Define and document a SMART_SUSPEND flag to instruct bus types and PM
domains that the system suspend callbacks provided by the driver can
cope with runtime-suspended devices, so from the driver's perspective
it should be safe to leave devices in runtime suspend during system
suspend.

Setting that flag may also cause middle-layer code (bus types,
PM domains etc.) to skip invocations of the ->suspend_late and
->suspend_noirq callbacks provided by the driver if the device
is in runtime suspend at the beginning of the "late" phase of
the system-wide suspend transition, in which case the driver's
system-wide resume callbacks may be invoked back-to-back with
its ->runtime_suspend callback, so the driver has to be able to
cope with that too.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Acked-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Reviewed-by: NUlf Hansson <ulf.hansson@linaro.org>

The motivation for this change is to provide a way to work around
a problem with the direct-complete mechanism used for avoiding
system suspend/resume handling for devices in runtime suspend.

The problem is that some middle layer code (the PCI bus type and
the ACPI PM domain in particular) returns positive values from its
system suspend ->prepare callbacks regardless of whether the driver's
->prepare returns a positive value or 0, which effectively prevents
drivers from being able to control the direct-complete feature.
Some drivers need that control, however, and the PCI bus type has
grown its own flag to deal with this issue, but since it is not
limited to PCI, it is better to address it by adding driver flags at
the core level.

To that end, add a driver_flags field to struct dev_pm_info for flags
that can be set by device drivers at the probe time to inform the PM
core and/or bus types, PM domains and so on on the capabilities and/or
preferences of device drivers.  Also add two static inline helpers
for setting that field and testing it against a given set of flags
and make the driver core clear it automatically on driver remove
and probe failures.

Define and document two PM driver flags related to the direct-
complete feature: NEVER_SKIP and SMART_PREPARE that can be used,
respectively, to indicate to the PM core that the direct-complete
mechanism should never be used for the device and to inform the
middle layer code (bus types, PM domains etc) that it can only
request the PM core to use the direct-complete mechanism for
the device (by returning a positive value from its ->prepare
callback) if it also has been requested by the driver.

While at it, make the core check pm_runtime_suspended() when
setting power.direct_complete so that it doesn't need to be
checked by ->prepare callbacks.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Acked-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Acked-by: NBjorn Helgaas <bhelgaas@google.com>
Reviewed-by: NUlf Hansson <ulf.hansson@linaro.org>

Fix the kerneldoc comments of __device_suspend_noirq(),
__device_suspend_late() and __device_suspend() where the function
names in kerneldoc don't match the actual names of the functions.

Also fix the device_resume_noirq() kerneldoc comment which mentions
"early resume" instead of "noirq resume" incorrectly.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: NUlf Hansson <ulf.hansson@linaro.org>

There are no classes using the legacy suspend/resume operations in
the tree any more, so drop these operations and update the code
referring to them accordingly.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Acked-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The device_pm_check_callbacks() function doesn't check legacy
->suspend and ->resume callback pointers under the device's
bus type, class and driver, so in some cases it may set the
no_pm_callbacks flag for the device incorrectly and then the
callbacks may be skipped during system suspend/resume, which
shouldn't happen.

Fixes: aa8e54b5 (PM / sleep: Go direct_complete if driver has no callbacks)
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: 4.5+ <stable@vger.kernel.org> # 4.5+

Restore the pm_wakeup_pending() check in __device_suspend_noirq()
removed by commit eed4d47e (ACPI / sleep: Ignore spurious SCI
wakeups from suspend-to-idle) as that allows the function to return
earlier if there's a wakeup event pending already (so that it may
spend less time on carrying out operations that will be reversed
shortly anyway) and rework the main suspend-to-idle loop to take
that optimization into account.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Make the core device suspend/resume code also call dpm_show_time()
on failures and add an error argument to this function so that the
message printed by it can reflect the success or failure condition.

