From: Peter Zijlstra <peterz@infradead.org>
To: tglx@linutronix.de, mingo@kernel.org
Cc: linux-kernel@vger.kernel.org, bigeasy@linutronix.de,
qais.yousef@arm.com, swood@redhat.com, peterz@infradead.org,
valentin.schneider@arm.com, juri.lelli@redhat.com,
vincent.guittot@linaro.org, dietmar.eggemann@arm.com,
rostedt@goodmis.org, bsegall@google.com, mgorman@suse.de,
bristot@redhat.com, vincent.donnefort@arm.com
Subject: [PATCH 4/9] sched/core: Wait for tasks being pushed away on hotplug
Date: Mon, 21 Sep 2020 18:36:01 +0200 [thread overview]
Message-ID: <20200921163845.582326981@infradead.org> (raw)
In-Reply-To: 20200921163557.234036895@infradead.org
From: Thomas Gleixner <tglx@linutronix.de>
RT kernels need to ensure that all tasks which are not per CPU kthreads
have left the outgoing CPU to guarantee that no tasks are force migrated
within a migrate disabled section.
There is also some desire to (ab)use fine grained CPU hotplug control to
clear a CPU from active state to force migrate tasks which are not per CPU
kthreads away for power control purposes.
Add a mechanism which waits until all tasks which should leave the CPU
after the CPU active flag is cleared have moved to a different online CPU.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
---
kernel/sched/core.c | 38 +++++++++++++++++++++++++++++++++++++-
kernel/sched/sched.h | 4 ++++
2 files changed, 41 insertions(+), 1 deletion(-)
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -2588,6 +2588,20 @@ void sched_ttwu_pending(void *arg)
rq_unlock_irqrestore(rq, &rf);
}
+/*
+ * Invoked from a CPUs hotplug control thread after the CPU has been marked
+ * inactive. All tasks which are not per CPU kernel threads are either
+ * pushed off this CPU now via balance_push() or placed on a different CPU
+ * during wakeup. Wait until the CPU is quiescent.
+ */
+static void balance_hotplug_wait(void)
+{
+ struct rq *rq = this_rq();
+
+ rcuwait_wait_event(&rq->hotplug_wait, rq->nr_running == 1,
+ TASK_UNINTERRUPTIBLE);
+}
+
void send_call_function_single_ipi(int cpu)
{
struct rq *rq = cpu_rq(cpu);
@@ -6898,8 +6912,21 @@ static bool balance_push(struct rq *rq)
* Both the cpu-hotplug and stop task are in this case and are
* required to complete the hotplug process.
*/
- if (is_per_cpu_kthread(push_task))
+ if (is_per_cpu_kthread(push_task)) {
+ /*
+ * If this is the idle task on the outgoing CPU try to wake
+ * up the hotplug control thread which might wait for the
+ * last task to vanish. The rcuwait_active() check is
+ * accurate here because the waiter is pinned on this CPU
+ * and can't obviously be running in parallel.
+ */
+ if (!rq->nr_running && rcuwait_active(&rq->hotplug_wait)) {
+ raw_spin_unlock(&rq->lock);
+ rcuwait_wake_up(&rq->hotplug_wait);
+ raw_spin_lock(&rq->lock);
+ }
return false;
+ }
get_task_struct(push_task);
/*
@@ -6939,6 +6966,8 @@ static inline bool balance_push(struct r
return false;
}
+static inline void balance_hotplug_wait(void) { }
+
#endif /* CONFIG_HOTPLUG_CPU */
void set_rq_online(struct rq *rq)
@@ -7093,6 +7122,10 @@ int sched_cpu_deactivate(unsigned int cp
return ret;
}
sched_domains_numa_masks_clear(cpu);
+
+ /* Wait for all non per CPU kernel threads to vanish. */
+ balance_hotplug_wait();
+
return 0;
}
@@ -7333,6 +7366,9 @@ void __init sched_init(void)
rq_csd_init(rq, &rq->nohz_csd, nohz_csd_func);
