All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Austin S. Hemmelgarn" <ahferroin7@gmail.com>
To: "Daniel Díaz" <daniel.diaz@linaro.org>, x86@kernel.org
Cc: anders.roxell@linaro.org, Thomas Gleixner <tglx@linutronix.de>,
	Ingo Molnar <mingo@redhat.com>, "H. Peter Anvin" <hpa@zytor.com>,
	Josh Poimboeuf <jpoimboe@redhat.com>,
	Krzysztof Kozlowski <krzk@kernel.org>,
	"open list:X86 ARCHITECTURE (32-BIT AND 64-BIT)" 
	<linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 1/2] x86: x86_64_defconfig: Enable KSM.
Date: Fri, 15 Jun 2018 08:23:39 -0400	[thread overview]
Message-ID: <bc4cf4cb-e2b5-47c3-e29f-8ddf27a6a307@gmail.com> (raw)
In-Reply-To: <1529016629-2948-1-git-send-email-daniel.diaz@linaro.org>

On 2018-06-14 18:50, Daniel Díaz wrote:
> As per the documentation, Kernel Samepage Merging (available
> since 2.6.32) is a memory-saving de-duplication feature,
> enabled by CONFIG_KSM=y and activated via sysfs. More
> information can be found here:
>    https://www.kernel.org/doc/Documentation/vm/ksm.txt
> 
> When enabled in the kernel, the default is to not do anything
> at all, until it is activated at run-time with:
>    echo 1 > /sys/kernel/mm/ksm/run
> 
As pointed out by a couple of others, this doesn't explain why this is a 
good idea.  All you're doing here is giving a reason that it won't have 
a negative impact on most users.

Two points that may be worth adding, but also don't really argue for it 
being a significant improvement:

* Pretty much all of the major distributions that use pre-built kernels 
have it enabled in their kernels (At minimum, Debian, Ubuntu, Fedora 
(and by extension RHEL and CentOS), openSUSE (and by extension SLES), 
Arch, and Alpine have it enabled), so enabling this in defconfig would 
bring it a bit closer to parity with distribution kernels in terms of 
core features.

* Software other than QEMU is starting to take advantage of it if 
available (for example, Netdata [1] can mark it's in-memory TSDB's for 
deduplication, which usually cuts it's memory usage roughly in half).


[1] https://my-netdata.io/

      parent reply	other threads:[~2018-06-15 12:23 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-06-14 22:50 [PATCH 1/2] x86: x86_64_defconfig: Enable KSM Daniel Díaz
2018-06-14 22:50 ` [PATCH 2/2] arm: multi_v7_defconfig: " Daniel Díaz
2018-06-14 22:50   ` Daniel Díaz
2018-06-15  6:09   ` Geert Uytterhoeven
2018-06-15  6:09     ` Geert Uytterhoeven
2018-06-15  8:27 ` [PATCH 1/2] x86: x86_64_defconfig: " Thomas Gleixner
2018-06-15 12:23 ` Austin S. Hemmelgarn [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=bc4cf4cb-e2b5-47c3-e29f-8ddf27a6a307@gmail.com \
    --to=ahferroin7@gmail.com \
    --cc=anders.roxell@linaro.org \
    --cc=daniel.diaz@linaro.org \
    --cc=hpa@zytor.com \
    --cc=jpoimboe@redhat.com \
    --cc=krzk@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=tglx@linutronix.de \
    --cc=x86@kernel.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 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.