linux-kernel - [PATCH 2/4] sched: Account per task_group nr

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 linux-cve-announce PHC
Open Source and information security mailing list archives

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

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

Message-ID: <150997923220.4082.7025655919924328239.stgit@localhost.localdomain>
Date:   Mon, 06 Nov 2017 17:40:32 +0300
From:   Kirill Tkhai <ktkhai@...tuozzo.com>
To:     mingo@...hat.com, peterz@...radead.org,
        linux-kernel@...r.kernel.org, ktkhai@...tuozzo.com
Subject: [PATCH 2/4] sched: Account per task_group nr_iowait

The patch makes number of task_group's tasks in iowait state
be tracked separately. This may be useful for containers to
check nr_iowait state of a single one.

Signed-off-by: Kirill Tkhai <ktkhai@...tuozzo.com>
---
 kernel/sched/core.c  |   45 +++++++++++++++++++++++++++++++++++++++++++++
 kernel/sched/sched.h |    5 +++++
 2 files changed, 50 insertions(+)

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 712ee54edaa1..86d1ad5f49bd 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -796,12 +796,32 @@ void deactivate_task(struct rq *rq, struct task_struct *p, int flags)
 
 static void task_iowait_start(struct rq *rq, struct task_struct *p)
 {
+#ifdef CONFIG_CGROUP_SCHED
+	struct task_group *tg = task_group(p);
+
+	/* Task's sched_task_group is changed under both of the below locks */
+	BUG_ON(!raw_spin_is_locked(&p->pi_lock) && !raw_spin_is_locked(&rq->lock));
+	while (task_group_is_autogroup(tg))
+		tg = tg->parent;
+	atomic_inc(&tg->stat[rq->cpu].nr_iowait);
+#endif
+
 	atomic_inc(&rq->nr_iowait);
 	delayacct_blkio_start();
 }
 
 static void task_iowait_end(struct rq *rq, struct task_struct *p)
 {
+#ifdef CONFIG_CGROUP_SCHED
+	struct task_group *tg = task_group(p);
+
+	/* Task's sched_task_group is changed under both of the below locks */
+	BUG_ON(!raw_spin_is_locked(&p->pi_lock) && !raw_spin_is_locked(&rq->lock));
+	while (task_group_is_autogroup(tg))
+		tg = tg->parent;
+	atomic_dec(&tg->stat[rq->cpu].nr_iowait);
+#endif
+
 	delayacct_blkio_end();
 	atomic_dec(&rq->nr_iowait);
 }
@@ -5805,6 +5825,9 @@ void __init sched_init(void)
 	sched_clock_init();
 	wait_bit_init();
 
+#ifdef CONFIG_CGROUP_SCHED
+	alloc_size += nr_cpu_ids * sizeof(struct tg_stat);
+#endif
 #ifdef CONFIG_FAIR_GROUP_SCHED
 	alloc_size += 2 * nr_cpu_ids * sizeof(void **);
 #endif
@@ -5814,6 +5837,10 @@ void __init sched_init(void)
 	if (alloc_size) {
 		ptr = (unsigned long)kzalloc(alloc_size, GFP_NOWAIT);
 
+#ifdef CONFIG_CGROUP_SCHED
+		root_task_group.stat = (struct tg_stat *)ptr;
+		ptr += nr_cpu_ids * sizeof(struct tg_stat);
+#endif
 #ifdef CONFIG_FAIR_GROUP_SCHED
 		root_task_group.se = (struct sched_entity **)ptr;
 		ptr += nr_cpu_ids * sizeof(void **);
@@ -6133,6 +6160,8 @@ static DEFINE_SPINLOCK(task_group_lock);
 
 static void sched_free_group(struct task_group *tg)
 {
+	if (tg->stat)
+		kfree(tg->stat);
 	free_fair_sched_group(tg);
 	free_rt_sched_group(tg);
 	autogroup_free(tg);
@@ -6154,6 +6183,10 @@ struct task_group *sched_create_group(struct task_group *parent)
 	if (!alloc_rt_sched_group(tg, parent))
 		goto err;
 
+	tg->stat = kzalloc(sizeof(struct tg_stat) * nr_cpu_ids, 0);
+	if (!tg->stat)
+		goto err;
+
 	return tg;
 
 err:
@@ -6207,8 +6240,16 @@ void sched_offline_group(struct task_group *tg)
 
 static void sched_change_group(struct task_struct *tsk, int type)
 {
+	int cpu = 0, queued = task_on_rq_queued(tsk);
 	struct task_group *tg;
 
+	if (!queued && tsk->in_iowait && type == TASK_MOVE_GROUP) {
+		cpu = task_cpu(tsk);
+		tg = task_group(tsk);
+		while (task_group_is_autogroup(tg))
+			tg = tg->parent;
+		atomic_dec(&tg->stat[cpu].nr_iowait);
+	}
 	/*
 	 * All callers are synchronized by task_rq_lock(); we do not use RCU
 	 * which is pointless here. Thus, we pass "true" to task_css_check()
@@ -6216,6 +6257,10 @@ static void sched_change_group(struct task_struct *tsk, int type)
 	 */
 	tg = container_of(task_css_check(tsk, cpu_cgrp_id, true),
 			  struct task_group, css);
+
+	if (!queued && tsk->in_iowait && type == TASK_MOVE_GROUP)
+		atomic_inc(&tg->stat[cpu].nr_iowait);
+
 	tg = autogroup_task_group(tsk, tg);
 	tsk->sched_task_group = tg;
 
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index 58787e3631c7..5e7cb18e1340 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -290,6 +290,10 @@ struct cfs_bandwidth {
 #endif
 };
 
+struct tg_stat {
+	atomic_t nr_iowait;
+};
+
 /* task group related information */
 struct task_group {
 	struct cgroup_subsys_state css;
@@ -330,6 +334,7 @@ struct task_group {
 #endif
 
 	struct cfs_bandwidth cfs_bandwidth;
+	struct tg_stat *stat;
 };
 
 #ifdef CONFIG_FAIR_GROUP_SCHED