[PATCH -next] PM: Simplify the new suspend/hibernation framework for devices

[PATCH -next] PM: Simplify the new suspend/hibernation framework for devices

[Rafael J. Wysocki]

Hi Greg,

I promised at the KS that I would simplify the new suspend/hibernation
framework for devices to avoid the confusion with two types of PM
operations and pointers to PM operations from too many places.

The appended patch is intended for this purpose.  Unfortunately, I can't
split it into subsystem-related patches, because compilation would be broken
between them.

The patch applies to linux-next, but it's trivial to make it apply to the
mainline.  It's been compiled on x86 (both 32-bit and 64-bit) and tested
on hp nx6325, doesn't appear to break anything.

Thanks,
Rafael


---
From: Rafael J. Wysocki <rjw@sisk.pl>

PM: Simplify the new suspend/hibernation framework for devices

Following the discussion at the Kernel Summit, simplify the new
device PM framework by merging 'struct pm_ops' and
'struct pm_ext_ops' and removing pointers to 'struct pm_ext_ops'
from 'struct platform_driver' and 'struct pci_driver'.

After this change, the suspend/hibernation callbacks will only
reside in 'struct device_driver' as well as at the bus type/
device class/device type level.  Accordingly, PCI and platform
device drivers are now expected to put their suspend/hibernation
callbacks into the 'struct device_driver' embedded in
'struct pci_driver' or 'struct platform_driver', respectively.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
---
 drivers/base/platform.c         |  115 ++++++++++++++++++++--------------------
 drivers/base/power/main.c       |   19 +++---
 drivers/pci/pci-driver.c        |   46 ++++++----------
 drivers/usb/core/usb.c          |    4 -
 include/linux/device.h          |    8 +-
 include/linux/pci.h             |    1 
 include/linux/platform_device.h |    1 
 include/linux/pm.h              |   76 ++++++++------------------
 8 files changed, 119 insertions(+), 151 deletions(-)

Index: linux-2.6/include/linux/pm.h
===================================================================
--- linux-2.6.orig/include/linux/pm.h
+++ linux-2.6/include/linux/pm.h
@@ -41,7 +41,7 @@ typedef struct pm_message {
 } pm_message_t;
 
 /**
- * struct pm_ops - device PM callbacks
+ * struct dev_pm_ops - device PM callbacks
  *
  * Several driver power state transitions are externally visible, affecting
  * the state of pending I/O queues and (for drivers that touch hardware)
@@ -126,46 +126,6 @@ typedef struct pm_message {
  *	On most platforms, there are no restrictions on availability of
  *	resources like clocks during @restore().
  *
- * All of the above callbacks, except for @complete(), return error codes.
- * However, the error codes returned by the resume operations, @resume(),
- * @thaw(), and @restore(), do not cause the PM core to abort the resume
- * transition during which they are returned.  The error codes returned in
- * that cases are only printed by the PM core to the system logs for debugging
- * purposes.  Still, it is recommended that drivers only return error codes
- * from their resume methods in case of an unrecoverable failure (i.e. when the
- * device being handled refuses to resume and becomes unusable) to allow us to
- * modify the PM core in the future, so that it can avoid attempting to handle
- * devices that failed to resume and their children.
- *
- * It is allowed to unregister devices while the above callbacks are being
- * executed.  However, it is not allowed to unregister a device from within any
- * of its own callbacks.
- */
-
-struct pm_ops {
-	int (*prepare)(struct device *dev);
-	void (*complete)(struct device *dev);
-	int (*suspend)(struct device *dev);
-	int (*resume)(struct device *dev);
-	int (*freeze)(struct device *dev);
-	int (*thaw)(struct device *dev);
-	int (*poweroff)(struct device *dev);
-	int (*restore)(struct device *dev);
-};
-
-/**
- * struct pm_ext_ops - extended device PM callbacks
- *
- * Some devices require certain operations related to suspend and hibernation
- * to be carried out with interrupts disabled.  Thus, 'struct pm_ext_ops' below
- * is defined, adding callbacks to be executed with interrupts disabled to
- * 'struct pm_ops'.
- *
- * The following callbacks included in 'struct pm_ext_ops' are executed with
- * the nonboot CPUs switched off and with interrupts disabled on the only
- * functional CPU.  They also are executed with the PM core list of devices
- * locked, so they must NOT unregister any devices.
- *
  * @suspend_noirq: Complete the operations of ->suspend() by carrying out any
  *	actions required for suspending the device that need interrupts to be
  *	disabled
@@ -190,18 +150,32 @@ struct pm_ops {
  *	actions required for restoring the operations of the device that need
  *	interrupts to be disabled
  *
- * All of the above callbacks return error codes, but the error codes returned
- * by the resume operations, @resume_noirq(), @thaw_noirq(), and
- * @restore_noirq(), do not cause the PM core to abort the resume transition
- * during which they are returned.  The error codes returned in that cases are
- * only printed by the PM core to the system logs for debugging purposes.
- * Still, as stated above, it is recommended that drivers only return error
- * codes from their resume methods if the device being handled fails to resume
- * and is not usable any more.
+ * All of the above callbacks, except for @complete(), return error codes.
+ * However, the error codes returned by the resume operations, @resume(),
+ * @thaw(), @restore(), @resume_noirq(), @thaw_noirq(), and @restore_noirq() do
+ * not cause the PM core to abort the resume transition during which they are
+ * returned.  The error codes returned in that cases are only printed by the PM
+ * core to the system logs for debugging purposes.  Still, it is recommended
+ * that drivers only return error codes from their resume methods in case of an
+ * unrecoverable failure (i.e. when the device being handled refuses to resume
+ * and becomes unusable) to allow us to modify the PM core in the future, so
+ * that it can avoid attempting to handle devices that failed to resume and
+ * their children.
+ *
+ * It is allowed to unregister devices while the above callbacks are being
+ * executed.  However, it is not allowed to unregister a device from within any
+ * of its own callbacks.
  */
 
-struct pm_ext_ops {
-	struct pm_ops base;
+struct dev_pm_ops {
+	int (*prepare)(struct device *dev);
+	void (*complete)(struct device *dev);
+	int (*suspend)(struct device *dev);
+	int (*resume)(struct device *dev);
+	int (*freeze)(struct device *dev);
+	int (*thaw)(struct device *dev);
+	int (*poweroff)(struct device *dev);
+	int (*restore)(struct device *dev);
 	int (*suspend_noirq)(struct device *dev);
 	int (*resume_noirq)(struct device *dev);
 	int (*freeze_noirq)(struct device *dev);
Index: linux-2.6/include/linux/device.h
===================================================================
--- linux-2.6.orig/include/linux/device.h
+++ linux-2.6/include/linux/device.h
@@ -65,7 +65,7 @@ struct bus_type {
 	int (*resume_early)(struct device *dev);
 	int (*resume)(struct device *dev);
 
-	struct pm_ext_ops *pm;
+	struct dev_pm_ops *pm;
 
 	struct bus_type_private *p;
 };
@@ -133,7 +133,7 @@ struct device_driver {
 	int (*resume) (struct device *dev);
 	struct attribute_group **groups;
 
-	struct pm_ops *pm;
+	struct dev_pm_ops *pm;
 
 	struct driver_private *p;
 };
@@ -201,7 +201,7 @@ struct class {
 	enum sysfs_tag_type tag_type;
 	const void *(*sysfs_tag)(struct device *dev);
 
-	struct pm_ops *pm;
+	struct dev_pm_ops *pm;
 	struct class_private *p;
 };
 
@@ -291,7 +291,7 @@ struct device_type {
 	int (*uevent)(struct device *dev, struct kobj_uevent_env *env);
 	void (*release)(struct device *dev);
 
-	struct pm_ops *pm;
+	struct dev_pm_ops *pm;
 };
 
 /* interface for exporting device attributes */
Index: linux-2.6/include/linux/platform_device.h
===================================================================
--- linux-2.6.orig/include/linux/platform_device.h
+++ linux-2.6/include/linux/platform_device.h
@@ -53,7 +53,6 @@ struct platform_driver {
 	int (*suspend_late)(struct platform_device *, pm_message_t state);
 	int (*resume_early)(struct platform_device *);
 	int (*resume)(struct platform_device *);
-	struct pm_ext_ops *pm;
 	struct device_driver driver;
 };
 
