--- zzzz-none-000/linux-3.10.107/kernel/time/tick-common.c 2017-06-27 09:49:32.000000000 +0000 +++ scorpion-7490-727/linux-3.10.107/kernel/time/tick-common.c 2021-02-04 17:41:59.000000000 +0000 @@ -19,6 +19,7 @@ #include #include #include +#include #include @@ -33,8 +34,22 @@ */ ktime_t tick_next_period; ktime_t tick_period; + +/* + * tick_do_timer_cpu is a timer core internal variable which holds the CPU NR + * which is responsible for calling do_timer(), i.e. the timekeeping stuff. This + * variable has two functions: + * + * 1) Prevent a thundering herd issue of a gazillion of CPUs trying to grab the + * timekeeping lock all at once. Only the CPU which is assigned to do the + * update is handling it. + * + * 2) Hand off the duty in the NOHZ idle case by setting the value to + * TICK_DO_TIMER_NONE, i.e. a non existing CPU. So the next cpu which looks + * at it will take over and keep the time keeping alive. The handover + * procedure also covers cpu hotplug. + */ int tick_do_timer_cpu __read_mostly = TICK_DO_TIMER_BOOT; -static DEFINE_RAW_SPINLOCK(tick_device_lock); /* * Debugging: see timer_list.c @@ -71,6 +86,7 @@ do_timer(1); write_sequnlock(&jiffies_lock); + update_wall_time(); } update_process_times(user_mode(get_irq_regs())); @@ -83,18 +99,29 @@ void tick_handle_periodic(struct clock_event_device *dev) { int cpu = smp_processor_id(); - ktime_t next; + ktime_t next = dev->next_event; tick_periodic(cpu); - if (dev->mode != CLOCK_EVT_MODE_ONESHOT) - return; +#if defined(CONFIG_HIGH_RES_TIMERS) || defined(CONFIG_NO_HZ_COMMON) /* - * Setup the next period for devices, which do not have - * periodic mode: + * The cpu might have transitioned to HIGHRES or NOHZ mode via + * update_process_times() -> run_local_timers() -> + * hrtimer_run_queues(). */ - next = ktime_add(dev->next_event, tick_period); + if (dev->event_handler != tick_handle_periodic) + return; +#endif + + if (!clockevent_state_oneshot(dev)) + return; for (;;) { + /* + * Setup the next period for devices, which do not have + * periodic mode: + */ + next = ktime_add(next, tick_period); + if (!clockevents_program_event(dev, next, false)) return; /* @@ -103,12 +130,11 @@ * to be sure we're using a real hardware clocksource. * Otherwise we could get trapped in an infinite * loop, as the tick_periodic() increments jiffies, - * when then will increment time, posibly causing + * which then will increment time, possibly causing * the loop to trigger again and again. */ if (timekeeping_valid_for_hres()) tick_periodic(cpu); - next = ktime_add(next, tick_period); } } @@ -125,7 +151,7 @@ if ((dev->features & CLOCK_EVT_FEAT_PERIODIC) && !tick_broadcast_oneshot_active()) { - clockevents_set_mode(dev, CLOCK_EVT_MODE_PERIODIC); + clockevents_switch_state(dev, CLOCK_EVT_STATE_PERIODIC); } else { unsigned long seq; ktime_t next; @@ -135,7 +161,7 @@ next = tick_next_period; } while (read_seqretry(&jiffies_lock, seq)); - clockevents_set_mode(dev, CLOCK_EVT_MODE_ONESHOT); + clockevents_switch_state(dev, CLOCK_EVT_STATE_ONESHOT); for (;;) { if (!clockevents_program_event(dev, next, false)) @@ -207,6 +233,17 @@ tick_setup_oneshot(newdev, handler, next_event); } +void tick_install_replacement(struct clock_event_device *newdev) +{ + struct tick_device *td = this_cpu_ptr(&tick_cpu_device); + int cpu = smp_processor_id(); + + clockevents_exchange_device(td->evtdev, newdev); + tick_setup_device(td, newdev, cpu, cpumask_of(cpu)); + if (newdev->features & CLOCK_EVT_FEAT_ONESHOT) + tick_oneshot_notify(); +} + static bool tick_check_percpu(struct clock_event_device *curdev, struct clock_event_device *newdev, int cpu) { @@ -244,21 +281,29 @@ } /* - * Check, if the new registered device should be used. + * Check whether the new device is a better fit than