linux-kernel - Re: [PATCH 09/12] remove race between use and initialization of locks

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <20070702213832.8864656c.akpm@linux-foundation.org>
Date:	Mon, 2 Jul 2007 21:38:32 -0700
From:	Andrew Morton <akpm@...ux-foundation.org>
To:	"Ed L. Cashin" <ecashin@...aid.com>
Cc:	linux-kernel@...r.kernel.org, Greg K-H <greg@...ah.com>
Subject: Re: [PATCH 09/12] remove race between use and initialization of
 locks

On Tue, 26 Jun 2007 14:50:12 -0400 "Ed L. Cashin" <ecashin@...aid.com> wrote:

> This change was originally submitted by Alexey Dobriyan in an email
> with ...
> 
>   Message-ID: <20070325190221.GA5308@...tell.zuzino.mipt.ru>
> 
> and the comment,
> 
>   Some drivers do register_chrdev() before lock or semaphore used in
>   corresponding file_operations is initialized.
> 
> Andrew Morton commented that these locks should be initialized at
> compile time, but Alexey Debriyan pointed out that the Documentation
> tells us to use dynamic initialization whenever possible, and then the
> discussion petered out.
> 
>   http://preview.tinyurl.com/2pxq6p
> 
> I believe we made these locks dynamic because of the notice in
> Documentation/spinlocks.txt, which says that static initializers are
> deprecated:
> 
>   UPDATE March 21 2005 Amit Gud <gud@....net>
> 
>   Macros SPIN_LOCK_UNLOCKED and RW_LOCK_UNLOCKED are deprecated and will be
>   removed soon. So for any new code dynamic initialization should be used:

The document is inaccurate.

Yes, the use of SPIN_LOCK_UNLOCKED should be avoided because it breaks
lockdep.  But it can be replaced with DEFINE_SPINLOCK: there is no need to
switch to runtime initialisation.



> ...
> 
> In any case, the patch below makes the code correct and in keeping
> with the existing documentation.  If the existing docs are wrong, I'd
> be happy to follow up with a patch that corrects them and makes these
> aoechr.c locks static.
> 
> Signed-off-by: Alexey Dobriyan <adobriyan@...il.com>
> Signed-off-by: Ed L. Cashin <ecashin@...aid.com>
> ---
>  drivers/block/aoe/aoechr.c |    4 ++--
>  1 files changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/block/aoe/aoechr.c b/drivers/block/aoe/aoechr.c
> index 10b38a7..2b4f873 100644
> --- a/drivers/block/aoe/aoechr.c
> +++ b/drivers/block/aoe/aoechr.c
> @@ -256,13 +256,13 @@ aoechr_init(void)
>  {
>  	int n, i;
>  
> +	sema_init(&emsgs_sema, 0);
> +	spin_lock_init(&emsgs_lock);
>  	n = register_chrdev(AOE_MAJOR, "aoechr", &aoe_fops);
>  	if (n < 0) { 
>  		printk(KERN_ERR "aoe: can't register char device\n");
>  		return n;
>  	}
> -	sema_init(&emsgs_sema, 0);
> -	spin_lock_init(&emsgs_lock);
>  	aoe_class = class_create(THIS_MODULE, "aoe");
>  	if (IS_ERR(aoe_class)) {
>  		unregister_chrdev(AOE_MAJOR, "aoechr");
> -- 
> 1.5.2.1
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@...r.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/