This makes the debug messages in question look less confusing in
the failing cases.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Put the device interrupts disabling and enabling as well as
cpuidle_pause() and cpuidle_resume() called during the "noirq"
stages of system suspend into separate functions to allow the
core suspend-to-idle code to be optimized (later).

The only functional difference this makes is that debug facilities
and diagnostic tools will not include the above operations into the
"noirq" device suspend/resume duration measurements.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Debug messages from the system suspend/hibernation infrastructure can
fill up the entire kernel log buffer in some cases and anyway they
are only useful for debugging.  They depend on CONFIG_PM_DEBUG, but
that is set as a rule as some generally useful diagnostic facilities
depend on it too.

For this reason, avoid printing those messages by default, but make
it possible to turn them on as needed with the help of a new sysfs
attribute under /sys/power/.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

The 'info' string appearing in many places points to a .rodata string so
it should be passes as pointer to const.
Signed-off-by: NKrzysztof Kozlowski <krzk@kernel.org>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

The pm_verb() returns a pointer to string from .rodata so it should be
marked as const.
Signed-off-by: NKrzysztof Kozlowski <krzk@kernel.org>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

The ACPI SCI (System Control Interrupt) is set up as a wakeup IRQ
during suspend-to-idle transitions and, consequently, any events
signaled through it wake up the system from that state.  However,
on some systems some of the events signaled via the ACPI SCI while
suspended to idle should not cause the system to wake up.  In fact,
quite often they should just be discarded.

Arguably, systems should not resume entirely on such events, but in
order to decide which events really should cause the system to resume
and which are spurious, it is necessary to resume up to the point
when ACPI SCIs are actually handled and processed, which is after
executing dpm_resume_noirq() in the system resume path.

For this reasons, add a loop around freeze_enter() in which the
platforms can process events signaled via multiplexed IRQ lines
like the ACPI SCI and add suspend-to-idle hooks that can be
used for this purpose to struct platform_freeze_ops.

In the ACPI case, the ->wake hook is used for checking if the SCI
has triggered while suspended and deferring the interrupt-induced
system wakeup until the events signaled through it are actually
processed sufficiently to decide whether or not the system should
resume.  In turn, the ->sync hook allows all of the relevant event
queues to be flushed so as to prevent events from being missed due
to race conditions.

In addition to that, some ACPI code processing wakeup events needs
to be modified to use the "hard" version of wakeup triggers, so that
it will cause a system resume to happen on device-induced wakeup
events even if the "soft" mechanism to prevent the system from
suspending is not enabled.  However, to preserve the existing
behavior with respect to suspend-to-RAM, this only is done in
the suspend-to-idle case and only if an SCI has occurred while
suspended.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Avoid printing the device suspend/resume timing information if
CONFIG_PM_DEBUG is not set to reduce the log noise level.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Revert commit eed4d47e (ACPI / sleep: Ignore spurious SCI wakeups
from suspend-to-idle) as it turned out to be premature and triggered
a number of different issues on various systems.

That includes, but is not limited to, premature suspend-to-RAM aborts
on Dell XPS 13 (9343) reported by Dominik.

The issue the commit in question attempted to address is real and
will need to be taken care of going forward, but evidently more work
is needed for this purpose.
Reported-by: NDominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

The ACPI SCI (System Control Interrupt) is set up as a wakeup IRQ
during suspend-to-idle transitions and, consequently, any events
signaled through it wake up the system from that state.  However,
on some systems some of the events signaled via the ACPI SCI while
suspended to idle should not cause the system to wake up.  In fact,
quite often they should just be discarded.

Arguably, systems should not resume entirely on such events, but in
order to decide which events really should cause the system to resume
and which are spurious, it is necessary to resume up to the point
when ACPI SCIs are actually handled and processed, which is after
executing dpm_resume_noirq() in the system resume path.

For this reasons, add a loop around freeze_enter() in which the
platforms can process events signaled via multiplexed IRQ lines
like the ACPI SCI and add suspend-to-idle hooks that can be
used for this purpose to struct platform_freeze_ops.

