Re: cpusets not cpu hotplug aware

[Paul Jackson <pj@sgi.com>]

Simon,

  Take note below of my proposal to make the 'cpus' and 'mems'
  of the top cpuset read-only.  Holler if that hurts.

Nathan wrote:
> +static int cpuset_handle_cpuhp(struct notifier_block *nb,
> +				unsigned long phase, void *_cpu)
> +{
> +	unsigned long cpu = (unsigned long)_cpu;
> +
> +	mutex_lock(&manage_mutex);
> +	mutex_lock(&callback_mutex);
> +
> +	switch (phase) {
> +	case CPU_ONLINE:
> +		cpu_set(cpu, top_cpuset.cpus_allowed);
> +		break;
> +	case CPU_DEAD:
> +		cpu_clear(cpu, top_cpuset.cpus_allowed);
> +		break;
> +	}
> +
> +	mutex_unlock(&callback_mutex);
> +	mutex_unlock(&manage_mutex);
> +
> +	return 0;
> +}
> +

Andrew commented:
> I must say, that's a pretty simple patch. 

Not simple enough ;).

How about (uncompiled, untested, unanything):

=================================================================
/*
 * The top_cpuset tracks what CPUs and Memory Nodes are online,
 * period.  This is necessary in order to make cpusets transparent
 * (of no affect) on systems that are actively using CPU hotplug
 * but making no active use of cpusets.
 *
 * This handles CPU hotplug (cpuhp) events.  If someday Memory
 * Nodes can be hotplugged (dynamically changing node_online_map)
 * then we should handle that too, perhaps in a similar way.
 */

#ifdef CONFIG_HOTPLUG_CPU
static int cpuset_handle_cpuhp(struct notifier_block *nb,
				unsigned long phase, void *cpu)
{
	mutex_lock(&manage_mutex);
	mutex_lock(&callback_mutex);

	top_cpuset.cpus_allowed = cpu_online_map;

	mutex_unlock(&callback_mutex);
	mutex_unlock(&manage_mutex);

	return 0;
}
#endif

... plus the hotcpu_notifier() initializer.
=================================================================

However I get to spend the few lines of code I saved here elsewhere,
adding special case code so that the top_cpuset's 'cpus' and 'mems'
files are read-only.

The new rule is simple:

  The top cpuset's cpus and mems track what is online.

The user is no longer in direct control of these two cpuset settings.

And I should add a few lines to Documentation/cpusets.txt, describing
this.

When I return from my son's 18-th birthday party this evening, I will
see what I can whip up.

Thanks, Nathan.

-- 
                  I won't rest till it's the best ...
                  Programmer, Linux Scalability
                  Paul Jackson <pj@sgi.com> 1.925.600.0401