All of lore.kernel.org
 help / color / mirror / Atom feed
From: Rik van Riel <riel@redhat.com>
To: Minchan Kim <minchan@kernel.org>
Cc: Rafael Aquini <aquini@redhat.com>,
	linux-mm@kvack.org, linux-kernel@vger.kernel.org,
	virtualization@lists.linux-foundation.org,
	Rusty Russell <rusty@rustcorp.com.au>,
	"Michael S. Tsirkin" <mst@redhat.com>, Mel Gorman <mel@csn.ul.ie>,
	Andi Kleen <andi@firstfloor.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>,
	Rafael Aquini <aquini@linux.com>
Subject: Re: [PATCH v4 1/3] mm: introduce compaction and migration for virtio ballooned pages
Date: Wed, 01 Aug 2012 16:51:25 -0400	[thread overview]
Message-ID: <501996CD.70007@redhat.com> (raw)
In-Reply-To: <20120723023332.GA6832@bbox>

On 07/22/2012 10:33 PM, Minchan Kim wrote:

> IMHO, better approach is that after we can get complete free pageblocks
> by compaction or reclaim, move balloon pages into that pageblocks and make
> that blocks to unmovable. It can prevent fragmentation and it makes
> current or future code don't need to consider balloon page.

I believe this is the wrong thing to do.

In a KVM guest, getting applications in transparent
huge pages can be a 10-25% performance benefit.

Therefore, we need to make all the 2MB pageblocks
we can available for use by userland.

Using 2MB blocks for the balloon (which is never
touched) is extremely wasteful and could result in
a large performance penalty, if we cannot defragment
the remaining memory enough to give 2MB pages to
applications.

The 2MB blocks are prime real estate. They should
remain available for applications.



WARNING: multiple messages have this Message-ID (diff)
From: Rik van Riel <riel@redhat.com>
To: Minchan Kim <minchan@kernel.org>
Cc: Rafael Aquini <aquini@redhat.com>,
	linux-mm@kvack.org, linux-kernel@vger.kernel.org,
	virtualization@lists.linux-foundation.org,
	Rusty Russell <rusty@rustcorp.com.au>,
	"Michael S. Tsirkin" <mst@redhat.com>, Mel Gorman <mel@csn.ul.ie>,
	Andi Kleen <andi@firstfloor.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>,
	Rafael Aquini <aquini@linux.com>
Subject: Re: [PATCH v4 1/3] mm: introduce compaction and migration for virtio ballooned pages
Date: Wed, 01 Aug 2012 16:51:25 -0400	[thread overview]
Message-ID: <501996CD.70007@redhat.com> (raw)
In-Reply-To: <20120723023332.GA6832@bbox>

On 07/22/2012 10:33 PM, Minchan Kim wrote:

> IMHO, better approach is that after we can get complete free pageblocks
> by compaction or reclaim, move balloon pages into that pageblocks and make
> that blocks to unmovable. It can prevent fragmentation and it makes
> current or future code don't need to consider balloon page.

I believe this is the wrong thing to do.

In a KVM guest, getting applications in transparent
huge pages can be a 10-25% performance benefit.

Therefore, we need to make all the 2MB pageblocks
we can available for use by userland.

Using 2MB blocks for the balloon (which is never
touched) is extremely wasteful and could result in
a large performance penalty, if we cannot defragment
the remaining memory enough to give 2MB pages to
applications.

The 2MB blocks are prime real estate. They should
remain available for applications.


--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  parent reply	other threads:[~2012-08-01 20:53 UTC|newest]

Thread overview: 62+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-07-17 16:50 [PATCH v4 0/3] make balloon pages movable by compaction Rafael Aquini
2012-07-17 16:50 ` Rafael Aquini
2012-07-17 16:50 ` [PATCH v4 1/3] mm: introduce compaction and migration for virtio ballooned pages Rafael Aquini
2012-07-17 16:50 ` Rafael Aquini
2012-07-17 16:50   ` Rafael Aquini
2012-07-18  5:48   ` Minchan Kim
2012-07-18  5:48   ` Minchan Kim
2012-07-18  5:48     ` Minchan Kim
2012-07-20 19:48     ` Rafael Aquini
2012-07-20 19:48       ` Rafael Aquini
2012-07-20 19:48       ` Rafael Aquini
2012-07-23  2:33       ` Minchan Kim
2012-07-23  2:33       ` Minchan Kim
2012-07-23  2:33         ` Minchan Kim
2012-07-23  2:35         ` Minchan Kim
2012-07-23  2:35         ` Minchan Kim
2012-07-23  2:35           ` Minchan Kim
2012-07-23 18:19         ` Rafael Aquini
2012-07-23 18:19           ` Rafael Aquini
2012-07-23 18:19           ` Rafael Aquini
2012-08-01 20:53           ` Rik van Riel
2012-08-01 20:53             ` Rik van Riel
2012-08-03 11:13             ` Rafael Aquini
2012-08-03 11:13               ` Rafael Aquini
2012-08-03 11:13               ` Rafael Aquini
2012-08-01 20:53           ` Rik van Riel
2012-08-01 20:51         ` Rik van Riel
2012-08-01 20:51         ` Rik van Riel [this message]
2012-08-01 20:51           ` Rik van Riel
2012-08-03 12:26         ` Rafael Aquini
2012-08-03 12:26           ` Rafael Aquini
2012-08-03 12:26         ` Rafael Aquini
2012-07-18 22:46   ` Andrew Morton
2012-07-18 22:46     ` Andrew Morton
2012-07-18 22:46     ` Andrew Morton
2012-07-18 23:07     ` Rafael Aquini
2012-07-18 23:07     ` Rafael Aquini
2012-07-18 23:07       ` Rafael Aquini
2012-07-18 23:12       ` Andrew Morton
2012-07-18 23:12         ` Andrew Morton
2012-07-18 23:12         ` Andrew Morton
2012-07-19  1:00         ` Rafael Aquini
2012-07-19  1:00         ` Rafael Aquini
2012-07-19  1:00           ` Rafael Aquini
2012-07-19  1:29           ` Andrew Morton
2012-07-19  1:29           ` Andrew Morton
2012-07-19  1:29             ` Andrew Morton
2012-07-19 12:32             ` Rafael Aquini
2012-07-19 12:32               ` Rafael Aquini
2012-07-19 12:32               ` Rafael Aquini
2012-07-17 16:50 ` [PATCH v4 2/3] virtio_balloon: introduce migration primitives to balloon pages Rafael Aquini
2012-07-17 16:50   ` Rafael Aquini
2012-07-18 22:49   ` Andrew Morton
2012-07-18 22:49     ` Andrew Morton
2012-07-18 22:49     ` Andrew Morton
2012-07-18 23:16     ` Rafael Aquini
2012-07-18 23:16     ` Rafael Aquini
2012-07-18 23:16       ` Rafael Aquini
2012-07-17 16:50 ` Rafael Aquini
2012-07-17 16:50 ` [PATCH v4 3/3] mm: add vm event counters for balloon pages compaction Rafael Aquini
2012-07-17 16:50 ` Rafael Aquini
2012-07-17 16:50   ` Rafael Aquini

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=501996CD.70007@redhat.com \
    --to=riel@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=andi@firstfloor.org \
    --cc=aquini@linux.com \
    --cc=aquini@redhat.com \
    --cc=konrad.wilk@oracle.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mel@csn.ul.ie \
    --cc=minchan@kernel.org \
    --cc=mst@redhat.com \
    --cc=rusty@rustcorp.com.au \
    --cc=virtualization@lists.linux-foundation.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.