mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: "Paul E. McKenney" <paulmck@kernel.org>
To: frederic@kernel.org, rcu@vger.kernel.org
Cc: linux-kernel@vger.kernel.org, kernel-team@meta.com,
	rostedt@goodmis.org, "Paul E. McKenney" <paulmck@kernel.org>,
	Joel Fernandes <joel@joelfernandes.org>
Subject: [PATCH v2 rcu 3/3] rcu: Finer-grained grace-period-end checks in rcu_dump_cpu_stacks()
Date: Wed, 16 Oct 2024 09:19:31 -0700	[thread overview]
Message-ID: <20241016161931.478592-3-paulmck@kernel.org> (raw)
In-Reply-To: <92193018-8624-495e-a685-320119f78db1@paulmck-laptop>

This commit pushes the grace-period-end checks further down into
rcu_dump_cpu_stacks(), and also uses lockless checks coupled with
finer-grained locking.

The result is that the current leaf rcu_node structure's ->lock is
acquired only if a stack backtrace might be needed from the current CPU,
and is held across only that CPU's backtrace.  As a result, if there are
no stalled CPUs associated with a given rcu_node structure, then its
->lock will not be acquired at all.  On large systems, it is usually
(though not always) the case that a small number of CPUs are stalling
the current grace period, which means that the ->lock need be acquired
only for a small fraction of the rcu_node structures.

Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org>
---
 kernel/rcu/tree_stall.h | 18 +++++++++++-------
 1 file changed, 11 insertions(+), 7 deletions(-)

diff --git a/kernel/rcu/tree_stall.h b/kernel/rcu/tree_stall.h
index b530844becf85..8994391b95c76 100644
--- a/kernel/rcu/tree_stall.h
+++ b/kernel/rcu/tree_stall.h
@@ -342,20 +342,24 @@ static void rcu_dump_cpu_stacks(unsigned long gp_seq)
 	struct rcu_node *rnp;
 
 	rcu_for_each_leaf_node(rnp) {
-		if (gp_seq != data_race(rcu_state.gp_seq)) {
-			pr_err("INFO: Stall ended during stack backtracing.\n");
-			return;
-		}
 		printk_deferred_enter();
-		raw_spin_lock_irqsave_rcu_node(rnp, flags);
-		for_each_leaf_node_possible_cpu(rnp, cpu)
+		for_each_leaf_node_possible_cpu(rnp, cpu) {
+			if (gp_seq != data_race(rcu_state.gp_seq)) {
+				printk_deferred_exit();
+				pr_err("INFO: Stall ended during stack backtracing.\n");
+				return;
+			}
+			if (!(data_race(rnp->qsmask) & leaf_node_cpu_bit(rnp, cpu)))
+				continue;
+			raw_spin_lock_irqsave_rcu_node(rnp, flags);
 			if (rnp->qsmask & leaf_node_cpu_bit(rnp, cpu)) {
 				if (cpu_is_offline(cpu))
 					pr_err("Offline CPU %d blocking current GP.\n", cpu);
 				else
 					dump_cpu_task(cpu);
+			raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 			}
-		raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
+		}
 		printk_deferred_exit();
 	}
 }
-- 
2.40.1


  parent reply	other threads:[~2024-10-16 16:19 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-10-09 18:05 [PATCH rcu 0/3] RCU CPU stall-warning changes for v6.13 Paul E. McKenney
2024-10-09 18:05 ` [PATCH rcu 1/3] rcu: Delete unused rcu_gp_might_be_stalled() function Paul E. McKenney
2024-10-09 18:05 ` [PATCH rcu 2/3] rcu: Stop stall warning from dumping stacks if grace period ends Paul E. McKenney
2024-10-15 18:48   ` Joel Fernandes
2024-10-09 18:05 ` [PATCH rcu 3/3] rcu: Finer-grained grace-period-end checks in rcu_dump_cpu_stacks() Paul E. McKenney
2024-10-15 18:49 ` [PATCH rcu 0/3] RCU CPU stall-warning changes for v6.13 Joel Fernandes
2024-10-15 23:02   ` Paul E. McKenney
2024-10-16  0:01     ` Joel Fernandes
2024-10-16 16:18 ` [PATCH v2 " Paul E. McKenney
2024-10-16 16:19   ` [PATCH v2 rcu 1/3] rcu: Delete unused rcu_gp_might_be_stalled() function Paul E. McKenney
2024-10-16 16:19   ` [PATCH v2 rcu 2/3] rcu: Stop stall warning from dumping stacks if grace period ends Paul E. McKenney
2024-10-16 16:19   ` Paul E. McKenney [this message]
2024-10-29  0:22     ` [PATCH v3 rcu 3/3] rcu: Finer-grained grace-period-end checks in rcu_dump_cpu_stacks() Paul E. McKenney
2024-10-29  2:20       ` Cheng-Jui Wang (王正睿)
2024-10-29 16:29         ` Paul E. McKenney
2024-10-30  3:55           ` Cheng-Jui Wang (王正睿)
2024-10-30 13:54             ` Paul E. McKenney
2024-10-30 20:12               ` Doug Anderson
2024-10-30 23:26                 ` Paul E. McKenney
2024-10-31  0:21                   ` Doug Anderson
2024-10-31  5:03                     ` Paul E. McKenney
2024-10-31 21:27                       ` Doug Anderson
2024-11-01  1:44                         ` Cheng-Jui Wang (王正睿)
2024-11-01 13:55                           ` Paul E. McKenney
2025-10-30  8:30                             ` Tze-nan Wu (吳澤南)
2024-11-01  7:41               ` Cheng-Jui Wang (王正睿)
2024-11-01 13:59                 ` Paul E. McKenney
2024-10-18 21:49   ` [PATCH v2 rcu 0/3] RCU CPU stall-warning changes for v6.13 Frederic Weisbecker

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=20241016161931.478592-3-paulmck@kernel.org \
    --to=paulmck@kernel.org \
    --cc=frederic@kernel.org \
    --cc=joel@joelfernandes.org \
    --cc=kernel-team@meta.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=rcu@vger.kernel.org \
    --cc=rostedt@goodmis.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

all inboxes | Powered by JetHome®