diff options
Diffstat (limited to 'kernel')
87 files changed, 3524 insertions, 2304 deletions
diff --git a/kernel/Kconfig.locks b/kernel/Kconfig.locks index 08561f1..ebdb004 100644 --- a/kernel/Kconfig.locks +++ b/kernel/Kconfig.locks @@ -235,9 +235,16 @@ config LOCK_SPIN_ON_OWNER def_bool y depends on MUTEX_SPIN_ON_OWNER || RWSEM_SPIN_ON_OWNER -config ARCH_USE_QUEUE_RWLOCK +config ARCH_USE_QUEUED_SPINLOCKS bool -config QUEUE_RWLOCK - def_bool y if ARCH_USE_QUEUE_RWLOCK +config QUEUED_SPINLOCKS + def_bool y if ARCH_USE_QUEUED_SPINLOCKS + depends on SMP + +config ARCH_USE_QUEUED_RWLOCKS + bool + +config QUEUED_RWLOCKS + def_bool y if ARCH_USE_QUEUED_RWLOCKS depends on SMP diff --git a/kernel/context_tracking.c b/kernel/context_tracking.c index 72d59a1..0a495ab 100644 --- a/kernel/context_tracking.c +++ b/kernel/context_tracking.c @@ -30,12 +30,23 @@ EXPORT_SYMBOL_GPL(context_tracking_enabled); DEFINE_PER_CPU(struct context_tracking, context_tracking); EXPORT_SYMBOL_GPL(context_tracking); -void context_tracking_cpu_set(int cpu) +static bool context_tracking_recursion_enter(void) { - if (!per_cpu(context_tracking.active, cpu)) { - per_cpu(context_tracking.active, cpu) = true; - static_key_slow_inc(&context_tracking_enabled); - } + int recursion; + + recursion = __this_cpu_inc_return(context_tracking.recursion); + if (recursion == 1) + return true; + + WARN_ONCE((recursion < 1), "Invalid context tracking recursion value %d\n", recursion); + __this_cpu_dec(context_tracking.recursion); + + return false; +} + +static void context_tracking_recursion_exit(void) +{ + __this_cpu_dec(context_tracking.recursion); } /** @@ -75,6 +86,9 @@ void context_tracking_enter(enum ctx_state state) WARN_ON_ONCE(!current->mm); local_irq_save(flags); + if (!context_tracking_recursion_enter()) + goto out_irq_restore; + if ( __this_cpu_read(context_tracking.state) != state) { if (__this_cpu_read(context_tracking.active)) { /* @@ -105,6 +119,8 @@ void context_tracking_enter(enum ctx_state state) */ __this_cpu_write(context_tracking.state, state); } + context_tracking_recursion_exit(); +out_irq_restore: local_irq_restore(flags); } NOKPROBE_SYMBOL(context_tracking_enter); @@ -139,6 +155,9 @@ void context_tracking_exit(enum ctx_state state) return; local_irq_save(flags); + if (!context_tracking_recursion_enter()) + goto out_irq_restore; + if (__this_cpu_read(context_tracking.state) == state) { if (__this_cpu_read(context_tracking.active)) { /* @@ -153,6 +172,8 @@ void context_tracking_exit(enum ctx_state state) } __this_cpu_write(context_tracking.state, CONTEXT_KERNEL); } + context_tracking_recursion_exit(); +out_irq_restore: local_irq_restore(flags); } NOKPROBE_SYMBOL(context_tracking_exit); @@ -164,24 +185,26 @@ void context_tracking_user_exit(void) } NOKPROBE_SYMBOL(context_tracking_user_exit); -/** - * __context_tracking_task_switch - context switch the syscall callbacks - * @prev: the task that is being switched out - * @next: the task that is being switched in - * - * The context tracking uses the syscall slow path to implement its user-kernel - * boundaries probes on syscalls. This way it doesn't impact the syscall fast - * path on CPUs that don't do context tracking. - * - * But we need to clear the flag on the previous task because it may later - * migrate to some CPU that doesn't do the context tracking. As such the TIF - * flag may not be desired there. - */ -void __context_tracking_task_switch(struct task_struct *prev, - struct task_struct *next) +void __init context_tracking_cpu_set(int cpu) { - clear_tsk_thread_flag(prev, TIF_NOHZ); - set_tsk_thread_flag(next, TIF_NOHZ); + static __initdata bool initialized = false; + + if (!per_cpu(context_tracking.active, cpu)) { + per_cpu(context_tracking.active, cpu) = true; + static_key_slow_inc(&context_tracking_enabled); + } + + if (initialized) + return; + + /* + * Set TIF_NOHZ to init/0 and let it propagate to all tasks through fork + * This assumes that init is the only task at this early boot stage. + */ + set_tsk_thread_flag(&init_task, TIF_NOHZ); + WARN_ON_ONCE(!tasklist_empty()); + + initialized = true; } #ifdef CONFIG_CONTEXT_TRACKING_FORCE diff --git a/kernel/cpu.c b/kernel/cpu.c index 94bbe46..9c9c9fa 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -398,7 +398,6 @@ static int __ref _cpu_down(unsigned int cpu, int tasks_frozen) err = __stop_machine(take_cpu_down, &tcd_param, cpumask_of(cpu)); if (err) { /* CPU didn't die: tell everyone. Can't complain. */ - smpboot_unpark_threads(cpu); cpu_notify_nofail(CPU_DOWN_FAILED | mod, hcpu); goto out_release; } @@ -463,6 +462,7 @@ static int smpboot_thread_call(struct notifier_block *nfb, switch (action & ~CPU_TASKS_FROZEN) { + case CPU_DOWN_FAILED: case CPU_ONLINE: smpboot_unpark_threads(cpu); break; @@ -479,7 +479,7 @@ static struct notifier_block smpboot_thread_notifier = { .priority = CPU_PRI_SMPBOOT, }; -void __cpuinit smpboot_thread_init(void) +void smpboot_thread_init(void) { register_cpu_notifier(&smpboot_thread_notifier); } diff --git a/kernel/events/core.c b/kernel/events/core.c index eddf1ed..8e13f3e 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -51,9 +51,11 @@ static struct workqueue_struct *perf_wq; +typedef int (*remote_function_f)(void *); + struct remote_function_call { struct task_struct *p; - int (*func)(void *info); + remote_function_f func; void *info; int ret; }; @@ -86,7 +88,7 @@ static void remote_function(void *data) * -EAGAIN - when the process moved away */ static int -task_function_call(struct task_struct *p, int (*func) (void *info), void *info) +task_function_call(struct task_struct *p, remote_function_f func, void *info) { struct remote_function_call data = { .p = p, @@ -110,7 +112,7 @@ task_function_call(struct task_struct *p, int (*func) (void *info), void *info) * * returns: @func return value or -ENXIO when the cpu is offline */ -static int cpu_function_call(int cpu, int (*func) (void *info), void *info) +static int cpu_function_call(int cpu, remote_function_f func, void *info) { struct remote_function_call data = { .p = NULL, @@ -747,62 +749,31 @@ perf_cgroup_mark_enabled(struct perf_event *event, /* * function must be called with interrupts disbled */ -static enum hrtimer_restart perf_cpu_hrtimer_handler(struct hrtimer *hr) +static enum hrtimer_restart perf_mux_hrtimer_handler(struct hrtimer *hr) { struct perf_cpu_context *cpuctx; - enum hrtimer_restart ret = HRTIMER_NORESTART; int rotations = 0; WARN_ON(!irqs_disabled()); cpuctx = container_of(hr, struct perf_cpu_context, hrtimer); - rotations = perf_rotate_context(cpuctx); - /* - * arm timer if needed - */ - if (rotations) { + raw_spin_lock(&cpuctx->hrtimer_lock); + if (rotations) hrtimer_forward_now(hr, cpuctx->hrtimer_interval); - ret = HRTIMER_RESTART; - } - - return ret; -} - -/* CPU is going down */ -void perf_cpu_hrtimer_cancel(int cpu) -{ - struct perf_cpu_context *cpuctx; - struct pmu *pmu; - unsigned long flags; - - if (WARN_ON(cpu != smp_processor_id())) - return; - - local_irq_save(flags); - - rcu_read_lock(); - - list_for_each_entry_rcu(pmu, &pmus, entry) { - cpuctx = this_cpu_ptr(pmu->pmu_cpu_context); - - if (pmu->task_ctx_nr == perf_sw_context) - continue; - - hrtimer_cancel(&cpuctx->hrtimer); - } - - rcu_read_unlock(); + else + cpuctx->hrtimer_active = 0; + raw_spin_unlock(&cpuctx->hrtimer_lock); - local_irq_restore(flags); + return rotations ? HRTIMER_RESTART : HRTIMER_NORESTART; } -static void __perf_cpu_hrtimer_init(struct perf_cpu_context *cpuctx, int cpu) +static void __perf_mux_hrtimer_init(struct perf_cpu_context *cpuctx, int cpu) { - struct hrtimer *hr = &cpuctx->hrtimer; + struct hrtimer *timer = &cpuctx->hrtimer; struct pmu *pmu = cpuctx->ctx.pmu; - int timer; + u64 interval; /* no multiplexing needed for SW PMU */ if (pmu->task_ctx_nr == perf_sw_context) @@ -812,31 +783,36 @@ static void __perf_cpu_hrtimer_init(struct perf_cpu_context *cpuctx, int cpu) * check default is sane, if not set then force to * default interval (1/tick) */ - timer = pmu->hrtimer_interval_ms; - if (timer < 1) - timer = pmu->hrtimer_interval_ms = PERF_CPU_HRTIMER; + interval = pmu->hrtimer_interval_ms; + if (interval < 1) + interval = pmu->hrtimer_interval_ms = PERF_CPU_HRTIMER; - cpuctx->hrtimer_interval = ns_to_ktime(NSEC_PER_MSEC * timer); + cpuctx->hrtimer_interval = ns_to_ktime(NSEC_PER_MSEC * interval); - hrtimer_init(hr, CLOCK_MONOTONIC, HRTIMER_MODE_REL_PINNED); - hr->function = perf_cpu_hrtimer_handler; + raw_spin_lock_init(&cpuctx->hrtimer_lock); + hrtimer_init(timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_PINNED); + timer->function = perf_mux_hrtimer_handler; } -static void perf_cpu_hrtimer_restart(struct perf_cpu_context *cpuctx) +static int perf_mux_hrtimer_restart(struct perf_cpu_context *cpuctx) { - struct hrtimer *hr = &cpuctx->hrtimer; + struct hrtimer *timer = &cpuctx->hrtimer; struct pmu *pmu = cpuctx->ctx.pmu; + unsigned long flags; /* not for SW PMU */ if (pmu->task_ctx_nr == perf_sw_context) - return; + return 0; - if (hrtimer_active(hr)) - return; + raw_spin_lock_irqsave(&cpuctx->hrtimer_lock, flags); + if (!cpuctx->hrtimer_active) { + cpuctx->hrtimer_active = 1; + hrtimer_forward_now(timer, cpuctx->hrtimer_interval); + hrtimer_start_expires(timer, HRTIMER_MODE_ABS_PINNED); + } + raw_spin_unlock_irqrestore(&cpuctx->hrtimer_lock, flags); - if (!hrtimer_callback_running(hr)) - __hrtimer_start_range_ns(hr, cpuctx->hrtimer_interval, - 0, HRTIMER_MODE_REL_PINNED, 0); + return 0; } void perf_pmu_disable(struct pmu *pmu) @@ -1935,7 +1911,7 @@ group_sched_in(struct perf_event *group_event, if (event_sched_in(group_event, cpuctx, ctx)) { pmu->cancel_txn(pmu); - perf_cpu_hrtimer_restart(cpuctx); + perf_mux_hrtimer_restart(cpuctx); return -EAGAIN; } @@ -1982,7 +1958,7 @@ group_error: pmu->cancel_txn(pmu); - perf_cpu_hrtimer_restart(cpuctx); + perf_mux_hrtimer_restart(cpuctx); return -EAGAIN; } @@ -2255,7 +2231,7 @@ static int __perf_event_enable(void *info) */ if (leader != event) { group_sched_out(leader, cpuctx, ctx); - perf_cpu_hrtimer_restart(cpuctx); + perf_mux_hrtimer_restart(cpuctx); } if (leader->attr.pinned) { update_group_times(leader); @@ -4331,20 +4307,20 @@ static void ring_buffer_attach(struct perf_event *event, WARN_ON_ONCE(event->rcu_pending); old_rb = event->rb; - event->rcu_batches = get_state_synchronize_rcu(); - event->rcu_pending = 1; - spin_lock_irqsave(&old_rb->event_lock, flags); list_del_rcu(&event->rb_entry); spin_unlock_irqrestore(&old_rb->event_lock, flags); - } - if (event->rcu_pending && rb) { - cond_synchronize_rcu(event->rcu_batches); - event->rcu_pending = 0; + event->rcu_batches = get_state_synchronize_rcu(); + event->rcu_pending = 1; } if (rb) { + if (event->rcu_pending) { + cond_synchronize_rcu(event->rcu_batches); + event->rcu_pending = 0; + } + spin_lock_irqsave(&rb->event_lock, flags); list_add_rcu(&event->rb_entry, &rb->event_list); spin_unlock_irqrestore(&rb->event_lock, flags); @@ -5381,9 +5357,9 @@ void perf_prepare_sample(struct perf_event_header *header, } } -static void perf_event_output(struct perf_event *event, - struct perf_sample_data *data, - struct pt_regs *regs) +void perf_event_output(struct perf_event *event, + struct perf_sample_data *data, + struct pt_regs *regs) { struct perf_output_handle handle; struct perf_event_header header; @@ -5975,6 +5951,39 @@ void perf_event_aux_event(struct perf_event *event, unsigned long head, } /* + * Lost/dropped samples logging + */ +void perf_log_lost_samples(struct perf_event *event, u64 lost) +{ + struct perf_output_handle handle; + struct perf_sample_data sample; + int ret; + + struct { + struct perf_event_header header; + u64 lost; + } lost_samples_event = { + .header = { + .type = PERF_RECORD_LOST_SAMPLES, + .misc = 0, + .size = sizeof(lost_samples_event), + }, + .lost = lost, + }; + + perf_event_header__init_id(&lost_samples_event.header, &sample, event); + + ret = perf_output_begin(&handle, event, + lost_samples_event.header.size); + if (ret) + return; + + perf_output_put(&handle, lost_samples_event); + perf_event__output_id_sample(event, &handle, &sample); + perf_output_end(&handle); +} + +/* * IRQ throttle logging */ @@ -6864,9 +6873,8 @@ static void perf_swevent_start_hrtimer(struct perf_event *event) } else { period = max_t(u64, 10000, hwc->sample_period); } - __hrtimer_start_range_ns(&hwc->hrtimer, - ns_to_ktime(period), 0, - HRTIMER_MODE_REL_PINNED, 0); + hrtimer_start(&hwc->hrtimer, ns_to_ktime(period), + HRTIMER_MODE_REL_PINNED); } static void perf_swevent_cancel_hrtimer(struct perf_event *event) @@ -7167,6 +7175,8 @@ perf_event_mux_interval_ms_show(struct device *dev, return snprintf(page, PAGE_SIZE-1, "%d\n", pmu->hrtimer_interval_ms); } +static DEFINE_MUTEX(mux_interval_mutex); + static ssize_t perf_event_mux_interval_ms_store(struct device *dev, struct device_attribute *attr, @@ -7186,17 +7196,21 @@ perf_event_mux_interval_ms_store(struct device *dev, if (timer == pmu->hrtimer_interval_ms) return count; + mutex_lock(&mux_interval_mutex); pmu->hrtimer_interval_ms = timer; /* update all cpuctx for this PMU */ - for_each_possible_cpu(cpu) { + get_online_cpus(); + for_each_online_cpu(cpu) { struct perf_cpu_context *cpuctx; cpuctx = per_cpu_ptr(pmu->pmu_cpu_context, cpu); cpuctx->hrtimer_interval = ns_to_ktime(NSEC_PER_MSEC * timer); - if (hrtimer_active(&cpuctx->hrtimer)) - hrtimer_forward_now(&cpuctx->hrtimer, cpuctx->hrtimer_interval); + cpu_function_call(cpu, + (remote_function_f)perf_mux_hrtimer_restart, cpuctx); } + put_online_cpus(); + mutex_unlock(&mux_interval_mutex); return count; } @@ -7301,7 +7315,7 @@ skip_type: lockdep_set_class(&cpuctx->ctx.lock, &cpuctx_lock); cpuctx->ctx.pmu = pmu; - __perf_cpu_hrtimer_init(cpuctx, cpu); + __perf_mux_hrtimer_init(cpuctx, cpu); cpuctx->unique_pmu = pmu; } diff --git a/kernel/events/internal.h b/kernel/events/internal.h index 9f6ce9b..2deb24c 100644 --- a/kernel/events/internal.h +++ b/kernel/events/internal.h @@ -72,15 +72,6 @@ static inline bool rb_has_aux(struct ring_buffer *rb) void perf_event_aux_event(struct perf_event *event, unsigned long head, unsigned long size, u64 flags); -extern void -perf_event_header__init_id(struct perf_event_header *header, - struct perf_sample_data *data, - struct perf_event *event); -extern void -perf_event__output_id_sample(struct perf_event *event, - struct perf_output_handle *handle, - struct perf_sample_data *sample); - extern struct page * perf_mmap_to_page(struct ring_buffer *rb, unsigned long pgoff); diff --git a/kernel/events/ring_buffer.c b/kernel/events/ring_buffer.c index 725c416..9647282 100644 --- a/kernel/events/ring_buffer.c +++ b/kernel/events/ring_buffer.c @@ -141,7 +141,7 @@ int perf_output_begin(struct perf_output_handle *handle, perf_output_get_handle(handle); do { - tail = ACCESS_ONCE(rb->user_page->data_tail); + tail = READ_ONCE_CTRL(rb->user_page->data_tail); offset = head = local_read(&rb->head); if (!rb->overwrite && unlikely(CIRC_SPACE(head, tail, perf_data_size(rb)) < size)) diff --git a/kernel/fork.c b/kernel/fork.c index 03c1eaa..0bb88b5 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1091,10 +1091,7 @@ static void posix_cpu_timers_init_group(struct signal_struct *sig) { unsigned long cpu_limit; - /* Thread group counters. */ - thread_group_cputime_init(sig); - - cpu_limit = ACCESS_ONCE(sig->rlim[RLIMIT_CPU].rlim_cur); + cpu_limit = READ_ONCE(sig->rlim[RLIMIT_CPU].rlim_cur); if (cpu_limit != RLIM_INFINITY) { sig->cputime_expires.prof_exp = secs_to_cputime(cpu_limit); sig->cputimer.running = 1; @@ -1396,6 +1393,9 @@ static struct task_struct *copy_process(unsigned long clone_flags, p->hardirq_context = 0; p->softirq_context = 0; #endif + + p->pagefault_disabled = 0; + #ifdef CONFIG_LOCKDEP p->lockdep_depth = 0; /* no locks held yet */ p->curr_chain_key = 0; diff --git a/kernel/futex.c b/kernel/futex.c index 2579e40..ea6ca0b 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -1090,9 +1090,11 @@ static void __unqueue_futex(struct futex_q *q) /* * The hash bucket lock must be held when this is called. - * Afterwards, the futex_q must not be accessed. + * Afterwards, the futex_q must not be accessed. Callers + * must ensure to later call wake_up_q() for the actual + * wakeups to occur. */ -static void wake_futex(struct futex_q *q) +static void mark_wake_futex(struct wake_q_head *wake_q, struct futex_q *q) { struct task_struct *p = q->task; @@ -1100,14 +1102,10 @@ static void wake_futex(struct futex_q *q) return; /* - * We set q->lock_ptr = NULL _before_ we wake up the task. If - * a non-futex wake up happens on another CPU then the task - * might exit and p would dereference a non-existing task - * struct. Prevent this by holding a reference on p across the - * wake up. + * Queue the task for later wakeup for after we've released + * the hb->lock. wake_q_add() grabs reference to p. */ - get_task_struct(p); - + wake_q_add(wake_q, p); __unqueue_futex(q); /* * The waiting task can free the futex_q as soon as @@ -1117,9 +1115,6 @@ static void wake_futex(struct futex_q *q) */ smp_wmb(); q->lock_ptr = NULL; - - wake_up_state(p, TASK_NORMAL); - put_task_struct(p); } static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_q *this) @@ -1217,6 +1212,7 @@ futex_wake(u32 __user *uaddr, unsigned int flags, int nr_wake, u32 bitset) struct futex_q *this, *next; union futex_key key = FUTEX_KEY_INIT; int ret; + WAKE_Q(wake_q); if (!bitset) return -EINVAL; @@ -1244,13 +1240,14 @@ futex_wake(u32 __user *uaddr, unsigned int flags, int nr_wake, u32 bitset) if (!(this->bitset & bitset)) continue; - wake_futex(this); + mark_wake_futex(&wake_q, this); if (++ret >= nr_wake) break; } } spin_unlock(&hb->lock); + wake_up_q(&wake_q); out_put_key: put_futex_key(&key); out: @@ -1269,6 +1266,7 @@ futex_wake_op(u32 __user *uaddr1, unsigned int flags, u32 __user *uaddr2, struct futex_hash_bucket *hb1, *hb2; struct futex_q *this, *next; int ret, op_ret; + WAKE_Q(wake_q); retry: ret = get_futex_key(uaddr1, flags & FLAGS_SHARED, &key1, VERIFY_READ); @@ -1320,7 +1318,7 @@ retry_private: ret = -EINVAL; goto out_unlock; } - wake_futex(this); + mark_wake_futex(&wake_q, this); if (++ret >= nr_wake) break; } @@ -1334,7 +1332,7 @@ retry_private: ret = -EINVAL; goto out_unlock; } - wake_futex(this); + mark_wake_futex(&wake_q, this); if (++op_ret >= nr_wake2) break; } @@ -1344,6 +1342,7 @@ retry_private: out_unlock: double_unlock_hb(hb1, hb2); + wake_up_q(&wake_q); out_put_keys: put_futex_key(&key2); out_put_key1: @@ -1503,6 +1502,7 @@ static int futex_requeue(u32 __user *uaddr1, unsigned int flags, struct futex_pi_state *pi_state = NULL; struct futex_hash_bucket *hb1, *hb2; struct futex_q *this, *next; + WAKE_Q(wake_q); if (requeue_pi) { /* @@ -1679,7 +1679,7 @@ retry_private: * woken by futex_unlock_pi(). */ if (++task_count <= nr_wake && !requeue_pi) { - wake_futex(this); + mark_wake_futex(&wake_q, this); continue; } @@ -1719,6 +1719,7 @@ retry_private: out_unlock: free_pi_state(pi_state); double_unlock_hb(hb1, hb2); + wake_up_q(&wake_q); hb_waiters_dec(hb2); /* @@ -2055,7 +2056,7 @@ static void futex_wait_queue_me(struct futex_hash_bucket *hb, struct futex_q *q, { /* * The task state is guaranteed to be set before another task can - * wake it. set_current_state() is implemented using set_mb() and + * wake it. set_current_state() is implemented using smp_store_mb() and * queue_me() calls spin_unlock() upon completion, both serializing * access to the hash list and forcing another memory barrier. */ @@ -2063,11 +2064,8 @@ static void futex_wait_queue_me(struct futex_hash_bucket *hb, struct futex_q *q, queue_me(q, hb); /* Arm the timer */ - if (timeout) { + if (timeout) hrtimer_start_expires(&timeout->timer, HRTIMER_MODE_ABS); - if (!hrtimer_active(&timeout->timer)) - timeout->task = NULL; - } /* * If we have been removed from the hash list, then another task diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c index eb9a4ea..27f4332 100644 --- a/kernel/irq/chip.c +++ b/kernel/irq/chip.c @@ -719,15 +719,9 @@ void handle_percpu_devid_irq(unsigned int irq, struct irq_desc *desc) } void -__irq_set_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained, - const char *name) +__irq_do_set_handler(struct irq_desc *desc, irq_flow_handler_t handle, + int is_chained, const char *name) { - unsigned long flags; - struct irq_desc *desc = irq_get_desc_buslock(irq, &flags, 0); - - if (!desc) - return; - if (!handle) { handle = handle_bad_irq; } else { @@ -749,13 +743,13 @@ __irq_set_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained, * right away. */ if (WARN_ON(is_chained)) - goto out; + return; /* Try the parent */ irq_data = irq_data->parent_data; } #endif if (WARN_ON(!irq_data || irq_data->chip == &no_irq_chip)) - goto out; + return; } /* Uninstall? */ @@ -774,12 +768,41 @@ __irq_set_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained, irq_settings_set_nothread(desc); irq_startup(desc, true); } -out: +} + +void +__irq_set_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained, + const char *name) +{ + unsigned long flags; + struct irq_desc *desc = irq_get_desc_buslock(irq, &flags, 0); + + if (!desc) + return; + + __irq_do_set_handler(desc, handle, is_chained, name); irq_put_desc_busunlock(desc, flags); } EXPORT_SYMBOL_GPL(__irq_set_handler); void +irq_set_chained_handler_and_data(unsigned int irq, irq_flow_handler_t handle, + void *data) +{ + unsigned long flags; + struct irq_desc *desc = irq_get_desc_buslock(irq, &flags, 0); + + if (!desc) + return; + + __irq_do_set_handler(desc, handle, 1, NULL); + desc->irq_data.handler_data = data; + + irq_put_desc_busunlock(desc, flags); +} +EXPORT_SYMBOL_GPL(irq_set_chained_handler_and_data); + +void irq_set_chip_and_handler_name(unsigned int irq, struct irq_chip *chip, irq_flow_handler_t handle, const char *name) { @@ -876,6 +899,34 @@ void irq_cpu_offline(void) #ifdef CONFIG_IRQ_DOMAIN_HIERARCHY /** + * irq_chip_enable_parent - Enable the parent interrupt (defaults to unmask if + * NULL) + * @data: Pointer to interrupt specific data + */ +void irq_chip_enable_parent(struct irq_data *data) +{ + data = data->parent_data; + if (data->chip->irq_enable) + data->chip->irq_enable(data); + else + data->chip->irq_unmask(data); +} + +/** + * irq_chip_disable_parent - Disable the parent interrupt (defaults to mask if + * NULL) + * @data: Pointer to interrupt specific data + */ +void irq_chip_disable_parent(struct irq_data *data) +{ + data = data->parent_data; + if (data->chip->irq_disable) + data->chip->irq_disable(data); + else + data->chip->irq_mask(data); +} + +/** * irq_chip_ack_parent - Acknowledge the parent interrupt * @data: Pointer to interrupt specific data */ @@ -950,6 +1001,20 @@ int irq_chip_retrigger_hierarchy(struct irq_data *data) } /** + * irq_chip_set_vcpu_affinity_parent - Set vcpu affinity on the parent interrupt + * @data: Pointer to interrupt specific data + * @dest: The vcpu affinity information + */ +int irq_chip_set_vcpu_affinity_parent(struct irq_data *data, void *vcpu_info) +{ + data = data->parent_data; + if (data->chip->irq_set_vcpu_affinity) + return data->chip->irq_set_vcpu_affinity(data, vcpu_info); + + 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 diff --git a/kernel/irq/devres.c b/kernel/irq/devres.c index d5d0f73..74d90a7 100644 --- a/kernel/irq/devres.c +++ b/kernel/irq/devres.c @@ -104,7 +104,7 @@ int devm_request_any_context_irq(struct device *dev, unsigned int irq, return -ENOMEM; rc = request_any_context_irq(irq, handler, irqflags, devname, dev_id); - if (rc) { + if (rc < 0) { devres_free(dr); return rc; } @@ -113,7 +113,7 @@ int devm_request_any_context_irq(struct device *dev, unsigned int irq, dr->dev_id = dev_id; devres_add(dev, dr); - return 0; + return rc; } EXPORT_SYMBOL(devm_request_any_context_irq); diff --git a/kernel/irq/dummychip.c b/kernel/irq/dummychip.c index 2feb6fe..326a67f 100644 --- a/kernel/irq/dummychip.c +++ b/kernel/irq/dummychip.c @@ -42,6 +42,7 @@ struct irq_chip no_irq_chip = { .irq_enable = noop, .irq_disable = noop, .irq_ack = ack_bad, + .flags = IRQCHIP_SKIP_SET_WAKE, }; /* diff --git a/kernel/irq/generic-chip.c b/kernel/irq/generic-chip.c index 61024e8..15b370d 100644 --- a/kernel/irq/generic-chip.c +++ b/kernel/irq/generic-chip.c @@ -360,7 +360,7 @@ static struct lock_class_key irq_nested_lock_class; int irq_map_generic_chip(struct irq_domain *d, unsigned int virq, irq_hw_number_t hw_irq) { - struct irq_data *data = irq_get_irq_data(virq); + struct irq_data *data = irq_domain_get_irq_data(d, virq); struct irq_domain_chip_generic *dgc = d->gc; struct irq_chip_generic *gc; struct irq_chip_type *ct; @@ -405,8 +405,7 @@ int irq_map_generic_chip(struct irq_domain *d, unsigned int virq, else data->mask = 1 << idx; - irq_set_chip_and_handler(virq, chip, ct->handler); - irq_set_chip_data(virq, gc); + irq_domain_set_info(d, virq, hw_irq, chip, gc, ct->handler, NULL, NULL); irq_modify_status(virq, dgc->irq_flags_to_clear, dgc->irq_flags_to_set); return 0; } diff --git a/kernel/irq/internals.h b/kernel/irq/internals.h index df553b0..4834ee8 100644 --- a/kernel/irq/internals.h +++ b/kernel/irq/internals.h @@ -59,8 +59,6 @@ enum { #include "debug.h" #include "settings.h" -#define irq_data_to_desc(data) container_of(data, struct irq_desc, irq_data) - extern int __irq_set_trigger(struct irq_desc *desc, unsigned int irq, unsigned long flags); extern void __disable_irq(struct irq_desc *desc, unsigned int irq); @@ -170,27 +168,27 @@ irq_put_desc_unlock(struct irq_desc *desc, unsigned long flags) */ static inline void irqd_set_move_pending(struct irq_data *d) { - d->state_use_accessors |= IRQD_SETAFFINITY_PENDING; + __irqd_to_state(d) |= IRQD_SETAFFINITY_PENDING; } static inline void irqd_clr_move_pending(struct irq_data *d) { - d->state_use_accessors &= ~IRQD_SETAFFINITY_PENDING; + __irqd_to_state(d) &= ~IRQD_SETAFFINITY_PENDING; } static inline void irqd_clear(struct irq_data *d, unsigned int mask) { - d->state_use_accessors &= ~mask; + __irqd_to_state(d) &= ~mask; } static inline void irqd_set(struct irq_data *d, unsigned int mask) { - d->state_use_accessors |= mask; + __irqd_to_state(d) |= mask; } static inline bool irqd_has_set(struct irq_data *d, unsigned int mask) { - return d->state_use_accessors & mask; + return __irqd_to_state(d) & mask; } static inline void kstat_incr_irqs_this_cpu(unsigned int irq, struct irq_desc *desc) @@ -199,6 +197,11 @@ static inline void kstat_incr_irqs_this_cpu(unsigned int irq, struct irq_desc *d __this_cpu_inc(kstat.irqs_sum); } +static inline int irq_desc_get_node(struct irq_desc *desc) +{ + return irq_data_get_node(&desc->irq_data); +} + #ifdef CONFIG_PM_SLEEP bool irq_pm_check_wakeup(struct irq_desc *desc); void irq_pm_install_action(struct irq_desc *desc, struct irqaction *action); diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c index 99793b9..4afc457 100644 --- a/kernel/irq/irqdesc.c +++ b/kernel/irq/irqdesc.c @@ -59,16 +59,10 @@ static void desc_smp_init(struct irq_desc *desc, int node) #endif } -static inline int desc_node(struct irq_desc *desc) -{ - return desc->irq_data.node; -} - #else static inline int alloc_masks(struct irq_desc *desc, gfp_t gfp, int node) { return 0; } static inline void desc_smp_init(struct irq_desc *desc, int node) { } -static inline int desc_node(struct irq_desc *desc) { return 0; } #endif static void desc_set_defaults(unsigned int irq, struct irq_desc *desc, int node, @@ -76,6 +70,7 @@ static void desc_set_defaults(unsigned int irq, struct irq_desc *desc, int node, { int cpu; + desc->irq_data.common = &desc->irq_common_data; desc->irq_data.irq = irq; desc->irq_data.chip = &no_irq_chip; desc->irq_data.chip_data = NULL; @@ -299,7 +294,7 @@ static void free_desc(unsigned int irq) unsigned long flags; raw_spin_lock_irqsave(&desc->lock, flags); - desc_set_defaults(irq, desc, desc_node(desc), NULL); + desc_set_defaults(irq, desc, irq_desc_get_node(desc), NULL); raw_spin_unlock_irqrestore(&desc->lock, flags); } @@ -619,7 +614,7 @@ unsigned int kstat_irqs(unsigned int irq) { struct irq_desc *desc = irq_to_desc(irq); int cpu; - int sum = 0; + unsigned int sum = 0; if (!desc || !desc->kstat_irqs) return 0; @@ -639,7 +634,7 @@ unsigned int kstat_irqs(unsigned int irq) */ unsigned int kstat_irqs_usr(unsigned int irq) { - int sum; + unsigned int sum; irq_lock_sparse(); sum = kstat_irqs(irq); diff --git a/kernel/irq/irqdomain.c b/kernel/irq/irqdomain.c index 7fac311..8c3577f 100644 --- a/kernel/irq/irqdomain.c +++ b/kernel/irq/irqdomain.c @@ -830,10 +830,12 @@ static struct irq_data *irq_domain_insert_irq_data(struct irq_domain *domain, { struct irq_data *irq_data; - irq_data = kzalloc_node(sizeof(*irq_data), GFP_KERNEL, child->node); + irq_data = kzalloc_node(sizeof(*irq_data), GFP_KERNEL, + irq_data_get_node(child)); if (irq_data) { child->parent_data = irq_data; irq_data->irq = child->irq; + irq_data->common = child->common; irq_data->node = child->node; irq_data->domain = domain; } @@ -1232,6 +1234,27 @@ struct irq_data *irq_domain_get_irq_data(struct irq_domain *domain, return (irq_data && irq_data->domain == domain) ? irq_data : NULL; } +/** + * irq_domain_set_info - Set the complete data for a @virq in @domain + * @domain: Interrupt domain to match + * @virq: IRQ number + * @hwirq: The hardware interrupt number + * @chip: The associated interrupt chip + * @chip_data: The associated interrupt chip data + * @handler: The interrupt flow handler + * @handler_data: The interrupt flow handler data + * @handler_name: The interrupt handler name + */ +void irq_domain_set_info(struct irq_domain *domain, unsigned int virq, + irq_hw_number_t hwirq, struct irq_chip *chip, + void *chip_data, irq_flow_handler_t handler, + void *handler_data, const char *handler_name) +{ + irq_set_chip_and_handler_name(virq, chip, handler, handler_name); + irq_set_chip_data(virq, chip_data); + irq_set_handler_data(virq, handler_data); +} + static void irq_domain_check_hierarchy(struct irq_domain *domain) { } diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index e68932b..f974485 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -256,6 +256,37 @@ int irq_set_affinity_hint(unsigned int irq, const struct cpumask *m) } EXPORT_SYMBOL_GPL(irq_set_affinity_hint); +/** + * irq_set_vcpu_affinity - Set vcpu affinity for the interrupt + * @irq: interrupt number to set affinity + * @vcpu_info: vCPU specific data + * + * This function uses the vCPU specific data to set the vCPU + * affinity for an irq. The vCPU specific data is passed from + * outside, such as KVM. One example code path is as below: + * KVM -> IOMMU -> irq_set_vcpu_affinity(). + */ +int irq_set_vcpu_affinity(unsigned int irq, void *vcpu_info) +{ + unsigned long flags; + struct irq_desc *desc = irq_get_desc_lock(irq, &flags, 0); + struct irq_data *data; + struct irq_chip *chip; + int ret = -ENOSYS; + + if (!desc) + return -EINVAL; + + data = irq_desc_get_irq_data(desc); + chip = irq_data_get_irq_chip(data); + if (chip && chip->irq_set_vcpu_affinity) + ret = chip->irq_set_vcpu_affinity(data, vcpu_info); + irq_put_desc_unlock(desc, flags); + + return ret; +} +EXPORT_SYMBOL_GPL(irq_set_vcpu_affinity); + static void irq_affinity_notify(struct work_struct *work) { struct irq_affinity_notify *notify = @@ -332,7 +363,7 @@ static int setup_affinity(unsigned int irq, struct irq_desc *desc, struct cpumask *mask) { struct cpumask *set = irq_default_affinity; - int node = desc->irq_data.node; + int node = irq_desc_get_node(desc); /* Excludes PER_CPU and NO_BALANCE interrupts */ if (!irq_can_set_affinity(irq)) diff --git a/kernel/irq/migration.c b/kernel/irq/migration.c index ca3f4aa..37ddb7b 100644 --- a/kernel/irq/migration.c +++ b/kernel/irq/migration.c @@ -7,21 +7,21 @@ void irq_move_masked_irq(struct irq_data *idata) { struct irq_desc *desc = irq_data_to_desc(idata); - struct irq_chip *chip = idata->chip; + struct irq_chip *chip = desc->irq_data.chip; if (likely(!irqd_is_setaffinity_pending(&desc->irq_data))) return; + irqd_clr_move_pending(&desc->irq_data); + /* * Paranoia: cpu-local interrupts shouldn't be calling in here anyway. */ - if (!irqd_can_balance(&desc->irq_data)) { + if (irqd_is_per_cpu(&desc->irq_data)) { WARN_ON(1); return; } - irqd_clr_move_pending(&desc->irq_data); - if (unlikely(cpumask_empty(desc->pending_mask))) return; @@ -52,6 +52,13 @@ void irq_move_irq(struct irq_data *idata) { bool masked; + /* + * Get top level irq_data when CONFIG_IRQ_DOMAIN_HIERARCHY is enabled, + * and it should be optimized away when CONFIG_IRQ_DOMAIN_HIERARCHY is + * disabled. So we avoid an "#ifdef CONFIG_IRQ_DOMAIN_HIERARCHY" here. + */ + idata = irq_desc_get_irq_data(irq_data_to_desc(idata)); + if (likely(!irqd_is_setaffinity_pending(idata))) return; diff --git a/kernel/irq/msi.c b/kernel/irq/msi.c index 474de5c..7bf1f1b 100644 --- a/kernel/irq/msi.c +++ b/kernel/irq/msi.c @@ -124,7 +124,7 @@ static void msi_domain_free(struct irq_domain *domain, unsigned int virq, irq_domain_free_irqs_top(domain, virq, nr_irqs); } -static struct irq_domain_ops msi_domain_ops = { +static const struct irq_domain_ops msi_domain_ops = { .alloc = msi_domain_alloc, .free = msi_domain_free, .activate = msi_domain_activate, diff --git a/kernel/irq/pm.c b/kernel/irq/pm.c index 5204a6d..d22786a 100644 --- a/kernel/irq/pm.c +++ b/kernel/irq/pm.c @@ -123,6 +123,8 @@ void suspend_device_irqs(void) unsigned long flags; bool sync; + if (irq_settings_is_nested_thread(desc)) + continue; raw_spin_lock_irqsave(&desc->lock, flags); sync = suspend_device_irq(desc, irq); raw_spin_unlock_irqrestore(&desc->lock, flags); @@ -163,6 +165,8 @@ static void resume_irqs(bool want_early) if (!is_early && want_early) continue; + if (irq_settings_is_nested_thread(desc)) + continue; raw_spin_lock_irqsave(&desc->lock, flags); resume_irq(desc, irq); diff --git a/kernel/irq/proc.c b/kernel/irq/proc.c index df2f464..0e97c14 100644 --- a/kernel/irq/proc.c +++ b/kernel/irq/proc.c @@ -241,7 +241,7 @@ static int irq_node_proc_show(struct seq_file *m, void *v) { struct irq_desc *desc = irq_to_desc((long) m->private); - seq_printf(m, "%d\n", desc->irq_data.node); + seq_printf(m, "%d\n", irq_desc_get_node(desc)); return 0; } diff --git a/kernel/livepatch/core.c b/kernel/livepatch/core.c index 284e269..c40ebcc 100644 --- a/kernel/livepatch/core.c +++ b/kernel/livepatch/core.c @@ -128,7 +128,7 @@ static bool klp_is_patch_registered(struct klp_patch *patch) static bool klp_initialized(void) { - return klp_root_kobj; + return !!klp_root_kobj; } struct klp_find_arg { @@ -179,7 +179,9 @@ static int klp_find_object_symbol(const char *objname, const char *name, .count = 0 }; + mutex_lock(&module_mutex); kallsyms_on_each_symbol(klp_find_callback, &args); + mutex_unlock(&module_mutex); if (args.count == 0) pr_err("symbol '%s' not found in symbol table\n", name); @@ -219,13 +221,19 @@ static int klp_verify_vmlinux_symbol(const char *name, unsigned long addr) .name = name, .addr = addr, }; + int ret; - if (kallsyms_on_each_symbol(klp_verify_callback, &args)) - return 0; + mutex_lock(&module_mutex); + ret = kallsyms_on_each_symbol(klp_verify_callback, &args); + mutex_unlock(&module_mutex); - pr_err("symbol '%s' not found at specified address 0x%016lx, kernel mismatch?\n", - name, addr); - return -EINVAL; + if (!ret) { + pr_err("symbol '%s' not found at specified address 0x%016lx, kernel mismatch?\n", + name, addr); + return -EINVAL; + } + + return 0; } static int klp_find_verify_func_addr(struct klp_object *obj, @@ -234,8 +242,9 @@ static int klp_find_verify_func_addr(struct klp_object *obj, int ret; #if defined(CONFIG_RANDOMIZE_BASE) - /* KASLR is enabled, disregard old_addr from user */ - func->old_addr = 0; + /* If KASLR has been enabled, adjust old_addr accordingly */ + if (kaslr_enabled() && func->old_addr) + func->old_addr += kaslr_offset(); #endif if (!func->old_addr || klp_is_module(obj)) @@ -422,7 +431,7 @@ static void klp_disable_object(struct klp_object *obj) { struct klp_func *func; - for (func = obj->funcs; func->old_name; func++) + klp_for_each_func(obj, func) if (func->state == KLP_ENABLED) klp_disable_func(func); @@ -440,7 +449,7 @@ static int klp_enable_object(struct klp_object *obj) if (WARN_ON(!klp_is_object_loaded(obj))) return -EINVAL; - for (func = obj->funcs; func->old_name; func++) { + klp_for_each_func(obj, func) { ret = klp_enable_func(func); if (ret) { klp_disable_object(obj); @@ -463,7 +472,7 @@ static int __klp_disable_patch(struct klp_patch *patch) pr_notice("disabling patch '%s'\n", patch->mod->name); - for (obj = patch->objs; obj->funcs; obj++) { + klp_for_each_object(patch, obj) { if (obj->state == KLP_ENABLED) klp_disable_object(obj); } @@ -523,7 +532,7 @@ static int __klp_enable_patch(struct klp_patch *patch) pr_notice("enabling patch '%s'\n", patch->mod->name); - for (obj = patch->objs; obj->funcs; obj++) { + klp_for_each_object(patch, obj) { if (!klp_is_object_loaded(obj)) continue; @@ -651,6 +660,15 @@ static struct kobj_type klp_ktype_patch = { .default_attrs = klp_patch_attrs, }; +static void klp_kobj_release_object(struct kobject *kobj) +{ +} + +static struct kobj_type klp_ktype_object = { + .release = klp_kobj_release_object, + .sysfs_ops = &kobj_sysfs_ops, +}; + static void klp_kobj_release_func(struct kobject *kobj) { } @@ -680,7 +698,7 @@ static void klp_free_object_loaded(struct klp_object *obj) obj->mod = NULL; - for (func = obj->funcs; func->old_name; func++) + klp_for_each_func(obj, func) func->old_addr = 0; } @@ -695,7 +713,7 @@ static void klp_free_objects_limited(struct klp_patch *patch, for (obj = patch->objs; obj->funcs && obj != limit; obj++) { klp_free_funcs_limited(obj, NULL); - kobject_put(obj->kobj); + kobject_put(&obj->kobj); } } @@ -713,7 +731,7 @@ 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, "%s", func->old_name); + &obj->kobj, "%s", func->old_name); } /* parts of the initialization that is done only when the object is loaded */ @@ -729,7 +747,7 @@ static int klp_init_object_loaded(struct klp_patch *patch, return ret; } - for (func = obj->funcs; func->old_name; func++) { + klp_for_each_func(obj, func) { ret = klp_find_verify_func_addr(obj, func); if (ret) return ret; @@ -753,11 +771,12 @@ static int klp_init_object(struct klp_patch *patch, struct klp_object *obj) klp_find_object_module(obj); name = klp_is_module(obj) ? obj->name : "vmlinux"; - obj->kobj = kobject_create_and_add(name, &patch->kobj); - if (!obj->kobj) - return -ENOMEM; + ret = kobject_init_and_add(&obj->kobj, &klp_ktype_object, + &patch->kobj, "%s", name); + if (ret) + return ret; - for (func = obj->funcs; func->old_name; func++) { + klp_for_each_func(obj, func) { ret = klp_init_func(obj, func); if (ret) goto free; @@ -773,7 +792,7 @@ static int klp_init_object(struct klp_patch *patch, struct klp_object *obj) free: klp_free_funcs_limited(obj, func); - kobject_put(obj->kobj); + kobject_put(&obj->kobj); return ret; } @@ -794,7 +813,7 @@ static int klp_init_patch(struct klp_patch *patch) if (ret) goto unlock; - for (obj = patch->objs; obj->funcs; obj++) { + klp_for_each_object(patch, obj) { ret = klp_init_object(patch, obj); if (ret) goto free; @@ -883,7 +902,7 @@ int klp_register_patch(struct klp_patch *patch) } EXPORT_SYMBOL_GPL(klp_register_patch); -static void klp_module_notify_coming(struct klp_patch *patch, +static int klp_module_notify_coming(struct klp_patch *patch, struct klp_object *obj) { struct module *pmod = patch->mod; @@ -891,22 +910,23 @@ static void klp_module_notify_coming(struct klp_patch *patch, int ret; ret = klp_init_object_loaded(patch, obj); - if (ret) - goto err; + if (ret) { + pr_warn("failed to initialize patch '%s' for module '%s' (%d)\n", + pmod->name, mod->name, ret); + return ret; + } if (patch->state == KLP_DISABLED) - return; + return 0; pr_notice("applying patch '%s' to loading module '%s'\n", pmod->name, mod->name); ret = klp_enable_object(obj); - if (!ret) - return; - -err: - pr_warn("failed to apply patch '%s' to module '%s' (%d)\n", - pmod->name, mod->name, ret); + if (ret) + pr_warn("failed to apply patch '%s' to module '%s' (%d)\n", + pmod->name, mod->name, ret); + return ret; } static void klp_module_notify_going(struct klp_patch *patch, @@ -930,6 +950,7 @@ disabled: static int klp_module_notify(struct notifier_block *nb, unsigned long action, void *data) { + int ret; struct module *mod = data; struct klp_patch *patch; struct klp_object *obj; @@ -949,13 +970,18 @@ static int klp_module_notify(struct notifier_block *nb, unsigned long action, mod->klp_alive = false; list_for_each_entry(patch, &klp_patches, list) { - for (obj = patch->objs; obj->funcs; obj++) { + klp_for_each_object(patch, obj) { if (!klp_is_module(obj) || strcmp(obj->name, mod->name)) continue; if (action == MODULE_STATE_COMING) { obj->mod = mod; - klp_module_notify_coming(patch, obj); + ret = klp_module_notify_coming(patch, obj); + if (ret) { + obj->mod = NULL; + pr_warn("patch '%s' is in an inconsistent state!\n", + patch->mod->name); + } } else /* MODULE_STATE_GOING */ klp_module_notify_going(patch, obj); @@ -973,7 +999,7 @@ static struct notifier_block klp_module_nb = { .priority = INT_MIN+1, /* called late but before ftrace notifier */ }; -static int klp_init(void) +static int __init klp_init(void) { int ret; diff --git a/kernel/locking/Makefile b/kernel/locking/Makefile index de7a416..7dd5c99 100644 --- a/kernel/locking/Makefile +++ b/kernel/locking/Makefile @@ -17,6 +17,7 @@ obj-$(CONFIG_SMP) += spinlock.o obj-$(CONFIG_LOCK_SPIN_ON_OWNER) += osq_lock.o obj-$(CONFIG_SMP) += lglock.o obj-$(CONFIG_PROVE_LOCKING) += spinlock.o +obj-$(CONFIG_QUEUED_SPINLOCKS) += qspinlock.o obj-$(CONFIG_RT_MUTEXES) += rtmutex.o obj-$(CONFIG_DEBUG_RT_MUTEXES) += rtmutex-debug.o obj-$(CONFIG_RT_MUTEX_TESTER) += rtmutex-tester.o @@ -25,5 +26,5 @@ obj-$(CONFIG_DEBUG_SPINLOCK) += spinlock_debug.o obj-$(CONFIG_RWSEM_GENERIC_SPINLOCK) += rwsem-spinlock.o obj-$(CONFIG_RWSEM_XCHGADD_ALGORITHM) += rwsem-xadd.o obj-$(CONFIG_PERCPU_RWSEM) += percpu-rwsem.o -obj-$(CONFIG_QUEUE_RWLOCK) += qrwlock.o +obj-$(CONFIG_QUEUED_RWLOCKS) += qrwlock.o obj-$(CONFIG_LOCK_TORTURE_TEST) += locktorture.o diff --git a/kernel/locking/lglock.c b/kernel/locking/lglock.c index 86ae2ae..951cfcd 100644 --- a/kernel/locking/lglock.c +++ b/kernel/locking/lglock.c @@ -60,6 +60,28 @@ void lg_local_unlock_cpu(struct lglock *lg, int cpu) } EXPORT_SYMBOL(lg_local_unlock_cpu); +void lg_double_lock(struct lglock *lg, int cpu1, int cpu2) +{ + BUG_ON(cpu1 == cpu2); + + /* lock in cpu order, just like lg_global_lock */ + if (cpu2 < cpu1) + swap(cpu1, cpu2); + + preempt_disable(); + lock_acquire_shared(&lg->lock_dep_map, 0, 0, NULL, _RET_IP_); + arch_spin_lock(per_cpu_ptr(lg->lock, cpu1)); + arch_spin_lock(per_cpu_ptr(lg->lock, cpu2)); +} + +void lg_double_unlock(struct lglock *lg, int cpu1, int cpu2) +{ + lock_release(&lg->lock_dep_map, 1, _RET_IP_); + arch_spin_unlock(per_cpu_ptr(lg->lock, cpu1)); + arch_spin_unlock(per_cpu_ptr(lg->lock, cpu2)); + preempt_enable(); +} + void lg_global_lock(struct lglock *lg) { int i; diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c index aaeae88..4566141 100644 --- a/kernel/locking/lockdep.c +++ b/kernel/locking/lockdep.c @@ -4067,8 +4067,7 @@ void __init lockdep_info(void) #ifdef CONFIG_DEBUG_LOCKDEP if (lockdep_init_error) { - printk("WARNING: lockdep init error! lock-%s was acquired" - "before lockdep_init\n", lock_init_error); + printk("WARNING: lockdep init error: lock '%s' was acquired before lockdep_init().