diff options
Diffstat (limited to 'kernel')
52 files changed, 2868 insertions, 1533 deletions
diff --git a/kernel/cpu.c b/kernel/cpu.c index 1972b161c61e..82eea9c5af61 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -20,6 +20,7 @@ #include <linux/gfp.h> #include <linux/suspend.h> #include <linux/lockdep.h> +#include <linux/tick.h> #include <trace/events/power.h> #include "smpboot.h" @@ -338,6 +339,8 @@ static int __ref take_cpu_down(void *_param) return err; cpu_notify(CPU_DYING | param->mod, param->hcpu); + /* Give up timekeeping duties */ + tick_handover_do_timer(); /* Park the stopper thread */ kthread_park(current); return 0; @@ -411,10 +414,12 @@ static int __ref _cpu_down(unsigned int cpu, int tasks_frozen) while (!idle_cpu(cpu)) cpu_relax(); + hotplug_cpu__broadcast_tick_pull(cpu); /* This actually kills the CPU. */ __cpu_die(cpu); /* CPU is completely dead: tell everyone. Too late to complain. */ + tick_cleanup_dead_cpu(cpu); cpu_notify_nofail(CPU_DEAD | mod, hcpu); check_for_tasks(cpu); diff --git a/kernel/events/core.c b/kernel/events/core.c index f04daabfd1cf..2fabc0627165 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -3591,7 +3591,7 @@ static void put_event(struct perf_event *event) ctx = perf_event_ctx_lock_nested(event, SINGLE_DEPTH_NESTING); WARN_ON_ONCE(ctx->parent_ctx); perf_remove_from_context(event, true); - mutex_unlock(&ctx->mutex); + perf_event_ctx_unlock(event, ctx); _free_event(event); } @@ -4574,6 +4574,13 @@ static void perf_pending_event(struct irq_work *entry) { struct perf_event *event = container_of(entry, struct perf_event, pending); + int rctx; + + rctx = perf_swevent_get_recursion_context(); + /* + * If we 'fail' here, that's OK, it means recursion is already disabled + * and we won't recurse 'further'. + */ if (event->pending_disable) { event->pending_disable = 0; @@ -4584,6 +4591,9 @@ static void perf_pending_event(struct irq_work *entry) event->pending_wakeup = 0; perf_event_wakeup(event); } + + if (rctx >= 0) + perf_swevent_put_recursion_context(rctx); } /* diff --git a/kernel/futex.c b/kernel/futex.c index 2a5e3830e953..2579e407ff67 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -900,7 +900,7 @@ static int attach_to_pi_owner(u32 uval, union futex_key *key, if (!p) return -ESRCH; - if (!p->mm) { + if (unlikely(p->flags & PF_KTHREAD)) { put_task_struct(p); return -EPERM; } diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c index 6f1c7a566b95..eb9a4ea394ab 100644 --- a/kernel/irq/chip.c +++ b/kernel/irq/chip.c @@ -948,6 +948,22 @@ int irq_chip_retrigger_hierarchy(struct irq_data *data) return -ENOSYS; } + +/** + * irq_chip_set_wake_parent - Set/reset wake-up on the parent interrupt + * @data: Pointer to interrupt specific data + * @on: Whether to set or reset the wake-up capability of this irq + * + * Conditional, as the underlying parent chip might not implement it. + */ +int irq_chip_set_wake_parent(struct irq_data *data, unsigned int on) +{ + data = data->parent_data; + if (data->chip->irq_set_wake) + return data->chip->irq_set_wake(data, on); + + return -ENOSYS; +} #endif /** diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 196a06fbc122..e68932bb308e 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -68,14 +68,20 @@ static void __synchronize_hardirq(struct irq_desc *desc) * Do not use this for shutdown scenarios where you must be sure * that all parts (hardirq and threaded handler) have completed. * + * Returns: false if a threaded handler is active. + * * This function may be called - with care - from IRQ context. */ -void synchronize_hardirq(unsigned int irq) +bool synchronize_hardirq(unsigned int irq) { struct irq_desc *desc = irq_to_desc(irq); - if (desc) + if (desc) { __synchronize_hardirq(desc); + return !atomic_read(&desc->threads_active); + } + + return true; } EXPORT_SYMBOL(synchronize_hardirq); @@ -440,6 +446,32 @@ void disable_irq(unsigned int irq) } EXPORT_SYMBOL(disable_irq); +/** + * disable_hardirq - disables an irq and waits for hardirq completion + * @irq: Interrupt to disable + * + * Disable the selected interrupt line. Enables and Disables are + * nested. + * This function waits for any pending hard IRQ handlers for this + * interrupt to complete before returning. If you use this function while + * holding a resource the hard IRQ handler may need you will deadlock. + * + * When used to optimistically disable an interrupt from atomic context + * the return value must be checked. + * + * Returns: false if a threaded handler is active. + * + * This function may be called - with care - from IRQ context. + */ +bool disable_hardirq(unsigned int irq) +{ + if (!__disable_irq_nosync(irq)) + return synchronize_hardirq(irq); + + return false; +} +EXPORT_SYMBOL_GPL(disable_hardirq); + void __enable_irq(struct irq_desc *desc, unsigned int irq) { switch (desc->depth) { @@ -1474,8 +1506,13 @@ int request_threaded_irq(unsigned int irq, irq_handler_t handler, * otherwise we'll have trouble later trying to figure out * which interrupt is which (messes up the interrupt freeing * logic etc). + * + * Also IRQF_COND_SUSPEND only makes sense for shared interrupts and + * it cannot be set along with IRQF_NO_SUSPEND. */ - if ((irqflags & IRQF_SHARED) && !dev_id) + if (((irqflags & IRQF_SHARED) && !dev_id) || + (!(irqflags & IRQF_SHARED) && (irqflags & IRQF_COND_SUSPEND)) || + ((irqflags & IRQF_NO_SUSPEND) && (irqflags & IRQF_COND_SUSPEND))) return -EINVAL; desc = irq_to_desc(irq); @@ -1761,3 +1798,94 @@ int request_percpu_irq(unsigned int irq, irq_handler_t handler, return retval; } + +/** + * irq_get_irqchip_state - returns the irqchip state of a interrupt. + * @irq: Interrupt line that is forwarded to a VM + * @which: One of IRQCHIP_STATE_* the caller wants to know about + * @state: a pointer to a boolean where the state is to be storeed + * + * This call snapshots the internal irqchip state of an + * interrupt, returning into @state the bit corresponding to + * stage @which + * + * This function should be called with preemption disabled if the + * interrupt controller has per-cpu registers. + */ +int irq_get_irqchip_state(unsigned int irq, enum irqchip_irq_state which, + bool *state) +{ + struct irq_desc *desc; + struct irq_data *data; + struct irq_chip *chip; + unsigned long flags; + int err = -EINVAL; + + desc = irq_get_desc_buslock(irq, &flags, 0); + if (!desc) + return err; + + data = irq_desc_get_irq_data(desc); + + do { + chip = irq_data_get_irq_chip(data); + if (chip->irq_get_irqchip_state) + break; +#ifdef CONFIG_IRQ_DOMAIN_HIERARCHY + data = data->parent_data; +#else + data = NULL; +#endif + } while (data); + + if (data) + err = chip->irq_get_irqchip_state(data, which, state); + + irq_put_desc_busunlock(desc, flags); + return err; +} + +/** + * irq_set_irqchip_state - set the state of a forwarded interrupt. + * @irq: Interrupt line that is forwarded to a VM + * @which: State to be restored (one of IRQCHIP_STATE_*) + * @val: Value corresponding to @which + * + * This call sets the internal irqchip state of an interrupt, + * depending on the value of @which. + * + * This function should be called with preemption disabled if the + * interrupt controller has per-cpu registers. + */ +int irq_set_irqchip_state(unsigned int irq, enum irqchip_irq_state which, + bool val) +{ + struct irq_desc *desc; + struct irq_data *data; + struct irq_chip *chip; + unsigned long flags; + int err = -EINVAL; + + desc = irq_get_desc_buslock(irq, &flags, 0); + if (!desc) + return err; + + data = irq_desc_get_irq_data(desc); + + do { + chip = irq_data_get_irq_chip(data); + if (chip->irq_set_irqchip_state) + break; +#ifdef CONFIG_IRQ_DOMAIN_HIERARCHY + data = data->parent_data; +#else + data = NULL; +#endif + } while (data); + + if (data) + err = chip->irq_set_irqchip_state(data, which, val); + + irq_put_desc_busunlock(desc, flags); + return err; +} diff --git a/kernel/irq/msi.c b/kernel/irq/msi.c index 3e18163f336f..474de5cb394d 100644 --- a/kernel/irq/msi.c +++ b/kernel/irq/msi.c @@ -310,8 +310,15 @@ void msi_domain_free_irqs(struct irq_domain *domain, struct device *dev) struct msi_desc *desc; for_each_msi_entry(desc, dev) { - irq_domain_free_irqs(desc->irq, desc->nvec_used); - desc->irq = 0; + /* + * We might have failed to allocate an MSI early + * enough that there is no IRQ associated to this + * entry. If that's the case, don't do anything. + */ + if (desc->irq) { + irq_domain_free_irqs(desc->irq, desc->nvec_used); + desc->irq = 0; + } } } diff --git a/kernel/irq/pm.c b/kernel/irq/pm.c index 3ca532592704..5204a6d1b985 100644 --- a/kernel/irq/pm.c +++ b/kernel/irq/pm.c @@ -43,9 +43,12 @@ void irq_pm_install_action(struct irq_desc *desc, struct irqaction *action) if (action->flags & IRQF_NO_SUSPEND) desc->no_suspend_depth++; + else if (action->flags & IRQF_COND_SUSPEND) + desc->cond_suspend_depth++; WARN_ON_ONCE(desc->no_suspend_depth && - desc->no_suspend_depth != desc->nr_actions); + (desc->no_suspend_depth + + desc->cond_suspend_depth) != desc->nr_actions); } /* @@ -61,6 +64,8 @@ void irq_pm_remove_action(struct irq_desc *desc, struct irqaction *action) if (action->flags & IRQF_NO_SUSPEND) desc->no_suspend_depth--; + else if (action->flags & IRQF_COND_SUSPEND) + desc->cond_suspend_depth--; } static bool suspend_device_irq(struct irq_desc *desc, int irq) diff --git a/kernel/livepatch/core.c b/kernel/livepatch/core.c index ff7f47d026ac..3f9f1d6b4c2e 100644 --- a/kernel/livepatch/core.c +++ b/kernel/livepatch/core.c @@ -89,16 +89,28 @@ static bool klp_is_object_loaded(struct klp_object *obj) /* sets obj->mod if object is not vmlinux and module is found */ static void klp_find_object_module(struct klp_object *obj) { + struct module *mod; + if (!klp_is_module(obj)) return; mutex_lock(&module_mutex); /* - * We don't need to take a reference on the module here because we have - * the klp_mutex, which is also taken by the module notifier. This - * prevents any module from unloading until we release the klp_mutex. + * We do not want to block removal of patched modules and therefore + * we do not take a reference here. The patches are removed by + * a going module handler instead. + */ + mod = find_module(obj->name); + /* + * Do not mess work of the module coming and going notifiers. + * Note that the patch might still be needed before the going handler + * is called. Module functions can be called even in the GOING state + * until mod->exit() finishes. This is especially important for + * patches that modify semantic of the functions. */ - obj->mod = find_module(obj->name); + if (mod && mod->klp_alive) + obj->mod = mod; + mutex_unlock(&module_mutex); } @@ -248,11 +260,12 @@ static int klp_find_external_symbol(struct module *pmod, const char *name, /* first, check if it's an exported symbol */ preempt_disable(); sym = find_symbol(name, NULL, NULL, true, true); - preempt_enable(); if (sym) { *addr = sym->value; + preempt_enable(); return 0; } + preempt_enable(); /* otherwise check if it's in another .o within the patch module */ return klp_find_object_symbol(pmod->name, name, addr); @@ -314,12 +327,12 @@ static void notrace klp_ftrace_handler(unsigned long ip, rcu_read_lock(); func = list_first_or_null_rcu(&ops->func_stack, struct klp_func, stack_node); - rcu_read_unlock(); - if (WARN_ON_ONCE(!func)) - return; + goto unlock; klp_arch_set_pc(regs, (unsigned long)func->new_func); +unlock: + rcu_read_unlock(); } static int klp_disable_func(struct klp_func *func) @@ -731,7 +744,7 @@ static int klp_init_func(struct klp_object *obj, struct klp_func *func) func->state = KLP_DISABLED; return kobject_init_and_add(&func->kobj, &klp_ktype_func, - obj->kobj, func->old_name); + obj->kobj, "%s", func->old_name); } /* parts of the initialization that is done only when the object is loaded */ @@ -766,6 +779,7 @@ static int klp_init_object(struct klp_patch *patch, struct klp_object *obj) return -EINVAL; obj->state = KLP_DISABLED; + obj->mod = NULL; klp_find_object_module(obj); @@ -807,7 +821,7 @@ static int klp_init_patch(struct klp_patch *patch) patch->state = KLP_DISABLED; ret = kobject_init_and_add(&patch->kobj, &klp_ktype_patch, - klp_root_kobj, patch->mod->name); + klp_root_kobj, "%s", patch->mod->name); if (ret) goto unlock; @@ -960,6 +974,15 @@ static int klp_module_notify(struct notifier_block *nb, unsigned long action, mutex_lock(&klp_mutex); + /* + * Each module has to know that the notifier has been called. + * We never know what module will get patched by a new patch. + */ + if (action == MODULE_STATE_COMING) + mod->klp_alive = true; + else /* MODULE_STATE_GOING */ + mod->klp_alive = false; + list_for_each_entry(patch, &klp_patches, list) { for (obj = patch->objs; obj->funcs; obj++) { if (!klp_is_module(obj) || strcmp(obj->name, mod->name)) diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c index 88d0d4420ad2..ba77ab5f64dd 100644 --- a/kernel/locking/lockdep.c +++ b/kernel/locking/lockdep.c @@ -633,7 +633,7 @@ static int count_matching_names(struct lock_class *new_class) if (!new_class->name) return 0; - list_for_each_entry(class, &all_lock_classes, lock_entry) { + list_for_each_entry_rcu(class, &all_lock_classes, lock_entry) { if (new_class->key - new_class->subclass == class->key) return class->name_version; if (class->name && !strcmp(class->name, new_class->name)) @@ -700,10 +700,12 @@ look_up_lock_class(struct lockdep_map *lock, unsigned int subclass) hash_head = classhashentry(key); /* - * We can walk the hash lockfree, because the hash only - * grows, and we are careful when adding entries to the end: + * We do an RCU walk of the hash, see lockdep_free_key_range(). */ - list_for_each_entry(class, hash_head, hash_entry) { + if (DEBUG_LOCKS_WARN_ON(!irqs_disabled())) + return NULL; + + list_for_each_entry_rcu(class, hash_head, hash_entry) { if (class->key == key) { /* * Huh! same key, different name? Did someone trample @@ -728,7 +730,8 @@ register_lock_class(struct lockdep_map *lock, unsigned int subclass, int force) struct lockdep_subclass_key *key; struct list_head *hash_head; struct lock_class *class; - unsigned long flags; + + DEBUG_LOCKS_WARN_ON(!irqs_disabled()); class = look_up_lock_class(lock, subclass); if (likely(class)) @@ -750,28 +753,26 @@ register_lock_class(struct lockdep_map *lock, unsigned int subclass, int force) key = lock->key->subkeys + subclass; hash_head = classhashentry(key); - raw_local_irq_save(flags); if (!graph_lock()) { - raw_local_irq_restore(flags); return NULL; } /* * We have to do the hash-walk again, to avoid races * with another CPU: */ - list_for_each_entry(class, hash_head, hash_entry) + list_for_each_entry_rcu(class, hash_head, hash_entry) { if (class->key == key) goto out_unlock_set; + } + /* * Allocate a new key from the static array, and add it to * the hash: */ if (nr_lock_classes >= MAX_LOCKDEP_KEYS) { if (!debug_locks_off_graph_unlock()) { - raw_local_irq_restore(flags); return NULL; } - raw_local_irq_restore(flags); print_lockdep_off("BUG: MAX_LOCKDEP_KEYS too low!"); dump_stack(); @@ -798,7 +799,6 @@ register_lock_class(struct lockdep_map *lock, unsigned int subclass, int force) if (verbose(class)) { graph_unlock(); - raw_local_irq_restore(flags); printk("\nnew class %p: %s", class->key, class->name); if (class->name_version > 1) @@ -806,15 +806,12 @@ register_lock_class(struct lockdep_map *lock, unsigned int subclass, int force) printk("\n"); dump_stack(); - raw_local_irq_save(flags); if (!graph_lock()) { - raw_local_irq_restore(flags); return NULL; } } out_unlock_set: graph_unlock(); - raw_local_irq_restore(flags); out_set_class_cache: if (!subclass || force) @@ -870,11 +867,9 @@ static int add_lock_to_list(struct lock_class *class, struct lock_class *this, entry->distance = distance; entry->trace = *trace; /* - * Since we never remove from the dependency list, the list can - * be walked lockless by other CPUs, it's only allocation - * that must be protected by the spinlock. But this also means - * we must make new entries visible only once writes to the - * entry become visible - hence the RCU op: + * Both allocation and removal are done under the graph lock; but + * iteration is under RCU-sched; see look_up_lock_class() and + * lockdep_free_key_range(). */ list_add_tail_rcu(&entry->entry, head); @@ -1025,7 +1020,9 @@ static int __bfs(struct lock_list *source_entry, else head = &lock->class->locks_before; - list_for_each_entry(entry, head, entry) { + DEBUG_LOCKS_WARN_ON(!irqs_disabled()); + + list_for_each_entry_rcu(entry, head, entry) { if (!lock_accessed(entry)) { unsigned int cq_depth; mark_lock_accessed(entry, lock); @@ -2022,7 +2019,7 @@ static inline int lookup_chain_cache(struct task_struct *curr, * We can walk it lock-free, because entries only get added * to the hash: */ - list_for_each_entry(chain, hash_head, entry) { + list_for_each_entry_rcu(chain, hash_head, entry) { if (chain->chain_key == chain_key) { cache_hit: debug_atomic_inc(chain_lookup_hits); @@ -2996,8 +2993,18 @@ void lockdep_init_map(struct lockdep_map *lock, const char *name, if (unlikely(!debug_locks)) return; - if (subclass) + if (subclass) { + unsigned long flags; + + if (DEBUG_LOCKS_WARN_ON(current->lockdep_recursion)) + return; + + raw_local_irq_save(flags); + current->lockdep_recursion = 1; register_lock_class(lock, subclass, 1); + current->lockdep_recursion = 0; + raw_local_irq_restore(flags); + } } EXPORT_SYMBOL_GPL(lockdep_init_map); @@ -3887,9 +3894,17 @@ static inline int within(const void *addr, void *start, unsigned long size) return addr >= start && addr < start + size; } +/* + * Used in module.c to remove lock classes from memory that is going to be + * freed; and possibly re-used by other modules. + * + * We will have had one sync_sched() before getting here, so we're guaranteed + * nobody will look up these exact classes -- they're properly dead but still + * allocated. + */ void lockdep_free_key_range(void *start, unsigned long size) { - struct lock_class *class, *next; + struct lock_class *class; struct list_head *head; unsigned long flags; int i; @@ -3905,7 +3920,7 @@ void lockdep_free_key_range(void *start, unsigned long size) head = classhash_table + i; if (list_empty(head)) continue; - list_for_each_entry_safe(class, next, head, hash_entry) { + list_for_each_entry_rcu(class, head, hash_entry) { if (within(class->key, start, size)) zap_class(class); else if (within(class->name, start, size)) @@ -3916,11 +3931,25 @@ void lockdep_free_key_range(void *start, unsigned long size) if (locked) graph_unlock(); raw_local_irq_restore(flags); + + /* + * Wait for any possible iterators from look_up_lock_class() to pass + * before continuing to free the memory they refer to. + * + * sync_sched() is sufficient because the read-side is IRQ disable. + */ + synchronize_sched(); + + /* + * XXX at this point we could return the resources to the pool; + * instead we leak them. We would need to change to bitmap allocators + * instead of the linear allocators we have now. + */ } void lockdep_reset_lock(struct lockdep_map *lock) { - struct lock_class *class, *next; + struct lock_class *class; struct list_head *head; unsigned long flags; int i, j; @@ -3948,7 +3977,7 @@ void lockdep_reset_lock(struct lockdep_map *lock) head = classhash_table + i; if (list_empty(head)) continue; - list_for_each_entry_safe(class, next, head, hash_entry) { + list_for_each_entry_rcu(class, head, hash_entry) { int match = 0; for (j = 0; j < NR_LOCKDEP_CACHING_CLASSES; j++) diff --git a/kernel/locking/mcs_spinlock.h b/kernel/locking/mcs_spinlock.h index d1fe2ba5bac9..75e114bdf3f2 100644 --- a/kernel/locking/mcs_spinlock.h +++ b/kernel/locking/mcs_spinlock.h @@ -78,7 +78,7 @@ void mcs_spin_lock(struct mcs_spinlock **lock, struct mcs_spinlock *node) */ return; } - ACCESS_ONCE(prev->next) = node; + WRITE_ONCE(prev->next, node); /* Wait until the lock holder passes the lock down. */ arch_mcs_spin_lock_contended(&node->locked); @@ -91,7 +91,7 @@ void mcs_spin_lock(struct mcs_spinlock **lock, struct mcs_spinlock *node) static inline void mcs_spin_unlock(struct mcs_spinlock **lock, struct mcs_spinlock *node) { - struct mcs_spinlock *next = ACCESS_ONCE(node->next); + struct mcs_spinlock *next = READ_ONCE(node->next); if (likely(!next)) { /* @@ -100,7 +100,7 @@ void mcs_spin_unlock(struct mcs_spinlock **lock, struct mcs_spinlock *node) if (likely(cmpxchg(lock, node, NULL) == node)) return; /* Wait until the next pointer is set */ - while (!(next = ACCESS_ONCE(node->next))) + while (!(next = READ_ONCE(node->next))) cpu_relax_lowlatency(); } diff --git a/kernel/locking/mutex.c b/kernel/locking/mutex.c index 94674e5919cb..4cccea6b8934 100644 --- a/kernel/locking/mutex.c +++ b/kernel/locking/mutex.c @@ -25,7 +25,7 @@ #include <linux/spinlock.h> #include <linux/interrupt.h> #include <linux/debug_locks.h> -#include "mcs_spinlock.h" +#include <linux/osq_lock.h> /* * In the DEBUG case we are using the "NULL fastpath" for mutexes, @@ -217,44 +217,35 @@ ww_mutex_set_context_slowpath(struct ww_mutex *lock, } #ifdef CONFIG_MUTEX_SPIN_ON_OWNER -static inline bool owner_running(struct mutex *lock, struct task_struct *owner) -{ - if (lock->owner != owner) - return false; - - /* - * Ensure we emit the owner->on_cpu, dereference _after_ checking - * lock->owner still matches owner, if that fails, owner might - * point to free()d memory, if it still matches, the rcu_read_lock() - * ensures the memory stays valid. - */ - barrier(); - - return owner->on_cpu; -} - /* * Look out! "owner" is an entirely speculative pointer * access and not reliable. */ static noinline -int mutex_spin_on_owner(struct mutex *lock, struct task_struct *owner) +bool mutex_spin_on_owner(struct mutex *lock, struct task_struct *owner) { + bool ret = true; + rcu_read_lock(); - while (owner_running(lock, owner)) { - if (need_resched()) + while (lock->owner == owner) { + /* + * Ensure we emit the owner->on_cpu, dereference _after_ + * checking lock->owner still matches owner. If that fails, + * owner might point to freed memory. If it still matches, + * the rcu_read_lock() ensures the memory stays valid. + */ + barrier(); + + if (!owner->on_cpu || need_resched()) { + ret = false; break; + } cpu_relax_lowlatency(); } rcu_read_unlock(); - /* - * We break out the loop above on need_resched() and when the - * owner changed, which is a sign for heavy contention. Return - * success only when lock->owner is NULL. - */ - return lock->owner == NULL; + return ret; } /* @@ -269,7 +260,7 @@ static inline int mutex_can_spin_on_owner(struct mutex *lock) return 0; rcu_read_lock(); - owner = ACCESS_ONCE(lock->owner); + owner = READ_ONCE(lock->owner); if (owner) retval = owner->on_cpu; rcu_read_unlock(); @@ -343,7 +334,7 @@ static bool mutex_optimistic_spin(struct mutex *lock, * As such, when deadlock detection needs to be * performed the optimistic spinning cannot be done. */ - if (ACCESS_ONCE(ww->ctx)) + if (READ_ONCE(ww->ctx)) break; } @@ -351,7 +342,7 @@ static bool mutex_optimistic_spin(struct mutex *lock, * If there's an owner, wait for it to either * release the lock or go to sleep. */ - owner = ACCESS_ONCE(lock->owner); + owner = READ_ONCE(lock->owner); if (owner && !mutex_spin_on_owner(lock, owner)) break; @@ -490,7 +481,7 @@ static inline int __sched __ww_mutex_lock_check_stamp(struct mutex *lock, struct ww_acquire_ctx *ctx) { struct ww_mutex *ww = container_of(lock, struct ww_mutex, base); - struct ww_acquire_ctx *hold_ctx = ACCESS_ONCE(ww->ctx); + struct ww_acquire_ctx *hold_ctx = READ_ONCE(ww->ctx); if (!hold_ctx) return 0; diff --git a/kernel/locking/osq_lock.c b/kernel/locking/osq_lock.c index c112d00341b0..dc85ee23a26f 100644 --- a/kernel/locking/osq_lock.c +++ b/kernel/locking/osq_lock.c @@ -98,7 +98,7 @@ bool osq_lock(struct optimistic_spin_queue *lock) prev = decode_cpu(old); node->prev = prev; - ACCESS_ONCE(prev->next) = node; + WRITE_ONCE(prev->next, node); /* * Normally @prev is untouchable after the above store; because at that @@ -109,7 +109,7 @@ bool osq_lock(struct optimistic_spin_queue *lock) * cmpxchg in an attempt to undo our queueing. */ - while (!ACCESS_ONCE(node->locked)) { + while (!READ_ONCE(node->locked)) { /* * If we need to reschedule bail... so we can block. */ @@ -148,7 +148,7 @@ unqueue: * Or we race against a concurrent unqueue()'s step-B, in which * case its step-C will write us a new @node->prev pointer. */ - prev = ACCESS_ONCE(node->prev); + prev = READ_ONCE(node->prev); } /* @@ -170,8 +170,8 @@ unqueue: * it will wait in Step-A. */ - ACCESS_ONCE(next->prev) = prev; - ACCESS_ONCE(prev->next) = next; + WRITE_ONCE(next->prev, prev); + WRITE_ONCE(prev->next, next); return false; } @@ -193,11 +193,11 @@ void osq_unlock(struct optimistic_spin_queue *lock) node = this_cpu_ptr(&osq_node); next = xchg(&node->next, NULL); if (next) { - ACCESS_ONCE(next->locked) = 1; + WRITE_ONCE(next->locked, 1); return; } next = osq_wait_next(lock, node, NULL); if (next) - ACCESS_ONCE(next->locked) = 1; + WRITE_ONCE(next->locked, 1); } diff --git a/kernel/locking/rtmutex.c b/kernel/locking/rtmutex.c index e16e5542bf13..b73279367087 100644 --- a/kernel/locking/rtmutex.c +++ b/kernel/locking/rtmutex.c @@ -349,7 +349,7 @@ static inline struct rt_mutex *task_blocked_on_lock(struct task_struct *p) * * @task: the task owning the mutex (owner) for which a chain walk is * probably needed - * @deadlock_detect: do we have to carry out deadlock detection? + * @chwalk: do we have to carry out deadlock detection? * @orig_lock: the mutex (can be NULL if we are walking the chain to recheck * things for a task that has just got its priority adjusted, and * is waiting on a mutex) @@ -1193,6 +1193,7 @@ rt_mutex_slowlock(struct rt_mutex *lock, int state, ret = __rt_mutex_slowlock(lock, state, timeout, &waiter); if (unlikely(ret)) { + __set_current_state(TASK_RUNNING); if (rt_mutex_has_waiters(lock)) remove_waiter(lock, &waiter); rt_mutex_handle_deadlock(ret, chwalk, &waiter); diff --git a/kernel/locking/rwsem-spinlock.c b/kernel/locking/rwsem-spinlock.c index 2555ae15ec14..3a5048572065 100644 --- a/kernel/locking/rwsem-spinlock.c +++ b/kernel/locking/rwsem-spinlock.c @@ -85,6 +85,13 @@ __rwsem_do_wake(struct rw_semaphore *sem, int wakewrite) list_del(&waiter->list); tsk = waiter->task; + /* + * Make sure we do not wakeup the next reader before + * setting the nil condition to grant the next reader; + * otherwise we could miss the wakeup on the other + * side and end up sleeping again. See the pairing + * in rwsem_down_read_failed(). + */ smp_mb(); waiter->task = NULL; wake_up_process(tsk); diff --git a/kernel/locking/rwsem-xadd.c b/kernel/locking/rwsem-xadd.c index 2f7cc4076f50..3417d0172a5d 100644 --- a/kernel/locking/rwsem-xadd.c +++ b/kernel/locking/rwsem-xadd.c @@ -14,8 +14,9 @@ #include <linux/init.h> #include <linux/export.h> #include <linux/sched/rt.h> +#include <linux/osq_lock.h> -#include "mcs_spinlock.h" +#include "rwsem.h" /* * Guide to the rw_semaphore's count field for common values. @@ -186,6 +187,13 @@ __rwsem_do_wake(struct rw_semaphore *sem, enum rwsem_wake_type wake_type) waiter = list_entry(next, struct rwsem_waiter, list); next = waiter->list.next; tsk = waiter->task; + /* + * Make sure we do not wakeup the next reader before + * setting the nil condition to grant the next reader; + * otherwise we could miss the wakeup on the other + * side and end up sleeping again. See the pairing + * in rwsem_down_read_failed(). + */ smp_mb(); waiter->task = NULL; wake_up_process(tsk); @@ -258,6 +266,7 @@ static inline bool rwsem_try_write_lock(long count, struct rw_semaphore *sem) RWSEM_ACTIVE_WRITE_BIAS) == RWSEM_WAITING_BIAS) { if (!list_is_singular(&sem->wait_list)) rwsem_atomic_update(RWSEM_WAITING_BIAS, sem); + rwsem_set_owner(sem); return true; } @@ -270,15 +279,17 @@ static inline bool rwsem_try_write_lock(long count, struct rw_semaphore *sem) */ static inline bool rwsem_try_write_lock_unqueued(struct rw_semaphore *sem) { - long old, count = ACCESS_ONCE(sem->count); + long old, count = READ_ONCE(sem->count); while (true) { if (!