From mboxrd@z Thu Jan  1 00:00:00 1970
From: Steven Whitehouse <swhiteho@redhat.com>
Date: Wed, 04 Sep 2013 16:27:57 +0100
Subject: [Cluster-devel] [GFS2 PATCH] GFS2: Remove unnecessary memory
 barrier
In-Reply-To: <1554021184.19774670.1374760429690.JavaMail.root@redhat.com>
References: <1554021184.19774670.1374760429690.JavaMail.root@redhat.com>
Message-ID: <1378308477.2740.32.camel@menhir>
List-Id: <cluster-devel.redhat.com>
To: cluster-devel.redhat.com
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit

Hi,


On Thu, 2013-07-25 at 09:53 -0400, Bob Peterson wrote:
> Hi,
> 
> Function test_and_clear_bit implies a memory barrier, so subsequent
> memory barriers are unnecessary.
> 
> Regards,
> 
> Bob Peterson
> Red Hat File Systems
> 
Now in the -nmw tree. Thanks,

Steve.

> Signed-off-by: Bob Peterson <rpeterso@redhat.com> 
> ---
> diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c
> index 9435384..a45b1e0 100644
> --- a/fs/gfs2/glock.c
> +++ b/fs/gfs2/glock.c
> @@ -1411,7 +1411,6 @@ __acquires(&lru_lock)
>  		if (demote_ok(gl))
>  			handle_callback(gl, LM_ST_UNLOCKED, 0, false);
>  		WARN_ON(!test_and_clear_bit(GLF_LOCK, &gl->gl_flags));
> -		smp_mb__after_clear_bit();
>  		if (queue_delayed_work(glock_workqueue, &gl->gl_work, 0) == 0)
>  			gfs2_glock_put_nolock(gl);
>  		spin_unlock(&gl->gl_spin);
>