linux-ext4.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Phillip Susi <psusi@cfl.rr.com>
To: Ted Ts'o <tytso@mit.edu>
Cc: Eric Sandeen <sandeen@redhat.com>,
	"linux-ext4@vger.kernel.org" <linux-ext4@vger.kernel.org>
Subject: Re: Large directories and poor order correlation
Date: Mon, 14 Mar 2011 19:43:57 -0400	[thread overview]
Message-ID: <4D7EA83D.20400@cfl.rr.com> (raw)
In-Reply-To: <20110314215249.GE8120@thunk.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 03/14/2011 05:52 PM, Ted Ts'o wrote:
> Unfortunately the kernel can't do it, because a directory could be
> arbitrarily big, and kernel memory is non-swappable.  In addition,

Buffers/cache is discardable though.  Or does the entire htree have to
be kept in slab or something?

> what if a process opens a directory, starts calling readdir, pauses in
> the middle, and then holds onto it for days, weeks, or months?

The same thing that happened before htree?

> It's not hard to provide library routines that do the right thing, and
> I have written an LD_PRELOAD which intercepts opendir() and readdir()
> calls and does the sorting in userspace.  Perhaps the right answer is
> getting this into libc, but I have exactly two words for you: "Uhlrich
> Drepper".

Wouldn't it be better to just have readdir() use the main directory,
which presumably is in a more sane ordering, instead of the htree?  That
way you don't have to burn cpu and ram sorting on every opendir().

Also, I have checked some smaller directories and lsattr reports they
are NOT using indexing, yet still display poor correlation.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAk1+qDkACgkQJ4UciIs+XuIktwCgi1u4T2x+igOw4feO0hNjzB9W
liIAmwRBdPiZMSfWpzu4+40xJsNXzouQ
=d4VX
-----END PGP SIGNATURE-----

  reply	other threads:[~2011-03-14 23:43 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-03-14 20:24 Large directories and poor order correlation Phillip Susi
2011-03-14 20:37 ` Eric Sandeen
2011-03-14 20:52   ` Phillip Susi
2011-03-14 21:12     ` Eric Sandeen
2011-03-14 21:52     ` Ted Ts'o
2011-03-14 23:43       ` Phillip Susi [this message]
2011-03-15  0:14         ` Ted Ts'o
2011-03-15 14:01           ` Phillip Susi
2011-03-15 14:33             ` Rogier Wolff
2011-03-15 14:36               ` Ric Wheeler
2011-03-15 17:08             ` Ted Ts'o
2011-03-15 19:08               ` Phillip Susi
2011-03-16  1:50                 ` Ted Ts'o
2011-03-15  7:59   ` Florian Weimer
2011-03-15 11:06     ` Theodore Tso
2011-03-15 11:23       ` Ric Wheeler
2011-03-15 11:38         ` Theodore Tso
2011-03-15 13:33       ` Rogier Wolff
2011-03-15 17:18         ` Ted Ts'o

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=4D7EA83D.20400@cfl.rr.com \
    --to=psusi@cfl.rr.com \
    --cc=linux-ext4@vger.kernel.org \
    --cc=sandeen@redhat.com \
    --cc=tytso@mit.edu \
    /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;
as well as URLs for NNTP newsgroup(s).