mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Joel Fernandes <joelagnelf@nvidia.com>
To: linux-kernel@vger.kernel.org
Cc: "Paul E . McKenney" <paulmck@kernel.org>,
	Frederic Weisbecker <frederic@kernel.org>,
	Neeraj Upadhyay <neeraj.upadhyay@kernel.org>,
	Joel Fernandes <joelagnelf@nvidia.com>,
	Josh Triplett <josh@joshtriplett.org>,
	Boqun Feng <boqun.feng@gmail.com>,
	Steven Rostedt <rostedt@goodmis.org>,
	Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
	Lai Jiangshan <jiangshanlai@gmail.com>,
	Zqiang <qiang.zhang@linux.dev>,
	Uladzislau Rezki <urezki@gmail.com>,
	joel@joelfernandes.org, rcu@vger.kernel.org
Subject: [PATCH RFC 02/14] rcu: Add per-CPU blocked task lists for PREEMPT_RCU
Date: Fri,  2 Jan 2026 19:23:31 -0500	[thread overview]
Message-ID: <20260103002343.6599-3-joelagnelf@nvidia.com> (raw)
In-Reply-To: <20260103002343.6599-1-joelagnelf@nvidia.com>

Add per-CPU tracking of tasks blocked in RCU read-side critical
sections. Each rcu_data gets a blkd_list protected by blkd_lock,
mirroring the rcu_node blkd_tasks list at per-CPU granularity.

Tasks are added on preemption and removed on rcu_read_unlock.
A WARN_ON_ONCE in rcu_gp_init verifies list consistency.

Signed-off-by: Joel Fernandes <joelagnelf@nvidia.com>
---
 include/linux/sched.h    |  4 ++++
 kernel/fork.c            |  4 ++++
 kernel/rcu/Kconfig       | 12 ++++++++++++
 kernel/rcu/tree.c        | 32 ++++++++++++++++++++++++++++++++
 kernel/rcu/tree.h        |  6 ++++++
 kernel/rcu/tree_plugin.h | 21 +++++++++++++++++++++
 6 files changed, 79 insertions(+)

diff --git a/include/linux/sched.h b/include/linux/sched.h
index d395f2810fac..90ce501a568e 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -931,6 +931,10 @@ struct task_struct {
 	union rcu_special		rcu_read_unlock_special;
 	struct list_head		rcu_node_entry;
 	struct rcu_node			*rcu_blocked_node;
+#ifdef CONFIG_RCU_PER_CPU_BLOCKED_LISTS
+	struct list_head		rcu_rdp_entry;
+	int				rcu_blocked_cpu;
+#endif
 #endif /* #ifdef CONFIG_PREEMPT_RCU */
 
 #ifdef CONFIG_TASKS_RCU
diff --git a/kernel/fork.c b/kernel/fork.c
index b1f3915d5f8e..7a5ba2d2c1b5 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -1819,6 +1819,10 @@ static inline void rcu_copy_process(struct task_struct *p)
 	p->rcu_read_unlock_special.s = 0;
 	p->rcu_blocked_node = NULL;
 	INIT_LIST_HEAD(&p->rcu_node_entry);
+#ifdef CONFIG_RCU_PER_CPU_BLOCKED_LISTS
+	INIT_LIST_HEAD(&p->rcu_rdp_entry);
+	p->rcu_blocked_cpu = -1;
+#endif
 #endif /* #ifdef CONFIG_PREEMPT_RCU */
 #ifdef CONFIG_TASKS_RCU
 	p->rcu_tasks_holdout = false;
diff --git a/kernel/rcu/Kconfig b/kernel/rcu/Kconfig
index 4d9b21f69eaa..4bb12f1fed09 100644
--- a/kernel/rcu/Kconfig
+++ b/kernel/rcu/Kconfig
@@ -248,6 +248,18 @@ config RCU_EXP_KTHREAD
 
 	  Accept the default if unsure.
 
+config RCU_PER_CPU_BLOCKED_LISTS
+	bool "Use per-CPU blocked task lists in PREEMPT_RCU"
+	depends on PREEMPT_RCU
+	default n
+	help
+	  Enable per-CPU tracking of tasks blocked in RCU read-side
+	  critical sections. This allows to quickly toggle the feature.
+	  Eventually the config will be removed, in favor of always keeping
+	  the optimization enabled.
+
+	  Accept the default if unsure.
+
 config RCU_NOCB_CPU
 	bool "Offload RCU callback processing from boot-selected CPUs"
 	depends on TREE_RCU
diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
index 293bbd9ac3f4..e2b6a4579086 100644
--- a/kernel/rcu/tree.c
+++ b/kernel/rcu/tree.c
@@ -1809,6 +1809,14 @@ static noinline_for_stack bool rcu_gp_init(void)
 	struct rcu_node *rnp = rcu_get_root();
 	bool start_new_poll;
 	unsigned long old_gp_seq;
+#ifdef CONFIG_RCU_PER_CPU_BLOCKED_LISTS
+	struct task_struct *t_verify;
+	int cpu_verify;
+	int rnp_count;
+	int rdp_total;
+	struct rcu_data *rdp_cpu;
+	struct task_struct *t_rdp;
+#endif
 
