ACPI / PM: Rework device power management to follow ACPI 6
diff mbox

Message ID 5141050.0s0C74Tu0K@vostro.rjw.lan
State Not Applicable
Headers show

Commit Message

Rafael J. Wysocki May 6, 2015, 10:37 p.m. UTC
From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>

The ACPI 6 specification has made some changes in the device power
management area.  In particular:

 * The D3hot power state is now supposed to be always available
   (instead of D3cold) and D3cold is only regarded as valid if the
   _PR3 object is present for the given device.

 * The required ordering of transitions into power states deeper than
   D0 is now such that for a transition into state Dx the _PSx method
   is supposed to be executed first, if present, and the states of
   the power resources the device depends on are supposed to be
   changed after that.

 * It is now explicitly forbidden to transition devices from
   lower-power (deeper) into higher-power (shallower) power states
   other than D0.

Those changes have been made so the specification reflects the
Windows' device power management code that the vast majority of
systems using ACPI is validated against.

To avoid artificial differences in ACPI device power management
between Windows and Linux, modify the ACPI device power management
code to follow the new specification.  Add comments explaining the
code flow in some unclear places.

This only may affect some real corner cases in which the OS behavior
expected by the firmware is different from the Windows one, but that's
quite unlikely.  The transition ordering change affects transitions
to D1 and D2 which are rarely used (if at all) and into D3hot and
D3cold for devices actually having _PR3, but those are likely to
be validated against Windows anyway.  The other changes may affect
code calling acpi_device_get_power() or acpi_device_update_power()
where ACPI_STATE_D3_HOT may be returned instead of ACPI_STATE_D3_COLD
(that's why the ACPI fan driver needs to be updated too) and since
transitions into ACPI_STATE_D3_HOT may remove power now, it is better
to avoid this one in acpi_pm_device_sleep_state() if the "no power
off" PM QoS flag is set.

The only existing user of acpi_device_can_poweroff() really cares
about the case when _PR3 is present, so the change in that function
should not cause any problems to happen too.

A plus is that PCI_D3hot can be mapped to ACPI_STATE_D3_HOT
now and the compatibility with older systems should be covered
automatically.

In any case, if any real problems result from this, it still will
be better to follow the Windows' behavior (which now is reflected
by the specification too) in general and handle the cases when it
doesn't work via quirks.

Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
---
 drivers/acpi/device_pm.c |   97 ++++++++++++++++++++++++++++-------------------
 drivers/acpi/fan.c       |    5 +-
 drivers/acpi/power.c     |    3 -
 drivers/acpi/scan.c      |   26 ++----------
 drivers/pci/pci-acpi.c   |    2 
 include/acpi/acpi_bus.h  |    3 -
 6 files changed, 71 insertions(+), 65 deletions(-)


--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Comments

Rafael J. Wysocki May 14, 2015, 10:06 p.m. UTC | #1
On Thursday, May 07, 2015 12:37:13 AM Rafael J. Wysocki wrote:
> From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> 
> The ACPI 6 specification has made some changes in the device power
> management area.  In particular:
> 
>  * The D3hot power state is now supposed to be always available
>    (instead of D3cold) and D3cold is only regarded as valid if the
>    _PR3 object is present for the given device.
> 
>  * The required ordering of transitions into power states deeper than
>    D0 is now such that for a transition into state Dx the _PSx method
>    is supposed to be executed first, if present, and the states of
>    the power resources the device depends on are supposed to be
>    changed after that.
> 
>  * It is now explicitly forbidden to transition devices from
>    lower-power (deeper) into higher-power (shallower) power states
>    other than D0.
> 
> Those changes have been made so the specification reflects the
> Windows' device power management code that the vast majority of
> systems using ACPI is validated against.
> 
> To avoid artificial differences in ACPI device power management
> between Windows and Linux, modify the ACPI device power management
> code to follow the new specification.  Add comments explaining the
> code flow in some unclear places.
> 
> This only may affect some real corner cases in which the OS behavior
> expected by the firmware is different from the Windows one, but that's
> quite unlikely.  The transition ordering change affects transitions
> to D1 and D2 which are rarely used (if at all) and into D3hot and
> D3cold for devices actually having _PR3, but those are likely to
> be validated against Windows anyway.  The other changes may affect
> code calling acpi_device_get_power() or acpi_device_update_power()
> where ACPI_STATE_D3_HOT may be returned instead of ACPI_STATE_D3_COLD
> (that's why the ACPI fan driver needs to be updated too) and since
> transitions into ACPI_STATE_D3_HOT may remove power now, it is better
> to avoid this one in acpi_pm_device_sleep_state() if the "no power
> off" PM QoS flag is set.
> 
> The only existing user of acpi_device_can_poweroff() really cares
> about the case when _PR3 is present, so the change in that function
> should not cause any problems to happen too.
> 
> A plus is that PCI_D3hot can be mapped to ACPI_STATE_D3_HOT
> now and the compatibility with older systems should be covered
> automatically.
> 
> In any case, if any real problems result from this, it still will
> be better to follow the Windows' behavior (which now is reflected
> by the specification too) in general and handle the cases when it
> doesn't work via quirks.
> 
> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>