\n", lock_init_error); printk("Call stack leading to lockdep invocation was:\n"); print_stack_trace(&lockdep_init_trace, 0); } diff --git a/kernel/locking/locktorture.c b/kernel/locking/locktorture.c index ec8cce2..3224418 100644 --- a/kernel/locking/locktorture.c +++ b/kernel/locking/locktorture.c @@ -122,12 +122,12 @@ static int torture_lock_busted_write_lock(void) static void torture_lock_busted_write_delay(struct torture_random_state *trsp) { - const unsigned long longdelay_us = 100; + const unsigned long longdelay_ms = 100; /* We want a long delay occasionally to force massive contention. */ if (!(torture_random(trsp) % - (cxt.nrealwriters_stress * 2000 * longdelay_us))) - mdelay(longdelay_us); + (cxt.nrealwriters_stress * 2000 * longdelay_ms))) + mdelay(longdelay_ms); #ifdef CONFIG_PREEMPT if (!(torture_random(trsp) % (cxt.nrealwriters_stress * 20000))) preempt_schedule(); /* Allow test to be preempted. */ @@ -160,14 +160,14 @@ static int torture_spin_lock_write_lock(void) __acquires(torture_spinlock) static void torture_spin_lock_write_delay(struct torture_random_state *trsp) { const unsigned long shortdelay_us = 2; - const unsigned long longdelay_us = 100; + const unsigned long longdelay_ms = 100; /* We want a short delay mostly to emulate likely code, and * we want a long delay occasionally to force massive contention. */ if (!(torture_random(trsp) % - (cxt.nrealwriters_stress * 2000 * longdelay_us))) - mdelay(longdelay_us); + (cxt.nrealwriters_stress * 2000 * longdelay_ms))) + mdelay(longdelay_ms); if (!(torture_random(trsp) % (cxt.nrealwriters_stress * 2 * shortdelay_us))) udelay(shortdelay_us); @@ -309,7 +309,7 @@ static int torture_rwlock_read_lock_irq(void) __acquires(torture_rwlock) static void torture_rwlock_read_unlock_irq(void) __releases(torture_rwlock) { - write_unlock_irqrestore(&torture_rwlock, cxt.cur_ops->flags); + read_unlock_irqrestore(&torture_rwlock, cxt.cur_ops->flags); } static struct lock_torture_ops rw_lock_irq_ops = { diff --git a/kernel/locking/mcs_spinlock.h b/kernel/locking/mcs_spinlock.h index 75e114b..fd91aaa 100644 --- a/kernel/locking/mcs_spinlock.h +++ b/kernel/locking/mcs_spinlock.h @@ -17,6 +17,7 @@ struct mcs_spinlock { struct mcs_spinlock *next; int locked; /* 1 if lock acquired */ + int count; /* nesting count, see qspinlock.c */ }; #ifndef arch_mcs_spin_lock_contended diff --git a/kernel/locking/qrwlock.c b/kernel/locking/qrwlock.c index f956ede..6c5da483 100644 --- a/kernel/locking/qrwlock.c +++ b/kernel/locking/qrwlock.c @@ -1,5 +1,5 @@ /* - * Queue read/write lock + * Queued read/write locks * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -22,6 +22,26 @@ #include <linux/hardirq.h> #include <asm/qrwlock.h> +/* + * This internal data structure is used for optimizing access to some of + * the subfields within the atomic_t cnts. + */ +struct __qrwlock { + union { + atomic_t cnts; + struct { +#ifdef __LITTLE_ENDIAN + u8 wmode; /* Writer mode */ + u8 rcnts[3]; /* Reader counts */ +#else + u8 rcnts[3]; /* Reader counts */ + u8 wmode; /* Writer mode */ +#endif + }; + }; + arch_spinlock_t lock; +}; + /** * rspin_until_writer_unlock - inc reader count & spin until writer is gone * @lock : Pointer to queue rwlock structure @@ -107,10 +127,10 @@ void queue_write_lock_slowpath(struct qrwlock *lock) * or wait for a previous writer to go away. */ for (;;) { - cnts = atomic_read(&lock->cnts); - if (!(cnts & _QW_WMASK) && - (atomic_cmpxchg(&lock->cnts, cnts, - cnts | _QW_WAITING) == cnts)) + struct __qrwlock *l = (struct __qrwlock *)lock; + + if (!READ_ONCE(l->wmode) && + (cmpxchg(&l->wmode, 0, _QW_WAITING) == 0)) break; cpu_relax_lowlatency(); diff --git a/kernel/locking/qspinlock.c b/kernel/locking/qspinlock.c new file mode 100644 index 0000000..38c4920 --- /dev/null +++ b/kernel/locking/qspinlock.c @@ -0,0 +1,473 @@ +/* + * Queued spinlock + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * (C) Copyright 2013-2015 Hewlett-Packard Development Company, L.P. + * (C) Copyright 2013-2014 Red Hat, Inc. + * (C) Copyright 2015 Intel Corp. + * + * Authors: Waiman Long <waiman.long@hp.com> + * Peter Zijlstra <peterz@infradead.org> + */ + +#ifndef _GEN_PV_LOCK_SLOWPATH + +#include <linux/smp.h> +#include <linux/bug.h> +#include <linux/cpumask.h> +#include <linux/percpu.h> +#include <linux/hardirq.h> +#include <linux/mutex.h> +#include <asm/byteorder.h> +#include <asm/qspinlock.h> + +/* + * The basic principle of a queue-based spinlock can best be understood + * by studying a classic queue-based spinlock implementation called the + * MCS lock. The paper below provides a good description for this kind + * of lock. + * + * http://www.cise.ufl.edu/tr/DOC/REP-1992-71.pdf + * + * This queued spinlock implementation is based on the MCS lock, however to make + * it fit the 4 bytes we assume spinlock_t to be, and preserve its existing + * API, we must modify it somehow. + * + * In particular; where the traditional MCS lock consists of a tail pointer + * (8 bytes) and needs the next pointer (another 8 bytes) of its own node to + * unlock the next pending (next->locked), we compress both these: {tail, + * next->locked} into a single u32 value. + * + * Since a spinlock disables recursion of its own context and there is a limit + * to the contexts that can nest; namely: task, softirq, hardirq, nmi. As there + * are at most 4 nesting levels, it can be encoded by a 2-bit number. Now + * we can encode the tail by combining the 2-bit nesting level with the cpu + * number. With one byte for the lock value and 3 bytes for the tail, only a + * 32-bit word is now needed. Even though we only need 1 bit for the lock, + * we extend it to a full byte to achieve better performance for architectures + * that support atomic byte write. + * + * We also change the first spinner to spin on the lock bit instead of its + * node; whereby avoiding the need to carry a node from lock to unlock, and + * preserving existing lock API. This also makes the unlock code simpler and + * faster. + * + * N.B. The current implementation only supports architectures that allow + * atomic operations on smaller 8-bit and 16-bit data types. + * + */ + +#include "mcs_spinlock.h" + +#ifdef CONFIG_PARAVIRT_SPINLOCKS +#define MAX_NODES 8 +#else +#define MAX_NODES 4 +#endif + +/* + * Per-CPU queue node structures; we can never have more than 4 nested + * contexts: task, softirq, hardirq, nmi. + * + * Exactly fits one 64-byte cacheline on a 64-bit architecture. + * + * PV doubles the storage and uses the second cacheline for PV state. + */ +static DEFINE_PER_CPU_ALIGNED(struct mcs_spinlock, mcs_nodes[MAX_NODES]); + +/* + * We must be able to distinguish between no-tail and the tail at 0:0, + * therefore increment the cpu number by one. + */ + +static inline u32 encode_tail(int cpu, int idx) +{ + u32 tail; + +#ifdef CONFIG_DEBUG_SPINLOCK + BUG_ON(idx > 3); +#endif + tail = (cpu + 1) << _Q_TAIL_CPU_OFFSET; + tail |= idx << _Q_TAIL_IDX_OFFSET; /* assume < 4 */ + + return tail; +} + +static inline struct mcs_spinlock *decode_tail(u32 tail) +{ + int cpu = (tail >> _Q_TAIL_CPU_OFFSET) - 1; + int idx = (tail & _Q_TAIL_IDX_MASK) >> _Q_TAIL_IDX_OFFSET; + + return per_cpu_ptr(&mcs_nodes[idx], cpu); +} + +#define _Q_LOCKED_PENDING_MASK (_Q_LOCKED_MASK | _Q_PENDING_MASK) + +/* + * By using the whole 2nd least significant byte for the pending bit, we + * can allow better optimization of the lock acquisition for the pending + * bit holder. + * + * This internal structure is also used by the set_locked function which + * is not restricted to _Q_PENDING_BITS == 8. + */ +struct __qspinlock { + union { + atomic_t val; +#ifdef __LITTLE_ENDIAN + struct { + u8 locked; + u8 pending; + }; + struct { + u16 locked_pending; + u16 tail; + }; +#else + struct { + u16 tail; + u16 locked_pending; + }; + struct { + u8 reserved[2]; + u8 pending; + u8 locked; + }; +#endif + }; +}; + +#if _Q_PENDING_BITS == 8 +/** + * clear_pending_set_locked - take ownership and clear the pending bit. + * @lock: Pointer to queued spinlock structure + * + * *,1,0 -> *,0,1 + * + * Lock stealing is not allowed if this function is used. + */ +static __always_inline void clear_pending_set_locked(struct qspinlock *lock) +{ + struct __qspinlock *l = (void *)lock; + + WRITE_ONCE(l->locked_pending, _Q_LOCKED_VAL); +} + +/* + * xchg_tail - Put in the new queue tail code word & retrieve previous one + * @lock : Pointer to queued spinlock structure + * @tail : The new queue tail code word + * Return: The previous queue tail code word + * + * xchg(lock, tail) + * + * p,*,* -> n,*,* ; prev = xchg(lock, node) + */ +static __always_inline u32 xchg_tail(struct qspinlock *lock, u32 tail) +{ + struct __qspinlock *l = (void *)lock; + + return (u32)xchg(&l->tail, tail >> _Q_TAIL_OFFSET) << _Q_TAIL_OFFSET; +} + +#else /* _Q_PENDING_BITS == 8 */ + +/** + * clear_pending_set_locked - take ownership and clear the pending bit. + * @lock: Pointer to queued spinlock structure + * + * *,1,0 -> *,0,1 + */ +static __always_inline void clear_pending_set_locked(struct qspinlock *lock) +{ + atomic_add(-_Q_PENDING_VAL + _Q_LOCKED_VAL, &lock->val); +} + +/** + * xchg_tail - Put in the new queue tail code word & retrieve previous one + * @lock : Pointer to queued spinlock structure + * @tail : The new queue tail code word + * Return: The previous queue tail code word + * + * xchg(lock, tail) + * + * p,*,* -> n,*,* ; prev = xchg(lock, node) + */ +static __always_inline u32 xchg_tail(struct qspinlock *lock, u32 tail) +{ + u32 old, new, val = atomic_read(&lock->val); + + for (;;) { + new = (val & _Q_LOCKED_PENDING_MASK) | tail; + old = atomic_cmpxchg(&lock->val, val, new); + if (old == val) + break; + + val = old; + } + return old; +} +#endif /* _Q_PENDING_BITS == 8 */ + +/** + * set_locked - Set the lock bit and own the lock + * @lock: Pointer to queued spinlock structure + * + * *,*,0 -> *,0,1 + */ +static __always_inline void set_locked(struct qspinlock *lock) +{ + struct __qspinlock *l = (void *)lock; + + WRITE_ONCE(l->locked, _Q_LOCKED_VAL); +} + + +/* + * Generate the native code for queued_spin_unlock_slowpath(); provide NOPs for + * all the PV callbacks. + */ + +static __always_inline void __pv_init_node(struct mcs_spinlock *node) { } +static __always_inline void __pv_wait_node(struct mcs_spinlock *node) { } +static __always_inline void __pv_kick_node(struct mcs_spinlock *node) { } + +static __always_inline void __pv_wait_head(struct qspinlock *lock, + struct mcs_spinlock *node) { } + +#define pv_enabled() false + +#define pv_init_node __pv_init_node +#define pv_wait_node __pv_wait_node +#define pv_kick_node __pv_kick_node +#define pv_wait_head __pv_wait_head + +#ifdef CONFIG_PARAVIRT_SPINLOCKS +#define queued_spin_lock_slowpath native_queued_spin_lock_slowpath +#endif + +#endif /* _GEN_PV_LOCK_SLOWPATH */ + +/** + * queued_spin_lock_slowpath - acquire the queued spinlock + * @lock: Pointer to queued spinlock structure + * @val: Current value of the queued spinlock 32-bit word + * + * (queue tail, pending bit, lock value) + * + * fast : slow : unlock + * : : + * uncontended (0,0,0) -:--> (0,0,1) ------------------------------:--> (*,*,0) + * : | ^--------.------. / : + * : v \ \ | : + * pending : (0,1,1) +--> (0,1,0) \ | : + * : | ^--' | | : + * : v | | : + * uncontended : (n,x,y) +--> (n,0,0) --' | : + * queue : | ^--' | : + * : v | : + * contended : (*,x,y) +--> (*,0,0) ---> (*,0,1) -' : + * queue : ^--' : + */ +void queued_spin_lock_slowpath(struct qspinlock *lock, u32 val) +{ + struct mcs_spinlock *prev, *next, *node; + u32 new, old, tail; + int idx; + + BUILD_BUG_ON(CONFIG_NR_CPUS >= (1U << _Q_TAIL_CPU_BITS)); + + if (pv_enabled()) + goto queue; + + if (virt_queued_spin_lock(lock)) + return; + + /* + * wait for in-progress pending->locked hand-overs + * + * 0,1,0 -> 0,0,1 + */ + if (val == _Q_PENDING_VAL) { + while ((val = atomic_read(&lock->val)) == _Q_PENDING_VAL) + cpu_relax(); + } + + /* + * trylock || pending + * + * 0,0,0 -> 0,0,1 ; trylock + * 0,0,1 -> 0,1,1 ; pending + */ + for (;;) { + /* + * If we observe any contention; queue. + */ + if (val & ~_Q_LOCKED_MASK) + goto queue; + + new = _Q_LOCKED_VAL; + if (val == new) + new |= _Q_PENDING_VAL; + + old = atomic_cmpxchg(&lock->val, val, new); + if (old == val) + break; + + val = old; + } + + /* + * we won the trylock + */ + if (new == _Q_LOCKED_VAL) + return; + + /* + * we're pending, wait for the owner to go away. + * + * *,1,1 -> *,1,0 + * + * this wait loop must be a load-acquire such that we match the + * store-release that clears the locked bit and create lock + * sequentiality; this is because not all clear_pending_set_locked() + * implementations imply full barriers. + */ + while ((val = smp_load_acquire(&lock->val.counter)) & _Q_LOCKED_MASK) + cpu_relax(); + + /* + * take ownership and clear the pending bit. + * + * *,1,0 -> *,0,1 + */ + clear_pending_set_locked(lock); + return; + + /* + * End of pending bit optimistic spinning and beginning of MCS + * queuing. + */ +queue: + node = this_cpu_ptr(&mcs_nodes[0]); + idx = node->count++; + tail = encode_tail(smp_processor_id(), idx); + + node += idx; + node->locked = 0; + node->next = NULL; + pv_init_node(node); + + /* + * We touched a (possibly) cold cacheline in the per-cpu queue node; + * attempt the trylock once more in the hope someone let go while we + * weren't watching. + */ + if (queued_spin_trylock(lock)) + goto release; + + /* + * We have already touched the queueing cacheline; don't bother with + * pending stuff. + * + * p,*,* -> n,*,* + */ + old = xchg_tail(lock, tail); + + /* + * if there was a previous node; link it and wait until reaching the + * head of the waitqueue. + */ + if (old & _Q_TAIL_MASK) { + prev = decode_tail(old); + WRITE_ONCE(prev->next, node); + + pv_wait_node(node); + arch_mcs_spin_lock_contended(&node->locked); + } + + /* + * we're at the head of the waitqueue, wait for the owner & pending to + * go away. + * + * *,x,y -> *,0,0 + * + * this wait loop must use a load-acquire such that we match the + * store-release that clears the locked bit and create lock + * sequentiality; this is because the set_locked() function below + * does not imply a full barrier. + * + */ + pv_wait_head(lock, node); + while ((val = smp_load_acquire(&lock->val.counter)) & _Q_LOCKED_PENDING_MASK) + cpu_relax(); + + /* + * claim the lock: + * + * n,0,0 -> 0,0,1 : lock, uncontended + * *,0,0 -> *,0,1 : lock, contended + * + * If the queue head is the only one in the queue (lock value == tail), + * clear the tail code and grab the lock. Otherwise, we only need + * to grab the lock. + */ + for (;;) { + if (val != tail) { + set_locked(lock); + break; + } + old = atomic_cmpxchg(&lock->val, val, _Q_LOCKED_VAL); + if (old == val) + goto release; /* No contention */ + + val = old; + } + + /* + * contended path; wait for next, release. + */ + while (!(next = READ_ONCE(node->next))) + cpu_relax(); + + arch_mcs_spin_unlock_contended(&next->locked); + pv_kick_node(next); + +release: + /* + * release the node + */ + this_cpu_dec(mcs_nodes[0].count); +} +EXPORT_SYMBOL(queued_spin_lock_slowpath); + +/* + * Generate the paravirt code for queued_spin_unlock_slowpath(). + */ +#if !defined(_GEN_PV_LOCK_SLOWPATH) && defined(CONFIG_PARAVIRT_SPINLOCKS) +#define _GEN_PV_LOCK_SLOWPATH + +#undef pv_enabled +#define pv_enabled() true + +#undef pv_init_node +#undef pv_wait_node +#undef pv_kick_node +#undef pv_wait_head + +#undef queued_spin_lock_slowpath +#define queued_spin_lock_slowpath __pv_queued_spin_lock_slowpath + +#include "qspinlock_paravirt.h" +#include "qspinlock.c" + +#endif diff --git a/kernel/locking/qspinlock_paravirt.h b/kernel/locking/qspinlock_paravirt.h new file mode 100644 index 0000000..04ab181 --- /dev/null +++ b/kernel/locking/qspinlock_paravirt.h @@ -0,0 +1,325 @@ +#ifndef _GEN_PV_LOCK_SLOWPATH +#error "do not include this file" +#endif + +#include <linux/hash.h> +#include <linux/bootmem.h> + +/* + * Implement paravirt qspinlocks; the general idea is to halt the vcpus instead + * of spinning them. + * + * This relies on the architecture to provide two paravirt hypercalls: + * + * pv_wait(u8 *ptr, u8 val) -- suspends the vcpu if *ptr == val + * pv_kick(cpu) -- wakes a suspended vcpu + * + * Using these we implement __pv_queued_spin_lock_slowpath() and + * __pv_queued_spin_unlock() to replace native_queued_spin_lock_slowpath() and + * native_queued_spin_unlock(). + */ + +#define _Q_SLOW_VAL (3U << _Q_LOCKED_OFFSET) + +enum vcpu_state { + vcpu_running = 0, + vcpu_halted, +}; + +struct pv_node { + struct mcs_spinlock mcs; + struct mcs_spinlock __res[3]; + + int cpu; + u8 state; +}; + +/* + * Lock and MCS node addresses hash table for fast lookup + * + * Hashing is done on a per-cacheline basis to minimize the need to access + * more than one cacheline. + * + * Dynamically allocate a hash table big enough to hold at least 4X the + * number of possible cpus in the system. Allocation is done on page + * granularity. So the minimum number of hash buckets should be at least + * 256 (64-bit) or 512 (32-bit) to fully utilize a 4k page. + * + * Since we should not be holding locks from NMI context (very rare indeed) the + * max load factor is 0.75, which is around the point where open addressing + * breaks down. + * + */ +struct pv_hash_entry { + struct qspinlock *lock; + struct pv_node *node; +}; + +#define PV_HE_PER_LINE (SMP_CACHE_BYTES / sizeof(struct pv_hash_entry)) +#define PV_HE_MIN (PAGE_SIZE / sizeof(struct pv_hash_entry)) + +static struct pv_hash_entry *pv_lock_hash; +static unsigned int pv_lock_hash_bits __read_mostly; + +/* + * Allocate memory for the PV qspinlock hash buckets + * + * This function should be called from the paravirt spinlock initialization + * routine. + */ +void __init __pv_init_lock_hash(void) +{ + int pv_hash_size = ALIGN(4 * num_possible_cpus(), PV_HE_PER_LINE); + + if (pv_hash_size < PV_HE_MIN) + pv_hash_size = PV_HE_MIN; + + /* + * Allocate space from bootmem which should be page-size aligned + * and hence cacheline aligned. + */ + pv_lock_hash = alloc_large_system_hash("PV qspinlock", + sizeof(struct pv_hash_entry), + pv_hash_size, 0, HASH_EARLY, + &pv_lock_hash_bits, NULL, + pv_hash_size, pv_hash_size); +} + +#define for_each_hash_entry(he, offset, hash) \ + for (hash &= ~(PV_HE_PER_LINE - 1), he = &pv_lock_hash[hash], offset = 0; \ + offset < (1 << pv_lock_hash_bits); \ + offset++, he = &pv_lock_hash[(hash + offset) & ((1 << pv_lock_hash_bits) - 1)]) + +static struct qspinlock **pv_hash(struct qspinlock *lock, struct pv_node *node) +{ + unsigned long offset, hash = hash_ptr(lock, pv_lock_hash_bits); + struct pv_hash_entry *he; + + for_each_hash_entry(he, offset, hash) { + if (!cmpxchg(&he->lock, NULL, lock)) { + WRITE_ONCE(he->node, node); + return &he->lock; + } + } + /* + * Hard assume there is a free entry for us. + * + * This is guaranteed by ensuring every blocked lock only ever consumes + * a single entry, and since we only have 4 nesting levels per CPU + * and allocated 4*nr_possible_cpus(), this must be so. + * + * The single entry is guaranteed by having the lock owner unhash + * before it releases. + */ + BUG(); +} + +static struct pv_node *pv_unhash(struct qspinlock *lock) +{ + unsigned long offset, hash = hash_ptr(lock, pv_lock_hash_bits); + struct pv_hash_entry *he; + struct pv_node *node; + + for_each_hash_entry(he, offset, hash) { + if (READ_ONCE(he->lock) == lock) { + node = READ_ONCE(he->node); + WRITE_ONCE(he->lock, NULL); + return node; + } + } + /* + * Hard assume we'll find an entry. + * + * This guarantees a limited lookup time and is itself guaranteed by + * having the lock owner do the unhash -- IFF the unlock sees the + * SLOW flag, there MUST be a hash entry. + */ + BUG(); +} + +/* + * Initialize the PV part of the mcs_spinlock node. + */ +static void pv_init_node(struct mcs_spinlock *node) +{ + struct pv_node *pn = (struct pv_node *)node; + + BUILD_BUG_ON(sizeof(struct pv_node) > 5*sizeof(struct mcs_spinlock)); + + pn->cpu = smp_processor_id(); + pn->state = vcpu_running; +} + +/* + * Wait for node->locked to become true, halt the vcpu after a short spin. + * pv_kick_node() is used to wake the vcpu again. + */ +static void pv_wait_node(struct mcs_spinlock *node) +{ + struct pv_node *pn = (struct pv_node *)node; + int loop; + + for (;;) { + for (loop = SPIN_THRESHOLD; loop; loop--) { + if (READ_ONCE(node->locked)) + return; + cpu_relax(); + } + + /* + * Order pn->state vs pn->locked thusly: + * + * [S] pn->state = vcpu_halted [S] next->locked = 1 + * MB MB + * [L] pn->locked [RmW] pn->state = vcpu_running + * + * Matches the xchg() from pv_kick_node(). + */ + smp_store_mb(pn->state, vcpu_halted); + + if (!READ_ONCE(node->locked)) + pv_wait(&pn->state, vcpu_halted); + + /* + * Reset the vCPU state to avoid unncessary CPU kicking + */ + WRITE_ONCE(pn->state, vcpu_running); + + /* + * If the locked flag is still not set after wakeup, it is a + * spurious wakeup and the vCPU should wait again. However, + * there is a pretty high overhead for CPU halting and kicking. + * So it is better to spin for a while in the hope that the + * MCS lock will be released soon. + */ + } + /* + * By now our node->locked should be 1 and our caller will not actually + * spin-wait for it. We do however rely on our caller to do a + * load-acquire for us. + */ +} + +/* + * Called after setting next->locked = 1, used to wake those stuck in + * pv_wait_node(). + */ +static void pv_kick_node(struct mcs_spinlock *node) +{ + struct pv_node *pn = (struct pv_node *)node; + + /* + * Note that because node->locked is already set, this actual + * mcs_spinlock entry could be re-used already. + * + * This should be fine however, kicking people for no reason is + * harmless. + * + * See the comment in pv_wait_node(). + */ + if (xchg(&pn->state, vcpu_running) == vcpu_halted) + pv_kick(pn->cpu); +} + +/* + * Wait for l->locked to become clear; halt the vcpu after a short spin. + * __pv_queued_spin_unlock() will wake us. + */ +static void pv_wait_head(struct qspinlock *lock, struct mcs_spinlock *node) +{ + struct pv_node *pn = (struct pv_node *)node; + struct __qspinlock *l = (void *)lock; + struct qspinlock **lp = NULL; + int loop; + + for (;;) { + for (loop = SPIN_THRESHOLD; loop; loop--) { + if (!READ_ONCE(l->locked)) + return; + cpu_relax(); + } + + WRITE_ONCE(pn->state, vcpu_halted); + if (!lp) { /* ONCE */ + lp = pv_hash(lock, pn); + /* + * lp must be set before setting _Q_SLOW_VAL + * + * [S] lp = lock [RmW] l = l->locked = 0 + * MB MB + * [S] l->locked = _Q_SLOW_VAL [L] lp + * + * Matches the cmpxchg() in __pv_queued_spin_unlock(). + */ + if (!cmpxchg(&l->locked, _Q_LOCKED_VAL, _Q_SLOW_VAL)) { + /* + * The lock is free and _Q_SLOW_VAL has never + * been set. Therefore we need to unhash before + * getting the lock. + */ + WRITE_ONCE(*lp, NULL); + return; + } + } + pv_wait(&l->locked, _Q_SLOW_VAL); + + /* + * The unlocker should have freed the lock before kicking the + * CPU. So if the lock is still not free, it is a spurious + * wakeup and so the vCPU should wait again after spinning for + * a while. + */ + } + + /* + * Lock is unlocked now; the caller will acquire it without waiting. + * As with pv_wait_node() we rely on the caller to do a load-acquire + * for us. + */ +} + +/* + * PV version of the unlock function to be used in stead of + * queued_spin_unlock(). + */ +__visible void __pv_queued_spin_unlock(struct qspinlock *lock) +{ + struct __qspinlock *l = (void *)lock; + struct pv_node *node; + + /* + * We must not unlock if SLOW, because in that case we must first + * unhash. Otherwise it would be possible to have multiple @lock + * entries, which would be BAD. + */ + if (likely(cmpxchg(&l->locked, _Q_LOCKED_VAL, 0) == _Q_LOCKED_VAL)) + return; + + /* + * Since the above failed to release, this must be the SLOW path. + * Therefore start by looking up the blocked node and unhashing it. + */ + node = pv_unhash(lock); + + /* + * Now that we have a reference to the (likely) blocked pv_node, + * release the lock. + */ + smp_store_release(&l->locked, 0); + + /* + * At this point the memory pointed at by lock can be freed/reused, + * however we can still use the pv_node to kick the CPU. + */ + if (READ_ONCE(node->state) == vcpu_halted) + pv_kick(node->cpu); +} +/* + * Include the architecture specific callee-save thunk of the + * __pv_queued_spin_unlock(). This thunk is put together with + * __pv_queued_spin_unlock() near the top of the file to make sure + * that the callee-save thunk and the real unlock function are close + * to each other sharing consecutive instruction cachelines. + */ +#include <asm/qspinlock_paravirt.h> + diff --git a/kernel/locking/rtmutex.c b/kernel/locking/rtmutex.c index b025295..36573e9 100644 --- a/kernel/locking/rtmutex.c +++ b/kernel/locking/rtmutex.c @@ -70,10 +70,10 @@ static void fixup_rt_mutex_waiters(struct rt_mutex *lock) } /* - * We can speed up the acquire/release, if the architecture - * supports cmpxchg and if there's no debugging state to be set up + * We can speed up the acquire/release, if there's no debugging state to be + * set up. */ -#if defined(__HAVE_ARCH_CMPXCHG) && !defined(CONFIG_DEBUG_RT_MUTEXES) +#ifndef CONFIG_DEBUG_RT_MUTEXES # define rt_mutex_cmpxchg(l,c,n) (cmpxchg(&l->owner, c, n) == c) static inline void mark_rt_mutex_waiters(struct rt_mutex *lock) { @@ -1182,11 +1182,8 @@ rt_mutex_slowlock(struct rt_mutex *lock, int state, set_current_state(state); /* Setup the timer, when timeout != NULL */ - if (unlikely(timeout)) { + if (unlikely(timeout)) hrtimer_start_expires(&timeout->timer, HRTIMER_MODE_ABS); - if (!hrtimer_active(&timeout->timer)) - timeout->task = NULL; - } ret = task_blocks_on_rt_mutex(lock, &waiter, current, chwalk); @@ -1443,10 +1440,17 @@ EXPORT_SYMBOL_GPL(rt_mutex_timed_lock); * * @lock: the rt_mutex to be locked * + * This function can only be called in thread context. It's safe to + * call it from atomic regions, but not from hard interrupt or soft + * interrupt context. + * * Returns 1 on success and 0 on contention */ int __sched rt_mutex_trylock(struct rt_mutex *lock) { + if (WARN_ON(in_irq() || in_nmi() || in_serving_softirq())) + return 0; + return rt_mutex_fasttrylock(lock, rt_mutex_slowtrylock); } EXPORT_SYMBOL_GPL(rt_mutex_trylock); diff --git a/kernel/locking/rwsem-xadd.c b/kernel/locking/rwsem-xadd.c index 3417d01..0f18971 100644 --- a/kernel/locking/rwsem-xadd.c +++ b/kernel/locking/rwsem-xadd.c @@ -409,11 +409,24 @@ done: return taken; } +/* + * Return true if the rwsem has active spinner + */ +static inline bool rwsem_has_spinner(struct rw_semaphore *sem) +{ + return osq_is_locked(&sem->osq); +} + #else static bool rwsem_optimistic_spin(struct rw_semaphore *sem) { return false; } + +static inline bool rwsem_has_spinner(struct rw_semaphore *sem) +{ + return false; +} #endif /* @@ -496,7 +509,38 @@ struct rw_semaphore *rwsem_wake(struct rw_semaphore *sem) { unsigned long flags; + /* + * If a spinner is present, it is not necessary to do the wakeup. + * Try to do wakeup only if the trylock succeeds to minimize + * spinlock contention which may introduce too much delay in the + * unlock operation. + * + * spinning writer up_write/up_read caller + * --------------- ----------------------- + * [S] osq_unlock() [L] osq + * MB RMB + * [RmW] rwsem_try_write_lock() [RmW] spin_trylock(wait_lock) + * + * Here, it is important to make sure that there won't be a missed + * wakeup while the rwsem is free and the only spinning writer goes + * to sleep without taking the rwsem. Even when the spinning writer + * is just going to break out of the waiting loop, it will still do + * a trylock in rwsem_down_write_failed() before sleeping. IOW, if + * rwsem_has_spinner() is true, it will guarantee at least one + * trylock attempt on the rwsem later on. + */ + if (rwsem_has_spinner(sem)) { + /* + * The smp_rmb() here is to make sure that the spinner + * state is consulted before reading the wait_lock. + */ + smp_rmb(); + if (!raw_spin_trylock_irqsave(&sem->wait_lock, flags)) + return sem; + goto locked; + } raw_spin_lock_irqsave(&sem->wait_lock, flags); +locked: /* do nothing if list empty */ if (!list_empty(&sem->wait_list)) diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig index 7e01f78..9e30231 100644 --- a/kernel/power/Kconfig +++ b/kernel/power/Kconfig @@ -187,7 +187,7 @@ config DPM_WATCHDOG config DPM_WATCHDOG_TIMEOUT int "Watchdog timeout in seconds" range 1 120 - default 12 + default 60 depends on DPM_WATCHDOG config PM_TRACE diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c index 2329daa..690f78f 100644 --- a/kernel/power/hibernate.c +++ b/kernel/power/hibernate.c @@ -552,7 +552,7 @@ int hibernation_platform_enter(void) error = disable_nonboot_cpus(); if (error) - goto Platform_finish; + goto Enable_cpus; local_irq_disable(); syscore_suspend(); @@ -568,6 +568,8 @@ int hibernation_platform_enter(void) Power_up: syscore_resume(); local_irq_enable(); + + Enable_cpus: enable_nonboot_cpus(); Platform_finish: diff --git a/kernel/power/main.c b/kernel/power/main.c index 86e8157..63d395b 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -272,7 +272,7 @@ static inline void pm_print_times_init(void) { pm_print_times_enabled = !!initcall_debug; } -#else /* !CONFIG_PP_SLEEP_DEBUG */ +#else /* !CONFIG_PM_SLEEP_DEBUG */ static inline void pm_print_times_init(void) {} #endif /* CONFIG_PM_SLEEP_DEBUG */ diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c index 8d7a1ef..53266b7 100644 --- a/kernel/power/suspend.c +++ b/kernel/power/suspend.c @@ -366,6 +366,8 @@ static int suspend_enter(suspend_state_t state, bool *wakeup) trace_suspend_resume(TPS("machine_suspend"), state, false); events_check_enabled = false; + } else if (*wakeup) { + error = -EBUSY; } syscore_resume(); } @@ -468,7 +470,7 @@ static int enter_state(suspend_state_t state) if (state == PM_SUSPEND_FREEZE) { #ifdef CONFIG_PM_DEBUG if (pm_test_level != TEST_NONE && pm_test_level <= TEST_CPUS) { - pr_warning("PM: Unsupported test mode for freeze state," + pr_warning("PM: Unsupported test mode for suspend to idle," "please choose none/freezer/devices/platform.\n"); return -EAGAIN; } @@ -488,7 +490,7 @@ static int enter_state(suspend_state_t state) printk("done.\n"); trace_suspend_resume(TPS("sync_filesystems"), 0, false); - pr_debug("PM: Preparing system for %s sleep\n", pm_states[state]); + pr_debug("PM: Preparing system for sleep (%s)\n", pm_states[state]); error = suspend_prepare(state); if (error) goto Unlock; @@ -497,7 +499,7 @@ static int enter_state(suspend_state_t state) goto Finish; trace_suspend_resume(TPS("suspend_enter"), state, false); - pr_debug("PM: Entering %s sleep\n", pm_states[state]); + pr_debug("PM: Suspending system (%s)\n", pm_states[state]); pm_restrict_gfp_mask(); error = suspend_devices_and_enter(state); pm_restore_gfp_mask(); diff --git a/kernel/rcu/rcutorture.c b/kernel/rcu/rcutorture.c index 8dbe276..59e3268 100644 --- a/kernel/rcu/rcutorture.c +++ b/kernel/rcu/rcutorture.c @@ -241,6 +241,7 @@ rcu_torture_free(struct rcu_torture *p) struct rcu_torture_ops { int ttype; void (*init)(void); + void (*cleanup)(void); int (*readlock)(void); void (*read_delay)(struct torture_random_state *rrsp); void (*readunlock)(int idx); @@ -477,10 +478,12 @@ static struct rcu_torture_ops rcu_busted_ops = { */ DEFINE_STATIC_SRCU(srcu_ctl); +static struct srcu_struct srcu_ctld; +static struct srcu_struct *srcu_ctlp = &srcu_ctl; -static int srcu_torture_read_lock(void) __acquires(&srcu_ctl) +static int srcu_torture_read_lock(void) __acquires(srcu_ctlp) { - return srcu_read_lock(&srcu_ctl); + return srcu_read_lock(srcu_ctlp); } static void srcu_read_delay(struct torture_random_state *rrsp) @@ -499,49 +502,49 @@ static void srcu_read_delay(struct torture_random_state *rrsp) rcu_read_delay(rrsp); } -static void srcu_torture_read_unlock(int idx) __releases(&srcu_ctl) +static void srcu_torture_read_unlock(int idx) __releases(srcu_ctlp) { - srcu_read_unlock(&srcu_ctl, idx); + srcu_read_unlock(srcu_ctlp, idx); } static unsigned long srcu_torture_completed(void) { - return srcu_batches_completed(&srcu_ctl); + return srcu_batches_completed(srcu_ctlp); } static void srcu_torture_deferred_free(struct rcu_torture *rp) { - call_srcu(&srcu_ctl, &rp->rtort_rcu, rcu_torture_cb); + call_srcu(srcu_ctlp, &rp->rtort_rcu, rcu_torture_cb); } static void srcu_torture_synchronize(void) { - synchronize_srcu(&srcu_ctl); + synchronize_srcu(srcu_ctlp); } static void srcu_torture_call(struct rcu_head *head, void (*func)(struct rcu_head *head)) { - call_srcu(&srcu_ctl, head, func); + call_srcu(srcu_ctlp, head, func); } static void srcu_torture_barrier(void) { - srcu_barrier(&srcu_ctl); + srcu_barrier(srcu_ctlp); } static void srcu_torture_stats(void) { int cpu; - int idx = srcu_ctl.completed & 0x1; + int idx = srcu_ctlp->completed & 0x1; pr_alert("%s%s per-CPU(idx=%d):", torture_type, TORTURE_FLAG, idx); for_each_possible_cpu(cpu) { long c0, c1; - c0 = (long)per_cpu_ptr(srcu_ctl.per_cpu_ref, cpu)->c[!idx]; - c1 = (long)per_cpu_ptr(srcu_ctl.per_cpu_ref, cpu)->c[idx]; + c0 = (long)per_cpu_ptr(srcu_ctlp->per_cpu_ref, cpu)->c[!idx]; + c1 = (long)per_cpu_ptr(srcu_ctlp->per_cpu_ref, cpu)->c[idx]; pr_cont(" %d(%ld,%ld)", cpu, c0, c1); } pr_cont("\n"); @@ -549,7 +552,7 @@ static void srcu_torture_stats(void) static void srcu_torture_synchronize_expedited(void) { - synchronize_srcu_expedited(&srcu_ctl); + synchronize_srcu_expedited(srcu_ctlp); } static struct rcu_torture_ops srcu_ops = { @@ -569,6 +572,38 @@ static struct rcu_torture_ops srcu_ops = { .name = "srcu" }; +static void srcu_torture_init(void) +{ + rcu_sync_torture_init(); + WARN_ON(init_srcu_struct(&srcu_ctld)); + srcu_ctlp = &srcu_ctld; +} + +static void srcu_torture_cleanup(void) +{ + cleanup_srcu_struct(&srcu_ctld); + srcu_ctlp = &srcu_ctl; /* In case of a later rcutorture run. */ +} + +/* As above, but dynamically allocated. */ +static struct rcu_torture_ops srcud_ops = { + .ttype = SRCU_FLAVOR, + .init = srcu_torture_init, + .cleanup = srcu_torture_cleanup, + .readlock = srcu_torture_read_lock, + .read_delay = srcu_read_delay, + .readunlock = srcu_torture_read_unlock, + .started = NULL, + .completed = srcu_torture_completed, + .deferred_free = srcu_torture_deferred_free, + .sync = srcu_torture_synchronize, + .exp_sync = srcu_torture_synchronize_expedited, + .call = srcu_torture_call, + .cb_barrier = srcu_torture_barrier, + .stats = srcu_torture_stats, + .name = "srcud" +}; + /* * Definitions for sched torture testing. */ @@ -672,8 +707,8 @@ static void rcu_torture_boost_cb(struct rcu_head *head) struct rcu_boost_inflight *rbip = container_of(head, struct rcu_boost_inflight, rcu); - smp_mb(); /* Ensure RCU-core accesses precede clearing ->inflight */ - rbip->inflight = 0; + /* Ensure RCU-core accesses precede clearing ->inflight */ + smp_store_release(&rbip->inflight, 0); } static int rcu_torture_boost(void *arg) @@ -710,9 +745,9 @@ static int rcu_torture_boost(void *arg) call_rcu_time = jiffies; while (ULONG_CMP_LT(jiffies, endtime)) { /* If we don't have a callback in flight, post one. */ - if (!rbi.inflight) { - smp_mb(); /* RCU core before ->inflight = 1. */ - rbi.inflight = 1; + if (!smp_load_acquire(&rbi.inflight)) { + /* RCU core before ->inflight = 1. */ + smp_store_release(&rbi.inflight, 1); call_rcu(&rbi.rcu, rcu_torture_boost_cb); if (jiffies - call_rcu_time > test_boost_duration * HZ - HZ / 2) { @@ -751,11 +786,10 @@ checkwait: stutter_wait("rcu_torture_boost"); } while (!torture_must_stop()); /* Clean up and exit. */ - while (!kthread_should_stop() || rbi.inflight) { + while (!kthread_should_stop() || smp_load_acquire(&rbi.inflight)) { torture_shutdown_absorb("rcu_torture_boost"); schedule_timeout_uninterruptible(1); } - smp_mb(); /* order accesses to ->inflight before stack-frame death. */ destroy_rcu_head_on_stack(&rbi.rcu); torture_kthread_stopping("rcu_torture_boost"); return 0; @@ -1054,7 +1088,7 @@ static void rcu_torture_timer(unsigned long unused) p = rcu_dereference_check(rcu_torture_current, rcu_read_lock_bh_held() || rcu_read_lock_sched_held() || - srcu_read_lock_held(&srcu_ctl)); + srcu_read_lock_held(srcu_ctlp)); if (p == NULL) { /* Leave because rcu_torture_writer is not yet underway */ cur_ops->readunlock(idx); @@ -1128,7 +1162,7 @@ rcu_torture_reader(void *arg) p = rcu_dereference_check(rcu_torture_current, rcu_read_lock_bh_held() || rcu_read_lock_sched_held() || - srcu_read_lock_held(&srcu_ctl)); + srcu_read_lock_held(srcu_ctlp)); if (p == NULL) { /* Wait for rcu_torture_writer to get underway */ cur_ops->readunlock(idx); @@ -1413,12 +1447,15 @@ static int rcu_torture_barrier_cbs(void *arg) do { wait_event(barrier_cbs_wq[myid], (newphase = - ACCESS_ONCE(barrier_phase)) != lastphase || + smp_load_acquire(&barrier_phase)) != lastphase || torture_must_stop()); lastphase = newphase; - smp_mb(); /* ensure barrier_phase load before ->call(). */ if (torture_must_stop()) break; + /* + * The above smp_load_acquire() ensures barrier_phase load + * is ordered before the folloiwng ->call(). + */ cur_ops->call(&rcu, rcu_torture_barrier_cbf); if (atomic_dec_and_test(&barrier_cbs_count)) wake_up(&barrier_wq); @@ -1439,8 +1476,8 @@ static int rcu_torture_barrier(void *arg) do { atomic_set(&barrier_cbs_invoked, 0); atomic_set(&barrier_cbs_count, n_barrier_cbs); - smp_mb(); /* Ensure barrier_phase after prior assignments. */ - barrier_phase = !barrier_phase; + /* Ensure barrier_phase ordered after prior assignments. */ + smp_store_release(&barrier_phase, !barrier_phase); for (i = 0; i < n_barrier_cbs; i++) wake_up(&barrier_cbs_wq[i]); wait_event(barrier_wq, @@ -1588,10 +1625,14 @@ rcu_torture_cleanup(void) rcutorture_booster_cleanup(i); } - /* Wait for all RCU callbacks to fire. */ - + /* + * Wait for all RCU callbacks to fire, then do flavor-specific + * cleanup operations. + */ if (cur_ops->cb_barrier != NULL) cur_ops->cb_barrier(); + if (cur_ops->cleanup != NULL) + cur_ops->cleanup(); rcu_torture_stats_print(); /* -After- the stats thread is stopped! */ @@ -1668,8 +1709,8 @@ rcu_torture_init(void) int cpu; int firsterr = 0; static struct rcu_torture_ops *torture_ops[] = { - &rcu_ops, &rcu_bh_ops, &rcu_busted_ops, &srcu_ops, &sched_ops, - RCUTORTURE_TASKS_OPS + &rcu_ops, &rcu_bh_ops, &rcu_busted_ops, &srcu_ops, &srcud_ops, + &sched_ops, RCUTORTURE_TASKS_OPS }; if (!torture_init_begin(torture_type, verbose, &torture_runnable)) @@ -1701,7 +1742,7 @@ rcu_torture_init(void) if (nreaders >= 0) { nrealreaders = nreaders; } else { - nrealreaders = num_online_cpus() - 1; + nrealreaders = num_online_cpus() - 2 - nreaders; if (nrealreaders <= 0) nrealreaders = 1; } diff --git a/kernel/rcu/srcu.c b/kernel/rcu/srcu.c index cad76e7..fb33d35 100644 --- a/kernel/rcu/srcu.c +++ b/kernel/rcu/srcu.c @@ -151,7 +151,7 @@ static unsigned long srcu_readers_seq_idx(struct srcu_struct *sp, int idx) unsigned long t; for_each_possible_cpu(cpu) { - t = ACCESS_ONCE(per_cpu_ptr(sp->per_cpu_ref, cpu)->seq[idx]); + t = READ_ONCE(per_cpu_ptr(sp->per_cpu_ref, cpu)->seq[idx]); sum += t; } return sum; @@ -168,7 +168,7 @@ static unsigned long srcu_readers_active_idx(struct srcu_struct *sp, int idx) unsigned long t; for_each_possible_cpu(cpu) { - t = ACCESS_ONCE(per_cpu_ptr(sp->per_cpu_ref, cpu)->c[idx]); + t = READ_ONCE(per_cpu_ptr(sp->per_cpu_ref, cpu)->c[idx]); sum += t; } return sum; @@ -265,8 +265,8 @@ static int srcu_readers_active(struct srcu_struct *sp) unsigned long sum = 0; for_each_possible_cpu(cpu) { - sum += ACCESS_ONCE(per_cpu_ptr(sp->per_cpu_ref, cpu)->c[0]); - sum += ACCESS_ONCE(per_cpu_ptr(sp->per_cpu_ref, cpu)->c[1]); + sum += READ_ONCE(per_cpu_ptr(sp->per_cpu_ref, cpu)->c[0]); + sum += READ_ONCE(per_cpu_ptr(sp->per_cpu_ref, cpu)->c[1]); } return sum; } @@ -296,7 +296,7 @@ int __srcu_read_lock(struct srcu_struct *sp) { int idx; - idx = ACCESS_ONCE(sp->completed) & 0x1; + idx = READ_ONCE(sp->completed) & 0x1; preempt_disable(); __this_cpu_inc(sp->per_cpu_ref->c[idx]); smp_mb(); /* B */ /* Avoid leaking the critical section. */ diff --git a/kernel/rcu/tiny.c b/kernel/rcu/tiny.c index 069742d..591af0c 100644 --- a/kernel/rcu/tiny.c +++ b/kernel/rcu/tiny.c @@ -49,39 +49,6 @@ static void __call_rcu(struct rcu_head *head, #include "tiny_plugin.h" -/* - * Enter idle, which is an extended quiescent state if we have fully - * entered that mode. - */ -void rcu_idle_enter(void) -{ -} -EXPORT_SYMBOL_GPL(rcu_idle_enter); - -/* - * Exit an interrupt handler towards idle. - */ -void rcu_irq_exit(void) -{ -} -EXPORT_SYMBOL_GPL(rcu_irq_exit); - -/* - * Exit idle, so that we are no longer in an extended quiescent state. - */ -void rcu_idle_exit(void) -{ -} -EXPORT_SYMBOL_GPL(rcu_idle_exit); - -/* - * Enter an interrupt handler, moving away from idle. - */ -void rcu_irq_enter(void) -{ -} -EXPORT_SYMBOL_GPL(rcu_irq_enter); - #if defined(CONFIG_DEBUG_LOCK_ALLOC) || defined(CONFIG_RCU_TRACE) /* @@ -170,6 +137,11 @@ static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp) /* Move the ready-to-invoke callbacks to a local list. */ local_irq_save(flags); + if (rcp->donetail == &rcp->rcucblist) { + /* No callbacks ready, so just leave. */ + local_irq_restore(flags); + return; + } RCU_TRACE(trace_rcu_batch_start(rcp->name, 0, rcp->qlen, -1)); list = rcp->rcucblist; rcp->rcucblist = *rcp->donetail; diff --git a/kernel/rcu/tiny_plugin.h b/kernel/rcu/tiny_plugin.h index f94e209..e492a52 100644 --- a/kernel/rcu/tiny_plugin.h +++ b/kernel/rcu/tiny_plugin.h @@ -144,16 +144,17 @@ static void check_cpu_stall(struct rcu_ctrlblk *rcp) return; rcp->ticks_this_gp++; j = jiffies; - js = ACCESS_ONCE(rcp->jiffies_stall); + js = READ_ONCE(rcp->jiffies_stall); if (rcp->rcucblist && ULONG_CMP_GE(j, js)) { pr_err("INFO: %s stall on CPU (%lu ticks this GP) idle=%llx (t=%lu jiffies q=%ld)\n", rcp->name, rcp->ticks_this_gp, DYNTICK_TASK_EXIT_IDLE, jiffies - rcp->gp_start, rcp->qlen); dump_stack(); - ACCESS_ONCE(rcp->jiffies_stall) = jiffies + - 3 * rcu_jiffies_till_stall_check() + 3; + WRITE_ONCE(rcp->jiffies_stall, + jiffies + 3 * rcu_jiffies_till_stall_check() + 3); } else if (ULONG_CMP_GE(j, js)) { - ACCESS_ONCE(rcp->jiffies_stall) = jiffies + rcu_jiffies_till_stall_check(); + WRITE_ONCE(rcp->jiffies_stall, + jiffies + rcu_jiffies_till_stall_check()); } } @@ -161,7 +162,8 @@ static void reset_cpu_stall_ticks(struct rcu_ctrlblk *rcp) { rcp->ticks_this_gp = 0; rcp->gp_start = jiffies; - ACCESS_ONCE(rcp->jiffies_stall) = jiffies + rcu_jiffies_till_stall_check(); + WRITE_ONCE(rcp->jiffies_stall, + jiffies + rcu_jiffies_till_stall_check()); } static void check_cpu_stalls(void) diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 8cf7304..add0429 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -91,7 +91,7 @@ static const char *tp_##sname##_varname __used __tracepoint_string = sname##_var #define RCU_STATE_INITIALIZER(sname, sabbr, cr) \ DEFINE_RCU_TPS(sname) \ -DEFINE_PER_CPU_SHARED_ALIGNED(struct rcu_data, sname##_data); \ +static DEFINE_PER_CPU_SHARED_ALIGNED(struct rcu_data, sname##_data); \ struct rcu_state sname##_state = { \ .level = { &sname##_state.node[0] }, \ .rda = &sname##_data, \ @@ -110,11 +110,18 @@ struct rcu_state sname##_state = { \ RCU_STATE_INITIALIZER(rcu_sched, 's', call_rcu_sched); RCU_STATE_INITIALIZER(rcu_bh, 'b', call_rcu_bh); -static struct rcu_state *rcu_state_p; +static struct rcu_state *const rcu_state_p; +static struct rcu_data __percpu *const rcu_data_p; LIST_HEAD(rcu_struct_flavors); -/* Increase (but not decrease) the CONFIG_RCU_FANOUT_LEAF at boot time. */ -static int rcu_fanout_leaf = CONFIG_RCU_FANOUT_LEAF; +/* Dump rcu_node combining tree at boot to verify correct setup. */ +static bool dump_tree; +module_param(dump_tree, bool, 0444); +/* Control rcu_node-tree auto-balancing at boot time. */ +static bool rcu_fanout_exact; +module_param(rcu_fanout_exact, bool, 0444); +/* Increase (but not decrease) the RCU_FANOUT_LEAF at boot time. */ +static int rcu_fanout_leaf = RCU_FANOUT_LEAF; module_param(rcu_fanout_leaf, int, 0444); int rcu_num_lvls __read_mostly = RCU_NUM_LVLS; static int num_rcu_lvl[] = { /* Number of rcu_nodes at specified level. */ @@ -159,17 +166,46 @@ static void invoke_rcu_core(void); static void invoke_rcu_callbacks(struct rcu_state *rsp, struct rcu_data *rdp); /* rcuc/rcub kthread realtime priority */ +#ifdef CONFIG_RCU_KTHREAD_PRIO static int kthread_prio = CONFIG_RCU_KTHREAD_PRIO; +#else /* #ifdef CONFIG_RCU_KTHREAD_PRIO */ +static int kthread_prio = IS_ENABLED(CONFIG_RCU_BOOST) ? 