linux-stable.git/kernel/sched/idle.c, branch v4.0

cpuidle / sleep: Use broadcast timer for states that stop local timer

2015-03-05T22:13:19+00:00

Commit 381063133246 (PM / sleep: Re-implement suspend-to-idle handling)
overlooked the fact that entering some sufficiently deep idle states
by CPUs may cause their local timers to stop and in those cases it
is necessary to switch over to a broadcast timer prior to entering
the idle state.  If the cpuidle driver in use does not provide
the new ->enter_freeze callback for any of the idle states, that
problem affects suspend-to-idle too, but it is not taken into account
after the changes made by commit 381063133246.

Fix that by changing the definition of cpuidle_enter_freeze() and
re-arranging of the code in cpuidle_idle_call(), so the former does
not call cpuidle_enter() any more and the fallback case is handled
by cpuidle_idle_call() directly.

Fixes: 381063133246 (PM / sleep: Re-implement suspend-to-idle handling)
Reported-and-tested-by: Lorenzo Pieralisi 
Signed-off-by: Rafael J. Wysocki 
Acked-by: Peter Zijlstra (Intel)

cpuidle: Clean up fallback handling in cpuidle_idle_call()

2015-03-02T21:25:37+00:00

Move the fallback code path in cpuidle_idle_call() to the end of the
function to avoid jumping to a label in an if () branch.

Signed-off-by: Rafael J. Wysocki

idle / sleep: Avoid excessive disabling and enabling interrupts

2015-02-28T22:46:24+00:00

Disabling interrupts at the end of cpuidle_enter_freeze() is not
useful, because its caller, cpuidle_idle_call(), re-enables them
right away after invoking it.

To avoid that unnecessary back and forth dance with interrupts,
make cpuidle_enter_freeze() enable interrupts after calling
enter_freeze_proper() and drop the local_irq_disable() at its
end, so that all of the code paths in it end up with interrupts
enabled.  Then, cpuidle_idle_call() will not need to re-enable
interrupts after calling cpuidle_enter_freeze() any more, because
the latter will return with interrupts enabled, in analogy with
cpuidle_enter().

Reported-by: Lorenzo Pieralisi 
Signed-off-by: Rafael J. Wysocki 
Acked-by: Peter Zijlstra (Intel)

PM / sleep: Re-implement suspend-to-idle handling

2015-02-13T22:49:36+00:00

In preparation for adding support for quiescing timers in the final
stage of suspend-to-idle transitions, rework the freeze_enter()
function making the system wait on a wakeup event, the freeze_wake()
function terminating the suspend-to-idle loop and the mechanism by
which deep idle states are entered during suspend-to-idle.

First of all, introduce a simple state machine for suspend-to-idle
and make the code in question use it.

Second, prevent freeze_enter() from losing wakeup events due to race
conditions and ensure that the number of online CPUs won't change
while it is being executed.  In addition to that, make it force
all of the CPUs re-enter the idle loop in case they are in idle
states already (so they can enter deeper idle states if possible).

Next, drop cpuidle_use_deepest_state() and replace use_deepest_state
checks in cpuidle_select() and cpuidle_reflect() with a single
suspend-to-idle state check in cpuidle_idle_call().

Finally, introduce cpuidle_enter_freeze() that will simply find the
deepest idle state available to the given CPU and enter it using
cpuidle_enter().

Signed-off-by: Rafael J. Wysocki 
Acked-by: Peter Zijlstra (Intel)

sched/idle: Add missing checks to the exit condition of cpu_idle_poll()

2015-01-30T18:38:52+00:00

cpu_idle_poll() is entered into when either the cpu_idle_force_poll is set or
tick_check_broadcast_expired() returns true. The exit condition from
cpu_idle_poll() is tif_need_resched().

However this does not take into account scenarios where cpu_idle_force_poll
changes or tick_check_broadcast_expired() returns false, without setting
the resched flag. So a cpu will be caught in cpu_idle_poll() needlessly,
thereby wasting power. Add an explicit check on cpu_idle_force_poll and
tick_check_broadcast_expired() to the exit condition of cpu_idle_poll()
to avoid this.

Signed-off-by: Preeti U Murthy 
Reviewed-by: Thomas Gleixner 
Signed-off-by: Peter Zijlstra (Intel) 
Cc: linuxppc-dev@lists.ozlabs.org
Cc: Linus Torvalds 
Link: http://lkml.kernel.org/r/20150121105655.15279.59626.stgit@preeti.in.ibm.com
Signed-off-by: Ingo Molnar

sched: Let the scheduler see CPU idle states

2014-09-24T12:46:58+00:00

When the cpu enters idle, it stores the cpuidle state pointer in its
struct rq instance which in turn could be used to make a better decision
when balancing tasks.