OK, so it looks like nobody has any comments, which in particular means
"no objections" to me.

I'm going to queue up this patch for 4.2, then.

> ---
>  drivers/acpi/device_pm.c |   97 ++++++++++++++++++++++++++++-------------------
>  drivers/acpi/fan.c       |    5 +-
>  drivers/acpi/power.c     |    3 -
>  drivers/acpi/scan.c      |   26 ++----------
>  drivers/pci/pci-acpi.c   |    2 
>  include/acpi/acpi_bus.h  |    3 -
>  6 files changed, 71 insertions(+), 65 deletions(-)
> 
> Index: linux-pm/drivers/acpi/power.c
> ===================================================================
> --- linux-pm.orig/drivers/acpi/power.c
> +++ linux-pm/drivers/acpi/power.c
> @@ -684,7 +684,8 @@ int acpi_power_get_inferred_state(struct
>  		}
>  	}
>  
> -	*state = ACPI_STATE_D3_COLD;
> +	*state = device->power.states[ACPI_STATE_D3_COLD].flags.valid ?
> +		ACPI_STATE_D3_COLD : ACPI_STATE_D3_HOT;
>  	return 0;
>  }
>  
> Index: linux-pm/include/acpi/acpi_bus.h
> ===================================================================
> --- linux-pm.orig/include/acpi/acpi_bus.h
> +++ linux-pm/include/acpi/acpi_bus.h
> @@ -272,7 +272,6 @@ struct acpi_device_power_flags {
>  struct acpi_device_power_state {
>  	struct {
>  		u8 valid:1;
> -		u8 os_accessible:1;
>  		u8 explicit_set:1;	/* _PSx present? */
>  		u8 reserved:6;
>  	} flags;
> @@ -602,7 +601,7 @@ static inline bool acpi_device_can_wakeu
>  
>  static inline bool acpi_device_can_poweroff(struct acpi_device *adev)
>  {
> -	return adev->power.states[ACPI_STATE_D3_COLD].flags.os_accessible;
> +	return adev->power.states[ACPI_STATE_D3_COLD].flags.valid;
>  }
>  
>  #else	/* CONFIG_ACPI */
> Index: linux-pm/drivers/acpi/scan.c
> ===================================================================
> --- linux-pm.orig/drivers/acpi/scan.c
> +++ linux-pm/drivers/acpi/scan.c
> @@ -1766,15 +1766,9 @@ static void acpi_bus_init_power_state(st
>  	if (acpi_has_method(device->handle, pathname))
>  		ps->flags.explicit_set = 1;
>  
> -	/*
> -	 * State is valid if there are means to put the device into it.
> -	 * D3hot is only valid if _PR3 present.
> -	 */
> -	if (!list_empty(&ps->resources)
> -	    || (ps->flags.explicit_set && state < ACPI_STATE_D3_HOT)) {
> +	/* State is valid if there are means to put the device into it. */
> +	if (!list_empty(&ps->resources) || ps->flags.explicit_set)
>  		ps->flags.valid = 1;
> -		ps->flags.os_accessible = 1;
> -	}
>  
>  	ps->power = -1;		/* Unknown - driver assigned */
>  	ps->latency = -1;	/* Unknown - driver assigned */
> @@ -1810,21 +1804,13 @@ static void acpi_bus_get_power_flags(str
>  		acpi_bus_init_power_state(device, i);
>  
>  	INIT_LIST_HEAD(&device->power.states[ACPI_STATE_D3_COLD].resources);
> +	if (!list_empty(&device->power.states[ACPI_STATE_D3_HOT].resources))
> +		device->power.states[ACPI_STATE_D3_COLD].flags.valid = 1;
>  
> -	/* Set defaults for D0 and D3 states (always valid) */
> +	/* Set defaults for D0 and D3hot states (always valid) */
>  	device->power.states[ACPI_STATE_D0].flags.valid = 1;
>  	device->power.states[ACPI_STATE_D0].power = 100;
> -	device->power.states[ACPI_STATE_D3_COLD].flags.valid = 1;
> -	device->power.states[ACPI_STATE_D3_COLD].power = 0;
> -
> -	/* Set D3cold's explicit_set flag if _PS3 exists. */
> -	if (device->power.states[ACPI_STATE_D3_HOT].flags.explicit_set)
> -		device->power.states[ACPI_STATE_D3_COLD].flags.explicit_set = 1;
> -
> -	/* Presence of _PS3 or _PRx means we can put the device into D3 cold */
> -	if (device->power.states[ACPI_STATE_D3_HOT].flags.explicit_set ||
> -			device->power.flags.power_resources)
