| Message ID | 5451A94F.1090200@linux.intel.com (mailing list archive) |
|---|---|
| State | Changes Requested, archived |
| Headers | show |
On Thursday, October 30, 2014 10:58:23 AM Li, Aubrey wrote: > The patch is based on v3.17, merged with Rafael's pm+acpi-3.18-rc1 tag from > linux-pm.git tree. > > The patch is based on the patch PeterZ initially wrote. > --- > Freeze is a general power saving state that processes are frozen, devices > are suspended and CPUs are in idle state. However, when the system enters > freeze state, there are a few timers keep ticking and hence consumes more > power unnecessarily. The observed timer events in freeze state are: > - tick_sched_timer > - watchdog lockup detector > - realtime scheduler period timer > > The system power consumption in freeze state will be reduced significantly > if we quiesce these timers. > > On Baytrail-T(ASUS_T100) platform, when the system is freezed to low power > idle state(S0ix), quiescing these timers saves 29.8% power(94.48mw -> 66.32mw). > > The patch is also tested on: > - Sandybrdige-EP system, both RTC alarm and power button are able to wake > the system up from freeze state. > - HP laptop EliteBook 8460p, both RTC alarm and power button are able to > wake the system up from freeze state. > > Signed-off-by: Aubrey Li <aubrey.li@linux.intel.com> > Signed-off-by: Peter Zijlstra <peterz@infradead.org> > Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com> > Cc: Len Brown <len.brown@intel.com> > Cc: Alan Cox <alan@linux.intel.com> Peter, Thomas, any comments here? > --- > arch/x86/kernel/apic/apic.c | 8 ++ > drivers/cpuidle/cpuidle.c | 12 +++ > kernel/power/suspend.c | 185 +++++++++++++++++++++++++++++++++++-- > kernel/time/timekeeping.c | 4 +- > kernel/time/timekeeping_internal.h | 3 + > 5 files changed, 204 insertions(+), 8 deletions(-) > > diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c > index 6776027..f2bb645 100644 > --- a/arch/x86/kernel/apic/apic.c > +++ b/arch/x86/kernel/apic/apic.c > @@ -917,6 +917,14 @@ static void local_apic_timer_interrupt(void) > */ > inc_irq_stat(apic_timer_irqs); > > + /* > + * if timekeeping is suspended, the clock event device will be > + * suspended as well, so we are not supposed to invoke the event > + * handler of clock event device. > + */ > + if (unlikely(timekeeping_suspended)) > + return; > + > evt->event_handler(evt); > } > > diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c > index ee9df5e..8f84f40 100644 > --- a/drivers/cpuidle/cpuidle.c > +++ b/drivers/cpuidle/cpuidle.c > @@ -119,6 +119,18 @@ int cpuidle_enter_state(struct cpuidle_device *dev, struct cpuidle_driver *drv, > ktime_t time_start, time_end; > s64 diff; > > + /* > + * under the scenario of use deepest idle state, the timekeeping > + * could be suspended as well as the clock source device, so we > + * bypass the idle counter update for this case > + */ > + if (unlikely(use_deepest_state)) { > + entered_state = target_state->enter(dev, drv, index); > + if (!cpuidle_state_is_coupled(dev, drv, entered_state)) > + local_irq_enable(); > + return entered_state; > + } > + > trace_cpu_idle_rcuidle(index, dev->cpu); > time_start = ktime_get(); > > diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c > index 4ca9a33..660fd15 100644 > --- a/kernel/power/suspend.c > +++ b/kernel/power/suspend.c > @@ -28,16 +28,20 @@ > #include <linux/ftrace.h> > #include <trace/events/power.h> > #include <linux/compiler.h> > +#include <linux/stop_machine.h> > +#include <linux/clockchips.h> > +#include <linux/hrtimer.h> > > #include "power.h" > +#include "../time/tick-internal.h" > +#include "../time/timekeeping_internal.h" > > const char *pm_labels[] = { "mem", "standby", "freeze", NULL }; > const char *pm_states[PM_SUSPEND_MAX]; > > static const struct platform_suspend_ops *suspend_ops; > static const struct platform_freeze_ops *freeze_ops; > -static DECLARE_WAIT_QUEUE_HEAD(suspend_freeze_wait_head); > -static bool suspend_freeze_wake; > +static int suspend_freeze_wake; > > void freeze_set_ops(const struct platform_freeze_ops *ops) > { > @@ -48,22 +52,191 @@ void freeze_set_ops(const struct platform_freeze_ops *ops) > > static void freeze_begin(void) > { > - suspend_freeze_wake = false; > + suspend_freeze_wake = -1; > +} > + > +enum freezer_state { > + FREEZER_NONE, > + FREEZER_PICK_TK, > + FREEZER_SUSPEND_CLKEVT, > + FREEZER_SUSPEND_TK, > + FREEZER_IDLE, > + FREEZER_RESUME_TK, > + FREEZER_RESUME_CLKEVT, > + FREEZER_EXIT, > +}; > + > +struct freezer_data { > + int thread_num; > + atomic_t thread_ack; > + enum freezer_state state; > +}; > + > +static void set_state(struct freezer_data *fd, enum freezer_state state) > +{ > + /* set ack counter */ > + atomic_set(&fd->thread_ack, fd->thread_num); > + /* guarantee the write ordering between ack counter and state */ > + smp_wmb(); > + fd->state = state; > +} > + > +static void ack_state(struct freezer_data *fd) > +{ > + if (atomic_dec_and_test(&fd->thread_ack)) > + set_state(fd, fd->state + 1); > +} > + > +static void freezer_pick_tk(int cpu) > +{ > + if (tick_do_timer_cpu == TICK_DO_TIMER_NONE) { > + static DEFINE_SPINLOCK(lock); > + > + spin_lock(&lock); > + if (tick_do_timer_cpu == TICK_DO_TIMER_NONE) > + tick_do_timer_cpu = cpu; > + spin_unlock(&lock); > + } > +} > + > +static void freezer_suspend_clkevt(int cpu) > +{ > + if (tick_do_timer_cpu == cpu) > + return; > + > + clockevents_notify(CLOCK_EVT_NOTIFY_SUSPEND, NULL); > +} > + > +static void freezer_suspend_tk(int cpu) > +{ > + if (tick_do_timer_cpu != cpu) > + return; > + > + timekeeping_suspend(); > + > +} > + > +static void freezer_idle(int cpu) > +{ > + struct cpuidle_device *dev = __this_cpu_read(cpuidle_devices); > + struct cpuidle_driver *drv = cpuidle_get_cpu_driver(dev); > + > + stop_critical_timings(); > + > + while (suspend_freeze_wake == -1) { > + int next_state; > + > + /* > + * interrupt must be disabled before cpu enters idle > + */ > + local_irq_disable(); > + > + next_state = cpuidle_select(drv, dev); > + if (next_state < 0) { > + arch_cpu_idle(); > + continue; > + } > + /* > + * cpuidle_enter will return with interrupt enabled > + */ > + cpuidle_enter(drv, dev, next_state); > + } > + > + if (suspend_freeze_wake == cpu) > + kick_all_cpus_sync(); > + > + /* > + * We disable interrupt here for the rest of resume operations > + */ > + local_irq_disable(); > + start_critical_timings(); > +} > + > +static void freezer_resume_tk(int cpu) > +{ > + if (tick_do_timer_cpu != cpu) > + return; > + > + timekeeping_resume(); > +} > + > +static void freezer_resume_clkevt(int cpu) > +{ > + if (tick_do_timer_cpu == cpu) { > + /* > + * Turn on the interrupt on the tick timer CPU as freezer > + * tasks are finished. > + */ > + local_irq_enable(); > + return; > + } > + > + touch_softlockup_watchdog(); > + clockevents_notify(CLOCK_EVT_NOTIFY_RESUME, NULL); > + hrtimers_resume(); > + /* > + * Turn on the interrupt on the non-tick-timer CPUs as freezer > + * tasks are finished > + */ > + local_irq_enable(); > +} > + > +typedef void (*freezer_fn)(int); > + > +static freezer_fn freezer_func[FREEZER_EXIT] = { > + NULL, > + freezer_pick_tk, > + freezer_suspend_clkevt, > + freezer_suspend_tk, > + freezer_idle, > + freezer_resume_tk, > + freezer_resume_clkevt, > +}; > + > +static int freezer_stopper_fn(void *arg) > +{ > + struct freezer_data *fd = arg; > + enum freezer_state state = FREEZER_NONE; > + int cpu = smp_processor_id(); > + > + do { > + cpu_relax(); > + if (fd->state != state) { > + state = fd->state; > + if (freezer_func[state]) > + (*freezer_func[state])(cpu); > + ack_state(fd); > + } > + } while (fd->state != FREEZER_EXIT); > + > + return 0; > } > > static void freeze_enter(void) > { > + struct freezer_data fd; > + > cpuidle_use_deepest_state(true); > cpuidle_resume(); > - wait_event(suspend_freeze_wait_head, suspend_freeze_wake); > + > + get_online_cpus(); > + > + fd.thread_num = num_online_cpus(); > + set_state(&fd, FREEZER_PICK_TK); > + > + __stop_machine(freezer_stopper_fn, &fd, cpu_online_mask); > + > + put_online_cpus(); > + > cpuidle_pause(); > cpuidle_use_deepest_state(false); > } > > void freeze_wake(void) > { > - suspend_freeze_wake = true; > - wake_up(&suspend_freeze_wait_head); > + if (suspend_freeze_wake != -1) > + return; > + suspend_freeze_wake = smp_processor_id(); > } > EXPORT_SYMBOL_GPL(freeze_wake); > > diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c > index ec1791f..23d8feb 100644 > --- a/kernel/time/timekeeping.c > +++ b/kernel/time/timekeeping.c > @@ -1114,7 +1114,7 @@ void timekeeping_inject_sleeptime(struct timespec *delta) > * xtime/wall_to_monotonic/jiffies/etc are > * still managed by arch specific suspend/resume code. > */ > -static void timekeeping_resume(void) > +void timekeeping_resume(void) > { > struct timekeeper *tk = &tk_core.timekeeper; > struct clocksource *clock = tk->tkr.clock; > @@ -1195,7 +1195,7 @@ static void timekeeping_resume(void) > hrtimers_resume(); > } > > -static int timekeeping_suspend(void) > +int timekeeping_suspend(void) > { > struct timekeeper *tk = &tk_core.timekeeper; > unsigned long flags; > diff --git a/kernel/time/timekeeping_internal.h b/kernel/time/timekeeping_internal.h > index 4ea005a..ed7a574 100644 > --- a/kernel/time/timekeeping_internal.h > +++ b/kernel/time/timekeeping_internal.h > @@ -26,4 +26,7 @@ static inline cycle_t clocksource_delta(cycle_t now, cycle_t last, cycle_t mask) > } > #endif > > +extern int timekeeping_suspend(void); > +extern void timekeeping_resume(void); > + > #endif /* _TIMEKEEPING_INTERNAL_H */ >
On Sat, Nov 08, 2014 at 03:05:56AM +0100, Rafael J. Wysocki wrote:
> Peter, Thomas, any comments here?
