On Fri, 24 Jun 2016, Richard Cochran wrote:
> On Fri, Jun 24, 2016 at 02:32:19PM -0000, Thomas Gleixner wrote:
> > @@ -809,6 +815,12 @@ static void tick_nohz_restart_sched_tick
> > tick_do_update_jiffies64(now);
> > cpu_load_update_nohz_stop();
> >
> > + /*
> > + * Clear the timer idle flag, so we avoid IPIs on remote queueing and
> > + * the clock forward checks in the enqueue path.
> > + */
> > + timer_clear_idle();
>
> Since function calls timer_clear_idle() already ...
>
> > calc_load_exit_idle();
> > touch_softlockup_watchdog_sched();
> > /*
> > @@ -1025,6 +1037,7 @@ void tick_nohz_idle_exit(void)
> > tick_nohz_stop_idle(ts, now);
> >
> > if (ts->tick_stopped) {
> > + timer_clear_idle();
> > tick_nohz_restart_sched_tick(ts, now);
> > tick_nohz_account_idle_ticks(ts);
> > }
>
> ... there is no need to call timer_clear_idle() here.
Indeed.
