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 <daniel.lezcano@linaro.org>
Signed-off-by: Nicolas Pitre <nico@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>
Cc: linux-pm@vger.kernel.org
Cc: linaro-kernel@lists.linaro.org
Cc: Daniel Lezcano <daniel.lezcano@linaro.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/n/tip-@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
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 <sandeep.tripathy@linaro.org>
Acked-by: Daniel Lezcano <daniel.lezcano@linaro.org>
[rjw: Subject, rebase]
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
If freeze_enter() is called, we want to bypass the current cpuidle
governor and always use the deepest available (that is, not disabled)
C-state, because we want to save as much energy as reasonably possible
then and runtime latency constraints don't matter at that point, since
the system is in a sleep state anyway.
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Tested-by: Aubrey Li <aubrey.li@linux.intel.com>
Since both cpuidle_enabled() and cpuidle_select() are only called by
cpuidle_idle_call(), it is not really useful to keep them separate
and combining them will help to avoid complicating cpuidle_idle_call()
even further if governors are changed to return error codes sometimes.
This code modification shouldn't lead to any functional changes.
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Pull sched/idle changes from Ingo Molnar:
"More idle code reorganization, to prepare for more integration.
(Sent separately because it depended on pending timer work, which is
now upstream)"
* 'sched-idle-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/idle: Add more comments to the code
sched/idle: Move idle conditions in cpuidle_idle main function
sched/idle: Reorganize the idle loop
cpuidle/idle: Move the cpuidle_idle_call function to idle.c
idle/cpuidle: Split cpuidle_idle_call main function into smaller functions
- Device PM QoS support for latency tolerance constraints on systems with
hardware interfaces allowing such constraints to be specified. That is
necessary to prevent hardware-driven power management from becoming
overly aggressive on some systems and to prevent power management
features leading to excessive latencies from being used in some cases.
- Consolidation of the handling of ACPI hotplug notifications for device
objects. This causes all device hotplug notifications to go through
the root notify handler (that was executed for all of them anyway
before) that propagates them to individual subsystems, if necessary,
by executing callbacks provided by those subsystems (those callbacks
are associated with struct acpi_device objects during device
enumeration). As a result, the code in question becomes both smaller
in size and more straightforward and all of those changes should not
affect users.
- ACPICA update, including fixes related to the handling of _PRT in cases
when it is broken and the addition of "Windows 2013" to the list of
supported "features" for _OSI (which is necessary to support systems
that work incorrectly or don't even boot without it). Changes from
Bob Moore and Lv Zheng.
- Consolidation of ACPI _OST handling from Jiang Liu.
- ACPI battery and AC fixes allowing unusual system configurations to
be handled by that code from Alexander Mezin.
- New device IDs for the ACPI LPSS driver from Chiau Ee Chew.
- ACPI fan and thermal optimizations related to system suspend and resume
from Aaron Lu.
- Cleanups related to ACPI video from Jean Delvare.
- Assorted ACPI fixes and cleanups from Al Stone, Hanjun Guo, Lan Tianyu,
Paul Bolle, Tomasz Nowicki.
- Intel RAPL (Running Average Power Limits) driver cleanups from Jacob Pan.
- intel_pstate fixes and cleanups from Dirk Brandewie.
- cpufreq fixes related to system suspend/resume handling from Viresh Kumar.
- cpufreq core fixes and cleanups from Viresh Kumar, Stratos Karafotis,
Saravana Kannan, Rashika Kheria, Joe Perches.
- cpufreq drivers updates from Viresh Kumar, Zhuoyu Zhang, Rob Herring.
- cpuidle fixes related to the menu governor from Tuukka Tikkanen.
- cpuidle fix related to coupled CPUs handling from Paul Burton.
- Asynchronous execution of all device suspend and resume callbacks,
except for ->prepare and ->complete, during system suspend and resume
from Chuansheng Liu.
- Delayed resuming of runtime-suspended devices during system suspend for
the PCI bus type and ACPI PM domain.
- New set of PM helper routines to allow device runtime PM callbacks to
be used during system suspend and resume more easily from Ulf Hansson.
- Assorted fixes and cleanups in the PM core from Geert Uytterhoeven,
Prabhakar Lad, Philipp Zabel, Rashika Kheria, Sebastian Capella.
- devfreq fix from Saravana Kannan.
/
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.22 (GNU/Linux)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=ywR6
-----END PGP SIGNATURE-----
Merge tag 'pm+acpi-3.15-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
Pull ACPI and power management updates from Rafael Wysocki:
"The majority of this material spent some time in linux-next, some of
it even several weeks. There are a few relatively fresh commits in
it, but they are mostly fixes and simple cleanups.
ACPI took the lead this time, both in terms of the number of commits
and the number of modified lines of code, cpufreq follows and there
are a few changes in the PM core and in cpuidle too.