(count == 0 || count == RWSEM_WAITING_BIAS)) return false; old = cmpxchg(&sem->count, count, count + RWSEM_ACTIVE_WRITE_BIAS); - if (old == count) + if (old == count) { + rwsem_set_owner(sem); return true; + } count = old; } @@ -287,60 +298,67 @@ static inline bool rwsem_try_write_lock_unqueued(struct rw_semaphore *sem) static inline bool rwsem_can_spin_on_owner(struct rw_semaphore *sem) { struct task_struct *owner; - bool on_cpu = false; + bool ret = true; if (need_resched()) return false; rcu_read_lock(); - owner = ACCESS_ONCE(sem->owner); - if (owner) - on_cpu = owner->on_cpu; - rcu_read_unlock(); - - /* - * If sem->owner is not set, yet we have just recently entered the - * slowpath, then there is a possibility reader(s) may have the lock. - * To be safe, avoid spinning in these situations. - */ - return on_cpu; -} - -static inline bool owner_running(struct rw_semaphore *sem, - struct task_struct *owner) -{ - if (sem->owner != owner) - return false; - - /* - * Ensure we emit the owner->on_cpu, dereference _after_ checking - * sem->owner still matches owner, if that fails, owner might - * point to free()d memory, if it still matches, the rcu_read_lock() - * ensures the memory stays valid. - */ - barrier(); + owner = READ_ONCE(sem->owner); + if (!owner) { + long count = READ_ONCE(sem->count); + /* + * If sem->owner is not set, yet we have just recently entered the + * slowpath with the lock being active, then there is a possibility + * reader(s) may have the lock. To be safe, bail spinning in these + * situations. + */ + if (count & RWSEM_ACTIVE_MASK) + ret = false; + goto done; + } - return owner->on_cpu; + ret = owner->on_cpu; +done: + rcu_read_unlock(); + return ret; } static noinline bool rwsem_spin_on_owner(struct rw_semaphore *sem, struct task_struct *owner) { + long count; + rcu_read_lock(); - while (owner_running(sem, owner)) { - if (need_resched()) - break; + while (sem->owner == owner) { + /* + * Ensure we emit the owner->on_cpu, dereference _after_ + * checking sem->owner still matches owner, if that fails, + * owner might point to free()d memory, if it still matches, + * the rcu_read_lock() ensures the memory stays valid. + */ + barrier(); + + /* abort spinning when need_resched or owner is not running */ + if (!owner->on_cpu || need_resched()) { + rcu_read_unlock(); + return false; + } cpu_relax_lowlatency(); } rcu_read_unlock(); + if (READ_ONCE(sem->owner)) + return true; /* new owner, continue spinning */ + /* - * We break out the loop above on need_resched() or when the - * owner changed, which is a sign for heavy contention. Return - * success only when sem->owner is NULL. + * When the owner is not set, the lock could be free or + * held by readers. Check the counter to verify the + * state. */ - return sem->owner == NULL; + count = READ_ONCE(sem->count); + return (count == 0 || count == RWSEM_WAITING_BIAS); } static bool rwsem_optimistic_spin(struct rw_semaphore *sem) @@ -358,7 +376,7 @@ static bool rwsem_optimistic_spin(struct rw_semaphore *sem) goto done; while (true) { - owner = ACCESS_ONCE(sem->owner); + owner = READ_ONCE(sem->owner); if (owner && !rwsem_spin_on_owner(sem, owner)) break; @@ -432,7 +450,7 @@ struct rw_semaphore __sched *rwsem_down_write_failed(struct rw_semaphore *sem) /* we're now waiting on the lock, but no longer actively locking */ if (waiting) { - count = ACCESS_ONCE(sem->count); + count = READ_ONCE(sem->count); /* * If there were already threads queued before us and there are diff --git a/kernel/locking/rwsem.c b/kernel/locking/rwsem.c index e2d3bc7f03b4..205be0ce34de 100644 --- a/kernel/locking/rwsem.c +++ b/kernel/locking/rwsem.c @@ -9,29 +9,9 @@ #include <linux/sched.h> #include <linux/export.h> #include <linux/rwsem.h> - #include <linux/atomic.h> -#ifdef CONFIG_RWSEM_SPIN_ON_OWNER -static inline void rwsem_set_owner(struct rw_semaphore *sem) -{ - sem->owner = current; -} - -static inline void rwsem_clear_owner(struct rw_semaphore *sem) -{ - sem->owner = NULL; -} - -#else -static inline void rwsem_set_owner(struct rw_semaphore *sem) -{ -} - -static inline void rwsem_clear_owner(struct rw_semaphore *sem) -{ -} -#endif +#include "rwsem.h" /* * lock for reading diff --git a/kernel/locking/rwsem.h b/kernel/locking/rwsem.h new file mode 100644 index 000000000000..870ed9a5b426 --- /dev/null +++ b/kernel/locking/rwsem.h @@ -0,0 +1,20 @@ +#ifdef CONFIG_RWSEM_SPIN_ON_OWNER +static inline void rwsem_set_owner(struct rw_semaphore *sem) +{ + sem->owner = current; +} + +static inline void rwsem_clear_owner(struct rw_semaphore *sem) +{ + sem->owner = NULL; +} + +#else +static inline void rwsem_set_owner(struct rw_semaphore *sem) +{ +} + +static inline void rwsem_clear_owner(struct rw_semaphore *sem) +{ +} +#endif diff --git a/kernel/module.c b/kernel/module.c index b34813f725e9..ec53f594e9c9 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -56,7 +56,6 @@ #include <linux/async.h> #include <linux/percpu.h> #include <linux/kmemleak.h> -#include <linux/kasan.h> #include <linux/jump_label.h> #include <linux/pfn.h> #include <linux/bsearch.h> @@ -1814,7 +1813,6 @@ static void unset_module_init_ro_nx(struct module *mod) { } void __weak module_memfree(void *module_region) { vfree(module_region); - kasan_module_free(module_region); } void __weak module_arch_cleanup(struct module *mod) @@ -1867,7 +1865,7 @@ static void free_module(struct module *mod) kfree(mod->args); percpu_modfree(mod); - /* Free lock-classes: */ + /* Free lock-classes; relies on the preceding sync_rcu(). */ lockdep_free_key_range(mod->module_core, mod->core_size); /* Finally, free the core (containing the module structure) */ @@ -2313,11 +2311,13 @@ static void layout_symtab(struct module *mod, struct load_info *info) info->symoffs = ALIGN(mod->core_size, symsect->sh_addralign ?: 1); info->stroffs = mod->core_size = info->symoffs + ndst * sizeof(Elf_Sym); mod->core_size += strtab_size; + mod->core_size = debug_align(mod->core_size); /* Put string table section at end of init part of module. */ strsect->sh_flags |= SHF_ALLOC; strsect->sh_entsize = get_offset(mod, &mod->init_size, strsect, info->index.str) | INIT_OFFSET_MASK; + mod->init_size = debug_align(mod->init_size); pr_debug("\t%s\n", info->secstrings + strsect->sh_name); } @@ -2479,6 +2479,23 @@ static int elf_header_check(struct load_info *info) return 0; } +#define COPY_CHUNK_SIZE (16*PAGE_SIZE) + +static int copy_chunked_from_user(void *dst, const void __user *usrc, unsigned long len) +{ + do { + unsigned long n = min(len, COPY_CHUNK_SIZE); + + if (copy_from_user(dst, usrc, n) != 0) + return -EFAULT; + cond_resched(); + dst += n; + usrc += n; + len -= n; + } while (len); + return 0; +} + /* Sets info->hdr and info->len. */ static int copy_module_from_user(const void __user *umod, unsigned long len, struct load_info *info) @@ -2498,7 +2515,7 @@ static int copy_module_from_user(const void __user *umod, unsigned long len, if (!info->hdr) return -ENOMEM; - if (copy_from_user(info->hdr, umod, info->len) != 0) { + if (copy_chunked_from_user(info->hdr, umod, info->len) != 0) { vfree(info->hdr); return -EFAULT; } @@ -3349,9 +3366,6 @@ static int load_module(struct load_info *info, const char __user *uargs, module_bug_cleanup(mod); mutex_unlock(&module_mutex); - /* Free lock-classes: */ - lockdep_free_key_range(mod->module_core, mod->core_size); - /* we can't deallocate the module until we clear memory protection */ unset_module_init_ro_nx(mod); unset_module_core_ro_nx(mod); @@ -3375,6 +3389,9 @@ static int load_module(struct load_info *info, const char __user *uargs, synchronize_rcu(); mutex_unlock(&module_mutex); free_module: + /* Free lock-classes; relies on the preceding sync_rcu() */ + lockdep_free_key_range(mod->module_core, mod->core_size); + module_deallocate(mod, info); free_copy: free_copy(info); diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c index c24d5a23bf93..5235dd4e1e2f 100644 --- a/kernel/power/snapshot.c +++ b/kernel/power/snapshot.c @@ -955,25 +955,6 @@ static void mark_nosave_pages(struct memory_bitmap *bm) } } -static bool is_nosave_page(unsigned long pfn) -{ - struct nosave_region *region; - - list_for_each_entry(region, &nosave_regions, list) { - if (pfn >= region->start_pfn && pfn < region->end_pfn) { - pr_err("PM: %#010llx in e820 nosave region: " - "[mem %#010llx-%#010llx]\n", - (unsigned long long) pfn << PAGE_SHIFT, - (unsigned long long) region->start_pfn << PAGE_SHIFT, - ((unsigned long long) region->end_pfn << PAGE_SHIFT) - - 1); - return true; - } - } - - return false; -} - /** * create_basic_memory_bitmaps - create bitmaps needed for marking page * frames that should not be saved and free page frames. The pointers @@ -2042,7 +2023,7 @@ static int mark_unsafe_pages(struct memory_bitmap *bm) do { pfn = memory_bm_next_pfn(bm); if (likely(pfn != BM_END_OF_MAP)) { - if (likely(pfn_valid(pfn)) && !is_nosave_page(pfn)) + if (likely(pfn_valid(pfn))) swsusp_set_page_free(pfn_to_page(pfn)); else return -EFAULT; diff --git a/kernel/printk/console_cmdline.h b/kernel/printk/console_cmdline.h index cbd69d842341..2ca4a8b5fe57 100644 --- a/kernel/printk/console_cmdline.h +++ b/kernel/printk/console_cmdline.h @@ -3,7 +3,7 @@ struct console_cmdline { - char name[8]; /* Name of the driver */ + char name[16]; /* Name of the driver */ int index; /* Minor dev. to use */ char *options; /* Options for the driver */ #ifdef CONFIG_A11Y_BRAILLE_CONSOLE diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c index 01cfd69c54c6..bb0635bd74f2 100644 --- a/kernel/printk/printk.c +++ b/kernel/printk/printk.c @@ -2464,6 +2464,7 @@ void register_console(struct console *newcon) for (i = 0, c = console_cmdline; i < MAX_CMDLINECONSOLES && c->name[0]; i++, c++) { + BUILD_BUG_ON(sizeof(c->name) != sizeof(newcon->name)); if (strcmp(c->name, newcon->name) != 0) continue; if (newcon->index >= 0 && diff --git a/kernel/sched/core.c b/kernel/sched/core.c index b578bb23410b..2f7937ee9e3a 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -693,6 +693,23 @@ static inline bool got_nohz_idle_kick(void) bool sched_can_stop_tick(void) { /* + * FIFO realtime policy runs the highest priority task. Other runnable + * tasks are of a lower priority. The scheduler tick does nothing. + */ + if (current->policy == SCHED_FIFO) + return true; + + /* + * Round-robin realtime tasks time slice with other tasks at the same + * realtime priority. Is this task the only one at this priority? + */ + if (current->policy == SCHED_RR) { + struct sched_rt_entity *rt_se = ¤t->rt; + + return rt_se->run_list.prev == rt_se->run_list.next; + } + + /* * More than one running task need preemption. * nr_running update is assumed to be visible * after IPI is sent from wakers. @@ -999,6 +1016,13 @@ void check_preempt_curr(struct rq *rq, struct task_struct *p, int flags) rq_clock_skip_update(rq, true); } +static ATOMIC_NOTIFIER_HEAD(task_migration_notifier); + +void register_task_migration_notifier(struct notifier_block *n) +{ + atomic_notifier_chain_register(&task_migration_notifier, n); +} + #ifdef CONFIG_SMP void set_task_cpu(struct task_struct *p, unsigned int new_cpu) { @@ -1029,10 +1053,18 @@ void set_task_cpu(struct task_struct *p, unsigned int new_cpu) trace_sched_migrate_task(p, new_cpu); if (task_cpu(p) != new_cpu) { + struct task_migration_notifier tmn; + if (p->sched_class->migrate_task_rq) p->sched_class->migrate_task_rq(p, new_cpu); p->se.nr_migrations++; perf_sw_event_sched(PERF_COUNT_SW_CPU_MIGRATIONS, 1, 0); + + tmn.task = p; + tmn.from_cpu = task_cpu(p); + tmn.to_cpu = new_cpu; + + atomic_notifier_call_chain(&task_migration_notifier, 0, &tmn); } __set_task_cpu(p, new_cpu); @@ -3037,6 +3069,8 @@ void rt_mutex_setprio(struct task_struct *p, int prio) } else { if (dl_prio(oldprio)) p->dl.dl_boosted = 0; + if (rt_prio(oldprio)) + p->rt.timeout = 0; p->sched_class = &fair_sched_class; } @@ -5321,36 +5355,13 @@ static int sched_cpu_active(struct notifier_block *nfb, static int sched_cpu_inactive(struct notifier_block *nfb, unsigned long action, void *hcpu) { - unsigned long flags; - long cpu = (long)hcpu; - struct dl_bw *dl_b; - switch (action & ~CPU_TASKS_FROZEN) { case CPU_DOWN_PREPARE: - set_cpu_active(cpu, false); - - /* explicitly allow suspend */ - if (!(action & CPU_TASKS_FROZEN)) { - bool overflow; - int cpus; - - rcu_read_lock_sched(); - dl_b = dl_bw_of(cpu); - - raw_spin_lock_irqsave(&dl_b->lock, flags); - cpus = dl_bw_cpus(cpu); - overflow = __dl_overflow(dl_b, cpus, 0, 0); - raw_spin_unlock_irqrestore(&dl_b->lock, flags); - - rcu_read_unlock_sched(); - - if (overflow) - return notifier_from_errno(-EBUSY); - } + set_cpu_active((long)hcpu, false); return NOTIFY_OK; + default: + return NOTIFY_DONE; } - - return NOTIFY_DONE; } static int __init migration_init(void) @@ -5431,17 +5442,6 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level, break; } - /* - * Even though we initialize ->capacity to something semi-sane, - * we leave capacity_orig unset. This allows us to detect if - * domain iteration is still funny without causing /0 traps. - */ - if (!group->sgc->capacity_orig) { - printk(KERN_CONT "\n"); - printk(KERN_ERR "ERROR: domain->cpu_capacity not set\n"); - break; - } - if (!cpumask_weight(sched_group_cpus(group))) { printk(KERN_CONT "\n"); printk(KERN_ERR "ERROR: empty group\n"); @@ -5922,7 +5922,6 @@ build_overlap_sched_groups(struct sched_domain *sd, int cpu) * die on a /0 trap. */ sg->sgc->capacity = SCHED_CAPACITY_SCALE * cpumask_weight(sg_span); - sg->sgc->capacity_orig = sg->sgc->capacity; /* * Make sure the first group of this domain contains the @@ -6233,6 +6232,7 @@ sd_init(struct sched_domain_topology_level *tl, int cpu) */ if (sd->flags & SD_SHARE_CPUCAPACITY) { + sd->flags |= SD_PREFER_SIBLING; sd->imbalance_pct = 110; sd->smt_gain = 1178; /* ~15% */ @@ -6998,7 +6998,6 @@ static int cpuset_cpu_active(struct notifier_block *nfb, unsigned long action, */ case CPU_ONLINE: - case CPU_DOWN_FAILED: cpuset_update_active_cpus(true); break; default: @@ -7010,8 +7009,30 @@ static int cpuset_cpu_active(struct notifier_block *nfb, unsigned long action, static int cpuset_cpu_inactive(struct notifier_block *nfb, unsigned long action, void *hcpu) { - switch (action) { + unsigned long flags; + long cpu = (long)hcpu; + struct dl_bw *dl_b; + + switch (action & ~CPU_TASKS_FROZEN) { case CPU_DOWN_PREPARE: + /* explicitly allow suspend */ + if (!(action & CPU_TASKS_FROZEN)) { + bool overflow; + int cpus; + + rcu_read_lock_sched(); + dl_b = dl_bw_of(cpu); + + raw_spin_lock_irqsave(&dl_b->lock, flags); + cpus = dl_bw_cpus(cpu); + overflow = __dl_overflow(dl_b, cpus, 0, 0); + raw_spin_unlock_irqrestore(&dl_b->lock, flags); + + rcu_read_unlock_sched(); + + if (overflow) + return notifier_from_errno(-EBUSY); + } cpuset_update_active_cpus(false); break; case CPU_DOWN_PREPARE_FROZEN: @@ -7156,8 +7177,8 @@ void __init sched_init(void) rq->calc_load_active = 0; rq->calc_load_update = jiffies + LOAD_FREQ; init_cfs_rq(&rq->cfs); - init_rt_rq(&rq->rt, rq); - init_dl_rq(&rq->dl, rq); + init_rt_rq(&rq->rt); + init_dl_rq(&rq->dl); #ifdef CONFIG_FAIR_GROUP_SCHED root_task_group.shares = ROOT_TASK_GROUP_LOAD; INIT_LIST_HEAD(&rq->leaf_cfs_rq_list); @@ -7197,7 +7218,7 @@ void __init sched_init(void) #ifdef CONFIG_SMP rq->sd = NULL; rq->rd = NULL; - rq->cpu_capacity = SCHED_CAPACITY_SCALE; + rq->cpu_capacity = rq->cpu_capacity_orig = SCHED_CAPACITY_SCALE; rq->post_schedule = 0; rq->active_balance = 0; rq->next_balance = jiffies; @@ -7796,7 +7817,7 @@ static int sched_rt_global_constraints(void) } #endif /* CONFIG_RT_GROUP_SCHED */ -static int sched_dl_global_constraints(void) +static int sched_dl_global_validate(void) { u64 runtime = global_rt_runtime(); u64 period = global_rt_period(); @@ -7897,11 +7918,11 @@ int sched_rt_handler(struct ctl_table *table, int write, if (ret) goto undo; - ret = sched_rt_global_constraints(); + ret = sched_dl_global_validate(); if (ret) goto undo; - ret = sched_dl_global_constraints(); + ret = sched_rt_global_constraints(); if (ret) goto undo; diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c index 3fa8fa6d9403..5e95145088fd 100644 --- a/kernel/sched/deadline.c +++ b/kernel/sched/deadline.c @@ -69,7 +69,7 @@ void init_dl_bw(struct dl_bw *dl_b) dl_b->total_bw = 0; } -void init_dl_rq(struct dl_rq *dl_rq, struct rq *rq) +void init_dl_rq(struct dl_rq *dl_rq) { dl_rq->rb_root = RB_ROOT; @@ -218,6 +218,52 @@ static inline void set_post_schedule(struct rq *rq) rq->post_schedule = has_pushable_dl_tasks(rq); } +static struct rq *find_lock_later_rq(struct task_struct *task, struct rq *rq); + +static void dl_task_offline_migration(struct rq *rq, struct task_struct *p) +{ + struct rq *later_rq = NULL; + bool fallback = false; + + later_rq = find_lock_later_rq(p, rq); + + if (!later_rq) { + int cpu; + + /* + * If we cannot preempt any rq, fall back to pick any + * online cpu. + */ + fallback = true; + cpu = cpumask_any_and(cpu_active_mask, tsk_cpus_allowed(p)); + if (cpu >= nr_cpu_ids) { + /* + * Fail to find any suitable cpu. + * The task will never come back! + */ + BUG_ON(dl_bandwidth_enabled()); + + /* + * If admission control is disabled we + * try a little harder to let the task + * run. + */ + cpu = cpumask_any(cpu_active_mask); + } + later_rq = cpu_rq(cpu); + double_lock_balance(rq, later_rq); + } + + deactivate_task(rq, p, 0); + set_task_cpu(p, later_rq->cpu); + activate_task(later_rq, p, ENQUEUE_REPLENISH); + + if (!fallback) + resched_curr(later_rq); + + double_unlock_balance(rq, later_rq); +} + #else static inline @@ -514,7 +560,7 @@ static enum hrtimer_restart dl_task_timer(struct hrtimer *timer) unsigned long flags; struct rq *rq; - rq = task_rq_lock(current, &flags); + rq = task_rq_lock(p, &flags); /* * We need to take care of several possible races here: @@ -536,6 +582,17 @@ static enum hrtimer_restart dl_task_timer(struct hrtimer *timer) sched_clock_tick(); update_rq_clock(rq); +#ifdef CONFIG_SMP + /* + * If we find that the rq the task was on is no longer + * available, we need to select a new rq. + */ + if (unlikely(!rq->online)) { + dl_task_offline_migration(rq, p); + goto unlock; + } +#endif + /* * If the throttle happened during sched-out; like: * @@ -569,7 +626,7 @@ static enum hrtimer_restart dl_task_timer(struct hrtimer *timer) push_dl_task(rq); #endif unlock: - task_rq_unlock(rq, current, &flags); + task_rq_unlock(rq, p, &flags); return HRTIMER_NORESTART; } @@ -914,6 +971,12 @@ static void yield_task_dl(struct rq *rq) } update_rq_clock(rq); update_curr_dl(rq); + /* + * Tell update_rq_clock() that we've just updated, + * so we don't do microscopic update in schedule() + * and double the fastpath cost. + */ + rq_clock_skip_update(rq, true); } #ifdef CONFIG_SMP @@ -1659,14 +1722,6 @@ static void switched_to_dl(struct rq *rq, struct task_struct *p) { int check_resched = 1; - /* - * If p is throttled, don't consider the possibility - * of preempting rq->curr, the check will be done right - * after its runtime will get replenished. - */ - if (unlikely(p->dl.dl_throttled)) - return; - if (task_on_rq_queued(p) && rq->curr != p) { #ifdef CONFIG_SMP if (p->nr_cpus_allowed > 1 && rq->dl.overloaded && diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c index 8baaf858d25c..a245c1fc6f0a 100644 --- a/kernel/sched/debug.c +++ b/kernel/sched/debug.c @@ -71,7 +71,7 @@ static void print_cfs_group_stats(struct seq_file *m, int cpu, struct task_group if (!se) { struct sched_avg *avg = &cpu_rq(cpu)->avg; P(avg->runnable_avg_sum); - P(avg->runnable_avg_period); + P(avg->avg_period); return; } @@ -94,8 +94,10 @@ static void print_cfs_group_stats(struct seq_file *m, int cpu, struct task_group P(se->load.weight); #ifdef CONFIG_SMP P(se->avg.runnable_avg_sum); - P(se->avg.runnable_avg_period); + P(se->avg.running_avg_sum); + P(se->avg.avg_period); P(se->avg.load_avg_contrib); + P(se->avg.utilization_avg_contrib); P(se->avg.decay_count); #endif #undef PN @@ -214,6 +216,8 @@ void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq) cfs_rq->runnable_load_avg); SEQ_printf(m, " .%-30s: %ld\n", "blocked_load_avg", cfs_rq->blocked_load_avg); + SEQ_printf(m, " .%-30s: %ld\n", "utilization_load_avg", + cfs_rq->utilization_load_avg); #ifdef CONFIG_FAIR_GROUP_SCHED SEQ_printf(m, " .%-30s: %ld\n", "tg_load_contrib", cfs_rq->tg_load_contrib); @@ -636,8 +640,10 @@ void proc_sched_show_task(struct task_struct *p, struct seq_file *m) P(se.load.weight); #ifdef CONFIG_SMP P(se.avg.runnable_avg_sum); - P(se.avg.runnable_avg_period); + P(se.avg.running_avg_sum); + P(se.avg.avg_period); P(se.avg.load_avg_contrib); + P(se.avg.utilization_avg_contrib); P(se.avg.decay_count); #endif P(policy); diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 7ce18f3c097a..ffeaa4105e48 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -670,6 +670,7 @@ static int select_idle_sibling(struct task_struct *p, int cpu); static unsigned long task_h_load(struct task_struct *p); static inline void __update_task_entity_contrib(struct sched_entity *se); +static inline void __update_task_entity_utilization(struct sched_entity *se); /* Give new task start runnable values to heavy its load in infant time */ void init_task_runnable_average(struct task_struct *p) @@ -677,9 +678,10 @@ void init_task_runnable_average(struct task_struct *p) u32 slice; slice = sched_slice(task_cfs_rq(p), &p->se) >> 10; - p->se.avg.runnable_avg_sum = slice; - p->se.avg.runnable_avg_period = slice; + p->se.avg.runnable_avg_sum = p->se.avg.running_avg_sum = slice; + p->se.avg.avg_period = slice; __update_task_entity_contrib(&p->se); + __update_task_entity_utilization(&p->se); } #else void init_task_runnable_average(struct task_struct *p) @@ -1196,9 +1198,11 @@ static void task_numa_assign(struct task_numa_env *env, static bool load_too_imbalanced(long src_load, long dst_load, struct task_numa_env *env) { - long imb, old_imb; - long orig_src_load, orig_dst_load; long src_capacity, dst_capacity; + long orig_src_load; + long load_a, load_b; + long moved_load; + long imb; /* * The load is corrected for the CPU capacity available on each node. @@ -1211,30 +1215,39 @@ static bool load_too_imbalanced(long src_load, long dst_load, dst_capacity = env->dst_stats.compute_capacity; /* We care about the slope of the imbalance, not the direction. */ - if (dst_load < src_load) - swap(dst_load, src_load); + load_a = dst_load; + load_b = src_load; + if (load_a < load_b) + swap(load_a, load_b); /* Is the difference below the threshold? */ - imb = dst_load * src_capacity * 100 - - src_load * dst_capacity * env->imbalance_pct; + imb = load_a * src_capacity * 100 - + load_b * dst_capacity * env->imbalance_pct; if (imb <= 0) return false; /* * The imbalance is above the allowed threshold. - * Compare it with the old imbalance. + * Allow a move that brings us closer to a balanced situation, + * without moving things past the point of balance. */ orig_src_load = env->src_stats.load; - orig_dst_load = env->dst_stats.load; - if (orig_dst_load < orig_src_load) - swap(orig_dst_load, orig_src_load); - - old_imb = orig_dst_load * src_capacity * 100 - - orig_src_load * dst_capacity * env->imbalance_pct; + /* + * In a task swap, there will be one load moving from src to dst, + * and another moving back. This is the net sum of both moves. + * A simple task move will always have a positive value. + * Allow the move if it brings the system closer to a balanced + * situation, without crossing over the balance point. + */ + moved_load = orig_src_load - src_load; - /* Would this change make things worse? */ - return (imb > old_imb); + if (moved_load > 0) + /* Moving src -> dst. Did we overshoot balance? */ + return src_load * dst_capacity < dst_load * src_capacity; + else + /* Moving dst -> src. Did we overshoot balance? */ + return dst_load * src_capacity < src_load * dst_capacity; } /* @@ -1609,9 +1622,11 @@ static void update_task_scan_period(struct task_struct *p, /* * If there were no record hinting faults then either the task is * completely idle or all activity is areas that are not of interest - * to automatic numa balancing. Scan slower + * to automatic numa balancing. Related to that, if there were failed + * migration then it implies we are migrating too quickly or the local + * node is overloaded. In either case, scan slower */ - if (local + shared == 0) { + if (local + shared == 0 || p->numa_faults_locality[2]) { p->numa_scan_period = min(p->numa_scan_period_max, p->numa_scan_period << 1); @@ -1673,7 +1688,7 @@ static u64 numa_get_avg_runtime(struct task_struct *p, u64 *period) *period = now - p->last_task_numa_placement; } else { delta = p->se.avg.runnable_avg_sum; - *period = p->se.avg.runnable_avg_period; + *period = p->se.avg.avg_period; } p->last_sum_exec_runtime = runtime; @@ -1763,6 +1778,8 @@ static int preferred_group_nid(struct task_struct *p, int nid) } } /* Next round, evaluate the nodes within max_group. */ + if (!max_faults) + break; nodes = max_group; } return nid; @@ -2080,6 +2097,8 @@ void task_numa_fault(int last_cpupid, int mem_node, int pages, int flags) if (migrated) p->numa_pages_migrated += pages; + if (flags & TNF_MIGRATE_FAIL) + p->numa_faults_locality[2] += pages; p->numa_faults[task_faults_idx(NUMA_MEMBUF, mem_node, priv)] += pages; p->numa_faults[task_faults_idx(NUMA_CPUBUF, cpu_node, priv)] += pages; @@ -2161,8 +2180,10 @@ void task_numa_work(struct callback_head *work) vma = mm->mmap; } for (; vma; vma = vma->vm_next) { - if (!vma_migratable(vma) || !vma_policy_mof(vma)) + if (!vma_migratable(vma) || !vma_policy_mof(vma) || + is_vm_hugetlb_page(vma)) { continue; + } /* * Shared library pages mapped by multiple processes are not @@ -2497,13 +2518,15 @@ static u32 __compute_runnable_contrib(u64 n) * load_avg = u_0` + y*(u_0 + u_1*y + u_2*y^2 + ... ) * = u_0 + u_1*y + u_2*y^2 + ... [re-labeling u_i --> u_{i+1}] */ -static __always_inline int __update_entity_runnable_avg(u64 now, +static __always_inline int __update_entity_runnable_avg(u64 now, int cpu, struct sched_avg *sa, - int runnable) + int runnable, + int running) { u64 delta, periods; u32 runnable_contrib; int delta_w, decayed = 0; + unsigned long scale_freq = arch_scale_freq_capacity(NULL, cpu); delta = now - sa->last_runnable_update; /* @@ -2525,7 +2548,7 @@ static __always_inline int __update_entity_runnable_avg(u64 now, sa->last_runnable_update = now; /* delta_w is the amount already accumulated against our next period */ - delta_w = sa->runnable_avg_period % 1024; + delta_w = sa->avg_period % 1024; if (delta + delta_w >= 1024) { /* period roll-over */ decayed = 1; @@ -2538,7 +2561,10 @@ static __always_inline int __update_entity_runnable_avg(u64 now, delta_w = 1024 - delta_w; if (runnable) sa->runnable_avg_sum += delta_w; - sa->runnable_avg_period += delta_w; + if (running) + sa->running_avg_sum += delta_w * scale_freq + >> SCHED_CAPACITY_SHIFT; + sa->avg_period += delta_w; delta -= delta_w; @@ -2548,20 +2574,28 @@ static __always_inline int __update_entity_runnable_avg(u64 now, sa->runnable_avg_sum = decay_load(sa->runnable_avg_sum, periods + 1); - sa->runnable_avg_period = decay_load(sa->runnable_avg_period, + sa->running_avg_sum = decay_load(sa->running_avg_sum, + periods + 1); + sa->avg_period = decay_load(sa->avg_period, periods + 1); /* Efficiently calculate \sum (1..n_period) 1024*y^i */ runnable_contrib = __compute_runnable_contrib(periods); if (runnable) sa->runnable_avg_sum += runnable_contrib; - sa->runnable_avg_period += runnable_contrib; + if (running) + sa->running_avg_sum += runnable_contrib * scale_freq + >> SCHED_CAPACITY_SHIFT; + sa->avg_period += runnable_contrib; } /* Remainder of delta accrued against u_0` */ if (runnable) sa->runnable_avg_sum += delta; - sa->runnable_avg_period += delta; + if (running) + sa->running_avg_sum += delta * scale_freq + >> SCHED_CAPACITY_SHIFT; + sa->avg_period += delta; return decayed; } @@ -2578,6 +2612,8 @@ static inline u64 __synchronize_entity_decay(struct sched_entity *se) return 0; se->avg.load_avg_contrib = decay_load(se->avg.load_avg_contrib, decays); + se->avg.utilization_avg_contrib = + decay_load(se->avg.utilization_avg_contrib, decays); return decays; } @@ -2613,7 +2649,7 @@ static inline void __update_tg_runnable_avg(struct sched_avg *sa, /* The fraction of a cpu used by this cfs_rq */ contrib = div_u64((u64)sa->runnable_avg_sum << NICE_0_SHIFT, - sa->runnable_avg_period + 1); + sa->avg_period + 1); contrib -= cfs_rq->tg_runnable_contrib; if (abs(contrib) > cfs_rq->tg_runnable_contrib / 64) { @@ -2666,7 +2702,8 @@ static inline void __update_group_entity_contrib(struct sched_entity *se) static inline void update_rq_runnable_avg(struct rq *rq, int runnable) { - __update_entity_runnable_avg(rq_clock_task(rq), &rq->avg, runnable); + __update_entity_runnable_avg(rq_clock_task(rq), cpu_of(rq), &rq->avg, + runnable, runnable); __update_tg_runnable_avg(&rq->avg, &rq->cfs); } #else /* CONFIG_FAIR_GROUP_SCHED */ @@ -2684,7 +2721,7 @@ static inline void __update_task_entity_contrib(struct sched_entity *se) /* avoid overflowing a 32-bit type w/ SCHED_LOAD_SCALE */ contrib = se->avg.runnable_avg_sum * scale_load_down(se->load.weight); - contrib /= (se->avg.runnable_avg_period + 1); + contrib /= (se->avg.avg_period + 1); se->avg.load_avg_contrib = scale_load(contrib); } @@ -2703,6 +2740,30 @@ static long __update_entity_load_avg_contrib(struct sched_entity *se) return se->avg.load_avg_contrib - old_contrib; } + +static inline void __update_task_entity_utilization(struct sched_entity *se) +{ + u32 contrib; + + /* avoid overflowing a 32-bit type w/ SCHED_LOAD_SCALE */ + contrib = se->avg.running_avg_sum * scale_load_down(SCHED_LOAD_SCALE); + contrib /= (se->avg.avg_period + 1); + se->avg.utilization_avg_contrib = scale_load(contrib); +} + +static long __update_entity_utilization_avg_contrib(struct sched_entity *se) +{ + long old_contrib = se->avg.utilization_avg_contrib; + + if (entity_is_task(se)) + __update_task_entity_utilization(se); + else + se->avg.utilization_avg_contrib = + group_cfs_rq(se)->utilization_load_avg; + + return se->avg.utilization_avg_contrib - old_contrib; +} + static inline void subtract_blocked_load_contrib(struct cfs_rq *cfs_rq, long load_contrib) { @@ -2719,7 +2780,8 @@ static inline void update_entity_load_avg(struct sched_entity *se, int update_cfs_rq) { struct cfs_rq *cfs_rq = cfs_rq_of(se); - long contrib_delta; + long contrib_delta, utilization_delta; + int cpu = cpu_of(rq_of(cfs_rq)); u64 now; /* @@ -2731,18 +2793,22 @@ static inline void update_entity_load_avg(struct sched_entity *se, else now = cfs_rq_clock_task(group_cfs_rq(se)); - if (!