2
0
mirror of https://github.com/edk2-porting/linux-next.git synced 2024-12-23 04:34:11 +08:00
linux-next/kernel/time/tick-broadcast.c
Linus Torvalds 534c97b095 Merge branch 'timers-nohz-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull 'full dynticks' support from Ingo Molnar:
 "This tree from Frederic Weisbecker adds a new, (exciting! :-) core
  kernel feature to the timer and scheduler subsystems: 'full dynticks',
  or CONFIG_NO_HZ_FULL=y.

  This feature extends the nohz variable-size timer tick feature from
  idle to busy CPUs (running at most one task) as well, potentially
  reducing the number of timer interrupts significantly.

  This feature got motivated by real-time folks and the -rt tree, but
  the general utility and motivation of full-dynticks runs wider than
  that:

   - HPC workloads get faster: CPUs running a single task should be able
     to utilize a maximum amount of CPU power.  A periodic timer tick at
     HZ=1000 can cause a constant overhead of up to 1.0%.  This feature
     removes that overhead - and speeds up the system by 0.5%-1.0% on
     typical distro configs even on modern systems.

   - Real-time workload latency reduction: CPUs running critical tasks
     should experience as little jitter as possible.  The last remaining
     source of kernel-related jitter was the periodic timer tick.

   - A single task executing on a CPU is a pretty common situation,
     especially with an increasing number of cores/CPUs, so this feature
     helps desktop and mobile workloads as well.

  The cost of the feature is mainly related to increased timer
  reprogramming overhead when a CPU switches its tick period, and thus
  slightly longer to-idle and from-idle latency.

  Configuration-wise a third mode of operation is added to the existing
  two NOHZ kconfig modes:

   - CONFIG_HZ_PERIODIC: [formerly !CONFIG_NO_HZ], now explicitly named
     as a config option.  This is the traditional Linux periodic tick
     design: there's a HZ tick going on all the time, regardless of
     whether a CPU is idle or not.

   - CONFIG_NO_HZ_IDLE: [formerly CONFIG_NO_HZ=y], this turns off the
     periodic tick when a CPU enters idle mode.

   - CONFIG_NO_HZ_FULL: this new mode, in addition to turning off the
     tick when a CPU is idle, also slows the tick down to 1 Hz (one
     timer interrupt per second) when only a single task is running on a
     CPU.

  The .config behavior is compatible: existing !CONFIG_NO_HZ and
  CONFIG_NO_HZ=y settings get translated to the new values, without the
  user having to configure anything.  CONFIG_NO_HZ_FULL is turned off by
  default.

  This feature is based on a lot of infrastructure work that has been
  steadily going upstream in the last 2-3 cycles: related RCU support
  and non-periodic cputime support in particular is upstream already.

  This tree adds the final pieces and activates the feature.  The pull
  request is marked RFC because:

   - it's marked 64-bit only at the moment - the 32-bit support patch is
     small but did not get ready in time.

   - it has a number of fresh commits that came in after the merge
     window.  The overwhelming majority of commits are from before the
     merge window, but still some aspects of the tree are fresh and so I
     marked it RFC.

   - it's a pretty wide-reaching feature with lots of effects - and
     while the components have been in testing for some time, the full
     combination is still not very widely used.  That it's default-off
     should reduce its regression abilities and obviously there are no
     known regressions with CONFIG_NO_HZ_FULL=y enabled either.

   - the feature is not completely idempotent: there is no 100%
     equivalent replacement for a periodic scheduler/timer tick.  In
     particular there's ongoing work to map out and reduce its effects
     on scheduler load-balancing and statistics.  This should not impact
     correctness though, there are no known regressions related to this
     feature at this point.

   - it's a pretty ambitious feature that with time will likely be
     enabled by most Linux distros, and we'd like you to make input on
     its design/implementation, if you dislike some aspect we missed.
     Without flaming us to crisp! :-)

  Future plans:

   - there's ongoing work to reduce 1Hz to 0Hz, to essentially shut off
     the periodic tick altogether when there's a single busy task on a
     CPU.  We'd first like 1 Hz to be exposed more widely before we go
     for the 0 Hz target though.

   - once we reach 0 Hz we can remove the periodic tick assumption from
     nr_running>=2 as well, by essentially interrupting busy tasks only
     as frequently as the sched_latency constraints require us to do -
     once every 4-40 msecs, depending on nr_running.

