Hello, Li.
On Wed, Aug 21, 2013 at 06:01:05PM +0800, Li Zefan wrote:
> Firstly offline cpu1:
>
> # echo 0-1 > cpuset.cpus
> # echo 0 > /sys/devices/system/cpu/cpu1/online
> # cat cpuset.cpus
> 0-1
> # cat cpuset.effective_cpus
> 0
>
> Then online it:
>
> # echo 1 > /sys/devices/system/cpu/cpu1/online
> # cat cpuset.cpus
> 0-1
> # cat cpuset.effective_cpus
> 0-1
>
> And cpuset will bring it back to the effective mask.
>
> This is a behavior change for sane_behavior.
It'd be great if the patch description also explains "how".
> +static void hotplug_update_tasks_insane(struct cpuset *cs,
> + struct cpumask *off_cpus,
> + nodemask_t *off_mems)
> +{
> + bool is_empty;
> +
> + cpumask_andnot(off_cpus, cs->real_cpus_allowed,
> + top_cpuset.real_cpus_allowed);
> + nodes_andnot(*off_mems, cs->real_mems_allowed,
> + top_cpuset.real_mems_allowed);
> +
> + mutex_lock(&callback_mutex);
> + cpumask_andnot(cs->cpus_allowed, cs->cpus_allowed, off_cpus);
> + cpumask_andnot(cs->real_cpus_allowed, cs->real_cpus_allowed, off_cpus);
> + nodes_andnot(cs->mems_allowed, cs->mems_allowed, *off_mems);
> + nodes_andnot(cs->real_mems_allowed, cs->real_mems_allowed, *off_mems);
> + mutex_unlock(&callback_mutex);
> +
> + /*
> + * Don't call update_tasks_cpumask() if the cpuset becomes empty,
> + * as the tasks will be migrated to an ancestor.
> + */
> + if (!cpumask_empty(off_cpus) && !cpumask_empty(cs->cpus_allowed))
> + update_tasks_cpumask(cs, NULL);
> +
> + if (!nodes_empty(*off_mems) && !cpumask_empty(cs->cpus_allowed))
> + update_tasks_nodemask(cs, NULL);
> +
> + is_empty = cpumask_empty(cs->cpus_allowed) ||
> + nodes_empty(cs->mems_allowed);
> +
> + mutex_unlock(&cpuset_mutex);
> + /*
> + * Move tasks to the nearest ancestor with execution resources,
> + * This is full cgroup operation which will also call back into
> + * cpuset. Should be don outside any lock.
> + */
> + if (is_empty)
> + remove_tasks_in_empty_cpuset(cs);
> + mutex_lock(&cpuset_mutex);
> +}
Maybe the patch would be easier to follow if factoring out the above
was in a separate patch?
Thanks.
--
tejun
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
