public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Daniel Phillips <phillips@bonn-fries.net>
To: Anton Altaparmakov <aia21@cam.ac.uk>
Cc: Rik van Riel <riel@conectiva.com.br>, <jlnance@intrex.net>,
	<linux-kernel@vger.kernel.org>
Subject: Re: [RFC] Optimization for use-once pages
Date: Wed, 25 Jul 2001 03:30:34 +0200	[thread overview]
Message-ID: <0107250330340D.00520@starship> (raw)
In-Reply-To: <Pine.LNX.4.33L.0107241355090.20326-100000@duckman.distro.conectiva> <5.1.0.14.2.20010725013436.00a91880@pop.cus.cam.ac.uk>
In-Reply-To: <5.1.0.14.2.20010725013436.00a91880@pop.cus.cam.ac.uk>

On Wednesday 25 July 2001 02:43, Anton Altaparmakov wrote:
> At 01:04 25/07/2001, Daniel Phillips wrote:
> >At that size you'd run a real risk of missing the tell-tale multiple
> >references that mark a page as frequently used.  Think about
> > metadata here (right now, that usually just means directory pages,
> > but later... who knows).
>
> This is not actually implemented yet, but NTFS TNG will use the page
> cache to hold both the LCN (physical clusters) and MFT (on disk
> inodes) allocation bitmaps in addition to file and directory pages.
> (Admittedly the LCN case folds into the file pages in page cache one
> as the LCN bitmap is just stored inside the usual data of a file
> called $Bitmap, but the MFT case is more complex as it is in an
> additional attribute inside the file $MFT so the normal file
> read/write functions definitely cannot be used. The usual data here
> is the actual on disk inodes...)
>
> Just FYI.

Yes, not a surprise.  Plus, I was working on an experimental patch to 
put Ext2 index blocks into the page cache just before I got off on this 
use-once tangent.  Time to go back to that...

--
Daniel

  reply	other threads:[~2001-07-25  1:26 UTC|newest]

Thread overview: 58+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2001-07-24  3:47 [RFC] Optimization for use-once pages Daniel Phillips
2001-07-24 12:38 ` jlnance
2001-07-24 16:56   ` Rik van Riel
2001-07-25  0:04     ` Daniel Phillips
2001-07-25  0:43     ` Anton Altaparmakov
2001-07-25  1:30       ` Daniel Phillips [this message]
2001-07-25 21:18         ` Steve Lord
2001-07-24 16:48 ` Linus Torvalds
2001-07-24 17:04   ` Rik van Riel
2001-07-24 18:14     ` Daniel Phillips
2001-07-24 18:15       ` Rik van Riel
2001-07-24 19:41         ` Is /dev/epoll scheduled for official inclusion anytime soon? David E. Weekly
2001-07-24 20:05           ` Rik van Riel
2001-07-24 20:26           ` linux partitioning in IA64 hiufungeric.tse
2001-07-25  9:29             ` Mike A. Harris
2001-07-24 20:24   ` [RFC] Optimization for use-once pages Patrick Dreker
2001-07-24 20:32     ` Rik van Riel
2001-07-24 22:16       ` Patrick Dreker
2001-07-24 22:25         ` Rik van Riel
2001-07-25  0:27           ` Linus Torvalds
2001-07-25  8:20             ` Patrick Dreker
2001-07-25 12:57               ` Martin Devera
2001-07-25 14:16               ` Daniel Phillips
2001-07-24 20:33     ` Linus Torvalds
2001-07-25  1:25       ` Daniel Phillips
2001-07-25  0:18     ` Daniel Phillips
2001-07-24 20:27 ` Marcelo Tosatti
2001-07-24 22:05   ` Rik van Riel
2001-07-24 20:53     ` Marcelo Tosatti
2001-07-24 22:27       ` Rik van Riel
2001-07-24 23:09       ` Daniel Phillips
2001-07-24 19:35         ` Rob Landley
2001-07-25  6:10           ` Marcelo Tosatti
2001-07-25  8:32           ` Eric W. Biederman
2001-07-25 12:53             ` Daniel Phillips
2001-07-25 16:05               ` Eric W. Biederman
2001-07-25 12:57           ` Daniel Phillips
2001-07-25  5:12       ` Andrew Morton
2001-07-25  6:33         ` Marcelo Tosatti
2001-07-30 18:09         ` Marcelo Tosatti
2001-07-24 22:41     ` Daniel Phillips
2001-07-24 22:22   ` Daniel Phillips
2001-07-25  2:31 ` Marcelo Tosatti
     [not found] <010301c11463$1ee00440$294b82ce@connecttech.com>
2001-07-24 17:07 ` Rik van Riel
2001-07-24 17:42   ` Stuart MacDonald
2001-07-24 17:51     ` Rik van Riel
2001-07-24 18:09       ` Stuart MacDonald
2001-07-24 18:15       ` Mike Castle
2001-07-24 18:21         ` Rik van Riel
2001-07-24 17:44   ` Mike Castle
2001-07-24 17:52     ` Rik van Riel
     [not found] <0107251802300B.00907@starship>
2001-07-25 16:41 ` Rik van Riel
2001-07-25 17:46   ` Daniel Phillips
2001-07-26  8:36   ` Eric W. Biederman
2001-07-26 12:06     ` Daniel Phillips
2001-07-26 10:38       ` Marcelo Tosatti
2001-07-26 12:17         ` Daniel Phillips
  -- strict thread matches above, loose matches on Subject: below --
2001-07-26  3:27 Ed Tomlinson

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=0107250330340D.00520@starship \
    --to=phillips@bonn-fries.net \
    --cc=aia21@cam.ac.uk \
    --cc=jlnance@intrex.net \
    --cc=linux-kernel@vger.kernel.org \
    --cc=riel@conectiva.com.br \
    /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