A new feature that already got some LWN.net's attention is the device
PM QoS extension allowing latency tolerance requirements to be
propagated from leaf devices to their ancestors with hardware
interfaces for specifying latency tolerance. That should help systems
with hardware-driven power management to avoid going too far with it
in cases when there are latency tolerance constraints.
There also are some significant changes in the ACPI core related to
the way in which hotplug notifications are handled. They affect PCI
hotplug (ACPIPHP) and the ACPI dock station code too. The bottom line
is that all those notification now go through the root notify handler
and are propagated to the interested subsystems by means of callbacks
instead of having to install a notify handler for each device object
that we can potentially get hotplug notifications for.
In addition to that ACPICA will now advertise "Windows 2013"
compatibility for _OSI, because some systems out there don't work
correctly if that is not done (some of them don't even boot).
On the system suspend side of things, all of the device suspend and
resume callbacks, except for ->prepare() and ->complete(), are now
going to be executed asynchronously as that turns out to speed up
system suspend and resume on some platforms quite significantly and we
have a few more optimizations in that area.
Apart from that, there are some new device IDs and fixes and cleanups
all over. In particular, the system suspend and resume handling by
cpufreq should be improved and the cpuidle menu governor should be a
bit more robust now.
Specifics:
- Device PM QoS support for latency tolerance constraints on systems
with hardware interfaces allowing such constraints to be specified.
That is necessary to prevent hardware-driven power management from
becoming overly aggressive on some systems and to prevent power
management features leading to excessive latencies from being used
in some cases.
- Consolidation of the handling of ACPI hotplug notifications for
device objects. This causes all device hotplug notifications to go
through the root notify handler (that was executed for all of them
anyway before) that propagates them to individual subsystems, if
necessary, by executing callbacks provided by those subsystems
(those callbacks are associated with struct acpi_device objects
during device enumeration). As a result, the code in question
becomes both smaller in size and more straightforward and all of
those changes should not affect users.
- ACPICA update, including fixes related to the handling of _PRT in
cases when it is broken and the addition of "Windows 2013" to the
list of supported "features" for _OSI (which is necessary to
support systems that work incorrectly or don't even boot without
it). Changes from Bob Moore and Lv Zheng.
- Consolidation of ACPI _OST handling from Jiang Liu.
- ACPI battery and AC fixes allowing unusual system configurations to
be handled by that code from Alexander Mezin.
- New device IDs for the ACPI LPSS driver from Chiau Ee Chew.
- ACPI fan and thermal optimizations related to system suspend and
resume from Aaron Lu.
- Cleanups related to ACPI video from Jean Delvare.
- Assorted ACPI fixes and cleanups from Al Stone, Hanjun Guo, Lan
Tianyu, Paul Bolle, Tomasz Nowicki.
- Intel RAPL (Running Average Power Limits) driver cleanups from
Jacob Pan.
- intel_pstate fixes and cleanups from Dirk Brandewie.
- cpufreq fixes related to system suspend/resume handling from Viresh
Kumar.
- cpufreq core fixes and cleanups from Viresh Kumar, Stratos
Karafotis, Saravana Kannan, Rashika Kheria, Joe Perches.
- cpufreq drivers updates from Viresh Kumar, Zhuoyu Zhang, Rob
Herring.
- cpuidle fixes related to the menu governor from Tuukka Tikkanen.
- cpuidle fix related to coupled CPUs handling from Paul Burton.
- Asynchronous execution of all device suspend and resume callbacks,
except for ->prepare and ->complete, during system suspend and
resume from Chuansheng Liu.
- Delayed resuming of runtime-suspended devices during system suspend
for the PCI bus type and ACPI PM domain.
- New set of PM helper routines to allow device runtime PM callbacks
to be used during system suspend and resume more easily from Ulf
Hansson.
- Assorted fixes and cleanups in the PM core from Geert Uytterhoeven,
Prabhakar Lad, Philipp Zabel, Rashika Kheria, Sebastian Capella.
- devfreq fix from Saravana Kannan"
* tag 'pm+acpi-3.15-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm: (162 commits)
PM / devfreq: Rewrite devfreq_update_status() to fix multiple bugs
PM / sleep: Correct whitespace errors in <linux/pm.h>
intel_pstate: Set core to min P state during core offline
cpufreq: Add stop CPU callback to cpufreq_driver interface
cpufreq: Remove unnecessary braces
cpufreq: Fix checkpatch errors and warnings
cpufreq: powerpc: add cpufreq transition latency for FSL e500mc SoCs
MAINTAINERS: Reorder maintainer addresses for PM and ACPI
PM / Runtime: Update runtime_idle() documentation for return value meaning
video / output: Drop display output class support
fujitsu-laptop: Drop unneeded include
acer-wmi: Stop selecting VIDEO_OUTPUT_CONTROL
ACPI / gpu / drm: Stop selecting VIDEO_OUTPUT_CONTROL
ACPI / video: fix ACPI_VIDEO dependencies
cpufreq: remove unused notifier: CPUFREQ_{SUSPENDCHANGE|RESUMECHANGE}
cpufreq: Do not allow ->setpolicy drivers to provide ->target
cpufreq: arm_big_little: set 'physical_cluster' for each CPU
cpufreq: arm_big_little: make vexpress driver depend on bL core driver
ACPI / button: Add ACPI Button event via netlink routine
ACPI: Remove duplicate definitions of PREFIX
...
