shmem: add mTHP size alignment in shmem_get_unmapped_area Baolin Wang
2024-06-11 10:11 ` [PATCH v5 6/6] mm: shmem: add mTHP counters for anonymous shmem Baolin Wang
2024-06-12 8:04 ` Lance Yang
2024-06-12 9:28 ` Baolin Wang
2024-06-12 14:16 ` Lance Yang
2024-06-12 13:46 ` Kefeng Wang
2024-06-13 1:00 ` Baolin Wang
2024-06-12 14:18 ` Lance Yang
2024-06-13 1:08 ` Baolin Wang
2024-07-04 18:43 ` Matthew Wilcox [this message]
2024-07-04 19:03 ` [PATCH v5 0/6] add mTHP support " David Hildenbrand
2024-07-04 19:19 ` David Hildenbrand
2024-07-04 19:49 ` Matthew Wilcox
2024-07-05 5:47 ` Baolin Wang
2024-07-05 8:45 ` Ryan Roberts
2024-07-05 8:59 ` David Hildenbrand
2024-07-05 9:13 ` Ryan Roberts
2024-07-05 9:16 ` David Hildenbrand
2024-07-05 9:23 ` Ryan Roberts
2024-07-07 16:39 ` Daniel Gomez
2024-07-09 8:28 ` Ryan Roberts
2024-07-16 13:11 ` Daniel Gomez
2024-07-16 13:22 ` David Hildenbrand
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=ZobtTmzj0AmNXcav@casper.infradead.org \
--to=willy@infradead.org \
--cc=21cnbao@gmail.com \
--cc=akpm@linux-foundation.org \
--cc=baolin.wang@linux.alibaba.com \
--cc=da.gomez@samsung.com \
--cc=david@redhat.com \
--cc=hughd@google.com \
--cc=ioworker0@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=p.raghav@samsung.com \
--cc=ryan.roberts@arm.com \
--cc=shy828301@gmail.com \
--cc=wangkefeng.wang@huawei.com \
--cc=ying.huang@intel.com \
--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).