Re: clocksource_watchdog causing scheduling of timers every second (was [v13] support "task_isolation" mode)

[Chris Metcalf]

On 7/27/2016 11:31 AM, Christoph Lameter wrote:
> Ok here is a possible patch that explicitly checks for housekeeping cpus:
>
> Subject: clocksource: Do not schedule watchdog on isolated or NOHZ cpus
>
> watchdog checks can only run on housekeeping capable cpus. Otherwise
> we will be generating noise that we would like to avoid on the isolated
> processors.
>
> Signed-off-by: Christoph Lameter <cl@xxxxxxxxx>
>
> Index: linux/kernel/time/clocksource.c
> ===================================================================
> --- linux.orig/kernel/time/clocksource.c	2016-07-27 08:41:17.109862517 -0500
> +++ linux/kernel/time/clocksource.c	2016-07-27 10:28:31.172447732 -0500
> @@ -269,9 +269,12 @@ static void clocksource_watchdog(unsigne
>   	 * Cycle through CPUs to check if the CPUs stay synchronized
>   	 * to each other.
>   	 */
> -	next_cpu = cpumask_next(raw_smp_processor_id(), cpu_online_mask);
> -	if (next_cpu >= nr_cpu_ids)
> -		next_cpu = cpumask_first(cpu_online_mask);
> +	do {
> +		next_cpu = cpumask_next(raw_smp_processor_id(), cpu_online_mask);
> +		if (next_cpu >= nr_cpu_ids)
> +			next_cpu = cpumask_first(cpu_online_mask);
> +	} while (!is_housekeeping_cpu(next_cpu));
> +
>   	watchdog_timer.expires += WATCHDOG_INTERVAL;
>   	add_timer_on(&watchdog_timer, next_cpu);
>   out:

How about using cpumask_next_and(raw_smp_processor_id(), cpu_online_mask,
housekeeping_cpumask()), likewise cpumask_first_and()?  Does that work?

Note that you should also  cpumask_first_and() in clocksource_start_watchdog(),
just to be complete.

Hopefully the init code runs after tick_init().  It seems like that's probably true.

--
Chris Metcalf, Mellanox Technologies
http://www.mellanox.com