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, "Paul E. McKenney" <paulmck@kernel.org>,
	Jens Axboe <axboe@kernel.dk>
Subject: [PATCH rcu 1/9] rcu: Add lockdep_assert_in_rcu_read_lock() and friends
Date: Tue,  4 Jun 2024 15:23:47 -0700	[thread overview]
Message-ID: <20240604222355.2370768-1-paulmck@kernel.org> (raw)
In-Reply-To: <657595c8-e86c-4594-a5b1-3c64a8275607@paulmck-laptop>

There is no direct RCU counterpart to lockdep_assert_irqs_disabled()
and friends.  Although it is possible to construct them, it would
be more convenient to have the following lockdep assertions:

lockdep_assert_in_rcu_read_lock()
lockdep_assert_in_rcu_read_lock_bh()
lockdep_assert_in_rcu_read_lock_sched()
lockdep_assert_in_rcu_reader()

This commit therefore creates them.

Reported-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
---
 include/linux/rcupdate.h | 60 ++++++++++++++++++++++++++++++++++++++++
 1 file changed, 60 insertions(+)

diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index dfd2399f2cde0..8470a85f65634 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -421,11 +421,71 @@ static inline void rcu_preempt_sleep_check(void) { }
 				 "Illegal context switch in RCU-sched read-side critical section"); \
 	} while (0)
 
+// See RCU_LOCKDEP_WARN() for an explanation of the double call to
+// debug_lockdep_rcu_enabled().
+static inline bool lockdep_assert_rcu_helper(bool c)
+{
+	return debug_lockdep_rcu_enabled() &&
+	       (c || !rcu_is_watching() || !rcu_lockdep_current_cpu_online()) &&
+	       debug_lockdep_rcu_enabled();
+}
+
+/**
+ * lockdep_assert_in_rcu_read_lock - WARN if not protected by rcu_read_lock()
+ *
+ * Splats if lockdep is enabled and there is no rcu_read_lock() in effect.
+ */
+#define lockdep_assert_in_rcu_read_lock() \
+	WARN_ON_ONCE(lockdep_assert_rcu_helper(!lock_is_held(&rcu_lock_map)))
+
+/**
+ * lockdep_assert_in_rcu_read_lock_bh - WARN if not protected by rcu_read_lock_bh()
+ *
+ * Splats if lockdep is enabled and there is no rcu_read_lock_bh() in effect.
+ * Note that local_bh_disable() and friends do not suffice here, instead an
+ * actual rcu_read_lock_bh() is required.
+ */
+#define lockdep_assert_in_rcu_read_lock_bh() \
+	WARN_ON_ONCE(lockdep_assert_rcu_helper(!lock_is_held(&rcu_bh_lock_map)))
+
+/**
+ * lockdep_assert_in_rcu_read_lock_sched - WARN if not protected by rcu_read_lock_sched()
+ *
+ * Splats if lockdep is enabled and there is no rcu_read_lock_sched()
+ * in effect.  Note that preempt_disable() and friends do not suffice here,
+ * instead an actual rcu_read_lock_sched() is required.
+ */
+#define lockdep_assert_in_rcu_read_lock_sched() \
+	WARN_ON_ONCE(lockdep_assert_rcu_helper(!lock_is_held(&rcu_sched_lock_map)))
+
+/**
+ * lockdep_assert_in_rcu_reader - WARN if not within some type of RCU reader
+ *
+ * Splats if lockdep is enabled and there is no RCU reader of any
+ * type in effect.  Note that regions of code protected by things like
+ * preempt_disable, local_bh_disable(), and local_irq_disable() all qualify
+ * as RCU readers.
+ *
+ * Note that this will never trigger in PREEMPT_NONE or PREEMPT_VOLUNTARY
+ * kernels that are not also built with PREEMPT_COUNT.  But if you have
+ * lockdep enabled, you might as well also enable PREEMPT_COUNT.
+ */
+#define lockdep_assert_in_rcu_reader()								\
+	WARN_ON_ONCE(lockdep_assert_rcu_helper(!lock_is_held(&rcu_lock_map) &&			\
+					       !lock_is_held(&rcu_bh_lock_map) &&		\
+					       !lock_is_held(&rcu_sched_lock_map) &&		\
+					       preemptible()))
+
 #else /* #ifdef CONFIG_PROVE_RCU */
 