Index: linux-2.6/include/linux/pci.h
===================================================================
--- linux-2.6.orig/include/linux/pci.h
+++ linux-2.6/include/linux/pci.h
@@ -414,7 +414,6 @@ struct pci_driver {
 	int  (*resume_early) (struct pci_dev *dev);
 	int  (*resume) (struct pci_dev *dev);	                /* Device woken up */
 	void (*shutdown) (struct pci_dev *dev);
-	struct pm_ext_ops *pm;
 	struct pci_error_handlers *err_handler;
 	struct device_driver	driver;
 	struct pci_dynids dynids;
Index: linux-2.6/drivers/pci/pci-driver.c
===================================================================
--- linux-2.6.orig/drivers/pci/pci-driver.c
+++ linux-2.6/drivers/pci/pci-driver.c
@@ -433,8 +433,7 @@ static int pci_pm_suspend(struct device 
 
 static int pci_pm_suspend_noirq(struct device *dev)
 {
-	struct pci_dev *pci_dev = to_pci_dev(dev);
-	struct pci_driver *drv = pci_dev->driver;
+	struct device_driver *drv = dev->driver;
 	int error = 0;
 
 	if (drv && drv->pm) {
@@ -469,11 +468,10 @@ static int pci_pm_resume(struct device *
 
 static int pci_pm_resume_noirq(struct device *dev)
 {
-	struct pci_dev *pci_dev = to_pci_dev(dev);
-	struct pci_driver *drv = pci_dev->driver;
+	struct device_driver *drv = dev->driver;
 	int error = 0;
 
-	pci_fixup_device(pci_fixup_resume_early, pci_dev);
+	pci_fixup_device(pci_fixup_resume_early, to_pci_dev(dev));
 
 	if (drv && drv->pm) {
 		if (drv->pm->resume_noirq)
@@ -519,8 +517,7 @@ static int pci_pm_freeze(struct device *
 
 static int pci_pm_freeze_noirq(struct device *dev)
 {
-	struct pci_dev *pci_dev = to_pci_dev(dev);
-	struct pci_driver *drv = pci_dev->driver;
+	struct device_driver *drv = dev->driver;
 	int error = 0;
 
 	if (drv && drv->pm) {
@@ -553,15 +550,14 @@ static int pci_pm_thaw(struct device *de
 
 static int pci_pm_thaw_noirq(struct device *dev)
 {
-	struct pci_dev *pci_dev = to_pci_dev(dev);
-	struct pci_driver *drv = pci_dev->driver;
+	struct device_driver *drv = dev->driver;
 	int error = 0;
 
 	if (drv && drv->pm) {
 		if (drv->pm->thaw_noirq)
 			error = drv->pm->thaw_noirq(dev);
 	} else {
-		pci_fixup_device(pci_fixup_resume_early, pci_dev);
+		pci_fixup_device(pci_fixup_resume_early, to_pci_dev(dev));
 		error = pci_legacy_resume_early(dev);
 	}
 
@@ -589,8 +585,7 @@ static int pci_pm_poweroff(struct device
 
 static int pci_pm_poweroff_noirq(struct device *dev)
 {
-	struct pci_dev *pci_dev = to_pci_dev(dev);
-	struct pci_driver *drv = pci_dev->driver;
+	struct device_driver *drv = dev->driver;
 	int error = 0;
 
 	if (drv && drv->pm) {
@@ -625,7 +620,7 @@ static int pci_pm_restore(struct device 
 static int pci_pm_restore_noirq(struct device *dev)
 {
 	struct pci_dev *pci_dev = to_pci_dev(dev);
-	struct pci_driver *drv = pci_dev->driver;
+	struct device_driver *drv = dev->driver;
 	int error = 0;
 
 	pci_fixup_device(pci_fixup_resume, pci_dev);
@@ -654,17 +649,15 @@ static int pci_pm_restore_noirq(struct d
 
 #endif /* !CONFIG_HIBERNATION */
 
-struct pm_ext_ops pci_pm_ops = {
-	.base = {
-		.prepare = pci_pm_prepare,
-		.complete = pci_pm_complete,
-		.suspend = pci_pm_suspend,
-		.resume = pci_pm_resume,
-		.freeze = pci_pm_freeze,
-		.thaw = pci_pm_thaw,
-		.poweroff = pci_pm_poweroff,
-		.restore = pci_pm_restore,
-	},
+struct dev_pm_ops pci_dev_pm_ops = {
+	.prepare = pci_pm_prepare,
+	.complete = pci_pm_complete,
+	.suspend = pci_pm_suspend,
+	.resume = pci_pm_resume,
+	.freeze = pci_pm_freeze,
+	.thaw = pci_pm_thaw,
+	.poweroff = pci_pm_poweroff,
+	.restore = pci_pm_restore,
 	.suspend_noirq = pci_pm_suspend_noirq,
 	.resume_noirq = pci_pm_resume_noirq,
 	.freeze_noirq = pci_pm_freeze_noirq,
@@ -673,7 +666,7 @@ struct pm_ext_ops pci_pm_ops = {
 	.restore_noirq = pci_pm_restore_noirq,
 };
 
-#define PCI_PM_OPS_PTR	&pci_pm_ops
+#define PCI_PM_OPS_PTR	&pci_dev_pm_ops
 
 #else /* !CONFIG_PM_SLEEP */
 
@@ -703,9 +696,6 @@ int __pci_register_driver(struct pci_dri
 	drv->driver.owner = owner;
 	drv->driver.mod_name = mod_name;
 
-	if (drv->pm)
-		drv->driver.pm = &drv->pm->base;
-
 	spin_lock_init(&drv->dynids.lock);
 	INIT_LIST_HEAD(&drv->dynids.list);
 
Index: linux-2.6/drivers/base/platform.c
===================================================================
--- linux-2.6.orig/drivers/base/platform.c
+++ linux-2.6/drivers/base/platform.c
@@ -453,8 +453,6 @@ int platform_driver_register(struct plat
 		drv->driver.suspend = platform_drv_suspend;
 	if (drv->resume)
 		drv->driver.resume = platform_drv_resume;
-	if (drv->pm)
-		drv->driver.pm = &drv->pm->base;
 	return driver_register(&drv->driver);
 }
 EXPORT_SYMBOL_GPL(platform_driver_register);
@@ -636,7 +634,10 @@ static int platform_pm_suspend(struct de
 	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (drv && drv->pm) {
+	if (!drv)
+		return 0;
+
+	if (drv->pm) {
 		if (drv->pm->suspend)
 			ret = drv->pm->suspend(dev);
 	} else {
@@ -648,16 +649,15 @@ static int platform_pm_suspend(struct de
 
 static int platform_pm_suspend_noirq(struct device *dev)
 {
-	struct platform_driver *pdrv;
+	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (!dev->driver)
+	if (!drv)
 		return 0;
 
-	pdrv = to_platform_driver(dev->driver);
-	if (pdrv->pm) {
-		if (pdrv->pm->suspend_noirq)
-			ret = pdrv->pm->suspend_noirq(dev);
+	if (drv->pm) {
+		if (drv->pm->suspend_noirq)
+			ret = drv->pm->suspend_noirq(dev);
 	} else {
 		ret = platform_legacy_suspend_late(dev, PMSG_SUSPEND);
 	}
@@ -670,7 +670,10 @@ static int platform_pm_resume(struct dev
 	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (drv && drv->pm) {
+	if (!drv)
+		return 0;
+
+	if (drv->pm) {
 		if (drv->pm->resume)
 			ret = drv->pm->resume(dev);
 	} else {
@@ -682,16 +685,15 @@ static int platform_pm_resume(struct dev
 
 static int platform_pm_resume_noirq(struct device *dev)
 {
-	struct platform_driver *pdrv;
+	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (!dev->driver)
+	if (!drv)
 		return 0;
 
