Messages in this thread | | | Date | Fri, 20 Feb 2015 17:26:41 +0100 | From | Peter Zijlstra <> | Subject | Re: [PATCH RESEND v9 10/10] sched: move cfs task on a CPU with higher capacity |
| |
On Fri, Feb 20, 2015 at 02:54:09PM +0100, Vincent Guittot wrote:
> >> if (rq->nr_running >= 2) > >> + return true; > > > > So this,
> >> + sd = rcu_dereference(rq->sd); > >> + if (sd) { > >> + if ((rq->cfs.h_nr_running >= 1) && > >> + check_cpu_capacity(rq, sd)) { > >> + kick = true; > >> + goto unlock; > >> + } > >> + } > > > > vs this: how would we ever get here? > > > > If h_nr_running > 1, must then not nr_running > 1 as well? > > you're right, > but the test above can trig a kick with h_nr_running == 1 whereas the > other tests may not
Duh, clearly I cannot read today.
| |