From 924df8a0117aa2725750f1ec4d282867534d9a89 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Wed, 29 Oct 2014 15:37:58 -0700 Subject: rcu: Fix invoke_rcu_callbacks() comment Despite what the comment says, it is only softirqs that are disabled, not interrupts. This commit therefore fixes the comment. Signed-off-by: Paul E. McKenney --- kernel/rcu/tree.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/rcu/tree.c') diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 4c106fcc0d54..9c25b99e2c33 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -2601,7 +2601,7 @@ static void rcu_process_callbacks(struct softirq_action *unused) * Schedule RCU callback invocation. If the specified type of RCU * does not support RCU priority boosting, just do a direct call, * otherwise wake up the per-CPU kernel kthread. Note that because we - * are running on the current CPU with interrupts disabled, the + * are running on the current CPU with softirqs disabled, the * rcu_cpu_kthread_task cannot disappear out from under us. */ static void invoke_rcu_callbacks(struct rcu_state *rsp, struct rcu_data *rdp) -- cgit v1.2.2 From 5f6130fa52ee0df0b1da4518c5bbef42bcfe7d83 Mon Sep 17 00:00:00 2001 From: Lai Jiangshan Date: Tue, 9 Dec 2014 17:53:34 +0800 Subject: tiny_rcu: Directly force QS when call_rcu_[bh|sched]() on idle_task For RCU in UP, context-switch = QS = GP, thus we can force a context-switch when any call_rcu_[bh|sched]() is happened on idle_task. After doing so, rcu_idle/irq_enter/exit() are useless, so we can simply make these functions empty. More important, this change does not change the functionality logically. Note: raise_softirq(RCU_SOFTIRQ)/rcu_sched_qs() in rcu_idle_enter() and outmost rcu_irq_exit() will have to wake up the ksoftirqd (due to in_interrupt() == 0). Before this patch After this patch: call_rcu_sched() in idle; call_rcu_sched() in idle set resched do other stuffs; do other stuffs outmost rcu_irq_exit() outmost rcu_irq_exit() (empty function) (or rcu_idle_enter()) (or rcu_idle_enter(), also empty function) start to resched. (see above) rcu_sched_qs() rcu_sched_qs() QS,and GP and advance cb QS,and GP and advance cb wake up the ksoftirqd wake up the ksoftirqd set resched resched to ksoftirqd (or other) resched to ksoftirqd (or other) These two code patches are almost the same. Size changed after patched: size kernel/rcu/tiny-old.o kernel/rcu/tiny-patched.o text data bss dec hex filename 3449 206 8 3663 e4f kernel/rcu/tiny-old.o 2406 144 8 2558 9fe kernel/rcu/tiny-patched.o Signed-off-by: Lai Jiangshan Signed-off-by: Paul E. McKenney --- kernel/rcu/tree.c | 6 ------ 1 file changed, 6 deletions(-) (limited to 'kernel/rcu/tree.c') diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 9c25b99e2c33..203b50d7ecbd 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -934,12 +934,6 @@ static int dyntick_save_progress_counter(struct rcu_data *rdp, } } -/* - * This function really isn't for public consumption, but RCU is special in - * that context switches can allow the state machine to make progress. - */ -extern void resched_cpu(int cpu); - /* * Return true if the specified CPU has passed through a quiescent * state by virtue of being in or having passed through an dynticks -- cgit v1.2.2 From 41050a009640f9f330a5b916563ca7faf853a98c Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Thu, 18 Dec 2014 12:31:27 -0800 Subject: rcu: Fix rcu_barrier() race that could result in too-short wait The rcu_barrier() no-callbacks check for no-CBs CPUs has race conditions. It checks a given CPU's lists of callbacks, and if all three no-CBs lists are empty, ignores that CPU. However, these three lists could potentially be empty even when callbacks are present if the check executed just as the callbacks were being moved from one list to another. It turns out that recent versions of rcutorture can spot this race. This commit plugs this hole by consolidating the per-list counts of no-CBs callbacks into a single count, which is incremented before the corresponding callback is posted and after it is invoked. Then rcu_barrier() checks this single count to reliably determine whether the corresponding CPU has no-CBs callbacks. Signed-off-by: Paul E. McKenney --- kernel/rcu/tree.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel/rcu/tree.c') diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 203b50d7ecbd..5b9f3b972a79 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -3344,6 +3344,7 @@ static void _rcu_barrier(struct rcu_state *rsp) } else { _rcu_barrier_trace(rsp, "OnlineNoCB", cpu, rsp->n_barrier_done); + smp_mb__before_atomic(); atomic_inc(&rsp->barrier_cpu_count); __call_rcu(&rdp->barrier_head, rcu_barrier_callback, rsp, cpu, 0); -- cgit v1.2.2 From 8af3a5e78cfb63abe8813743946b7bd5a8a3134c Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 31 Oct 2014 11:22:37 -0700 Subject: rcu: Abstract rcu_cleanup_dead_rnp() from rcu_cleanup_dead_cpu() This commit abstracts rcu_cleanup_dead_rnp() from rcu_cleanup_dead_cpu() in preparation for the rework of RCU priority boosting. This new function will be invoked from rcu_read_unlock_special() in the reworked scheme, which is why rcu_cleanup_dead_rnp() assumes that the leaf rcu_node structure's ->qsmaskinit field has already been updated. Signed-off-by: Paul E. McKenney --- kernel/rcu/tree.c | 67 +++++++++++++++++++++++++++++++++++++++---------------- 1 file changed, 48 insertions(+), 19 deletions(-) (limited to 'kernel/rcu/tree.c') diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 4c106fcc0d54..75c6b3301abb 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -2226,6 +2226,46 @@ static void rcu_cleanup_dying_cpu(struct rcu_state *rsp) TPS("cpuofl")); } +/* + * All CPUs for the specified rcu_node structure have gone offline, + * and all tasks that were preempted within an RCU read-side critical + * section while running on one of those CPUs have since exited their RCU + * read-side critical section. Some other CPU is reporting this fact with + * the specified rcu_node structure's ->lock held and interrupts disabled. + * This function therefore goes up the tree of rcu_node structures, + * clearing the corresponding bits in the ->qsmaskinit fields. Note that + * the leaf rcu_node structure's ->qsmaskinit field has already been + * updated + * + * This function does check that the specified rcu_node structure has + * all CPUs offline and no blocked tasks, so it is OK to invoke it + * prematurely. That said, invoking it after the fact will cost you + * a needless lock acquisition. So once it has done its work, don't + * invoke it again. + */ +static void rcu_cleanup_dead_rnp(struct rcu_node *rnp_leaf) +{ + long mask; + struct rcu_node *rnp = rnp_leaf; + + if (rnp->qsmaskinit || rcu_preempt_has_tasks(rnp)) + return; + for (;;) { + mask = rnp->grpmask; + rnp = rnp->parent; + if (!rnp) + break; + raw_spin_lock(&rnp->lock); /* irqs already disabled. */ + smp_mb__after_unlock_lock(); /* GP memory ordering. */ + rnp->qsmaskinit &= ~mask; + if (rnp->qsmaskinit) { + raw_spin_unlock(&rnp->lock); /* irqs remain disabled. */ + return; + } + raw_spin_unlock(&rnp->lock); /* irqs remain disabled. */ + } +} + /* * The CPU has been completely removed, and some other CPU is reporting * this fact from process context. Do the remainder of the cleanup, @@ -2236,7 +2276,6 @@ static void rcu_cleanup_dying_cpu(struct rcu_state *rsp) static void rcu_cleanup_dead_cpu(int cpu, struct rcu_state *rsp) { unsigned long flags; - unsigned long mask; int need_report = 0; struct rcu_data *rdp = per_cpu_ptr(rsp->rda, cpu); struct rcu_node *rnp = rdp->mynode; /* Outgoing CPU's rdp & rnp. */ @@ -2252,24 +2291,14 @@ static void rcu_cleanup_dead_cpu(int cpu, struct rcu_state *rsp) rcu_send_cbs_to_orphanage(cpu, rsp, rnp, rdp); rcu_adopt_orphan_cbs(rsp, flags); - /* Remove the outgoing CPU from the masks in the rcu_node hierarchy. */ - mask = rdp->grpmask; /* rnp->grplo is constant. */ - do { - raw_spin_lock(&rnp->lock); /* irqs already disabled. */ - smp_mb__after_unlock_lock(); - rnp->qsmaskinit &= ~mask; - if (rnp->qsmaskinit != 0) { - if (rnp != rdp->mynode) - raw_spin_unlock(&rnp->lock); /* irqs remain disabled. */ - break; - } - if (rnp == rdp->mynode) - need_report = rcu_preempt_offline_tasks(rsp, rnp, rdp); - else - raw_spin_unlock(&rnp->lock); /* irqs remain disabled. */ - mask = rnp->grpmask; - rnp = rnp->parent; - } while (rnp != NULL); + /* Remove outgoing CPU from mask in the leaf rcu_node structure. */ + raw_spin_lock(&rnp->lock); /* irqs already disabled. */ + smp_mb__after_unlock_lock(); /* Enforce GP memory-order guarantee. */ + rnp->qsmaskinit &= ~rdp->grpmask; + if (rnp->qsmaskinit == 0) { + need_report = rcu_preempt_offline_tasks(rsp, rnp, rdp); + rcu_cleanup_dead_rnp(rnp); + } /* * We still hold the leaf rcu_node structure lock here, and -- cgit v1.2.2 From b6a932d1d9840727eee619d455bdeeedaa205be9 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 31 Oct 2014 12:05:04 -0700 Subject: rcu: Make rcu_read_unlock_special() propagate ->qsmaskinit bit clearing This commit causes rcu_read_unlock_special() to propagate ->qsmaskinit bit clearing up the rcu_node tree once a given rcu_node structure's blkd_tasks list becomes empty. This is the final commit in preparation for the rework of RCU priority boosting: It enables preempted tasks to remain queued on their rcu_node structure even after all of that rcu_node structure's CPUs have gone offline. Signed-off-by: Paul E. McKenney --- kernel/rcu/tree.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'kernel/rcu/tree.c') diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 75c6b3301abb..6625a1b5d9a1 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -2329,6 +2329,10 @@ static void rcu_cleanup_dying_cpu(struct rcu_state *rsp) { } +static void __maybe_unused rcu_cleanup_dead_rnp(struct rcu_node *rnp_leaf) +{ +} + static void rcu_cleanup_dead_cpu(int cpu, struct rcu_state *rsp) { } -- cgit v1.2.2 From d19fb8d1f3f66cc342d30aa48f090c70afb753ed Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 31 Oct 2014 12:56:16 -0700 Subject: rcu: Don't migrate blocked tasks even if all corresponding CPUs offline When the last CPU associated with a given leaf rcu_node structure goes offline, something must be done about the tasks queued on that rcu_node structure. Each of these tasks has been preempted on one of the leaf rcu_node structure's CPUs while in an RCU read-side critical section that it have not yet exited. Handling these tasks is the job of rcu_preempt_offline_tasks(), which migrates them from the leaf rcu_node structure to the root rcu_node structure. Unfortunately, this migration has to be done one task at a time because each tasks allegiance must be shifted from the original leaf rcu_node to the root, so that future attempts to deal with these tasks will acquire the root rcu_node structure's ->lock rather than that of the leaf. Worse yet, this migration must be done with interrupts disabled, which is not so good for realtime response, especially given that there is no bound on the number of tasks on a given rcu_node structure's list. (OK, OK, there is a bound, it is just that it is unreasonably large, especially on 64-bit systems.) This was not considered a problem back when rcu_preempt_offline_tasks() was first written because realtime systems were assumed not to do CPU-hotplug operations while real-time