As described by a comment at the end of cpuidle_enter_state_coupled it
can be inefficient for coupled idle states to return with IRQs enabled
since they may proceed to service an interrupt instead of clearing the
coupled idle state. Until they have finished & cleared the idle state
all CPUs coupled with them will spin rather than being able to enter a
safe idle state.
Commits e1689795a7 "cpuidle: Add common time keeping and irq
enabling" and 554c06ba3e "cpuidle: remove en_core_tk_irqen flag" led
to the cpuidle_enter_state enabling interrupts for all idle states,
including coupled ones, making this inefficiency unavoidable by drivers
& the local_irq_enable near the end of cpuidle_enter_state_coupled
redundant. This patch avoids enabling interrupts in cpuidle_enter_state
after a coupled state has been entered, allowing them to remain disabled
until all coupled CPUs have exited the idle state and
cpuidle_enter_state_coupled re-enables them.
Cc: Daniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: Paul Burton <paul.burton@imgtec.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
The cpuidle_idle_call does nothing more than calling the three individuals
function and is no longer used by any arch specific code but only in the
cpuidle framework code.
We can move this function into the idle task code to ensure better
proximity to the scheduler code.
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Acked-by: Nicolas Pitre <nicolas.pitre@linaro.org>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: rjw@rjwysocki.net
Cc: preeti@linux.vnet.ibm.com
Link: http://lkml.kernel.org/r/1393832934-11625-2-git-send-email-daniel.lezcano@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
In order to allow better integration between the cpuidle framework and the
scheduler, reducing the distance between these two sub-components will
facilitate this integration by moving part of the cpuidle code in the idle
task file and, because idle.c is in the sched directory, we have access to
the scheduler's private structures.
This patch splits the cpuidle_idle_call main entry function into 3 calls
to a newly added API:
1. select the idle state
2. enter the idle state
3. reflect the idle state
The cpuidle_idle_call calls these three functions to implement the main
idle entry function.
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Acked-by: Nicolas Pitre <nico@linaro.org>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: rjw@rjwysocki.net
Cc: preeti@linux.vnet.ibm.com
Link: http://lkml.kernel.org/r/1393832934-11625-1-git-send-email-daniel.lezcano@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Some archs set the CPUIDLE_FLAG_TIMER_STOP flag for idle states in which the
local timers stop. The cpuidle_idle_call() currently handles such idle states
by calling into the broadcast framework so as to wakeup CPUs at their next
wakeup event. With the hrtimer mode of broadcast, the BROADCAST_ENTER call
into the broadcast frameowork can fail for archs that do not have an external
clock device to handle wakeups and the CPU in question has thus to be made
the stand by CPU. This patch handles such cases by failing the call into
cpuidle so that the arch can take some default action. The arch will certainly
not enter a similar idle state because a failed cpuidle call will also implicitly
indicate that the broadcast framework has not registered this CPU to be woken up.
Hence we are safe if we fail the cpuidle call.
In the process move the functions that trace idle statistics just before and
after the entry and exit into idle states respectively. In other
scenarios where the call to cpuidle fails, we end up not tracing idle
entry and exit since a decision on an idle state could not be taken. Similarly
when the call to broadcast framework fails, we skip tracing idle statistics
because we are in no further position to take a decision on an alternative
idle state to enter into.
Signed-off-by: Preeti U Murthy <preeti@linux.vnet.ibm.com>
Cc: deepthi@linux.vnet.ibm.com
Cc: paulmck@linux.vnet.ibm.com
Cc: fweisbec@gmail.com
Cc: paulus@samba.org
Cc: srivatsa.bhat@linux.vnet.ibm.com
Cc: svaidy@linux.vnet.ibm.com
Cc: peterz@infradead.org
Cc: benh@kernel.crashing.org
Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: linuxppc-dev@lists.ozlabs.org
Link: http://lkml.kernel.org/r/20140207080652.17187.66344.stgit@preeti.in.ibm.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
If not, we could end up in the unfortunate situation where
we dereference a NULL pointer b/c we have cpuidle disabled.
This is the case when booting under Xen (which uses the
ACPI P/C states but disables the CPU idle driver) - and can
be easily reproduced when booting with cpuidle.off=1.
BUG: unable to handle kernel NULL pointer dereference at (null)
IP: [<ffffffff8156db4a>] cpuidle_unregister_device+0x2a/0x90
.. snip..
