diff options
Diffstat (limited to 'drivers/base/power/main.c')
| -rw-r--r-- | drivers/base/power/main.c | 1713 |
1 files changed, 1350 insertions, 363 deletions
diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index 5a9b6569dd74..97a8b4fcf471 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -1,12 +1,10 @@ +// SPDX-License-Identifier: GPL-2.0 /* * drivers/base/power/main.c - Where the driver meets power management. * * Copyright (c) 2003 Patrick Mochel * Copyright (c) 2003 Open Source Development Lab * - * This file is released under the GPLv2 - * - * * The driver model core calls device_pm_add() when a device is registered. * This will initialize the embedded device_pm_info object in the device * and add it to the list of power-controlled devices. sysfs entries for @@ -17,18 +15,27 @@ * subsystem list maintains. */ +#define pr_fmt(fmt) "PM: " fmt +#define dev_fmt pr_fmt + #include <linux/device.h> -#include <linux/kallsyms.h> #include <linux/export.h> #include <linux/mutex.h> #include <linux/pm.h> #include <linux/pm_runtime.h> -#include <linux/resume-trace.h> +#include <linux/pm-trace.h> +#include <linux/pm_wakeirq.h> #include <linux/interrupt.h> #include <linux/sched.h> +#include <linux/sched/debug.h> #include <linux/async.h> #include <linux/suspend.h> -#include <linux/cpuidle.h> +#include <trace/events/power.h> +#include <linux/cpufreq.h> +#include <linux/devfreq.h> +#include <linux/timer.h> +#include <linux/nmi.h> + #include "../base.h" #include "power.h" @@ -50,13 +57,53 @@ static LIST_HEAD(dpm_suspended_list); static LIST_HEAD(dpm_late_early_list); static LIST_HEAD(dpm_noirq_list); -struct suspend_stats suspend_stats; static DEFINE_MUTEX(dpm_list_mtx); static pm_message_t pm_transition; +static DEFINE_MUTEX(async_wip_mtx); static int async_error; /** + * pm_hibernate_is_recovering - if recovering from hibernate due to error. + * + * Used to query if dev_pm_ops.thaw() is called for normal hibernation case or + * recovering from some error. + * + * Return: true for error case, false for normal case. + */ +bool pm_hibernate_is_recovering(void) +{ + return pm_transition.event == PM_EVENT_RECOVER; +} +EXPORT_SYMBOL_GPL(pm_hibernate_is_recovering); + +static const char *pm_verb(int event) +{ + switch (event) { + case PM_EVENT_SUSPEND: + return "suspend"; + case PM_EVENT_RESUME: + return "resume"; + case PM_EVENT_FREEZE: + return "freeze"; + case PM_EVENT_QUIESCE: + return "quiesce"; + case PM_EVENT_HIBERNATE: + return "hibernate"; + case PM_EVENT_THAW: + return "thaw"; + case PM_EVENT_RESTORE: + return "restore"; + case PM_EVENT_RECOVER: + return "recover"; + case PM_EVENT_POWEROFF: + return "poweroff"; + default: + return "(unknown PM event)"; + } +} + +/** * device_pm_sleep_init - Initialize system suspend-related device fields. * @dev: Device object being initialized. */ @@ -64,6 +111,8 @@ void device_pm_sleep_init(struct device *dev) { dev->power.is_prepared = false; dev->power.is_suspended = false; + dev->power.is_noirq_suspended = false; + dev->power.is_late_suspended = false; init_completion(&dev->power.completion); complete_all(&dev->power.completion); dev->power.wakeup = NULL; @@ -92,13 +141,19 @@ void device_pm_unlock(void) */ void device_pm_add(struct device *dev) { - pr_debug("PM: Adding info for %s:%s\n", + /* Skip PM setup/initialization. */ + if (device_pm_not_required(dev)) + return; + + pr_debug("Adding info for %s:%s\n", dev->bus ? dev->bus->name : "No Bus", dev_name(dev)); + device_pm_check_callbacks(dev); mutex_lock(&dpm_list_mtx); if (dev->parent && dev->parent->power.is_prepared) dev_warn(dev, "parent %s should not be sleeping\n", dev_name(dev->parent)); list_add_tail(&dev->power.entry, &dpm_list); + dev->power.in_dpm_list = true; mutex_unlock(&dpm_list_mtx); } @@ -108,14 +163,19 @@ void device_pm_add(struct device *dev) */ void device_pm_remove(struct device *dev) { - pr_debug("PM: Removing info for %s:%s\n", + if (device_pm_not_required(dev)) + return; + + pr_debug("Removing info for %s:%s\n", dev->bus ? dev->bus->name : "No Bus", dev_name(dev)); complete_all(&dev->power.completion); mutex_lock(&dpm_list_mtx); list_del_init(&dev->power.entry); + dev->power.in_dpm_list = false; mutex_unlock(&dpm_list_mtx); device_wakeup_disable(dev); pm_runtime_remove(dev); + device_pm_check_callbacks(dev); } /** @@ -125,7 +185,7 @@ void device_pm_remove(struct device *dev) */ void device_pm_move_before(struct device *deva, struct device *devb) { - pr_debug("PM: Moving %s:%s before %s:%s\n", + pr_debug("Moving %s:%s before %s:%s\n", deva->bus ? deva->bus->name : "No Bus", dev_name(deva), devb->bus ? devb->bus->name : "No Bus", dev_name(devb)); /* Delete deva from dpm_list and reinsert before devb. */ @@ -139,7 +199,7 @@ void device_pm_move_before(struct device *deva, struct device *devb) */ void device_pm_move_after(struct device *deva, struct device *devb) { - pr_debug("PM: Moving %s:%s after %s:%s\n", + pr_debug("Moving %s:%s after %s:%s\n", deva->bus ? deva->bus->name : "No Bus", dev_name(deva), devb->bus ? devb->bus->name : "No Bus", dev_name(devb)); /* Delete deva from dpm_list and reinsert after devb. */ @@ -152,36 +212,33 @@ void device_pm_move_after(struct device *deva, struct device *devb) */ void device_pm_move_last(struct device *dev) { - pr_debug("PM: Moving %s:%s to end of list\n", + pr_debug("Moving %s:%s to end of list\n", dev->bus ? dev->bus->name : "No Bus", dev_name(dev)); list_move_tail(&dev->power.entry, &dpm_list); } -static ktime_t initcall_debug_start(struct device *dev) +static ktime_t initcall_debug_start(struct device *dev, void *cb) { - ktime_t calltime = ktime_set(0, 0); - - if (pm_print_times_enabled) { - pr_info("calling %s+ @ %i, parent: %s\n", - dev_name(dev), task_pid_nr(current), - dev->parent ? dev_name(dev->parent) : "none"); - calltime = ktime_get(); - } + if (!pm_print_times_enabled) + return 0; - return calltime; + dev_info(dev, "calling %ps @ %i, parent: %s\n", cb, + task_pid_nr(current), + dev->parent ? dev_name(dev->parent) : "none"); + return ktime_get(); } static void initcall_debug_report(struct device *dev, ktime_t calltime, - int error) + void *cb, int error) { - ktime_t delta, rettime; + ktime_t rettime; - if (pm_print_times_enabled) { - rettime = ktime_get(); - delta = ktime_sub(rettime, calltime); - pr_info("call %s+ returned %d after %Ld usecs\n", dev_name(dev), - error, (unsigned long long)ktime_to_ns(delta) >> 10); - } + if (!pm_print_times_enabled) + return; + + rettime = ktime_get(); + dev_info(dev, "%ps returned %d after %Ld usecs\n", cb, error, + (unsigned long long)ktime_us_delta(rettime, calltime)); } /** @@ -206,7 +263,93 @@ static int dpm_wait_fn(struct device *dev, void *async_ptr) static void dpm_wait_for_children(struct device *dev, bool async) { - device_for_each_child(dev, &async, dpm_wait_fn); + device_for_each_child(dev, &async, dpm_wait_fn); +} + +static void dpm_wait_for_suppliers(struct device *dev, bool async) +{ + struct device_link *link; + int idx; + + idx = device_links_read_lock(); + + /* + * If the supplier goes away right after we've checked the link to it, + * we'll wait for its completion to change the state, but that's fine, + * because the only things that will block as a result are the SRCU + * callbacks freeing the link objects for the links in the list we're + * walking. + */ + dev_for_each_link_to_supplier(link, dev) + if (READ_ONCE(link->status) != DL_STATE_DORMANT && + !device_link_flag_is_sync_state_only(link->flags)) + dpm_wait(link->supplier, async); + + device_links_read_unlock(idx); +} + +static bool dpm_wait_for_superior(struct device *dev, bool async) +{ + struct device *parent; + + /* + * If the device is resumed asynchronously and the parent's callback + * deletes both the device and the parent itself, the parent object may + * be freed while this function is running, so avoid that by reference + * counting the parent once more unless the device has been deleted + * already (in which case return right away). + */ + mutex_lock(&dpm_list_mtx); + + if (!device_pm_initialized(dev)) { + mutex_unlock(&dpm_list_mtx); + return