Re: [RFC][PATCH] introduce task cgroup (#task restrictioon for prevent fork bomb by cgroup)

[Dhaval Giani <dhaval@linux.vnet.ibm.com>]

> +static int task_cgroup_max_tasks_write(struct cgroup *cgrp,
> +				       struct cftype *cftype,
> +				       s64 max_tasks)
> +{
> +	struct task_cgroup *taskcg;
> +
> +	if ((max_tasks > INT_MAX) ||
> +	    (max_tasks < INT_MIN))

It should be < -1 I think.

> +		return -EINVAL;
> +
> +	taskcg = task_cgroup_from_cgrp(cgrp);
> +
> +	spin_lock(&taskcg->lock);
> +	if (max_tasks < taskcg->nr_tasks)
> +		return -EBUSY;

Shouldn't you drop the lock here?

> +	taskcg->max_tasks = max_tasks;
> +	spin_unlock(&taskcg->lock);
> +
> +	return 0;
> +}

How does this controller affect performance? Do you have some numbers?

Thanks,
-- 
regards,
Dhaval