1 : 0; +#endif /* #else #ifdef CONFIG_RCU_KTHREAD_PRIO */ module_param(kthread_prio, int, 0644); /* Delay in jiffies for grace-period initialization delays, debug only. */ + +#ifdef CONFIG_RCU_TORTURE_TEST_SLOW_PREINIT +static int gp_preinit_delay = CONFIG_RCU_TORTURE_TEST_SLOW_PREINIT_DELAY; +module_param(gp_preinit_delay, int, 0644); +#else /* #ifdef CONFIG_RCU_TORTURE_TEST_SLOW_PREINIT */ +static const int gp_preinit_delay; +#endif /* #else #ifdef CONFIG_RCU_TORTURE_TEST_SLOW_PREINIT */ + #ifdef CONFIG_RCU_TORTURE_TEST_SLOW_INIT static int gp_init_delay = CONFIG_RCU_TORTURE_TEST_SLOW_INIT_DELAY; module_param(gp_init_delay, int, 0644); #else /* #ifdef CONFIG_RCU_TORTURE_TEST_SLOW_INIT */ static const int gp_init_delay; #endif /* #else #ifdef CONFIG_RCU_TORTURE_TEST_SLOW_INIT */ -#define PER_RCU_NODE_PERIOD 10 /* Number of grace periods between delays. */ + +#ifdef CONFIG_RCU_TORTURE_TEST_SLOW_CLEANUP +static int gp_cleanup_delay = CONFIG_RCU_TORTURE_TEST_SLOW_CLEANUP_DELAY; +module_param(gp_cleanup_delay, int, 0644); +#else /* #ifdef CONFIG_RCU_TORTURE_TEST_SLOW_CLEANUP */ +static const int gp_cleanup_delay; +#endif /* #else #ifdef CONFIG_RCU_TORTURE_TEST_SLOW_CLEANUP */ + +/* + * Number of grace periods between delays, normalized by the duration of + * the delay. The longer the the delay, the more the grace periods between + * each delay. The reason for this normalization is that it means that, + * for non-zero delays, the overall slowdown of grace periods is constant + * regardless of the duration of the delay. This arrangement balances + * the need for long delays to increase some race probabilities with the + * need for fast grace periods to increase other race probabilities. + */ +#define PER_RCU_NODE_PERIOD 3 /* Number of grace periods between delays. */ /* * Track the rcutorture test sequence number and the update version @@ -191,17 +227,17 @@ unsigned long rcutorture_vernum; */ unsigned long rcu_rnp_online_cpus(struct rcu_node *rnp) { - return ACCESS_ONCE(rnp->qsmaskinitnext); + return READ_ONCE(rnp->qsmaskinitnext); } /* - * Return true if an RCU grace period is in progress. The ACCESS_ONCE()s + * Return true if an RCU grace period is in progress. The READ_ONCE()s * permit this function to be invoked without holding the root rcu_node * structure's ->lock, but of course results can be subject to change. */ static int rcu_gp_in_progress(struct rcu_state *rsp) { - return ACCESS_ONCE(rsp->completed) != ACCESS_ONCE(rsp->gpnum); + return READ_ONCE(rsp->completed) != READ_ONCE(rsp->gpnum); } /* @@ -278,8 +314,8 @@ static void rcu_momentary_dyntick_idle(void) if (!(resched_mask & rsp->flavor_mask)) continue; smp_mb(); /* rcu_sched_qs_mask before cond_resched_completed. */ - if (ACCESS_ONCE(rdp->mynode->completed) != - ACCESS_ONCE(rdp->cond_resched_completed)) + if (READ_ONCE(rdp->mynode->completed) != + READ_ONCE(rdp->cond_resched_completed)) continue; /* @@ -491,9 +527,9 @@ void rcutorture_get_gp_data(enum rcutorture_type test_type, int *flags, break; } if (rsp != NULL) { - *flags = ACCESS_ONCE(rsp->gp_flags); - *gpnum = ACCESS_ONCE(rsp->gpnum); - *completed = ACCESS_ONCE(rsp->completed); + *flags = READ_ONCE(rsp->gp_flags); + *gpnum = READ_ONCE(rsp->gpnum); + *completed = READ_ONCE(rsp->completed); return; } *flags = 0; @@ -539,10 +575,10 @@ static struct rcu_node *rcu_get_root(struct rcu_state *rsp) static int rcu_future_needs_gp(struct rcu_state *rsp) { struct rcu_node *rnp = rcu_get_root(rsp); - int idx = (ACCESS_ONCE(rnp->completed) + 1) & 0x1; + int idx = (READ_ONCE(rnp->completed) + 1) & 0x1; int *fp = &rnp->need_future_gp[idx]; - return ACCESS_ONCE(*fp); + return READ_ONCE(*fp); } /* @@ -565,7 +601,7 @@ cpu_needs_another_gp(struct rcu_state *rsp, struct rcu_data *rdp) return 1; /* Yes, this CPU has newly registered callbacks. */ for (i = RCU_WAIT_TAIL; i < RCU_NEXT_TAIL; i++) if (rdp->nxttail[i - 1] != rdp->nxttail[i] && - ULONG_CMP_LT(ACCESS_ONCE(rsp->completed), + ULONG_CMP_LT(READ_ONCE(rsp->completed), rdp->nxtcompleted[i])) return 1; /* Yes, CBs for future grace period. */ return 0; /* No grace period needed. */ @@ -585,7 +621,8 @@ static void rcu_eqs_enter_common(long long oldval, bool user) struct rcu_dynticks *rdtp = this_cpu_ptr(&rcu_dynticks); trace_rcu_dyntick(TPS("Start"), oldval, rdtp->dynticks_nesting); - if (!user && !is_idle_task(current)) { + if (IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && + !user && !is_idle_task(current)) { struct task_struct *idle __maybe_unused = idle_task(smp_processor_id()); @@ -604,7 +641,8 @@ static void rcu_eqs_enter_common(long long oldval, bool user) smp_mb__before_atomic(); /* See above. */ atomic_inc(&rdtp->dynticks); smp_mb__after_atomic(); /* Force ordering with next sojourn. */ - WARN_ON_ONCE(atomic_read(&rdtp->dynticks) & 0x1); + WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && + atomic_read(&rdtp->dynticks) & 0x1); rcu_dynticks_task_enter(); /* @@ -630,7 +668,8 @@ static void rcu_eqs_enter(bool user) rdtp = this_cpu_ptr(&rcu_dynticks); oldval = rdtp->dynticks_nesting; - WARN_ON_ONCE((oldval & DYNTICK_TASK_NEST_MASK) == 0); + WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && + (oldval & DYNTICK_TASK_NEST_MASK) == 0); if ((oldval & DYNTICK_TASK_NEST_MASK) == DYNTICK_TASK_NEST_VALUE) { rdtp->dynticks_nesting = 0; rcu_eqs_enter_common(oldval, user); @@ -703,7 +742,8 @@ void rcu_irq_exit(void) rdtp = this_cpu_ptr(&rcu_dynticks); oldval = rdtp->dynticks_nesting; rdtp->dynticks_nesting--; - WARN_ON_ONCE(rdtp->dynticks_nesting < 0); + WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && + rdtp->dynticks_nesting < 0); if (rdtp->dynticks_nesting) trace_rcu_dyntick(TPS("--="), oldval, rdtp->dynticks_nesting); else @@ -728,10 +768,12 @@ static void rcu_eqs_exit_common(long long oldval, int user) atomic_inc(&rdtp->dynticks); /* CPUs seeing atomic_inc() must see later RCU read-side crit sects */ smp_mb__after_atomic(); /* See above. */ - WARN_ON_ONCE(!(atomic_read(&rdtp->dynticks) & 0x1)); + WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && + !(atomic_read(&rdtp->dynticks) & 0x1)); rcu_cleanup_after_idle(); trace_rcu_dyntick(TPS("End"), oldval, rdtp->dynticks_nesting); - if (!user && !is_idle_task(current)) { + if (IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && + !user && !is_idle_task(current)) { struct task_struct *idle __maybe_unused = idle_task(smp_processor_id()); @@ -755,7 +797,7 @@ static void rcu_eqs_exit(bool user) rdtp = this_cpu_ptr(&rcu_dynticks); oldval = rdtp->dynticks_nesting; - WARN_ON_ONCE(oldval < 0); + WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && oldval < 0); if (oldval & DYNTICK_TASK_NEST_MASK) { rdtp->dynticks_nesting += DYNTICK_TASK_NEST_VALUE; } else { @@ -828,7 +870,8 @@ void rcu_irq_enter(void) rdtp = this_cpu_ptr(&rcu_dynticks); oldval = rdtp->dynticks_nesting; rdtp->dynticks_nesting++; - WARN_ON_ONCE(rdtp->dynticks_nesting == 0); + WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && + rdtp->dynticks_nesting == 0); if (oldval) trace_rcu_dyntick(TPS("++="), oldval, rdtp->dynticks_nesting); else @@ -1011,9 +1054,9 @@ static int dyntick_save_progress_counter(struct rcu_data *rdp, trace_rcu_fqs(rdp->rsp->name, rdp->gpnum, rdp->cpu, TPS("dti")); return 1; } else { - if (ULONG_CMP_LT(ACCESS_ONCE(rdp->gpnum) + ULONG_MAX / 4, + if (ULONG_CMP_LT(READ_ONCE(rdp->gpnum) + ULONG_MAX / 4, rdp->mynode->gpnum)) - ACCESS_ONCE(rdp->gpwrap) = true; + WRITE_ONCE(rdp->gpwrap, true); return 0; } } @@ -1093,12 +1136,12 @@ static int rcu_implicit_dynticks_qs(struct rcu_data *rdp, if (ULONG_CMP_GE(jiffies, rdp->rsp->gp_start + jiffies_till_sched_qs) || ULONG_CMP_GE(jiffies, rdp->rsp->jiffies_resched)) { - if (!(ACCESS_ONCE(*rcrmp) & rdp->rsp->flavor_mask)) { - ACCESS_ONCE(rdp->cond_resched_completed) = - ACCESS_ONCE(rdp->mynode->completed); + if (!(READ_ONCE(*rcrmp) & rdp->rsp->flavor_mask)) { + WRITE_ONCE(rdp->cond_resched_completed, + READ_ONCE(rdp->mynode->completed)); smp_mb(); /* ->cond_resched_completed before *rcrmp. */ - ACCESS_ONCE(*rcrmp) = - ACCESS_ONCE(*rcrmp) + rdp->rsp->flavor_mask; + WRITE_ONCE(*rcrmp, + READ_ONCE(*rcrmp) + rdp->rsp->flavor_mask); resched_cpu(rdp->cpu); /* Force CPU into scheduler. */ rdp->rsp->jiffies_resched += 5; /* Enable beating. */ } else if (ULONG_CMP_GE(jiffies, rdp->rsp->jiffies_resched)) { @@ -1119,9 +1162,9 @@ static void record_gp_stall_check_time(struct rcu_state *rsp) rsp->gp_start = j; smp_wmb(); /* Record start time before stall time. */ j1 = rcu_jiffies_till_stall_check(); - ACCESS_ONCE(rsp->jiffies_stall) = j + j1; + WRITE_ONCE(rsp->jiffies_stall, j + j1); rsp->jiffies_resched = j + j1 / 2; - rsp->n_force_qs_gpstart = ACCESS_ONCE(rsp->n_force_qs); + rsp->n_force_qs_gpstart = READ_ONCE(rsp->n_force_qs); } /* @@ -1133,10 +1176,11 @@ static void rcu_check_gp_kthread_starvation(struct rcu_state *rsp) unsigned long j; j = jiffies; - gpa = ACCESS_ONCE(rsp->gp_activity); + gpa = READ_ONCE(rsp->gp_activity); if (j - gpa > 2 * HZ) - pr_err("%s kthread starved for %ld jiffies!\n", - rsp->name, j - gpa); + pr_err("%s kthread starved for %ld jiffies! g%lu c%lu f%#x\n", + rsp->name, j - gpa, + rsp->gpnum, rsp->completed, rsp->gp_flags); } /* @@ -1173,12 +1217,13 @@ static void print_other_cpu_stall(struct rcu_state *rsp, unsigned long gpnum) /* Only let one CPU complain about others per time interval. */ raw_spin_lock_irqsave(&rnp->lock, flags); - delta = jiffies - ACCESS_ONCE(rsp->jiffies_stall); + delta = jiffies - READ_ONCE(rsp->jiffies_stall); if (delta < RCU_STALL_RAT_DELAY || !rcu_gp_in_progress(rsp)) { raw_spin_unlock_irqrestore(&rnp->lock, flags); return; } - ACCESS_ONCE(rsp->jiffies_stall) = jiffies + 3 * rcu_jiffies_till_stall_check() + 3; + WRITE_ONCE(rsp->jiffies_stall, + jiffies + 3 * rcu_jiffies_till_stall_check() + 3); raw_spin_unlock_irqrestore(&rnp->lock, flags); /* @@ -1212,12 +1257,12 @@ static void print_other_cpu_stall(struct rcu_state *rsp, unsigned long gpnum) if (ndetected) { rcu_dump_cpu_stacks(rsp); } else { - if (ACCESS_ONCE(rsp->gpnum) != gpnum || - ACCESS_ONCE(rsp->completed) == gpnum) { + if (READ_ONCE(rsp->gpnum) != gpnum || + READ_ONCE(rsp->completed) == gpnum) { pr_err("INFO: Stall ended before state dump start\n"); } else { j = jiffies; - gpa = ACCESS_ONCE(rsp->gp_activity); + gpa = READ_ONCE(rsp->gp_activity); pr_err("All QSes seen, last %s kthread activity %ld (%ld-%ld), jiffies_till_next_fqs=%ld, root ->qsmask %#lx\n", rsp->name, j - gpa, j, gpa, jiffies_till_next_fqs, @@ -1262,9 +1307,9 @@ static void print_cpu_stall(struct rcu_state *rsp) rcu_dump_cpu_stacks(rsp); raw_spin_lock_irqsave(&rnp->lock, flags); - if (ULONG_CMP_GE(jiffies, ACCESS_ONCE(rsp->jiffies_stall))) - ACCESS_ONCE(rsp->jiffies_stall) = jiffies + - 3 * rcu_jiffies_till_stall_check() + 3; + if (ULONG_CMP_GE(jiffies, READ_ONCE(rsp->jiffies_stall))) + WRITE_ONCE(rsp->jiffies_stall, + jiffies + 3 * rcu_jiffies_till_stall_check() + 3); raw_spin_unlock_irqrestore(&rnp->lock, flags); /* @@ -1307,20 +1352,20 @@ static void check_cpu_stall(struct rcu_state *rsp, struct rcu_data *rdp) * Given this check, comparisons of jiffies, rsp->jiffies_stall, * and rsp->gp_start suffice to forestall false positives. */ - gpnum = ACCESS_ONCE(rsp->gpnum); + gpnum = READ_ONCE(rsp->gpnum); smp_rmb(); /* Pick up ->gpnum first... */ - js = ACCESS_ONCE(rsp->jiffies_stall); + js = READ_ONCE(rsp->jiffies_stall); smp_rmb(); /* ...then ->jiffies_stall before the rest... */ - gps = ACCESS_ONCE(rsp->gp_start); + gps = READ_ONCE(rsp->gp_start); smp_rmb(); /* ...and finally ->gp_start before ->completed. */ - completed = ACCESS_ONCE(rsp->completed); + completed = READ_ONCE(rsp->completed); if (ULONG_CMP_GE(completed, gpnum) || ULONG_CMP_LT(j, js) || ULONG_CMP_GE(gps, js)) return; /* No stall or GP completed since entering function. */ rnp = rdp->mynode; if (rcu_gp_in_progress(rsp) && - (ACCESS_ONCE(rnp->qsmask) & rdp->grpmask)) { + (READ_ONCE(rnp->qsmask) & rdp->grpmask)) { /* We haven't checked in, so go dump stack. */ print_cpu_stall(rsp); @@ -1347,7 +1392,7 @@ void rcu_cpu_stall_reset(void) struct rcu_state *rsp; for_each_rcu_flavor(rsp) - ACCESS_ONCE(rsp->jiffies_stall) = jiffies + ULONG_MAX / 2; + WRITE_ONCE(rsp->jiffies_stall, jiffies + ULONG_MAX / 2); } /* @@ -1457,7 +1502,7 @@ rcu_start_future_gp(struct rcu_node *rnp, struct rcu_data *rdp, * doing some extra useless work. */ if (rnp->gpnum != rnp->completed || - ACCESS_ONCE(rnp_root->gpnum) != ACCESS_ONCE(rnp_root->completed)) { + READ_ONCE(rnp_root->gpnum) != READ_ONCE(rnp_root->completed)) { rnp->need_future_gp[c & 0x1]++; trace_rcu_future_gp(rnp, rdp, c, TPS("Startedleaf")); goto out; @@ -1542,7 +1587,7 @@ static int rcu_future_gp_cleanup(struct rcu_state *rsp, struct rcu_node *rnp) static void rcu_gp_kthread_wake(struct rcu_state *rsp) { if (current == rsp->gp_kthread || - !ACCESS_ONCE(rsp->gp_flags) || + !READ_ONCE(rsp->gp_flags) || !rsp->gp_kthread) return; wake_up(&rsp->gp_wq); @@ -1677,7 +1722,7 @@ static bool __note_gp_changes(struct rcu_state *rsp, struct rcu_node *rnp, /* Handle the ends of any preceding grace periods first. */ if (rdp->completed == rnp->completed && - !unlikely(ACCESS_ONCE(rdp->gpwrap))) { + !unlikely(READ_ONCE(rdp->gpwrap))) { /* No grace period end, so just accelerate recent callbacks. */ ret = rcu_accelerate_cbs(rsp, rnp, rdp); @@ -1692,7 +1737,7 @@ static bool __note_gp_changes(struct rcu_state *rsp, struct rcu_node *rnp, trace_rcu_grace_period(rsp->name, rdp->gpnum, TPS("cpuend")); } - if (rdp->gpnum != rnp->gpnum || unlikely(ACCESS_ONCE(rdp->gpwrap))) { + if (rdp->gpnum != rnp->gpnum || unlikely(READ_ONCE(rdp->gpwrap))) { /* * If the current grace period is waiting for this CPU, * set up to detect a quiescent state, otherwise don't @@ -1704,7 +1749,7 @@ static bool __note_gp_changes(struct rcu_state *rsp, struct rcu_node *rnp, rdp->rcu_qs_ctr_snap = __this_cpu_read(rcu_qs_ctr); rdp->qs_pending = !!(rnp->qsmask & rdp->grpmask); zero_cpu_stall_ticks(rdp); - ACCESS_ONCE(rdp->gpwrap) = false; + WRITE_ONCE(rdp->gpwrap, false); } return ret; } @@ -1717,9 +1762,9 @@ static void note_gp_changes(struct rcu_state *rsp, struct rcu_data *rdp) local_irq_save(flags); rnp = rdp->mynode; - if ((rdp->gpnum == ACCESS_ONCE(rnp->gpnum) && - rdp->completed == ACCESS_ONCE(rnp->completed) && - !unlikely(ACCESS_ONCE(rdp->gpwrap))) || /* w/out lock. */ + if ((rdp->gpnum == READ_ONCE(rnp->gpnum) && + rdp->completed == READ_ONCE(rnp->completed) && + !unlikely(READ_ONCE(rdp->gpwrap))) || /* w/out lock. */ !raw_spin_trylock(&rnp->lock)) { /* irqs already off, so later. */ local_irq_restore(flags); return; @@ -1731,6 +1776,13 @@ static void note_gp_changes(struct rcu_state *rsp, struct rcu_data *rdp) rcu_gp_kthread_wake(rsp); } +static void rcu_gp_slow(struct rcu_state *rsp, int delay) +{ + if (delay > 0 && + !(rsp->gpnum % (rcu_num_nodes * PER_RCU_NODE_PERIOD * delay))) + schedule_timeout_uninterruptible(delay); +} + /* * Initialize a new grace period. Return 0 if no grace period required. */ @@ -1740,15 +1792,15 @@ static int rcu_gp_init(struct rcu_state *rsp) struct rcu_data *rdp; struct rcu_node *rnp = rcu_get_root(rsp); - ACCESS_ONCE(rsp->gp_activity) = jiffies; + WRITE_ONCE(rsp->gp_activity, jiffies); raw_spin_lock_irq(&rnp->lock); smp_mb__after_unlock_lock(); - if (!ACCESS_ONCE(rsp->gp_flags)) { + if (!READ_ONCE(rsp->gp_flags)) { /* Spurious wakeup, tell caller to go back to sleep. */ raw_spin_unlock_irq(&rnp->lock); return 0; } - ACCESS_ONCE(rsp->gp_flags) = 0; /* Clear all flags: New grace period. */ + WRITE_ONCE(rsp->gp_flags, 0); /* Clear all flags: New grace period. */ if (WARN_ON_ONCE(rcu_gp_in_progress(rsp))) { /* @@ -1773,6 +1825,7 @@ static int rcu_gp_init(struct rcu_state *rsp) * will handle subsequent offline CPUs. */ rcu_for_each_leaf_node(rsp, rnp) { + rcu_gp_slow(rsp, gp_preinit_delay); raw_spin_lock_irq(&rnp->lock); smp_mb__after_unlock_lock(); if (rnp->qsmaskinit == rnp->qsmaskinitnext && @@ -1829,14 +1882,15 @@ static int rcu_gp_init(struct rcu_state *rsp) * process finishes, because this kthread handles both. */ rcu_for_each_node_breadth_first(rsp, rnp) { + rcu_gp_slow(rsp, gp_init_delay); raw_spin_lock_irq(&rnp->lock); smp_mb__after_unlock_lock(); rdp = this_cpu_ptr(rsp->rda); rcu_preempt_check_blocked_tasks(rnp); rnp->qsmask = rnp->qsmaskinit; - ACCESS_ONCE(rnp->gpnum) = rsp->gpnum; + WRITE_ONCE(rnp->gpnum, rsp->gpnum); if (WARN_ON_ONCE(rnp->completed != rsp->completed)) - ACCESS_ONCE(rnp->completed) = rsp->completed; + WRITE_ONCE(rnp->completed, rsp->completed); if (rnp == rdp->mynode) (void)__note_gp_changes(rsp, rnp, rdp); rcu_preempt_boost_start_gp(rnp); @@ -1845,10 +1899,7 @@ static int rcu_gp_init(struct rcu_state *rsp) rnp->grphi, rnp->qsmask); raw_spin_unlock_irq(&rnp->lock); cond_resched_rcu_qs(); - ACCESS_ONCE(rsp->gp_activity) = jiffies; - if (gp_init_delay > 0 && - !(rsp->gpnum % (rcu_num_nodes * PER_RCU_NODE_PERIOD))) - schedule_timeout_uninterruptible(gp_init_delay); + WRITE_ONCE(rsp->gp_activity, jiffies); } return 1; @@ -1864,7 +1915,7 @@ static int rcu_gp_fqs(struct rcu_state *rsp, int fqs_state_in) unsigned long maxj; struct rcu_node *rnp = rcu_get_root(rsp); - ACCESS_ONCE(rsp->gp_activity) = jiffies; + WRITE_ONCE(rsp->gp_activity, jiffies); rsp->n_force_qs++; if (fqs_state == RCU_SAVE_DYNTICK) { /* Collect dyntick-idle snapshots. */ @@ -1882,11 +1933,11 @@ static int rcu_gp_fqs(struct rcu_state *rsp, int fqs_state_in) force_qs_rnp(rsp, rcu_implicit_dynticks_qs, &isidle, &maxj); } /* Clear flag to prevent immediate re-entry. */ - if (ACCESS_ONCE(rsp->gp_flags) & RCU_GP_FLAG_FQS) { + if (READ_ONCE(rsp->gp_flags) & RCU_GP_FLAG_FQS) { raw_spin_lock_irq(&rnp->lock); smp_mb__after_unlock_lock(); - ACCESS_ONCE(rsp->gp_flags) = - ACCESS_ONCE(rsp->gp_flags) & ~RCU_GP_FLAG_FQS; + WRITE_ONCE(rsp->gp_flags, + READ_ONCE(rsp->gp_flags) & ~RCU_GP_FLAG_FQS); raw_spin_unlock_irq(&rnp->lock); } return fqs_state; @@ -1903,7 +1954,7 @@ static void rcu_gp_cleanup(struct rcu_state *rsp) struct rcu_data *rdp; struct rcu_node *rnp = rcu_get_root(rsp); - ACCESS_ONCE(rsp->gp_activity) = jiffies; + WRITE_ONCE(rsp->gp_activity, jiffies); raw_spin_lock_irq(&rnp->lock); smp_mb__after_unlock_lock(); gp_duration = jiffies - rsp->gp_start; @@ -1934,7 +1985,7 @@ static void rcu_gp_cleanup(struct rcu_state *rsp) smp_mb__after_unlock_lock(); WARN_ON_ONCE(rcu_preempt_blocked_readers_cgp(rnp)); WARN_ON_ONCE(rnp->qsmask); - ACCESS_ONCE(rnp->completed) = rsp->gpnum; + WRITE_ONCE(rnp->completed, rsp->gpnum); rdp = this_cpu_ptr(rsp->rda); if (rnp == rdp->mynode) needgp = __note_gp_changes(rsp, rnp, rdp) || needgp; @@ -1942,7 +1993,8 @@ static void rcu_gp_cleanup(struct rcu_state *rsp) nocb += rcu_future_gp_cleanup(rsp, rnp); raw_spin_unlock_irq(&rnp->lock); cond_resched_rcu_qs(); - ACCESS_ONCE(rsp->gp_activity) = jiffies; + WRITE_ONCE(rsp->gp_activity, jiffies); + rcu_gp_slow(rsp, gp_cleanup_delay); } rnp = rcu_get_root(rsp); raw_spin_lock_irq(&rnp->lock); @@ -1950,16 +2002,16 @@ static void rcu_gp_cleanup(struct rcu_state *rsp) rcu_nocb_gp_set(rnp, nocb); /* Declare grace period done. */ - ACCESS_ONCE(rsp->completed) = rsp->gpnum; + WRITE_ONCE(rsp->completed, rsp->gpnum); trace_rcu_grace_period(rsp->name, rsp->completed, TPS("end")); rsp->fqs_state = RCU_GP_IDLE; rdp = this_cpu_ptr(rsp->rda); /* Advance CBs to reduce false positives below. */ needgp = rcu_advance_cbs(rsp, rnp, rdp) || needgp; if (needgp || cpu_needs_another_gp(rsp, rdp)) { - ACCESS_ONCE(rsp->gp_flags) = RCU_GP_FLAG_INIT; + WRITE_ONCE(rsp->gp_flags, RCU_GP_FLAG_INIT); trace_rcu_grace_period(rsp->name, - ACCESS_ONCE(rsp->gpnum), + READ_ONCE(rsp->gpnum), TPS("newreq")); } raw_spin_unlock_irq(&rnp->lock); @@ -1983,20 +2035,20 @@ static int __noreturn rcu_gp_kthread(void *arg) /* Handle grace-period start. */ for (;;) { trace_rcu_grace_period(rsp->name, - ACCESS_ONCE(rsp->gpnum), + READ_ONCE(rsp->gpnum), TPS("reqwait")); rsp->gp_state = RCU_GP_WAIT_GPS; wait_event_interruptible(rsp->gp_wq, - ACCESS_ONCE(rsp->gp_flags) & + READ_ONCE(rsp->gp_flags) & RCU_GP_FLAG_INIT); /* Locking provides needed memory barrier. */ if (rcu_gp_init(rsp)) break; cond_resched_rcu_qs(); - ACCESS_ONCE(rsp->gp_activity) = jiffies; + WRITE_ONCE(rsp->gp_activity, jiffies); WARN_ON(signal_pending(current)); trace_rcu_grace_period(rsp->name, - ACCESS_ONCE(rsp->gpnum), + READ_ONCE(rsp->gpnum), TPS("reqwaitsig")); } @@ -2012,39 +2064,39 @@ static int __noreturn rcu_gp_kthread(void *arg) if (!ret) rsp->jiffies_force_qs = jiffies + j; trace_rcu_grace_period(rsp->name, - ACCESS_ONCE(rsp->gpnum), + READ_ONCE(rsp->gpnum), TPS("fqswait")); rsp->gp_state = RCU_GP_WAIT_FQS; ret = wait_event_interruptible_timeout(rsp->gp_wq, - ((gf = ACCESS_ONCE(rsp->gp_flags)) & + ((gf = READ_ONCE(rsp->gp_flags)) & RCU_GP_FLAG_FQS) || - (!ACCESS_ONCE(rnp->qsmask) && + (!READ_ONCE(rnp->qsmask) && !rcu_preempt_blocked_readers_cgp(rnp)), j); /* Locking provides needed memory barriers. */ /* If grace period done, leave loop. */ - if (!ACCESS_ONCE(rnp->qsmask) && + if (!READ_ONCE(rnp->qsmask) && !rcu_preempt_blocked_readers_cgp(rnp)) break; /* If time for quiescent-state forcing, do it. */ if (ULONG_CMP_GE(jiffies, rsp->jiffies_force_qs) || (gf & RCU_GP_FLAG_FQS)) { trace_rcu_grace_period(rsp->name, - ACCESS_ONCE(rsp->gpnum), + READ_ONCE(rsp->gpnum), TPS("fqsstart")); fqs_state = rcu_gp_fqs(rsp, fqs_state); trace_rcu_grace_period(rsp->name, - ACCESS_ONCE(rsp->gpnum), + READ_ONCE(rsp->gpnum), TPS("fqsend")); cond_resched_rcu_qs(); - ACCESS_ONCE(rsp->gp_activity) = jiffies; + WRITE_ONCE(rsp->gp_activity, jiffies); } else { /* Deal with stray signal. */ cond_resched_rcu_qs(); - ACCESS_ONCE(rsp->gp_activity) = jiffies; + WRITE_ONCE(rsp->gp_activity, jiffies); WARN_ON(signal_pending(current)); trace_rcu_grace_period(rsp->name, - ACCESS_ONCE(rsp->gpnum), + READ_ONCE(rsp->gpnum), TPS("fqswaitsig")); } j = jiffies_till_next_fqs; @@ -2086,8 +2138,8 @@ rcu_start_gp_advanced(struct rcu_state *rsp, struct rcu_node *rnp, */ return false; } - ACCESS_ONCE(rsp->gp_flags) = RCU_GP_FLAG_INIT; - trace_rcu_grace_period(rsp->name, ACCESS_ONCE(rsp->gpnum), + WRITE_ONCE(rsp->gp_flags, RCU_GP_FLAG_INIT); + trace_rcu_grace_period(rsp->name, READ_ONCE(rsp->gpnum), TPS("newreq")); /* @@ -2137,6 +2189,7 @@ static void rcu_report_qs_rsp(struct rcu_state *rsp, unsigned long flags) __releases(rcu_get_root(rsp)->lock) { WARN_ON_ONCE(!rcu_gp_in_progress(rsp)); + WRITE_ONCE(rsp->gp_flags, READ_ONCE(rsp->gp_flags) | RCU_GP_FLAG_FQS); raw_spin_unlock_irqrestore(&rcu_get_root(rsp)->lock, flags); rcu_gp_kthread_wake(rsp); } @@ -2334,8 +2387,6 @@ rcu_check_quiescent_state(struct rcu_state *rsp, struct rcu_data *rdp) rcu_report_qs_rdp(rdp->cpu, rsp, rdp); } -#ifdef CONFIG_HOTPLUG_CPU - /* * Send the specified CPU's RCU callbacks to the orphanage. The * specified CPU must be offline, and the caller must hold the @@ -2346,7 +2397,7 @@ rcu_send_cbs_to_orphanage(int cpu, struct rcu_state *rsp, struct rcu_node *rnp, struct rcu_data *rdp) { /* No-CBs CPUs do not have orphanable callbacks. */ - if (rcu_is_nocb_cpu(rdp->cpu)) + if (!IS_ENABLED(CONFIG_HOTPLUG_CPU) || rcu_is_nocb_cpu(rdp->cpu)) return; /* @@ -2359,7 +2410,7 @@ rcu_send_cbs_to_orphanage(int cpu, struct rcu_state *rsp, rsp->qlen += rdp->qlen; rdp->n_cbs_orphaned += rdp->qlen; rdp->qlen_lazy = 0; - ACCESS_ONCE(rdp->qlen) = 0; + WRITE_ONCE(rdp->qlen, 0); } /* @@ -2405,7 +2456,8 @@ static void rcu_adopt_orphan_cbs(struct rcu_state *rsp, unsigned long flags) struct rcu_data *rdp = raw_cpu_ptr(rsp->rda); /* No-CBs CPUs are handled specially. */ - if (rcu_nocb_adopt_orphan_cbs(rsp, rdp, flags)) + if (!IS_ENABLED(CONFIG_HOTPLUG_CPU) || + rcu_nocb_adopt_orphan_cbs(rsp, rdp, flags)) return; /* Do the accounting first. */ @@ -2452,6 +2504,9 @@ static void rcu_cleanup_dying_cpu(struct rcu_state *rsp) RCU_TRACE(struct rcu_data *rdp = this_cpu_ptr(rsp->rda)); RCU_TRACE(struct rcu_node *rnp = rdp->mynode); + if (!IS_ENABLED(CONFIG_HOTPLUG_CPU)) + return; + RCU_TRACE(mask = rdp->grpmask); trace_rcu_grace_period(rsp->name, rnp->gpnum + 1 - !!(rnp->qsmask & mask), @@ -2480,7 +2535,8 @@ static void rcu_cleanup_dead_rnp(struct rcu_node *rnp_leaf) long mask; struct rcu_node *rnp = rnp_leaf; - if (rnp->qsmaskinit || rcu_preempt_has_tasks(rnp)) + if (!IS_ENABLED(CONFIG_HOTPLUG_CPU) || + rnp->qsmaskinit || rcu_preempt_has_tasks(rnp)) return; for (;;) { mask = rnp->grpmask; @@ -2511,6 +2567,9 @@ static void rcu_cleanup_dying_idle_cpu(int cpu, struct rcu_state *rsp) struct rcu_data *rdp = per_cpu_ptr(rsp->rda, cpu); struct rcu_node *rnp = rdp->mynode; /* Outgoing CPU's rdp & rnp. */ + if (!IS_ENABLED(CONFIG_HOTPLUG_CPU)) + return; + /* Remove outgoing CPU from mask in the leaf rcu_node structure. */ mask = rdp->grpmask; raw_spin_lock_irqsave(&rnp->lock, flags); @@ -2532,6 +2591,9 @@ static void rcu_cleanup_dead_cpu(int cpu, struct rcu_state *rsp) struct rcu_data *rdp = per_cpu_ptr(rsp->rda, cpu); struct rcu_node *rnp = rdp->mynode; /* Outgoing CPU's rdp & rnp. */ + if (!IS_ENABLED(CONFIG_HOTPLUG_CPU)) + return; + /* Adjust any no-longer-needed kthreads. */ rcu_boost_kthread_setaffinity(rnp, -1); @@ -2546,26 +2608,6 @@ static void rcu_cleanup_dead_cpu(int cpu, struct rcu_state *rsp) cpu, rdp->qlen, rdp->nxtlist); } -#else /* #ifdef CONFIG_HOTPLUG_CPU */ - -static void rcu_cleanup_dying_cpu(struct rcu_state *rsp) -{ -} - -static void __maybe_unused rcu_cleanup_dead_rnp(struct rcu_node *rnp_leaf) -{ -} - -static void rcu_cleanup_dying_idle_cpu(int cpu, struct rcu_state *rsp) -{ -} - -static void rcu_cleanup_dead_cpu(int cpu, struct rcu_state *rsp) -{ -} - -#endif /* #else #ifdef CONFIG_HOTPLUG_CPU */ - /* * Invoke any RCU callbacks that have made it to the end of their grace * period. Thottle as specified by rdp->blimit. @@ -2580,7 +2622,7 @@ static void rcu_do_batch(struct rcu_state *rsp, struct rcu_data *rdp) /* If no callbacks are ready, just return. */ if (!cpu_has_callbacks_ready_to_invoke(rdp)) { trace_rcu_batch_start(rsp->name, rdp->qlen_lazy, rdp->qlen, 0); - trace_rcu_batch_end(rsp->name, 0, !!ACCESS_ONCE(rdp->nxtlist), + trace_rcu_batch_end(rsp->name, 0, !!READ_ONCE(rdp->nxtlist), need_resched(), is_idle_task(current), rcu_is_callbacks_kthread()); return; @@ -2636,7 +2678,7 @@ static void rcu_do_batch(struct rcu_state *rsp, struct rcu_data *rdp) } smp_mb(); /* List handling before counting for rcu_barrier(). */ rdp->qlen_lazy -= count_lazy; - ACCESS_ONCE(rdp->qlen) = rdp->qlen - count; + WRITE_ONCE(rdp->qlen, rdp->qlen - count); rdp->n_cbs_invoked += count; /* Reinstate batch limit if we have worked down the excess. */ @@ -2730,10 +2772,6 @@ static void force_qs_rnp(struct rcu_state *rsp, mask = 0; raw_spin_lock_irqsave(&rnp->lock, flags); smp_mb__after_unlock_lock(); - if (!rcu_gp_in_progress(rsp)) { - raw_spin_unlock_irqrestore(&rnp->lock, flags); - return; - } if (rnp->qsmask == 0) { if (rcu_state_p == &rcu_sched_state || rsp != rcu_state_p || @@ -2763,8 +2801,6 @@ static void force_qs_rnp(struct rcu_state *rsp, bit = 1; for (; cpu <= rnp->grphi; cpu++, bit <<= 1) { if ((rnp->qsmask & bit) != 0) { - if ((rnp->qsmaskinit & bit) == 0) - *isidle = false; /* Pending hotplug. */ if (f(per_cpu_ptr(rsp->rda, cpu), isidle, maxj)) mask |= bit; } @@ -2793,7 +2829,7 @@ static void force_quiescent_state(struct rcu_state *rsp) /* Funnel through hierarchy to reduce memory contention. */ rnp = __this_cpu_read(rsp->rda->mynode); for (; rnp != NULL; rnp = rnp->parent) { - ret = (ACCESS_ONCE(rsp->gp_flags) & RCU_GP_FLAG_FQS) || + ret = (READ_ONCE(rsp->gp_flags) & RCU_GP_FLAG_FQS) || !raw_spin_trylock(&rnp->fqslock); if (rnp_old != NULL) raw_spin_unlock(&rnp_old->fqslock); @@ -2809,13 +2845,12 @@ static void force_quiescent_state(struct rcu_state *rsp) raw_spin_lock_irqsave(&rnp_old->lock, flags); smp_mb__after_unlock_lock(); raw_spin_unlock(&rnp_old->fqslock); - if (ACCESS_ONCE(rsp->gp_flags) & RCU_GP_FLAG_FQS) { + if (READ_ONCE(rsp->gp_flags) & RCU_GP_FLAG_FQS) { rsp->n_force_qs_lh++; raw_spin_unlock_irqrestore(&rnp_old->lock, flags); return; /* Someone beat us to it. */ } - ACCESS_ONCE(rsp->gp_flags) = - ACCESS_ONCE(rsp->gp_flags) | RCU_GP_FLAG_FQS; + WRITE_ONCE(rsp->gp_flags, READ_ONCE(rsp->gp_flags) | RCU_GP_FLAG_FQS); raw_spin_unlock_irqrestore(&rnp_old->lock, flags); rcu_gp_kthread_wake(rsp); } @@ -2881,7 +2916,7 @@ static void rcu_process_callbacks(struct softirq_action *unused) */ static void invoke_rcu_callbacks(struct rcu_state *rsp, struct rcu_data *rdp) { - if (unlikely(!ACCESS_ONCE(rcu_scheduler_fully_active))) + if (unlikely(!READ_ONCE(rcu_scheduler_fully_active))) return; if (likely(!rsp->boost)) { rcu_do_batch(rsp, rdp); @@ -2972,7 +3007,7 @@ __call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu), WARN_ON_ONCE((unsigned long)head & 0x1); /* Misaligned rcu_head! */ if (debug_rcu_head_queue(head)) { /* Probable double call_rcu(), so leak the callback. */ - ACCESS_ONCE(head->func) = rcu_leak_callback; + WRITE_ONCE(head->func, rcu_leak_callback); WARN_ONCE(1, "__call_rcu(): Leaked duplicate callback\n"); return; } @@ -3011,7 +3046,7 @@ __call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu), if (!likely(rdp->nxtlist)) init_default_callback_list(rdp); } - ACCESS_ONCE(rdp->qlen) = rdp->qlen + 1; + WRITE_ONCE(rdp->qlen, rdp->qlen + 1); if (lazy) rdp->qlen_lazy++; else @@ -3287,7 +3322,7 @@ void synchronize_sched_expedited(void) if (ULONG_CMP_GE((ulong)atomic_long_read(&rsp->expedited_start), (ulong)atomic_long_read(&rsp->expedited_done) + ULONG_MAX / 8)) { - synchronize_sched(); + wait_rcu_gp(call_rcu_sched); atomic_long_inc(&rsp->expedited_wrap); return; } @@ -3450,14 +3485,14 @@ static int __rcu_pending(struct rcu_state *rsp, struct rcu_data *rdp) } /* Has another RCU grace period completed? */ - if (ACCESS_ONCE(rnp->completed) != rdp->completed) { /* outside lock */ + if (READ_ONCE(rnp->completed) != rdp->completed) { /* outside lock */ rdp->n_rp_gp_completed++; return 1; } /* Has a new RCU grace period started? */ - if (ACCESS_ONCE(rnp->gpnum) != rdp->gpnum || - unlikely(ACCESS_ONCE(rdp->gpwrap))) { /* outside lock */ + if (READ_ONCE(rnp->gpnum) != rdp->gpnum || + unlikely(READ_ONCE(rdp->gpwrap))) { /* outside lock */ rdp->n_rp_gp_started++; return 1; } @@ -3493,7 +3528,7 @@ static int rcu_pending(void) * non-NULL, store an indication of whether all callbacks are lazy. * (If there are no callbacks, all of them are deemed to be lazy.) */ -static int __maybe_unused rcu_cpu_has_callbacks(bool *all_lazy) +static bool __maybe_unused rcu_cpu_has_callbacks(bool *all_lazy) { bool al = true; bool hc = false; @@ -3564,7 +3599,7 @@ static void _rcu_barrier(struct rcu_state *rsp) { int cpu; struct rcu_data *rdp; - unsigned long snap = ACCESS_ONCE(rsp->n_barrier_done); + unsigned long snap = READ_ONCE(rsp->n_barrier_done); unsigned long snap_done; _rcu_barrier_trace(rsp, "Begin", -1, snap); @@ -3606,10 +3641,10 @@ static void _rcu_barrier(struct rcu_state *rsp) /* * Increment ->n_barrier_done to avoid duplicate work. Use - * ACCESS_ONCE() to prevent the compiler from speculating + * WRITE_ONCE() to prevent the compiler from speculating * the increment to precede the early-exit check. */ - ACCESS_ONCE(rsp->n_barrier_done) = rsp->n_barrier_done + 1; + WRITE_ONCE(rsp->n_barrier_done, rsp->n_barrier_done + 1); WARN_ON_ONCE((rsp->n_barrier_done & 0x1) != 1); _rcu_barrier_trace(rsp, "Inc1", -1, rsp->n_barrier_done); smp_mb(); /* Order ->n_barrier_done increment with below mechanism. */ @@ -3645,7 +3680,7 @@ static void _rcu_barrier(struct rcu_state *rsp) __call_rcu(&rdp->barrier_head, rcu_barrier_callback, rsp, cpu, 0); } - } else if (ACCESS_ONCE(rdp->qlen)) { + } else if (READ_ONCE(rdp->qlen)) { _rcu_barrier_trace(rsp, "OnlineQ", cpu, rsp->n_barrier_done); smp_call_function_single(cpu, rcu_barrier_func, rsp, 1); @@ -3665,7 +3700,7 @@ static void _rcu_barrier(struct rcu_state *rsp) /* Increment ->n_barrier_done to prevent duplicate work. */ smp_mb(); /* Keep increment after above mechanism. */ - ACCESS_ONCE(rsp->n_barrier_done) = rsp->n_barrier_done + 1; + WRITE_ONCE(rsp->n_barrier_done, rsp->n_barrier_done + 1); WARN_ON_ONCE((rsp->n_barrier_done & 0x1) != 0); _rcu_barrier_trace(rsp, "Inc2", -1, rsp->n_barrier_done); smp_mb(); /* Keep increment before caller's subsequent code. */ @@ -3780,7 +3815,7 @@ rcu_init_percpu_data(int cpu, struct rcu_state *rsp) rdp->gpnum = rnp->completed; /* Make CPU later note any new GP. */ rdp->completed = rnp->completed; rdp->passed_quiesce = false; - rdp->rcu_qs_ctr_snap = __this_cpu_read(rcu_qs_ctr); + rdp->rcu_qs_ctr_snap = per_cpu(rcu_qs_ctr, cpu); rdp->qs_pending = false; trace_rcu_grace_period(rsp->name, rdp->gpnum, TPS("cpuonl")); raw_spin_unlock_irqrestore(&rnp->lock, flags); @@ -3924,16 +3959,16 @@ void rcu_scheduler_starting(void) /* * Compute the per-level fanout, either using the exact fanout specified - * or balancing the tree, depending on CONFIG_RCU_FANOUT_EXACT. + * or balancing the tree, depending on the rcu_fanout_exact boot parameter. */ static void __init rcu_init_levelspread(struct rcu_state *rsp) { int i; - if (IS_ENABLED(CONFIG_RCU_FANOUT_EXACT)) { + if (rcu_fanout_exact) { rsp->levelspread[rcu_num_lvls - 1] = rcu_fanout_leaf; for (i = rcu_num_lvls - 2; i >= 0; i--) - rsp->levelspread[i] = CONFIG_RCU_FANOUT; + rsp->levelspread[i] = RCU_FANOUT; } else { int ccur; int cprv; @@ -3971,9 +4006,9 @@ static void __init rcu_init_one(struct rcu_state *rsp, BUILD_BUG_ON(MAX_RCU_LVLS > ARRAY_SIZE(buf)); /* Fix buf[] init! */ - /* Silence gcc 4.8 warning about array index out of range. */ - if (rcu_num_lvls > RCU_NUM_LVLS) - panic("rcu_init_one: rcu_num_lvls overflow"); + /* Silence gcc 4.8 false positive about array index out of range. */ + if (rcu_num_lvls <= 0 || rcu_num_lvls > RCU_NUM_LVLS) + panic("rcu_init_one: rcu_num_lvls out of range"); /* Initialize the level-tracking arrays. */ @@ -4059,7 +4094,7 @@ static void __init rcu_init_geometry(void) jiffies_till_next_fqs = d; /* If the compile-time values are accurate, just leave. */ - if (rcu_fanout_leaf == CONFIG_RCU_FANOUT_LEAF && + if (rcu_fanout_leaf == RCU_FANOUT_LEAF && nr_cpu_ids == NR_CPUS) return; pr_info("RCU: Adjusting geometry for rcu_fanout_leaf=%d, nr_cpu_ids=%d\n", @@ -4073,7 +4108,7 @@ static void __init rcu_init_geometry(void) rcu_capacity[0] = 1; rcu_capacity[1] = rcu_fanout_leaf; for (i = 2; i <= MAX_RCU_LVLS; i++) - rcu_capacity[i] = rcu_capacity[i - 1] * CONFIG_RCU_FANOUT; + rcu_capacity[i] = rcu_capacity[i - 1] * RCU_FANOUT; /* * The boot-time rcu_fanout_leaf parameter is only permitted @@ -4083,7 +4118,7 @@ static void __init rcu_init_geometry(void) * the configured number of CPUs. Complain and fall back to the * compile-time values if these limits are exceeded. */ - if (rcu_fanout_leaf < CONFIG_RCU_FANOUT_LEAF || + if (rcu_fanout_leaf < RCU_FANOUT_LEAF || rcu_fanout_leaf > sizeof(unsigned long) * 8 || n > rcu_capacity[MAX_RCU_LVLS]) { WARN_ON(1); @@ -4109,6 +4144,28 @@ static void __init rcu_init_geometry(void) rcu_num_nodes -= n; } +/* + * Dump out the structure of the rcu_node combining tree associated + * with the rcu_state structure referenced by rsp. + */ +static void __init rcu_dump_rcu_node_tree(struct rcu_state *rsp) +{ + int level = 0; + struct rcu_node *rnp; + + pr_info("rcu_node tree layout dump\n"); + pr_info(" "); + rcu_for_each_node_breadth_first(rsp, rnp) { + if (rnp->level != level) { + pr_cont("\n"); + pr_info(" "); + level = rnp->level; + } + pr_cont("%d:%d ^%d ", rnp->grplo, rnp->grphi, rnp->grpnum); + } + pr_cont("\n"); +} + void __init rcu_init(void) { int cpu; @@ -4119,6 +4176,8 @@ void __init rcu_init(void) rcu_init_geometry(); rcu_init_one(&rcu_bh_state, &rcu_bh_data); rcu_init_one(&rcu_sched_state, &rcu_sched_data); + if (dump_tree) + rcu_dump_rcu_node_tree(&rcu_sched_state); __rcu_init_preempt(); open_softirq(RCU_SOFTIRQ, rcu_process_callbacks); diff --git a/kernel/rcu/tree.h b/kernel/rcu/tree.h index a69d3da..4adb7ca 100644 --- a/kernel/rcu/tree.h +++ b/kernel/rcu/tree.h @@ -35,11 +35,33 @@ * In practice, this did work well going from three levels to four. * Of course, your mileage may vary. */ + #define MAX_RCU_LVLS 4 -#define RCU_FANOUT_1 (CONFIG_RCU_FANOUT_LEAF) -#define RCU_FANOUT_2 (RCU_FANOUT_1 * CONFIG_RCU_FANOUT) -#define RCU_FANOUT_3 (RCU_FANOUT_2 * CONFIG_RCU_FANOUT) -#define RCU_FANOUT_4 (RCU_FANOUT_3 * CONFIG_RCU_FANOUT) + +#ifdef CONFIG_RCU_FANOUT +#define RCU_FANOUT CONFIG_RCU_FANOUT +#else /* #ifdef CONFIG_RCU_FANOUT */ +# ifdef CONFIG_64BIT +# define RCU_FANOUT 64 +# else +# define RCU_FANOUT 32 +# endif +#endif /* #else #ifdef CONFIG_RCU_FANOUT */ + +#ifdef CONFIG_RCU_FANOUT_LEAF +#define RCU_FANOUT_LEAF CONFIG_RCU_FANOUT_LEAF +#else /* #ifdef CONFIG_RCU_FANOUT_LEAF */ +# ifdef CONFIG_64BIT +# define RCU_FANOUT_LEAF 64 +# else +# define RCU_FANOUT_LEAF 32 +# endif +#endif /* #else #ifdef CONFIG_RCU_FANOUT_LEAF */ + +#define RCU_FANOUT_1 (RCU_FANOUT_LEAF) +#define RCU_FANOUT_2 (RCU_FANOUT_1 * RCU_FANOUT) +#define RCU_FANOUT_3 (RCU_FANOUT_2 * RCU_FANOUT) +#define RCU_FANOUT_4 (RCU_FANOUT_3 * RCU_FANOUT) #if NR_CPUS <= RCU_FANOUT_1 # define RCU_NUM_LVLS 1 @@ -170,7 +192,6 @@ struct rcu_node { /* if there is no such task. If there */ /* is no current expedited grace period, */ /* then there can cannot be any such task. */ -#ifdef CONFIG_RCU_BOOST struct list_head *boost_tasks; /* Pointer to first task that needs to be */ /* priority boosted, or NULL if no priority */ @@ -208,7 +229,6 @@ struct rcu_node { unsigned long n_balk_nos; /* Refused to boost: not sure why, though. */ /* This can happen due to race conditions. */ -#endif /* #ifdef CONFIG_RCU_BOOST */ #ifdef CONFIG_RCU_NOCB_CPU wait_queue_head_t nocb_gp_wq[2]; /* Place for rcu_nocb_kthread() to wait GP. */ @@ -519,14 +539,11 @@ extern struct list_head rcu_struct_flavors; * RCU implementation internal declarations: */ extern struct rcu_state rcu_sched_state; -DECLARE_PER_CPU(struct rcu_data, rcu_sched_data); extern struct rcu_state rcu_bh_state; -DECLARE_PER_CPU(struct rcu_data, rcu_bh_data); #ifdef CONFIG_PREEMPT_RCU extern struct rcu_state rcu_preempt_state; -DECLARE_PER_CPU(struct rcu_data, rcu_preempt_data); #endif /* #ifdef CONFIG_PREEMPT_RCU */ #ifdef CONFIG_RCU_BOOST diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h index 8c0ec0f..013485f 100644 --- a/kernel/rcu/tree_plugin.h +++ b/kernel/rcu/tree_plugin.h @@ -43,7 +43,17 @@ DEFINE_PER_CPU(unsigned int, rcu_cpu_kthread_status); DEFINE_PER_CPU(unsigned int, rcu_cpu_kthread_loops); DEFINE_PER_CPU(char, rcu_cpu_has_work); -#endif /* #ifdef CONFIG_RCU_BOOST */ +#else /* #ifdef CONFIG_RCU_BOOST */ + +/* + * Some architectures do not define rt_mutexes, but if !CONFIG_RCU_BOOST, + * all uses are in dead code. Provide a definition to keep the compiler + * happy, but add WARN_ON_ONCE() to complain if used in the wrong place. + * This probably needs to be excluded from -rt builds. + */ +#define rt_mutex_owner(a) ({ WARN_ON_ONCE(1); NULL; }) + +#endif /* #else #ifdef CONFIG_RCU_BOOST */ #ifdef CONFIG_RCU_NOCB_CPU static cpumask_var_t rcu_nocb_mask; /* CPUs to have callbacks offloaded. */ @@ -60,11 +70,11 @@ static void __init rcu_bootup_announce_oddness(void) { if (IS_ENABLED(CONFIG_RCU_TRACE)) pr_info("\tRCU debugfs-based tracing is enabled.\n"); - if ((IS_ENABLED(CONFIG_64BIT) && CONFIG_RCU_FANOUT != 64) || - (!IS_ENABLED(CONFIG_64BIT) && CONFIG_RCU_FANOUT != 32)) + if ((IS_ENABLED(CONFIG_64BIT) && RCU_FANOUT != 64) || + (!IS_ENABLED(CONFIG_64BIT) && RCU_FANOUT != 32)) pr_info("\tCONFIG_RCU_FANOUT set to non-default value of %d\n", - CONFIG_RCU_FANOUT); - if (IS_ENABLED(CONFIG_RCU_FANOUT_EXACT)) + RCU_FANOUT); + if (rcu_fanout_exact) pr_info("\tHierarchical RCU autobalancing is disabled.\n"); if (IS_ENABLED(CONFIG_RCU_FAST_NO_HZ)) pr_info("\tRCU dyntick-idle grace-period acceleration is enabled.\n"); @@ -76,10 +86,10 @@ static void __init rcu_bootup_announce_oddness(void) pr_info("\tAdditional per-CPU info printed with stalls.\n"); if (NUM_RCU_LVL_4 != 0) pr_info("\tFour-level hierarchy is enabled.\n"); - if (CONFIG_RCU_FANOUT_LEAF != 16) + if (RCU_FANOUT_LEAF != 16) pr_info("\tBuild-time adjustment of leaf fanout to %d.\n", - CONFIG_RCU_FANOUT_LEAF); - if (rcu_fanout_leaf != CONFIG_RCU_FANOUT_LEAF) + RCU_FANOUT_LEAF); + if (rcu_fanout_leaf != RCU_FANOUT_LEAF) pr_info("\tBoot-time adjustment of leaf fanout to %d.\n", rcu_fanout_leaf); if (nr_cpu_ids != NR_CPUS) pr_info("\tRCU restricting CPUs from NR_CPUS=%d to nr_cpu_ids=%d.\n", NR_CPUS, nr_cpu_ids); @@ -90,7 +100,8 @@ static void __init rcu_bootup_announce_oddness(void) #ifdef CONFIG_PREEMPT_RCU RCU_STATE_INITIALIZER(rcu_preempt, 'p', call_rcu); -static struct rcu_state *rcu_state_p = &rcu_preempt_state; +static struct rcu_state *const rcu_state_p = &rcu_preempt_state; +static struct rcu_data __percpu *const rcu_data_p = &rcu_preempt_data; static int rcu_preempted_readers_exp(struct rcu_node *rnp); static void rcu_report_exp_rnp(struct rcu_state *rsp, struct rcu_node *rnp, @@ -116,11 +127,11 @@ static void __init rcu_bootup_announce(void) */ static void rcu_preempt_qs(void) { - if (!__this_cpu_read(rcu_preempt_data.passed_quiesce)) { + if (!