Call Trace:
[<ffffffff813b15b4>] acpi_processor_power_exit+0x3c/0x5c
[<ffffffff813af0a9>] acpi_processor_stop+0x61/0xb6
[<ffffffff814215bf>] __device_release_driver+0fffff81421653>] device_release_driver+0x23/0x30
[<ffffffff81420ed8>] bus_remove_device+0x108/0x180
[<ffffffff8141d9d9>] device_del+0x129/0x1c0
[<ffffffff813cb4b0>] ? unregister_xenbus_watch+0x1f0/0x1f0
[<ffffffff8141da8e>] device_unregister+0x1e/0x60
[<ffffffff814243e9>] unregister_cpu+0x39/0x60
[<ffffffff81019e03>] arch_unregister_cpu+0x23/0x30
[<ffffffff813c3c51>] handle_vcpu_hotplug_event+0xc1/0xe0
[<ffffffff813cb4f5>] xenwatch_thread+0x45/0x120
[<ffffffff810af010>] ? abort_exclusive_wait+0xb0/0xb0
[<ffffffff8108ec42>] kthread+0xd2/0xf0
[<ffffffff8108eb70>] ? kthread_create_on_node+0x180/0x180
[<ffffffff816ce17c>] ret_from_fork+0x7c/0xb0
[<ffffffff8108eb70>] ? kthread_create_on_node+0x180/0x180
This problem also appears in 3.12 and could be a candidate for backport.
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Cc: All applicable <stable@vger.kernel.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
poll_idle_init() just initializes drv->states[0] and so that is
required to be done only once for each driver. Currently, it is
called from cpuidle_enable_device() which is called for every CPU
that the driver supports. That is not required, so move it to a
better place and call it from __cpuidle_register_driver() so that
the initialization is carried out only once.
Acked-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Few statements in cpuidle_idle_call() are broken into multiple lines,
although that isn't really necessary. Convert those to single line.
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
We are doing this twice in cpuidle_idle_call() routine:
drv->states[next_state].flags & CPUIDLE_FLAG_TIMER_STOP
Would be better if we actually store this in a local variable and
use that. That reduces code duplication and likely makes this piece
of code run faster (in case the compiler wasn't able to optimize it
earlier)
[rjw: Cast the result of bitwise AND to bool explicitly using !!]
Acked-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Two checks cpuidle_idle_call() cause the same error code to be
returned if they fail, so merge them for clarity.
[rjw: Changelog]
Acked-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
This patch rearranges __cpuidle_register_device() a bit in order to
reduce the number of exit points in that function.
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
The only value returned by __cpuidle_device_init() is 0, so it very
well may be a void function. Make that happen.
[rjw: Changelog]
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Some comments in cpuidle core files contain trivial mistakes.
This patch fixes them.
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Make __cpuidle_register_device() check whether or not the device has
been registered already and return -EBUSY immediately if that's the
case.
[rjw: Changelog]
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Add __cpuidle_device_init() for initializing the cpuidle_device
structure.
[rjw: Changelog]
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
To reduce code duplication related to the unregistration of cpuidle
devices, introduce __cpuidle_unregister_device() and move all of the
unregistration code to that function.
[rjw: Changelog]
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
We previously changed the ordering of the cpuidle framework
initialization so that the governors are registered before the
drivers which can register their devices right from the start.
Now, we can safely remove the __cpuidle_register_device() call hack
in cpuidle_enable_device() and check if the driver has been
registered before enabling it. Then, cpuidle_register_device() can
consistently check the cpuidle_enable_device() return value when
enabling the device.
[rjw: Changelog]
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Commit bf4d1b5 (cpuidle: support multiple drivers) introduced support
for using multiple cpuidle drivers at the same time. It added a
couple of new APIs to register the driver per CPU, but that led to
some unnecessary code complexity related to the kernel config options
deciding whether or not the multiple driver support is enabled. The
code has to work as it did before when the multiple driver support is
not enabled and the multiple driver support has to be compatible with
the previously existing API.
Remove the new API, not used by any driver in the tree yet (but
needed for the HMP cpuidle drivers that will be submitted soon), and
add a new cpumask pointer to the cpuidle driver structure that will
point to the mask of CPUs handled by the given driver. That will
allow the cpuidle_[un]register_driver() API to be used for the
multiple driver support along with the cpuidle_[un]register()
functions added recently.
[rjw: Changelog]
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Fix comment format for the kernel doc script.
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
The usual scheme to initialize a cpuidle driver on a SMP is:
cpuidle_register_driver(drv);
for_each_possible_cpu(cpu) {
device = &per_cpu(cpuidle_dev, cpu);
cpuidle_register_device(device);
}
This code is duplicated in each cpuidle driver.