In the ACPI case, the ->wake hook is used for checking if the SCI
has triggered while suspended and deferring the interrupt-induced
system wakeup until the events signaled through it are actually
processed sufficiently to decide whether or not the system should
resume.  In turn, the ->sync hook allows all of the relevant event
queues to be flushed so as to prevent events from being missed due
to race conditions.

In addition to that, some ACPI code processing wakeup events needs
to be modified to use the "hard" version of wakeup triggers, so that
it will cause a system resume to happen on device-induced wakeup
events even if the "soft" mechanism to prevent the system from
suspending is not enabled (that also helps to catch device-induced
wakeup events occurring during suspend transitions in progress).
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

We are going to split <linux/sched/debug.h> out of <linux/sched.h>, which
will have to be picked up from other headers and a couple of .c files.

Create a trivial placeholder <linux/sched/debug.h> file that just
maps to <linux/sched.h> to make this patch obviously correct and
bisectable.

Include the new header in the files that are going to need it.
Acked-by: NLinus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: NIngo Molnar <mingo@kernel.org>

ktime_set(S,N) was required for the timespec storage type and is still
useful for situations where a Seconds and Nanoseconds part of a time value
needs to be converted. For anything where the Seconds argument is 0, this
is pointless and can be replaced with a simple assignment.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Cc: Peter Zijlstra <peterz@infradead.org>

If async_suspend is enabled for parent and child devices, then
PM framework has to ensure that parent's async suspend gets called
only after child's async suspend is done. In case if child's async
suspend fails with error, then parent's async suspend must not be
invoked. The current code uses async_error to ensure this but there
is a problem with it in __device_suspend(). This function notifies
the completion of child's async suspend before updating its error
via async_error variable. As a result, parent's async suspend gets
invoked even though it's child suspend has failed. Fix this bug by
updating the async_error before notifying the child's completion.
Signed-off-by: NSahitya Tummala <stummala@codeaurora.org>
[ rjw: Rearranged wthitespace ]
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Consider two devices, A and B, where B is a child of A, and B utilizes
asynchronous suspend (it does not matter whether A is sync or async). If
B fails to suspend_noirq() or suspend_late(), or is interrupted by a
wakeup (pm_wakeup_pending()), then it aborts and sets the async_error
variable. However, device A does not (immediately) check the async_error
variable; it may continue to run its own suspend_noirq()/suspend_late()
callback. This is bad.

We can resolve this problem by doing our error and wakeup checking
(particularly, for the async_error flag) after waiting for children to
suspend, instead of before. This also helps align the logic for the noirq and
late suspend cases with the logic in __device_suspend().

It's easy to observe this erroneous behavior by, for example, forcing a
device to sleep a bit in its suspend_noirq() (to ensure the parent is
waiting for the child to complete), then return an error, and watch the
parent suspend_noirq() still get called. (Or similarly, fake a wakeup
event at the right (or is it wrong?) time.)

Fixes: de377b39 (PM / sleep: Asynchronous threads for suspend_late)
Fixes: 28b6fd6e (PM / sleep: Asynchronous threads for suspend_noirq)
Reported-by: NJeffy Chen <jeffy.chen@rock-chips.com>
Signed-off-by: NBrian Norris <briannorris@chromium.org>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Make the device suspend/resume part of the core system
suspend/resume code use device links to ensure that supplier
and consumer devices will be suspended and resumed in the right
order in case of async suspend/resume.

The idea, roughly, is to use dpm_wait() to wait for all consumers
before a supplier device suspend and to wait for all suppliers
before a consumer device resume.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Tested-by: NMarek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Currently, there is a problem with taking functional dependencies
between devices into account.

What I mean by a "functional dependency" is when the driver of device
B needs device A to be functional and (generally) its driver to be
present in order to work properly.  This has certain consequences
for power management (suspend/resume and runtime PM ordering) and
shutdown ordering of these devices.  In general, it also implies that
the driver of A needs to be working for B to be probed successfully
and it cannot be unbound from the device before the B's driver.

