From 107d47b2b95ef478d71f3bf36201886d7475427a Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sat, 18 Apr 2020 18:29:30 +0200 Subject: PM: sleep: core: Simplify the SMART_SUSPEND flag handling The code to handle the SMART_SUSPEND driver PM flag is hard to follow and somewhat inconsistent with respect to devices without middle-layer (subsystem) callbacks. Namely, for those devices the core takes the role of a middle layer in providing the expected ordering of execution of callbacks (under the assumption that the drivers setting SMART_SUSPEND can reuse their PM-runtime callbacks directly for system-wide suspend). To that end, it prevents driver ->suspend_late and ->suspend_noirq callbacks from being executed for devices that are still runtime-suspended in __device_suspend_late(), because running the same callback funtion that was previously run by PM-runtime for them may be invalid. However, it does that only for devices without any middle-layer callbacks for the late/noirq/early suspend/resume phases even though it would be simpler and more consistent to skip the driver-lavel callbacks for all devices with SMART_SUSPEND set that are runtime-suspended in __device_suspend_late(). Simplify the code in accordance with the above observation. Suggested-by: Alan Stern Signed-off-by: Rafael J. Wysocki Acked-by: Alan Stern --- drivers/base/power/main.c | 118 +++++++++++++++------------------------------- 1 file changed, 39 insertions(+), 79 deletions(-) diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index fdd508a78ffd..5d0225573bbe 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -561,24 +561,6 @@ static void dpm_watchdog_clear(struct dpm_watchdog *wd) /*------------------------- Resume routines -------------------------*/ -/** - * suspend_event - Return a "suspend" message for given "resume" one. - * @resume_msg: PM message representing a system-wide resume transition. - */ -static pm_message_t suspend_event(pm_message_t resume_msg) -{ - switch (resume_msg.event) { - case PM_EVENT_RESUME: - return PMSG_SUSPEND; - case PM_EVENT_THAW: - case PM_EVENT_RESTORE: - return PMSG_FREEZE; - case PM_EVENT_RECOVER: - return PMSG_HIBERNATE; - } - return PMSG_ON; -} - /** * dev_pm_may_skip_resume - System-wide device resume optimization check. * @dev: Target device. @@ -656,37 +638,36 @@ static int device_resume_noirq(struct device *dev, pm_message_t state, bool asyn if (!dpm_wait_for_superior(dev, async)) goto Out; - skip_resume = dev_pm_may_skip_resume(dev); - callback = dpm_subsys_resume_noirq_cb(dev, state, &info); - if (callback) + if (callback) { + skip_resume = false; goto Run; + } + skip_resume = dev_pm_may_skip_resume(dev); if (skip_resume) goto Skip; - if (dev_pm_smart_suspend_and_suspended(dev)) { - pm_message_t suspend_msg = suspend_event(state); - - /* - * If "freeze" callbacks have been skipped during a transition - * related to hibernation, the subsequent "thaw" callbacks must - * be skipped too or bad things may happen. Otherwise, resume - * callbacks are going to be run for the device, so its runtime - * PM status must be changed to reflect the new state after the - * transition under way. - */ - if (!dpm_subsys_suspend_late_cb(dev, suspend_msg, NULL) && - !dpm_subsys_suspend_noirq_cb(dev, suspend_msg, NULL)) { - if (state.event == PM_EVENT_THAW) { - skip_resume = true; - goto Skip; - } else { - pm_runtime_set_active(dev); - } - } + /* + * If "freeze" driver callbacks have been skipped during hibernation, + * because the device was runtime-suspended in __device_suspend_late(), + * the corresponding "thaw" callbacks must be skipped too, because + * running them for a runtime-suspended device may not be valid. + */ + if (dev_pm_smart_suspend_and_suspended(dev) && + state.event == PM_EVENT_THAW) { + skip_resume = true; + goto Skip; } + /* + * The device is going to be resumed, so set its PM-runtime status to + * "active", but do that only if DPM_FLAG_SMART_SUSPEND is set to avoid + * confusing drivers that don't use it. + */ + if (dev_pm_smart_suspend_and_suspended(dev)) + pm_runtime_set_active(dev); + if (dev->driver && dev->driver->pm) { info = "noirq driver "; callback = pm_noirq_op(dev->driver->pm, state); @@ -1274,32 +1255,6 @@ static pm_callback_t dpm_subsys_suspend_noirq_cb(struct device *dev, return callback; } -static bool device_must_resume(struct device *dev, pm_message_t state, - bool no_subsys_suspend_noirq) -{ - pm_message_t resume_msg = resume_event(state); - - /* - * If all of the device driver's "noirq", "late" and "early" callbacks - * are invoked directly by the core, the decision to allow the device to - * stay in suspend can be based on its current runtime PM status and its - * wakeup settings. - */ - if (no_subsys_suspend_noirq && - !dpm_subsys_suspend_late_cb(dev, state, NULL) && - !dpm_subsys_resume_early_cb(dev, resume_msg, NULL) && - !dpm_subsys_resume_noirq_cb(dev, resume_msg, NULL)) - return !pm_runtime_status_suspended(dev) && - (resume_msg.event != PM_EVENT_RESUME || - (device_can_wakeup(dev) && !device_may_wakeup(dev))); - - /* - * The only safe strategy here is to require that if the device may not - * be left in suspend, resume callbacks must be invoked for it. - */ - return !dev->power.may_skip_resume; -} - /** * __device_suspend_noirq - Execute a "noirq suspend" callback for given device. * @dev: Device to handle. @@ -1313,7 +1268,6 @@ static int __device_suspend_noirq(struct device *dev, pm_message_t state, bool a { pm_callback_t callback; const char *info; - bool no_subsys_cb = false; int error = 0; TRACE_DEVICE(dev); @@ -1331,9 +1285,7 @@ static int __device_suspend_noirq(struct device *dev, pm_message_t state, bool a if (callback) goto Run; - no_subsys_cb = !dpm_subsys_suspend_late_cb(dev, state, NULL); - - if (dev_pm_smart_suspend_and_suspended(dev) && no_subsys_cb) + if (dev_pm_smart_suspend_and_suspended(dev)) goto Skip; if (dev->driver && dev->driver->pm) { @@ -1351,13 +1303,16 @@ Run: Skip: dev->power.is_noirq_suspended = true; - if (dev_pm_test_driver_flags(dev, DPM_FLAG_LEAVE_SUSPENDED)) { - dev->power.must_resume = dev->power.must_resume || - atomic_read(&dev->power.usage_count) > 1 || - device_must_resume(dev, state, no_subsys_cb); - } else { + /* + * Skipping the resume of devices that were in use right before the + * system suspend (as indicated by their PM-runtime usage counters) + * would be suboptimal. Also resume them if doing that is not allowed + * to be skipped. + */ + if (atomic_read(&dev->power.usage_count) > 1 || + !(dev_pm_test_driver_flags(dev, DPM_FLAG_LEAVE_SUSPENDED) && + dev->power.may_skip_resume)) dev->power.must_resume = true; - } if (dev->power.must_resume) dpm_superior_set_must_resume(dev); @@ -1539,9 +1494,14 @@ static int __device_suspend_late(struct device *dev, pm_message_t state, bool as if (callback) goto Run; - if (dev_pm_smart_suspend_and_suspended(dev) && - !dpm_subsys_suspend_noirq_cb(dev, state, NULL)) + if (dev_pm_smart_suspend_and_suspended(dev)) { + /* + * In principle, the resume of the device may be skippend if it + * remains in runtime suspend at this point. + */ + dev->power.may_skip_resume = true; goto Skip; + } if (dev->driver && dev->driver->pm) { info = "late driver "; -- cgit v1.2.3 From 30205377ddbb717ee451e872fd59511f4f76373d Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sat, 18 Apr 2020 18:51:44 +0200 Subject: PM: sleep: core: Fold functions into their callers Fold four functions in the PM core that each have only one caller now into their callers. No intentional functional impact. Signed-off-by: Rafael J. Wysocki Acked-by: Alan Stern --- drivers/base/power/main.c | 198 ++++++++++++++-------------------------------- 1 file changed, 60 insertions(+), 138 deletions(-) diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index 5d0225573bbe..75d7cdb4de9c 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -573,43 +573,6 @@ bool dev_pm_may_skip_resume(struct device *dev) return !dev->power.must_resume && pm_transition.event != PM_EVENT_RESTORE; } -static pm_callback_t dpm_subsys_resume_noirq_cb(struct device *dev, - pm_message_t state, - const char **info_p) -{ - pm_callback_t callback; - const char *info; - - if (dev->pm_domain) { - info = "noirq power domain "; - callback = pm_noirq_op(&dev->pm_domain->ops, state); - } else if (dev->type && dev->type->pm) { - info = "noirq type "; - callback = pm_noirq_op(dev->type->pm, state); - } else if (dev->class && dev->class->pm) { - info = "noirq class "; - callback = pm_noirq_op(dev->class->pm, state); - } else if (dev->bus && dev->bus->pm) { - info = "noirq bus "; - callback = pm_noirq_op(dev->bus->pm, state); - } else { - return NULL; - } - - if (info_p) - *info_p = info; - - return callback; -} - -static pm_callback_t dpm_subsys_suspend_noirq_cb(struct device *dev, - pm_message_t state, - const char **info_p); - -static pm_callback_t dpm_subsys_suspend_late_cb(struct device *dev, - pm_message_t state, - const char **info_p); - /** * device_resume_noirq - Execute a "noirq resume" callback for given device. * @dev: Device to handle. @@ -621,8 +584,8 @@ static pm_callback_t dpm_subsys_suspend_late_cb(struct device *dev, */ static int device_resume_noirq(struct device *dev, pm_message_t state, bool async) { - pm_callback_t callback; - const char *info; + pm_callback_t callback = NULL; + const char *info = NULL; bool skip_resume; int error = 0; @@ -638,7 +601,19 @@ static int device_resume_noirq(struct device *dev, pm_message_t state, bool asyn if (!dpm_wait_for_superior(dev, async)) goto Out; - callback = dpm_subsys_resume_noirq_cb(dev, state, &info); + if (dev->pm_domain) { + info = "noirq power domain "; + callback = pm_noirq_op(&dev->pm_domain->ops, state); + } else if (dev->type && dev->type->pm) { + info = "noirq type "; + callback = pm_noirq_op(dev->type->pm, state); + } else if (dev->class && dev->class->pm) { + info = "noirq class "; + callback = pm_noirq_op(dev->class->pm, state); + } else if (dev->bus && dev->bus->pm) { + info = "noirq bus "; + callback = pm_noirq_op(dev->bus->pm, state); + } if (callback) { skip_resume = false; goto Run; @@ -791,35 +766,6 @@ void dpm_resume_noirq(pm_message_t state) cpuidle_resume(); } -static pm_callback_t dpm_subsys_resume_early_cb(struct device *dev, - pm_message_t state, - const char **info_p) -{ - pm_callback_t callback; - const char *info; - - if (dev->pm_domain) { - info = "early power domain "; - callback = pm_late_early_op(&dev->pm_domain->ops, state); - } else if (dev->type && dev->type->pm) { - info = "early type "; - callback = pm_late_early_op(dev->type->pm, state); - } else if (dev->class && dev->class->pm) { - info = "early class "; - callback = pm_late_early_op(dev->class->pm, state); - } else if (dev->bus && dev->bus->pm) { - info = "early bus "; - callback = pm_late_early_op(dev->bus->pm, state); - } else { - return NULL; - } - - if (info_p) - *info_p = info; - - return callback; -} - /** * device_resume_early - Execute an "early resume" callback for given device. * @dev: Device to handle. @@ -830,8 +776,8 @@ static pm_callback_t dpm_subsys_resume_early_cb(struct device *dev, */ static int device_resume_early(struct device *dev, pm_message_t state, bool async) { - pm_callback_t callback; - const char *info; + pm_callback_t callback = NULL; + const char *info = NULL; int error = 0; TRACE_DEVICE(dev); @@ -846,9 +792,19 @@ static int device_resume_early(struct device *dev, pm_message_t state, bool asyn if (!dpm_wait_for_superior(dev, async)) goto Out; - callback = dpm_subsys_resume_early_cb(dev, state, &info); - - if (!callback && dev->driver && dev->driver->pm) { + if (dev->pm_domain) { + info = "early power domain "; + callback = pm_late_early_op(&dev->pm_domain->ops, state); + } else if (dev->type && dev->type->pm) { + info = "early type "; + callback = pm_late_early_op(dev->type->pm, state); + } else if (dev->class && dev->class->pm) { + info = "early class "; + callback = pm_late_early_op(dev->class->pm, state); + } else if (dev->bus && dev->bus->pm) { + info = "early bus "; + callback = pm_late_early_op(dev->bus->pm, state); + } else if (dev->driver && dev->driver->pm) { info = "early driver "; callback = pm_late_early_op(dev->driver->pm, state); } @@ -1226,35 +1182,6 @@ static void dpm_superior_set_must_resume(struct device *dev) device_links_read_unlock(idx); } -static pm_callback_t dpm_subsys_suspend_noirq_cb(struct device *dev, - pm_message_t state, - const char **info_p) -{ - pm_callback_t callback; - const char *info; - - if (dev->pm_domain) { - info = "noirq power domain "; - callback = pm_noirq_op(&dev->pm_domain->ops, state); - } else if (dev->type && dev->type->pm) { - info = "noirq type "; - callback = pm_noirq_op(dev->type->pm, state); - } else if (dev->class && dev->class->pm) { - info = "noirq class "; - callback = pm_noirq_op(dev->class->pm, state); - } else if (dev->bus && dev->bus->pm) { - info = "noirq bus "; - callback = pm_noirq_op(dev->bus->pm, state); - } else { - return NULL; - } - - if (info_p) - *info_p = info; - - return callback; -} - /** * __device_suspend_noirq - Execute a "noirq suspend" callback for given device. * @dev: Device to handle. @@ -1266,8 +1193,8 @@ static pm_callback_t dpm_subsys_suspend_noirq_cb(struct device *dev, */ static int __device_suspend_noirq(struct device *dev, pm_message_t state, bool async) { - pm_callback_t callback; - const char *info; + pm_callback_t callback = NULL; + const char *info = NULL; int error = 0; TRACE_DEVICE(dev); @@ -1281,7 +1208,19 @@ static int __device_suspend_noirq(struct device *dev, pm_message_t state, bool a if (dev->power.syscore || dev->power.direct_complete) goto Complete; - callback = dpm_subsys_suspend_noirq_cb(dev, state, &info); + if (dev->pm_domain) { + info = "noirq power domain "; + callback = pm_noirq_op(&dev->pm_domain->ops, state); + } else if (dev->type && dev->type->pm) { + info = "noirq type "; + callback = pm_noirq_op(dev->type->pm, state); + } else if (dev->class && dev->class->pm) { + info = "noirq class "; + callback = pm_noirq_op(dev->class->pm, state); + } else if (dev->bus && dev->bus->pm) { + info = "noirq bus "; + callback = pm_noirq_op(dev->bus->pm, state); + } if (callback) goto Run; @@ -1429,35 +1368,6 @@ static void dpm_propagate_wakeup_to_parent(struct device *dev) spin_unlock_irq(&parent->power.lock); } -static pm_callback_t dpm_subsys_suspend_late_cb(struct device *dev, - pm_message_t state, - const char **info_p) -{ - pm_callback_t callback; - const char *info; - - if (dev->pm_domain) { - info = "late power domain "; - callback = pm_late_early_op(&dev->pm_domain->ops, state); - } else if (dev->type && dev->type->pm) { - info = "late type "; - callback = pm_late_early_op(dev->type->pm, state); - } else if (dev->class && dev->class->pm) { - info = "late class "; - callback = pm_late_early_op(dev->class->pm, state); - } else if (dev->bus && dev->bus->pm) { - info = "late bus "; - callback = pm_late_early_op(dev->bus->pm, state); - } else { - return NULL; - } - - if (info_p) - *info_p = info; - - return callback; -} - /** * __device_suspend_late - Execute a "late suspend" callback for given device. * @dev: Device to handle. @@ -1468,8 +1378,8 @@ static pm_callback_t dpm_subsys_suspend_late_cb(struct device *dev, */ static int __device_suspend_late(struct device *dev, pm_message_t state, bool async) { - pm_callback_t callback; - const char *info; + pm_callback_t callback = NULL; + const char *info = NULL; int error = 0; TRACE_DEVICE(dev); @@ -1490,7 +1400,19 @@ static int __device_suspend_late(struct device *dev, pm_message_t state, bool as if (dev->power.syscore || dev->power.direct_complete) goto Complete; - callback = dpm_subsys_suspend_late_cb(dev, state, &info); + if (dev->pm_domain) { + info = "late power domain "; + callback = pm_late_early_op(&dev->pm_domain->ops, state); + } else if (dev->type && dev->type->pm) { + info = "late type "; + callback = pm_late_early_op(dev->type->pm, state); + } else if (dev->class && dev->class->pm) { + info = "late class "; + callback = pm_late_early_op(dev->class->pm, state); + } else if (dev->bus && dev->bus->pm) { + info = "late bus "; + callback = pm_late_early_op(dev->bus->pm, state); + } if (callback) goto Run; -- cgit v1.2.3 From 6e176bf8d46194353163c2cb660808bc633b45d9 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sat, 18 Apr 2020 18:52:08 +0200 Subject: PM: sleep: core: Do not skip callbacks in the resume phase The current code in device_resume_noirq() causes the entire early resume and resume phases of device suspend to be skipped for devices for which the noirq resume phase have been skipped (due to the LEAVE_SUSPENDED flag being set) on the premise that those devices should stay in runtime-suspend after system-wide resume. However, that may not be correct in two situations. First, the middle layer (subsystem) noirq resume callback may be missing for a given device, but its early resume callback may be present and it may need to do something even if it decides to skip the driver callback. Second, if the device's wakeup settings were adjusted in the suspend phase without resuming the device (that was in runtime suspend at that time), they most likely need to be adjusted again in the resume phase and so the driver callback in that phase needs to be run. For the above reason, modify the core to allow the middle layer ->resume_late callback to run even if its ->resume_noirq callback is missing (and the core has skipped the driver-level callback in that phase) and to allow all device callbacks to run in the resume phase. Also make the core set the PM-runtime status of devices with SMART_SUSPEND set whose resume callbacks are not skipped to "active" in the "noirq" resume phase and update the affected subsystems (PCI and ACPI) accordingly. After this change, middle-layer (subsystem) callbacks will always be invoked in all phases of system suspend and resume and driver callbacks will always run in the prepare, suspend, resume, and complete phases for all devices. For devices with SMART_SUSPEND set, driver callbacks will be skipped in the late and noirq phases of system suspend if those devices remain in runtime suspend in __device_suspend_late(). Driver callbacks will also be skipped for them during the noirq and early phases of the "thaw" transition related to hibernation in that case. Setting LEAVE_SUSPENDED means that the driver allows its callbacks to be skipped in the noirq and early phases of system resume, but some additional conditions need to be met for that to happen (among other things, the power.may_skip_resume flag needs to be set for the device during system suspend for the driver callbacks to be skipped during the subsequent resume transition). For all devices with SMART_SUSPEND set whose driver callbacks are invoked during system resume, the PM-runtime status will be set to "active" (by the core). Signed-off-by: Rafael J. Wysocki Acked-by: Alan Stern Acked-by: Bjorn Helgaas --- Documentation/power/pci.rst | 5 +-- drivers/acpi/acpi_lpss.c | 6 ++-- drivers/acpi/device_pm.c | 15 ++++---- drivers/base/power/main.c | 85 ++++++++++++++++++++++----------------------- drivers/pci/pci-driver.c | 18 +++++----- 5 files changed, 62 insertions(+), 67 deletions(-) diff --git a/Documentation/power/pci.rst b/Documentation/power/pci.rst index 0924d29636ad..a39b2461919a 100644 --- a/Documentation/power/pci.rst +++ b/Documentation/power/pci.rst @@ -1035,10 +1035,7 @@ This flag is checked by the PM core, but the PCI bus type informs the PM core which devices may be left in suspend from its perspective (that happens during the "noirq" phase of system-wide suspend and analogous transitions) and next it uses the dev_pm_may_skip_resume() helper to decide whether or not to return from -pci_pm_resume_noirq() early, as the PM core will skip the remaining resume -callbacks for the device during the transition under way and will set its -runtime PM status to "suspended" if dev_pm_may_skip_resume() returns "true" for -it. +pci_pm_resume_noirq() and pci_pm_resume_early() upfront. 3.2. Device Runtime Power Management ------------------------------------ diff --git a/drivers/acpi/acpi_lpss.c b/drivers/acpi/acpi_lpss.c index dee999938213..c4a84df6cc98 100644 --- a/drivers/acpi/acpi_lpss.c +++ b/drivers/acpi/acpi_lpss.c @@ -1093,6 +1093,9 @@ static int acpi_lpss_resume_early(struct device *dev) if (pdata->dev_desc->resume_from_noirq) return 0; + if (dev_pm_may_skip_resume(dev)) + return 0; + return acpi_lpss_do_resume_early(dev); } @@ -1105,9 +1108,6 @@ static int acpi_lpss_resume_noirq(struct device *dev) if (dev_pm_may_skip_resume(dev)) return 0; - if (dev_pm_smart_suspend_and_suspended(dev)) - pm_runtime_set_active(dev); - ret = pm_generic_resume_noirq(dev); if (ret) return ret; diff --git a/drivers/acpi/device_pm.c b/drivers/acpi/device_pm.c index b2263ec67b43..399684085f85 100644 --- a/drivers/acpi/device_pm.c +++ b/drivers/acpi/device_pm.c @@ -1132,14 +1132,6 @@ static int acpi_subsys_resume_noirq(struct device *dev) if (dev_pm_may_skip_resume(dev)) return 0; - /* - * Devices with DPM_FLAG_SMART_SUSPEND may be left in runtime suspend - * during system suspend, so update their runtime PM status to "active" - * as they will be put into D0 going forward. - */ - if (dev_pm_smart_suspend_and_suspended(dev)) - pm_runtime_set_active(dev); - return pm_generic_resume_noirq(dev); } @@ -1153,7 +1145,12 @@ static int acpi_subsys_resume_noirq(struct device *dev) */ static int acpi_subsys_resume_early(struct device *dev) { - int ret = acpi_dev_resume(dev); + int ret; + + if (dev_pm_may_skip_resume(dev)) + return 0; + + ret = acpi_dev_resume(dev); return ret ? ret : pm_generic_resume_early(dev); } diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index 75d7cdb4de9c..25b0302188d8 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -565,12 +565,22 @@ static void dpm_watchdog_clear(struct dpm_watchdog *wd) * dev_pm_may_skip_resume - System-wide device resume optimization check. * @dev: Target device. * - * Checks whether or not the device may be left in suspend after a system-wide - * transition to the working state. + * Return: + * - %false if the transition under way is RESTORE. + * - The return value of dev_pm_smart_suspend_and_suspended() 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_may_skip_resume(struct device *dev) { - return !dev->power.must_resume && pm_transition.event != PM_EVENT_RESTORE; + if (pm_transition.event == PM_EVENT_RESTORE) + return false; + + if (pm_transition.event == PM_EVENT_THAW) + return dev_pm_smart_suspend_and_suspended(dev); + + return !dev->power.must_resume; } /** @@ -601,6 +611,22 @@ static int device_resume_noirq(struct device *dev, pm_message_t state, bool asyn if (!dpm_wait_for_superior(dev, async)) goto Out; + skip_resume = dev_pm_may_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", but do that only if DPM_FLAG_SMART_SUSPEND is set + * to avoid confusing drivers that don't use it. + */ + if (skip_resume) + pm_runtime_set_suspended(dev); + else if (dev_pm_smart_suspend_and_suspended(dev)) + pm_runtime_set_active(dev); + if (dev->pm_domain) { info = "noirq power domain "; callback = pm_noirq_op(&dev->pm_domain->ops, state); @@ -614,35 +640,12 @@ static int device_resume_noirq(struct device *dev, pm_message_t state, bool asyn info = "noirq bus "; callback = pm_noirq_op(dev->bus->pm, state); } - if (callback) { - skip_resume = false; + if (callback) goto Run; - } - skip_resume = dev_pm_may_skip_resume(dev); if (skip_resume) goto Skip; - /* - * If "freeze" driver callbacks have been skipped during hibernation, - * because the device was runtime-suspended in __device_suspend_late(), - * the corresponding "thaw" callbacks must be skipped too, because - * running them for a runtime-suspended device may not be valid. - */ - if (dev_pm_smart_suspend_and_suspended(dev) && - state.event == PM_EVENT_THAW) { - skip_resume = true; - goto Skip; - } - - /* - * The device is going to be resumed, so set its PM-runtime status to - * "active", but do that only if DPM_FLAG_SMART_SUSPEND is set to avoid - * confusing drivers that don't use it. - */ - if (dev_pm_smart_suspend_and_suspended(dev)) - pm_runtime_set_active(dev); - if (dev->driver && dev->driver->pm) { info = "noirq driver "; callback = pm_noirq_op(dev->driver->pm, state); @@ -654,20 +657,6 @@ Run: Skip: dev->power.is_noirq_suspended = false; - if (skip_resume) { - /* Make the next phases of resume skip the device. */ - dev->power.is_late_suspended = false; - dev->power.is_suspended = false; - /* - * The device is going to be left in suspend, but it might not - * have been in runtime suspend before the system suspended, so - * its runtime PM status needs to be updated to avoid confusing - * the runtime PM framework when runtime PM is enabled for the - * device again. - */ - pm_runtime_set_suspended(dev); - } - Out: complete_all(&dev->power.completion); TRACE_RESUME(error); @@ -804,15 +793,25 @@ static int device_resume_early(struct device *dev, pm_message_t state, bool asyn } else if (dev->bus && dev->bus->pm) { info = "early bus "; callback = pm_late_early_op(dev->bus->pm, state); - } else if (dev->driver && dev->driver->pm) { + } + if (callback) + goto Run; + + if (dev_pm_may_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); + +Skip: dev->power.is_late_suspended = false; - Out: +Out: TRACE_RESUME(error); pm_runtime_enable(dev); diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index 0454ca0e4e3f..685fbf044911 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -896,14 +896,6 @@ static int pci_pm_resume_noirq(struct device *dev) if (dev_pm_may_skip_resume(dev)) return 0; - /* - * Devices with DPM_FLAG_SMART_SUSPEND may be left in runtime suspend - * during system suspend, so update their runtime PM status to "active" - * as they are going to be put into D0 shortly. - */ - if (dev_pm_smart_suspend_and_suspended(dev)) - pm_runtime_set_active(dev); - /* * In the suspend-to-idle case, devices left in D0 during suspend will * stay in D0, so it is not necessary to restore or update their @@ -928,6 +920,14 @@ static int pci_pm_resume_noirq(struct device *dev) return 0; } +static int pci_pm_resume_early(struct device *dev) +{ + if (dev_pm_may_skip_resume(dev)) + return 0; + + return pm_generic_resume_early(dev); +} + static int pci_pm_resume(struct device *dev) { struct pci_dev *pci_dev = to_pci_dev(dev); @@ -961,6 +961,7 @@ static int pci_pm_resume(struct device *dev) #define pci_pm_suspend_late NULL #define pci_pm_suspend_noirq NULL #define pci_pm_resume NULL +#define pci_pm_resume_early NULL #define pci_pm_resume_noirq NULL #endif /* !CONFIG_SUSPEND */ @@ -1358,6 +1359,7 @@ static const struct dev_pm_ops pci_dev_pm_ops = { .suspend = pci_pm_suspend, .suspend_late = pci_pm_suspend_late, .resume = pci_pm_resume, + .resume_early = pci_pm_resume_early, .freeze = pci_pm_freeze, .thaw = pci_pm_thaw, .poweroff = pci_pm_poweroff, -- cgit v1.2.3 From 0fe8a1be599ab97f840ba22d98cb8f24a9f9e872 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sat, 18 Apr 2020 18:52:19 +0200 Subject: PM: sleep: core: Rework the power.may_skip_resume handling Because the power.may_skip_resume device status bit is taken into account in combination with the DPM_FLAG_LEAVE_SUSPENDED driver flag, it can be set to 'true' for all devices in the "suspend" phase of a suspend-resume cycle, so do that. Then, neither the PM core nor the middle-layer (sybsystem) code handling it needs to set it to 'true' any more and it just has to be cleared if there is a reason to avoid skipping the "noirq" and "early" resume callbacks provided by the driver, so update the code in question accordingly. Suggested-by: Alan Stern Signed-off-by: Rafael J. Wysocki Acked-by: Alan Stern Acked-by: Bjorn Helgaas --- drivers/acpi/device_pm.c | 8 +++----- drivers/base/power/main.c | 10 ++-------- drivers/pci/pci-driver.c | 8 +++----- 3 files changed, 8 insertions(+), 18 deletions(-) diff --git a/drivers/acpi/device_pm.c b/drivers/acpi/device_pm.c index 399684085f85..1b02d7dc7d34 100644 --- a/drivers/acpi/device_pm.c +++ b/drivers/acpi/device_pm.c @@ -1100,10 +1100,8 @@ int acpi_subsys_suspend_noirq(struct device *dev) { int ret; - if (dev_pm_smart_suspend_and_suspended(dev)) { - dev->power.may_skip_resume = true; + if (dev_pm_smart_suspend_and_suspended(dev)) return 0; - } ret = pm_generic_suspend_noirq(dev); if (ret) @@ -1116,8 +1114,8 @@ int acpi_subsys_suspend_noirq(struct device *dev) * acpi_subsys_complete() to take care of fixing up the device's state * anyway, if need be. */ - dev->power.may_skip_resume = device_may_wakeup(dev) || - !device_can_wakeup(dev); + if (device_can_wakeup(dev) && !device_may_wakeup(dev)) + dev->power.may_skip_resume = false; return 0; } diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index 25b0302188d8..5adf0be6aa47 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -1415,14 +1415,8 @@ static int __device_suspend_late(struct device *dev, pm_message_t state, bool as if (callback) goto Run; - if (dev_pm_smart_suspend_and_suspended(dev)) { - /* - * In principle, the resume of the device may be skippend if it - * remains in runtime suspend at this point. - */ - dev->power.may_skip_resume = true; + if (dev_pm_smart_suspend_and_suspended(dev)) goto Skip; - } if (dev->driver && dev->driver->pm) { info = "late driver "; @@ -1647,7 +1641,7 @@ static int __device_suspend(struct device *dev, pm_message_t state, bool async) dev->power.direct_complete = false; } - dev->power.may_skip_resume = false; + dev->power.may_skip_resume = true; dev->power.must_resume = false; dpm_watchdog_set(&wd, dev); diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index 685fbf044911..ce220b1987df 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -789,10 +789,8 @@ static int pci_pm_suspend_noirq(struct device *dev) struct pci_dev *pci_dev = to_pci_dev(dev); const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; - if (dev_pm_smart_suspend_and_suspended(dev)) { - dev->power.may_skip_resume = true; + if (dev_pm_smart_suspend_and_suspended(dev)) return 0; - } if (pci_has_legacy_pm_support(pci_dev)) return pci_legacy_suspend_late(dev, PMSG_SUSPEND); @@ -880,8 +878,8 @@ Fixup: * pci_pm_complete() to take care of fixing up the device's state * anyway, if need be. */ - dev->power.may_skip_resume = device_may_wakeup(dev) || - !device_can_wakeup(dev); + if (device_can_wakeup(dev) && !device_may_wakeup(dev)) + dev->power.may_skip_resume = false; return 0; } -- cgit v1.2.3 From 76c70cb58ce30264af4b714109ee756da25d830a Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sat, 18 Apr 2020 18:52:30 +0200 Subject: PM: sleep: core: Rename dev_pm_may_skip_resume() The name of dev_pm_may_skip_resume() may be easily confused with the power.may_skip_resume flag which is not checked by that function, so rename the former as dev_pm_skip_resume(). No functional impact. Suggested-by: Alan Stern Signed-off-by: Rafael J. Wysocki Acked-by: Alan Stern Acked-by: Bjorn Helgaas --- Documentation/power/pci.rst | 2 +- drivers/acpi/acpi_lpss.c | 4 ++-- drivers/acpi/device_pm.c | 4 ++-- drivers/base/power/main.c | 8 ++++---- drivers/pci/pci-driver.c | 4 ++-- include/linux/pm.h | 2 +- 6 files changed, 12 insertions(+), 12 deletions(-) diff --git a/Documentation/power/pci.rst b/Documentation/power/pci.rst index a39b2461919a..aa1c7fce6cd0 100644 --- a/Documentation/power/pci.rst +++ b/Documentation/power/pci.rst @@ -1034,7 +1034,7 @@ device to be left in suspend after system-wide transitions to the working state. This flag is checked by the PM core, but the PCI bus type informs the PM core which devices may be left in suspend from its perspective (that happens during the "noirq" phase of system-wide suspend and analogous transitions) and next it -uses the dev_pm_may_skip_resume() helper to decide whether or not to return from +uses the dev_pm_skip_resume() helper to decide whether or not to return from pci_pm_resume_noirq() and pci_pm_resume_early() upfront. 3.2. Device Runtime Power Management diff --git a/drivers/acpi/acpi_lpss.c b/drivers/acpi/acpi_lpss.c index c4a84df6cc98..7632df1a5be3 100644 --- a/drivers/acpi/acpi_lpss.c +++ b/drivers/acpi/acpi_lpss.c @@ -1093,7 +1093,7 @@ static int acpi_lpss_resume_early(struct device *dev) if (pdata->dev_desc->resume_from_noirq) return 0; - if (dev_pm_may_skip_resume(dev)) + if (dev_pm_skip_resume(dev)) return 0; return acpi_lpss_do_resume_early(dev); @@ -1105,7 +1105,7 @@ static int acpi_lpss_resume_noirq(struct device *dev) int ret; /* Follow acpi_subsys_resume_noirq(). */ - if (dev_pm_may_skip_resume(dev)) + if (dev_pm_skip_resume(dev)) return 0; ret = pm_generic_resume_noirq(dev); diff --git a/drivers/acpi/device_pm.c b/drivers/acpi/device_pm.c index 1b02d7dc7d34..8c2a091728a9 100644 --- a/drivers/acpi/device_pm.c +++ b/drivers/acpi/device_pm.c @@ -1127,7 +1127,7 @@ EXPORT_SYMBOL_GPL(acpi_subsys_suspend_noirq); */ static int acpi_subsys_resume_noirq(struct device *dev) { - if (dev_pm_may_skip_resume(dev)) + if (dev_pm_skip_resume(dev)) return 0; return pm_generic_resume_noirq(dev); @@ -1145,7 +1145,7 @@ static int acpi_subsys_resume_early(struct device *dev) { int ret; - if (dev_pm_may_skip_resume(dev)) + if (dev_pm_skip_resume(dev)) return 0; ret = acpi_dev_resume(dev); diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index 5adf0be6aa47..f98eced0f200 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -562,7 +562,7 @@ static void dpm_watchdog_clear(struct dpm_watchdog *wd) /*------------------------- Resume routines -------------------------*/ /** - * dev_pm_may_skip_resume - System-wide device resume optimization check. + * dev_pm_skip_resume - System-wide device resume optimization check. * @dev: Target device. * * Return: @@ -572,7 +572,7 @@ static void dpm_watchdog_clear(struct dpm_watchdog *wd) * - The logical negation of %power.must_resume otherwise (that is, when the * transition under way is RESUME). */ -bool dev_pm_may_skip_resume(struct device *dev) +bool dev_pm_skip_resume(struct device *dev) { if (pm_transition.event == PM_EVENT_RESTORE) return false; @@ -611,7 +611,7 @@ static int device_resume_noirq(struct device *dev, pm_message_t state, bool asyn if (!dpm_wait_for_superior(dev, async)) goto Out; - skip_resume = dev_pm_may_skip_resume(dev); + 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 @@ -797,7 +797,7 @@ static int device_resume_early(struct device *dev, pm_message_t state, bool asyn if (callback) goto Run; - if (dev_pm_may_skip_resume(dev)) + if (dev_pm_skip_resume(dev)) goto Skip; if (dev->driver && dev->driver->pm) { diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index ce220b1987df..decf82595340 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -891,7 +891,7 @@ static int pci_pm_resume_noirq(struct device *dev) pci_power_t prev_state = pci_dev->current_state; bool skip_bus_pm = pci_dev->skip_bus_pm; - if (dev_pm_may_skip_resume(dev)) + if (dev_pm_skip_resume(dev)) return 0; /* @@ -920,7 +920,7 @@ static int pci_pm_resume_noirq(struct device *dev) static int pci_pm_resume_early(struct device *dev) { - if (dev_pm_may_skip_resume(dev)) + if (dev_pm_skip_resume(dev)) return 0; return pm_generic_resume_early(dev); diff --git a/include/linux/pm.h b/include/linux/pm.h index e057d1fa2469..d89b7099f241 100644 --- a/include/linux/pm.h +++ b/include/linux/pm.h @@ -758,7 +758,7 @@ extern int pm_generic_poweroff_late(struct device *dev); extern int pm_generic_poweroff(struct device *dev); extern void pm_generic_complete(struct device *dev); -extern bool dev_pm_may_skip_resume(struct device *dev); +extern bool dev_pm_skip_resume(struct device *dev); extern bool dev_pm_smart_suspend_and_suspended(struct device *dev); #else /* !CONFIG_PM_SLEEP */ -- cgit v1.2.3 From fa2bfead910322e44e7e0bb74364ac198a2abd32 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sat, 18 Apr 2020 18:52:48 +0200 Subject: PM: sleep: core: Rename dev_pm_smart_suspend_and_suspended() Because all callers of dev_pm_smart_suspend_and_suspended use it only for checking whether or not to skip driver suspend callbacks for a device, rename it to dev_pm_skip_suspend() in analogy with dev_pm_skip_resume(). No functional impact. Suggested-by: Alan Stern Signed-off-by: Rafael J. Wysocki Acked-by: Alan Stern Acked-by: Bjorn Helgaas --- drivers/acpi/acpi_lpss.c | 6 +++--- drivers/acpi/device_pm.c | 8 ++++---- drivers/base/power/main.c | 13 ++++++------- drivers/pci/hotplug/pciehp_core.c | 2 +- drivers/pci/pci-driver.c | 8 ++++---- include/linux/pm.h | 2 +- 6 files changed, 19 insertions(+), 20 deletions(-) diff --git a/drivers/acpi/acpi_lpss.c b/drivers/acpi/acpi_lpss.c index 7632df1a5be3..5e2bfbcf526f 100644 --- a/drivers/acpi/acpi_lpss.c +++ b/drivers/acpi/acpi_lpss.c @@ -1041,7 +1041,7 @@ static int acpi_lpss_do_suspend_late(struct device *dev) { int ret; - if (dev_pm_smart_suspend_and_suspended(dev)) + if (dev_pm_skip_suspend(dev)) return 0; ret = pm_generic_suspend_late(dev); @@ -1169,7 +1169,7 @@ static int acpi_lpss_poweroff_late(struct device *dev) { struct lpss_private_data *pdata = acpi_driver_data(ACPI_COMPANION(dev)); - if (dev_pm_smart_suspend_and_suspended(dev)) + if (dev_pm_skip_suspend(dev)) return 0; if (pdata->dev_desc->resume_from_noirq) @@ -1182,7 +1182,7 @@ static int acpi_lpss_poweroff_noirq(struct device *dev) { struct lpss_private_data *pdata = acpi_driver_data(ACPI_COMPANION(dev)); - if (dev_pm_smart_suspend_and_suspended(dev)) + if (dev_pm_skip_suspend(dev)) return 0; if (pdata->dev_desc->resume_from_noirq) { diff --git a/drivers/acpi/device_pm.c b/drivers/acpi/device_pm.c index 8c2a091728a9..ae234d731d42 100644 --- a/drivers/acpi/device_pm.c +++ b/drivers/acpi/device_pm.c @@ -1084,7 +1084,7 @@ int acpi_subsys_suspend_late(struct device *dev) { int ret; - if (dev_pm_smart_suspend_and_suspended(dev)) + if (dev_pm_skip_suspend(dev)) return 0; ret = pm_generic_suspend_late(dev); @@ -1100,7 +1100,7 @@ int acpi_subsys_suspend_noirq(struct device *dev) { int ret; - if (dev_pm_smart_suspend_and_suspended(dev)) + if (dev_pm_skip_suspend(dev)) return 0; ret = pm_generic_suspend_noirq(dev); @@ -1213,7 +1213,7 @@ static int acpi_subsys_poweroff_late(struct device *dev) { int ret; - if (dev_pm_smart_suspend_and_suspended(dev)) + if (dev_pm_skip_suspend(dev)) return 0; ret = pm_generic_poweroff_late(dev); @@ -1229,7 +1229,7 @@ static int acpi_subsys_poweroff_late(struct device *dev) */ static int acpi_subsys_poweroff_noirq(struct device *dev) { - if (dev_pm_smart_suspend_and_suspended(dev)) + if (dev_pm_skip_suspend(dev)) return 0; return pm_generic_poweroff_noirq(dev); diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index f98eced0f200..3170d93e29f9 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -567,8 +567,7 @@ static void dpm_watchdog_clear(struct dpm_watchdog *wd) * * Return: * - %false if the transition under way is RESTORE. - * - The return value of dev_pm_smart_suspend_and_suspended() if the transition - * under way is THAW. + * - 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). */ @@ -578,7 +577,7 @@ bool dev_pm_skip_resume(struct device *dev) return false; if (pm_transition.event == PM_EVENT_THAW) - return dev_pm_smart_suspend_and_suspended(dev); + return dev_pm_skip_suspend(dev); return !dev->power.must_resume; } @@ -624,7 +623,7 @@ static int device_resume_noirq(struct device *dev, pm_message_t state, bool asyn */ if (skip_resume) pm_runtime_set_suspended(dev); - else if (dev_pm_smart_suspend_and_suspended(dev)) + else if (dev_pm_skip_suspend(dev)) pm_runtime_set_active(dev); if (dev->pm_domain) { @@ -1223,7 +1222,7 @@ static int __device_suspend_noirq(struct device *dev, pm_message_t state, bool a if (callback) goto Run; - if (dev_pm_smart_suspend_and_suspended(dev)) + if (dev_pm_skip_suspend(dev)) goto Skip; if (dev->driver && dev->driver->pm) { @@ -1415,7 +1414,7 @@ static int __device_suspend_late(struct device *dev, pm_message_t state, bool as if (callback) goto Run; - if (dev_pm_smart_suspend_and_suspended(dev)) + if (dev_pm_skip_suspend(dev)) goto Skip; if (dev->driver && dev->driver->pm) { @@ -2003,7 +2002,7 @@ void device_pm_check_callbacks(struct device *dev) spin_unlock_irq(&dev->power.lock); } -bool dev_pm_smart_suspend_and_suspended(struct device *dev) +bool dev_pm_skip_suspend(struct device *dev) { return dev_pm_test_driver_flags(dev, DPM_FLAG_SMART_SUSPEND) && pm_runtime_status_suspended(dev); diff --git a/drivers/pci/hotplug/pciehp_core.c b/drivers/pci/hotplug/pciehp_core.c index 312cc45c44c7..bf779f291f15 100644 --- a/drivers/pci/hotplug/pciehp_core.c +++ b/drivers/pci/hotplug/pciehp_core.c @@ -275,7 +275,7 @@ static int pciehp_suspend(struct pcie_device *dev) * If the port is already runtime suspended we can keep it that * way. */ - if (dev_pm_smart_suspend_and_suspended(&dev->port->dev)) + if (dev_pm_skip_suspend(&dev->port->dev)) return 0; pciehp_disable_interrupt(dev); diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index decf82595340..da6510af1221 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -776,7 +776,7 @@ static int pci_pm_suspend(struct device *dev) static int pci_pm_suspend_late(struct device *dev) { - if (dev_pm_smart_suspend_and_suspended(dev)) + if (dev_pm_skip_suspend(dev)) return 0; pci_fixup_device(pci_fixup_suspend, to_pci_dev(dev)); @@ -789,7 +789,7 @@ static int pci_pm_suspend_noirq(struct device *dev) struct pci_dev *pci_dev = to_pci_dev(dev); const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; - if (dev_pm_smart_suspend_and_suspended(dev)) + if (dev_pm_skip_suspend(dev)) return 0; if (pci_has_legacy_pm_support(pci_dev)) @@ -1126,7 +1126,7 @@ static int pci_pm_poweroff(struct device *dev) static int pci_pm_poweroff_late(struct device *dev) { - if (dev_pm_smart_suspend_and_suspended(dev)) + if (dev_pm_skip_suspend(dev)) return 0; pci_fixup_device(pci_fixup_suspend, to_pci_dev(dev)); @@ -1139,7 +1139,7 @@ static int pci_pm_poweroff_noirq(struct device *dev) struct pci_dev *pci_dev = to_pci_dev(dev); const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; - if (dev_pm_smart_suspend_and_suspended(dev)) + if (dev_pm_skip_suspend(dev)) return 0; if (pci_has_legacy_pm_support(pci_dev)) diff --git a/include/linux/pm.h b/include/linux/pm.h index d89b7099f241..8c59a7f0bcf4 100644 --- a/include/linux/pm.h +++ b/include/linux/pm.h @@ -759,7 +759,7 @@ extern int pm_generic_poweroff(struct device *dev); extern void pm_generic_complete(struct device *dev); extern bool dev_pm_skip_resume(struct device *dev); -extern bool dev_pm_smart_suspend_and_suspended(struct device *dev); +extern bool dev_pm_skip_suspend(struct device *dev); #else /* !CONFIG_PM_SLEEP */ -- cgit v1.2.3 From e07515563d010d8b32967634e8dc2fdc732c1aa6 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sat, 18 Apr 2020 18:53:01 +0200 Subject: PM: sleep: core: Rename DPM_FLAG_NEVER_SKIP Rename DPM_FLAG_NEVER_SKIP to DPM_FLAG_NO_DIRECT_COMPLETE which matches its purpose more closely. No functional impact. Suggested-by: Alan Stern Signed-off-by: Rafael J. Wysocki Acked-by: Bjorn Helgaas # for PCI parts Acked-by: Jeff Kirsher Acked-by: Alan Stern Acked-by: Bjorn Helgaas Acked-by: Alex Deucher --- Documentation/driver-api/pm/devices.rst | 6 +++--- Documentation/power/pci.rst | 10 +++++----- drivers/base/power/main.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 2 +- drivers/gpu/drm/i915/intel_runtime_pm.c | 2 +- drivers/gpu/drm/radeon/radeon_kms.c | 2 +- drivers/misc/mei/pci-me.c | 2 +- drivers/misc/mei/pci-txe.c | 2 +- drivers/net/ethernet/intel/e1000e/netdev.c | 2 +- drivers/net/ethernet/intel/igb/igb_main.c | 2 +- drivers/net/ethernet/intel/igc/igc_main.c | 2 +- drivers/pci/pcie/portdrv_pci.c | 2 +- include/linux/pm.h | 6 +++--- 13 files changed, 21 insertions(+), 21 deletions(-) diff --git a/Documentation/driver-api/pm/devices.rst b/Documentation/driver-api/pm/devices.rst index f66c7b9126ea..4ace0eba4506 100644 --- a/Documentation/driver-api/pm/devices.rst +++ b/Documentation/driver-api/pm/devices.rst @@ -361,9 +361,9 @@ the phases are: ``prepare``, ``suspend``, ``suspend_late``, ``suspend_noirq``. runtime PM disabled. This feature also can be controlled by device drivers by using the - ``DPM_FLAG_NEVER_SKIP`` and ``DPM_FLAG_SMART_PREPARE`` driver power - management flags. [Typically, they are set at the time the driver is - probed against the device in question by passing them to the + ``DPM_FLAG_NO_DIRECT_COMPLETE`` and ``DPM_FLAG_SMART_PREPARE`` driver + power management flags. [Typically, they are set at the time the driver + is probed against the device in question by passing them to the :c:func:`dev_pm_set_driver_flags` helper function.] If the first of these flags is set, the PM core will not apply the direct-complete procedure described above to the given device and, consequenty, to any diff --git a/Documentation/power/pci.rst b/Documentation/power/pci.rst index aa1c7fce6cd0..9e1408121bea 100644 --- a/Documentation/power/pci.rst +++ b/Documentation/power/pci.rst @@ -1004,11 +1004,11 @@ including the PCI bus type. The flags should be set once at the driver probe time with the help of the dev_pm_set_driver_flags() function and they should not be updated directly afterwards. -The DPM_FLAG_NEVER_SKIP flag prevents the PM core from using the direct-complete -mechanism allowing device suspend/resume callbacks to be skipped if the device -is in runtime suspend when the system suspend starts. That also affects all of -the ancestors of the device, so this flag should only be used if absolutely -necessary. +The DPM_FLAG_NO_DIRECT_COMPLETE flag prevents the PM core from using the +direct-complete mechanism allowing device suspend/resume callbacks to be skipped +if the device is in runtime suspend when the system suspend starts. That also +affects all of the ancestors of the device, so this flag should only be used if +absolutely necessary. The DPM_FLAG_SMART_PREPARE flag instructs the PCI bus type to only return a positive value from pci_pm_prepare() if the ->prepare callback provided by the diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index 3170d93e29f9..dbc1e5e7346b 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -1844,7 +1844,7 @@ unlock: spin_lock_irq(&dev->power.lock); dev->power.direct_complete = state.event == PM_EVENT_SUSPEND && (ret > 0 || dev->power.no_pm_callbacks) && - !dev_pm_test_driver_flags(dev, DPM_FLAG_NEVER_SKIP); + !dev_pm_test_driver_flags(dev, DPM_FLAG_NO_DIRECT_COMPLETE); spin_unlock_irq(&dev->power.lock); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index fd1dc3236eca..a9086ea1ab60 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -191,7 +191,7 @@ int amdgpu_driver_load_kms(struct drm_device *dev, unsigned long flags) } if (adev->runpm) { - dev_pm_set_driver_flags(dev->dev, DPM_FLAG_NEVER_SKIP); + dev_pm_set_driver_flags(dev->dev, DPM_FLAG_NO_DIRECT_COMPLETE); pm_runtime_use_autosuspend(dev->dev); pm_runtime_set_autosuspend_delay(dev->dev, 5000); pm_runtime_set_active(dev->dev); diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c index ad719c9602af..9cb2d7548daa 100644 --- a/drivers/gpu/drm/i915/intel_runtime_pm.c +++ b/drivers/gpu/drm/i915/intel_runtime_pm.c @@ -549,7 +549,7 @@ void intel_runtime_pm_enable(struct intel_runtime_pm *rpm) * becaue the HDA driver may require us to enable the audio power * domain during system suspend. */ - dev_pm_set_driver_flags(kdev, DPM_FLAG_NEVER_SKIP); + dev_pm_set_driver_flags(kdev, DPM_FLAG_NO_DIRECT_COMPLETE); pm_runtime_set_autosuspend_delay(kdev, 10000); /* 10s */ pm_runtime_mark_last_busy(kdev); diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c index 58176db85952..372962358a18 100644 --- a/drivers/gpu/drm/radeon/radeon_kms.c +++ b/drivers/gpu/drm/radeon/radeon_kms.c @@ -158,7 +158,7 @@ int radeon_driver_load_kms(struct drm_device *dev, unsigned long flags) } if (radeon_is_px(dev)) { - dev_pm_set_driver_flags(dev->dev, DPM_FLAG_NEVER_SKIP); + dev_pm_set_driver_flags(dev->dev, DPM_FLAG_NO_DIRECT_COMPLETE); pm_runtime_use_autosuspend(dev->dev); pm_runtime_set_autosuspend_delay(dev->dev, 5000); pm_runtime_set_active(dev->dev); diff --git a/drivers/misc/mei/pci-me.c b/drivers/misc/mei/pci-me.c index 3d21c38e2dbb..53f16f3bd091 100644 --- a/drivers/misc/mei/pci-me.c +++ b/drivers/misc/mei/pci-me.c @@ -240,7 +240,7 @@ static int mei_me_probe(struct pci_dev *pdev, const struct pci_device_id *ent) * MEI requires to resume from runtime suspend mode * in order to perform link reset flow upon system suspend. */ - dev_pm_set_driver_flags(&pdev->dev, DPM_FLAG_NEVER_SKIP); + dev_pm_set_driver_flags(&pdev->dev, DPM_FLAG_NO_DIRECT_COMPLETE); /* * ME maps runtime suspend/resume to D0i states, diff --git a/drivers/misc/mei/pci-txe.c b/drivers/misc/mei/pci-txe.c index beacf2a2f2b5..4bf26ce61044 100644 --- a/drivers/misc/mei/pci-txe.c +++ b/drivers/misc/mei/pci-txe.c @@ -128,7 +128,7 @@ static int mei_txe_probe(struct pci_dev *pdev, const struct pci_device_id *ent) * MEI requires to resume from runtime suspend mode * in order to perform link reset flow upon system suspend. */ - dev_pm_set_driver_flags(&pdev->dev, DPM_FLAG_NEVER_SKIP); + dev_pm_set_driver_flags(&pdev->dev, DPM_FLAG_NO_DIRECT_COMPLETE); /* * TXE maps runtime suspend/resume to own power gating states, diff --git a/drivers/net/ethernet/intel/e1000e/netdev.c b/drivers/net/ethernet/intel/e1000e/netdev.c index 177c6da80c57..2730b1c7dddb 100644 --- a/drivers/net/ethernet/intel/e1000e/netdev.c +++ b/drivers/net/ethernet/intel/e1000e/netdev.c @@ -7549,7 +7549,7 @@ static int e1000_probe(struct pci_dev *pdev, const struct pci_device_id *ent) e1000_print_device_info(adapter); - dev_pm_set_driver_flags(&pdev->dev, DPM_FLAG_NEVER_SKIP); + dev_pm_set_driver_flags(&pdev->dev, DPM_FLAG_NO_DIRECT_COMPLETE); if (pci_dev_run_wake(pdev) && hw->mac.type < e1000_pch_cnp) pm_runtime_put_noidle(&pdev->dev); diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c index b46bff8fe056..8bb3db2cbd41 100644 --- a/drivers/net/ethernet/intel/igb/igb_main.c +++ b/drivers/net/ethernet/intel/igb/igb_main.c @@ -3445,7 +3445,7 @@ static int igb_probe(struct pci_dev *pdev, const struct pci_device_id *ent) } } - dev_pm_set_driver_flags(&pdev->dev, DPM_FLAG_NEVER_SKIP); + dev_pm_set_driver_flags(&pdev->dev, DPM_FLAG_NO_DIRECT_COMPLETE); pm_runtime_put_noidle(&pdev->dev); return 0; diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c index 69fa1ce1f927..59fc0097438f 100644 --- a/drivers/net/ethernet/intel/igc/igc_main.c +++ b/drivers/net/ethernet/intel/igc/igc_main.c @@ -4825,7 +4825,7 @@ static int igc_probe(struct pci_dev *pdev, pcie_print_link_status(pdev); netdev_info(netdev, "MAC: %pM\n", netdev->dev_addr); - dev_pm_set_driver_flags(&pdev->dev, DPM_FLAG_NEVER_SKIP); + dev_pm_set_driver_flags(&pdev->dev, DPM_FLAG_NO_DIRECT_COMPLETE); pm_runtime_put_noidle(&pdev->dev); diff --git a/drivers/pci/pcie/portdrv_pci.c b/drivers/pci/pcie/portdrv_pci.c index 160d67c59310..3acf151ae015 100644 --- a/drivers/pci/pcie/portdrv_pci.c +++ b/drivers/pci/pcie/portdrv_pci.c @@ -115,7 +115,7 @@ static int pcie_portdrv_probe(struct pci_dev *dev, pci_save_state(dev); - dev_pm_set_driver_flags(&dev->dev, DPM_FLAG_NEVER_SKIP | + dev_pm_set_driver_flags(&dev->dev, DPM_FLAG_NO_DIRECT_COMPLETE | DPM_FLAG_SMART_SUSPEND); if (pci_bridge_d3_possible(dev)) { diff --git a/include/linux/pm.h b/include/linux/pm.h index 8c59a7f0bcf4..cdb8fbd6ab18 100644 --- a/include/linux/pm.h +++ b/include/linux/pm.h @@ -544,7 +544,7 @@ struct pm_subsys_data { * These flags can be set by device drivers at the probe time. They need not be * cleared by the drivers as the driver core will take care of that. * - * NEVER_SKIP: Do not skip all system suspend/resume callbacks for the device. + * NO_DIRECT_COMPLETE: Do not apply direct-complete optimization to the device. * SMART_PREPARE: Check the return value of the driver's ->prepare callback. * SMART_SUSPEND: No need to resume the device from runtime suspend. * LEAVE_SUSPENDED: Avoid resuming the device during system resume if possible. @@ -554,7 +554,7 @@ struct pm_subsys_data { * their ->prepare callbacks if the driver's ->prepare callback returns 0 (in * other words, the system suspend/resume callbacks can only be skipped for the * device if its driver doesn't object against that). This flag has no effect - * if NEVER_SKIP is set. + * if NO_DIRECT_COMPLETE is set. * * Setting SMART_SUSPEND instructs bus types and PM domains which may want to * runtime resume the device upfront during system suspend that doing so is not @@ -565,7 +565,7 @@ struct pm_subsys_data { * Setting LEAVE_SUSPENDED informs the PM core and middle-layer code that the * driver prefers the device to be left in suspend after system resume. */ -#define DPM_FLAG_NEVER_SKIP BIT(0) +#define DPM_FLAG_NO_DIRECT_COMPLETE BIT(0) #define DPM_FLAG_SMART_PREPARE BIT(1) #define DPM_FLAG_SMART_SUSPEND BIT(2) #define DPM_FLAG_LEAVE_SUSPENDED BIT(3) -- cgit v1.2.3 From 2a3f34750b8b07df42ab4b30b70e029d46e0d7f3 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sat, 18 Apr 2020 18:53:20 +0200 Subject: PM: sleep: core: Rename DPM_FLAG_LEAVE_SUSPENDED Rename DPM_FLAG_LEAVE_SUSPENDED to DPM_FLAG_MAY_SKIP_RESUME which matches its purpose more closely. No functional impact. Suggested-by: Alan Stern Signed-off-by: Rafael J. Wysocki Acked-by: Wolfram Sang # for I2C Acked-by: Alan Stern Acked-by: Bjorn Helgaas --- Documentation/driver-api/pm/devices.rst | 4 ++-- Documentation/power/pci.rst | 2 +- drivers/acpi/acpi_tad.c | 2 +- drivers/base/power/main.c | 2 +- drivers/i2c/busses/i2c-designware-platdrv.c | 4 ++-- include/linux/pm.h | 6 +++--- 6 files changed, 10 insertions(+), 10 deletions(-) diff --git a/Documentation/driver-api/pm/devices.rst b/Documentation/driver-api/pm/devices.rst index 4ace0eba4506..f342c7549b4c 100644 --- a/Documentation/driver-api/pm/devices.rst +++ b/Documentation/driver-api/pm/devices.rst @@ -803,7 +803,7 @@ general.] However, it often is desirable to leave devices in suspend after system transitions to the working state, especially if those devices had been in runtime suspend before the preceding system-wide suspend (or analogous) -transition. Device drivers can use the ``DPM_FLAG_LEAVE_SUSPENDED`` flag to +transition. Device drivers can use the ``DPM_FLAG_MAY_SKIP_RESUME`` flag to indicate to the PM core (and middle-layer code) that they prefer the specific devices handled by them to be left suspended and they have no problems with skipping their system-wide resume callbacks for this reason. Whether or not the @@ -825,7 +825,7 @@ device really can be left in suspend. For devices whose "noirq", "late" and "early" driver callbacks are invoked directly by the PM core, all of the system-wide resume callbacks are skipped if -``DPM_FLAG_LEAVE_SUSPENDED`` is set and the device is in runtime suspend during +``DPM_FLAG_MAY_SKIP_RESUME`` is set and the device is in runtime suspend during the ``suspend_noirq`` (or analogous) phase or the transition under way is a proper system suspend (rather than anything related to hibernation) and the device's wakeup settings are suitable for runtime PM (that is, it cannot diff --git a/Documentation/power/pci.rst b/Documentation/power/pci.rst index 9e1408121bea..f09b382b4621 100644 --- a/Documentation/power/pci.rst +++ b/Documentation/power/pci.rst @@ -1029,7 +1029,7 @@ into D0 going forward), but if it is in runtime suspend in pci_pm_thaw_noirq(), the function will set the power.direct_complete flag for it (to make the PM core skip the subsequent "thaw" callbacks for it) and return. -Setting the DPM_FLAG_LEAVE_SUSPENDED flag means that the driver prefers the +Setting the DPM_FLAG_MAY_SKIP_RESUME flag means that the driver prefers the device to be left in suspend after system-wide transitions to the working state. This flag is checked by the PM core, but the PCI bus type informs the PM core which devices may be left in suspend from its perspective (that happens during diff --git a/drivers/acpi/acpi_tad.c b/drivers/acpi/acpi_tad.c index 33a4bcdaa4d7..7d45cce0c3c1 100644 --- a/drivers/acpi/acpi_tad.c +++ b/drivers/acpi/acpi_tad.c @@ -624,7 +624,7 @@ static int acpi_tad_probe(struct platform_device *pdev) */ device_init_wakeup(dev, true); dev_pm_set_driver_flags(dev, DPM_FLAG_SMART_SUSPEND | - DPM_FLAG_LEAVE_SUSPENDED); + DPM_FLAG_MAY_SKIP_RESUME); /* * The platform bus type layer tells the ACPI PM domain powers up the * device, so set the runtime PM status of it to "active". diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index dbc1e5e7346b..aaa4aaf41d27 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -1247,7 +1247,7 @@ Skip: * to be skipped. */ if (atomic_read(&dev->power.usage_count) > 1 || - !(dev_pm_test_driver_flags(dev, DPM_FLAG_LEAVE_SUSPENDED) && + !(dev_pm_test_driver_flags(dev, DPM_FLAG_MAY_SKIP_RESUME) && dev->power.may_skip_resume)) dev->power.must_resume = true; diff --git a/drivers/i2c/busses/i2c-designware-platdrv.c b/drivers/i2c/busses/i2c-designware-platdrv.c index 5536673060cc..c429d664f655 100644 --- a/drivers/i2c/busses/i2c-designware-platdrv.c +++ b/drivers/i2c/busses/i2c-designware-platdrv.c @@ -357,12 +357,12 @@ static int dw_i2c_plat_probe(struct platform_device *pdev) if (dev->flags & ACCESS_NO_IRQ_SUSPEND) { dev_pm_set_driver_flags(&pdev->dev, DPM_FLAG_SMART_PREPARE | - DPM_FLAG_LEAVE_SUSPENDED); + DPM_FLAG_MAY_SKIP_RESUME); } else { dev_pm_set_driver_flags(&pdev->dev, DPM_FLAG_SMART_PREPARE | DPM_FLAG_SMART_SUSPEND | - DPM_FLAG_LEAVE_SUSPENDED); + DPM_FLAG_MAY_SKIP_RESUME); } /* The code below assumes runtime PM to be disabled. */ diff --git a/include/linux/pm.h b/include/linux/pm.h index cdb8fbd6ab18..35796fc49e7a 100644 --- a/include/linux/pm.h +++ b/include/linux/pm.h @@ -547,7 +547,7 @@ struct pm_subsys_data { * NO_DIRECT_COMPLETE: Do not apply direct-complete optimization to the device. * SMART_PREPARE: Check the return value of the driver's ->prepare callback. * SMART_SUSPEND: No need to resume the device from runtime suspend. - * LEAVE_SUSPENDED: Avoid resuming the device during system resume if possible. + * MAY_SKIP_RESUME: Avoid resuming the device during system resume if possible. * * Setting SMART_PREPARE instructs bus types and PM domains which may want * system suspend/resume callbacks to be skipped for the device to return 0 from @@ -562,13 +562,13 @@ struct pm_subsys_data { * invocations of the ->suspend_late and ->suspend_noirq callbacks provided by * the driver if they decide to leave the device in runtime suspend. * - * Setting LEAVE_SUSPENDED informs the PM core and middle-layer code that the + * Setting MAY_SKIP_RESUME informs the PM core and middle-layer code that the * driver prefers the device to be left in suspend after system resume. */ #define DPM_FLAG_NO_DIRECT_COMPLETE BIT(0) #define DPM_FLAG_SMART_PREPARE BIT(1) #define DPM_FLAG_SMART_SUSPEND BIT(2) -#define DPM_FLAG_LEAVE_SUSPENDED BIT(3) +#define DPM_FLAG_MAY_SKIP_RESUME BIT(3) struct dev_pm_info { pm_message_t power_state; -- cgit v1.2.3 From 2fff3f73e8c27801b84d2315e1a49bce96b00eff Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sat, 18 Apr 2020 18:55:32 +0200 Subject: Documentation: PM: sleep: Update driver flags documentation Update the documentation of the driver flags for system-wide power management to reflect the current code flows and be more consistent. Signed-off-by: Rafael J. Wysocki --- Documentation/driver-api/pm/devices.rst | 139 +++++++++++++++++++++----------- Documentation/power/pci.rst | 43 +++++----- include/linux/pm.h | 24 ++---- 3 files changed, 119 insertions(+), 87 deletions(-) diff --git a/Documentation/driver-api/pm/devices.rst b/Documentation/driver-api/pm/devices.rst index f342c7549b4c..782cb37073a3 100644 --- a/Documentation/driver-api/pm/devices.rst +++ b/Documentation/driver-api/pm/devices.rst @@ -772,62 +772,107 @@ the state of devices (possibly except for resuming them from runtime suspend) from their ``->prepare`` and ``->suspend`` callbacks (or equivalent) *before* invoking device drivers' ``->suspend`` callbacks (or equivalent). +.. _smart_suspend_flag: + +The ``DPM_FLAG_SMART_SUSPEND`` Driver Flag +------------------------------------------ + Some bus types and PM domains have a policy to resume all devices from runtime suspend upfront in their ``->suspend`` callbacks, but that may not be really necessary if the driver of the device can cope with runtime-suspended devices. The driver can indicate that by setting ``DPM_FLAG_SMART_SUSPEND`` in -:c:member:`power.driver_flags` at the probe time, by passing it to the -:c:func:`dev_pm_set_driver_flags` helper. That also may cause middle-layer code +:c:member:`power.driver_flags` at the probe time with the help of the +:c:func:`dev_pm_set_driver_flags` helper routine. + +However, setting that flag also causes the PM core and middle-layer code (bus types, PM domains etc.) to skip the ``->suspend_late`` and ``->suspend_noirq`` callbacks provided by the driver if the device remains in -runtime suspend at the beginning of the ``suspend_late`` phase of system-wide -suspend (or in the ``poweroff_late`` phase of hibernation), when runtime PM -has been disabled for it, under the assumption that its state should not change -after that point until the system-wide transition is over (the PM core itself -does that for devices whose "noirq", "late" and "early" system-wide PM callbacks -are executed directly by it). If that happens, the driver's system-wide resume -callbacks, if present, may still be invoked during the subsequent system-wide -resume transition and the device's runtime power management status may be set -to "active" before enabling runtime PM for it, so the driver must be prepared to -cope with the invocation of its system-wide resume callbacks back-to-back with -its ``->runtime_suspend`` one (without the intervening ``->runtime_resume`` and -so on) and the final state of the device must reflect the "active" runtime PM -status in that case. +runtime suspend during the ``suspend_late`` phase of system-wide suspend (or +during the ``poweroff_late`` or ``freeze_late`` phase of hibernation), +after runtime PM was disabled for it. [Without doing that, the same driver +callback might be executed twice in a row for the same device, which would not +be valid in general.] If the middle-layer system-wide PM callbacks are present +for the device, they are responsible for doing the above, and the PM core takes +care of it otherwise. + +In addition, with ``DPM_FLAG_SMART_SUSPEND`` set, the driver's ``->thaw_late`` +and ``->thaw_noirq`` callbacks are skipped if the device remained in runtime +suspend during the preceding "freeze" transition related to hibernation. +Again, if the middle-layer callbacks are present for the device, they are +responsible for doing that, or the PM core takes care of it otherwise. + + +The ``DPM_FLAG_MAY_SKIP_RESUME`` Driver Flag +-------------------------------------------- During system-wide resume from a sleep state it's easiest to put devices into the full-power state, as explained in :file:`Documentation/power/runtime_pm.rst`. [Refer to that document for more information regarding this particular issue as well as for information on the device runtime power management framework in -general.] - -However, it often is desirable to leave devices in suspend after system -transitions to the working state, especially if those devices had been in +general.] However, it often is desirable to leave devices in suspend after +system transitions to the working state, especially if those devices had been in runtime suspend before the preceding system-wide suspend (or analogous) -transition. Device drivers can use the ``DPM_FLAG_MAY_SKIP_RESUME`` flag to -indicate to the PM core (and middle-layer code) that they prefer the specific -devices handled by them to be left suspended and they have no problems with -skipping their system-wide resume callbacks for this