__update_entity_runnable_avg(now, &se->avg, se->on_rq)) + if (!__update_entity_runnable_avg(now, cpu, &se->avg, se->on_rq, + cfs_rq->curr == se)) return; contrib_delta = __update_entity_load_avg_contrib(se); + utilization_delta = __update_entity_utilization_avg_contrib(se); if (!update_cfs_rq) return; - if (se->on_rq) + if (se->on_rq) { cfs_rq->runnable_load_avg += contrib_delta; - else + cfs_rq->utilization_load_avg += utilization_delta; + } else { subtract_blocked_load_contrib(cfs_rq, -contrib_delta); + } } /* @@ -2817,6 +2883,7 @@ static inline void enqueue_entity_load_avg(struct cfs_rq *cfs_rq, } cfs_rq->runnable_load_avg += se->avg.load_avg_contrib; + cfs_rq->utilization_load_avg += se->avg.utilization_avg_contrib; /* we force update consideration on load-balancer moves */ update_cfs_rq_blocked_load(cfs_rq, !wakeup); } @@ -2835,6 +2902,7 @@ static inline void dequeue_entity_load_avg(struct cfs_rq *cfs_rq, update_cfs_rq_blocked_load(cfs_rq, !sleep); cfs_rq->runnable_load_avg -= se->avg.load_avg_contrib; + cfs_rq->utilization_load_avg -= se->avg.utilization_avg_contrib; if (sleep) { cfs_rq->blocked_load_avg += se->avg.load_avg_contrib; se->avg.decay_count = atomic64_read(&cfs_rq->decay_counter); @@ -3172,6 +3240,7 @@ set_next_entity(struct cfs_rq *cfs_rq, struct sched_entity *se) */ update_stats_wait_end(cfs_rq, se); __dequeue_entity(cfs_rq, se); + update_entity_load_avg(se, 1); } update_stats_curr_start(cfs_rq, se); @@ -4298,6 +4367,11 @@ static unsigned long capacity_of(int cpu) return cpu_rq(cpu)->cpu_capacity; } +static unsigned long capacity_orig_of(int cpu) +{ + return cpu_rq(cpu)->cpu_capacity_orig; +} + static unsigned long cpu_avg_load_per_task(int cpu) { struct rq *rq = cpu_rq(cpu); @@ -4711,6 +4785,33 @@ next: done: return target; } +/* + * get_cpu_usage returns the amount of capacity of a CPU that is used by CFS + * tasks. The unit of the return value must be the one of capacity so we can + * compare the usage with the capacity of the CPU that is available for CFS + * task (ie cpu_capacity). + * cfs.utilization_load_avg is the sum of running time of runnable tasks on a + * CPU. It represents the amount of utilization of a CPU in the range + * [0..SCHED_LOAD_SCALE]. The usage of a CPU can't be higher than the full + * capacity of the CPU because it's about the running time on this CPU. + * Nevertheless, cfs.utilization_load_avg can be higher than SCHED_LOAD_SCALE + * because of unfortunate rounding in avg_period and running_load_avg or just + * after migrating tasks until the average stabilizes with the new running + * time. So we need to check that the usage stays into the range + * [0..cpu_capacity_orig] and cap if necessary. + * Without capping the usage, a group could be seen as overloaded (CPU0 usage + * at 121% + CPU1 usage at 80%) whereas CPU1 has 20% of available capacity + */ +static int get_cpu_usage(int cpu) +{ + unsigned long usage = cpu_rq(cpu)->cfs.utilization_load_avg; + unsigned long capacity = capacity_orig_of(cpu); + + if (usage >= SCHED_LOAD_SCALE) + return capacity; + + return (usage * capacity) >> SCHED_LOAD_SHIFT; +} /* * select_task_rq_fair: Select target runqueue for the waking task in domains @@ -5837,12 +5938,12 @@ struct sg_lb_stats { unsigned long sum_weighted_load; /* Weighted load of group's tasks */ unsigned long load_per_task; unsigned long group_capacity; + unsigned long group_usage; /* Total usage of the group */ unsigned int sum_nr_running; /* Nr tasks running in the group */ - unsigned int group_capacity_factor; unsigned int idle_cpus; unsigned int group_weight; enum group_type group_type; - int group_has_free_capacity; + int group_no_capacity; #ifdef CONFIG_NUMA_BALANCING unsigned int nr_numa_running; unsigned int nr_preferred_running; @@ -5913,16 +6014,6 @@ static inline int get_sd_load_idx(struct sched_domain *sd, return load_idx; } -static unsigned long default_scale_capacity(struct sched_domain *sd, int cpu) -{ - return SCHED_CAPACITY_SCALE; -} - -unsigned long __weak arch_scale_freq_capacity(struct sched_domain *sd, int cpu) -{ - return default_scale_capacity(sd, cpu); -} - static unsigned long default_scale_cpu_capacity(struct sched_domain *sd, int cpu) { if ((sd->flags & SD_SHARE_CPUCAPACITY) && (sd->span_weight > 1)) @@ -5939,7 +6030,7 @@ unsigned long __weak arch_scale_cpu_capacity(struct sched_domain *sd, int cpu) static unsigned long scale_rt_capacity(int cpu) { struct rq *rq = cpu_rq(cpu); - u64 total, available, age_stamp, avg; + u64 total, used, age_stamp, avg; s64 delta; /* @@ -5955,19 +6046,12 @@ static unsigned long scale_rt_capacity(int cpu) total = sched_avg_period() + delta; - if (unlikely(total < avg)) { - /* Ensures that capacity won't end up being negative */ - available = 0; - } else { - available = total - avg; - } - - if (unlikely((s64)total < SCHED_CAPACITY_SCALE)) - total = SCHED_CAPACITY_SCALE; + used = div_u64(avg, total); - total >>= SCHED_CAPACITY_SHIFT; + if (likely(used < SCHED_CAPACITY_SCALE)) + return SCHED_CAPACITY_SCALE - used; - return div_u64(available, total); + return 1; } static void update_cpu_capacity(struct sched_domain *sd, int cpu) @@ -5982,14 +6066,7 @@ static void update_cpu_capacity(struct sched_domain *sd, int cpu) capacity >>= SCHED_CAPACITY_SHIFT; - sdg->sgc->capacity_orig = capacity; - - if (sched_feat(ARCH_CAPACITY)) - capacity *= arch_scale_freq_capacity(sd, cpu); - else - capacity *= default_scale_capacity(sd, cpu); - - capacity >>= SCHED_CAPACITY_SHIFT; + cpu_rq(cpu)->cpu_capacity_orig = capacity; capacity *= scale_rt_capacity(cpu); capacity >>= SCHED_CAPACITY_SHIFT; @@ -6005,7 +6082,7 @@ void update_group_capacity(struct sched_domain *sd, int cpu) { struct sched_domain *child = sd->child; struct sched_group *group, *sdg = sd->groups; - unsigned long capacity, capacity_orig; + unsigned long capacity; unsigned long interval; interval = msecs_to_jiffies(sd->balance_interval); @@ -6017,7 +6094,7 @@ void update_group_capacity(struct sched_domain *sd, int cpu) return; } - capacity_orig = capacity = 0; + capacity = 0; if (child->flags & SD_OVERLAP) { /* @@ -6037,19 +6114,15 @@ void update_group_capacity(struct sched_domain *sd, int cpu) * Use capacity_of(), which is set irrespective of domains * in update_cpu_capacity(). * - * This avoids capacity/capacity_orig from being 0 and + * This avoids capacity from being 0 and * causing divide-by-zero issues on boot. - * - * Runtime updates will correct capacity_orig. */ if (unlikely(!rq->sd)) { - capacity_orig += capacity_of(cpu); capacity += capacity_of(cpu); continue; } sgc = rq->sd->groups->sgc; - capacity_orig += sgc->capacity_orig; capacity += sgc->capacity; } } else { @@ -6060,39 +6133,24 @@ void update_group_capacity(struct sched_domain *sd, int cpu) group = child->groups; do { - capacity_orig += group->sgc->capacity_orig; capacity += group->sgc->capacity; group = group->next; } while (group != child->groups); } - sdg->sgc->capacity_orig = capacity_orig; sdg->sgc->capacity = capacity; } /* - * Try and fix up capacity for tiny siblings, this is needed when - * things like SD_ASYM_PACKING need f_b_g to select another sibling - * which on its own isn't powerful enough. - * - * See update_sd_pick_busiest() and check_asym_packing(). + * Check whether the capacity of the rq has been noticeably reduced by side + * activity. The imbalance_pct is used for the threshold. + * Return true is the capacity is reduced */ static inline int -fix_small_capacity(struct sched_domain *sd, struct sched_group *group) +check_cpu_capacity(struct rq *rq, struct sched_domain *sd) { - /* - * Only siblings can have significantly less than SCHED_CAPACITY_SCALE - */ - if (!(sd->flags & SD_SHARE_CPUCAPACITY)) - return 0; - - /* - * If ~90% of the cpu_capacity is still there, we're good. - */ - if (group->sgc->capacity * 32 > group->sgc->capacity_orig * 29) - return 1; - - return 0; + return ((rq->cpu_capacity * sd->imbalance_pct) < + (rq->cpu_capacity_orig * 100)); } /* @@ -6130,37 +6188,56 @@ static inline int sg_imbalanced(struct sched_group *group) } /* - * Compute the group capacity factor. - * - * Avoid the issue where N*frac(smt_capacity) >= 1 creates 'phantom' cores by - * first dividing out the smt factor and computing the actual number of cores - * and limit unit capacity with that. + * group_has_capacity returns true if the group has spare capacity that could + * be used by some tasks. + * We consider that a group has spare capacity if the * number of task is + * smaller than the number of CPUs or if the usage is lower than the available + * capacity for CFS tasks. + * For the latter, we use a threshold to stabilize the state, to take into + * account the variance of the tasks' load and to return true if the available + * capacity in meaningful for the load balancer. + * As an example, an available capacity of 1% can appear but it doesn't make + * any benefit for the load balance. */ -static inline int sg_capacity_factor(struct lb_env *env, struct sched_group *group) +static inline bool +group_has_capacity(struct lb_env *env, struct sg_lb_stats *sgs) { - unsigned int capacity_factor, smt, cpus; - unsigned int capacity, capacity_orig; + if (sgs->sum_nr_running < sgs->group_weight) + return true; - capacity = group->sgc->capacity; - capacity_orig = group->sgc->capacity_orig; - cpus = group->group_weight; + if ((sgs->group_capacity * 100) > + (sgs->group_usage * env->sd->imbalance_pct)) + return true; + + return false; +} - /* smt := ceil(cpus / capacity), assumes: 1 < smt_capacity < 2 */ - smt = DIV_ROUND_UP(SCHED_CAPACITY_SCALE * cpus, capacity_orig); - capacity_factor = cpus / smt; /* cores */ +/* + * group_is_overloaded returns true if the group has more tasks than it can + * handle. + * group_is_overloaded is not equals to !group_has_capacity because a group + * with the exact right number of tasks, has no more spare capacity but is not + * overloaded so both group_has_capacity and group_is_overloaded return + * false. + */ +static inline bool +group_is_overloaded(struct lb_env *env, struct sg_lb_stats *sgs) +{ + if (sgs->sum_nr_running <= sgs->group_weight) + return false; - capacity_factor = min_t(unsigned, - capacity_factor, DIV_ROUND_CLOSEST(capacity, SCHED_CAPACITY_SCALE)); - if (!capacity_factor) - capacity_factor = fix_small_capacity(env->sd, group); + if ((sgs->group_capacity * 100) < + (sgs->group_usage * env->sd->imbalance_pct)) + return true; - return capacity_factor; + return false; } -static enum group_type -group_classify(struct sched_group *group, struct sg_lb_stats *sgs) +static enum group_type group_classify(struct lb_env *env, + struct sched_group *group, + struct sg_lb_stats *sgs) { - if (sgs->sum_nr_running > sgs->group_capacity_factor) + if (sgs->group_no_capacity) return group_overloaded; if (sg_imbalanced(group)) @@ -6198,6 +6275,7 @@ static inline void update_sg_lb_stats(struct lb_env *env, load = source_load(i, load_idx); sgs->group_load += load; + sgs->group_usage += get_cpu_usage(i); sgs->sum_nr_running += rq->cfs.h_nr_running; if (rq->nr_running > 1) @@ -6220,11 +6298,9 @@ static inline void update_sg_lb_stats(struct lb_env *env, sgs->load_per_task = sgs->sum_weighted_load / sgs->sum_nr_running; sgs->group_weight = group->group_weight; - sgs->group_capacity_factor = sg_capacity_factor(env, group); - sgs->group_type = group_classify(group, sgs); - if (sgs->group_capacity_factor > sgs->sum_nr_running) - sgs->group_has_free_capacity = 1; + sgs->group_no_capacity = group_is_overloaded(env, sgs); + sgs->group_type = group_classify(env, group, sgs); } /** @@ -6346,18 +6422,19 @@ static inline void update_sd_lb_stats(struct lb_env *env, struct sd_lb_stats *sd /* * In case the child domain prefers tasks go to siblings - * first, lower the sg capacity factor to one so that we'll try + * first, lower the sg capacity so that we'll try * and move all the excess tasks away. We lower the capacity * of a group only if the local group has the capacity to fit - * these excess tasks, i.e. nr_running < group_capacity_factor. The - * extra check prevents the case where you always pull from the - * heaviest group when it is already under-utilized (possible - * with a large weight task outweighs the tasks on the system). + * these excess tasks. The extra check prevents the case where + * you always pull from the heaviest group when it is already + * under-utilized (possible with a large weight task outweighs + * the tasks on the system). */ if (prefer_sibling && sds->local && - sds->local_stat.group_has_free_capacity) { - sgs->group_capacity_factor = min(sgs->group_capacity_factor, 1U); - sgs->group_type = group_classify(sg, sgs); + group_has_capacity(env, &sds->local_stat) && + (sgs->sum_nr_running > 1)) { + sgs->group_no_capacity = 1; + sgs->group_type = group_overloaded; } if (update_sd_pick_busiest(env, sds, sg, sgs)) { @@ -6537,11 +6614,12 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s */ if (busiest->group_type == group_overloaded && local->group_type == group_overloaded) { - load_above_capacity = - (busiest->sum_nr_running - busiest->group_capacity_factor); - - load_above_capacity *= (SCHED_LOAD_SCALE * SCHED_CAPACITY_SCALE); - load_above_capacity /= busiest->group_capacity; + load_above_capacity = busiest->sum_nr_running * + SCHED_LOAD_SCALE; + if (load_above_capacity > busiest->group_capacity) + load_above_capacity -= busiest->group_capacity; + else + load_above_capacity = ~0UL; } /* @@ -6604,6 +6682,7 @@ static struct sched_group *find_busiest_group(struct lb_env *env) local = &sds.local_stat; busiest = &sds.busiest_stat; + /* ASYM feature bypasses nice load balance check */ if ((env->idle == CPU_IDLE || env->idle == CPU_NEWLY_IDLE) && check_asym_packing(env, &sds)) return sds.busiest; @@ -6624,8 +6703,8 @@ static struct sched_group *find_busiest_group(struct lb_env *env) goto force_balance; /* SD_BALANCE_NEWIDLE trumps SMP nice when underutilized */ - if (env->idle == CPU_NEWLY_IDLE && local->group_has_free_capacity && - !busiest->group_has_free_capacity) + if (env->idle == CPU_NEWLY_IDLE && group_has_capacity(env, local) && + busiest->group_no_capacity) goto force_balance; /* @@ -6684,7 +6763,7 @@ static struct rq *find_busiest_queue(struct lb_env *env, int i; for_each_cpu_and(i, sched_group_cpus(group), env->cpus) { - unsigned long capacity, capacity_factor, wl; + unsigned long capacity, wl; enum fbq_type rt; rq = cpu_rq(i); @@ -6713,9 +6792,6 @@ static struct rq *find_busiest_queue(struct lb_env *env, continue; capacity = capacity_of(i); - capacity_factor = DIV_ROUND_CLOSEST(capacity, SCHED_CAPACITY_SCALE); - if (!capacity_factor) - capacity_factor = fix_small_capacity(env->sd, group); wl = weighted_cpuload(i); @@ -6723,7 +6799,9 @@ static struct rq *find_busiest_queue(struct lb_env *env, * When comparing with imbalance, use weighted_cpuload() * which is not scaled with the cpu capacity. */ - if (capacity_factor && rq->nr_running == 1 && wl > env->imbalance) + + if (rq->nr_running == 1 && wl > env->imbalance && + !check_cpu_capacity(rq, env->sd)) continue; /* @@ -6771,6 +6849,19 @@ static int need_active_balance(struct lb_env *env) return 1; } + /* + * The dst_cpu is idle and the src_cpu CPU has only 1 CFS task. + * It's worth migrating the task if the src_cpu's capacity is reduced + * because of other sched_class or IRQs if more capacity stays + * available on dst_cpu. + */ + if ((env->idle != CPU_NOT_IDLE) && + (env->src_rq->cfs.h_nr_running == 1)) { + if ((check_cpu_capacity(env->src_rq, sd)) && + (capacity_of(env->src_cpu)*sd->imbalance_pct < capacity_of(env->dst_cpu)*100)) + return 1; + } + return unlikely(sd->nr_balance_failed > sd->cache_nice_tries+2); } @@ -6870,6 +6961,9 @@ redo: schedstat_add(sd, lb_imbalance[idle], env.imbalance); + env.src_cpu = busiest->cpu; + env.src_rq = busiest; + ld_moved = 0; if (busiest->nr_running > 1) { /* @@ -6879,8 +6973,6 @@ redo: * correctly treated as an imbalance. */ env.flags |= LBF_ALL_PINNED; - env.src_cpu = busiest->cpu; - env.src_rq = busiest; env.loop_max = min(sysctl_sched_nr_migrate, busiest->nr_running); more_balance: @@ -7580,22 +7672,25 @@ end: /* * Current heuristic for kicking the idle load balancer in the presence - * of an idle cpu is the system. + * of an idle cpu in the system. * - This rq has more than one task. - * - At any scheduler domain level, this cpu's scheduler group has multiple - * busy cpu's exceeding the group's capacity. + * - This rq has at least one CFS task and the capacity of the CPU is + * significantly reduced because of RT tasks or IRQs. + * - At parent of LLC scheduler domain level, this cpu's scheduler group has + * multiple busy cpu. * - For SD_ASYM_PACKING, if the lower numbered cpu's in the scheduler * domain span are idle. */ -static inline int nohz_kick_needed(struct rq *rq) +static inline bool nohz_kick_needed(struct rq *rq) { unsigned long now = jiffies; struct sched_domain *sd; struct sched_group_capacity *sgc; int nr_busy, cpu = rq->cpu; + bool kick = false; if (unlikely(rq->idle_balance)) - return 0; + return false; /* * We may be recently in ticked or tickless idle mode. At the first @@ -7609,38 +7704,46 @@ static inline int nohz_kick_needed(struct rq *rq) * balancing. */ if (likely(!atomic_read(&nohz.nr_cpus))) - return 0; + return false; if (time_before(now, nohz.next_balance)) - return 0; + return false; if (rq->nr_running >= 2) - goto need_kick; + return true; rcu_read_lock(); sd = rcu_dereference(per_cpu(sd_busy, cpu)); - if (sd) { sgc = sd->groups->sgc; nr_busy = atomic_read(&sgc->nr_busy_cpus); - if (nr_busy > 1) - goto need_kick_unlock; + if (nr_busy > 1) { + kick = true; + goto unlock; + } + } - sd = rcu_dereference(per_cpu(sd_asym, cpu)); + sd = rcu_dereference(rq->sd); + if (sd) { + if ((rq->cfs.h_nr_running >= 1) && + check_cpu_capacity(rq, sd)) { + kick = true; + goto unlock; + } + } + sd = rcu_dereference(per_cpu(sd_asym, cpu)); if (sd && (cpumask_first_and(nohz.idle_cpus_mask, - sched_domain_span(sd)) < cpu)) - goto need_kick_unlock; - - rcu_read_unlock(); - return 0; + sched_domain_span(sd)) < cpu)) { + kick = true; + goto unlock; + } -need_kick_unlock: +unlock: rcu_read_unlock(); -need_kick: - return 1; + return kick; } #else static void nohz_idle_balance(struct rq *this_rq, enum cpu_idle_type idle) { } @@ -7656,14 +7759,16 @@ static void run_rebalance_domains(struct softirq_action *h) enum cpu_idle_type idle = this_rq->idle_balance ? CPU_IDLE : CPU_NOT_IDLE; - rebalance_domains(this_rq, idle); - /* * If this cpu has a pending nohz_balance_kick, then do the * balancing on behalf of the other idle cpus whose ticks are - * stopped. + * stopped. Do nohz_idle_balance *before* rebalance_domains to + * give the idle cpus a chance to load balance. Else we may + * load balance only within the local sched_domain hierarchy + * and abort nohz_idle_balance altogether if we pull some load. */ nohz_idle_balance(this_rq, idle); + rebalance_domains(this_rq, idle); } /* diff --git a/kernel/sched/features.h b/kernel/sched/features.h index 90284d117fe6..91e33cd485f6 100644 --- a/kernel/sched/features.h +++ b/kernel/sched/features.h @@ -56,6 +56,19 @@ SCHED_FEAT(NONTASK_CAPACITY, true) */ SCHED_FEAT(TTWU_QUEUE, true) +#ifdef HAVE_RT_PUSH_IPI +/* + * In order to avoid a thundering herd attack of CPUs that are + * lowering their priorities at the same time, and there being + * a single CPU that has an RT task that can migrate and is waiting + * to run, where the other CPUs will try to take that CPUs + * rq lock and possibly create a large contention, sending an + * IPI to that CPU and let that CPU push the RT task to where + * it should go may be a better scenario. + */ +SCHED_FEAT(RT_PUSH_IPI, true) +#endif + SCHED_FEAT(FORCE_SD_OVERLAP, false) SCHED_FEAT(RT_RUNTIME_SHARE, true) SCHED_FEAT(LB_MIN, false) diff --git a/kernel/sched/idle.c b/kernel/sched/idle.c index 94b2d7b88a27..4d207d2abcbd 100644 --- a/kernel/sched/idle.c +++ b/kernel/sched/idle.c @@ -82,6 +82,7 @@ static void cpuidle_idle_call(void) struct cpuidle_driver *drv = cpuidle_get_cpu_driver(dev); int next_state, entered_state; unsigned int broadcast; + bool reflect; /* * Check if the idle task must be rescheduled. If it is the @@ -105,6 +106,9 @@ static void cpuidle_idle_call(void) */ rcu_idle_enter(); + if (cpuidle_not_available(drv, dev)) + goto use_default; + /* * Suspend-to-idle ("freeze") is a system state in which all user space * has been frozen, all I/O devices have been suspended and the only @@ -115,30 +119,24 @@ static void cpuidle_idle_call(void) * until a proper wakeup interrupt happens. */ if (idle_should_freeze()) { - cpuidle_enter_freeze(); - local_irq_enable(); - goto exit_idle; - } + entered_state = cpuidle_enter_freeze(drv, dev); + if (entered_state >= 0) { + local_irq_enable(); + goto exit_idle; + } - /* - * Ask the cpuidle framework to choose a convenient idle state. - * Fall back to the default arch idle method on errors. - */ - next_state = cpuidle_select(drv, dev); - if (next_state < 0) { -use_default: + reflect = false; + next_state = cpuidle_find_deepest_state(drv, dev); + } else { + reflect = true; /* - * We can't use the cpuidle framework, let's use the default - * idle routine. + * Ask the cpuidle framework to choose a convenient idle state. */ - if (current_clr_polling_and_test()) - local_irq_enable(); - else - arch_cpu_idle(); - - goto exit_idle; + next_state = cpuidle_select(drv, dev); } - + /* Fall back to the default arch idle method on errors. */ + if (next_state < 0) + goto use_default; /* * The idle task must be scheduled, it is pointless to @@ -160,8 +158,7 @@ use_default: * is used from another cpu as a broadcast timer, this call may * fail if it is not available */ - if (broadcast && - clockevents_notify(CLOCK_EVT_NOTIFY_BROADCAST_ENTER, &dev->cpu)) + if (broadcast && tick_broadcast_enter()) goto use_default; /* Take note of the planned idle state. */ @@ -178,12 +175,13 @@ use_default: idle_set_state(this_rq(), NULL); if (broadcast) - clockevents_notify(CLOCK_EVT_NOTIFY_BROADCAST_EXIT, &dev->cpu); + tick_broadcast_exit(); /* * Give the governor an opportunity to reflect on the outcome */ - cpuidle_reflect(dev, entered_state); + if (reflect) + cpuidle_reflect(dev, entered_state); exit_idle: __current_set_polling(); @@ -196,6 +194,19 @@ exit_idle: rcu_idle_exit(); start_critical_timings(); + return; + +use_default: + /* + * We can't use the cpuidle framework, let's use the default + * idle routine. + */ + if (current_clr_polling_and_test()) + local_irq_enable(); + else + arch_cpu_idle(); + + goto exit_idle; } /* diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c index f4d4b077eba0..575da76a3874 100644 --- a/kernel/sched/rt.c +++ b/kernel/sched/rt.c @@ -6,6 +6,7 @@ #include "sched.h" #include <linux/slab.h> +#include <linux/irq_work.h> int sched_rr_timeslice = RR_TIMESLICE; @@ -59,7 +60,11 @@ static void start_rt_bandwidth(struct rt_bandwidth *rt_b) raw_spin_unlock(&rt_b->rt_runtime_lock); } -void init_rt_rq(struct rt_rq *rt_rq, struct rq *rq) +#ifdef CONFIG_SMP +static void push_irq_work_func(struct irq_work *work); +#endif + +void init_rt_rq(struct rt_rq *rt_rq) { struct rt_prio_array *array; int i; @@ -78,7 +83,14 @@ void init_rt_rq(struct rt_rq *rt_rq, struct rq *rq) rt_rq->rt_nr_migratory = 0; rt_rq->overloaded = 0; plist_head_init(&rt_rq->pushable_tasks); + +#ifdef HAVE_RT_PUSH_IPI + rt_rq->push_flags = 0; + rt_rq->push_cpu = nr_cpu_ids; + raw_spin_lock_init(&rt_rq->push_lock); + init_irq_work(&rt_rq->push_work, push_irq_work_func); #endif +#endif /* CONFIG_SMP */ /* We start is dequeued state, because no RT tasks are queued */ rt_rq->rt_queued = 0; @@ -193,7 +205,7 @@ int alloc_rt_sched_group(struct task_group *tg, struct task_group *parent) if (!rt_se) goto err_free_rq; - init_rt_rq(rt_rq, cpu_rq(i)); + init_rt_rq(rt_rq); rt_rq->rt_runtime = tg->rt_bandwidth.rt_runtime; init_tg_rt_entry(tg, rt_rq, rt_se, i, parent->rt_se[i]); } @@ -1778,6 +1790,164 @@ static void push_rt_tasks(struct rq *rq) ; } +#ifdef HAVE_RT_PUSH_IPI +/* + * The search for the next cpu always starts at rq->cpu and ends + * when we reach rq->cpu again. It will never return rq->cpu. + * This returns the next cpu to check, or nr_cpu_ids if the loop + * is complete. + * + * rq->rt.push_cpu holds the last cpu returned by this function, + * or if this is the first instance, it must hold rq->cpu. + */ +static int rto_next_cpu(struct rq *rq) +{ + int prev_cpu = rq->rt.push_cpu; + int cpu; + + cpu = cpumask_next(prev_cpu, rq->rd->rto_mask); + + /* + * If the previous cpu is less than the rq's CPU, then it already + * passed the end of the mask, and has started from the beginning. + * We end if the next CPU is greater or equal to rq's CPU. + */ + if (prev_cpu < rq->cpu) { + if (cpu >= rq->cpu) + return nr_cpu_ids; + + } else if (cpu >= nr_cpu_ids) { + /* + * We passed the end of the mask, start at the beginning. + * If the result is greater or equal to the rq's CPU, then + * the loop is finished. + */ + cpu = cpumask_first(rq->rd->rto_mask); + if (cpu >= rq->cpu) + return nr_cpu_ids; + } + rq->rt.push_cpu = cpu; + + /* Return cpu to let the caller know if the loop is finished or not */ + return cpu; +} + +static int find_next_push_cpu(struct rq *rq) +{ + struct rq *next_rq; + int cpu; + + while (1) { + cpu = rto_next_cpu(rq); + if (cpu >= nr_cpu_ids) + break; + next_rq = cpu_rq(cpu); + + /* Make sure the next rq can push to this rq */ + if (next_rq->rt.highest_prio.next < rq->rt.highest_prio.curr) + break; + } + + return cpu; +} + +#define RT_PUSH_IPI_EXECUTING 1 +#define RT_PUSH_IPI_RESTART 2 + +static void tell_cpu_to_push(struct rq *rq) +{ + int cpu; + + if (rq->rt.push_flags & RT_PUSH_IPI_EXECUTING) { + raw_spin_lock(&rq->rt.push_lock); + /* Make sure it's still executing */ + if (rq->rt.push_flags & RT_PUSH_IPI_EXECUTING) { + /* + * Tell the IPI to restart the loop as things have + * changed since it started. + */ + rq->rt.push_flags |= RT_PUSH_IPI_RESTART; + raw_spin_unlock(&rq->rt.push_lock); + return; + } + raw_spin_unlock(&rq->rt.push_lock); + } + + /* When here, there's no IPI going around */ + + rq->rt.push_cpu = rq->cpu; + cpu = find_next_push_cpu(rq); + if (cpu >= nr_cpu_ids) + return; + + rq->rt.push_flags = RT_PUSH_IPI_EXECUTING; + + irq_work_queue_on(&rq->rt.push_work, cpu); +} + +/* Called from