From mboxrd@z Thu Jan  1 00:00:00 1970
From: Bob Peterson <rpeterso@redhat.com>
Date: Fri, 17 Jul 2015 09:49:41 -0400 (EDT)
Subject: [Cluster-devel] GFS2: Use resizable hash table for glocks
In-Reply-To: <20150715111349.GB16569@mwanda>
References: <20150715111349.GB16569@mwanda>
Message-ID: <1441826102.40023485.1437140981704.JavaMail.zimbra@redhat.com>
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

----- Original Message -----
> Hello Bob Peterson,
> 
> The patch 6afe4c76af37: "GFS2: Use resizable hash table for glocks"
> from Mar 16, 2015, leads to the following static checker warning:
> 
> 	fs/gfs2/glock.c:1798 gfs2_glock_iter_next()
> 	error: 'gi->gl' dereferencing possible ERR_PTR()
> 
> fs/gfs2/glock.c
>   1793  static void gfs2_glock_iter_next(struct gfs2_glock_rht_iter *gi)
>   1794  {
>   1795          do {
>   1796                  gi->gl = rhashtable_walk_next(&gi->hti);
>                                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> can return -EAGAIN sometimes.
> 
>   1797          /* Skip entries for other sb and dead entries */
>   1798          } while ((gi->gl) && ((gi->sdp != gi->gl->gl_name.ln_sbd) ||
>                                                   ^^^^^^^^
> Might be an issue here, might not.  I don't know the subsystem well
> enough to know when it returns NULL vs -EAGAIN.
> 
>   1799
>   __lockref_is_dead(&gi->gl->gl_lockref)));
>   1800  }
> 
> regards,
> dan carpenter
> 

Thanks, Dan. I'll craft a fix.

Bob Peterson
Red Hat File Systems