linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Lai Jiangshan <laijs@cn.fujitsu.com>
To: <jjherne@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>,
	Sasha Levin <sasha.levin@oracle.com>, Tejun Heo <tj@kernel.org>,
	LKML <linux-kernel@vger.kernel.org>,
	Dave Jones <davej@redhat.com>, Ingo Molnar <mingo@redhat.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	Steven Rostedt <rostedt@goodmis.org>
Subject: Re: workqueue: WARN at at kernel/workqueue.c:2176
Date: Tue, 3 Jun 2014 20:45:39 +0800	[thread overview]
Message-ID: <538DC373.9@cn.fujitsu.com> (raw)
In-Reply-To: <538DB076.4090704@cn.fujitsu.com>


Hi, Peter,

I rewrote the analyse. (scheduler_ipi() must be called before stopper-task,
so the part for workqueue of the old analyse maybe be wrong.)


I found something strange by review (just by review, no test yet)

int set_cpus_allowed_ptr(struct task_struct *p, const struct cpumask *new_mask)
{
...
	if (p->on_rq) {
		struct migration_arg arg = { p, dest_cpu };
		/* Need help from migration thread: drop lock and wait. */
		task_rq_unlock(rq, p, &flags);
		stop_one_cpu(cpu_of(rq), migration_cpu_stop, &arg);
		tlb_migrate_finish(p->mm);
		return 0;
	}
...
}

The branch failed to migrate a waken-up task with p->on_rq==0 if TTWU_QUEUE is enabled:
	p->wake_entry is added to the rq,
	p->state is TASK_WAKING
	p->on_rq is 0

In this case set_cpus_allowed_ptr() fails to migrate the waken-up task!!!.

Go back to workqueue for higher level analysing.

task1					task2					cpu#4
					wake_up_process(worker1)
					  ttwu_queue_remote() #queue worker1
					  to cpu#4
workqueue_cpu_up_callback(cpu=5)

set_cpus_allowed_ptr(worker1)
  set worker's cpuallowed to
  cpumask_of(5)
  see worker1->on_rq = 0,
  do not migrate it.
										scheduler_ipi()
										set worker1->on_rq = 1

										wq_worker_waking_up(worker1)
										fail to hit the WARN_ON()
										  due to WORKER_UNBOUND is not cleared.
set_cpus_allowed_ptr() return 0
clear WORKER_UNBOUND for worker1

In this case, the WARN_ON() in process_one_work() hit.

Thanks,
Lai


The following code maybe help.
---
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 268a45e..1a198a5 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -4530,7 +4530,7 @@ int set_cpus_allowed_ptr(struct task_struct *p, const struct cpumask *new_mask)
 		goto out;
 
 	dest_cpu = cpumask_any_and(cpu_active_mask, new_mask);
-	if (p->on_rq) {
+	if (p->on_rq || p->state == TASK_WAKING) {
 		struct migration_arg arg = { p, dest_cpu };
 		/* Need help from migration thread: drop lock and wait. */
 		task_rq_unlock(rq, p, &flags);


  reply	other threads:[~2014-06-03 12:41 UTC|newest]

Thread overview: 50+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-05-12 18:58 workqueue: WARN at at kernel/workqueue.c:2176 Sasha Levin
2014-05-12 20:01 ` Tejun Heo
2014-05-13  2:19   ` Lai Jiangshan
2014-05-13  2:17     ` Sasha Levin
2014-05-14 16:52       ` Jason J. Herne
2014-05-16  3:50         ` Lai Jiangshan
2014-05-16  9:35           ` Peter Zijlstra
2014-05-16  9:56             ` Lai Jiangshan
2014-05-16 10:29               ` Peter Zijlstra
2014-05-16 10:15             ` Peter Zijlstra
2014-05-16 10:16               ` Peter Zijlstra
2014-05-16 10:39                 ` Peter Zijlstra
2014-05-16 11:57           ` Peter Zijlstra
2014-05-16 12:08             ` Tejun Heo
2014-05-16 12:14               ` Thomas Gleixner
2014-05-16 12:16                 ` Tejun Heo
2014-05-16 16:18             ` Lai Jiangshan
2014-05-16 16:29               ` Peter Zijlstra
2014-05-27 14:18                 ` Jason J. Herne
2014-05-27 14:26                   ` Peter Zijlstra
2014-05-29 16:23                     ` Jason J. Herne
2014-06-03 11:24                       ` Lai Jiangshan
2014-06-03 12:45                         ` Lai Jiangshan [this message]
2014-06-03 14:28                           ` Peter Zijlstra
2014-06-04  1:47                             ` Lai Jiangshan
2014-06-03 14:16                         ` Peter Zijlstra
2014-06-04  2:27                           ` Lai Jiangshan
2014-06-04  6:49                             ` Peter Zijlstra
2014-06-04  8:25                               ` Lai Jiangshan
2014-06-04  9:39                                 ` Peter Zijlstra
2014-06-05 10:54                                   ` Lai Jiangshan
2014-06-05 15:22                                     ` Jason J. Herne
2014-06-06 12:39                                     ` Jason J. Herne
2014-06-06 13:36                                     ` Peter Zijlstra
2014-06-08  2:50                                       ` Lai Jiangshan
2014-09-01  3:04                                       ` Lai Jiangshan
2014-09-03 15:15                                         ` Peter Zijlstra
2014-09-04  2:22                                           ` Lai Jiangshan
2014-09-04  6:39                                             ` Peter Zijlstra
2014-06-09 14:01                                     ` Jason J. Herne
2014-06-10  1:21                                       ` Lai Jiangshan
2014-06-16  1:30                                         ` Lai Jiangshan
2014-09-09 14:52                                 ` [tip:sched/core] sched: Migrate waking tasks tip-bot for Lai Jiangshan
2014-09-10  7:38                                   ` Kirill Tkhai
2014-09-10  7:53                                     ` Peter Zijlstra
2014-06-04  2:28                         ` workqueue: WARN at at kernel/workqueue.c:2176 Lai Jiangshan
2014-06-04  6:48                           ` Peter Zijlstra
2014-05-19 13:07           ` [tip:sched/core] sched: Fix hotplug vs set_cpus_allowed_ptr() tip-bot for Lai Jiangshan
2014-05-22 12:26           ` [tip:sched/core] sched: Fix hotplug vs. set_cpus_allowed_ptr() tip-bot for Lai Jiangshan
2014-05-22 22:02             ` Srivatsa S. Bhat

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=538DC373.9@cn.fujitsu.com \
    --to=laijs@cn.fujitsu.com \
    --cc=davej@redhat.com \
    --cc=jjherne@linux.vnet.ibm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=peterz@infradead.org \
    --cc=rostedt@goodmis.org \
    --cc=sasha.levin@oracle.com \
    --cc=tglx@linutronix.de \
    --cc=tj@kernel.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).