applications were running. This assumption has proved of dubious validity given that people are starting to run multiple realtime applications on a single SMP system and that it is common practice to offline then online a CPU before starting its real-time application in order to clear extraneous processing off of that CPU. So we now need CPU hotplug operations to avoid undue latencies. This commit therefore avoids migrating these tasks, instead letting them be dequeued one by one from the original leaf rcu_node structure by rcu_read_unlock_special(). This means that the clearing of bits from the upper-level rcu_node structures must be deferred until the last such task has been dequeued, because otherwise subsequent grace periods won't wait on them. This commit has the beneficial side effect of simplifying the CPU-hotplug code for TREE_PREEMPT_RCU, especially in CONFIG_RCU_BOOST builds. Signed-off-by: Paul E. McKenney --- kernel/rcu/tree.c | 20 ++------------------ 1 file changed, 2 insertions(+), 18 deletions(-) (limited to 'kernel/rcu/tree.c') diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 6625a1b5d9a1..84f16cf05991 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -2276,7 +2276,6 @@ static void rcu_cleanup_dead_rnp(struct rcu_node *rnp_leaf) static void rcu_cleanup_dead_cpu(int cpu, struct rcu_state *rsp) { unsigned long flags; - int need_report = 0; struct rcu_data *rdp = per_cpu_ptr(rsp->rda, cpu); struct rcu_node *rnp = rdp->mynode; /* Outgoing CPU's rdp & rnp. */ @@ -2295,25 +2294,10 @@ static void rcu_cleanup_dead_cpu(int cpu, struct rcu_state *rsp) raw_spin_lock(&rnp->lock); /* irqs already disabled. */ smp_mb__after_unlock_lock(); /* Enforce GP memory-order guarantee. */ rnp->qsmaskinit &= ~rdp->grpmask; - if (rnp->qsmaskinit == 0) { - need_report = rcu_preempt_offline_tasks(rsp, rnp, rdp); + if (rnp->qsmaskinit == 0 && !rcu_preempt_has_tasks(rnp)) rcu_cleanup_dead_rnp(rnp); - } - - /* - * We still hold the leaf rcu_node structure lock here, and - * irqs are still disabled. The reason for this subterfuge is - * because invoking rcu_report_unblock_qs_rnp() with ->orphan_lock - * held leads to deadlock. - */ raw_spin_unlock(&rsp->orphan_lock); /* irqs remain disabled. */ - rnp = rdp->mynode; - if (need_report & RCU_OFL_TASKS_NORM_GP) - rcu_report_unblock_qs_rnp(rnp, flags); - else - raw_spin_unlock_irqrestore(&rnp->lock, flags); - if (need_report & RCU_OFL_TASKS_EXP_GP) - rcu_report_exp_rnp(rsp, rnp, true); + raw_spin_unlock_irqrestore(&rnp->lock, flags); WARN_ONCE(rdp->qlen != 0 || rdp->nxtlist != NULL, "rcu_cleanup_dead_cpu: Callbacks on offline CPU %d: qlen=%lu, nxtlist=%p\n", cpu, rdp->qlen, rdp->nxtlist); -- cgit v1.2.2 From a8f4cbadfb5c5f7f5d9eef70821276a4e2e9289d Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 31 Oct 2014 13:48:41 -0700 Subject: rcu: Shorten irq-disable region in rcu_cleanup_dead_cpu() Now that we are not migrating callbacks, there is no need to hold the ->orphan_lock across the the ->qsmaskinit bit-clearing process. This commit therefore releases ->orphan_lock immediately after adopting the orphaned RCU callbacks. Signed-off-by: Paul E. McKenney --- kernel/rcu/tree.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel/rcu/tree.c') diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 84f16cf05991..990b406faf4e 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -2289,14 +2289,14 @@ static void rcu_cleanup_dead_cpu(int cpu, struct rcu_state *rsp) /* Orphan the dead CPU's callbacks, and adopt them if appropriate. */ rcu_send_cbs_to_orphanage(cpu, rsp, rnp, rdp); rcu_adopt_orphan_cbs(rsp, flags); + raw_spin_unlock_irqrestore(&rsp->orphan_lock, flags); /* Remove outgoing CPU from mask in the leaf rcu_node structure. */ - raw_spin_lock(&rnp->lock); /* irqs already disabled. */ + raw_spin_lock_irqsave(&rnp->lock, flags); smp_mb__after_unlock_lock(); /* Enforce GP memory-order guarantee. */ rnp->qsmaskinit &= ~rdp->grpmask; if (rnp->qsmaskinit == 0 && !rcu_preempt_has_tasks(rnp)) rcu_cleanup_dead_rnp(rnp); - raw_spin_unlock(&rsp->orphan_lock); /* irqs remain disabled. */ raw_spin_unlock_irqrestore(&rnp->lock, flags); WARN_ONCE(rdp->qlen != 0 || rdp->nxtlist != NULL, "rcu_cleanup_dead_cpu: Callbacks on offline CPU %d: qlen=%lu, nxtlist=%p\n", -- cgit v1.2.2 From 1be0085b515e786e962b9f2d03616209eb6eb9a7 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Mon, 3 Nov 2014 18:20:20 -0800 Subject: rcu: Don't initiate RCU priority boosting on root rcu_node Because there is no longer any preempted tasks on the root rcu_node, and because there is no longer ever an rcub kthread for the root rcu_node, this commit drops the code in force_qs_rnp() that attempts to awaken the non-existent root rcub kthread. This is strictly a performance enhancement, removing a root rcu_node ->lock acquisition and release along with some tests in rcu_initiate_boost(), ending with the test that notes that there is no rcub kthread. Signed-off-by: Paul E. McKenney --- kernel/rcu/tree.c | 6 ------ 1 file changed, 6 deletions(-) (limited to 'kernel/rcu/tree.c') diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 990b406faf4e..5a0a4c969d38 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -2513,12 +2513,6 @@ static void force_qs_rnp(struct rcu_state *rsp, } raw_spin_unlock_irqrestore(&rnp->lock, flags); } - rnp = rcu_get_root(rsp); - if (rnp->qsmask == 0) { - raw_spin_lock_irqsave(&rnp->lock, flags); - smp_mb__after_unlock_lock(); - rcu_initiate_boost(rnp, flags); /* releases rnp->lock. */ - } } /* -- cgit v1.2.2 From 3ba4d0e09bf965297e97adf195e0ea246cfe5c74 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Wed, 12 Nov 2014 09:57:51 -0800 Subject: rcu: Note quiescent state when CPU goes offline The rcu_cleanup_dead_cpu() function (called after a CPU has gone completely offline) has not reported a quiescent state because there was probably at least one synchronize_rcu() between the time the CPU went offline and the CPU_DEAD notifier, and this would have detected the CPU's offline state via quiescent-state forcing. However, the plan is for CPUs to take themselves offline, at which point it makes sense for them to report their own quiescent state. This commit makes this change in preparation for the new CPU-hotplug setup. Signed-off-by: Paul E. McKenney --- kernel/rcu/tree.