__this_cpu_read(rcu_data_p->passed_quiesce)) { trace_rcu_grace_period(TPS("rcu_preempt"), - __this_cpu_read(rcu_preempt_data.gpnum), + __this_cpu_read(rcu_data_p->gpnum), TPS("cpuqs")); - __this_cpu_write(rcu_preempt_data.passed_quiesce, 1); + __this_cpu_write(rcu_data_p->passed_quiesce, 1); barrier(); /* Coordinate with rcu_preempt_check_callbacks(). */ current->rcu_read_unlock_special.b.need_qs = false; } @@ -150,7 +161,7 @@ static void rcu_preempt_note_context_switch(void) !t->rcu_read_unlock_special.b.blocked) { /* Possibly blocking in an RCU read-side critical section. */ - rdp = this_cpu_ptr(rcu_preempt_state.rda); + rdp = this_cpu_ptr(rcu_state_p->rda); rnp = rdp->mynode; raw_spin_lock_irqsave(&rnp->lock, flags); smp_mb__after_unlock_lock(); @@ -180,10 +191,9 @@ static void rcu_preempt_note_context_switch(void) if ((rnp->qsmask & rdp->grpmask) && rnp->gp_tasks != NULL) { list_add(&t->rcu_node_entry, rnp->gp_tasks->prev); rnp->gp_tasks = &t->rcu_node_entry; -#ifdef CONFIG_RCU_BOOST - if (rnp->boost_tasks != NULL) + if (IS_ENABLED(CONFIG_RCU_BOOST) && + rnp->boost_tasks != NULL) rnp->boost_tasks = rnp->gp_tasks; -#endif /* #ifdef CONFIG_RCU_BOOST */ } else { list_add(&t->rcu_node_entry, &rnp->blkd_tasks); if (rnp->qsmask & rdp->grpmask) @@ -263,9 +273,7 @@ void rcu_read_unlock_special(struct task_struct *t) bool empty_exp_now; unsigned long flags; struct list_head *np; -#ifdef CONFIG_RCU_BOOST bool drop_boost_mutex = false; -#endif /* #ifdef CONFIG_RCU_BOOST */ struct rcu_node *rnp; union rcu_special special; @@ -307,9 +315,11 @@ void rcu_read_unlock_special(struct task_struct *t) t->rcu_read_unlock_special.b.blocked = false; /* - * Remove this task from the list it blocked on. The - * task can migrate while we acquire the lock, but at - * most one time. So at most two passes through loop. + * Remove this task from the list it blocked on. The task + * now remains queued on the rcu_node corresponding to + * the CPU it first blocked on, so the first attempt to + * acquire the task's rcu_node's ->lock will succeed. + * Keep the loop and add a WARN_ON() out of sheer paranoia. */ for (;;) { rnp = t->rcu_blocked_node; @@ -317,6 +327,7 @@ void rcu_read_unlock_special(struct task_struct *t) smp_mb__after_unlock_lock(); if (rnp == t->rcu_blocked_node) break; + WARN_ON_ONCE(1); raw_spin_unlock(&rnp->lock); /* irqs remain disabled. */ } empty_norm = !rcu_preempt_blocked_readers_cgp(rnp); @@ -331,12 +342,12 @@ void rcu_read_unlock_special(struct task_struct *t) rnp->gp_tasks = np; if (&t->rcu_node_entry == rnp->exp_tasks) rnp->exp_tasks = np; -#ifdef CONFIG_RCU_BOOST - if (&t->rcu_node_entry == rnp->boost_tasks) - rnp->boost_tasks = np; - /* Snapshot ->boost_mtx ownership with rcu_node lock held. */ - drop_boost_mutex = rt_mutex_owner(&rnp->boost_mtx) == t; -#endif /* #ifdef CONFIG_RCU_BOOST */ + if (IS_ENABLED(CONFIG_RCU_BOOST)) { + if (&t->rcu_node_entry == rnp->boost_tasks) + rnp->boost_tasks = np; + /* Snapshot ->boost_mtx ownership w/rnp->lock held. */ + drop_boost_mutex = rt_mutex_owner(&rnp->boost_mtx) == t; + } /* * If this was the last task on the current list, and if @@ -353,24 +364,21 @@ void rcu_read_unlock_special(struct task_struct *t) rnp->grplo, rnp->grphi, !!rnp->gp_tasks); - rcu_report_unblock_qs_rnp(&rcu_preempt_state, - rnp, flags); + rcu_report_unblock_qs_rnp(rcu_state_p, rnp, flags); } else { raw_spin_unlock_irqrestore(&rnp->lock, flags); } -#ifdef CONFIG_RCU_BOOST /* Unboost if we were boosted. */ - if (drop_boost_mutex) + if (IS_ENABLED(CONFIG_RCU_BOOST) && drop_boost_mutex) rt_mutex_unlock(&rnp->boost_mtx); -#endif /* #ifdef CONFIG_RCU_BOOST */ /* * If this was the last task on the expedited lists, * then we need to report up the rcu_node hierarchy. */ if (!empty_exp && empty_exp_now) - rcu_report_exp_rnp(&rcu_preempt_state, rnp, true); + rcu_report_exp_rnp(rcu_state_p, rnp, true); } else { local_irq_restore(flags); } @@ -390,7 +398,7 @@ static void rcu_print_detail_task_stall_rnp(struct rcu_node *rnp) raw_spin_unlock_irqrestore(&rnp->lock, flags); return; } - t = list_entry(rnp->gp_tasks, + t = list_entry(rnp->gp_tasks->prev, struct task_struct, rcu_node_entry); list_for_each_entry_continue(t, &rnp->blkd_tasks, rcu_node_entry) sched_show_task(t); @@ -447,7 +455,7 @@ static int rcu_print_task_stall(struct rcu_node *rnp) if (!rcu_preempt_blocked_readers_cgp(rnp)) return 0; rcu_print_task_stall_begin(rnp); - t = list_entry(rnp->gp_tasks, + t = list_entry(rnp->gp_tasks->prev, struct task_struct, rcu_node_entry); list_for_each_entry_continue(t, &rnp->blkd_tasks, rcu_node_entry) { pr_cont(" P%d", t->pid); @@ -491,8 +499,8 @@ static void rcu_preempt_check_callbacks(void) return; } if (t->rcu_read_lock_nesting > 0 && - __this_cpu_read(rcu_preempt_data.qs_pending) && - !__this_cpu_read(rcu_preempt_data.passed_quiesce)) + __this_cpu_read(rcu_data_p->qs_pending) && + !__this_cpu_read(rcu_data_p->passed_quiesce)) t->rcu_read_unlock_special.b.need_qs = true; } @@ -500,7 +508,7 @@ static void rcu_preempt_check_callbacks(void) static void rcu_preempt_do_callbacks(void) { - rcu_do_batch(&rcu_preempt_state, this_cpu_ptr(&rcu_preempt_data)); + rcu_do_batch(rcu_state_p, this_cpu_ptr(rcu_data_p)); } #endif /* #ifdef CONFIG_RCU_BOOST */ @@ -510,7 +518,7 @@ static void rcu_preempt_do_callbacks(void) */ void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu)) { - __call_rcu(head, func, &rcu_preempt_state, -1, 0); + __call_rcu(head, func, rcu_state_p, -1, 0); } EXPORT_SYMBOL_GPL(call_rcu); @@ -570,7 +578,7 @@ static int rcu_preempted_readers_exp(struct rcu_node *rnp) static int sync_rcu_preempt_exp_done(struct rcu_node *rnp) { return !rcu_preempted_readers_exp(rnp) && - ACCESS_ONCE(rnp->expmask) == 0; + READ_ONCE(rnp->expmask) == 0; } /* @@ -711,12 +719,12 @@ sync_rcu_preempt_exp_init2(struct rcu_state *rsp, struct rcu_node *rnp) void synchronize_rcu_expedited(void) { struct rcu_node *rnp; - struct rcu_state *rsp = &rcu_preempt_state; + struct rcu_state *rsp = rcu_state_p; unsigned long snap; int trycount = 0; smp_mb(); /* Caller's modifications seen first by other CPUs. */ - snap = ACCESS_ONCE(sync_rcu_preempt_exp_count) + 1; + snap = READ_ONCE(sync_rcu_preempt_exp_count) + 1; smp_mb(); /* Above access cannot bleed into critical section. */ /* @@ -740,7 +748,7 @@ void synchronize_rcu_expedited(void) */ while (!mutex_trylock(&sync_rcu_preempt_exp_mutex)) { if (ULONG_CMP_LT(snap, - ACCESS_ONCE(sync_rcu_preempt_exp_count))) { + READ_ONCE(sync_rcu_preempt_exp_count))) { put_online_cpus(); goto mb_ret; /* Others did our work for us. */ } @@ -752,7 +760,7 @@ void synchronize_rcu_expedited(void) return; } } - if (ULONG_CMP_LT(snap, ACCESS_ONCE(sync_rcu_preempt_exp_count))) { + if (ULONG_CMP_LT(snap, READ_ONCE(sync_rcu_preempt_exp_count))) { put_online_cpus(); goto unlock_mb_ret; /* Others did our work for us. */ } @@ -780,8 +788,7 @@ void synchronize_rcu_expedited(void) /* Clean up and exit. */ smp_mb(); /* ensure expedited GP seen before counter increment. */ - ACCESS_ONCE(sync_rcu_preempt_exp_count) = - sync_rcu_preempt_exp_count + 1; + WRITE_ONCE(sync_rcu_preempt_exp_count, sync_rcu_preempt_exp_count + 1); unlock_mb_ret: mutex_unlock(&sync_rcu_preempt_exp_mutex); mb_ret: @@ -799,7 +806,7 @@ EXPORT_SYMBOL_GPL(synchronize_rcu_expedited); */ void rcu_barrier(void) { - _rcu_barrier(&rcu_preempt_state); + _rcu_barrier(rcu_state_p); } EXPORT_SYMBOL_GPL(rcu_barrier); @@ -808,7 +815,7 @@ EXPORT_SYMBOL_GPL(rcu_barrier); */ static void __init __rcu_init_preempt(void) { - rcu_init_one(&rcu_preempt_state, &rcu_preempt_data); + rcu_init_one(rcu_state_p, rcu_data_p); } /* @@ -831,7 +838,8 @@ void exit_rcu(void) #else /* #ifdef CONFIG_PREEMPT_RCU */ -static struct rcu_state *rcu_state_p = &rcu_sched_state; +static struct rcu_state *const rcu_state_p = &rcu_sched_state; +static struct rcu_data __percpu *const rcu_data_p = &rcu_sched_data; /* * Tell them what RCU they are running. @@ -994,8 +1002,8 @@ static int rcu_boost(struct rcu_node *rnp) struct task_struct *t; struct list_head *tb; - if (ACCESS_ONCE(rnp->exp_tasks) == NULL && - ACCESS_ONCE(rnp->boost_tasks) == NULL) + if (READ_ONCE(rnp->exp_tasks) == NULL && + READ_ONCE(rnp->boost_tasks) == NULL) return 0; /* Nothing left to boost. */ raw_spin_lock_irqsave(&rnp->lock, flags); @@ -1048,8 +1056,8 @@ static int rcu_boost(struct rcu_node *rnp) rt_mutex_lock(&rnp->boost_mtx); rt_mutex_unlock(&rnp->boost_mtx); /* Then keep lockdep happy. */ - return ACCESS_ONCE(rnp->exp_tasks) != NULL || - ACCESS_ONCE(rnp->boost_tasks) != NULL; + return READ_ONCE(rnp->exp_tasks) != NULL || + READ_ONCE(rnp->boost_tasks) != NULL; } /* @@ -1173,7 +1181,7 @@ static int rcu_spawn_one_boost_kthread(struct rcu_state *rsp, struct sched_param sp; struct task_struct *t; - if (&rcu_preempt_state != rsp) + if (rcu_state_p != rsp) return 0; if (!rcu_scheduler_fully_active || rcu_rnp_online_cpus(rnp) == 0) @@ -1367,13 +1375,12 @@ static void rcu_prepare_kthreads(int cpu) * Because we not have RCU_FAST_NO_HZ, just check whether this CPU needs * any flavor of RCU. */ -#ifndef CONFIG_RCU_NOCB_CPU_ALL -int rcu_needs_cpu(unsigned long *delta_jiffies) +int rcu_needs_cpu(u64 basemono, u64 *nextevt) { - *delta_jiffies = ULONG_MAX; - return rcu_cpu_has_callbacks(NULL); + *nextevt = KTIME_MAX; + return IS_ENABLED(CONFIG_RCU_NOCB_CPU_ALL) + ? 0 : rcu_cpu_has_callbacks(NULL); } -#endif /* #ifndef CONFIG_RCU_NOCB_CPU_ALL */ /* * Because we do not have RCU_FAST_NO_HZ, don't bother cleaning up @@ -1432,8 +1439,6 @@ module_param(rcu_idle_gp_delay, int, 0644); static int rcu_idle_lazy_gp_delay = RCU_IDLE_LAZY_GP_DELAY; module_param(rcu_idle_lazy_gp_delay, int, 0644); -extern int tick_nohz_active; - /* * Try to advance callbacks for all flavors of RCU on the current CPU, but * only if it has been awhile since the last time we did so. Afterwards, @@ -1462,7 +1467,7 @@ static bool __maybe_unused rcu_try_advance_all_cbs(void) * callbacks not yet ready to invoke. */ if ((rdp->completed != rnp->completed || - unlikely(ACCESS_ONCE(rdp->gpwrap))) && + unlikely(READ_ONCE(rdp->gpwrap))) && rdp->nxttail[RCU_DONE_TAIL] != rdp->nxttail[RCU_NEXT_TAIL]) note_gp_changes(rsp, rdp); @@ -1480,17 +1485,22 @@ static bool __maybe_unused rcu_try_advance_all_cbs(void) * * The caller must have disabled interrupts. */ -#ifndef CONFIG_RCU_NOCB_CPU_ALL -int rcu_needs_cpu(unsigned long *dj) +int rcu_needs_cpu(u64 basemono, u64 *nextevt) { struct rcu_dynticks *rdtp = this_cpu_ptr(&rcu_dynticks); + unsigned long dj; + + if (IS_ENABLED(CONFIG_RCU_NOCB_CPU_ALL)) { + *nextevt = KTIME_MAX; + return 0; + } /* Snapshot to detect later posting of non-lazy callback. */ rdtp->nonlazy_posted_snap = rdtp->nonlazy_posted; /* If no callbacks, RCU doesn't need the CPU. */ if (!rcu_cpu_has_callbacks(&rdtp->all_lazy)) { - *dj = ULONG_MAX; + *nextevt = KTIME_MAX; return 0; } @@ -1504,14 +1514,14 @@ int rcu_needs_cpu(unsigned long *dj) /* Request timer delay depending on laziness, and round. */ if (!rdtp->all_lazy) { - *dj = round_up(rcu_idle_gp_delay + jiffies, + dj = round_up(rcu_idle_gp_delay + jiffies, rcu_idle_gp_delay) - jiffies; } else { - *dj = round_jiffies(rcu_idle_lazy_gp_delay + jiffies) - jiffies; + dj = round_jiffies(rcu_idle_lazy_gp_delay + jiffies) - jiffies; } + *nextevt = basemono + dj * TICK_NSEC; return 0; } -#endif /* #ifndef CONFIG_RCU_NOCB_CPU_ALL */ /* * Prepare a CPU for idle from an RCU perspective. The first major task @@ -1525,7 +1535,6 @@ int rcu_needs_cpu(unsigned long *dj) */ static void rcu_prepare_for_idle(void) { -#ifndef CONFIG_RCU_NOCB_CPU_ALL bool needwake; struct rcu_data *rdp; struct rcu_dynticks *rdtp = this_cpu_ptr(&rcu_dynticks); @@ -1533,8 +1542,11 @@ static void rcu_prepare_for_idle(void) struct rcu_state *rsp; int tne; + if (IS_ENABLED(CONFIG_RCU_NOCB_CPU_ALL)) + return; + /* Handle nohz enablement switches conservatively. */ - tne = ACCESS_ONCE(tick_nohz_active); + tne = READ_ONCE(tick_nohz_active); if (tne != rdtp->tick_nohz_enabled_snap) { if (rcu_cpu_has_callbacks(NULL)) invoke_rcu_core(); /* force nohz to see update. */ @@ -1580,7 +1592,6 @@ static void rcu_prepare_for_idle(void) if (needwake) rcu_gp_kthread_wake(rsp); } -#endif /* #ifndef CONFIG_RCU_NOCB_CPU_ALL */ } /* @@ -1590,12 +1601,11 @@ static void rcu_prepare_for_idle(void) */ static void rcu_cleanup_after_idle(void) { -#ifndef CONFIG_RCU_NOCB_CPU_ALL - if (rcu_is_nocb_cpu(smp_processor_id())) + if (IS_ENABLED(CONFIG_RCU_NOCB_CPU_ALL) || + rcu_is_nocb_cpu(smp_processor_id())) return; if (rcu_try_advance_all_cbs()) invoke_rcu_core(); -#endif /* #ifndef CONFIG_RCU_NOCB_CPU_ALL */ } /* @@ -1760,7 +1770,7 @@ static void print_cpu_stall_info(struct rcu_state *rsp, int cpu) atomic_read(&rdtp->dynticks) & 0xfff, rdtp->dynticks_nesting, rdtp->dynticks_nmi_nesting, rdp->softirq_snap, kstat_softirqs_cpu(RCU_SOFTIRQ, cpu), - ACCESS_ONCE(rsp->n_force_qs) - rsp->n_force_qs_gpstart, + READ_ONCE(rsp->n_force_qs) - rsp->n_force_qs_gpstart, fast_no_hz); } @@ -1898,11 +1908,11 @@ static void wake_nocb_leader(struct rcu_data *rdp, bool force) { struct rcu_data *rdp_leader = rdp->nocb_leader; - if (!ACCESS_ONCE(rdp_leader->nocb_kthread)) + if (!READ_ONCE(rdp_leader->nocb_kthread)) return; - if (ACCESS_ONCE(rdp_leader->nocb_leader_sleep) || force) { + if (READ_ONCE(rdp_leader->nocb_leader_sleep) || force) { /* Prior smp_mb__after_atomic() orders against prior enqueue. */ - ACCESS_ONCE(rdp_leader->nocb_leader_sleep) = false; + WRITE_ONCE(rdp_leader->nocb_leader_sleep, false); wake_up(&rdp_leader->nocb_wq); } } @@ -1934,14 +1944,14 @@ static bool rcu_nocb_cpu_needs_barrier(struct rcu_state *rsp, int cpu) ret = atomic_long_read(&rdp->nocb_q_count); #ifdef CONFIG_PROVE_RCU - rhp = ACCESS_ONCE(rdp->nocb_head); + rhp = READ_ONCE(rdp->nocb_head); if (!rhp) - rhp = ACCESS_ONCE(rdp->nocb_gp_head); + rhp = READ_ONCE(rdp->nocb_gp_head); if (!rhp) - rhp = ACCESS_ONCE(rdp->nocb_follower_head); + rhp = READ_ONCE(rdp->nocb_follower_head); /* Having no rcuo kthread but CBs after scheduler starts is bad! */ - if (!ACCESS_ONCE(rdp->nocb_kthread) && rhp && + if (!READ_ONCE(rdp->nocb_kthread) && rhp && rcu_scheduler_fully_active) { /* RCU callback enqueued before CPU first came online??? */ pr_err("RCU: Never-onlined no-CBs CPU %d has CB %p\n", @@ -1975,12 +1985,12 @@ static void __call_rcu_nocb_enqueue(struct rcu_data *rdp, atomic_long_add(rhcount, &rdp->nocb_q_count); /* rcu_barrier() relies on ->nocb_q_count add before xchg. */ old_rhpp = xchg(&rdp->nocb_tail, rhtp); - ACCESS_ONCE(*old_rhpp) = rhp; + WRITE_ONCE(*old_rhpp, rhp); atomic_long_add(rhcount_lazy, &rdp->nocb_q_count_lazy); smp_mb__after_atomic(); /* Store *old_rhpp before _wake test. */ /* If we are not being polled and there is a kthread, awaken it ... */ - t = ACCESS_ONCE(rdp->nocb_kthread); + t = READ_ONCE(rdp->nocb_kthread); if (rcu_nocb_poll || !t) { trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu, TPS("WakeNotPoll")); @@ -2118,7 +2128,7 @@ static void rcu_nocb_wait_gp(struct rcu_data *rdp) for (;;) { wait_event_interruptible( rnp->nocb_gp_wq[c & 0x1], - (d = ULONG_CMP_GE(ACCESS_ONCE(rnp->completed), c))); + (d = ULONG_CMP_GE(READ_ONCE(rnp->completed), c))); if (likely(d)) break; WARN_ON(signal_pending(current)); @@ -2145,7 +2155,7 @@ wait_again: if (!rcu_nocb_poll) { trace_rcu_nocb_wake(my_rdp->rsp->name, my_rdp->cpu, "Sleep"); wait_event_interruptible(my_rdp->nocb_wq, - !ACCESS_ONCE(my_rdp->nocb_leader_sleep)); + !READ_ONCE(my_rdp->nocb_leader_sleep)); /* Memory barrier handled by smp_mb() calls below and repoll. */ } else if (firsttime) { firsttime = false; /* Don't drown trace log with "Poll"! */ @@ -2159,12 +2169,12 @@ wait_again: */ gotcbs = false; for (rdp = my_rdp; rdp; rdp = rdp->nocb_next_follower) { - rdp->nocb_gp_head = ACCESS_ONCE(rdp->nocb_head); + rdp->nocb_gp_head = READ_ONCE(rdp->nocb_head); if (!rdp->nocb_gp_head) continue; /* No CBs here, try next follower. */ /* Move callbacks to wait-for-GP list, which is empty. */ - ACCESS_ONCE(rdp->nocb_head) = NULL; + WRITE_ONCE(rdp->nocb_head, NULL); rdp->nocb_gp_tail = xchg(&rdp->nocb_tail, &rdp->nocb_head); gotcbs = true; } @@ -2184,7 +2194,7 @@ wait_again: my_rdp->nocb_leader_sleep = true; smp_mb(); /* Ensure _sleep true before scan. */ for (rdp = my_rdp; rdp; rdp = rdp->nocb_next_follower) - if (ACCESS_ONCE(rdp->nocb_head)) { + if (READ_ONCE(rdp->nocb_head)) { /* Found CB, so short-circuit next wait. */ my_rdp->nocb_leader_sleep = false; break; @@ -2205,7 +2215,7 @@ wait_again: /* Each pass through the following loop wakes a follower, if needed. */ for (rdp = my_rdp; rdp; rdp = rdp->nocb_next_follower) { - if (ACCESS_ONCE(rdp->nocb_head)) + if (READ_ONCE(rdp->nocb_head)) my_rdp->nocb_leader_sleep = false;/* No need to sleep.*/ if (!rdp->nocb_gp_head) continue; /* No CBs, so no need to wake follower. */ @@ -2241,7 +2251,7 @@ static void nocb_follower_wait(struct rcu_data *rdp) trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu, "FollowerSleep"); wait_event_interruptible(rdp->nocb_wq, - ACCESS_ONCE(rdp->nocb_follower_head)); + READ_ONCE(rdp->nocb_follower_head)); } else if (firsttime) { /* Don't drown trace log with "Poll"! */ firsttime = false; @@ -2282,10 +2292,10 @@ static int rcu_nocb_kthread(void *arg) nocb_follower_wait(rdp); /* Pull the ready-to-invoke callbacks onto local list. */ - list = ACCESS_ONCE(rdp->nocb_follower_head); + list = READ_ONCE(rdp->nocb_follower_head); BUG_ON(!list); trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu, "WokeNonEmpty"); - ACCESS_ONCE(rdp->nocb_follower_head) = NULL; + WRITE_ONCE(rdp->nocb_follower_head, NULL); tail = xchg(&rdp->nocb_follower_tail, &rdp->nocb_follower_head); /* Each pass through the following loop invokes a callback. */ @@ -2324,7 +2334,7 @@ static int rcu_nocb_kthread(void *arg) /* Is a deferred wakeup of rcu_nocb_kthread() required? */ static int rcu_nocb_need_deferred_wakeup(struct rcu_data *rdp) { - return ACCESS_ONCE(rdp->nocb_defer_wakeup); + return READ_ONCE(rdp->nocb_defer_wakeup); } /* Do a deferred wakeup of rcu_nocb_kthread(). */ @@ -2334,8 +2344,8 @@ static void do_nocb_deferred_wakeup(struct rcu_data *rdp) if (!rcu_nocb_need_deferred_wakeup(rdp)) return; - ndw = ACCESS_ONCE(rdp->nocb_defer_wakeup); - ACCESS_ONCE(rdp->nocb_defer_wakeup) = RCU_NOGP_WAKE_NOT; + ndw = READ_ONCE(rdp->nocb_defer_wakeup); + WRITE_ONCE(rdp->nocb_defer_wakeup, RCU_NOGP_WAKE_NOT); wake_nocb_leader(rdp, ndw == RCU_NOGP_WAKE_FORCE); trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu, TPS("DeferredWake")); } @@ -2448,7 +2458,7 @@ static void rcu_spawn_one_nocb_kthread(struct rcu_state *rsp, int cpu) t = kthread_run(rcu_nocb_kthread, rdp_spawn, "rcuo%c/%d", rsp->abbr, cpu); BUG_ON(IS_ERR(t)); - ACCESS_ONCE(rdp_spawn->nocb_kthread) = t; + WRITE_ONCE(rdp_spawn->nocb_kthread, t); } /* @@ -2663,7 +2673,7 @@ static void rcu_sysidle_enter(int irq) /* Record start of fully idle period. */ j = jiffies; - ACCESS_ONCE(rdtp->dynticks_idle_jiffies) = j; + WRITE_ONCE(rdtp->dynticks_idle_jiffies, j); smp_mb__before_atomic(); atomic_inc(&rdtp->dynticks_idle); smp_mb__after_atomic(); @@ -2681,7 +2691,7 @@ static void rcu_sysidle_enter(int irq) */ void rcu_sysidle_force_exit(void) { - int oldstate = ACCESS_ONCE(full_sysidle_state); + int oldstate = READ_ONCE(full_sysidle_state); int newoldstate; /* @@ -2794,7 +2804,7 @@ static void rcu_sysidle_check_cpu(struct rcu_data *rdp, bool *isidle, smp_mb(); /* Read counters before timestamps. */ /* Pick up timestamps. */ - j = ACCESS_ONCE(rdtp->dynticks_idle_jiffies); + j = READ_ONCE(rdtp->dynticks_idle_jiffies); /* If this CPU entered idle more recently, update maxj timestamp. */ if (ULONG_CMP_LT(*maxj, j)) *maxj = j; @@ -2831,11 +2841,11 @@ static unsigned long rcu_sysidle_delay(void) static void rcu_sysidle(unsigned long j) { /* Check the current state. */ - switch (ACCESS_ONCE(full_sysidle_state)) { + switch (READ_ONCE(full_sysidle_state)) { case RCU_SYSIDLE_NOT: /* First time all are idle, so note a short idle period. */ - ACCESS_ONCE(full_sysidle_state) = RCU_SYSIDLE_SHORT; + WRITE_ONCE(full_sysidle_state, RCU_SYSIDLE_SHORT); break; case RCU_SYSIDLE_SHORT: @@ -2873,7 +2883,7 @@ static void rcu_sysidle_cancel(void) { smp_mb(); if (full_sysidle_state > RCU_SYSIDLE_SHORT) - ACCESS_ONCE(full_sysidle_state) = RCU_SYSIDLE_NOT; + WRITE_ONCE(full_sysidle_state, RCU_SYSIDLE_NOT); } /* @@ -2925,7 +2935,7 @@ static void rcu_sysidle_cb(struct rcu_head *rhp) smp_mb(); /* grace period precedes setting inuse. */ rshp = container_of(rhp, struct rcu_sysidle_head, rh); - ACCESS_ONCE(rshp->inuse) = 0; + WRITE_ONCE(rshp->inuse, 0); } /* @@ -2936,7 +2946,7 @@ static void rcu_sysidle_cb(struct rcu_head *rhp) bool rcu_sys_is_idle(void) { static struct rcu_sysidle_head rsh; - int rss = ACCESS_ONCE(full_sysidle_state); + int rss = READ_ONCE(full_sysidle_state); if (WARN_ON_ONCE(smp_processor_id() != tick_do_timer_cpu)) return false; @@ -2964,7 +2974,7 @@ bool rcu_sys_is_idle(void) } rcu_sysidle_report(rcu_state_p, isidle, maxj, false); oldrss = rss; - rss = ACCESS_ONCE(full_sysidle_state); + rss = READ_ONCE(full_sysidle_state); } } @@ -3048,10 +3058,10 @@ static bool rcu_nohz_full_cpu(struct rcu_state *rsp) #ifdef CONFIG_NO_HZ_FULL if (tick_nohz_full_cpu(smp_processor_id()) && (!rcu_gp_in_progress(rsp) || - ULONG_CMP_LT(jiffies, ACCESS_ONCE(rsp->gp_start) + HZ))) - return 1; + ULONG_CMP_LT(jiffies, READ_ONCE(rsp->gp_start) + HZ))) + return true; #endif /* #ifdef CONFIG_NO_HZ_FULL */ - return 0; + return false; } /* @@ -3077,7 +3087,7 @@ static void rcu_bind_gp_kthread(void) static void rcu_dynticks_task_enter(void) { #if defined(CONFIG_TASKS_RCU) && defined(CONFIG_NO_HZ_FULL) - ACCESS_ONCE(current->rcu_tasks_idle_cpu) = smp_processor_id(); + WRITE_ONCE(current->rcu_tasks_idle_cpu, smp_processor_id()); #endif /* #if defined(CONFIG_TASKS_RCU) && defined(CONFIG_NO_HZ_FULL) */ } @@ -3085,6 +3095,6 @@ static void rcu_dynticks_task_enter(void) static void rcu_dynticks_task_exit(void) { #if defined(CONFIG_TASKS_RCU) && defined(CONFIG_NO_HZ_FULL) - ACCESS_ONCE(current->rcu_tasks_idle_cpu) = -1; + WRITE_ONCE(current->rcu_tasks_idle_cpu, -1); #endif /* #if defined(CONFIG_TASKS_RCU) && defined(CONFIG_NO_HZ_FULL) */ } diff --git a/kernel/rcu/tree_trace.c b/kernel/rcu/tree_trace.c index f92361e..3ea7ffc 100644 --- a/kernel/rcu/tree_trace.c +++ b/kernel/rcu/tree_trace.c @@ -277,7 +277,7 @@ static void print_one_rcu_state(struct seq_file *m, struct rcu_state *rsp) seq_printf(m, "nfqs=%lu/nfqsng=%lu(%lu) fqlh=%lu oqlen=%ld/%ld\n", rsp->n_force_qs, rsp->n_force_qs_ngp, rsp->n_force_qs - rsp->n_force_qs_ngp, - ACCESS_ONCE(rsp->n_force_qs_lh), rsp->qlen_lazy, rsp->qlen); + READ_ONCE(rsp->n_force_qs_lh), rsp->qlen_lazy, rsp->qlen); for (rnp = &rsp->node[0]; rnp - &rsp->node[0] < rcu_num_nodes; rnp++) { if (rnp->level != level) { seq_puts(m, "\n"); @@ -323,8 +323,8 @@ static void show_one_rcugp(struct seq_file *m, struct rcu_state *rsp) struct rcu_node *rnp = &rsp->node[0]; raw_spin_lock_irqsave(&rnp->lock, flags); - completed = ACCESS_ONCE(rsp->completed); - gpnum = ACCESS_ONCE(rsp->gpnum); + completed = READ_ONCE(rsp->completed); + gpnum = READ_ONCE(rsp->gpnum); if (completed == gpnum) gpage = 0; else diff --git a/kernel/rcu/update.c b/kernel/rcu/update.c index 1f13335..afaecb7 100644 --- a/kernel/rcu/update.c +++ b/kernel/rcu/update.c @@ -150,14 +150,14 @@ void __rcu_read_unlock(void) barrier(); /* critical section before exit code. */ t->rcu_read_lock_nesting = INT_MIN; barrier(); /* assign before ->rcu_read_unlock_special load */ - if (unlikely(ACCESS_ONCE(t->rcu_read_unlock_special.s))) + if (unlikely(READ_ONCE(t->rcu_read_unlock_special.s))) rcu_read_unlock_special(t); barrier(); /* ->rcu_read_unlock_special load before assign */ t->rcu_read_lock_nesting = 0; } #ifdef CONFIG_PROVE_LOCKING { - int rrln = ACCESS_ONCE(t->rcu_read_lock_nesting); + int rrln = READ_ONCE(t->rcu_read_lock_nesting); WARN_ON_ONCE(rrln < 0 && rrln > INT_MIN / 2); } @@ -389,17 +389,17 @@ module_param(rcu_cpu_stall_timeout, int, 0644); int rcu_jiffies_till_stall_check(void) { - int till_stall_check = ACCESS_ONCE(rcu_cpu_stall_timeout); + int till_stall_check = READ_ONCE(rcu_cpu_stall_timeout); /* * Limit check must be consistent with the Kconfig limits * for CONFIG_RCU_CPU_STALL_TIMEOUT. */ if (till_stall_check < 3) { - ACCESS_ONCE(rcu_cpu_stall_timeout) = 3; + WRITE_ONCE(rcu_cpu_stall_timeout, 3); till_stall_check = 3; } else if (till_stall_check > 300) { - ACCESS_ONCE(rcu_cpu_stall_timeout) = 300; + WRITE_ONCE(rcu_cpu_stall_timeout, 300); till_stall_check = 300; } return till_stall_check * HZ + RCU_STALL_DELAY_DELTA; @@ -550,12 +550,12 @@ static void check_holdout_task(struct task_struct *t, { int cpu; - if (!ACCESS_ONCE(t->rcu_tasks_holdout) || - t->rcu_tasks_nvcsw != ACCESS_ONCE(t->nvcsw) || - !ACCESS_ONCE(t->on_rq) || + if (!READ_ONCE(t->rcu_tasks_holdout) || + t->rcu_tasks_nvcsw != READ_ONCE(t->nvcsw) || + !READ_ONCE(t->on_rq) || (IS_ENABLED(CONFIG_NO_HZ_FULL) && !is_idle_task(t) && t->rcu_tasks_idle_cpu >= 0)) { - ACCESS_ONCE(t->rcu_tasks_holdout) = false; + WRITE_ONCE(t->rcu_tasks_holdout, false); list_del_init(&t->rcu_tasks_holdout_list); put_task_struct(t); return; @@ -639,11 +639,11 @@ static int __noreturn rcu_tasks_kthread(void *arg) */ rcu_read_lock(); for_each_process_thread(g, t) { - if (t != current && ACCESS_ONCE(t->on_rq) && + if (t != current && READ_ONCE(t->on_rq) && !is_idle_task(t)) { get_task_struct(t); - t->rcu_tasks_nvcsw = ACCESS_ONCE(t->nvcsw); - ACCESS_ONCE(t->rcu_tasks_holdout) = true; + t->rcu_tasks_nvcsw = READ_ONCE(t->nvcsw); + WRITE_ONCE(t->rcu_tasks_holdout, true); list_add(&t->rcu_tasks_holdout_list, &rcu_tasks_holdouts); } @@ -672,7 +672,7 @@ static int __noreturn rcu_tasks_kthread(void *arg) struct task_struct *t1; schedule_timeout_interruptible(HZ); - rtst = ACCESS_ONCE(rcu_task_stall_timeout); + rtst = READ_ONCE(rcu_task_stall_timeout); needreport = rtst > 0 && time_after(jiffies, lastreport + rtst); if (needreport) @@ -728,7 +728,7 @@ static void rcu_spawn_tasks_kthread(void) static struct task_struct *rcu_tasks_kthread_ptr; struct task_struct *t; - if (ACCESS_ONCE(rcu_tasks_kthread_ptr)) { + if (READ_ONCE(rcu_tasks_kthread_ptr)) { smp_mb(); /* Ensure caller sees full kthread. */ return; } @@ -740,7 +740,7 @@ static void rcu_spawn_tasks_kthread(void) t = kthread_run(rcu_tasks_kthread, NULL, "rcu_tasks_kthread"); BUG_ON(IS_ERR(t)); smp_mb(); /* Ensure others see full kthread. */ - ACCESS_ONCE(rcu_tasks_kthread_ptr) = t; + WRITE_ONCE(rcu_tasks_kthread_ptr, t); mutex_unlock(&rcu_tasks_kthread_mutex); } diff --git a/kernel/sched/Makefile b/kernel/sched/Makefile index 46be870..6768797 100644 --- a/kernel/sched/Makefile +++ b/kernel/sched/Makefile @@ -11,7 +11,7 @@ ifneq ($(CONFIG_SCHED_OMIT_FRAME_POINTER),y) CFLAGS_core.o := $(PROFILING) -fno-omit-frame-pointer endif -obj-y += core.o proc.o clock.o cputime.o +obj-y += core.o loadavg.o clock.o cputime.o obj-y += idle_task.o fair.o rt.o deadline.o stop_task.o obj-y += wait.o completion.o idle.o obj-$(CONFIG_SMP) += cpupri.o cpudeadline.o diff --git a/kernel/sched/auto_group.c b/kernel/sched/auto_group.c index eae160d..750ed60 100644 --- a/kernel/sched/auto_group.c +++ b/kernel/sched/auto_group.c @@ -1,5 +1,3 @@ -#ifdef CONFIG_SCHED_AUTOGROUP - #include "sched.h" #include <linux/proc_fs.h> @@ -141,7 +139,7 @@ autogroup_move_group(struct task_struct *p, struct autogroup *ag) p->signal->autogroup = autogroup_kref_get(ag); - if (!ACCESS_ONCE(sysctl_sched_autogroup_enabled)) + if (!READ_ONCE(sysctl_sched_autogroup_enabled)) goto out; for_each_thread(p, t) @@ -249,5 +247,3 @@ int autogroup_path(struct task_group *tg, char *buf, int buflen) return snprintf(buf, buflen, "%s-%ld", "/autogroup", tg->autogroup->id); } #endif /* CONFIG_SCHED_DEBUG */ - -#endif /* CONFIG_SCHED_AUTOGROUP */ diff --git a/kernel/sched/auto_group.h b/kernel/sched/auto_group.h index 8bd0471..890c95f 100644 --- a/kernel/sched/auto_group.h +++ b/kernel/sched/auto_group.h @@ -29,7 +29,7 @@ extern bool task_wants_autogroup(struct task_struct *p, struct task_group *tg); static inline struct task_group * autogroup_task_group(struct task_struct *p, struct task_group *tg) { - int enabled = ACCESS_ONCE(sysctl_sched_autogroup_enabled); + int enabled = READ_ONCE(sysctl_sched_autogroup_enabled); if (enabled && task_wants_autogroup(p, tg)) return p->signal->autogroup->tg; diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 1236732..c86935a 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -90,26 +90,6 @@ #define CREATE_TRACE_POINTS #include <trace/events/sched.h> -void start_bandwidth_timer(struct hrtimer *period_timer, ktime_t period) -{ - unsigned long delta; - ktime_t soft, hard, now; - - for (;;) { - if (hrtimer_active(period_timer)) - break; - - now = hrtimer_cb_get_time(period_timer); - hrtimer_forward(period_timer, now, period); - - soft = hrtimer_get_softexpires(period_timer); - hard = hrtimer_get_expires(period_timer); - delta = ktime_to_ns(ktime_sub(hard, soft)); - __hrtimer_start_range_ns(period_timer, soft, delta, - HRTIMER_MODE_ABS_PINNED, 0); - } -} - DEFINE_MUTEX(sched_domains_mutex); DEFINE_PER_CPU_SHARED_ALIGNED(struct rq, runqueues); @@ -355,12 +335,11 @@ static enum hrtimer_restart hrtick(struct hrtimer *timer) #ifdef CONFIG_SMP -static int __hrtick_restart(struct rq *rq) +static void __hrtick_restart(struct rq *rq) { struct hrtimer *timer = &rq->hrtick_timer; - ktime_t time = hrtimer_get_softexpires(timer); - return __hrtimer_start_range_ns(timer, time, 0, HRTIMER_MODE_ABS_PINNED, 0); + hrtimer_start_expires(timer, HRTIMER_MODE_ABS_PINNED); } /* @@ -440,8 +419,8 @@ void hrtick_start(struct rq *rq, u64 delay) * doesn't make sense. Rely on vruntime for fairness. */ delay = max_t(u64, delay, 10000LL); - __hrtimer_start_range_ns(&rq->hrtick_timer, ns_to_ktime(delay), 0, - HRTIMER_MODE_REL_PINNED, 0); + hrtimer_start(&rq->hrtick_timer, ns_to_ktime(delay), + HRTIMER_MODE_REL_PINNED); } static inline void init_hrtick(void) @@ -511,7 +490,7 @@ static bool set_nr_and_not_polling(struct task_struct *p) static bool set_nr_if_polling(struct task_struct *p) { struct thread_info *ti = task_thread_info(p); - typeof(ti->flags) old, val = ACCESS_ONCE(ti->flags); + typeof(ti->flags) old, val = READ_ONCE(ti->flags); for (;;) { if (!(val & _TIF_POLLING_NRFLAG)) @@ -541,6 +520,52 @@ static bool set_nr_if_polling(struct task_struct *p) #endif #endif +void wake_q_add(struct wake_q_head *head, struct task_struct *task) +{ + struct wake_q_node *node = &task->wake_q; + + /* + * Atomically grab the task, if ->wake_q is !nil already it means + * its already queued (either by us or someone else) and will get the + * wakeup due to that. + * + * This cmpxchg() implies a full barrier, which pairs with the write + * barrier implied by the wakeup in wake_up_list(). + */ + if (cmpxchg(&node->next, NULL, WAKE_Q_TAIL)) + return; + + get_task_struct(task); + + /* + * The head is context local, there can be no concurrency. + */ + *head->lastp = node; + head->lastp = &node->next; +} + +void wake_up_q(struct wake_q_head *head) +{ + struct wake_q_node *node = head->first; + + while (node != WAKE_Q_TAIL) { + struct task_struct *task; + + task = container_of(node, struct task_struct, wake_q); + BUG_ON(!task); + /* task can safely be re-inserted now */ + node = node->next; + task->wake_q.next = NULL; + + /* + * wake_up_process() implies a wmb() to pair with the queueing + * in wake_q_add() so as not to miss wakeups. + */ + wake_up_process(task); + put_task_struct(task); + } +} + /* * resched_curr - mark rq's current task 'to be rescheduled now'. * @@ -593,13 +618,12 @@ void resched_cpu(int cpu) * selecting an idle cpu will add more delays to the timers than intended * (as that cpu's timer base may not be uptodate wrt jiffies etc). */ -int get_nohz_timer_target(int pinned) +int get_nohz_timer_target(void) { - int cpu = smp_processor_id(); - int i; + int i, cpu = smp_processor_id(); struct sched_domain *sd; - if (pinned || !get_sysctl_timer_migration() || !idle_cpu(cpu)) + if (!idle_cpu(cpu)) return cpu; rcu_read_lock(); @@ -1049,7 +1073,7 @@ void set_task_cpu(struct task_struct *p, unsigned int new_cpu) 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); + perf_event_task_migrate(p); } __set_task_cpu(p, new_cpu); @@ -2105,12 +2129,15 @@ void wake_up_new_task(struct task_struct *p) #ifdef CONFIG_PREEMPT_NOTIFIERS +static struct static_key preempt_notifier_key = STATIC_KEY_INIT_FALSE; + /** * preempt_notifier_register - tell me when current is being preempted & rescheduled * @notifier: notifier struct to register */ void preempt_notifier_register(struct preempt_notifier *notifier) { + static_key_slow_inc(&preempt_notifier_key); hlist_add_head(¬ifier->link, ¤t->preempt_notifiers); } EXPORT_SYMBOL_GPL(preempt_notifier_register); @@ -2119,15 +2146,16 @@ EXPORT_SYMBOL_GPL(preempt_notifier_register); * preempt_notifier_unregister - no longer interested in preemption notifications * @notifier: notifier struct to unregister * - * This is safe to call from within a preemption notifier. + * This is *not* safe to call from within a preemption notifier. */ void preempt_notifier_unregister(struct preempt_notifier *notifier) { hlist_del(¬ifier->link); + static_key_slow_dec(&preempt_notifier_key); } EXPORT_SYMBOL_GPL(preempt_notifier_unregister); -static void fire_sched_in_preempt_notifiers(struct task_struct *curr) +static void __fire_sched_in_preempt_notifiers(struct task_struct *curr) { struct preempt_notifier *notifier; @@ -2135,9 +2163,15 @@ static void fire_sched_in_preempt_notifiers(struct task_struct *curr) notifier->ops->sched_in(notifier, raw_smp_processor_id()); } +static __always_inline void fire_sched_in_preempt_notifiers(struct task_struct *curr) +{ + if (static_key_false(&preempt_notifier_key)) + __fire_sched_in_preempt_notifiers(curr); +} + static void -fire_sched_out_preempt_notifiers(struct task_struct *curr, - struct task_struct *next) +__fire_sched_out_preempt_notifiers(struct task_struct *curr, + struct task_struct *next) { struct preempt_notifier *notifier; @@ -2145,13 +2179,21 @@ fire_sched_out_preempt_notifiers(struct task_struct *curr, notifier->ops->sched_out(notifier, next); } +static __always_inline void +fire_sched_out_preempt_notifiers(struct task_struct *curr, + struct task_struct *next) +{ + if (static_key_false(&preempt_notifier_key)) + __fire_sched_out_preempt_notifiers(curr, next); +} + #else /* !CONFIG_PREEMPT_NOTIFIERS */ -static void fire_sched_in_preempt_notifiers(struct task_struct *curr) +static inline void fire_sched_in_preempt_notifiers(struct task_struct *curr) { } -static void +static inline void fire_sched_out_preempt_notifiers(struct task_struct *curr, struct task_struct *next) { @@ -2332,7 +2374,6 @@ context_switch(struct rq *rq, struct task_struct *prev, */ spin_release(&rq->lock.dep_map, 1, _THIS_IP_); - context_tracking_task_switch(prev, next); /* Here we just switch the register state and the stack. */ switch_to(prev, next, prev); barrier(); @@ -2397,9 +2438,9 @@ unsigned long nr_iowait_cpu(int cpu) void get_iowait_load(unsigned long *nr_waiters, unsigned long *load) { - struct rq *this = this_rq(); - *nr_waiters = atomic_read(&this->nr_iowait); - *load = this->cpu_load[0]; + struct rq *rq = this_rq(); + *nr_waiters = atomic_read(&rq->nr_iowait); + *load = rq->load.weight; } #ifdef CONFIG_SMP @@ -2497,6 +2538,7 @@ void scheduler_tick(void) update_rq_clock(rq); curr->sched_class->task_tick(rq, curr, 0); update_cpu_load_active(rq); + calc_global_load_tick(rq); raw_spin_unlock(&rq->lock); perf_event_task_tick(); @@ -2525,7 +2567,7 @@ void scheduler_tick(void) u64 scheduler_tick_max_deferment(void) { struct rq *rq = this_rq(); - unsigned long next, now = ACCESS_ONCE(jiffies); + unsigned long next, now = READ_ONCE(jiffies); next = rq->last_sched_tick + HZ; @@ -2726,9 +2768,7 @@ again: * - return from syscall or exception to user-space * - return from interrupt-handler to user-space * - * WARNING: all callers must re-check need_resched() afterward and reschedule - * accordingly in case an event triggered the need for rescheduling (such as - * an interrupt waking up a task) while preemption was disabled in __schedule(). + * WARNING: must be called with preemption disabled! */ static void __sched __schedule(void) { @@ -2737,7 +2777,6 @@ static void __sched __schedule(void) struct rq *rq; int cpu; - preempt_disable(); cpu = smp_processor_id(); rq = cpu_rq(cpu); rcu_note_context_switch(); @@ -2801,8 +2840,6 @@ static void __sched __schedule(void) raw_spin_unlock_irq(&rq->lock); post_schedule(rq); - - sched_preempt_enable_no_resched(); } static inline void sched_submit_work(struct task_struct *tsk) @@ -2823,7 +2860,9 @@ asmlinkage __visible void __sched schedule(void) sched_submit_work(tsk); do { + preempt_disable(); __schedule(); + sched_preempt_enable_no_resched(); } while (need_resched()); } EXPORT_SYMBOL(schedule); @@ -2862,15 +2901,14 @@ void __sched schedule_preempt_disabled(void) static void __sched notrace preempt_schedule_common(void) { do { - __preempt_count_add(PREEMPT_ACTIVE); + preempt_active_enter(); __schedule(); - __preempt_count_sub(PREEMPT_ACTIVE); + preempt_active_exit(); /* * Check again in case we missed a preemption opportunity * between schedule and now. */ - barrier(); } while (need_resched()); } @@ -2894,9 +2932,8 @@ asmlinkage __visible void __sched notrace preempt_schedule(void) NOKPROBE_SYMBOL(preempt_schedule); EXPORT_SYMBOL(preempt_schedule); -#ifdef CONFIG_CONTEXT_TRACKING /** - * preempt_schedule_context - preempt_schedule called by tracing + * preempt_schedule_notrace - preempt_schedule called by tracing * * The tracing infrastructure uses preempt_enable_notrace to prevent * recursion and tracing preempt enabling caused by the tracing @@ -2909,7 +2946,7 @@ EXPORT_SYMBOL(preempt_schedule); * instead of preempt_schedule() to exit user context if needed before * calling the scheduler. */ -asmlinkage __visible void __sched notrace preempt_schedule_context(void) +asmlinkage __visible void __sched notrace preempt_schedule_notrace(void) { enum ctx_state prev_ctx; @@ -2917,7 +2954,13 @@ asmlinkage __visible void __sched notrace preempt_schedule_context(void) return; do { - __preempt_count_add(PREEMPT_ACTIVE); + /* + * Use raw __prempt_count() ops that don't call function. + * We can't call functions before disabling preemption which + * disarm preemption tracing recursions. + */ + __preempt_count_add(PREEMPT_ACTIVE + PREEMPT_DISABLE_OFFSET); + barrier(); /* * Needs preempt disabled in case user_exit() is traced * and the tracer calls preempt_enable_notrace() causing @@ -2927,12 +2970,11 @@ asmlinkage __visible void __sched notrace preempt_schedule_context(void) __schedule(); exception_exit(prev_ctx); - __preempt_count_sub(PREEMPT_ACTIVE); barrier(); + __preempt_count_sub(PREEMPT_ACTIVE + PREEMPT_DISABLE_OFFSET); } while (need_resched()); } -EXPORT_SYMBOL_GPL(preempt_schedule_context); -#endif /* CONFIG_CONTEXT_TRACKING */ +EXPORT_SYMBOL_GPL(preempt_schedule_notrace); #endif /* CONFIG_PREEMPT */ @@ -2952,17 +2994,11 @@ asmlinkage __visible void __sched preempt_schedule_irq(void) prev_state = exception_enter(); do { - __preempt_count_add(PREEMPT_ACTIVE); + preempt_active_enter(); local_irq_enable(); __schedule(); local_irq_disable(); - __preempt_count_sub(PREEMPT_ACTIVE); - - /* - * Check again in case we missed a preemption opportunity - * between schedule and now. - */ - barrier(); + preempt_active_exit(); } while (need_resched()); exception_exit(prev_state); @@ -3040,7 +3076,6 @@ void rt_mutex_setprio(struct task_struct *p, int prio) if (!dl_prio(p->normal_prio) || (pi_task && dl_entity_preempt(&pi_task->dl, &p->dl))) { p->dl.dl_boosted = 1; - p->dl.dl_throttled = 0; enqueue_flag = ENQUEUE_REPLENISH; } else p->dl.dl_boosted = 0; @@ -5314,7 +5349,7 @@ static struct notifier_block migration_notifier = { .priority = CPU_PRI_MIGRATION, }; -static void __cpuinit set_cpu_rq_start_time(void) +static void set_cpu_rq_start_time(void) { int cpu = smp_processor_id(); struct rq *rq = cpu_rq(cpu); @@ -7032,6 +7067,9 @@ void __init sched_init_smp(void) alloc_cpumask_var(&non_isolated_cpus, GFP_KERNEL); alloc_cpumask_var(&fallback_doms, GFP_KERNEL); + /* nohz_full won't take effect without isolating the cpus. */ + tick_nohz_full_add_cpus_to(cpu_isolated_map); + sched_init_numa(); /* @@ -7068,8 +7106,6 @@ void __init sched_init_smp(void) } #endif /* CONFIG_SMP */ -const_debug unsigned int sysctl_timer_migration = 1; - int in_sched_functions(unsigned long addr) { return in_lock_functions(addr) || @@ -7734,11 +7770,11 @@ static long sched_group_rt_runtime(struct task_group *tg) return rt_runtime_us; } -static int sched_group_set_rt_period(struct task_group *tg, long rt_period_us) +static int sched_group_set_rt_period(struct task_group *tg, u64 rt_period_us) { u64 rt_runtime, rt_period; - rt_period = (u64)rt_period_us * NSEC_PER_USEC; + rt_period = rt_period_us * NSEC_PER_USEC; rt_runtime = tg->rt_bandwidth.rt_runtime; return tg_set_rt_bandwidth(tg, rt_period, rt_runtime); @@ -8105,10 +8141,8 @@ static int tg_set_cfs_bandwidth(struct task_group *tg, u64 period, u64 quota) __refill_cfs_bandwidth_runtime(cfs_b); /* restart the period timer (if active) to handle new period expiry */ - if (runtime_enabled && cfs_b->timer_active) { - /* force a reprogram */ - __start_cfs_bandwidth(cfs_b, true); - } + if (runtime_enabled) + start_cfs_bandwidth(cfs_b); raw_spin_unlock_irq(&cfs_b->lock); for_each_online_cpu(i) { diff --git a/kernel/sched/cputime.c b/kernel/sched/cputime.c index 8394b1e..f5a64ff 100644 --- a/kernel/sched/cputime.c +++ b/kernel/sched/cputime.c @@ -567,7 +567,7 @@ static void cputime_advance(cputime_t *counter, cputime_t new) { cputime_t old; - while (new > (old = ACCESS_ONCE(*counter))) + while (new > (old = READ_ONCE(*counter))) cmpxchg_cputime(counter, old, new); } diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c index 5e95145..eac20c5 100644 --- a/kernel/sched/deadline.c +++ b/kernel/sched/deadline.c @@ -503,8 +503,6 @@ static int start_dl_timer(struct sched_dl_entity *dl_se, bool boosted) struct dl_rq *dl_rq = dl_rq_of_se(dl_se); struct rq *rq = rq_of_dl_rq(dl_rq); ktime_t now, act; - ktime_t soft, hard; - unsigned long range; s64 delta; if (boosted) @@ -527,15 +525,9 @@ static int start_dl_timer(struct sched_dl_entity *dl_se, bool boosted) if (ktime_us_delta(act, now) < 0) return 0; - hrtimer_set_expires(&dl_se->dl_timer, act); + hrtimer_start(&dl_se->dl_timer, act, HRTIMER_MODE_ABS); - soft = hrtimer_get_softexpires(&dl_se->dl_timer); - hard = hrtimer_get_expires(&dl_se->dl_timer); - range = ktime_to_ns(ktime_sub(hard, soft)); - __hrtimer_start_range_ns(&dl_se->dl_timer, soft, - range, HRTIMER_MODE_ABS, 0); - - return hrtimer_active(&dl_se->dl_timer); + return 1; } /* @@ -640,7 +632,7 @@ void init_dl_task_timer(struct sched_dl_entity *dl_se) } static -int dl_runtime_exceeded(struct rq *rq, struct sched_dl_entity *dl_se) +int dl_runtime_exceeded(struct sched_dl_entity *dl_se) { return (dl_se->runtime <= 0); } @@ -684,7 +676,7 @@ static void update_curr_dl(struct rq *rq) sched_rt_avg_update(rq, delta_exec); dl_se->runtime -= dl_se->dl_yielded ? 0 : delta_exec; - if (dl_runtime_exceeded(rq, dl_se)) { + if (dl_runtime_exceeded(dl_se)) { dl_se->dl_throttled = 1; __dequeue_task_dl(rq, curr, 0); if (unlikely(!start_dl_timer(dl_se, curr->dl.dl_boosted))) @@ -995,7 +987,7 @@ select_task_rq_dl(struct task_struct *p, int cpu, int sd_flag, int flags) rq = cpu_rq(cpu); rcu_read_lock(); - curr = ACCESS_ONCE(rq->curr); /* unlocked access */ + curr = READ_ONCE(rq->curr); /* unlocked access */ /* * If we are dealing with a -deadline task, we must @@ -1012,7 +1004,9 @@ select_task_rq_dl(struct task_struct *p, int cpu, int sd_flag, int flags) (p->nr_cpus_allowed > 1)) { int target = find_later_rq(p); - if (target != -1) + if (target != -1 && + dl_time_before(p->dl.deadline, + cpu_rq(target)->dl.earliest_dl.curr)) cpu = target; } rcu_read_unlock(); @@ -1230,6 +1224,32 @@ next_node: return NULL; } +/* + * Return the earliest pushable rq's task, which is suitable to be executed + * on the CPU, NULL otherwise: + */ +static struct task_struct *pick_earliest_pushable_dl_task(struct rq *rq, int cpu) +{ + struct rb_node *next_node = rq->dl.pushable_dl_tasks_leftmost; + struct task_struct *p = NULL; + + if (!has_pushable_dl_tasks(rq)) + return NULL; + +next_node: + if (next_node) { + p = rb_entry(next_node, struct task_struct, pushable_dl_tasks); + + if (pick_dl_task(rq, p, cpu)) + return p; + + next_node = rb_next(next_node); + goto next_node; + } + + return NULL; +} + static DEFINE_PER_CPU(cpumask_var_t, local_cpu_mask_dl); static int find_later_rq(struct task_struct *task) @@ -1333,6 +1353,17 @@ static struct rq *find_lock_later_rq(struct task_struct *task, struct rq *rq) later_rq = cpu_rq(cpu); + if (!dl_time_before(task->dl.deadline, + later_rq->dl.earliest_dl.curr)) { + /* + * Target rq has tasks of equal or earlier deadline, + * retrying does not release any lock and is unlikely + * to yield a different result. + */ + later_rq = NULL; + break; + } + /* Retry if something changed. */ if (double_lock_balance(rq, later_rq)) { if (unlikely(task_rq(task) != rq || @@ -1514,7 +1545,7 @@ static int pull_dl_task(struct rq *this_rq) if (src_rq->dl.dl_nr_running <= 1) goto skip; - p = pick_next_earliest_dl_task(src_rq, this_cpu); + p = pick_earliest_pushable_dl_task(src_rq, this_cpu); /* * We found a task to be pulled if: @@ -1659,7 +1690,7 @@ static void rq_offline_dl(struct rq *rq) cpudl_clear_freecpu(&rq->rd->cpudl, rq->cpu); } -void init_sched_dl_class(void) +void __init init_sched_dl_class(void) { unsigned int i; diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c index a245c1f..315c68e 100644 --- a/kernel/sched/debug.c +++ b/kernel/sched/debug.c @@ -132,12 +132,14 @@ print_task(struct seq_file *m, struct rq *rq, struct task_struct *p) p->prio); #ifdef CONFIG_SCHEDSTATS SEQ_printf(m, "%9Ld.