As soon as the cpu exits its idle state, the struct rq reference is
cleared.

There are a couple of situations where the idle state pointer could be changed
while it is being consulted:

1. For x86/acpi with dynamic c-states, when a laptop switches from battery
   to AC that could result on removing the deeper idle state. The acpi driver
   triggers:
	'acpi_processor_cst_has_changed'
		'cpuidle_pause_and_lock'
			'cpuidle_uninstall_idle_handler'
				'kick_all_cpus_sync'.

All cpus will exit their idle state and the pointed object will be set to
NULL.

2. The cpuidle driver is unloaded. Logically that could happen but not
in practice because the drivers are always compiled in and 95% of them are
not coded to unregister themselves.  In any case, the unloading code must
call 'cpuidle_unregister_device', that calls 'cpuidle_pause_and_lock'
leading to 'kick_all_cpus_sync' as mentioned above.

A race can happen if we use the pointer and then one of these two scenarios
occurs at the same moment.

In order to be safe, the idle state pointer stored in the rq must be
used inside a rcu_read_lock section where we are protected with the
'rcu_barrier' in the 'cpuidle_uninstall_idle_handler' function. The
idle_get_state() and idle_put_state() accessors should be used to that
effect.

Signed-off-by: Daniel Lezcano 
Signed-off-by: Nicolas Pitre 
Signed-off-by: Peter Zijlstra (Intel) 
Cc: "Rafael J. Wysocki" 
Cc: linux-pm@vger.kernel.org
Cc: linaro-kernel@lists.linaro.org
Cc: Daniel Lezcano 
Cc: Linus Torvalds 
Link: http://lkml.kernel.org/n/tip-@git.kernel.org
Signed-off-by: Ingo Molnar

Merge tag 'pm+acpi-3.17-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm

2014-08-07T03:34:19+00:00

Pull ACPI and power management updates from Rafael Wysocki:
"Again, ACPICA leads the pack (47 commits), followed by cpufreq (18
commits) and system suspend/hibernation (9 commits).

From the new code perspective, the ACPICA update brings ACPI 5.1 to
the table, including a new device configuration object called _DSD
(Device Specific Data) that will hopefully help us to operate device
properties like Device Trees do (at least to some extent) and changes
related to supporting ACPI on ARM.

Apart from that we have hibernation changes making it use radix trees
to store memory bitmaps which should speed up some operations carried
out by it quite significantly. We also have some power management
changes related to suspend-to-idle (the "freeze" sleep state) support
and more preliminary changes needed to support ACPI on ARM (outside of
ACPICA).

The rest is fixes and cleanups pretty much everywhere.

Specifics:

- ACPICA update to upstream version 20140724. That includes ACPI 5.1
material (support for the _CCA and _DSD predefined names, changes
related to the DMAR and PCCT tables and ARM support among other
things) and cleanups related to using ACPICA's header files. A
major part of it is related to acpidump and the core code used by
that utility. Changes from Bob Moore, David E Box, Lv Zheng,
Sascha Wildner, Tomasz Nowicki, Hanjun Guo.

- Radix trees for memory bitmaps used by the hibernation core from
Joerg Roedel.

- Support for waking up the system from suspend-to-idle (also known
as the "freeze" sleep state) using ACPI-based PCI wakeup signaling
(Rafael J Wysocki).

- Fixes for issues related to ACPI button events (Rafael J Wysocki).

- New device ID for an ACPI-enumerated device included into the
Wildcat Point PCH from Jie Yang.

- ACPI video updates related to backlight handling from Hans de Goede
and Linus Torvalds.

- Preliminary changes needed to support ACPI on ARM from Hanjun Guo
and Graeme Gregory.

- ACPI PNP core cleanups from Arjun Sreedharan and Zhang Rui.

- Cleanups related to ACPI_COMPANION() and ACPI_HANDLE() macros
(Rafael J Wysocki).

- ACPI-based device hotplug cleanups from Wei Yongjun and Rafael J
Wysocki.

- Cleanups and improvements related to system suspend from Lan
Tianyu, Randy Dunlap and Rafael J Wysocki.

- ACPI battery cleanup from Wei Yongjun.

- cpufreq core fixes from Viresh Kumar.

- Elimination of a deadband effect from the cpufreq ondemand governor
and intel_pstate driver cleanups from Stratos Karafotis.

- 350MHz CPU support for the powernow-k6 cpufreq driver from Mikulas
Patocka.

- Fix for the imx6 cpufreq driver from Anson Huang.

- cpuidle core and governor cleanups from Daniel Lezcano, Sandeep
Tripathy and Mohammad Merajul Islam Molla.