I'm fine with this; but Thomas needs to ack, lets give him a few more
days to reply with this reminder.
--
To unsubscribe from this list: send the line "unsubscribe linux-pm" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
On Thu, 30 Oct 2014, Li, Aubrey wrote: > Freeze is a general power saving state that processes are frozen, devices > are suspended and CPUs are in idle state. However, when the system enters > freeze state, there are a few timers keep ticking and hence consumes more > power unnecessarily. The observed timer events in freeze state are: > - tick_sched_timer > - watchdog lockup detector > - realtime scheduler period timer > > The system power consumption in freeze state will be reduced significantly > if we quiesce these timers. So the obvious question is why dont we quiesce these timers by telling the subsystems which manage these timers to shut them down? I really want a proper answer for this in the first place, but let me look at the proposed "solution" as well. > diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c > index 6776027..f2bb645 100644 > --- a/arch/x86/kernel/apic/apic.c > +++ b/arch/x86/kernel/apic/apic.c > @@ -917,6 +917,14 @@ static void local_apic_timer_interrupt(void) > */ > inc_irq_stat(apic_timer_irqs); > > + /* > + * if timekeeping is suspended, the clock event device will be > + * suspended as well, so we are not supposed to invoke the event > + * handler of clock event device. > + */ > + if (unlikely(timekeeping_suspended)) > + return; Why do you need that if you already suspended the clock event device? The above comment does not explain that at all. So if there is a proper reason to do so, we rather do the following in tick_suspend(): td->evtdev.real_handler = td->evtdev.event_handler; td->evtdev.event_handler = clockevents_handle_noop; and restore that on resume instead of sprinkling if (tk_suspended) checks all over the place. x86/apic is probably not the only one which wants that treatment. But before we do that we want a proper explanation why the interrupt fires at all. The lack of explanation cleary documents that this is a 'hacked it into submission' approach. > diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c > index 4ca9a33..660fd15 100644 > --- a/kernel/power/suspend.c > +++ b/kernel/power/suspend.c > @@ -28,16 +28,20 @@ > #include <linux/ftrace.h> > #include <trace/events/power.h> > #include <linux/compiler.h> > +#include <linux/stop_machine.h> > +#include <linux/clockchips.h> > +#include <linux/hrtimer.h> > > #include "power.h" > +#include "../time/tick-internal.h" > +#include "../time/timekeeping_internal.h" Eew. > +static void freezer_pick_tk(int cpu) > +{ > + if (tick_do_timer_cpu == TICK_DO_TIMER_NONE) { > + static DEFINE_SPINLOCK(lock); > + > + spin_lock(&lock); > + if (tick_do_timer_cpu == TICK_DO_TIMER_NONE) > + tick_do_timer_cpu = cpu; > + spin_unlock(&lock); > + } > +} > +static void freezer_suspend_clkevt(int cpu) > +{ > + if (tick_do_timer_cpu == cpu) > + return; > + > + clockevents_notify(CLOCK_EVT_NOTIFY_SUSPEND, NULL); > +} > + > +static void freezer_suspend_tk(int cpu) > +{ > + if (tick_do_timer_cpu != cpu) > + return; > + > + timekeeping_suspend(); > + > +} So you export the world and some more from timekeeping and the tick code and fiddle with it randomly just to do: 1) Suspend clock event devices 2) Suspend timekeeping 3) Resume timekeeping 4) Resume clock event devices And for that you kick the frozen cpus out of idle into the stomp_machine task and let them enter deep idle from there. stomp_machine() is in 99% of all use cases a clear indicator for a complete design failure. It's not that hard to solve that problem, w/o stomp_machine and w/o all the tick_do_timer_cpu mess. 1) Run the freeze code until freeze_enter() 2) Prevent CPU hotplug and switch state. That tells the cpu idle code to enter the deepest idle state and also tells the clock events code about the desire to freeze everything. clock_events_set_freeze_state(true); And let that be: clock_events_set_freeze_state(bool on) { raw_spin_lock_irq(&clockevents_lock); if (on) tobefrozen_cpus = num_online_cpus(); idle_freeze = on; raw_spin_unlock_irq(&clockevents_lock); } So the generic idle task needs a check like this: if (idle_should_freeze()) frozen_idle(); with the implementation: bool idle_should_freeze() { return clock_events_get_freeze_state(); } which resolves to: bool clock_events_get_freeze_state() { /* * Lockfree access because it does not matter. * * See below at CLOCK_EVT_NOTIFY_FREEZE */ return idle_freeze; } 4) Kick all cpus out of idle, so they enter the deep idle state via frozen_idle() frozen_idle() { if (clock_events_notify(CLOCK_EVT_NOTIFY_FREEZE)) return; while (idle_should_freeze()) magic_frozen_idle(); clock_events_notify(CLOCK_EVT_NOTIFY_UNFREEZE); } Let