Re: [PATCH 1/2] sched/fair: optimization of update_blocked_averages()

[Peter Zijlstra]

On Fri, Feb 08, 2019 at 05:47:53PM +0100, Vincent Guittot wrote:
> On Fri, 8 Feb 2019 at 17:30, Peter Zijlstra <peterz@xxxxxxxxxxxxx> wrote:
> > On Fri, Feb 08, 2019 at 04:44:53PM +0100, Vincent Guittot wrote:
> > > On Fri, 8 Feb 2019 at 16:40, Peter Zijlstra <peterz@xxxxxxxxxxxxx> wrote:

> > > Good point but this should go after the for_each_sched_entity() loop
> >
> > Indeed, but that loop does enqueue and can throttle again, should that
> > not also get that additional list_add_leaf_cfs_rq() loop we added to
> > enqueue_task_fair() to finish the add?
> 
> Initially, I added this additional loop but finally removed it because
> I didn't hit it during my tests. IIRC, we are protected by
> throttle_count in such case, which is not the case when we enqueue a
> task

Fair enough; and the to-be added assert will notify us if we got that
wrong :-)

I'll add the assert, no need to re-send.

Thanks!