[Cluster-devel] GFS2: Fix panic in glock memory shrinker

[Steven Whitehouse <swhiteho@redhat.com>]

Hi,

Now in the -nmw git tree. Thanks,

Steve.

On Tue, 2009-06-30 at 13:51 -0500, Benjamin Marzinski wrote:
> To: cluster-devel at redhat.com
> Subject: GFS2: Fix panic in glock memory shrinker
> 
> It is possible for gfs2_shrink_glock_memory() to check a glock for
> demotion
> that's in the process of being freed by gfs2_glock_put().  In this case,
> gfs2_shrink_glock_memory() will acquire a new reference to this glock,
> and
> then try to free the glock itself when it drops the refernce.  To solve
> this, gfs2_shrink_glock_memory() just needs to check if the glock is in
> the process of being freed, and if so skip it without ever unlocking the
> lru_lock.
> 
> Signed-off-by: Benjamin Marzinski <bmarzins@redhat.com>
> ---
>  fs/gfs2/glock.c |    4 ++++
>  1 file changed, 4 insertions(+)
> 
> Index: kernel-upstream/fs/gfs2/glock.c
> ===================================================================
> --- kernel-upstream.orig/fs/gfs2/glock.c
> +++ kernel-upstream/fs/gfs2/glock.c
> @@ -1314,6 +1314,10 @@ static int gfs2_shrink_glock_memory(int 
>  		list_del_init(&gl->gl_lru);
>  		atomic_dec(&lru_count);
>  
> +		/* Check if glock is about to be freed */
> +		if (atomic_read(&gl->gl_ref) == 0)
> +			continue;
> +
>  		/* Test for being demotable */
>  		if (!test_and_set_bit(GLF_LOCK, &gl->gl_flags)) {
>  			gfs2_glock_hold(gl);
>