The bypass timer calls __call_rcu_nocb_wake() instead of directly calling __wake_nocb_gp(). The only difference here is that rdp->qlen_last_fqs_check gets overridden. But resetting the deferred force quiescent state base shouldn't be relevant for that timer. In fact the bypass queue in question can be for any rdp from the group and not necessarily the rdp leader on which the bypass timer is attached. This commit therefore calls __wake_nocb_gp() directly. This way we don't even need to lock the ->nocb_lock.
Showing the changes to kernel (1 of 1 file). Show all
Unified · Split
| @@ -2015,9 +2015,10 @@ static void do_nocb_bypass_wakeup_timer(struct timer_list *t) · 2014 unmodified lines | |||
| 2015 | struct rcu_data *rdp = from_timer(rdp, t, nocb_bypass_timer); | 2015 | struct rcu_data *rdp = from_timer(rdp, t, nocb_bypass_timer); |
| 2016 | 2016 | ||
| 2017 | trace_rcu_nocb_wake(rcu_state.name, rdp->cpu, TPS("Timer")); | 2017 | trace_rcu_nocb_wake(rcu_state.name, rdp->cpu, TPS("Timer")); |
| 2018 | rcu_nocb_lock_irqsave(rdp, flags); | 2018 | |
| 2019 | raw_spin_lock_irqsave(&rdp->nocb_gp_lock, flags); | ||
| 2019 | smp_mb__after_spinlock(); /* Timer expire before wakeup. */ | 2020 | smp_mb__after_spinlock(); /* Timer expire before wakeup. */ |
| 2020 | __call_rcu_nocb_wake(rdp, true, flags); | 2021 | __wake_nocb_gp(rdp, rdp, false, flags); |
| 2021 | } | 2022 | } |
| 2022 | 2023 | ||
| 2023 | /* | 2024 | /* |