false; + } + + parent = get_device(dev->parent); + + mutex_unlock(&dpm_list_mtx); + + dpm_wait(parent, async); + put_device(parent); + + dpm_wait_for_suppliers(dev, async); + + /* + * If the parent's callback has deleted the device, attempting to resume + * it would be invalid, so avoid doing that then. + */ + return device_pm_initialized(dev); +} + +static void dpm_wait_for_consumers(struct device *dev, bool async) +{ + struct device_link *link; + int idx; + + idx = device_links_read_lock(); + + /* + * The status of a device link can only be changed from "dormant" by a + * probe, but that cannot happen during system suspend/resume. In + * theory it can change to "dormant" at that time, but then it is + * reasonable to wait for the target device anyway (eg. if it goes + * away, it's better to wait for it to go away completely and then + * continue instead of trying to continue in parallel with its + * unregistration). + */ + dev_for_each_link_to_consumer(link, dev) + if (READ_ONCE(link->status) != DL_STATE_DORMANT && + !device_link_flag_is_sync_state_only(link->flags)) + dpm_wait(link->consumer, async); + + device_links_read_unlock(idx); +} + +static void dpm_wait_for_subordinate(struct device *dev, bool async) +{ + dpm_wait_for_children(dev, async); + dpm_wait_for_consumers(dev, async); } /** @@ -227,12 +370,12 @@ static pm_callback_t pm_op(const struct dev_pm_ops *ops, pm_message_t state) case PM_EVENT_FREEZE: case PM_EVENT_QUIESCE: return ops->freeze; + case PM_EVENT_POWEROFF: case PM_EVENT_HIBERNATE: return ops->poweroff; case PM_EVENT_THAW: case PM_EVENT_RECOVER: return ops->thaw; - break; case PM_EVENT_RESTORE: return ops->restore; #endif /* CONFIG_HIBERNATE_CALLBACKS */ @@ -262,6 +405,7 @@ static pm_callback_t pm_late_early_op(const struct dev_pm_ops *ops, case PM_EVENT_FREEZE: case PM_EVENT_QUIESCE: return ops->freeze_late; + case PM_EVENT_POWEROFF: case PM_EVENT_HIBERNATE: return ops->poweroff_late; case PM_EVENT_THAW: @@ -296,6 +440,7 @@ static pm_callback_t pm_noirq_op(const struct dev_pm_ops *ops, pm_message_t stat case PM_EVENT_FREEZE: case PM_EVENT_QUIESCE: return ops->freeze_noirq; + case PM_EVENT_POWEROFF: case PM_EVENT_HIBERNATE: return ops->poweroff_noirq; case PM_EVENT_THAW: @@ -309,45 +454,22 @@ static pm_callback_t pm_noirq_op(const struct dev_pm_ops *ops, pm_message_t stat return NULL; } -static char *pm_verb(int event) +static void pm_dev_dbg(struct device *dev, pm_message_t state, const char *info) { - switch (event) { - case PM_EVENT_SUSPEND: - return "suspend"; - case PM_EVENT_RESUME: - return "resume"; - case PM_EVENT_FREEZE: - return "freeze"; - case PM_EVENT_QUIESCE: - return "quiesce"; - case PM_EVENT_HIBERNATE: - return "hibernate"; - case PM_EVENT_THAW: - return "thaw"; - case PM_EVENT_RESTORE: - return "restore"; - case PM_EVENT_RECOVER: - return "recover"; - default: - return "(unknown PM event)"; - } -} - -static void pm_dev_dbg(struct device *dev, pm_message_t state, char *info) -{ - dev_dbg(dev, "%s%s%s\n", info, pm_verb(state.event), + dev_dbg(dev, "%s%s%s driver flags: %x\n", info, pm_verb(state.event), ((state.event & PM_EVENT_SLEEP) && device_may_wakeup(dev)) ? - ", may wakeup" : ""); + ", may wakeup" : "", dev->power.driver_flags); } -static void pm_dev_err(struct device *dev, pm_message_t state, char *info, +static void pm_dev_err(struct device *dev, pm_message_t state, const char *info, int error) { - printk(KERN_ERR "PM: Device %s failed to %s%s: error %d\n", - dev_name(dev), pm_verb(state.event), info, error); + dev_err(dev, "failed to %s%s: error %d\n", pm_verb(state.event), info, + error); } -static void dpm_show_time(ktime_t starttime, pm_message_t state, char *info) +static void dpm_show_time(ktime_t starttime, pm_message_t state, int error, + const char *info) { ktime_t calltime; u64 usecs64; @@ -359,13 +481,15 @@ static void dpm_show_time(ktime_t starttime, pm_message_t state, char *info) usecs = usecs64; if (usecs == 0) usecs = 1; - pr_info("PM: %s%s%s of devices complete after %ld.%03ld msecs\n", - info ?: "", info ? " " : "", pm_verb(state.event), - usecs / USEC_PER_MSEC, usecs % USEC_PER_MSEC); + + pm_pr_dbg("%s%s%s of devices %s after %ld.%03ld msecs\n", + info ?: "", info ? " " : "", pm_verb(state.event), + error ? "aborted" : "complete", + usecs / USEC_PER_MSEC, usecs % USEC_PER_MSEC); } static int dpm_run_callback(pm_callback_t cb, struct device *dev, - pm_message_t state, char *info) + pm_message_t state, const char *info) { ktime_t calltime; int error; @@ -373,38 +497,279 @@ static int dpm_run_callback(pm_callback_t cb, struct device *dev, if (!cb) return 0; - calltime = initcall_debug_start(dev); + calltime = initcall_debug_start(dev, cb); pm_dev_dbg(dev, state, info); + trace_device_pm_callback_start(dev, info, state.event); error = cb(dev); - suspend_report_result(cb, error); + trace_device_pm_callback_end(dev, error); + suspend_report_result(dev, cb, error); - initcall_debug_report(dev, calltime, error); + initcall_debug_report(dev, calltime, cb, error); return error; } +#ifdef CONFIG_DPM_WATCHDOG +struct dpm_watchdog { + struct device *dev; + struct task_struct *tsk; + struct timer_list timer; + bool fatal; +}; + +#define DECLARE_DPM_WATCHDOG_ON_STACK(wd) \ + struct dpm_watchdog wd + +static bool __read_mostly dpm_watchdog_all_cpu_backtrace; +module_param(dpm_watchdog_all_cpu_backtrace, bool, 0644); +MODULE_PARM_DESC(dpm_watchdog_all_cpu_backtrace, + "Backtrace all CPUs on DPM watchdog timeout"); + +/** + * dpm_watchdog_handler - Driver suspend / resume watchdog handler. + * @t: The timer that PM watchdog depends on. + * + * Called when a driver has timed out suspending or resuming. + * There's not much we can do here to recover so panic() to + * capture a crash-dump in pstore. + */ +static void dpm_watchdog_handler(struct timer_list *t) +{ + struct dpm_watchdog *wd = timer_container_of(wd, t, timer); + struct timer_list *timer = &wd->timer; + unsigned int time_left; + + if (wd->fatal) { + unsigned int this_cpu = smp_processor_id(); + + dev_emerg(wd->dev, "**** DPM device timeout ****\n"); + show_stack(wd->tsk, NULL, KERN_EMERG); + if (dpm_watchdog_all_cpu_backtrace) + trigger_allbutcpu_cpu_backtrace(this_cpu); + panic("%s %s: unrecoverable failure\n", + dev_driver_string(wd->dev), dev_name(wd->dev)); + } + + time_left = CONFIG_DPM_WATCHDOG_TIMEOUT - CONFIG_DPM_WATCHDOG_WARNING_TIMEOUT; + dev_warn(wd->dev, "**** DPM device timeout after %u seconds; %u seconds until panic ****\n", + CONFIG_DPM_WATCHDOG_WARNING_TIMEOUT, time_left); + show_stack(wd->tsk, NULL, KERN_WARNING); + + wd->fatal = true; + mod_timer(timer, jiffies + HZ * time_left); +} + +/** + * dpm_watchdog_set - Enable pm watchdog for given device. + * @wd: Watchdog. Must be allocated on the stack. + * @dev: Device to handle. + */ +static void dpm_watchdog_set(struct dpm_watchdog *wd, struct device *dev) +{ + struct timer_list *timer = &wd->timer; + + wd->dev = dev; + wd->tsk = current; + wd->fatal = CONFIG_DPM_WATCHDOG_TIMEOUT == CONFIG_DPM_WATCHDOG_WARNING_TIMEOUT; + + timer_setup_on_stack(timer, dpm_watchdog_handler, 0); + /* use same timeout value for both suspend and resume */ + timer->expires = jiffies + HZ * CONFIG_DPM_WATCHDOG_WARNING_TIMEOUT; + add_timer(timer); +} + +/** + * dpm_watchdog_clear - Disable suspend/resume watchdog. + * @wd: Watchdog to disable. + */ +static void dpm_watchdog_clear(struct dpm_watchdog *wd) +{ + struct timer_list *timer = &wd->timer; + + timer_delete_sync(timer); + timer_destroy_on_stack(timer); +} +#else +#define DECLARE_DPM_WATCHDOG_ON_STACK(wd) +#define dpm_watchdog_set(x, y) +#define dpm_watchdog_clear(x) +#endif + /*------------------------- Resume routines -------------------------*/ /** - * device_resume_noirq - Execute an "early resume" callback for given device. + * dev_pm_skip_resume - System-wide device resume optimization check. + * @dev: Target device. + * + * Return: + * - %false if the transition under way is RESTORE. + * - Return value of dev_pm_skip_suspend() if the transition under way is THAW. + * - The logical negation of %power.must_resume otherwise (that is, when the + * transition under way is RESUME). + */ +bool dev_pm_skip_resume(struct device *dev) +{ + if (pm_transition.event == PM_EVENT_RESTORE) + return false; + + if (pm_transition.event == PM_EVENT_THAW) + return dev_pm_skip_suspend(dev); + + return !dev->power.must_resume; +} + +static bool is_async(struct device *dev) +{ + return dev->power.async_suspend && pm_async_enabled + && !pm_trace_is_enabled(); +} + +static bool __dpm_async(struct device *dev, async_func_t func) +{ + if (dev->power.work_in_progress) + return true; + + if (!is_async(dev)) + return false; + + dev->power.work_in_progress = true; + + get_device(dev); + + if (async_schedule_dev_nocall(func, dev)) + return true; + + put_device(dev); + + return false; +} + +static bool dpm_async_fn(struct device *dev, async_func_t func) +{ + guard(mutex)(&async_wip_mtx); + + return __dpm_async(dev, func); +} + +static int dpm_async_with_cleanup(struct device *dev, void *fn) +{ + guard(mutex)(&async_wip_mtx); + + if (!