hardirq context */ +static void try_to_push_tasks(void *arg) +{ + struct rt_rq *rt_rq = arg; + struct rq *rq, *src_rq; + int this_cpu; + int cpu; + + this_cpu = rt_rq->push_cpu; + + /* Paranoid check */ + BUG_ON(this_cpu != smp_processor_id()); + + rq = cpu_rq(this_cpu); + src_rq = rq_of_rt_rq(rt_rq); + +again: + if (has_pushable_tasks(rq)) { + raw_spin_lock(&rq->lock); + push_rt_task(rq); + raw_spin_unlock(&rq->lock); + } + + /* Pass the IPI to the next rt overloaded queue */ + raw_spin_lock(&rt_rq->push_lock); + /* + * If the source queue changed since the IPI went out, + * we need to restart the search from that CPU again. + */ + if (rt_rq->push_flags & RT_PUSH_IPI_RESTART) { + rt_rq->push_flags &= ~RT_PUSH_IPI_RESTART; + rt_rq->push_cpu = src_rq->cpu; + } + + cpu = find_next_push_cpu(src_rq); + + if (cpu >= nr_cpu_ids) + rt_rq->push_flags &= ~RT_PUSH_IPI_EXECUTING; + raw_spin_unlock(&rt_rq->push_lock); + + if (cpu >= nr_cpu_ids) + return; + + /* + * It is possible that a restart caused this CPU to be + * chosen again. Don't bother with an IPI, just see if we + * have more to push. + */ + if (unlikely(cpu == rq->cpu)) + goto again; + + /* Try the next RT overloaded CPU */ + irq_work_queue_on(&rt_rq->push_work, cpu); +} + +static void push_irq_work_func(struct irq_work *work) +{ + struct rt_rq *rt_rq = container_of(work, struct rt_rq, push_work); + + try_to_push_tasks(rt_rq); +} +#endif /* HAVE_RT_PUSH_IPI */ + static int pull_rt_task(struct rq *this_rq) { int this_cpu = this_rq->cpu, ret = 0, cpu; @@ -1793,6 +1963,13 @@ static int pull_rt_task(struct rq *this_rq) */ smp_rmb(); +#ifdef HAVE_RT_PUSH_IPI + if (sched_feat(RT_PUSH_IPI)) { + tell_cpu_to_push(this_rq); + return 0; + } +#endif + for_each_cpu(cpu, this_rq->rd->rto_mask) { if (this_cpu == cpu) continue; diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index dc0f435a2779..e0e129993958 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -6,6 +6,7 @@ #include <linux/mutex.h> #include <linux/spinlock.h> #include <linux/stop_machine.h> +#include <linux/irq_work.h> #include <linux/tick.h> #include <linux/slab.h> @@ -362,8 +363,14 @@ struct cfs_rq { * Under CFS, load is tracked on a per-entity basis and aggregated up. * This allows for the description of both thread and group usage (in * the FAIR_GROUP_SCHED case). + * runnable_load_avg is the sum of the load_avg_contrib of the + * sched_entities on the rq. + * blocked_load_avg is similar to runnable_load_avg except that its + * the blocked sched_entities on the rq. + * utilization_load_avg is the sum of the average running time of the + * sched_entities on the rq. */ - unsigned long runnable_load_avg, blocked_load_avg; + unsigned long runnable_load_avg, blocked_load_avg, utilization_load_avg; atomic64_t decay_counter; u64 last_decay; atomic_long_t removed_load; @@ -418,6 +425,11 @@ static inline int rt_bandwidth_enabled(void) return sysctl_sched_rt_runtime >= 0; } +/* RT IPI pull logic requires IRQ_WORK */ +#ifdef CONFIG_IRQ_WORK +# define HAVE_RT_PUSH_IPI +#endif + /* Real-Time classes' related field in a runqueue: */ struct rt_rq { struct rt_prio_array active; @@ -435,7 +447,13 @@ struct rt_rq { unsigned long rt_nr_total; int overloaded; struct plist_head pushable_tasks; +#ifdef HAVE_RT_PUSH_IPI + int push_flags; + int push_cpu; + struct irq_work push_work; + raw_spinlock_t push_lock; #endif +#endif /* CONFIG_SMP */ int rt_queued; int rt_throttled; @@ -597,6 +615,7 @@ struct rq { struct sched_domain *sd; unsigned long cpu_capacity; + unsigned long cpu_capacity_orig; unsigned char idle_balance; /* For active balancing */ @@ -807,7 +826,7 @@ struct sched_group_capacity { * CPU capacity of this group, SCHED_LOAD_SCALE being max capacity * for a single CPU. */ - unsigned int capacity, capacity_orig; + unsigned int capacity; unsigned long next_update; int imbalance; /* XXX unrelated to capacity but shared group state */ /* @@ -1368,9 +1387,18 @@ static inline int hrtick_enabled(struct rq *rq) #ifdef CONFIG_SMP extern void sched_avg_update(struct rq *rq); + +#ifndef arch_scale_freq_capacity +static __always_inline +unsigned long arch_scale_freq_capacity(struct sched_domain *sd, int cpu) +{ + return SCHED_CAPACITY_SCALE; +} +#endif + static inline void sched_rt_avg_update(struct rq *rq, u64 rt_delta) { - rq->rt_avg += rt_delta; + rq->rt_avg += rt_delta * arch_scale_freq_capacity(NULL, cpu_of(rq)); sched_avg_update(rq); } #else @@ -1643,8 +1671,8 @@ extern void print_rt_stats(struct seq_file *m, int cpu); extern void print_dl_stats(struct seq_file *m, int cpu); extern void init_cfs_rq(struct cfs_rq *cfs_rq); -extern void init_rt_rq(struct rt_rq *rt_rq, struct rq *rq); -extern void init_dl_rq(struct dl_rq *dl_rq, struct rq *rq); +extern void init_rt_rq(struct rt_rq *rt_rq); +extern void init_dl_rq(struct dl_rq *dl_rq); extern void cfs_bandwidth_usage_inc(void); extern void cfs_bandwidth_usage_dec(void); diff --git a/kernel/sys.c b/kernel/sys.c index 667b2e62fad2..a03d9cd23ed7 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -1108,6 +1108,7 @@ DECLARE_RWSEM(uts_sem); /* * Work around broken programs that cannot handle "Linux 3.0". * Instead we map 3.x to 2.6.40+x, so e.g. 3.0 would be 2.6.40 + * And we map 4.x to 2.6.60+x, so 4.0 would be 2.6.60. */ static int override_release(char __user *release, size_t len) { @@ -1127,7 +1128,7 @@ static int override_release(char __user *release, size_t len) break; rest++; } - v = ((LINUX_VERSION_CODE >> 8) & 0xff) + 40; + v = ((LINUX_VERSION_CODE >> 8) & 0xff) + 60; copy = clamp_t(size_t, len, 1, sizeof(buf)); copy = scnprintf(buf, copy, "2.6.%u%s", v, rest); ret = copy_to_user(release, buf, copy + 1); diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 88ea2d6e0031..ce410bb9f2e1 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -1228,6 +1228,14 @@ static struct ctl_table vm_table[] = { .extra1 = &zero, }, { + .procname = "dirtytime_expire_seconds", + .data = &dirtytime_expire_interval, + .maxlen = sizeof(dirty_expire_interval), + .mode = 0644, + .proc_handler = dirtytime_interval_handler, + .extra1 = &zero, + }, + { .procname = "nr_pdflush_threads", .mode = 0444 /* read-only */, .proc_handler = pdflush_proc_obsolete, diff --git a/kernel/time/Kconfig b/kernel/time/Kconfig index d626dc98e8df..579ce1b929af 100644 --- a/kernel/time/Kconfig +++ b/kernel/time/Kconfig @@ -33,12 +33,6 @@ config ARCH_USES_GETTIMEOFFSET config GENERIC_CLOCKEVENTS bool -# Migration helper. Builds, but does not invoke -config GENERIC_CLOCKEVENTS_BUILD - bool - default y - depends on GENERIC_CLOCKEVENTS - # Architecture can handle broadcast in a driver-agnostic way config ARCH_HAS_TICK_BROADCAST bool diff --git a/kernel/time/Makefile b/kernel/time/Makefile index c09c07817d7a..01f0312419b3 100644 --- a/kernel/time/Makefile +++ b/kernel/time/Makefile @@ -2,15 +2,13 @@ obj-y += time.o timer.o hrtimer.o itimer.o posix-timers.o posix-cpu-timers.o obj-y += timekeeping.o ntp.o clocksource.o jiffies.o timer_list.o obj-y += timeconv.o timecounter.o posix-clock.o alarmtimer.o -obj-$(CONFIG_GENERIC_CLOCKEVENTS_BUILD) += clockevents.o -obj-$(CONFIG_GENERIC_CLOCKEVENTS) += tick-common.o +obj-$(CONFIG_GENERIC_CLOCKEVENTS) += clockevents.o tick-common.o ifeq ($(CONFIG_GENERIC_CLOCKEVENTS_BROADCAST),y) obj-y += tick-broadcast.o obj-$(CONFIG_TICK_ONESHOT) += tick-broadcast-hrtimer.o endif obj-$(CONFIG_GENERIC_SCHED_CLOCK) += sched_clock.o -obj-$(CONFIG_TICK_ONESHOT) += tick-oneshot.o -obj-$(CONFIG_TICK_ONESHOT) += tick-sched.o +obj-$(CONFIG_TICK_ONESHOT) += tick-oneshot.o tick-sched.o obj-$(CONFIG_TIMER_STATS) += timer_stats.o obj-$(CONFIG_DEBUG_FS) += timekeeping_debug.o obj-$(CONFIG_TEST_UDELAY) += test_udelay.o diff --git a/kernel/time/clockevents.c b/kernel/time/clockevents.c index 55449909f114..25d942d1da27 100644 --- a/kernel/time/clockevents.c +++ b/kernel/time/clockevents.c @@ -94,25 +94,76 @@ u64 clockevent_delta2ns(unsigned long latch, struct clock_event_device *evt) } EXPORT_SYMBOL_GPL(clockevent_delta2ns); +static int __clockevents_set_state(struct clock_event_device *dev, + enum clock_event_state state) +{ + /* Transition with legacy set_mode() callback */ + if (dev->set_mode) { + /* Legacy callback doesn't support new modes */ + if (state > CLOCK_EVT_STATE_ONESHOT) + return -ENOSYS; + /* + * 'clock_event_state' and 'clock_event_mode' have 1-to-1 + * mapping until *_ONESHOT, and so a simple cast will work. + */ + dev->set_mode((enum clock_event_mode)state, dev); + dev->mode = (enum clock_event_mode)state; + return 0; + } + + if (dev->features & CLOCK_EVT_FEAT_DUMMY) + return 0; + + /* Transition with new state-specific callbacks */ + switch (state) { + case CLOCK_EVT_STATE_DETACHED: + /* + * This is an internal state, which is guaranteed to go from + * SHUTDOWN to DETACHED. No driver interaction required. + */ + return 0; + + case CLOCK_EVT_STATE_SHUTDOWN: + return dev->set_state_shutdown(dev); + + case CLOCK_EVT_STATE_PERIODIC: + /* Core internal bug */ + if (!(dev->features & CLOCK_EVT_FEAT_PERIODIC)) + return -ENOSYS; + return dev->set_state_periodic(dev); + + case CLOCK_EVT_STATE_ONESHOT: + /* Core internal bug */ + if (!(dev->features & CLOCK_EVT_FEAT_ONESHOT)) + return -ENOSYS; + return dev->set_state_oneshot(dev); + + default: + return -ENOSYS; + } +} + /** - * clockevents_set_mode - set the operating mode of a clock event device + * clockevents_set_state - set the operating state of a clock event device * @dev: device to modify - * @mode: new mode + * @state: new state * * Must be called with interrupts disabled ! */ -void clockevents_set_mode(struct clock_event_device *dev, - enum clock_event_mode mode) +void clockevents_set_state(struct clock_event_device *dev, + enum clock_event_state state) { - if (dev->mode != mode) { - dev->set_mode(mode, dev); - dev->mode = mode; + if (dev->state != state) { + if (__clockevents_set_state(dev, state)) + return; + + dev->state = state; /* * A nsec2cyc multiplicator of 0 is invalid and we'd crash * on it, so fix it up and emit a warning: */ - if (mode == CLOCK_EVT_MODE_ONESHOT) { + if (state == CLOCK_EVT_STATE_ONESHOT) { if (unlikely(!dev->mult)) { dev->mult = 1; WARN_ON(1); @@ -127,10 +178,28 @@ void clockevents_set_mode(struct clock_event_device *dev, */ void clockevents_shutdown(struct clock_event_device *dev) { - clockevents_set_mode(dev, CLOCK_EVT_MODE_SHUTDOWN); + clockevents_set_state(dev, CLOCK_EVT_STATE_SHUTDOWN); dev->next_event.tv64 = KTIME_MAX; } +/** + * clockevents_tick_resume - Resume the tick device before using it again + * @dev: device to resume + */ +int clockevents_tick_resume(struct clock_event_device *dev) +{ + int ret = 0; + + if (dev->set_mode) { + dev->set_mode(CLOCK_EVT_MODE_RESUME, dev); + dev->mode = CLOCK_EVT_MODE_RESUME; + } else if (dev->tick_resume) { + ret = dev->tick_resume(dev); + } + + return ret; +} + #ifdef CONFIG_GENERIC_CLOCKEVENTS_MIN_ADJUST /* Limit min_delta to a jiffie */ @@ -183,7 +252,7 @@ static int clockevents_program_min_delta(struct clock_event_device *dev) delta = dev->min_delta_ns; dev->next_event = ktime_add_ns(ktime_get(), delta); - if (dev->mode == CLOCK_EVT_MODE_SHUTDOWN) + if (dev->state == CLOCK_EVT_STATE_SHUTDOWN) return 0; dev->retries++; @@ -220,7 +289,7 @@ static int clockevents_program_min_delta(struct clock_event_device *dev) delta = dev->min_delta_ns; dev->next_event = ktime_add_ns(ktime_get(), delta); - if (dev->mode == CLOCK_EVT_MODE_SHUTDOWN) + if (dev->state == CLOCK_EVT_STATE_SHUTDOWN) return 0; dev->retries++; @@ -252,7 +321,7 @@ int clockevents_program_event(struct clock_event_device *dev, ktime_t expires, dev->next_event = expires; - if (dev->mode == CLOCK_EVT_MODE_SHUTDOWN) + if (dev->state == CLOCK_EVT_STATE_SHUTDOWN) return 0; /* Shortcut for clockevent devices that can deal with ktime. */ @@ -297,7 +366,7 @@ static int clockevents_replace(struct clock_event_device *ced) struct clock_event_device *dev, *newdev = NULL; list_for_each_entry(dev, &clockevent_devices, list) { - if (dev == ced || dev->mode != CLOCK_EVT_MODE_UNUSED) + if (dev == ced || dev->state != CLOCK_EVT_STATE_DETACHED) continue; if (!tick_check_replacement(newdev, dev)) @@ -323,7 +392,7 @@ static int clockevents_replace(struct clock_event_device *ced) static int __clockevents_try_unbind(struct clock_event_device *ced, int cpu) { /* Fast track. Device is unused */ - if (ced->mode == CLOCK_EVT_MODE_UNUSED) { + if (ced->state == CLOCK_EVT_STATE_DETACHED) { list_del_init(&ced->list); return 0; } @@ -373,6 +442,37 @@ int clockevents_unbind_device(struct clock_event_device *ced, int cpu) } EXPORT_SYMBOL_GPL(clockevents_unbind); +/* Sanity check of state transition callbacks */ +static int clockevents_sanity_check(struct clock_event_device *dev) +{ + /* Legacy set_mode() callback */ + if (dev->set_mode) { + /* We shouldn't be supporting new modes now */ + WARN_ON(dev->set_state_periodic || dev->set_state_oneshot || + dev->set_state_shutdown || dev->tick_resume); + + BUG_ON(dev->mode != CLOCK_EVT_MODE_UNUSED); + return 0; + } + + if (dev->features & CLOCK_EVT_FEAT_DUMMY) + return 0; + + /* New state-specific callbacks */ + if (!dev->set_state_shutdown) + return -EINVAL; + + if ((dev->features & CLOCK_EVT_FEAT_PERIODIC) && + !dev->set_state_periodic) + return -EINVAL; + + if ((dev->features & CLOCK_EVT_FEAT_ONESHOT) && + !dev->set_state_oneshot) + return -EINVAL; + + return 0; +} + /** * clockevents_register_device - register a clock event device * @dev: device to register @@ -381,7 +481,11 @@ void clockevents_register_device(struct clock_event_device *dev) { unsigned long flags; - BUG_ON(dev->mode != CLOCK_EVT_MODE_UNUSED); + BUG_ON(clockevents_sanity_check(dev)); + + /* Initialize state to DETACHED */ + dev->state = CLOCK_EVT_STATE_DETACHED; + if (!dev->cpumask) { WARN_ON(num_possible_cpus() > 1); dev->cpumask = cpumask_of(smp_processor_id()); @@ -445,11 +549,11 @@ int __clockevents_update_freq(struct clock_event_device *dev, u32 freq) { clockevents_config(dev, freq); - if (dev->mode == CLOCK_EVT_MODE_ONESHOT) + if (dev->state == CLOCK_EVT_STATE_ONESHOT) return clockevents_program_event(dev, dev->next_event, false); - if (dev->mode == CLOCK_EVT_MODE_PERIODIC) - dev->set_mode(CLOCK_EVT_MODE_PERIODIC, dev); + if (dev->state == CLOCK_EVT_STATE_PERIODIC) + return __clockevents_set_state(dev, CLOCK_EVT_STATE_PERIODIC); return 0; } @@ -491,30 +595,27 @@ void clockevents_handle_noop(struct clock_event_device *dev) * @old: device to release (can be NULL) * @new: device to request (can be NULL) * - * Called from the notifier chain. clockevents_lock is held already + * Called from various tick functions with clockevents_lock held and + * interrupts disabled. */ void clockevents_exchange_device(struct clock_event_device *old, struct clock_event_device *new) { - unsigned long flags; - - local_irq_save(flags); /* * Caller releases a clock event device. We queue it into the * released list and do a notify add later. */ if (old) { module_put(old->owner); - clockevents_set_mode(old, CLOCK_EVT_MODE_UNUSED); + clockevents_set_state(old, CLOCK_EVT_STATE_DETACHED); list_del(&old->list); list_add(&old->list, &clockevents_released); } if (new) { - BUG_ON(new->mode != CLOCK_EVT_MODE_UNUSED); + BUG_ON(new->state != CLOCK_EVT_STATE_DETACHED); clockevents_shutdown(new); } - local_irq_restore(flags); } /** @@ -541,74 +642,40 @@ void clockevents_resume(void) dev->resume(dev); } -#ifdef CONFIG_GENERIC_CLOCKEVENTS +#ifdef CONFIG_HOTPLUG_CPU /** - * clockevents_notify - notification about relevant events - * Returns 0 on success, any other value on error + * tick_cleanup_dead_cpu - Cleanup the tick and clockevents of a dead cpu */ -int clockevents_notify(unsigned long reason, void *arg) +void tick_cleanup_dead_cpu(int cpu) { struct clock_event_device *dev, *tmp; unsigned long flags; - int cpu, ret = 0; raw_spin_lock_irqsave(&clockevents_lock, flags); - switch (reason) { - case CLOCK_EVT_NOTIFY_BROADCAST_ON: - case CLOCK_EVT_NOTIFY_BROADCAST_OFF: - case CLOCK_EVT_NOTIFY_BROADCAST_FORCE: - tick_broadcast_on_off(reason, arg); - break; - - case CLOCK_EVT_NOTIFY_BROADCAST_ENTER: - case CLOCK_EVT_NOTIFY_BROADCAST_EXIT: - ret = tick_broadcast_oneshot_control(reason); - break; - - case CLOCK_EVT_NOTIFY_CPU_DYING: - tick_handover_do_timer(arg); - break; - - case CLOCK_EVT_NOTIFY_SUSPEND: - tick_suspend(); - tick_suspend_broadcast(); - break; - - case CLOCK_EVT_NOTIFY_RESUME: - tick_resume(); - break; - - case CLOCK_EVT_NOTIFY_CPU_DEAD: - tick_shutdown_broadcast_oneshot(arg); - tick_shutdown_broadcast(arg); - tick_shutdown(arg); - /* - * Unregister the clock event devices which were - * released from the users in the notify chain. - */ - list_for_each_entry_safe(dev, tmp, &clockevents_released, list) + tick_shutdown_broadcast_oneshot(cpu); + tick_shutdown_broadcast(cpu); + tick_shutdown(cpu); + /* + * Unregister the clock event devices which were + * released from the users in the notify chain. + */ + list_for_each_entry_safe(dev, tmp, &clockevents_released, list) + list_del(&dev->list); + /* + * Now check whether the CPU has left unused per cpu devices + */ + list_for_each_entry_safe(dev, tmp, &clockevent_devices, list) { + if (cpumask_test_cpu(cpu, dev->cpumask) && + cpumask_weight(dev->cpumask) == 1 && + !tick_is_broadcast_device(dev)) { + BUG_ON(dev->state != CLOCK_EVT_STATE_DETACHED); list_del(&dev->list); - /* - * Now check whether the CPU has left unused per cpu devices - */ - cpu = *((int *)arg); - list_for_each_entry_safe(dev, tmp, &clockevent_devices, list) { - if (cpumask_test_cpu(cpu, dev->cpumask) && - cpumask_weight(dev->cpumask) == 1 && - !tick_is_broadcast_device(dev)) { - BUG_ON(dev->mode != CLOCK_EVT_MODE_UNUSED); - list_del(&dev->list); - } } - break; - default: - break; } raw_spin_unlock_irqrestore(&clockevents_lock, flags); - return ret; } -EXPORT_SYMBOL_GPL(clockevents_notify); +#endif #ifdef CONFIG_SYSFS struct bus_type clockevents_subsys = { @@ -727,5 +794,3 @@ static int __init clockevents_init_sysfs(void) } device_initcall(clockevents_init_sysfs); #endif /* SYSFS */ - -#endif /* GENERIC_CLOCK_EVENTS */ diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c index 4892352f0e49..15facb1b9c60 100644 --- a/kernel/time/clocksource.c +++ b/kernel/time/clocksource.c @@ -142,13 +142,6 @@ static void __clocksource_unstable(struct clocksource *cs) schedule_work(&watchdog_work); } -static void clocksource_unstable(struct clocksource *cs, int64_t delta) -{ - printk(KERN_WARNING "Clocksource %s unstable (delta = %Ld ns)\n", - cs->name, delta); - __clocksource_unstable(cs); -} - /** * clocksource_mark_unstable - mark clocksource unstable via watchdog * @cs: clocksource to be marked unstable @@ -174,7 +167,7 @@ void clocksource_mark_unstable(struct clocksource *cs) static void clocksource_watchdog(unsigned long data) { struct clocksource *cs; - cycle_t csnow, wdnow, delta; + cycle_t csnow, wdnow, cslast, wdlast, delta; int64_t wd_nsec, cs_nsec; int next_cpu, reset_pending; @@ -213,6 +206,8 @@ static void clocksource_watchdog(unsigned long data) delta = clocksource_delta(csnow, cs->cs_last, cs->mask); cs_nsec = clocksource_cyc2ns(delta, cs->mult, cs->shift); + wdlast = cs->wd_last; /* save these in case we print them */ + cslast = cs->cs_last; cs->cs_last = csnow; cs->wd_last = wdnow; @@ -221,7 +216,12 @@ static void clocksource_watchdog(unsigned long data) /* Check the deviation from the watchdog clocksource. */ if ((abs(cs_nsec - wd_nsec) > WATCHDOG_THRESHOLD)) { - clocksource_unstable(cs, cs_nsec - wd_nsec); + pr_warn("timekeeping watchdog: Marking clocksource '%s' as unstable, because the skew is too large:\n", cs->name); + pr_warn(" '%s' wd_now: %llx wd_last: %llx mask: %llx\n", + watchdog->name, wdnow, wdlast, watchdog->mask); + pr_warn(" '%s' cs_now: %llx cs_last: %llx mask: %llx\n", + cs->name, csnow, cslast, cs->mask); + __clocksource_unstable(cs); continue; } @@ -469,26 +469,25 @@ static u32 clocksource_max_adjustment(struct clocksource *cs) * @shift: cycle to nanosecond divisor (power of two) * @maxadj: maximum adjustment value to mult (~11%) * @mask: bitmask for two's complement subtraction of non 64 bit counters + * @max_cyc: maximum cycle value before potential overflow (does not include + * any safety margin) + * + * NOTE: This function includes a safety margin of 50%, in other words, we + * return half the number of nanoseconds the hardware counter can technically + * cover. This is done so that we can potentially detect problems caused by + * delayed timers or bad hardware, which might result in time intervals that + * are larger then what the math used can handle without overflows. */ -u64 clocks_calc_max_nsecs(u32 mult, u32 shift, u32 maxadj, u64 mask) +u64 clocks_calc_max_nsecs(u32 mult, u32 shift, u32 maxadj, u64 mask, u64 *max_cyc) { u64 max_nsecs, max_cycles; /* * Calculate the maximum number of cycles that we can pass to the - * cyc2ns function without overflowing a 64-bit signed result. The - * maximum number of cycles is equal to ULLONG_MAX/(mult+maxadj) - * which is equivalent to the below. - * max_cycles < (2^63)/(mult + maxadj) - * max_cycles < 2^(log2((2^63)/(mult + maxadj))) - * max_cycles < 2^(log2(2^63) - log2(mult + maxadj)) - * max_cycles < 2^(63 - log2(mult + maxadj)) - * max_cycles < 1 << (63 - log2(mult + maxadj)) - * Please note that we add 1 to the result of the log2 to account for - * any rounding errors, ensure the above inequality is satisfied and - * no overflow will occur. + * cyc2ns() function without overflowing a 64-bit result. */ - max_cycles = 1ULL << (63 - (ilog2(mult + maxadj) + 1)); + max_cycles = ULLONG_MAX; + do_div(max_cycles, mult+maxadj); /* * The actual maximum number of cycles we can defer the clocksource is @@ -499,27 +498,26 @@ u64 clocks_calc_max_nsecs(u32 mult, u32 shift, u32 maxadj, u64 mask) max_cycles = min(max_cycles, mask); max_nsecs = clocksource_cyc2ns(max_cycles, mult - maxadj, shift); + /* return the max_cycles value as well if requested */ + if (max_cyc) + *max_cyc = max_cycles; + + /* Return 50% of the actual maximum, so we can detect bad values */ + max_nsecs >>= 1; + return max_nsecs; } /** - * clocksource_max_deferment - Returns max time the clocksource can be deferred - * @cs: Pointer to clocksource + * clocksource_update_max_deferment - Updates the clocksource max_idle_ns & max_cycles + * @cs: Pointer to clocksource to be updated * */ -static u64 clocksource_max_deferment(struct clocksource *cs) +static inline void clocksource_update_max_deferment(struct clocksource *cs) { - u64 max_nsecs; - - max_nsecs = clocks_calc_max_nsecs(cs->mult, cs->shift, cs->maxadj, - cs->mask); - /* - * To ensure that the clocksource does not wrap whilst we are idle, - * limit the time the clocksource can be deferred by 12.5%. Please - * note a margin of 12.5% is used because this can be computed with - * a shift, versus say 10% which would require division. - */ - return max_nsecs - (max_nsecs >> 3); + cs->max_idle_ns = clocks_calc_max_nsecs(cs->mult, cs->shift, + cs->maxadj, cs->mask, + &cs->max_cycles); } #ifndef CONFIG_ARCH_USES_GETTIMEOFFSET @@ -648,7 +646,7 @@ static void clocksource_enqueue(struct clocksource *cs) } /** - * __clocksource_updatefreq_scale - Used update clocksource with new freq + * __clocksource_update_freq_scale - Used update clocksource with new freq * @cs: clocksource to be registered * @scale: Scale factor multiplied against freq to get clocksource hz * @freq: clocksource frequency (cycles per second) divided by scale @@ -656,48 +654,64 @@ static void clocksource_enqueue(struct clocksource *cs) * This should only be called from the clocksource->enable() method. * * This *SHOULD NOT* be called directly! Please use the - * clocksource_updatefreq_hz() or clocksource_updatefreq_khz helper functions. + * __clocksource_update_freq_hz() or __clocksource_update_freq_khz() helper + * functions. */ -void __clocksource_updatefreq_scale(struct clocksource *cs, u32 scale, u32 freq) +void __clocksource_update_freq_scale(struct clocksource *cs, u32 scale, u32 freq) { u64 sec; + /* - * Calc the maximum number of seconds which we can run before - * wrapping around. For clocksources which have a mask > 32bit - * we need to limit the max sleep time to have a good - * conversion precision. 10 minutes is still a reasonable - * amount. That results in a shift value of 24 for a - * clocksource with mask >= 40bit and f >= 4GHz. That maps to - * ~ 0.06ppm granularity for NTP. We apply the same 12.5% - * margin as we do in clocksource_max_deferment() + * Default clocksources are *special* and self-define their mult/shift. + * But, you're not special, so you should specify a freq value. */ - sec = (cs->mask - (cs->mask >> 3)); - do_div(sec, freq); - do_div(sec, scale); - if (!sec) - sec = 1; - else if (sec > 600 && cs->mask > UINT_MAX) - sec = 600; - - clocks_calc_mult_shift(&cs->mult, &cs->shift, freq, - NSEC_PER_SEC / scale, sec * scale); - + if (freq) { + /* + * Calc the maximum number of seconds which we can run before + * wrapping around. For clocksources which have a mask > 32-bit + * we need to limit the max sleep time to have a good + * conversion precision. 