Commit 20dacb71 ("ACPI / PM: Rework device power management to follow
ACPI 6") changed the device power management to use D3hot if the device
in question does not have _PR3 method even if D3cold was requested by the
caller.

However, if the device has _PR3 device->power.state is also set to D3hot
instead of D3Cold after power resources have been turned off because
device->power.state will be assigned from "state" instead of
"target_state".

Next time the device is transitioned to D0, acpi_power_transition() will
find that the current power state of the device is D3hot instead of D3cold
which causes it to power down all resources required for the current
(wrong) state D3hot.

Below is a simplified ASL example of a real touch panel device which
triggers the problem:

  Scope (TPL1)
  {
      Name (_PR0, Package (1) { \_SB.PCI0.I2C1.PXTC })
      Name (_PR3, Package (1) { \_SB.PCI0.I2C1.PXTC })
      ...
  }

In both D0 and D3hot the same power resource is required. However, when
acpi_power_transition() turns off power resources required for D3hot (as
the device is transitioned to D0) it powers down PXTC which then makes the
device to lose its power.

Fix this by assigning "target_state" to the device power state instead of
"state" that is always D3hot even for devices with valid _PR3.

Fixes: 20dacb71 (ACPI / PM: Rework device power management to follow ACPI 6)
Signed-off-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Add missing invocation of pm_generic_complete() to
acpi_subsys_complete() to allow ->complete callbacks provided
by the drivers of devices using the ACPI PM domain to be executed
during system resume.

Fixes: f25c0ae2 (ACPI / PM: Avoid resuming devices in ACPI PM domain during system suspend)
Cc: 3.16+ <stable@vger.kernel.org> # 3.16+
Signed-off-by: NRafael 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: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

In commit 5de21bb9 ("ACPI / PM: Drop CONFIG_PM_RUNTIME from the
ACPI core"), all occurrences of CONFIG_PM_RUNTIME were replaced with
CONFIG_PM. This created the following structure of #ifdef blocks in
the code:

 [...]
 #ifdef CONFIG_PM
 #ifdef CONFIG_PM
 /* always on / undead */
 #ifdef CONFIG_PM_SLEEP
 [...]
 #endif
 #endif
 [...]
 #endif

This patch removes the inner "#ifdef CONFIG_PM" block as it will
always be enabled when the outer block is enabled. This inconsistency
was found using the undertaker-checkpatch tool.
Signed-off-by: NAndreas Ruprecht <rupran@einserver.de>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

If an ACPI device object whose _STA returns 0 (not present and not
functional) has _PR0 or _PS0, its power_manageable flag will be set
and acpi_bus_init_power() will return 0 for it.  Consequently, if
such a device object is passed to the ACPI device PM functions, they
will attempt to carry out the requested operation on the device,
although they should not do that for devices that are not present.

To fix that problem make acpi_bus_init_power() return an error code
for devices that are not present which will cause power_manageable to
be cleared for them as appropriate in acpi_bus_get_power_flags().
However, the lists of power resources should not be freed for the
device in that case, so modify acpi_bus_get_power_flags() to keep
those lists even if acpi_bus_init_power() returns an error.
Accordingly, when deciding whether or not the lists of power
resources need to be freed, acpi_free_power_resources_lists()
should check the power.flags.power_resources flag instead of
flags.power_manageable, so make that change too.

Furthermore, if acpi_bus_attach() sees that flags.initialized is
unset for the given device, it should reset the power management
settings of the device and re-initialize them from scratch instead
of relying on the previous settings (the device may have appeared
after being not present previously, for example), so make it use
the 'valid' flag of the D0 power state as the initial value of
flags.power_manageable for it and call acpi_bus_init_power() to
discover its current power state.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Cc: 3.10+ <stable@vger.kernel.org> # 3.10+

In some cases acpi_device_wakeup() may be called to ensure wakeup
power to be off for a given device even though that device's wakeup
GPE has not been enabled so far.  It calls acpi_disable_gpe() on a
GPE that's not enabled and this causes ACPICA to return the AE_LIMIT
status code from that call which then is reported as an error by the
ACPICA's debug facilities (if enabled).  This may lead to a fair
amount of confusion, so introduce a new ACPI device wakeup flag
to store the wakeup GPE status and avoid disabling wakeup GPEs
that have not been enabled.
Reported-and-tested-by: NVenkat Raghavulu <venkat.raghavulu@intel.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

After commit b2b49ccb (PM: Kconfig: Set PM_RUNTIME if PM_SLEEP is
selected) PM_RUNTIME is always set if PM is set, so quite a few
depend on CONFIG_PM.