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/rcu/tree.c') diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 5a0a4c969d38..24d3c67f7be7 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -2297,7 +2297,7 @@ static void rcu_cleanup_dead_cpu(int cpu, struct rcu_state *rsp) rnp->qsmaskinit &= ~rdp->grpmask; if (rnp->qsmaskinit == 0 && !rcu_preempt_has_tasks(rnp)) rcu_cleanup_dead_rnp(rnp); - raw_spin_unlock_irqrestore(&rnp->lock, flags); + rcu_report_qs_rnp(rdp->grpmask, rsp, rnp, flags); /* Rlses rnp->lock. */ WARN_ONCE(rdp->qlen != 0 || rdp->nxtlist != NULL, "rcu_cleanup_dead_cpu: Callbacks on offline CPU %d: qlen=%lu, nxtlist=%p\n", cpu, rdp->qlen, rdp->nxtlist); -- cgit v1.2.2 From 6cd534ef8bfe422a5a847b953d1039841509c374 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Wed, 10 Dec 2014 07:45:13 -0800 Subject: rcu: Don't scan root rcu_node structure for stalled tasks Now that blocked tasks are no longer migrated to the root rcu_node structure, there is no need to scan the root rcu_node structure for blocked tasks stalling the current grace period. This commit therefore removes this scan. Signed-off-by: Paul E. McKenney --- kernel/rcu/tree.c | 9 --------- 1 file changed, 9 deletions(-) (limited to 'kernel/rcu/tree.c') diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 24d3c67f7be7..4ed2c2842103 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -1107,15 +1107,6 @@ static void print_other_cpu_stall(struct rcu_state *rsp) raw_spin_unlock_irqrestore(&rnp->lock, flags); } - /* - * Now rat on any tasks that got kicked up to the root rcu_node - * due to CPU offlining. - */ - rnp = rcu_get_root(rsp); - raw_spin_lock_irqsave(&rnp->lock, flags); - ndetected += rcu_print_task_stall(rnp); - raw_spin_unlock_irqrestore(&rnp->lock, flags); - print_cpu_stall_info_end(); for_each_possible_cpu(cpu) totqlen += per_cpu_ptr(rsp->rda, cpu)->qlen; -- cgit v1.2.2 From ab954c167ed9ac107a8beb1d6e1745ae698a3421 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Wed, 17 Dec 2014 18:25:59 -0800 Subject: rcu: Remove redundant callback-list initialization The RCU callback lists are initialized in both rcu_boot_init_percpu_data() and rcu_init_percpu_data(). The former is intended for initializing immutable data, so this commit removes the initialization from rcu_boot_init_percpu_data() and leaves it in rcu_init_percpu_data(). This change prepares for permitting callbacks to be queued very early in boot. Signed-off-by: Paul E. McKenney --- kernel/rcu/tree.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'kernel/rcu/tree.c') diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 4ed2c2842103..febd0f72a3c9 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -3419,9 +3419,6 @@ rcu_boot_init_percpu_data(int cpu, struct rcu_state *rsp) /* Set up local state, ensuring consistent view of global state. */ raw_spin_lock_irqsave(&rnp->lock, flags); rdp->grpmask = 1UL << (cpu - rdp->mynode->grplo); - init_callback_list(rdp); - rdp->qlen_lazy = 0; - ACCESS_ONCE(rdp->qlen) = 0; rdp->dynticks = &per_cpu(rcu_dynticks, cpu); WARN_ON_ONCE(rdp->dynticks->dynticks_nesting != DYNTICK_TASK_EXIT_IDLE); WARN_ON_ONCE(atomic_read(&rdp->dynticks->dynticks) != 1); -- cgit v1.2.2 From fc908ed33e7c1428f799abb12399f906da03b397 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Mon, 8 Dec 2014 09:57:48 -0800 Subject: rcu: Make RCU_CPU_STALL_INFO include number of fqs attempts One way that an RCU CPU stall warning can happen is if the grace-period kthread is not allowed to execute. One proxy for this kthread's forward progress is the number of force-quiescent-state (fqs) scans. This commit therefore adds the number of fqs scans to the RCU CPU stall warning printouts when CONFIG_RCU_CPU_STALL_INFO=y. Signed-off-by: Paul E. McKenney --- kernel/rcu/tree.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel/rcu/tree.c') diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 4c106fcc0d54..654b15be1e36 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -1043,6 +1043,7 @@ static void record_gp_stall_check_time(struct rcu_state *rsp) j1 = rcu_jiffies_till_stall_check(); ACCESS_ONCE(rsp->jiffies_stall) = j + j1; rsp->jiffies_resched = j + j1 / 2; + rsp->n_force_qs_gpstart = ACCESS_ONCE(rsp->n_force_qs); } /* -- cgit v1.2.2 From 6ccd2ecd422644277b7d8b37222e3af3f43ea9ae Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Thu, 11 Dec 2014 10:20:59 -0800 Subject: rcu: Improve diagnostics for spurious RCU CPU stall warnings The current RCU CPU stall warning code will print "Stall ended before state dump start" any time that the stall-warning code is triggered on a CPU that has already reported a quiescent state for the current grace period and if all quiescent states have been reported for the current grace period. However, a true stall can result in these symptoms, for example, by preventing RCU's grace-period kthreads from ever running This commit therefore checks for this condition, reporting the end of the stall only if one of the grace-period counters has actually advanced. Otherwise, it reports the last time that