Support for representing those functional dependencies between
devices is added here to allow the driver core to track them and act
on them in certain cases where applicable.

The argument for doing that in the driver core is that there are
quite a few distinct use cases involving device dependencies, they
are relatively hard to get right in a driver (if one wants to
address all of them properly) and it only gets worse if multiplied
by the number of drivers potentially needing to do it.  Morever, at
least one case (asynchronous system suspend/resume) cannot be handled
in a single driver at all, because it requires the driver of A to
wait for B to suspend (during system suspend) and the driver of B to
wait for A to resume (during system resume).

For this reason, represent dependencies between devices as "links",
with the help of struct device_link objects each containing pointers
to the "linked" devices, a list node for each of them, status
information, flags, and an RCU head for synchronization.

Also add two new list heads, representing the lists of links to the
devices that depend on the given one (consumers) and to the devices
depended on by it (suppliers), and a "driver presence status" field
(needed for figuring out initial states of device links) to struct
device.

The entire data structure consisting of all of the lists of link
objects for all devices is protected by a mutex (for link object
addition/removal and for list walks during device driver probing
and removal) and by SRCU (for list walking in other case that will
be introduced by subsequent change sets).  If CONFIG_SRCU is not
selected, however, an rwsem is used for protecting the entire data
structure.

In addition, each link object has an internal status field whose
value reflects whether or not drivers are bound to the devices
pointed to by the link or probing/removal of their drivers is in
progress etc.  That field is only modified under the device links
mutex, but it may be read outside of it in some cases (introduced by
subsequent change sets), so modifications of it are annotated with
WRITE_ONCE().

New links are added by calling device_link_add() which takes three
arguments: pointers to the devices in question and flags.  In
particular, if DL_FLAG_STATELESS is set in the flags, the link status
is not to be taken into account for this link and the driver core
will not manage it.  In turn, if DL_FLAG_AUTOREMOVE is set in the
flags, the driver core will remove the link automatically when the
consumer device driver unbinds from it.

One of the actions carried out by device_link_add() is to reorder
the lists used for device shutdown and system suspend/resume to
put the consumer device along with all of its children and all of
its consumers (and so on, recursively) to the ends of those lists
in order to ensure the right ordering between all of the supplier
and consumer devices.

For this reason, it is not possible to create a link between two
devices if the would-be supplier device already depends on the
would-be consumer device as either a direct descendant of it or a
consumer of one of its direct descendants or one of its consumers
and so on.

There are two types of link objects, persistent and non-persistent.
The persistent ones stay around until one of the target devices is
deleted, while the non-persistent ones are removed automatically when
the consumer driver unbinds from its device (ie. they are assumed to
be valid only as long as the consumer device has a driver bound to
it).  Persistent links are created by default and non-persistent
links are created when the DL_FLAG_AUTOREMOVE flag is passed
to device_link_add().

Both persistent and non-persistent device links can be deleted
with an explicit call to device_link_del().

Links created without the DL_FLAG_STATELESS flag set are managed
by the driver core using a simple state machine.  There are 5 states
each link can be in: DORMANT (unused), AVAILABLE (the supplier driver
is present and functional), CONSUMER_PROBE (the consumer driver is
probing), ACTIVE (both supplier and consumer drivers are present and
functional), and SUPPLIER_UNBIND (the supplier driver is unbinding).
The driver core updates the link state automatically depending on
what happens to the linked devices and for each link state specific
actions are taken in addition to that.

For example, if the supplier driver unbinds from its device, the
driver core will also unbind the drivers of all of its consumers
automatically under the assumption that they cannot function
properly without the supplier.  Analogously, the driver core will
only allow the consumer driver to bind to its device if the
supplier driver is present and functional (ie. the link is in
the AVAILABLE state).  If that's not the case, it will rely on
the existing deferred probing mechanism to wait for the supplier
driver to become available.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Grygorii Strashko reports:

 The PM runtime will be left disabled for the device if its
 .suspend_late() callback fails and async suspend is not allowed
 for this device. In this case device will not be added in
 dpm_late_early_list and dpm_resume_early() will ignore this
 device, as result PM runtime will be disabled for it forever
 (side effect: after 8 subsequent failures for the same device
 the PM runtime will be reenabled due to disable_depth overflow).