__dpm_async(dev, fn)) + dev->power.work_in_progress = false; + + return 0; +} + +static void dpm_async_resume_children(struct device *dev, async_func_t func) +{ + /* + * Prevent racing with dpm_clear_async_state() during initial list + * walks in dpm_noirq_resume_devices(), dpm_resume_early(), and + * dpm_resume(). + */ + guard(mutex)(&dpm_list_mtx); + + /* + * Start processing "async" children of the device unless it's been + * started already for them. + */ + device_for_each_child(dev, func, dpm_async_with_cleanup); +} + +static void dpm_async_resume_subordinate(struct device *dev, async_func_t func) +{ + struct device_link *link; + int idx; + + dpm_async_resume_children(dev, func); + + idx = device_links_read_lock(); + + /* Start processing the device's "async" consumers. */ + dev_for_each_link_to_consumer(link, dev) + if (READ_ONCE(link->status) != DL_STATE_DORMANT) + dpm_async_with_cleanup(link->consumer, func); + + device_links_read_unlock(idx); +} + +static void dpm_clear_async_state(struct device *dev) +{ + reinit_completion(&dev->power.completion); + dev->power.work_in_progress = false; +} + +static bool dpm_root_device(struct device *dev) +{ + lockdep_assert_held(&dpm_list_mtx); + + /* + * Since this function is required to run under dpm_list_mtx, the + * list_empty() below will only return true if the device's list of + * consumers is actually empty before calling it. + */ + return !dev->parent && list_empty(&dev->links.suppliers); +} + +static void async_resume_noirq(void *data, async_cookie_t cookie); + +/** + * device_resume_noirq - Execute a "noirq resume" callback for given device. * @dev: Device to handle. * @state: PM transition of the system being carried out. + * @async: If true, the device is being resumed asynchronously. * * The driver of @dev will not receive interrupts while this function is being * executed. */ -static int device_resume_noirq(struct device *dev, pm_message_t state) +static void device_resume_noirq(struct device *dev, pm_message_t state, bool async) { pm_callback_t callback = NULL; - char *info = NULL; + const char *info = NULL; + bool skip_resume; int error = 0; TRACE_DEVICE(dev); TRACE_RESUME(0); - if (dev->power.syscore) + if (dev->power.syscore || dev->power.direct_complete) + goto Out; + + if (!dev->power.is_noirq_suspended) { + /* + * This means that system suspend has been aborted in the noirq + * phase before invoking the noirq suspend callback for the + * device, so if device_suspend_late() has left it in suspend, + * device_resume_early() should leave it in suspend either in + * case the early resume of it depends on the noirq resume that + * has not run. + */ + if (dev_pm_skip_suspend(dev)) + dev->power.must_resume = false; + goto Out; + } + + if (!dpm_wait_for_superior(dev, async)) + goto Out; + + skip_resume = dev_pm_skip_resume(dev); + /* + * If the driver callback is skipped below or by the middle layer + * callback and device_resume_early() also skips the driver callback for + * this device later, it needs to appear as "suspended" to PM-runtime, + * so change its status accordingly. + * + * Otherwise, the device is going to be resumed, so set its PM-runtime + * status to "active" unless its power.smart_suspend flag is clear, in + * which case it is not necessary to update its PM-runtime status. + */ + if (skip_resume) + pm_runtime_set_suspended(dev); + else if (dev_pm_smart_suspend(dev)) + pm_runtime_set_active(dev); if (dev->pm_domain) { info = "noirq power domain "; @@ -419,73 +784,135 @@ static int device_resume_noirq(struct device *dev, pm_message_t state) info = "noirq bus "; callback = pm_noirq_op(dev->bus->pm, state); } + if (callback) + goto Run; - if (!callback && dev->driver && dev->driver->pm) { + if (skip_resume) + goto Skip; + + if (dev->driver && dev->driver->pm) { info = "noirq driver "; callback = pm_noirq_op(dev->driver->pm, state); } +Run: error = dpm_run_callback(callback, dev, state, info); - Out: +Skip: + dev->power.is_noirq_suspended = false; + +Out: + complete_all(&dev->power.completion); TRACE_RESUME(error); - return error; + + if (error) { + WRITE_ONCE(async_error, error); + dpm_save_failed_dev(dev_name(dev)); + pm_dev_err(dev, state, async ? " async noirq" : " noirq", error); + } + + dpm_async_resume_subordinate(dev, async_resume_noirq); } -/** - * dpm_resume_noirq - Execute "noirq resume" callbacks for all devices. - * @state: PM transition of the system being carried out. - * - * Call the "noirq" resume handlers for all devices in dpm_noirq_list and - * enable device drivers to receive interrupts. - */ -static void dpm_resume_noirq(pm_message_t state) +static void async_resume_noirq(void *data, async_cookie_t cookie) { + struct device *dev = data; + + device_resume_noirq(dev, pm_transition, true); + put_device(dev); +} + +static void dpm_noirq_resume_devices(pm_message_t state) +{ + struct device *dev; ktime_t starttime = ktime_get(); + trace_suspend_resume(TPS("dpm_resume_noirq"), state.event, true); + + async_error = 0; + pm_transition = state; + mutex_lock(&dpm_list_mtx); - while (!list_empty(&dpm_noirq_list)) { - struct device *dev = to_device(dpm_noirq_list.next); - int error; - get_device(dev); + /* + * Start processing "async" root devices upfront so they don't wait for + * the "sync" devices they don't depend on. + */ + list_for_each_entry(dev, &dpm_noirq_list, power.entry) { + dpm_clear_async_state(dev); + if (dpm_root_device(dev)) + dpm_async_with_cleanup(dev, async_resume_noirq); + } + + while (!list_empty(&dpm_noirq_list)) { + dev = to_device(dpm_noirq_list.next); list_move_tail(&dev->power.entry, &dpm_late_early_list); - mutex_unlock(&dpm_list_mtx); - error = device_resume_noirq(dev, state); - if (error) { - suspend_stats.failed_resume_noirq++; - dpm_save_failed_step(SUSPEND_RESUME_NOIRQ); - dpm_save_failed_dev(dev_name(dev)); - pm_dev_err(dev, state, " noirq", error); - } + if (!dpm_async_fn(dev, async_resume_noirq)) { + get_device(dev); - mutex_lock(&dpm_list_mtx); - put_device(dev); + mutex_unlock(&dpm_list_mtx); + + device_resume_noirq(dev, state, false); + + put_device(dev); + + mutex_lock(&dpm_list_mtx); + } } mutex_unlock(&dpm_list_mtx); - dpm_show_time(starttime, state, "noirq"); + async_synchronize_full(); + dpm_show_time(starttime, state, 0, "noirq"); + if (READ_ONCE(async_error)) + dpm_save_failed_step(SUSPEND_RESUME_NOIRQ); + + trace_suspend_resume(TPS("dpm_resume_noirq"), state.event, false); +} + +/** + * dpm_resume_noirq - Execute "noirq resume" callbacks for all devices. + * @state: PM transition of the system being carried out. + * + * Invoke the "noirq" resume callbacks for all devices in dpm_noirq_list and + * allow device drivers' interrupt handlers to be called. + */ +void dpm_resume_noirq(pm_message_t state) +{ + dpm_noirq_resume_devices(state); + resume_device_irqs(); - cpuidle_resume(); + device_wakeup_disarm_wake_irqs(); } +static void async_resume_early(void *data, async_cookie_t cookie); + /** * device_resume_early - Execute an "early resume" callback for given device. * @dev: Device to handle. * @state: PM transition of the system being