Replace CONFIG_PM_RUNTIME with CONFIG_PM in the ACPI core code.
Reviewed-by: NUlf Hansson <ulf.hansson@linaro.org>
Acked-by: NKevin Hilman <khilman@linaro.org>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Signed-off-by: NHuang Rui <ray.huang@amd.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

As reported by Dmitry, on some Chromebooks there are devices with
corresponding ACPI objects and with unusual system wakeup
configuration.  Namely, they technically are wakeup-capable, but the
wakeup is handled via a platform-specific out-of-band mechanism and
the ACPI PM layer has no information on the wakeup capability.  As
a result, device_may_wakeup(dev) called from acpi_dev_suspend_late()
returns 'true' for those devices, but the wakeup.flags.valid flag is
unset for the corresponding ACPI device objects, so acpi_device_wakeup()
reproducibly fails for them causing acpi_dev_suspend_late() to return
an error code.  The entire system suspend is then aborted and the
machines in question cannot suspend at all.

Address the problem by ignoring the device_may_wakeup(dev) return
value in acpi_dev_suspend_late() if the ACPI companion of the device
being handled has wakeup.flags.valid unset (in which case it is clear
that the wakeup is supposed to be handled by other means).

This fixes a regression introduced by commit a76e9bd8 (i2c:
attach/detach I2C client device to the ACPI power domain) as the
affected systems could suspend and resume successfully before that
commit.

Fixes: a76e9bd8 (i2c: attach/detach I2C client device to the ACPI power domain)
Reported-by: NDmitry Torokhov <dtor@chromium.org>
Reviewed-by: NDmitry Torokhov <dtor@chromium.org>
Cc: 3.13+ <stable@vger.kernel.org> # 3.13+
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Fix a bug that invokes acpi_device_wakeup() with wrong parameters.

Fixes: f35cec25 (ACPI / PM: Always enable wakeup GPEs when enabling device wakeup)
Signed-off-by: NZhang Rui <rui.zhang@intel.com>
Cc: 3.17+ <stable@vger.kernel.org> # 3.17+
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

When we have the acpi_device pointer, there is no need to pass the
device's handle to the acpi_bus_xxx_power functions to get/set/update
the device's power state, instead, use the acpi_device_xxx_power
functions directly.

To make this happen for fan module, export acpi_device_update_power.
Signed-off-by: NAaron Lu <aaron.lu@intel.com>
Signed-off-by: NZhang Rui <rui.zhang@intel.com>

The ->detach() callback for the PM domain has now been fully adopted,
thus there no users left of the acpi_dev_pm_detach() API. This allow us
to convert it into a static function.
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>
Reviewed-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

As as preparation to simplify the detachment of devices from their PM
domains, we assign the ->detach() callback to genpd_dev_pm_detach().
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>
Reviewed-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

The ACPI_HANDLE() macro evaluates ACPI_COMPANION() internally to
return the handle of the device's ACPI companion, so it is much
more straightforward and efficient to use ACPI_COMPANION()
directly to obtain the device's ACPI companion object instead of
using ACPI_HANDLE() and acpi_bus_get_device() on the returned
handle for the same thing.

Do that in three places in the ACPI device PM code.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Wakeup GPEs are currently only enabled when setting up devices for
remote wakeup at run time.  During system-wide transitions they are
enabled by ACPICA at the very last stage of suspend (before asking
the BIOS to take over).  Of course, that only works for system
sleep states supported by ACPI, so in particular it doesn't work
for the "freeze" sleep state.

