From: Yilei Zheng <trilith.zheng@gmail.com>
To: Muchun Song <muchun.song@linux.dev>
Cc: Muchun Song <songmuchun@bytedance.com>,
Andrew Morton <akpm@linux-foundation.org>,
Dan Williams <djbw@kernel.org>,
David Hildenbrand <david@kernel.org>,
Oscar Salvador <osalvador@suse.de>,
Pankaj Gupta <pankaj.gupta@amd.com>,
linux-mm@kvack.org, nvdimm@lists.linux.dev,
linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org,
linux-cxl@vger.kernel.org,
Vishal Verma <vishal.l.verma@intel.com>,
Dave Jiang <dave.jiang@intel.com>,
Alison Schofield <alison.schofield@intel.com>,
Mike Rapoport <rppt@kernel.org>, Ira Weiny <iweiny@kernel.org>,
Jan Kara <jack@suse.cz>, Matthew Wilcox <willy@infradead.org>,
Lorenzo Stoakes <ljs@kernel.org>,
Vlastimil Babka <vbabka@kernel.org>,
Michal Hocko <mhocko@suse.com>, Qi Zheng <qi.zheng@linux.dev>
Subject: Re: [PATCH 3/4] mm: add shared read-only vmemmap support for FS-DAX
Date: Thu, 29 Oct 2026 14:45:10 +0800 [thread overview]
Message-ID: <20261029064510.17667-1-trilith.zheng@gmail.com> (raw)
In-Reply-To: <2E5C10CF-F3E0-4694-965C-B0E432A1A962@linux.dev>
> What's the size of the roots image? And I just want to know how much
> memory this series could save in your workload.
In our setup, the rootfs is an EROFS image exported from the OCI
container image, so the pmem device size follows the image size.
Most of our workloads are simple today, with images around 1 to 2 GiB,
which means 16 to 32 MiB of vmemmap overhead per guest. And we expect
the average image size to reach about 5 GiB as we add heavier desktop
and agentic RL environments. That would save roughly 80 MiB per guest.
At our target density of about 200 sandboxes per node, this adds up
to around 16 GiB of guest RAM saved per node. Memory is what caps
density here, so this saving helps a lot.
next prev parent reply other threads:[~2026-10-29 6:45 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-09-03 12:21 [PATCH 0/4] mm: Reduce struct page overhead for FS-DAX pmem Muchun Song
2026-09-03 12:21 ` [PATCH 1/4] mm: generalize vmemmap remap architecture support Muchun Song
2026-09-21 12:25 ` Oscar Salvador (SUSE)
2026-10-29 7:03 ` Mike Rapoport
2026-09-03 12:21 ` [PATCH 2/4] nvdimm/pmem: avoid HWPoison flag updates for clean pages Muchun Song
2026-09-21 7:33 ` Gupta, Pankaj
2026-09-21 12:53 ` Oscar Salvador (SUSE)
2026-09-22 2:33 ` Muchun Song
2026-09-03 12:21 ` [PATCH 3/4] mm: add shared read-only vmemmap support for FS-DAX Muchun Song
2026-09-21 6:42 ` Gupta, Pankaj
2026-09-21 9:33 ` Muchun Song
2026-09-21 11:05 ` Gupta, Pankaj
2026-09-21 13:09 ` Oscar Salvador (SUSE)
2026-09-22 3:17 ` Muchun Song
2026-10-27 6:57 ` Yilei Zheng
2026-10-28 9:15 ` Muchun Song
2026-10-29 3:05 ` Yilei Zheng
2026-10-29 5:35 ` Muchun Song
2026-10-29 6:45 ` Yilei Zheng [this message]
2026-10-29 7:12 ` Yilei Zheng
2026-10-29 7:31 ` Muchun Song
2026-09-03 12:21 ` [PATCH 4/4] fsdax: materialize pmem vmemmap metadata on faults Muchun Song
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=20261029064510.17667-1-trilith.zheng@gmail.com \
--to=trilith.zheng@gmail.com \
--cc=akpm@linux-foundation.org \
--cc=alison.schofield@intel.com \
--cc=dave.jiang@intel.com \
--cc=david@kernel.org \
--cc=djbw@kernel.org \
--cc=iweiny@kernel.org \
--cc=jack@suse.cz \
--cc=linux-cxl@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=ljs@kernel.org \
--cc=mhocko@suse.com \
--cc=muchun.song@linux.dev \
--cc=nvdimm@lists.linux.dev \
--cc=osalvador@suse.de \
--cc=pankaj.gupta@amd.com \
--cc=qi.zheng@linux.dev \
--cc=rppt@kernel.org \
--cc=songmuchun@bytedance.com \
--cc=vbabka@kernel.org \
--cc=vishal.l.verma@intel.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
all inboxes | Powered by JetHome®