clock_events_notify() have these new cases: CLOCK_EVT_NOTIFY_FREEZE: ret = tick_freeze(); break; CLOCK_EVT_NOTIFY_UNFREEZE: tick_unfreeze(); break; and tick_freeze() { /* * This is serialized against a concurrent wakeup * via clockevents_lock! */ if (!idle_freeze) return -EBUSY; if (--tobefrozen_cpus) { tick_suspend(); } else /* * Needs to be a seperate interface due to * clockevents_lock being held in clock_events_notify() */ timekeeping_freeze(); } } and tick_unfreeze() { if (!timekeeping_frozen) tick_resume(); else timekeeping_unfreeze(); } and the wakeup notification wants to have a proper interface as well: wakeup_the_whole_thing() { do_whatever_unfreeze_needs(); clock_events_set_freeze_state(false); } 5) Reenable cpu hotplug when the freezer task returns. No stomp_machine, no tick_do_timer_cpu() abuse. All nicely serialized via clockevents_lock. All abortable at any given point in time and not dependend on running through another state machine nested into the stomp_machine() state machine. Thoughts? Thanks, tglx -- To unsubscribe from this list: send the line "unsubscribe linux-pm" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Wed, Nov 12, 2014 at 10:09:47PM +0100, Thomas Gleixner wrote: > On Thu, 30 Oct 2014, Li, Aubrey wrote: > > > Freeze is a general power saving state that processes are frozen, devices > > are suspended and CPUs are in idle state. However, when the system enters > > freeze state, there are a few timers keep ticking and hence consumes more > > power unnecessarily. The observed timer events in freeze state are: > > - tick_sched_timer > > - watchdog lockup detector > > - realtime scheduler period timer > > > > The system power consumption in freeze state will be reduced significantly > > if we quiesce these timers. > > So the obvious question is why dont we quiesce these timers by telling > the subsystems which manage these timers to shut them down? > > I really want a proper answer for this in the first place, but let me > look at the proposed "solution" as well. Two arguments here: 1) the current suspend modes don't care, so if this suspend mode starts to care, its likely to 'break' in the future simply because people never cared about timers. 2) there could be userland timers, userland is frozen but they'll still have their timers set and those can and will fire. But sure, we can add suspend notifiers to stuff to shut down timers; I should have a patch for at least one of the offenders somewhere. But I really think that we should not be looking at the individual timers for this, none of the other suspend modes care about active timers. > But before we do that we want a proper explanation why the interrupt > fires at all. The lack of explanation cleary documents that this is a > 'hacked it into submission' approach. From what I remember its the waking interrupt that ends up in the timekeeping code, Li should have a backtrace somwhere. > > +#include "../time/tick-internal.h" > > +#include "../time/timekeeping_internal.h" > > Eew. I knew you'd love that :-) > So you export the world and some more from timekeeping and the tick > code and fiddle with it randomly just to do: > > 1) Suspend clock event devices > 2) Suspend timekeeping > 3) Resume timekeeping > 4) Resume clock event devices Sure, we can add some exports and clean that up, but.. > stomp_machine() is in 99% of all use cases a clear indicator for a > complete design failure. > So the generic idle task needs a check like this: > > if (idle_should_freeze()) > frozen_idle(); So that is adding extra code to fairly common/hot paths just for this one extra special case. I tried to avoid doing that. But I suppose we can try and merge that with the offline case and guard both special cases with a single variable or so. -- To unsubscribe from this list: send the line "unsubscribe linux-pm" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On 2014/11/13 9:37, Peter Zijlstra wrote: > On Wed, Nov 12, 2014 at 10:09:47PM +0100, Thomas Gleixner wrote: >> On Thu, 30 Oct 2014, Li, Aubrey wrote: >> >>> Freeze is a general power saving state that processes are frozen, devices >>> are suspended and CPUs are in idle state. However, when the system enters >>> freeze state, there are a few timers keep ticking and hence consumes more >>> power unnecessarily. The observed timer events in freeze state are: >>> - tick_sched_timer >>> - watchdog lockup detector >>> - realtime scheduler period timer >>> >>> The system power consumption in freeze state will be reduced significantly >>> if we quiesce these timers. >> >> So the obvious question is why dont we quiesce these timers by telling >> the subsystems which manage these timers to shut them down? >> >> I really want a proper answer for this in the first place, but let me >> look at the proposed "solution" as well. > > Two