Re: [PATCH 2/3] cgroup, sched: let cpu serve the same files as cpuacct

[Glauber Costa <glommer-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>]

On 09/20/2012 02:43 AM, Tejun Heo wrote:
> +
> +void task_group_charge(struct task_struct *tsk, u64 cputime)
> +{
> +	struct task_group *tg;
> +	int cpu = task_cpu(tsk);
> +
> +	rcu_read_lock();
> +
> +	tg = container_of(task_subsys_state(tsk, cpu_cgroup_subsys_id),
> +			  struct task_group, css);
> +
> +	for (; tg; tg = tg->parent) {
> +		u64 *cpuusage = per_cpu_ptr(tg->cpuusage, cpu);
> +		*cpuusage += cputime;
> +	}
> +
> +	rcu_read_unlock();
> +}
>  #endif /* CONFIG_CGROUP_SCHED */

The whole point of this merge is that this is not needed.
This information is already available from exec_clock for fair tasks.
for rt tasks, we have no exec clock, but do have a hierarchy walk a bit
below the current cpuacct charge, that can be used for that purpose.