All of lore.kernel.org
 help / color / mirror / Atom feed
From: Johannes Weiner <jweiner@redhat.com>
To: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Konstantin Khlebnikov <khlebnikov@parallels.com>,
	Pekka Enberg <penberg@kernel.org>,
	"linux-mm@kvack.org" <linux-mm@kvack.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	Wu Fengguang <fengguang.wu@intel.com>,
	Johannes Weiner <hannes@cmpxchg.org>,
	Rik van Riel <riel@redhat.com>, Mel Gorman <mel@csn.ul.ie>,
	Minchan Kim <minchan.kim@gmail.com>,
	Gene Heskett <gene.heskett@gmail.com>
Subject: Re: [rfc 3/3] mm: vmscan: revert file list boost on lru addition
Date: Thu, 10 Nov 2011 17:12:44 +0100	[thread overview]
Message-ID: <20111110161244.GN3153@redhat.com> (raw)
In-Reply-To: <20111107114520.33050d75.kamezawa.hiroyu@jp.fujitsu.com>

On Mon, Nov 07, 2011 at 11:45:20AM +0900, KAMEZAWA Hiroyuki wrote:
> On Wed, 2 Nov 2011 17:32:47 +0100
> Johannes Weiner <jweiner@redhat.com> wrote:
> 
> > The idea in 9ff473b 'vmscan: evict streaming IO first' was to steer
> > reclaim focus onto file pages with every new file page that hits the
> > lru list, so that an influx of used-once file pages does not lead to
> > swapping of anonymous pages.
> > 
> > The problem is that nobody is fixing up the balance if the pages in
> > fact become part of the resident set.
> > 
> > Anonymous page creation is neutral to the inter-lru balance, so even a
> > comparably tiny number of heavily used file pages tip the balance in
> > favor of the file list.
> > 
> > In addition, there is no refault detection, and every refault will
> > bias the balance even more.  A thrashing file working set will be
> > mistaken for a very lucrative source of reclaimable pages.
> > 
> > As anonymous pages are no longer swapped above a certain priority
> > level, this mechanism is no longer needed.  Used-once file pages
> > should get reclaimed before the VM even considers swapping.
> > 
> > Signed-off-by: Johannes Weiner <jweiner@redhat.com>
> 
> Do you have some results ?

Not yet, sorry, I had to drop it all and do something else.

This change relies on the VM having a different mechanism to go for
one-shot file cache first, so I need to address Kosaki-san's concerns
about 1/3 before pursuing this patch.

WARNING: multiple messages have this Message-ID (diff)
From: Johannes Weiner <jweiner@redhat.com>
To: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Konstantin Khlebnikov <khlebnikov@parallels.com>,
	Pekka Enberg <penberg@kernel.org>,
	"linux-mm@kvack.org" <linux-mm@kvack.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	Wu Fengguang <fengguang.wu@intel.com>,
	Johannes Weiner <hannes@cmpxchg.org>,
	Rik van Riel <riel@redhat.com>, Mel Gorman <mel@csn.ul.ie>,
	Minchan Kim <minchan.kim@gmail.com>,
	Gene Heskett <gene.heskett@gmail.com>
Subject: Re: [rfc 3/3] mm: vmscan: revert file list boost on lru addition
Date: Thu, 10 Nov 2011 17:12:44 +0100	[thread overview]
Message-ID: <20111110161244.GN3153@redhat.com> (raw)
In-Reply-To: <20111107114520.33050d75.kamezawa.hiroyu@jp.fujitsu.com>

On Mon, Nov 07, 2011 at 11:45:20AM +0900, KAMEZAWA Hiroyuki wrote:
> On Wed, 2 Nov 2011 17:32:47 +0100
> Johannes Weiner <jweiner@redhat.com> wrote:
> 
> > The idea in 9ff473b 'vmscan: evict streaming IO first' was to steer
> > reclaim focus onto file pages with every new file page that hits the
> > lru list, so that an influx of used-once file pages does not lead to
> > swapping of anonymous pages.
> > 
> > The problem is that nobody is fixing up the balance if the pages in
> > fact become part of the resident set.
> > 
> > Anonymous page creation is neutral to the inter-lru balance, so even a
> > comparably tiny number of heavily used file pages tip the balance in
> > favor of the file list.
> > 
> > In addition, there is no refault detection, and every refault will
> > bias the balance even more.  A thrashing file working set will be
> > mistaken for a very lucrative source of reclaimable pages.
> > 
> > As anonymous pages are no longer swapped above a certain priority
> > level, this mechanism is no longer needed.  Used-once file pages
> > should get reclaimed before the VM even considers swapping.
> > 
> > Signed-off-by: Johannes Weiner <jweiner@redhat.com>
> 
> Do you have some results ?

Not yet, sorry, I had to drop it all and do something else.

This change relies on the VM having a different mechanism to go for
one-shot file cache first, so I need to address Kosaki-san's concerns
about 1/3 before pursuing this patch.