the grace-period kthread made meaningful progress. (In normal situations, the grace-period kthread should make meaningful progress at least every jiffies_till_next_fqs jiffies.) Reported-by: Miroslav Benes Signed-off-by: Paul E. McKenney Tested-by: Miroslav Benes --- kernel/rcu/tree.c | 32 +++++++++++++++++++++++++++----- 1 file changed, 27 insertions(+), 5 deletions(-) (limited to 'kernel/rcu/tree.c') diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 654b15be1e36..a2ceb66bcd67 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -1066,11 +1066,13 @@ static void rcu_dump_cpu_stacks(struct rcu_state *rsp) } } -static void print_other_cpu_stall(struct rcu_state *rsp) +static void print_other_cpu_stall(struct rcu_state *rsp, unsigned long gpnum) { int cpu; long delta; unsigned long flags; + unsigned long gpa; + unsigned long j; int ndetected = 0; struct rcu_node *rnp = rcu_get_root(rsp); long totqlen = 0; @@ -1123,10 +1125,22 @@ static void print_other_cpu_stall(struct rcu_state *rsp) pr_cont("(detected by %d, t=%ld jiffies, g=%ld, c=%ld, q=%lu)\n", smp_processor_id(), (long)(jiffies - rsp->gp_start), (long)rsp->gpnum, (long)rsp->completed, totqlen); - if (ndetected == 0) - pr_err("INFO: Stall ended before state dump start\n"); - else + if (ndetected) { rcu_dump_cpu_stacks(rsp); + } else { + if (ACCESS_ONCE(rsp->gpnum) != gpnum || + ACCESS_ONCE(rsp->completed) == gpnum) { + pr_err("INFO: Stall ended before state dump start\n"); + } else { + j = jiffies; + gpa = ACCESS_ONCE(rsp->gp_activity); + pr_err("All QSes seen, last %s kthread activity %ld (%ld-%ld), jiffies_till_next_fqs=%ld\n", + rsp->name, j - gpa, j, gpa, + jiffies_till_next_fqs); + /* In this case, the current CPU might be at fault. */ + sched_show_task(current); + } + } /* Complain about tasks blocking the grace period. */ @@ -1226,7 +1240,7 @@ static void check_cpu_stall(struct rcu_state *rsp, struct rcu_data *rdp) ULONG_CMP_GE(j, js + RCU_STALL_RAT_DELAY)) { /* They had a few time units to dump stack, so complain. */ - print_other_cpu_stall(rsp); + print_other_cpu_stall(rsp, gpnum); } } @@ -1622,6 +1636,7 @@ static int rcu_gp_init(struct rcu_state *rsp) struct rcu_data *rdp; struct rcu_node *rnp = rcu_get_root(rsp); + ACCESS_ONCE(rsp->gp_activity) = jiffies; rcu_bind_gp_kthread(); raw_spin_lock_irq(&rnp->lock); smp_mb__after_unlock_lock(); @@ -1682,6 +1697,7 @@ static int rcu_gp_init(struct rcu_state *rsp) rnp->grphi, rnp->qsmask); raw_spin_unlock_irq(&rnp->lock); cond_resched_rcu_qs(); + ACCESS_ONCE(rsp->gp_activity) = jiffies; } mutex_unlock(&rsp->onoff_mutex); @@ -1698,6 +1714,7 @@ static int rcu_gp_fqs(struct rcu_state *rsp, int fqs_state_in) unsigned long maxj; struct rcu_node *rnp = rcu_get_root(rsp); + ACCESS_ONCE(rsp->gp_activity) = jiffies; rsp->n_force_qs++; if (fqs_state == RCU_SAVE_DYNTICK) { /* Collect dyntick-idle snapshots. */ @@ -1736,6 +1753,7 @@ static void rcu_gp_cleanup(struct rcu_state *rsp) struct rcu_data *rdp; struct rcu_node *rnp = rcu_get_root(rsp); + ACCESS_ONCE(rsp->gp_activity) = jiffies; raw_spin_lock_irq(&rnp->lock); smp_mb__after_unlock_lock(); gp_duration = jiffies - rsp->gp_start; @@ -1772,6 +1790,7 @@ static void rcu_gp_cleanup(struct rcu_state *rsp) nocb += rcu_future_gp_cleanup(rsp, rnp); raw_spin_unlock_irq(&rnp->lock); cond_resched_rcu_qs(); + ACCESS_ONCE(rsp->gp_activity) = jiffies; } rnp = rcu_get_root(rsp); raw_spin_lock_irq(&rnp->lock); @@ -1821,6 +1840,7 @@ static int __noreturn rcu_gp_kthread(void *arg) if (rcu_gp_init(rsp)) break; cond_resched_rcu_qs(); + ACCESS_ONCE(rsp->gp_activity) = jiffies; WARN_ON(signal_pending(current)); trace_rcu_grace_period(rsp->name, ACCESS_ONCE(rsp->gpnum), @@ -1864,9 +1884,11 @@ static int __noreturn rcu_gp_kthread(void *arg) ACCESS_ONCE(rsp->gpnum), TPS("fqsend")); cond_resched_rcu_qs(); + ACCESS_ONCE(rsp->gp_activity) = jiffies; } else { /* Deal with stray signal. */ cond_resched_rcu_qs(); + ACCESS_ONCE(rsp->gp_activity) = jiffies; WARN_ON(signal_pending(current)); trace_rcu_grace_period(rsp->name, ACCESS_ONCE(rsp->gpnum), -- cgit v1.2.2 From e3663b1024d1f94688e5233440ad67a9bc10b94e Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Mon, 8 Dec 2014 20:26:55 -0800 Subject: rcu: Handle gpnum/completed wrap while dyntick idle Subtle race conditions can result if a CPU stays in dyntick-idle mode long enough for the ->gpnum and ->completed fields to wrap. For example, consider the following sequence of events: o CPU 1 encounters a quiescent state while waiting for grace period 5 to complete, but then enters dyntick-idle mode. o While CPU 1 is in dyntick-idle mode, the grace-period counters wrap around so that the grace period number is now 4. o Just as CPU 1 exits dyntick-idle mode, grace period 4 completes and grace period 5 begins. o The quiescent state that CPU 1 passed through during the old grace period 5 looks like it applies to the new grace period 5. Therefore, the new grace period 5 completes without CPU 1 having passed through a quiescent state. This could clearly be a fatal surprise to any long-running RCU read-side critical section that happened to be running on CPU 1 at the time. At one time, this was not a problem, given that it takes significant time for the grace-period counters to overflow even on 32-bit systems. However, with the advent of NO_HZ_FULL and SMP embedded systems, arbitrarily long idle periods are now becoming quite feasible. It is therefore time to close this race. This commit therefore avoids this race condition by having the quiescent-state forcing code detect when a CPU is falling too far behind, and setting a new rcu_data field ->gpwrap when this happens. Whenever this new ->gpwrap field is set, the CPU's ->gpnum