Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.31 as permitted sender) client-ip=23.128.96.31;
MIME-Version: 1.0
References: <CAOgh=Fwb+JCTQ-iqzjq8st9qbvauxc4gqqafjWG2Xc08MeBabQ@mail.gmail.com>
 <941aff31-6aa4-4c37-bb94-547c46250304@linux.alibaba.com> <ZXgNQ85PdUKrQU1j@infradead.org>
 <58d175f8-a06e-4b00-95fe-1bd5a79106df@linux.alibaba.com> <ZXha1IxzRfhsRNOu@infradead.org>
In-Reply-To: <ZXha1IxzRfhsRNOu@infradead.org>
From:   Eric Curtin <ecurtin@redhat.com>
Date:   Tue, 12 Dec 2023 21:17:29 +0000
Message-ID: <CAOgh=Fw2TcOFgCz1HbU1E=_HGRnf1PTTG2Qp_nD1D9f083RwUA@mail.gmail.com>
Subject: Re: [RFC KERNEL] initoverlayfs - a scalable initial filesystem
To:     Christoph Hellwig <hch@infradead.org>
Cc:     Gao Xiang <hsiangkao@linux.alibaba.com>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        linux-unionfs@vger.kernel.org, linux-erofs@lists.ozlabs.org,
        Daan De Meyer <daan.j.demeyer@gmail.com>,
        Stephen Smoogen <ssmoogen@redhat.com>,
        Yariv Rachmani <yrachman@redhat.com>,
        Daniel Walsh <dwalsh@redhat.com>,
        Douglas Landgraf <dlandgra@redhat.com>,
        Alexander Larsson <alexl@redhat.com>,
        Colin Walters <walters@redhat.com>,
        Brian Masney <bmasney@redhat.com>,
        Eric Chanudet <echanude@redhat.com>,
        Pavol Brilla <pbrilla@redhat.com>,
        Lokesh Mandvekar <lmandvek@redhat.com>,
        =?UTF-8?B?UGV0ciDFoGFiYXRh?= <psabata@redhat.com>,
        Lennart Poettering <lennart@poettering.net>,
        Luca Boccassi <bluca@debian.org>, Neal Gompa <neal@gompa.dev>,
        nvdimm@lists.linux.dev
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Tue, 12 Dec 2023 at 13:06, Christoph Hellwig <hch@infradead.org> wrote:
>
> On Tue, Dec 12, 2023 at 03:50:25PM +0800, Gao Xiang wrote:
> > I have no idea how it's faster than the current initramfs or initrd.
> > So if it's really useful, maybe some numbers can be posted first
> > with the current `memmap` hack and see it's worth going further with
> > some new infrastructure like initdax.
>
> Agreed.
>

I was politely poked this morning to highlight the graphs on the
initoverlayfs page, so as promised highlighting. That's not to say
this is either kernelspace's or userspace's role to optimize, but it
does prove there are benefits if we put some effort into optimizing
early boot.

https://github.com/containers/initoverlayfs

With this approach systemd starts ~300ms faster on a Raspberry Pi 4
with sd card, and this systemd instance has access to all the files
that a traditional initramfs would. I did this test on a Raspberry Pi
4 with NVMe drive over USB and the results were closer to a 500ms
benefit in systemd start time.

Is mise le meas/Regards,

Eric Curtin