From: byungchul.park@lge.com
To: mingo@kernel.org, peterz@infradead.org
Cc: linux-kernel@vger.kernel.org, yuyang.du@intel.com,
        Byungchul Park <byungchul.park@lge.com>
Subject: [PATCH 2/2] sched: make task_move_group_fair simple by using switched_to(from)_fair
Date: Thu, 13 Aug 2015 19:49:19 +0900
Message-Id: <1439462959-2312-2-git-send-email-byungchul.park@lge.com>
In-Reply-To: <1439462959-2312-1-git-send-email-byungchul.park@lge.com>
References: <1439462959-2312-1-git-send-email-byungchul.park@lge.com>
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 4911
Lines: 141

From: Byungchul Park <byungchul.park@lge.com>

i introduced need_vruntime_adjust() to check if do or not adjust vruntime
when attaching/detaching a se to/from its cfs_rq, and use it.

i made the best use of switched_to(from)_fair for attach(detach)
operations in task_move_group_fair().

Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
 kernel/sched/fair.c |   80 +++++++++++++++++++--------------------------------
 1 file changed, 29 insertions(+), 51 deletions(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index d91e4da..5ec846f 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -7904,6 +7904,31 @@ prio_changed_fair(struct rq *rq, struct task_struct *p, int oldprio)
 		check_preempt_curr(rq, p, 0);
 }
 
+static inline int need_vruntime_adjust(struct task_struct *p, int queued)
+{
+	struct sched_entity *se = &p->se;
+
+	/*
+	 * When !queued, 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().
+	 * - Moving a task which has been woken up by try_to_wake_up() and
+	 *   waiting for actually being woken up by sched_ttwu_pending().
+	 */
+	if (queued)
+		return 0;
+
+	if (!se->sum_exec_runtime || p->state == TASK_WAKING)
+		return 0;
+
+	if (p->state != TASK_RUNNING)
+		return 1;
+
+	return 0;
+}
+
 static void switched_from_fair(struct rq *rq, struct task_struct *p)
 {
 	struct sched_entity *se = &p->se;
@@ -7918,7 +7943,7 @@ static void switched_from_fair(struct rq *rq, struct task_struct *p)
 	 * have normalized the vruntime, if it's !queued, then only when
 	 * the task is sleeping will it still have non-normalized vruntime.
 	 */
-	if (!task_on_rq_queued(p) && p->state != TASK_RUNNING) {
+	if (need_vruntime_adjust(p, task_on_rq_queued(p))) {
 		/*
 		 * Fix up our vruntime so that the current sleep doesn't
 		 * cause 'unlimited' sleep bonus.
@@ -7939,7 +7964,6 @@ static void switched_from_fair(struct rq *rq, struct task_struct *p)
 static void switched_to_fair(struct rq *rq, struct task_struct *p)
 {
 	struct sched_entity *se = &p->se;
-
 #ifdef CONFIG_FAIR_GROUP_SCHED
 	/*
 	 * Since the real-depth could have been changed (only FAIR
@@ -7964,7 +7988,7 @@ static void switched_to_fair(struct rq *rq, struct task_struct *p)
 		 * has non-normalized vruntime, if it's !queued, then it still has
 		 * normalized vruntime.
 		 */
-		if (p->state != TASK_RUNNING)
+		if (need_vruntime_adjust(p, task_on_rq_queued(p)))
 			se->vruntime += cfs_rq_of(se)->min_vruntime;
 		return;
 	}
@@ -8014,55 +8038,9 @@ void init_cfs_rq(struct cfs_rq *cfs_rq)
 #ifdef CONFIG_FAIR_GROUP_SCHED
 static void task_move_group_fair(struct task_struct *p, int queued)
 {
-	struct sched_entity *se = &p->se;
-	struct cfs_rq *cfs_rq;
-
-	/*
-	 * If the task was not on the rq at the time of this cgroup movement
-	 * it must have been asleep, sleeping tasks keep their ->vruntime
-	 * absolute on their old rq until wakeup (needed for the fair sleeper
-	 * bonus in place_entity()).
-	 *
-	 * If it was on the rq, we've just 'preempted' it, which does convert
-	 * ->vruntime to a relative base.
-	 *
-	 * Make sure both cases convert their relative position when migrating
-	 * to another cgroup's rq. This does somewhat interfere with the
-	 * fair sleeper stuff for the first placement, but who cares.
-	 */
-	/*
-	 * When !queued, 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().
-	 * - Moving a task which has been woken up by try_to_wake_up() and
-	 *   waiting for actually being woken up by sched_ttwu_pending().
-	 *
-	 * 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 (!queued && (!se->sum_exec_runtime || p->state == TASK_WAKING))
-		queued = 1;
-
-	cfs_rq = cfs_rq_of(se);
-	if (!queued)
-		se->vruntime -= cfs_rq->min_vruntime;
-
-#ifdef CONFIG_SMP
-	/* synchronize task with its prev cfs_rq */
-	detach_entity_load_avg(cfs_rq, se);
-#endif
+	switched_from_fair(task_rq(p), p);
 	set_task_rq(p, task_cpu(p));
-	se->depth = se->parent ? se->parent->depth + 1 : 0;
-	cfs_rq = cfs_rq_of(se);
-	if (!queued)
-		se->vruntime += cfs_rq->min_vruntime;
-
-#ifdef CONFIG_SMP
-	/* Virtually synchronize task with its new cfs_rq */
-	attach_entity_load_avg(cfs_rq, se);
-#endif
+	switched_to_fair(task_rq(p), p);
 }
 
 void free_fair_sched_group(struct task_group *tg)
-- 
1.7.9.5

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/