and ->completed fields are known to be untrustworthy, and can be ignored, along with any associated quiescent states. Signed-off-by: Paul E. McKenney --- kernel/rcu/tree.c | 17 ++++++++++++----- 1 file changed, 12 insertions(+), 5 deletions(-) (limited to 'kernel/rcu/tree.c') diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index a2ceb66bcd67..5987fdc85fc4 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -930,6 +930,9 @@ static int dyntick_save_progress_counter(struct rcu_data *rdp, trace_rcu_fqs(rdp->rsp->name, rdp->gpnum, rdp->cpu, TPS("dti")); return 1; } else { + if (ULONG_CMP_LT(ACCESS_ONCE(rdp->gpnum) + ULONG_MAX / 4, + rdp->mynode->gpnum)) + ACCESS_ONCE(rdp->gpwrap) = true; return 0; } } @@ -1577,7 +1580,8 @@ static bool __note_gp_changes(struct rcu_state *rsp, struct rcu_node *rnp, bool ret; /* Handle the ends of any preceding grace periods first. */ - if (rdp->completed == rnp->completed) { + if (rdp->completed == rnp->completed && + !unlikely(ACCESS_ONCE(rdp->gpwrap))) { /* No grace period end, so just accelerate recent callbacks. */ ret = rcu_accelerate_cbs(rsp, rnp, rdp); @@ -1592,7 +1596,7 @@ static bool __note_gp_changes(struct rcu_state *rsp, struct rcu_node *rnp, trace_rcu_grace_period(rsp->name, rdp->gpnum, TPS("cpuend")); } - if (rdp->gpnum != rnp->gpnum) { + if (rdp->gpnum != rnp->gpnum || unlikely(ACCESS_ONCE(rdp->gpwrap))) { /* * If the current grace period is waiting for this CPU, * set up to detect a quiescent state, otherwise don't @@ -1603,6 +1607,7 @@ static bool __note_gp_changes(struct rcu_state *rsp, struct rcu_node *rnp, rdp->passed_quiesce = 0; rdp->qs_pending = !!(rnp->qsmask & rdp->grpmask); zero_cpu_stall_ticks(rdp); + ACCESS_ONCE(rdp->gpwrap) = false; } return ret; } @@ -1616,7 +1621,8 @@ static void note_gp_changes(struct rcu_state *rsp, struct rcu_data *rdp) local_irq_save(flags); rnp = rdp->mynode; if ((rdp->gpnum == ACCESS_ONCE(rnp->gpnum) && - rdp->completed == ACCESS_ONCE(rnp->completed)) || /* w/out lock. */ + rdp->completed == ACCESS_ONCE(rnp->completed) && + !unlikely(ACCESS_ONCE(rdp->gpwrap))) || /* w/out lock. */ !raw_spin_trylock(&rnp->lock)) { /* irqs already off, so later. */ local_irq_restore(flags); return; @@ -2066,7 +2072,7 @@ rcu_report_qs_rdp(int cpu, struct rcu_state *rsp, struct rcu_data *rdp) raw_spin_lock_irqsave(&rnp->lock, flags); smp_mb__after_unlock_lock(); if (rdp->passed_quiesce == 0 || rdp->gpnum != rnp->gpnum || - rnp->completed == rnp->gpnum) { + rnp->completed == rnp->gpnum || rdp->gpwrap) { /* * The grace period in which this quiescent state was @@ -3190,7 +3196,8 @@ static int __rcu_pending(struct rcu_state *rsp, struct rcu_data *rdp) } /* Has a new RCU grace period started? */ - if (ACCESS_ONCE(rnp->gpnum) != rdp->gpnum) { /* outside lock */ + if (ACCESS_ONCE(rnp->gpnum) != rdp->gpnum || + unlikely(ACCESS_ONCE(rdp->gpwrap))) { /* outside lock */ rdp->n_rp_gp_started++; return 1; } -- cgit v1.2.2 From 9733e4f0a973a354034f5dd603b4142a3095c85f Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 21 Nov 2014 12:49:13 -0800 Subject: rcu: Make _batches_completed() functions return unsigned long Long ago, the various ->completed fields were of type long, but now are unsigned long due to signed-integer-overflow concerns. However, the various _batches_completed() functions remained of type long, even though their only purpose in life is to return the corresponding ->completed field. This patch cleans this up by changing these functions' return types to unsigned long. Signed-off-by: Paul E. McKenney --- kernel/rcu/tree.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel/rcu/tree.c') diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 4c106fcc0d54..e26d78712e16 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -317,7 +317,7 @@ static int rcu_pending(void); /* * Return the number of RCU-sched batches processed thus far for debug & stats. */ -long rcu_batches_completed_sched(void) +unsigned long rcu_batches_completed_sched(void) { return rcu_sched_state.completed; } @@ -326,7 +326,7 @@ EXPORT_SYMBOL_GPL(rcu_batches_completed_sched); /* * Return the number of RCU BH batches processed thus far for debug & stats. */ -long rcu_batches_completed_bh(void) +unsigned long rcu_batches_completed_bh(void) { return rcu_bh_state.completed; } -- cgit v1.2.2 From 917963d0b30f9c4153c372c165178501d97b6b55 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 21 Nov 2014 17:10:16 -0800 Subject: rcutorture: Check from beginning to end of grace period Currently, rcutorture's Reader Batch checks measure from the end of the previous grace period to the end of the current one. This commit tightens up these checks by measuring from the start and end of the same grace period. This involves adding rcu_batches_started() and friends corresponding to the existing rcu_batches_completed() and friends. We leave SRCU alone for the moment, as it does not yet have a way of tracking both ends of its grace periods. Signed-off-by: Paul E. McKenney --- kernel/rcu/tree.c | 40 ++++++++++++++++++++++++++++++++++++++-- 1 file changed, 38 insertions(+), 2 deletions(-) (limited to 'kernel/rcu/tree.c') diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index e26d78712e16..c0faad51ae87 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -315,7 +315,43 @@ static void force_quiescent_state(struct rcu_state *rsp); static int rcu_pending(void); /* - * Return the number of RCU-sched batches processed thus far for debug & stats. + * Return the number of RCU batches started thus far for debug & stats. + */ +unsigned long rcu_batches_started(void) +{ + return rcu_state_p->gpnum; +} +EXPORT_SYMBOL_GPL(rcu_batches_started); + +/* + * Return the number of RCU-sched