%06ld %9Ld.%06ld %9Ld.%06ld", - SPLIT_NS(p->se.vruntime), + SPLIT_NS(p->se.statistics.wait_sum), SPLIT_NS(p->se.sum_exec_runtime), SPLIT_NS(p->se.statistics.sum_sleep_runtime)); #else - SEQ_printf(m, "%15Ld %15Ld %15Ld.%06ld %15Ld.%06ld %15Ld.%06ld", - 0LL, 0LL, 0LL, 0L, 0LL, 0L, 0LL, 0L); + SEQ_printf(m, "%9Ld.%06ld %9Ld.%06ld %9Ld.%06ld", + 0LL, 0L, + SPLIT_NS(p->se.sum_exec_runtime), + 0LL, 0L); #endif #ifdef CONFIG_NUMA_BALANCING SEQ_printf(m, " %d", task_node(p)); @@ -156,7 +158,7 @@ static void print_rq(struct seq_file *m, struct rq *rq, int rq_cpu) SEQ_printf(m, "\nrunnable tasks:\n" " task PID tree-key switches prio" - " exec-runtime sum-exec sum-sleep\n" + " wait-time sum-exec sum-sleep\n" "------------------------------------------------------" "----------------------------------------------------\n"); @@ -230,8 +232,6 @@ void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq) #endif #endif #ifdef CONFIG_CFS_BANDWIDTH - SEQ_printf(m, " .%-30s: %d\n", "tg->cfs_bandwidth.timer_active", - cfs_rq->tg->cfs_bandwidth.timer_active); SEQ_printf(m, " .%-30s: %d\n", "throttled", cfs_rq->throttled); SEQ_printf(m, " .%-30s: %d\n", "throttle_count", @@ -582,6 +582,7 @@ void proc_sched_show_task(struct task_struct *p, struct seq_file *m) nr_switches = p->nvcsw + p->nivcsw; #ifdef CONFIG_SCHEDSTATS + PN(se.statistics.sum_sleep_runtime); PN(se.statistics.wait_start); PN(se.statistics.sleep_start); PN(se.statistics.block_start); diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index c2980e8..40a7fcb 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -141,9 +141,9 @@ static inline void update_load_set(struct load_weight *lw, unsigned long w) * * This idea comes from the SD scheduler of Con Kolivas: */ -static int get_update_sysctl_factor(void) +static unsigned int get_update_sysctl_factor(void) { - unsigned int cpus = min_t(int, num_online_cpus(), 8); + unsigned int cpus = min_t(unsigned int, num_online_cpus(), 8); unsigned int factor; switch (sysctl_sched_tunable_scaling) { @@ -576,7 +576,7 @@ int sched_proc_update_handler(struct ctl_table *table, int write, loff_t *ppos) { int ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos); - int factor = get_update_sysctl_factor(); + unsigned int factor = get_update_sysctl_factor(); if (ret || !write) return ret; @@ -834,7 +834,7 @@ static unsigned int task_nr_scan_windows(struct task_struct *p) static unsigned int task_scan_min(struct task_struct *p) { - unsigned int scan_size = ACCESS_ONCE(sysctl_numa_balancing_scan_size); + unsigned int scan_size = READ_ONCE(sysctl_numa_balancing_scan_size); unsigned int scan, floor; unsigned int windows = 1; @@ -1198,11 +1198,9 @@ 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. @@ -1215,39 +1213,30 @@ 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. */ - load_a = dst_load; - load_b = src_load; - if (load_a < load_b) - swap(load_a, load_b); + if (dst_load < src_load) + swap(dst_load, src_load); /* Is the difference below the threshold? */ - imb = load_a * src_capacity * 100 - - load_b * dst_capacity * env->imbalance_pct; + imb = dst_load * src_capacity * 100 - + src_load * dst_capacity * env->imbalance_pct; if (imb <= 0) return false; /* * The imbalance is above the allowed threshold. - * Allow a move that brings us closer to a balanced situation, - * without moving things past the point of balance. + * Compare it with the old imbalance. */ orig_src_load = env->src_stats.load; + orig_dst_load = env->dst_stats.load; - /* - * 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; + if (orig_dst_load < orig_src_load) + swap(orig_dst_load, orig_src_load); - 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; + old_imb = orig_dst_load * src_capacity * 100 - + orig_src_load * dst_capacity * env->imbalance_pct; + + /* Would this change make things worse? */ + return (imb > old_imb); } /* @@ -1409,6 +1398,30 @@ static void task_numa_find_cpu(struct task_numa_env *env, } } +/* Only move tasks to a NUMA node less busy than the current node. */ +static bool numa_has_capacity(struct task_numa_env *env) +{ + struct numa_stats *src = &env->src_stats; + struct numa_stats *dst = &env->dst_stats; + + if (src->has_free_capacity && !dst->has_free_capacity) + return false; + + /* + * Only consider a task move if the source has a higher load + * than the destination, corrected for CPU capacity on each node. + * + * src->load dst->load + * --------------------- vs --------------------- + * src->compute_capacity dst->compute_capacity + */ + if (src->load * dst->compute_capacity > + dst->load * src->compute_capacity) + return true; + + return false; +} + static int task_numa_migrate(struct task_struct *p) { struct task_numa_env env = { @@ -1463,7 +1476,8 @@ static int task_numa_migrate(struct task_struct *p) update_numa_stats(&env.dst_stats, env.dst_nid); /* Try to find a spot on the preferred nid. */ - task_numa_find_cpu(&env, taskimp, groupimp); + if (numa_has_capacity(&env)) + task_numa_find_cpu(&env, taskimp, groupimp); /* * Look at other nodes in these cases: @@ -1494,7 +1508,8 @@ static int task_numa_migrate(struct task_struct *p) env.dist = dist; env.dst_nid = nid; update_numa_stats(&env.dst_stats, env.dst_nid); - task_numa_find_cpu(&env, taskimp, groupimp); + if (numa_has_capacity(&env)) + task_numa_find_cpu(&env, taskimp, groupimp); } } @@ -1794,7 +1809,12 @@ static void task_numa_placement(struct task_struct *p) u64 runtime, period; spinlock_t *group_lock = NULL; - seq = ACCESS_ONCE(p->mm->numa_scan_seq); + /* + * The p->mm->numa_scan_seq field gets updated without + * exclusive access. Use READ_ONCE() here to ensure + * that the field is read in a single access: + */ + seq = READ_ONCE(p->mm->numa_scan_seq); if (p->numa_scan_seq == seq) return; p->numa_scan_seq = seq; @@ -1938,7 +1958,7 @@ static void task_numa_group(struct task_struct *p, int cpupid, int flags, } rcu_read_lock(); - tsk = ACCESS_ONCE(cpu_rq(cpu)->curr); + tsk = READ_ONCE(cpu_rq(cpu)->curr); if (!cpupid_match_pid(tsk, cpupid)) goto no_join; @@ -2107,7 +2127,15 @@ void task_numa_fault(int last_cpupid, int mem_node, int pages, int flags) static void reset_ptenuma_scan(struct task_struct *p) { - ACCESS_ONCE(p->mm->numa_scan_seq)++; + /* + * We only did a read acquisition of the mmap sem, so + * p->mm->numa_scan_seq is written to without exclusive access + * and the update is not guaranteed to be atomic. That's not + * much of an issue though, since this is just used for + * statistical sampling. Use READ_ONCE/WRITE_ONCE, which are not + * expensive, to avoid any form of compiler optimizations: + */ + WRITE_ONCE(p->mm->numa_scan_seq, READ_ONCE(p->mm->numa_scan_seq) + 1); p->mm->numa_scan_offset = 0; } @@ -3476,16 +3504,7 @@ static int assign_cfs_rq_runtime(struct cfs_rq *cfs_rq) if (cfs_b->quota == RUNTIME_INF) amount = min_amount; else { - /* - * If the bandwidth pool has become inactive, then at least one - * period must have elapsed since the last consumption. - * Refresh the global state and ensure bandwidth timer becomes - * active. - */ - if (!cfs_b->timer_active) { - __refill_cfs_bandwidth_runtime(cfs_b); - __start_cfs_bandwidth(cfs_b, false); - } + start_cfs_bandwidth(cfs_b); if (cfs_b->runtime > 0) { amount = min(cfs_b->runtime, min_amount); @@ -3634,6 +3653,7 @@ static void throttle_cfs_rq(struct cfs_rq *cfs_rq) struct cfs_bandwidth *cfs_b = tg_cfs_bandwidth(cfs_rq->tg); struct sched_entity *se; long task_delta, dequeue = 1; + bool empty; se = cfs_rq->tg->se[cpu_of(rq_of(cfs_rq))]; @@ -3663,13 +3683,21 @@ static void throttle_cfs_rq(struct cfs_rq *cfs_rq) cfs_rq->throttled = 1; cfs_rq->throttled_clock = rq_clock(rq); raw_spin_lock(&cfs_b->lock); + empty = list_empty(&cfs_rq->throttled_list); + /* * Add to the _head_ of the list, so that an already-started * distribute_cfs_runtime will not see us */ list_add_rcu(&cfs_rq->throttled_list, &cfs_b->throttled_cfs_rq); - if (!cfs_b->timer_active) - __start_cfs_bandwidth(cfs_b, false); + + /* + * If we're the first throttled task, make sure the bandwidth + * timer is running. + */ + if (empty) + start_cfs_bandwidth(cfs_b); + raw_spin_unlock(&cfs_b->lock); } @@ -3784,13 +3812,6 @@ static int do_sched_cfs_period_timer(struct cfs_bandwidth *cfs_b, int overrun) if (cfs_b->idle && !throttled) goto out_deactivate; - /* - * if we have relooped after returning idle once, we need to update our - * status as actually running, so that other cpus doing - * __start_cfs_bandwidth will stop trying to cancel us. - */ - cfs_b->timer_active = 1; - __refill_cfs_bandwidth_runtime(cfs_b); if (!throttled) { @@ -3835,7 +3856,6 @@ static int do_sched_cfs_period_timer(struct cfs_bandwidth *cfs_b, int overrun) return 0; out_deactivate: - cfs_b->timer_active = 0; return 1; } @@ -3850,7 +3870,7 @@ static const u64 cfs_bandwidth_slack_period = 5 * NSEC_PER_MSEC; * Are we near the end of the current quota period? * * Requires cfs_b->lock for hrtimer_expires_remaining to be safe against the - * hrtimer base being cleared by __hrtimer_start_range_ns. In the case of + * hrtimer base being cleared by hrtimer_start. In the case of * migrate_hrtimers, base is never cleared, so we are fine. */ static int runtime_refresh_within(struct cfs_bandwidth *cfs_b, u64 min_expire) @@ -3878,8 +3898,9 @@ static void start_cfs_slack_bandwidth(struct cfs_bandwidth *cfs_b) if (runtime_refresh_within(cfs_b, min_left)) return; - start_bandwidth_timer(&cfs_b->slack_timer, - ns_to_ktime(cfs_bandwidth_slack_period)); + hrtimer_start(&cfs_b->slack_timer, + ns_to_ktime(cfs_bandwidth_slack_period), + HRTIMER_MODE_REL); } /* we know any runtime found here is valid as update_curr() precedes return */ @@ -3999,6 +4020,7 @@ static enum hrtimer_restart sched_cfs_slack_timer(struct hrtimer *timer) { struct cfs_bandwidth *cfs_b = container_of(timer, struct cfs_bandwidth, slack_timer); + do_sched_cfs_slack_timer(cfs_b); return HRTIMER_NORESTART; @@ -4008,20 +4030,19 @@ static enum hrtimer_restart sched_cfs_period_timer(struct hrtimer *timer) { struct cfs_bandwidth *cfs_b = container_of(timer, struct cfs_bandwidth, period_timer); - ktime_t now; int overrun; int idle = 0; raw_spin_lock(&cfs_b->lock); for (;;) { - now = hrtimer_cb_get_time(timer); - overrun = hrtimer_forward(timer, now, cfs_b->period); - + overrun = hrtimer_forward_now(timer, cfs_b->period); if (!overrun) break; idle = do_sched_cfs_period_timer(cfs_b, overrun); } + if (idle) + cfs_b->period_active = 0; raw_spin_unlock(&cfs_b->lock); return idle ? HRTIMER_NORESTART : HRTIMER_RESTART; @@ -4035,7 +4056,7 @@ void init_cfs_bandwidth(struct cfs_bandwidth *cfs_b) cfs_b->period = ns_to_ktime(default_cfs_period()); INIT_LIST_HEAD(&cfs_b->throttled_cfs_rq); - hrtimer_init(&cfs_b->period_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); + hrtimer_init(&cfs_b->period_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_PINNED); cfs_b->period_timer.function = sched_cfs_period_timer; hrtimer_init(&cfs_b->slack_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); cfs_b->slack_timer.function = sched_cfs_slack_timer; @@ -4047,28 +4068,15 @@ static void init_cfs_rq_runtime(struct cfs_rq *cfs_rq) INIT_LIST_HEAD(&cfs_rq->throttled_list); } -/* requires cfs_b->lock, may release to reprogram timer */ -void __start_cfs_bandwidth(struct cfs_bandwidth *cfs_b, bool force) +void start_cfs_bandwidth(struct cfs_bandwidth *cfs_b) { - /* - * The timer may be active because we're trying to set a new bandwidth - * period or because we're racing with the tear-down path - * (timer_active==0 becomes visible before the hrtimer call-back - * terminates). In either case we ensure that it's re-programmed - */ - while (unlikely(hrtimer_active(&cfs_b->period_timer)) && - hrtimer_try_to_cancel(&cfs_b->period_timer) < 0) { - /* bounce the lock to allow do_sched_cfs_period_timer to run */ - raw_spin_unlock(&cfs_b->lock); - cpu_relax(); - raw_spin_lock(&cfs_b->lock); - /* if someone else restarted the timer then we're done */ - if (!force && cfs_b->timer_active) - return; - } + lockdep_assert_held(&cfs_b->lock); - cfs_b->timer_active = 1; - start_bandwidth_timer(&cfs_b->period_timer, cfs_b->period); + if (!cfs_b->period_active) { + cfs_b->period_active = 1; + hrtimer_forward_now(&cfs_b->period_timer, cfs_b->period); + hrtimer_start_expires(&cfs_b->period_timer, HRTIMER_MODE_ABS_PINNED); + } } static void destroy_cfs_bandwidth(struct cfs_bandwidth *cfs_b) @@ -4323,6 +4331,189 @@ static void dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags) } #ifdef CONFIG_SMP + +/* + * per rq 'load' arrray crap; XXX kill this. + */ + +/* + * The exact cpuload at various idx values, calculated at every tick would be + * load = (2^idx - 1) / 2^idx * load + 1 / 2^idx * cur_load + * + * If a cpu misses updates for n-1 ticks (as it was idle) and update gets called + * on nth tick when cpu may be busy, then we have: + * load = ((2^idx - 1) / 2^idx)^(n-1) * load + * load = (2^idx - 1) / 2^idx) * load + 1 / 2^idx * cur_load + * + * decay_load_missed() below does efficient calculation of + * load = ((2^idx - 1) / 2^idx)^(n-1) * load + * avoiding 0..n-1 loop doing load = ((2^idx - 1) / 2^idx) * load + * + * The calculation is approximated on a 128 point scale. + * degrade_zero_ticks is the number of ticks after which load at any + * particular idx is approximated to be zero. + * degrade_factor is a precomputed table, a row for each load idx. + * Each column corresponds to degradation factor for a power of two ticks, + * based on 128 point scale. + * Example: + * row 2, col 3 (=12) says that the degradation at load idx 2 after + * 8 ticks is 12/128 (which is an approximation of exact factor 3^8/4^8). + * + * With this power of 2 load factors, we can degrade the load n times + * by looking at 1 bits in n and doing as many mult/shift instead of + * n mult/shifts needed by the exact degradation. + */ +#define DEGRADE_SHIFT 7 +static const unsigned char + degrade_zero_ticks[CPU_LOAD_IDX_MAX] = {0, 8, 32, 64, 128}; +static const unsigned char + degrade_factor[CPU_LOAD_IDX_MAX][DEGRADE_SHIFT + 1] = { + {0, 0, 0, 0, 0, 0, 0, 0}, + {64, 32, 8, 0, 0, 0, 0, 0}, + {96, 72, 40, 12, 1, 0, 0}, + {112, 98, 75, 43, 15, 1, 0}, + {120, 112, 98, 76, 45, 16, 2} }; + +/* + * Update cpu_load for any missed ticks, due to tickless idle. The backlog + * would be when CPU is idle and so we just decay the old load without + * adding any new load. + */ +static unsigned long +decay_load_missed(unsigned long load, unsigned long missed_updates, int idx) +{ + int j = 0; + + if (!missed_updates) + return load; + + if (missed_updates >= degrade_zero_ticks[idx]) + return 0; + + if (idx == 1) + return load >> missed_updates; + + while (missed_updates) { + if (missed_updates % 2) + load = (load * degrade_factor[idx][j]) >> DEGRADE_SHIFT; + + missed_updates >>= 1; + j++; + } + return load; +} + +/* + * Update rq->cpu_load[] statistics. This function is usually called every + * scheduler tick (TICK_NSEC). With tickless idle this will not be called + * every tick. We fix it up based on jiffies. + */ +static void __update_cpu_load(struct rq *this_rq, unsigned long this_load, + unsigned long pending_updates) +{ + int i, scale; + + this_rq->nr_load_updates++; + + /* Update our load: */ + this_rq->cpu_load[0] = this_load; /* Fasttrack for idx 0 */ + for (i = 1, scale = 2; i < CPU_LOAD_IDX_MAX; i++, scale += scale) { + unsigned long old_load, new_load; + + /* scale is effectively 1 << i now, and >> i divides by scale */ + + old_load = this_rq->cpu_load[i]; + old_load = decay_load_missed(old_load, pending_updates - 1, i); + new_load = this_load; + /* + * Round up the averaging division if load is increasing. This + * prevents us from getting stuck on 9 if the load is 10, for + * example. + */ + if (new_load > old_load) + new_load += scale - 1; + + this_rq->cpu_load[i] = (old_load * (scale - 1) + new_load) >> i; + } + + sched_avg_update(this_rq); +} + +#ifdef CONFIG_NO_HZ_COMMON +/* + * There is no sane way to deal with nohz on smp when using jiffies because the + * cpu doing the jiffies update might drift wrt the cpu doing the jiffy reading + * causing off-by-one errors in observed deltas; {0,2} instead of {1,1}. + * + * Therefore we cannot use the delta approach from the regular tick since that + * would seriously skew the load calculation. However we'll make do for those + * updates happening while idle (nohz_idle_balance) or coming out of idle + * (tick_nohz_idle_exit). + * + * This means we might still be one tick off for nohz periods. + */ + +/* + * Called from nohz_idle_balance() to update the load ratings before doing the + * idle balance. + */ +static void update_idle_cpu_load(struct rq *this_rq) +{ + unsigned long curr_jiffies = READ_ONCE(jiffies); + unsigned long load = this_rq->cfs.runnable_load_avg; + unsigned long pending_updates; + + /* + * bail if there's load or we're actually up-to-date. + */ + if (load || curr_jiffies == this_rq->last_load_update_tick) + return; + + pending_updates = curr_jiffies - this_rq->last_load_update_tick; + this_rq->last_load_update_tick = curr_jiffies; + + __update_cpu_load(this_rq, load, pending_updates); +} + +/* + * Called from tick_nohz_idle_exit() -- try and fix up the ticks we missed. + */ +void update_cpu_load_nohz(void) +{ + struct rq *this_rq = this_rq(); + unsigned long curr_jiffies = READ_ONCE(jiffies); + unsigned long pending_updates; + + if (curr_jiffies == this_rq->last_load_update_tick) + return; + + raw_spin_lock(&this_rq->lock); + pending_updates = curr_jiffies - this_rq->last_load_update_tick; + if (pending_updates) { + this_rq->last_load_update_tick = curr_jiffies; + /* + * We were idle, this means load 0, the current load might be + * !0 due to remote wakeups and the sort. + */ + __update_cpu_load(this_rq, 0, pending_updates); + } + raw_spin_unlock(&this_rq->lock); +} +#endif /* CONFIG_NO_HZ */ + +/* + * Called from scheduler_tick() + */ +void update_cpu_load_active(struct rq *this_rq) +{ + unsigned long load = this_rq->cfs.runnable_load_avg; + /* + * See the mess around update_idle_cpu_load() / update_cpu_load_nohz(). + */ + this_rq->last_load_update_tick = jiffies; + __update_cpu_load(this_rq, load, 1); +} + /* Used instead of source_load when we know the type == 0 */ static unsigned long weighted_cpuload(const int cpu) { @@ -4375,7 +4566,7 @@ static unsigned long capacity_orig_of(int cpu) static unsigned long cpu_avg_load_per_task(int cpu) { struct rq *rq = cpu_rq(cpu); - unsigned long nr_running = ACCESS_ONCE(rq->cfs.h_nr_running); + unsigned long nr_running = READ_ONCE(rq->cfs.h_nr_running); unsigned long load_avg = rq->cfs.runnable_load_avg; if (nr_running) @@ -5126,18 +5317,21 @@ again: * entity, update_curr() will update its vruntime, otherwise * forget we've ever seen it. */ - if (curr && curr->on_rq) - update_curr(cfs_rq); - else - curr = NULL; + if (curr) { + if (curr->on_rq) + update_curr(cfs_rq); + else + curr = NULL; - /* - * This call to check_cfs_rq_runtime() will do the throttle and - * dequeue its entity in the parent(s). Therefore the 'simple' - * nr_running test will indeed be correct. - */ - if (unlikely(check_cfs_rq_runtime(cfs_rq))) - goto simple; + /* + * This call to check_cfs_rq_runtime() will do the + * throttle and dequeue its entity in the parent(s). + * Therefore the 'simple' nr_running test will indeed + * be correct. + */ + if (unlikely(check_cfs_rq_runtime(cfs_rq))) + goto simple; + } se = pick_next_entity(cfs_rq, curr); cfs_rq = group_cfs_rq(se); @@ -5467,10 +5661,15 @@ static int task_hot(struct task_struct *p, struct lb_env *env) } #ifdef CONFIG_NUMA_BALANCING -/* Returns true if the destination node has incurred more faults */ +/* + * Returns true if the destination node is the preferred node. + * Needs to match fbq_classify_rq(): if there is a runnable task + * that is not on its preferred node, we should identify it. + */ static bool migrate_improves_locality(struct task_struct *p, struct lb_env *env) { struct numa_group *numa_group = rcu_dereference(p->numa_group); + unsigned long src_faults, dst_faults; int src_nid, dst_nid; if (!sched_feat(NUMA_FAVOUR_HIGHER) || !p->numa_faults || @@ -5484,29 +5683,30 @@ static bool migrate_improves_locality(struct task_struct *p, struct lb_env *env) if (src_nid == dst_nid) return false; - if (numa_group) { - /* Task is already in the group's interleave set. */ - if (node_isset(src_nid, numa_group->active_nodes)) - return false; - - /* Task is moving into the group's interleave set. */ - if (node_isset(dst_nid, numa_group->active_nodes)) - return true; - - return group_faults(p, dst_nid) > group_faults(p, src_nid); - } - /* Encourage migration to the preferred node. */ if (dst_nid == p->numa_preferred_nid) return true; - return task_faults(p, dst_nid) > task_faults(p, src_nid); + /* Migrating away from the preferred node is bad. */ + if (src_nid == p->numa_preferred_nid) + return false; + + if (numa_group) { + src_faults = group_faults(p, src_nid); + dst_faults = group_faults(p, dst_nid); + } else { + src_faults = task_faults(p, src_nid); + dst_faults = task_faults(p, dst_nid); + } + + return dst_faults > src_faults; } static bool migrate_degrades_locality(struct task_struct *p, struct lb_env *env) { struct numa_group *numa_group = rcu_dereference(p->numa_group); + unsigned long src_faults, dst_faults; int src_nid, dst_nid; if (!sched_feat(NUMA) || !sched_feat(NUMA_RESIST_LOWER)) @@ -5521,23 +5721,23 @@ static bool migrate_degrades_locality(struct task_struct *p, struct lb_env *env) if (src_nid == dst_nid) return false; - if (numa_group) { - /* Task is moving within/into the group's interleave set. */ - if (node_isset(dst_nid, numa_group->active_nodes)) - return false; + /* Migrating away from the preferred node is bad. */ + if (src_nid == p->numa_preferred_nid) + return true; - /* Task is moving out of the group's interleave set. */ - if (node_isset(src_nid, numa_group->active_nodes)) - return true; + /* Encourage migration to the preferred node. */ + if (dst_nid == p->numa_preferred_nid) + return false; - return group_faults(p, dst_nid) < group_faults(p, src_nid); + if (numa_group) { + src_faults = group_faults(p, src_nid); + dst_faults = group_faults(p, dst_nid); + } else { + src_faults = task_faults(p, src_nid); + dst_faults = task_faults(p, dst_nid); } - /* Migrating away from the preferred node is always bad. */ - if (src_nid == p->numa_preferred_nid) - return true; - - return task_faults(p, dst_nid) < task_faults(p, src_nid); + return dst_faults < src_faults; } #else @@ -6037,8 +6237,8 @@ static unsigned long scale_rt_capacity(int cpu) * Since we're reading these variables without serialization make sure * we read them once before doing sanity checks on them. */ - age_stamp = ACCESS_ONCE(rq->age_stamp); - avg = ACCESS_ONCE(rq->rt_avg); + age_stamp = READ_ONCE(rq->age_stamp); + avg = READ_ONCE(rq->rt_avg); delta = __rq_clock_broken(rq) - age_stamp; if (unlikely(delta < 0)) diff --git a/kernel/sched/idle.c b/kernel/sched/idle.c index fefcb1f..594275e 100644 --- a/kernel/sched/idle.c +++ b/kernel/sched/idle.c @@ -15,6 +15,15 @@ #include "sched.h" +/** + * sched_idle_set_state - Record idle state for the current CPU. + * @idle_state: State to record. + */ +void sched_idle_set_state(struct cpuidle_state *idle_state) +{ + idle_set_state(this_rq(), idle_state); +} + static int __read_mostly cpu_idle_force_poll; void cpu_idle_poll_ctrl(bool enable) @@ -68,6 +77,46 @@ void __weak arch_cpu_idle(void) } /** + * default_idle_call - Default CPU idle routine. + * + * To use when the cpuidle framework cannot be used. + */ +void default_idle_call(void) +{ + if (current_clr_polling_and_test()) + local_irq_enable(); + else + arch_cpu_idle(); +} + +static int call_cpuidle(struct cpuidle_driver *drv, struct cpuidle_device *dev, + int next_state) +{ + /* Fall back to the default arch idle method on errors. */ + if (next_state < 0) { + default_idle_call(); + return next_state; + } + + /* + * The idle task must be scheduled, it is pointless to go to idle, just + * update no idle residency and return. + */ + if (current_clr_polling_and_test()) { + dev->last_residency = 0; + local_irq_enable(); + return -EBUSY; + } + + /* + * Enter the idle state previously returned by the governor decision. + * This function will block until an interrupt occurs and will take + * care of re-enabling the local interrupts + */ + return cpuidle_enter(drv, dev, next_state); +} + +/** * cpuidle_idle_call - the main idle function * * NOTE: no locks or semaphores should be used here @@ -81,7 +130,6 @@ static void cpuidle_idle_call(void) struct cpuidle_device *dev = __this_cpu_read(cpuidle_devices); struct cpuidle_driver *drv = cpuidle_get_cpu_driver(dev); int next_state, entered_state; - bool reflect; /* * Check if the idle task must be rescheduled. If it is the @@ -105,8 +153,10 @@ static void cpuidle_idle_call(void) */ rcu_idle_enter(); - if (cpuidle_not_available(drv, dev)) - goto use_default; + if (cpuidle_not_available(drv, dev)) { + default_idle_call(); + goto exit_idle; + } /* * Suspend-to-idle ("freeze") is a system state in which all user space @@ -124,52 +174,19 @@ static void cpuidle_idle_call(void) goto exit_idle; } - reflect = false; next_state = cpuidle_find_deepest_state(drv, dev); + call_cpuidle(drv, dev, next_state); } else { - reflect = true; /* * Ask the cpuidle framework to choose a convenient idle state. */ 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 - * go to idle, just update no idle residency and get - * out of this function - */ - if (current_clr_polling_and_test()) { - dev->last_residency = 0; - entered_state = next_state; - local_irq_enable(); - goto exit_idle; - } - - /* Take note of the planned idle state. */ - idle_set_state(this_rq(), &drv->states[next_state]); - - /* - * Enter the idle state previously returned by the governor decision. - * This function will block until an interrupt occurs and will take - * care of re-enabling the local interrupts - */ - entered_state = cpuidle_enter(drv, dev, next_state); - - /* The cpu is no longer idle or about to enter idle. */ - idle_set_state(this_rq(), NULL); - - if (entered_state == -EBUSY) - goto use_default; - - /* - * Give the governor an opportunity to reflect on the outcome - */ - if (reflect) + entered_state = call_cpuidle(drv, dev, next_state); + /* + * Give the governor an opportunity to reflect on the outcome + */ cpuidle_reflect(dev, entered_state); + } exit_idle: __current_set_polling(); @@ -182,19 +199,6 @@ 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; } DEFINE_PER_CPU(bool, cpu_dead_idle); diff --git a/kernel/sched/proc.c b/kernel/sched/loadavg.c index 8ecd552..ef71590 100644 --- a/kernel/sched/proc.c +++ b/kernel/sched/loadavg.c @@ -1,7 +1,9 @@ /* - * kernel/sched/proc.c + * kernel/sched/loadavg.c * - * Kernel load calculations, forked from sched/core.c + * This file contains the magic bits required to compute the global loadavg + * figure. Its a silly number but people think its important. We go through + * great pains to make it work on big machines and tickless kernels. */ #include <linux/export.h> @@ -81,7 +83,7 @@ long calc_load_fold_active(struct rq *this_rq) long nr_active, delta = 0; nr_active = this_rq->nr_running; - nr_active += (long) this_rq->nr_uninterruptible; + nr_active += (long)this_rq->nr_uninterruptible; if (nr_active != this_rq->calc_load_active) { delta = nr_active - this_rq->calc_load_active; @@ -186,6 +188,7 @@ void calc_load_enter_idle(void) delta = calc_load_fold_active(this_rq); if (delta) { int idx = calc_load_write_idx(); + atomic_long_add(delta, &calc_load_idle[idx]); } } @@ -241,18 +244,20 @@ fixed_power_int(unsigned long x, unsigned int frac_bits, unsigned int n) { unsigned long result = 1UL << frac_bits; - if (n) for (;;) { - if (n & 1) { - result *= x; - result += 1UL << (frac_bits - 1); - result >>= frac_bits; + if (n) { + for (;;) { + if (n & 1) { + result *= x; + result += 1UL << (frac_bits - 1); + result >>= frac_bits; + } + n >>= 1; + if (!n) + break; + x *= x; + x += 1UL << (frac_bits - 1); + x >>= frac_bits; } - n >>= 1; - if (!n) - break; - x *= x; - x += 1UL << (frac_bits - 1); - x >>= frac_bits; } return result; @@ -285,7 +290,6 @@ static unsigned long calc_load_n(unsigned long load, unsigned long exp, unsigned long active, unsigned int n) { - return calc_load(load, fixed_power_int(exp, FSHIFT, n), active); } @@ -339,6 +343,8 @@ static inline void calc_global_nohz(void) { } /* * calc_load - update the avenrun load estimates 10 ticks after the * CPUs have updated calc_load_tasks. + * + * Called from the global timer code. */ void calc_global_load(unsigned long ticks) { @@ -370,10 +376,10 @@ void calc_global_load(unsigned long ticks) } /* - * Called from update_cpu_load() to periodically update this CPU's + * Called from scheduler_tick() to periodically update this CPU's * active count. */ -static void calc_load_account_active(struct rq *this_rq) +void calc_global_load_tick(struct rq *this_rq) { long delta; @@ -386,199 +392,3 @@ static void calc_load_account_active(struct rq *this_rq) this_rq->calc_load_update += LOAD_FREQ; } - -/* - * End of global load-average stuff - */ - -/* - * The exact cpuload at various idx values, calculated at every tick would be - * load = (2^idx - 1) / 2^idx * load + 1 / 2^idx * cur_load - * - * If a cpu misses updates for n-1 ticks (as it was idle) and update gets called - * on nth tick when cpu may be busy, then we have: - * load = ((2^idx - 1) / 2^idx)^(n-1) * load - * load = (2^idx - 1) / 2^idx) * load + 1 / 2^idx * cur_load - * - * decay_load_missed() below does efficient calculation of - * load = ((2^idx - 1) / 2^idx)^(n-1) * load - * avoiding 0..n-1 loop doing load = ((2^idx - 1) / 2^idx) * load - * - * The calculation is approximated on a 128 point scale. - * degrade_zero_ticks is the number of ticks after which load at any - * particular idx is approximated to be zero. - * degrade_factor is a precomputed table, a row for each load idx. - * Each column corresponds to degradation factor for a power of two ticks, - * based on 128 point scale. - * Example: - * row 2, col 3 (=12) says that the degradation at load idx 2 after - * 8 ticks is 12/128 (which is an approximation of exact factor 3^8/4^8). - * - * With this power of 2 load factors, we can degrade the load n times - * by looking at 1 bits in n and doing as many mult/shift instead of - * n mult/shifts needed by the exact degradation. - */ -#define DEGRADE_SHIFT 7 -static const unsigned char - degrade_zero_ticks[CPU_LOAD_IDX_MAX] = {0, 8, 32, 64, 128}; -static const unsigned char - degrade_factor[CPU_LOAD_IDX_MAX][DEGRADE_SHIFT + 1] = { - {0, 0, 0, 0, 0, 0, 0, 0}, - {64, 32, 8, 0, 0, 0, 0, 0}, - {96, 72, 40, 12, 1, 0, 0}, - {112, 98, 75, 43, 15, 1, 0}, - {120, 112, 98, 76, 45, 16, 2} }; - -/* - * Update cpu_load for any missed ticks, due to tickless idle. The backlog - * would be when CPU is idle and so we just decay the old load without - * adding any new load. - */ -static unsigned long -decay_load_missed(unsigned long load, unsigned long missed_updates, int idx) -{ - int j = 0; - - if (!missed_updates) - return load; - - if (missed_updates >= degrade_zero_ticks[idx]) - return 0; - - if (idx == 1) - return load >> missed_updates; - - while (missed_updates) { - if (missed_updates % 2) - load = (load * degrade_factor[idx][j]) >> DEGRADE_SHIFT; - - missed_updates >>= 1; - j++; - } - return load; -} - -/* - * Update rq->cpu_load[] statistics. This function is usually called every - * scheduler tick (TICK_NSEC). With tickless idle this will not be called - * every tick. We fix it up based on jiffies. - */ -static void __update_cpu_load(struct rq *this_rq, unsigned long this_load, - unsigned long pending_updates) -{ - int i, scale; - - this_rq->nr_load_updates++; - - /* Update our load: */ - this_rq->cpu_load[0] = this_load; /* Fasttrack for idx 0 */ - for (i = 1, scale = 2; i < CPU_LOAD_IDX_MAX; i++, scale += scale) { - unsigned long old_load, new_load; - - /* scale is effectively 1 << i now, and >> i divides by scale */ - - old_load = this_rq->cpu_load[i]; - old_load = decay_load_missed(old_load, pending_updates - 1, i); - new_load = this_load; - /* - * Round up the averaging division if load is increasing. This - * prevents us from getting stuck on 9 if the load is 10, for - * example. - */ - if (new_load > old_load) - new_load += scale - 1; - - this_rq->cpu_load[i] = (old_load * (scale - 1) + new_load) >> i; - } - - sched_avg_update(this_rq); -} - -#ifdef CONFIG_SMP -static inline unsigned long get_rq_runnable_load(struct rq *rq) -{ - return rq->cfs.runnable_load_avg; -} -#else -static inline unsigned long get_rq_runnable_load(struct rq *rq) -{ - return rq->load.weight; -} -#endif - -#ifdef CONFIG_NO_HZ_COMMON -/* - * There is no sane way to deal with nohz on smp when using jiffies because the - * cpu doing the jiffies update might drift wrt the cpu doing the jiffy reading - * causing off-by-one errors in observed deltas; {0,2} instead of {1,1}. - * - * Therefore we cannot use the delta approach from the regular tick since that - * would seriously skew the load calculation. However we'll make do for those - * updates happening while idle (nohz_idle_balance) or coming out of idle - * (tick_nohz_idle_exit). - * - * This means we might still be one tick off for nohz periods. - */ - -/* - * Called from nohz_idle_balance() to update the load ratings before doing the - * idle balance. - */ -void update_idle_cpu_load(struct rq *this_rq) -{ - unsigned long curr_jiffies = ACCESS_ONCE(jiffies); - unsigned long load = get_rq_runnable_load(this_rq); - unsigned long pending_updates; - - /* - * bail if there's load or we're actually up-to-date. - */ - if (load || curr_jiffies == this_rq->last_load_update_tick) - return; - - pending_updates = curr_jiffies - this_rq->last_load_update_tick; - this_rq->last_load_update_tick = curr_jiffies; - - __update_cpu_load(this_rq, load, pending_updates); -} - -/* - * Called from tick_nohz_idle_exit() -- try and fix up the ticks we missed. - */ -void update_cpu_load_nohz(void) -{ - struct rq *this_rq = this_rq(); - unsigned long curr_jiffies = ACCESS_ONCE(jiffies); - unsigned long pending_updates; - - if (curr_jiffies == this_rq->last_load_update_tick) - return; - - raw_spin_lock(&this_rq->lock); - pending_updates = curr_jiffies - this_rq->last_load_update_tick; - if (pending_updates) { - this_rq->last_load_update_tick = curr_jiffies; - /* - * We were idle, this means load 0, the current load might be - * !0 due to remote wakeups and the sort. - */ - __update_cpu_load(this_rq, 0, pending_updates); - } - raw_spin_unlock(&this_rq->lock); -} -#endif /* CONFIG_NO_HZ */ - -/* - * Called from scheduler_tick() - */ -void update_cpu_load_active(struct rq *this_rq) -{ - unsigned long load = get_rq_runnable_load(this_rq); - /* - * See the mess around update_idle_cpu_load() / update_cpu_load_nohz(). - */ - this_rq->last_load_update_tick = jiffies; - __update_cpu_load(this_rq, load, 1); - - calc_load_account_active(this_rq); -} diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c index 575da76..7d7093c5 100644 --- a/kernel/sched/rt.c +++ b/kernel/sched/rt.c @@ -18,19 +18,22 @@ static enum hrtimer_restart sched_rt_period_timer(struct hrtimer *timer) { struct rt_bandwidth *rt_b = container_of(timer, struct rt_bandwidth, rt_period_timer); - ktime_t now; - int overrun; int idle = 0; + int overrun; + raw_spin_lock(&rt_b->rt_runtime_lock); for (;;) { - now = hrtimer_cb_get_time(timer); - overrun = hrtimer_forward(timer, now, rt_b->rt_period); - + overrun = hrtimer_forward_now(timer, rt_b->rt_period); if (!overrun) break; + raw_spin_unlock(&rt_b->rt_runtime_lock); idle = do_sched_rt_period_timer(rt_b, overrun); + raw_spin_lock(&rt_b->rt_runtime_lock); } + if (idle) + rt_b->rt_period_active = 0; + raw_spin_unlock(&rt_b->rt_runtime_lock); return idle ? HRTIMER_NORESTART : HRTIMER_RESTART; } @@ -52,11 +55,12 @@ static void start_rt_bandwidth(struct rt_bandwidth *rt_b) if (!rt_bandwidth_enabled() || rt_b->rt_runtime == RUNTIME_INF) return; - if (hrtimer_active(&rt_b->rt_period_timer)) - return; - raw_spin_lock(&rt_b->rt_runtime_lock); - start_bandwidth_timer(&rt_b->rt_period_timer, rt_b->rt_period); + if (!rt_b->rt_period_active) { + rt_b->rt_period_active = 1; + hrtimer_forward_now(&rt_b->rt_period_timer, rt_b->rt_period); + hrtimer_start_expires(&rt_b->rt_period_timer, HRTIMER_MODE_ABS_PINNED); + } raw_spin_unlock(&rt_b->rt_runtime_lock); } @@ -1323,7 +1327,7 @@ select_task_rq_rt(struct task_struct *p, int cpu, int sd_flag, int flags) rq = cpu_rq(cpu); rcu_read_lock(); - curr = ACCESS_ONCE(rq->curr); /* unlocked access */ + curr = READ_ONCE(rq->curr); /* unlocked access */ /* * If the current task on @p's runqueue is an RT task, then diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index e0e1299..aea7c1f 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -26,8 +26,14 @@ extern __read_mostly int scheduler_running; extern unsigned long calc_load_update; extern atomic_long_t calc_load_tasks; +extern void calc_global_load_tick(struct rq *this_rq); extern long calc_load_fold_active(struct rq *this_rq); + +#ifdef CONFIG_SMP extern void update_cpu_load_active(struct rq *this_rq); +#else +static inline void update_cpu_load_active(struct rq *this_rq) { } +#endif /* * Helpers for converting nanosecond timing to jiffy resolution @@ -131,6 +137,7 @@ struct rt_bandwidth { ktime_t rt_period; u64 rt_runtime; struct hrtimer rt_period_timer; + unsigned int rt_period_active; }; void __dl_clear_params(struct task_struct *p); @@ -215,7 +222,7 @@ struct cfs_bandwidth { s64 hierarchical_quota; u64 runtime_expires; - int idle, timer_active; + int idle, period_active; struct hrtimer period_timer, slack_timer; struct list_head throttled_cfs_rq; @@ -306,7 +313,7 @@ extern void init_cfs_bandwidth(struct cfs_bandwidth *cfs_b); extern int sched_group_set_shares(struct task_group *tg, unsigned long shares); extern void __refill_cfs_bandwidth_runtime(struct cfs_bandwidth *cfs_b); -extern void __start_cfs_bandwidth(struct cfs_bandwidth *cfs_b, bool force); +extern void start_cfs_bandwidth(struct cfs_bandwidth *cfs_b); extern void unthrottle_cfs_rq(struct cfs_rq *cfs_rq); extern void free_rt_sched_group(struct task_group *tg); @@ -707,7 +714,7 @@ DECLARE_PER_CPU_SHARED_ALIGNED(struct rq, runqueues); static inline u64 __rq_clock_broken(struct rq *rq) { - return ACCESS_ONCE(rq->clock); + return READ_ONCE(rq->clock); } static inline u64 rq_clock(struct rq *rq) @@ -1284,7 +1291,6 @@ extern void update_max_interval(void); extern void init_sched_dl_class(void); extern void init_sched_rt_class(void); extern void init_sched_fair_class(void); -extern void init_sched_dl_class(void); extern void resched_curr(struct rq *rq); extern void resched_cpu(int cpu); @@ -1298,8 +1304,6 @@ extern void init_dl_task_timer(struct sched_dl_entity *dl_se); unsigned long to_ratio(u64 period, u64 runtime); -extern void update_idle_cpu_load(struct rq *this_rq); - extern void init_task_runnable_average(struct task_struct *p); static inline void add_nr_running(struct rq *rq, unsigned count) @@ -1406,8 +1410,6 @@ static inline void sched_rt_avg_update(struct rq *rq, u64 rt_delta) { } static inline void sched_avg_update(struct rq *rq) { } #endif -extern void start_bandwidth_timer(struct hrtimer *period_timer, ktime_t period); - /* * __task_rq_lock - lock the rq @p resides on. */ diff --git a/kernel/sched/stats.h b/kernel/sched/stats.h index 4ab7043..077ebbd 100644 --- a/kernel/sched/stats.h +++ b/kernel/sched/stats.h @@ -174,7 +174,8 @@ static inline bool cputimer_running(struct task_struct *tsk) { struct thread_group_cputimer *cputimer = &tsk->signal->cputimer; - if (!cputimer->running) + /* Check if cputimer isn't running. This is accessed without locking. */ + if (!READ_ONCE(cputimer->running)) return false; /* @@ -215,9 +216,7 @@ static inline void account_group_user_time(struct task_struct *tsk, if (!cputimer_running(tsk)) return; - raw_spin_lock(&cputimer->lock); - cputimer->cputime.utime += cputime; - raw_spin_unlock(&cputimer->lock); + atomic64_add(cputime, &cputimer->cputime_atomic.utime); } /** @@ -238,9 +237,7 @@ static inline void account_group_system_time(struct task_struct *tsk, if (!cputimer_running(tsk)) return; - raw_spin_lock(&cputimer->lock); - cputimer->cputime.stime += cputime; - raw_spin_unlock(&cputimer->lock); + atomic64_add(cputime, &cputimer->cputime_atomic.stime); } /** @@ -261,7 +258,5 @@ static inline void account_group_exec_runtime(struct task_struct *tsk, if (!cputimer_running(tsk)) return; - raw_spin_lock(&cputimer->lock); - cputimer->cputime.sum_exec_runtime += ns; - raw_spin_unlock(&cputimer->lock); + atomic64_add(ns, &cputimer->cputime_atomic.sum_exec_runtime); } diff --git a/kernel/sched/wait.c b/kernel/sched/wait.c index 852143a..052e026 100644 --- a/kernel/sched/wait.c +++ b/kernel/sched/wait.c @@ -341,7 +341,7 @@ long wait_woken(wait_queue_t *wait, unsigned mode, long timeout) * condition being true _OR_ WQ_FLAG_WOKEN such that we will not miss * an event. */ - set_mb(wait->flags, wait->flags & ~WQ_FLAG_WOKEN); /* B */ + smp_store_mb(wait->flags, wait->flags & ~WQ_FLAG_WOKEN); /* B */ return timeout; } @@ -354,7 +354,7 @@ int woken_wake_function(wait_queue_t *wait, unsigned mode, int sync, void *key) * doesn't imply write barrier and the users expects write * barrier semantics on wakeup functions. The following * smp_wmb() is equivalent to smp_wmb() in try_to_wake_up() - * and is paired with set_mb() in wait_woken(). + * and is paired with smp_store_mb() in wait_woken(). */ smp_wmb(); /* C */ wait->flags |= WQ_FLAG_WOKEN; @@ -601,7 +601,7 @@ EXPORT_SYMBOL(bit_wait_io); __sched int bit_wait_timeout(struct wait_bit_key *word) { - unsigned long now = ACCESS_ONCE(jiffies); + unsigned long now = READ_ONCE(jiffies); if (signal_pending_state(current->state, current)) return 1; if (time_after_eq(now, word->timeout)) @@ -613,7 +613,7 @@ EXPORT_SYMBOL_GPL(bit_wait_timeout); __sched int bit_wait_io_timeout(struct wait_bit_key *word) { - unsigned long now = ACCESS_ONCE(jiffies); + unsigned long now = READ_ONCE(jiffies); if (signal_pending_state(current->state, current)) return 1; if (time_after_eq(now, word->timeout)) diff --git a/kernel/signal.c b/kernel/signal.c index d51c5dd..f19833b 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -245,7 +245,7 @@ static inline void print_dropped_signal(int sig) * RETURNS: * %true if @mask is set, %false if made noop because @task was dying. */ -bool task_set_jobctl_pending(struct task_struct *task, unsigned int mask) +bool task_set_jobctl_pending(struct task_struct *task, unsigned long mask) { BUG_ON(mask & ~(JOBCTL_PENDING_MASK | JOBCTL_STOP_CONSUME | JOBCTL_STOP_SIGMASK | JOBCTL_TRAPPING)); @@ -297,7 +297,7 @@ void task_clear_jobctl_trapping(struct task_struct *task) * CONTEXT: * Must be called with @task->sighand->siglock held. */ -void task_clear_jobctl_pending(struct task_struct *task, unsigned int mask) +void task_clear_jobctl_pending(struct task_struct *task, unsigned long mask) { BUG_ON(mask & ~JOBCTL_PENDING_MASK); @@ -2000,7 +2000,7 @@ static bool do_signal_stop(int signr) struct signal_struct *sig = current->signal; if (!