Merge branch 'acpi-pm'

* acpi-pm:
  ACPI / PM: Rework and clean up acpi_dev_pm_get_state()
  ACPI / PM: Replace ACPI_STATE_D3 with ACPI_STATE_D3_COLD in device_pm.c
  ACPI / PM: Rename function acpi_device_power_state() and make it static
  ACPI / PM: acpi_processor_suspend() can be static
  xen / ACPI / sleep: Register an acpi_suspend_lowlevel callback.
  x86 / ACPI / sleep: Provide registration for acpi_suspend_lowlevel.

arch/x86/include/asm/acpi.h

@@ -111,7 +111,7 @@ static inline void acpi_disable_pci(void)
 }
 
 /* Low-level suspend routine. */
-extern int acpi_suspend_lowlevel(void);
+extern int (*acpi_suspend_lowlevel)(void);
 
 /* Physical address to resume after wakeup */
 #define acpi_wakeup_address ((unsigned long)(real_mode_header->wakeup_start))

arch/x86/kernel/acpi/boot.c

@@ -44,6 +44,7 @@
 #include <asm/mpspec.h>
 #include <asm/smp.h>
 
+#include "sleep.h" /* To include x86_acpi_suspend_lowlevel */
 static int __initdata acpi_force = 0;
 u32 acpi_rsdt_forced;
 int acpi_disabled;
@@ -559,6 +560,12 @@ static int acpi_register_gsi_ioapic(struct device *dev, u32 gsi,
 int (*__acpi_register_gsi)(struct device *dev, u32 gsi,
 			   int trigger, int polarity) = acpi_register_gsi_pic;
 
+#ifdef CONFIG_ACPI_SLEEP
+int (*acpi_suspend_lowlevel)(void) = x86_acpi_suspend_lowlevel;
+#else
+int (*acpi_suspend_lowlevel)(void);
+#endif
+
 /*
  * success: return IRQ number (>=0)
  * failure: return < 0

arch/x86/kernel/acpi/sleep.c

@@ -26,12 +26,12 @@ static char temp_stack[4096];
 #endif
 
 /**
- * acpi_suspend_lowlevel - save kernel state
+ * x86_acpi_suspend_lowlevel - save kernel state
  *
  * Create an identity mapped page table and copy the wakeup routine to
  * low memory.
  */
-int acpi_suspend_lowlevel(void)
+int x86_acpi_suspend_lowlevel(void)
 {
 	struct wakeup_header *header =
 		(struct wakeup_header *) __va(real_mode_header->wakeup_header);

arch/x86/kernel/acpi/sleep.h

@@ -15,3 +15,5 @@ extern unsigned long acpi_copy_wakeup_routine(unsigned long);
 extern void wakeup_long64(void);
 
 extern void do_suspend_lowlevel(void);
+
+extern int x86_acpi_suspend_lowlevel(void);

drivers/acpi/device_pm.c

@@ -419,62 +419,73 @@ bool acpi_bus_can_wakeup(acpi_handle handle)
 EXPORT_SYMBOL(acpi_bus_can_wakeup);
 
 /**
- * acpi_device_power_state - Get preferred power state of ACPI device.
+ * acpi_dev_pm_get_state - Get preferred power state of ACPI device.
  * @dev: Device whose preferred target power state to return.
  * @adev: ACPI device node corresponding to @dev.
  * @target_state: System state to match the resultant device state.
- * @d_max_in: Deepest low-power state to take into consideration.
+ * @d_min_p: Location to store the highest power state available to the device.
- * @d_min_p: Location to store the upper limit of the allowed states range.
+ * @d_max_p: Location to store the lowest power state available to the device.
- * Return value: Preferred power state of the device on success, -ENODEV
- * (if there's no 'struct acpi_device' for @dev) or -EINVAL on failure
  *
- * Find the lowest power (highest number) ACPI device power state that the
+ * Find the lowest power (highest number) and highest power (lowest number) ACPI
- * device can be in while the system is in the state represented by
+ * device power states that the device can be in while the system is in the
- * @target_state.  If @d_min_p is set, the highest power (lowest number) device
+ * state represented by @target_state.  Store the integer numbers representing
- * power state that @dev can be in for the given system sleep state is stored
+ * those stats in the memory locations pointed to by @d_max_p and @d_min_p,
- * at the location pointed to by it.
+ * respectively.
  *
  * Callers must ensure that @dev and @adev are valid pointers and that @adev
  * actually corresponds to @dev before using this function.
+ *
+ * Returns 0 on success or -ENODATA when one of the ACPI methods fails or
+ * returns a value that doesn't make sense.  The memory locations pointed to by
+ * @d_max_p and @d_min_p are only modified on success.
  */
-int acpi_device_power_state(struct device *dev, struct acpi_device *adev,
+static int acpi_dev_pm_get_state(struct device *dev, struct acpi_device *adev,
-			    u32 target_state, int d_max_in, int *d_min_p)
+				 u32 target_state, int *d_min_p, int *d_max_p)
 {
-	char acpi_method[] = "_SxD";
+	char method[] = { '_', 'S', '0' + target_state, 'D', '\0' };
-	unsigned long long d_min, d_max;
+	acpi_handle handle = adev->handle;
+	unsigned long long ret;
+	int d_min, d_max;
 	bool wakeup = false;
+	acpi_status status;
 