-	pdrv = to_platform_driver(dev->driver);
-	if (pdrv->pm) {
-		if (pdrv->pm->resume_noirq)
-			ret = pdrv->pm->resume_noirq(dev);
+	if (drv->pm) {
+		if (drv->pm->resume_noirq)
+			ret = drv->pm->resume_noirq(dev);
 	} else {
 		ret = platform_legacy_resume_early(dev);
 	}
@@ -730,16 +732,15 @@ static int platform_pm_freeze(struct dev
 
 static int platform_pm_freeze_noirq(struct device *dev)
 {
-	struct platform_driver *pdrv;
+	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (!dev->driver)
+	if (!drv)
 		return 0;
 
-	pdrv = to_platform_driver(dev->driver);
-	if (pdrv->pm) {
-		if (pdrv->pm->freeze_noirq)
-			ret = pdrv->pm->freeze_noirq(dev);
+	if (drv->pm) {
+		if (drv->pm->freeze_noirq)
+			ret = drv->pm->freeze_noirq(dev);
 	} else {
 		ret = platform_legacy_suspend_late(dev, PMSG_FREEZE);
 	}
@@ -752,7 +753,10 @@ static int platform_pm_thaw(struct devic
 	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (drv && drv->pm) {
+	if (!drv)
+		return 0;
+
+	if (drv->pm) {
 		if (drv->pm->thaw)
 			ret = drv->pm->thaw(dev);
 	} else {
@@ -764,16 +768,15 @@ static int platform_pm_thaw(struct devic
 
 static int platform_pm_thaw_noirq(struct device *dev)
 {
-	struct platform_driver *pdrv;
+	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (!dev->driver)
+	if (!drv)
 		return 0;
 
-	pdrv = to_platform_driver(dev->driver);
-	if (pdrv->pm) {
-		if (pdrv->pm->thaw_noirq)
-			ret = pdrv->pm->thaw_noirq(dev);
+	if (drv->pm) {
+		if (drv->pm->thaw_noirq)
+			ret = drv->pm->thaw_noirq(dev);
 	} else {
 		ret = platform_legacy_resume_early(dev);
 	}
@@ -786,7 +789,10 @@ static int platform_pm_poweroff(struct d
 	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (drv && drv->pm) {
+	if (!drv)
+		return 0;
+
+	if (drv->pm) {
 		if (drv->pm->poweroff)
 			ret = drv->pm->poweroff(dev);
 	} else {
@@ -798,16 +804,15 @@ static int platform_pm_poweroff(struct d
 
 static int platform_pm_poweroff_noirq(struct device *dev)
 {
-	struct platform_driver *pdrv;
+	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (!dev->driver)
+	if (!drv)
 		return 0;
 
-	pdrv = to_platform_driver(dev->driver);
-	if (pdrv->pm) {
-		if (pdrv->pm->poweroff_noirq)
-			ret = pdrv->pm->poweroff_noirq(dev);
+	if (drv->pm) {
+		if (drv->pm->poweroff_noirq)
+			ret = drv->pm->poweroff_noirq(dev);
 	} else {
 		ret = platform_legacy_suspend_late(dev, PMSG_HIBERNATE);
 	}
@@ -820,7 +825,10 @@ static int platform_pm_restore(struct de
 	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (drv && drv->pm) {
+	if (!drv)
+		return 0;
+
+	if (drv->pm) {
 		if (drv->pm->restore)
 			ret = drv->pm->restore(dev);
 	} else {
@@ -832,16 +840,15 @@ static int platform_pm_restore(struct de
 
 static int platform_pm_restore_noirq(struct device *dev)
 {
-	struct platform_driver *pdrv;
+	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (!dev->driver)
+	if (!drv)
 		return 0;
 
-	pdrv = to_platform_driver(dev->driver);
-	if (pdrv->pm) {
-		if (pdrv->pm->restore_noirq)
-			ret = pdrv->pm->restore_noirq(dev);
+	if (drv->pm) {
+		if (drv->pm->restore_noirq)
+			ret = drv->pm->restore_noirq(dev);
 	} else {
 		ret = platform_legacy_resume_early(dev);
 	}
@@ -862,17 +869,15 @@ static int platform_pm_restore_noirq(str
 
 #endif /* !CONFIG_HIBERNATION */
 
-static struct pm_ext_ops platform_pm_ops = {
-	.base = {
-		.prepare = platform_pm_prepare,
-		.complete = platform_pm_complete,
-		.suspend = platform_pm_suspend,
-		.resume = platform_pm_resume,
-		.freeze = platform_pm_freeze,
-		.thaw = platform_pm_thaw,
-		.poweroff = platform_pm_poweroff,
-		.restore = platform_pm_restore,
-	},
+static struct dev_pm_ops platform_dev_pm_ops = {
+	.prepare = platform_pm_prepare,
+	.complete = platform_pm_complete,
+	.suspend = platform_pm_suspend,
+	.resume = platform_pm_resume,
+	.freeze = platform_pm_freeze,
+	.thaw = platform_pm_thaw,
+	.poweroff = platform_pm_poweroff,
+	.restore = platform_pm_restore,
 	.suspend_noirq = platform_pm_suspend_noirq,
 	.resume_noirq = platform_pm_resume_noirq,
 	.freeze_noirq = platform_pm_freeze_noirq,
@@ -881,7 +886,7 @@ static struct pm_ext_ops platform_pm_ops
 	.restore_noirq = platform_pm_restore_noirq,
 };
 
-#define PLATFORM_PM_OPS_PTR	&platform_pm_ops
+#define PLATFORM_PM_OPS_PTR	&platform_dev_pm_ops
 
 #else /* !CONFIG_PM_SLEEP */
 
Index: linux-2.6/drivers/usb/core/usb.c
===================================================================
--- linux-2.6.orig/drivers/usb/core/usb.c
+++ linux-2.6/drivers/usb/core/usb.c
@@ -286,7 +286,7 @@ static int usb_dev_restore(struct device
 	return usb_resume(dev);
 }
 
-static struct pm_ops usb_device_pm_ops = {
+static struct dev_pm_ops usb_device_pm_ops = {
 	.prepare =	usb_dev_prepare,
 	.complete =	usb_dev_complete,
 	.suspend =	usb_dev_suspend,
@@ -301,7 +301,7 @@ static struct pm_ops usb_device_pm_ops =
 
 #define ksuspend_usb_init()	0
 #define ksuspend_usb_cleanup()	do {} while (0)
-#define usb_device_pm_ops	(*(struct pm_ops *)0)
+#define usb_device_pm_ops	(*(struct dev_pm_ops *)0)
 
 #endif	/* CONFIG_PM */
 
Index: linux-2.6/drivers/base/power/main.c
===================================================================
--- linux-2.6.orig/drivers/base/power/main.c
+++ linux-2.6/drivers/base/power/main.c
@@ -112,7 +112,8 @@ void device_pm_remove(struct device *dev
  *	@ops:	PM operations to choose from.
  *	@state:	PM transition of the system being carried out.
  */
-static int pm_op(struct device *dev, struct pm_ops *ops, pm_message_t state)
+static int pm_op(struct device *dev, struct dev_pm_ops *ops,
+			pm_message_t state)
 {
 	int error = 0;
 
@@ -174,7 +175,7 @@ static int pm_op(struct device *dev, str
  *	The operation is executed with interrupts disabled by the only remaining
  *	functional CPU in the system.
  */
-static int pm_noirq_op(struct device *dev, struct pm_ext_ops *ops,
+static int pm_noirq_op(struct device *dev, struct dev_pm_ops *ops,
 			pm_message_t state)
 {
 	int error = 0;
@@ -355,7 +356,7 @@ static int resume_device(struct device *
 	if (dev->bus) {
 		if (dev->bus->pm) {
 			pm_dev_dbg(dev, state, "");
-			error = pm_op(dev, &dev->bus->pm->base, state);
+			error = pm_op(dev, dev->bus->pm, state);
 		} else if (dev->bus->resume) {
 			pm_dev_dbg(dev, state, "legacy ");
 			error = dev->bus->resume(dev);
@@ -440,9 +441,9 @@ static void complete_device(struct devic
 		dev->type->pm->complete(dev);
 	}
 
