Explain why we need to disable softirqs while processing callbacks in an offline fashion. The subtle reason doesn't want to be forgotten. Reported-by: Boqun Feng <boqun.feng@xxxxxxxxx> Reported-by: Paul E. McKenney <paulmck@xxxxxxxxxx> Cc: Josh Triplett <josh@xxxxxxxxxxxxxxxx> Cc: Lai Jiangshan <jiangshanlai@xxxxxxxxx> Cc: Joel Fernandes <joel@xxxxxxxxxxxxxxxxx> Cc: Neeraj Upadhyay <neeraju@xxxxxxxxxxxxxx> Cc: Boqun Feng <boqun.feng@xxxxxxxxx> Signed-off-by: Frederic Weisbecker <frederic@xxxxxxxxxx> --- kernel/rcu/tree_plugin.h | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h index a44f80d7661b..dcfae03eb9e9 100644 --- a/kernel/rcu/tree_plugin.h +++ b/kernel/rcu/tree_plugin.h @@ -2235,6 +2235,12 @@ static void nocb_cb_wait(struct rcu_data *rdp) local_irq_save(flags); rcu_momentary_dyntick_idle(); local_irq_restore(flags); + /* + * While transitioning to/from NOCB mode, a CPU might execute the same + * callback concurrently if it requeues itself and the softirq interrupts + * the offloaded callback processing. Make sure we disable BH to prevent + * from that. + */ local_bh_disable(); rcu_do_batch(rdp); local_bh_enable(); -- 2.25.1