curdev. curdev + * can be NULL ! + */ +bool tick_check_replacement(struct clock_event_device *curdev, + struct clock_event_device *newdev) +{ + if (!tick_check_percpu(curdev, newdev, smp_processor_id())) + return false; + + return tick_check_preferred(curdev, newdev); +} + +/* + * Check, if the new registered device should be used. Called with + * clockevents_lock held and interrupts disabled. */ void tick_check_new_device(struct clock_event_device *newdev) { struct clock_event_device *curdev; struct tick_device *td; int cpu; - unsigned long flags; - - raw_spin_lock_irqsave(&tick_device_lock, flags); cpu = smp_processor_id(); - if (!cpumask_test_cpu(cpu, newdev->cpumask)) - goto out_bc; - td = &per_cpu(tick_cpu_device, cpu); curdev = td->evtdev; @@ -286,8 +331,6 @@ tick_setup_device(td, newdev, cpu, cpumask_of(cpu)); if (newdev->features & CLOCK_EVT_FEAT_ONESHOT) tick_oneshot_notify(); - - raw_spin_unlock_irqrestore(&tick_device_lock, flags); return; out_bc: @@ -295,17 +338,40 @@ * Can the new device be used as a broadcast device ? */ tick_install_broadcast_device(newdev); - raw_spin_unlock_irqrestore(&tick_device_lock, flags); } +/** + * tick_broadcast_oneshot_control - Enter/exit broadcast oneshot mode + * @state: The target state (enter/exit) + * + * The system enters/leaves a state, where affected devices might stop + * Returns 0 on success, -EBUSY if the cpu is used to broadcast wakeups. + * + * Called with interrupts disabled, so clockevents_lock is not + * required here because the local clock event device cannot go away + * under us. + */ +int tick_broadcast_oneshot_control(enum tick_broadcast_state state) +{ + struct tick_device *td = this_cpu_ptr(&tick_cpu_device); + + if (!(td->evtdev->features & CLOCK_EVT_FEAT_C3STOP)) + return 0; + + return __tick_broadcast_oneshot_control(state); +} +EXPORT_SYMBOL_GPL(tick_broadcast_oneshot_control); + +#ifdef CONFIG_HOTPLUG_CPU /* * Transfer the do_timer job away from a dying cpu. * - * Called with interrupts disabled. + * Called with interrupts disabled. Not locking required. If + * tick_do_timer_cpu is owned by this cpu, nothing can change it. */ -static void tick_handover_do_timer(int *cpup) +void tick_handover_do_timer(void) { - if (*cpup == tick_do_timer_cpu) { + if (tick_do_timer_cpu == smp_processor_id()) { int cpu = cpumask_first(cpu_online_mask); tick_do_timer_cpu = (cpu < nr_cpu_ids) ? cpu : @@ -320,93 +386,144 @@ * access the hardware device itself. * We just set the mode and remove it from the lists. */ -static void tick_shutdown(unsigned int *cpup) +void tick_shutdown(unsigned int cpu) { - struct tick_device *td = &per_cpu(tick_cpu_device, *cpup); + struct tick_device *td = &per_cpu(tick_cpu_device, cpu); struct clock_event_device *dev = td->evtdev; - unsigned long flags; - raw_spin_lock_irqsave(&tick_device_lock, flags); td->mode = TICKDEV_MODE_PERIODIC; if (dev) { /* * Prevent that the clock events layer tries to call * the set mode function! */ - dev->mode = CLOCK_EVT_MODE_UNUSED; + clockevent_set_state(dev, CLOCK_EVT_STATE_DETACHED); clockevents_exchange_device(dev, NULL); dev->event_handler = clockevents_handle_noop; td->evtdev = NULL; } - raw_spin_unlock_irqrestore(&tick_device_lock, flags); } +#endif -static void tick_suspend(void) +/** + * tick_suspend_local - Suspend the local tick device + * + * Called from the local cpu for freeze with interrupts disabled. + * + * No locks required. Nothing can change the per cpu device. + */ +void tick_suspend_local(void) { - struct tick_device *td = &__get_cpu_var(tick_cpu_device); - unsigned long flags; + struct tick_device *td = this_cpu_ptr(&tick_cpu_device); - raw_spin_lock_irqsave(&tick_device_lock, flags); clockevents_shutdown(td->evtdev); - raw_spin_unlock_irqrestore(&tick_device_lock, flags); } -static