-	if (dev->bus && dev->bus->pm && dev->bus->pm->base.complete) {
+	if (dev->bus && dev->bus->pm && dev->bus->pm->complete) {
 		pm_dev_dbg(dev, state, "completing ");
-		dev->bus->pm->base.complete(dev);
+		dev->bus->pm->complete(dev);
 	}
 
 	up(&dev->sem);
@@ -600,7 +601,7 @@ static int suspend_device(struct device 
 	if (dev->bus) {
 		if (dev->bus->pm) {
 			pm_dev_dbg(dev, state, "");
-			error = pm_op(dev, &dev->bus->pm->base, state);
+			error = pm_op(dev, dev->bus->pm, state);
 		} else if (dev->bus->suspend) {
 			pm_dev_dbg(dev, state, "legacy ");
 			error = dev->bus->suspend(dev, state);
@@ -661,10 +662,10 @@ static int prepare_device(struct device 
 
 	down(&dev->sem);
 
-	if (dev->bus && dev->bus->pm && dev->bus->pm->base.prepare) {
+	if (dev->bus && dev->bus->pm && dev->bus->pm->prepare) {
 		pm_dev_dbg(dev, state, "preparing ");
-		error = dev->bus->pm->base.prepare(dev);
-		suspend_report_result(dev->bus->pm->base.prepare, error);
+		error = dev->bus->pm->prepare(dev);
+		suspend_report_result(dev->bus->pm->prepare, error);
 		if (error)
 			goto End;
 	}

Re: [PATCH -next] PM: Simplify the new suspend/hibernation framework for devices

[Rafael J. Wysocki]

On Monday, 6 of October 2008, Rafael J. Wysocki wrote:
> Hi Greg,
> 
> I promised at the KS that I would simplify the new suspend/hibernation
> framework for devices to avoid the confusion with two types of PM
> operations and pointers to PM operations from too many places.
> 
> The appended patch is intended for this purpose.  Unfortunately, I can't
> split it into subsystem-related patches, because compilation would be broken
> between them.
> 
> The patch applies to linux-next, but it's trivial to make it apply to the
> mainline.  It's been compiled on x86 (both 32-bit and 64-bit) and tested
> on hp nx6325, doesn't appear to break anything.

This one had a checkpatch.pl problem, sorry for that.  Updated patch is
appended.

Thanks,
Rafael


---
From: Rafael J. Wysocki <rjw@sisk.pl>

PM: Simplify the new suspend/hibernation framework for devices

Following the discussion at the Kernel Summit, simplify the new
device PM framework by merging 'struct pm_ops' and
'struct pm_ext_ops' and removing pointers to 'struct pm_ext_ops'
from 'struct platform_driver' and 'struct pci_driver'.

After this change, the suspend/hibernation callbacks will only
reside in 'struct device_driver' as well as at the bus type/
device class/device type level.  Accordingly, PCI and platform
device drivers are now expected to put their suspend/hibernation
callbacks into the 'struct device_driver' embedded in
'struct pci_driver' or 'struct platform_driver', respectively.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
---
 drivers/base/platform.c         |  115 ++++++++++++++++++++--------------------
 drivers/base/power/main.c       |   19 +++---
 drivers/pci/pci-driver.c        |   46 ++++++----------
 drivers/usb/core/usb.c          |    4 -
 include/linux/device.h          |    8 +-
 include/linux/pci.h             |    1 
 include/linux/platform_device.h |    1 
 include/linux/pm.h              |   76 ++++++++------------------
 8 files changed, 119 insertions(+), 151 deletions(-)

Index: linux-2.6/include/linux/pm.h
===================================================================
--- linux-2.6.orig/include/linux/pm.h
+++ linux-2.6/include/linux/pm.h
@@ -41,7 +41,7 @@ typedef struct pm_message {
 } pm_message_t;
 
 /**
- * struct pm_ops - device PM callbacks
+ * struct dev_pm_ops - device PM callbacks
  *
  * Several driver power state transitions are externally visible, affecting
  * the state of pending I/O queues and (for drivers that touch hardware)
@@ -126,46 +126,6 @@ typedef struct pm_message {
  *	On most platforms, there are no restrictions on availability of
  *	resources like clocks during @restore().
  *
- * All of the above callbacks, except for @complete(), return error codes.
- * However, the error codes returned by the resume operations, @resume(),
- * @thaw(), and @restore(), do not cause the PM core to abort the resume
- * transition during which they are returned.  The error codes returned in
- * that cases are only printed by the PM core to the system logs for debugging
- * purposes.  Still, it is recommended that drivers only return error codes
- * from their resume methods in case of an unrecoverable failure (i.e. when the
- * device being handled refuses to resume and becomes unusable) to allow us to
- * modify the PM core in the future, so that it can avoid attempting to handle
- * devices that failed to resume and their children.
- *
- * It is allowed to unregister devices while the above callbacks are being
- * executed.  However, it is not allowed to unregister a device from within any
- * of its own callbacks.
- */
-
-struct pm_ops {
-	int (*prepare)(struct device *dev);
-	void (*complete)(struct device *dev);
-	int (*suspend)(struct device *dev);
-	int (*resume)(struct device *dev);
-	int (*freeze)(struct device *dev);
-	int (*thaw)(struct device *dev);
-	int (*poweroff)(struct device *dev);
-	int (*restore)(struct device *dev);
-};
-
-/**
- * struct pm_ext_ops - extended device PM callbacks
- *
- * Some devices require certain operations related to suspend and hibernation
- * to be carried out with interrupts disabled.  Thus, 'struct pm_ext_ops' below
- * is defined, adding callbacks to be executed with interrupts disabled to
- * 'struct pm_ops'.
- *
- * The following callbacks included in 'struct pm_ext_ops' are executed with
- * the nonboot CPUs switched off and with interrupts disabled on the only
- * functional CPU.  They also are executed with the PM core list of devices
- * locked, so they must NOT unregister any devices.
- *
  * @suspend_noirq: Complete the operations of ->suspend() by carrying out any
  *	actions required for suspending the device that need interrupts to be
  *	disabled
@@ -190,18 +150,32 @@ struct pm_ops {
  *	actions required for restoring the operations of the device that need
  *	interrupts to be disabled
  *
- * All of the above callbacks return error codes, but the error codes returned
- * by the resume operations, @resume_noirq(), @thaw_noirq(), and
- * @restore_noirq(), do not cause the PM core to abort the resume transition
- * during which they are returned.  The error codes returned in that cases are
- * only printed by the PM core to the system logs for debugging purposes.
- * Still, as stated above, it is recommended that drivers only return error
- * codes from their resume methods if the device being handled fails to resume
- * and is not usable any more.
+ * All of the above callbacks, except for @complete(), return error codes.
+ * However, the error codes returned by the resume operations, @resume(),
+ * @thaw(), @restore(), @resume_noirq(), @thaw_noirq(), and @restore_noirq() do
+ * not cause the PM core to abort the resume transition during which they are
+ * returned.  The error codes returned in that cases are only printed by the PM
+ * core to the system logs for debugging purposes.  Still, it is recommended
+ * that drivers only return error codes from their resume methods in case of an
+ * unrecoverable failure (i.e. when the device being handled refuses to resume
+ * and becomes unusable) to allow us to modify the PM core in the future, so
+ * that it can avoid attempting to handle devices that failed to resume and
+ * their children.
+ *
+ * It is allowed to unregister devices while the above callbacks are being
+ * executed.  However, it is not allowed to unregister a device from within any
+ * of its own callbacks.
  */
 
-struct pm_ext_ops {
-	struct pm_ops base;
+struct dev_pm_ops {
+	int (*prepare)(struct device *dev);
+	void (*complete)(struct device *dev);
+	int (*suspend)(struct device *dev);
+	int (*resume)(struct device *dev);
+	int (*freeze)(struct device *dev);
+	int (*thaw)(struct device *dev);
+	int (*poweroff)(struct device *dev);
+	int (*restore)(struct device *dev);
 	int (*suspend_noirq)(struct device *dev);
 	int (*resume_noirq)(struct device *dev);
 	int (*freeze_noirq)(struct device *dev);
Index: linux-2.6/include/linux/device.h
===================================================================
--- linux-2.6.orig/include/linux/device.h
+++ linux-2.6/include/linux/device.h
@@ -65,7 +65,7 @@ struct bus_type {
 	int (*resume_early)(struct device *dev);
 	int (*resume)(struct device *dev);
 
