Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
From:   Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To:     linux-kernel@vger.kernel.org
Cc:     Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        stable@vger.kernel.org, Gaurav Kohli <gkohli@codeaurora.org>,
        "Peter Zijlstra (Intel)" <peterz@infradead.org>,
        Oleg Nesterov <oleg@redhat.com>,
        Linus Torvalds <torvalds@linux-foundation.org>,
        Thomas Gleixner <tglx@linutronix.de>,
        Ingo Molnar <mingo@kernel.org>,
        Sasha Levin <alexander.levin@microsoft.com>
Subject: [PATCH 3.18 23/85] kthread, sched/wait: Fix kthread_parkme() wait-loop
Date:   Sun,  1 Jul 2018 18:01:41 +0200
Message-Id: <20180701153123.280903344@linuxfoundation.org>
In-Reply-To: <20180701153122.365061142@linuxfoundation.org>
References: <20180701153122.365061142@linuxfoundation.org>
User-Agent: quilt/0.65
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

3.18-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Peter Zijlstra <peterz@infradead.org>

[ Upstream commit 741a76b350897604c48fb12beff1c9b77724dc96 ]

Gaurav reported a problem with __kthread_parkme() where a concurrent
try_to_wake_up() could result in competing stores to ->state which,
when the TASK_PARKED store got lost bad things would happen.

The comment near set_current_state() actually mentions this competing
store, but only mentions the case against TASK_RUNNING. This same
store, with different timing, can happen against a subsequent !RUNNING
store.

This normally is not a problem, because as per that same comment, the
!RUNNING state store is inside a condition based wait-loop:

  for (;;) {
    set_current_state(TASK_UNINTERRUPTIBLE);
    if (!need_sleep)
      break;
    schedule();
  }
  __set_current_state(TASK_RUNNING);

If we loose the (first) TASK_UNINTERRUPTIBLE store to a previous
(concurrent) wakeup, the schedule() will NO-OP and we'll go around the
loop once more.

The problem here is that the TASK_PARKED store is not inside the
KTHREAD_SHOULD_PARK condition wait-loop.

There is a genuine issue with sleeps that do not have a condition;
this is addressed in a subsequent patch.

Reported-by: Gaurav Kohli <gkohli@codeaurora.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 kernel/kthread.c |    7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

--- a/kernel/kthread.c
+++ b/kernel/kthread.c
@@ -156,12 +156,13 @@ void *probe_kthread_data(struct task_str
 
 static void __kthread_parkme(struct kthread *self)
 {
-	__set_current_state(TASK_PARKED);
-	while (test_bit(KTHREAD_SHOULD_PARK, &self->flags)) {
+	for (;;) {
+		set_current_state(TASK_PARKED);
+		if (!test_bit(KTHREAD_SHOULD_PARK, &self->flags))
+			break;
 		if (!test_and_set_bit(KTHREAD_IS_PARKED, &self->flags))
 			complete(&self->parked);
 		schedule();
-		__set_current_state(TASK_PARKED);
 	}
 	clear_bit(KTHREAD_IS_PARKED, &self->flags);
 	__set_current_state(TASK_RUNNING);