> -		device->power.states[ACPI_STATE_D3_COLD].flags.os_accessible = 1;
> +	device->power.states[ACPI_STATE_D3_HOT].flags.valid = 1;
>  
>  	if (acpi_bus_init_power(device))
>  		device->flags.power_manageable = 0;
> Index: linux-pm/drivers/acpi/device_pm.c
> ===================================================================
> --- linux-pm.orig/drivers/acpi/device_pm.c
> +++ linux-pm/drivers/acpi/device_pm.c
> @@ -98,17 +98,16 @@ int acpi_device_get_power(struct acpi_de
>  
>  		/*
>  		 * The power resources settings may indicate a power state
> -		 * shallower than the actual power state of the device.
> +		 * shallower than the actual power state of the device, because
> +		 * the same power resources may be referenced by other devices.
>  		 *
> -		 * Moreover, on systems predating ACPI 4.0, if the device
> -		 * doesn't depend on any power resources and _PSC returns 3,
> -		 * that means "power off".  We need to maintain compatibility
> -		 * with those systems.
> +		 * For systems predating ACPI 4.0 we assume that D3hot is the
> +		 * deepest state that can be supported.
>  		 */
>  		if (psc > result && psc < ACPI_STATE_D3_COLD)
>  			result = psc;
>  		else if (result == ACPI_STATE_UNKNOWN)
> -			result = psc > ACPI_STATE_D2 ? ACPI_STATE_D3_COLD : psc;
> +			result = psc > ACPI_STATE_D2 ? ACPI_STATE_D3_HOT : psc;
>  	}
>  
>  	/*
> @@ -153,8 +152,8 @@ static int acpi_dev_pm_explicit_set(stru
>   */
>  int acpi_device_set_power(struct acpi_device *device, int state)
>  {
> +	int target_state = state;
>  	int result = 0;
> -	bool cut_power = false;
>  
>  	if (!device || !device->flags.power_manageable
>  	    || (state < ACPI_STATE_D0) || (state > ACPI_STATE_D3_COLD))
> @@ -169,11 +168,21 @@ int acpi_device_set_power(struct acpi_de
>  		return 0;
>  	}
>  
> -	if (!device->power.states[state].flags.valid) {
> +	if (state == ACPI_STATE_D3_COLD) {
> +		/*
> +		 * For transitions to D3cold we need to execute _PS3 and then
> +		 * possibly drop references to the power resources in use.
> +		 */
> +		state = ACPI_STATE_D3_HOT;
> +		/* If _PR3 is not available, use D3hot as the target state. */
> +		if (!device->power.states[ACPI_STATE_D3_COLD].flags.valid)
> +			target_state = state;
> +	} else if (!device->power.states[state].flags.valid) {
>  		dev_warn(&device->dev, "Power state %s not supported\n",
>  			 acpi_power_state_string(state));
>  		return -ENODEV;
>  	}
> +
>  	if (!device->power.flags.ignore_parent &&
>  	    device->parent && (state < device->parent->power.state)) {
>  		dev_warn(&device->dev,
> @@ -183,39 +192,38 @@ int acpi_device_set_power(struct acpi_de
>  		return -ENODEV;
>  	}
>  
> -	/* For D3cold we should first transition into D3hot. */
> -	if (state == ACPI_STATE_D3_COLD
> -	    && device->power.states[ACPI_STATE_D3_COLD].flags.os_accessible) {
> -		state = ACPI_STATE_D3_HOT;
> -		cut_power = true;
> -	}
> -
> -	if (state < device->power.state && state != ACPI_STATE_D0
> -	    && device->power.state >= ACPI_STATE_D3_HOT) {
> -		dev_warn(&device->dev,
> -			 "Cannot transition to non-D0 state from D3\n");
> -		return -ENODEV;
> -	}
> -
>  	/*
>  	 * Transition Power
>  	 * ----------------
> -	 * In accordance with the ACPI specification first apply power (via
> -	 * power resources) and then evaluate _PSx.
> +	 * In accordance with ACPI 6, _PSx is executed before manipulating power
> +	 * resources, unless the target state is D0, in which case _PS0 is
> +	 * supposed to be executed after turning the power resources on.
>  	 */
> -	if (device->power.flags.power_resources) {