-	struct pm_ext_ops *pm;
+	struct dev_pm_ops *pm;
 
 	struct bus_type_private *p;
 };
@@ -133,7 +133,7 @@ struct device_driver {
 	int (*resume) (struct device *dev);
 	struct attribute_group **groups;
 
-	struct pm_ops *pm;
+	struct dev_pm_ops *pm;
 
 	struct driver_private *p;
 };
@@ -201,7 +201,7 @@ struct class {
 	enum sysfs_tag_type tag_type;
 	const void *(*sysfs_tag)(struct device *dev);
 
-	struct pm_ops *pm;
+	struct dev_pm_ops *pm;
 	struct class_private *p;
 };
 
@@ -291,7 +291,7 @@ struct device_type {
 	int (*uevent)(struct device *dev, struct kobj_uevent_env *env);
 	void (*release)(struct device *dev);
 
-	struct pm_ops *pm;
+	struct dev_pm_ops *pm;
 };
 
 /* interface for exporting device attributes */
Index: linux-2.6/include/linux/platform_device.h
===================================================================
--- linux-2.6.orig/include/linux/platform_device.h
+++ linux-2.6/include/linux/platform_device.h
@@ -53,7 +53,6 @@ struct platform_driver {
 	int (*suspend_late)(struct platform_device *, pm_message_t state);
 	int (*resume_early)(struct platform_device *);
 	int (*resume)(struct platform_device *);
-	struct pm_ext_ops *pm;
 	struct device_driver driver;
 };
 
Index: linux-2.6/include/linux/pci.h
===================================================================
--- linux-2.6.orig/include/linux/pci.h
+++ linux-2.6/include/linux/pci.h
@@ -414,7 +414,6 @@ struct pci_driver {
 	int  (*resume_early) (struct pci_dev *dev);
 	int  (*resume) (struct pci_dev *dev);	                /* Device woken up */
 	void (*shutdown) (struct pci_dev *dev);
-	struct pm_ext_ops *pm;
 	struct pci_error_handlers *err_handler;
 	struct device_driver	driver;
 	struct pci_dynids dynids;
Index: linux-2.6/drivers/pci/pci-driver.c
===================================================================
--- linux-2.6.orig/drivers/pci/pci-driver.c
+++ linux-2.6/drivers/pci/pci-driver.c
@@ -433,8 +433,7 @@ static int pci_pm_suspend(struct device 
 
 static int pci_pm_suspend_noirq(struct device *dev)
 {
-	struct pci_dev *pci_dev = to_pci_dev(dev);
-	struct pci_driver *drv = pci_dev->driver;
+	struct device_driver *drv = dev->driver;
 	int error = 0;
 
 	if (drv && drv->pm) {
@@ -469,11 +468,10 @@ static int pci_pm_resume(struct device *
 
 static int pci_pm_resume_noirq(struct device *dev)
 {
-	struct pci_dev *pci_dev = to_pci_dev(dev);
-	struct pci_driver *drv = pci_dev->driver;
+	struct device_driver *drv = dev->driver;
 	int error = 0;
 
-	pci_fixup_device(pci_fixup_resume_early, pci_dev);
+	pci_fixup_device(pci_fixup_resume_early, to_pci_dev(dev));
 
 	if (drv && drv->pm) {
 		if (drv->pm->resume_noirq)
@@ -519,8 +517,7 @@ static int pci_pm_freeze(struct device *
 
 static int pci_pm_freeze_noirq(struct device *dev)
 {
-	struct pci_dev *pci_dev = to_pci_dev(dev);
-	struct pci_driver *drv = pci_dev->driver;
+	struct device_driver *drv = dev->driver;
 	int error = 0;
 
 	if (drv && drv->pm) {
@@ -553,15 +550,14 @@ static int pci_pm_thaw(struct device *de
 
 static int pci_pm_thaw_noirq(struct device *dev)
 {
-	struct pci_dev *pci_dev = to_pci_dev(dev);
-	struct pci_driver *drv = pci_dev->driver;
+	struct device_driver *drv = dev->driver;
 	int error = 0;
 
 	if (drv && drv->pm) {
 		if (drv->pm->thaw_noirq)
 			error = drv->pm->thaw_noirq(dev);
 	} else {
-		pci_fixup_device(pci_fixup_resume_early, pci_dev);
+		pci_fixup_device(pci_fixup_resume_early, to_pci_dev(dev));
 		error = pci_legacy_resume_early(dev);
 	}
 
@@ -589,8 +585,7 @@ static int pci_pm_poweroff(struct device
 
 static int pci_pm_poweroff_noirq(struct device *dev)
 {
-	struct pci_dev *pci_dev = to_pci_dev(dev);
-	struct pci_driver *drv = pci_dev->driver;
+	struct device_driver *drv = dev->driver;
 	int error = 0;
 
 	if (drv && drv->pm) {
@@ -625,7 +620,7 @@ static int pci_pm_restore(struct device 
 static int pci_pm_restore_noirq(struct device *dev)
 {
 	struct pci_dev *pci_dev = to_pci_dev(dev);
-	struct pci_driver *drv = pci_dev->driver;
+	struct device_driver *drv = dev->driver;
 	int error = 0;
 
 	pci_fixup_device(pci_fixup_resume, pci_dev);
@@ -654,17 +649,15 @@ static int pci_pm_restore_noirq(struct d
 
 #endif /* !CONFIG_HIBERNATION */
 
-struct pm_ext_ops pci_pm_ops = {
-	.base = {
-		.prepare = pci_pm_prepare,
-		.complete = pci_pm_complete,
-		.suspend = pci_pm_suspend,
-		.resume = pci_pm_resume,
-		.freeze = pci_pm_freeze,
-		.thaw = pci_pm_thaw,
-		.poweroff = pci_pm_poweroff,
-		.restore = pci_pm_restore,
-	},
+struct dev_pm_ops pci_dev_pm_ops = {
+	.prepare = pci_pm_prepare,
+	.complete = pci_pm_complete,
+	.suspend = pci_pm_suspend,
+	.resume = pci_pm_resume,
+	.freeze = pci_pm_freeze,
+	.thaw = pci_pm_thaw,
+	.poweroff = pci_pm_poweroff,
+	.restore = pci_pm_restore,
 	.suspend_noirq = pci_pm_suspend_noirq,
 	.resume_noirq = pci_pm_resume_noirq,
 	.freeze_noirq = pci_pm_freeze_noirq,
@@ -673,7 +666,7 @@ struct pm_ext_ops pci_pm_ops = {
 	.restore_noirq = pci_pm_restore_noirq,
 };
 
-#define PCI_PM_OPS_PTR	&pci_pm_ops
+#define PCI_PM_OPS_PTR	(&pci_dev_pm_ops)
 
 #else /* !CONFIG_PM_SLEEP */
 
@@ -703,9 +696,6 @@ int __pci_register_driver(struct pci_dri
 	drv->driver.owner = owner;
 	drv->driver.mod_name = mod_name;
 
-	if (drv->pm)
-		drv->driver.pm = &drv->pm->base;
-
 	spin_lock_init(&drv->dynids.lock);
 	INIT_LIST_HEAD(&drv->dynids.list);
 
Index: linux-2.6/drivers/base/platform.c
===================================================================
--- linux-2.6.orig/drivers/base/platform.c
+++ linux-2.6/drivers/base/platform.c
@@ -453,8 +453,6 @@ int platform_driver_register(struct plat
 		drv->driver.suspend = platform_drv_suspend;
 	if (drv->resume)
 		drv->driver.resume = platform_drv_resume;
-	if (drv->pm)
-		drv->driver.pm = &drv->pm->base;
 	return driver_register(&drv->driver);
 }
 EXPORT_SYMBOL_GPL(platform_driver_register);
@@ -636,7 +634,10 @@ static int platform_pm_suspend(struct de
 	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (drv && drv->pm) {
+	if (!drv)
+		return 0;
+
+	if (drv->pm) {
 		if (drv->pm->suspend)
 			ret = drv->pm->suspend(dev);
 	} else {
@@ -648,16 +649,15 @@ static int platform_pm_suspend(struct de
 
 static int platform_pm_suspend_noirq(struct device *dev)
 {
-	struct platform_driver *pdrv;
+	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (!dev->driver)
+	if (!drv)
 		return 0;
 
