Re: [RFC PATCH 5/6] sched/fair: trigger the update of blocked load on newly idle cpu

[Dietmar Eggemann]

On 05/02/2021 12:48, Vincent Guittot wrote:
> Instead of waking up a random and already idle CPU, we can take advantage
> of this_cpu being about to enter idle to run the ILB and update the
> blocked load.
> 
> Signed-off-by: Vincent Guittot <vincent.guittot@xxxxxxxxxx>
> ---
>  include/linux/sched/nohz.h |  2 ++
>  kernel/sched/fair.c        | 11 ++++++++---
>  kernel/sched/idle.c        |  6 ++++++
>  3 files changed, 16 insertions(+), 3 deletions(-)
> 
> diff --git a/include/linux/sched/nohz.h b/include/linux/sched/nohz.h
> index 6d67e9a5af6b..74cdc4e87310 100644
> --- a/include/linux/sched/nohz.h
> +++ b/include/linux/sched/nohz.h
> @@ -9,8 +9,10 @@
>  #if defined(CONFIG_SMP) && defined(CONFIG_NO_HZ_COMMON)
>  extern void nohz_balance_enter_idle(int cpu);
>  extern int get_nohz_timer_target(void);
> +extern void nohz_run_idle_balance(int cpu);
>  #else
>  static inline void nohz_balance_enter_idle(int cpu) { }
> +static inline void nohz_run_idle_balance(int cpu) { }
>  #endif

(1) Since nohz_run_idle_balance() would be an interface one sched class
(fair) exports to another (idle) I wonder if kernel/sched/sched.h would
be the more appropriate include file to export/define it?

nohz_balance_exit_idle() is exported via kernel/sched/sched.h (used only
within the scheduler) whereas nohz_balance_enter_idle() is exported via
include/linux/sched/nohz.h (used in kernel/time/tick-sched.c).

Isn't include/linux/sched/nohz.h the interface between kernel/sched/ and
kernel/time?

There is one exception already though: calc_load_nohz_remote() defined
in kernel/sched/loadavg.c and (only) used in kernel/sched/core.c.


(2) Is there a need for an extra function nohz_run_idle_balance()?
do_idle() could call nohz_idle_balance() directly in case in would be
exported instead.

[...]