reason. Whether or not the -devices will actually be left in suspend may depend on their state before the -given system suspend-resume cycle and on the type of the system transition under -way. In particular, devices are not left suspended if that transition is a -restore from hibernation, as device states are not guaranteed to be reflected -by the information stored in the hibernation image in that case. - -The middle-layer code involved in the handling of the device is expected to -indicate to the PM core if the device may be left in suspend by setting its -:c:member:`power.may_skip_resume` status bit which is checked by the PM core -during the "noirq" phase of the preceding system-wide suspend (or analogous) -transition. The middle layer is then responsible for handling the device as -appropriate in its "noirq" resume callback, which is executed regardless of -whether or not the device is left suspended, but the other resume callbacks -(except for ``->complete``) will be skipped automatically by the PM core if the -device really can be left in suspend. - -For devices whose "noirq", "late" and "early" driver callbacks are invoked -directly by the PM core, all of the system-wide resume callbacks are skipped if -``DPM_FLAG_MAY_SKIP_RESUME`` is set and the device is in runtime suspend during -the ``suspend_noirq`` (or analogous) phase or the transition under way is a -proper system suspend (rather than anything related to hibernation) and the -device's wakeup settings are suitable for runtime PM (that is, it cannot -generate wakeup signals at all or it is allowed to wake up the system from -sleep). +transition. + +To that end, device drivers can use the ``DPM_FLAG_MAY_SKIP_RESUME`` flag to +indicate to the PM core and middle-layer code that they allow their "noirq" and +"early" resume callbacks to be skipped if the device can be left in suspend +after system-wide PM transitions to the working state. Whether or not that is +the case generally depends on the state of the device before the given system +suspend-resume cycle and on the type of the system transition under way. +In particular, the "restore" and "thaw" transitions related to hibernation are +not affected by ``DPM_FLAG_MAY_SKIP_RESUME`` at all. [All devices are always +resumed during the "restore" transition and whether or not any driver callbacks +are skipped during the "freeze" transition depends whether or not the +``DPM_FLAG_SMART_SUSPEND`` flag is set (see `above `_).] + +The ``DPM_FLAG_MAY_SKIP_RESUME`` flag is taken into account in combination with +the :c:member:`power.may_skip_resume` status bit set by the PM core during the +"suspend" phase of suspend-type transitions. If the driver or the middle layer +has a reason to prevent the driver's "noirq" and "early" resume callbacks from +being skipped during the subsequent resume transition of the system, it should +clear :c:member:`power.may_skip_resume` in its ``->suspend``, ``->suspend_late`` +or ``->suspend_noirq`` callback. [Note that the drivers setting +``DPM_FLAG_SMART_SUSPEND`` need to clear :c:member:`power.may_skip_resume` in +their ``->suspend`` callback in case the other two are skipped.] + +Setting the :c:member:`power.may_skip_resume` status bit along with the +``DPM_FLAG_MAY_SKIP_RESUME`` flag is necessary, but generally not sufficient, +for the driver's "noirq" and "early" resume callbacks to be skipped. Whether or +not they should be skipped can be determined by evaluating the +:c:func:`dev_pm_skip_resume` helper function. + +If that function returns ``true``, the driver's "noirq" and "early" resume +callbacks should be skipped and the device's runtime PM status will be set to +"suspended" by the PM core. Otherwise, if the device was runtime-suspended +during the preceding system-wide suspend transition and +``DPM_FLAG_SMART_SUSPEND`` is set for it, its runtime PM status will be set to +"active" by the PM core. [Hence, the drivers that do not set +``DPM_FLAG_SMART_SUSPEND`` should not expect the runtime PM status of their +devices to be changed from "suspended" to "active" by the PM core during +system-wide resume-type transitions.] + +If the ``DPM_FLAG_MAY_SKIP_RESUME`` flag is not set for a device, but +``DPM_FLAG_SMART_SUSPEND`` is set and the driver's "late" and "noirq" suspend +callbacks are skipped, its system-wide "noirq" and "early" resume callbacks, if +present, are invoked as usual and the device's runtime PM status is set to +"active" by the PM core before enabling runtime PM for it. In that case, the +driver must be prepared to cope with the invocation of its system-wide resume +callbacks back-to-back with its ``->runtime_suspend`` one (without the +intervening ``->runtime_resume`` and system-wide suspend callbacks) and the +final state of the device must reflect the "active" runtime PM status in that +case. [Note that this is not a problem at all if the driver's +``->suspend_late`` callback pointer points to the same function as its +``->runtime_suspend`` one and its ``->resume_early`` callback pointer points to +the same function as the ``->runtime_resume`` one, while none of the other +system-wide suspend-resume callbacks of the driver are present, for example.] + +Likewise, if ``DPM_FLAG_MAY_SKIP_RESUME`` is set for a device, its driver's +system-wide "noirq" and "early" resume callbacks may be skipped while its "late" +and "noirq" suspend callbacks may have been executed (in principle, regardless +of whether or not ``DPM_FLAG_SMART_SUSPEND`` is set). In that case, the driver +needs to be able to cope with the invocation of its ``->runtime_resume`` +callback back-to-back with its "late" and "noirq" suspend ones. [For instance, +that is not a concern if the driver sets both ``DPM_FLAG_SMART_SUSPEND`` and +``DPM_FLAG_MAY_SKIP_RESUME`` and uses the same pair of suspend/resume callback +functions for runtime PM and system-wide suspend/resume.] diff --git a/Documentation/power/pci.rst b/Documentation/power/pci.rst index f09b382b4621..1831e431f725 100644 --- a/Documentation/power/pci.rst +++ b/Documentation/power/pci.rst @@ -1010,32 +1010,33 @@ if the device is in runtime suspend when the system suspend starts. That also affects all of the ancestors of the device, so this flag should only be used if absolutely necessary. -The DPM_FLAG_SMART_PREPARE flag instructs the PCI bus type to only return a -positive value from pci_pm_prepare() if the ->prepare callback provided by the +The DPM_FLAG_SMART_PREPARE flag causes the PCI bus type to return a positive +value from pci_pm_prepare() only if the ->prepare callback provided by the driver of the device returns a positive value. That allows the driver to opt -out from using the direct-complete mechanism dynamically. +out from using the direct-complete mechanism dynamically (whereas setting +DPM_FLAG_NO_DIRECT_COMPLETE means permanent opt-out). The DPM_FLAG_SMART_SUSPEND flag tells the PCI bus type that from the driver's perspective the device can be safely left in runtime suspend during system suspend. That causes pci_pm_suspend(), pci_pm_freeze() and pci_pm_poweroff() -to skip resuming the device from runtime suspend unless there are PCI-specific -reasons for doing that. Also, it causes pci_pm_suspend_late/noirq(), -pci_pm_freeze_late/noirq() and pci_pm_poweroff_late/noirq() to return early -if the device remains in runtime suspend in the beginning of the "late" phase -of the system-wide transition under way. Moreover, if the device is in -runtime suspend in pci_pm_resume_noirq() or pci_pm_restore_noirq(), its runtime -power management status will be changed to "active" (as it is going to be put -into D0 going forward), but if it is in runtime suspend in pci_pm_thaw_noirq(), -the function will set the power.direct_complete flag for it (to make the PM core -skip the subsequent "thaw" callbacks for it) and return. - -Setting the DPM_FLAG_MAY_SKIP_RESUME flag means that the driver prefers the -device to be left in suspend after system-wide transitions to the working state. -This flag is checked by the PM core, but the PCI bus type informs the PM core -which devices may be left in suspend from its perspective (that happens during -the "noirq" phase of system-wide suspend and analogous transitions) and next it -uses the dev_pm_skip_resume() helper to decide whether or not to return from -pci_pm_resume_noirq() and pci_pm_resume_early() upfront. +to avoid resuming the device from runtime suspend unless there are PCI-specific +reasons for doing that. Also, it causes pci_pm_suspend_late/noirq() and +pci_pm_poweroff_late/noirq() to return early if the device remains in runtime +suspend during the "late" phase of the system-wide transition under way. +Moreover, if the device is in runtime suspend in pci_pm_resume_noirq() or +pci_pm_restore_noirq(), its runtime PM status will be changed to "active" (as it +is going to be put into D0 going forward). + +Setting the DPM_FLAG_MAY_SKIP_RESUME flag means that the driver allows its +"noirq" and "early" resume callbacks to be skipped if the device can be left +in suspend after a system-wide transition into the working state. This flag is +taken into consideration by the PM core along with the power.may_skip_resume +status bit of the device which is set by pci_pm_suspend_noirq() in certain +situations. If the PM core determines that the driver's "noirq" and "early" +resume callbacks should be skipped, the dev_pm_skip_resume() helper function +will return "true" and that will cause pci_pm_resume_noirq() and +pci_pm_resume_early() to return upfront without touching the device and +executing the driver callbacks. 3.2. Device Runtime Power Management ------------------------------------ diff --git a/include/linux/pm.h b/include/linux/pm.h index 35796fc49e7a..121c104a4090 100644 --- a/include/linux/pm.h +++ b/include/linux/pm.h @@ -545,25 +545,11 @@ struct pm_subsys_data { * cleared by the drivers as the driver core will take care of that. * * NO_DIRECT_COMPLETE: Do not apply direct-complete optimization to the device. - * SMART_PREPARE: Check the return value of the driver's ->prepare callback. - * SMART_SUSPEND: No need to resume the device from runtime suspend. - * MAY_SKIP_RESUME: Avoid resuming the device during system resume if possible. - * - * Setting SMART_PREPARE instructs bus types and PM domains which may want - * system suspend/resume callbacks to be skipped for the device to return 0 from - * their ->prepare callbacks if the driver's ->prepare callback returns 0 (in - * other words, the system suspend/resume callbacks can only be skipped for the - * device if its driver doesn't object against that). This flag has no effect - * if NO_DIRECT_COMPLETE is set. - * - * Setting SMART_SUSPEND instructs bus types and PM domains which may want to - * runtime resume the device upfront during system suspend that doing so is not - * necessary from the driver's perspective. It also may cause them to skip - * invocations of the ->suspend_late and ->suspend_noirq callbacks provided by - * the driver if they decide to leave the device in runtime suspend. - * - * Setting MAY_SKIP_RESUME informs the PM core and middle-layer code that the - * driver prefers the device to be left in suspend after system resume. + * SMART_PREPARE: Take the driver ->prepare callback return value into account. + * SMART_SUSPEND: Avoid resuming the device from runtime suspend. + * MAY_SKIP_RESUME: Allow driver "noirq" and "early" callbacks to be skipped. + * + * See Documentation/driver-api/pm/devices.rst for details. */ #define DPM_FLAG_NO_DIRECT_COMPLETE BIT(0) #define DPM_FLAG_SMART_PREPARE BIT(1) -- cgit v1.2.3 From 598cc93005636e32a98ed003b0158329827c0ccb Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Sat, 25 Apr 2020 16:35:40 -0400 Subject: PM: sleep: Helpful edits for devices.rst documentation Here are some minor edits of the devices.rst documentation file, intended to improve the clarity and add a couple of missing details. Signed-off-by: Alan Stern Signed-off-by: Rafael J. Wysocki --- Documentation/driver-api/pm/devices.rst | 94 ++++++++++++++++++--------------- 1 file changed, 52 insertions(+), 42 deletions(-) diff --git a/Documentation/driver-api/pm/devices.rst b/Documentation/driver-api/pm/devices.rst index 782cb37073a3..946ad0b94e31 100644 --- a/Documentation/driver-api/pm/devices.rst +++ b/Documentation/driver-api/pm/devices.rst @@ -349,7 +349,7 @@ the phases are: ``prepare``, ``suspend``, ``suspend_late``, ``suspend_noirq``. PM core will skip the ``suspend``, ``suspend_late`` and ``suspend_noirq`` phases as well as all of the corresponding phases of the subsequent device resume for all of these devices. In that case, - the ``->complete`` callback will be invoked directly after the + the ``->complete`` callback will be the next one invoked after the ``->prepare`` callback and is entirely responsible for putting the device into a consistent state as appropriate. @@ -383,11 +383,15 @@ the phases are: ``prepare``, ``suspend``, ``suspend_late``, ``suspend_noirq``. ``->suspend`` methods provided by subsystems (bus types and PM domains in particular) must follow an additional rule regarding what can be done to the devices before their drivers' ``->suspend`` methods are called. - Namely, they can only resume the devices from runtime suspend by - calling :c:func:`pm_runtime_resume` for them, if that is necessary, and + Namely, they may resume the devices from runtime suspend by + calling :c:func:`pm_runtime_resume` for them, if that is necessary, but they must not update the state of the devices in any other way at that time (in case the drivers need to resume the devices from runtime - suspend in their ``->suspend`` methods). + suspend in their ``->suspend`` methods). In fact, the PM core prevents + subsystems or drivers from putting devices into runtime suspend at + these times by calling :c:func:`pm_runtime_get_noresume` before issuing + the ``->prepare`` callback (and calling :c:func:`pm_runtime_put` after + issuing the ``->complete`` callback). 