-	if (d_max_in < ACPI_STATE_D0 || d_max_in > ACPI_STATE_D3)
-		return -EINVAL;
-
-	if (d_max_in > ACPI_STATE_D3_HOT) {
-		enum pm_qos_flags_status stat;
-
-		stat = dev_pm_qos_flags(dev, PM_QOS_FLAG_NO_POWER_OFF);
-		if (stat == PM_QOS_FLAGS_ALL)
-			d_max_in = ACPI_STATE_D3_HOT;
-	}
-
-	acpi_method[2] = '0' + target_state;
 	/*
-	 * If the sleep state is S0, the lowest limit from ACPI is D3,
+	 * If the system state is S0, the lowest power state the device can be
-	 * but if the device has _S0W, we will use the value from _S0W
+	 * in is D3cold, unless the device has _S0W and is supposed to signal
-	 * as the lowest limit from ACPI.  Finally, we will constrain
+	 * wakeup, in which case the return value of _S0W has to be used as the
-	 * the lowest limit with the specified one.
+	 * lowest power state available to the device.
 	 */
 	d_min = ACPI_STATE_D0;
-	d_max = ACPI_STATE_D3;
+	d_max = ACPI_STATE_D3_COLD;
 
 	/*
 	 * If present, _SxD methods return the minimum D-state (highest power
 	 * state) we can use for the corresponding S-states.  Otherwise, the
 	 * minimum D-state is D0 (ACPI 3.x).
-	 *
-	 * NOTE: We rely on acpi_evaluate_integer() not clobbering the integer
-	 * provided -- that's our fault recovery, we ignore retval.
 	 */
 	if (target_state > ACPI_STATE_S0) {
-		acpi_evaluate_integer(adev->handle, acpi_method, NULL, &d_min);
+		/*
+		 * We rely on acpi_evaluate_integer() not clobbering the integer
+		 * provided if AE_NOT_FOUND is returned.
+		 */
+		ret = d_min;
+		status = acpi_evaluate_integer(handle, method, NULL, &ret);
+		if ((ACPI_FAILURE(status) && status != AE_NOT_FOUND)
+		    || ret > ACPI_STATE_D3_COLD)
+			return -ENODATA;
+
+		/*
+		 * We need to handle legacy systems where D3hot and D3cold are
+		 * the same and 3 is returned in both cases, so fall back to
+		 * D3cold if D3hot is not a valid state.
+		 */
+		if (!adev->power.states[ret].flags.valid) {
+			if (ret == ACPI_STATE_D3_HOT)
+				ret = ACPI_STATE_D3_COLD;
+			else
+				return -ENODATA;
+		}
+		d_min = ret;
 		wakeup = device_may_wakeup(dev) && adev->wakeup.flags.valid
 			&& adev->wakeup.sleep_state >= target_state;
 	} else if (dev_pm_qos_flags(dev, PM_QOS_FLAG_REMOTE_WAKEUP) !=
@@ -490,38 +501,30 @@ int acpi_device_power_state(struct device *dev, struct acpi_device *adev,
 	 * can wake the system.  _S0W may be valid, too.
 	 */
 	if (wakeup) {
-		acpi_status status;
+		method[3] = 'W';
-
+		status = acpi_evaluate_integer(handle, method, NULL, &ret);
-		acpi_method[3] = 'W';
+		if (status == AE_NOT_FOUND) {
-		status = acpi_evaluate_integer(adev->handle, acpi_method, NULL,
+			if (target_state > ACPI_STATE_S0)
-						&d_max);
-		if (ACPI_FAILURE(status)) {
-			if (target_state != ACPI_STATE_S0 ||
-			    status != AE_NOT_FOUND)
 				d_max = d_min;
-		} else if (d_max < d_min) {
+		} else if (ACPI_SUCCESS(status) && ret <= ACPI_STATE_D3_COLD) {
-			/* Warn the user of the broken DSDT */
+			/* Fall back to D3cold if ret is not a valid state. */
-			printk(KERN_WARNING "ACPI: Wrong value from %s\n",
+			if (!adev->power.states[ret].flags.valid)
-				acpi_method);
+				ret = ACPI_STATE_D3_COLD;
-			/* Sanitize it */
+
-			d_min = d_max;
+			d_max = ret > d_min ? ret : d_min;
+		} else {
+			return -ENODATA;
 		}
 	}
 
