Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1458548 > unrolled thread

[PATCH RESEND v4] locking/pvqspinlock: Fix double hash race

Back to article view | Back to linux.kernel

Contents

  [PATCH RESEND v4] locking/pvqspinlock: Fix double hash race Wanpeng Li <kernellwp@gmail.com> - 2016-08-09 11:40 +0200
    Re: [PATCH RESEND v4] locking/pvqspinlock: Fix double hash race Peter Zijlstra <peterz@infradead.org> - 2016-08-09 12:50 +0200
      Re: [PATCH RESEND v4] locking/pvqspinlock: Fix double hash race Wanpeng Li <kernellwp@gmail.com> - 2016-08-09 13:40 +0200

#1458548 — [PATCH RESEND v4] locking/pvqspinlock: Fix double hash race

From: Wanpeng Li <wanpeng.li@hotmail.com>

When the lock holder vCPU is racing with the queue head vCPU:

lock holder vCPU             queue head vCPU
=====================        ==================

node->locked = 1;
<preemption>                 READ_ONCE(node->locked)
   ...                       pv_wait_head_or_lock():
                               SPIN_THRESHOLD loop;
                               pv_hash();
                               lock->locked = _Q_SLOW_VAL;
                               node->state  = vcpu_hashed;
pv_kick_node():
  cmpxchg(node->state,
     vcpu_halted, vcpu_hashed);
  lock->locked = _Q_SLOW_VAL;
  pv_hash();

With preemption at the right moment, it is possible that both the
lock holder and queue head vCPUs can be racing to set node->state
which can result in hash entry race. Making sure the state is never 
set to vcpu_halted will prevent this racing from happening.
 
This patch fix it by setting vcpu_hashed after we did all hash thing.

Acked-by: Waiman Long <Waiman.Long@hpe.com>
Reviewed-by: Davidlohr Bueso <dave@stgolabs.net>
Reviewed-by: Pan Xinhui <xinhui.pan@linux.vnet.ibm.com>
Cc: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Waiman Long <Waiman.Long@hpe.com>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
---
v3 -> v4:
 * update patch subject
 * add code comments
v2 -> v3:
 * fix typo in patch description
v1 -> v2:
 * adjust patch description

 kernel/locking/qspinlock_paravirt.h | 23 ++++++++++++++++++++++-
 1 file changed, 22 insertions(+), 1 deletion(-)

diff --git a/kernel/locking/qspinlock_paravirt.h b/kernel/locking/qspinlock_paravirt.h
index 21ede57..ca96db4 100644
--- a/kernel/locking/qspinlock_paravirt.h
+++ b/kernel/locking/qspinlock_paravirt.h
@@ -450,7 +450,28 @@ pv_wait_head_or_lock(struct qspinlock *lock, struct mcs_spinlock *node)
 				goto gotlock;
 			}
 		}
-		WRITE_ONCE(pn->state, vcpu_halted);
+		/*
+		 * lock holder vCPU             queue head vCPU
+		 * ----------------             ---------------
+		 * node->locked = 1;
+		 * <preemption>                 READ_ONCE(node->locked)
+		 *    ...                       pv_wait_head_or_lock():
+		 *                                SPIN_THRESHOLD loop;
+		 *                                pv_hash();
+		 *                                lock->locked = _Q_SLOW_VAL;
+		 *                                node->state  = vcpu_hashed;
+		 * pv_kick_node():
+		 *   cmpxchg(node->state,
+		 *      vcpu_halted, vcpu_hashed);
+		 *   lock->locked = _Q_SLOW_VAL;
+		 *   pv_hash();
+		 *
+		 * With preemption at the right moment, it is possible that both the
+		 * lock holder and queue head vCPUs can be racing to set node->state.
+		 * Making sure the state is never set to vcpu_halted will prevent this
+		 * racing from happening.
+		 */
+		WRITE_ONCE(pn->state, vcpu_hashed);
 		qstat_inc(qstat_pv_wait_head, true);
 		qstat_inc(qstat_pv_wait_again, waitcnt);
 		pv_wait(&l->locked, _Q_SLOW_VAL);
-- 
2.1.0

[toc] | [next] | [standalone]

#1458595

On Tue, Aug 09, 2016 at 05:37:47PM +0800, Wanpeng Li wrote:
> From: Wanpeng Li <wanpeng.li@hotmail.com>
> 
> When the lock holder vCPU is racing with the queue head vCPU:
> 
> lock holder vCPU             queue head vCPU
> =====================        ==================
> 
> node->locked = 1;
> <preemption>                 READ_ONCE(node->locked)
>    ...                       pv_wait_head_or_lock():
>                                SPIN_THRESHOLD loop;
>                                pv_hash();
>                                lock->locked = _Q_SLOW_VAL;
>                                node->state  = vcpu_hashed;
> pv_kick_node():
>   cmpxchg(node->state,
>      vcpu_halted, vcpu_hashed);
>   lock->locked = _Q_SLOW_VAL;
>   pv_hash();

So here the example is 'wrong' in that it doesn't illustrate the fail
case, namely having vcpu_halted win while we're hashed.

> +++ b/kernel/locking/qspinlock_paravirt.h
> @@ -450,7 +450,28 @@ pv_wait_head_or_lock(struct qspinlock *lock, struct mcs_spinlock *node)
>  				goto gotlock;
>  			}
>  		}
> -		WRITE_ONCE(pn->state, vcpu_halted);
> +		/*
> +		 * lock holder vCPU             queue head vCPU
> +		 * ----------------             ---------------
> +		 * node->locked = 1;
> +		 * <preemption>                 READ_ONCE(node->locked)
> +		 *    ...                       pv_wait_head_or_lock():
> +		 *                                SPIN_THRESHOLD loop;
> +		 *                                pv_hash();
> +		 *                                lock->locked = _Q_SLOW_VAL;
> +		 *                                node->state  = vcpu_hashed;
> +		 * pv_kick_node():
> +		 *   cmpxchg(node->state,
> +		 *      vcpu_halted, vcpu_hashed);
> +		 *   lock->locked = _Q_SLOW_VAL;
> +		 *   pv_hash();
> +		 *
> +		 * With preemption at the right moment, it is possible that both the
> +		 * lock holder and queue head vCPUs can be racing to set node->state.
> +		 * Making sure the state is never set to vcpu_halted will prevent this
> +		 * racing from happening.
> +		 */
> +		WRITE_ONCE(pn->state, vcpu_hashed);
>  		qstat_inc(qstat_pv_wait_head, true);
>  		qstat_inc(qstat_pv_wait_again, waitcnt);
>  		pv_wait(&l->locked, _Q_SLOW_VAL);

And I completely fail to see the point of this comment, still.

Yes, if we would have used vcpu_halted, there would be a problem, but we
don't so there isn't. What does this comment tell us about the current
code?

In any case, I have an older version (possibly v1) queued up. That fixes
the bug.

[toc] | [prev] | [next] | [standalone]

#1458629

2016-08-09 18:49 GMT+08:00 Peter Zijlstra <peterz@infradead.org>:
[...]
> In any case, I have an older version (possibly v1) queued up. That fixes
> the bug.

Thanks. :)

Regards,
Wanpeng Li

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web