Re: [PATCH] rcu: Avoid kthread_stop on invalid pointer if rcutorture reader startup fails

["Paul E. McKenney" <paulmck@us.ibm.com>]

On Wed, Aug 16, 2006 at 03:24:02PM -0700, Josh Triplett wrote:
> rcu_torture_init kmallocs the array of reader threads, then creates each one
> with kthread_run, cleaning up with rcu_torture_cleanup if this fails.
> rcu_torture_cleanup calls kthread_stop on any non-NULL pointer in the array;
> however, any readers after the one that failed to start up will have invalid
> pointers, not null pointers.  Avoid this by using kzalloc instead.

Good catch!!!

Acked-by: Paul E. McKenney <paulmck@us.ibm.com>
> Signed-off-by: Josh Triplett <josh@freedesktop.org>
> ---
>  kernel/rcutorture.c |    2 +-
>  1 files changed, 1 insertions(+), 1 deletions(-)
> 
> diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c
> index aff0064..8b09c95 100644
> --- a/kernel/rcutorture.c
> +++ b/kernel/rcutorture.c
> @@ -779,7 +779,7 @@ rcu_torture_init(void)
>  		writer_task = NULL;
>  		goto unwind;
>  	}
> -	reader_tasks = kmalloc(nrealreaders * sizeof(reader_tasks[0]),
> +	reader_tasks = kzalloc(nrealreaders * sizeof(reader_tasks[0]),
>  			       GFP_KERNEL);
>  	if (reader_tasks == NULL) {
>  		VERBOSE_PRINTK_ERRSTRING("out of memory");
> -- 
> 1.4.1.1
> 
>