 	WRITE_ONCE(rcu_state.gp_activity, jiffies);
 	raw_spin_lock_irq_rcu_node(rnp);
@@ -1891,6 +1899,26 @@ static noinline_for_stack bool rcu_gp_init(void)
 		 */
 		arch_spin_lock(&rcu_state.ofl_lock);
 		raw_spin_lock_rcu_node(rnp);
+#ifdef CONFIG_RCU_PER_CPU_BLOCKED_LISTS
+		/*
+		 * Verify rdp lists consistent with rnp list. Since the unlock
+		 * path removes from rdp before rnp, we can have tasks that are
+		 * on rnp but not on rdp (in the middle of being removed).
+		 * Therefore rnp_count >= rdp_total is the expected invariant.
+		 */
+		rnp_count = 0;
+		rdp_total = 0;
+		list_for_each_entry(t_verify, &rnp->blkd_tasks, rcu_node_entry)
+			rnp_count++;
+		for (cpu_verify = rnp->grplo; cpu_verify <= rnp->grphi; cpu_verify++) {
+			rdp_cpu = per_cpu_ptr(&rcu_data, cpu_verify);
+			raw_spin_lock(&rdp_cpu->blkd_lock);
+			list_for_each_entry(t_rdp, &rdp_cpu->blkd_list, rcu_rdp_entry)
+				rdp_total++;
+			raw_spin_unlock(&rdp_cpu->blkd_lock);
+		}
+		WARN_ON_ONCE(rnp_count < rdp_total);
+#endif
 		if (rnp->qsmaskinit == rnp->qsmaskinitnext &&
 		    !rnp->wait_blkd_tasks) {
 			/* Nothing to do on this leaf rcu_node structure. */
@@ -4143,6 +4171,10 @@ rcu_boot_init_percpu_data(int cpu)
 	rdp->rcu_onl_gp_state = RCU_GP_CLEANED;
 	rdp->last_sched_clock = jiffies;
 	rdp->cpu = cpu;
+#ifdef CONFIG_RCU_PER_CPU_BLOCKED_LISTS
+	raw_spin_lock_init(&rdp->blkd_lock);
+	INIT_LIST_HEAD(&rdp->blkd_list);
+#endif
 	rcu_boot_init_nocb_percpu_data(rdp);
 }
 
diff --git a/kernel/rcu/tree.h b/kernel/rcu/tree.h
index b8bbe7960cda..13d5649a80fb 100644
--- a/kernel/rcu/tree.h
+++ b/kernel/rcu/tree.h
@@ -294,6 +294,12 @@ struct rcu_data {
 
 	long lazy_len;			/* Length of buffered lazy callbacks. */
 	int cpu;
+
+#ifdef CONFIG_RCU_PER_CPU_BLOCKED_LISTS
+	/* 8) Per-CPU blocked task tracking. */
+	raw_spinlock_t blkd_lock;	/* Protects blkd_list. */
+	struct list_head blkd_list;	/* Tasks blocked on this CPU. */
+#endif
 };
 
 /* Values for nocb_defer_wakeup field in struct rcu_data. */
diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h
index 73ba5f4a968d..5d2bde19131a 100644
--- a/kernel/rcu/tree_plugin.h
+++ b/kernel/rcu/tree_plugin.h
@@ -338,6 +338,12 @@ void rcu_note_context_switch(bool preempt)
 		raw_spin_lock_rcu_node(rnp);
 		t->rcu_read_unlock_special.b.blocked = true;
 		t->rcu_blocked_node = rnp;
+#ifdef CONFIG_RCU_PER_CPU_BLOCKED_LISTS
+		t->rcu_blocked_cpu = rdp->cpu;
+		raw_spin_lock(&rdp->blkd_lock);
+		list_add(&t->rcu_rdp_entry, &rdp->blkd_list);
+		raw_spin_unlock(&rdp->blkd_lock);
+#endif
 
 		/*
 		 * Verify the CPU's sanity, trace the preemption, and
@@ -485,6 +491,10 @@ rcu_preempt_deferred_qs_irqrestore(struct task_struct *t, unsigned long flags)
 	struct rcu_data *rdp;
 	struct rcu_node *rnp;
 	union rcu_special special;
+#ifdef CONFIG_RCU_PER_CPU_BLOCKED_LISTS
+	int blocked_cpu;
+	struct rcu_data *blocked_rdp;
+#endif
 