(current->jobctl & JOBCTL_STOP_PENDING)) { - unsigned int gstop = JOBCTL_STOP_PENDING | JOBCTL_STOP_CONSUME; + unsigned long gstop = JOBCTL_STOP_PENDING | JOBCTL_STOP_CONSUME; struct task_struct *t; /* signr will be recorded in task->jobctl for retries */ diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c index 695f0c6..fd643d8 100644 --- a/kernel/stop_machine.c +++ b/kernel/stop_machine.c @@ -211,25 +211,6 @@ static int multi_cpu_stop(void *data) return err; } -struct irq_cpu_stop_queue_work_info { - int cpu1; - int cpu2; - struct cpu_stop_work *work1; - struct cpu_stop_work *work2; -}; - -/* - * This function is always run with irqs and preemption disabled. - * This guarantees that both work1 and work2 get queued, before - * our local migrate thread gets the chance to preempt us. - */ -static void irq_cpu_stop_queue_work(void *arg) -{ - struct irq_cpu_stop_queue_work_info *info = arg; - cpu_stop_queue_work(info->cpu1, info->work1); - cpu_stop_queue_work(info->cpu2, info->work2); -} - /** * stop_two_cpus - stops two cpus * @cpu1: the cpu to stop @@ -245,7 +226,6 @@ int stop_two_cpus(unsigned int cpu1, unsigned int cpu2, cpu_stop_fn_t fn, void * { struct cpu_stop_done done; struct cpu_stop_work work1, work2; - struct irq_cpu_stop_queue_work_info call_args; struct multi_stop_data msdata; preempt_disable(); @@ -262,13 +242,6 @@ int stop_two_cpus(unsigned int cpu1, unsigned int cpu2, cpu_stop_fn_t fn, void * .done = &done }; - call_args = (struct irq_cpu_stop_queue_work_info){ - .cpu1 = cpu1, - .cpu2 = cpu2, - .work1 = &work1, - .work2 = &work2, - }; - cpu_stop_init_done(&done, 2); set_state(&msdata, MULTI_STOP_PREPARE); @@ -285,16 +258,11 @@ int stop_two_cpus(unsigned int cpu1, unsigned int cpu2, cpu_stop_fn_t fn, void * return -ENOENT; } - lg_local_lock(&stop_cpus_lock); - /* - * Queuing needs to be done by the lowest numbered CPU, to ensure - * that works are always queued in the same order on every CPU. - * This prevents deadlocks. - */ - smp_call_function_single(min(cpu1, cpu2), - &irq_cpu_stop_queue_work, - &call_args, 1); - lg_local_unlock(&stop_cpus_lock); + lg_double_lock(&stop_cpus_lock, cpu1, cpu2); + cpu_stop_queue_work(cpu1, &work1); + cpu_stop_queue_work(cpu2, &work2); + lg_double_unlock(&stop_cpus_lock, cpu1, cpu2); + preempt_enable(); wait_for_completion(&done.completion); diff --git a/kernel/sys.c b/kernel/sys.c index a4e372b..8571296 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -92,10 +92,10 @@ # define SET_TSC_CTL(a) (-EINVAL) #endif #ifndef MPX_ENABLE_MANAGEMENT -# define MPX_ENABLE_MANAGEMENT(a) (-EINVAL) +# define MPX_ENABLE_MANAGEMENT() (-EINVAL) #endif #ifndef MPX_DISABLE_MANAGEMENT -# define MPX_DISABLE_MANAGEMENT(a) (-EINVAL) +# define MPX_DISABLE_MANAGEMENT() (-EINVAL) #endif #ifndef GET_FP_MODE # define GET_FP_MODE(a) (-EINVAL) @@ -2230,12 +2230,12 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, case PR_MPX_ENABLE_MANAGEMENT: if (arg2 || arg3 || arg4 || arg5) return -EINVAL; - error = MPX_ENABLE_MANAGEMENT(me); + error = MPX_ENABLE_MANAGEMENT(); break; case PR_MPX_DISABLE_MANAGEMENT: if (arg2 || arg3 || arg4 || arg5) return -EINVAL; - error = MPX_DISABLE_MANAGEMENT(me); + error = MPX_DISABLE_MANAGEMENT(); break; case PR_SET_FP_MODE: error = SET_FP_MODE(me, arg2); diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 2082b1a..b13e9d2 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -349,15 +349,6 @@ static struct ctl_table kern_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, - { - .procname = "timer_migration", - .data = &sysctl_timer_migration, - .maxlen = sizeof(unsigned int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &zero, - .extra2 = &one, - }, #endif /* CONFIG_SMP */ #ifdef CONFIG_NUMA_BALANCING { @@ -1132,6 +1123,15 @@ static struct ctl_table kern_table[] = { .extra1 = &zero, .extra2 = &one, }, +#if defined(CONFIG_SMP) && defined(CONFIG_NO_HZ_COMMON) + { + .procname = "timer_migration", + .data = &sysctl_timer_migration, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = timer_migration_handler, + }, +#endif { } }; diff --git a/kernel/time/Makefile b/kernel/time/Makefile index 01f0312..ffc4cc3 100644 --- a/kernel/time/Makefile +++ b/kernel/time/Makefile @@ -13,19 +13,4 @@ obj-$(CONFIG_TIMER_STATS) += timer_stats.o obj-$(CONFIG_DEBUG_FS) += timekeeping_debug.o obj-$(CONFIG_TEST_UDELAY) += test_udelay.o -$(obj)/time.o: $(obj)/timeconst.h - -quiet_cmd_hzfile = HZFILE $@ - cmd_hzfile = echo "hz=$(CONFIG_HZ)" > $@ - -targets += hz.bc -$(obj)/hz.bc: $(objtree)/include/config/hz.h FORCE - $(call if_changed,hzfile) - -quiet_cmd_bc = BC $@ - cmd_bc = bc -q $(filter-out FORCE,$^) > $@ - -targets += timeconst.h -$(obj)/timeconst.h: $(obj)/hz.bc $(src)/timeconst.bc FORCE - $(call if_changed,bc) - +$(obj)/time.o: $(objtree)/include/config/ diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index 1b001ed..7fbba63 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -317,19 +317,16 @@ EXPORT_SYMBOL_GPL(alarm_init); * @alarm: ptr to alarm to set * @start: time to run the alarm */ -int alarm_start(struct alarm *alarm, ktime_t start) +void alarm_start(struct alarm *alarm, ktime_t start) { struct alarm_base *base = &alarm_bases[alarm->type]; unsigned long flags; - int ret; spin_lock_irqsave(&base->lock, flags); alarm->node.expires = start; alarmtimer_enqueue(base, alarm); - ret = hrtimer_start(&alarm->timer, alarm->node.expires, - HRTIMER_MODE_ABS); + hrtimer_start(&alarm->timer, alarm->node.expires, HRTIMER_MODE_ABS); spin_unlock_irqrestore(&base->lock, flags); - return ret; } EXPORT_SYMBOL_GPL(alarm_start); @@ -338,12 +335,12 @@ EXPORT_SYMBOL_GPL(alarm_start); * @alarm: ptr to alarm to set * @start: time relative to now to run the alarm */ -int alarm_start_relative(struct alarm *alarm, ktime_t start) +void alarm_start_relative(struct alarm *alarm, ktime_t start) { struct alarm_base *base = &alarm_bases[alarm->type]; start = ktime_add(start, base->gettime()); - return alarm_start(alarm, start); + alarm_start(alarm, start); } EXPORT_SYMBOL_GPL(alarm_start_relative); @@ -495,12 +492,12 @@ static enum alarmtimer_restart alarm_handle_timer(struct alarm *alarm, */ static int alarm_clock_getres(const clockid_t which_clock, struct timespec *tp) { - clockid_t baseid = alarm_bases[clock2alarm(which_clock)].base_clockid; - if (!alarmtimer_get_rtcdev()) return -EINVAL; - return hrtimer_get_res(baseid, tp); + tp->tv_sec = 0; + tp->tv_nsec = hrtimer_resolution; + return 0; } /** diff --git a/kernel/time/clockevents.c b/kernel/time/clockevents.c index 637a094..08ccc3d 100644 --- a/kernel/time/clockevents.c +++ b/kernel/time/clockevents.c @@ -94,8 +94,8 @@ 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) +static int __clockevents_switch_state(struct clock_event_device *dev, + enum clock_event_state state) { /* Transition with legacy set_mode() callback */ if (dev->set_mode) { @@ -134,32 +134,44 @@ static int __clockevents_set_state(struct clock_event_device *dev, return -ENOSYS; return dev->set_state_oneshot(dev); + case CLOCK_EVT_STATE_ONESHOT_STOPPED: + /* Core internal bug */ + if (WARN_ONCE(!clockevent_state_oneshot(dev), + "Current state: %d\n", + clockevent_get_state(dev))) + return -EINVAL; + + if (dev->set_state_oneshot_stopped) + return dev->set_state_oneshot_stopped(dev); + else + return -ENOSYS; + default: return -ENOSYS; } } /** - * clockevents_set_state - set the operating state of a clock event device + * clockevents_switch_state - set the operating state of a clock event device * @dev: device to modify * @state: new state * * Must be called with interrupts disabled ! */ -void clockevents_set_state(struct clock_event_device *dev, - enum clock_event_state state) +void clockevents_switch_state(struct clock_event_device *dev, + enum clock_event_state state) { - if (dev->state != state) { - if (__clockevents_set_state(dev, state)) + if (clockevent_get_state(dev) != state) { + if (__clockevents_switch_state(dev, state)) return; - dev->state = state; + clockevent_set_state(dev, state); /* * A nsec2cyc multiplicator of 0 is invalid and we'd crash * on it, so fix it up and emit a warning: */ - if (state == CLOCK_EVT_STATE_ONESHOT) { + if (clockevent_state_oneshot(dev)) { if (unlikely(!dev->mult)) { dev->mult = 1; WARN_ON(1); @@ -174,7 +186,7 @@ void clockevents_set_state(struct clock_event_device *dev, */ void clockevents_shutdown(struct clock_event_device *dev) { - clockevents_set_state(dev, CLOCK_EVT_STATE_SHUTDOWN); + clockevents_switch_state(dev, CLOCK_EVT_STATE_SHUTDOWN); dev->next_event.tv64 = KTIME_MAX; } @@ -248,7 +260,7 @@ 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->state == CLOCK_EVT_STATE_SHUTDOWN) + if (clockevent_state_shutdown(dev)) return 0; dev->retries++; @@ -285,7 +297,7 @@ 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->state == CLOCK_EVT_STATE_SHUTDOWN) + if (clockevent_state_shutdown(dev)) return 0; dev->retries++; @@ -317,9 +329,13 @@ int clockevents_program_event(struct clock_event_device *dev, ktime_t expires, dev->next_event = expires; - if (dev->state == CLOCK_EVT_STATE_SHUTDOWN) + if (clockevent_state_shutdown(dev)) return 0; + /* We must be in ONESHOT state here */ + WARN_ONCE(!clockevent_state_oneshot(dev), "Current state: %d\n", + clockevent_get_state(dev)); + /* Shortcut for clockevent devices that can deal with ktime. */ if (dev->features & CLOCK_EVT_FEAT_KTIME) return dev->set_next_ktime(expires, dev); @@ -362,7 +378,7 @@ 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->state != CLOCK_EVT_STATE_DETACHED) + if (dev == ced || !clockevent_state_detached(dev)) continue; if (!tick_check_replacement(newdev, dev)) @@ -388,7 +404,7 @@ 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->state == CLOCK_EVT_STATE_DETACHED) { + if (clockevent_state_detached(ced)) { list_del_init(&ced->list); return 0; } @@ -445,7 +461,8 @@ static int clockevents_sanity_check(struct clock_event_device *dev) 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); + dev->set_state_shutdown || dev->tick_resume || + dev->set_state_oneshot_stopped); BUG_ON(dev->mode != CLOCK_EVT_MODE_UNUSED); return 0; @@ -480,7 +497,7 @@ void clockevents_register_device(struct clock_event_device *dev) BUG_ON(clockevents_sanity_check(dev)); /* Initialize state to DETACHED */ - dev->state = CLOCK_EVT_STATE_DETACHED; + clockevent_set_state(dev, CLOCK_EVT_STATE_DETACHED); if (!dev->cpumask) { WARN_ON(num_possible_cpus() > 1); @@ -545,11 +562,11 @@ int __clockevents_update_freq(struct clock_event_device *dev, u32 freq) { clockevents_config(dev, freq); - if (dev->state == CLOCK_EVT_STATE_ONESHOT) + if (clockevent_state_oneshot(dev)) return clockevents_program_event(dev, dev->next_event, false); - if (dev->state == CLOCK_EVT_STATE_PERIODIC) - return __clockevents_set_state(dev, CLOCK_EVT_STATE_PERIODIC); + if (clockevent_state_periodic(dev)) + return __clockevents_switch_state(dev, CLOCK_EVT_STATE_PERIODIC); return 0; } @@ -603,13 +620,13 @@ void clockevents_exchange_device(struct clock_event_device *old, */ if (old) { module_put(old->owner); - clockevents_set_state(old, CLOCK_EVT_STATE_DETACHED); + clockevents_switch_state(old, CLOCK_EVT_STATE_DETACHED); list_del(&old->list); list_add(&old->list, &clockevents_released); } if (new) { - BUG_ON(new->state != CLOCK_EVT_STATE_DETACHED); + BUG_ON(!clockevent_state_detached(new)); clockevents_shutdown(new); } } @@ -622,7 +639,7 @@ void clockevents_suspend(void) struct clock_event_device *dev; list_for_each_entry_reverse(dev, &clockevent_devices, list) - if (dev->suspend) + if (dev->suspend && !clockevent_state_detached(dev)) dev->suspend(dev); } @@ -634,7 +651,7 @@ void clockevents_resume(void) struct clock_event_device *dev; list_for_each_entry(dev, &clockevent_devices, list) - if (dev->resume) + if (dev->resume && !clockevent_state_detached(dev)) dev->resume(dev); } @@ -665,7 +682,7 @@ void tick_cleanup_dead_cpu(int cpu) 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); + BUG_ON(!clockevent_state_detached(dev)); list_del(&dev->list); } } diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c index 15facb1..841b72f 100644 --- a/kernel/time/clocksource.c +++ b/kernel/time/clocksource.c @@ -23,6 +23,8 @@ * o Allow clocksource drivers to be unregistered */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/device.h> #include <linux/clocksource.h> #include <linux/init.h> @@ -216,10 +218,11 @@ static void clocksource_watchdog(unsigned long data) /* Check the deviation from the watchdog clocksource. */ if ((abs(cs_nsec - wd_nsec) > WATCHDOG_THRESHOLD)) { - 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", + 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", + pr_warn(" '%s' cs_now: %llx cs_last: %llx mask: %llx\n", cs->name, csnow, cslast, cs->mask); __clocksource_unstable(cs); continue; @@ -567,9 +570,8 @@ static void __clocksource_select(bool skipcur) */ if (!(cs->flags & CLOCK_SOURCE_VALID_FOR_HRES) && oneshot) { /* Override clocksource cannot be used. */ - printk(KERN_WARNING "Override clocksource %s is not " - "HRT compatible. Cannot switch while in " - "HRT/NOHZ mode\n", cs->name); + pr_warn("Override clocksource %s is not HRT compatible - cannot switch while in HRT/NOHZ mode\n", + cs->name); override_name[0] = 0; } else /* Override clocksource can be used. */ @@ -708,8 +710,8 @@ void __clocksource_update_freq_scale(struct clocksource *cs, u32 scale, u32 freq 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); + pr_info("%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_update_freq_scale); @@ -1008,12 +1010,10 @@ __setup("clocksource=", boot_override_clocksource); static int __init boot_override_clock(char* str) { if (!strcmp(str, "pmtmr")) { - printk("Warning: clock=pmtmr is deprecated. " - "Use clocksource=acpi_pm.\n"); + pr_warn("clock=pmtmr is deprecated - use clocksource=acpi_pm\n"); return boot_override_clocksource("acpi_pm"); } - printk("Warning! clock= boot option is deprecated. " - "Use clocksource=xyz\n"); + pr_warn("clock= boot option is deprecated - use clocksource=xyz\n"); return boot_override_clocksource(str); } diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c index 93ef7190..5c7ae4b 100644 --- a/kernel/time/hrtimer.c +++ b/kernel/time/hrtimer.c @@ -66,33 +66,29 @@ */ DEFINE_PER_CPU(struct hrtimer_cpu_base, hrtimer_bases) = { - .lock = __RAW_SPIN_LOCK_UNLOCKED(hrtimer_bases.lock), + .seq = SEQCNT_ZERO(hrtimer_bases.seq), .clock_base = { { .index = HRTIMER_BASE_MONOTONIC, .clockid = CLOCK_MONOTONIC, .get_time = &ktime_get, - .resolution = KTIME_LOW_RES, }, { .index = HRTIMER_BASE_REALTIME, .clockid = CLOCK_REALTIME, .get_time = &ktime_get_real, - .resolution = KTIME_LOW_RES, }, { .index = HRTIMER_BASE_BOOTTIME, .clockid = CLOCK_BOOTTIME, .get_time = &ktime_get_boottime, - .resolution = KTIME_LOW_RES, }, { .index = HRTIMER_BASE_TAI, .clockid = CLOCK_TAI, .get_time = &ktime_get_clocktai, - .resolution = KTIME_LOW_RES, }, } }; @@ -109,27 +105,6 @@ static inline int hrtimer_clockid_to_base(clockid_t clock_id) return hrtimer_clock_to_base_table[clock_id]; } - -/* - * Get the coarse grained time at the softirq based on xtime and - * wall_to_monotonic. - */ -static void hrtimer_get_softirq_time(struct hrtimer_cpu_base *base) -{ - ktime_t xtim, mono, boot, tai; - ktime_t off_real, off_boot, off_tai; - - mono = ktime_get_update_offsets_tick(&off_real, &off_boot, &off_tai); - boot = ktime_add(mono, off_boot); - xtim = ktime_add(mono, off_real); - tai = ktime_add(mono, off_tai); - - base->clock_base[HRTIMER_BASE_REALTIME].softirq_time = xtim; - base->clock_base[HRTIMER_BASE_MONOTONIC].softirq_time = mono; - base->clock_base[HRTIMER_BASE_BOOTTIME].softirq_time = boot; - base->clock_base[HRTIMER_BASE_TAI].softirq_time = tai; -} - /* * Functions and macros which are different for UP/SMP systems are kept in a * single place @@ -137,6 +112,18 @@ static void hrtimer_get_softirq_time(struct hrtimer_cpu_base *base) #ifdef CONFIG_SMP /* + * We require the migration_base for lock_hrtimer_base()/switch_hrtimer_base() + * such that hrtimer_callback_running() can unconditionally dereference + * timer->base->cpu_base + */ +static struct hrtimer_cpu_base migration_cpu_base = { + .seq = SEQCNT_ZERO(migration_cpu_base), + .clock_base = { { .cpu_base = &migration_cpu_base, }, }, +}; + +#define migration_base migration_cpu_base.clock_base[0] + +/* * We are using hashed locking: holding per_cpu(hrtimer_bases)[n].lock * means that all timers which are tied to this base via timer->base are * locked, and the base itself is locked too. @@ -145,8 +132,8 @@ static void hrtimer_get_softirq_time(struct hrtimer_cpu_base *base) * be found on the lists/queues. * * When the timer's base is locked, and the timer removed from list, it is - * possible to set timer->base = NULL and drop the lock: the timer remains - * locked. + * possible to set timer->base = &migration_base and drop the lock: the timer + * remains locked. */ static struct hrtimer_clock_base *lock_hrtimer_base(const struct hrtimer *timer, @@ -156,7 +143,7 @@ struct hrtimer_clock_base *lock_hrtimer_base(const struct hrtimer *timer, for (;;) { base = timer->base; - if (likely(base != NULL)) { + if (likely(base != &migration_base)) { raw_spin_lock_irqsave(&base->cpu_base->lock, *flags); if (likely(base == timer->base)) return base; @@ -190,6 +177,24 @@ hrtimer_check_target(struct hrtimer *timer, struct hrtimer_clock_base *new_base) #endif } +#if defined(CONFIG_SMP) && defined(CONFIG_NO_HZ_COMMON) +static inline +struct hrtimer_cpu_base *get_target_base(struct hrtimer_cpu_base *base, + int pinned) +{ + if (pinned || !base->migration_enabled) + return this_cpu_ptr(&hrtimer_bases); + return &per_cpu(hrtimer_bases, get_nohz_timer_target()); +} +#else +static inline +struct hrtimer_cpu_base *get_target_base(struct hrtimer_cpu_base *base, + int pinned) +{ + return this_cpu_ptr(&hrtimer_bases); +} +#endif + /* * Switch the timer base to the current CPU when possible. */ @@ -197,14 +202,13 @@ static inline struct hrtimer_clock_base * switch_hrtimer_base(struct hrtimer *timer, struct hrtimer_clock_base *base, int pinned) { + struct hrtimer_cpu_base *new_cpu_base, *this_base; struct hrtimer_clock_base *new_base; - struct hrtimer_cpu_base *new_cpu_base; - int this_cpu = smp_processor_id(); - int cpu = get_nohz_timer_target(pinned); int basenum = base->index; + this_base = this_cpu_ptr(&hrtimer_bases); + new_cpu_base = get_target_base(this_base, pinned); again: - new_cpu_base = &per_cpu(hrtimer_bases, cpu); new_base = &new_cpu_base->clock_base[basenum]; if (base != new_base) { @@ -220,22 +224,24 @@ again: if (unlikely(hrtimer_callback_running(timer))) return base; - /* See the comment in lock_timer_base() */ - timer->base = NULL; + /* See the comment in lock_hrtimer_base() */ + timer->base = &migration_base; raw_spin_unlock(&base->cpu_base->lock); raw_spin_lock(&new_base->cpu_base->lock); - if (cpu != this_cpu && hrtimer_check_target(timer, new_base)) { - cpu = this_cpu; + if (new_cpu_base != this_base && + hrtimer_check_target(timer, new_base)) { raw_spin_unlock(&new_base->cpu_base->lock); raw_spin_lock(&base->cpu_base->lock); + new_cpu_base = this_base; timer->base = base; goto again; } timer->base = new_base; } else { - if (cpu != this_cpu && hrtimer_check_target(timer, new_base)) { - cpu = this_cpu; + if (new_cpu_base != this_base && + hrtimer_check_target(timer, new_base)) { + new_cpu_base = this_base; goto again; } } @@ -443,24 +449,35 @@ static inline void debug_deactivate(struct hrtimer *timer) } #if defined(CONFIG_NO_HZ_COMMON) || defined(CONFIG_HIGH_RES_TIMERS) +static inline void hrtimer_update_next_timer(struct hrtimer_cpu_base *cpu_base, + struct hrtimer *timer) +{ +#ifdef CONFIG_HIGH_RES_TIMERS + cpu_base->next_timer = timer; +#endif +} + static ktime_t __hrtimer_get_next_event(struct hrtimer_cpu_base *cpu_base) { struct hrtimer_clock_base *base = cpu_base->clock_base; ktime_t expires, expires_next = { .tv64 = KTIME_MAX }; - int i; + unsigned int active = cpu_base->active_bases; - for (i = 0; i < HRTIMER_MAX_CLOCK_BASES; i++, base++) { + hrtimer_update_next_timer(cpu_base, NULL); + for (; active; base++, active >>= 1) { struct timerqueue_node *next; struct hrtimer *timer; - next = timerqueue_getnext(&base->active); - if (!next) + if (!(active & 0x01)) continue; + next = timerqueue_getnext(&base->active); timer = container_of(next, struct hrtimer, node); expires = ktime_sub(hrtimer_get_expires(timer), base->offset); - if (expires.tv64 < expires_next.tv64) + if (expires.tv64 < expires_next.tv64) { expires_next = expires; + hrtimer_update_next_timer(cpu_base, timer); + } } /* * clock_was_set() might have changed base->offset of any of @@ -473,6 +490,16 @@ static ktime_t __hrtimer_get_next_event(struct hrtimer_cpu_base *cpu_base) } #endif +static inline ktime_t hrtimer_update_base(struct hrtimer_cpu_base *base) +{ + ktime_t *offs_real = &base->clock_base[HRTIMER_BASE_REALTIME].offset; + ktime_t *offs_boot = &base->clock_base[HRTIMER_BASE_BOOTTIME].offset; + ktime_t *offs_tai = &base->clock_base[HRTIMER_BASE_TAI].offset; + + return ktime_get_update_offsets_now(&base->clock_was_set_seq, + offs_real, offs_boot, offs_tai); +} + /* High resolution timer related functions */ #ifdef CONFIG_HIGH_RES_TIMERS @@ -480,6 +507,8 @@ static ktime_t __hrtimer_get_next_event(struct hrtimer_cpu_base *cpu_base) * High resolution timer enabled ? */ static int hrtimer_hres_enabled __read_mostly = 1; +unsigned int hrtimer_resolution __read_mostly = LOW_RES_NSEC; +EXPORT_SYMBOL_GPL(hrtimer_resolution); /* * Enable / Disable high resolution mode @@ -508,9 +537,14 @@ static inline int hrtimer_is_hres_enabled(void) /* * Is the high resolution mode active ? */ +static inline int __hrtimer_hres_active(struct hrtimer_cpu_base *cpu_base) +{ + return cpu_base->hres_active; +} + static inline int hrtimer_hres_active(void) { - return __this_cpu_read(hrtimer_bases.hres_active); + return __hrtimer_hres_active(this_cpu_ptr(&hrtimer_bases)); } /* @@ -521,7 +555,12 @@ static inline int hrtimer_hres_active(void) static void hrtimer_force_reprogram(struct hrtimer_cpu_base *cpu_base, int skip_equal) { - ktime_t expires_next = __hrtimer_get_next_event(cpu_base); + ktime_t expires_next; + + if (!cpu_base->hres_active) + return; + + expires_next = __hrtimer_get_next_event(cpu_base); if (skip_equal && expires_next.tv64 == cpu_base->expires_next.tv64) return; @@ -545,63 +584,53 @@ hrtimer_force_reprogram(struct hrtimer_cpu_base *cpu_base, int skip_equal) if (cpu_base->hang_detected) return; - if (cpu_base->expires_next.tv64 != KTIME_MAX) - tick_program_event(cpu_base->expires_next, 1); + tick_program_event(cpu_base->expires_next, 1); } /* - * Shared reprogramming for clock_realtime and clock_monotonic - * * When a timer is enqueued and expires earlier than the already enqueued * timers, we have to check, whether it expires earlier than the timer for * which the clock event device was armed. * - * Note, that in case the state has HRTIMER_STATE_CALLBACK set, no reprogramming - * and no expiry check happens. The timer gets enqueued into the rbtree. The - * reprogramming and expiry check is done in the hrtimer_interrupt or in the - * softirq. - * * Called with interrupts disabled and base->cpu_base.lock held */ -static int hrtimer_reprogram(struct hrtimer *timer, - struct hrtimer_clock_base *base) +static void hrtimer_reprogram(struct hrtimer *timer, + struct hrtimer_clock_base *base) { struct hrtimer_cpu_base *cpu_base = this_cpu_ptr(&hrtimer_bases); ktime_t expires = ktime_sub(hrtimer_get_expires(timer), base->offset); - int res; WARN_ON_ONCE(hrtimer_get_expires_tv64(timer) < 0); /* - * When the callback is running, we do not reprogram the clock event - * device. The timer callback is either running on a different CPU or - * the callback is executed in the hrtimer_interrupt context. The - * reprogramming is handled either by the softirq, which called the - * callback or at the end of the hrtimer_interrupt. + * If the timer is not on the current cpu, we cannot reprogram + * the other cpus clock event device. */ - if (hrtimer_callback_running(timer)) - return 0; + if (base->cpu_base != cpu_base) + return; + + /* + * If the hrtimer interrupt is running, then it will + * reevaluate the clock bases and reprogram the clock event + * device. The callbacks are always executed in hard interrupt + * context so we don't need an extra check for a running + * callback. + */ + if (cpu_base->in_hrtirq) + return; /* * CLOCK_REALTIME timer might be requested with an absolute - * expiry time which is less than base->offset. Nothing wrong - * about that, just avoid to call into the tick code, which - * has now objections against negative expiry values. + * expiry time which is less than base->offset. Set it to 0. */ if (expires.tv64 < 0) - return -ETIME; + expires.tv64 = 0; if (expires.tv64 >= cpu_base->expires_next.tv64) - return 0; + return; - /* - * When the target cpu of the timer is currently executing - * hrtimer_interrupt(), then we do not touch the clock event - * device. hrtimer_interrupt() will reevaluate all clock bases - * before reprogramming the device. - */ - if (cpu_base->in_hrtirq) - return 0; + /* Update the pointer to the next expiring timer */ + cpu_base->next_timer = timer; /* * If a hang was detected in the last timer interrupt then we @@ -610,15 +639,14 @@ static int hrtimer_reprogram(struct hrtimer *timer, * to make progress. */ if (cpu_base->hang_detected) - return 0; + return; /* - * Clockevents returns -ETIME, when the event was in the past. + * Program the timer hardware. We enforce the expiry for + * events which are already in the past. */ - res = tick_program_event(expires, 0); - if (!IS_ERR_VALUE(res)) - cpu_base->expires_next = expires; - return res; + cpu_base->expires_next = expires; + tick_program_event(expires, 1); } /* @@ -630,15 +658,6 @@ static inline void hrtimer_init_hres(struct hrtimer_cpu_base *base) base->hres_active = 0; } -static inline ktime_t hrtimer_update_base(struct hrtimer_cpu_base *base) -{ - ktime_t *offs_real = &base->clock_base[HRTIMER_BASE_REALTIME].offset; - ktime_t *offs_boot = &base->clock_base[HRTIMER_BASE_BOOTTIME].offset; - ktime_t *offs_tai = &base->clock_base[HRTIMER_BASE_TAI].offset; - - return ktime_get_update_offsets_now(offs_real, offs_boot, offs_tai); -} - /* * Retrigger next event is called after clock was set * @@ -648,7 +667,7 @@ static void retrigger_next_event(void *arg) { struct hrtimer_cpu_base *base = this_cpu_ptr(&hrtimer_bases); - if (!hrtimer_hres_active()) + if (!base->hres_active) return; raw_spin_lock(&base->lock); @@ -662,29 +681,19 @@ static void retrigger_next_event(void *arg) */ static int hrtimer_switch_to_hres(void) { - int i, cpu = smp_processor_id(); - struct hrtimer_cpu_base *base = &per_cpu(hrtimer_bases, cpu); - unsigned long flags; - - if (base->hres_active) - return 1; - - local_irq_save(flags); + struct hrtimer_cpu_base *base = this_cpu_ptr(&hrtimer_bases); if (tick_init_highres()) { - local_irq_restore(flags); printk(KERN_WARNING "Could not switch to high resolution " - "mode on CPU %d\n", cpu); + "mode on CPU %d\n", base->cpu); return 0; } base->hres_active = 1; - for (i = 0; i < HRTIMER_MAX_CLOCK_BASES; i++) - base->clock_base[i].resolution = KTIME_HIGH_RES; + hrtimer_resolution = HIGH_RES_NSEC; tick_setup_sched_timer(); /* "Retrigger" the interrupt to get things going */ retrigger_next_event(NULL); - local_irq_restore(flags); return 1; } @@ -706,6 +715,7 @@ void clock_was_set_delayed(void) #else +static inline int __hrtimer_hres_active(struct hrtimer_cpu_base *b) { return 0; } static inline int hrtimer_hres_active(void) { return 0; } static inline int hrtimer_is_hres_enabled(void) { return 0; } static inline int hrtimer_switch_to_hres(void) { return 0; } @@ -803,6 +813,14 @@ void unlock_hrtimer_base(const struct hrtimer *timer, unsigned long *flags) * * Forward the timer expiry so it will expire in the future. * Returns the number of overruns. + * + * Can be safely called from the callback function of @timer. If + * called from other contexts @timer must neither be enqueued nor + * running the callback and the caller needs to take care of + * serialization. + * + * Note: This only updates the timer expiry value and does not requeue + * the timer. */ u64 hrtimer_forward(struct hrtimer *timer, ktime_t now, ktime_t interval) { @@ -814,8 +832,11 @@ u64 hrtimer_forward(struct hrtimer *timer, ktime_t now, ktime_t interval) if (delta.tv64 < 0) return 0; - if (interval.tv64 < timer->base->resolution.tv64) - interval.tv64 = timer->base->resolution.tv64; + if (WARN_ON(timer->state & HRTIMER_STATE_ENQUEUED)) + return 0; + + if (interval.tv64 < hrtimer_resolution) + interval.tv64 = hrtimer_resolution; if (unlikely(delta.tv64 >= interval.tv64)) { s64 incr = ktime_to_ns(interval); @@ -849,16 +870,11 @@ static int enqueue_hrtimer(struct hrtimer *timer, { debug_activate(timer); - timerqueue_add(&base->active, &timer->node); base->cpu_base->active_bases |= 1 << base->index; - /* - * HRTIMER_STATE_ENQUEUED is or'ed to the current state to preserve the - * state of a possibly running callback. - */ - timer->state |= HRTIMER_STATE_ENQUEUED; + timer->state = HRTIMER_STATE_ENQUEUED; - return (&timer->node == base->active.next); + return timerqueue_add(&base->active, &timer->node); } /* @@ -875,39 +891,38 @@ static void __remove_hrtimer(struct hrtimer *timer, struct hrtimer_clock_base *base, unsigned long newstate, int reprogram) { - struct timerqueue_node *next_timer; - if (!(timer->state & HRTIMER_STATE_ENQUEUED)) - goto out; + struct hrtimer_cpu_base *cpu_base = base->cpu_base; + unsigned int state = timer->state; + + timer->state = newstate; + if (!(state & HRTIMER_STATE_ENQUEUED)) + return; + + if (!timerqueue_del(&base->active, &timer->node)) + cpu_base->active_bases &= ~(1 << base->index); - next_timer = timerqueue_getnext(&base->active); - timerqueue_del(&base->active, &timer->node); - if (&timer->node == next_timer) { #ifdef CONFIG_HIGH_RES_TIMERS - /* Reprogram the clock event device. if enabled */ - if (reprogram && hrtimer_hres_active()) { - ktime_t expires; - - expires = ktime_sub(hrtimer_get_expires(timer), - base->offset); - if (base->cpu_base->expires_next.tv64 == expires.tv64) - hrtimer_force_reprogram(base->cpu_base, 1); - } + /* + * Note: If reprogram is false we do not update + * cpu_base->next_timer. This happens when we remove the first + * timer on a remote cpu. No harm as we never dereference + * cpu_base->next_timer. So the worst thing what can happen is + * an superflous call to hrtimer_force_reprogram() on the + * remote cpu later on if the same timer gets enqueued again. + */ + if (reprogram && timer == cpu_base->next_timer) + hrtimer_force_reprogram(cpu_base, 1); #endif - } - if (!timerqueue_getnext(&base->active)) - base->cpu_base->active_bases &= ~(1 << base->index); -out: - timer->state = newstate; } /* * remove hrtimer, called with base lock held */ static inline int -remove_hrtimer(struct hrtimer *timer, struct hrtimer_clock_base *base) +remove_hrtimer(struct hrtimer *timer, struct hrtimer_clock_base *base, bool restart) { if (hrtimer_is_queued(timer)) { - unsigned long state; + unsigned long state = timer->state; int reprogram; /* @@ -921,30 +936,35 @@ remove_hrtimer(struct hrtimer *timer, struct hrtimer_clock_base *base) debug_deactivate(timer); timer_stats_hrtimer_clear_start_info(timer); reprogram = base->cpu_base == this_cpu_ptr(&hrtimer_bases); - /* - * We must preserve the CALLBACK state flag here, - * otherwise we could move the timer base in - * switch_hrtimer_base. - */ - state = timer->state & HRTIMER_STATE_CALLBACK; + + if (!restart) + state = HRTIMER_STATE_INACTIVE; + __remove_hrtimer(timer, base, state, reprogram); return 1; } return 0; } -int __hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, - unsigned long delta_ns, const enum hrtimer_mode mode, - int wakeup) +/** + * hrtimer_start_range_ns - (re)start an hrtimer on the current CPU + * @timer: the timer to be added + * @tim: expiry time + * @delta_ns: "slack" range for the timer + * @mode: expiry mode: absolute (HRTIMER_MODE_ABS) or + * relative (HRTIMER_MODE_REL) + */ +void hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, + unsigned long delta_ns, const enum hrtimer_mode mode) { struct hrtimer_clock_base *base, *new_base; unsigned long flags; - int ret, leftmost; + int leftmost; base = lock_hrtimer_base(timer, &flags); /* Remove an active timer from the queue: */ - ret = remove_hrtimer(timer, base); + remove_hrtimer(timer, base, true); if (mode & HRTIMER_MODE_REL) { tim = ktime_add_safe(tim, base->get_time()); @@ -956,7 +976,7 @@ int __hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, * timeouts. This will go away with the GTOD framework. */ #ifdef CONFIG_TIME_LOW_RES - tim = ktime_add_safe(tim, base->resolution); + tim = ktime_add_safe(tim, ktime_set(0, hrtimer_resolution)); #endif } @@ -968,85 +988,25 @@ int __hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, timer_stats_hrtimer_set_start_info(timer); leftmost = enqueue_hrtimer(timer, new_base); - - if (!leftmost) { - unlock_hrtimer_base(timer, &flags); - return ret; - } + if (!leftmost) + goto unlock; if (!hrtimer_is_hres_active(timer)) { /* * Kick to reschedule the next tick to handle the new timer * on dynticks target. */ - wake_up_nohz_cpu(new_base->cpu_base->cpu); - } else if (new_base->cpu_base == this_cpu_ptr(&hrtimer_bases) && - hrtimer_reprogram(timer, new_base)) { - /* - * Only allow reprogramming if the new base is on this CPU. - * (it might still be on another CPU if the timer was pending) - * - * XXX send_remote_softirq() ? - */ - if (wakeup) { - /* - * We need to drop cpu_base->lock to avoid a - * lock ordering issue vs. rq->lock. - */ - raw_spin_unlock(&new_base->cpu_base->lock); - raise_softirq_irqoff(HRTIMER_SOFTIRQ); - local_irq_restore(flags); - return ret; - } else { - __raise_softirq_irqoff(HRTIMER_SOFTIRQ); - } + if (new_base->cpu_base->nohz_active) + wake_up_nohz_cpu(new_base->cpu_base->cpu); + } else { + hrtimer_reprogram(timer, new_base); } - +unlock: unlock_hrtimer_base(timer, &flags); - - return ret; -} -EXPORT_SYMBOL_GPL(__hrtimer_start_range_ns); - -/** - * hrtimer_start_range_ns - (re)start an hrtimer on the current CPU - * @timer: the timer to be added - * @tim: expiry time - * @delta_ns: "slack" range for the timer - * @mode: expiry mode: absolute (HRTIMER_MODE_ABS) or - * relative (HRTIMER_MODE_REL) - * - * Returns: - * 0 on success - * 1 when the timer was active - */ -int hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, - unsigned long delta_ns, const enum hrtimer_mode mode) -{ - return __hrtimer_start_range_ns(timer, tim, delta_ns, mode, 1); } EXPORT_SYMBOL_GPL(hrtimer_start_range_ns); /** - * hrtimer_start - (re)start an hrtimer on the current CPU - * @timer: the timer to be added - * @tim: expiry time - * @mode: expiry mode: absolute (HRTIMER_MODE_ABS) or - * relative (HRTIMER_MODE_REL) - * - * Returns: - * 0 on success - * 1 when the timer was active - */ -int -hrtimer_start(struct hrtimer *timer, ktime_t tim, const enum hrtimer_mode mode) -{ - return __hrtimer_start_range_ns(timer, tim, 0, mode, 1); -} -EXPORT_SYMBOL_GPL(hrtimer_start); - - -/** * hrtimer_try_to_cancel - try to deactivate a timer * @timer: hrtimer to stop * @@ -1062,10 +1022,19 @@ int hrtimer_try_to_cancel(struct hrtimer *timer) unsigned long flags; int ret = -1; + /* + * Check lockless first. If the timer is not active (neither + * enqueued nor running the callback, nothing to do here. The + * base lock does not serialize against a concurrent enqueue, + * so we can avoid taking it. + */ + if (!hrtimer_active(timer)) + return 0; + base = lock_hrtimer_base(timer, &flags); if (!hrtimer_callback_running(timer)) - ret = remove_hrtimer(timer, base); + ret = remove_hrtimer(timer, base, false); unlock_hrtimer_base(timer, &flags); @@ -1115,26 +1084,22 @@ EXPORT_SYMBOL_GPL(hrtimer_get_remaining); /** * hrtimer_get_next_event - get the time until next expiry event * - * Returns the delta to the next expiry event or KTIME_MAX if no timer - * is pending. + * Returns the next expiry time or KTIME_MAX if no timer is pending. */ -ktime_t hrtimer_get_next_event(void) +u64 hrtimer_get_next_event(void) { struct hrtimer_cpu_base *cpu_base = this_cpu_ptr(&hrtimer_bases); - ktime_t mindelta = { .tv64 = KTIME_MAX }; + u64 expires = KTIME_MAX; unsigned long flags; raw_spin_lock_irqsave(&cpu_base->lock, flags); - if (!hrtimer_hres_active()) - mindelta = ktime_sub(__hrtimer_get_next_event(cpu_base), - ktime_get()); + if (!__hrtimer_hres_active(cpu_base)) + expires = __hrtimer_get_next_event(cpu_base).tv64; raw_spin_unlock_irqrestore(&cpu_base->lock, flags); - if (mindelta.tv64 < 0) - mindelta.tv64 = 0; - return mindelta; + return expires; } #endif @@ -1176,37 +1141,73 @@ void hrtimer_init(struct hrtimer *timer, clockid_t clock_id, } EXPORT_SYMBOL_GPL(hrtimer_init); -/** - * hrtimer_get_res - get the timer resolution for a clock - * @which_clock: which clock to query - * @tp: pointer to timespec variable to store the resolution +/* + * A timer is active, when it is enqueued into the rbtree or the + * callback function is running or it's in the state of being migrated + * to another cpu. * - * Store the resolution of the clock selected by @which_clock in the - * variable pointed to by @tp. + * It is important for this function to not return a false negative. */ -int hrtimer_get_res(const clockid_t which_clock, struct timespec *tp) +bool hrtimer_active(const struct hrtimer *timer) { struct hrtimer_cpu_base *cpu_base; - int base = hrtimer_clockid_to_base(which_clock); + unsigned int seq; - cpu_base = raw_cpu_ptr(&hrtimer_bases); - *tp = ktime_to_timespec(cpu_base->clock_base[base].resolution); + do { + cpu_base = READ_ONCE(timer->base->cpu_base); + seq = raw_read_seqcount_begin(&cpu_base->seq); - return 0; + if (timer->state != HRTIMER_STATE_INACTIVE || + cpu_base->running == timer) + return true; + + } while (read_seqcount_retry(&cpu_base->seq, seq) || + cpu_base != READ_ONCE(timer->base->cpu_base)); + + return false; } -EXPORT_SYMBOL_GPL(hrtimer_get_res); +EXPORT_SYMBOL_GPL(hrtimer_active); -static void __run_hrtimer(struct hrtimer *timer, ktime_t *now) +/* + * The write_seqcount_barrier()s in __run_hrtimer() split the thing into 3 + * distinct sections: + * + * - queued: the timer is queued + * - callback: the timer is being ran + * - post: the timer is inactive or (re)queued + * + * On the read side we ensure we observe timer->state and cpu_base->running + * from the same section, if anything changed while we looked at it, we retry. + * This includes timer->base changing because sequence numbers alone are + * insufficient for that. + * + * The sequence numbers are required because otherwise we could still observe + * a false negative if the read side got smeared over multiple consequtive + * __run_hrtimer() invocations. + */ + +static void __run_hrtimer(struct hrtimer_cpu_base *cpu_base, + struct hrtimer_clock_base *base, + struct hrtimer *timer, ktime_t *now) { - struct hrtimer_clock_base *base = timer->base; - struct hrtimer_cpu_base *cpu_base = base->cpu_base; enum hrtimer_restart (*fn)(struct hrtimer *); int restart; - WARN_ON(!irqs_disabled()); + lockdep_assert_held(&cpu_base->lock); debug_deactivate(timer); - __remove_hrtimer(timer, base, HRTIMER_STATE_CALLBACK, 0); + cpu_base->running = timer; + + /* + * Separate the ->running assignment from the ->state assignment. + * + * As with a regular write barrier, this ensures the read side in + * hrtimer_active() cannot observe cpu_base->running == NULL && + * timer->state == INACTIVE. + */ + raw_write_seqcount_barrier(&cpu_base->seq); + + __remove_hrtimer(timer, base, HRTIMER_STATE_INACTIVE, 0); timer_stats_account_hrtimer(timer); fn = timer->function; @@ -1222,58 +1223,43 @@ static void __run_hrtimer(struct hrtimer *timer, ktime_t *now) raw_spin_lock(&cpu_base->lock); /* - * Note: We clear the CALLBACK bit after enqueue_hrtimer and + * Note: We clear the running state after enqueue_hrtimer and * we do not reprogramm the event hardware. Happens either in * hrtimer_start_range_ns() or in hrtimer_interrupt() + * + * Note: Because we dropped the cpu_base->lock above, + * hrtimer_start_range_ns() can have popped in and enqueued the timer + * for us already. */ - if (restart != HRTIMER_NORESTART) { - BUG_ON(timer->state != HRTIMER_STATE_CALLBACK); + if (restart != HRTIMER_NORESTART && + !(timer->state & HRTIMER_STATE_ENQUEUED)) enqueue_hrtimer(timer, base); - } - WARN_ON_ONCE(!(timer->state & HRTIMER_STATE_CALLBACK)); + /* + * Separate the ->running assignment from the ->state assignment. + * + * As with a regular write barrier, this ensures the read side in + * hrtimer_active() cannot observe cpu_base->running == NULL && + * timer->state == INACTIVE. + */ + raw_write_seqcount_barrier(&cpu_base->seq); - timer->state &= ~HRTIMER_STATE_CALLBACK; + WARN_ON_ONCE(cpu_base->running != timer); + cpu_base->running = NULL; } -#ifdef CONFIG_HIGH_RES_TIMERS - -/* - * High resolution timer interrupt - * Called with interrupts disabled - */ -void hrtimer_interrupt(struct clock_event_device *dev) +static void __hrtimer_run_queues(struct hrtimer_cpu_base *cpu_base, ktime_t now) { - struct hrtimer_cpu_base *cpu_base = this_cpu_ptr(&hrtimer_bases); - ktime_t expires_next, now, entry_time, delta; - int i, retries = 0; - - BUG_ON(!cpu_base->hres_active); - cpu_base->nr_events++; - dev->next_event.tv64 = KTIME_MAX; - - raw_spin_lock(&cpu_base->lock); - entry_time = now = hrtimer_update_base(cpu_base); -retry: - cpu_base->in_hrtirq = 1; - /* - * We set expires_next to KTIME_MAX here with cpu_base->lock - * held to prevent that a timer is enqueued in our queue via - * the migration code. This does not affect enqueueing of - * timers which run their callback and need to be requeued on - * this CPU. - */ - cpu_base->expires_next.tv64 = KTIME_MAX; + struct hrtimer_clock_base *base = cpu_base->clock_base; + unsigned int active = cpu_base->active_bases; - for (i = 0; i < HRTIMER_MAX_CLOCK_BASES; i++) { - struct hrtimer_clock_base *base; + for (; active; base++, active >>= 1) { struct timerqueue_node *node; ktime_t basenow; - if (!(cpu_base->active_bases & (1 << i))) + if (!