For this reason, modify the ACPI core device PM code to enable wakeup
GPEs for devices when setting them up for wakeup regardless of whether
that is remote wakeup at runtime or system wakeup.  That allows the
same device wakeup setup routine to be used for both runtime PM and
system-wide PM and makes it possible to reduce code size quite a bit.

This make ACPI-based PCI Wake-on-LAN work with the "freeze" sleep
state on my venerable Toshiba Portege R500 and should help other
systems too.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Since ACPI wakeup GPEs are going to be enabled during system suspend
as well as for runtime wakeup by a subsequent patch and the same
notify handlers will be used in both cases, rework the ACPI device
wakeup notification framework so that the part specific to physical
devices is always run asynchronously from the PM workqueue.  This
prevents runtime resume callbacks for those devices from being
run during system suspend and resume which may not be appropriate,
among other things.

Also make ACPI device wakeup notification handling a bit more robust
agaist subsequent removal of ACPI device objects, whould that ever
happen, and create a wakeup source object for each ACPI device
configured for wakeup so that wakeup notifications for those
devices can wake up the system from the "freeze" sleep state.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

No reason for excluding the remaining ones.
Signed-off-by: NHeikki Krogerus <heikki.krogerus@linux.intel.com>
[rjw: Rebased and exported the new acpi_subsys_complete() too.]
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Rework the ACPI PM domain's PM callbacks to avoid resuming devices
during system suspend (in order to modify their wakeup settings etc.)
if that isn't necessary.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Runtime-suspended devices are resumed during system suspend by
acpi_subsys_prepare() for two reasons: First, because they may need
to be reprogrammed in order to change their wakeup settings and,
second, because they may need to be operatonal for their children
to be successfully suspended.  That is a problem, though, if there
are many runtime-suspended devices that need to be resumed this
way during system suspend, because the .prepare() PM callbacks of
devices are executed sequentially and the times taken by them
accumulate, which may increase the total system suspend time quite
a bit.

For this reason, move the resume of runtime-suspended devices up
to the next phase of device suspend (during system suspend), except
for the ones that have power.ignore_children set.  The exception is
made, because the devices with power.ignore_children set may still
be necessary for their children to be successfully suspended (during
system suspend) and they won't be resumed automatically as a result
of the runtime resume of their children.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

The ACPI device PM code in device_pm.c uses a special function,
acpi_dev_pm_get_node(), to obtain an ACPI companion object of a given
device.  However, that is not necessary any more after recent changes
that introduced the ACPI_COMPANION() macro serving exactly the same
purpose, but working in a much more straightforward way.  For this
reason, drop acpi_dev_pm_get_node() and use ACPI_COMPANION() instead
of it everywhere.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: NMika Westerberg <mika.westerberg@linux.intel.com>

Modify the ACPI namespace scanning code to register a struct
acpi_device object for every namespace node representing a device,
processor and so on, even if the device represented by that namespace
node is reported to be not present and not functional by _STA.

There are multiple reasons to do that.  First of all, it avoids
quite a lot of overhead when struct acpi_device objects are
deleted every time acpi_bus_trim() is run and then added again
by a subsequent acpi_bus_scan() for the same scope, although the
namespace objects they correspond to stay in memory all the time
(which always is the case on a vast majority of systems).

Second, it will allow user space to see that there are namespace
nodes representing devices that are not present at the moment and may
be added to the system.  It will also allow user space to evaluate
_SUN for those nodes to check what physical slots the "missing"
devices may be put into and it will make sense to add a sysfs
attribute for _STA evaluation after this change (that will be
useful for thermal management on some systems).

Next, it will help to consolidate the ACPI hotplug handling among
subsystems by making it possible to store hotplug-related information
in struct acpi_device objects in a standard common way.