-	pdrv = to_platform_driver(dev->driver);
-	if (pdrv->pm) {
-		if (pdrv->pm->suspend_noirq)
-			ret = pdrv->pm->suspend_noirq(dev);
+	if (drv->pm) {
+		if (drv->pm->suspend_noirq)
+			ret = drv->pm->suspend_noirq(dev);
 	} else {
 		ret = platform_legacy_suspend_late(dev, PMSG_SUSPEND);
 	}
@@ -670,7 +670,10 @@ static int platform_pm_resume(struct dev
 	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (drv && drv->pm) {
+	if (!drv)
+		return 0;
+
+	if (drv->pm) {
 		if (drv->pm->resume)
 			ret = drv->pm->resume(dev);
 	} else {
@@ -682,16 +685,15 @@ static int platform_pm_resume(struct dev
 
 static int platform_pm_resume_noirq(struct device *dev)
 {
-	struct platform_driver *pdrv;
+	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (!dev->driver)
+	if (!drv)
 		return 0;
 
-	pdrv = to_platform_driver(dev->driver);
-	if (pdrv->pm) {
-		if (pdrv->pm->resume_noirq)
-			ret = pdrv->pm->resume_noirq(dev);
+	if (drv->pm) {
+		if (drv->pm->resume_noirq)
+			ret = drv->pm->resume_noirq(dev);
 	} else {
 		ret = platform_legacy_resume_early(dev);
 	}
@@ -730,16 +732,15 @@ static int platform_pm_freeze(struct dev
 
 static int platform_pm_freeze_noirq(struct device *dev)
 {
-	struct platform_driver *pdrv;
+	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (!dev->driver)
+	if (!drv)
 		return 0;
 
-	pdrv = to_platform_driver(dev->driver);
-	if (pdrv->pm) {
-		if (pdrv->pm->freeze_noirq)
-			ret = pdrv->pm->freeze_noirq(dev);
+	if (drv->pm) {
+		if (drv->pm->freeze_noirq)
+			ret = drv->pm->freeze_noirq(dev);
 	} else {
 		ret = platform_legacy_suspend_late(dev, PMSG_FREEZE);
 	}
@@ -752,7 +753,10 @@ static int platform_pm_thaw(struct devic
 	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (drv && drv->pm) {
+	if (!drv)
+		return 0;
+
+	if (drv->pm) {
 		if (drv->pm->thaw)
 			ret = drv->pm->thaw(dev);
 	} else {
@@ -764,16 +768,15 @@ static int platform_pm_thaw(struct devic
 
 static int platform_pm_thaw_noirq(struct device *dev)
 {
-	struct platform_driver *pdrv;
+	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (!dev->driver)
+	if (!drv)
 		return 0;
 
-	pdrv = to_platform_driver(dev->driver);
-	if (pdrv->pm) {
-		if (pdrv->pm->thaw_noirq)
-			ret = pdrv->pm->thaw_noirq(dev);
+	if (drv->pm) {
+		if (drv->pm->thaw_noirq)
+			ret = drv->pm->thaw_noirq(dev);
 	} else {
 		ret = platform_legacy_resume_early(dev);
 	}
@@ -786,7 +789,10 @@ static int platform_pm_poweroff(struct d
 	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (drv && drv->pm) {
+	if (!drv)
+		return 0;
+
+	if (drv->pm) {
 		if (drv->pm->poweroff)
 			ret = drv->pm->poweroff(dev);
 	} else {
@@ -798,16 +804,15 @@ static int platform_pm_poweroff(struct d
 
 static int platform_pm_poweroff_noirq(struct device *dev)
 {
-	struct platform_driver *pdrv;
+	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (!dev->driver)
+	if (!drv)
 		return 0;
 
-	pdrv = to_platform_driver(dev->driver);
-	if (pdrv->pm) {
-		if (pdrv->pm->poweroff_noirq)
-			ret = pdrv->pm->poweroff_noirq(dev);
+	if (drv->pm) {
+		if (drv->pm->poweroff_noirq)
+			ret = drv->pm->poweroff_noirq(dev);
 	} else {
 		ret = platform_legacy_suspend_late(dev, PMSG_HIBERNATE);
 	}
@@ -820,7 +825,10 @@ static int platform_pm_restore(struct de
 	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (drv && drv->pm) {
+	if (!drv)
+		return 0;
+
+	if (drv->pm) {
 		if (drv->pm->restore)
 			ret = drv->pm->restore(dev);
 	} else {
@@ -832,16 +840,15 @@ static int platform_pm_restore(struct de
 
 static int platform_pm_restore_noirq(struct device *dev)
 {
-	struct platform_driver *pdrv;
+	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (!dev->driver)
+	if (!drv)
 		return 0;
 
-	pdrv = to_platform_driver(dev->driver);
-	if (pdrv->pm) {
-		if (pdrv->pm->restore_noirq)
-			ret = pdrv->pm->restore_noirq(dev);
+	if (drv->pm) {
+		if (drv->pm->restore_noirq)
+			ret = drv->pm->restore_noirq(dev);
 	} else {
 		ret = platform_legacy_resume_early(dev);
 	}
@@ -862,17 +869,15 @@ static int platform_pm_restore_noirq(str
 
 #endif /* !CONFIG_HIBERNATION */
 
-static struct pm_ext_ops platform_pm_ops = {
-	.base = {
-		.prepare = platform_pm_prepare,
-		.complete = platform_pm_complete,
-		.suspend = platform_pm_suspend,
-		.resume = platform_pm_resume,
-		.freeze = platform_pm_freeze,
-		.thaw = platform_pm_thaw,
-		.poweroff = platform_pm_poweroff,
-		.restore = platform_pm_restore,
-	},
+static struct dev_pm_ops platform_dev_pm_ops = {
+	.prepare = platform_pm_prepare,
+	.complete = platform_pm_complete,
+	.suspend = platform_pm_suspend,
+	.resume = platform_pm_resume,
+	.freeze = platform_pm_freeze,
+	.thaw = platform_pm_thaw,
+	.poweroff = platform_pm_poweroff,
+	.restore = platform_pm_restore,
 	.suspend_noirq = platform_pm_suspend_noirq,
 	.resume_noirq = platform_pm_resume_noirq,
 	.freeze_noirq = platform_pm_freeze_noirq,
@@ -881,7 +886,7 @@ static struct pm_ext_ops platform_pm_ops
 	.restore_noirq = platform_pm_restore_noirq,
 };
 
-#define PLATFORM_PM_OPS_PTR	&platform_pm_ops
+#define PLATFORM_PM_OPS_PTR	(&platform_dev_pm_ops)
 
 #else /* !CONFIG_PM_SLEEP */
 
Index: linux-2.6/drivers/usb/core/usb.c
===================================================================
--- linux-2.6.orig/drivers/usb/core/usb.c
+++ linux-2.6/drivers/usb/core/usb.c
@@ -286,7 +286,7 @@ static int usb_dev_restore(struct device
 	return usb_resume(dev);
 }
 
-static struct pm_ops usb_device_pm_ops = {
+static struct dev_pm_ops usb_device_pm_ops = {
 	.prepare =	usb_dev_prepare,
 	.complete =	usb_dev_complete,
 	.suspend =	usb_dev_suspend,
@@ -301,7 +301,7 @@ static struct pm_ops usb_device_pm_ops =
 
 #define ksuspend_usb_init()	0
 #define ksuspend_usb_cleanup()	do {} while (0)
-#define usb_device_pm_ops	(*(struct pm_ops *)0)
+#define usb_device_pm_ops	(*(struct dev_pm_ops *)0)
 
 #endif	/* CONFIG_PM */
 
Index: linux-2.6/drivers/base/power/main.c
===================================================================
--- linux-2.6.orig/drivers/base/power/main.c
+++ linux-2.6/drivers/base/power/main.c
@@ -112,7 +112,8 @@ void device_pm_remove(struct device *dev
  *	@ops:	PM operations to choose from.
  *	@state:	PM transition of the system being carried out.
  */
-static int pm_op(struct device *dev, struct pm_ops *ops, pm_message_t state)
+static int pm_op(struct device *dev, struct dev_pm_ops *ops,
+			pm_message_t state)
 {
 	int error = 0;
 