--
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/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  reply	other threads:[~2011-11-10 16:14 UTC|newest]

Thread overview: 64+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-08-08 11:06 [PATCH 1/2] vmscan: promote shared file mapped pages Konstantin Khlebnikov
2011-08-08 11:06 ` Konstantin Khlebnikov
2011-08-08 11:07 ` [PATCH 2/2] vmscan: activate executable pages after first usage Konstantin Khlebnikov
2011-08-08 11:07   ` Konstantin Khlebnikov
2011-08-08 23:58   ` KAMEZAWA Hiroyuki
2011-08-08 23:58     ` KAMEZAWA Hiroyuki
2011-08-09  0:02   ` Minchan Kim
2011-08-09  0:02     ` Minchan Kim
2011-08-09  0:04     ` KAMEZAWA Hiroyuki
2011-08-09  0:04       ` KAMEZAWA Hiroyuki
2011-08-09  0:26       ` Minchan Kim
2011-08-09  0:26         ` Minchan Kim
2011-08-09  1:23   ` Shaohua Li
2011-08-09  1:23     ` Shaohua Li
2011-08-08 11:37 ` [PATCH 1/2] vmscan: promote shared file mapped pages Pekka Enberg
2011-08-08 11:37   ` Pekka Enberg
2011-08-08 12:18   ` Konstantin Khlebnikov
2011-08-08 12:18     ` Konstantin Khlebnikov
2011-08-08 12:40     ` Pekka Enberg
2011-08-08 12:40       ` Pekka Enberg
2011-08-08 12:51       ` Konstantin Khlebnikov
2011-08-08 12:51         ` Konstantin Khlebnikov
2011-08-18  9:09     ` Johannes Weiner
2011-08-18  9:09       ` Johannes Weiner
2011-11-02 16:30     ` Johannes Weiner
2011-11-02 16:30       ` Johannes Weiner
2011-11-02 16:31       ` [rfc 1/3] mm: vmscan: never swap under low memory pressure Johannes Weiner
2011-11-02 16:31         ` Johannes Weiner
2011-11-02 17:54         ` KOSAKI Motohiro
2011-11-02 17:54           ` KOSAKI Motohiro
2011-11-03 15:51           ` Johannes Weiner
2011-11-03 15:51             ` Johannes Weiner
2011-11-08  0:16             ` KOSAKI Motohiro
2011-11-08  0:16               ` KOSAKI Motohiro
2011-11-07  2:29         ` KAMEZAWA Hiroyuki
2011-11-07  2:29           ` KAMEZAWA Hiroyuki
2011-11-10 15:29           ` Johannes Weiner
2011-11-10 15:29             ` Johannes Weiner
2011-11-02 16:32       ` [rfc 2/3] mm: vmscan: treat inactive cycling as neutral Johannes Weiner
2011-11-02 16:32         ` Johannes Weiner
2011-11-02 18:04         ` KOSAKI Motohiro
2011-11-02 18:04           ` KOSAKI Motohiro
2011-11-03 12:49           ` Johannes Weiner
2011-11-03 12:49             ` Johannes Weiner
2011-11-07  2:34         ` KAMEZAWA Hiroyuki
2011-11-07  2:34           ` KAMEZAWA Hiroyuki
2011-11-10 16:06           ` Johannes Weiner
2011-11-10 16:06             ` Johannes Weiner
2011-11-11  0:05             ` KAMEZAWA Hiroyuki
2011-11-11  0:05               ` KAMEZAWA Hiroyuki
2011-11-02 16:32       ` [rfc 3/3] mm: vmscan: revert file list boost on lru addition Johannes Weiner
2011-11-02 16:32         ` Johannes Weiner
2011-11-07  2:45         ` KAMEZAWA Hiroyuki
2011-11-07  2:45           ` KAMEZAWA Hiroyuki
2011-11-10 16:12           ` Johannes Weiner [this message]
2011-11-10 16:12             ` Johannes Weiner
2011-11-02 16:35       ` [PATCH 1/2] vmscan: promote shared file mapped pages Johannes Weiner
2011-11-02 16:35         ` Johannes Weiner
2011-08-08 23:36 ` Minchan Kim
2011-08-08 23:36   ` Minchan Kim
2011-08-08 23:51 ` KAMEZAWA Hiroyuki
2011-08-08 23:51   ` KAMEZAWA Hiroyuki
2011-10-31 20:12 ` Andrew Morton
2011-10-31 20:12   ` Andrew Morton

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=20111110161244.GN3153@redhat.com \
    --to=jweiner@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=fengguang.wu@intel.com \
    --cc=gene.heskett@gmail.com \
    --cc=hannes@cmpxchg.org \
    --cc=kamezawa.hiroyu@jp.fujitsu.com \
    --cc=khlebnikov@parallels.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mel@csn.ul.ie \
    --cc=minchan.kim@gmail.com \
    --cc=penberg@kernel.org \
    --cc=riel@redhat.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.