linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Hans Reiser <hans@reiser.to>
To: "Stephen C. Tweedie" <sct@redhat.com>
Cc: "Quintela Carreira Juan J." <quintela@fi.udc.es>,
	Rik van Riel <riel@conectiva.com.br>, bert hubert <ahu@ds9a.nl>,
	linux-kernel@vger.rutgers.edu, Chris Mason <mason@suse.com>,
	linux-mm@kvack.org,
	Alexander Zarochentcev <zam@odintsovo.comcor.ru>
Subject: Re: journaling & VM
Date: Wed, 07 Jun 2000 15:20:23 -0700	[thread overview]
Message-ID: <393ECAA7.AD0DC0D6@reiser.to> (raw)
In-Reply-To: <20000607223352.J30951@redhat.com>

"Stephen C. Tweedie" wrote:
> 
> Hi,
> 
> On Wed, Jun 07, 2000 at 02:29:25PM -0700, Hans Reiser wrote:
> >
> > If I understand Juan correctly, they fixed this issue.  Aging 1/64th of the
> > cache for every cache evenly at every round of trying to free pages should be an
> > excellent fix.  It should do just fine at the task of handling a system with
> > both ext3 and reiserfs running.
> 
> That is _exactly_ what breaks the VM balance!  The net result of
> an algorithm like that is that all caches are shrunk at the same
> rate regardless of which ones are busy.  The "shrink everything
> at once" principle is what used to cause large filesystem scans
> (such as find|grep over a large source tree) to swap all our
> running processes out.
> 
> There _has_ to be a way to allow the relative ages of the different
> pages to influence the reclamation of pages from different sources.
> 
> Cheers,
>  Stephen

I am confused, if a page is accessed the aging is undone.  Aging 1/64th is not
the same as flushing 1/64th.  If cache A is not used the aging process gradually
shrinks it to nothing because its pages aren't unaged, if cache B is heavily
used the aging process doesn't age fast enough to overcome the unaging and new
pages get added and it grows.  I am missing something....

Hans
--
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.eu.org/Linux-MM/

  reply	other threads:[~2000-06-07 22:20 UTC|newest]

Thread overview: 60+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <Pine.LNX.4.10.10006060811120.15888-100000@dax.joh.cam.ac.uk>
     [not found] ` <393CA40C.648D3261@reiser.to>
     [not found]   ` <20000606114851.A30672@home.ds9a.nl>
     [not found]     ` <393CBBB8.554A0D2A@reiser.to>
     [not found]       ` <20000606172606.I25794@redhat.com>
     [not found]         ` <393D37D1.1BC61DC3@reiser.to>
     [not found]           ` <20000606205447.T23701@redhat.com>
2000-06-06 23:06             ` journaling & VM (was: Re: reiserfs being part of the kernel: it's not just the code) Rik van Riel
2000-06-07  1:19               ` journaling & VM (was: Re: reiserfs being part of the kernel: it'snot " Hans Reiser
2000-06-07  1:46                 ` Quintela Carreira Juan J.
2000-06-07  3:45                   ` Hans Reiser
2000-06-07 11:15                     ` Stephen C. Tweedie
2000-06-07 13:23                       ` Rik van Riel
2000-06-07 13:41                         ` Stephen C. Tweedie
2000-06-07 14:27                           ` Rik van Riel
2000-06-07 14:46                             ` Stephen C. Tweedie
2000-06-07 14:51                               ` bert hubert
2000-06-07 15:20                               ` Quintela Carreira Juan J.
2000-06-07 15:35                                 ` Stephen C. Tweedie
2000-06-07 15:41                                   ` Rik van Riel
2000-06-07 15:44                                   ` Juan J. Quintela
2000-06-07 17:10                                   ` Jeff V. Merkey
2000-06-07 17:14                                     ` Stephen C. Tweedie
2000-06-07 17:21                                       ` Jeff V. Merkey
2000-06-07 20:16                                   ` Hans Reiser
2000-06-07 21:20                                     ` Rik van Riel
2000-06-07 21:52                                       ` journaling & VM Hans Reiser
2000-06-07 22:11                                         ` James Sutherland
2000-06-07 22:29                                           ` Rik van Riel
2000-06-08  1:11                                         ` Neil Schemenauer
2000-06-08  1:29                                           ` Rik van Riel
2000-06-07 20:16                                 ` journaling & VM (was: Re: reiserfs being part of the kernel: it'snot just the code) Hans Reiser
2000-06-07 20:54                                   ` Stephen C. Tweedie
2000-06-07 21:29                                     ` Hans Reiser
2000-06-07 21:31                                       ` Rik van Riel
2000-06-07 21:33                                       ` Stephen C. Tweedie
2000-06-07 22:20                                         ` Hans Reiser [this message]
2000-06-07 21:50                                       ` Juan J. Quintela
2000-06-07 19:02                         ` journaling & VM (was: Re: reiserfs being part of the kernel:it'snot " Hans Reiser
2000-06-07 13:40                       ` journaling & VM (was: Re: reiserfs being part of the kernel: it'snot " Chris Mason
2000-06-07 13:47                         ` Stephen C. Tweedie
2000-06-07 11:12                 ` Stephen C. Tweedie
2000-06-07 16:35                   ` journaling & VM John Fremlin
2000-06-07 17:11                     ` Stephen C. Tweedie
     [not found]                       ` <20000608114435.A15433@uni-koblenz.de>
2000-06-08 21:29                         ` Stephen C. Tweedie
2000-06-09 11:53                           ` Ralf Baechle
2000-06-07 17:48                   ` journaling & VM (was: Re: reiserfs being part of the kernel: it'snot just the code) Hans Reiser
2000-06-07 18:01                     ` Rik van Riel
2000-06-07 19:58                       ` Stephen C. Tweedie
2000-06-07 20:56                         ` Juan J. Quintela
2000-06-07 21:14                           ` Rik van Riel
2000-06-07 21:24                           ` Stephen C. Tweedie
2000-06-07 21:40                             ` Juan J. Quintela
2000-06-07 21:49                               ` Stephen C. Tweedie
2000-06-07 22:00                                 ` Juan J. Quintela
2000-06-07 22:22                                 ` Manfred Spraul
2000-06-09 15:08                                   ` Rik van Riel
2000-06-09 16:52                                     ` Manfred Spraul
2000-06-09 17:23                                       ` Rik van Riel
2000-06-09 18:26                                         ` journaling & VM (was: Re: reiserfs being part of the kernel:it'snot " Manfred Spraul
2000-06-07 22:28                                 ` journaling & VM (was: Re: reiserfs being part of the kernel: it'snot " Hans Reiser
2000-06-07 10:10               ` journaling & VM (was: Re: reiserfs being part of the kernel: it's not " Stephen C. Tweedie
     [not found]             ` <393DACC8.5DB60A81@reiser.to>
2000-06-07 11:00               ` reiserfs being part of the kernel: it's not just the code Stephen C. Tweedie
2000-06-07 17:11                 ` Rik van Riel
2000-06-07 17:13                   ` Stephen C. Tweedie
2000-06-07 17:46                 ` Hans Reiser
2000-06-07 19:53                   ` Stephen C. Tweedie

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=393ECAA7.AD0DC0D6@reiser.to \
    --to=hans@reiser.to \
    --cc=ahu@ds9a.nl \
    --cc=linux-kernel@vger.rutgers.edu \
    --cc=linux-mm@kvack.org \
    --cc=mason@suse.com \
    --cc=quintela@fi.udc.es \
    --cc=riel@conectiva.com.br \
    --cc=sct@redhat.com \
    --cc=zam@odintsovo.comcor.ru \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox