From: David Hildenbrand <david@redhat.com>
To: "Yin, Fengwei" <fengwei.yin@intel.com>, Zi Yan <ziy@nvidia.com>
Cc: Matthew Wilcox <willy@infradead.org>,
linux-mm@kvack.org, Vishal Moola <vishal.moola@gmail.com>,
Hugh Dickins <hughd@google.com>, Rik van Riel <riel@surriel.com>
Subject: Re: Folio mapcount
Date: Sun, 2 Jul 2023 13:45:48 +0200 [thread overview]
Message-ID: <6cec6f68-248e-63b4-5615-9e0f3f819a0a@redhat.com> (raw)
In-Reply-To: <957ea888-a96b-89cc-29e2-973bb9e36f40@intel.com>
On 02.07.23 11:50, Yin, Fengwei wrote:
>
>
> On 7/1/2023 9:17 AM, Zi Yan wrote:
>> In kernel, almost all code only cares: 1) if a page/folio has extra pins
>> by checking if mapcount is equal to refcount + extra, and 2)
>> if a page/folio is mapped multiple times. A single mapcount can meet
>> these two needs.
> For 2, how can we know whether a page/folio is mapped multiple times from
> single mapcount? My understanding is we need two counts as folio could be
> partial mapped.
Yes, a single mapcount is most probably insufficient. I started
analyzing all existing users and use cases, trying to avoid walking page
tables.
If we want to get rid of all of (most) sub-page mapcounts, we'd probably
want:
(1) Total mapcount (compound + any sub-page): page_mapped(), pagecount
vs. refcount games, ...
(2) Compound mapcount (for PMD/PUD-mappale THP only): (2) - (1) tells
you if it's only PMD mapped or also PTE-mapped. For example, for
statistics but also swapout code.
(3) Mapcount of first (or any other) subpage (compount+subpage): for
folio_estimated_sharers().
For anon pages, I'm thinking about remembering an additional
(1) Page/folio creator (MM pointer/identification)
(2) Page/folio creator mapcount
When optimizing a PTE-mapped THP (especially not- pmd-mappale) for the
fork()+exec() case, we'd have to walk page tables to see if all folio
references come from this MM. The page/folio creator exactly avoids that
completely. We might need a mechanism to synchronize against
mapping/unmapping of this folio from the creator concurrently (relevant
when mapped into multiple page tables).
Further, for (1) we'd want a 64bit mapcount for large folios, which
implies a 64bit refcount. For smallish folios, we don't really care.
We should most probably use a bi-weekly MM meeting to discuss that.
Hopefully, I have a full understanding of all use cases and requirements
until then. Don't have sufficient time to look into all the nasty
details right now.
--
Cheers,
David / dhildenb
next prev parent reply other threads:[~2023-07-02 11:45 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-01-24 18:13 Folio mapcount Matthew Wilcox
2023-01-24 18:35 ` David Hildenbrand
2023-01-24 18:37 ` David Hildenbrand
2023-01-24 18:35 ` Yang Shi
2023-02-02 3:45 ` Mike Kravetz
2023-02-02 15:31 ` Matthew Wilcox
2023-02-07 16:19 ` Zi Yan
2023-02-07 16:44 ` Matthew Wilcox
2023-02-06 20:34 ` Matthew Wilcox
2023-02-06 22:55 ` Yang Shi
2023-02-06 23:09 ` Matthew Wilcox
2023-02-07 3:06 ` Yin, Fengwei
2023-02-07 4:08 ` Matthew Wilcox
2023-02-07 22:39 ` Peter Xu
2023-02-07 23:27 ` Matthew Wilcox
2023-02-08 19:40 ` Peter Xu
2023-02-08 20:25 ` Matthew Wilcox
2023-02-08 20:58 ` Peter Xu
2023-02-09 15:10 ` Chih-En Lin
2023-02-09 15:43 ` Peter Xu
2023-02-07 22:56 ` James Houghton
2023-02-07 23:08 ` Matthew Wilcox
2023-02-07 23:27 ` James Houghton
2023-02-07 23:35 ` Matthew Wilcox
2023-02-08 0:35 ` James Houghton
2023-02-08 2:26 ` Matthew Wilcox
2023-02-07 16:23 ` Zi Yan
2023-02-07 16:51 ` Matthew Wilcox
2023-02-08 19:36 ` Zi Yan
2023-02-08 19:54 ` Matthew Wilcox
2023-02-10 15:15 ` Zi Yan
2023-03-29 14:02 ` Yin, Fengwei
2023-07-01 1:17 ` Zi Yan
2023-07-02 9:50 ` Yin, Fengwei
2023-07-02 11:45 ` David Hildenbrand [this message]
2023-07-02 12:26 ` Matthew Wilcox
2023-07-03 20:54 ` David Hildenbrand
2023-07-02 19:51 ` Zi Yan
2023-07-03 1:09 ` Yin, Fengwei
2023-07-03 13:24 ` Zi Yan
2023-07-03 20:46 ` David Hildenbrand
2023-07-04 1:22 ` Yin, Fengwei
2023-07-04 2:25 ` Matthew Wilcox
2023-07-03 21:09 ` David Hildenbrand
-- strict thread matches above, loose matches on Subject: below --
2021-12-15 21:55 folio mapcount Matthew Wilcox
2021-12-16 9:37 ` Kirill A. Shutemov
2021-12-16 13:56 ` Matthew Wilcox
2021-12-16 15:19 ` Jason Gunthorpe
2021-12-16 15:54 ` Matthew Wilcox
2021-12-16 16:45 ` David Hildenbrand
2021-12-16 17:01 ` Jason Gunthorpe
2021-12-16 18:56 ` Kirill A. Shutemov
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=6cec6f68-248e-63b4-5615-9e0f3f819a0a@redhat.com \
--to=david@redhat.com \
--cc=fengwei.yin@intel.com \
--cc=hughd@google.com \
--cc=linux-mm@kvack.org \
--cc=riel@surriel.com \
--cc=vishal.moola@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).