diff mbox series

[3/4] rcu/nocb: Recheck lazy callbacks under the ->nocb_lock from shrinker

Message ID 20230322194456.2331527-4-frederic@kernel.org
State New
Headers
Series rcu/nocb: Shrinker related boring fixes |

Commit Message

Frederic Weisbecker March 22, 2023, 7:44 p.m. UTC 
  The ->lazy_len is only checked locklessly. Recheck again under the
->nocb_lock to avoid spending more time on flushing/waking if not
necessary. The ->lazy_len can still increment concurrently (from 1 to
infinity) but under the ->nocb_lock we at least know for sure if there
are lazy callbacks at all (->lazy_len > 0).

Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
---
 kernel/rcu/tree_nocb.h | 16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

Comments

Paul E. McKenney March 22, 2023, 11:21 p.m. UTC | #1 
On Wed, Mar 22, 2023 at 08:44:55PM +0100, Frederic Weisbecker wrote:
> The ->lazy_len is only checked locklessly. Recheck again under the
> ->nocb_lock to avoid spending more time on flushing/waking if not
> necessary. The ->lazy_len can still increment concurrently (from 1 to
> infinity) but under the ->nocb_lock we at least know for sure if there
> are lazy callbacks at all (->lazy_len > 0).
> 
> Signed-off-by: Frederic Weisbecker <frederic@kernel.org>

This one looks plausible, and might hold the answer to earlier
questions.

							Thanx, Paul

> ---
>  kernel/rcu/tree_nocb.h | 16 ++++++++++++----
>  1 file changed, 12 insertions(+), 4 deletions(-)
> 
> diff --git a/kernel/rcu/tree_nocb.h b/kernel/rcu/tree_nocb.h
> index cb57e8312231..a3dc7465b0b2 100644
> --- a/kernel/rcu/tree_nocb.h
> +++ b/kernel/rcu/tree_nocb.h
> @@ -1350,12 +1350,20 @@ lazy_rcu_shrink_scan(struct shrinker *shrink, struct shrink_control *sc)
>  		if (!rcu_rdp_is_offloaded(rdp))
>  			continue;
>  
> +		if (!READ_ONCE(rdp->lazy_len))
> +			continue;
> +
> +		rcu_nocb_lock_irqsave(rdp, flags);
> +		/*
> +		 * Recheck under the nocb lock. Since we are not holding the bypass
> +		 * lock we may still race with increments from the enqueuer but still
> +		 * we know for sure if there is at least one lazy callback.
> +		 */
>  		_count = READ_ONCE(rdp->lazy_len);
> -
> -		if (_count == 0)
> +		if (!_count) {
> +			rcu_nocb_unlock_irqrestore(rdp, flags);
>  			continue;
> -
> -		rcu_nocb_lock_irqsave(rdp, flags);
> +		}
>  		WARN_ON_ONCE(!rcu_nocb_flush_bypass(rdp, NULL, jiffies, false));
>  		rcu_nocb_unlock_irqrestore(rdp, flags);
>  		wake_nocb_gp(rdp, false);
> -- 
> 2.34.1
>
diff mbox series

Patch

diff --git a/kernel/rcu/tree_nocb.h b/kernel/rcu/tree_nocb.h
index cb57e8312231..a3dc7465b0b2 100644
--- a/kernel/rcu/tree_nocb.h
+++ b/kernel/rcu/tree_nocb.h
@@ -1350,12 +1350,20 @@  lazy_rcu_shrink_scan(struct shrinker *shrink, struct shrink_control *sc)
 		if (!rcu_rdp_is_offloaded(rdp))
 			continue;
 
+		if (!READ_ONCE(rdp->lazy_len))
+			continue;
+
+		rcu_nocb_lock_irqsave(rdp, flags);
+		/*
+		 * Recheck under the nocb lock. Since we are not holding the bypass
+		 * lock we may still race with increments from the enqueuer but still
+		 * we know for sure if there is at least one lazy callback.
+		 */
 		_count = READ_ONCE(rdp->lazy_len);
-
-		if (_count == 0)
+		if (!_count) {
+			rcu_nocb_unlock_irqrestore(rdp, flags);
 			continue;
-
-		rcu_nocb_lock_irqsave(rdp, flags);
+		}
 		WARN_ON_ONCE(!rcu_nocb_flush_bypass(rdp, NULL, jiffies, false));
 		rcu_nocb_unlock_irqrestore(rdp, flags);
 		wake_nocb_gp(rdp, false);