mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Roman Zippel <zippel@linux-m68k.org>
To: Alexander Viro <viro@math.psu.edu>
Cc: Daniel Phillips <phillips@arcor.de>,
	Rusty Russell <rusty@rustcorp.com.au>,
	"David S. Miller" <davem@redhat.com>, <adam@yggdrasil.com>,
	<R.E.Wolff@bitwizard.nl>, <linux-kernel@vger.kernel.org>
Subject: Re: Rusty's module talk at the Kernel Summit
Date: Fri, 12 Jul 2002 13:30:17 +0200 (CEST)	[thread overview]
Message-ID: <Pine.LNX.4.44.0207121257470.28515-100000@serv> (raw)
In-Reply-To: <Pine.GSO.4.21.0207111928390.9488-100000@weyl.math.psu.edu>

Hi,

On Thu, 11 Jul 2002, Alexander Viro wrote:

> For filesystems the only currently existing race is in the case when
> init_module() registers one, then decides to bail out and unregisters it.
> If somebody finds the thing between register/unregister the current code
> is screwed.  And no, "don't block in between" is not viable - typically
> the reason of failure is failing allocation and/or timeouts on some sort
> of probing.

In the load path there are more races. Read access to the module list is
only protected by the BKL. This makes it possible to create multiple
modules with the same name in sys_create_module (module_map() can sleep).

> As for determining the loading/normal/unloading - we _already_ have that
> state, no need to introduce new fields.  How do you think try_inc_mod_count()
> manages to work?  Exactly - there's a field of struct module that contains
> a bunch of flags.  And no, Daniel's ramblings (from what I've seen quoted)
> are pure BS - there's no need to mess with "oh, but I refuse to be
> unregistered"; proper refcounting is easy for normal cases.

normal cases or simple cases? The filesystem interface is a very simple
one, now e.g. add a proc interface to it. The current module interface
requires that any first reference to a module must be protected against
module load/unload and unregister, what requires two locks. It could be
reduced to a single lock, if the module was allowed to let the unload
fail, because there are still users. The module actually knows best,
whether there are still references to the module, but it can't tell that
the kernel.

bye, Roman


  parent reply	other threads:[~2002-07-12 11:27 UTC|newest]

Thread overview: 48+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2002-07-04 17:24 Adam J. Richter
2002-07-11  2:48 ` Rusty Russell
2002-07-11  2:45   ` David S. Miller
2002-07-11  3:30     ` Alexander Viro
2002-07-11  5:13       ` Rusty Russell
2002-07-11  6:37         ` Alexander Viro
2002-07-11  7:14           ` Rusty Russell
2002-07-11 10:54             ` Daniel Phillips
2002-07-11 17:37               ` Roman Zippel
2002-07-11 18:01                 ` Thunder from the hill
2002-07-11 18:50                   ` Daniel Phillips
2002-07-17 18:16                   ` bill davidsen
2002-07-17 19:35                     ` Thunder from the hill
2002-07-11 18:28                 ` Daniel Phillips
2002-07-11 19:48                   ` Roman Zippel
2002-07-11 20:29                     ` Daniel Phillips
2002-07-11 23:37                     ` Alexander Viro
2002-07-12  1:54                       ` Daniel Phillips
2002-07-12  3:53                       ` Rusty Russell
2002-07-12  6:49                         ` Kai Henningsen
2002-07-12 11:30                       ` Roman Zippel [this message]
2002-07-12  0:00               ` Rusty Russell
2002-07-12  6:57                 ` Kai Henningsen
2002-07-19  0:19           ` Richard Gooch
2002-07-22 16:29             ` Alexander Viro
2002-07-23  4:37               ` Richard Gooch
2002-07-11  4:02     ` Cort Dougan
2002-07-11  4:19       ` Arnaldo Carvalho de Melo
2002-07-11  4:46       ` Cort Dougan
2002-07-11  2:55   ` Arnaldo Carvalho de Melo
2002-07-11  3:01     ` Arnaldo Carvalho de Melo
2002-07-11  5:16     ` Rusty Russell
  -- strict thread matches above, loose matches on Subject: below --
2002-07-11  5:44 Adam J. Richter
2002-07-11  5:07 Adam J. Richter
2002-07-03 15:53 Adam J. Richter
2002-07-03 17:07 ` Hugh Dickins
2002-07-03 18:46   ` Oliver Neukum
2002-07-03 23:25     ` Keith Owens
2002-07-03 23:09 ` Keith Owens
2002-07-03  7:31 Adam J. Richter
2002-07-03  8:54 ` Rogier Wolff
2002-07-03 12:27 ` Keith Owens
2002-07-03 14:10   ` Keith Owens
2002-07-01 17:20 Adam J. Richter
2002-07-01 16:12 Adam J. Richter
2002-07-01 17:02 ` jlnance
2002-07-03  5:01 ` Keith Owens
2002-07-01  8:45 Keith Owens

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=Pine.LNX.4.44.0207121257470.28515-100000@serv \
    --to=zippel@linux-m68k.org \
    --cc=R.E.Wolff@bitwizard.nl \
    --cc=adam@yggdrasil.com \
    --cc=davem@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=phillips@arcor.de \
    --cc=rusty@rustcorp.com.au \
    --cc=viro@math.psu.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

all inboxes | Powered by JetHome®