3. For a number of devices it is convenient to split suspend into the "quiesce device" and "save device state" phases, in which cases @@ -459,22 +463,22 @@ When resuming from freeze, standby or memory sleep, the phases are: Note, however, that new children may be registered below the device as soon as the ``->resume`` callbacks occur; it's not necessary to wait - until the ``complete`` phase with that. + until the ``complete`` phase runs. Moreover, if the preceding ``->prepare`` callback returned a positive number, the device may have been left in runtime suspend throughout the - whole system suspend and resume (the ``suspend``, ``suspend_late``, - ``suspend_noirq`` phases of system suspend and the ``resume_noirq``, - ``resume_early``, ``resume`` phases of system resume may have been - skipped for it). In that case, the ``->complete`` callback is entirely + whole system suspend and resume (its ``->suspend``, ``->suspend_late``, + ``->suspend_noirq``, ``->resume_noirq``, + ``->resume_early``, and ``->resume`` callbacks may have been + skipped). In that case, the ``->complete`` callback is entirely responsible for putting the device into a consistent state after system suspend if necessary. [For example, it may need to queue up a runtime resume request for the device for this purpose.] To check if that is the case, the ``->complete`` callback can consult the device's - ``power.direct_complete`` flag. Namely, if that flag is set when the - ``->complete`` callback is being run, it has been called directly after - the preceding ``->prepare`` and special actions may be required - to make the device work correctly afterward. + ``power.direct_complete`` flag. If that flag is set when the + ``->complete`` callback is being run then the direct-complete mechanism + was used, and special actions may be required to make the device work + correctly afterward. At the end of these phases, drivers should be as functional as they were before suspending: I/O can be performed using DMA and IRQs, and the relevant clocks are @@ -575,10 +579,12 @@ and the phases are similar. The ``->poweroff``, ``->poweroff_late`` and ``->poweroff_noirq`` callbacks should do essentially the same things as the ``->suspend``, ``->suspend_late`` -and ``->suspend_noirq`` callbacks, respectively. The only notable difference is +and ``->suspend_noirq`` callbacks, respectively. A notable difference is that they need not store the device register values, because the registers should already have been stored during the ``freeze``, ``freeze_late`` or -``freeze_noirq`` phases. +``freeze_noirq`` phases. Also, on many machines the firmware will power-down +the entire system, so it is not necessary for the callback to put the device in +a low-power state. Leaving Hibernation @@ -764,11 +770,10 @@ device driver in question. If it is necessary to resume a device from runtime suspend during a system-wide transition into a sleep state, that can be done by calling -:c:func:`pm_runtime_resume` for it from the ``->suspend`` callback (or its -couterpart for transitions related to hibernation) of either the device's driver -or a subsystem responsible for it (for example, a bus type or a PM domain). -That is guaranteed to work by the requirement that subsystems must not change -the state of devices (possibly except for resuming them from runtime suspend) +:c:func:`pm_runtime_resume` from the ``->suspend`` callback (or the ``->freeze`` +or ``->poweroff`` callback for transitions related to hibernation) of either the +device's driver or its subsystem (for example, a bus type or a PM domain). +However, subsystems must not otherwise change the runtime status of devices from their ``->prepare`` and ``->suspend`` callbacks (or equivalent) *before* invoking device drivers' ``->suspend`` callbacks (or equivalent). @@ -779,27 +784,29 @@ The ``DPM_FLAG_SMART_SUSPEND`` Driver Flag Some bus types and PM domains have a policy to resume all devices from runtime suspend upfront in their ``->suspend`` callbacks, but that may not be really -necessary if the driver of the device can cope with runtime-suspended devices. -The driver can indicate that by setting ``DPM_FLAG_SMART_SUSPEND`` in -:c:member:`power.driver_flags` at the probe time with the help of the +necessary if the device's driver can cope with runtime-suspended devices. +The driver can indicate this by setting ``DPM_FLAG_SMART_SUSPEND`` in +:c:member:`power.driver_flags` at probe time, with the assistance of the :c:func:`dev_pm_set_driver_flags` helper routine. -However, setting that flag also causes the PM core and middle-layer code +Setting that flag causes the PM core and middle-layer code (bus types, PM domains etc.) to skip the ``->suspend_late`` and ``->suspend_noirq`` callbacks provided by the driver if the device remains in -runtime suspend during the ``suspend_late`` phase of system-wide suspend (or -during the ``poweroff_late`` or ``freeze_late`` phase of hibernation), -after runtime PM was disabled for it. [Without doing that, the same driver +runtime suspend throughout those phases of the system-wide suspend (and +similarly for the "freeze" and "poweroff" parts of system hibernation). +[Otherwise the same driver callback might be executed twice in a row for the same device, which would not be valid in general.] If the middle-layer system-wide PM callbacks are present -for the device, they are responsible for doing the above, and the PM core takes -care of it otherwise. +for the device then they are responsible for skipping these driver callbacks; +if not then the PM core skips them. The subsystem callback routines can +determine whether they need to skip the driver callbacks by testing the return +value from the :c:func:`dev_pm_skip_suspend` helper function. -In addition, with ``DPM_FLAG_SMART_SUSPEND`` set, the driver's ``->thaw_late`` -and ``->thaw_noirq`` callbacks are skipped if the device remained in runtime -suspend during the preceding "freeze" transition related to hibernation. -Again, if the middle-layer callbacks are present for the device, they are -responsible for doing that, or the PM core takes care of it otherwise. +In addition, with ``DPM_FLAG_SMART_SUSPEND`` set, the driver's ``->thaw_noirq`` +and ``->thaw_early`` callbacks are skipped in hibernation if the device remained +in runtime suspend throughout the preceding "freeze" transition. Again, if the +middle-layer callbacks are present for the device, they are responsible for +doing this, otherwise the PM core takes care of it. The ``DPM_FLAG_MAY_SKIP_RESUME`` Driver Flag @@ -820,17 +827,20 @@ indicate to the PM core and middle-layer code that they allow their "noirq" and after system-wide PM transitions to the working state. Whether or not that is the case generally depends on the state of the device before the given system suspend-resume cycle and on the type of the system transition under way. -In particular, the "restore" and "thaw" transitions related to hibernation are -not affected by ``DPM_FLAG_MAY_SKIP_RESUME`` at all. [All devices are always -resumed during the "restore" transition and whether or not any driver callbacks -are skipped during the "freeze" transition depends whether or not the -``DPM_FLAG_SMART_SUSPEND`` flag is set (see `above `_).] +In particular, the "thaw" and "restore" transitions related to hibernation are +not affected by ``DPM_FLAG_MAY_SKIP_RESUME`` at all. [All callbacks are +issued during the "restore" transition regardless of the flag settings, +and whether or not any driver callbacks +are skipped during the "thaw" transition depends whether or not the +``DPM_FLAG_SMART_SUSPEND`` flag is set (see `above `_). +In addition, a device is not allowed to remain in runtime suspend if any of its +children will be returned to full power.] The ``DPM_FLAG_MAY_SKIP_RESUME`` flag is taken into account in combination with the :c:member:`power.may_skip_resume` status bit set by the PM core during the "suspend" phase of suspend-type transitions. If the driver or the middle layer has a reason to prevent the driver's "noirq" and "early" resume callbacks from -being skipped during the subsequent resume transition of the system, it should +being skipped during the subsequent system resume transition, it should clear :c:member:`power.may_skip_resume` in its ``->suspend``, ``->suspend_late`` or ``->suspend_noirq`` callback. [Note that the drivers setting ``DPM_FLAG_SMART_SUSPEND`` need to clear :c:member:`power.may_skip_resume` in @@ -845,8 +855,8 @@ not they should be skipped can be determined by evaluating the If that function returns ``true``, the driver's "noirq" and "early" resume callbacks should be skipped and the device's runtime PM status will be set to "suspended" by the PM core. Otherwise, if the device was runtime-suspended -during the preceding system-wide suspend transition and -``DPM_FLAG_SMART_SUSPEND`` is set for it, its runtime PM status will be set to +during the preceding system-wide suspend transition and its +``DPM_FLAG_SMART_SUSPEND`` is set, its runtime PM status will be set to "active" by the PM core. [Hence, the drivers that do not set ``DPM_FLAG_SMART_SUSPEND`` should not expect the runtime PM status of their devices to be changed from "suspended" to "active" by the PM core during -- cgit v1.2.3 From ab7e9b067f3d9cbec28cfca51d341efb421b7a51 Mon Sep 17 00:00:00 2001 From: Domenico Andreoli Date: Thu, 7 May 2020 09:19:52 +0200 Subject: PM: hibernate: Incorporate concurrency handling Hibernation concurrency handling is currently delegated to user.c, where it's also used for regulating the access to the snapshot device. In the prospective of making user.c a separate configuration option, such mutual exclusion is brought into hibernate.c and made available through accessor helpers hereby introduced. Signed-off-by: Domenico Andreoli Signed-off-by: Rafael J. Wysocki --- kernel/power/hibernate.c | 20 ++++++++++++++++---- kernel/power/power.h | 4 ++-- kernel/power/user.c | 10 ++++------ 3 files changed, 22 insertions(+), 12 deletions(-) diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c index 30bd28d1d418..02ec716a4927 100644 --- a/kernel/power/hibernate.c +++ b/kernel/power/hibernate.c @@ -67,6 +67,18 @@ bool freezer_test_done; static const struct platform_hibernation_ops *hibernation_ops; +static atomic_t hibernate_atomic = ATOMIC_INIT(1); + +bool hibernate_acquire(void) +{ + return atomic_add_unless(&hibernate_atomic, -1, 0); +} + +void hibernate_release(void) +{ + atomic_inc(&hibernate_atomic); +} + bool hibernation_available(void) { return nohibernate == 0 && !security_locked_down(LOCKDOWN_HIBERNATION); @@ -704,7 +716,7 @@ int hibernate(void) lock_system_sleep(); /* The snapshot device should not be opened while we're running */ - if (!atomic_add_unless(&snapshot_device_available, -1, 0)) { + if (!hibernate_acquire()) { error = -EBUSY; goto Unlock; } @@ -775,7 +787,7 @@ int hibernate(void) Exit: __pm_notifier_call_chain(PM_POST_HIBERNATION, nr_calls, NULL); pm_restore_console(); - atomic_inc(&snapshot_device_available); + hibernate_release(); Unlock: unlock_system_sleep(); pr_info("hibernation exit\n"); @@ -880,7 +892,7 @@ static int software_resume(void) goto Unlock; /* The snapshot device should not be opened while we're running */ - if (!atomic_add_unless(&snapshot_device_available, -1, 0)) { + if (!hibernate_acquire()) { error = -EBUSY; swsusp_close(FMODE_READ); goto Unlock; @@ -911,7 +923,7 @@ static int software_resume(void) __pm_notifier_call_chain(PM_POST_RESTORE, nr_calls, NULL); pm_restore_console(); pr_info("resume