Re: rcu, sched: WARNING: CPU: 30 PID: 23771 at kernel/rcu/tree_plugin.h:337 rcu_read_unlock_special+0x369/0x550()

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

On Mon, Jan 26, 2015 at 10:08:04AM +0800, Lai Jiangshan wrote:
> On 01/25/2015 05:18 AM, Paul E. McKenney wrote:
> 
> > 
> > Good point!  In my scenario, CPU 0 would not yet have switched away from
> > Task A.  Hmmm...  Yet Sasha really does see this failure.  Will give it
> > some more thought.
> > 
> > Any ideas?
> 
> I don't known which commit was merged from the rcu-git-tree in Sasha's test
> I try to review it.

If I had to guess, it would be 1d082fd06188 (Remove local_irq_disable()
in rcu_preempt_note_context_switch()), though his finding this might be
more directly related to increases in trinity's levels of stress.

> We can fallback to git-bitsect if the reviews fails.

One (very unlikely) possibility is that Sasha's compiler is ignoring the
barrier() in rcu_preempt_qs().

							Thanx, Paul

> Thanks,
> Lai
> 
> > 
> > 							Thanx, Paul
> > 
> >> Thanks,
> >> Lai
> >>
> >>>     
> >>> 6.	Once in rcu_read_unlock_special(), the fact that
> >>>     	current->rcu_read_unlock_special.b.need_qs is true becomes
> >>>     	apparent, so rcu_read_unlock_special() invokes rcu_preempt_qs().
> >>>     	Recursively, given that we interrupted out of that same
> >>>     	function in the preceding step.
> >>>     
> >>> 7.	Because rcu_preempt_data.passed_quiesce is now true,
> >>>     	rcu_preempt_qs() does nothing, and simply returns.
> >>>     
> >>> 8.	Upon return to rcu_read_unlock_special(), it is noted that
> >>>     	current->rcu_read_unlock_special is still nonzero (because
> >>>     	the interrupted rcu_preempt_qs() had not yet gotten around
> >>>     	to clearing current->rcu_read_unlock_special.b.need_qs).
> >>>     
> >>> 9.	Execution proceeds to the WARN_ON_ONCE(), which notes that
> >>>     	we are in an interrupt handler and thus duly splats.
> >>>
> >>> The solution, as noted above, is to make rcu_read_unlock_special()
> >>> clear out current->rcu_read_unlock_special.b.need_qs after calling
> >>> rcu_preempt_qs().  The interrupted rcu_preempt_qs() will clear it again,
> >>> but this is harmless.  The worst that happens is that we clobber another
> >>> attempt to set this field, but this is not a problem because we just
> >>> got done reporting a quiescent state.
> >>>
> >>> Reported-by: Sasha Levin <sasha.levin@oracle.com>
> >>> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
> >>>
> >>> diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h
> >>> index 8669de884445..ec99dc16aa38 100644
> >>> --- a/kernel/rcu/tree_plugin.h
> >>> +++ b/kernel/rcu/tree_plugin.h
> >>> @@ -322,6 +322,7 @@ void rcu_read_unlock_special(struct task_struct *t)
> >>>  	special = t->rcu_read_unlock_special;
> >>>  	if (special.b.need_qs) {
> >>>  		rcu_preempt_qs();
> >>> +		t->rcu_read_unlock_special.need_qs = false;
> >>>  		if (!t->rcu_read_unlock_special.s) {
> >>>  			local_irq_restore(flags);
> >>>  			return;
> >>>
> >>> .
> >>>
> >>
> > 
> > .
> > 
>