From: Gleb Natapov <gleb@redhat.com>
To: Stefan Hajnoczi <stefanha@gmail.com>
Cc: Peter Lieven <pl@dlh.net>, qemu-devel@nongnu.org, kvm@vger.kernel.org
Subject: Re: linux guests and ksm performance
Date: Fri, 24 Feb 2012 09:04:25 +0200 [thread overview]
Message-ID: <20120224070425.GB27946@redhat.com> (raw)
In-Reply-To: <CAJSP0QXPzBHopybzkncedQjzGiakbizs3nQMc0R7s2TNqwDOew@mail.gmail.com>
On Thu, Feb 23, 2012 at 04:42:54PM +0000, Stefan Hajnoczi wrote:
> On Thu, Feb 23, 2012 at 3:40 PM, Peter Lieven <pl@dlh.net> wrote:
> > However, in a virtual machine I have not observed the above slow down to
> > that extend
> > while the benefit of zero after free in a virtualisation environment is
> > obvious:
> >
> > 1) zero pages can easily be merged by ksm or other technique.
> > 2) zero (dup) pages are a lot faster to transfer in case of migration.
>
> The other approach is a memory page "discard" mechanism - which
> obviously requires more code changes than zeroing freed pages.
>
> The advantage is that we don't take the brute-force and CPU intensive
> approach of zeroing pages. It would be like a fine-grained ballooning
> feature.
>
> I hope someone will follow up saying this has already been done or
> prototyped :).
>
That was attempted. It is called "page hinting", but AFAIK due to
complex locking issue attempt was abandoned.
--
Gleb.
WARNING: multiple messages have this Message-ID (diff)
From: Gleb Natapov <gleb@redhat.com>
To: Stefan Hajnoczi <stefanha@gmail.com>
Cc: Peter Lieven <pl@dlh.net>, qemu-devel@nongnu.org, kvm@vger.kernel.org
Subject: Re: [Qemu-devel] linux guests and ksm performance
Date: Fri, 24 Feb 2012 09:04:25 +0200 [thread overview]
Message-ID: <20120224070425.GB27946@redhat.com> (raw)
In-Reply-To: <CAJSP0QXPzBHopybzkncedQjzGiakbizs3nQMc0R7s2TNqwDOew@mail.gmail.com>
On Thu, Feb 23, 2012 at 04:42:54PM +0000, Stefan Hajnoczi wrote:
> On Thu, Feb 23, 2012 at 3:40 PM, Peter Lieven <pl@dlh.net> wrote:
> > However, in a virtual machine I have not observed the above slow down to
> > that extend
> > while the benefit of zero after free in a virtualisation environment is
> > obvious:
> >
> > 1) zero pages can easily be merged by ksm or other technique.
> > 2) zero (dup) pages are a lot faster to transfer in case of migration.
>
> The other approach is a memory page "discard" mechanism - which
> obviously requires more code changes than zeroing freed pages.
>
> The advantage is that we don't take the brute-force and CPU intensive
> approach of zeroing pages. It would be like a fine-grained ballooning
> feature.
>
> I hope someone will follow up saying this has already been done or
> prototyped :).
>
That was attempted. It is called "page hinting", but AFAIK due to
complex locking issue attempt was abandoned.
--
Gleb.
next prev parent reply other threads:[~2012-02-24 7:04 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-02-23 15:40 linux guests and ksm performance Peter Lieven
2012-02-23 15:40 ` [Qemu-devel] " Peter Lieven
2012-02-23 16:42 ` Stefan Hajnoczi
2012-02-23 16:42 ` [Qemu-devel] " Stefan Hajnoczi
2012-02-23 18:39 ` Javier Guerra Giraldez
2012-02-23 18:39 ` [Qemu-devel] " Javier Guerra Giraldez
2012-02-23 19:08 ` peter.lieven@gmail.com
2012-02-23 19:08 ` [Qemu-devel] " peter.lieven@gmail.com
2012-02-24 6:41 ` Stefan Hajnoczi
2012-02-24 6:41 ` [Qemu-devel] " Stefan Hajnoczi
2012-02-24 6:53 ` Stefan Hajnoczi
2012-02-24 6:53 ` [Qemu-devel] " Stefan Hajnoczi
2012-02-24 7:23 ` Stefan Hajnoczi
2012-02-24 7:23 ` [Qemu-devel] " Stefan Hajnoczi
2012-02-24 7:43 ` Peter Lieven
2012-02-24 7:43 ` [Qemu-devel] " Peter Lieven
2012-02-28 11:45 ` Peter Lieven
2012-02-28 11:46 ` Peter Lieven
2012-02-28 12:05 ` Stefan Hajnoczi
2012-02-28 12:16 ` Peter Lieven
2012-02-28 12:16 ` Peter Lieven
2012-02-28 13:16 ` Avi Kivity
2012-02-28 13:16 ` [Qemu-devel] " Avi Kivity
2012-02-28 13:20 ` Peter Lieven
2012-02-28 13:20 ` [Qemu-devel] " Peter Lieven
2012-02-28 13:56 ` Avi Kivity
2012-02-28 13:56 ` [Qemu-devel] " Avi Kivity
2012-02-24 7:04 ` Gleb Natapov [this message]
2012-02-24 7:04 ` Gleb Natapov
2012-02-28 13:14 ` Avi Kivity
2012-02-28 13:14 ` [Qemu-devel] " Avi Kivity
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=20120224070425.GB27946@redhat.com \
--to=gleb@redhat.com \
--cc=kvm@vger.kernel.org \
--cc=pl@dlh.net \
--cc=qemu-devel@nongnu.org \
--cc=stefanha@gmail.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.