 #define RCU_LOCKDEP_WARN(c, s) do { } while (0 && (c))
 #define rcu_sleep_check() do { } while (0)
 
+#define lockdep_assert_in_rcu_read_lock() do { } while (0)
+#define lockdep_assert_in_rcu_read_lock_bh() do { } while (0)
+#define lockdep_assert_in_rcu_read_lock_sched() do { } while (0)
+#define lockdep_assert_in_rcu_reader() do { } while (0)
+
 #endif /* #else #ifdef CONFIG_PROVE_RCU */
 
 /*
-- 
2.40.1


  reply	other threads:[~2024-06-04 22:23 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-06-04 22:23 [PATCH rcu 0/9] Miscellaneous fixes for v6.11 Paul E. McKenney
2024-06-04 22:23 ` Paul E. McKenney [this message]
2025-02-20 19:38   ` [PATCH rcu 1/9] rcu: Add lockdep_assert_in_rcu_read_lock() and friends Jeff Johnson
2025-02-20 22:04     ` Paul E. McKenney
2025-02-20 23:51       ` Jeff Johnson
2024-06-04 22:23 ` [PATCH rcu 2/9] rcu: Reduce synchronize_rcu() delays when all wait heads are in use Paul E. McKenney
2024-06-05 12:09   ` Frederic Weisbecker
2024-06-05 18:38     ` Paul E. McKenney
2024-06-06  3:46       ` Neeraj Upadhyay
2024-06-06 16:49         ` Paul E. McKenney
2024-06-11 10:12           ` Uladzislau Rezki
2024-06-04 22:23 ` [PATCH rcu 3/9] rcu/tree: Reduce wake up for synchronize_rcu() common case Paul E. McKenney
2024-06-05 16:35   ` Frederic Weisbecker
2024-06-05 18:42     ` Paul E. McKenney
2024-06-06  5:58     ` Neeraj upadhyay
2024-06-06 18:12       ` Paul E. McKenney
2024-06-07  1:51         ` Neeraj upadhyay
2024-06-10 15:12           ` Paul E. McKenney
2024-06-11 13:46             ` Neeraj upadhyay
2024-06-11 16:17               ` Paul E. McKenney
2024-06-04 22:23 ` [PATCH rcu 4/9] rcu: Disable interrupts directly in rcu_gp_init() Paul E. McKenney
2024-06-04 22:23 ` [PATCH rcu 5/9] srcu: Disable interrupts directly in srcu_gp_end() Paul E. McKenney
2024-06-04 22:23 ` [PATCH rcu 6/9] rcu: Add rcutree.nocb_patience_delay to reduce nohz_full OS jitter Paul E. McKenney
2024-06-10  5:05   ` Leonardo Bras
2024-06-10 15:10     ` Paul E. McKenney
2024-07-03 16:21   ` Frederic Weisbecker
2024-07-03 17:25     ` Paul E. McKenney
2024-07-04 22:18       ` Frederic Weisbecker
2024-07-05  0:26         ` Paul E. McKenney
2024-06-04 22:23 ` [PATCH rcu 7/9] MAINTAINERS: Add Uladzislau Rezki as RCU maintainer Paul E. McKenney
2024-06-04 22:23 ` [PATCH rcu 8/9] rcu: Eliminate lockless accesses to rcu_sync->gp_count Paul E. McKenney
2024-06-04 22:23 ` [PATCH rcu 9/9] rcu: Fix rcu_barrier() VS post CPUHP_TEARDOWN_CPU invocation 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=20240604222355.2370768-1-paulmck@kernel.org \
    --to=paulmck@kernel.org \
    --cc=axboe@kernel.dk \
    --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