From: Kalesh Singh <kaleshsingh@google.com>
To: anthony.yznaga@oracle.com
Cc: lsf-pc@lists.linux-foundation.org, linux-mm@kvack.org,
Isaac Manjarres <isaacmanjarres@google.com>,
Suren Baghdasaryan <surenb@google.com>,
android-mm <android-mm@google.com>
Subject: Re: [LSF/MM/BPF TOPIC] mshare - sharing page tables across processes
Date: Wed, 25 Feb 2026 16:59:32 -0800 [thread overview]
Message-ID: <CAC_TJvdzgOdvf9=uDaF=zBFWdZ5J9Qh=04Y9OHry6V+YiA-3EQ@mail.gmail.com> (raw)
In-Reply-To: <ea6c914d-44a4-4958-adb6-642e86488045@oracle.com>
On Mon, Feb 23, 2026 at 6:12 PM <anthony.yznaga@oracle.com> wrote:
>
> mshare is a new form of shared memory that shares resources between
> processes, namely the page tables entries and VMAs of the shared
> objects. Advantages include reduced memory consumption through
> eliminating the need for duplicated PTEs in each process, reduced
> overhead needed to fault shared pages into multiple processes, and the
> ability for updates to the memory such as protection changes to be
> immdediately visible to all sharing processes.
>
> For some time now the proposed API has continued to evolve. With
> feedback from mm maintainers, a new approach using syscalls is being
> taken to create, attach, and deattach mshare regions and to map and
> manipulate the memory objects within a region.
>
> Let's discuss the current state of the API and technical challenges of
> managing and accomodating shared page tables in the kernel.
>
> Anthony
>
>
> References:
>
> https://lore.kernel.org/linux-mm/20250820010415.699353-1-anthony.yznaga@oracle.com/
> https://lore.kernel.org/linux-mm/20250404021902.48863-1-anthony.yznaga@oracle.com/
Hi Anthony,
I would be very interested in participating in this discussion at LSF/MM.
As mentioned on the mailing list, we are currently exploring similar
optimizations to deduplicate page tables for Android's Zygote process.
I'd love to join the session to discuss how the proposed sharing
mechanisms might align.
Thanks,
Kalesh
>
>
prev parent reply other threads:[~2026-02-26 0:59 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-02-24 2:11 anthony.yznaga
2026-02-26 0:59 ` Kalesh Singh [this message]
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='CAC_TJvdzgOdvf9=uDaF=zBFWdZ5J9Qh=04Y9OHry6V+YiA-3EQ@mail.gmail.com' \
--to=kaleshsingh@google.com \
--cc=android-mm@google.com \
--cc=anthony.yznaga@oracle.com \
--cc=isaacmanjarres@google.com \
--cc=linux-mm@kvack.org \
--cc=lsf-pc@lists.linux-foundation.org \
--cc=surenb@google.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