Re: NULL pointer dereference in pick_next_task_fair

[Quentin Perret <qperret@google.com>]

On Friday 08 Nov 2019 at 13:00:35 (+0100), Peter Zijlstra wrote:
> On Fri, Nov 08, 2019 at 11:02:12AM +0000, Quentin Perret wrote:
> > On Thursday 07 Nov 2019 at 20:29:07 (+0100), Peter Zijlstra wrote:
> > > I still havne't had food, but this here compiles...
> > 
> > And it seems to work, too :)
> 
> Excellent!
> 
> > > @@ -3929,13 +3929,17 @@ pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
> > >  	}
> > > 
> > >  restart:
> > > -	/*
> > > -	 * Ensure that we put DL/RT tasks before the pick loop, such that they
> > > -	 * can PULL higher prio tasks when we lower the RQ 'priority'.
> > > -	 */
> > > -	prev->sched_class->put_prev_task(rq, prev, rf);
> > > -	if (!rq->nr_running)
> > > -		newidle_balance(rq, rf);
> > > +#ifdef CONFIG_SMP
> > > +	for (class = prev->sched_class;
> > > +	     class != &idle_sched_class;
> > > +	     class = class->next) {
> > > +
> > > +		if (class->balance(rq, prev, rf))
> > > +			break;
> > > +	}
> > > +#endif
> > > +
> > > +	put_prev_task(rq, prev);
> > 
> > Right, that looks much cleaner IMO. I'm thinking if we killed the
> > special case for CFS above we could do with a single loop to iterate the
> > classes, and you could fold ->balance() in ->pick_next_task() ...
> 
> No, you can't, because then you're back to having to restart the pick
> when something happens when we drop the rq halfway down the pick.  Which
> is something I really wanted to get rid of.

Right, with a single loop you'll have to re-iterate the classes from
the start in case of RETRY_TASK, but you're re-iterating all the classes
too with this patch. You're doing a little less work in the second loop
though, so maybe it's worth it. And I was the one worried about
refactoring the code too much close to the release, so maybe that's for
another time ;)

Thanks,
Quentin