10 minutes is still a reasonable + * amount. That results in a shift value of 24 for a + * clocksource with mask >= 40-bit and f >= 4GHz. That maps to + * ~ 0.06ppm granularity for NTP. + */ + sec = cs->mask; + do_div(sec, freq); + do_div(sec, scale); + if (!sec) + sec = 1; + else if (sec > 600 && cs->mask > UINT_MAX) + sec = 600; + + clocks_calc_mult_shift(&cs->mult, &cs->shift, freq, + NSEC_PER_SEC / scale, sec * scale); + } /* - * for clocksources that have large mults, to avoid overflow. - * Since mult may be adjusted by ntp, add an safety extra margin - * + * Ensure clocksources that have large 'mult' values don't overflow + * when adjusted. */ cs->maxadj = clocksource_max_adjustment(cs); - while ((cs->mult + cs->maxadj < cs->mult) - || (cs->mult - cs->maxadj > cs->mult)) { + while (freq && ((cs->mult + cs->maxadj < cs->mult) + || (cs->mult - cs->maxadj > cs->mult))) { cs->mult >>= 1; cs->shift--; cs->maxadj = clocksource_max_adjustment(cs); } - cs->max_idle_ns = clocksource_max_deferment(cs); + /* + * Only warn for *special* clocksources that self-define + * their mult/shift values and don't specify a freq. + */ + WARN_ONCE(cs->mult + cs->maxadj < cs->mult, + "timekeeping: Clocksource %s might overflow on 11%% adjustment\n", + cs->name); + + clocksource_update_max_deferment(cs); + + pr_info("clocksource %s: mask: 0x%llx max_cycles: 0x%llx, max_idle_ns: %lld ns\n", + cs->name, cs->mask, cs->max_cycles, cs->max_idle_ns); } -EXPORT_SYMBOL_GPL(__clocksource_updatefreq_scale); +EXPORT_SYMBOL_GPL(__clocksource_update_freq_scale); /** * __clocksource_register_scale - Used to install new clocksources @@ -714,7 +728,7 @@ int __clocksource_register_scale(struct clocksource *cs, u32 scale, u32 freq) { /* Initialize mult/shift and max_idle_ns */ - __clocksource_updatefreq_scale(cs, scale, freq); + __clocksource_update_freq_scale(cs, scale, freq); /* Add clocksource to the clocksource list */ mutex_lock(&clocksource_mutex); @@ -726,33 +740,6 @@ int __clocksource_register_scale(struct clocksource *cs, u32 scale, u32 freq) } EXPORT_SYMBOL_GPL(__clocksource_register_scale); - -/** - * clocksource_register - Used to install new clocksources - * @cs: clocksource to be registered - * - * Returns -EBUSY if registration fails, zero otherwise. - */ -int clocksource_register(struct clocksource *cs) -{ - /* calculate max adjustment for given mult/shift */ - cs->maxadj = clocksource_max_adjustment(cs); - WARN_ONCE(cs->mult + cs->maxadj < cs->mult, - "Clocksource %s might overflow on 11%% adjustment\n", - cs->name); - - /* calculate max idle time permitted for this clocksource */ - cs->max_idle_ns = clocksource_max_deferment(cs); - - mutex_lock(&clocksource_mutex); - clocksource_enqueue(cs); - clocksource_enqueue_watchdog(cs); - clocksource_select(); - mutex_unlock(&clocksource_mutex); - return 0; -} -EXPORT_SYMBOL(clocksource_register); - static void __clocksource_change_rating(struct clocksource *cs, int rating) { list_del(&cs->list); diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c index bee0c1f78091..76d4bd962b19 100644 --- a/kernel/time/hrtimer.c +++ b/kernel/time/hrtimer.c @@ -54,7 +54,7 @@ #include <trace/events/timer.h> -#include "timekeeping.h" +#include "tick-internal.h" /* * The timer bases: @@ -1707,17 +1707,10 @@ static int hrtimer_cpu_notify(struct notifier_block *self, break; #ifdef CONFIG_HOTPLUG_CPU - case CPU_DYING: - case CPU_DYING_FROZEN: - clockevents_notify(CLOCK_EVT_NOTIFY_CPU_DYING, &scpu); - break; case CPU_DEAD: case CPU_DEAD_FROZEN: - { - clockevents_notify(CLOCK_EVT_NOTIFY_CPU_DEAD, &scpu); migrate_hrtimers(scpu); break; - } #endif default: diff --git a/kernel/time/jiffies.c b/kernel/time/jiffies.c index a6a5bf53e86d..347fecf86a3f 100644 --- a/kernel/time/jiffies.c +++ b/kernel/time/jiffies.c @@ -25,7 +25,7 @@ #include <linux/module.h> #include <linux/init.h> -#include "tick-internal.h" +#include "timekeeping.h" /* The Jiffies based clocksource is the lowest common * denominator clock source which should function on @@ -71,6 +71,7 @@ static struct clocksource clocksource_jiffies = { .mask = 0xffffffff, /*32bits*/ .mult = NSEC_PER_JIFFY << JIFFIES_SHIFT, /* details above */ .shift = JIFFIES_SHIFT, + .max_cycles = 10, }; __cacheline_aligned_in_smp DEFINE_SEQLOCK(jiffies_lock); @@ -94,7 +95,7 @@ EXPORT_SYMBOL(jiffies); static int __init init_jiffies_clocksource(void) { - return clocksource_register(&clocksource_jiffies); + return __clocksource_register(&clocksource_jiffies); } core_initcall(init_jiffies_clocksource); @@ -130,6 +131,6 @@ int register_refined_jiffies(long cycles_per_second) refined_jiffies.mult = ((u32)nsec_per_tick) << JIFFIES_SHIFT; - clocksource_register(&refined_jiffies); + __clocksource_register(&refined_jiffies); return 0; } diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c index 0f60b08a4f07..7a681003001c 100644 --- a/kernel/time/ntp.c +++ b/kernel/time/ntp.c @@ -17,7 +17,6 @@ #include <linux/module.h> #include <linux/rtc.h> -#include "tick-internal.h" #include "ntp_internal.h" /* @@ -459,6 +458,16 @@ out: return leap; } +#ifdef CONFIG_GENERIC_CMOS_UPDATE +int __weak update_persistent_clock64(struct timespec64 now64) +{ + struct timespec now; + + now = timespec64_to_timespec(now64); + return update_persistent_clock(now); +} +#endif + #if defined(CONFIG_GENERIC_CMOS_UPDATE) || defined(CONFIG_RTC_SYSTOHC) static void sync_cmos_clock(struct work_struct *work); @@ -494,8 +503,9 @@ static void sync_cmos_clock(struct work_struct *work) if (persistent_clock_is_local) adjust.tv_sec -= (sys_tz.tz_minuteswest * 60); #ifdef CONFIG_GENERIC_CMOS_UPDATE - fail = update_persistent_clock(timespec64_to_timespec(adjust)); + fail = update_persistent_clock64(adjust); #endif + #ifdef CONFIG_RTC_SYSTOHC if (fail == -ENODEV) fail = rtc_set_ntp_time(adjust); diff --git a/kernel/time/sched_clock.c b/kernel/time/sched_clock.c index 01d2d15aa662..a26036d37a38 100644 --- a/kernel/time/sched_clock.c +++ b/kernel/time/sched_clock.c @@ -1,5 +1,6 @@ /* - * sched_clock.c: support for extending counters to full 64-bit ns counter + * sched_clock.c: Generic sched_clock() support, to extend low level + * hardware time counters to full 64-bit ns values. * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -18,15 +19,53 @@ #include <linux/seqlock.h> #include <linux/bitops.h> -struct clock_data { - ktime_t wrap_kt; +/** + * struct clock_read_data - data required to read from sched_clock() + * + * @epoch_ns: sched_clock() value at last update + * @epoch_cyc: Clock cycle value at last update. + * @sched_clock_mask: Bitmask for two's complement subtraction of non 64bit + * clocks. + * @read_sched_clock: Current clock source (or dummy source when suspended). + * @mult: Multipler for scaled math conversion. + * @shift: Shift value for scaled math conversion. + * + * Care must be taken when updating this structure; it is read by + * some very hot code paths. It occupies <=40 bytes and, when combined + * with the seqcount used to synchronize access, comfortably fits into + * a 64 byte cache line. + */ +struct clock_read_data { u64 epoch_ns; u64 epoch_cyc; - seqcount_t seq; - unsigned long rate; + u64 sched_clock_mask; + u64 (*read_sched_clock)(void); u32 mult; u32 shift; - bool suspended; +}; + +/** + * struct clock_data - all data needed for sched_clock() (including + * registration of a new clock source) + * + * @seq: Sequence counter for protecting updates. The lowest + * bit is the index for @read_data. + * @read_data: Data required to read from sched_clock. + * @wrap_kt: Duration for which clock can run before wrapping. + * @rate: Tick rate of the registered clock. + * @actual_read_sched_clock: Registered hardware level clock read function. + * + * The ordering of this structure has been chosen to optimize cache + * performance. In particular 'seq' and 'read_data[0]' (combined) should fit + * into a single 64-byte cache line. + */ +struct clock_data { + seqcount_t seq; + struct clock_read_data read_data[2]; + ktime_t wrap_kt; + unsigned long rate; + + u64 (*actual_read_sched_clock)(void); }; static struct hrtimer sched_clock_timer; @@ -34,12 +73,6 @@ static int irqtime = -1; core_param(irqtime, irqtime, int, 0400); -static struct clock_data cd = { - .mult = NSEC_PER_SEC / HZ, -}; - -static u64 __read_mostly sched_clock_mask; - static u64 notrace jiffy_sched_clock_read(void) { /* @@ -49,7 +82,11 @@ static u64 notrace jiffy_sched_clock_read(void) return (u64)(jiffies - INITIAL_JIFFIES); } -static u64 __read_mostly (*read_sched_clock)(void) = jiffy_sched_clock_read; +static struct clock_data cd ____cacheline_aligned = { + .read_data[0] = { .mult = NSEC_PER_SEC / HZ, + .read_sched_clock = jiffy_sched_clock_read, }, + .actual_read_sched_clock = jiffy_sched_clock_read, +}; static inline u64 notrace cyc_to_ns(u64 cyc, u32 mult, u32 shift) { @@ -58,111 +95,136 @@ static inline u64 notrace cyc_to_ns(u64 cyc, u32 mult, u32 shift) unsigned long long notrace sched_clock(void) { - u64 epoch_ns; - u64 epoch_cyc; - u64 cyc; + u64 cyc, res; unsigned long seq; - - if (cd.suspended) - return cd.epoch_ns; + struct clock_read_data *rd; do { - seq = raw_read_seqcount_begin(&cd.seq); - epoch_cyc = cd.epoch_cyc; - epoch_ns = cd.epoch_ns; + seq = raw_read_seqcount(&cd.seq); + rd = cd.read_data + (seq & 1); + + cyc = (rd->read_sched_clock() - rd->epoch_cyc) & + rd->sched_clock_mask; + res = rd->epoch_ns + cyc_to_ns(cyc, rd->mult, rd->shift); } while (read_seqcount_retry(&cd.seq, seq)); - cyc = read_sched_clock(); - cyc = (cyc - epoch_cyc) & sched_clock_mask; - return epoch_ns + cyc_to_ns(cyc, cd.mult, cd.shift); + return res; +} + +/* + * Updating the data required to read the clock. + * + * sched_clock() will never observe mis-matched data even if called from + * an NMI. We do this by maintaining an odd/even copy of the data and + * steering sched_clock() to one or the other using a sequence counter. + * In order to preserve the data cache profile of sched_clock() as much + * as possible the system reverts back to the even copy when the update + * completes; the odd copy is used *only* during an update. + */ +static void update_clock_read_data(struct clock_read_data *rd) +{ + /* update the backup (odd) copy with the new data */ + cd.read_data[1] = *rd; + + /* steer readers towards the odd copy */ + raw_write_seqcount_latch(&cd.seq); + + /* now its safe for us to update the normal (even) copy */ + cd.read_data[0] = *rd; + + /* switch readers back to the even copy */ + raw_write_seqcount_latch(&cd.seq); } /* - * Atomically update the sched_clock epoch. + * Atomically update the sched_clock() epoch. */ -static void notrace update_sched_clock(void) +static void update_sched_clock(void) { - unsigned long flags; u64 cyc; u64 ns; + struct clock_read_data rd; + + rd = cd.read_data[0]; + + cyc = cd.actual_read_sched_clock(); + ns = rd.epoch_ns + cyc_to_ns((cyc - rd.epoch_cyc) & rd.sched_clock_mask, rd.mult, rd.shift); + + rd.epoch_ns = ns; + rd.epoch_cyc = cyc; - cyc = read_sched_clock(); - ns = cd.epoch_ns + - cyc_to_ns((cyc - cd.epoch_cyc) & sched_clock_mask, - cd.mult, cd.shift); - - raw_local_irq_save(flags); - raw_write_seqcount_begin(&cd.seq); - cd.epoch_ns = ns; - cd.epoch_cyc = cyc; - raw_write_seqcount_end(&cd.seq); - raw_local_irq_restore(flags); + update_clock_read_data(&rd); } static enum hrtimer_restart sched_clock_poll(struct hrtimer *hrt) { update_sched_clock(); hrtimer_forward_now(hrt, cd.wrap_kt); + return HRTIMER_RESTART; } -void __init sched_clock_register(u64 (*read)(void), int bits, - unsigned long rate) +void __init +sched_clock_register(u64 (*read)(void), int bits, unsigned long rate) { u64 res, wrap, new_mask, new_epoch, cyc, ns; u32 new_mult, new_shift; - ktime_t new_wrap_kt; unsigned long r; char r_unit; + struct clock_read_data rd; if (cd.rate > rate) return; WARN_ON(!irqs_disabled()); - /* calculate the mult/shift to convert counter ticks to ns. */ + /* Calculate the mult/shift to convert counter ticks to ns. */ clocks_calc_mult_shift(&new_mult, &new_shift, rate, NSEC_PER_SEC, 3600); new_mask = CLOCKSOURCE_MASK(bits); + cd.rate = rate; + + /* Calculate how many nanosecs until we risk wrapping */ + wrap = clocks_calc_max_nsecs(new_mult, new_shift, 0, new_mask, NULL); + cd.wrap_kt = ns_to_ktime(wrap); - /* calculate how many ns until we wrap */ - wrap = clocks_calc_max_nsecs(new_mult, new_shift, 0, new_mask); - new_wrap_kt = ns_to_ktime(wrap - (wrap >> 3)); + rd = cd.read_data[0]; - /* update epoch for new counter and update epoch_ns from old counter*/ + /* Update epoch for new counter and update 'epoch_ns' from old counter*/ new_epoch = read(); - cyc = read_sched_clock(); - ns = cd.epoch_ns + cyc_to_ns((cyc - cd.epoch_cyc) & sched_clock_mask, - cd.mult, cd.shift); + cyc = cd.actual_read_sched_clock(); + ns = rd.epoch_ns + cyc_to_ns((cyc - rd.epoch_cyc) & rd.sched_clock_mask, rd.mult, rd.shift); + cd.actual_read_sched_clock = read; - raw_write_seqcount_begin(&cd.seq); - read_sched_clock = read; - sched_clock_mask = new_mask; - cd.rate = rate; - cd.wrap_kt = new_wrap_kt; - cd.mult = new_mult; - cd.shift = new_shift; - cd.epoch_cyc = new_epoch; - cd.epoch_ns = ns; - raw_write_seqcount_end(&cd.seq); + rd.read_sched_clock = read; + rd.sched_clock_mask = new_mask; + rd.mult = new_mult; + rd.shift = new_shift; + rd.epoch_cyc = new_epoch; + rd.epoch_ns = ns; + + update_clock_read_data(&rd); r = rate; if (r >= 4000000) { r /= 1000000; r_unit = 'M'; - } else if (r >= 1000) { - r /= 1000; - r_unit = 'k'; - } else - r_unit = ' '; - - /* calculate the ns resolution of this counter */ + } else { + if (r >= 1000) { + r /= 1000; + r_unit = 'k'; + } else { + r_unit = ' '; + } + } + + /* Calculate the ns resolution of this counter */ res = cyc_to_ns(1ULL, new_mult, new_shift); pr_info("sched_clock: %u bits at %lu%cHz, resolution %lluns, wraps every %lluns\n", bits, r, r_unit, res, wrap); - /* Enable IRQ time accounting if we have a fast enough sched_clock */ + /* Enable IRQ time accounting if we have a fast enough sched_clock() */ if (irqtime > 0 || (irqtime == -1 && rate >= 1000000)) enable_sched_clock_irqtime(); @@ -172,10 +234,10 @@ void __init sched_clock_register(u64 (*read)(void), int bits, void __init sched_clock_postinit(void) { /* - * If no sched_clock function has been provided at that point, + * If no sched_clock() function has been provided at that point, * make it the final one one. */ - if (read_sched_clock == jiffy_sched_clock_read) + if (cd.actual_read_sched_clock == jiffy_sched_clock_read) sched_clock_register(jiffy_sched_clock_read, BITS_PER_LONG, HZ); update_sched_clock(); @@ -189,29 +251,53 @@ void __init sched_clock_postinit(void) hrtimer_start(&sched_clock_timer, cd.wrap_kt, HRTIMER_MODE_REL); } +/* + * Clock read function for use when the clock is suspended. + * + * This function makes it appear to sched_clock() as if the clock + * stopped counting at its last update. + * + * This function must only be called from the critical + * section in sched_clock(). It relies on the read_seqcount_retry() + * at the end of the critical section to be sure we observe the + * correct copy of 'epoch_cyc'. + */ +static u64 notrace suspended_sched_clock_read(void) +{ + unsigned long seq = raw_read_seqcount(&cd.seq); + + return cd.read_data[seq & 1].epoch_cyc; +} + static int sched_clock_suspend(void) { + struct clock_read_data *rd = &cd.read_data[0]; + update_sched_clock(); hrtimer_cancel(&sched_clock_timer); - cd.suspended = true; + rd->read_sched_clock = suspended_sched_clock_read; + return 0; } static void sched_clock_resume(void) { - cd.epoch_cyc = read_sched_clock(); + struct clock_read_data *rd = &cd.read_data[0]; + + rd->epoch_cyc = cd.actual_read_sched_clock(); hrtimer_start(&sched_clock_timer, cd.wrap_kt, HRTIMER_MODE_REL); - cd.suspended = false; + rd->read_sched_clock = cd.actual_read_sched_clock; } static struct syscore_ops sched_clock_ops = { - .suspend = sched_clock_suspend, - .resume = sched_clock_resume, + .suspend = sched_clock_suspend, + .resume = sched_clock_resume, }; static int __init sched_clock_syscore_init(void) { register_syscore_ops(&sched_clock_ops); + return 0; } device_initcall(sched_clock_syscore_init); diff --git a/kernel/time/tick-broadcast-hrtimer.c b/kernel/time/tick-broadcast-hrtimer.c index eb682d5c697c..6aac4beedbbe 100644 --- a/kernel/time/tick-broadcast-hrtimer.c +++ b/kernel/time/tick-broadcast-hrtimer.c @@ -49,6 +49,7 @@ static void bc_set_mode(enum clock_event_mode mode, */ static int bc_set_next(ktime_t expires, struct clock_event_device *bc) { + int bc_moved; /* * We try to cancel the timer first. If the callback is on * flight on some other cpu then we let it handle it. If we @@ -60,9 +61,15 @@ static int bc_set_next(ktime_t expires, struct clock_event_device *bc) * restart the timer because we are in the callback, but we * can set the expiry time and let the callback return * HRTIMER_RESTART. + * + * Since we are in the idle loop at this point and because + * hrtimer_{start/cancel} functions call into tracing, + * calls to these functions must be bound within RCU_NONIDLE. */ - if (hrtimer_try_to_cancel(&bctimer) >= 0) { - hrtimer_start(&bctimer, expires, HRTIMER_MODE_ABS_PINNED); + RCU_NONIDLE(bc_moved = (hrtimer_try_to_cancel(&bctimer) >= 0) ? + !hrtimer_start(&bctimer, expires, HRTIMER_MODE_ABS_PINNED) : + 0); + if (bc_moved) { /* Bind the "device" to the cpu */ bc->bound_on = smp_processor_id(); } else if (bc->bound_on == smp_processor_id()) { diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c index 066f0ec05e48..7e8ca4f448a8 100644 --- a/kernel/time/tick-broadcast.c +++ b/kernel/time/tick-broadcast.c @@ -33,12 +33,14 @@ static cpumask_var_t tick_broadcast_mask; static cpumask_var_t tick_broadcast_on; static cpumask_var_t tmpmask; static DEFINE_RAW_SPINLOCK(tick_broadcast_lock); -static int tick_broadcast_force; +static int tick_broadcast_forced; #ifdef CONFIG_TICK_ONESHOT static void tick_broadcast_clear_oneshot(int cpu); +static void tick_resume_broadcast_oneshot(struct clock_event_device *bc); #else static inline void tick_broadcast_clear_oneshot(int cpu) { } +static inline void tick_resume_broadcast_oneshot(struct clock_event_device *bc) { } #endif /* @@ -303,7 +305,7 @@ static void tick_handle_periodic_broadcast(struct clock_event_device *dev) /* * The device is in periodic mode. No reprogramming necessary: */ - if (dev->mode == CLOCK_EVT_MODE_PERIODIC) + if (dev->state == CLOCK_EVT_STATE_PERIODIC) goto unlock; /* @@ -324,49 +326,54 @@ unlock: raw_spin_unlock(&tick_broadcast_lock); } -/* - * Powerstate information: The system enters/leaves a state, where - * affected devices might stop +/** + * tick_broadcast_control - Enable/disable or force broadcast mode + * @mode: The selected broadcast mode + * + * Called when the system enters a state where affected tick devices + * might stop. Note: TICK_BROADCAST_FORCE cannot be undone. + * + * Called with interrupts disabled, so clockevents_lock is not + * required here because the local clock event device cannot go away + * under us. */ -static void tick_do_broadcast_on_off(unsigned long *reason) +void tick_broadcast_control(enum tick_broadcast_mode mode) { struct clock_event_device *bc, *dev; struct tick_device *td; - unsigned long flags; int cpu, bc_stopped; - raw_spin_lock_irqsave(&tick_broadcast_lock, flags); - - cpu = smp_processor_id(); - td = &per_cpu(tick_cpu_device, cpu); + td = this_cpu_ptr(&tick_cpu_device); dev = td->evtdev; - bc = tick_broadcast_device.evtdev; /* * Is the device not affected by the powerstate ? */ if (!dev || !(dev->features & CLOCK_EVT_FEAT_C3STOP)) - goto out; + return; if (!tick_device_is_functional(dev)) - goto out; + return; + raw_spin_lock(&tick_broadcast_lock); + cpu = smp_processor_id(); + bc = tick_broadcast_device.evtdev; bc_stopped = cpumask_empty(tick_broadcast_mask); - switch (*reason) { - case CLOCK_EVT_NOTIFY_BROADCAST_ON: - case CLOCK_EVT_NOTIFY_BROADCAST_FORCE: + switch (mode) { + case TICK_BROADCAST_FORCE: + tick_broadcast_forced = 1; + case TICK_BROADCAST_ON: cpumask_set_cpu(cpu, tick_broadcast_on); if (!cpumask_test_and_set_cpu(cpu, tick_broadcast_mask)) { if (tick_broadcast_device.mode == TICKDEV_MODE_PERIODIC) clockevents_shutdown(dev); } - if (*reason == CLOCK_EVT_NOTIFY_BROADCAST_FORCE) - tick_broadcast_force = 1; break; - case CLOCK_EVT_NOTIFY_BROADCAST_OFF: - if (tick_broadcast_force) + + case TICK_BROADCAST_OFF: + if (tick_broadcast_forced) break; cpumask_clear_cpu(cpu, tick_broadcast_on); if (!tick_device_is_functional(dev)) @@ -388,22 +395,9 @@ static void tick_do_broadcast_on_off(unsigned long *reason) else tick_broadcast_setup_oneshot(bc); } -out: - raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags); -} - -/* - * Powerstate information: The system enters/leaves a state, where - * affected devices might stop. - */ -void tick_broadcast_on_off(unsigned long reason, int *oncpu) -{ - if (!cpumask_test_cpu(*oncpu, cpu_online_mask)) - printk(KERN_ERR "tick-broadcast: ignoring broadcast for " - "offline CPU #%d\n", *oncpu); - else - tick_do_broadcast_on_off(&reason); + raw_spin_unlock(&tick_broadcast_lock); } +EXPORT_SYMBOL_GPL(tick_broadcast_control); /* * Set the periodic handler depending on broadcast on/off @@ -416,14 +410,14 @@ void tick_set_periodic_handler(struct clock_event_device *dev, int broadcast) dev->event_handler = tick_handle_periodic_broadcast; } +#ifdef CONFIG_HOTPLUG_CPU /* * Remove a CPU from broadcasting */ -void tick_shutdown_broadcast(unsigned int *cpup) +void tick_shutdown_broadcast(unsigned int cpu) { struct clock_event_device *bc; unsigned long flags; - unsigned int cpu = *cpup; raw_spin_lock_irqsave(&tick_broadcast_lock, flags); @@ -438,6 +432,7 @@ void tick_shutdown_broadcast(unsigned int *cpup) raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags); } +#endif void tick_suspend_broadcast(void) { @@ -453,38 +448,48 @@ void tick_suspend_broadcast(void) raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags); } -int tick_resume_broadcast(void) +/* + * This is called from tick_resume_local() on a resuming CPU. That's + * called from the core resume function, tick_unfreeze() and the magic XEN + * resume hackery. + * + * In none of these cases the broadcast device mode can change and the + * bit of the resuming CPU in the broadcast mask is safe as well. + */ +bool tick_resume_check_broadcast(void) +{ + if (tick_broadcast_device.mode == TICKDEV_MODE_ONESHOT) + return false; + else + return cpumask_test_cpu(smp_processor_id(), tick_broadcast_mask); +} + +void tick_resume_broadcast(void) { struct clock_event_device *bc; unsigned long flags; - int broadcast = 0; raw_spin_lock_irqsave(&tick_broadcast_lock, flags); bc = tick_broadcast_device.evtdev; if (bc) { - clockevents_set_mode(bc, CLOCK_EVT_MODE_RESUME); + clockevents_tick_resume(bc); switch (tick_broadcast_device.mode) { case TICKDEV_MODE_PERIODIC: if (!cpumask_empty(tick_broadcast_mask)) tick_broadcast_start_periodic(bc); - broadcast = cpumask_test_cpu(smp_processor_id(), - tick_broadcast_mask); break; case TICKDEV_MODE_ONESHOT: if (!cpumask_empty(tick_broadcast_mask)) - broadcast = tick_resume_broadcast_oneshot(bc); + tick_resume_broadcast_oneshot(bc); break; } } raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags); - - return broadcast; } - #ifdef CONFIG_TICK_ONESHOT static cpumask_var_t tick_broadcast_oneshot_mask; @@ -532,8 +537,8 @@ static int tick_broadcast_set_event(struct clock_event_device *bc, int cpu, { int ret; - if (bc->mode != CLOCK_EVT_MODE_ONESHOT) - clockevents_set_mode(bc, CLOCK_EVT_MODE_ONESHOT); + if (bc->state != CLOCK_EVT_STATE_ONESHOT) + clockevents_set_state(bc, CLOCK_EVT_STATE_ONESHOT); ret = clockevents_program_event(bc, expires, force); if (!ret) @@ -541,10 +546,9 @@ static int tick_broadcast_set_event(struct clock_event_device *bc, int cpu, return ret; } -int tick_resume_broadcast_oneshot(struct clock_event_device *bc) +static void tick_resume_broadcast_oneshot(struct clock_event_device *bc) { - clockevents_set_mode(bc, CLOCK_EVT_MODE_ONESHOT); - return 0; + clockevents_set_state(bc, CLOCK_EVT_STATE_ONESHOT); } /* @@ -562,8 +566,8 @@ void tick_check_oneshot_broadcast_this_cpu(void) * switched over, leave the device alone. */ if (td->mode == TICKDEV_MODE_ONESHOT) { - clockevents_set_mode(td->evtdev, - CLOCK_EVT_MODE_ONESHOT); + clockevents_set_state(td->evtdev, + CLOCK_EVT_STATE_ONESHOT); } } } @@ -666,31 +670,26 @@ static void broadcast_shutdown_local(struct clock_event_device *bc, if (dev->next_event.tv64 < bc->next_event.tv64) return; } - clockevents_set_mode(dev, CLOCK_EVT_MODE_SHUTDOWN); + clockevents_set_state(dev, CLOCK_EVT_STATE_SHUTDOWN); } -static void broadcast_move_bc(int deadcpu) -{ - struct clock_event_device *bc = tick_broadcast_device.evtdev; - - if (!bc || !broadcast_needs_cpu(bc, deadcpu)) - return; - /* This moves the broadcast assignment to this cpu */ - clockevents_program_event(bc, bc->next_event, 1); -} - -/* - * Powerstate information: The system enters/leaves a state, where - * affected devices might stop +/** + * tick_broadcast_oneshot_control - Enter/exit broadcast oneshot mode + * @state: The target state (enter/exit) + * + * The system enters/leaves a state, where affected devices might stop * Returns 0 on success, -EBUSY if the cpu is used to broadcast wakeups. + * + * Called with interrupts disabled, so clockevents_lock is not + * required here because the local clock event device cannot go away + * under us. */ -int tick_broadcast_oneshot_control(unsigned long reason) +int tick_broadcast_oneshot_control(enum tick_broadcast_state state) { struct clock_event_device *bc, *dev; struct tick_device *td; - unsigned long flags; - ktime_t now; int cpu, ret = 0; + ktime_t now; /* * Periodic mode does not care about the enter/exit of power @@ -703,17 +702,17 @@ int tick_broadcast_oneshot_control(unsigned long reason) * We are called with preemtion disabled from the depth of the * idle code, so we can't be moved away. */ - cpu = smp_processor_id(); - td = &per_cpu(tick_cpu_device, cpu); + td = this_cpu_ptr(&tick_cpu_device); dev = td->evtdev; if (!(dev->features & CLOCK_EVT_FEAT_C3STOP)) return 0; + raw_spin_lock(&tick_broadcast_lock); bc = tick_broadcast_device.evtdev; + cpu = smp_processor_id(); - raw_spin_lock_irqsave(&tick_broadcast_lock, flags); - if (reason == CLOCK_EVT_NOTIFY_BROADCAST_ENTER) { + if (state == TICK_BROADCAST_ENTER) { if (!cpumask_test_and_set_cpu(cpu, tick_broadcast_oneshot_mask)) { WARN_ON_ONCE(cpumask_test_cpu(cpu, tick_broadcast_pending_mask)); broadcast_shutdown_local(bc, dev); @@ -741,7 +740,7 @@ int tick_broadcast_oneshot_control(unsigned long reason) cpumask_clear_cpu(cpu, tick_broadcast_oneshot_mask); } else { if (cpumask_test_and_clear_cpu(cpu, tick_broadcast_oneshot_mask)) { - clockevents_set_mode(dev, CLOCK_EVT_MODE_ONESHOT); + clockevents_set_state(dev, CLOCK_EVT_STATE_ONESHOT); /* * The cpu which was handling the broadcast * timer marked this cpu in the broadcast @@ -805,9 +804,10 @@ int tick_broadcast_oneshot_control(unsigned long reason) } } out: - raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags); + raw_spin_unlock(&tick_broadcast_lock); return ret; } +EXPORT_SYMBOL_GPL(tick_broadcast_oneshot_control); /* * Reset the one shot broadcast for a cpu @@ -842,7 +842,7 @@ void tick_broadcast_setup_oneshot(struct clock_event_device *bc) /* Set it up only once ! */ if (bc->event_handler != tick_handle_oneshot_broadcast) { - int was_periodic = bc->mode == CLOCK_EVT_MODE_PERIODIC; + int was_periodic = bc->state == CLOCK_EVT_STATE_PERIODIC; bc->event_handler = tick_handle_oneshot_broadcast; @@ -858,7 +858,7 @@ void tick_broadcast_setup_oneshot(struct clock_event_device *bc) tick_broadcast_oneshot_mask, tmpmask); if (was_periodic && !cpumask_empty(tmpmask)) { - clockevents_set_mode(bc, CLOCK_EVT_MODE_ONESHOT); + clockevents_set_state(bc, CLOCK_EVT_STATE_ONESHOT); tick_broadcast_init_next_event(tmpmask, tick_next_period); tick_broadcast_set_event(bc, cpu, tick_next_period, 1); @@ -894,14 +894,28 @@ void tick_broadcast_switch_to_oneshot(void) raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags); } +#ifdef CONFIG_HOTPLUG_CPU +void hotplug_cpu__broadcast_tick_pull(int deadcpu) +{ + struct clock_event_device *bc; + unsigned long flags; + + raw_spin_lock_irqsave(&tick_broadcast_lock, flags); + bc = tick_broadcast_device.evtdev; + + if (bc && broadcast_needs_cpu(bc, deadcpu)) { + /* This moves the broadcast assignment to this CPU: */ + clockevents_program_event(bc, bc->next_event, 1); + } + raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags); +} /* * Remove a dead CPU from broadcasting */ -void tick_shutdown_broadcast_oneshot(unsigned int *cpup) +void tick_shutdown_broadcast_oneshot(unsigned int cpu) { unsigned long flags; - unsigned int cpu = *cpup; raw_spin_lock_irqsave(&tick_broadcast_lock, flags); @@ -913,10 +927,9 @@ void tick_shutdown_broadcast_oneshot(unsigned int *cpup) cpumask_clear_cpu(cpu, tick_broadcast_pending_mask); cpumask_clear_cpu(cpu, tick_broadcast_force_mask); - broadcast_move_bc(cpu); - raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags); } +#endif /* * Check, whether the broadcast device is in one shot mode diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c index f7c515595b42..3ae6afa1eb98 100644 --- a/kernel/time/tick-common.c +++ b/kernel/time/tick-common.c @@ -102,7 +102,7 @@ void tick_handle_periodic(struct clock_event_device *dev) tick_periodic(cpu); - if (dev->mode != CLOCK_EVT_MODE_ONESHOT) + if (dev->state != CLOCK_EVT_STATE_ONESHOT) return; for (;;) { /* @@ -140,7 +140,7 @@ void tick_setup_periodic(struct clock_event_device *dev, int broadcast) if ((dev->features & CLOCK_EVT_FEAT_PERIODIC) && !tick_broadcast_oneshot_active()) { - clockevents_set_mode(dev, CLOCK_EVT_MODE_PERIODIC); + clockevents_set_state(dev, CLOCK_EVT_STATE_PERIODIC); } else { unsigned long seq; ktime_t next; @@ -150,7 +150,7 @@ void tick_setup_periodic(struct clock_event_device *dev, int broadcast) next = tick_next_period; } while (read_seqretry(&jiffies_lock, seq)); - clockevents_set_mode(dev, CLOCK_EVT_MODE_ONESHOT); + clockevents_set_state(dev, CLOCK_EVT_STATE_ONESHOT); for (;;) { if (!clockevents_program_event(dev, next, false)) @@ -332,14 +332,16 @@ out_bc: tick_install_broadcast_device(newdev); } +#ifdef CONFIG_HOTPLUG_CPU /* * Transfer the do_timer job away from a dying cpu. * - * Called with interrupts disabled. + * Called with interrupts disabled. Not locking required. If + * tick_do_timer_cpu is owned by this cpu, nothing can change it. */ -void tick_handover_do_timer(int *cpup) +void tick_handover_do_timer(void) { - if (*cpup == tick_do_timer_cpu) { + if (tick_do_timer_cpu == smp_processor_id()) { int cpu = cpumask_first(cpu_online_mask); tick_do_timer_cpu = (cpu < nr_cpu_ids) ? cpu : @@ -354,9 +356,9 @@ void tick_handover_do_timer(int *cpup) * access the hardware device itself. * We just set the mode and remove it from the lists. */ -void tick_shutdown(unsigned int *cpup) +void tick_shutdown(unsigned int cpu) { - struct tick_device *td = &per_cpu(tick_cpu_device, *cpup); + struct tick_device *td = &per_cpu(tick_cpu_device, cpu); struct clock_event_device *dev = td->evtdev; td->mode = TICKDEV_MODE_PERIODIC; @@ -365,27 +367,42 @@ void tick_shutdown(unsigned int *cpup) * Prevent that the clock events layer tries to call * the set mode function! */ + dev->state = CLOCK_EVT_STATE_DETACHED; dev->mode = CLOCK_EVT_MODE_UNUSED; clockevents_exchange_device(dev, NULL); dev->event_handler = clockevents_handle_noop; td->evtdev = NULL; } } +#endif -void tick_suspend(void) +/** + * tick_suspend_local - Suspend the local tick device + * + * Called from the local cpu for freeze with interrupts disabled. + * + * No locks required. Nothing can change the per cpu device. + */ +void tick_suspend_local(void) { struct tick_device *td = this_cpu_ptr(&tick_cpu_device); clockevents_shutdown(td->evtdev); } -void tick_resume(void) +/** + * tick_resume_local - Resume the local tick device + * + * Called from the local CPU for unfreeze or XEN resume magic. + * + * No locks required. Nothing can change the per cpu device. + */ +void tick_resume_local(void) { struct tick_device *td = this_cpu_ptr(&tick_cpu_device); - int broadcast = tick_resume_broadcast(); - - clockevents_set_mode(td->evtdev, CLOCK_EVT_MODE_RESUME); + bool broadcast = tick_resume_check_broadcast(); + clockevents_tick_resume(td->evtdev); if (!broadcast) { if (td->mode == TICKDEV_MODE_PERIODIC) tick_setup_periodic(td->evtdev, 0); @@ -394,6 +411,35 @@ void tick_resume(void) } } +/** + * tick_suspend - Suspend the tick and the broadcast device + * + * Called from syscore_suspend() via timekeeping_suspend with only one + * CPU online and interrupts disabled or from tick_unfreeze() under + * tick_freeze_lock. + * + * No locks required. Nothing can change the per cpu device. + */ +void tick_suspend(void) +{ + tick_suspend_local(); + tick_suspend_broadcast(); +} + +/** + * tick_resume - Resume the tick and the broadcast device + * + * Called from syscore_resume() via timekeeping_resume with only one + * CPU online and interrupts disabled. + * + * No locks required. Nothing can change the per cpu device. + */ +void tick_resume(void) +{ + tick_resume_broadcast(); + tick_resume_local(); +} + static DEFINE_RAW_SPINLOCK(tick_freeze_lock); static unsigned int tick_freeze_depth; @@ -411,12 +457,10 @@ void tick_freeze(void) raw_spin_lock(&tick_freeze_lock); tick_freeze_depth++; - if (tick_freeze_depth == num_online_cpus()) { + if (tick_freeze_depth == num_online_cpus()) timekeeping_suspend(); - } else { - tick_suspend(); - tick_suspend_broadcast(); - } + else + tick_suspend_local(); raw_spin_unlock(&tick_freeze_lock); } @@ -437,7 +481,7 @@ void tick_unfreeze(void) if (tick_freeze_depth == num_online_cpus()) timekeeping_resume(); else - tick_resume(); + tick_resume_local(); tick_freeze_depth--; diff --git a/kernel/time/tick-internal.h b/kernel/time/tick-internal.h index 366aeb4f2c66..b64fdd8054c5 100644 --- a/kernel/time/tick-internal.h +++ b/kernel/time/tick-internal.h @@ -5,15 +5,12 @@ #include <linux/tick.h> #include "timekeeping.h" +#include "tick-sched.h" -extern seqlock_t jiffies_lock; +#ifdef CONFIG_GENERIC_CLOCKEVENTS -#define CS_NAME_LEN 32 - -#ifdef CONFIG_GENERIC_CLOCKEVENTS_BUILD - -#define TICK_DO_TIMER_NONE -1 -#define TICK_DO_TIMER_BOOT -2 +# define TICK_DO_TIMER_NONE -1 +# define TICK_DO_TIMER_BOOT -2 DECLARE_PER_CPU(struct tick_device, tick_cpu_device); extern ktime_t tick_next_period; @@ -23,21 +20,72 @@ extern int tick_do_timer_cpu __read_mostly; extern void tick_setup_periodic(struct clock_event_device *dev, int broadcast); extern void tick_handle_periodic(struct clock_event_device *dev); extern void tick_check_new_device(struct clock_event_device *dev); -extern void tick_handover_do_timer(int *cpup); -extern void tick_shutdown(unsigned int *cpup); +extern void tick_shutdown(unsigned int cpu); extern void tick_suspend(void); extern void tick_resume(void); extern bool tick_check_replacement(struct clock_event_device *curdev, struct clock_event_device *newdev); extern void tick_install_replacement(struct clock_event_device *dev); +extern int tick_is_oneshot_available(void); +extern struct tick_device *tick_get_device(int cpu); -extern void clockevents_shutdown(struct clock_event_device *dev); +extern int clockevents_tick_resume(struct clock_event_device *dev); +/* Check, if the device is functional or a dummy for broadcast */ +static inline int tick_device_is_functional(struct clock_event_device *dev) +{ + return !