[PATCH rcu 5/8] rcu: Make preemptible rcu_exp_handler() check idempotency

["Paul E. McKenney" <paulmck@kernel.org>]

Although the non-preemptible implementation of rcu_exp_handler()
contains checks to enforce idempotency, the preemptible version does not.
The reason for this omission is that in preemptible kernels, there is
no reporting of quiescent states from CPU hotplug notifiers, and thus
no need for idempotency.

In theory, anyway.

In practice, accidents happen.  This commit therefore adds checks under
WARN_ON_ONCE() to catch any such accidents.

Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Cc: Frederic Weisbecker <frederic@kernel.org>
---
 kernel/rcu/tree_exp.h | 18 +++++++++++++++---
 1 file changed, 15 insertions(+), 3 deletions(-)

diff --git a/kernel/rcu/tree_exp.h b/kernel/rcu/tree_exp.h
index f3884393b9477..6985c998fe6b1 100644
--- a/kernel/rcu/tree_exp.h
+++ b/kernel/rcu/tree_exp.h
@@ -746,7 +746,19 @@ static void rcu_exp_handler(void *unused)
 	struct task_struct *t = current;
 
 	/*
-	 * First, the common case of not being in an RCU read-side
+	 * First, is there no need for a quiescent state from this CPU,
+	 * or is this CPU already looking for a quiescent state for the
+	 * current grace period?  If either is the case, just leave.
+	 * However, this should not happen due to the preemptible
+	 * sync_sched_exp_online_cleanup() implementation being a no-op,
+	 * so warn if this does happen.
+	 */
+	if (WARN_ON_ONCE(!(READ_ONCE(rnp->expmask) & rdp->grpmask) ||
+			 READ_ONCE(rdp->cpu_no_qs.b.exp)))
+		return;
+
+	/*
+	 * Second, the common case of not being in an RCU read-side
 	 * critical section.  If also enabled or idle, immediately
 	 * report the quiescent state, otherwise defer.
 	 */
@@ -760,7 +772,7 @@ static void rcu_exp_handler(void *unused)
 	}
 
 	/*
-	 * Second, the less-common case of being in an RCU read-side
+	 * Third, the less-common case of being in an RCU read-side
 	 * critical section.  In this case we can count on a future
 	 * rcu_read_unlock().  However, this rcu_read_unlock() might
 	 * execute on some other CPU, but in that case there will be
@@ -781,7 +793,7 @@ static void rcu_exp_handler(void *unused)
 		return;
 	}
 
-	// Finally, negative nesting depth should not happen.
+	// Fourth and finally, negative nesting depth should not happen.
 	WARN_ON_ONCE(1);
 }
 
-- 
2.40.1