> -		result = acpi_power_transition(device, state);
> +	if (state > ACPI_STATE_D0) {
> +		/*
> +		 * According to ACPI 6, devices cannot go from lower-power
> +		 * (deeper) states to higher-power (shallower) states.
> +		 */
> +		if (state < device->power.state) {
> +			dev_warn(&device->dev, "Cannot transition from %s to %s\n",
> +				 acpi_power_state_string(device->power.state),
> +				 acpi_power_state_string(state));
> +			return -ENODEV;
> +		}
> +
> +		result = acpi_dev_pm_explicit_set(device, state);
>  		if (result)
>  			goto end;
> -	}
> -	result = acpi_dev_pm_explicit_set(device, state);
> -	if (result)
> -		goto end;
>  
> -	if (cut_power) {
> -		device->power.state = state;
> -		state = ACPI_STATE_D3_COLD;
> -		result = acpi_power_transition(device, state);
> +		if (device->power.flags.power_resources)
> +			result = acpi_power_transition(device, target_state);
> +	} else {
> +		if (device->power.flags.power_resources) {
> +			result = acpi_power_transition(device, ACPI_STATE_D0);
> +			if (result)
> +				goto end;
> +		}
> +		result = acpi_dev_pm_explicit_set(device, ACPI_STATE_D0);
>  	}
>  
>   end:
> @@ -264,13 +272,24 @@ int acpi_bus_init_power(struct acpi_devi
>  		return result;
>  
>  	if (state < ACPI_STATE_D3_COLD && device->power.flags.power_resources) {
> +		/* Reference count the power resources. */
>  		result = acpi_power_on_resources(device, state);
>  		if (result)
>  			return result;
>  
> -		result = acpi_dev_pm_explicit_set(device, state);
> -		if (result)
> -			return result;
> +		if (state == ACPI_STATE_D0) {
> +			/*
> +			 * If _PSC is not present and the state inferred from
> +			 * power resources appears to be D0, it still may be
> +			 * necessary to execute _PS0 at this point, because
> +			 * another device using the same power resources may
> +			 * have been put into D0 previously and that's why we
> +			 * see D0 here.
> +			 */
> +			result = acpi_dev_pm_explicit_set(device, state);
> +			if (result)
> +				return result;
> +		}
>  	} else if (state == ACPI_STATE_UNKNOWN) {
>  		/*
>  		 * No power resources and missing _PSC?  Cross fingers and make
> @@ -603,12 +622,12 @@ int acpi_pm_device_sleep_state(struct de
>  	if (d_max_in < ACPI_STATE_D0 || d_max_in > ACPI_STATE_D3_COLD)
>  		return -EINVAL;
>  
> -	if (d_max_in > ACPI_STATE_D3_HOT) {
> +	if (d_max_in > ACPI_STATE_D2) {
>  		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;
> +			d_max_in = ACPI_STATE_D2;
>  	}
>  
>  	adev = ACPI_COMPANION(dev);
> Index: linux-pm/drivers/acpi/fan.c
> ===================================================================
> --- linux-pm.orig/drivers/acpi/fan.c
> +++ linux-pm/drivers/acpi/fan.c
> @@ -158,8 +158,9 @@ static int fan_get_state(struct acpi_dev
>  	if (result)
>  		return result;
>  
> -	*state = (acpi_state == ACPI_STATE_D3_COLD ? 0 :
> -		 (acpi_state == ACPI_STATE_D0 ? 1 : -1));
> +	*state = acpi_state == ACPI_STATE_D3_COLD
> +			|| acpi_state == ACPI_STATE_D3_HOT ?
> +		0 : (acpi_state == ACPI_STATE_D0 ? 1 : -1);
>  	return 0;
>  }
>  
> Index: linux-pm/drivers/pci/pci-acpi.c
> ===================================================================
> --- linux-pm.orig/drivers/pci/pci-acpi.c
> +++ linux-pm/drivers/pci/pci-acpi.c
> @@ -420,7 +420,7 @@ static int acpi_pci_set_power_state(stru
>  		[PCI_D0] = ACPI_STATE_D0,
>  		[PCI_D1] = ACPI_STATE_D1,
>  		[PCI_D2] = ACPI_STATE_D2,
> -		[PCI_D3hot] = ACPI_STATE_D3_COLD,
> +		[PCI_D3hot] = ACPI_STATE_D3_HOT,
>  		[PCI_D3cold] = ACPI_STATE_D3_COLD,
>  	};
>  	int error = -EINVAL;
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Patch
diff mbox

