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 Tue, Jan 27, 2015 at 05:08:21PM -0500, Sasha Levin wrote:
> On 01/27/2015 05:03 PM, Paul E. McKenney wrote:
> > 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.
> 
> Quick update from my end: I've stopped seeing this warning, but I've also stopped
> seeing warnings for the other RCU issue I've reported (https://lkml.org/lkml/2015/1/22/676)
> so I'm slightly unhappy about that.

Another approach would be to remove that patch and then revert 1d082fd06188.

Either way, may I have your Tested-by?

> >> > 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().
> 
> I'm actually running the latest gcc (trunk) as well, so it's very possible that it was
> doing something stupid.

Hmmmm...  Could you please send along the assembly output for rcu_preempt_qs()?

							Thanx, Paul