To fix this problem, add devices to dpm_late_early_list regardless
of whether or not device_suspend_late() returns errors for them.

That will ensure failures in there to be handled consistently for
all devices regardless of their async suspend/resume status.
Reported-by: NGrygorii Strashko <grygorii.strashko@ti.com>
Tested-by: NGrygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: All applicable <stable@vger.kernel.org>

Commit 32e8d689 (PM / sleep: trace_device_pm_callback coverage in
dpm_prepare/complete) removed all users of this variable but forgot to
remove the variable itself.
Signed-off-by: NThierry Reding <treding@nvidia.com>
[ rjw: Subject ]
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

If a suitable prepare callback cannot be found for a given device and
its driver has no PM callbacks at all, assume that it can go direct to
complete when the system goes to sleep.

The reason for this is that there's lots of devices in a system that do
no PM at all and there's no reason for them to prevent their ancestors
to do direct_complete if they can support it.
Signed-off-by: NTomeu Vizoso <tomeu.vizoso@collabora.com>
Reviewed-by: NUlf Hansson <ulf.hansson@linaro.org>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

It is unsafe [1] if probing of devices will happen during suspend or
hibernation and system behavior will be unpredictable in this case.
So, let's prohibit device's probing in dpm_prepare() and defer their
probing instead. The normal behavior will be restored in
dpm_complete().

This patch introduces new DD core APIs:
 device_block_probing()
   It will disable probing of devices and defer their probes instead.
 device_unblock_probing()
   It will restore normal behavior and trigger re-probing of deferred
   devices.

[1] https://lkml.org/lkml/2015/9/11/554Signed-off-by: NGrygorii Strashko <grygorii.strashko@ti.com>
Acked-by: NPavel Machek <pavel@ucw.cz>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Don't unset the direct_complete flag on devices that have runtime PM
disabled, if they are runtime suspended.

This is needed because otherwise ancestor devices wouldn't be able to
do direct_complete without adding runtime PM support to all its
descendants.

Also removes pm_runtime_suspended_if_enabled() because it's now unused.
Signed-off-by: NTomeu Vizoso <tomeu.vizoso@collabora.com>
Signed-off-by: NAlan Stern <stern@rowland.harvard.edu>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Move the trace_device_pm_callback locations for dpm_prepare and dpm_complete
to encompass the attempt to capture the device mutex prior to callback. This
is needed by analyze_suspend to identify gaps in the trace output caused by
the delay in locking the mutex for a device.
Signed-off-by: NTodd Brandt <todd.e.brandt@linux.intel.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Turns out we can automate the handling for the device_may_wakeup()
quite a bit by using the kernel wakeup source list as suggested
by Rafael J. Wysocki <rjw@rjwysocki.net>.

And as some hardware has separate dedicated wake-up interrupt
in addition to the IO interrupt, we can automate the handling by
adding a generic threaded interrupt handler that just calls the
device PM runtime to wake up the device.

This allows dropping code from device drivers as we currently
are doing it in multiple ways, and often wrong.

For most drivers, we should be able to drop the following
boilerplate code from runtime_suspend and runtime_resume
functions:

	...
	device_init_wakeup(dev, true);
	...
	if (device_may_wakeup(dev))
		enable_irq_wake(irq);
	...
	if (device_may_wakeup(dev))
		disable_irq_wake(irq);
	...
	device_init_wakeup(dev, false);
	...

We can replace it with just the following init and exit
time code:

	...
	device_init_wakeup(dev, true);
	dev_pm_set_wake_irq(dev, irq);
	...
	dev_pm_clear_wake_irq(dev);
	device_init_wakeup(dev, false);
	...