On UP systems, it is done this way:
cpuidle_register_driver(drv);
device = &per_cpu(cpuidle_dev, cpu);
cpuidle_register_device(device);
On UP, the macro 'for_each_cpu' does one iteration:
#define for_each_cpu(cpu, mask) \
for ((cpu) = 0; (cpu) < 1; (cpu)++, (void)mask)
Hence, the initialization loop is the same for UP than SMP.
Beside, we saw different bugs / mis-initialization / return code unchecked in
the different drivers, the code is duplicated including bugs. After fixing all
these ones, it appears the initialization pattern is the same for everyone.
Please note, some drivers are doing dev->state_count = drv->state_count. This is
not necessary because it is done by the cpuidle_enable_device function in the
cpuidle framework. This is true, until you have the same states for all your
devices. Otherwise, the 'low level' API should be used instead with the specific
initialization for the driver.
Let's add a wrapper function doing this initialization with a cpumask parameter
for the coupled idle states and use it for all the drivers.
That will save a lot of LOC, consolidate the code, and the modifications in the
future could be done in a single place. Another benefit is the consolidation of
the cpuidle_device variable which is now in the cpuidle framework and no longer
spread accross the different arch specific drivers.
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
The en_core_tk_irqen flag is set in all the cpuidle driver which
means it is not necessary to specify this flag.
Remove the flag and the code related to it.
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Acked-by: Kevin Hilman <khilman@linaro.org> # for mach-omap2/*
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
When a cpu enters a deep idle state, the local timers are stopped and
the time framework falls back to the timer device used as a broadcast
timer.
The different cpuidle drivers are calling clockevents_notify ENTER/EXIT
when the idle state stops the local timer.
Add a new flag CPUIDLE_FLAG_TIMER_STOP which can be set by the cpuidle
drivers. If the flag is set, the cpuidle core code takes care of the
notification on behalf of the driver to avoid pointless code duplication.
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
The text in Documentation said it would be removed in 2.6.41;
the text in the Kconfig said removal in the 3.1 release. Either
way you look at it, we are well past both, so push it off a cliff.
Note that the POWER_CSTATE and the POWER_PSTATE are part of the
legacy tracing API. Remove all tracepoints which use these flags.
As can be seen from context, most already have a trace entry via
trace_cpu_idle anyways.
Also, the cpufreq/cpufreq.c PSTATE one is actually unpaired, as
compared to the CSTATE ones which all have a clear start/stop.
As part of this, the trace_power_frequency also becomes orphaned,
so it too is deleted.
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
We realized that the power usage field is never filled and when it
is filled for tegra, the power_specified flag is not set causing all
of these values to be reset when the driver is initialized with
set_power_state().
However, the power_specified flag can be simply removed under the
assumption that the states are always backward sorted, which is the
case with the current code.
This change allows the menu governor select function and the
cpuidle_play_dead() to be simplified. Moreover, the
set_power_states() function can removed as it does not make sense
any more.
Drop the power_specified flag from struct cpuidle_driver and make
the related changes as described above.
As a consequence, this also fixes the bug where on the dynamic
C-states system, the power fields are not initialized.
[rjw: Changelog]
References: https://bugzilla.kernel.org/show_bug.cgi?id=42870
References: https://bugzilla.kernel.org/show_bug.cgi?id=43349
References: https://lkml.org/lkml/2012/10/16/518
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Since cpuidle_state.power_usage is a signed value, use INT_MAX (instead
of -1) to init the local copies so that functions that tries to find
cpuidle states with minimum power usage works correctly even if they use
non-negative values.
Signed-off-by: Sivaram Nair <sivaramn@nvidia.com>
Reviewed-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Many cpuidle drivers measure their time spent in an idle state by
reading the wallclock time before and after idling and calculating the
difference. This leads to erroneous results when the wallclock time gets
updated by another processor in the meantime, adding that clock
adjustment to the idle state's time counter.
If the clock adjustment was negative, the result is even worse due to an
erroneous cast from int to unsigned long long of the last_residency
variable. The negative 32 bit integer will zero-extend and result in a
forward time jump of roughly four billion milliseconds or 1.3 hours on
the idle state residency counter.
This patch changes all affected cpuidle drivers to either use the
monotonic clock for their measurements or make use of the generic time
measurement wrapper in cpuidle.c, which was already working correctly.
Some superfluous CLIs/STIs in the ACPI code are removed (interrupts
should always already be disabled before entering the idle function, and
not get reenabled until the generic wrapper has performed its second
measurement). It also removes the erroneous cast, making sure that
negative residency values are applied correctly even though they should
not appear anymore.
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-by: Preeti U Murthy <preeti@linux.vnet.ibm.com>
Tested-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Acked-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Acked-by: Len Brown <len.brown@intel.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
With the tegra3 and the big.LITTLE [1] new architectures, several cpus
with different characteristics (latencies and states) can co-exists on the
system.
The cpuidle framework has the limitation of handling only identical cpus.
