Messages in this thread | | | Date | Tue, 10 Dec 2013 15:02:19 +0100 | From | Frederic Weisbecker <> | Subject | Re: [PATCH v2 3/3] nohz_full: update cpu load fix in nohz_full |
| |
On Tue, Dec 03, 2013 at 08:35:12PM +0800, Alex Shi wrote: > We are not always 0 when update nohz cpu load, after nohz_full enabled. > But current code still treat the cpu as idle. that is incorrect. > Fix it to use correct cpu_load. > > Signed-off-by: Alex Shi <alex.shi@linaro.org> > --- > kernel/sched/proc.c | 8 +++++++- > 1 file changed, 7 insertions(+), 1 deletion(-) > > diff --git a/kernel/sched/proc.c b/kernel/sched/proc.c > index 16f5a30..f1441f0 100644 > --- a/kernel/sched/proc.c > +++ b/kernel/sched/proc.c > @@ -568,8 +568,14 @@ void update_cpu_load_nohz(void) > /* > * We were idle, this means load 0, the current load might be > * !0 due to remote wakeups and the sort. > + * or we may has only one task and in NO_HZ_FULL, then still use > + * normal cpu load. > */ > - __update_cpu_load(this_rq, 0, pending_updates); > + if (this_rq->cfs.h_nr_running) { > + unsigned load = get_rq_runnable_load(this_rq); > + __update_cpu_load(this_rq, load, pending_updates); > + } else > + __update_cpu_load(this_rq, 0, pending_updates);
But decay_load_missed() doesnt handle non 0 loads, right? It probably make more sense to first fix __update_cpu_load() to make it handle this kind of thing before fixing the caller.
Now you had patches that remove the cpu_load secondary idx I think? You should move this patch to that series.
> } > raw_spin_unlock(&this_rq->lock); > } > -- > 1.8.1.2 >
| |