- Build fix for the big_little cpuidle driver from Sachin Kamat.

- Configuration fix for the Operation Performance Points (OPP)
framework from Mark Brown.

- APM cleanup from Jean Delvare.

- cpupower utility fixes and cleanups from Peter Senna Tschudin,
Andrey Utkin, Himangi Saraogi, Rickard Strandqvist, Thomas
Renninger"

* tag 'pm+acpi-3.17-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm: (118 commits)
ACPI / LPSS: add LPSS device for Wildcat Point PCH
ACPI / PNP: Replace faulty is_hex_digit() by isxdigit()
ACPICA: Update version to 20140724.
ACPICA: ACPI 5.1: Update for PCCT table changes.
ACPICA/ARM: ACPI 5.1: Update for GTDT table changes.
ACPICA/ARM: ACPI 5.1: Update for MADT changes.
ACPICA/ARM: ACPI 5.1: Update for FADT changes.
ACPICA: ACPI 5.1: Support for the _CCA predifined name.
ACPICA: ACPI 5.1: New notify value for System Affinity Update.
ACPICA: ACPI 5.1: Support for the _DSD predefined name.
ACPICA: Debug object: Add current value of Timer() to debug line prefix.
ACPICA: acpihelp: Add UUID support, restructure some existing files.
ACPICA: Utilities: Fix local printf issue.
ACPICA: Tables: Update for DMAR table changes.
ACPICA: Remove some extraneous printf arguments.
ACPICA: Update for comments/formatting. No functional changes.
ACPICA: Disassembler: Add support for the ToUUID opererator (macro).
ACPICA: Remove a redundant cast to acpi_size for ACPI_OFFSET() macro.
ACPICA: Work around an ancient GCC bug.
ACPI / processor: Make it possible to get local x2apic id via _MAT
...

cpuidle: move idle traces to cpuidle_enter_state()

2014-07-09T13:45:23+00:00

idle_exit event is the first event after a core exits
idle state. So this should be traced before local irq
is ebabled. Likewise idle_entry is the last event before
a core enters idle state. This will ease visualising the
cpu idle state from kernel traces.

Signed-off-by: Sandeep Tripathy 
Acked-by: Daniel Lezcano 
[rjw: Subject, rebase]
Signed-off-by: Rafael J. Wysocki

sched/idle: Drop !! while calculating 'broadcast'

2014-07-05T09:17:31+00:00

We don't need 'broadcast' to be set to 'zero or one', but to 'zero or non-zero'
and so the extra operation to convert it to 'zero or one' can be skipped.

Also change type of 'broadcast' to unsigned int, i.e. type of
drv->states[*].flags.

Signed-off-by: Viresh Kumar 
Cc: linaro-kernel@lists.linaro.org
Signed-off-by: Peter Zijlstra 
Link: http://lkml.kernel.org/r/0dfbe2976aa108c53e08d3477ea90f6360c1f54c.1403584026.git.viresh.kumar@linaro.org
Signed-off-by: Ingo Molnar

sched/idle: Optimize try-to-wake-up IPI

2014-06-05T10:09:53+00:00

[ This series reduces the number of IPIs on Andy's workload by something like
  99%. It's down from many hundreds per second to very few.

  The basic idea behind this series is to make TIF_POLLING_NRFLAG be a
  reliable indication that the idle task is polling.  Once that's done,
  the rest is reasonably straightforward. ]

When enqueueing tasks on remote LLC domains, we send an IPI to do the
work 'locally' and avoid bouncing all the cachelines over.

However, when the remote CPU is idle (and polling, say x86 mwait), we
don't need to send an IPI, we can simply kick the TIF word to wake it
up and have the 'idle' loop do the work.

So when _TIF_POLLING_NRFLAG is set, but _TIF_NEED_RESCHED is not (yet)
set, set _TIF_NEED_RESCHED and avoid sending the IPI.

Much-requested-by: Andy Lutomirski 
Signed-off-by: Peter Zijlstra 
[Edited by Andy Lutomirski, but this is mostly Peter Zijlstra's code.]
Signed-off-by: Andy Lutomirski 
Cc: nicolas.pitre@linaro.org
Cc: daniel.lezcano@linaro.org
Cc: Mike Galbraith 
Cc: umgwanakikbuti@gmail.com
Cc: Rafael J. Wysocki 
Cc: Linus Torvalds 
Cc: linux-kernel@vger.kernel.org
Link: http://lkml.kernel.org/r/ce06f8b02e7e337be63e97597fc4b248d3aa6f9b.1401902905.git.luto@amacapital.net
Signed-off-by: Ingo Molnar