(dev->features & CLOCK_EVT_FEAT_DUMMY); +} +extern void clockevents_shutdown(struct clock_event_device *dev); +extern void clockevents_exchange_device(struct clock_event_device *old, + struct clock_event_device *new); +extern void clockevents_set_state(struct clock_event_device *dev, + enum clock_event_state state); +extern int clockevents_program_event(struct clock_event_device *dev, + ktime_t expires, bool force); +extern void clockevents_handle_noop(struct clock_event_device *dev); +extern int __clockevents_update_freq(struct clock_event_device *dev, u32 freq); extern ssize_t sysfs_get_uname(const char *buf, char *dst, size_t cnt); -/* - * NO_HZ / high resolution timer shared code - */ +/* Broadcasting support */ +# ifdef CONFIG_GENERIC_CLOCKEVENTS_BROADCAST +extern int tick_device_uses_broadcast(struct clock_event_device *dev, int cpu); +extern void tick_install_broadcast_device(struct clock_event_device *dev); +extern int tick_is_broadcast_device(struct clock_event_device *dev); +extern void tick_shutdown_broadcast(unsigned int cpu); +extern void tick_suspend_broadcast(void); +extern void tick_resume_broadcast(void); +extern bool tick_resume_check_broadcast(void); +extern void tick_broadcast_init(void); +extern void tick_set_periodic_handler(struct clock_event_device *dev, int broadcast); +extern int tick_broadcast_update_freq(struct clock_event_device *dev, u32 freq); +extern struct tick_device *tick_get_broadcast_device(void); +extern struct cpumask *tick_get_broadcast_mask(void); +# else /* !CONFIG_GENERIC_CLOCKEVENTS_BROADCAST: */ +static inline void tick_install_broadcast_device(struct clock_event_device *dev) { } +static inline int tick_is_broadcast_device(struct clock_event_device *dev) { return 0; } +static inline int tick_device_uses_broadcast(struct clock_event_device *dev, int cpu) { return 0; } +static inline void tick_do_periodic_broadcast(struct clock_event_device *d) { } +static inline void tick_shutdown_broadcast(unsigned int cpu) { } +static inline void tick_suspend_broadcast(void) { } +static inline void tick_resume_broadcast(void) { } +static inline bool tick_resume_check_broadcast(void) { return false; } +static inline void tick_broadcast_init(void) { } +static inline int tick_broadcast_update_freq(struct clock_event_device *dev, u32 freq) { return -ENODEV; } + +/* Set the periodic handler in non broadcast mode */ +static inline void tick_set_periodic_handler(struct clock_event_device *dev, int broadcast) +{ + dev->event_handler = tick_handle_periodic; +} +# endif /* !CONFIG_GENERIC_CLOCKEVENTS_BROADCAST */ + +#else /* !GENERIC_CLOCKEVENTS: */ +static inline void tick_suspend(void) { } +static inline void tick_resume(void) { } +#endif /* !GENERIC_CLOCKEVENTS */ + +/* Oneshot related functions */ #ifdef CONFIG_TICK_ONESHOT extern void tick_setup_oneshot(struct clock_event_device *newdev, void (*handler)(struct clock_event_device *), @@ -46,58 +94,42 @@ extern int tick_program_event(ktime_t expires, int force); extern void tick_oneshot_notify(void); extern int tick_switch_to_oneshot(void (*handler)(struct clock_event_device *)); extern void tick_resume_oneshot(void); -# ifdef CONFIG_GENERIC_CLOCKEVENTS_BROADCAST +static inline bool tick_oneshot_possible(void) { return true; } +extern int tick_oneshot_mode_active(void); +extern void tick_clock_notify(void); +extern int tick_check_oneshot_change(int allow_nohz); +extern int tick_init_highres(void); +#else /* !CONFIG_TICK_ONESHOT: */ +static inline +void tick_setup_oneshot(struct clock_event_device *newdev, + void (*handler)(struct clock_event_device *), + ktime_t nextevt) { BUG(); } +static inline void tick_resume_oneshot(void) { BUG(); } +static inline int tick_program_event(ktime_t expires, int force) { return 0; } +static inline void tick_oneshot_notify(void) { } +static inline bool tick_oneshot_possible(void) { return false; } +static inline int tick_oneshot_mode_active(void) { return 0; } +static inline void tick_clock_notify(void) { } +static inline int tick_check_oneshot_change(int allow_nohz) { return 0; } +#endif /* !CONFIG_TICK_ONESHOT */ + +/* Functions related to oneshot broadcasting */ +#if defined(CONFIG_GENERIC_CLOCKEVENTS_BROADCAST) && defined(CONFIG_TICK_ONESHOT) extern void tick_broadcast_setup_oneshot(struct clock_event_device *bc); -extern int tick_broadcast_oneshot_control(unsigned long reason); extern void tick_broadcast_switch_to_oneshot(void); -extern void tick_shutdown_broadcast_oneshot(unsigned int *cpup); -extern int tick_resume_broadcast_oneshot(struct clock_event_device *bc); +extern void tick_shutdown_broadcast_oneshot(unsigned int cpu); extern int tick_broadcast_oneshot_active(void); extern void tick_check_oneshot_broadcast_this_cpu(void); bool tick_broadcast_oneshot_available(void); -# else /* BROADCAST */ -static inline void tick_broadcast_setup_oneshot(struct clock_event_device *bc) -{ - BUG(); -} -static inline int tick_broadcast_oneshot_control(unsigned long reason) { return 0; } +extern struct cpumask *tick_get_broadcast_oneshot_mask(void); +#else /* !(BROADCAST && ONESHOT): */ +static inline void tick_broadcast_setup_oneshot(struct clock_event_device *bc) { BUG(); } static inline void tick_broadcast_switch_to_oneshot(void) { } -static inline void tick_shutdown_broadcast_oneshot(unsigned int *cpup) { } +static inline void tick_shutdown_broadcast_oneshot(unsigned int cpu) { } static inline int tick_broadcast_oneshot_active(void) { return 0; } static inline void tick_check_oneshot_broadcast_this_cpu(void) { } -static inline bool tick_broadcast_oneshot_available(void) { return true; } -# endif /* !BROADCAST */ - -#else /* !ONESHOT */ -static inline -void tick_setup_oneshot(struct clock_event_device *newdev, - void (*handler)(struct clock_event_device *), - ktime_t nextevt) -{ - BUG(); -} -static inline void tick_resume_oneshot(void) -{ - BUG(); -} -static inline int tick_program_event(ktime_t expires, int force) -{ - return 0; -} -static inline void tick_oneshot_notify(void) { } -static inline void tick_broadcast_setup_oneshot(struct clock_event_device *bc) -{ - BUG(); -} -static inline int tick_broadcast_oneshot_control(unsigned long reason) { return 0; } -static inline void tick_shutdown_broadcast_oneshot(unsigned int *cpup) { } -static inline int tick_resume_broadcast_oneshot(struct clock_event_device *bc) -{ - return 0; -} -static inline int tick_broadcast_oneshot_active(void) { return 0; } -static inline bool tick_broadcast_oneshot_available(void) { return false; } -#endif /* !TICK_ONESHOT */ +static inline bool tick_broadcast_oneshot_available(void) { return tick_oneshot_possible(); } +#endif /* !(BROADCAST && ONESHOT) */ /* NO_HZ_FULL internal */ #ifdef CONFIG_NO_HZ_FULL @@ -105,68 +137,3 @@ extern void tick_nohz_init(void); # else static inline void tick_nohz_init(void) { } #endif - -/* - * Broadcasting support - */ -#ifdef CONFIG_GENERIC_CLOCKEVENTS_BROADCAST -extern int tick_device_uses_broadcast(struct clock_event_device *dev, int cpu); -extern void tick_install_broadcast_device(struct clock_event_device *dev); -extern int tick_is_broadcast_device(struct clock_event_device *dev); -extern void tick_broadcast_on_off(unsigned long reason, int *oncpu); -extern void tick_shutdown_broadcast(unsigned int *cpup); -extern void tick_suspend_broadcast(void); -extern int tick_resume_broadcast(void); -extern void tick_broadcast_init(void); -extern void -tick_set_periodic_handler(struct clock_event_device *dev, int broadcast); -int tick_broadcast_update_freq(struct clock_event_device *dev, u32 freq); - -#else /* !BROADCAST */ - -static inline void tick_install_broadcast_device(struct clock_event_device *dev) -{ -} - -static inline int tick_is_broadcast_device(struct clock_event_device *dev) -{ - return 0; -} -static inline int tick_device_uses_broadcast(struct clock_event_device *dev, - int cpu) -{ - return 0; -} -static inline void tick_do_periodic_broadcast(struct clock_event_device *d) { } -static inline void tick_broadcast_on_off(unsigned long reason, int *oncpu) { } -static inline void tick_shutdown_broadcast(unsigned int *cpup) { } -static inline void tick_suspend_broadcast(void) { } -static inline int tick_resume_broadcast(void) { return 0; } -static inline void tick_broadcast_init(void) { } -static inline int tick_broadcast_update_freq(struct clock_event_device *dev, - u32 freq) { return -ENODEV; } - -/* - * Set the periodic handler in non broadcast mode - */ -static inline void tick_set_periodic_handler(struct clock_event_device *dev, - int broadcast) -{ - dev->event_handler = tick_handle_periodic; -} -#endif /* !BROADCAST */ - -/* - * Check, if the device is functional or a dummy for broadcast - */ -static inline int tick_device_is_functional(struct clock_event_device *dev) -{ - return !(dev->features & CLOCK_EVT_FEAT_DUMMY); -} - -int __clockevents_update_freq(struct clock_event_device *dev, u32 freq); - -#endif - -extern void do_timer(unsigned long ticks); -extern void update_wall_time(void); diff --git a/kernel/time/tick-oneshot.c b/kernel/time/tick-oneshot.c index 7ce740e78e1b..67a64b1670bf 100644 --- a/kernel/time/tick-oneshot.c +++ b/kernel/time/tick-oneshot.c @@ -38,7 +38,7 @@ void tick_resume_oneshot(void) { struct clock_event_device *dev = __this_cpu_read(tick_cpu_device.evtdev); - clockevents_set_mode(dev, CLOCK_EVT_MODE_ONESHOT); + clockevents_set_state(dev, CLOCK_EVT_STATE_ONESHOT); clockevents_program_event(dev, ktime_get(), true); } @@ -50,7 +50,7 @@ void tick_setup_oneshot(struct clock_event_device *newdev, ktime_t next_event) { newdev->event_handler = handler; - clockevents_set_mode(newdev, CLOCK_EVT_MODE_ONESHOT); + clockevents_set_state(newdev, CLOCK_EVT_STATE_ONESHOT); clockevents_program_event(newdev, next_event, true); } @@ -81,7 +81,7 @@ int tick_switch_to_oneshot(void (*handler)(struct clock_event_device *)) td->mode = TICKDEV_MODE_ONESHOT; dev->event_handler = handler; - clockevents_set_mode(dev, CLOCK_EVT_MODE_ONESHOT); + clockevents_set_state(dev, CLOCK_EVT_STATE_ONESHOT); tick_broadcast_switch_to_oneshot(); return 0; } diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c index a4c4edac4528..914259128145 100644 --- a/kernel/time/tick-sched.c +++ b/kernel/time/tick-sched.c @@ -34,7 +34,7 @@ /* * Per cpu nohz control structure */ -DEFINE_PER_CPU(struct tick_sched, tick_cpu_sched); +static DEFINE_PER_CPU(struct tick_sched, tick_cpu_sched); /* * The time, when the last jiffy update happened. Protected by jiffies_lock. @@ -416,6 +416,11 @@ static int __init setup_tick_nohz(char *str) __setup("nohz=", setup_tick_nohz); +int tick_nohz_tick_stopped(void) +{ + return __this_cpu_read(tick_cpu_sched.tick_stopped); +} + /** * tick_nohz_update_jiffies - update jiffies when idle was interrupted * diff --git a/kernel/time/tick-sched.h b/kernel/time/tick-sched.h new file mode 100644 index 000000000000..28b5da3e1a17 --- /dev/null +++ b/kernel/time/tick-sched.h @@ -0,0 +1,74 @@ +#ifndef _TICK_SCHED_H +#define _TICK_SCHED_H + +#include <linux/hrtimer.h> + +enum tick_device_mode { + TICKDEV_MODE_PERIODIC, + TICKDEV_MODE_ONESHOT, +}; + +struct tick_device { + struct clock_event_device *evtdev; + enum tick_device_mode mode; +}; + +enum tick_nohz_mode { + NOHZ_MODE_INACTIVE, + NOHZ_MODE_LOWRES, + NOHZ_MODE_HIGHRES, +}; + +/** + * struct tick_sched - sched tick emulation and no idle tick control/stats + * @sched_timer: hrtimer to schedule the periodic tick in high + * resolution mode + * @last_tick: Store the last tick expiry time when the tick + * timer is modified for nohz sleeps. This is necessary + * to resume the tick timer operation in the timeline + * when the CPU returns from nohz sleep. + * @tick_stopped: Indicator that the idle tick has been stopped + * @idle_jiffies: jiffies at the entry to idle for idle time accounting + * @idle_calls: Total number of idle calls + * @idle_sleeps: Number of idle calls, where the sched tick was stopped + * @idle_entrytime: Time when the idle call was entered + * @idle_waketime: Time when the idle was interrupted + * @idle_exittime: Time when the idle state was left + * @idle_sleeptime: Sum of the time slept in idle with sched tick stopped + * @iowait_sleeptime: Sum of the time slept in idle with sched tick stopped, with IO outstanding + * @sleep_length: Duration of the current idle sleep + * @do_timer_lst: CPU was the last one doing do_timer before going idle + */ +struct tick_sched { + struct hrtimer sched_timer; + unsigned long check_clocks; + enum tick_nohz_mode nohz_mode; + ktime_t last_tick; + int inidle; + int tick_stopped; + unsigned long idle_jiffies; + unsigned long idle_calls; + unsigned long idle_sleeps; + int idle_active; + ktime_t idle_entrytime; + ktime_t idle_waketime; + ktime_t idle_exittime; + ktime_t idle_sleeptime; + ktime_t iowait_sleeptime; + ktime_t sleep_length; + unsigned long last_jiffies; + unsigned long next_jiffies; + ktime_t idle_expires; + int do_timer_last; +}; + +extern struct tick_sched *tick_get_tick_sched(int cpu); + +extern void tick_setup_sched_timer(void); +#if defined CONFIG_NO_HZ_COMMON || defined CONFIG_HIGH_RES_TIMERS +extern void tick_cancel_sched_timer(int cpu); +#else +static inline void tick_cancel_sched_timer(int cpu) { } +#endif + +#endif diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index 91db94136c10..946acb72179f 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -59,17 +59,15 @@ struct tk_fast { }; static struct tk_fast tk_fast_mono ____cacheline_aligned; +static struct tk_fast tk_fast_raw ____cacheline_aligned; /* flag for if timekeeping is suspended */ int __read_mostly timekeeping_suspended; -/* Flag for if there is a persistent clock on this platform */ -bool __read_mostly persistent_clock_exist = false; - static inline void tk_normalize_xtime(struct timekeeper *tk) { - while (tk->tkr.xtime_nsec >= ((u64)NSEC_PER_SEC << tk->tkr.shift)) { - tk->tkr.xtime_nsec -= (u64)NSEC_PER_SEC << tk->tkr.shift; + while (tk->tkr_mono.xtime_nsec >= ((u64)NSEC_PER_SEC << tk->tkr_mono.shift)) { + tk->tkr_mono.xtime_nsec -= (u64)NSEC_PER_SEC << tk->tkr_mono.shift; tk->xtime_sec++; } } @@ -79,20 +77,20 @@ static inline struct timespec64 tk_xtime(struct timekeeper *tk) struct timespec64 ts; ts.tv_sec = tk->xtime_sec; - ts.tv_nsec = (long)(tk->tkr.xtime_nsec >> tk->tkr.shift); + ts.tv_nsec = (long)(tk->tkr_mono.xtime_nsec >> tk->tkr_mono.shift); return ts; } static void tk_set_xtime(struct timekeeper *tk, const struct timespec64 *ts) { tk->xtime_sec = ts->tv_sec; - tk->tkr.xtime_nsec = (u64)ts->tv_nsec << tk->tkr.shift; + tk->tkr_mono.xtime_nsec = (u64)ts->tv_nsec << tk->tkr_mono.shift; } static void tk_xtime_add(struct timekeeper *tk, const struct timespec64 *ts) { tk->xtime_sec += ts->tv_sec; - tk->tkr.xtime_nsec += (u64)ts->tv_nsec << tk->tkr.shift; + tk->tkr_mono.xtime_nsec += (u64)ts->tv_nsec << tk->tkr_mono.shift; tk_normalize_xtime(tk); } @@ -118,6 +116,117 @@ static inline void tk_update_sleep_time(struct timekeeper *tk, ktime_t delta) tk->offs_boot = ktime_add(tk->offs_boot, delta); } +#ifdef CONFIG_DEBUG_TIMEKEEPING +#define WARNING_FREQ (HZ*300) /* 5 minute rate-limiting */ +/* + * These simple flag variables are managed + * without locks, which is racy, but ok since + * we don't really care about being super + * precise about how many events were seen, + * just that a problem was observed. + */ +static int timekeeping_underflow_seen; +static int timekeeping_overflow_seen; + +/* last_warning is only modified under the timekeeping lock */ +static long timekeeping_last_warning; + +static void timekeeping_check_update(struct timekeeper *tk, cycle_t offset) +{ + + cycle_t max_cycles = tk->tkr_mono.clock->max_cycles; + const char *name = tk->tkr_mono.clock->name; + + if (offset > max_cycles) { + printk_deferred("WARNING: timekeeping: Cycle offset (%lld) is larger than allowed by the '%s' clock's max_cycles value (%lld): time overflow danger\n", + offset, name, max_cycles); + printk_deferred(" timekeeping: Your kernel is sick, but tries to cope by capping time updates\n"); + } else { + if (offset > (max_cycles >> 1)) { + printk_deferred("INFO: timekeeping: Cycle offset (%lld) is larger than the the '%s' clock's 50%% safety margin (%lld)\n", + offset, name, max_cycles >> 1); + printk_deferred(" timekeeping: Your kernel is still fine, but is feeling a bit nervous\n"); + } + } + + if (timekeeping_underflow_seen) { + if (jiffies - timekeeping_last_warning > WARNING_FREQ) { + printk_deferred("WARNING: Underflow in clocksource '%s' observed, time update ignored.\n", name); + printk_deferred(" Please report this, consider using a different clocksource, if possible.\n"); + printk_deferred(" Your kernel is probably still fine.\n"); + timekeeping_last_warning = jiffies; + } + timekeeping_underflow_seen = 0; + } + + if (timekeeping_overflow_seen) { + if (jiffies - timekeeping_last_warning > WARNING_FREQ) { + printk_deferred("WARNING: Overflow in clocksource '%s' observed, time update capped.\n", name); + printk_deferred(" Please report this, consider using a different clocksource, if possible.\n"); + printk_deferred(" Your kernel is probably still fine.\n"); + timekeeping_last_warning = jiffies; + } + timekeeping_overflow_seen = 0; + } +} + +static inline cycle_t timekeeping_get_delta(struct tk_read_base *tkr) +{ + cycle_t now, last, mask, max, delta; + unsigned int seq; + + /* + * Since we're called holding a seqlock, the data may shift + * under us while we're doing the calculation. This can cause + * false positives, since we'd note a problem but throw the + * results away. So nest another seqlock here to atomically + * grab the points we are checking with. + */ + do { + seq = read_seqcount_begin(&tk_core.seq); + now = tkr->read(tkr->clock); + last = tkr->cycle_last; + mask = tkr->mask; + max = tkr->clock->max_cycles; + } while (read_seqcount_retry(&tk_core.seq, seq)); + + delta = clocksource_delta(now, last, mask); + + /* + * Try to catch underflows by checking if we are seeing small + * mask-relative negative values. + */ + if (unlikely((~delta & mask) < (mask >> 3))) { + timekeeping_underflow_seen = 1; + delta = 0; + } + + /* Cap delta value to the max_cycles values to avoid mult overflows */ + if (unlikely(delta > max)) { + timekeeping_overflow_seen = 1; + delta = tkr->clock->max_cycles; + } + + return delta; +} +#else +static inline void timekeeping_check_update(struct timekeeper *tk, cycle_t offset) +{ +} +static inline cycle_t timekeeping_get_delta(struct tk_read_base *tkr) +{ + cycle_t cycle_now, delta; + + /* read clocksource */ + cycle_now = tkr->read(tkr->clock); + + /* calculate the delta since the last update_wall_time */ + delta = clocksource_delta(cycle_now, tkr->cycle_last, tkr->mask); + + return delta; +} +#endif + /** * tk_setup_internals - Set up internals to use clocksource clock. * @@ -135,11 +244,16 @@ static void tk_setup_internals(struct timekeeper *tk, struct clocksource *clock) u64 tmp, ntpinterval; struct clocksource *old_clock; - old_clock = tk->tkr.clock; - tk->tkr.clock = clock; - tk->tkr.read = clock->read; - tk->tkr.mask = clock->mask; - tk->tkr.cycle_last = tk->tkr.read(clock); + old_clock = tk->tkr_mono.clock; + tk->tkr_mono.clock = clock; + tk->tkr_mono.read = clock->read; + tk->tkr_mono.mask = clock->mask; + tk->tkr_mono.cycle_last = tk->tkr_mono.read(clock); + + tk->tkr_raw.clock = clock; + tk->tkr_raw.read = clock->read; + tk->tkr_raw.mask = clock->mask; + tk->tkr_raw.cycle_last = tk->tkr_mono.cycle_last; /* Do the ns -> cycle conversion first, using original mult */ tmp = NTP_INTERVAL_LENGTH; @@ -163,11 +277,14 @@ static void tk_setup_internals(struct timekeeper *tk, struct clocksource *clock) if (old_clock) { int shift_change = clock->shift - old_clock->shift; if (shift_change < 0) - tk->tkr.xtime_nsec >>= -shift_change; + tk->tkr_mono.xtime_nsec >>= -shift_change; else - tk->tkr.xtime_nsec <<= shift_change; + tk->tkr_mono.xtime_nsec <<= shift_change; } - tk->tkr.shift = clock->shift; + tk->tkr_raw.xtime_nsec = 0; + + tk->tkr_mono.shift = clock->shift; + tk->tkr_raw.shift = clock->shift; tk->ntp_error = 0; tk->ntp_error_shift = NTP_SCALE_SHIFT - clock->shift; @@ -178,7 +295,8 @@ static void tk_setup_internals(struct timekeeper *tk, struct clocksource *clock) * active clocksource. These value will be adjusted via NTP * to counteract clock drifting. */ - tk->tkr.mult = clock->mult; + tk->tkr_mono.mult = clock->mult; + tk->tkr_raw.mult = clock->mult; tk->ntp_err_mult = 0; } @@ -193,14 +311,10 @@ static inline u32 arch_gettimeoffset(void) { return 0; } static inline s64 timekeeping_get_ns(struct tk_read_base *tkr) { - cycle_t cycle_now, delta; + cycle_t delta; s64 nsec; - /* read clocksource: */ - cycle_now = tkr->read(tkr->clock); - - /* calculate the delta since the last update_wall_time: */ - delta = clocksource_delta(cycle_now, tkr->cycle_last, tkr->mask); + delta = timekeeping_get_delta(tkr); nsec = delta * tkr->mult + tkr->xtime_nsec; nsec >>= tkr->shift; @@ -209,25 +323,6 @@ static inline s64 timekeeping_get_ns(struct tk_read_base *tkr) return nsec + arch_gettimeoffset(); } -static inline s64 timekeeping_get_ns_raw(struct timekeeper *tk) -{ - struct clocksource *clock = tk->tkr.clock; - cycle_t cycle_now, delta; - s64 nsec; - - /* read clocksource: */ - cycle_now = tk->tkr.read(clock); - - /* calculate the delta since the last update_wall_time: */ - delta = clocksource_delta(cycle_now, tk->tkr.cycle_last, tk->tkr.mask); - - /* convert delta to nanoseconds. */ - nsec = clocksource_cyc2ns(delta, clock->mult, clock->shift); - - /* If arch requires, add in get_arch_timeoffset() */ - return nsec + arch_gettimeoffset(); -} - /** * update_fast_timekeeper - Update the fast and NMI safe monotonic timekeeper. * @tkr: Timekeeping readout base from which we take the update @@ -267,18 +362,18 @@ static inline s64 timekeeping_get_ns_raw(struct timekeeper *tk) * slightly wrong timestamp (a few nanoseconds). See * @ktime_get_mono_fast_ns. */ -static void update_fast_timekeeper(struct tk_read_base *tkr) +static void update_fast_timekeeper(struct tk_read_base *tkr, struct tk_fast *tkf) { - struct tk_read_base *base = tk_fast_mono.base; + struct tk_read_base *base = tkf->base; /* Force readers off to base[1] */ - raw_write_seqcount_latch(&tk_fast_mono.seq); + raw_write_seqcount_latch(&tkf->seq); /* Update base[0] */ memcpy(base, tkr, sizeof(*base)); /* Force readers back to base[0] */ - raw_write_seqcount_latch(&tk_fast_mono.seq); + raw_write_seqcount_latch(&tkf->seq); /* Update base[1] */ memcpy(base + 1, base, sizeof(*base)); @@ -316,22 +411,33 @@ static void update_fast_timekeeper(struct tk_read_base *tkr) * of the following timestamps. Callers need to be aware of that and * deal with it. */ -u64 notrace ktime_get_mono_fast_ns(void) +static __always_inline u64 __ktime_get_fast_ns(struct tk_fast *tkf) { struct tk_read_base *tkr; unsigned int seq; u64 now; do { - seq = raw_read_seqcount(&tk_fast_mono.seq); - tkr = tk_fast_mono.base + (seq & 0x01); - now = ktime_to_ns(tkr->base_mono) + timekeeping_get_ns(tkr); + seq = raw_read_seqcount(&tkf->seq); + tkr = tkf->base + (seq & 0x01); + now = ktime_to_ns(tkr->base) + timekeeping_get_ns(tkr); + } while (read_seqcount_retry(&tkf->seq, seq)); - } while (read_seqcount_retry(&tk_fast_mono.seq, seq)); return now; } + +u64 ktime_get_mono_fast_ns(void) +{ + return __ktime_get_fast_ns(&tk_fast_mono); +} EXPORT_SYMBOL_GPL(ktime_get_mono_fast_ns); +u64 ktime_get_raw_fast_ns(void) +{ + return __ktime_get_fast_ns(&tk_fast_raw); +} +EXPORT_SYMBOL_GPL(ktime_get_raw_fast_ns); + /* Suspend-time cycles value for halted fast timekeeper. */ static cycle_t cycles_at_suspend; @@ -353,12 +459,17 @@ static cycle_t dummy_clock_read(struct clocksource *cs) static void halt_fast_timekeeper(struct timekeeper *tk) { static struct tk_read_base tkr_dummy; - struct tk_read_base *tkr = &tk->tkr; + struct tk_read_base *tkr = &tk->tkr_mono; memcpy(&tkr_dummy, tkr, sizeof(tkr_dummy)); cycles_at_suspend = tkr->read(tkr->clock); tkr_dummy.read = dummy_clock_read; - update_fast_timekeeper(&tkr_dummy); + update_fast_timekeeper(&tkr_dummy, &tk_fast_mono); + + tkr = &tk->tkr_raw; + memcpy(&tkr_dummy, tkr, sizeof(tkr_dummy)); + tkr_dummy.read = dummy_clock_read; + update_fast_timekeeper(&tkr_dummy, &tk_fast_raw); } #ifdef