All of lore.kernel.org
 help / color / mirror / Atom feed
From: Tejun Heo <tj@kernel.org>
To: Peter Zijlstra <peterz@infradead.org>
Cc: Ulrich Obergfell <uobergfe@redhat.com>,
	Ingo Molnar <mingo@redhat.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	linux-kernel@vger.kernel.org, kernel-team@fb.com
Subject: [PATCH] workqueue: warn if memory reclaim tries to flush !WQ_MEM_RECLAIM workqueue
Date: Thu, 3 Dec 2015 14:26:16 -0500	[thread overview]
Message-ID: <20151203192616.GJ27463@mtj.duckdns.org> (raw)
In-Reply-To: <20151203150604.GC27463@mtj.duckdns.org>

Task or work item involved in memory reclaim trying to flush a
non-WQ_MEM_RECLAIM workqueue or one of its work items can lead to
deadlock.  Trigger WARN_ONCE() if such conditions are detected.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
---
Hello,

So, something like this.  Seems to work fine here.  If there's no
objection, I'm gonna push it through wq/for-4.5.

Thanks.

 kernel/workqueue.c |   35 +++++++++++++++++++++++++++++++++++
 1 file changed, 35 insertions(+)

--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -2330,6 +2330,37 @@ repeat:
 	goto repeat;
 }
 
+/**
+ * check_flush_dependency - check for flush dependency sanity
+ * @target_wq: workqueue being flushed
+ * @target_work: work item being flushed (NULL for workqueue flushes)
+ *
+ * %current is trying to flush the whole @target_wq or @target_work on it.
+ * If @target_wq doesn't have %WQ_MEM_RECLAIM, verify that %current is not
+ * reclaiming memory or running on a workqueue which doesn't have
+ * %WQ_MEM_RECLAIM as that can break forward-progress guarantee leading to
+ * a deadlock.
+ */
+static void check_flush_dependency(struct workqueue_struct *target_wq,
+				   struct work_struct *target_work)
+{
+	work_func_t target_func = target_work ? target_work->func : NULL;
+	struct worker *worker;
+
+	if (target_wq->flags & WQ_MEM_RECLAIM)
+		return;
+
+	worker = current_wq_worker();
+
+	WARN_ONCE(current->flags & PF_MEMALLOC,
+		  "workqueue: PF_MEMALLOC task %d(%s) is flushing !WQ_MEM_RECLAIM %s:%pf",
+		  current->pid, current->comm, target_wq->name, target_func);
+	WARN_ONCE(worker && (worker->current_pwq->wq->flags & WQ_MEM_RECLAIM),
+		  "workqueue: WQ_MEM_RECLAIM %s:%pf is flushing !WQ_MEM_RECLAIM %s:%pf",
+		  worker->current_pwq->wq->name, worker->current_func,
+		  target_wq->name, target_func);
+}
+
 struct wq_barrier {
 	struct work_struct	work;
 	struct completion	done;
@@ -2539,6 +2570,8 @@ void flush_workqueue(struct workqueue_st
 		list_add_tail(&this_flusher.list, &wq->flusher_overflow);
 	}
 
+	check_flush_dependency(wq, NULL);
+
 	mutex_unlock(&wq->mutex);
 
 	wait_for_completion(&this_flusher.done);
@@ -2711,6 +2744,8 @@ static bool start_flush_work(struct work
 		pwq = worker->current_pwq;
 	}
 
+	check_flush_dependency(pwq->wq, work);
+
 	insert_wq_barrier(pwq, barr, work, worker);
 	spin_unlock_irq(&pool->lock);
 

  reply	other threads:[~2015-12-03 19:26 UTC|newest]