This patch removes this limitation by introducing the multiple driver support
for cpuidle.
This option is configurable at compile time and should be enabled for the
architectures mentioned above. So there is no impact for the other platforms
if the option is disabled. The option defaults to 'n'. Note the multiple drivers
support is also compatible with the existing drivers, even if just one driver is
needed, all the cpu will be tied to this driver using an extra small chunk of
processor memory.
The multiple driver support use a per-cpu driver pointer instead of a global
variable and the accessor to this variable are done from a cpu context.
In order to keep the compatibility with the existing drivers, the function
'cpuidle_register_driver' and 'cpuidle_unregister_driver' will register
the specified driver for all the cpus.
The semantic for the output of /sys/devices/system/cpu/cpuidle/current_driver
remains the same except the driver name will be related to the current cpu.
The /sys/devices/system/cpu/cpu[0-9]/cpuidle/driver/name files are added
allowing to read the per cpu driver name.
[1] http://lwn.net/Articles/481055/
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Acked-by: Peter De Schrijver <pdeschrijver@nvidia.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
When cpuidle governor choose a C-state to enter for idle CPU, but it notice that
there is tasks request to be executed. So the idle CPU will not really enter
the target C-state and go to run task.
In this situation, it will use the residency of previous really entered target
C-states. Obviously, it is not reasonable.
So, this patch fix it by set the target C-state residency to 0.
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Youquan Song <youquan.song@intel.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Move the kobj initialization and completion in the sysfs.c
and encapsulate the code more.
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
The function needs the cpuidle_device which is initially passed to the
caller.
The current code gets the struct device from the struct cpuidle_device,
pass it the cpuidle_add_sysfs function. This function calls
per_cpu(cpuidle_devices, cpu) to get the cpuidle_device.
This patch pass the cpuidle_device instead and simplify the code.
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Pull ACPI & power management update from Len Brown:
"Re-write of the turbostat tool.
lower overhead was necessary for measuring very large system when
they are very idle.
IVB support in intel_idle
It's what I run on my IVB, others should be able to also:-)
ACPICA core update
We have found some bugs due to divergence between Linux and the
upstream ACPICA base. Most of these patches are to reduce that
divergence to reduce the risk of future bugs.
Some cpuidle updates, mostly for non-Intel
More will be coming, as they depend on this part.
Some thermal management changes needed by non-ACPI systems.
Some _OST (OS Status Indication) updates for hot ACPI hot-plug."
* 'release' of git://git.kernel.org/pub/scm/linux/kernel/git/lenb/linux: (51 commits)
Thermal: Documentation update
Thermal: Add Hysteresis attributes
Thermal: Make Thermal trip points writeable
ACPI/AC: prevent OOPS on some boxes due to missing check power_supply_register() return value check
tools/power: turbostat: fix large c1% issue
tools/power: turbostat v2 - re-write for efficiency
ACPICA: Update to version 20120711
ACPICA: AcpiSrc: Fix some translation issues for Linux conversion
ACPICA: Update header files copyrights to 2012
ACPICA: Add new ACPI table load/unload external interfaces
ACPICA: Split file: tbxface.c -> tbxfload.c
ACPICA: Add PCC address space to space ID decode function
ACPICA: Fix some comment fields
ACPICA: Table manager: deploy new firmware error/warning interfaces
ACPICA: Add new interfaces for BIOS(firmware) errors and warnings
ACPICA: Split exception code utilities to a new file, utexcep.c
ACPI: acpi_pad: tune round_robin_time
ACPICA: Update to version 20120620
ACPICA: Add support for implicit notify on multiple devices
ACPICA: Update comments; no functional change
...
* pm-domains:
PM / Domains: Fix build warning for CONFIG_PM_RUNTIME unset
PM / Domains: Replace plain integer with NULL pointer in domain.c file
PM / Domains: Add missing static storage class specifier in domain.c file
PM / Domains: Allow device callbacks to be added at any time
PM / Domains: Add device domain data reference counter
PM / Domains: Add preliminary support for cpuidle, v2
PM / Domains: Do not stop devices after restoring their states
PM / Domains: Use subsystem runtime suspend/resume callbacks by default
On certain bios, resume hangs if cpus are allowed to enter idle states
during suspend [1].
This was fixed in apci idle driver [2].But intel_idle driver does not
have this fix. Thus instead of replicating the fix in both the idle
drivers, or in more platform specific idle drivers if needed, the
more general cpuidle infrastructure could handle this.
A suspend callback in cpuidle_driver could handle this fix. But
a cpuidle_driver provides only basic functionalities like platform idle
state detection capability and mechanisms to support entry and exit
into CPU idle states. All other cpuidle functions are found in the
cpuidle generic infrastructure for good reason that all cpuidle
drivers, irrepective of their platforms will support these functions.