carried out. + * @async: If true, the device is being resumed asynchronously. * * Runtime PM is disabled for @dev while this function is being executed. */ -static int device_resume_early(struct device *dev, pm_message_t state) +static void device_resume_early(struct device *dev, pm_message_t state, bool async) { pm_callback_t callback = NULL; - char *info = NULL; + const char *info = NULL; int error = 0; TRACE_DEVICE(dev); TRACE_RESUME(0); + if (dev->power.direct_complete) + goto Out; + + if (!dev->power.is_late_suspended) + goto Out; + if (dev->power.syscore) + goto Skip; + + if (!dpm_wait_for_superior(dev, async)) goto Out; if (dev->pm_domain) { @@ -501,51 +928,95 @@ static int device_resume_early(struct device *dev, pm_message_t state) info = "early bus "; callback = pm_late_early_op(dev->bus->pm, state); } + if (callback) + goto Run; - if (!callback && dev->driver && dev->driver->pm) { + if (dev_pm_skip_resume(dev)) + goto Skip; + + if (dev->driver && dev->driver->pm) { info = "early driver "; callback = pm_late_early_op(dev->driver->pm, state); } +Run: error = dpm_run_callback(callback, dev, state, info); - Out: +Skip: + dev->power.is_late_suspended = false; + pm_runtime_enable(dev); + +Out: TRACE_RESUME(error); - pm_runtime_enable(dev); - return error; + complete_all(&dev->power.completion); + + if (error) { + WRITE_ONCE(async_error, error); + dpm_save_failed_dev(dev_name(dev)); + pm_dev_err(dev, state, async ? " async early" : " early", error); + } + + dpm_async_resume_subordinate(dev, async_resume_early); +} + +static void async_resume_early(void *data, async_cookie_t cookie) +{ + struct device *dev = data; + + device_resume_early(dev, pm_transition, true); + put_device(dev); } /** * dpm_resume_early - Execute "early resume" callbacks for all devices. * @state: PM transition of the system being carried out. */ -static void dpm_resume_early(pm_message_t state) +void dpm_resume_early(pm_message_t state) { + struct device *dev; ktime_t starttime = ktime_get(); + trace_suspend_resume(TPS("dpm_resume_early"), state.event, true); + + async_error = 0; + pm_transition = state; + mutex_lock(&dpm_list_mtx); - while (!list_empty(&dpm_late_early_list)) { - struct device *dev = to_device(dpm_late_early_list.next); - int error; - get_device(dev); + /* + * Start processing "async" root devices upfront so they don't wait for + * the "sync" devices they don't depend on. + */ + list_for_each_entry(dev, &dpm_late_early_list, power.entry) { + dpm_clear_async_state(dev); + if (dpm_root_device(dev)) + dpm_async_with_cleanup(dev, async_resume_early); + } + + while (!list_empty(&dpm_late_early_list)) { + dev = to_device(dpm_late_early_list.next); list_move_tail(&dev->power.entry, &dpm_suspended_list); - mutex_unlock(&dpm_list_mtx); - error = device_resume_early(dev, state); - if (error) { - suspend_stats.failed_resume_early++; - dpm_save_failed_step(SUSPEND_RESUME_EARLY); - dpm_save_failed_dev(dev_name(dev)); - pm_dev_err(dev, state, " early", error); - } + if (!dpm_async_fn(dev, async_resume_early)) { + get_device(dev); - mutex_lock(&dpm_list_mtx); - put_device(dev); + mutex_unlock(&dpm_list_mtx); + + device_resume_early(dev, state, false); + + put_device(dev); + + mutex_lock(&dpm_list_mtx); + } } mutex_unlock(&dpm_list_mtx); - dpm_show_time(starttime, state, "early"); + async_synchronize_full(); + dpm_show_time(starttime, state, 0, "early"); + if (READ_ONCE(async_error)) + dpm_save_failed_step(SUSPEND_RESUME_EARLY); + + trace_suspend_resume(TPS("dpm_resume_early"), state.event, false); } /** @@ -559,17 +1030,20 @@ void dpm_resume_start(pm_message_t state) } EXPORT_SYMBOL_GPL(dpm_resume_start); +static void async_resume(void *data, async_cookie_t cookie); + /** * device_resume - Execute "resume" callbacks for given device. * @dev: Device to handle. * @state: PM transition of the system being carried out. * @async: If true, the device is being resumed asynchronously. */ -static int device_resume(struct device *dev, pm_message_t state, bool async) +static void device_resume(struct device *dev, pm_message_t state, bool async) { pm_callback_t callback = NULL; - char *info = NULL; + const char *info = NULL; int error = 0; + DECLARE_DPM_WATCHDOG_ON_STACK(wd); TRACE_DEVICE(dev); TRACE_RESUME(0); @@ -577,7 +1051,28 @@ static int device_resume(struct device *dev, pm_message_t state, bool async) if (dev->power.syscore) goto Complete; - dpm_wait(dev->parent, async); + if (!dev->power.is_suspended) + goto Complete; + + dev->power.is_suspended = false; + + if (dev->power.direct_complete) { + /* + * Allow new children to be added under the device after this + * point if it has no PM callbacks. + */ + if (dev->power.no_pm_callbacks) + dev->power.is_prepared = false; + + /* Match the pm_runtime_disable() in device_suspend(). */ + pm_runtime_enable(dev); + goto Complete; + } + + if (!dpm_wait_for_superior(dev, async)) + goto Complete; + + dpm_watchdog_set(&wd, dev); device_lock(dev); /* @@ -586,9 +1081,6 @@ static int device_resume(struct device *dev, pm_message_t state, bool async) */ dev->power.is_prepared = false; - if (!dev->power.is_suspended) - goto Unlock; - if (dev->pm_domain) { info = "power domain "; callback = pm_op(&dev->pm_domain->ops, state); @@ -601,16 +1093,10 @@ static int device_resume(struct device *dev, pm_message_t state, bool async) goto Driver; } - if (dev->class) { - if (dev->class->pm) { - info = "class "; - callback = pm_op(dev->class->pm, state); - goto Driver; - } else if (dev->class->resume) { - info = "legacy class "; - callback = dev->class->resume; - goto End; - } + if (dev->class && dev->class->pm) { + info = "class "; + callback = pm_op(dev->class->pm, state); + goto Driver; } if (dev->bus) { @@ -632,36 +1118,32 @@ static int device_resume(struct device *dev, pm_message_t state, bool async) End: error = dpm_run_callback(callback, dev, state, info); - dev->power.is_suspended = false; - Unlock: device_unlock(dev); + dpm_watchdog_clear(&wd); Complete: complete_all(&dev->power.completion); TRACE_RESUME(error); - return error; + if (error) { + WRITE_ONCE(async_error, error); + dpm_save_failed_dev(dev_name(dev)); + pm_dev_err(dev, state, async ? " async" : "", error); + } + + dpm_async_resume_subordinate(dev, async_resume); } static void async_resume(void *data, async_cookie_t cookie) { - struct device *dev = (struct device *)data; - int error; + struct device *dev = data; - error = device_resume(dev, pm_transition, true); - if (error) - pm_dev_err(dev, pm_transition, " async", error); + device_resume(dev, pm_transition, true); put_device(dev); } -static bool is_async(struct device *dev) -{ - return dev->power.async_suspend && pm_async_enabled - && !pm_trace_is_enabled(); -} - /** * dpm_resume - Execute "resume" callbacks for non-sysdev devices. * @state: PM transition of the system being carried out. @@ -674,45 +1156,48 @@ void dpm_resume(pm_message_t state) struct device *dev; ktime_t starttime = ktime_get(); - might_sleep(); + trace_suspend_resume(TPS("dpm_resume"), state.event, true); - mutex_lock(&dpm_list_mtx); pm_transition = state; async_error = 0; + mutex_lock(&dpm_list_mtx); + + /* + * Start processing "async" root devices upfront so they don't wait for + * the "sync" devices they don't depend on. + */ list_for_each_entry(dev, &dpm_suspended_list, power.entry) { - INIT_COMPLETION(dev->power.completion); - if (is_async(dev)) { - get_device(dev); - async_schedule(async_resume, dev); - } + dpm_clear_async_state(dev); + if (dpm_root_device(dev)) + dpm_async_with_cleanup(dev, async_resume); } while (!list_empty(&dpm_suspended_list)) { dev = to_device(dpm_suspended_list.next); - get_device(dev); - if (!is_async(dev)) { - int error; + list_move_tail(&dev->power.entry, &dpm_prepared_list); + + if (!dpm_async_fn(dev, async_resume)) { + get_device(dev); mutex_unlock(&dpm_list_mtx); - error = device_resume(dev, state, false); - if (error) { - suspend_stats.failed_resume++; - dpm_save_failed_step(SUSPEND_RESUME); - dpm_save_failed_dev(dev_name(dev)); - pm_dev_err(dev, state, "", error); - } + device_resume(dev, state, false); + + put_device(dev); mutex_lock(&dpm_list_mtx); } - if (!list_empty(&dev->power.entry)) - list_move_tail(&dev->power.entry, &dpm_prepared_list); - put_device(dev); } mutex_unlock(&dpm_list_mtx); async_synchronize_full(); - dpm_show_time(starttime, state, NULL); + dpm_show_time(starttime, state, 0, NULL); + if (READ_ONCE(async_error)) + dpm_save_failed_step(SUSPEND_RESUME); + + cpufreq_resume(); + devfreq_resume(); + trace_suspend_resume(TPS("dpm_resume"), state.event, false); } /** @@ -723,10 +1208,10 @@ void dpm_resume(pm_message_t state) static void device_complete(struct device *dev, pm_message_t state) { void (*callback)(struct device *) = NULL; - char *info = NULL; + const char *info = NULL; if (dev->power.syscore) - return; + goto out; device_lock(dev); @@ -756,6 +1241,9 @@ static void device_complete(struct device *dev, pm_message_t state) device_unlock(dev); +out: + /* If enabling runtime PM for the device is blocked, unblock it. */ + pm_runtime_unblock(dev); pm_runtime_put(dev); } @@ -770,7 +1258,7 @@ void dpm_complete(pm_message_t state) { struct list_head list; - might_sleep(); + trace_suspend_resume(TPS("dpm_complete"), state.event, true); INIT_LIST_HEAD(&list); mutex_lock(&dpm_list_mtx); @@ -780,15 +1268,23 @@ void dpm_complete(pm_message_t state) get_device(dev); dev->power.is_prepared = false; list_move(&dev->power.entry, &list); + mutex_unlock(&dpm_list_mtx); + trace_device_pm_callback_start(dev, "", state.event); device_complete(dev, state); + trace_device_pm_callback_end(dev, 0); - mutex_lock(&dpm_list_mtx); put_device(dev); + + mutex_lock(&dpm_list_mtx); } list_splice(&list, &dpm_list); mutex_unlock(&dpm_list_mtx); + + /* Allow device probing and trigger re-probing of deferred devices */ + device_unblock_probing(); + trace_suspend_resume(TPS("dpm_complete"), state.event, false); } /** @@ -801,6 +1297,7 @@ void dpm_complete(pm_message_t state) void dpm_resume_end(pm_message_t state) { dpm_resume(state); + pm_restore_gfp_mask(); dpm_complete(state); } EXPORT_SYMBOL_GPL(dpm_resume_end); @@ -808,6 +1305,82 @@ EXPORT_SYMBOL_GPL(dpm_resume_end); /*------------------------- Suspend routines -------------------------*/ +static bool dpm_leaf_device(struct device *dev) +{ + struct device *child; + + lockdep_assert_held(&dpm_list_mtx); + + child = device_find_any_child(dev); + if (child) { + put_device(child); + + return false; + } + + /* + * Since this function is required to run under dpm_list_mtx, the + * list_empty() below will only return true if the device's list of + * consumers is actually empty before calling it. + */ + return list_empty(&dev->links.consumers); +} + +static bool dpm_async_suspend_parent(struct device *dev, async_func_t func) +{ + guard(mutex)(&dpm_list_mtx); + + /* + * If the device is suspended asynchronously and the parent's callback + * deletes both the device and the parent itself, the parent object may + * be freed while this function is running, so avoid that by checking + * if the device has been deleted already as the parent cannot be + * deleted before it. + */ + if (!device_pm_initialized(dev)) + return false; + + /* Start processing the device's parent if it is "async". */ + if (dev->parent) + dpm_async_with_cleanup(dev->parent, func); + + return true; +} + +static void dpm_async_suspend_superior(struct device *dev, async_func_t func) +{ + struct device_link *link; + int idx; + + if (!dpm_async_suspend_parent(dev, func)) + return; + + idx = device_links_read_lock(); + + /* Start processing the device's "async" suppliers. */ + dev_for_each_link_to_supplier(link, dev) + if (READ_ONCE(link->status) != DL_STATE_DORMANT) + dpm_async_with_cleanup(link->supplier, func); + + device_links_read_unlock(idx); +} + +static void dpm_async_suspend_complete_all(struct list_head *device_list) +{ + struct device *dev; + + guard(mutex)(&async_wip_mtx); + + list_for_each_entry_reverse(dev, device_list, power.entry) { + /* + * In case the device is being waited for and async processing + * has not started for it yet, let the waiters make progress. + */ + if (!dev->power.work_in_progress) + complete_all(&dev->power.completion); + } +} + /** * resume_event - Return a "resume" message for given "suspend" sleep state. * @sleep_state: PM message representing a sleep state. @@ -829,21 +1402,49 @@ static pm_message_t resume_event(pm_message_t sleep_state) return PMSG_ON; } +static void dpm_superior_set_must_resume(struct device *dev) +{ + struct device_link *link; + int idx; + + if (dev->parent) + dev->parent->power.must_resume = true; + + idx = device_links_read_lock(); + + dev_for_each_link_to_supplier(link, dev) + link->supplier->power.must_resume = true; + + device_links_read_unlock(idx); +} + +static void async_suspend_noirq(void *data, async_cookie_t cookie); + /** - * device_suspend_noirq - Execute a "late suspend" callback for given device. + * device_suspend_noirq - Execute a "noirq suspend" callback for given device. * @dev: Device to handle. * @state: PM transition of the system being carried out. + * @async: If true, the device is being suspended asynchronously. * * The driver of @dev will not receive interrupts while this function is being * executed. */ -static int device_suspend_noirq(struct device *dev, pm_message_t state) +static void device_suspend_noirq(struct device *dev, pm_message_t state, bool async) { pm_callback_t callback = NULL; - char *info = NULL; + const char *info = NULL; + int error = 0; - if (dev->power.syscore) - return 0; + TRACE_DEVICE(dev); + TRACE_SUSPEND(0); + + dpm_wait_for_subordinate(dev, async); + + if (READ_ONCE(async_error)) + goto Complete; + + if (dev->power.syscore || dev->power.direct_complete) + goto Complete; if (dev->pm_domain) { info = "noirq power domain "; @@ -858,80 +1459,201 @@ static int device_suspend_noirq(struct device *dev, pm_message_t state) info = "noirq bus "; callback = pm_noirq_op(dev->bus->pm, state); } + if (callback) + goto Run; - if (!callback && dev->driver && dev->driver->pm) { + if (dev_pm_skip_suspend(dev)) + goto Skip; + + if (dev->driver && dev->driver->pm) { info = "noirq driver "; callback = pm_noirq_op(dev->driver->pm, state); } - return dpm_run_callback(callback, dev, state, info); +Run: + error = dpm_run_callback(callback, dev, state, info); + if (error) { + WRITE_ONCE(async_error, error); + dpm_save_failed_dev(dev_name(dev)); + pm_dev_err(dev, state, async ? " async noirq" : " noirq", error); + goto Complete; + } + +Skip: + dev->power.is_noirq_suspended = true; + + /* + * Devices must be resumed unless they are explicitly allowed to be left + * in suspend, but even in that case skipping the resume of devices that + * were in use right before the system suspend (as indicated by their + * runtime PM usage counters and child counters) would be suboptimal. + */ + if (!(dev_pm_test_driver_flags(dev, DPM_FLAG_MAY_SKIP_RESUME) && + dev->power.may_skip_resume) || !pm_runtime_need_not_resume(dev)) + dev->power.must_resume = true; + + if (dev->power.must_resume) + dpm_superior_set_must_resume(dev); + +Complete: + complete_all(&dev->power.completion); + TRACE_SUSPEND(error); + + if (error || READ_ONCE(async_error)) + return; + + dpm_async_suspend_superior(dev, async_suspend_noirq); } -/** - * dpm_suspend_noirq - Execute "noirq suspend" callbacks for all devices. - * @state: PM transition of the system being carried out. - * - * Prevent device drivers from receiving interrupts and call the "noirq" suspend - * handlers for all non-sysdev devices. - */ -static int dpm_suspend_noirq(pm_message_t state) +static void async_suspend_noirq(void *data, async_cookie_t cookie) +{ + struct device *dev = data; + + device_suspend_noirq(dev, pm_transition, true); + put_device(dev); +} + +static int dpm_noirq_suspend_devices(pm_message_t state) { ktime_t starttime = ktime_get(); - int error = 0; + struct device *dev; + int error; + + trace_suspend_resume(TPS("dpm_suspend_noirq"), state.event, true); + + pm_transition = state; + async_error = 0; - cpuidle_pause(); - suspend_device_irqs(); mutex_lock(&dpm_list_mtx); + + /* + * Start processing "async" leaf devices upfront so