And for hardware with dedicated wake-up interrupts:

	...
	device_init_wakeup(dev, true);
	dev_pm_set_dedicated_wake_irq(dev, irq);
	...
	dev_pm_clear_wake_irq(dev);
	device_init_wakeup(dev, false);
	...
Signed-off-by: NTony Lindgren <tony@atomide.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Occasionally, the system can't come back up after suspend/resume
due to problems of device suspending phase. This patch make
PM_TRACE infrastructure cover device suspending phase of
suspend/resume process, and the information in RTC can tell
developers which device suspending function make system hang.
Signed-off-by: NZhonghui Fu <zhonghui.fu@linux.intel.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

If an asynchronous suspend_late or freeze_late callback fails
during the SUSPEND, FREEZE or QUIESCE phases, we don't propagate the
corresponding error correctly, in effect ignoring the error and
continuing the suspend-to-ram/hibernation. During suspend-to-ram this
could leave some devices without a valid saved context, leading to a
failure to reinitialize them during resume. During hibernation this
could leave some devices active interfeering with the creation /
restoration of the hibernation image. Also this could leave the
corresponding devices without a valid saved context and failure to
reinitialize them during resume.

Fixes: de377b39 (PM / sleep: Asynchronous threads for suspend_late)
Signed-off-by: NImre Deak <imre.deak@intel.com>
Cc: 3.15+ <stable@vger.kernel.org> # 3.15+
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Subsequent change sets will add platform-related operations between
dpm_suspend_late() and dpm_suspend_noirq() as well as between
dpm_resume_noirq() and dpm_resume_early() in suspend_enter(), so
export these functions for suspend_enter() to be able to call them
separately and split the invocations of dpm_suspend_end() and
dpm_resume_start() in there accordingly.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Fix kernel-doc warnings in drivers/base/power/main.c:

Warning(..//drivers/base/power/main.c:473): No description found for parameter 'async'
Warning(..//drivers/base/power/main.c:601): No description found for parameter 'async'
Warning(..//drivers/base/power/main.c:1012): No description found for parameter 'async'
Warning(..//drivers/base/power/main.c:1151): No description found for parameter 'async'
Warning(..//drivers/base/power/main.c:1305): No description found for parameter 'info'
Signed-off-by: NRandy Dunlap <rdunlap@infradead.org>
Acked-by: NPavel Machek <pavel@ucw.cz>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Adds two trace events which supply the same info that initcall_debug
provides, but via ftrace instead of dmesg. The existing initcall_debug
calls require the pm_print_times_enabled var to be set (either via
sysfs or via the kernel cmd line). The new trace events provide all the
same info as the initcall_debug prints but with less overhead, and also
with coverage of device prepare and complete device callbacks.

These events replace the device_pm_report_time event (which has been
removed). device_pm_callback_start is called first and provides the device
and callback info. device_pm_callback_end is called after with the
device name and error info. The time and pid are gathered from the trace
data headers.
Signed-off-by: NTodd Brandt <todd.e.brandt@intel.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Adds trace events that give finer resolution into suspend/resume. These
events are graphed in the timelines generated by the analyze_suspend.py
script. They represent large areas of time consumed that are typical to
suspend and resume.

The event is triggered by calling the function "trace_suspend_resume"
with three arguments: a string (the name of the event to be displayed
in the timeline), an integer (case specific number, such as the power
state or cpu number), and a boolean (where true is used to denote the start
of the timeline event, and false to denote the end).

The suspend_resume trace event reproduces the data that the machine_suspend
trace event did, so the latter has been removed.
Signed-off-by: NTodd Brandt <todd.e.brandt@intel.com>
Acked-by: NSteven Rostedt <rostedt@goodmis.org>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Currently, some subsystems (e.g. PCI and the ACPI PM domain) have to
resume all runtime-suspended devices during system suspend, mostly
because those devices may need to be reprogrammed due to different
wakeup settings for system sleep and for runtime PM.