One option therefore would be to register a suspend callback in cpuidle
which handles this fix. This could be called through a PM_SUSPEND_PREPARE
notifier. But this is too generic a notfier for a driver to handle.
Also, ideally the job of cpuidle is not to handle side effects of suspend.
It should expose the interfaces which "handle cpuidle 'during' suspend"
or any other operation, which the subsystems call during that respective
operation.
The fix demands that during suspend, no cpus should be allowed to enter
deep C-states. The interface cpuidle_uninstall_idle_handler() in cpuidle
ensures that. Not just that it also kicks all the cpus which are already
in idle out of their idle states which was being done during cpu hotplug
through a CPU_DYING_FROZEN callbacks.
Now the question arises about when during suspend should
cpuidle_uninstall_idle_handler() be called. Since we are dealing with
drivers it seems best to call this function during dpm_suspend().
Delaying the call till dpm_suspend_noirq() does no harm, as long as it is
before cpu_hotplug_begin() to avoid race conditions with cpu hotpulg
operations. In dpm_suspend_noirq(), it would be wise to place this call
before suspend_device_irqs() to avoid ugly interactions with the same.
Ananlogously, during resume.
References:
[1] https://bugs.launchpad.net/ubuntu/+source/linux/+bug/674075.
[2] http://marc.info/?l=linux-pm&m=133958534231884&w=2
Reported-and-tested-by: Dave Hansen <dave@linux.vnet.ibm.com>
Signed-off-by: Preeti U Murthy <preeti@linux.vnet.ibm.com>
Reviewed-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
On some systems there are CPU cores located in the same power
domains as I/O devices. Then, power can only be removed from the
domain if all I/O devices in it are not in use and the CPU core
is idle. Add preliminary support for that to the generic PM domains
framework.
First, the platform is expected to provide a cpuidle driver with one
extra state designated for use with the generic PM domains code.
This state should be initially disabled and its exit_latency value
should be set to whatever time is needed to bring up the CPU core
itself after restoring power to it, not including the domain's
power on latency. Its .enter() callback should point to a procedure
that will remove power from the domain containing the CPU core at
the end of the CPU power transition.
The remaining characteristics of the extra cpuidle state, referred to
as the "domain" cpuidle state below, (e.g. power usage, target
residency) should be populated in accordance with the properties of
the hardware.
Next, the platform should execute genpd_attach_cpuidle() on the PM
domain containing the CPU core. That will cause the generic PM
domains framework to treat that domain in a special way such that:
* When all devices in the domain have been suspended and it is about
to be turned off, the states of the devices will be saved, but
power will not be removed from the domain. Instead, the "domain"
cpuidle state will be enabled so that power can be removed from
the domain when the CPU core is idle and the state has been chosen
as the target by the cpuidle governor.
* When the first I/O device in the domain is resumed and
__pm_genpd_poweron(() is called for the first time after
power has been removed from the domain, the "domain" cpuidle
state will be disabled to avoid subsequent surprise power removals
via cpuidle.
The effective exit_latency value of the "domain" cpuidle state
depends on the time needed to bring up the CPU core itself after
restoring power to it as well as on the power on latency of the
domain containing the CPU core. Thus the "domain" cpuidle state's
exit_latency has to be recomputed every time the domain's power on
latency is updated, which may happen every time power is restored
to the domain, if the measured power on latency is greater than
the latency stored in the corresponding generic_pm_domain structure.
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Reviewed-by: Kevin Hilman <khilman@ti.com>
Andrew J.Schorr raises a question. When he changes the disable setting on
a single CPU, it affects all the other CPUs. Basically, currently, the
disable field is per-driver instead of per-cpu. All the C states of the
same driver are shared by all CPU in the same machine.
The patch changes the `disable' field to per-cpu, so we could set this
separately for each cpu.
Signed-off-by: ShuoX Liu <shuox.liu@intel.com>
Reported-by: Andrew J.Schorr <aschorr@telemetry-investments.com>
Reviewed-by: Yanmin Zhang <yanmin_zhang@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
On some ARM SMP SoCs (OMAP4460, Tegra 2, and probably more), the
cpus cannot be independently powered down, either due to
sequencing restrictions (on Tegra 2, cpu 0 must be the last to
power down), or due to HW bugs (on OMAP4460, a cpu powering up
will corrupt the gic state unless the other cpu runs a work
around). Each cpu has a power state that it can enter without
coordinating with the other cpu (usually Wait For Interrupt, or
WFI), and one or more "coupled" power states that affect blocks
shared between the cpus (L2 cache, interrupt controller, and
sometimes the whole SoC). Entering a coupled power state must
be tightly controlled on both cpus.
The easiest solution to implementing coupled cpu power states is
to hotplug all but one cpu whenever possible, usually using a
cpufreq governor that looks at cpu load to determine when to
enable the secondary cpus. This causes problems, as hotplug is an
expensive operation, so the number of hotplug transitions must be
minimized, leading to very slow response to loads, often on the
order of seconds.