Index: linux-pm/drivers/acpi/power.c
===================================================================
--- linux-pm.orig/drivers/acpi/power.c
+++ linux-pm/drivers/acpi/power.c
@@ -684,7 +684,8 @@  int acpi_power_get_inferred_state(struct
 		}
 	}
 
-	*state = ACPI_STATE_D3_COLD;
+	*state = device->power.states[ACPI_STATE_D3_COLD].flags.valid ?
+		ACPI_STATE_D3_COLD : ACPI_STATE_D3_HOT;
 	return 0;
 }
 
Index: linux-pm/include/acpi/acpi_bus.h
===================================================================
--- linux-pm.orig/include/acpi/acpi_bus.h
+++ linux-pm/include/acpi/acpi_bus.h
@@ -272,7 +272,6 @@  struct acpi_device_power_flags {
 struct acpi_device_power_state {
 	struct {
 		u8 valid:1;
-		u8 os_accessible:1;
 		u8 explicit_set:1;	/* _PSx present? */
 		u8 reserved:6;
 	} flags;
@@ -602,7 +601,7 @@  static inline bool acpi_device_can_wakeu
 
 static inline bool acpi_device_can_poweroff(struct acpi_device *adev)
 {
-	return adev->power.states[ACPI_STATE_D3_COLD].flags.os_accessible;
+	return adev->power.states[ACPI_STATE_D3_COLD].flags.valid;
 }
 
 #else	/* CONFIG_ACPI */
Index: linux-pm/drivers/acpi/scan.c
===================================================================
--- linux-pm.orig/drivers/acpi/scan.c
+++ linux-pm/drivers/acpi/scan.c
@@ -1766,15 +1766,9 @@  static void acpi_bus_init_power_state(st
 	if (acpi_has_method(device->handle, pathname))
 		ps->flags.explicit_set = 1;
 
-	/*
-	 * State is valid if there are means to put the device into it.
-	 * D3hot is only valid if _PR3 present.
-	 */
-	if (!list_empty(&ps->resources)
-	    || (ps->flags.explicit_set && state < ACPI_STATE_D3_HOT)) {
+	/* State is valid if there are means to put the device into it. */
+	if (!list_empty(&ps->resources) || ps->flags.explicit_set)
 		ps->flags.valid = 1;
-		ps->flags.os_accessible = 1;
-	}
 
 	ps->power = -1;		/* Unknown - driver assigned */
 	ps->latency = -1;	/* Unknown - driver assigned */
@@ -1810,21 +1804,13 @@  static void acpi_bus_get_power_flags(str
 		acpi_bus_init_power_state(device, i);
 
 	INIT_LIST_HEAD(&device->power.states[ACPI_STATE_D3_COLD].resources);
+	if (!list_empty(&device->power.states[ACPI_STATE_D3_HOT].resources))
+		device->power.states[ACPI_STATE_D3_COLD].flags.valid = 1;
 
