Re: [PATCH] kernel/sched: run nohz idle load balancer on HK_FLAG_MISC CPUs

From: Peter Zijlstra
Date: Thu Apr 25 2019 - 07:56:29 EST


On Fri, Apr 12, 2019 at 02:26:13PM +1000, Nicholas Piggin wrote:
> The nohz idle balancer runs on the lowest idle CPU. This can
> interfere with isolated CPUs, so confine it to HK_FLAG_MISC
> housekeeping CPUs.
>
> HK_FLAG_SCHED is not used for this because it is not set anywhere
> at the moment. This could be folded into HK_FLAG_SCHED once that
> option is fixed.

Frederic? Anyway, I thnk I'll take this patch as is.

Thanks!

> The problem was observed with increased jitter on an application
> running on CPU0, caused by nohz idle load balancing being run on
> CPU1 (an SMT sibling).
>
> Signed-off-by: Nicholas Piggin <npiggin@xxxxxxxxx>
> ---
> kernel/sched/fair.c | 16 ++++++++++------
> 1 file changed, 10 insertions(+), 6 deletions(-)
>
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index fdab7eb6f351..d29ca323214d 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -9522,22 +9522,26 @@ static inline int on_null_domain(struct rq *rq)
> * - When one of the busy CPUs notice that there may be an idle rebalancing
> * needed, they will kick the idle load balancer, which then does idle
> * load balancing for all the idle CPUs.
> + * - HK_FLAG_MISC CPUs are used for this task, because HK_FLAG_SCHED not set
> + * anywhere yet.
> */
>
> static inline int find_new_ilb(void)
> {
> - int ilb = cpumask_first(nohz.idle_cpus_mask);
> + int ilb;
>
> - if (ilb < nr_cpu_ids && idle_cpu(ilb))
> - return ilb;
> + for_each_cpu_and(ilb, nohz.idle_cpus_mask,
> + housekeeping_cpumask(HK_FLAG_MISC)) {
> + if (idle_cpu(ilb))
> + return ilb;
> + }
>
> return nr_cpu_ids;
> }
>
> /*
> - * Kick a CPU to do the nohz balancing, if it is time for it. We pick the
> - * nohz_load_balancer CPU (if there is one) otherwise fallback to any idle
> - * CPU (if there is one).
> + * Kick a CPU to do the nohz balancing, if it is time for it. We pick any
> + * idle CPU in the HK_FLAG_MISC housekeeping set (if there is one).
> */
> static void kick_ilb(unsigned int flags)
> {
> --
> 2.20.1
>