All of lore.kernel.org
 help / color / mirror / Atom feed
From: Mark Lord <lkml@rtr.ca>
To: Greg KH <gregkh@suse.de>
Cc: Linux Kernel <linux-kernel@vger.kernel.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	linux-usb-devel@lists.sourceforge.net
Subject: [PATCH] base/class.c: prevent ooops due to insert/remove race (v3)
Date: Thu, 29 Nov 2007 13:09:53 -0500	[thread overview]
Message-ID: <474F0071.5010507@rtr.ca> (raw)
In-Reply-To: <474EFBFF.9090400@rtr.ca>

Mark Lord wrote:
> Mark Lord wrote:
>> Mark Lord wrote:
>> ...
>> And here is a "prevented" oops, courtesy of the patch (2.6.23.8).
>> These are easy to reproduce (just jiggle the connection on an
>> attached USB multi-card reader with a CF card inserted):
> ...
>> [  347.099562] usb 5-6: USB disconnect, address 10
>> [  347.101077] BUG: unable to handle kernel NULL pointer dereference 
>> at virtual address 00000000
> 
> ...
> 
> 
> Mmmm.. that's odd.  It *did* Oops there.  I wonder how/why ?
...

Ahh.. it was the *other* call to strlen() that was crapping out.
Here's an updated patch to prevent the Oops:

* * * *

While doing insert/remove (quickly) tests on USB,
I managed to trigger an Oops on 2.6.23.8 on a call
to strlen() in make_class_name().

USB maintainers can try this themselves, by plugging in an
external USB XX-in-1 flash reader, with a CF card inserted.
Then just jiggle the connection so that the device connects
and disconnects rapidly.  This may not sound realistic,
but there's definitely a race in there, and this is a quick
way to reproduce it if need be.

The patch below prevents the oops, but still keeps the bug visible.

Signed-off-by:  Mark Lord <mlord@pobox.com> 
---

Patch applies to both 2.6.24 and 2.6.23.

--- old/drivers/base/class.c	2007-11-29 10:51:43.000000000 -0500
+++ linux/drivers/base/class.c	2007-11-29 13:00:15.000000000 -0500
@@ -352,9 +352,22 @@
 char *make_class_name(const char *name, struct kobject *kobj)
 {
 	char *class_name;
+	const char *kname;
 	int size;
 
-	size = strlen(name) + strlen(kobject_name(kobj)) + 2;
+	/* Rapidly inserting/removing a USB device (others?)
+	 * can trigger an Oops on the strlen() call.
+	 * Cause unknown yet, so prevent the Oops
+	 * but don't mask the issue.
+	 */
+	kname = kobject_name(kobj);
+	if (!kname || !name) {
+		printk(KERN_ERR "make_class_name: name=%p kname=%p\n",
+						name, kname);
+		BUG_ON(1);
+		return NULL;
+	}
+	size = strlen(name) + strlen(kname) + 2;
 
 	class_name = kmalloc(size, GFP_KERNEL);
 	if (!class_name)

  reply	other threads:[~2007-11-29 18:10 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-11-29  4:00 [PATCH] base/class.c: prevent ooops due to insert/remove race Mark Lord
2007-11-29  4:33 ` Greg KH
2007-11-29 15:41   ` Mark Lord
2007-11-29 16:27     ` Greg KH
2007-11-29 17:45       ` Mark Lord
2007-11-29 17:48         ` Mark Lord
2007-11-29 17:50           ` Mark Lord
2007-11-29 18:09             ` Mark Lord [this message]
2007-11-29 18:12               ` [PATCH] base/class.c: prevent ooops due to insert/remove race (v3) Mark Lord
2007-11-29 19:20                 ` [linux-usb-devel] " Alan Stern
2007-11-29 20:01                   ` Mark Lord
2007-11-29 20:12                     ` Greg KH
     [not found]                       ` <474F1DB3.4030900@rtr.ca>
2007-11-29 20:25                         ` Mark Lord
2007-11-29 20:32                           ` Greg KH
2007-11-29 20:46                             ` Mark Lord
2007-11-29 21:17                               ` Greg KH
2007-11-29 22:07                               ` Alan Stern
2007-11-29 22:11                                 ` Greg KH
2007-11-29 22:28                                 ` Mark Lord
2007-11-29 22:43                                   ` Alan Stern
2007-11-29 23:09                                 ` Linus Torvalds
2007-11-30  2:58                                   ` Alan Stern
2007-11-29 20:13                     ` Alan Stern
2007-11-29 20:17                       ` Mark Lord
2007-11-29 20:20                       ` Mark Lord
2007-11-29 21:59                         ` Alan Stern
2007-11-29 15:59   ` [PATCH] base/class.c: prevent ooops due to insert/remove race (v2) Mark Lord

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=474F0071.5010507@rtr.ca \
    --to=lkml@rtr.ca \
    --cc=akpm@linux-foundation.org \
    --cc=gregkh@suse.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-usb-devel@lists.sourceforge.net \
    /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.