[Cluster-devel] Re: [PATCH] [GFS2] Fix format string bug

Steven Whitehouse swhiteho at redhat.com
Fri Jun 13 11:37:39 UTC 2008


Hi,

On Thu, 2008-06-12 at 06:50 +0200, maximilian attems wrote:
> similar to 326f6a5c9c9e1a62aec37bdc0c3f8d53adabe77b,
> only writable by root but worth fixing.
> 
> Signed-off-by: maximilian attems <max at stro.at>
> ---
>  fs/gfs2/locking.c |    2 +-
>  1 files changed, 1 insertions(+), 1 deletions(-)
> 
> diff --git a/fs/gfs2/locking.c b/fs/gfs2/locking.c
> index 663fee7..4857024 100644
> --- a/fs/gfs2/locking.c
> +++ b/fs/gfs2/locking.c
> @@ -131,7 +131,7 @@ retry:
>  		if (!try && capable(CAP_SYS_MODULE)) {
>  			try = 1;
>  			mutex_unlock(&lmh_lock);
> -			request_module(proto_name);
> +			request_module("%s", proto_name);
>  			goto retry;
>  		}
>  		printk(KERN_INFO "GFS2: can't find protocol %s\n", proto_name);

I'm not quite sure whats wrong with the original here, but I'm currently
working upon a patch to remove the lock_dlm module (the lock_nolock
module is already gone) by merging it into the core of GFS2. So that
this module loading code will be disappearing very shortly.

So on that basis, I don't feel that its worth applying this patch at the
moment anyway,

Steve.





More information about the Cluster-devel mailing list