-	/* Set defaults for D0 and D3 states (always valid) */
+	/* Set defaults for D0 and D3hot states (always valid) */
 	device->power.states[ACPI_STATE_D0].flags.valid = 1;
 	device->power.states[ACPI_STATE_D0].power = 100;
-	device->power.states[ACPI_STATE_D3_COLD].flags.valid = 1;
-	device->power.states[ACPI_STATE_D3_COLD].power = 0;
-
-	/* Set D3cold's explicit_set flag if _PS3 exists. */
-	if (device->power.states[ACPI_STATE_D3_HOT].flags.explicit_set)
-		device->power.states[ACPI_STATE_D3_COLD].flags.explicit_set = 1;
-
-	/* Presence of _PS3 or _PRx means we can put the device into D3 cold */
-	if (device->power.states[ACPI_STATE_D3_HOT].flags.explicit_set ||
-			device->power.flags.power_resources)
-		device->power.states[ACPI_STATE_D3_COLD].flags.os_accessible = 1;
+	device->power.states[ACPI_STATE_D3_HOT].flags.valid = 1;
 
 	if (acpi_bus_init_power(device))
 		device->flags.power_manageable = 0;
Index: linux-pm/drivers/acpi/device_pm.c
===================================================================
--- linux-pm.orig/drivers/acpi/device_pm.c
+++ linux-pm/drivers/acpi/device_pm.c
@@ -98,17 +98,16 @@  int acpi_device_get_power(struct acpi_de
 
 		/*
 		 * The power resources settings may indicate a power state
-		 * shallower than the actual power state of the device.
+		 * shallower than the actual power state of the device, because
+		 * the same power resources may be referenced by other devices.
 		 *
-		 * Moreover, on systems predating ACPI 4.0, if the device
-		 * doesn't depend on any power resources and _PSC returns 3,
-		 * that means "power off".  We need to maintain compatibility
-		 * with those systems.
+		 * For systems predating ACPI 4.0 we assume that D3hot is the
+		 * deepest state that can be supported.
 		 */
 		if (psc > result && psc < ACPI_STATE_D3_COLD)
 			result = psc;
 		else if (result == ACPI_STATE_UNKNOWN)
-			result = psc > ACPI_STATE_D2 ? ACPI_STATE_D3_COLD : psc;
+			result = psc > ACPI_STATE_D2 ? ACPI_STATE_D3_HOT : psc;
 	}
 
 	/*
@@ -153,8 +152,8 @@  static int acpi_dev_pm_explicit_set(stru
  */
 int acpi_device_set_power(struct acpi_device *device, int state)
 {
+	int target_state = state;
 	int result = 0;
-	bool cut_power = false;
 
 	if (!device || !device->flags.power_manageable
 	    || (state < ACPI_STATE_D0) || (state > ACPI_STATE_D3_COLD))
@@ -169,11 +168,21 @@  int acpi_device_set_power(struct acpi_de
 		return 0;
 	}
 
-	if (!device->power.states[state].flags.valid) {
+	if (state == ACPI_STATE_D3_COLD) {
+		/*
+		 * For transitions to D3cold we need to execute _PS3 and then
+		 * possibly drop references to the power resources in use.
+		 */
+		state = ACPI_STATE_D3_HOT;
+		/* If _PR3 is not available, use D3hot as the target state. */
+		if (!device->power.states[ACPI_STATE_D3_COLD].flags.valid)
+			target_state = state;
+	} else if (!device->power.states[state].flags.valid) {
 		dev_warn(&device->dev, "Power state %s not supported\n",
 			 acpi_power_state_string(state));
 		return -ENODEV;
 	}
+
 	if (!device->power.flags.ignore_parent &&
 	    device->parent && (state < device->parent->power.state)) {
 		dev_warn(&device->dev,
@@ -183,39 +192,38 @@  int acpi_device_set_power(struct acpi_de
 		return -ENODEV;
 	}
 
-	/* For D3cold we should first transition into D3hot. */
-	if (state == ACPI_STATE_D3_COLD
-	    && device->power.states[ACPI_STATE_D3_COLD].flags.os_accessible) {
-		state = ACPI_STATE_D3_HOT;
-		cut_power = true;
-	}
-
-	if (state < device->power.state && state != ACPI_STATE_D0
-	    && device->power.state >= ACPI_STATE_D3_HOT) {
-		dev_warn(&device->dev,
-			 "Cannot transition to non-D0 state from D3\n");
-		return -ENODEV;
-	}
-
 	/*
 	 * Transition Power
 	 * ----------------
-	 * In accordance with the ACPI specification first apply power (via
-	 * power resources) and then evaluate _PSx.
+	 * In accordance with ACPI 6, _PSx is executed before manipulating power
+	 * resources, unless the target state is D0, in which case _PS0 is
+	 * supposed to be executed after turning the power resources on.
 	 */
-	if (device->power.flags.power_resources) {
-		result = acpi_power_transition(device, state);
+	if (state > ACPI_STATE_D0) {
+		/*
+		 * According to ACPI 6, devices cannot go from lower-power
+		 * (deeper) states to higher-power (shallower) states.
+		 */
+		if (state < device->power.state) {
+			dev_warn(&device->dev, "Cannot transition from %s to %s\n",
+				 acpi_power_state_string(device->power.state),
+				 acpi_power_state_string(state));
+			return -ENODEV;
+		}
+
+		result = acpi_dev_pm_explicit_set(device, state);
 		if (result)
 			goto end;
-	}
-	result = acpi_dev_pm_explicit_set(device, state);
-	if (result)
-		goto end;
 
