sched: Fix cgroup movement of newly created process

There is a small race between do_fork() and sched_move_task(), which is
trying to move the child.

            do_fork()                 sched_move_task()
--------------------------------+---------------------------------
  copy_process()
    sched_fork()
      task_fork_fair()
        -> vruntime of the child is initialized
           based on that of the parent.
  -> we can see the child in "tasks" file now.
                                    task_rq_lock()
                                    task_move_group_fair()
                                      -> child.se.vruntime
                                           -= (old)cfs_rq->min_vruntime
                                           += (new)cfs_rq->min_vruntime
                                    task_rq_unlock()
  wake_up_new_task()
    ...
    enqueue_entity()
      child.se.vruntime += cfs_rq->min_vruntime

As a result, vruntime of the child becomes far bigger than min_vruntime,
if (new)cfs_rq->min_vruntime >> (old)cfs_rq->min_vruntime.

This patch fixes this problem by just ignoring such process in
task_move_group_fair(), because the vruntime has already been normalized in
task_fork_fair().

Signed-off-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Tejun Heo <tj@kernel.org>
Link: http://lkml.kernel.org/r/20111215143607.2ee12c5d.nishimura@mxp.nes.nec.co.jp
Signed-off-by: Ingo Molnar <mingo@elte.hu>

kernel/sched/fair.c

@@ -5338,6 +5338,19 @@ static void task_move_group_fair(struct task_struct *p, int on_rq)
 	 * to another cgroup's rq. This does somewhat interfere with the
 	 * fair sleeper stuff for the first placement, but who cares.
 	 */
+	/*
+	 * When !on_rq, vruntime of the task has usually NOT been normalized.
+	 * But there are some cases where it has already been normalized:
+	 *
+	 * - Moving a forked child which is waiting for being woken up by
+	 *   wake_up_new_task().
+	 *
+	 * To prevent boost or penalty in the new cfs_rq caused by delta
+	 * min_vruntime between the two cfs_rqs, we skip vruntime adjustment.
+	 */
+	if (!on_rq && !p->se.sum_exec_runtime)
+		on_rq = 1;
+
 	if (!on_rq)
 		p->se.vruntime -= cfs_rq_of(&p->se)->min_vruntime;
 	set_task_rq(p, task_cpu(p));