For some devices, though, it's OK to remain in runtime suspend
throughout a complete system suspend/resume cycle (if the device was in
runtime suspend at the start of the cycle).  We would like to do this
whenever possible, to avoid the overhead of extra power-up and power-down
events.

However, problems may arise because the device's descendants may require
it to be at full power at various points during the cycle.  Therefore the
most straightforward way to do this safely is if the device and all its
descendants can remain runtime suspended until the complete stage of
system resume.

To this end, introduce a new device PM flag, power.direct_complete
and modify the PM core to use that flag as follows.

If the ->prepare() callback of a device returns a positive number,
the PM core will regard that as an indication that it may leave the
device runtime-suspended.  It will then check if the system power
transition in progress is a suspend (and not hibernation in particular)
and if the device is, indeed, runtime-suspended.  In that case, the PM
core will set the device's power.direct_complete flag.  Otherwise it
will clear power.direct_complete for the device and it also will later
clear it for the device's parent (if there's one).

Next, the PM core will not invoke the ->suspend() ->suspend_late(),
->suspend_irq(), ->resume_irq(), ->resume_early(), or ->resume()
callbacks for all devices having power.direct_complete set.  It
will invoke their ->complete() callbacks, however, and those
callbacks are then responsible for resuming the devices as
appropriate, if necessary.  For example, in some cases they may
need to queue up runtime resume requests for the devices using
pm_request_resume().

Changelog partly based on an Alan Stern's description of the idea
(http://marc.info/?l=linux-pm&m=139940466625569&w=2).
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Acked-by: NAlan Stern <stern@rowland.harvard.edu>

This patch adds cpufreq suspend/resume calls to dpm_{suspend|resume}()
for handling suspend/resume of cpufreq governors.

Lan Tianyu (Intel) & Jinhyuk Choi (Broadcom) found an issue where the
tunables configuration for clusters/sockets with non-boot CPUs was
lost after system suspend/resume, as we were notifying governors with
CPUFREQ_GOV_POLICY_EXIT on removal of the last CPU for that policy
which caused the tunables memory to be freed.

This is fixed by preventing any governor operations from being
carried out between the device suspend and device resume stages of
system suspend and resume, respectively.

We could have added these callbacks at dpm_{suspend|resume}_noirq()
level, but there is an additional problem that the majority of I/O
devices is already suspended at that point and if cpufreq drivers
want to change the frequency before suspending, then that not be
possible on some platforms (which depend on peripherals like i2c,
regulators, etc).
Reported-and-tested-by: NLan Tianyu <tianyu.lan@intel.com>
Reported-by: NJinhyuk Choi <jinchoi@broadcom.com>
Signed-off-by: NViresh Kumar <viresh.kumar@linaro.org>
[rjw: Changelog]
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

In analogy with commits 5af84b82 and 97df8c12, using
asynchronous threads can improve the overall suspend_late
time significantly.

This patch is for suspend_late phase.
Signed-off-by: NChuansheng Liu <chuansheng.liu@intel.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

In analogy with commits 5af84b82 and 97df8c12, using
asynchronous threads can improve the overall suspend_noirq
time significantly.

This patch is for suspend_noirq phase.
Signed-off-by: NChuansheng Liu <chuansheng.liu@intel.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

In analogy with commits 5af84b82 and 97df8c12, using
asynchronous threads can improve the overall resume_early
time significantly.

This patch is for resume_early phase.
Signed-off-by: NChuansheng Liu <chuansheng.liu@intel.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

In analogy with commits 5af84b82 and 97df8c12, using
asynchronous threads can improve the overall resume_noirq time
significantly.

One typical case is:
In resume_noirq phase and for the PCI devices, the function
pci_pm_resume_noirq() will be called, and there is one d3_delay
(10ms) at least.

With the way of asynchronous threads, we just need wait d3_delay
time once in parallel for each calling, which saves much time to
resume quickly.
Signed-off-by: NChuansheng Liu <chuansheng.liu@intel.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>