(active & 0x01)) continue; - base = cpu_base->clock_base + i; basenow = ktime_add(now, base->offset); while ((node = timerqueue_getnext(&base->active))) { @@ -1296,9 +1282,42 @@ retry: if (basenow.tv64 < hrtimer_get_softexpires_tv64(timer)) break; - __run_hrtimer(timer, &basenow); + __run_hrtimer(cpu_base, base, timer, &basenow); } } +} + +#ifdef CONFIG_HIGH_RES_TIMERS + +/* + * High resolution timer interrupt + * Called with interrupts disabled + */ +void hrtimer_interrupt(struct clock_event_device *dev) +{ + struct hrtimer_cpu_base *cpu_base = this_cpu_ptr(&hrtimer_bases); + ktime_t expires_next, now, entry_time, delta; + int retries = 0; + + BUG_ON(!cpu_base->hres_active); + cpu_base->nr_events++; + dev->next_event.tv64 = KTIME_MAX; + + raw_spin_lock(&cpu_base->lock); + entry_time = now = hrtimer_update_base(cpu_base); +retry: + cpu_base->in_hrtirq = 1; + /* + * We set expires_next to KTIME_MAX here with cpu_base->lock + * held to prevent that a timer is enqueued in our queue via + * the migration code. This does not affect enqueueing of + * timers which run their callback and need to be requeued on + * this CPU. + */ + cpu_base->expires_next.tv64 = KTIME_MAX; + + __hrtimer_run_queues(cpu_base, now); + /* Reevaluate the clock bases for the next expiry */ expires_next = __hrtimer_get_next_event(cpu_base); /* @@ -1310,8 +1329,7 @@ retry: raw_spin_unlock(&cpu_base->lock); /* Reprogramming necessary ? */ - if (expires_next.tv64 == KTIME_MAX || - !tick_program_event(expires_next, 0)) { + if (!tick_program_event(expires_next, 0)) { cpu_base->hang_detected = 0; return; } @@ -1344,8 +1362,8 @@ retry: cpu_base->hang_detected = 1; raw_spin_unlock(&cpu_base->lock); delta = ktime_sub(now, entry_time); - if (delta.tv64 > cpu_base->max_hang_time.tv64) - cpu_base->max_hang_time = delta; + if ((unsigned int)delta.tv64 > cpu_base->max_hang_time) + cpu_base->max_hang_time = (unsigned int) delta.tv64; /* * Limit it to a sensible value as we enforce a longer * delay. Give the CPU at least 100ms to catch up. @@ -1363,7 +1381,7 @@ retry: * local version of hrtimer_peek_ahead_timers() called with interrupts * disabled. */ -static void __hrtimer_peek_ahead_timers(void) +static inline void __hrtimer_peek_ahead_timers(void) { struct tick_device *td; @@ -1375,29 +1393,6 @@ static void __hrtimer_peek_ahead_timers(void) hrtimer_interrupt(td->evtdev); } -/** - * hrtimer_peek_ahead_timers -- run soft-expired timers now - * - * hrtimer_peek_ahead_timers will peek at the timer queue of - * the current cpu and check if there are any timers for which - * the soft expires time has passed. If any such timers exist, - * they are run immediately and then removed from the timer queue. - * - */ -void hrtimer_peek_ahead_timers(void) -{ - unsigned long flags; - - local_irq_save(flags); - __hrtimer_peek_ahead_timers(); - local_irq_restore(flags); -} - -static void run_hrtimer_softirq(struct softirq_action *h) -{ - hrtimer_peek_ahead_timers(); -} - #else /* CONFIG_HIGH_RES_TIMERS */ static inline void __hrtimer_peek_ahead_timers(void) { } @@ -1405,66 +1400,32 @@ static inline void __hrtimer_peek_ahead_timers(void) { } #endif /* !CONFIG_HIGH_RES_TIMERS */ /* - * Called from timer softirq every jiffy, expire hrtimers: - * - * For HRT its the fall back code to run the softirq in the timer - * softirq context in case the hrtimer initialization failed or has - * not been done yet. + * Called from run_local_timers in hardirq context every jiffy */ -void hrtimer_run_pending(void) +void hrtimer_run_queues(void) { - if (hrtimer_hres_active()) + struct hrtimer_cpu_base *cpu_base = this_cpu_ptr(&hrtimer_bases); + ktime_t now; + + if (__hrtimer_hres_active(cpu_base)) return; /* - * This _is_ ugly: We have to check in the softirq context, - * whether we can switch to highres and / or nohz mode. The - * clocksource switch happens in the timer interrupt with - * xtime_lock held. Notification from there only sets the - * check bit in the tick_oneshot code, otherwise we might - * deadlock vs. xtime_lock. + * This _is_ ugly: We have to check periodically, whether we + * can switch to highres and / or nohz mode. The clocksource + * switch happens with xtime_lock held. Notification from + * there only sets the check bit in the tick_oneshot code, + * otherwise we might deadlock vs. xtime_lock. */ - if (tick_check_oneshot_change(!hrtimer_is_hres_enabled())) + if (tick_check_oneshot_change(!hrtimer_is_hres_enabled())) { hrtimer_switch_to_hres(); -} - -/* - * Called from hardirq context every jiffy - */ -void hrtimer_run_queues(void) -{ - struct timerqueue_node *node; - struct hrtimer_cpu_base *cpu_base = this_cpu_ptr(&hrtimer_bases); - struct hrtimer_clock_base *base; - int index, gettime = 1; - - if (hrtimer_hres_active()) return; - - for (index = 0; index < HRTIMER_MAX_CLOCK_BASES; index++) { - base = &cpu_base->clock_base[index]; - if (!timerqueue_getnext(&base->active)) - continue; - - if (gettime) { - hrtimer_get_softirq_time(cpu_base); - gettime = 0; - } - - raw_spin_lock(&cpu_base->lock); - - while ((node = timerqueue_getnext(&base->active))) { - struct hrtimer *timer; - - timer = container_of(node, struct hrtimer, node); - if (base->softirq_time.tv64 <= - hrtimer_get_expires_tv64(timer)) - break; - - __run_hrtimer(timer, &base->softirq_time); - } - raw_spin_unlock(&cpu_base->lock); } + + raw_spin_lock(&cpu_base->lock); + now = hrtimer_update_base(cpu_base); + __hrtimer_run_queues(cpu_base, now); + raw_spin_unlock(&cpu_base->lock); } /* @@ -1497,8 +1458,6 @@ static int __sched do_nanosleep(struct hrtimer_sleeper *t, enum hrtimer_mode mod do { set_current_state(TASK_INTERRUPTIBLE); hrtimer_start_expires(&t->timer, mode); - if (!hrtimer_active(&t->timer)) - t->task = NULL; if (likely(t->task)) freezable_schedule(); @@ -1642,11 +1601,11 @@ static void migrate_hrtimer_list(struct hrtimer_clock_base *old_base, debug_deactivate(timer); /* - * Mark it as STATE_MIGRATE not INACTIVE otherwise the + * Mark it as ENQUEUED not INACTIVE otherwise the * timer could be seen as !active and just vanish away * under us on another CPU */ - __remove_hrtimer(timer, old_base, HRTIMER_STATE_MIGRATE, 0); + __remove_hrtimer(timer, old_base, HRTIMER_STATE_ENQUEUED, 0); timer->base = new_base; /* * Enqueue the timers on the new cpu. This does not @@ -1657,9 +1616,6 @@ static void migrate_hrtimer_list(struct hrtimer_clock_base *old_base, * event device. */ enqueue_hrtimer(timer, new_base); - - /* Clear the migration state bit */ - timer->state &= ~HRTIMER_STATE_MIGRATE; } } @@ -1731,9 +1687,6 @@ void __init hrtimers_init(void) hrtimer_cpu_notify(&hrtimers_nb, (unsigned long)CPU_UP_PREPARE, (void *)(long)smp_processor_id()); register_cpu_notifier(&hrtimers_nb); -#ifdef CONFIG_HIGH_RES_TIMERS - open_softirq(HRTIMER_SOFTIRQ, run_hrtimer_softirq); -#endif } /** @@ -1772,8 +1725,6 @@ schedule_hrtimeout_range_clock(ktime_t *expires, unsigned long delta, hrtimer_init_sleeper(&t, current); hrtimer_start_expires(&t.timer, mode); - if (!hrtimer_active(&t.timer)) - t.task = NULL; if (likely(t.task)) schedule(); diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c index 7a68100..fb4d98c 100644 --- a/kernel/time/ntp.c +++ b/kernel/time/ntp.c @@ -35,6 +35,7 @@ unsigned long tick_nsec; static u64 tick_length; static u64 tick_length_base; +#define SECS_PER_DAY 86400 #define MAX_TICKADJ 500LL /* usecs */ #define MAX_TICKADJ_SCALED \ (((MAX_TICKADJ * NSEC_PER_USEC) << NTP_SCALE_SHIFT) / NTP_INTERVAL_FREQ) @@ -76,6 +77,9 @@ static long time_adjust; /* constant (boot-param configurable) NTP tick adjustment (upscaled) */ static s64 ntp_tick_adj; +/* second value of the next pending leapsecond, or TIME64_MAX if no leap */ +static time64_t ntp_next_leap_sec = TIME64_MAX; + #ifdef CONFIG_NTP_PPS /* @@ -349,6 +353,7 @@ void ntp_clear(void) tick_length = tick_length_base; time_offset = 0; + ntp_next_leap_sec = TIME64_MAX; /* Clear PPS state variables */ pps_clear(); } @@ -359,6 +364,21 @@ u64 ntp_tick_length(void) return tick_length; } +/** + * ntp_get_next_leap - Returns the next leapsecond in CLOCK_REALTIME ktime_t + * + * Provides the time of the next leapsecond against CLOCK_REALTIME in + * a ktime_t format. Returns KTIME_MAX if no leapsecond is pending. + */ +ktime_t ntp_get_next_leap(void) +{ + ktime_t ret; + + if ((time_state == TIME_INS) && (time_status & STA_INS)) + return ktime_set(ntp_next_leap_sec, 0); + ret.tv64 = KTIME_MAX; + return ret; +} /* * this routine handles the overflow of the microsecond field @@ -382,15 +402,21 @@ int second_overflow(unsigned long secs) */ switch (time_state) { case TIME_OK: - if (time_status & STA_INS) + if (time_status & STA_INS) { time_state = TIME_INS; - else if (time_status & STA_DEL) + ntp_next_leap_sec = secs + SECS_PER_DAY - + (secs % SECS_PER_DAY); + } else if (time_status & STA_DEL) { time_state = TIME_DEL; + ntp_next_leap_sec = secs + SECS_PER_DAY - + ((secs+1) % SECS_PER_DAY); + } break; case TIME_INS: - if (!(time_status & STA_INS)) + if (!(time_status & STA_INS)) { + ntp_next_leap_sec = TIME64_MAX; time_state = TIME_OK; - else if (secs % 86400 == 0) { + } else if (secs % SECS_PER_DAY == 0) { leap = -1; time_state = TIME_OOP; printk(KERN_NOTICE @@ -398,19 +424,21 @@ int second_overflow(unsigned long secs) } break; case TIME_DEL: - if (!(time_status & STA_DEL)) + if (!(time_status & STA_DEL)) { + ntp_next_leap_sec = TIME64_MAX; time_state = TIME_OK; - else if ((secs + 1) % 86400 == 0) { + } else if ((secs + 1) % SECS_PER_DAY == 0) { leap = 1; + ntp_next_leap_sec = TIME64_MAX; time_state = TIME_WAIT; printk(KERN_NOTICE "Clock: deleting leap second 23:59:59 UTC\n"); } break; case TIME_OOP: + ntp_next_leap_sec = TIME64_MAX; time_state = TIME_WAIT; break; - case TIME_WAIT: if (!(time_status & (STA_INS | STA_DEL))) time_state = TIME_OK; @@ -547,6 +575,7 @@ static inline void process_adj_status(struct timex *txc, struct timespec64 *ts) if ((time_status & STA_PLL) && !(txc->status & STA_PLL)) { time_state = TIME_OK; time_status = STA_UNSYNC; + ntp_next_leap_sec = TIME64_MAX; /* restart PPS frequency calibration */ pps_reset_freq_interval(); } @@ -711,6 +740,24 @@ int __do_adjtimex(struct timex *txc, struct timespec64 *ts, s32 *time_tai) if (!(time_status & STA_NANO)) txc->time.tv_usec /= NSEC_PER_USEC; + /* Handle leapsec adjustments */ + if (unlikely(ts->tv_sec >= ntp_next_leap_sec)) { + if ((time_state == TIME_INS) && (time_status & STA_INS)) { + result = TIME_OOP; + txc->tai++; + txc->time.tv_sec--; + } + if ((time_state == TIME_DEL) && (time_status & STA_DEL)) { + result = TIME_WAIT; + txc->tai--; + txc->time.tv_sec++; + } + if ((time_state == TIME_OOP) && + (ts->tv_sec == ntp_next_leap_sec)) { + result = TIME_WAIT; + } + } + return result; } diff --git a/kernel/time/ntp_internal.h b/kernel/time/ntp_internal.h index bbd102a..6543050 100644 --- a/kernel/time/ntp_internal.h +++ b/kernel/time/ntp_internal.h @@ -5,6 +5,7 @@ extern void ntp_init(void); extern void ntp_clear(void); /* Returns how long ticks are at present, in ns / 2^NTP_SCALE_SHIFT. */ extern u64 ntp_tick_length(void); +extern ktime_t ntp_get_next_leap(void); extern int second_overflow(unsigned long secs); extern int ntp_validate_timex(struct timex *); extern int __do_adjtimex(struct timex *, struct timespec64 *, s32 *); diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c index 0075da7..892e3da 100644 --- a/kernel/time/posix-cpu-timers.c +++ b/kernel/time/posix-cpu-timers.c @@ -196,39 +196,62 @@ static int cpu_clock_sample(const clockid_t which_clock, struct task_struct *p, return 0; } -static void update_gt_cputime(struct task_cputime *a, struct task_cputime *b) +/* + * Set cputime to sum_cputime if sum_cputime > cputime. Use cmpxchg + * to avoid race conditions with concurrent updates to cputime. + */ +static inline void __update_gt_cputime(atomic64_t *cputime, u64 sum_cputime) { - if (b->utime > a->utime) - a->utime = b->utime; + u64 curr_cputime; +retry: + curr_cputime = atomic64_read(cputime); + if (sum_cputime > curr_cputime) { + if (atomic64_cmpxchg(cputime, curr_cputime, sum_cputime) != curr_cputime) + goto retry; + } +} - if (b->stime > a->stime) - a->stime = b->stime; +static void update_gt_cputime(struct task_cputime_atomic *cputime_atomic, struct task_cputime *sum) +{ + __update_gt_cputime(&cputime_atomic->utime, sum->utime); + __update_gt_cputime(&cputime_atomic->stime, sum->stime); + __update_gt_cputime(&cputime_atomic->sum_exec_runtime, sum->sum_exec_runtime); +} - if (b->sum_exec_runtime > a->sum_exec_runtime) - a->sum_exec_runtime = b->sum_exec_runtime; +/* Sample task_cputime_atomic values in "atomic_timers", store results in "times". */ +static inline void sample_cputime_atomic(struct task_cputime *times, + struct task_cputime_atomic *atomic_times) +{ + times->utime = atomic64_read(&atomic_times->utime); + times->stime = atomic64_read(&atomic_times->stime); + times->sum_exec_runtime = atomic64_read(&atomic_times->sum_exec_runtime); } void thread_group_cputimer(struct task_struct *tsk, struct task_cputime *times) { struct thread_group_cputimer *cputimer = &tsk->signal->cputimer; struct task_cputime sum; - unsigned long flags; - if (!cputimer->running) { + /* Check if cputimer isn't running. This is accessed without locking. */ + if (!READ_ONCE(cputimer->running)) { /* * The POSIX timer interface allows for absolute time expiry * values through the TIMER_ABSTIME flag, therefore we have - * to synchronize the timer to the clock every time we start - * it. + * to synchronize the timer to the clock every time we start it. */ thread_group_cputime(tsk, &sum); - raw_spin_lock_irqsave(&cputimer->lock, flags); - cputimer->running = 1; - update_gt_cputime(&cputimer->cputime, &sum); - } else - raw_spin_lock_irqsave(&cputimer->lock, flags); - *times = cputimer->cputime; - raw_spin_unlock_irqrestore(&cputimer->lock, flags); + update_gt_cputime(&cputimer->cputime_atomic, &sum); + + /* + * We're setting cputimer->running without a lock. Ensure + * this only gets written to in one operation. We set + * running after update_gt_cputime() as a small optimization, + * but barriers are not required because update_gt_cputime() + * can handle concurrent updates. + */ + WRITE_ONCE(cputimer->running, 1); + } + sample_cputime_atomic(times, &cputimer->cputime_atomic); } /* @@ -582,7 +605,8 @@ bool posix_cpu_timers_can_stop_tick(struct task_struct *tsk) if (!task_cputime_zero(&tsk->cputime_expires)) return false; - if (tsk->signal->cputimer.running) + /* Check if cputimer is running. This is accessed without locking. */ + if (READ_ONCE(tsk->signal->cputimer.running)) return false; return true; @@ -852,10 +876,10 @@ static void check_thread_timers(struct task_struct *tsk, /* * Check for the special case thread timers. */ - soft = ACCESS_ONCE(sig->rlim[RLIMIT_RTTIME].rlim_cur); + soft = READ_ONCE(sig->rlim[RLIMIT_RTTIME].rlim_cur); if (soft != RLIM_INFINITY) { unsigned long hard = - ACCESS_ONCE(sig->rlim[RLIMIT_RTTIME].rlim_max); + READ_ONCE(sig->rlim[RLIMIT_RTTIME].rlim_max); if (hard != RLIM_INFINITY && tsk->rt.timeout > DIV_ROUND_UP(hard, USEC_PER_SEC/HZ)) { @@ -882,14 +906,12 @@ static void check_thread_timers(struct task_struct *tsk, } } -static void stop_process_timers(struct signal_struct *sig) +static inline void stop_process_timers(struct signal_struct *sig) { struct thread_group_cputimer *cputimer = &sig->cputimer; - unsigned long flags; - raw_spin_lock_irqsave(&cputimer->lock, flags); - cputimer->running = 0; - raw_spin_unlock_irqrestore(&cputimer->lock, flags); + /* Turn off cputimer->running. This is done without locking. */ + WRITE_ONCE(cputimer->running, 0); } static u32 onecputick; @@ -958,11 +980,11 @@ static void check_process_timers(struct task_struct *tsk, SIGPROF); check_cpu_itimer(tsk, &sig->it[CPUCLOCK_VIRT], &virt_expires, utime, SIGVTALRM); - soft = ACCESS_ONCE(sig->rlim[RLIMIT_CPU].rlim_cur); + soft = READ_ONCE(sig->rlim[RLIMIT_CPU].rlim_cur); if (soft != RLIM_INFINITY) { unsigned long psecs = cputime_to_secs(ptime); unsigned long hard = - ACCESS_ONCE(sig->rlim[RLIMIT_CPU].rlim_max); + READ_ONCE(sig->rlim[RLIMIT_CPU].rlim_max); cputime_t x; if (psecs >= hard) { /* @@ -1111,12 +1133,11 @@ static inline int fastpath_timer_check(struct task_struct *tsk) } sig = tsk->signal; - if (sig->cputimer.running) { + /* Check if cputimer is running. This is accessed without locking. */ + if (READ_ONCE(sig->cputimer.running)) { struct task_cputime group_sample; - raw_spin_lock(&sig->cputimer.lock); - group_sample = sig->cputimer.cputime; - raw_spin_unlock(&sig->cputimer.lock); + sample_cputime_atomic(&group_sample, &sig->cputimer.cputime_atomic); if (task_cputime_expired(&group_sample, &sig->cputime_expires)) return 1; @@ -1157,7 +1178,7 @@ void run_posix_cpu_timers(struct task_struct *tsk) * If there are any active process wide timers (POSIX 1.b, itimers, * RLIMIT_CPU) cputimer must be running. */ - if (tsk->signal->cputimer.running) + if (READ_ONCE(tsk->signal->cputimer.running)) check_process_timers(tsk, &firing); /* diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index 31ea01f..31d11ac 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -272,13 +272,20 @@ static int posix_get_tai(clockid_t which_clock, struct timespec *tp) return 0; } +static int posix_get_hrtimer_res(clockid_t which_clock, struct timespec *tp) +{ + tp->tv_sec = 0; + tp->tv_nsec = hrtimer_resolution; + return 0; +} + /* * Initialize everything, well, just everything in Posix clocks/timers ;) */ static __init int init_posix_timers(void) { struct k_clock clock_realtime = { - .clock_getres = hrtimer_get_res, + .clock_getres = posix_get_hrtimer_res, .clock_get = posix_clock_realtime_get, .clock_set = posix_clock_realtime_set, .clock_adj = posix_clock_realtime_adj, @@ -290,7 +297,7 @@ static __init int init_posix_timers(void) .timer_del = common_timer_del, }; struct k_clock clock_monotonic = { - .clock_getres = hrtimer_get_res, + .clock_getres = posix_get_hrtimer_res, .clock_get = posix_ktime_get_ts, .nsleep = common_nsleep, .nsleep_restart = hrtimer_nanosleep_restart, @@ -300,7 +307,7 @@ static __init int init_posix_timers(void) .timer_del = common_timer_del, }; struct k_clock clock_monotonic_raw = { - .clock_getres = hrtimer_get_res, + .clock_getres = posix_get_hrtimer_res, .clock_get = posix_get_monotonic_raw, }; struct k_clock clock_realtime_coarse = { @@ -312,7 +319,7 @@ static __init int init_posix_timers(void) .clock_get = posix_get_monotonic_coarse, }; struct k_clock clock_tai = { - .clock_getres = hrtimer_get_res, + .clock_getres = posix_get_hrtimer_res, .clock_get = posix_get_tai, .nsleep = common_nsleep, .nsleep_restart = hrtimer_nanosleep_restart, @@ -322,7 +329,7 @@ static __init int init_posix_timers(void) .timer_del = common_timer_del, }; struct k_clock clock_boottime = { - .clock_getres = hrtimer_get_res, + .clock_getres = posix_get_hrtimer_res, .clock_get = posix_get_boottime, .nsleep = common_nsleep, .nsleep_restart = hrtimer_nanosleep_restart, diff --git a/kernel/time/tick-broadcast-hrtimer.c b/kernel/time/tick-broadcast-hrtimer.c index 6aac4be..3e7db49 100644 --- a/kernel/time/tick-broadcast-hrtimer.c +++ b/kernel/time/tick-broadcast-hrtimer.c @@ -22,6 +22,7 @@ static void bc_set_mode(enum clock_event_mode mode, struct clock_event_device *bc) { switch (mode) { + case CLOCK_EVT_MODE_UNUSED: case CLOCK_EVT_MODE_SHUTDOWN: /* * Note, we cannot cancel the timer here as we might @@ -66,9 +67,11 @@ static int bc_set_next(ktime_t expires, struct clock_event_device *bc) * hrtimer_{start/cancel} functions call into tracing, * calls to these functions must be bound within RCU_NONIDLE. */ - RCU_NONIDLE(bc_moved = (hrtimer_try_to_cancel(&bctimer) >= 0) ? - !hrtimer_start(&bctimer, expires, HRTIMER_MODE_ABS_PINNED) : - 0); + RCU_NONIDLE({ + bc_moved = hrtimer_try_to_cancel(&bctimer) >= 0; + if (bc_moved) + hrtimer_start(&bctimer, expires, + HRTIMER_MODE_ABS_PINNED);}); if (bc_moved) { /* Bind the "device" to the cpu */ bc->bound_on = smp_processor_id(); @@ -99,10 +102,13 @@ static enum hrtimer_restart bc_handler(struct hrtimer *t) { ce_broadcast_hrtimer.event_handler(&ce_broadcast_hrtimer); - if (ce_broadcast_hrtimer.next_event.tv64 == KTIME_MAX) + switch (ce_broadcast_hrtimer.mode) { + case CLOCK_EVT_MODE_ONESHOT: + if (ce_broadcast_hrtimer.next_event.tv64 != KTIME_MAX) + return HRTIMER_RESTART; + default: return HRTIMER_NORESTART; - - return HRTIMER_RESTART; + } } void tick_setup_hrtimer_broadcast(void) diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c index 7e8ca4f..d39f32c 100644 --- a/kernel/time/tick-broadcast.c +++ b/kernel/time/tick-broadcast.c @@ -255,18 +255,18 @@ int tick_receive_broadcast(void) /* * Broadcast the event to the cpus, which are set in the mask (mangled). */ -static void tick_do_broadcast(struct cpumask *mask) +static bool tick_do_broadcast(struct cpumask *mask) { int cpu = smp_processor_id(); struct tick_device *td; + bool local = false; /* * Check, if the current cpu is in the mask */ if (cpumask_test_cpu(cpu, mask)) { cpumask_clear_cpu(cpu, mask); - td = &per_cpu(tick_cpu_device, cpu); - td->evtdev->event_handler(td->evtdev); + local = true; } if (!cpumask_empty(mask)) { @@ -279,16 +279,17 @@ static void tick_do_broadcast(struct cpumask *mask) td = &per_cpu(tick_cpu_device, cpumask_first(mask)); td->evtdev->broadcast(mask); } + return local; } /* * Periodic broadcast: * - invoke the broadcast handlers */ -static void tick_do_periodic_broadcast(void) +static bool tick_do_periodic_broadcast(void) { cpumask_and(tmpmask, cpu_online_mask, tick_broadcast_mask); - tick_do_broadcast(tmpmask); + return tick_do_broadcast(tmpmask); } /* @@ -296,34 +297,26 @@ static void tick_do_periodic_broadcast(void) */ static void tick_handle_periodic_broadcast(struct clock_event_device *dev) { - ktime_t next; + struct tick_device *td = this_cpu_ptr(&tick_cpu_device); + bool bc_local; raw_spin_lock(&tick_broadcast_lock); + bc_local = tick_do_periodic_broadcast(); - tick_do_periodic_broadcast(); + if (clockevent_state_oneshot(dev)) { + ktime_t next = ktime_add(dev->next_event, tick_period); - /* - * The device is in periodic mode. No reprogramming necessary: - */ - if (dev->state == CLOCK_EVT_STATE_PERIODIC) - goto unlock; + clockevents_program_event(dev, next, true); + } + raw_spin_unlock(&tick_broadcast_lock); /* - * Setup the next period for devices, which do not have - * periodic mode. We read dev->next_event first and add to it - * when the event already expired. clockevents_program_event() - * sets dev->next_event only when the event is really - * programmed to the device. + * We run the handler of the local cpu after dropping + * tick_broadcast_lock because the handler might deadlock when + * trying to switch to oneshot mode. */ - for (next = dev->next_event; ;) { - next = ktime_add(next, tick_period); - - if (!clockevents_program_event(dev, next, false)) - goto unlock; - tick_do_periodic_broadcast(); - } -unlock: - raw_spin_unlock(&tick_broadcast_lock); + if (bc_local) + td->evtdev->event_handler(td->evtdev); } /** @@ -532,23 +525,19 @@ static void tick_broadcast_set_affinity(struct clock_event_device *bc, irq_set_affinity(bc->irq, bc->cpumask); } -static int tick_broadcast_set_event(struct clock_event_device *bc, int cpu, - ktime_t expires, int force) +static void tick_broadcast_set_event(struct clock_event_device *bc, int cpu, + ktime_t expires) { - int ret; - - if (bc->state != CLOCK_EVT_STATE_ONESHOT) - clockevents_set_state(bc, CLOCK_EVT_STATE_ONESHOT); + if (!clockevent_state_oneshot(bc)) + clockevents_switch_state(bc, CLOCK_EVT_STATE_ONESHOT); - ret = clockevents_program_event(bc, expires, force); - if (!ret) - tick_broadcast_set_affinity(bc, cpumask_of(cpu)); - return ret; + clockevents_program_event(bc, expires, 1); + tick_broadcast_set_affinity(bc, cpumask_of(cpu)); } static void tick_resume_broadcast_oneshot(struct clock_event_device *bc) { - clockevents_set_state(bc, CLOCK_EVT_STATE_ONESHOT); + clockevents_switch_state(bc, CLOCK_EVT_STATE_ONESHOT); } /* @@ -566,7 +555,7 @@ void tick_check_oneshot_broadcast_this_cpu(void) * switched over, leave the device alone. */ if (td->mode == TICKDEV_MODE_ONESHOT) { - clockevents_set_state(td->evtdev, + clockevents_switch_state(td->evtdev, CLOCK_EVT_STATE_ONESHOT); } } @@ -580,9 +569,9 @@ static void tick_handle_oneshot_broadcast(struct clock_event_device *dev) struct tick_device *td; ktime_t now, next_event; int cpu, next_cpu = 0; + bool bc_local; raw_spin_lock(&tick_broadcast_lock); -again: dev->next_event.tv64 = KTIME_MAX; next_event.tv64 = KTIME_MAX; cpumask_clear(tmpmask); @@ -624,7 +613,7 @@ again: /* * Wakeup the cpus which have an expired event. */ - tick_do_broadcast(tmpmask); + bc_local = tick_do_broadcast(tmpmask); /* * Two reasons for reprogram: @@ -636,15 +625,15 @@ again: * - There are pending events on sleeping CPUs which were not * in the event mask */ - if (next_event.tv64 != KTIME_MAX) { - /* - * Rearm the broadcast device. If event expired, - * repeat the above - */ - if (tick_broadcast_set_event(dev, next_cpu, next_event, 0)) - goto again; - } + if (next_event.tv64 != KTIME_MAX) + tick_broadcast_set_event(dev, next_cpu, next_event); + raw_spin_unlock(&tick_broadcast_lock); + + if (bc_local) { + td = this_cpu_ptr(&tick_cpu_device); + td->evtdev->event_handler(td->evtdev); + } } static int broadcast_needs_cpu(struct clock_event_device *bc, int cpu) @@ -670,7 +659,7 @@ static void broadcast_shutdown_local(struct clock_event_device *bc, if (dev->next_event.tv64 < bc->next_event.tv64) return; } - clockevents_set_state(dev, CLOCK_EVT_STATE_SHUTDOWN); + clockevents_switch_state(dev, CLOCK_EVT_STATE_SHUTDOWN); } /** @@ -726,7 +715,7 @@ int tick_broadcast_oneshot_control(enum tick_broadcast_state state) */ if (!cpumask_test_cpu(cpu, tick_broadcast_force_mask) && dev->next_event.tv64 < bc->next_event.tv64) - tick_broadcast_set_event(bc, cpu, dev->next_event, 1); + tick_broadcast_set_event(bc, cpu, dev->next_event); } /* * If the current CPU owns the hrtimer broadcast @@ -740,7 +729,7 @@ int tick_broadcast_oneshot_control(enum tick_broadcast_state state) cpumask_clear_cpu(cpu, tick_broadcast_oneshot_mask); } else { if (cpumask_test_and_clear_cpu(cpu, tick_broadcast_oneshot_mask)) { - clockevents_set_state(dev, CLOCK_EVT_STATE_ONESHOT); + clockevents_switch_state(dev, CLOCK_EVT_STATE_ONESHOT); /* * The cpu which was handling the broadcast * timer marked this cpu in the broadcast @@ -842,7 +831,7 @@ 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->state == CLOCK_EVT_STATE_PERIODIC; + int was_periodic = clockevent_state_periodic(bc); bc->event_handler = tick_handle_oneshot_broadcast; @@ -858,10 +847,10 @@ void tick_broadcast_setup_oneshot(struct clock_event_device *bc) tick_broadcast_oneshot_mask, tmpmask); if (was_periodic && !cpumask_empty(tmpmask)) { - clockevents_set_state(bc, CLOCK_EVT_STATE_ONESHOT); + clockevents_switch_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); + tick_broadcast_set_event(bc, cpu, tick_next_period); } else bc->next_event.tv64 = KTIME_MAX; } else { diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c index 3ae6afa..76446cb 100644 --- a/kernel/time/tick-common.c +++ b/kernel/time/tick-common.c @@ -19,6 +19,7 @@ #include <linux/profile.h> #include <linux/sched.h> #include <linux/module.h> +#include <trace/events/power.h> #include <asm/irq_regs.h> @@ -102,7 +103,17 @@ void tick_handle_periodic(struct clock_event_device *dev) tick_periodic(cpu); - if (dev->state != CLOCK_EVT_STATE_ONESHOT) +#if defined(CONFIG_HIGH_RES_TIMERS) || defined(CONFIG_NO_HZ_COMMON) + /* + * The cpu might have transitioned to HIGHRES or NOHZ mode via + * update_process_times() -> run_local_timers() -> + * hrtimer_run_queues(). + */ + if (dev->event_handler != tick_handle_periodic) + return; +#endif + + if (!clockevent_state_oneshot(dev)) return; for (;;) { /* @@ -140,7 +151,7 @@ void tick_setup_periodic(struct clock_event_device *dev, int broadcast) if ((dev->features & CLOCK_EVT_FEAT_PERIODIC) && !tick_broadcast_oneshot_active()) { - clockevents_set_state(dev, CLOCK_EVT_STATE_PERIODIC); + clockevents_switch_state(dev, CLOCK_EVT_STATE_PERIODIC); } else { unsigned long seq; ktime_t next; @@ -150,7 +161,7 @@ void tick_setup_periodic(struct clock_event_device *dev, int broadcast) next = tick_next_period; } while (read_seqretry(&jiffies_lock, seq)); - clockevents_set_state(dev, CLOCK_EVT_STATE_ONESHOT); + clockevents_switch_state(dev, CLOCK_EVT_STATE_ONESHOT); for (;;) { if (!clockevents_program_event(dev, next, false)) @@ -367,7 +378,7 @@ void tick_shutdown(unsigned int cpu) * Prevent that the clock events layer tries to call * the set mode function! */ - dev->state = CLOCK_EVT_STATE_DETACHED; + clockevent_set_state(dev, CLOCK_EVT_STATE_DETACHED); dev->mode = CLOCK_EVT_MODE_UNUSED; clockevents_exchange_device(dev, NULL); dev->event_handler = clockevents_handle_noop; @@ -440,6 +451,7 @@ void tick_resume(void) tick_resume_local(); } +#ifdef CONFIG_SUSPEND static DEFINE_RAW_SPINLOCK(tick_freeze_lock); static unsigned int tick_freeze_depth; @@ -457,10 +469,13 @@ 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()) { + trace_suspend_resume(TPS("timekeeping_freeze"), + smp_processor_id(), true); timekeeping_suspend(); - else + } else { tick_suspend_local(); + } raw_spin_unlock(&tick_freeze_lock); } @@ -478,15 +493,19 @@ void tick_unfreeze(void) { raw_spin_lock(&tick_freeze_lock); - if (tick_freeze_depth == num_online_cpus()) + if (tick_freeze_depth == num_online_cpus()) { timekeeping_resume(); - else + trace_suspend_resume(TPS("timekeeping_freeze"), + smp_processor_id(), false); + } else { tick_resume_local(); + } tick_freeze_depth--; raw_spin_unlock(&tick_freeze_lock); } +#endif /* CONFIG_SUSPEND */ /** * tick_init - initialize the tick control diff --git a/kernel/time/tick-internal.h b/kernel/time/tick-internal.h index b64fdd8..966a5a6 100644 --- a/kernel/time/tick-internal.h +++ b/kernel/time/tick-internal.h @@ -36,11 +36,22 @@ static inline int tick_device_is_functional(struct clock_event_device *dev) return !(dev->features & CLOCK_EVT_FEAT_DUMMY); } +static inline enum clock_event_state clockevent_get_state(struct clock_event_device *dev) +{ + return dev->state_use_accessors; +} + +static inline void clockevent_set_state(struct clock_event_device *dev, + enum clock_event_state state) +{ + dev->state_use_accessors = state; +} + 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 void clockevents_switch_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); @@ -137,3 +148,19 @@ extern void tick_nohz_init(void); # else static inline void tick_nohz_init(void) { } #endif + +#ifdef CONFIG_NO_HZ_COMMON +extern unsigned long tick_nohz_active; +#else +#define tick_nohz_active (0) +#endif + +#if defined(CONFIG_SMP) && defined(CONFIG_NO_HZ_COMMON) +extern void timers_update_migration(bool update_nohz); +#else +static inline void timers_update_migration(bool update_nohz) { } +#endif + +DECLARE_PER_CPU(struct hrtimer_cpu_base, hrtimer_bases); + +extern u64 get_next_timer_interrupt(unsigned long basej, u64 basem); diff --git a/kernel/time/tick-oneshot.c b/kernel/time/tick-oneshot.c index 67a64b1..b513446 100644 --- a/kernel/time/tick-oneshot.c +++ b/kernel/time/tick-oneshot.c @@ -28,6 +28,22 @@ int tick_program_event(ktime_t expires, int force) { struct clock_event_device *dev = __this_cpu_read(tick_cpu_device.evtdev); + if (unlikely(expires.tv64 == KTIME_MAX)) { + /* + * We don't need the clock event device any more, stop it. + */ + clockevents_switch_state(dev, CLOCK_EVT_STATE_ONESHOT_STOPPED); + return 0; + } + + if (unlikely(clockevent_state_oneshot_stopped(dev))) { + /* + * We need the clock event again, configure it in ONESHOT mode + * before using it. + */ + clockevents_switch_state(dev, CLOCK_EVT_STATE_ONESHOT); + } + return clockevents_program_event(dev, expires, force); } @@ -38,7 +54,7 @@ void tick_resume_oneshot(void) { struct clock_event_device *dev = __this_cpu_read(tick_cpu_device.evtdev); - clockevents_set_state(dev, CLOCK_EVT_STATE_ONESHOT); + clockevents_switch_state(dev, CLOCK_EVT_STATE_ONESHOT); clockevents_program_event(dev, ktime_get(), true); } @@ -50,7 +66,7 @@ void tick_setup_oneshot(struct clock_event_device *newdev, ktime_t next_event) { newdev->event_handler = handler; - clockevents_set_state(newdev, CLOCK_EVT_STATE_ONESHOT); + clockevents_switch_state(newdev, CLOCK_EVT_STATE_ONESHOT); clockevents_program_event(newdev, next_event, true); } @@ -81,7 +97,7 @@ int tick_switch_to_oneshot(void (*handler)(struct clock_event_device *)) td->mode = TICKDEV_MODE_ONESHOT; dev->event_handler = handler; - clockevents_set_state(dev, CLOCK_EVT_STATE_ONESHOT); + clockevents_switch_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 9142591..c792429 100644 --- a/kernel/time/tick-sched.c +++ b/kernel/time/tick-sched.c @@ -399,7 +399,7 @@ void __init tick_nohz_init(void) * NO HZ enabled ? */ static int tick_nohz_enabled __read_mostly = 1; -int tick_nohz_active __read_mostly; +unsigned long tick_nohz_active __read_mostly; /* * Enable / Disable tickless mode */ @@ -565,156 +565,144 @@ u64 get_cpu_iowait_time_us(int cpu, u64 *last_update_time) } EXPORT_SYMBOL_GPL(get_cpu_iowait_time_us); +static void tick_nohz_restart(struct tick_sched *ts, ktime_t now) +{ + hrtimer_cancel(&ts->sched_timer); + hrtimer_set_expires(&ts->sched_timer, ts->last_tick); + + /* Forward the time to expire in the future */ + hrtimer_forward(&ts->sched_timer, now, tick_period); + + if (ts->nohz_mode == NOHZ_MODE_HIGHRES) + hrtimer_start_expires(&ts->sched_timer, HRTIMER_MODE_ABS_PINNED); + else + tick_program_event(hrtimer_get_expires(&ts->sched_timer), 1); +} + static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts, ktime_t now, int cpu) { - unsigned long seq, last_jiffies, next_jiffies, delta_jiffies; - ktime_t last_update, expires, ret = { .tv64 = 0 }; - unsigned long rcu_delta_jiffies; struct clock_event_device *dev = __this_cpu_read(tick_cpu_device.evtdev); - u64 time_delta; - - time_delta = timekeeping_max_deferment(); + u64 basemono, next_tick, next_tmr, next_rcu, delta, expires; + unsigned long seq, basejiff; + ktime_t tick; /* Read jiffies and the time when jiffies were updated last */ do { seq = read_seqbegin(&jiffies_lock); - last_update = last_jiffies_update; - last_jiffies = jiffies; + basemono = last_jiffies_update.tv64; + basejiff = jiffies; } while (read_seqretry(&jiffies_lock, seq)); + ts->last_jiffies = basejiff; - if (rcu_needs_cpu(&rcu_delta_jiffies) || + if (rcu_needs_cpu(basemono, &next_rcu) || arch_needs_cpu() || irq_work_needs_cpu()) { - next_jiffies = last_jiffies + 1; - delta_jiffies = 1; + next_tick = basemono + TICK_NSEC; } else { - /* Get the next timer wheel timer */ - next_jiffies = get_next_timer_interrupt(last_jiffies); - delta_jiffies = next_jiffies - last_jiffies; - if (rcu_delta_jiffies < delta_jiffies) { - next_jiffies = last_jiffies + rcu_delta_jiffies; - delta_jiffies = rcu_delta_jiffies; - } + /* + * Get the next pending timer. If high resolution + * timers are enabled this only takes the timer wheel + * timers into account. If high resolution timers are + * disabled this also looks at the next expiring + * hrtimer. + */ + next_tmr = get_next_timer_interrupt(basejiff, basemono); + ts->next_timer = next_tmr; + /* Take the next rcu event into account */ + next_tick = next_rcu < next_tmr ? next_rcu : next_tmr; } /* - * Do not stop the tick, if we are only one off (or less) - * or if the cpu is required for RCU: + * If the tick is due in the next period, keep it ticking or + * restart it proper. */ - if (!ts->tick_stopped && delta_jiffies <= 1) - goto out; - - /* Schedule the tick, if we are at least one jiffie off */ - if ((long)delta_jiffies >= 1) { - - /* - * If this cpu is the one which updates jiffies, then - * give up the assignment and let it be taken by the - * cpu which runs the tick timer next, which might be - * this cpu as well. If we don't drop this here the - * jiffies might be stale and do_timer() never - * invoked. Keep track of the fact that it was the one - * which had the do_timer() duty last. If this cpu is - * the one which had the do_timer() duty last, we - * limit the sleep time to the timekeeping - * max_deferement value which we retrieved - * above. Otherwise we can sleep as long as we want. - */ - if (cpu == tick_do_timer_cpu) { - tick_do_timer_cpu = TICK_DO_TIMER_NONE; - ts->do_timer_last = 1; - } else if (tick_do_timer_cpu != TICK_DO_TIMER_NONE) { - time_delta = KTIME_MAX; - ts->do_timer_last = 0; - } else if (!ts->do_timer_last) { - time_delta = KTIME_MAX; + delta = next_tick - basemono; + if (delta <= (u64)TICK_NSEC) { + tick.tv64 = 0; + if (!ts->tick_stopped) + goto out; + if (delta == 0) { + /* Tick is stopped, but required now. Enforce it */ + tick_nohz_restart(ts, now); + goto out; } + } + + /* + * If this cpu is the one which updates jiffies, then give up + * the assignment and let it be taken by the cpu which runs + * the tick timer next, which might be this cpu as well. If we + * don't drop this here the jiffies might be stale and + * do_timer() never invoked. Keep track of the fact that it + * was the one which had the do_timer() duty last. If this cpu + * is the one which had the do_timer() duty last, we limit the + * sleep time to the timekeeping max_deferement value. + * Otherwise we can sleep as long as we want. + */ + delta = timekeeping_max_deferment(); + if (cpu == tick_do_timer_cpu) { + tick_do_timer_cpu = TICK_DO_TIMER_NONE; + ts->do_timer_last = 1; + } else if (tick_do_timer_cpu != TICK_DO_TIMER_NONE) { + delta = KTIME_MAX; + ts->do_timer_last = 0; + } else if (!ts->do_timer_last) { + delta = KTIME_MAX; + } #ifdef CONFIG_NO_HZ_FULL - if (!ts->inidle) { - time_delta = min(time_delta, - scheduler_tick_max_deferment()); - } + /* Limit the tick delta to the maximum scheduler deferment */ + if (!ts->inidle) + delta = min(delta, scheduler_tick_max_deferment()); #endif - /* - * calculate the expiry time for the next timer wheel - * timer. delta_jiffies >= NEXT_TIMER_MAX_DELTA signals - * that there is no timer pending or at least extremely - * far into the future (12 days for HZ=1000). In this - * case we set the expiry to the end of time. - */ - if (likely(delta_jiffies < NEXT_TIMER_MAX_DELTA)) { - /* - * Calculate the time delta for the next timer event. - * If the time delta exceeds the maximum time delta - * permitted by the current clocksource then adjust - * the time delta accordingly to ensure the - * clocksource does not wrap. - */ - time_delta = min_t(u64, time_delta, - tick_period.tv64 * delta_jiffies); - } - - if (time_delta < KTIME_MAX) - expires = ktime_add_ns(last_update, time_delta); - else - expires.tv64 = KTIME_MAX; - - /* Skip reprogram of event if its not changed */ - if (ts->tick_stopped && ktime_equal(expires, dev->next_event)) - goto out; + /* Calculate the next expiry time */ + if (delta < (KTIME_MAX - basemono)) + expires = basemono + delta; + else + expires = KTIME_MAX; - ret = expires; + expires = min_t(u64, expires, next_tick); + tick.tv64 = expires; - /* - * nohz_stop_sched_tick can be called several times before - * the nohz_restart_sched_tick is called. This happens when - * interrupts arrive which do not cause a reschedule. In the - * first call we save the current tick time, so we can restart - * the scheduler tick in nohz_restart_sched_tick. - */ - if (!ts->tick_stopped) { - nohz_balance_enter_idle(cpu); - calc_load_enter_idle(); + /* Skip reprogram of event if its not changed */ + if (ts->tick_stopped && (expires == dev->next_event.tv64)) + goto out; - ts->last_tick = hrtimer_get_expires(&ts->sched_timer); - ts->tick_stopped = 1; - trace_tick_stop(1, " "); - } + /* + * nohz_stop_sched_tick can be called several times before + * the nohz_restart_sched_tick is called. This happens when + * interrupts arrive which do not cause a reschedule. In the + * first call we save the current tick time, so we can restart + * the scheduler tick in nohz_restart_sched_tick. + */ + if (!ts->tick_stopped) { + nohz_balance_enter_idle(cpu); + calc_load_enter_idle(); - /* - * If the expiration time == KTIME_MAX, then - * in this case we simply stop the tick timer. - */ - if (unlikely(expires.tv64 == KTIME_MAX)) { - if (ts->nohz_mode == NOHZ_MODE_HIGHRES) - hrtimer_cancel(&ts->sched_timer); - goto out; - } + ts->last_tick = hrtimer_get_expires(&ts->sched_timer); + ts->tick_stopped = 1; + trace_tick_stop(1, " "); + } - if (ts->nohz_mode == NOHZ_MODE_HIGHRES) { - hrtimer_start(&ts->sched_timer, expires, - HRTIMER_MODE_ABS_PINNED); - /* Check, if the timer was already in the past */ - if (hrtimer_active(&ts->sched_timer)) - goto out; - } else if (!tick_program_event(expires, 0)) - goto out; - /* - * We are past the event already. So we crossed a - * jiffie boundary. Update jiffies and raise the - * softirq. - */ - tick_do_update_jiffies64(ktime_get()); + /* + * If the expiration time == KTIME_MAX, then we simply stop + * the tick timer. + */ + if (unlikely(expires == KTIME_MAX)) { + if (ts->nohz_mode == NOHZ_MODE_HIGHRES) + hrtimer_cancel(&ts->sched_timer); + goto out; } - raise_softirq_irqoff(TIMER_SOFTIRQ); + + if (ts->nohz_mode == NOHZ_MODE_HIGHRES) + hrtimer_start(&ts->sched_timer, tick, HRTIMER_MODE_ABS_PINNED); + else + tick_program_event(tick, 1); out: - ts->next_jiffies = next_jiffies; - ts->last_jiffies = last_jiffies; + /* Update the estimated sleep length */ ts->sleep_length = ktime_sub(dev->next_event, now); - - return ret; + return tick; } static void tick_nohz_full_stop_tick(struct tick_sched *ts) @@ -876,32 +864,6 @@ ktime_t tick_nohz_get_sleep_length(void) return ts->sleep_length; } -static void tick_nohz_restart(struct tick_sched *ts, ktime_t now) -{ - hrtimer_cancel(&ts->sched_timer); - hrtimer_set_expires(&ts->sched_timer, ts->last_tick); - - while (1) { - /* Forward the time to expire in the future */ - hrtimer_forward(&ts->sched_timer, now, tick_period); - - if (ts->nohz_mode == NOHZ_MODE_HIGHRES) { - hrtimer_start_expires(&ts->sched_timer, - HRTIMER_MODE_ABS_PINNED); - /* Check, if the timer was already in the past */ - if (hrtimer_active(&ts->sched_timer)) - break; - } else { - if (!tick_program_event( - hrtimer_get_expires(&ts->sched_timer), 0)) - break; - } - /* Reread time and update jiffies */ - now = ktime_get(); - tick_do_update_jiffies64(now); - } -} - static void tick_nohz_restart_sched_tick(struct tick_sched *ts, ktime_t now) { /* Update jiffies first */ @@ -972,12 +934,6 @@ void tick_nohz_idle_exit(void) local_irq_enable(); } -static int tick_nohz_reprogram(struct tick_sched *ts, ktime_t now) -{ - hrtimer_forward(&ts->sched_timer, now, tick_period); - return tick_program_event(hrtimer_get_expires(&ts->sched_timer), 0); -} - /* * The nohz low res interrupt handler */ @@ -996,10 +952,18 @@ static void tick_nohz_handler(struct clock_event_device *dev) if (unlikely(ts->tick_stopped)) return; - while (tick_nohz_reprogram(ts, now)) { - now = ktime_get(); - tick_do_update_jiffies64(now); - } + hrtimer_forward(&ts->sched_timer, now, tick_period); + tick_program_event(hrtimer_get_expires(&ts->sched_timer), 1); +} + +static inline void tick_nohz_activate(struct tick_sched *ts, int mode) +{ + if (!tick_nohz_enabled) + return; + ts->nohz_mode = mode; + /* One update is enough */ + if (!test_and_set_bit(0, &tick_nohz_active)) + timers_update_migration(true); } /** @@ -1013,13 +977,8 @@ static void tick_nohz_switch_to_nohz(void) if (!tick_nohz_enabled) return; - local_irq_disable(); - if (tick_switch_to_oneshot(tick_nohz_handler)) { - local_irq_enable(); + if (tick_switch_to_oneshot(tick_nohz_handler)) return; - } - tick_nohz_active = 1; - ts->nohz_mode = NOHZ_MODE_LOWRES; /* * Recycle the hrtimer in ts, so we can share the @@ -1029,13 +988,10 @@ static void tick_nohz_switch_to_nohz(void) /* Get the next period */ next = tick_init_jiffy_update(); - for (;;) { - hrtimer_set_expires(&ts->sched_timer, next); - if (!tick_program_event(next, 0)) - break; - next = ktime_add(next, tick_period); - } - local_irq_enable(); + hrtimer_forward_now(&ts->sched_timer, tick_period); + hrtimer_set_expires(&ts->sched_timer, next); + tick_program_event(next, 1); + tick_nohz_activate(ts, NOHZ_MODE_LOWRES); } /* @@ -1087,6 +1043,7 @@ static inline void tick_nohz_irq_enter(void) static inline void tick_nohz_switch_to_nohz(void) { } static inline void tick_nohz_irq_enter(void) { } +static inline void tick_nohz_activate(struct tick_sched *ts, int mode) { } #endif /* CONFIG_NO_HZ_COMMON */ @@ -1167,22 +1124,9 @@ void tick_setup_sched_timer(void) hrtimer_add_expires_ns(&ts->sched_timer, offset); } - for (;;) { - hrtimer_forward(&ts->sched_timer, now, tick_period); - hrtimer_start_expires(&ts->sched_timer, - HRTIMER_MODE_ABS_PINNED); - /* Check, if the timer was already in the past */ - if (hrtimer_active(&ts->sched_timer)) - break; - now = ktime_get(); - } - -#ifdef CONFIG_NO_HZ_COMMON - if (tick_nohz_enabled) { - ts->nohz_mode = NOHZ_MODE_HIGHRES; - tick_nohz_active = 1; - } -#endif + hrtimer_forward(&ts->sched_timer, now, tick_period); + hrtimer_start_expires(&ts->sched_timer, HRTIMER_MODE_ABS_PINNED); + tick_nohz_activate(ts, NOHZ_MODE_HIGHRES); } #endif /* HIGH_RES_TIMERS */ @@ -1227,7 +1171,7 @@ void tick_oneshot_notify(void) * Called cyclic from the hrtimer softirq (driven by the timer * softirq) allow_nohz signals, that we can switch into low-res nohz * mode, because high resolution timers are disabled (either compile - * or runtime). + * or runtime). Called with interrupts disabled. */ int tick_check_oneshot_change(int allow_nohz) { diff --git a/kernel/time/tick-sched.h b/kernel/time/tick-sched.h index 28b5da3..42fdf49 100644 --- a/kernel/time/tick-sched.h +++ b/kernel/time/tick-sched.h @@ -57,7 +57,7 @@ struct tick_sched { ktime_t iowait_sleeptime; ktime_t sleep_length; unsigned long last_jiffies; - unsigned long next_jiffies; + u64 next_timer; ktime_t idle_expires; int do_timer_last; }; diff --git a/kernel/time/time.c b/kernel/time/time.c index 2c85b77..85d5bb1 100644 --- a/kernel/time/time.c +++ b/kernel/time/time.c @@ -41,7 +41,7 @@ #include <asm/uaccess.h> #include <asm/unistd.h> -#include "timeconst.h" +#include <generated/timeconst.h> #include "timekeeping.h" /* @@ -173,6 +173,10 @@ int do_sys_settimeofday(const struct timespec *tv, const struct timezone *tz) return error; if (tz) { + /* Verify we're witin the +-15 hrs range */ + if (tz->tz_minuteswest > 15*60 || tz->tz_minuteswest < -15*60) + return -EINVAL; + sys_tz = *tz; update_vsyscall_tz(); if (firsttime) { @@ -483,9 +487,11 @@ struct timespec64 ns_to_timespec64(const s64 nsec) } EXPORT_SYMBOL(ns_to_timespec64); #endif -/* - * When we convert to jiffies then we interpret incoming values - * the following way: +/** + * msecs_to_jiffies: - convert milliseconds to jiffies + * @m: time in milliseconds + * + * conversion is done as follows: * * - negative values mean 'infinite timeout' (MAX_JIFFY_OFFSET) * @@ -493,66 +499,36 @@ EXPORT_SYMBOL(ns_to_timespec64); * MAX_JIFFY_OFFSET values] mean 'infinite timeout' too. * * - all other values are converted to jiffies by either multiplying - * the input value by a factor or dividing it with a factor - * - * We must also be careful about 32-bit overflows. + * the input value by a factor or dividing it with a factor and + * handling any 32-bit overflows. + * for the details see __msecs_to_jiffies() + * + * msecs_to_jiffies() checks for the passed in value being a constant + * via __builtin_constant_p() allowing gcc to eliminate most of the + * code, __msecs_to_jiffies() is called if the value passed does not + * allow constant folding and the actual conversion must be done at + * runtime. + * the _msecs_to_jiffies helpers are the HZ dependent conversion + * routines found in include/linux/jiffies.h */ -unsigned long msecs_to_jiffies(const unsigned int m) +unsigned long __msecs_to_jiffies(const unsigned int m) { /* * Negative value, means infinite timeout: */ if ((int)m < 0) return MAX_JIFFY_OFFSET; - -#if HZ <= MSEC_PER_SEC && !