#endif
+#ifdef CONFIG_HOTPLUG_CPU
+ rcuwait_init(&rq->hotplug_wait);
+#endif
#endif /* CONFIG_SMP */
hrtick_rq_init(rq);
atomic_set(&rq->nr_iowait, 0);
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -1004,6 +1004,10 @@ struct rq {
/* This is used to determine avg_idle's max value */
u64 max_idle_balance_cost;
+
+#ifdef CONFIG_HOTPLUG_CPU
+ struct rcuwait hotplug_wait;
+#endif
#endif /* CONFIG_SMP */
#ifdef CONFIG_IRQ_TIME_ACCOUNTING
next prev parent reply other threads:[~2020-09-21 17:01 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-09-21 16:35 [PATCH 0/9] sched: Migrate disable support Peter Zijlstra
2020-09-21 16:35 ` [PATCH 1/9] stop_machine: Add function and caller debug info Peter Zijlstra
2020-09-21 16:35 ` [PATCH 2/9] sched: Fix balance_callback() Peter Zijlstra
2020-09-23 14:08 ` Thomas Gleixner
2020-09-21 16:36 ` [PATCH 3/9] sched/hotplug: Ensure only per-cpu kthreads run during hotplug Peter Zijlstra
2020-09-25 16:38 ` Dietmar Eggemann
2020-10-02 14:20 ` Peter Zijlstra
2020-09-21 16:36 ` Peter Zijlstra [this message]
2020-09-21 16:36 ` [PATCH 5/9] sched/hotplug: Consolidate task migration on CPU unplug Peter Zijlstra
2020-10-01 17:12 ` Vincent Donnefort
2020-10-02 14:17 ` Peter Zijlstra
2020-09-21 16:36 ` [PATCH 6/9] sched: Massage set_cpus_allowed Peter Zijlstra
2020-09-23 14:07 ` Thomas Gleixner
2020-09-21 16:36 ` [PATCH 7/9] sched: Add migrate_disable() Peter Zijlstra
2020-09-21 19:16 ` Thomas Gleixner
2020-09-21 20:42 ` Daniel Bristot de Oliveira
2020-09-23 8:31 ` Thomas Gleixner
2020-09-23 10:51 ` Daniel Bristot de Oliveira
2020-09-23 17:08 ` peterz
2020-09-23 17:54 ` Daniel Bristot de Oliveira
2020-09-23 7:48 ` peterz
2020-09-24 11:53 ` Valentin Schneider
2020-09-24 12:29 ` Peter Zijlstra
2020-09-24 12:33 ` Valentin Schneider
2020-09-24 12:35 ` Peter Zijlstra
2020-09-25 16:50 ` Sebastian Andrzej Siewior
2020-10-02 14:21 ` Peter Zijlstra
2020-10-02 14:36 ` Sebastian Andrzej Siewior
2020-09-21 16:36 ` [PATCH 8/9] sched: Fix migrate_disable() vs set_cpus_allowed_ptr() Peter Zijlstra
2020-09-24 19:59 ` Valentin Schneider
2020-09-25 8:43 ` Peter Zijlstra
2020-09-25 10:07 ` Valentin Schneider
2020-09-25 9:05 ` Peter Zijlstra
2020-09-25 9:56 ` Peter Zijlstra
2020-09-25 10:09 ` Valentin Schneider
2020-09-21 16:36 ` [PATCH 9/9] sched/core: Make migrate disable and CPU hotplug cooperative Peter Zijlstra
2020-09-25 9:12 ` [PATCH 0/9] sched: Migrate disable support Dietmar Eggemann
2020-09-25 10:10 ` Peter Zijlstra
2020-09-25 11:58 ` Dietmar Eggemann
2020-09-25 12:19 ` Valentin Schneider
2020-09-25 17:49 ` Valentin Schneider
2020-09-29 9:15 ` Dietmar Eggemann
2020-09-25 18:17 ` Sebastian Andrzej Siewior
2020-09-25 19:32 ` Valentin Schneider
2020-10-02 14:30 ` Peter Zijlstra
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20200921163845.582326981@infradead.org \
--to=peterz@infradead.org \
--cc=bigeasy@linutronix.de \
--cc=bristot@redhat.com \
--cc=bsegall@google.com \
--cc=dietmar.eggemann@arm.com \
--cc=juri.lelli@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mgorman@suse.de \
--cc=mingo@kernel.org \
--cc=qais.yousef@arm.com \
--cc=rostedt@goodmis.org \
--cc=swood@redhat.com \
--cc=tglx@linutronix.de \
--cc=valentin.schneider@arm.com \
--cc=vincent.donnefort@arm.com \
--cc=vincent.guittot@linaro.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).