[RFC PATCH 09/12] rcu: Set SEGCBLIST_SOFTIRQ_ONLY at the very last stage of de-offloading

[Frederic Weisbecker]

Set SEGCBLIST_SOFTIRQ_ONLY once everything is settled. After that, the
callbacks are handled locklessly and locally.

Inspired-by: Paul E. McKenney <paulmck@xxxxxxxxxx>
Signed-off-by: Frederic Weisbecker <frederic@xxxxxxxxxx>
Cc: Paul E. McKenney <paulmck@xxxxxxxxxx>
Cc: Josh Triplett <josh@xxxxxxxxxxxxxxxx>
Cc: Steven Rostedt <rostedt@xxxxxxxxxxx>
Cc: Mathieu Desnoyers <mathieu.desnoyers@xxxxxxxxxxxx>
Cc: Lai Jiangshan <jiangshanlai@xxxxxxxxx>
Cc: Joel Fernandes <joel@xxxxxxxxxxxxxxxxx>
---
 kernel/rcu/tree_plugin.h | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h
index cee15841c178..ef09897c8f2c 100644
--- a/kernel/rcu/tree_plugin.h
+++ b/kernel/rcu/tree_plugin.h
@@ -2312,7 +2312,14 @@ static void __rcu_nocb_rdp_deoffload(struct rcu_data *rdp)
 	 * enqueued on bypass.
 	 */
 	rcu_nocb_flush_bypass(rdp, NULL, jiffies);
-	rcu_nocb_unlock_irqrestore(rdp, flags);
+	rcu_segcblist_set_flags(cblist, SEGCBLIST_SOFTIRQ_ONLY);
+	/*
+	 * With SEGCBLIST_SOFTIRQ_ONLY, we can't use
+	 * rcu_nocb_unlock_irqrestore() anymore. Theoretically we
+	 * could set SEGCBLIST_SOFTIRQ_ONLY with cb unlocked and IRQs
+	 * disabled now, but let's be paranoid.
+	 */
+	raw_spin_unlock_irqrestore(&rdp->nocb_lock, flags);
 }
 
 static long rcu_nocb_rdp_deoffload(void *arg)
-- 
2.28.0