linux-kernel - [PATCH] sched/core: Reduce cost of sched_move

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Date:   Sun, 19 Mar 2023 15:56:43 +0800
From:   wuchi <wuchi.zero@...il.com>
To:     mingo@...hat.com, peterz@...radead.org, juri.lelli@...hat.com,
        vincent.guittot@...aro.org, dietmar.eggemann@....com,
        rostedt@...dmis.org, bsegall@...gle.com, mgorman@...e.de,
        bristot@...hat.com, vschneid@...hat.com
Cc:     linux-kernel@...r.kernel.org
Subject: [PATCH] sched/core: Reduce cost of sched_move_task when config autogroup

Some sched_move_task calls of autogroup is useless when the
task_struct->sched_task_group isn't changed because of task_group
of cpu_cgroup overlay task_group of autogroup. The overlay key codes
are as follows:

sched_cgroup_fork->autogroup_task_group->task_wants_autogroup
sched_change_group->autogroup_task_group->autogroup_task_group

sched_move_task eg:
task A belongs to cpu_cgroup0 and autogroup0, it will always to
cpu_cgroup0 when doing exit. So there is no need to do {de|en}queue.
The call graph is as follow.

do_exit
  sched_autogroup_exit_task
    sched_move_task
      dequeue_task
        sched_change_group
	  A.sched_task_group = sched_get_task_group
      enqueue_task

So do some check before dequeue task in sched_move_task.

Signed-off-by: wuchi <wuchi.zero@...il.com>
---
 kernel/sched/core.c | 29 +++++++++++++++++++++++++++--
 1 file changed, 27 insertions(+), 2 deletions(-)

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index a380f34789a2..acc9a0e391f4 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -10330,7 +10330,7 @@ void sched_release_group(struct task_group *tg)
 	spin_unlock_irqrestore(&task_group_lock, flags);
 }
 
-static void sched_change_group(struct task_struct *tsk)
+static struct task_group *sched_get_task_group(struct task_struct *tsk)
 {
 	struct task_group *tg;
 
@@ -10342,7 +10342,28 @@ static void sched_change_group(struct task_struct *tsk)
 	tg = container_of(task_css_check(tsk, cpu_cgrp_id, true),
 			  struct task_group, css);
 	tg = autogroup_task_group(tsk, tg);
-	tsk->sched_task_group = tg;
+
+	return tg;
+}
+
+static bool sched_task_group_changed(struct task_struct *tsk)
+{
+	/*
+	 * Some sched_move_task calls of autogroup is useless when the
+	 * task_struct->sched_task_group isn't changed because of task_group
+	 * of cpu_cgroup overlay task_group of autogroup. so do some check
+	 * before dequeue task in sched_move_task.
+	 */
+#ifdef CONFIG_SCHED_AUTOGROUP
+	return sched_get_task_group(tsk) != tsk->sched_task_group;
+#else
+	return true;
+#endif /* CONFIG_SCHED_AUTOGROUP */
+}
+
+static void sched_change_group(struct task_struct *tsk)
+{
+	tsk->sched_task_group = sched_get_task_group(tsk);
 
 #ifdef CONFIG_FAIR_GROUP_SCHED
 	if (tsk->sched_class->task_change_group)
@@ -10369,6 +10390,9 @@ void sched_move_task(struct task_struct *tsk)
 	rq = task_rq_lock(tsk, &rf);
 	update_rq_clock(rq);
 
+	if (!sched_task_group_changed(tsk))
+		goto unlock;
+
 	running = task_current(rq, tsk);
 	queued = task_on_rq_queued(tsk);
 
@@ -10391,6 +10415,7 @@ void sched_move_task(struct task_struct *tsk)
 		resched_curr(rq);
 	}
 
+unlock:
 	task_rq_unlock(rq, tsk, &rf);
 }
 
-- 
2.20.1