they don't need to + * wait for the "sync" devices they don't depend on. + */ + list_for_each_entry_reverse(dev, &dpm_late_early_list, power.entry) { + dpm_clear_async_state(dev); + if (dpm_leaf_device(dev)) + dpm_async_with_cleanup(dev, async_suspend_noirq); + } + while (!list_empty(&dpm_late_early_list)) { - struct device *dev = to_device(dpm_late_early_list.prev); + dev = to_device(dpm_late_early_list.prev); + + list_move(&dev->power.entry, &dpm_noirq_list); + + if (dpm_async_fn(dev, async_suspend_noirq)) + continue; get_device(dev); + mutex_unlock(&dpm_list_mtx); - error = device_suspend_noirq(dev, state); + device_suspend_noirq(dev, state, false); - mutex_lock(&dpm_list_mtx); - if (error) { - pm_dev_err(dev, state, " noirq", error); - suspend_stats.failed_suspend_noirq++; - dpm_save_failed_step(SUSPEND_SUSPEND_NOIRQ); - dpm_save_failed_dev(dev_name(dev)); - put_device(dev); - break; - } - if (!list_empty(&dev->power.entry)) - list_move(&dev->power.entry, &dpm_noirq_list); put_device(dev); - if (pm_wakeup_pending()) { - error = -EBUSY; + mutex_lock(&dpm_list_mtx); + + if (READ_ONCE(async_error)) { + dpm_async_suspend_complete_all(&dpm_late_early_list); + /* + * Move all devices to the target list to resume them + * properly. + */ + list_splice_init(&dpm_late_early_list, &dpm_noirq_list); break; } } + mutex_unlock(&dpm_list_mtx); + + async_synchronize_full(); + + error = READ_ONCE(async_error); if (error) - dpm_resume_noirq(resume_event(state)); - else - dpm_show_time(starttime, state, "noirq"); + dpm_save_failed_step(SUSPEND_SUSPEND_NOIRQ); + + dpm_show_time(starttime, state, error, "noirq"); + trace_suspend_resume(TPS("dpm_suspend_noirq"), state.event, false); return error; } /** + * dpm_suspend_noirq - Execute "noirq suspend" callbacks for all devices. + * @state: PM transition of the system being carried out. + * + * Prevent device drivers' interrupt handlers from being called and invoke + * "noirq" suspend callbacks for all non-sysdev devices. + */ +int dpm_suspend_noirq(pm_message_t state) +{ + int ret; + + device_wakeup_arm_wake_irqs(); + suspend_device_irqs(); + + ret = dpm_noirq_suspend_devices(state); + if (ret) + dpm_resume_noirq(resume_event(state)); + + return ret; +} + +static void dpm_propagate_wakeup_to_parent(struct device *dev) +{ + struct device *parent = dev->parent; + + if (!parent) + return; + + spin_lock_irq(&parent->power.lock); + + if (device_wakeup_path(dev) && !parent->power.ignore_children) + parent->power.wakeup_path = true; + + spin_unlock_irq(&parent->power.lock); +} + +static void async_suspend_late(void *data, async_cookie_t cookie); + +/** * device_suspend_late - Execute a "late suspend" callback for given device. * @dev: Device to handle. * @state: PM transition of the system being carried out. + * @async: If true, the device is being suspended asynchronously. * * Runtime PM is disabled for @dev while this function is being executed. */ -static int device_suspend_late(struct device *dev, pm_message_t state) +static void device_suspend_late(struct device *dev, pm_message_t state, bool async) { pm_callback_t callback = NULL; - char *info = NULL; + const char *info = NULL; + int error = 0; + + TRACE_DEVICE(dev); + TRACE_SUSPEND(0); + + dpm_wait_for_subordinate(dev, async); + + if (READ_ONCE(async_error)) + goto Complete; + + if (pm_wakeup_pending()) { + WRITE_ONCE(async_error, -EBUSY); + goto Complete; + } + + if (dev->power.direct_complete) + goto Complete; + /* + * Disable runtime PM for the device without checking if there is a + * pending resume request for it. + */ __pm_runtime_disable(dev, false); if (dev->power.syscore) - return 0; + goto Skip; if (dev->pm_domain) { info = "late power domain "; @@ -946,57 +1668,118 @@ static int device_suspend_late(struct device *dev, pm_message_t state) info = "late bus "; callback = pm_late_early_op(dev->bus->pm, state); } + if (callback) + goto Run; - if (!callback && dev->driver && dev->driver->pm) { + if (dev_pm_skip_suspend(dev)) + goto Skip; + + if (dev->driver && dev->driver->pm) { info = "late driver "; callback = pm_late_early_op(dev->driver->pm, state); } - return dpm_run_callback(callback, dev, state, info); +Run: + error = dpm_run_callback(callback, dev, state, info); + if (error) { + WRITE_ONCE(async_error, error); + dpm_save_failed_dev(dev_name(dev)); + pm_dev_err(dev, state, async ? " async late" : " late", error); + pm_runtime_enable(dev); + goto Complete; + } + dpm_propagate_wakeup_to_parent(dev); + +Skip: + dev->power.is_late_suspended = true; + +Complete: + TRACE_SUSPEND(error); + complete_all(&dev->power.completion); + + if (error || READ_ONCE(async_error)) + return; + + dpm_async_suspend_superior(dev, async_suspend_late); +} + +static void async_suspend_late(void *data, async_cookie_t cookie) +{ + struct device *dev = data; + + device_suspend_late(dev, pm_transition, true); + put_device(dev); } /** * dpm_suspend_late - Execute "late suspend" callbacks for all devices. * @state: PM transition of the system being carried out. */ -static int dpm_suspend_late(pm_message_t state) +int dpm_suspend_late(pm_message_t state) { ktime_t starttime = ktime_get(); - int error = 0; + struct device *dev; + int error; + + trace_suspend_resume(TPS("dpm_suspend_late"), state.event, true); + + pm_transition = state; + async_error = 0; + + wake_up_all_idle_cpus(); mutex_lock(&dpm_list_mtx); + + /* + * Start processing "async" leaf devices upfront so they don't need to + * wait for the "sync" devices they don't depend on. + */ + list_for_each_entry_reverse(dev, &dpm_suspended_list, power.entry) { + dpm_clear_async_state(dev); + if (dpm_leaf_device(dev)) + dpm_async_with_cleanup(dev, async_suspend_late); + } + while (!list_empty(&dpm_suspended_list)) { - struct device *dev = to_device(dpm_suspended_list.prev); + dev = to_device(dpm_suspended_list.prev); + + list_move(&dev->power.entry, &dpm_late_early_list); + + if (dpm_async_fn(dev, async_suspend_late)) + continue; get_device(dev); + mutex_unlock(&dpm_list_mtx); - error = device_suspend_late(dev, state); + device_suspend_late(dev, state, false); - mutex_lock(&dpm_list_mtx); - if (error) { - pm_dev_err(dev, state, " late", error); - suspend_stats.failed_suspend_late++; - dpm_save_failed_step(SUSPEND_SUSPEND_LATE); - dpm_save_failed_dev(dev_name(dev)); - put_device(dev); - break; - } - if (!list_empty(&dev->power.entry)) - list_move(&dev->power.entry, &dpm_late_early_list); put_device(dev); - if (pm_wakeup_pending()) { - error = -EBUSY; + mutex_lock(&dpm_list_mtx); + + if (READ_ONCE(async_error)) { + dpm_async_suspend_complete_all(&dpm_suspended_list); + /* + * Move all devices to the target list to resume them + * properly. + */ + list_splice_init(&dpm_suspended_list, &dpm_late_early_list); break; } } + mutex_unlock(&dpm_list_mtx); - if (error) - dpm_resume_early(resume_event(state)); - else - dpm_show_time(starttime, state, "late"); + async_synchronize_full(); + + error = READ_ONCE(async_error); + if (error) { + dpm_save_failed_step(SUSPEND_SUSPEND_LATE); + dpm_resume_early(resume_event(state)); + } + dpm_show_time(starttime, state, error, "late"); + trace_suspend_resume(TPS("dpm_suspend_late"), state.event, false); return error; } @@ -1006,17 +1789,20 @@ static int dpm_suspend_late(pm_message_t state) */ int dpm_suspend_end(pm_message_t state) { - int error = dpm_suspend_late(state); + ktime_t starttime = ktime_get(); + int error; + + error = dpm_suspend_late(state); if (error) - return error; + goto out; error = dpm_suspend_noirq(state); - if (error) { + if (error) dpm_resume_early(resume_event(state)); - return error; - } - return 0; +out: + dpm_show_time(starttime, state, error, "end"); + return error; } EXPORT_SYMBOL_GPL(dpm_suspend_end); @@ -1025,57 +1811,118 @@ EXPORT_SYMBOL_GPL(dpm_suspend_end); * @dev: Device to suspend. * @state: PM transition of the system being carried out. * @cb: Suspend callback to execute. + * @info: string description of caller. */ static int legacy_suspend(struct device *dev, pm_message_t state, - int (*cb)(struct device *dev, pm_message_t state)) + int (*cb)(struct device *dev, pm_message_t state), + const char *info) { int error; ktime_t calltime; - calltime = initcall_debug_start(dev); + calltime = initcall_debug_start(dev, cb); + trace_device_pm_callback_start(dev, info, state.event); error = cb(dev, state); - suspend_report_result(cb, error); + trace_device_pm_callback_end(dev, error); + suspend_report_result(dev, cb, error); - initcall_debug_report(dev, calltime, error); + initcall_debug_report(dev, calltime, cb, error); return error; } +static void dpm_clear_superiors_direct_complete(struct device *dev) +{ + struct device_link *link; + int idx; + + if (dev->parent) { + spin_lock_irq(&dev->parent->power.lock); + dev->parent->power.direct_complete = false; + spin_unlock_irq(&dev->parent->power.lock); + } + + idx = device_links_read_lock(); + + dev_for_each_link_to_supplier(link, dev) { + spin_lock_irq(&link->supplier->power.lock); + link->supplier->power.direct_complete = false; + spin_unlock_irq(&link->supplier->power.lock); + } + + device_links_read_unlock(idx); +} + +static void async_suspend(void *data, async_cookie_t cookie); + /** * device_suspend - Execute "suspend" callbacks for given device. * @dev: Device to handle. * @state: PM transition of the system being carried out. * @async: If true, the device is being suspended asynchronously. */ -static int __device_suspend(struct device *dev, pm_message_t state, bool async) +static void device_suspend(struct device *dev, pm_message_t state, bool async) { pm_callback_t callback = NULL; - char *info = NULL; + const char *info = NULL; int error = 0; + DECLARE_DPM_WATCHDOG_ON_STACK(wd); - dpm_wait_for_children(dev, async); + TRACE_DEVICE(dev); + TRACE_SUSPEND(0); + + dpm_wait_for_subordinate(dev, async); - if (async_error) + if (READ_ONCE(async_error)) { + dev->power.direct_complete = false; goto Complete; + } /* - * If a device configured to wake up the system from sleep states - * has been suspended at run time and there's a resume request pending - * for it, this is equivalent to the device signaling wakeup, so the - * system suspend operation should be aborted. + * Wait for possible runtime PM transitions of the device in progress + * to complete and if there's a runtime resume request pending for it, + * resume it before proceeding with invoking the system-wide suspend + * callbacks for it. + * + * If the system-wide suspend callbacks below change the configuration + * of the device, they must disable runtime PM for it or otherwise + * ensure that its runtime-resume callbacks will not be confused by that + * change in case they are invoked going forward. */ - if (pm_runtime_barrier(dev) && device_may_wakeup(dev)) - pm_wakeup_event(dev, 0); + pm_runtime_barrier(dev); if (pm_wakeup_pending()) { - async_error = -EBUSY; + dev->power.direct_complete = false; + WRITE_ONCE(async_error, -EBUSY); goto Complete; } if (dev->power.syscore) goto Complete; + /* Avoid direct_complete to let wakeup_path propagate. */ + if (device_may_wakeup(dev) || device_wakeup_path(dev)) + dev->power.direct_complete = false; + + if (dev->power.direct_complete) { + if (pm_runtime_status_suspended(dev)) { + pm_runtime_disable(dev); + if (pm_runtime_status_suspended(dev)) { + pm_dev_dbg(dev, state, "direct-complete "); + dev->power.is_suspended = true; + goto Complete; + } + + pm_runtime_enable(dev); + } + dev->power.direct_complete = false; + } + + dev->power.may_skip_resume = true; + dev->power.must_resume = !dev_pm_test_driver_flags(dev, DPM_FLAG_MAY_SKIP_RESUME); + + dpm_watchdog_set(&wd, dev); device_lock(dev); if (dev->pm_domain) { @@ -1090,16 +1937,10 @@ static int __device_suspend(struct device *dev, pm_message_t state, bool async) goto Run; } - if (dev->class) { - if (dev->class->pm) { - info = "class "; - callback = pm_op(dev->class->pm, state); - goto Run; - } else if (dev->class->suspend) { - pm_dev_dbg(dev, state, "legacy class "); - error = legacy_suspend(dev, state, dev->class->suspend); - goto End; - } + if (dev->class && dev->class->pm) { + info = "class "; + callback = pm_op(dev->class->pm, state); + goto Run; } if (dev->bus) { @@ -1108,7 +1949,8 @@ static int __device_suspend(struct device *dev, pm_message_t state, bool async) callback = pm_op(dev->bus->pm, state); } else if (dev->bus->suspend) { pm_dev_dbg(dev, state, "legacy bus "); - error = legacy_suspend(dev, state, dev->bus->suspend); + error = legacy_suspend(dev, state, dev->bus->suspend, + "legacy bus "); goto End; } } @@ -1124,46 +1966,38 @@ static int __device_suspend(struct device *dev, pm_message_t state, bool async) End: if (!error) { dev->power.is_suspended = true; - if (dev->power.wakeup_path - && dev->parent && !dev->parent->power.ignore_children) - dev->parent->power.wakeup_path = true; + if (device_may_wakeup(dev)) + dev->power.wakeup_path = true; + + dpm_propagate_wakeup_to_parent(dev); + dpm_clear_superiors_direct_complete(dev); } device_unlock(dev); + dpm_watchdog_clear(&wd); Complete: - complete_all(&dev->power.completion); - if (error) - async_error = error; - - return error; -} - -static void async_suspend(void *data, async_cookie_t cookie) -{ - struct device *dev = (struct device *)data; - int error; - - error = __device_suspend(dev, pm_transition, true); if (error) { + WRITE_ONCE(async_error, error); dpm_save_failed_dev(dev_name(dev)); - pm_dev_err(dev, pm_transition, " async", error); + pm_dev_err(dev, state, async ? " async" : "", error); } - put_device(dev); + complete_all(&dev->power.completion); + TRACE_SUSPEND(error); + + if (error || READ_ONCE(async_error)) + return; + + dpm_async_suspend_superior(dev, async_suspend); } -static int device_suspend(struct device *dev) +static void async_suspend(void *data, async_cookie_t cookie) { - INIT_COMPLETION(dev->power.completion); + struct device *dev = data; - if (pm_async_enabled && dev->power.async_suspend) { - get_device(dev); - async_schedule(async_suspend, dev); - return 0; - } - - return __device_suspend(dev, pm_transition, false); + device_suspend(dev, pm_transition, true); + put_device(dev); } /** @@ -1173,46 +2007,112 @@ static int device_suspend(struct device *dev) int dpm_suspend(pm_message_t state) { ktime_t starttime = ktime_get(); - int error = 0; + struct device *dev; + int error; + trace_suspend_resume(TPS("dpm_suspend"), state.event, true); might_sleep(); - mutex_lock(&dpm_list_mtx); + devfreq_suspend(); + cpufreq_suspend(); + pm_transition = state; async_error = 0; + + mutex_lock(&dpm_list_mtx); + + /* + * Start processing "async" leaf devices upfront so they don't need to + * wait for the "sync" devices they don't depend on. + */ + list_for_each_entry_reverse(dev, &dpm_prepared_list, power.entry) { + dpm_clear_async_state(dev); + if (dpm_leaf_device(dev)) + dpm_async_with_cleanup(dev, async_suspend); + } + while (!list_empty(&dpm_prepared_list)) { - struct device *dev = to_device(dpm_prepared_list.prev); + dev = to_device(dpm_prepared_list.prev); + + list_move(&dev->power.entry, &dpm_suspended_list); + + if (dpm_async_fn(dev, async_suspend)) + continue; get_device(dev); + mutex_unlock(&dpm_list_mtx); - error = device_suspend(dev); + device_suspend(dev, state, false); + + put_device(dev); mutex_lock(&dpm_list_mtx); - if (error) { - pm_dev_err(dev, state, "", error); - dpm_save_failed_dev(dev_name(dev)); - put_device(dev); + + if (READ_ONCE(async_error)) { + dpm_async_suspend_complete_all(&dpm_prepared_list); + /* + * Move all devices to the target list to resume them + * properly. + */ + list_splice_init(&dpm_prepared_list, &dpm_suspended_list); break; } - if (!list_empty(&dev->power.entry)) - list_move(&dev->power.entry, &dpm_suspended_list); - put_device(dev); - if (async_error) - break; } + mutex_unlock(&dpm_list_mtx); + async_synchronize_full(); - if (!error) - error = async_error; - if (error) { - suspend_stats.failed_suspend++; + + error = READ_ONCE(async_error); + if (error) dpm_save_failed_step(SUSPEND_SUSPEND); - } else - dpm_show_time(starttime, state, NULL); + + dpm_show_time(starttime, state, error, NULL); + trace_suspend_resume(TPS("dpm_suspend"), state.event, false); return error; } +static bool