-	if (d_max_in < d_min)
-		return -EINVAL;
 	if (d_min_p)
 		*d_min_p = d_min;
-	/* constrain d_max with specified lowest limit (max number) */
+
-	if (d_max > d_max_in) {
+	if (d_max_p)
-		for (d_max = d_max_in; d_max > d_min; d_max--) {
+		*d_max_p = d_max;
-			if (adev->power.states[d_max].flags.valid)
+
-				break;
+	return 0;
-		}
-	}
-	return d_max;
 }
-EXPORT_SYMBOL_GPL(acpi_device_power_state);
 
 /**
  * acpi_pm_device_sleep_state - Get preferred power state of ACPI device.
@@ -529,7 +532,8 @@ EXPORT_SYMBOL_GPL(acpi_device_power_state);
  * @d_min_p: Location to store the upper limit of the allowed states range.
  * @d_max_in: Deepest low-power state to take into consideration.
  * Return value: Preferred power state of the device on success, -ENODEV
- * (if there's no 'struct acpi_device' for @dev) or -EINVAL on failure
+ * if there's no 'struct acpi_device' for @dev, -EINVAL if @d_max_in is
+ * incorrect, or -ENODATA on ACPI method failure.
  *
  * The caller must ensure that @dev is valid before using this function.
  */
@@ -537,14 +541,39 @@ int acpi_pm_device_sleep_state(struct device *dev, int *d_min_p, int d_max_in)
 {
 	acpi_handle handle = DEVICE_ACPI_HANDLE(dev);
 	struct acpi_device *adev;
+	int ret, d_max;
+
+	if (d_max_in < ACPI_STATE_D0 || d_max_in > ACPI_STATE_D3_COLD)
+		return -EINVAL;
+
+	if (d_max_in > ACPI_STATE_D3_HOT) {
+		enum pm_qos_flags_status stat;
+
+		stat = dev_pm_qos_flags(dev, PM_QOS_FLAG_NO_POWER_OFF);
+		if (stat == PM_QOS_FLAGS_ALL)
+			d_max_in = ACPI_STATE_D3_HOT;
+	}
 
 	if (!handle || acpi_bus_get_device(handle, &adev)) {
 		dev_dbg(dev, "ACPI handle without context in %s!\n", __func__);
 		return -ENODEV;
 	}
 
-	return acpi_device_power_state(dev, adev, acpi_target_system_state(),
+	ret = acpi_dev_pm_get_state(dev, adev, acpi_target_system_state(),
-				       d_max_in, d_min_p);
+				    d_min_p, &d_max);
+	if (ret)
+		return ret;
+
+	if (d_max_in < *d_min_p)
+		return -EINVAL;
+
+	if (d_max > d_max_in) {
+		for (d_max = d_max_in; d_max > *d_min_p; d_max--) {
+			if (adev->power.states[d_max].flags.valid)
+				break;
+		}
+	}
+	return d_max;
 }
 EXPORT_SYMBOL(acpi_pm_device_sleep_state);
 
@@ -695,17 +724,13 @@ struct acpi_device *acpi_dev_pm_get_node(struct device *dev)
 static int acpi_dev_pm_low_power(struct device *dev, struct acpi_device *adev,
 				 u32 system_state)
 {
-	int power_state;
+	int ret, state;
 
 	if (!acpi_device_power_manageable(adev))
 		return 0;
 