(MSEC_PER_SEC % HZ) - /* - * HZ is equal to or smaller than 1000, and 1000 is a nice - * round multiple of HZ, divide with the factor between them, - * but round upwards: - */ - return (m + (MSEC_PER_SEC / HZ) - 1) / (MSEC_PER_SEC / HZ); -#elif HZ > MSEC_PER_SEC && !(HZ % MSEC_PER_SEC) - /* - * HZ is larger than 1000, and HZ is a nice round multiple of - * 1000 - simply multiply with the factor between them. - * - * But first make sure the multiplication result cannot - * overflow: - */ - if (m > jiffies_to_msecs(MAX_JIFFY_OFFSET)) - return MAX_JIFFY_OFFSET; - - return m * (HZ / MSEC_PER_SEC); -#else - /* - * Generic case - multiply, round and divide. But first - * check that if we are doing a net multiplication, that - * we wouldn't overflow: - */ - if (HZ > MSEC_PER_SEC && m > jiffies_to_msecs(MAX_JIFFY_OFFSET)) - return MAX_JIFFY_OFFSET; - - return (MSEC_TO_HZ_MUL32 * m + MSEC_TO_HZ_ADJ32) - >> MSEC_TO_HZ_SHR32; -#endif + return _msecs_to_jiffies(m); } -EXPORT_SYMBOL(msecs_to_jiffies); +EXPORT_SYMBOL(__msecs_to_jiffies); -unsigned long usecs_to_jiffies(const unsigned int u) +unsigned long __usecs_to_jiffies(const unsigned int u) { if (u > jiffies_to_usecs(MAX_JIFFY_OFFSET)) return MAX_JIFFY_OFFSET; -#if HZ <= USEC_PER_SEC && !(USEC_PER_SEC % HZ) - return (u + (USEC_PER_SEC / HZ) - 1) / (USEC_PER_SEC / HZ); -#elif HZ > USEC_PER_SEC && !(HZ % USEC_PER_SEC) - return u * (HZ / USEC_PER_SEC); -#else - return (USEC_TO_HZ_MUL32 * u + USEC_TO_HZ_ADJ32) - >> USEC_TO_HZ_SHR32; -#endif + return _usecs_to_jiffies(u); } -EXPORT_SYMBOL(usecs_to_jiffies); +EXPORT_SYMBOL(__usecs_to_jiffies); /* * The TICK_NSEC - 1 rounds up the value to the next resolution. Note diff --git a/kernel/time/timeconst.bc b/kernel/time/timeconst.bc index 511bdf2..c7388de 100644 --- a/kernel/time/timeconst.bc +++ b/kernel/time/timeconst.bc @@ -50,7 +50,7 @@ define timeconst(hz) { print "#include <linux/types.h>\n\n" print "#if HZ != ", hz, "\n" - print "#error \qkernel/timeconst.h has the wrong HZ value!\q\n" + print "#error \qinclude/generated/timeconst.h has the wrong HZ value!\q\n" print "#endif\n\n" if (hz < 2) { @@ -105,4 +105,5 @@ define timeconst(hz) { halt } +hz = read(); timeconst(hz) diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index 946acb7..30b7a40 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -118,18 +118,6 @@ static inline void tk_update_sleep_time(struct timekeeper *tk, ktime_t 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) { @@ -149,29 +137,30 @@ static void timekeeping_check_update(struct timekeeper *tk, cycle_t offset) } } - if (timekeeping_underflow_seen) { - if (jiffies - timekeeping_last_warning > WARNING_FREQ) { + if (tk->underflow_seen) { + if (jiffies - tk->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; + tk->last_warning = jiffies; } - timekeeping_underflow_seen = 0; + tk->underflow_seen = 0; } - if (timekeeping_overflow_seen) { - if (jiffies - timekeeping_last_warning > WARNING_FREQ) { + if (tk->overflow_seen) { + if (jiffies - tk->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; + tk->last_warning = jiffies; } - timekeeping_overflow_seen = 0; + tk->overflow_seen = 0; } } static inline cycle_t timekeeping_get_delta(struct tk_read_base *tkr) { + struct timekeeper *tk = &tk_core.timekeeper; cycle_t now, last, mask, max, delta; unsigned int seq; @@ -197,13 +186,13 @@ static inline cycle_t timekeeping_get_delta(struct tk_read_base *tkr) * mask-relative negative values. */ if (unlikely((~delta & mask) < (mask >> 3))) { - timekeeping_underflow_seen = 1; + tk->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; + tk->overflow_seen = 1; delta = tkr->clock->max_cycles; } @@ -551,6 +540,17 @@ int pvclock_gtod_unregister_notifier(struct notifier_block *nb) EXPORT_SYMBOL_GPL(pvclock_gtod_unregister_notifier); /* + * tk_update_leap_state - helper to update the next_leap_ktime + */ +static inline void tk_update_leap_state(struct timekeeper *tk) +{ + tk->next_leap_ktime = ntp_get_next_leap(); + if (tk->next_leap_ktime.tv64 != KTIME_MAX) + /* Convert to monotonic time */ + tk->next_leap_ktime = ktime_sub(tk->next_leap_ktime, tk->offs_real); +} + +/* * Update the ktime_t based scalar nsec members of the timekeeper */ static inline void tk_update_ktime_data(struct timekeeper *tk) @@ -591,17 +591,25 @@ static void timekeeping_update(struct timekeeper *tk, unsigned int action) ntp_clear(); } + tk_update_leap_state(tk); tk_update_ktime_data(tk); update_vsyscall(tk); update_pvclock_gtod(tk, action & TK_CLOCK_WAS_SET); + update_fast_timekeeper(&tk->tkr_mono, &tk_fast_mono); + update_fast_timekeeper(&tk->tkr_raw, &tk_fast_raw); + + if (action & TK_CLOCK_WAS_SET) + tk->clock_was_set_seq++; + /* + * The mirroring of the data to the shadow-timekeeper needs + * to happen last here to ensure we don't over-write the + * timekeeper structure on the next update with stale data + */ if (action & TK_MIRROR) memcpy(&shadow_timekeeper, &tk_core.timekeeper, sizeof(tk_core.timekeeper)); - - update_fast_timekeeper(&tk->tkr_mono, &tk_fast_mono); - update_fast_timekeeper(&tk->tkr_raw, &tk_fast_raw); } /** @@ -699,6 +707,23 @@ ktime_t ktime_get(void) } EXPORT_SYMBOL_GPL(ktime_get); +u32 ktime_get_resolution_ns(void) +{ + struct timekeeper *tk = &tk_core.timekeeper; + unsigned int seq; + u32 nsecs; + + WARN_ON(timekeeping_suspended); + + do { + seq = read_seqcount_begin(&tk_core.seq); + nsecs = tk->tkr_mono.mult >> tk->tkr_mono.shift; + } while (read_seqcount_retry(&tk_core.seq, seq)); + + return nsecs; +} +EXPORT_SYMBOL_GPL(ktime_get_resolution_ns); + static ktime_t *offsets[TK_OFFS_MAX] = { [TK_OFFS_REAL] = &tk_core.timekeeper.offs_real, [TK_OFFS_BOOT] = &tk_core.timekeeper.offs_boot, @@ -1179,28 +1204,20 @@ void __weak read_persistent_clock64(struct timespec64 *ts64) } /** - * read_boot_clock - Return time of the system start. + * read_boot_clock64 - Return time of the system start. * * Weak dummy function for arches that do not yet support it. * Function to read the exact time the system has been started. - * Returns a timespec with tv_sec=0 and tv_nsec=0 if unsupported. + * Returns a timespec64 with tv_sec=0 and tv_nsec=0 if unsupported. * * XXX - Do be sure to remove it once all arches implement it. */ -void __weak read_boot_clock(struct timespec *ts) +void __weak read_boot_clock64(struct timespec64 *ts) { ts->tv_sec = 0; 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; @@ -1836,8 +1853,9 @@ void update_wall_time(void) * memcpy under the tk_core.seq against one before we start * updating. */ + timekeeping_update(tk, clock_set); memcpy(real_tk, tk, sizeof(*tk)); - timekeeping_update(real_tk, clock_set); + /* The memcpy must come last. Do not put anything here! */ write_seqcount_end(&tk_core.seq); out: raw_spin_unlock_irqrestore(&timekeeper_lock, flags); @@ -1926,47 +1944,20 @@ void do_timer(unsigned long ticks) } /** - * ktime_get_update_offsets_tick - hrtimer helper - * @offs_real: pointer to storage for monotonic -> realtime offset - * @offs_boot: pointer to storage for monotonic -> boottime offset - * @offs_tai: pointer to storage for monotonic -> clock tai offset - * - * Returns monotonic time at last tick and various offsets - */ -ktime_t ktime_get_update_offsets_tick(ktime_t *offs_real, ktime_t *offs_boot, - ktime_t *offs_tai) -{ - struct timekeeper *tk = &tk_core.timekeeper; - unsigned int seq; - ktime_t base; - u64 nsecs; - - do { - seq = read_seqcount_begin(&tk_core.seq); - - 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; - *offs_tai = tk->offs_tai; - } while (read_seqcount_retry(&tk_core.seq, seq)); - - return ktime_add_ns(base, nsecs); -} - -#ifdef CONFIG_HIGH_RES_TIMERS -/** * ktime_get_update_offsets_now - hrtimer helper + * @cwsseq: pointer to check and store the clock was set sequence number * @offs_real: pointer to storage for monotonic -> realtime offset * @offs_boot: pointer to storage for monotonic -> boottime offset * @offs_tai: pointer to storage for monotonic -> clock tai offset * - * Returns current monotonic time and updates the offsets + * Returns current monotonic time and updates the offsets if the + * sequence number in @cwsseq and timekeeper.clock_was_set_seq are + * different. + * * Called from hrtimer_interrupt() or retrigger_next_event() */ -ktime_t ktime_get_update_offsets_now(ktime_t *offs_real, ktime_t *offs_boot, - ktime_t *offs_tai) +ktime_t ktime_get_update_offsets_now(unsigned int *cwsseq, ktime_t *offs_real, + ktime_t *offs_boot, ktime_t *offs_tai) { struct timekeeper *tk = &tk_core.timekeeper; unsigned int seq; @@ -1978,15 +1969,23 @@ ktime_t ktime_get_update_offsets_now(ktime_t *offs_real, ktime_t *offs_boot, base = tk->tkr_mono.base; nsecs = timekeeping_get_ns(&tk->tkr_mono); + base = ktime_add_ns(base, nsecs); + + if (*cwsseq != tk->clock_was_set_seq) { + *cwsseq = tk->clock_was_set_seq; + *offs_real = tk->offs_real; + *offs_boot = tk->offs_boot; + *offs_tai = tk->offs_tai; + } + + /* Handle leapsecond insertion adjustments */ + if (unlikely(base.tv64 >= tk->next_leap_ktime.tv64)) + *offs_real = ktime_sub(tk->offs_real, ktime_set(1, 0)); - *offs_real = tk->offs_real; - *offs_boot = tk->offs_boot; - *offs_tai = tk->offs_tai; } while (read_seqcount_retry(&tk_core.seq, seq)); - return ktime_add_ns(base, nsecs); + return base; } -#endif /** * do_adjtimex() - Accessor function to NTP __do_adjtimex function @@ -2027,6 +2026,8 @@ int do_adjtimex(struct timex *txc) __timekeeping_set_tai_offset(tk, tai); timekeeping_update(tk, TK_MIRROR | TK_CLOCK_WAS_SET); } + tk_update_leap_state(tk); + write_seqcount_end(&tk_core.seq); raw_spin_unlock_irqrestore(&timekeeper_lock, flags); diff --git a/kernel/time/timekeeping.h b/kernel/time/timekeeping.h index ead8794..704f595 100644 --- a/kernel/time/timekeeping.h +++ b/kernel/time/timekeeping.h @@ -3,19 +3,16 @@ /* * Internal interfaces for kernel/time/ */ -extern ktime_t ktime_get_update_offsets_tick(ktime_t *offs_real, - ktime_t *offs_boot, - ktime_t *offs_tai); -extern ktime_t ktime_get_update_offsets_now(ktime_t *offs_real, - ktime_t *offs_boot, - ktime_t *offs_tai); +extern ktime_t ktime_get_update_offsets_now(unsigned int *cwsseq, + ktime_t *offs_real, + ktime_t *offs_boot, + ktime_t *offs_tai); extern int timekeeping_valid_for_hres(void); extern u64 timekeeping_max_deferment(void); extern int timekeeping_inject_offset(struct timespec *ts); extern s32 timekeeping_get_tai_offset(void); extern void timekeeping_set_tai_offset(s32 tai_offset); -extern void timekeeping_clocktai(struct timespec *ts); extern int timekeeping_suspend(void); extern void timekeeping_resume(void); diff --git a/kernel/time/timer.c b/kernel/time/timer.c index 2ece3aa..520499d 100644 --- a/kernel/time/timer.c +++ b/kernel/time/timer.c @@ -49,6 +49,8 @@ #include <asm/timex.h> #include <asm/io.h> +#include "tick-internal.h" + #define CREATE_TRACE_POINTS #include <trace/events/timer.h> @@ -68,11 +70,11 @@ EXPORT_SYMBOL(jiffies_64); #define MAX_TVAL ((unsigned long)((1ULL << (TVR_BITS + 4*TVN_BITS)) - 1)) struct tvec { - struct list_head vec[TVN_SIZE]; + struct hlist_head vec[TVN_SIZE]; }; struct tvec_root { - struct list_head vec[TVR_SIZE]; + struct hlist_head vec[TVR_SIZE]; }; struct tvec_base { @@ -83,6 +85,8 @@ struct tvec_base { unsigned long active_timers; unsigned long all_timers; int cpu; + bool migration_enabled; + bool nohz_active; struct tvec_root tv1; struct tvec tv2; struct tvec tv3; @@ -90,43 +94,60 @@ 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; +static DEFINE_PER_CPU(struct tvec_base, tvec_bases); + +#if defined(CONFIG_SMP) && defined(CONFIG_NO_HZ_COMMON) +unsigned int sysctl_timer_migration = 1; -/* Functions below help us manage 'deferrable' flag */ -static inline unsigned int tbase_get_deferrable(struct tvec_base *base) +void timers_update_migration(bool update_nohz) { - return ((unsigned int)(unsigned long)base & TIMER_DEFERRABLE); + bool on = sysctl_timer_migration && tick_nohz_active; + unsigned int cpu; + + /* Avoid the loop, if nothing to update */ + if (this_cpu_read(tvec_bases.migration_enabled) == on) + return; + + for_each_possible_cpu(cpu) { + per_cpu(tvec_bases.migration_enabled, cpu) = on; + per_cpu(hrtimer_bases.migration_enabled, cpu) = on; + if (!update_nohz) + continue; + per_cpu(tvec_bases.nohz_active, cpu) = true; + per_cpu(hrtimer_bases.nohz_active, cpu) = true; + } } -static inline unsigned int tbase_get_irqsafe(struct tvec_base *base) +int timer_migration_handler(struct ctl_table *table, int write, + void __user *buffer, size_t *lenp, + loff_t *ppos) { - return ((unsigned int)(unsigned long)base & TIMER_IRQSAFE); + static DEFINE_MUTEX(mutex); + int ret; + + mutex_lock(&mutex); + ret = proc_dointvec(table, write, buffer, lenp, ppos); + if (!ret && write) + timers_update_migration(false); + mutex_unlock(&mutex); + return ret; } -static inline struct tvec_base *tbase_get_base(struct tvec_base *base) +static inline struct tvec_base *get_target_base(struct tvec_base *base, + int pinned) { - return ((struct tvec_base *)((unsigned long)base & ~TIMER_FLAG_MASK)); + if (pinned || !base->migration_enabled) + return this_cpu_ptr(&tvec_bases); + return per_cpu_ptr(&tvec_bases, get_nohz_timer_target()); } - -static inline void -timer_set_base(struct timer_list *timer, struct tvec_base *new_base) +#else +static inline struct tvec_base *get_target_base(struct tvec_base *base, + int pinned) { - unsigned long flags = (unsigned long)timer->base & TIMER_FLAG_MASK; - - timer->base = (struct tvec_base *)((unsigned long)(new_base) | flags); + return this_cpu_ptr(&tvec_bases); } +#endif static unsigned long round_jiffies_common(unsigned long j, int cpu, bool force_up) @@ -349,26 +370,12 @@ void set_timer_slack(struct timer_list *timer, int slack_hz) } EXPORT_SYMBOL_GPL(set_timer_slack); -/* - * If the list is empty, catch up ->timer_jiffies to the current time. - * The caller must hold the tvec_base lock. Returns true if the list - * was empty and therefore ->timer_jiffies was updated. - */ -static bool catchup_timer_jiffies(struct tvec_base *base) -{ - if (!base->all_timers) { - base->timer_jiffies = jiffies; - return true; - } - return false; -} - static void __internal_add_timer(struct tvec_base *base, struct timer_list *timer) { unsigned long expires = timer->expires; unsigned long idx = expires - base->timer_jiffies; - struct list_head *vec; + struct hlist_head *vec; if (idx < TVR_SIZE) { int i = expires & TVR_MASK; @@ -401,25 +408,25 @@ __internal_add_timer(struct tvec_base *base, struct timer_list *timer) i = (expires >> (TVR_BITS + 3 * TVN_BITS)) & TVN_MASK; vec = base->tv5.vec + i; } - /* - * Timers are FIFO: - */ - list_add_tail(&timer->entry, vec); + + hlist_add_head(&timer->entry, vec); } static void internal_add_timer(struct tvec_base *base, struct timer_list *timer) { - (void)catchup_timer_jiffies(base); + /* Advance base->jiffies, if the base is empty */ + if (!base->all_timers++) + base->timer_jiffies = jiffies; + __internal_add_timer(base, timer); /* * Update base->active_timers and base->next_timer */ - if (!tbase_get_deferrable(timer->base)) { + if (!(timer->flags & TIMER_DEFERRABLE)) { if (!base->active_timers++ || time_before(timer->expires, base->next_timer)) base->next_timer = timer->expires; } - base->all_timers++; /* * Check whether the other CPU is in dynticks mode and needs @@ -434,8 +441,11 @@ static void internal_add_timer(struct tvec_base *base, struct timer_list *timer) * require special care against races with idle_cpu(), lets deal * with that later. */ - if (!tbase_get_deferrable(base) || tick_nohz_full_cpu(base->cpu)) - wake_up_nohz_cpu(base->cpu); + if (base->nohz_active) { + if (!(timer->flags & TIMER_DEFERRABLE) || + tick_nohz_full_cpu(base->cpu)) + wake_up_nohz_cpu(base->cpu); + } } #ifdef CONFIG_TIMER_STATS @@ -451,15 +461,12 @@ void __timer_stats_timer_set_start_info(struct timer_list *timer, void *addr) static void timer_stats_account_timer(struct timer_list *timer) { - unsigned int flag = 0; - if (likely(!timer->start_site)) return; - if (unlikely(tbase_get_deferrable(timer->base))) - flag |= TIMER_STATS_FLAG_DEFERRABLE; timer_stats_update_stats(timer, timer->start_pid, timer->start_site, - timer->function, timer->start_comm, flag); + timer->function, timer->start_comm, + timer->flags); } #else @@ -516,8 +523,8 @@ static int timer_fixup_activate(void *addr, enum debug_obj_state state) * statically initialized. We just make sure that it * is tracked in the object tracker. */ - if (timer->entry.next == NULL && - timer->entry.prev == TIMER_ENTRY_STATIC) { + if (timer->entry.pprev == NULL && + timer->entry.next == TIMER_ENTRY_STATIC) { debug_object_init(timer, &timer_debug_descr); debug_object_activate(timer, &timer_debug_descr); return 0; @@ -563,7 +570,7 @@ static int timer_fixup_assert_init(void *addr, enum debug_obj_state state) switch (state) { case ODEBUG_STATE_NOTAVAILABLE: - if (timer->entry.prev == TIMER_ENTRY_STATIC) { + if (timer->entry.next == TIMER_ENTRY_STATIC) { /* * This is not really a fixup. The timer was * statically initialized. We just make sure that it @@ -648,7 +655,7 @@ static inline void debug_activate(struct timer_list *timer, unsigned long expires) { debug_timer_activate(timer); - trace_timer_start(timer, expires); + trace_timer_start(timer, expires, timer->flags); } static inline void debug_deactivate(struct timer_list *timer) @@ -665,10 +672,8 @@ static inline void debug_assert_init(struct timer_list *timer) static void do_init_timer(struct timer_list *timer, unsigned int flags, const char *name, struct lock_class_key *key) { - struct tvec_base *base = raw_cpu_read(tvec_bases); - - timer->entry.next = NULL; - timer->base = (void *)((unsigned long)base | flags); + timer->entry.pprev = NULL; + timer->flags = flags | raw_smp_processor_id(); timer->slack = -1; #ifdef CONFIG_TIMER_STATS timer->start_site = NULL; @@ -699,24 +704,23 @@ EXPORT_SYMBOL(init_timer_key); static inline void detach_timer(struct timer_list *timer, bool clear_pending) { - struct list_head *entry = &timer->entry; + struct hlist_node *entry = &timer->entry; debug_deactivate(timer); - __list_del(entry->prev, entry->next); + __hlist_del(entry); if (clear_pending) - entry->next = NULL; - entry->prev = LIST_POISON2; + entry->pprev = NULL; + entry->next = LIST_POISON2; } static inline void detach_expired_timer(struct timer_list *timer, struct tvec_base *base) { detach_timer(timer, true); - if (!tbase_get_deferrable(timer->base)) + if (!(timer->flags & TIMER_DEFERRABLE)) base->active_timers--; base->all_timers--; - (void)catchup_timer_jiffies(base); } static int detach_if_pending(struct timer_list *timer, struct tvec_base *base, @@ -726,13 +730,14 @@ static int detach_if_pending(struct timer_list *timer, struct tvec_base *base, return 0; detach_timer(timer, clear_pending); - if (!tbase_get_deferrable(timer->base)) { + if (!(timer->flags & TIMER_DEFERRABLE)) { base->active_timers--; if (timer->expires == base->next_timer) base->next_timer = base->timer_jiffies; } - base->all_timers--; - (void)catchup_timer_jiffies(base); + /* If this was the last timer, advance base->jiffies */ + if (!--base->all_timers) + base->timer_jiffies = jiffies; return 1; } @@ -744,24 +749,22 @@ static int detach_if_pending(struct timer_list *timer, struct tvec_base *base, * So __run_timers/migrate_timers can safely modify all timers which could * be found on ->tvX lists. * - * When the timer's base is locked, and the timer removed from list, it is - * possible to set timer->base = NULL and drop the lock: the timer remains - * locked. + * When the timer's base is locked and removed from the list, the + * TIMER_MIGRATING flag is set, FIXME */ static struct tvec_base *lock_timer_base(struct timer_list *timer, unsigned long *flags) __acquires(timer->base->lock) { - struct tvec_base *base; - for (;;) { - struct tvec_base *prelock_base = timer->base; - base = tbase_get_base(prelock_base); - if (likely(base != NULL)) { + u32 tf = timer->flags; + struct tvec_base *base; + + if (!(tf & TIMER_MIGRATING)) { + base = per_cpu_ptr(&tvec_bases, tf & TIMER_CPUMASK); spin_lock_irqsave(&base->lock, *flags); - if (likely(prelock_base == timer->base)) + if (timer->flags == tf) return base; - /* The timer has migrated to another CPU */ spin_unlock_irqrestore(&base->lock, *flags); } cpu_relax(); @@ -770,11 +773,11 @@ static struct tvec_base *lock_timer_base(struct timer_list *timer, static inline int __mod_timer(struct timer_list *timer, unsigned long expires, - bool pending_only, int pinned) + bool pending_only, int pinned) { struct tvec_base *base, *new_base; unsigned long flags; - int ret = 0 , cpu; + int ret = 0; timer_stats_timer_set_start_info(timer); BUG_ON(!timer->function); @@ -787,8 +790,7 @@ __mod_timer(struct timer_list *timer, unsigned long expires, debug_activate(timer, expires); - cpu = get_nohz_timer_target(pinned); - new_base = per_cpu(tvec_bases, cpu); + new_base = get_target_base(base, pinned); if (base != new_base) { /* @@ -800,11 +802,13 @@ __mod_timer(struct timer_list *timer, unsigned long expires, */ if (likely(base->running_timer != timer)) { /* See the comment in lock_timer_base() */ - timer_set_base(timer, NULL); + timer->flags |= TIMER_MIGRATING; + spin_unlock(&base->lock); base = new_base; spin_lock(&base->lock); - timer_set_base(timer, base); + timer->flags &= ~TIMER_BASEMASK; + timer->flags |= base->cpu; } } @@ -966,13 +970,13 @@ EXPORT_SYMBOL(add_timer); */ void add_timer_on(struct timer_list *timer, int cpu) { - struct tvec_base *base = per_cpu(tvec_bases, cpu); + struct tvec_base *base = per_cpu_ptr(&tvec_bases, cpu); unsigned long flags; timer_stats_timer_set_start_info(timer); BUG_ON(timer_pending(timer) || !timer->function); spin_lock_irqsave(&base->lock, flags); - timer_set_base(timer, base); + timer->flags = (timer->flags & ~TIMER_BASEMASK) | cpu; debug_activate(timer, timer->expires); internal_add_timer(base, timer); spin_unlock_irqrestore(&base->lock, flags); @@ -1037,8 +1041,6 @@ 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 @@ -1093,7 +1095,7 @@ int del_timer_sync(struct timer_list *timer) * don't use it in hardirq context, because it * could lead to deadlock. */ - WARN_ON(in_irq() && !tbase_get_irqsafe(timer->base)); + WARN_ON(in_irq() && !(timer->flags & TIMER_IRQSAFE)); for (;;) { int ret = try_to_del_timer_sync(timer); if (ret >= 0) @@ -1107,17 +1109,17 @@ EXPORT_SYMBOL(del_timer_sync); static int cascade(struct tvec_base *base, struct tvec *tv, int index) { /* cascade all the timers from tv up one level */ - struct timer_list *timer, *tmp; - struct list_head tv_list; + struct timer_list *timer; + struct hlist_node *tmp; + struct hlist_head tv_list; - list_replace_init(tv->vec + index, &tv_list); + hlist_move_list(tv->vec + index, &tv_list); /* * We are removing _all_ timers from the list, so we * don't have to detach them individually. */ - list_for_each_entry_safe(timer, tmp, &tv_list, entry) { - BUG_ON(tbase_get_base(timer->base) != base); + hlist_for_each_entry_safe(timer, tmp, &tv_list, entry) { /* No accounting, while moving them */ __internal_add_timer(base, timer); } @@ -1182,14 +1184,18 @@ static inline void __run_timers(struct tvec_base *base) struct timer_list *timer; spin_lock_irq(&base->lock); - if (catchup_timer_jiffies(base)) { - spin_unlock_irq(&base->lock); - return; - } + while (time_after_eq(jiffies, base->timer_jiffies)) { - struct list_head work_list; - struct list_head *head = &work_list; - int index = base->timer_jiffies & TVR_MASK; + struct hlist_head work_list; + struct hlist_head *head = &work_list; + int index; + + if (!base->all_timers) { + base->timer_jiffies = jiffies; + break; + } + + index = base->timer_jiffies & TVR_MASK; /* * Cascade timers: @@ -1200,16 +1206,16 @@ static inline void __run_timers(struct tvec_base *base) !cascade(base, &base->tv4, INDEX(2))) cascade(base, &base->tv5, INDEX(3)); ++base->timer_jiffies; - list_replace_init(base->tv1.vec + index, head); - while (!list_empty(head)) { + hlist_move_list(base->tv1.vec + index, head); + while (!hlist_empty(head)) { void (*fn)(unsigned long); unsigned long data; bool irqsafe; - timer = list_first_entry(head, struct timer_list,entry); + timer = hlist_entry(head->first, struct timer_list, entry); fn = timer->function; data = timer->data; - irqsafe = tbase_get_irqsafe(timer->base); + irqsafe = timer->flags & TIMER_IRQSAFE; timer_stats_account_timer(timer); @@ -1248,8 +1254,8 @@ static unsigned long __next_timer_interrupt(struct tvec_base *base) /* Look for timer events in tv1. */ index = slot = timer_jiffies & TVR_MASK; do { - list_for_each_entry(nte, base->tv1.vec + slot, entry) { - if (tbase_get_deferrable(nte->base)) + hlist_for_each_entry(nte, base->tv1.vec + slot, entry) { + if (nte->flags & TIMER_DEFERRABLE) continue; found = 1; @@ -1279,8 +1285,8 @@ cascade: index = slot = timer_jiffies & TVN_MASK; do { - list_for_each_entry(nte, varp->vec + slot, entry) { - if (tbase_get_deferrable(nte->base)) + hlist_for_each_entry(nte, varp->vec + slot, entry) { + if (nte->flags & TIMER_DEFERRABLE) continue; found = 1; @@ -1311,54 +1317,48 @@ cascade: * Check, if the next hrtimer event is before the next timer wheel * event: */ -static unsigned long cmp_next_hrtimer_event(unsigned long now, - unsigned long expires) +static u64 cmp_next_hrtimer_event(u64 basem, u64 expires) { - ktime_t hr_delta = hrtimer_get_next_event(); - struct timespec tsdelta; - unsigned long delta; - - if (hr_delta.tv64 == KTIME_MAX) - return expires; + u64 nextevt = hrtimer_get_next_event(); /* - * Expired timer available, let it expire in the next tick + * If high resolution timers are enabled + * hrtimer_get_next_event() returns KTIME_MAX. */ - if (hr_delta.tv64 <= 0) - return now + 1; - - tsdelta = ktime_to_timespec(hr_delta); - delta = timespec_to_jiffies(&tsdelta); + if (expires <= nextevt) + return expires; /* - * Limit the delta to the max value, which is checked in - * tick_nohz_stop_sched_tick(): + * If the next timer is already expired, return the tick base + * time so the tick is fired immediately. */ - if (delta > NEXT_TIMER_MAX_DELTA) - delta = NEXT_TIMER_MAX_DELTA; + if (nextevt <= basem) + return basem; /* - * Take rounding errors in to account and make sure, that it - * expires in the next tick. Otherwise we go into an endless - * ping pong due to tick_nohz_stop_sched_tick() retriggering - * the timer softirq + * Round up to the next jiffie. High resolution timers are + * off, so the hrtimers are expired in the tick and we need to + * make sure that this tick really expires the timer to avoid + * a ping pong of the nohz stop code. + * + * Use DIV_ROUND_UP_ULL to prevent gcc calling __divdi3 */ - if (delta < 1) - delta = 1; - now += delta; - if (time_before(now, expires)) - return now; - return expires; + return DIV_ROUND_UP_ULL(nextevt, TICK_NSEC) * TICK_NSEC; } /** - * get_next_timer_interrupt - return the jiffy of the next pending timer - * @now: current time (in jiffies) + * get_next_timer_interrupt - return the time (clock mono) of the next timer + * @basej: base time jiffies + * @basem: base time clock monotonic + * + * Returns the tick aligned clock monotonic time of the next pending + * timer or KTIME_MAX if no timer is pending. */ -unsigned long get_next_timer_interrupt(unsigned long now) +u64 get_next_timer_interrupt(unsigned long basej, u64 basem) { - struct tvec_base *base = __this_cpu_read(tvec_bases); - unsigned long expires = now + NEXT_TIMER_MAX_DELTA; + struct tvec_base *base = this_cpu_ptr(&tvec_bases); + u64 expires = KTIME_MAX; + unsigned long nextevt; /* * Pretend that there is no timer pending if the cpu is offline. @@ -1371,14 +1371,15 @@ unsigned long get_next_timer_interrupt(unsigned long now) if (base->active_timers) { if (time_before_eq(base->next_timer, base->timer_jiffies)) base->next_timer = __next_timer_interrupt(base); - expires = base->next_timer; + nextevt = base->next_timer; + if (time_before_eq(nextevt, basej)) + expires = basem; + else + expires = basem + (nextevt - basej) * TICK_NSEC; } spin_unlock(&base->lock); - if (time_before_eq(expires, now)) - return now; - - return cmp_next_hrtimer_event(now, expires); + return cmp_next_hrtimer_event(basem, expires); } #endif @@ -1407,9 +1408,7 @@ void update_process_times(int user_tick) */ static void run_timer_softirq(struct softirq_action *h) { - struct tvec_base *base = __this_cpu_read(tvec_bases); - - hrtimer_run_pending(); + struct tvec_base *base = this_cpu_ptr(&tvec_bases); if (time_after_eq(jiffies, base->timer_jiffies)) __run_timers(base); @@ -1545,15 +1544,16 @@ signed long __sched schedule_timeout_uninterruptible(signed long timeout) EXPORT_SYMBOL(schedule_timeout_uninterruptible); #ifdef CONFIG_HOTPLUG_CPU -static void migrate_timer_list(struct tvec_base *new_base, struct list_head *head) +static void migrate_timer_list(struct tvec_base *new_base, struct hlist_head *head) { struct timer_list *timer; + int cpu = new_base->cpu; - while (!list_empty(head)) { - timer = list_first_entry(head, struct timer_list, entry); + while (!hlist_empty(head)) { + timer = hlist_entry(head->first, struct timer_list, entry); /* We ignore the accounting on the dying cpu */ detach_timer(timer, false); - timer_set_base(timer, new_base); + timer->flags = (timer->flags & ~TIMER_BASEMASK) | cpu; internal_add_timer(new_base, timer); } } @@ -1565,8 +1565,8 @@ static void migrate_timers(int cpu) int i; BUG_ON(cpu_online(cpu)); - old_base = per_cpu(tvec_bases, cpu); - new_base = get_cpu_var(tvec_bases); + old_base = per_cpu_ptr(&tvec_bases, cpu); + new_base = this_cpu_ptr(&tvec_bases); /* * The caller is globally serialized and nobody else * takes two locks at once, deadlock is not possible. @@ -1590,7 +1590,6 @@ static void migrate_timers(int cpu) spin_unlock(&old_base->lock); spin_unlock_irq(&new_base->lock); - put_cpu_var(tvec_bases); } static int timer_cpu_notify(struct notifier_block *self, @@ -1616,52 +1615,27 @@ static inline void timer_register_cpu_notifier(void) static inline void timer_register_cpu_notifier(void) { } #endif /* CONFIG_HOTPLUG_CPU */ -static void __init init_timer_cpu(struct tvec_base *base, int cpu) +static void __init init_timer_cpu(int cpu) { - int j; - - BUG_ON(base != tbase_get_base(base)); + struct tvec_base *base = per_cpu_ptr(&tvec_bases, cpu); 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) { - 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); - } + for_each_possible_cpu(cpu) + init_timer_cpu(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); - init_timer_cpus(); init_timer_stats(); timer_register_cpu_notifier(); @@ -1697,14 +1671,14 @@ unsigned long msleep_interruptible(unsigned int msecs) EXPORT_SYMBOL(msleep_interruptible); -static int __sched do_usleep_range(unsigned long min, unsigned long max) +static void __sched do_usleep_range(unsigned long min, unsigned long max) { ktime_t kmin; unsigned long delta; kmin = ktime_set(0, min * NSEC_PER_USEC); delta = (max - min) * NSEC_PER_USEC; - return schedule_hrtimeout_range(&kmin, delta, HRTIMER_MODE_REL); + schedule_hrtimeout_range(&kmin, delta, HRTIMER_MODE_REL); } /** @@ -1712,7 +1686,7 @@ static int __sched do_usleep_range(unsigned long min, unsigned long max) * @min: Minimum time in usecs to sleep * @max: Maximum time in usecs to sleep */ -void usleep_range(unsigned long min, unsigned long max) +void __sched usleep_range(unsigned long min, unsigned long max) { __set_current_state(TASK_UNINTERRUPTIBLE); do_usleep_range(min, max); diff --git a/kernel/time/timer_list.c b/kernel/time/timer_list.c index e878c2e..a4536e1 100644 --- a/kernel/time/timer_list.c +++ b/kernel/time/timer_list.c @@ -29,19 +29,24 @@ struct timer_list_iter { typedef void (*print_fn_t)(struct seq_file *m, unsigned int *classes); -DECLARE_PER_CPU(struct hrtimer_cpu_base, hrtimer_bases); - /* * This allows printing both to /proc/timer_list and * to the console (on SysRq-Q): */ -#define SEQ_printf(m, x...) \ - do { \ - if (m) \ - seq_printf(m, x); \ - else \ - printk(x); \ - } while (0) +__printf(2, 3) +static void SEQ_printf(struct seq_file *m, const char *fmt, ...) +{ + va_list args; + + va_start(args, fmt); + + if (m) + seq_vprintf(m, fmt, args); + else + vprintk(fmt, args); + + va_end(args); +} static void print_name_offset(struct seq_file *m, void *sym) { @@ -120,10 +125,10 @@ static void print_base(struct seq_file *m, struct hrtimer_clock_base *base, u64 now) { SEQ_printf(m, " .base: %pK\n", base); - SEQ_printf(m, " .index: %d\n", - base->index); - SEQ_printf(m, " .resolution: %Lu nsecs\n", - (unsigned long long)ktime_to_ns(base->resolution)); + SEQ_printf(m, " .index: %d\n", base->index); + + SEQ_printf(m, " .resolution: %u nsecs\n", (unsigned) hrtimer_resolution); + SEQ_printf(m, " .get_time: "); print_name_offset(m, base->get_time); SEQ_printf(m, "\n"); @@ -158,7 +163,7 @@ static void print_cpu(struct seq_file *m, int cpu, u64 now) P(nr_events); P(nr_retries); P(nr_hangs); - P_ns(max_hang_time); + P(max_hang_time); #endif #undef P #undef P_ns @@ -184,7 +189,7 @@ static void print_cpu(struct seq_file *m, int cpu, u64 now) P_ns(idle_sleeptime); P_ns(iowait_sleeptime); P(last_jiffies); - P(next_jiffies); + P(next_timer); P_ns(idle_expires); SEQ_printf(m, "jiffies: %Lu\n", (unsigned long long)jiffies); @@ -251,6 +256,12 @@ print_tickdevice(struct seq_file *m, struct tick_device *td, int cpu) SEQ_printf(m, "\n"); } + if (dev->set_state_oneshot_stopped) { + SEQ_printf(m, " oneshot stopped: "); + print_name_offset(m, dev->set_state_oneshot_stopped); + SEQ_printf(m, "\n"); + } + if (dev->tick_resume) { SEQ_printf(m, " resume: "); print_name_offset(m, dev->tick_resume); @@ -269,11 +280,11 @@ static void timer_list_show_tickdevices_header(struct seq_file *m) { #ifdef CONFIG_GENERIC_CLOCKEVENTS_BROADCAST print_tickdevice(m, tick_get_broadcast_device(), -1); - SEQ_printf(m, "tick_broadcast_mask: %08lx\n", - cpumask_bits(tick_get_broadcast_mask())[0]); + SEQ_printf(m, "tick_broadcast_mask: %*pb\n", + cpumask_pr_args(tick_get_broadcast_mask())); #ifdef CONFIG_TICK_ONESHOT - SEQ_printf(m, "tick_broadcast_oneshot_mask: %08lx\n", - cpumask_bits(tick_get_broadcast_oneshot_mask())[0]); + SEQ_printf(m, "tick_broadcast_oneshot_mask: %*pb\n", + cpumask_pr_args(tick_get_broadcast_oneshot_mask())); #endif SEQ_printf(m, "\n"); #endif @@ -282,7 +293,7 @@ static void timer_list_show_tickdevices_header(struct seq_file *m) static inline void timer_list_header(struct seq_file *m, u64 now) { - SEQ_printf(m, "Timer List Version: v0.7\n"); + SEQ_printf(m, "Timer List Version: v0.8\n"); SEQ_printf(m, "HRTIMER_MAX_CLOCK_BASES: %d\n", HRTIMER_MAX_CLOCK_BASES); SEQ_printf(m, "now at %Ld nsecs\n", (unsigned long long)now); SEQ_printf(m, "\n"); diff --git a/kernel/time/timer_stats.c b/kernel/time/timer_stats.c index 1fb08f2..1adecb4 100644 --- a/kernel/time/timer_stats.c +++ b/kernel/time/timer_stats.c @@ -68,7 +68,7 @@ struct entry { * Number of timeout events: */ unsigned long count; - unsigned int timer_flag; + u32 flags; /* * We save the command-line string to preserve @@ -227,13 +227,13 @@ static struct entry *tstat_lookup(struct entry *entry, char *comm) * @startf: pointer to the function which did the timer setup * @timerf: pointer to the timer callback function of the timer * @comm: name of the process which set up the timer + * @tflags: The flags field of the timer * * When the timer is already registered, then the event counter is * incremented. Otherwise the timer is registered in a free slot. */ void timer_stats_update_stats(void *timer, pid_t pid, void *startf, - void *timerf, char *comm, - unsigned int timer_flag) + void *timerf, char *comm, u32 tflags) { /* * It doesn't matter which lock we take: @@ -251,7 +251,7 @@ void timer_stats_update_stats(void *timer, pid_t pid, void *startf, input.start_func = startf; input.expire_func = timerf; input.pid = pid; - input.timer_flag = timer_flag; + input.flags = tflags; raw_spin_lock_irqsave(lock, flags); if (!timer_stats_active) @@ -306,7 +306,7 @@ static int tstats_show(struct seq_file *m, void *v) for (i = 0; i < nr_entries; i++) { entry = entries + i; - if (entry->timer_flag & TIMER_STATS_FLAG_DEFERRABLE) { + if (entry->flags & TIMER_DEFERRABLE) { seq_printf(m, "%4luD, %5d %-16s ", entry->count, entry->pid, entry->comm); } else { diff --git a/kernel/torture.c b/kernel/torture.c index dd70993..3e48406 100644 --- a/kernel/torture.c +++ b/kernel/torture.c @@ -409,7 +409,7 @@ static void (*torture_shutdown_hook)(void); */ void torture_shutdown_absorb(const char *title) { - while (ACCESS_ONCE(fullstop) == FULLSTOP_SHUTDOWN) { + while (READ_ONCE(fullstop) == FULLSTOP_SHUTDOWN) { pr_notice("torture thread %s parking due to system shutdown\n", title); schedule_timeout_uninterruptible(MAX_SCHEDULE_TIMEOUT); @@ -480,9 +480,9 @@ static int torture_shutdown_notify(struct notifier_block *unused1, unsigned long unused2, void *unused3) { mutex_lock(&fullstop_mutex); - if (ACCESS_ONCE(fullstop) == FULLSTOP_DONTSTOP) { + if (READ_ONCE(fullstop) == FULLSTOP_DONTSTOP) { VERBOSE_TOROUT_STRING("Unscheduled system shutdown detected"); - ACCESS_ONCE(fullstop) = FULLSTOP_SHUTDOWN; + WRITE_ONCE(fullstop, FULLSTOP_SHUTDOWN); } else { pr_warn("Concurrent rmmod and shutdown illegal!\n"); } @@ -523,13 +523,13 @@ static int stutter; */ void stutter_wait(const char *title) { - while (ACCESS_ONCE(stutter_pause_test) || - (torture_runnable && !ACCESS_ONCE(*torture_runnable))) { + while (READ_ONCE(stutter_pause_test) || + (torture_runnable && !READ_ONCE(*torture_runnable))) { if (stutter_pause_test) - if (ACCESS_ONCE(stutter_pause_test) == 1) + if (READ_ONCE(stutter_pause_test) == 1) schedule_timeout_interruptible(1); else - while (ACCESS_ONCE(stutter_pause_test)) + while (READ_ONCE(stutter_pause_test)) cond_resched(); else schedule_timeout_interruptible(round_jiffies_relative(HZ)); @@ -549,14 +549,14 @@ static int torture_stutter(void *arg) if (!torture_must_stop()) { if (stutter > 1) { schedule_timeout_interruptible(stutter - 1); - ACCESS_ONCE(stutter_pause_test) = 2; + WRITE_ONCE(stutter_pause_test, 2); } schedule_timeout_interruptible(1); - ACCESS_ONCE(stutter_pause_test) = 1; + WRITE_ONCE(stutter_pause_test, 1); } if (!torture_must_stop()) schedule_timeout_interruptible(stutter); - ACCESS_ONCE(stutter_pause_test) = 0; + WRITE_ONCE(stutter_pause_test, 0); torture_shutdown_absorb("torture_stutter"); } while (!torture_must_stop()); torture_kthread_stopping("torture_stutter"); @@ -642,13 +642,13 @@ EXPORT_SYMBOL_GPL(torture_init_end); bool torture_cleanup_begin(void) { mutex_lock(&fullstop_mutex); - if (ACCESS_ONCE(fullstop) == FULLSTOP_SHUTDOWN) { + if (READ_ONCE(fullstop) == FULLSTOP_SHUTDOWN) { pr_warn("Concurrent rmmod and shutdown illegal!\n"); mutex_unlock(&fullstop_mutex); schedule_timeout_uninterruptible(10); return true; } - ACCESS_ONCE(fullstop) = FULLSTOP_RMMOD; + WRITE_ONCE(fullstop, FULLSTOP_RMMOD); mutex_unlock(&fullstop_mutex); torture_shutdown_cleanup(); torture_shuffle_cleanup(); @@ -681,7 +681,7 @@ EXPORT_SYMBOL_GPL(torture_must_stop); */ bool torture_must_stop_irq(void) { - return ACCESS_ONCE(fullstop) != FULLSTOP_DONTSTOP; + return READ_ONCE(fullstop) != FULLSTOP_DONTSTOP; } EXPORT_SYMBOL_GPL(torture_must_stop_irq); diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index ced69da..7f2e97c 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -1369,19 +1369,26 @@ static int check_preds(struct filter_parse_state *ps) { int n_normal_preds = 0, n_logical_preds = 0; struct postfix_elt *elt; + int cnt = 0; list_for_each_entry(elt, &ps->postfix, list) { - if (elt->op == OP_NONE) + if (elt->op == OP_NONE) { + cnt++; continue; + } if (elt->op == OP_AND || elt->op == OP_OR) { n_logical_preds++; + cnt--; continue; } + if (elt->op != OP_NOT) + cnt--; n_normal_preds++; + WARN_ON_ONCE(cnt < 0); } - if (!n_normal_preds || n_logical_preds >= n_normal_preds) { + if (cnt != 1 || !n_normal_preds || n_logical_preds >= n_normal_preds) { parse_error(ps, FILT_ERR_INVALID_FILTER, 0); return -EINVAL; } |