arguments here: > > 1) the current suspend modes don't care, so if this suspend mode starts > to care, its likely to 'break' in the future simply because people > never cared about timers. > > 2) there could be userland timers, userland is frozen but they'll still > have their timers set and those can and will fire. > > But sure, we can add suspend notifiers to stuff to shut down timers; I > should have a patch for at least one of the offenders somewhere. But I > really think that we should not be looking at the individual timers for > this, none of the other suspend modes care about active timers. > >> But before we do that we want a proper explanation why the interrupt >> fires at all. The lack of explanation cleary documents that this is a >> 'hacked it into submission' approach. > >>From what I remember its the waking interrupt that ends up in the > timekeeping code, Li should have a backtrace somwhere. There are two race conditions: The first one occurs after the interrupt is disabled and before we suspend lapic. In this time slot, if apic timer interrupt occurs, the interrupt is pending there because the interrupt is disabled. Then we suspend timekeeping, and then we enter idle and exit idle with interrupt re-enabled, the timer interrupt is handled with timekeeping is suspended. The other occurs after timekeeping_suspended = 1 and before we suspend lapic. In this time slot, if apic timer interrupt occurs, we invoke the timer interrupt while timekeeping is suspended as above. Thanks, -Aubrey > >>> +#include "../time/tick-internal.h" >>> +#include "../time/timekeeping_internal.h" >> >> Eew. > > I knew you'd love that :-) > >> So you export the world and some more from timekeeping and the tick >> code and fiddle with it randomly just to do: >> >> 1) Suspend clock event devices >> 2) Suspend timekeeping >> 3) Resume timekeeping >> 4) Resume clock event devices > > Sure, we can add some exports and clean that up, but.. > >> stomp_machine() is in 99% of all use cases a clear indicator for a >> complete design failure. > >> So the generic idle task needs a check like this: >> >> if (idle_should_freeze()) >> frozen_idle(); > > So that is adding extra code to fairly common/hot paths just for this > one extra special case. I tried to avoid doing that. > > But I suppose we can try and merge that with the offline case and guard > both special cases with a single variable or so. > > -- To unsubscribe from this list: send the line "unsubscribe linux-pm" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Thu, 13 Nov 2014, Peter Zijlstra wrote: > On Wed, Nov 12, 2014 at 10:09:47PM +0100, Thomas Gleixner wrote: > But sure, we can add suspend notifiers to stuff to shut down timers; I > should have a patch for at least one of the offenders somewhere. But I > really think that we should not be looking at the individual timers for > this, none of the other suspend modes care about active timers. Fair enough. > > But before we do that we want a proper explanation why the interrupt > > fires at all. The lack of explanation cleary documents that this is a > > 'hacked it into submission' approach. > > >From what I remember its the waking interrupt that ends up in the > timekeeping code, Li should have a backtrace somwhere. I can imagine what happens :) > > stomp_machine() is in 99% of all use cases a clear indicator for a > > complete design failure. > > > So the generic idle task needs a check like this: > > > > if (idle_should_freeze()) > > frozen_idle(); > > So that is adding extra code to fairly common/hot paths just for this > one extra special case. I tried to avoid doing that. idle enter is not that much of a hot path, really. Thanks, tglx -- To unsubscribe from this list: send the line "unsubscribe linux-pm" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Thu, 13 Nov 2014, Li, Aubrey wrote: > On 2014/11/13 9:37, Peter Zijlstra wrote: > > On Wed, Nov 12, 2014 at 10:09:47PM +0100, Thomas Gleixner wrote: > >> On Thu, 30 Oct 2014, Li, Aubrey wrote: > >> > >>> Freeze is a general power saving state that processes are frozen, devices > >>> are suspended and CPUs are in idle state. However, when the system enters > >>> freeze state, there are a few timers keep ticking and hence consumes more > >>> power unnecessarily. The observed timer events in freeze state are: > >>> - tick_sched_timer > >>> - watchdog lockup detector > >>> - realtime scheduler period timer > >>> > >>> The system power consumption in freeze state will be reduced significantly > >>> if we quiesce these timers. > >> > >> So the obvious question is why dont we quiesce these timers by telling > >> the subsystems which manage these timers to shut them down? > >> > >> I really want a proper answer for this in the first place, but let me > >> look at the proposed "solution" as