failed (%d)\n", error); - atomic_inc(&snapshot_device_available); + hibernate_release(); /* For success case, the suspend path will release the lock */ Unlock: mutex_unlock(&system_transition_mutex); diff --git a/kernel/power/power.h b/kernel/power/power.h index 7cdc64dc2373..ba2094db6294 100644 --- a/kernel/power/power.h +++ b/kernel/power/power.h @@ -154,8 +154,8 @@ extern int snapshot_write_next(struct snapshot_handle *handle); extern void snapshot_write_finalize(struct snapshot_handle *handle); extern int snapshot_image_loaded(struct snapshot_handle *handle); -/* If unset, the snapshot device cannot be open. */ -extern atomic_t snapshot_device_available; +extern bool hibernate_acquire(void); +extern void hibernate_release(void); extern sector_t alloc_swapdev_block(int swap); extern void free_all_swap_pages(int swap); diff --git a/kernel/power/user.c b/kernel/power/user.c index 7959449765d9..98548d1cf8a6 100644 --- a/kernel/power/user.c +++ b/kernel/power/user.c @@ -37,8 +37,6 @@ static struct snapshot_data { bool free_bitmaps; } snapshot_state; -atomic_t snapshot_device_available = ATOMIC_INIT(1); - static int snapshot_open(struct inode *inode, struct file *filp) { struct snapshot_data *data; @@ -49,13 +47,13 @@ static int snapshot_open(struct inode *inode, struct file *filp) lock_system_sleep(); - if (!atomic_add_unless(&snapshot_device_available, -1, 0)) { + if (!hibernate_acquire()) { error = -EBUSY; goto Unlock; } if ((filp->f_flags & O_ACCMODE) == O_RDWR) { - atomic_inc(&snapshot_device_available); + hibernate_release(); error = -ENOSYS; goto Unlock; } @@ -92,7 +90,7 @@ static int snapshot_open(struct inode *inode, struct file *filp) __pm_notifier_call_chain(PM_POST_RESTORE, nr_calls, NULL); } if (error) - atomic_inc(&snapshot_device_available); + hibernate_release(); data->frozen = false; data->ready = false; @@ -122,7 +120,7 @@ static int snapshot_release(struct inode *inode, struct file *filp) } pm_notifier_call_chain(data->mode == O_RDONLY ? PM_POST_HIBERNATION : PM_POST_RESTORE); - atomic_inc(&snapshot_device_available); + hibernate_release(); unlock_system_sleep(); -- cgit v1.2.3 From c4f39a6c74389fcc93ac39056ef342f32ab57a23 Mon Sep 17 00:00:00 2001 From: Domenico Andreoli Date: Thu, 7 May 2020 09:19:53 +0200 Subject: PM: hibernate: Split off snapshot dev option Make it possible to reduce the attack surface in case the snapshot device is not to be used from userspace. Signed-off-by: Domenico Andreoli Signed-off-by: Rafael J. Wysocki --- kernel/power/Kconfig | 12 ++++++++++++ kernel/power/Makefile | 3 ++- 2 files changed, 14 insertions(+), 1 deletion(-) diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig index c208566c844b..4d0e6e815a2b 100644 --- a/kernel/power/Kconfig +++ b/kernel/power/Kconfig @@ -80,6 +80,18 @@ config HIBERNATION For more information take a look at . +config HIBERNATION_SNAPSHOT_DEV + bool "Userspace snapshot device" + depends on HIBERNATION + default y + ---help--- + Device used by the uswsusp tools. + + Say N if no snapshotting from userspace is needed, this also + reduces the attack surface of the kernel. + + If in doubt, say Y. + config PM_STD_PARTITION string "Default resume partition" depends on HIBERNATION diff --git a/kernel/power/Makefile b/kernel/power/Makefile index e7e47d9be1e5..5899260a8bef 100644 --- a/kernel/power/Makefile +++ b/kernel/power/Makefile @@ -10,7 +10,8 @@ obj-$(CONFIG_VT_CONSOLE_SLEEP) += console.o obj-$(CONFIG_FREEZER) += process.o obj-$(CONFIG_SUSPEND) += suspend.o obj-$(CONFIG_PM_TEST_SUSPEND) += suspend_test.o -obj-$(CONFIG_HIBERNATION) += hibernate.o snapshot.o swap.o user.o +obj-$(CONFIG_HIBERNATION) += hibernate.o snapshot.o swap.o +obj-$(CONFIG_HIBERNATION_SNAPSHOT_DEV) += user.o obj-$(CONFIG_PM_AUTOSLEEP) += autosleep.o obj-$(CONFIG_PM_WAKELOCKS) += wakelock.o -- cgit v1.2.3 From 3618bbaaa898cf48e859736120a775fcb56f3838 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Fri, 22 May 2020 18:09:55 +0300 Subject: PM: runtime: Make clear what we do when conditions are wrong in rpm_suspend() rpm_suspend() simple bails out when conditions are wrong. But this is not immediately obvious from the code. Make it clear what we do when conditions are wrong in rpm_suspend(). Signed-off-by: Andy Shevchenko Signed-off-by: Rafael J. Wysocki --- drivers/base/power/runtime.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/drivers/base/power/runtime.c b/drivers/base/power/runtime.c index 99c7da112c95..9f62790f644c 100644 --- a/drivers/base/power/runtime.c +++ b/drivers/base/power/runtime.c @@ -523,13 +523,11 @@ static int rpm_suspend(struct device *dev, int rpmflags) repeat: retval = rpm_check_suspend_allowed(dev); - if (retval < 0) - ; /* Conditions are wrong. */ + goto out; /* Conditions are wrong. */ /* Synchronous suspends are not allowed in the RPM_RESUMING state. */ - else if (dev->power.runtime_status == RPM_RESUMING && - !(rpmflags & RPM_ASYNC)) + if (dev->power.runtime_status == RPM_RESUMING && !(rpmflags & RPM_ASYNC)) retval = -EAGAIN; if (retval) goto out; -- cgit v1.2.3 From 64c7d7ea22d86cacb65d0c097cc447bc0e6d8abd Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Thu, 21 May 2020 19:08:09 +0200 Subject: PM: runtime: clk: Fix clk_pm_runtime_get() error path clk_pm_runtime_get() assumes that the PM-runtime usage counter will be dropped by pm_runtime_get_sync() on errors, which is not the case, so PM-runtime references to devices acquired by the former are leaked on errors returned by the latter. Fix this by modifying clk_pm_runtime_get() to drop the reference if pm_runtime_get_sync() returns an error. Fixes: 9a34b45397e5 clk: Add support for runtime PM Cc: 4.15+ # 4.15+ Signed-off-by: Rafael J. Wysocki Reviewed-by: Ulf Hansson --- drivers/clk/clk.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/clk/clk.c b/drivers/clk/clk.c index 2dfb30b963c4..407f6919604c 100644 --- a/drivers/clk/clk.c +++ b/drivers/clk/clk.c @@ -114,7 +114,11 @@ static int clk_pm_runtime_get(struct clk_core *core) return 0; ret = pm_runtime_get_sync(core->dev); - return ret < 0 ? ret : 0; + if (ret < 0) { + pm_runtime_put_noidle(core->dev); + return ret; + } + return 0; } static void clk_pm_runtime_put(struct clk_core *core) -- cgit v1.2.3 From ad1e4f74c072eaa2c6d77dd710db31aafecd614f Mon Sep 17 00:00:00 2001 From: Domenico Andreoli Date: Tue, 19 May 2020 20:14:10 +0200 Subject: PM: hibernate: Restrict writes to the resume device Hibernation via snapshot device requires write permission to the swap block device, the one that more often (but not necessarily) is used to store the hibernation image. With this patch, such permissions are granted iff: 1) snapshot device config option is enabled 2) swap partition is used as resume device In other circumstances the swap device is not writable from userspace. In order to achieve this, every write attempt to a swap device is checked against the device configured as part of the uswsusp API [0] using a pointer to the inode struct in memory. If the swap device being written was not configured for resuming, the write request is denied. NOTE: this implementation works only for swap block devices, where the inode configured by swapon (which sets S_SWAPFILE) is the same used by SNAPSHOT_SET_SWAP_AREA. In case of swap file, SNAPSHOT_SET_SWAP_AREA indeed receives the inode of the block device containing the filesystem where the swap file is located (+ offset in it) which is never passed to swapon and then has not set S_SWAPFILE. As result, the swap file itself (as a file) has never an option to be written from userspace. Instead it remains writable if accessed directly from the containing block device, which is always writeable from root. [0] Documentation/power/userland-swsusp.rst v2: - rename is_hibernate_snapshot_dev() to is_hibernate_resume_dev() - fix description so to correctly refer to the resume device Signed-off-by: Domenico Andreoli Acked-by: Darrick J. Wong Signed-off-by: Rafael J. Wysocki --- fs/block_dev.c | 3 +-- include/linux/suspend.h | 6 ++++++ kernel/power/user.c | 14 +++++++++++++- 3 files changed, 20 insertions(+), 3 deletions(-) diff --git a/fs/block_dev.c b/fs/block_dev.c index 93672c3f1c78..608a7b9173f4 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -2023,8 +2023,7 @@ ssize_t blkdev_write_iter(struct kiocb *iocb, struct iov_iter *from) if (bdev_read_only(I_BDEV(bd_inode))) return -EPERM; - /* uswsusp needs write permission to the swap */ - if (IS_SWAPFILE(bd_inode) && !hibernation_available()) + if (IS_SWAPFILE(bd_inode) && !is_hibernate_resume_dev(bd_inode)) return -ETXTBSY; if (!iov_iter_count(from)) diff --git a/include/linux/suspend.h b/include/linux/suspend.h index 4fcc6fd0cbd6..b960098acfb0 100644 --- a/include/linux/suspend.h +++ b/include/linux/suspend.h @@ -466,6 +466,12 @@ static inline bool system_entering_hibernation(void) { return false; } static inline bool hibernation_available(void) { return false; } #endif /* CONFIG_HIBERNATION */ +#ifdef CONFIG_HIBERNATION_SNAPSHOT_DEV +int is_hibernate_resume_dev(const struct inode *); +#else +static inline int is_hibernate_resume_dev(const struct inode *i) { return 0; } +#endif + /* Hibernation and suspend events */ #define PM_HIBERNATION_PREPARE 0x0001 /* Going to hibernate */ #define PM_POST_HIBERNATION 0x0002 /* Hibernation finished */ diff --git a/kernel/power/user.c b/kernel/power/user.c index 98548d1cf8a6..d5eedc2baa2a 100644 --- a/kernel/power/user.c +++ b/kernel/power/user.c @@ -35,8 +35,14 @@ static struct snapshot_data { bool ready; bool platform_support; bool free_bitmaps; + struct inode *bd_inode; } snapshot_state; +int is_hibernate_resume_dev(const struct inode *bd_inode) +{ + return hibernation_available() && snapshot_state.bd_inode == bd_inode; +} + static int snapshot_open(struct inode *inode, struct file *filp) { struct snapshot_data *data; @@ -95,6 +101,7 @@ static int snapshot_open(struct inode *inode, struct file *filp) data->frozen = false; data->ready = false; data->platform_support = false; + data->bd_inode = NULL; Unlock: unlock_system_sleep(); @@ -110,6 +117,7 @@ static int snapshot_release(struct inode *inode, struct file *filp) swsusp_free(); data = filp->private_data; + data->bd_inode = NULL; free_all_swap_pages(data->swap); if (data->frozen) { pm_restore_gfp_mask(); @@ -202,6 +210,7 @@ struct compat_resume_swap_area { static int snapshot_set_swap_area(struct snapshot_data *data, void __user *argp) { + struct block_device *bdev; sector_t offset; dev_t swdev; @@ -232,9 +241,12 @@ static int snapshot_set_swap_area(struct snapshot_data *data, data->swap = -1; return -EINVAL; } - data->swap = swap_type_of(swdev, offset, NULL); + data->swap = swap_type_of(swdev, offset, &bdev); if (data->swap < 0) return -ENODEV; + + data->bd_inode = bdev->bd_inode; + bdput(bdev); return 0; } -- cgit v1.2.3 From 9a7875461fd0427dc86e3a87e93bd5723679b8b1 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Thu, 28 May 2020 16:45:14 +0200 Subject: PM: runtime: Replace pm_runtime_callbacks_present() The name of pm_runtime_callbacks_present() is confusing, because it suggests that the device has PM-runtime callbacks if 'true' is returned by that function, but in fact that may not be the case, so replace it with pm_runtime_has_no_callbacks() which is not ambiguous. No functional impact. Signed-off-by: Rafael J. Wysocki Reviewed-by: Ulf Hansson --- drivers/base/power/sysfs.c | 4 ++-- include/linux/pm_runtime.h | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/base/power/sysfs.c b/drivers/base/power/sysfs.c index 2b99fe1eb207..24d25cf8ab14 100644 --- a/drivers/base/power/sysfs.c +++ b/drivers/base/power/sysfs.c @@ -666,7 +666,7 @@ int dpm_sysfs_add(struct device *dev) if (rc) return rc; - if (pm_runtime_callbacks_present(dev)) { + if (!pm_runtime_has_no_callbacks(dev)) { rc = sysfs_merge_group(&dev->kobj, &pm_runtime_attr_group); if (rc) goto err_out; @@ -709,7 +709,7 @@ int dpm_sysfs_change_owner(struct device *dev, kuid_t kuid, kgid_t kgid) if (rc) return rc; - if (pm_runtime_callbacks_present(dev)) { + if (!pm_runtime_has_no_callbacks(dev)) { rc = sysfs_group_change_owner( &dev->kobj, &pm_runtime_attr_group, kuid, kgid); if (rc) diff --git a/include/linux/pm_runtime.h b/include/linux/pm_runtime.h index 3bdcbce8141a..3dbc207bff53 100644 --- a/include/linux/pm_runtime.h +++ b/include/linux/pm_runtime.h @@ -102,9 +102,9 @@ static inline bool pm_runtime_enabled(struct device *dev) return !dev->power.disable_depth; } -static inline bool pm_runtime_callbacks_present(struct device *dev) +static inline bool pm_runtime_has_no_callbacks(struct device *dev) { - return !dev->power.no_callbacks; + return dev->power.no_callbacks; } static inline void pm_runtime_mark_last_busy(struct device *dev) -- cgit v1.2.3