From aaf2bc50df1f4bfc6857fc601fc7b21d5a18c6a1 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Thu, 21 May 2020 22:05:15 +0200 Subject: [PATCH 1/2] rcu: Abstract out rcu_irq_enter_check_tick() from rcu_nmi_enter() There will likely be exception handlers that can sleep, which rules out the usual approach of invoking rcu_nmi_enter() on entry and also rcu_nmi_exit() on all exit paths. However, the alternative approach of just not calling anything can prevent RCU from coaxing quiescent states from nohz_full CPUs that are looping in the kernel: RCU must instead IPI them explicitly. It would be better to enable the scheduler tick on such CPUs to interact with RCU in a lighter-weight manner, and this enabling is one of the things that rcu_nmi_enter() currently does. What is needed is something that helps RCU coax quiescent states while not preventing subsequent sleeps. This commit therefore splits out the nohz_full scheduler-tick enabling from the rest of the rcu_nmi_enter() logic into a new function named rcu_irq_enter_check_tick(). [ tglx: Renamed the function and made it a nop when context tracking is off ] [ mingo: Fixed a CONFIG_NO_HZ_FULL assumption, harmonized and fixed all the comment blocks and cleaned up rcu_nmi_enter()/exit() definitions. ] Suggested-by: Andy Lutomirski Signed-off-by: Paul E. McKenney Signed-off-by: Thomas Gleixner Signed-off-by: Ingo Molnar Link: https://lore.kernel.org/r/20200521202116.996113173@linutronix.de --- include/linux/hardirq.h | 31 +++++++++------- kernel/rcu/tree.c | 82 +++++++++++++++++++++++++++++++---------- 2 files changed, 80 insertions(+), 33 deletions(-) diff --git a/include/linux/hardirq.h b/include/linux/hardirq.h index 621556efe45f..e07cf853aa16 100644 --- a/include/linux/hardirq.h +++ b/include/linux/hardirq.h @@ -2,31 +2,28 @@ #ifndef LINUX_HARDIRQ_H #define LINUX_HARDIRQ_H +#include #include #include #include #include #include - extern void synchronize_irq(unsigned int irq); extern bool synchronize_hardirq(unsigned int irq); -#if defined(CONFIG_TINY_RCU) - -static inline void rcu_nmi_enter(void) -{ -} - -static inline void rcu_nmi_exit(void) -{ -} - +#ifdef CONFIG_NO_HZ_FULL +void __rcu_irq_enter_check_tick(void); #else -extern void rcu_nmi_enter(void); -extern void rcu_nmi_exit(void); +static inline void __rcu_irq_enter_check_tick(void) { } #endif +static __always_inline void rcu_irq_enter_check_tick(void) +{ + if (context_tracking_enabled()) + __rcu_irq_enter_check_tick(); +} + /* * It is safe to do non-atomic ops on ->hardirq_context, * because NMI handlers may not preempt and the ops are @@ -65,6 +62,14 @@ extern void irq_exit(void); #define arch_nmi_exit() do { } while (0) #endif +#ifdef CONFIG_TINY_RCU +static inline void rcu_nmi_enter(void) { } +static inline void rcu_nmi_exit(void) { } +#else +extern void rcu_nmi_enter(void); +extern void rcu_nmi_exit(void); +#endif + /* * NMI vs Tracing * -------------- diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 90c8be22d57a..b7f8c494d1d1 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -848,6 +848,67 @@ void noinstr rcu_user_exit(void) { rcu_eqs_exit(1); } + +/** + * __rcu_irq_enter_check_tick - Enable scheduler tick on CPU if RCU needs it. + * + * The scheduler tick is not normally enabled when CPUs enter the kernel + * from nohz_full userspace execution. After all, nohz_full userspace + * execution is an RCU quiescent state and the time executing in the kernel + * is quite short. Except of course when it isn't. And it is not hard to + * cause a large system to spend tens of seconds or even minutes looping + * in the kernel, which can cause a number of problems, include RCU CPU + * stall warnings. + * + * Therefore, if a nohz_full CPU fails to report a quiescent state + * in a timely manner, the RCU grace-period kthread sets that CPU's + * ->rcu_urgent_qs flag with the expectation that the next interrupt or + * exception will invoke this function, which will turn on the scheduler + * tick, which will enable RCU to detect that CPU's quiescent states, + * for example, due to cond_resched() calls in CONFIG_PREEMPT=n kernels. + * The tick will be disabled once a quiescent state is reported for + * this CPU. + * + * Of course, in carefully tuned systems, there might never be an + * interrupt or exception. In that case, the RCU grace-period kthread + * will eventually cause one to happen. However, in less carefully + * controlled environments, this function allows RCU to get what it + * needs without creating otherwise useless interruptions. + */ +void __rcu_irq_enter_check_tick(void) +{ + struct rcu_data *rdp = this_cpu_ptr(&rcu_data); + + // Enabling the tick is unsafe in NMI handlers. + if (WARN_ON_ONCE(in_nmi())) + return; + + RCU_LOCKDEP_WARN(rcu_dynticks_curr_cpu_in_eqs(), + "Illegal rcu_irq_enter_check_tick() from extended quiescent state"); + + if (!tick_nohz_full_cpu(rdp->cpu) || + !READ_ONCE(rdp->rcu_urgent_qs) || + READ_ONCE(rdp->rcu_forced_tick)) { + // RCU doesn't need nohz_full help from this CPU, or it is + // already getting that help. + return; + } + + // We get here only when not in an extended quiescent state and + // from interrupts (as opposed to NMIs). Therefore, (1) RCU is + // already watching and (2) The fact that we are in an interrupt + // handler and that the rcu_node lock is an irq-disabled lock + // prevents self-deadlock. So we can safely recheck under the lock. + // Note that the nohz_full state currently cannot change. + raw_spin_lock_rcu_node(rdp->mynode); + if (rdp->rcu_urgent_qs && !rdp->rcu_forced_tick) { + // A nohz_full CPU is in the kernel and RCU needs a + // quiescent state. Turn on the tick! + WRITE_ONCE(rdp->rcu_forced_tick, true); + tick_dep_set_cpu(rdp->cpu, TICK_DEP_BIT_RCU); + } + raw_spin_unlock_rcu_node(rdp->mynode); +} #endif /* CONFIG_NO_HZ_FULL */ /** @@ -894,26 +955,7 @@ noinstr void rcu_nmi_enter(void) incby = 1; } else if (!in_nmi()) { instrumentation_begin(); - if (tick_nohz_full_cpu(rdp->cpu) && - rdp->dynticks_nmi_nesting == DYNTICK_IRQ_NONIDLE && - READ_ONCE(rdp->rcu_urgent_qs) && - !READ_ONCE(rdp->rcu_forced_tick)) { - // We get here only if we had already exited the - // extended quiescent state and this was an - // interrupt (not an NMI). Therefore, (1) RCU is - // already watching and (2) The fact that we are in - // an interrupt handler and that the rcu_node lock - // is an irq-disabled lock prevents self-deadlock. - // So we can safely recheck under the lock. - raw_spin_lock_rcu_node(rdp->mynode); - if (rdp->rcu_urgent_qs && !rdp->rcu_forced_tick) { - // A nohz_full CPU is in the kernel and RCU - // needs a quiescent state. Turn on the tick! - WRITE_ONCE(rdp->rcu_forced_tick, true); - tick_dep_set_cpu(rdp->cpu, TICK_DEP_BIT_RCU); - } - raw_spin_unlock_rcu_node(rdp->mynode); - } + rcu_irq_enter_check_tick(); instrumentation_end(); } instrumentation_begin(); From 07325d4a90d2d84de45cc07b134fd0f023dbb971 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 21 May 2020 22:05:16 +0200 Subject: [PATCH 2/2] rcu: Provide rcu_irq_exit_check_preempt() Provide a debug check which can be invoked from exception return to kernel mode before an attempt is made to schedule. Warn if RCU is not ready for this. Signed-off-by: Thomas Gleixner Signed-off-by: Ingo Molnar Cc: Paul E. McKenney Link: https://lore.kernel.org/r/20200521202117.089709607@linutronix.de --- include/linux/rcutiny.h | 1 + include/linux/rcutree.h | 6 ++++++ kernel/rcu/tree.c | 18 ++++++++++++++++++ 3 files changed, 25 insertions(+) diff --git a/include/linux/rcutiny.h b/include/linux/rcutiny.h index c869fb20cc51..8512caeb7682 100644 --- a/include/linux/rcutiny.h +++ b/include/linux/rcutiny.h @@ -72,6 +72,7 @@ static inline void rcu_irq_exit_irqson(void) { } static inline void rcu_irq_enter_irqson(void) { } static inline void rcu_irq_exit(void) { } static inline void rcu_irq_exit_preempt(void) { } +static inline void rcu_irq_exit_check_preempt(void) { } static inline void exit_rcu(void) { } static inline bool rcu_preempt_need_deferred_qs(struct task_struct *t) { diff --git a/include/linux/rcutree.h b/include/linux/rcutree.h index 9366fa4d0717..d5cc9d675987 100644 --- a/include/linux/rcutree.h +++ b/include/linux/rcutree.h @@ -51,6 +51,12 @@ void rcu_irq_exit_preempt(void); void rcu_irq_enter_irqson(void); void rcu_irq_exit_irqson(void); +#ifdef CONFIG_PROVE_RCU +void rcu_irq_exit_check_preempt(void); +#else +static inline void rcu_irq_exit_check_preempt(void) { } +#endif + void exit_rcu(void); void rcu_scheduler_starting(void); diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index b7f8c494d1d1..d8e9dbbefcfa 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -765,6 +765,24 @@ void rcu_irq_exit_preempt(void) "RCU in extended quiescent state!"); } +#ifdef CONFIG_PROVE_RCU +/** + * rcu_irq_exit_check_preempt - Validate that scheduling is possible + */ +void rcu_irq_exit_check_preempt(void) +{ + lockdep_assert_irqs_disabled(); + + RCU_LOCKDEP_WARN(__this_cpu_read(rcu_data.dynticks_nesting) <= 0, + "RCU dynticks_nesting counter underflow/zero!"); + RCU_LOCKDEP_WARN(__this_cpu_read(rcu_data.dynticks_nmi_nesting) != + DYNTICK_IRQ_NONIDLE, + "Bad RCU dynticks_nmi_nesting counter\n"); + RCU_LOCKDEP_WARN(rcu_dynticks_curr_cpu_in_eqs(), + "RCU in extended quiescent state!"); +} +#endif /* #ifdef CONFIG_PROVE_RCU */ + /* * Wrapper for rcu_irq_exit() where interrupts are enabled. *