From: David Hildenbrand <david@redhat.com>
To: Matthew Wilcox <willy@infradead.org>
Cc: Alex Shi <seakeel@gmail.com>,
alexs@kernel.org, Andrew Morton <akpm@linux-foundation.org>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org,
izik.eidus@ravellosystems.com, aarcange@redhat.com,
chrisw@sous-sol.org, hughd@google.com
Subject: Re: [PATCH 02/10] mm/ksm: skip subpages of compound pages
Date: Thu, 6 Jun 2024 09:11:59 +0200 [thread overview]
Message-ID: <6c1fdc0a-98dd-47b5-933e-e2f6a91bc31f@redhat.com> (raw)
In-Reply-To: <ZmDUpryUaVraNF6m@casper.infradead.org>
On 05.06.24 23:12, Matthew Wilcox wrote:
> On Wed, Jun 05, 2024 at 09:47:10AM +0200, David Hildenbrand wrote:
>> On 05.06.24 08:14, Alex Shi wrote:
>>>
>>>
>>> On 6/5/24 11:52 AM, Matthew Wilcox wrote:
>>>> On Tue, Jun 04, 2024 at 12:24:44PM +0800, alexs@kernel.org wrote:
>>>>> From: "Alex Shi (tencent)" <alexs@kernel.org>
>>>>>
>>>>> When a folio isn't fit for KSM, the subpages are unlikely to be good,
>>>>> So let's skip the rest page checking to save some actions.
>>>>
>>>> Why would you say that is true? We have plenty of evidence that
>>>> userspace allocators can allocate large folios, then use only the first
>>>> few bytes, leaving many tail pages full of zeroes.
>>>
>>> Um, that do need tail pages...
>>> Is there some way to use more folio in ksm?
>>
>> My take, and Willy can correct me if I am wrong:
>>
>> "struct page" is not going to away any time soon, but it might shrink at
>> some point.
>>
>> That is, we can use the "struct page" pointer to point at a page frame, and
>> use "struct folio" to lookup/manage the metadata.
>
> Right.
>
>> That is, use "struct page" when accessing the actual memory content
>> (checksum, testing for identical content), but use the folio part when
>> looking up metadata (folio_test_anon() etc). In the future we might want to
>> replace the "struct page" pointer by an index into the folio, but that
>> doesn't have to happen right now.
>
> My current thinking is that folio->pfn is how we know where the memory
> described by the folio is. Using an index would be memmap[folio->pfn +
> index] which isn't terribly expensive, but we may as well pass around the
> (folio, page) pair and save the reference to memmap.
Right, as soon as the folio does not overlay the head page it's going to
be a bit different.
A (folio,page) pair, like we use in the RMAP code, is likely the best
option for now and gives us sufficient flexibility for the future design.
--
Cheers,
David / dhildenb
next prev parent reply other threads:[~2024-06-06 7:12 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-06-04 4:24 [PATCH 00/10] use folio in ksm alexs
2024-06-04 4:24 ` [PATCH 01/10] mm/ksm: reduce the flush action for ksm merging page alexs
2024-06-04 8:07 ` David Hildenbrand
2024-06-04 10:26 ` Alex Shi
2024-06-04 10:45 ` David Hildenbrand
2024-06-04 13:02 ` Alex Shi
2024-06-05 7:26 ` David Hildenbrand
2024-06-05 9:10 ` Alex Shi
2024-06-05 9:14 ` David Hildenbrand
2024-06-05 9:49 ` Alex Shi
2024-06-05 10:00 ` David Hildenbrand
2024-06-04 4:24 ` [PATCH 02/10] mm/ksm: skip subpages of compound pages alexs
2024-06-04 8:12 ` David Hildenbrand
2024-06-04 10:31 ` Alex Shi
2024-06-04 10:43 ` David Hildenbrand
2024-06-04 13:10 ` Alex Shi
2024-06-04 13:14 ` David Hildenbrand
2024-06-05 3:58 ` Alex Shi
2024-06-05 7:40 ` David Hildenbrand
2024-06-05 3:52 ` Matthew Wilcox
2024-06-05 6:14 ` Alex Shi
2024-06-05 7:47 ` David Hildenbrand
2024-06-05 21:12 ` Matthew Wilcox
2024-06-06 7:11 ` David Hildenbrand [this message]
2024-06-04 4:24 ` [PATCH 03/10] mm/ksm: use folio in try_to_merge_one_page alexs
2024-06-04 8:19 ` David Hildenbrand
2024-06-05 3:38 ` Alex Shi
2024-06-04 4:24 ` [PATCH 04/10] mm/ksm: add identical_folio func alexs
2024-06-04 8:25 ` David Hildenbrand
2024-06-04 4:24 ` [PATCH 05/10] mm/ksm: use folio in stable_tree_search alexs
2024-06-04 4:24 ` [PATCH 06/10] mm/ksm: remove page_stable_node alexs
2024-06-04 4:24 ` [PATCH 07/10] mm/ksm: use folio in unstable_tree_search_insert alexs
2024-06-04 4:24 ` [PATCH 08/10] mm/ksm: use folio in try_to_merge_xx serie funcs alexs
2024-06-04 4:24 ` [PATCH 09/10] mm/ksm: calc_checksum for folio alexs
2024-06-04 13:18 ` David Hildenbrand
2024-06-05 3:44 ` Alex Shi
2024-06-05 7:53 ` David Hildenbrand
2024-06-04 4:24 ` [PATCH 10/10] m/ksm: use folio in ksm scan path alexs
2024-06-04 13:28 ` [PATCH 00/10] use folio in ksm David Hildenbrand
2024-06-05 3:46 ` Alex Shi
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=6c1fdc0a-98dd-47b5-933e-e2f6a91bc31f@redhat.com \
--to=david@redhat.com \
--cc=aarcange@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=alexs@kernel.org \
--cc=chrisw@sous-sol.org \
--cc=hughd@google.com \
--cc=izik.eidus@ravellosystems.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=seakeel@gmail.com \
--cc=willy@infradead.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