Thread overview: 51+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-12-03  0:28 [PATCH 1/2] watchdog: introduce touch_softlockup_watchdog_sched() Tejun Heo
2015-12-03  0:28 ` [PATCH 2/2] workqueue: implement lockup detector Tejun Heo
2015-12-03 14:49   ` Tejun Heo
2015-12-03 17:50   ` Don Zickus
2015-12-03 19:43     ` Tejun Heo
2015-12-03 20:12       ` Ulrich Obergfell
2015-12-03 20:54         ` Tejun Heo
2015-12-04  8:02           ` Ingo Molnar
2015-12-04 16:52             ` Don Zickus
2015-12-04 13:19           ` Ulrich Obergfell
2015-12-07 19:06   ` [PATCH v2 " Tejun Heo
2015-12-07 21:38     ` Don Zickus
2015-12-07 21:39       ` Tejun Heo
2015-12-08 16:00         ` Don Zickus
2015-12-08 16:31           ` Tejun Heo
2015-12-03  9:33 ` [PATCH 1/2] watchdog: introduce touch_softlockup_watchdog_sched() Peter Zijlstra
2015-12-03 10:00   ` Peter Zijlstra
2015-12-03 14:48     ` Tejun Heo
2015-12-03 15:04       ` Peter Zijlstra
2015-12-03 15:06         ` Tejun Heo
2015-12-03 19:26           ` Tejun Heo [this message]
2015-12-03 20:43             ` [PATCH] workqueue: warn if memory reclaim tries to flush !WQ_MEM_RECLAIM workqueue Peter Zijlstra
2015-12-03 20:56               ` Tejun Heo
2015-12-03 21:09                 ` Peter Zijlstra
2015-12-03 22:04                   ` Tejun Heo
2015-12-04 12:51                     ` Peter Zijlstra
2015-12-07 15:58             ` Tejun Heo
     [not found]             ` <20151203192616.GJ27463-qYNAdHglDFBN0TnZuCh8vA@public.gmane.org>
2016-01-26 17:38               ` Thierry Reding
2016-01-26 17:38                 ` Thierry Reding
     [not found]                 ` <20160126173843.GA11115-AwZRO8vwLAwmlAP/+Wk3EA@public.gmane.org>
2016-01-28 10:12                   ` Peter Zijlstra
2016-01-28 10:12                     ` Peter Zijlstra
     [not found]                     ` <20160128101210.GC6357-ndre7Fmf5hadTX5a5knrm8zTDFooKrT+cvkQGrU6aU0@public.gmane.org>
2016-01-28 12:47                       ` Thierry Reding
2016-01-28 12:47                         ` Thierry Reding
2016-01-28 12:48                         ` Thierry Reding
2016-01-28 12:48                           ` Thierry Reding
2016-01-29 11:09                       ` Tejun Heo
2016-01-29 11:09                         ` Tejun Heo
2016-01-29 11:09                         ` Tejun Heo
     [not found]                         ` <20160129110941.GK32380-piEFEHQLUPpN0TnZuCh8vA@public.gmane.org>
2016-01-29 15:17                           ` Peter Zijlstra
2016-01-29 15:17                             ` Peter Zijlstra
2016-01-29 15:17                             ` Peter Zijlstra
2016-01-29 18:28                             ` Tejun Heo
2016-01-29 18:28                               ` Tejun Heo
2016-01-29 10:59                   ` [PATCH wq/for-4.5-fixes] workqueue: skip flush dependency checks for legacy workqueues Tejun Heo
2016-01-29 10:59                     ` Tejun Heo
2016-01-29 15:07                     ` Thierry Reding
2016-01-29 18:32                     ` Tejun Heo
     [not found]                     ` <20160129105946.GJ32380-piEFEHQLUPpN0TnZuCh8vA@public.gmane.org>
2016-02-02  6:54                       ` Archit Taneja
2016-02-02  6:54                         ` Archit Taneja
2016-03-10 15:12             ` [PATCH] workqueue: warn if memory reclaim tries to flush !WQ_MEM_RECLAIM workqueue Adrian Hunter
2016-03-11 17:52               ` Tejun Heo

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=20151203192616.GJ27463@mtj.duckdns.org \
    --to=tj@kernel.org \
    --cc=akpm@linux-foundation.org \
    --cc=kernel-team@fb.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=peterz@infradead.org \
    --cc=uobergfe@redhat.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.