device_prepare_smart_suspend(struct device *dev) +{ + struct device_link *link; + bool ret = true; + int idx; + + /* + * The "smart suspend" feature is enabled for devices whose drivers ask + * for it and for devices without PM callbacks. + * + * However, if "smart suspend" is not enabled for the device's parent + * or any of its suppliers that take runtime PM into account, it cannot + * be enabled for the device either. + */ + if (!dev->power.no_pm_callbacks && + !dev_pm_test_driver_flags(dev, DPM_FLAG_SMART_SUSPEND)) + return false; + + if (dev->parent && !dev_pm_smart_suspend(dev->parent) && + !dev->parent->power.ignore_children && !pm_runtime_blocked(dev->parent)) + return false; + + idx = device_links_read_lock(); + + dev_for_each_link_to_supplier(link, dev) { + if (!device_link_test(link, DL_FLAG_PM_RUNTIME)) + continue; + + if (!dev_pm_smart_suspend(link->supplier) && + !pm_runtime_blocked(link->supplier)) { + ret = false; + break; + } + } + + device_links_read_unlock(idx); + + return ret; +} + /** * device_prepare - Prepare a device for system power transition. * @dev: Device to handle. @@ -1224,11 +2124,8 @@ int dpm_suspend(pm_message_t state) static int device_prepare(struct device *dev, pm_message_t state) { int (*callback)(struct device *) = NULL; - char *info = NULL; - int error = 0; - - if (dev->power.syscore) - return 0; + bool smart_suspend; + int ret = 0; /* * If a device's parent goes into runtime suspend at the wrong time, @@ -1237,38 +2134,69 @@ static int device_prepare(struct device *dev, pm_message_t state) * it again during the complete phase. */ pm_runtime_get_noresume(dev); + /* + * If runtime PM is disabled for the device at this point and it has + * never been enabled so far, it should not be enabled until this system + * suspend-resume cycle is complete, so prepare to trigger a warning on + * subsequent attempts to enable it. + */ + smart_suspend = !pm_runtime_block_if_disabled(dev); + + if (dev->power.syscore) + return 0; device_lock(dev); - dev->power.wakeup_path = device_may_wakeup(dev); + dev->power.wakeup_path = false; + dev->power.out_band_wakeup = false; - if (dev->pm_domain) { - info = "preparing power domain "; + if (dev->power.no_pm_callbacks) + goto unlock; + + if (dev->pm_domain) callback = dev->pm_domain->ops.prepare; - } else if (dev->type && dev->type->pm) { - info = "preparing type "; + else if (dev->type && dev->type->pm) callback = dev->type->pm->prepare; - } else if (dev->class && dev->class->pm) { - info = "preparing class "; + else if (dev->class && dev->class->pm) callback = dev->class->pm->prepare; - } else if (dev->bus && dev->bus->pm) { - info = "preparing bus "; + else if (dev->bus && dev->bus->pm) callback = dev->bus->pm->prepare; - } - if (!callback && dev->driver && dev->driver->pm) { - info = "preparing driver "; + if (!callback && dev->driver && dev->driver->pm) callback = dev->driver->pm->prepare; - } - if (callback) { - error = callback(dev); - suspend_report_result(callback, error); - } + if (callback) + ret = callback(dev); +unlock: device_unlock(dev); - return error; + if (ret < 0) { + suspend_report_result(dev, callback, ret); + pm_runtime_put(dev); + return ret; + } + /* Do not enable "smart suspend" for devices with disabled runtime PM. */ + if (smart_suspend) + smart_suspend = device_prepare_smart_suspend(dev); + + spin_lock_irq(&dev->power.lock); + + dev->power.smart_suspend = smart_suspend; + /* + * A positive return value from ->prepare() means "this device appears + * to be runtime-suspended and its state is fine, so if it really is + * runtime-suspended, you can leave it in that state provided that you + * will do the same thing with all of its descendants". This only + * applies to suspend transitions, however. + */ + dev->power.direct_complete = state.event == PM_EVENT_SUSPEND && + (ret > 0 || dev->power.no_pm_callbacks) && + !dev_pm_test_driver_flags(dev, DPM_FLAG_NO_DIRECT_COMPLETE); + + spin_unlock_irq(&dev->power.lock); + + return 0; } /** @@ -1281,36 +2209,55 @@ int dpm_prepare(pm_message_t state) { int error = 0; - might_sleep(); + trace_suspend_resume(TPS("dpm_prepare"), state.event, true); + + /* + * Give a chance for the known devices to complete their probes, before + * disable probing of devices. This sync point is important at least + * at boot time + hibernation restore. + */ + wait_for_device_probe(); + /* + * It is unsafe if probing of devices will happen during suspend or + * hibernation and system behavior will be unpredictable in this case. + * So, let's prohibit device's probing here and defer their probes + * instead. The normal behavior will be restored in dpm_complete(). + */ + device_block_probing(); mutex_lock(&dpm_list_mtx); - while (!list_empty(&dpm_list)) { + while (!list_empty(&dpm_list) && !error) { struct device *dev = to_device(dpm_list.next); get_device(dev); + mutex_unlock(&dpm_list_mtx); + trace_device_pm_callback_start(dev, "", state.event); error = device_prepare(dev, state); + trace_device_pm_callback_end(dev, error); mutex_lock(&dpm_list_mtx); - if (error) { - if (error == -EAGAIN) { - put_device(dev); - error = 0; - continue; - } - printk(KERN_INFO "PM: Device %s not prepared " - "for power transition: code %d\n", - dev_name(dev), error); - put_device(dev); - break; + + if (!error) { + dev->power.is_prepared = true; + if (!list_empty(&dev->power.entry)) + list_move_tail(&dev->power.entry, &dpm_prepared_list); + } else if (error == -EAGAIN) { + error = 0; + } else { + dev_info(dev, "not prepared for power transition: code %d\n", + error); } - dev->power.is_prepared = true; - if (!list_empty(&dev->power.entry)) - list_move_tail(&dev->power.entry, &dpm_prepared_list); + + mutex_unlock(&dpm_list_mtx); + put_device(dev); + + mutex_lock(&dpm_list_mtx); } mutex_unlock(&dpm_list_mtx); + trace_suspend_resume(TPS("dpm_prepare"), state.event, false); return error; } @@ -1323,29 +2270,33 @@ int dpm_prepare(pm_message_t state) */ int dpm_suspend_start(pm_message_t state) { + ktime_t starttime = ktime_get(); int error; error = dpm_prepare(state); - if (error) { - suspend_stats.failed_prepare++; + if (error) dpm_save_failed_step(SUSPEND_PREPARE); - } else + else { + pm_restrict_gfp_mask(); error = dpm_suspend(state); + } + + dpm_show_time(starttime, state, error, "start"); return error; } EXPORT_SYMBOL_GPL(dpm_suspend_start); -void __suspend_report_result(const char *function, void *fn, int ret) +void __suspend_report_result(const char *function, struct device *dev, void *fn, int ret) { if (ret) - printk(KERN_ERR "%s(): %pF returns %d\n", function, fn, ret); + dev_err(dev, "%s(): %ps returns %d\n", function, fn, ret); } EXPORT_SYMBOL_GPL(__suspend_report_result); /** * device_pm_wait_for_dev - Wait for suspend/resume of a device to complete. - * @dev: Device to wait for. * @subordinate: Device that needs to wait for @dev. + * @dev: Device to wait for. */ int device_pm_wait_for_dev(struct device *subordinate, struct device *dev) { @@ -1375,3 +2326,39 @@ void dpm_for_each_dev(void *data, void (*fn)(struct device *, void *)) device_pm_unlock(); } EXPORT_SYMBOL_GPL(dpm_for_each_dev); + +static bool pm_ops_is_empty(const struct dev_pm_ops *ops) +{ + if (!ops) + return true; + + return !ops->prepare && + !ops->suspend && + !ops->suspend_late && + !ops->suspend_noirq && + !ops->resume_noirq && + !ops->resume_early && + !ops->resume && + !ops->complete; +} + +void device_pm_check_callbacks(struct device *dev) +{ + unsigned long flags; + + spin_lock_irqsave(&dev->power.lock, flags); + dev->power.no_pm_callbacks = + (!dev->bus || (pm_ops_is_empty(dev->bus->pm) && + !dev->bus->suspend && !dev->bus->resume)) && + (!dev->class || pm_ops_is_empty(dev->class->pm)) && + (!dev->type || pm_ops_is_empty(dev->type->pm)) && + (!dev->pm_domain || pm_ops_is_empty(&dev->pm_domain->ops)) && + (!dev->driver || (pm_ops_is_empty(dev->driver->pm) && + !dev->driver->suspend && !dev->driver->resume)); + spin_unlock_irqrestore(&dev->power.lock, flags); +} + +bool dev_pm_skip_suspend(struct device *dev) +{ + return dev_pm_smart_suspend(dev) && pm_runtime_status_suspended(dev); +} |