CONFIG_GENERIC_TIME_VSYSCALL_OLD @@ -369,8 +480,8 @@ static inline void update_vsyscall(struct timekeeper *tk) xt = timespec64_to_timespec(tk_xtime(tk)); wm = timespec64_to_timespec(tk->wall_to_monotonic); - update_vsyscall_old(&xt, &wm, tk->tkr.clock, tk->tkr.mult, - tk->tkr.cycle_last); + update_vsyscall_old(&xt, &wm, tk->tkr_mono.clock, tk->tkr_mono.mult, + tk->tkr_mono.cycle_last); } static inline void old_vsyscall_fixup(struct timekeeper *tk) @@ -387,11 +498,11 @@ static inline void old_vsyscall_fixup(struct timekeeper *tk) * (shifted nanoseconds), and CONFIG_GENERIC_TIME_VSYSCALL_OLD * users are removed, this can be killed. */ - remainder = tk->tkr.xtime_nsec & ((1ULL << tk->tkr.shift) - 1); - tk->tkr.xtime_nsec -= remainder; - tk->tkr.xtime_nsec += 1ULL << tk->tkr.shift; + remainder = tk->tkr_mono.xtime_nsec & ((1ULL << tk->tkr_mono.shift) - 1); + tk->tkr_mono.xtime_nsec -= remainder; + tk->tkr_mono.xtime_nsec += 1ULL << tk->tkr_mono.shift; tk->ntp_error += remainder << tk->ntp_error_shift; - tk->ntp_error -= (1ULL << tk->tkr.shift) << tk->ntp_error_shift; + tk->ntp_error -= (1ULL << tk->tkr_mono.shift) << tk->ntp_error_shift; } #else #define old_vsyscall_fixup(tk) @@ -456,17 +567,17 @@ static inline void tk_update_ktime_data(struct timekeeper *tk) */ seconds = (u64)(tk->xtime_sec + tk->wall_to_monotonic.tv_sec); nsec = (u32) tk->wall_to_monotonic.tv_nsec; - tk->tkr.base_mono = ns_to_ktime(seconds * NSEC_PER_SEC + nsec); + tk->tkr_mono.base = ns_to_ktime(seconds * NSEC_PER_SEC + nsec); /* Update the monotonic raw base */ - tk->base_raw = timespec64_to_ktime(tk->raw_time); + tk->tkr_raw.base = timespec64_to_ktime(tk->raw_time); /* * The sum of the nanoseconds portions of xtime and * wall_to_monotonic can be greater/equal one second. Take * this into account before updating tk->ktime_sec. */ - nsec += (u32)(tk->tkr.xtime_nsec >> tk->tkr.shift); + nsec += (u32)(tk->tkr_mono.xtime_nsec >> tk->tkr_mono.shift); if (nsec >= NSEC_PER_SEC) seconds++; tk->ktime_sec = seconds; @@ -489,7 +600,8 @@ static void timekeeping_update(struct timekeeper *tk, unsigned int action) memcpy(&shadow_timekeeper, &tk_core.timekeeper, sizeof(tk_core.timekeeper)); - update_fast_timekeeper(&tk->tkr); + update_fast_timekeeper(&tk->tkr_mono, &tk_fast_mono); + update_fast_timekeeper(&tk->tkr_raw, &tk_fast_raw); } /** @@ -501,22 +613,23 @@ static void timekeeping_update(struct timekeeper *tk, unsigned int action) */ static void timekeeping_forward_now(struct timekeeper *tk) { - struct clocksource *clock = tk->tkr.clock; + struct clocksource *clock = tk->tkr_mono.clock; cycle_t cycle_now, delta; s64 nsec; - cycle_now = tk->tkr.read(clock); - delta = clocksource_delta(cycle_now, tk->tkr.cycle_last, tk->tkr.mask); - tk->tkr.cycle_last = cycle_now; + cycle_now = tk->tkr_mono.read(clock); + delta = clocksource_delta(cycle_now, tk->tkr_mono.cycle_last, tk->tkr_mono.mask); + tk->tkr_mono.cycle_last = cycle_now; + tk->tkr_raw.cycle_last = cycle_now; - tk->tkr.xtime_nsec += delta * tk->tkr.mult; + tk->tkr_mono.xtime_nsec += delta * tk->tkr_mono.mult; /* If arch requires, add in get_arch_timeoffset() */ - tk->tkr.xtime_nsec += (u64)arch_gettimeoffset() << tk->tkr.shift; + tk->tkr_mono.xtime_nsec += (u64)arch_gettimeoffset() << tk->tkr_mono.shift; tk_normalize_xtime(tk); - nsec = clocksource_cyc2ns(delta, clock->mult, clock->shift); + nsec = clocksource_cyc2ns(delta, tk->tkr_raw.mult, tk->tkr_raw.shift); timespec64_add_ns(&tk->raw_time, nsec); } @@ -537,7 +650,7 @@ int __getnstimeofday64(struct timespec64 *ts) seq = read_seqcount_begin(&tk_core.seq); ts->tv_sec = tk->xtime_sec; - nsecs = timekeeping_get_ns(&tk->tkr); + nsecs = timekeeping_get_ns(&tk->tkr_mono); } while (read_seqcount_retry(&tk_core.seq, seq)); @@ -577,8 +690,8 @@ ktime_t ktime_get(void) do { seq = read_seqcount_begin(&tk_core.seq); - base = tk->tkr.base_mono; - nsecs = timekeeping_get_ns(&tk->tkr); + base = tk->tkr_mono.base; + nsecs = timekeeping_get_ns(&tk->tkr_mono); } while (read_seqcount_retry(&tk_core.seq, seq)); @@ -603,8 +716,8 @@ ktime_t ktime_get_with_offset(enum tk_offsets offs) do { seq = read_seqcount_begin(&tk_core.seq); - base = ktime_add(tk->tkr.base_mono, *offset); - nsecs = timekeeping_get_ns(&tk->tkr); + base = ktime_add(tk->tkr_mono.base, *offset); + nsecs = timekeeping_get_ns(&tk->tkr_mono); } while (read_seqcount_retry(&tk_core.seq, seq)); @@ -645,8 +758,8 @@ ktime_t ktime_get_raw(void) do { seq = read_seqcount_begin(&tk_core.seq); - base = tk->base_raw; - nsecs = timekeeping_get_ns_raw(tk); + base = tk->tkr_raw.base; + nsecs = timekeeping_get_ns(&tk->tkr_raw); } while (read_seqcount_retry(&tk_core.seq, seq)); @@ -674,7 +787,7 @@ void ktime_get_ts64(struct timespec64 *ts) do { seq = read_seqcount_begin(&tk_core.seq); ts->tv_sec = tk->xtime_sec; - nsec = timekeeping_get_ns(&tk->tkr); + nsec = timekeeping_get_ns(&tk->tkr_mono); tomono = tk->wall_to_monotonic; } while (read_seqcount_retry(&tk_core.seq, seq)); @@ -759,8 +872,8 @@ void getnstime_raw_and_real(struct timespec *ts_raw, struct timespec *ts_real) ts_real->tv_sec = tk->xtime_sec; ts_real->tv_nsec = 0; - nsecs_raw = timekeeping_get_ns_raw(tk); - nsecs_real = timekeeping_get_ns(&tk->tkr); + nsecs_raw = timekeeping_get_ns(&tk->tkr_raw); + nsecs_real = timekeeping_get_ns(&tk->tkr_mono); } while (read_seqcount_retry(&tk_core.seq, seq)); @@ -943,7 +1056,7 @@ static int change_clocksource(void *data) */ if (try_module_get(new->owner)) { if (!new->enable || new->enable(new) == 0) { - old = tk->tkr.clock; + old = tk->tkr_mono.clock; tk_setup_internals(tk, new); if (old->disable) old->disable(old); @@ -971,11 +1084,11 @@ int timekeeping_notify(struct clocksource *clock) { struct timekeeper *tk = &tk_core.timekeeper; - if (tk->tkr.clock == clock) + if (tk->tkr_mono.clock == clock) return 0; stop_machine(change_clocksource, clock, NULL); tick_clock_notify(); - return tk->tkr.clock == clock ? 0 : -1; + return tk->tkr_mono.clock == clock ? 0 : -1; } /** @@ -993,7 +1106,7 @@ void getrawmonotonic64(struct timespec64 *ts) do { seq = read_seqcount_begin(&tk_core.seq); - nsecs = timekeeping_get_ns_raw(tk); + nsecs = timekeeping_get_ns(&tk->tkr_raw); ts64 = tk->raw_time; } while (read_seqcount_retry(&tk_core.seq, seq)); @@ -1016,7 +1129,7 @@ int timekeeping_valid_for_hres(void) do { seq = read_seqcount_begin(&tk_core.seq); - ret = tk->tkr.clock->flags & CLOCK_SOURCE_VALID_FOR_HRES; + ret = tk->tkr_mono.clock->flags & CLOCK_SOURCE_VALID_FOR_HRES; } while (read_seqcount_retry(&tk_core.seq, seq)); @@ -1035,7 +1148,7 @@ u64 timekeeping_max_deferment(void) do { seq = read_seqcount_begin(&tk_core.seq); - ret = tk->tkr.clock->max_idle_ns; + ret = tk->tkr_mono.clock->max_idle_ns; } while (read_seqcount_retry(&tk_core.seq, seq)); @@ -1057,6 +1170,14 @@ void __weak read_persistent_clock(struct timespec *ts) ts->tv_nsec = 0; } +void __weak read_persistent_clock64(struct timespec64 *ts64) +{ + struct timespec ts; + + read_persistent_clock(&ts); + *ts64 = timespec_to_timespec64(ts); +} + /** * read_boot_clock - Return time of the system start. * @@ -1072,6 +1193,20 @@ void __weak read_boot_clock(struct timespec *ts) ts->tv_nsec = 0; } +void __weak read_boot_clock64(struct timespec64 *ts64) +{ + struct timespec ts; + + read_boot_clock(&ts); + *ts64 = timespec_to_timespec64(ts); +} + +/* Flag for if timekeeping_resume() has injected sleeptime */ +static bool sleeptime_injected; + +/* Flag for if there is a persistent clock on this platform */ +static bool persistent_clock_exists; + /* * timekeeping_init - Initializes the clocksource and common timekeeping values */ @@ -1081,20 +1216,17 @@ void __init timekeeping_init(void) struct clocksource *clock; unsigned long flags; struct timespec64 now, boot, tmp; - struct timespec ts; - read_persistent_clock(&ts); - now = timespec_to_timespec64(ts); + read_persistent_clock64(&now); if (!timespec64_valid_strict(&now)) { pr_warn("WARNING: Persistent clock returned invalid value!\n" " Check your CMOS/BIOS settings.\n"); now.tv_sec = 0; now.tv_nsec = 0; } else if (now.tv_sec || now.tv_nsec) - persistent_clock_exist = true; + persistent_clock_exists = true; - read_boot_clock(&ts); - boot = timespec_to_timespec64(ts); + read_boot_clock64(&boot); if (!timespec64_valid_strict(&boot)) { pr_warn("WARNING: Boot clock returned invalid value!\n" " Check your CMOS/BIOS settings.\n"); @@ -1114,7 +1246,6 @@ void __init timekeeping_init(void) tk_set_xtime(tk, &now); tk->raw_time.tv_sec = 0; tk->raw_time.tv_nsec = 0; - tk->base_raw.tv64 = 0; if (boot.tv_sec == 0 && boot.tv_nsec == 0) boot = tk_xtime(tk); @@ -1127,7 +1258,7 @@ void __init timekeeping_init(void) raw_spin_unlock_irqrestore(&timekeeper_lock, flags); } -/* time in seconds when suspend began */ +/* time in seconds when suspend began for persistent clock */ static struct timespec64 timekeeping_suspend_time; /** @@ -1152,12 +1283,49 @@ static void __timekeeping_inject_sleeptime(struct timekeeper *tk, tk_debug_account_sleep_time(delta); } +#if defined(CONFIG_PM_SLEEP) && defined(CONFIG_RTC_HCTOSYS_DEVICE) +/** + * We have three kinds of time sources to use for sleep time + * injection, the preference order is: + * 1) non-stop clocksource + * 2) persistent clock (ie: RTC accessible when irqs are off) + * 3) RTC + * + * 1) and 2) are used by timekeeping, 3) by RTC subsystem. + * If system has neither 1) nor 2), 3) will be used finally. + * + * + * If timekeeping has injected sleeptime via either 1) or 2), + * 3) becomes needless, so in this case we don't need to call + * rtc_resume(), and this is what timekeeping_rtc_skipresume() + * means. + */ +bool timekeeping_rtc_skipresume(void) +{ + return sleeptime_injected; +} + +/** + * 1) can be determined whether to use or not only when doing + * timekeeping_resume() which is invoked after rtc_suspend(), + * so we can't skip rtc_suspend() surely if system has 1). + * + * But if system has 2), 2) will definitely be used, so in this + * case we don't need to call rtc_suspend(), and this is what + * timekeeping_rtc_skipsuspend() means. + */ +bool timekeeping_rtc_skipsuspend(void) +{ + return persistent_clock_exists; +} + /** * timekeeping_inject_sleeptime64 - Adds suspend interval to timeekeeping values * @delta: pointer to a timespec64 delta value * - * This hook is for architectures that cannot support read_persistent_clock + * This hook is for architectures that cannot support read_persistent_clock64 * because their RTC/persistent clock is only accessible when irqs are enabled. + * and also don't have an effective nonstop clocksource. * * This function should only be called by rtc_resume(), and allows * a suspend offset to be injected into the timekeeping values. @@ -1167,13 +1335,6 @@ void timekeeping_inject_sleeptime64(struct timespec64 *delta) struct timekeeper *tk = &tk_core.timekeeper; unsigned long flags; - /* - * Make sure we don't set the clock twice, as timekeeping_resume() - * already did it - */ - if (has_persistent_clock()) - return; - raw_spin_lock_irqsave(&timekeeper_lock, flags); write_seqcount_begin(&tk_core.seq); @@ -1189,26 +1350,21 @@ void timekeeping_inject_sleeptime64(struct timespec64 *delta) /* signal hrtimers about time change */ clock_was_set(); } +#endif /** * timekeeping_resume - Resumes the generic timekeeping subsystem. - * - * This is for the generic clocksource timekeeping. - * xtime/wall_to_monotonic/jiffies/etc are - * still managed by arch specific suspend/resume code. */ void timekeeping_resume(void) { struct timekeeper *tk = &tk_core.timekeeper; - struct clocksource *clock = tk->tkr.clock; + struct clocksource *clock = tk->tkr_mono.clock; unsigned long flags; struct timespec64 ts_new, ts_delta; - struct timespec tmp; cycle_t cycle_now, cycle_delta; - bool suspendtime_found = false; - read_persistent_clock(&tmp); - ts_new = timespec_to_timespec64(tmp); + sleeptime_injected = false; + read_persistent_clock64(&ts_new); clockevents_resume(); clocksource_resume(); @@ -1228,16 +1384,16 @@ void timekeeping_resume(void) * The less preferred source will only be tried if there is no better * usable source. The rtc part is handled separately in rtc core code. */ - cycle_now = tk->tkr.read(clock); + cycle_now = tk->tkr_mono.read(clock); if ((clock->flags & CLOCK_SOURCE_SUSPEND_NONSTOP) && - cycle_now > tk->tkr.cycle_last) { + cycle_now > tk->tkr_mono.cycle_last) { u64 num, max = ULLONG_MAX; u32 mult = clock->mult; u32 shift = clock->shift; s64 nsec = 0; - cycle_delta = clocksource_delta(cycle_now, tk->tkr.cycle_last, - tk->tkr.mask); + cycle_delta = clocksource_delta(cycle_now, tk->tkr_mono.cycle_last, + tk->tkr_mono.mask); /* * "cycle_delta * mutl" may cause 64 bits overflow, if the @@ -1253,17 +1409,19 @@ void timekeeping_resume(void) nsec += ((u64) cycle_delta * mult) >> shift; ts_delta = ns_to_timespec64(nsec); - suspendtime_found = true; + sleeptime_injected = true; } else if (timespec64_compare(&ts_new, &timekeeping_suspend_time) > 0) { ts_delta = timespec64_sub(ts_new, timekeeping_suspend_time); - suspendtime_found = true; + sleeptime_injected = true; } - if (suspendtime_found) + if (sleeptime_injected) __timekeeping_inject_sleeptime(tk, &ts_delta); /* Re-base the last cycle value */ - tk->tkr.cycle_last = cycle_now; + tk->tkr_mono.cycle_last = cycle_now; + tk->tkr_raw.cycle_last = cycle_now; + tk->ntp_error = 0; timekeeping_suspended = 0; timekeeping_update(tk, TK_MIRROR | TK_CLOCK_WAS_SET); @@ -1272,9 +1430,7 @@ void timekeeping_resume(void) touch_softlockup_watchdog(); - clockevents_notify(CLOCK_EVT_NOTIFY_RESUME, NULL); - - /* Resume hrtimers */ + tick_resume(); hrtimers_resume(); } @@ -1284,10 +1440,8 @@ int timekeeping_suspend(void) unsigned long flags; struct timespec64 delta, delta_delta; static struct timespec64 old_delta; - struct timespec tmp; - read_persistent_clock(&tmp); - timekeeping_suspend_time = timespec_to_timespec64(tmp); + read_persistent_clock64(&timekeeping_suspend_time); /* * On some systems the persistent_clock can not be detected at @@ -1295,31 +1449,33 @@ int timekeeping_suspend(void) * value returned, update the persistent_clock_exists flag. */ if (timekeeping_suspend_time.tv_sec || timekeeping_suspend_time.tv_nsec) - persistent_clock_exist = true; + persistent_clock_exists = true; raw_spin_lock_irqsave(&timekeeper_lock, flags); write_seqcount_begin(&tk_core.seq); timekeeping_forward_now(tk); timekeeping_suspended = 1; - /* - * To avoid drift caused by repeated suspend/resumes, - * which each can add ~1 second drift error, - * try to compensate so the difference in system time - * and persistent_clock time stays close to constant. - */ - delta = timespec64_sub(tk_xtime(tk), timekeeping_suspend_time); - delta_delta = timespec64_sub(delta, old_delta); - if (abs(delta_delta.tv_sec) >= 2) { + if (persistent_clock_exists) { /* - * if delta_delta is too large, assume time correction - * has occured and set old_delta to the current delta. + * To avoid drift caused by repeated suspend/resumes, + * which each can add ~1 second drift error, + * try to compensate so the difference in system time + * and persistent_clock time stays close to constant. */ - old_delta = delta; - } else { - /* Otherwise try to adjust old_system to compensate */ - timekeeping_suspend_time = - timespec64_add(timekeeping_suspend_time, delta_delta); + delta = timespec64_sub(tk_xtime(tk), timekeeping_suspend_time); + delta_delta = timespec64_sub(delta, old_delta); + if (abs(delta_delta.tv_sec) >= 2) { + /* + * if delta_delta is too large, assume time correction + * has occurred and set old_delta to the current delta. + */ + old_delta = delta; + } else { + /* Otherwise try to adjust old_system to compensate */ + timekeeping_suspend_time = + timespec64_add(timekeeping_suspend_time, delta_delta); + } } timekeeping_update(tk, TK_MIRROR); @@ -1327,7 +1483,7 @@ int timekeeping_suspend(void) write_seqcount_end(&tk_core.seq); raw_spin_unlock_irqrestore(&timekeeper_lock, flags); - clockevents_notify(CLOCK_EVT_NOTIFY_SUSPEND, NULL); + tick_suspend(); clocksource_suspend(); clockevents_suspend(); @@ -1416,15 +1572,15 @@ static __always_inline void timekeeping_apply_adjustment(struct timekeeper *tk, * * XXX - TODO: Doc ntp_error calculation. */ - if ((mult_adj > 0) && (tk->tkr.mult + mult_adj < mult_adj)) { + if ((mult_adj > 0) && (tk->tkr_mono.mult + mult_adj < mult_adj)) { /* NTP adjustment caused clocksource mult overflow */ WARN_ON_ONCE(1); return; } - tk->tkr.mult += mult_adj; + tk->tkr_mono.mult += mult_adj; tk->xtime_interval += interval; - tk->tkr.xtime_nsec -= offset; + tk->tkr_mono.xtime_nsec -= offset; tk->ntp_error -= (interval - offset) << tk->ntp_error_shift; } @@ -1486,13 +1642,13 @@ static void timekeeping_adjust(struct timekeeper *tk, s64 offset) tk->ntp_err_mult = 0; } - if (unlikely(tk->tkr.clock->maxadj && - (abs(tk->tkr.mult - tk->tkr.clock->mult) - > tk->tkr.clock->maxadj))) { + if (unlikely(tk->tkr_mono.clock->maxadj && + (abs(tk->tkr_mono.mult - tk->tkr_mono.clock->mult) + > tk->tkr_mono.clock->maxadj))) { printk_once(KERN_WARNING "Adjusting %s more than 11%% (%ld vs %ld)\n", - tk->tkr.clock->name, (long)tk->tkr.mult, - (long)tk->tkr.clock->mult + tk->tkr.clock->maxadj); + tk->tkr_mono.clock->name, (long)tk->tkr_mono.mult, + (long)tk->tkr_mono.clock->mult + tk->tkr_mono.clock->maxadj); } /* @@ -1509,9 +1665,9 @@ static void timekeeping_adjust(struct timekeeper *tk, s64 offset) * We'll correct this error next time through this function, when * xtime_nsec is not as small. */ - if (unlikely((s64)tk->tkr.xtime_nsec < 0)) { - s64 neg = -(s64)tk->tkr.xtime_nsec; - tk->tkr.xtime_nsec = 0; + if (unlikely((s64)tk->tkr_mono.xtime_nsec < 0)) { + s64 neg = -(s64)tk->tkr_mono.xtime_nsec; + tk->tkr_mono.xtime_nsec = 0; tk->ntp_error += neg << tk->ntp_error_shift; } } @@ -1526,13 +1682,13 @@ static void timekeeping_adjust(struct timekeeper *tk, s64 offset) */ static inline unsigned int accumulate_nsecs_to_secs(struct timekeeper *tk) { - u64 nsecps = (u64)NSEC_PER_SEC << tk->tkr.shift; + u64 nsecps = (u64)NSEC_PER_SEC << tk->tkr_mono.shift; unsigned int clock_set = 0; - while (tk->tkr.xtime_nsec >= nsecps) { + while (tk->tkr_mono.xtime_nsec >= nsecps) { int leap; - tk->tkr.xtime_nsec -= nsecps; + tk->tkr_mono.xtime_nsec -= nsecps; tk->xtime_sec++; /* Figure out if its a leap sec and apply if needed */ @@ -1577,9 +1733,10 @@ static cycle_t logarithmic_accumulation(struct timekeeper *tk, cycle_t offset, /* Accumulate one shifted interval */ offset -= interval; - tk->tkr.cycle_last += interval; + tk->tkr_mono.cycle_last += interval; + tk->tkr_raw.cycle_last += interval; - tk->tkr.xtime_nsec += tk->xtime_interval << shift; + tk->tkr_mono.xtime_nsec += tk->xtime_interval << shift; *clock_set |= accumulate_nsecs_to_secs(tk); /* Accumulate raw time */ @@ -1622,14 +1779,17 @@ void update_wall_time(void) #ifdef CONFIG_ARCH_USES_GETTIMEOFFSET offset = real_tk->cycle_interval; #else - offset = clocksource_delta(tk->tkr.read(tk->tkr.clock), - tk->tkr.cycle_last, tk->tkr.mask); + offset = clocksource_delta(tk->tkr_mono.read(tk->tkr_mono.clock), + tk->tkr_mono.cycle_last, tk->tkr_mono.mask); #endif /* Check if there's really nothing to do */ if (offset < real_tk->cycle_interval) goto out; + /* Do some additional sanity checking */ + timekeeping_check_update(real_tk, offset); + /* * With NO_HZ we may have to accumulate many cycle_intervals * (think "ticks") worth of time at once. To do this efficiently, @@ -1784,8 +1944,8 @@ ktime_t ktime_get_update_offsets_tick(ktime_t *offs_real, ktime_t *offs_boot, do { seq = read_seqcount_begin(&tk_core.seq); - base = tk->tkr.base_mono; - nsecs = tk->tkr.xtime_nsec >> tk->tkr.shift; + base = tk->tkr_mono.base; + nsecs = tk->tkr_mono.xtime_nsec >> tk->tkr_mono.shift; *offs_real = tk->offs_real; *offs_boot = tk->offs_boot; @@ -1816,8 +1976,8 @@ ktime_t ktime_get_update_offsets_now(ktime_t *offs_real, ktime_t *offs_boot, do { seq = read_seqcount_begin(&tk_core.seq); - base = tk->tkr.base_mono; - nsecs = timekeeping_get_ns(&tk->tkr); + base = tk->tkr_mono.base; + nsecs = timekeeping_get_ns(&tk->tkr_mono); *offs_real = tk->offs_real; *offs_boot = tk->offs_boot; diff --git a/kernel/time/timekeeping.h b/kernel/time/timekeeping.h index 1d91416055d5..ead8794b9a4e 100644 --- a/kernel/time/timekeeping.h +++ b/kernel/time/timekeeping.h @@ -19,4 +19,11 @@ extern void timekeeping_clocktai(struct timespec *ts); extern int timekeeping_suspend(void); extern void timekeeping_resume(void); +extern void do_timer(unsigned long ticks); +extern void update_wall_time(void); + +extern seqlock_t jiffies_lock; + +#define CS_NAME_LEN 32 + #endif diff --git a/kernel/time/timer.c b/kernel/time/timer.c index 2d3f5c504939..2ece3aa5069c 100644 --- a/kernel/time/timer.c +++ b/kernel/time/timer.c @@ -90,8 +90,18 @@ struct tvec_base { struct tvec tv5; } ____cacheline_aligned; +/* + * __TIMER_INITIALIZER() needs to set ->base to a valid pointer (because we've + * made NULL special, hint: lock_timer_base()) and we cannot get a compile time + * pointer to per-cpu entries because we don't know where we'll map the section, + * even for the boot cpu. + * + * And so we use boot_tvec_bases for boot CPU and per-cpu __tvec_bases for the + * rest of them. + */ struct tvec_base boot_tvec_bases; EXPORT_SYMBOL(boot_tvec_bases); + static DEFINE_PER_CPU(struct tvec_base *, tvec_bases) = &boot_tvec_bases; /* Functions below help us manage 'deferrable' flag */ @@ -1027,6 +1037,8 @@ int try_to_del_timer_sync(struct timer_list *timer) EXPORT_SYMBOL(try_to_del_timer_sync); #ifdef CONFIG_SMP +static DEFINE_PER_CPU(struct tvec_base, __tvec_bases); + /** * del_timer_sync - deactivate a timer and wait for the handler to finish. * @timer: the timer to be deactivated @@ -1532,64 +1544,6 @@ signed long __sched schedule_timeout_uninterruptible(signed long timeout) } EXPORT_SYMBOL(schedule_timeout_uninterruptible); -static int init_timers_cpu(int cpu) -{ - int j; - struct tvec_base *base; - static char tvec_base_done[NR_CPUS]; - - if (!tvec_base_done[cpu]) { - static char boot_done; - - if (boot_done) { - /* - * The APs use this path later in boot - */ - base = kzalloc_node(sizeof(*base), GFP_KERNEL, - cpu_to_node(cpu)); - if (!base) - return -ENOMEM; - - /* Make sure tvec_base has TIMER_FLAG_MASK bits free */ - if (WARN_ON(base != tbase_get_base(base))) { - kfree(base); - return -ENOMEM; - } - per_cpu(tvec_bases, cpu) = base; - } else { - /* - * This is for the boot CPU - we use compile-time - * static initialisation because per-cpu memory isn't - * ready yet and because the memory allocators are not - * initialised either. - */ - boot_done = 1; - base = &boot_tvec_bases; - } - spin_lock_init(&base->lock); - tvec_base_done[cpu] = 1; - base->cpu = cpu; - } else { - base = per_cpu(tvec_bases, cpu); - } - - - for (j = 0; j < TVN_SIZE; j++) { - INIT_LIST_HEAD(base->tv5.vec + j); - INIT_LIST_HEAD(base->tv4.vec + j); - INIT_LIST_HEAD(base->tv3.vec + j); - INIT_LIST_HEAD(base->tv2.vec + j); - } - for (j = 0; j < TVR_SIZE; j++) - INIT_LIST_HEAD(base->tv1.vec + j); - - base->timer_jiffies = jiffies; - base->next_timer = base->timer_jiffies; - base->active_timers = 0; - base->all_timers = 0; - return 0; -} - #ifdef CONFIG_HOTPLUG_CPU static void migrate_timer_list(struct tvec_base *new_base, struct list_head *head) { @@ -1631,55 +1585,86 @@ static void migrate_timers(int cpu) migrate_timer_list(new_base, old_base->tv5.vec + i); } + old_base->active_timers = 0; + old_base->all_timers = 0; + spin_unlock(&old_base->lock); spin_unlock_irq(&new_base->lock); put_cpu_var(tvec_bases); } -#endif /* CONFIG_HOTPLUG_CPU */ static int timer_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) { - long cpu = (long)hcpu; - int err; - - switch(action) { - case CPU_UP_PREPARE: - case CPU_UP_PREPARE_FROZEN: - err = init_timers_cpu(cpu); - if (err < 0) - return notifier_from_errno(err); - break; -#ifdef CONFIG_HOTPLUG_CPU + switch (action) { case CPU_DEAD: case CPU_DEAD_FROZEN: - migrate_timers(cpu); + migrate_timers((long)hcpu); break; -#endif default: break; } + return NOTIFY_OK; } -static struct notifier_block timers_nb = { - .notifier_call = timer_cpu_notify, -}; +static inline void timer_register_cpu_notifier(void) +{ + cpu_notifier(timer_cpu_notify, 0); +} +#else +static inline void timer_register_cpu_notifier(void) { } +#endif /* CONFIG_HOTPLUG_CPU */ +static void __init init_timer_cpu(struct tvec_base *base, int cpu) +{ + int j; -void __init init_timers(void) + BUG_ON(base != tbase_get_base(base)); + + base->cpu = cpu; + per_cpu(tvec_bases, cpu) = base; + spin_lock_init(&base->lock); + + for (j = 0; j < TVN_SIZE; j++) { + INIT_LIST_HEAD(base->tv5.vec + j); + INIT_LIST_HEAD(base->tv4.vec + j); + INIT_LIST_HEAD(base->tv3.vec + j); + INIT_LIST_HEAD(base->tv2.vec + j); + } + for (j = 0; j < TVR_SIZE; j++) + INIT_LIST_HEAD(base->tv1.vec + j); + + base->timer_jiffies = jiffies; + base->next_timer = base->timer_jiffies; +} + +static void __init init_timer_cpus(void) { - int err; + struct tvec_base *base; + int local_cpu = smp_processor_id(); + int cpu; + for_each_possible_cpu(cpu) { + if (cpu == local_cpu) + base = &boot_tvec_bases; +#ifdef CONFIG_SMP + else + base = per_cpu_ptr(&__tvec_bases, cpu); +#endif + + init_timer_cpu(base, cpu); + } +} + +void __init init_timers(void) +{ /* ensure there are enough low bits for flags in timer->base pointer */ BUILD_BUG_ON(__alignof__(struct tvec_base) & TIMER_FLAG_MASK); - err = timer_cpu_notify(&timers_nb, (unsigned long)CPU_UP_PREPARE, - (void *)(long)smp_processor_id()); - BUG_ON(err != NOTIFY_OK); - + init_timer_cpus(); init_timer_stats(); - register_cpu_notifier(&timers_nb); + timer_register_cpu_notifier(); open_softirq(TIMER_SOFTIRQ, run_timer_softirq); } diff --git a/kernel/time/timer_list.c b/kernel/time/timer_list.c index 61ed862cdd37..e878c2e0ba45 100644 --- a/kernel/time/timer_list.c +++ b/kernel/time/timer_list.c @@ -16,10 +16,10 @@ #include <linux/sched.h> #include <linux/seq_file.h> #include <linux/kallsyms.h> -#include <linux/tick.h> #include <asm/uaccess.h> +#include "tick-internal.h" struct timer_list_iter { int cpu; @@ -228,9 +228,35 @@ print_tickdevice(struct seq_file *m, struct tick_device *td, int cpu) print_name_offset(m, dev->set_next_event); SEQ_printf(m, "\n"); - SEQ_printf(m, " set_mode: "); - print_name_offset(m, dev->set_mode); - SEQ_printf(m, "\n"); + if (dev->set_mode) { + SEQ_printf(m, " set_mode: "); + print_name_offset(m, dev->set_mode); + SEQ_printf(m, "\n"); + } else { + if (dev->set_state_shutdown) { + SEQ_printf(m, " shutdown: "); + print_name_offset(m, dev->set_state_shutdown); + SEQ_printf(m, "\n"); + } + + if (dev->set_state_periodic) { + SEQ_printf(m, " periodic: "); + print_name_offset(m, dev->set_state_periodic); + SEQ_printf(m, "\n"); + } + + if (dev->set_state_oneshot) { + SEQ_printf(m, " oneshot: "); + print_name_offset(m, dev->set_state_oneshot); + SEQ_printf(m, "\n"); + } + + if (dev->tick_resume) { + SEQ_printf(m, " resume: "); + print_name_offset(m, dev->tick_resume); + SEQ_printf(m, "\n"); + } + } SEQ_printf(m, " event_handler: "); print_name_offset(m, dev->event_handler); diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 45e5cb143d17..4f228024055b 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -1059,6 +1059,12 @@ static __init void ftrace_profile_debugfs(struct dentry *d_tracer) static struct pid * const ftrace_swapper_pid = &init_struct_pid; +#ifdef CONFIG_FUNCTION_GRAPH_TRACER +static int ftrace_graph_active; +#else +# define ftrace_graph_active 0 +#endif + #ifdef CONFIG_DYNAMIC_FTRACE static struct ftrace_ops *removed_ops; @@ -2041,8 +2047,12 @@ static int ftrace_check_record(struct dyn_ftrace *rec, int enable, int update) if (!ftrace_rec_count(rec)) rec->flags = 0; else - /* Just disable the record (keep REGS state) */ - rec->flags &= ~FTRACE_FL_ENABLED; + /* + * Just disable the record, but keep the ops TRAMP + * and REGS states. The _EN flags must be disabled though. + */ + rec->flags &= ~(FTRACE_FL_ENABLED | FTRACE_FL_TRAMP_EN | + FTRACE_FL_REGS_EN); } return FTRACE_UPDATE_MAKE_NOP; @@ -2688,24 +2698,36 @@ static int ftrace_shutdown(struct ftrace_ops *ops, int command) static void ftrace_startup_sysctl(void) { + int command; + if (unlikely(ftrace_disabled)) return; /* Force update next time */ saved_ftrace_func = NULL; /* ftrace_start_up is true if we want ftrace running */ - if (ftrace_start_up) - ftrace_run_update_code(FTRACE_UPDATE_CALLS); + if (ftrace_start_up) { + command = FTRACE_UPDATE_CALLS; + if (ftrace_graph_active) + command |= FTRACE_START_FUNC_RET; + ftrace_startup_enable(command); + } } static void ftrace_shutdown_sysctl(void) { + int command; + if (unlikely(ftrace_disabled)) return; /* ftrace_start_up is true if ftrace is running */ - if (ftrace_start_up) - ftrace_run_update_code(FTRACE_DISABLE_CALLS); + if (ftrace_start_up) { + command = FTRACE_DISABLE_CALLS; + if (ftrace_graph_active) + command |= FTRACE_STOP_FUNC_RET; + ftrace_run_update_code(command); + } } static cycle_t ftrace_update_time; @@ -5558,12 +5580,12 @@ ftrace_enable_sysctl(struct ctl_table *table, int write, if (ftrace_enabled) { - ftrace_startup_sysctl(); - /* we are starting ftrace again */ if (ftrace_ops_list != &ftrace_list_end) update_ftrace_function(); + ftrace_startup_sysctl(); + } else { /* stopping ftrace calls (just send to ftrace_stub) */ ftrace_trace_function = ftrace_stub; @@ -5590,8 +5612,6 @@ static struct ftrace_ops graph_ops = { ASSIGN_OPS_HASH(graph_ops, &global_ops.local_hash) }; -static int ftrace_graph_active; - int ftrace_graph_entry_stub(struct ftrace_graph_ent *trace) { return 0; diff --git a/kernel/workqueue.c b/kernel/workqueue.c index f28849394791..586ad91300b0 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -159,6 +159,7 @@ struct worker_pool { /* see manage_workers() for details on the two manager mutexes */ struct mutex manager_arb; /* manager arbitration */ + struct worker *manager; /* L: purely informational */ struct mutex attach_mutex; /* attach/detach exclusion */ struct list_head workers; /* A: attached workers */ struct completion *detach_completion; /* all workers detached */ @@ -230,7 +231,7 @@ struct wq_device; */ struct workqueue_struct { struct list_head pwqs; /* WR: all pwqs of this wq */ - struct list_head list; /* PL: list of all workqueues */ + struct list_head list; /* PR: list of all workqueues */ struct mutex mutex; /* protects this wq */ int work_color; /* WQ: current work color */ @@ -257,6 +258,13 @@ struct workqueue_struct { #endif char name[WQ_NAME_LEN]; /* I: workqueue name */ + /* + * Destruction of workqueue_struct is sched-RCU protected to allow + * walking the workqueues list without grabbing wq_pool_mutex. + * This is used to dump all workqueues from sysrq. + */ + struct rcu_head rcu; + /* hot fields used during command issue, aligned to cacheline */ unsigned int flags ____cacheline_aligned; /* WQ: WQ_* flags */ struct pool_workqueue __percpu *cpu_pwqs; /* I: per-cpu pwqs */ @@ -288,7 +296,7 @@ static struct workqueue_attrs *wq_update_unbound_numa_attrs_buf; static DEFINE_MUTEX(wq_pool_mutex); /* protects pools and workqueues list */ static DEFINE_SPINLOCK(wq_mayday_lock); /* protects wq->maydays list */ -static LIST_HEAD(workqueues); /* PL: list of all workqueues */ +static LIST_HEAD(workqueues); /* PR: list of all workqueues */ static bool workqueue_freezing; /* PL: have wqs started freezing? */ /* the per-cpu worker pools */ @@ -324,6 +332,7 @@ EXPORT_SYMBOL_GPL(system_freezable_power_efficient_wq); static int worker_thread(void *__worker); static void copy_workqueue_attrs(struct workqueue_attrs *to, const struct workqueue_attrs *from); +static void workqueue_sysfs_unregister(struct workqueue_struct *wq); #define CREATE_TRACE_POINTS #include <trace/events/workqueue.h> @@ -1911,9 +1920,11 @@ static bool manage_workers(struct worker *worker) */ if (!mutex_trylock(&pool->manager_arb)) return false; + pool->manager = worker; maybe_create_worker(pool); + pool->manager = NULL; mutex_unlock(&pool->manager_arb); return true; } @@ -2303,6 +2314,7 @@ repeat: struct wq_barrier { struct work_struct work; struct completion done; + struct task_struct *task; /* purely informational */ }; static void wq_barrier_func(struct work_struct *work) @@ -2351,6 +2363,7 @@ static void insert_wq_barrier(struct pool_workqueue *pwq, INIT_WORK_ONSTACK(&barr->work, wq_barrier_func); __set_bit(WORK_STRUCT_PENDING_BIT, work_data_bits(&barr->work)); init_completion(&barr->done); + barr->task = current; /* * If @target is currently being executed, schedule the @@ -2728,19 +2741,57 @@ bool flush_work(struct work_struct *work) } EXPORT_SYMBOL_GPL(flush_work); +struct cwt_wait { + wait_queue_t wait; + struct work_struct *work; +}; + +static int cwt_wakefn(wait_queue_t *wait, unsigned mode, int sync, void *key) +{ + struct cwt_wait *cwait = container_of(wait, struct cwt_wait, wait); + + if (cwait->work != key) + return 0; + return autoremove_wake_function(wait, mode, sync, key); +} + static bool __cancel_work_timer(struct work_struct *work, bool is_dwork) { + static DECLARE_WAIT_QUEUE_HEAD(cancel_waitq); unsigned long flags; int ret; do { ret = try_to_grab_pending(work, is_dwork, &flags); /* - * If someone else is canceling, wait for the same event it - * would be waiting for before retrying. + * If someone else is already canceling, wait for it to + * finish. flush_work() doesn't work for PREEMPT_NONE + * because we may get scheduled between @work's completion + * and the other canceling task resuming and clearing + * CANCELING - flush_work() will return false immediately + * as @work is no longer busy, try_to_grab_pending() will + * return -ENOENT as @work is still being canceled and the + * other canceling task won't be able to clear CANCELING as + * we're hogging the CPU. + * + * Let's wait for completion using a waitqueue. As this + * may lead to the thundering herd problem, use a custom + * wake function which matches @work along with exclusive + * wait and wakeup. */ - if (unlikely(ret == -ENOENT)) - flush_work(work); + if (unlikely(ret == -ENOENT)) { + struct cwt_wait cwait; + + init_wait(&cwait.wait); + cwait.wait.func = cwt_wakefn; + cwait.work = work; + + prepare_to_wait_exclusive(&cancel_waitq, &cwait.wait, + TASK_UNINTERRUPTIBLE); + if (work_is_canceling(work)) + schedule(); + finish_wait(&cancel_waitq, &cwait.wait); + } } while (unlikely(ret < 0)); /* tell other tasks trying to grab @work to back off */ @@ -2749,6 +2800,16 @@ static bool __cancel_work_timer(struct work_struct *work, bool is_dwork) flush_work(work); clear_work_data(work); + + /* + * Paired with prepare_to_wait() above so that either + * waitqueue_active() is visible here or !work_is_canceling() is + * visible there. + */ + smp_mb(); + if (waitqueue_active(&cancel_waitq)) + __wake_up(&cancel_waitq, TASK_NORMAL, 1, work); + return ret; } @@ -2941,323 +3002,6 @@ int execute_in_process_context(work_func_t fn, struct execute_work *ew) } EXPORT_SYMBOL_GPL(execute_in_process_context); -#ifdef CONFIG_SYSFS -/* - * Workqueues with WQ_SYSFS flag set is visible to userland via - * /sys/bus/workqueue/devices/WQ_NAME. All visible workqueues have the - * following attributes. - * - * per_cpu RO bool : whether the workqueue is per-cpu or unbound - * max_active RW int : maximum number of in-flight work items - * - * Unbound workqueues have the following extra attributes. - * - * id RO int : the associated pool ID - * nice RW int : nice value of the workers - * cpumask RW mask : bitmask of allowed CPUs for the workers - */ -struct wq_device { - struct workqueue_struct *wq; - struct device dev; -}; - -static struct workqueue_struct *dev_to_wq(struct device *dev) -{ - struct wq_device *wq_dev = container_of(dev, struct wq_device, dev); - - return wq_dev->wq; -} - -static ssize_t per_cpu_show(struct device *dev, struct device_attribute *attr, - char *buf) -{ - struct workqueue_struct *wq = dev_to_wq(dev); - - return scnprintf(buf, PAGE_SIZE, "%d\n", (bool)!(wq->flags & WQ_UNBOUND)); -} -static DEVICE_ATTR_RO(per_cpu); - -static ssize_t max_active_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct workqueue_struct *wq = dev_to_wq(dev); - - return scnprintf(buf, PAGE_SIZE, "%d\n", wq->saved_max_active); -} - -static ssize_t max_active_store(struct device *dev, - struct device_attribute *attr, const char *buf, - size_t count) -{ - struct workqueue_struct *wq = dev_to_wq(dev); - int val; - - if (sscanf(buf, "%d", &val) != 1 || val <= 0) - return -EINVAL; - - workqueue_set_max_active(wq, val); - return count; -} -static DEVICE_ATTR_RW(max_active); - -static struct attribute *wq_sysfs_attrs[] = { - &dev_attr_per_cpu.attr, - &dev_attr_max_active.attr, - NULL, -}; -ATTRIBUTE_GROUPS(wq_sysfs); - -static ssize_t wq_pool_ids_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct workqueue_struct *wq = dev_to_wq(dev); - const char *delim = ""; - int node, written = 0; - - rcu_read_lock_sched(); - for_each_node(node) { - written += scnprintf(buf + written, PAGE_SIZE - written, - "%s%d:%d", delim, node, - unbound_pwq_by_node(wq, node)->pool->id); - delim = " "; - } - written += scnprintf(buf + written, PAGE_SIZE - written, "\n"); - rcu_read_unlock_sched(); - - return written; -} - -static ssize_t wq_nice_show(struct device *dev, struct device_attribute *attr, - char *buf) -{ - struct workqueue_struct *wq = dev_to_wq(dev); - int written; - - mutex_lock(&wq->mutex); - written = scnprintf(buf, PAGE_SIZE, "%d\n", wq->unbound_attrs->nice); - mutex_unlock(&wq->mutex); - - return written; -} - -/* prepare workqueue_attrs for sysfs store operations */ -static struct workqueue_attrs *wq_sysfs_prep_attrs(struct workqueue_struct *wq) -{ - struct workqueue_attrs *attrs; - - attrs = alloc_workqueue_attrs(GFP_KERNEL); - if (!attrs) - return NULL; - - mutex_lock(&wq->mutex); - copy_workqueue_attrs(attrs, wq->unbound_attrs); - mutex_unlock(&wq->mutex); - return attrs; -} - -static ssize_t wq_nice_store(struct device *dev, struct device_attribute *attr, - const char *buf, size_t count) -{ - struct workqueue_struct *wq = dev_to_wq(dev); - struct workqueue_attrs *attrs; - int ret; - - attrs = wq_sysfs_prep_attrs(wq); - if (!attrs) - return -ENOMEM; - - if (sscanf(buf, "%d", &attrs->nice) == 1 && - attrs->nice >= MIN_NICE && attrs->nice <= MAX_NICE) - ret = apply_workqueue_attrs(wq, attrs); - else - ret = -EINVAL; - - free_workqueue_attrs(attrs); - return ret ?: count; -} - -static ssize_t wq_cpumask_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct workqueue_struct *wq = dev_to_wq(dev); - int written; - - mutex_lock(&wq->mutex); - written = scnprintf(buf, PAGE_SIZE, "%*pb\n", - cpumask_pr_args(wq->unbound_attrs->cpumask)); - mutex_unlock(&wq->mutex); - return written; -} - -static ssize_t wq_cpumask_store(struct device *dev, - struct device_attribute *attr, - const char *buf, size_t count) -{ - struct workqueue_struct *wq = dev_to_wq(dev); - struct workqueue_attrs *attrs; - int ret; - - attrs = wq_sysfs_prep_attrs(wq); - if (!attrs) - return -ENOMEM; - - ret = cpumask_parse(buf, attrs->cpumask); - if (!ret) - ret = apply_workqueue_attrs(wq, attrs); - - free_workqueue_attrs(attrs); - return ret ?: count; -} - -static ssize_t wq_numa_show(struct device *dev, struct device_attribute *attr, - char *buf) -{ - struct workqueue_struct *wq = dev_to_wq(dev); - int written; - - mutex_lock(&wq->mutex); - written = scnprintf(buf, PAGE_SIZE, "%d\n", - !wq->unbound_attrs->no_numa); - mutex_unlock(&wq->mutex); - - return written; -} - -static ssize_t wq_numa_store(struct device *dev, struct device_attribute *attr, - const char *buf, size_t count) -{ - struct workqueue_struct *wq = dev_to_wq(dev); - struct workqueue_attrs *attrs; - int v, ret; - - attrs = wq_sysfs_prep_attrs(wq); - if (!attrs) - return -ENOMEM; - - ret = -EINVAL; - if (sscanf(buf, "%d", &v) == 1) { - attrs->no_numa = !v; - ret = apply_workqueue_attrs(wq, attrs); - } - - free_workqueue_attrs(attrs); - return ret ?: count; -} - -static struct device_attribute wq_sysfs_unbound_attrs[] = { - __ATTR(pool_ids, 0444, wq_pool_ids_show, NULL), - __ATTR(nice, 0644, wq_nice_show, wq_nice_store), - __ATTR(cpumask, 0644, wq_cpumask_show, wq_cpumask_store), - __ATTR(numa, 0644, wq_numa_show, wq_numa_store), - __ATTR_NULL, -}; - -static struct bus_type wq_subsys = { - .name = "workqueue", - .dev_groups = wq_sysfs_groups, -}; - -static int __init wq_sysfs_init(void) -{ - return subsys_virtual_register(&wq_subsys, NULL); -} -core_initcall(wq_sysfs_init); - -static void wq_device_release(struct device *dev) -{ - struct wq_device *wq_dev = container_of(dev, struct wq_device, dev); - - kfree(wq_dev); -} - -/** - * workqueue_sysfs_register - make a workqueue visible in sysfs - * @wq: the workqueue to register - * - * Expose @wq in sysfs under /sys/bus/workqueue/devices. - * alloc_workqueue*() automatically calls this function if WQ_SYSFS is set - * which is the preferred method. - * - * Workqueue user should use this function directly iff it wants to apply - * workqueue_attrs before making the workqueue visible in sysfs; otherwise, - * apply_workqueue_attrs() may race against userland updating the - * attributes. - * - * Return: 0 on success, -errno on failure. - */ -int workqueue_sysfs_register(struct workqueue_struct *wq) -{ - struct wq_device *wq_dev; - int ret; - - /* - * Adjusting max_active or creating new pwqs by applyting - * attributes breaks ordering guarantee. Disallow exposing ordered - * workqueues. - */ - if (WARN_ON(wq->flags & __WQ_ORDERED)) - return -EINVAL; - - wq->wq_dev = wq_dev = kzalloc(sizeof(*wq_dev), GFP_KERNEL); - if (!wq_dev) - return -ENOMEM; - - wq_dev->wq = wq; - wq_dev->dev.bus = &wq_subsys; - wq_dev->dev.init_name = wq->name; - wq_dev->dev.release = wq_device_release; - - /* - * unbound_attrs are created separately. Suppress uevent until - * everything is ready. - */ - dev_set_uevent_suppress(&wq_dev->dev, true); - - ret = device_register(&wq_dev->dev); - if (ret) { - kfree(wq_dev); - wq->wq_dev = NULL; - return ret; - } - - if (wq->flags & WQ_UNBOUND) { - struct device_attribute *attr; - - for (attr = wq_sysfs_unbound_attrs; attr->attr.name; attr++) { - ret = device_create_file(&wq_dev->dev, attr); - if (ret) { - device_unregister(&wq_dev->dev); - wq->wq_dev = NULL; - return ret; - } - } - } - - dev_set_uevent_suppress(&wq_dev->dev, false); - kobject_uevent(&wq_dev->dev.kobj, KOBJ_ADD); - return 0; -} - -/** - * workqueue_sysfs_unregister - undo workqueue_sysfs_register() - * @wq: the workqueue to unregister - * - * If @wq is registered to sysfs by workqueue_sysfs_register(), unregister. - */ -static void workqueue_sysfs_unregister(struct workqueue_struct *wq) -{ - struct wq_device *wq_dev = wq->wq_dev; - - if (!wq->wq_dev) - return; - - wq->wq_dev = NULL; - device_unregister(&wq_dev->dev); -} -#else /* CONFIG_SYSFS */ -static void workqueue_sysfs_unregister(struct workqueue_struct *wq) { } -#endif /* CONFIG_SYSFS */ - /** * free_workqueue_attrs - free a workqueue_attrs * @attrs: workqueue_attrs to free @@ -3376,6 +3120,20 @@ static int init_worker_pool(struct worker_pool *pool) return 0; } +static void rcu_free_wq(struct rcu_head *rcu) +{ + struct workqueue_struct *wq = + container_of(rcu, struct workqueue_struct, rcu); + + if (!(wq->flags & WQ_UNBOUND)) + free_percpu(wq->cpu_pwqs); + else + free_workqueue_attrs(wq->unbound_attrs); + + kfree(wq->rescuer); + kfree(wq); +} + static void rcu_free_pool(struct rcu_head *rcu) { struct worker_pool *pool = container_of(rcu, struct worker_pool, rcu); @@ -3553,12 +3311,10 @@ static void pwq_unbound_release_workfn(struct work_struct *work) /* * If we're the last pwq going away, @wq is already dead and no one - * is gonna access it anymore. Free it. + * is gonna access it anymore. Schedule RCU free. */ - if (is_last) { - free_workqueue_attrs(wq->unbound_attrs); - kfree(wq); - } + if (is_last) + call_rcu_sched(&wq->rcu, rcu_free_wq); } /** @@ -4095,7 +3851,7 @@ struct workqueue_struct *__alloc_workqueue_key(const char *fmt, pwq_adjust_max_active(pwq); mutex_unlock(&wq->mutex); - list_add(&wq->list, &workqueues); + list_add_tail_rcu(&wq->list, &workqueues); mutex_unlock(&wq_pool_mutex); @@ -4151,24 +3907,20 @@ void destroy_workqueue(struct workqueue_struct *wq) * flushing is complete in case freeze races us. */ mutex_lock(&wq_pool_mutex); - list_del_init(&wq->list); + list_del_rcu(&wq->list); mutex_unlock(&wq_pool_mutex); workqueue_sysfs_unregister(wq); - if (wq->rescuer) { + if (wq->rescuer) kthread_stop(wq->rescuer->task); - kfree(wq->rescuer); - wq->rescuer = NULL; - } if (!(wq->flags & WQ_UNBOUND)) { /* * The base ref is never dropped on per-cpu pwqs. Directly - * free the pwqs and wq. + * schedule RCU free. */ - free_percpu(wq->cpu_pwqs); - kfree(wq); + call_rcu_sched(&wq->rcu, rcu_free_wq); } else { /* * We're the sole accessor of @wq at this point. Directly @@ -4389,6 +4141,166 @@ void print_worker_info(const char *log_lvl, struct task_struct *task) } } +static void pr_cont_pool_info(struct worker_pool *pool) +{ + pr_cont(" cpus=%*pbl", nr_cpumask_bits, pool->attrs->cpumask); + if (pool->node != NUMA_NO_NODE) + pr_cont(" node=%d", pool->node); + pr_cont(" flags=0x%x nice=%d", pool->flags, pool->attrs->nice); +} + +static void pr_cont_work(bool comma, struct work_struct *work) +{ + if (work->func == wq_barrier_func) { + struct wq_barrier *barr; + + barr = container_of(work, struct wq_barrier, work); + + pr_cont("%s BAR(%d)", comma ? "," : "", + task_pid_nr(barr->task)); + } else { + pr_cont("%s %pf", comma ? "," : "", work->func); + } +} + +static void show_pwq(struct pool_workqueue *pwq) +{ + struct worker_pool *pool = pwq->pool; + struct work_struct *work; + struct worker *worker; + bool has_in_flight = false, has_pending = false; + int bkt; + + pr_info(" pwq %d:", pool->id); + pr_cont_pool_info(pool); + + pr_cont(" active=%d/%d%s\n", pwq->nr_active, pwq->max_active, + !list_empty(&pwq->mayday_node) ? " MAYDAY" : ""); + + hash_for_each(pool->busy_hash, bkt, worker, hentry) { + if (worker->current_pwq == pwq) { + has_in_flight = true; + break; + } + } + if (has_in_flight) { + bool comma = false; + + pr_info(" in-flight:"); + hash_for_each(pool->busy_hash, bkt, worker, hentry) { + if (worker->current_pwq != pwq) + continue; + + pr_cont("%s %d%s:%pf", comma ? "," : "", + task_pid_nr(worker->task), + worker == pwq->wq->rescuer ? "(RESCUER)" : "", + worker->current_func); + list_for_each_entry(work, &worker->scheduled, entry) + pr_cont_work(false, work); + comma = true; + } + pr_cont("\n"); + } + + list_for_each_entry(work, &pool->worklist, entry) { + if (get_work_pwq(work) == pwq) { + has_pending = true; + break; + } + } + if (has_pending) { + bool comma = false; + + pr_info(" pending:"); + list_for_each_entry(work, &pool->worklist, entry) { + if (get_work_pwq(work) != pwq) + continue; + + pr_cont_work(comma, work); + comma = !(*work_data_bits(work) & WORK_STRUCT_LINKED); + } + pr_cont("\n"); + } + + if (!list_empty(&pwq->delayed_works)) { + bool comma = false; + + pr_info(" delayed:"); + list_for_each_entry(work, &pwq->delayed_works, entry) { + pr_cont_work(comma, work); + comma = !(*work_data_bits(work) & WORK_STRUCT_LINKED); + } + pr_cont("\n"); + } +} + +/** + * show_workqueue_state - dump workqueue state + * + * Called from a sysrq handler and prints out all busy workqueues and + * pools. + */ +void show_workqueue_state(void) +{ + struct workqueue_struct *wq; + struct worker_pool *pool; + unsigned long flags; + int pi; + + rcu_read_lock_sched(); + + pr_info("Showing busy workqueues and worker pools:\n"); + + list_for_each_entry_rcu(wq, &workqueues, list) { + struct pool_workqueue *pwq; + bool idle = true; + + for_each_pwq(pwq, wq) { + if (pwq->nr_active || !list_empty(&pwq->delayed_works)) { + idle = false; + break; + } + } + if (idle) + continue; + + pr_info("workqueue %s: flags=0x%x\n", wq->name, wq->flags); + + for_each_pwq(pwq, wq) { + spin_lock_irqsave(&pwq->pool->lock, flags); + if (pwq->nr_active || !list_empty(&pwq->delayed_works)) + show_pwq(pwq); + spin_unlock_irqrestore(&pwq->pool->lock, flags); + } + } + + for_each_pool(pool, pi) { + struct worker *worker; + bool first = true; + + spin_lock_irqsave(&pool->lock, flags); + if (pool->nr_workers == pool->nr_idle) + goto next_pool; + + pr_info("pool %d:", pool->id); + pr_cont_pool_info(pool); + pr_cont(" workers=%d", pool->nr_workers); + if (pool->manager) + pr_cont(" manager: %d", + task_pid_nr(pool->manager->task)); + list_for_each_entry(worker, &pool->idle_list, entry) { + pr_cont(" %s%d", first ? "idle: " : "", + task_pid_nr(worker->task)); + first = false; + } + pr_cont("\n"); + next_pool: + spin_unlock_irqrestore(&pool->lock, flags); + } + + rcu_read_unlock_sched(); +} + /* * CPU hotplug. * @@ -4786,6 +4698,323 @@ out_unlock: } #endif /* CONFIG_FREEZER */ +#ifdef CONFIG_SYSFS +/* + * Workqueues with WQ_SYSFS flag set is visible to userland via + * /sys/bus/workqueue/devices/WQ_NAME. All visible workqueues have the + * following attributes. + * + * per_cpu RO bool : whether the workqueue is per-cpu or unbound + * max_active RW int : maximum number of in-flight work items + * + * Unbound workqueues have the following extra attributes. + * + * id RO int : the associated pool ID + * nice RW int : nice value of the workers + * cpumask RW mask : bitmask of allowed CPUs for the workers + */ +struct wq_device { + struct workqueue_struct *wq; + struct device dev; +}; + +static struct workqueue_struct *dev_to_wq(struct device *dev) +{ + struct wq_device *wq_dev = container_of(dev, struct wq_device, dev); + + return wq_dev->wq; +} + +static ssize_t per_cpu_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct workqueue_struct *wq = dev_to_wq(dev); + + return scnprintf(buf, PAGE_SIZE, "%d\n", (bool)!(wq->flags & WQ_UNBOUND)); +} +static DEVICE_ATTR_RO(per_cpu); + +static ssize_t max_active_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct workqueue_struct *wq = dev_to_wq(dev); + + return scnprintf(buf, PAGE_SIZE, "%d\n", wq->saved_max_active); +} + +static ssize_t max_active_store(struct device *dev, + struct device_attribute *attr, const char *buf, + size_t count) +{ + struct workqueue_struct *wq = dev_to_wq(dev); + int val; + + if (sscanf(buf, "%d", &val) != 1 || val <= 0) + return -EINVAL; + + workqueue_set_max_active(wq, val); + return count; +} +static DEVICE_ATTR_RW(max_active); + +static struct attribute *wq_sysfs_attrs[] = { + &dev_attr_per_cpu.attr, + &dev_attr_max_active.attr, + NULL, +}; +ATTRIBUTE_GROUPS(wq_sysfs); + +static ssize_t wq_pool_ids_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct workqueue_struct *wq = dev_to_wq(dev); + const char *delim = ""; + int node, written = 0; + + rcu_read_lock_sched(); + for_each_node(node) { + written += scnprintf(buf + written, PAGE_SIZE - written, + "%s%d:%d", delim, node, + unbound_pwq_by_node(wq, node)->pool->id); + delim = " "; + } + written += scnprintf(buf + written, PAGE_SIZE - written, "\n"); + rcu_read_unlock_sched(); + + return written; +} + +static ssize_t wq_nice_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct workqueue_struct *wq = dev_to_wq(dev); + int written; + + mutex_lock(&wq->mutex); + written = scnprintf(buf, PAGE_SIZE, "%d\n", wq->unbound_attrs->nice); + mutex_unlock(&wq->mutex); + + return written; +} + +/* prepare workqueue_attrs for sysfs store operations */ +static struct workqueue_attrs *wq_sysfs_prep_attrs(struct workqueue_struct *wq) +{ + struct workqueue_attrs *attrs; + + attrs = alloc_workqueue_attrs(GFP_KERNEL); + if (!attrs) + return NULL; + + mutex_lock(&wq->mutex); + copy_workqueue_attrs(attrs, wq->unbound_attrs); + mutex_unlock(&wq->mutex); + return attrs; +} + +static ssize_t wq_nice_store(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct workqueue_struct *wq = dev_to_wq(dev); + struct workqueue_attrs *attrs; + int ret; + + attrs = wq_sysfs_prep_attrs(wq); + if (!attrs) + return -ENOMEM; + + if (sscanf(buf, "%d", &attrs->nice) == 1 && + attrs->nice >= MIN_NICE && attrs->nice <= MAX_NICE) + ret = apply_workqueue_attrs(wq, attrs); + else + ret = -EINVAL; + + free_workqueue_attrs(attrs); + return ret ?: count; +} + +static ssize_t wq_cpumask_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct workqueue_struct *wq = dev_to_wq(dev); + int written; + + mutex_lock(&wq->mutex); + written = scnprintf(buf, PAGE_SIZE, "%*pb\n", + cpumask_pr_args(wq->unbound_attrs->cpumask)); + mutex_unlock(&wq->mutex); + return written; +} + +static ssize_t wq_cpumask_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + struct workqueue_struct *wq = dev_to_wq(dev); + struct workqueue_attrs *attrs; + int ret; + + attrs = wq_sysfs_prep_attrs(wq); + if (!attrs) + return -ENOMEM; + + ret = cpumask_parse(buf, attrs->cpumask); + if (!ret) + ret = apply_workqueue_attrs(wq, attrs); + + free_workqueue_attrs(attrs); + return ret ?: count; +} + +static ssize_t wq_numa_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct workqueue_struct *wq = dev_to_wq(dev); + int written; + + mutex_lock(&wq->mutex); + written = scnprintf(buf, PAGE_SIZE, "%d\n", + !wq->unbound_attrs->no_numa); + mutex_unlock(&wq->mutex); + + return written; +} + +static ssize_t wq_numa_store(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct workqueue_struct *wq = dev_to_wq(dev); + struct workqueue_attrs *attrs; + int v, ret; + + attrs = wq_sysfs_prep_attrs(wq); + if (!attrs) + return -ENOMEM; + + ret = -EINVAL; + if (sscanf(buf, "%d", &v) == 1) { + attrs->no_numa = !v; + ret = apply_workqueue_attrs(wq, attrs); + } + + free_workqueue_attrs(attrs); + return ret ?: count; +} + +static struct device_attribute wq_sysfs_unbound_attrs[] = { + __ATTR(pool_ids, 0444, wq_pool_ids_show, NULL), + __ATTR(nice, 0644, wq_nice_show, wq_nice_store), + __ATTR(cpumask, 0644, wq_cpumask_show, wq_cpumask_store), + __ATTR(numa, 0644, wq_numa_show, wq_numa_store), + __ATTR_NULL, +}; + +static struct bus_type wq_subsys = { + .name = "workqueue", + .dev_groups = wq_sysfs_groups, +}; + +static int __init wq_sysfs_init(void) +{ + return subsys_virtual_register(&wq_subsys, NULL); +} +core_initcall(wq_sysfs_init); + +static void wq_device_release(struct device *dev) +{ + struct wq_device *wq_dev = container_of(dev, struct wq_device, dev); + + kfree(wq_dev); +} + +/** + * workqueue_sysfs_register - make a workqueue visible in sysfs + * @wq: the workqueue to register + * + * Expose @wq in sysfs under /sys/bus/workqueue/devices. + * alloc_workqueue*() automatically calls this function if WQ_SYSFS is set + * which is the preferred method. + * + * Workqueue user should use this function directly iff it wants to apply + * workqueue_attrs before making the workqueue visible in sysfs; otherwise, + * apply_workqueue_attrs() may race against userland updating the + * attributes. + * + * Return: 0 on success, -errno on failure. + */ +int workqueue_sysfs_register(struct workqueue_struct *wq) +{ + struct wq_device *wq_dev; + int ret; + + /* + * Adjusting max_active or creating new pwqs by applyting + * attributes breaks ordering guarantee. Disallow exposing ordered + * workqueues. + */ + if (WARN_ON(wq->flags & __WQ_ORDERED)) + return -EINVAL; + + wq->wq_dev = wq_dev = kzalloc(sizeof(*wq_dev), GFP_KERNEL); + if (!wq_dev) + return -ENOMEM; + + wq_dev->wq = wq; + wq_dev->dev.bus = &wq_subsys; + wq_dev->dev.init_name = wq->name; + wq_dev->dev.release = wq_device_release; + + /* + * unbound_attrs are created separately. Suppress uevent until + * everything is ready. + */ + dev_set_uevent_suppress(&wq_dev->dev, true); + + ret = device_register(&wq_dev->dev); + if (ret) { + kfree(wq_dev); + wq->wq_dev = NULL; + return ret; + } + + if (wq->flags & WQ_UNBOUND) { + struct device_attribute *attr; + + for (attr = wq_sysfs_unbound_attrs; attr->attr.name; attr++) { + ret = device_create_file(&wq_dev->dev, attr); + if (ret) { + device_unregister(&wq_dev->dev); + wq->wq_dev = NULL; + return ret; + } + } + } + + dev_set_uevent_suppress(&wq_dev->dev, false); + kobject_uevent(&wq_dev->dev.kobj, KOBJ_ADD); + return 0; +} + +/** + * workqueue_sysfs_unregister - undo workqueue_sysfs_register() + * @wq: the workqueue to unregister + * + * If @wq is registered to sysfs by workqueue_sysfs_register(), unregister. + */ +static void workqueue_sysfs_unregister(struct workqueue_struct *wq) +{ + struct wq_device *wq_dev = wq->wq_dev; + + if (!wq->wq_dev) + return; + + wq->wq_dev = NULL; + device_unregister(&wq_dev->dev); +} +#else /* CONFIG_SYSFS */ +static void workqueue_sysfs_unregister(struct workqueue_struct *wq) { } +#endif /* CONFIG_SYSFS */ + static void __init wq_numa_init(void) { cpumask_var_t *tbl; |