public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: "Paul E. McKenney" <paulmck@kernel.org>
To: rcu@vger.kernel.org
Cc: linux-kernel@vger.kernel.org, kernel-team@meta.com,
	rostedt@goodmis.org, Ankur Arora <ankur.a.arora@oracle.com>,
	"Paul E . McKenney" <paulmck@kernel.org>,
	Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
	Frederic Weisbecker <frederic@kernel.org>
Subject: [PATCH rcu v2] 5/9] rcu: handle quiescent states for PREEMPT_RCU=n, PREEMPT_COUNT=y
Date: Thu, 30 Jan 2025 10:55:18 -0800	[thread overview]
Message-ID: <20250130185522.1652093-5-paulmck@kernel.org> (raw)
In-Reply-To: <9bc6e5fe-d49f-495a-9f55-537ed97a3615@paulmck-laptop>

From: Ankur Arora <ankur.a.arora@oracle.com>

With PREEMPT_RCU=n, cond_resched() provides urgently needed quiescent
states for read-side critical sections via rcu_all_qs().
One reason why this was needed: lacking preempt-count, the tick
handler has no way of knowing whether it is executing in a
read-side critical section or not.

With (PREEMPT_LAZY=y, PREEMPT_DYNAMIC=n), we get (PREEMPT_COUNT=y,
PREEMPT_RCU=n). In this configuration cond_resched() is a stub and
does not provide quiescent states via rcu_all_qs().
(PREEMPT_RCU=y provides this information via rcu_read_unlock() and
its nesting counter.)

So, use the availability of preempt_count() to report quiescent states
in rcu_flavor_sched_clock_irq().

Suggested-by: Paul E. McKenney <paulmck@kernel.org>
Reviewed-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Ankur Arora <ankur.a.arora@oracle.com>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
---
 kernel/rcu/tree_plugin.h | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h
index b025e95ede05..f444e8d3f68b 100644
--- a/kernel/rcu/tree_plugin.h
+++ b/kernel/rcu/tree_plugin.h
@@ -984,13 +984,16 @@ static void rcu_preempt_check_blocked_tasks(struct rcu_node *rnp)
  */
 static void rcu_flavor_sched_clock_irq(int user)
 {
-	if (user || rcu_is_cpu_rrupt_from_idle()) {
+	if (user || rcu_is_cpu_rrupt_from_idle() ||
+	     (IS_ENABLED(CONFIG_PREEMPT_COUNT) &&
+	      (preempt_count() == HARDIRQ_OFFSET))) {
 
 		/*
 		 * Get here if this CPU took its interrupt from user
-		 * mode or from the idle loop, and if this is not a
-		 * nested interrupt.  In this case, the CPU is in
-		 * a quiescent state, so note it.
+		 * mode, from the idle loop without this being a nested
+		 * interrupt, or while not holding the task preempt count
+		 * (with PREEMPT_COUNT=y). In this case, the CPU is in a
+		 * quiescent state, so note it.
 		 *
 		 * No memory barrier is required here because rcu_qs()
 		 * references only CPU-local variables that other CPUs
-- 
2.40.1


  parent reply	other threads:[~2025-01-30 18:55 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-01-16 20:13 [PATCH rcu 0/9] Lazy-preemption-related updates Paul E. McKenney
2025-01-16 20:13 ` [PATCH rcu 1/9] rcu: fix header guard for rcu_all_qs() Paul E. McKenney
2025-01-16 20:13 ` [PATCH rcu 2/9] rcu: rename PREEMPT_AUTO to PREEMPT_LAZY Paul E. McKenney
2025-01-16 20:13 ` [PATCH rcu 3/9] sched: update __cond_resched comment about RCU quiescent states Paul E. McKenney
2025-01-16 20:13 ` [PATCH rcu 4/9] rcu: handle unstable rdp in rcu_read_unlock_strict() Paul E. McKenney
2025-01-16 20:13 ` [PATCH rcu 5/9] rcu: handle quiescent states for PREEMPT_RCU=n, PREEMPT_COUNT=y Paul E. McKenney
2025-01-16 20:13 ` [PATCH rcu 6/9] osnoise: provide quiescent states Paul E. McKenney
2025-01-16 20:13 ` [PATCH rcu 7/9] rcu: limit PREEMPT_RCU configurations Paul E. McKenney
2025-01-16 20:13 ` [PATCH rcu 8/9] rcutorture: Make scenario TREE10 build CONFIG_PREEMPT_LAZY=y Paul E. McKenney
2025-01-16 20:13 ` [PATCH rcu 9/9] rcutorture: Make scenario TREE07 " Paul E. McKenney
2025-01-30 18:55 ` [PATCH rcu 0/9] Lazy-preemption-related updates Paul E. McKenney
2025-01-30 18:55   ` [PATCH rcu v2] 1/9] rcu: fix header guard for rcu_all_qs() Paul E. McKenney
2025-01-30 18:55   ` [PATCH rcu v2] 2/9] rcu: rename PREEMPT_AUTO to PREEMPT_LAZY Paul E. McKenney
2025-01-30 18:55   ` [PATCH rcu v2] 3/9] sched: update __cond_resched comment about RCU quiescent states Paul E. McKenney
2025-01-30 18:55   ` [PATCH rcu v2] 4/9] rcu: handle unstable rdp in rcu_read_unlock_strict() Paul E. McKenney
2025-01-30 18:55   ` Paul E. McKenney [this message]
2025-01-30 18:55   ` [PATCH rcu v2] 6/9] osnoise: provide quiescent states Paul E. McKenney
2025-01-30 18:55   ` [PATCH rcu v2] 7/9] rcu: limit PREEMPT_RCU configurations Paul E. McKenney
2025-01-30 18:55   ` [PATCH rcu v2] 8/9] rcutorture: Make scenario TREE10 build CONFIG_PREEMPT_LAZY=y Paul E. McKenney
2025-01-30 18:55   ` [PATCH rcu v2] 9/9] rcutorture: Make scenario TREE07 " Paul E. McKenney

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=20250130185522.1652093-5-paulmck@kernel.org \
    --to=paulmck@kernel.org \
    --cc=ankur.a.arora@oracle.com \
    --cc=bigeasy@linutronix.de \
    --cc=frederic@kernel.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