All of lore.kernel.org
 help / color / mirror / Atom feed
From: Sedat Dilek <sedat.dilek@gmail.com>
To: efremov@linux.com
Cc: linux-kbuild@vger.kernel.org,
	Masahiro Yamada <masahiroy@kernel.org>,
	Nick Terrell <terrelln@fb.com>
Subject: Re: [PATCH] kbuild: add variables for compression tools
Date: Thu, 4 Jun 2020 18:07:56 +0200	[thread overview]
Message-ID: <CA+icZUUGgtsxw7UXDQYpouRuRhFWnfA3-XN7zp3OK7Yv2Ase+w@mail.gmail.com> (raw)
In-Reply-To: <635d02d6-05a9-7620-c386-dac766cd49da@linux.com>

On Thu, Jun 4, 2020 at 3:55 PM Denis Efremov <efremov@linux.com> wrote:
>
> Hi,
>
> On 6/4/20 4:05 PM, Sedat Dilek wrote:
> > Hi Denis,
> >
> > is it possible to add ZSTD compression support with the possibility to
> > add - for example multithreading - options?
>
> Where do you want to use zstd in kernel?
>
> For example, to compress headers xz is used, to compress /proc/config.gz
> gzip is used. It could be relatively easy patched to use zstd for it and
> it could be easy to add tarzstd-pkg target for building the kernel
> as a zstd compressed tarball. However, I doubt that someone really needs it.
> xz is best for headers because of compression ratio, gzip is best for config
> because gzip is installed by default on most systems.
> To compress modules, ramdisk or kernel image with zstd one needs
> to implement zstd decompression algorithm in kernel. Given that there
> are already gzip,bzip2,lzma,xz,lzo,lz4 I really doubt that it will be
> useful. There should be a good comparison showing that kernel image
> with zstd, for example, beats gzip and couple of others in comression ratio
> and decompression speed to properly position the zstd algo.
>

[ CC Nick ]

Hi,

me and other people use "support for ZSTD-compressed kernel and
initramfs" for a long time successfully.

I have modified Debian's initramfs-tools package to support creating
such an initrd.img and boot it successfully.
See Debian Bug #955469.

There is a pull-request sent out to integrate into Linus tree.

Some numbers from Nick Terell in his pull-request:
"
The zstd compressed kernel is smaller than the gzip compressed kernel but larger
than the xz or lzma compressed kernels, and it decompresses faster than
everything except lz4. See the table below for the measurement of an x86_64
kernel ordered by compressed size:

algo size
xz   6,509,792
lzma 6,856,576
zstd 7,399,157
gzip 8,522,527
bzip 8,629,603
lzo 9,808,035
lz4 10,705,570
none 32,565,672
"

More numbers in [0].

Hope this helps you to position the zstd algo.

Maybe, Nick can give you some more details.

Thanks.

Regards,
- Sedat -

[0] https://lwn.net/Articles/817134/
[1] https://github.com/terrelln/linux/commits/zstd-v5
[2] https://lkml.org/lkml/2020/6/1/1590
[3] https://bugs.debian.org/955469

> > For the documentation - which I have not checked - are the tools and
> > libs mentioned you will need to have installed for certain compression
> > tools?
>
> gzip, bzip2 are already installed on most systems. It's highly likely that
> lzo, lzma, xz will require you to install them.
>
> Thanks,
> Denis

  reply	other threads:[~2020-06-04 16:08 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-06-04 13:05 [PATCH] kbuild: add variables for compression tools Sedat Dilek
2020-06-04 13:55 ` Denis Efremov
2020-06-04 16:07   ` Sedat Dilek [this message]
2020-06-04 17:01     ` Denis Efremov
2020-06-04 17:05       ` Sedat Dilek
2020-06-04 17:18         ` Sedat Dilek

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=CA+icZUUGgtsxw7UXDQYpouRuRhFWnfA3-XN7zp3OK7Yv2Ase+w@mail.gmail.com \
    --to=sedat.dilek@gmail.com \
    --cc=efremov@linux.com \
    --cc=linux-kbuild@vger.kernel.org \
    --cc=masahiroy@kernel.org \
    --cc=terrelln@fb.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.