well. > > > > Two arguments here: > > > > 1) the current suspend modes don't care, so if this suspend mode starts > > to care, its likely to 'break' in the future simply because people > > never cared about timers. > > > > 2) there could be userland timers, userland is frozen but they'll still > > have their timers set and those can and will fire. > > > > But sure, we can add suspend notifiers to stuff to shut down timers; I > > should have a patch for at least one of the offenders somewhere. But I > > really think that we should not be looking at the individual timers for > > this, none of the other suspend modes care about active timers. > > > >> But before we do that we want a proper explanation why the interrupt > >> fires at all. The lack of explanation cleary documents that this is a > >> 'hacked it into submission' approach. > > > >>From what I remember its the waking interrupt that ends up in the > > timekeeping code, Li should have a backtrace somwhere. > > There are two race conditions: > > The first one occurs after the interrupt is disabled and before we > suspend lapic. In this time slot, if apic timer interrupt occurs, the > interrupt is pending there because the interrupt is disabled. Then we > suspend timekeeping, and then we enter idle and exit idle with interrupt > re-enabled, the timer interrupt is handled with timekeeping is > suspended. > > The other occurs after timekeeping_suspended = 1 and before we suspend > lapic. In this time slot, if apic timer interrupt occurs, we invoke the > timer interrupt while timekeeping is suspended as above. And that race exists for every implementation and is not at all apic timer specific. So we fix it at the core and not at some random place in the architecture code. Thanks, tglx -- To unsubscribe from this list: send the line "unsubscribe linux-pm" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On 2014/11/13 17:10, Thomas Gleixner wrote: > On Thu, 13 Nov 2014, Peter Zijlstra wrote: >> On Wed, Nov 12, 2014 at 10:09:47PM +0100, Thomas Gleixner wrote: >> But sure, we can add suspend notifiers to stuff to shut down timers; I >> should have a patch for at least one of the offenders somewhere. But I >> really think that we should not be looking at the individual timers for >> this, none of the other suspend modes care about active timers. > > Fair enough. > If you are okay with the current method to suspend timekeeping entirely, then we can go further to fix the rest concerns. Thanks, -Aubrey -- To unsubscribe from this list: send the line "unsubscribe linux-pm" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On 2014/11/13 17:19, Thomas Gleixner wrote: > On Thu, 13 Nov 2014, Li, Aubrey wrote: >> On 2014/11/13 9:37, Peter Zijlstra wrote: >>> On Wed, Nov 12, 2014 at 10:09:47PM +0100, Thomas Gleixner wrote: >>>> On Thu, 30 Oct 2014, Li, Aubrey wrote: >>>> >>>>> Freeze is a general power saving state that processes are frozen, devices >>>>> are suspended and CPUs are in idle state. However, when the system enters >>>>> freeze state, there are a few timers keep ticking and hence consumes more >>>>> power unnecessarily. The observed timer events in freeze state are: >>>>> - tick_sched_timer >>>>> - watchdog lockup detector >>>>> - realtime scheduler period timer >>>>> >>>>> The system power consumption in freeze state will be reduced significantly >>>>> if we quiesce these timers. >>>> >>>> So the obvious question is why dont we quiesce these timers by telling >>>> the subsystems which manage these timers to shut them down? >>>> >>>> I really want a proper answer for this in the first place, but let me >>>> look at the proposed "solution" as well. >>> >>> Two arguments here: >>> >>> 1) the current suspend modes don't care, so if this suspend mode starts >>> to care, its likely to 'break' in the future simply because people >>> never cared about timers. >>> >>> 2) there could be userland timers, userland is frozen but they'll still >>> have their timers set and those can and will fire. >>> >>> But sure, we can add suspend notifiers to stuff to shut down timers; I >>> should have a patch for at least one of the offenders somewhere. But I >>> really think that we should not be looking at the individual timers for >>> this, none of the other suspend modes care about active timers. >>> >>>> But before we do that we want a proper explanation why the interrupt >>>> fires at all. The lack of explanation cleary documents that this is a >>>> 'hacked it into submission' approach. >>> >>> >From what I remember its the waking interrupt that ends up in the >>> timekeeping code, Li should have a backtrace somwhere. >> >> There are two race conditions: >> >> The first one occurs after the interrupt is disabled and before we >> suspend lapic. In this time slot, if apic timer interrupt occurs, the >> interrupt is pending there because the interrupt is disabled. Then we >> suspend timekeeping, and then we enter idle and exit idle with interrupt >> re-enabled, the timer interrupt is handled with timekeeping is >> suspended. >> >> The other occurs after timekeeping_suspended = 1 and before we suspend >> lapic. In this time slot, if apic timer interrupt occurs, we invoke the >> timer interrupt while timekeeping is suspended as above. > > And that race exists for every implementation and is not at all apic > timer specific. So we fix it at the core and not at some random place > in the architecture code. > You're right, will refine this in the next patch version. Thanks, -Aubrey -- To unsubscribe from this list: send the line "unsubscribe linux-pm" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Thu, 13 Nov 2014, Li, Aubrey wrote: > On 2014/11/13 17:10, Thomas Gleixner wrote: > > On Thu, 13 Nov 2014, Peter Zijlstra wrote: > >> On Wed, Nov 12, 2014 at 10:09:47PM +0100, Thomas Gleixner wrote: > >> But sure, we can add suspend notifiers to stuff to shut down timers; I > >> should have a patch for at least one of the offenders somewhere. But I > >> really think that we should not be looking at the individual timers for > >> this, none of the other suspend modes care about active timers. > > > > Fair enough. > > > > If you are okay with the current method to suspend timekeeping entirely, > then we can go further to fix the rest concerns. I'm fine with that when it's done proper :) Thanks, tglx -- To unsubscribe from this list: send the line "unsubscribe linux-pm" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On 2014/11/13 21:06, Thomas Gleixner wrote: > On Thu, 13 Nov 2014, Li, Aubrey wrote: > >> On 2014/11/13 17:10, Thomas Gleixner wrote: >>> On Thu, 13 Nov 2014, Peter Zijlstra wrote: >>>> On Wed, Nov 12, 2014 at 10:09:47PM +0100, Thomas Gleixner wrote: >>>> But sure, we can add suspend notifiers to stuff to shut down timers; I >>>> should have a patch for at least one of the offenders somewhere. But I >>>> really think that we should not be looking at the individual timers for >>>> this, none of the other suspend modes care about active timers. >>> >>> Fair enough. >>> >> >> If you are okay with the current method to suspend timekeeping entirely, >> then we can go further to fix the rest concerns. > > I'm fine with that when it's done proper :) > Sure, thanks for the suggestion, let me try my best to make you happy, ;) > Thanks, > > tglx > > -- To unsubscribe from this list: send the line "unsubscribe linux-pm" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c index 6776027..f2bb645 100644 --- a/arch/x86/kernel/apic/apic.c +++ b/arch/x86/kernel/apic/apic.c @@ -917,6 +917,14 @@ static void local_apic_timer_interrupt(void) */ inc_irq_stat(apic_timer_irqs); + /* + * if timekeeping is suspended, the clock event device will be + * suspended as well, so we are not supposed to invoke the event + * handler of clock event device. + */ + if (unlikely(timekeeping_suspended)) + return; + evt->event_handler(evt); } diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c index ee9df5e..8f84f40 100644 --- a/drivers/cpuidle/cpuidle.c +++ b/drivers/cpuidle/cpuidle.c @@ -119,6 +119,18 @@ int cpuidle_enter_state(struct cpuidle_device *dev, struct cpuidle_driver *drv, ktime_t time_start, time_end; s64 diff; + /* + * under the scenario of use deepest idle state, the timekeeping + * could be suspended as well as the clock source device, so we + * bypass the idle counter update for this case + */ + if (unlikely(use_deepest_state)) { + entered_state = target_state->enter(dev, drv, index); + if (!cpuidle_state_is_coupled(dev, drv, entered_state)) + local_irq_enable(); + return entered_state; + } + trace_cpu_idle_rcuidle(index, dev->cpu); time_start = ktime_get(); diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c index 4ca9a33..660fd15 100644 --- a/kernel/power/suspend.c +++ b/kernel/power/suspend.c @@ -28,16 +28,20 @@ #include <linux/ftrace.h> #include <trace/events/power.h> #include <linux/compiler.h> +#include <linux/stop_machine.h> +#include <linux/clockchips.h> +#include <linux/hrtimer.h> #include "power.h" +#include "../time/tick-internal.h" +#include "../time/timekeeping_internal.h" const char *pm_labels[] = { "mem", "standby", "freeze", NULL }; const char *pm_states[PM_SUSPEND_MAX]; static const struct platform_suspend_ops *suspend_ops; static const struct platform_freeze_ops *freeze_ops; -static DECLARE_WAIT_QUEUE_HEAD(suspend_freeze_wait_head); -static bool suspend_freeze_wake; +static int suspend_freeze_wake; void freeze_set_ops(const struct platform_freeze_ops *ops) { @@ -48,22 +52,191 @@ void freeze_set_ops(const struct platform_freeze_ops *ops) static void freeze_begin(void) { - suspend_freeze_wake = false; + suspend_freeze_wake = -1; +} + +enum freezer_state { + FREEZER_NONE, + FREEZER_PICK_TK, + FREEZER_SUSPEND_CLKEVT, + FREEZER_SUSPEND_TK, + FREEZER_IDLE, + FREEZER_RESUME_TK, + FREEZER_RESUME_CLKEVT, + FREEZER_EXIT, +}; + +struct freezer_data { + int thread_num; + atomic_t thread_ack; + enum freezer_state state; +}; + +static void set_state(struct freezer_data *fd, enum freezer_state state) +{ + /* set ack counter */ + atomic_set(&fd->thread_ack, fd->thread_num); + /* guarantee the write ordering between ack counter and state */ + smp_wmb(); + fd->state = state; +} + +static void ack_state(struct freezer_data *fd) +{ + if (atomic_dec_and_test(&fd->thread_ack)) + set_state(fd, fd->state + 1); +} + +static void freezer_pick_tk(int cpu) +{ + if (tick_do_timer_cpu == TICK_DO_TIMER_NONE) { + static DEFINE_SPINLOCK(lock); + + spin_lock(&lock); + if (tick_do_timer_cpu == TICK_DO_TIMER_NONE) + tick_do_timer_cpu = cpu; + spin_unlock(&lock); + } +} + +static void freezer_suspend_clkevt(int cpu) +{ + if (tick_do_timer_cpu == cpu) + return; + + clockevents_notify(CLOCK_EVT_NOTIFY_SUSPEND, NULL); +} + +static void freezer_suspend_tk(int cpu) +{ + if (tick_do_timer_cpu != cpu) + return; + + timekeeping_suspend(); + +} + +static void freezer_idle(int cpu) +{ + struct cpuidle_device *dev = __this_cpu_read(cpuidle_devices); + struct cpuidle_driver *drv = cpuidle_get_cpu_driver(dev); + + stop_critical_timings(); + + while (suspend_freeze_wake == -1) { + int next_state; + + /* + * interrupt must be disabled before cpu enters idle + */ + local_irq_disable(); + + next_state = cpuidle_select(drv, dev); + if (next_state < 0) { + arch_cpu_idle(); + continue; + } + /* + * cpuidle_enter will return with interrupt enabled + */ + cpuidle_enter(drv, dev, next_state); + } + + if (suspend_freeze_wake == cpu) + kick_all_cpus_sync(); + + /* + * We disable interrupt here for the rest of resume operations + */ + local_irq_disable(); + start_critical_timings(); +} + +static void freezer_resume_tk(int cpu) +{ + if (tick_do_timer_cpu != cpu) + return; + + timekeeping_resume(); +} + +static void freezer_resume_clkevt(int cpu) +{ + if (tick_do_timer_cpu == cpu) { + /* + * Turn on the interrupt on the tick timer CPU as freezer + * tasks are finished. + */ + local_irq_enable(); + return; + } + + touch_softlockup_watchdog(); + clockevents_notify(CLOCK_EVT_NOTIFY_RESUME, NULL); + hrtimers_resume(); + /* + * Turn on the interrupt on the non-tick-timer CPUs as freezer + * tasks are finished + */ + local_irq_enable(); +} + +typedef void (*freezer_fn)(int); + +static freezer_fn freezer_func[FREEZER_EXIT] = { + NULL, + freezer_pick_tk, + freezer_suspend_clkevt, + freezer_suspend_tk, + freezer_idle, + freezer_resume_tk, + freezer_resume_clkevt, +}; + +static int freezer_stopper_fn(void *arg) +{ + struct freezer_data *fd = arg; + enum freezer_state state = FREEZER_NONE; + int cpu = smp_processor_id(); + + do { + cpu_relax(); + if (fd->state != state) { + state = fd->state; + if (freezer_func[state]) + (*freezer_func[state])(cpu); + ack_state(fd); + } + } while (fd->state != FREEZER_EXIT); + + return 0; } static void freeze_enter(void) { + struct freezer_data fd; + cpuidle_use_deepest_state(true); cpuidle_resume(); - wait_event(suspend_freeze_wait_head, suspend_freeze_wake); + + get_online_cpus(); + + fd.thread_num = num_online_cpus(); + set_state(&fd, FREEZER_PICK_TK); + + __stop_machine(freezer_stopper_fn, &fd, cpu_online_mask); + + put_online_cpus(); + cpuidle_pause(); cpuidle_use_deepest_state(false); } void freeze_wake(void) { - suspend_freeze_wake = true; - wake_up(&suspend_freeze_wait_head); + if (suspend_freeze_wake != -1) + return; + suspend_freeze_wake = smp_processor_id(); } EXPORT_SYMBOL_GPL(freeze_wake); diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index ec1791f..23d8feb 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -1114,7 +1114,7 @@ void timekeeping_inject_sleeptime(struct timespec *delta) * xtime/wall_to_monotonic/jiffies/etc are * still managed by arch specific suspend/resume code. */ -static void timekeeping_resume(void) +void timekeeping_resume(void) { struct timekeeper *tk = &tk_core.timekeeper; struct clocksource *clock = tk->tkr.clock; @@ -1195,7 +1195,7 @@ static void timekeeping_resume(void) hrtimers_resume(); } -static int timekeeping_suspend(void) +int timekeeping_suspend(void) { struct timekeeper *tk = &tk_core.timekeeper; unsigned long flags; diff --git a/kernel/time/timekeeping_internal.h b/kernel/time/timekeeping_internal.h index 4ea005a..ed7a574 100644 --- a/kernel/time/timekeeping_internal.h +++ b/kernel/time/timekeeping_internal.h @@ -26,4 +26,7 @@ static inline cycle_t clocksource_delta(cycle_t now, cycle_t last, cycle_t mask) } #endif +extern int timekeeping_suspend(void); +extern void timekeeping_resume(void); + #endif /* _TIMEKEEPING_INTERNAL_H */