batches started thus far for debug & stats. + */ +unsigned long rcu_batches_started_sched(void) +{ + return rcu_sched_state.gpnum; +} +EXPORT_SYMBOL_GPL(rcu_batches_started_sched); + +/* + * Return the number of RCU BH batches started thus far for debug & stats. + */ +unsigned long rcu_batches_started_bh(void) +{ + return rcu_bh_state.gpnum; +} +EXPORT_SYMBOL_GPL(rcu_batches_started_bh); + +/* + * Return the number of RCU batches completed thus far for debug & stats. + */ +unsigned long rcu_batches_completed(void) +{ + return rcu_state_p->completed; +} +EXPORT_SYMBOL_GPL(rcu_batches_completed); + +/* + * Return the number of RCU-sched batches completed thus far for debug & stats. */ unsigned long rcu_batches_completed_sched(void) { @@ -324,7 +360,7 @@ unsigned long rcu_batches_completed_sched(void) EXPORT_SYMBOL_GPL(rcu_batches_completed_sched); /* - * Return the number of RCU BH batches processed thus far for debug & stats. + * Return the number of RCU BH batches completed thus far for debug & stats. */ unsigned long rcu_batches_completed_bh(void) { -- cgit v1.2.2 From a94844b22a2e2b9155bbc0878c507850477221c2 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 12 Dec 2014 07:37:48 -0800 Subject: rcu: Optionally run grace-period kthreads at real-time priority Recent testing has shown that under heavy load, running RCU's grace-period kthreads at real-time priority can improve performance (according to 0day test robot) and reduce the incidence of RCU CPU stall warnings. However, most systems do just fine with the default non-realtime priorities for these kthreads, and it does not make sense to expose the entire user base to any risk stemming from this change, given that this change is of use only to a few users running extremely heavy workloads. Therefore, this commit allows users to specify realtime priorities for the grace-period kthreads, but leaves them running SCHED_OTHER by default. The realtime priority may be specified at build time via the RCU_KTHREAD_PRIO Kconfig parameter, or at boot time via the rcutree.kthread_prio parameter. Either way, 0 says to continue the default SCHED_OTHER behavior and values from 1-99 specify that priority of SCHED_FIFO behavior. Note that a value of 0 is not permitted when the RCU_BOOST Kconfig parameter is specified. Signed-off-by: Paul E. McKenney --- kernel/rcu/tree.c | 24 +++++++++++++++++++++++- 1 file changed, 23 insertions(+), 1 deletion(-) (limited to 'kernel/rcu/tree.c') diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 5987fdc85fc4..75ce12316b4c 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -156,6 +156,10 @@ static void rcu_boost_kthread_setaffinity(struct rcu_node *rnp, int outgoingcpu) static void invoke_rcu_core(void); static void invoke_rcu_callbacks(struct rcu_state *rsp, struct rcu_data *rdp); +/* rcuc/rcub kthread realtime priority */ +static int kthread_prio = CONFIG_RCU_KTHREAD_PRIO; +module_param(kthread_prio, int, 0644); + /* * Track the rcutorture test sequence number and the update version * number within a given test. The rcutorture_testseq is incremented @@ -3597,17 +3601,35 @@ static int rcu_pm_notify(struct notifier_block *self, static int __init rcu_spawn_gp_kthread(void) { unsigned long flags; + int kthread_prio_in = kthread_prio; struct rcu_node *rnp; struct rcu_state *rsp; + struct sched_param sp; struct task_struct *t; + /* Force priority into range. */ + if (IS_ENABLED(CONFIG_RCU_BOOST) && kthread_prio < 1) + kthread_prio = 1; + else if (kthread_prio < 0) + kthread_prio = 0; + else if (kthread_prio > 99) + kthread_prio = 99; + if (kthread_prio != kthread_prio_in) + pr_alert("rcu_spawn_gp_kthread(): Limited prio to %d from %d\n", + kthread_prio, kthread_prio_in); + rcu_scheduler_fully_active = 1; for_each_rcu_flavor(rsp) { - t = kthread_run(rcu_gp_kthread, rsp, "%s", rsp->name); + t = kthread_create(rcu_gp_kthread, rsp, "%s", rsp->name); BUG_ON(IS_ERR(t)); rnp = rcu_get_root(rsp); raw_spin_lock_irqsave(&rnp->lock, flags); rsp->gp_kthread = t; + if (kthread_prio) { + sp.sched_priority = kthread_prio; + sched_setscheduler_nocheck(t, SCHED_FIFO, &sp); + } + wake_up_process(t); raw_spin_unlock_irqrestore(&rnp->lock, flags); } rcu_spawn_nocb_kthreads(); -- cgit v1.2.2 From 5cd37193ce8539be1e6ef76be226f4bcc984e0f5 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sat, 13 Dec 2014 20:32:04 -0800 Subject: rcu: Make cond_resched_rcu_qs() apply to normal RCU flavors Although cond_resched_rcu_qs() only applies to TASKS_RCU, it is used in places where it would be useful for it to apply to the normal RCU flavors, rcu_preempt, rcu_sched, and rcu_bh. This is especially the case for workloads that aggressively overload the system, particularly those that generate large numbers of RCU updates on systems running NO_HZ_FULL CPUs. This commit therefore communicates quiescent states from cond_resched_rcu_qs() to the normal RCU flavors. Note that it is unfortunately necessary to leave the old ->passed_quiesce mechanism in place to allow quiescent states that apply to only one flavor to be recorded. (Yes, we could decrement ->rcu_qs_ctr_snap in that case, but that is not so good for debugging of RCU internals.) In addition, if one of the RCU flavor's grace period has stalled, this will invoke rcu_momentary_dyntick_idle(), resulting in a heavy-weight quiescent state visible from other CPUs. Reported-by: Sasha Levin Reported-by: Dave Jones Signed-off-by: Paul E. McKenney [ paulmck: Merge commit from Sasha Levin fixing a bug where __this_cpu() was used in preemptible code. ] --- kernel/rcu/tree.c | 38 +++++++++++++++++++++++++++++++++----- 1 file changed, 33 insertions(+), 5 deletions(-) (limited to 