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
next prev parent 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 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.