-	if (cut_power) {
-		device->power.state = state;
-		state = ACPI_STATE_D3_COLD;
-		result = acpi_power_transition(device, state);
+		if (device->power.flags.power_resources)
+			result = acpi_power_transition(device, target_state);
+	} else {
+		if (device->power.flags.power_resources) {
+			result = acpi_power_transition(device, ACPI_STATE_D0);
+			if (result)
+				goto end;
+		}
+		result = acpi_dev_pm_explicit_set(device, ACPI_STATE_D0);
 	}
 
  end:
@@ -264,13 +272,24 @@  int acpi_bus_init_power(struct acpi_devi
 		return result;
 
 	if (state < ACPI_STATE_D3_COLD && device->power.flags.power_resources) {
+		/* Reference count the power resources. */
 		result = acpi_power_on_resources(device, state);
 		if (result)
 			return result;
 
-		result = acpi_dev_pm_explicit_set(device, state);
-		if (result)
-			return result;
+		if (state == ACPI_STATE_D0) {
+			/*
+			 * If _PSC is not present and the state inferred from
+			 * power resources appears to be D0, it still may be
+			 * necessary to execute _PS0 at this point, because
+			 * another device using the same power resources may
+			 * have been put into D0 previously and that's why we
+			 * see D0 here.
+			 */
+			result = acpi_dev_pm_explicit_set(device, state);
+			if (result)
+				return result;
+		}
 	} else if (state == ACPI_STATE_UNKNOWN) {
 		/*
 		 * No power resources and missing _PSC?  Cross fingers and make
@@ -603,12 +622,12 @@  int acpi_pm_device_sleep_state(struct de
 	if (d_max_in < ACPI_STATE_D0 || d_max_in > ACPI_STATE_D3_COLD)
 		return -EINVAL;
 
-	if (d_max_in > ACPI_STATE_D3_HOT) {
+	if (d_max_in > ACPI_STATE_D2) {
 		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;
+			d_max_in = ACPI_STATE_D2;
 	}
 
 	adev = ACPI_COMPANION(dev);
Index: linux-pm/drivers/acpi/fan.c
===================================================================
--- linux-pm.orig/drivers/acpi/fan.c
+++ linux-pm/drivers/acpi/fan.c
@@ -158,8 +158,9 @@  static int fan_get_state(struct acpi_dev
 	if (result)
 		return result;
 
-	*state = (acpi_state == ACPI_STATE_D3_COLD ? 0 :
-		 (acpi_state == ACPI_STATE_D0 ? 1 : -1));
+	*state = acpi_state == ACPI_STATE_D3_COLD
+			|| acpi_state == ACPI_STATE_D3_HOT ?
+		0 : (acpi_state == ACPI_STATE_D0 ? 1 : -1);
 	return 0;
 }
 
Index: linux-pm/drivers/pci/pci-acpi.c
===================================================================
--- linux-pm.orig/drivers/pci/pci-acpi.c
+++ linux-pm/drivers/pci/pci-acpi.c
@@ -420,7 +420,7 @@  static int acpi_pci_set_power_state(stru
 		[PCI_D0] = ACPI_STATE_D0,
 		[PCI_D1] = ACPI_STATE_D1,
 		[PCI_D2] = ACPI_STATE_D2,
-		[PCI_D3hot] = ACPI_STATE_D3_COLD,
+		[PCI_D3hot] = ACPI_STATE_D3_HOT,
 		[PCI_D3cold] = ACPI_STATE_D3_COLD,
 	};
 	int error = -EINVAL;