Finally, it will help to avoid a race condition related to the
deletion of ACPI namespace nodes.  Namely, namespace nodes may be
deleted as a result of a table unload triggered by _EJ0 or _DCK.
If a hotplug notification for one of those nodes is triggered
right before the deletion and it executes a hotplug callback
via acpi_hotplug_execute(), the ACPI handle passed to that
callback may be stale when the callback actually runs.  One way
to work around that is to always pass struct acpi_device pointers
to hotplug callbacks after doing a get_device() on the objects in
question which eliminates the use-after-free possibility (the ACPI
handles in those objects are invalidated by acpi_scan_drop_device(),
so they will trigger ACPICA errors on attempts to use them).
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Tested-by: NMika Westerberg <mika.westerberg@linux.intel.com>

Since DEVICE_ACPI_HANDLE() is now literally identical to
ACPI_HANDLE(), replace it with the latter everywhere and drop its
definition from include/acpi.h.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Acked-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Modify struct acpi_dev_node to contain a pointer to struct acpi_device
associated with the given device object (that is, its ACPI companion
device) instead of an ACPI handle corresponding to it.  Introduce two
new macros for manipulating that pointer in a CONFIG_ACPI-safe way,
ACPI_COMPANION() and ACPI_COMPANION_SET(), and rework the
ACPI_HANDLE() macro to take the above changes into account.
Drop the ACPI_HANDLE_SET() macro entirely and rework its users to
use ACPI_COMPANION_SET() instead.  For some of them who used to
pass the result of acpi_get_child() directly to ACPI_HANDLE_SET()
introduce a helper routine acpi_preset_companion() doing an
equivalent thing.

The main motivation for doing this is that there are things
represented by struct acpi_device objects that don't have valid
ACPI handles (so called fixed ACPI hardware features, such as
power and sleep buttons) and we would like to create platform
device objects for them and "glue" them to their ACPI companions
in the usual way (which currently is impossible due to the
lack of valid ACPI handles).  However, there are more reasons
why it may be useful.

First, struct acpi_device pointers allow of much better type checking
than void pointers which are ACPI handles, so it should be more
difficult to write buggy code using modified struct acpi_dev_node
and the new macros.  Second, the change should help to reduce (over
time) the number of places in which the result of ACPI_HANDLE() is
passed to acpi_bus_get_device() in order to obtain a pointer to the
struct acpi_device associated with the given "physical" device,
because now that pointer is returned by ACPI_COMPANION() directly.
Finally, the change should make it easier to write generic code that
will build both for CONFIG_ACPI set and unset without adding explicit
compiler directives to it.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Acked-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> # on Haswell
Reviewed-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Reviewed-by: Aaron Lu <aaron.lu@intel.com> # for ATA and SDIO part

Two functions defined in device_pm.c, acpi_dev_pm_add_dependent()
and acpi_dev_pm_remove_dependent(), have no callers and may be
dropped, so drop them.

Moreover, they are the only functions adding entries to and removing
entries from the power_dependent list in struct acpi_device, so drop
that list too.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Some serial buses like I2C and SPI don't require that the parent device is
in D0 before any of its children transitions to D0, but instead the parent
device can control its own power independently from the children.

This does not follow the ACPI specification as it requires the parent to be
powered on before its children. However, Windows seems to ignore this
requirement so I think we can do the same in Linux.

Implement this by adding a new power flag 'ignore_parent' to struct
acpi_device.  If this flag is set the ACPI core ignores checking of the
parent device power state when the device is powered on/off.
Signed-off-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

The state information can be useful to know what the problem is when
an error message about a device can not being set to a higher power
state than its parent appeared, so this patch adds such state
information for both the target state of the device and the current
state of its parent.
Signed-off-by: NAaron Lu <aaron.lu@intel.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Now that acpi_device_set_power() checks whether or not the given
device is power manageable, it is not necessary to do this check in
acpi_bus_set_power() any more, so remove it.
Signed-off-by: NAaron Lu <aaron.lu@intel.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Modify acpi_device_set_power() so that diagnostic messages printed by
it to the kernel log always contain the name of the device concerned
to make it possible to identify the device that triggered the message
if need be.