void tick_resume(void) +/** + * tick_resume_local - Resume the local tick device + * + * Called from the local CPU for unfreeze or XEN resume magic. + * + * No locks required. Nothing can change the per cpu device. + */ +void tick_resume_local(void) { - struct tick_device *td = &__get_cpu_var(tick_cpu_device); - unsigned long flags; - int broadcast = tick_resume_broadcast(); - - raw_spin_lock_irqsave(&tick_device_lock, flags); - clockevents_set_mode(td->evtdev, CLOCK_EVT_MODE_RESUME); + struct tick_device *td = this_cpu_ptr(&tick_cpu_device); + bool broadcast = tick_resume_check_broadcast(); + clockevents_tick_resume(td->evtdev); if (!broadcast) { if (td->mode == TICKDEV_MODE_PERIODIC) tick_setup_periodic(td->evtdev, 0); else tick_resume_oneshot(); } - raw_spin_unlock_irqrestore(&tick_device_lock, flags); } -void tick_notify(unsigned long reason, void *dev) +/** + * tick_suspend - Suspend the tick and the broadcast device + * + * Called from syscore_suspend() via timekeeping_suspend with only one + * CPU online and interrupts disabled or from tick_unfreeze() under + * tick_freeze_lock. + * + * No locks required. Nothing can change the per cpu device. + */ +void tick_suspend(void) { - switch (reason) { + tick_suspend_local(); + tick_suspend_broadcast(); +} - case CLOCK_EVT_NOTIFY_BROADCAST_ON: - case CLOCK_EVT_NOTIFY_BROADCAST_OFF: - case CLOCK_EVT_NOTIFY_BROADCAST_FORCE: - tick_broadcast_on_off(reason, dev); - break; +/** + * tick_resume - Resume the tick and the broadcast device + * + * Called from syscore_resume() via timekeeping_resume with only one + * CPU online and interrupts disabled. + * + * No locks required. Nothing can change the per cpu device. + */ +void tick_resume(void) +{ + tick_resume_broadcast(); + tick_resume_local(); +} - case CLOCK_EVT_NOTIFY_BROADCAST_ENTER: - case CLOCK_EVT_NOTIFY_BROADCAST_EXIT: - tick_broadcast_oneshot_control(reason); - break; +#ifdef CONFIG_SUSPEND +static DEFINE_RAW_SPINLOCK(tick_freeze_lock); +static unsigned int tick_freeze_depth; - case CLOCK_EVT_NOTIFY_CPU_DYING: - tick_handover_do_timer(dev); - break; +/** + * tick_freeze - Suspend the local tick and (possibly) timekeeping. + * + * Check if this is the last online CPU executing the function and if so, + * suspend timekeeping. Otherwise suspend the local tick. + * + * Call with interrupts disabled. Must be balanced with %tick_unfreeze(). + * Interrupts must not be enabled before the subsequent %tick_unfreeze(). + */ +void tick_freeze(void) +{ + raw_spin_lock(&tick_freeze_lock); - case CLOCK_EVT_NOTIFY_CPU_DEAD: - tick_shutdown_broadcast_oneshot(dev); - tick_shutdown_broadcast(dev); - tick_shutdown(dev); - break; + tick_freeze_depth++; + if (tick_freeze_depth == num_online_cpus()) { + trace_suspend_resume(TPS("timekeeping_freeze"), + smp_processor_id(), true); + timekeeping_suspend(); + } else { + tick_suspend_local(); + } - case CLOCK_EVT_NOTIFY_SUSPEND: - tick_suspend(); - tick_suspend_broadcast(); - break; + raw_spin_unlock(&tick_freeze_lock); +} - case CLOCK_EVT_NOTIFY_RESUME: - tick_resume(); - break; +/** + * tick_unfreeze - Resume the local tick and (possibly) timekeeping. + * + * Check if this is the first CPU executing the function and if so, resume + * timekeeping. Otherwise resume the local tick. + * + * Call with interrupts disabled. Must be balanced with %tick_freeze(). + * Interrupts must not be enabled after the preceding %tick_freeze(). + */ +void tick_unfreeze(void) +{ + raw_spin_lock(&tick_freeze_lock); - default: - break; + if (tick_freeze_depth == num_online_cpus()) { + timekeeping_resume(); + 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 @@ -414,4 +531,5 @@ void __init tick_init(void) { tick_broadcast_init(); + tick_nohz_init(); }