@@ -174,7 +175,7 @@ static int pm_op(struct device *dev, str
  *	The operation is executed with interrupts disabled by the only remaining
  *	functional CPU in the system.
  */
-static int pm_noirq_op(struct device *dev, struct pm_ext_ops *ops,
+static int pm_noirq_op(struct device *dev, struct dev_pm_ops *ops,
 			pm_message_t state)
 {
 	int error = 0;
@@ -355,7 +356,7 @@ static int resume_device(struct device *
 	if (dev->bus) {
 		if (dev->bus->pm) {
 			pm_dev_dbg(dev, state, "");
-			error = pm_op(dev, &dev->bus->pm->base, state);
+			error = pm_op(dev, dev->bus->pm, state);
 		} else if (dev->bus->resume) {
 			pm_dev_dbg(dev, state, "legacy ");
 			error = dev->bus->resume(dev);
@@ -440,9 +441,9 @@ static void complete_device(struct devic
 		dev->type->pm->complete(dev);
 	}
 
-	if (dev->bus && dev->bus->pm && dev->bus->pm->base.complete) {
+	if (dev->bus && dev->bus->pm && dev->bus->pm->complete) {
 		pm_dev_dbg(dev, state, "completing ");
-		dev->bus->pm->base.complete(dev);
+		dev->bus->pm->complete(dev);
 	}
 
 	up(&dev->sem);
@@ -600,7 +601,7 @@ static int suspend_device(struct device 
 	if (dev->bus) {
 		if (dev->bus->pm) {
 			pm_dev_dbg(dev, state, "");
-			error = pm_op(dev, &dev->bus->pm->base, state);
+			error = pm_op(dev, dev->bus->pm, state);
 		} else if (dev->bus->suspend) {
 			pm_dev_dbg(dev, state, "legacy ");
 			error = dev->bus->suspend(dev, state);
@@ -661,10 +662,10 @@ static int prepare_device(struct device 
 
 	down(&dev->sem);
 
-	if (dev->bus && dev->bus->pm && dev->bus->pm->base.prepare) {
+	if (dev->bus && dev->bus->pm && dev->bus->pm->prepare) {
 		pm_dev_dbg(dev, state, "preparing ");
-		error = dev->bus->pm->base.prepare(dev);
-		suspend_report_result(dev->bus->pm->base.prepare, error);
+		error = dev->bus->pm->prepare(dev);
+		suspend_report_result(dev->bus->pm->prepare, error);
 		if (error)
 			goto End;
 	}

Re: [PATCH -next] PM: Simplify the new suspend/hibernation framework for devices

[Pavel Machek]

On Mon 2008-10-06 22:46:05, Rafael J. Wysocki wrote:
> On Monday, 6 of October 2008, Rafael J. Wysocki wrote:
> > Hi Greg,
> > 
> > I promised at the KS that I would simplify the new suspend/hibernation
> > framework for devices to avoid the confusion with two types of PM
> > operations and pointers to PM operations from too many places.
> > 
> > The appended patch is intended for this purpose.  Unfortunately, I can't
> > split it into subsystem-related patches, because compilation would be broken
> > between them.
> > 
> > The patch applies to linux-next, but it's trivial to make it apply to the
> > mainline.  It's been compiled on x86 (both 32-bit and 64-bit) and tested
> > on hp nx6325, doesn't appear to break anything.
> 
> This one had a checkpatch.pl problem, sorry for that.  Updated patch is
> appended.

Seems ok to me. ACK.

-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

Re: [PATCH -next] PM: Simplify the new suspend/hibernation framework for devices

[Greg KH]

On Mon, Oct 06, 2008 at 10:46:05PM +0200, Rafael J. Wysocki wrote:
> On Monday, 6 of October 2008, Rafael J. Wysocki wrote:
> > Hi Greg,
> > 
> > I promised at the KS that I would simplify the new suspend/hibernation
> > framework for devices to avoid the confusion with two types of PM
> > operations and pointers to PM operations from too many places.
> > 
> > The appended patch is intended for this purpose.  Unfortunately, I can't
> > split it into subsystem-related patches, because compilation would be broken
> > between them.
> > 
> > The patch applies to linux-next, but it's trivial to make it apply to the
> > mainline.  It's been compiled on x86 (both 32-bit and 64-bit) and tested
> > on hp nx6325, doesn't appear to break anything.
> 
> This one had a checkpatch.pl problem, sorry for that.  Updated patch is
> appended.

I've added this to my tree (Jesse, is this ok, as it does have a PCI
portion?)

But it's too late for .28, especially due to the -next tree not up and
running right now.  I'll let it bake in -mm and -next and it should go
into .29.

Is that ok?

thanks,

greg k-h

Re: [PATCH -next] PM: Simplify the new suspend/hibernation framework for devices

[Greg KH]

On Sat, Oct 11, 2008 at 12:47:34AM +0200, Rafael J. Wysocki wrote:
> On Friday, 10 of October 2008, Greg KH wrote:
> > On Mon, Oct 06, 2008 at 10:46:05PM +0200, Rafael J. Wysocki wrote:
> > > On Monday, 6 of October 2008, Rafael J. Wysocki wrote:
> > > > Hi Greg,
> > > > 
> > > > I promised at the KS that I would simplify the new suspend/hibernation
> > > > framework for devices to avoid the confusion with two types of PM
> > > > operations and pointers to PM operations from too many places.
> > > > 
> > > > The appended patch is intended for this purpose.  Unfortunately, I can't
> > > > split it into subsystem-related patches, because compilation would be broken
> > > > between them.
> > > > 
> > > > The patch applies to linux-next, but it's trivial to make it apply to the
> > > > mainline.  It's been compiled on x86 (both 32-bit and 64-bit) and tested
> > > > on hp nx6325, doesn't appear to break anything.
> > > 
> > > This one had a checkpatch.pl problem, sorry for that.  Updated patch is
> > > appended.
> > 
> > I've added this to my tree (Jesse, is this ok, as it does have a PCI
> > portion?)
> > 
> > But it's too late for .28, especially due to the -next tree not up and
> > running right now.  I'll let it bake in -mm and -next and it should go
> > into .29.
> > 
> > Is that ok?
> 
> Well, if anyone pushes anything depending on this framework for .27, that will
> become a !@#$%^&* mess (we've had this problem once already).

Do we have any drivers that depend on this framework in any tree?  Is
anyone needing this now?

thanks,

greg k-h

Re: [PATCH -next] PM: Simplify the new suspend/hibernation framework for devices

[Rafael J. Wysocki]

On Friday, 10 of October 2008, Greg KH wrote:
> On Mon, Oct 06, 2008 at 10:46:05PM +0200, Rafael J. Wysocki wrote:
> > On Monday, 6 of October 2008, Rafael J. Wysocki wrote:
> > > Hi Greg,
> > > 
> > > I promised at the KS that I would simplify the new suspend/hibernation
> > > framework for devices to avoid the confusion with two types of PM
> > > operations and pointers to PM operations from too many places.
> > > 
> > > The appended patch is intended for this purpose.  Unfortunately, I can't
> > > split it into subsystem-related patches, because compilation would be broken
> > > between them.
> > > 
> > > The patch applies to linux-next, but it's trivial to make it apply to the
> > > mainline.  It's been compiled on x86 (both 32-bit and 64-bit) and tested
> > > on hp nx6325, doesn't appear to break anything.
> > 
> > This one had a checkpatch.pl problem, sorry for that.  Updated patch is
> > appended.
> 
> I've added this to my tree (Jesse, is this ok, as it does have a PCI
> portion?)
> 
> But it's too late for .28, especially due to the -next tree not up and
> running right now.  I'll let it bake in -mm and -next and it should go
> into .29.
> 
> Is that ok?

Well, if anyone pushes anything depending on this framework for .27, that will
become a !@#$%^&* mess (we've had this problem once already).

Other than that, fine by me.

