Re: [PATCH 2/2] sched: Check if lowest_mask is initialized in find_lowest_rq

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

On Tue, Jun 14, 2011 at 06:36:25PM -0400, Steven Rostedt wrote:
> From: Steven Rostedt <rostedt@goodmis.org>
> 
> On system boot up, the lowest_mask is initialized with early_initcall.
> But RT tasks may wake up on other early_initcall callers before the
> lowest_mask is initialized, causing a system crash.
> 
> The commit d72bce0e67 rcu: Cure load woes
> was the first commit to wake up RT tasks in early init. Before this
> commit this bug should not happen.

I never was able to reproduce this, but after adding Shaohua's patch, I
need this one too on some systems.  So:

Tested-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

> Cc: Peter Zijlstra <peterz@infradead.org>
> Cc: Paul McKenney <paulmck@linux.vnet.ibm.com>
> Reported-by: Andrew Theurer <habanero@linux.vnet.ibm.com>
> Tested-by: Andrew Theurer <habanero@linux.vnet.ibm.com>
> Link: http://lkml.kernel.org/r/1308014683.9218.107.camel@gandalf.stny.rr.com
> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
> ---
>  kernel/sched_rt.c |    4 ++++
>  1 files changed, 4 insertions(+), 0 deletions(-)
> 
> diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
> index 9b8d5dc..10d0182 100644
> --- a/kernel/sched_rt.c
> +++ b/kernel/sched_rt.c
> @@ -1239,6 +1239,10 @@ static int find_lowest_rq(struct task_struct *task)
>  	int this_cpu = smp_processor_id();
>  	int cpu      = task_cpu(task);
> 
> +	/* Make sure the mask is initialized first */
> +	if (unlikely(!lowest_mask))
> +		return -1;
> +
>  	if (task->rt.nr_cpus_allowed == 1)
>  		return -1; /* No other targets possible */
> 
> -- 
> 1.7.4.4
> 
>