-	power_state = acpi_device_power_state(dev, adev, system_state,
+	ret = acpi_dev_pm_get_state(dev, adev, system_state, NULL, &state);
-					      ACPI_STATE_D3, NULL);
+	return ret ? ret : acpi_device_set_power(adev, state);
-	if (power_state < ACPI_STATE_D0 || power_state > ACPI_STATE_D3)
-		return -EIO;
-
-	return acpi_device_set_power(adev, power_state);
 }
 
 /**

drivers/acpi/processor_idle.c

@@ -214,13 +214,13 @@ static void lapic_timer_state_broadcast(struct acpi_processor *pr,
 #ifdef CONFIG_PM_SLEEP
 static u32 saved_bm_rld;
 
-int acpi_processor_suspend(void)
+static int acpi_processor_suspend(void)
 {
 	acpi_read_bit_register(ACPI_BITREG_BUS_MASTER_RLD, &saved_bm_rld);
 	return 0;
 }
 
-void acpi_processor_resume(void)
+static void acpi_processor_resume(void)
 {
 	u32 resumed_bm_rld;
 

drivers/acpi/sleep.c

@@ -494,6 +494,8 @@ static int acpi_suspend_enter(suspend_state_t pm_state)
 		break;
 
 	case ACPI_STATE_S3:
+		if (!acpi_suspend_lowlevel)
+			return -ENOSYS;
 		error = acpi_suspend_lowlevel();
 		if (error)
 			return error;

include/acpi/acpi_bus.h

@@ -461,8 +461,6 @@ acpi_status acpi_add_pm_notifier(struct acpi_device *adev,
 				 acpi_notify_handler handler, void *context);
 acpi_status acpi_remove_pm_notifier(struct acpi_device *adev,
 				    acpi_notify_handler handler);
-int acpi_device_power_state(struct device *dev, struct acpi_device *adev,
-			    u32 target_state, int d_max_in, int *d_min_p);
 int acpi_pm_device_sleep_state(struct device *, int *, int);
 void acpi_dev_pm_add_dependent(acpi_handle handle, struct device *depdev);
 void acpi_dev_pm_remove_dependent(acpi_handle handle, struct device *depdev);
@@ -478,23 +476,13 @@ static inline acpi_status acpi_remove_pm_notifier(struct acpi_device *adev,
 {
 	return AE_SUPPORT;
 }
-static inline int __acpi_device_power_state(int m, int *p)
+static inline int acpi_pm_device_sleep_state(struct device *d, int *p, int m)
 {
 	if (p)
 		*p = ACPI_STATE_D0;
+
 	return (m >= ACPI_STATE_D0 && m <= ACPI_STATE_D3) ? m : ACPI_STATE_D0;
 }
-static inline int acpi_device_power_state(struct device *dev,
-					  struct acpi_device *adev,
-					  u32 target_state, int d_max_in,
-					  int *d_min_p)
-{
-	return __acpi_device_power_state(d_max_in, d_min_p);
-}
-static inline int acpi_pm_device_sleep_state(struct device *d, int *p, int m)
-{
-	return __acpi_device_power_state(m, p);
-}
 static inline void acpi_dev_pm_add_dependent(acpi_handle handle,
 					     struct device *depdev) {}
 static inline void acpi_dev_pm_remove_dependent(acpi_handle handle,

include/xen/acpi.h

@@ -78,11 +78,25 @@ static inline int xen_acpi_get_pxm(acpi_handle h)
 int xen_acpi_notify_hypervisor_state(u8 sleep_state,
 				     u32 pm1a_cnt, u32 pm1b_cnd);
 
+static inline int xen_acpi_suspend_lowlevel(void)
+{
+	/*
+	* Xen will save and restore CPU context, so
+	* we can skip that and just go straight to
+	* the suspend.
+	*/
+	acpi_enter_sleep_state(ACPI_STATE_S3);
+	return 0;
+}
+
 static inline void xen_acpi_sleep_register(void)
 {
-	if (xen_initial_domain())
+	if (xen_initial_domain()) {
 		acpi_os_set_prepare_sleep(
 			&xen_acpi_notify_hypervisor_state);
+
+		acpi_suspend_lowlevel = xen_acpi_suspend_lowlevel;
+	}
 }
 #else
 static inline void xen_acpi_sleep_register(void)