Thanks,
Rafael

Re: [PATCH -next] PM: Simplify the new suspend/hibernation framework for devices

[Rafael J. Wysocki]

On Saturday, 11 of October 2008, Greg KH wrote:
> On Sat, Oct 11, 2008 at 12:47:34AM +0200, Rafael J. Wysocki wrote:
> > On Friday, 10 of October 2008, Greg KH wrote:
> > > On Mon, Oct 06, 2008 at 10:46:05PM +0200, Rafael J. Wysocki wrote:
> > > > On Monday, 6 of October 2008, Rafael J. Wysocki wrote:
> > > > > Hi Greg,
> > > > > 
> > > > > I promised at the KS that I would simplify the new suspend/hibernation
> > > > > framework for devices to avoid the confusion with two types of PM
> > > > > operations and pointers to PM operations from too many places.
> > > > > 
> > > > > The appended patch is intended for this purpose.  Unfortunately, I can't
> > > > > split it into subsystem-related patches, because compilation would be broken
> > > > > between them.
> > > > > 
> > > > > The patch applies to linux-next, but it's trivial to make it apply to the
> > > > > mainline.  It's been compiled on x86 (both 32-bit and 64-bit) and tested
> > > > > on hp nx6325, doesn't appear to break anything.
> > > > 
> > > > This one had a checkpatch.pl problem, sorry for that.  Updated patch is
> > > > appended.
> > > 
> > > I've added this to my tree (Jesse, is this ok, as it does have a PCI
> > > portion?)
> > > 
> > > But it's too late for .28, especially due to the -next tree not up and
> > > running right now.  I'll let it bake in -mm and -next and it should go
> > > into .29.
> > > 
> > > Is that ok?
> > 
> > Well, if anyone pushes anything depending on this framework for .27, that will
> > become a !@#$%^&* mess (we've had this problem once already).
> 
> Do we have any drivers that depend on this framework in any tree?  Is
> anyone needing this now?

Well, I thought that the Hannes' work would depend on it quite a bit. :-)

Apart from this no one should really need it right now.  I hope.

Thanks,
Rafael

Re: [PATCH -next] PM: Simplify the new suspend/hibernation framework for devices

[Greg KH]

On Sat, Oct 11, 2008 at 01:04:17AM +0200, Rafael J. Wysocki wrote:
> On Saturday, 11 of October 2008, Greg KH wrote:
> > On Sat, Oct 11, 2008 at 12:47:34AM +0200, Rafael J. Wysocki wrote:
> > > On Friday, 10 of October 2008, Greg KH wrote:
> > > > On Mon, Oct 06, 2008 at 10:46:05PM +0200, Rafael J. Wysocki wrote:
> > > > > On Monday, 6 of October 2008, Rafael J. Wysocki wrote:
> > > > > > Hi Greg,
> > > > > > 
> > > > > > I promised at the KS that I would simplify the new suspend/hibernation
> > > > > > framework for devices to avoid the confusion with two types of PM
> > > > > > operations and pointers to PM operations from too many places.
> > > > > > 
> > > > > > The appended patch is intended for this purpose.  Unfortunately, I can't
> > > > > > split it into subsystem-related patches, because compilation would be broken
> > > > > > between them.
> > > > > > 
> > > > > > The patch applies to linux-next, but it's trivial to make it apply to the
> > > > > > mainline.  It's been compiled on x86 (both 32-bit and 64-bit) and tested
> > > > > > on hp nx6325, doesn't appear to break anything.
> > > > > 
> > > > > This one had a checkpatch.pl problem, sorry for that.  Updated patch is
> > > > > appended.
> > > > 
> > > > I've added this to my tree (Jesse, is this ok, as it does have a PCI
> > > > portion?)
> > > > 
> > > > But it's too late for .28, especially due to the -next tree not up and
> > > > running right now.  I'll let it bake in -mm and -next and it should go
> > > > into .29.
> > > > 
> > > > Is that ok?
> > > 
> > > Well, if anyone pushes anything depending on this framework for .27, that will
> > > become a !@#$%^&* mess (we've had this problem once already).
> > 
> > Do we have any drivers that depend on this framework in any tree?  Is
> > anyone needing this now?
> 
> Well, I thought that the Hannes' work would depend on it quite a bit. :-)

Hannes's work is still in my TO-APPLY mbox, only his cleanup patches are
in the tree and queued up for .28.  Now that your patch is in, his can
be based off of this, and all of them go into .29.

Sound good?

thanks,

greg k-h

Re: [PATCH -next] PM: Simplify the new suspend/hibernation framework for devices

[Rafael J. Wysocki]

On Saturday, 11 of October 2008, Greg KH wrote:
> On Sat, Oct 11, 2008 at 01:04:17AM +0200, Rafael J. Wysocki wrote:
> > On Saturday, 11 of October 2008, Greg KH wrote:
> > > On Sat, Oct 11, 2008 at 12:47:34AM +0200, Rafael J. Wysocki wrote:
> > > > On Friday, 10 of October 2008, Greg KH wrote:
> > > > > On Mon, Oct 06, 2008 at 10:46:05PM +0200, Rafael J. Wysocki wrote:
> > > > > > On Monday, 6 of October 2008, Rafael J. Wysocki wrote:
> > > > > > > Hi Greg,
> > > > > > > 
> > > > > > > I promised at the KS that I would simplify the new suspend/hibernation
> > > > > > > framework for devices to avoid the confusion with two types of PM
> > > > > > > operations and pointers to PM operations from too many places.
> > > > > > > 
> > > > > > > The appended patch is intended for this purpose.  Unfortunately, I can't
> > > > > > > split it into subsystem-related patches, because compilation would be broken
> > > > > > > between them.
> > > > > > > 
> > > > > > > The patch applies to linux-next, but it's trivial to make it apply to the
> > > > > > > mainline.  It's been compiled on x86 (both 32-bit and 64-bit) and tested
> > > > > > > on hp nx6325, doesn't appear to break anything.
> > > > > > 
> > > > > > This one had a checkpatch.pl problem, sorry for that.  Updated patch is
> > > > > > appended.
> > > > > 
> > > > > I've added this to my tree (Jesse, is this ok, as it does have a PCI
> > > > > portion?)
> > > > > 
> > > > > But it's too late for .28, especially due to the -next tree not up and
> > > > > running right now.  I'll let it bake in -mm and -next and it should go
> > > > > into .29.
> > > > > 
> > > > > Is that ok?
> > > > 
> > > > Well, if anyone pushes anything depending on this framework for .27, that will
> > > > become a !@#$%^&* mess (we've had this problem once already).
> > > 
> > > Do we have any drivers that depend on this framework in any tree?  Is
> > > anyone needing this now?
> > 
> > Well, I thought that the Hannes' work would depend on it quite a bit. :-)
> 
> Hannes's work is still in my TO-APPLY mbox, only his cleanup patches are
> in the tree and queued up for .28.  Now that your patch is in, his can
> be based off of this, and all of them go into .29.
> 
> Sound good?

Yes, fine by me.

Thanks,
Rafael

Re: [PATCH -next] PM: Simplify the new suspend/hibernation framework for devices

[Jesse Barnes]

On Friday, October 10, 2008 2:52 pm Greg KH wrote:
> On Mon, Oct 06, 2008 at 10:46:05PM +0200, Rafael J. Wysocki wrote:
> > On Monday, 6 of October 2008, Rafael J. Wysocki wrote:
> > > Hi Greg,
> > >
> > > I promised at the KS that I would simplify the new suspend/hibernation
> > > framework for devices to avoid the confusion with two types of PM
> > > operations and pointers to PM operations from too many places.
> > >
> > > The appended patch is intended for this purpose.  Unfortunately, I
> > > can't split it into subsystem-related patches, because compilation
> > > would be broken between them.
> > >
> > > The patch applies to linux-next, but it's trivial to make it apply to
> > > the mainline.  It's been compiled on x86 (both 32-bit and 64-bit) and
> > > tested on hp nx6325, doesn't appear to break anything.
> >
> > This one had a checkpatch.pl problem, sorry for that.  Updated patch is
> > appended.
>
> I've added this to my tree (Jesse, is this ok, as it does have a PCI
> portion?)
>
> But it's too late for .28, especially due to the -next tree not up and
> running right now.  I'll let it bake in -mm and -next and it should go
> into .29.
>
> Is that ok?

There are a couple of trivial PCI bits, but it's fine if they come in through 
your tree for 2.6.29.

Thanks,
Jesse