Also replace printk(KERN_WARNING ) with dev_warn() everywhere in that
function.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: NAaron Lu <aaron.lu@intel.com>

Make acpi_device_set_power() check if the given device is power
manageable before checking if the given power state is valid for that
device.  Otherwise it will print that "Device does not support" that
power state into the kernel log, which may not make sense for some
power states (D0 and D3cold are supported by all devices by
definition).
Tested-by: NYinghai Lu <yinghai@kernel.org>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

The role of acpi_bus_update_power() is to update the given ACPI
device object's power.state field to reflect the current physical
state of the device (as inferred from the configuration of power
resources and _PSC, if available).  For this purpose it calls
acpi_device_set_power() that should update the power resources'
reference counters and set power.state as appropriate.  However,
that doesn't work if the "new" state is D1, D2 or D3hot and the
the current value of power.state means D3cold, because in that
case acpi_device_set_power() will refuse to transition the device
from D3cold to non-D0.

To address this problem, make acpi_bus_update_power() call
acpi_power_transition() directly to update the power resources'
reference counters and only use acpi_device_set_power() to put
the device into D0 if the current physical state of it cannot
be determined.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: 3.9+ <stable@vger.kernel.org>

After commit fa1675b5 (ACPI / PM: Rework and clean up
acpi_dev_pm_get_state()) a NULL pointer dereference will take place
if NULL is passed to acpi_pm_device_sleep_state() as the second
argument.

Fix that by avoiding to use the pointer that may be NULL until
it's necessary to store a return value at the location pointed to
by it (if not NULL).
Reported-and-tested-by: NAaron Lu <aaron.lu@intel.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Commit 7cd8407d (ACPI / PM: Do not execute _PS0 for devices without
_PSC during initialization) introduced a regression on some systems
with Intel Lynxpoint Low-Power Subsystem (LPSS) where some devices
need to be powered up during initialization, but their device objects
in the ACPI namespace have _PS0 and _PS3 only (without _PSC or power
resources).

To work around this problem, make the ACPI LPSS driver power up
devices it knows about by using a new helper function
acpi_device_fix_up_power() that does all of the necessary
sanity checks and calls acpi_dev_pm_explicit_set() to put the
device into D0.
Reported-and-tested-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

The acpi_dev_pm_get_state() function defined in device_pm.c is quite
convoluted, which isn't really necessary, and it doesn't validate the
values returned by the ACPI methods executed by it appropriately.

To address these shortcomings modify it in the following way.

 (1) Make its return value only mean whether or not it succeeded and
     pass the device power states determined by it through pointers.

 (2) Drop the d_max_in argument, used by only one of its callers,
     from it, and move the code related to d_max_in into that caller,
     acpi_pm_device_sleep_state().

 (3) Make it always check the return value of acpi_evaluate_integer()
     and handle failures as appropriate.  Moreover, make it check if
     the values returned by the executed ACPI methods are not out of
     range.

 (4) Make it check if the values returned by the executed ACPI
     methods represent valid power states of the given device and
     handle situations in which that's not the case gracefully.

Also update the kerneldoc comments of acpi_dev_pm_get_state() and
acpi_pm_device_sleep_state() to reflect the code changes.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

The two symbols ACPI_STATE_D3 and ACPI_STATE_D3_COLD actually
represent the same number (4), but ACPI_STATE_D3 is slightly
ambigugous, because it may not be clear that it really means D3cold
and not D3hot at first sight.

Remove that ambiguity from drivers/acpi/device_pm.c by making it
use ACPI_STATE_D3_COLD everywhere instead of ACPI_STATE_D3.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

There is a name clash between function acpi_device_power_state()
defined in drivers/acpi/device_pm.c and structure type
acpi_device_power_state defined in include/acpi/acpi_bus.h, which
may be resolved by renaming the function.  Additionally, that
funtion may be made static, because it is not used anywhere outside
of the file it is defined in.

Rename acpi_device_power_state() to acpi_dev_pm_get_state(), which
better reflects its purpose, and make it static.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Commit b3785492 (ACPI / PM: Do not power manage devices in unknown
initial states) added code to force devices without _PSC, but having
_PS0 defined in the ACPI namespace, into ACPI power state D0 by
executing _PS0 for them.  That turned out to break Toshiba P870-303,
however, so revert that code.

