Re: perf related lockdep bug

[Paul E. McKenney]

On Wed, Nov 04, 2015 at 04:55:28PM -0800, Paul E. McKenney wrote:
> On Wed, Nov 04, 2015 at 09:58:36PM +0100, Andi Kleen wrote:
> > 
> > I tested my perf stress workload with the patch applied on 4.3,
> > unfortunately got a hang again :-/
> 
> Any diagnostics, sysrq-T output, or whatever?

Given that it looks like your hang is happening at runtime, I would guess
that the following patch won't help, but who knows?

							Thanx, Paul

------------------------------------------------------------------------

commit 05faf451f1239a28fcd63bf4b66c0db57d7b13f9
Author: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
Date:   Wed Nov 4 08:22:05 2015 -0800

    rcu: Move wakeup out from under rnp->lock
    
    This patch removes a potential deadlock hazard by moving the
    wake_up_process() in rcu_spawn_gp_kthread() out from under rnp->lock.
    
    Signed-off-by: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
    Signed-off-by: Paul E. McKenney <paulmck@xxxxxxxxxxxxxxxxxx>

diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
index caf3651fa5c9..183445959d00 100644
--- a/kernel/rcu/tree.c
+++ b/kernel/rcu/tree.c
@@ -4323,8 +4323,8 @@ static int __init rcu_spawn_gp_kthread(void)
 			sp.sched_priority = kthread_prio;
 			sched_setscheduler_nocheck(t, SCHED_FIFO, &sp);
 		}
-		wake_up_process(t);
 		raw_spin_unlock_irqrestore(&rnp->lock, flags);
+		wake_up_process(t);
 	}
 	rcu_spawn_nocb_kthreads();
 	rcu_spawn_boost_kthreads();

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/