Re: [PATCH 2/2] sched: Fix next_balance logic in rebalance_domains() and idle_balance()

[Jason Low <jason.low2@hp.com>]

On Thu, 2014-05-08 at 19:38 +0200, Ingo Molnar wrote:
> * Jason Low <jason.low2@hp.com> wrote:
> 
> > On Mon, 2014-04-28 at 15:45 -0700, Jason Low wrote:
> > > Currently, in idle_balance(), we update rq->next_balance when we pull_tasks. 
> > > However, it is also important to update this in the !pulled_tasks case too.
> > > 
> > > When the CPU is "busy" (the CPU isn't idle), rq->next_balance gets computed
> > > using sd->busy_factor (so we increase the balance interval when the CPU is
> > > busy). However, when the CPU goes idle, rq->next_balance could still be set
> > > to a large value that was computed with the sd->busy_factor.
> > > 
> > > Thus, we need to also update rq->next_balance in idle_balance() in the cases
> > > where !pulled_tasks too, so that rq->next_balance gets updated without taking
> > > the busy_factor into account when the CPU is about to go idle.
> > > 
> > > This patch makes rq->next_balance get updated independently of whether or
> > > not we pulled_task. Also, we add logic to ensure that we always traverse
> > > at least 1 of the sched domains to get a proper next_balance value for
> > > updating rq->next_balance.
> > > 
> > > Additionally, since load_balance() modifies the sd->balance_interval, we
> > > need to re-obtain the sched domain's interval after the call to
> > > load_balance() in rebalance_domains() before we update rq->next_balance.
> > > 
> > > This patch adds and uses 2 new helper functions, update_next_balance() and
> > > get_sd_balance_interval() to update next_balance and obtain the sched
> > > domain's balance_interval. 
> > 
> > 
> > Hi Peter,
> > 
> > I noticed that patch 1 is in tip, but not this patch 2. I was wondering
> > what the current status with this [PATCH 2/2] is at the moment.
> 
> It was crashing the bootup with the attached config, it gave the splat 
> attached below. (ignore the line duplication, it's a serial logging 
> artifact.)

Hi Ingo, Peter,

Were there NULL domains on the test system? If so, I think we can
address the problem by doing update_next_balance() only if the below
rcu_dereference_check_sched_domain() returns a non-null domain.

@@ -6665,8 +6692,14 @@ static int idle_balance(struct rq *this_rq)
>  	 */
>  	this_rq->idle_stamp = rq_clock(this_rq);
>  
> -	if (this_rq->avg_idle < sysctl_sched_migration_cost)
> +	if (this_rq->avg_idle < sysctl_sched_migration_cost) {
> +		rcu_read_lock();
> +		sd = rcu_dereference_check_sched_domain(this_rq->sd);
> +		update_next_balance(sd, 0, &next_balance);
> +		rcu_read_unlock();
> +
>  		goto out;
> +	}