Re: Nohz_full on boot CPU is broken (was: Re: [PATCH v2 1/1] wq: Avoid using isolated cpus' timers on queue_delayed_work)

[Oleg Nesterov]

On 04/10, Oleg Nesterov wrote:
>
> Will you agree with the change above or what do you suggest instead as
> a simple workaround?

OK, let me send the patch.

Nicholas, Frederic, please review.

> Again, I am not sure I understand, but I too thought that something like
>
> 	housekeeping_check(void)
> 	{
> 		for_each_set_bit(type, &housekeeping.flags, HK_TYPE_MAX) {
> 			if (!cpumask_intersects(cpu_online, housekeeping.cpumasks[type]))
> 				panic();
> 		}
>
> after bringup_nonboot_cpus(setup_max_cpus).

It seems we can make a simpler patch for this. It will (should) fix another
problem when maxcpus < first-housekeeping.

I'll recheck and try to do this later, lets solve the reported problem first.

Oleg.