 	rdp = this_cpu_ptr(&rcu_data);
 	if (rdp->defer_qs_iw_pending == DEFER_QS_PENDING)
@@ -530,6 +540,17 @@ rcu_preempt_deferred_qs_irqrestore(struct task_struct *t, unsigned long flags)
 		 * to loop.  Retain a WARN_ON_ONCE() out of sheer paranoia.
 		 */
 		rnp = t->rcu_blocked_node;
+#ifdef CONFIG_RCU_PER_CPU_BLOCKED_LISTS
+		/* Remove from per-CPU list if task was added to it. */
+		blocked_cpu = t->rcu_blocked_cpu;
+		if (blocked_cpu != -1) {
+			blocked_rdp = per_cpu_ptr(&rcu_data, blocked_cpu);
+			raw_spin_lock(&blocked_rdp->blkd_lock);
+			list_del_init(&t->rcu_rdp_entry);
+			t->rcu_blocked_cpu = -1;
+			raw_spin_unlock(&blocked_rdp->blkd_lock);
+		}
+#endif
 		raw_spin_lock_rcu_node(rnp); /* irqs already disabled. */
 		WARN_ON_ONCE(rnp != t->rcu_blocked_node);
 		WARN_ON_ONCE(!rcu_is_leaf_node(rnp));
-- 
2.34.1


  parent reply	other threads:[~2026-01-03  0:23 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-01-03  0:23 [PATCH RFC 00/14] rcu: Reduce rnp->lock contention with per-CPU blocked task lists Joel Fernandes
2026-01-03  0:23 ` [PATCH RFC 01/14] rcu: Add WARN_ON_ONCE for blocked flag invariant in exit_rcu() Joel Fernandes
2026-01-05 15:31   ` Steven Rostedt
2026-01-05 15:44     ` Joel Fernandes
2026-01-03  0:23 ` Joel Fernandes [this message]
2026-01-05 15:48   ` [PATCH RFC 02/14] rcu: Add per-CPU blocked task lists for PREEMPT_RCU Steven Rostedt
2026-01-03  0:23 ` [PATCH RFC 03/14] rcu: Early return during unlock for tasks only on per-CPU blocked list Joel Fernandes
2026-01-03  0:23 ` [PATCH RFC 04/14] rcu: Promote blocked tasks from per-CPU to rnp lists Joel Fernandes
2026-01-05 15:59   ` Steven Rostedt
2026-01-09  3:52     ` Joel Fernandes
2026-01-03  0:23 ` [PATCH RFC 05/14] rcu: Promote blocked tasks for expedited GPs Joel Fernandes
2026-01-03  0:23 ` [PATCH RFC 06/14] rcu: Promote per-CPU blocked tasks before checking for blocked readers Joel Fernandes
2026-01-03  0:23 ` [PATCH RFC 07/14] rcu: Promote late-arriving blocked tasks before reporting QS Joel Fernandes
2026-01-03  0:23 ` [PATCH RFC 08/14] rcu: Promote blocked tasks before QS report in force_qs_rnp() Joel Fernandes
2026-01-03  0:23 ` [PATCH RFC 09/14] rcu: Promote blocked tasks before QS report in rcutree_report_cpu_dead() Joel Fernandes
2026-01-03  0:23 ` [PATCH RFC 10/14] rcu: Promote blocked tasks before QS report in rcu_gp_init() Joel Fernandes
2026-01-03  0:23 ` [PATCH RFC 11/14] rcu: Add per-CPU blocked list check in exit_rcu() Joel Fernandes
2026-01-03  0:23 ` [PATCH RFC 12/14] rcu: Skip per-CPU list addition when GP already started Joel Fernandes
2026-01-03  0:23 ` [PATCH RFC 13/14] rcu: Skip rnp addition when no grace period waiting Joel Fernandes
2026-01-03  0:23 ` [PATCH RFC 14/14] rcu: Remove checking of per-cpu blocked list against the node list Joel Fernandes
2026-01-05 16:46 ` [PATCH RFC 00/14] rcu: Reduce rnp->lock contention with per-CPU blocked task lists Paul E. McKenney
2026-01-06  0:55   ` Joel Fernandes
2026-01-06 15:08     ` Joel Fernandes
2026-01-06 19:24       ` Paul E. McKenney
2026-01-06 21:24         ` Joel Fernandes
2026-01-09  2:00           ` Paul E. McKenney
2026-01-06 19:17     ` Paul E. McKenney
2026-01-06 20:19       ` Steven Rostedt
2026-01-06 20:35         ` Paul E. McKenney
2026-01-06 20:49           ` Joel Fernandes
2026-01-09  1:55             ` Paul E. McKenney
2026-01-06 20:40       ` Joel Fernandes
2026-01-09  1:52         ` 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=20260103002343.6599-3-joelagnelf@nvidia.com \
    --to=joelagnelf@nvidia.com \
    --cc=boqun.feng@gmail.com \
    --cc=frederic@kernel.org \
    --cc=jiangshanlai@gmail.com \
    --cc=joel@joelfernandes.org \
    --cc=josh@joshtriplett.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mathieu.desnoyers@efficios.com \
    --cc=neeraj.upadhyay@kernel.org \
    --cc=paulmck@kernel.org \
    --cc=qiang.zhang@linux.dev \
    --cc=rcu@vger.kernel.org \
    --cc=rostedt@goodmis.org \
    --cc=urezki@gmail.com \
    /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®