[tip/core/rcu,03/11] rcu: Use lockdep rather than comment to enforce lock held
diff mbox series

Message ID 20191210042629.3808-3-paulmck@kernel.org
State In Next
Commit 03bd2983d7a9f898fd89f8f7215c3e56732d8ecd
Headers show
Series
  • Preemptible-RCU updates for v5.6
Related show

Commit Message

Paul E. McKenney Dec. 10, 2019, 4:26 a.m. UTC
From: "Paul E. McKenney" <paulmck@kernel.org>

The rcu_preempt_check_blocked_tasks() function has a comment
that states that the rcu_node structure's ->lock must be held,
which might be informative, but which carries little weight if
not read.  This commit therefore removes this comment in favor of
raw_lockdep_assert_held_rcu_node(), which will complain quite
visibly if the required lock is not held.

Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
---
 kernel/rcu/tree_plugin.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

Patch
diff mbox series

diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h
index fe5f448..ed54d36 100644
--- a/kernel/rcu/tree_plugin.h
+++ b/kernel/rcu/tree_plugin.h
@@ -648,8 +648,7 @@  static void rcu_read_unlock_special(struct task_struct *t)
  * Check that the list of blocked tasks for the newly completed grace
  * period is in fact empty.  It is a serious bug to complete a grace
  * period that still has RCU readers blocked!  This function must be
- * invoked -before- updating this rnp's ->gp_seq, and the rnp's ->lock
- * must be held by the caller.
+ * invoked -before- updating this rnp's ->gp_seq.
  *
  * Also, if there are blocked tasks on the list, they automatically
  * block the newly created grace period, so set up ->gp_tasks accordingly.
@@ -659,6 +658,7 @@  static void rcu_preempt_check_blocked_tasks(struct rcu_node *rnp)
 	struct task_struct *t;
 
 	RCU_LOCKDEP_WARN(preemptible(), "rcu_preempt_check_blocked_tasks() invoked with preemption enabled!!!\n");
+	raw_lockdep_assert_held_rcu_node(rnp);
 	if (WARN_ON_ONCE(rcu_preempt_blocked_readers_cgp(rnp)))
 		dump_blkd_tasks(rnp, 10);
 	if (rcu_preempt_has_tasks(rnp) &&