This file implements an alternative solution, where each cpu will
wait in the WFI state until all cpus are ready to enter a coupled
state, at which point the coupled state function will be called
on all cpus at approximately the same time.
Once all cpus are ready to enter idle, they are woken by an smp
cross call. At this point, there is a chance that one of the
cpus will find work to do, and choose not to enter idle. A
final pass is needed to guarantee that all cpus will call the
power state enter function at the same time. During this pass,
each cpu will increment the ready counter, and continue once the
ready counter matches the number of online coupled cpus. If any
cpu exits idle, the other cpus will decrement their counter and
retry.
To use coupled cpuidle states, a cpuidle driver must:
Set struct cpuidle_device.coupled_cpus to the mask of all
coupled cpus, usually the same as cpu_possible_mask if all cpus
are part of the same cluster. The coupled_cpus mask must be
set in the struct cpuidle_device for each cpu.
Set struct cpuidle_device.safe_state to a state that is not a
coupled state. This is usually WFI.
Set CPUIDLE_FLAG_COUPLED in struct cpuidle_state.flags for each
state that affects multiple cpus.
Provide a struct cpuidle_state.enter function for each state
that affects multiple cpus. This function is guaranteed to be
called on all cpus at approximately the same time. The driver
should ensure that the cpus all abort together if any cpu tries
to abort once the function is called.
update1:
cpuidle: coupled: fix count of online cpus
online_count was never incremented on boot, and was also counting
cpus that were not part of the coupled set. Fix both issues by
introducting a new function that counts online coupled cpus, and
call it from register as well as the hotplug notifier.
update2:
cpuidle: coupled: fix decrementing ready count
cpuidle_coupled_set_not_ready sometimes refuses to decrement the
ready count in order to prevent a race condition. This makes it
unsuitable for use when finished with idle. Add a new function
cpuidle_coupled_set_done that decrements both the ready count and
waiting count, and call it after idle is complete.
Cc: Amit Kucheria <amit.kucheria@linaro.org>
Cc: Arjan van de Ven <arjan@linux.intel.com>
Cc: Trinabh Gupta <g.trinabh@gmail.com>
Cc: Deepthi Dharwar <deepthi@linux.vnet.ibm.com>
Reviewed-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Tested-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Reviewed-by: Kevin Hilman <khilman@ti.com>
Tested-by: Kevin Hilman <khilman@ti.com>
Signed-off-by: Colin Cross <ccross@android.com>
Acked-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Len Brown <len.brown@intel.com>
Fix the error handling in __cpuidle_register_device to include
the missing list_del. Move it to a label, which will simplify
the error handling when coupled states are added.
Reviewed-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Tested-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Reviewed-by: Kevin Hilman <khilman@ti.com>
Tested-by: Kevin Hilman <khilman@ti.com>
Signed-off-by: Colin Cross <ccross@android.com>
Reviewed-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Len Brown <len.brown@intel.com>
Split the code to enter a state and update the stats into a helper
function, cpuidle_enter_state, and export it. This function will
be called by the coupled state code to handle entering the safe
state and the final coupled state.
Reviewed-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Tested-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Reviewed-by: Kevin Hilman <khilman@ti.com>
Tested-by: Kevin Hilman <khilman@ti.com>
Signed-off-by: Colin Cross <ccross@android.com>
Reviewed-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Len Brown <len.brown@intel.com>
The existing check for dev == NULL in __cpuidle_register_device() is
rendered useless because dev is dereferenced before the check itself.
Moreover, correctly speaking, it is the job of the callers of this
function, i.e., cpuidle_register_device() & cpuidle_enable_device() (which
also happen to be exported functions) to ensure that
__cpuidle_register_device() is called with a non-NULL dev.
So add the necessary dev == NULL checks in the two callers and remove the
(useless) check from __cpuidle_register_device().
Signed-off-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
Acked-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Len Brown <len.brown@intel.com>
commit 9a6558371b
Author: Arjan van de Ven <arjan@linux.intel.com>
Date: Sun Nov 9 12:45:10 2008 -0800
regression: disable timer peek-ahead for 2.6.28
It's showing up as regressions; disabling it very likely just papers
over an underlying issue, but time is running out for 2.6.28, lets get
back to this for 2.6.29
Many years has passed since 2008, so it seems ok to remove whole `#if 0' block.
Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Cc: Kevin Hilman <khilman@ti.com>
Cc: Trinabh Gupta <g.trinabh@gmail.com>
Cc: Deepthi Dharwar <deepthi@linux.vnet.ibm.com>
Cc: Arjan van de Ven <arjan@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Len Brown <len.brown@intel.com>
kick_all_cpus_sync() is the core implementation of cpu_idle_wait()
which is copied all over the arch code.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120507175652.119842173@linutronix.de
Daniel Lezcano