References: https://bugzilla.kernel.org/show_bug.cgi?id=58201Reported-and-tested-by: NJerome Cantenot <jerome.cantenot@gmail.com>
Tracked-down-by: NLan Tianyu <tianyu.lan@intel.com>
Cc: 3.9+ <stable@vger.kernel.org>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

The "runtime idle" helper routine, rpm_idle(), currently ignores
return values from .runtime_idle() callbacks executed by it.
However, it turns out that many subsystems use
pm_generic_runtime_idle() which checks the return value of the
driver's callback and executes pm_runtime_suspend() for the device
unless that value is not 0.  If that logic is moved to rpm_idle()
instead, pm_generic_runtime_idle() can be dropped and its users
will not need any .runtime_idle() callbacks any more.

Moreover, the PCI, SCSI, and SATA subsystems' .runtime_idle()
routines, pci_pm_runtime_idle(), scsi_runtime_idle(), and
ata_port_runtime_idle(), respectively, as well as a few drivers'
ones may be simplified if rpm_idle() calls rpm_suspend() after 0 has
been returned by the .runtime_idle() callback executed by it.

To reduce overall code bloat, make the changes described above.
Tested-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Tested-by: NKevin Hilman <khilman@linaro.org>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Acked-by: NKevin Hilman <khilman@linaro.org>
Reviewed-by: NUlf Hansson <ulf.hansson@linaro.org>
Acked-by: NAlan Stern <stern@rowland.harvard.edu>

Currently, drivers/acpi/device_pm.c depends on CONFIG_PM and all of
the functions defined in there are replaced with static inline stubs
if that option is unset.  However, CONFIG_PM means, roughly, "runtime
PM or suspend/hibernation support" and some of those functions are
useful regardless of that.  For example, they are used by the ACPI
fan driver for controlling fans and acpi_device_set_power() is called
during device removal.  Moreover, device initialization may depend on
setting device power states properly.

For these reasons, make the routines manipulating ACPI device power
states defined in drivers/acpi/device_pm.c available for CONFIG_PM
unset too.
Reported-by: NZhang Rui <rui.zhang@intel.com>
Reported-and-tested-by: NMichel Lespinasse <walken@google.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: 3.9+ <stable@vger.kernel.org>

Theoretically, in some situations acpi_device_get_power() may return
an incorrect result, because the settings of the power resources
depended on by the device may indicate a power state shallower than
the actual power state of the device.

Say that two devices, A and B, depend on two power resources, X and
Y, in such a way that _PR0 for both A and B list both X and Y and
_PR3 for both A and B list power resource Y alone.  Also suppose
that _PS0 and _PS3 are present for both A and B.  Then, if devices
A and B are initially in D0, power resources X and Y are initially
"on" and their reference counters are equal to 2.  To put device A
into power state D3hot the kernel will decrement the reference
counter of power resource X, but that power resource won't be turned
off, because it is still in use by device B (its reference counter is
equal to 1).  Next, _PS3 will be executed for device A.  Afterward
the configuration of the power resources will indicate that device
A is in power state D0 (both X and Y are "on"), but in fact it is
in D3hot (because _PS3 has been executed for it).

In that situation, if acpi_device_get_power() is called to get the
power state of device A, it will first execute _PSC for it which
should return 3.  That will cause acpi_device_get_power() to run
acpi_power_get_inferred_state() for device A and the resultant power
state will be D0, which is incorrect.

To fix that change acpi_device_get_power() to first execute
acpi_power_get_inferred_state() for the given device (if it
depends on power resources) and to evaluate _PSC for it subsequently,
so that the result inferred from the power resources configuration
can be amended by the _PSC return value.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: NAaron Lu <aaron.lu@intel.com>

If _PS0 is defined for an ACPI device node, but _PSC isn't and
the device node doesn't use power resources for power management,
acpi_bus_update_power() will fail to update the power state of it,
because acpi_device_get_power() returns ACPI_STATE_UNKNOWN in that
case.

To handle that situation make acpi_bus_update_power() follow
acpi_bus_init_power() and try to force the given device node into
power state D0.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>