'kernel/rcu/tree.c') diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 75ce12316b4c..cb00e038c2f2 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -219,6 +219,9 @@ static DEFINE_PER_CPU(struct rcu_dynticks, rcu_dynticks) = { #endif /* #ifdef CONFIG_NO_HZ_FULL_SYSIDLE */ }; +DEFINE_PER_CPU_SHARED_ALIGNED(unsigned long, rcu_qs_ctr); +EXPORT_PER_CPU_SYMBOL_GPL(rcu_qs_ctr); + /* * Let the RCU core know that this CPU has gone through the scheduler, * which is a quiescent state. This is called when the need for a @@ -288,6 +291,22 @@ void rcu_note_context_switch(void) } EXPORT_SYMBOL_GPL(rcu_note_context_switch); +/* + * Register a quiesecent state for all RCU flavors. If there is an + * emergency, invoke rcu_momentary_dyntick_idle() to do a heavy-weight + * dyntick-idle quiescent state visible to other CPUs (but only for those + * RCU flavors in desparate need of a quiescent state, which will normally + * be none of them). Either way, do a lightweight quiescent state for + * all RCU flavors. + */ +void rcu_all_qs(void) +{ + if (unlikely(raw_cpu_read(rcu_sched_qs_mask))) + rcu_momentary_dyntick_idle(); + this_cpu_inc(rcu_qs_ctr); +} +EXPORT_SYMBOL_GPL(rcu_all_qs); + static long blimit = 10; /* Maximum callbacks per rcu_do_batch. */ static long qhimark = 10000; /* If this many pending, ignore blimit. */ static long qlowmark = 100; /* Once only this many pending, use blimit. */ @@ -1609,6 +1628,7 @@ static bool __note_gp_changes(struct rcu_state *rsp, struct rcu_node *rnp, rdp->gpnum = rnp->gpnum; trace_rcu_grace_period(rsp->name, rdp->gpnum, TPS("cpustart")); rdp->passed_quiesce = 0; + rdp->rcu_qs_ctr_snap = __this_cpu_read(rcu_qs_ctr); rdp->qs_pending = !!(rnp->qsmask & rdp->grpmask); zero_cpu_stall_ticks(rdp); ACCESS_ONCE(rdp->gpwrap) = false; @@ -2075,8 +2095,10 @@ rcu_report_qs_rdp(int cpu, struct rcu_state *rsp, struct rcu_data *rdp) rnp = rdp->mynode; raw_spin_lock_irqsave(&rnp->lock, flags); smp_mb__after_unlock_lock(); - if (rdp->passed_quiesce == 0 || rdp->gpnum != rnp->gpnum || - rnp->completed == rnp->gpnum || rdp->gpwrap) { + if ((rdp->passed_quiesce == 0 && + rdp->rcu_qs_ctr_snap == __this_cpu_read(rcu_qs_ctr)) || + rdp->gpnum != rnp->gpnum || rnp->completed == rnp->gpnum || + rdp->gpwrap) { /* * The grace period in which this quiescent state was @@ -2085,6 +2107,7 @@ rcu_report_qs_rdp(int cpu, struct rcu_state *rsp, struct rcu_data *rdp) * within the current grace period. */ rdp->passed_quiesce = 0; /* need qs for new gp. */ + rdp->rcu_qs_ctr_snap = __this_cpu_read(rcu_qs_ctr); raw_spin_unlock_irqrestore(&rnp->lock, flags); return; } @@ -2129,7 +2152,8 @@ rcu_check_quiescent_state(struct rcu_state *rsp, struct rcu_data *rdp) * Was there a quiescent state since the beginning of the grace * period? If no, then exit and wait for the next call. */ - if (!rdp->passed_quiesce) + if (!rdp->passed_quiesce && + rdp->rcu_qs_ctr_snap == __this_cpu_read(rcu_qs_ctr)) return; /* @@ -3174,9 +3198,12 @@ static int __rcu_pending(struct rcu_state *rsp, struct rcu_data *rdp) /* Is the RCU core waiting for a quiescent state from this CPU? */ if (rcu_scheduler_fully_active && - rdp->qs_pending && !rdp->passed_quiesce) { + rdp->qs_pending && !rdp->passed_quiesce && + rdp->rcu_qs_ctr_snap == __this_cpu_read(rcu_qs_ctr)) { rdp->n_rp_qs_pending++; - } else if (rdp->qs_pending && rdp->passed_quiesce) { + } else if (rdp->qs_pending && + (rdp->passed_quiesce || + rdp->rcu_qs_ctr_snap != __this_cpu_read(rcu_qs_ctr))) { rdp->n_rp_report_qs++; return 1; } @@ -3510,6 +3537,7 @@ rcu_init_percpu_data(int cpu, struct rcu_state *rsp) rdp->gpnum = rnp->completed; rdp->completed = rnp->completed; rdp->passed_quiesce = 0; + rdp->rcu_qs_ctr_snap = __this_cpu_read(rcu_qs_ctr); rdp->qs_pending = 0; trace_rcu_grace_period(rsp->name, rdp->gpnum, TPS("cpuonl")); } -- cgit v1.2.2 From fb81a44b88e6173ed0f6e9d6a1afa5305fb63f6e Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Wed, 17 Dec 2014 08:35:02 -0800 Subject: rcu: Add GP-kthread-starvation checks to CPU stall warnings This commit adds a message that is printed if the relevant grace-period kthread has not been able to run for the two seconds preceding the stall warning. (The two seconds is double the maximum interval between successive bouts of quiescent-state forcing.) Signed-off-by: Paul E. McKenney --- kernel/rcu/tree.c | 21 ++++++++++++++++++++- 1 file changed, 20 insertions(+), 1 deletion(-) (limited to 'kernel/rcu/tree.c') diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index cb00e038c2f2..e335f33d0b9f 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -1072,6 +1072,21 @@ static void record_gp_stall_check_time(struct rcu_state *rsp) rsp->n_force_qs_gpstart = ACCESS_ONCE(rsp->n_force_qs); } +/* + * Complain about starvation of grace-period kthread. + */ +static void rcu_check_gp_kthread_starvation(struct rcu_state *rsp) +{ + unsigned long gpa; + unsigned long j; + + j = jiffies; + gpa = ACCESS_ONCE(rsp->gp_activity); + if (j - gpa > 2 * HZ) + pr_err("%s kthread starved for %ld jiffies!\n", + rsp->name, j - gpa); +} + /* * Dump stacks of all tasks running on stalled CPUs. */ @@ -1169,9 +1184,10 @@ static void print_other_cpu_stall(struct rcu_state *rsp, unsigned long gpnum) } /* Complain about tasks blocking the grace period. */ - rcu_print_detail_task_stall(rsp); + rcu_check_gp_kthread_starvation(rsp); + force_quiescent_state(rsp); /* Kick them all. */ } @@ -1196,6 +1212,9 @@ static void print_cpu_stall(struct rcu_state *rsp) pr_cont(" (t=%lu jiffies g=%ld c=%ld q=%lu)\n", jiffies - rsp->gp_start, (long)rsp->gpnum, (long)rsp->completed, totqlen); + + rcu_check_gp_kthread_starvation(rsp); + rcu_dump_cpu_stacks(rsp); raw_spin_lock_irqsave(&rnp->lock, flags); -- cgit v1.2.2