netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: ebiederm@xmission.com (Eric W. Biederman)
To: David Woodhouse <dwmw2@infradead.org>
Cc: netdev@vger.kernel.org, johannes@sipsolutions.net
Subject: Re: tun netns BUG()
Date: Fri, 05 Jun 2009 16:24:29 -0700	[thread overview]
Message-ID: <m1ljo61cya.fsf@fess.ebiederm.org> (raw)
In-Reply-To: <1244193463.3751.129.camel@macbook.infradead.org> (David Woodhouse's message of "Fri\, 05 Jun 2009 10\:17\:43 +0100")


The patch below fixes the one bug I know of in with just my changes
applied.  I don't think it fixes everything with Herberts changes.

I expect some of this will need to be moved to release to fix the
select problem on 2.6.30-rc8.

Eric


From: Eric W. Biederman <ebiederm@aristanetworks.com>
Subject: [PATCH] tun: Fix unregister race

It is possible for tun_chr_close to race with dellink on the
a tun device.  In which case if __tun_get runs before dellink
but dellink runs before tun_chr_close calls unregister_netdevice
we will attempt to unregister the netdevice after it is already
gone.  

The two cases are already serialized on the rtnl_lock, so I have
gone for the cheap simple fix of moving rtnl_lock to cover __tun_get
in tun_chr_close.  Eliminating the possibility of the tun device
being unregistered between __tun_get and unregister_netdevice in
tun_chr_close.

Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com>
---

--- linux-2.6.28.x86_64-old/drivers/net/tun.c	2009-05-06 15:01:56.000000000 -0700
+++ linux-2.6.28.x86_64/drivers/net/tun.c	2009-05-06 15:10:09.000000000 -0700
@@ -1194,21 +1194,22 @@
 static int tun_chr_close(struct inode *inode, struct file *file)
 {
 	struct tun_file *tfile = file->private_data;
-	struct tun_struct *tun = __tun_get(tfile);
+	struct tun_struct *tun;
 
 
+	rtnl_lock();
+	tun = __tun_get(tfile);
 	if (tun) {
 		DBG(KERN_INFO "%s: tun_chr_close\n", tun->dev->name);
 
-		rtnl_lock();
 		__tun_detach(tun);
 
 		/* If desireable, unregister the netdevice. */
 		if (!(tun->flags & TUN_PERSIST))
 			unregister_netdevice(tun->dev);
 
-		rtnl_unlock();
 	}
+	rtnl_unlock();
 
 	put_net(tfile->net);
 	kfree(tfile);


  reply	other threads:[~2009-06-05 23:24 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-06-04 19:05 tun netns BUG() David Woodhouse
2009-06-05  1:19 ` Eric W. Biederman
2009-06-05  8:42   ` David Woodhouse
2009-06-05  9:17     ` David Woodhouse
2009-06-05 23:24       ` Eric W. Biederman [this message]
2009-06-06  7:06         ` David Woodhouse
2009-06-08  7:44           ` David Miller
2009-07-03  8:35         ` Herbert Xu
2009-07-03  9:03           ` Herbert Xu
2009-07-03 14:52             ` Eric W. Biederman
2009-07-03 15:25               ` Herbert Xu
2009-07-06  2:01                 ` David Miller
2009-06-05 23:20     ` Eric W. Biederman

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=m1ljo61cya.fsf@fess.ebiederm.org \
    --to=ebiederm@xmission.com \
    --cc=dwmw2@infradead.org \
    --cc=johannes@sipsolutions.net \
    --cc=netdev@vger.kernel.org \
    /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).