From: Frederic Weisbecker <frederic@kernel.org>
To: Peter Zijlstra <peterz@infradead.org>,
Thomas Gleixner <tglx@linutronix.de>
Cc: LKML <linux-kernel@vger.kernel.org>,
Frederic Weisbecker <frederic@kernel.org>,
"Rafael J . Wysocki" <rafael.j.wysocki@intel.com>,
Ti Zhou <x2019cwm@stfx.ca>, Yunfeng Ye <yeyunfeng@huawei.com>,
"Paul E . McKenney" <paulmck@kernel.org>,
Marcelo Tosatti <mtosatti@redhat.com>,
Ingo Molnar <mingo@kernel.org>
Subject: [PATCH 08/10] tick/nohz: Only wakeup a single target cpu when kicking a task
Date: Thu, 11 Mar 2021 13:37:06 +0100 [thread overview]
Message-ID: <20210311123708.23501-9-frederic@kernel.org> (raw)
In-Reply-To: <20210311123708.23501-1-frederic@kernel.org>
When adding a tick dependency to a task, its necessary to
wakeup the CPU where the task resides to reevaluate tick
dependencies on that CPU.
However the current code wakes up all nohz_full CPUs, which
is unnecessary.
Switch to waking up a single CPU, by using ordering of writes
to task->cpu and task->tick_dep_mask.
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Cc: Yunfeng Ye <yeyunfeng@huawei.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
---
kernel/time/tick-sched.c | 40 +++++++++++++++++++++++++++-------------
1 file changed, 27 insertions(+), 13 deletions(-)
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index 2a041d0dc3eb..8457f15a5073 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -322,6 +322,31 @@ void tick_nohz_full_kick_cpu(int cpu)
irq_work_queue_on(&per_cpu(nohz_full_kick_work, cpu), cpu);
}
+static void tick_nohz_kick_task(struct task_struct *tsk)
+{
+ int cpu = task_cpu(tsk);
+
+ /*
+ * If the task concurrently migrates to another cpu,
+ * we guarantee it sees the new tick dependency upon
+ * schedule.
+ *
+ *
+ * set_task_cpu(p, cpu);
+ * STORE p->cpu = @cpu
+ * __schedule() (switch to task 'p')
+ * LOCK rq->lock
+ * smp_mb__after_spin_lock() STORE p->tick_dep_mask
+ * tick_nohz_task_switch() smp_mb() (atomic_fetch_or())
+ * LOAD p->tick_dep_mask LOAD p->cpu
+ */
+
+ preempt_disable();
+ if (cpu_online(cpu))
+ tick_nohz_full_kick_cpu(cpu);
+ preempt_enable();
+}
+
/*
* Kick all full dynticks CPUs in order to force these to re-evaluate
* their dependency on the tick and restart it if necessary.
@@ -404,19 +429,8 @@ EXPORT_SYMBOL_GPL(tick_nohz_dep_clear_cpu);
*/
void tick_nohz_dep_set_task(struct task_struct *tsk, enum tick_dep_bits bit)
{
- if (!atomic_fetch_or(BIT(bit), &tsk->tick_dep_mask)) {
- if (tsk == current) {
- preempt_disable();
- tick_nohz_full_kick();
- preempt_enable();
- } else {
- /*
- * Some future tick_nohz_full_kick_task()
- * should optimize this.
- */
- tick_nohz_full_kick_all();
- }
- }
+ if (!atomic_fetch_or(BIT(bit), &tsk->tick_dep_mask))
+ tick_nohz_kick_task(tsk);
}
EXPORT_SYMBOL_GPL(tick_nohz_dep_set_task);
--
2.25.1
next prev parent reply other threads:[~2021-03-11 12:38 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-11 12:36 [PATCH 00/10] tick/nohz updates Frederic Weisbecker
2021-03-11 12:36 ` [PATCH 01/10] tick/nohz: Prevent tick_nohz_get_sleep_length() from returning negative value Frederic Weisbecker
2021-03-16 12:21 ` Peter Zijlstra
2021-03-16 13:37 ` Frederic Weisbecker
2021-03-16 14:35 ` Peter Zijlstra
2021-03-16 14:53 ` Frederic Weisbecker
2021-03-16 15:26 ` Rafael J. Wysocki
2021-03-16 15:57 ` 回复: " Zhou Ti (x2019cwm)
2021-03-16 16:08 ` Zhou Ti (x2019cwm)
2021-03-16 16:25 ` Peter Zijlstra
2021-03-17 21:49 ` Zhou Ti (x2019cwm)
2021-03-25 13:14 ` Frederic Weisbecker
2021-03-25 18:56 ` Rafael J. Wysocki
2021-03-25 19:18 ` Zhou Ti (x2019cwm)
2021-03-25 19:50 ` Rafael J. Wysocki
2021-03-25 20:37 ` Zhou Ti (x2019cwm)
2021-03-26 17:01 ` Rafael J. Wysocki
2021-03-26 17:53 ` Zhou Ti (x2019cwm)
2021-03-26 18:54 ` Rafael J. Wysocki
2021-03-26 22:53 ` Zhou Ti (x2019cwm)
2021-03-29 12:44 ` Rafael J. Wysocki
2021-03-29 14:49 ` Zhou Ti (x2019cwm)
2021-03-11 12:37 ` [PATCH 02/10] tick/nohz: Add tick_nohz_full_this_cpu() Frederic Weisbecker
2021-03-16 12:28 ` Peter Zijlstra
2021-03-16 13:05 ` Frederic Weisbecker
2021-03-11 12:37 ` [PATCH 03/10] tick/nohz: Conditionally restart tick on idle exit Frederic Weisbecker
2021-03-11 12:37 ` [PATCH 04/10] tick/nohz: Remove superflous check for CONFIG_VIRT_CPU_ACCOUNTING_NATIVE Frederic Weisbecker
2021-03-11 12:37 ` [PATCH 05/10] tick/nohz: Update idle_exittime on actual idle exit Frederic Weisbecker
2021-03-11 12:37 ` [PATCH 06/10] timer: Report ignored local enqueue in nohz mode Frederic Weisbecker
2021-03-16 15:27 ` Peter Zijlstra
2021-03-25 13:07 ` Frederic Weisbecker
2021-03-11 12:37 ` [PATCH 07/10] tick/nohz: Update nohz_full Kconfig help Frederic Weisbecker
2021-03-11 12:37 ` Frederic Weisbecker [this message]
2021-03-11 12:37 ` [PATCH 09/10] tick/nohz: Change signal tick dependency to wakeup CPUs of member tasks Frederic Weisbecker
2021-03-11 12:37 ` [PATCH 10/10] tick/nohz: Kick only _queued_ task whose tick dependency is updated Frederic Weisbecker
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=20210311123708.23501-9-frederic@kernel.org \
--to=frederic@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=mtosatti@redhat.com \
--cc=paulmck@kernel.org \
--cc=peterz@infradead.org \
--cc=rafael.j.wysocki@intel.com \
--cc=tglx@linutronix.de \
--cc=x2019cwm@stfx.ca \
--cc=yeyunfeng@huawei.com \
/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