  I am personally leaning towards biting the bullet and doing this in
  v3.10, like the -rt tree this effort has been going on for too long -
  but the final word is up to you as usual.

  More technical details can be found in Documentation/timers/NO_HZ.txt"

* 'timers-nohz-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (39 commits)
  sched: Keep at least 1 tick per second for active dynticks tasks
  rcu: Fix full dynticks' dependency on wide RCU nocb mode
  nohz: Protect smp_processor_id() in tick_nohz_task_switch()
  nohz_full: Add documentation.
  cputime_nsecs: use math64.h for nsec resolution conversion helpers
  nohz: Select VIRT_CPU_ACCOUNTING_GEN from full dynticks config
  nohz: Reduce overhead under high-freq idling patterns
  nohz: Remove full dynticks' superfluous dependency on RCU tree
  nohz: Fix unavailable tick_stop tracepoint in dynticks idle
  nohz: Add basic tracing
  nohz: Select wide RCU nocb for full dynticks
  nohz: Disable the tick when irq resume in full dynticks CPU
  nohz: Re-evaluate the tick for the new task after a context switch
  nohz: Prepare to stop the tick on irq exit
  nohz: Implement full dynticks kick
  nohz: Re-evaluate the tick from the scheduler IPI
  sched: New helper to prevent from stopping the tick in full dynticks
  sched: Kick full dynticks CPU that have more than one task enqueued.
  perf: New helper to prevent full dynticks CPUs from stopping tick
  perf: Kick full dynticks CPU if events rotation is needed
  ...
2013-05-05 13:23:27 -07:00

797 lines
20 KiB
C

/*
* linux/kernel/time/tick-broadcast.c
*
* This file contains functions which emulate a local clock-event
* device via a broadcast event source.
*
* Copyright(C) 2005-2006, Thomas Gleixner <tglx@linutronix.de>
* Copyright(C) 2005-2007, Red Hat, Inc., Ingo Molnar
* Copyright(C) 2006-2007, Timesys Corp., Thomas Gleixner
*
* This code is licenced under the GPL version 2. For details see
* kernel-base/COPYING.
*/
#include <linux/cpu.h>
#include <linux/err.h>
#include <linux/hrtimer.h>
#include <linux/interrupt.h>
#include <linux/percpu.h>
#include <linux/profile.h>
#include <linux/sched.h>
#include <linux/smp.h>
#include "tick-internal.h"
/*
* Broadcast support for broken x86 hardware, where the local apic
* timer stops in C3 state.
*/
static struct tick_device tick_broadcast_device;
static cpumask_var_t tick_broadcast_mask;
static cpumask_var_t tmpmask;
static DEFINE_RAW_SPINLOCK(tick_broadcast_lock);
static int tick_broadcast_force;
#ifdef CONFIG_TICK_ONESHOT
static void tick_broadcast_clear_oneshot(int cpu);
#else
static inline void tick_broadcast_clear_oneshot(int cpu) { }
#endif
/*
* Debugging: see timer_list.c
*/
struct tick_device *tick_get_broadcast_device(void)
{
return &tick_broadcast_device;
}
struct cpumask *tick_get_broadcast_mask(void)
{
return tick_broadcast_mask;
}
/*
* Start the device in periodic mode
*/
static void tick_broadcast_start_periodic(struct clock_event_device *bc)
{
if (bc)
tick_setup_periodic(bc, 1);
}
/*
* Check, if the device can be utilized as broadcast device:
*/
int tick_check_broadcast_device(struct clock_event_device *dev)
{
struct clock_event_device *cur = tick_broadcast_device.evtdev;
if ((dev->features & CLOCK_EVT_FEAT_DUMMY) ||
(tick_broadcast_device.evtdev &&
tick_broadcast_device.evtdev->rating >= dev->rating) ||
(dev->features & CLOCK_EVT_FEAT_C3STOP))
return 0;
clockevents_exchange_device(tick_broadcast_device.evtdev, dev);
if (cur)
cur->event_handler = clockevents_handle_noop;
tick_broadcast_device.evtdev = dev;
if (!cpumask_empty(tick_broadcast_mask))
tick_broadcast_start_periodic(dev);
/*
* Inform all cpus about this. We might be in a situation
* where we did not switch to oneshot mode because the per cpu
* devices are affected by CLOCK_EVT_FEAT_C3STOP and the lack
* of a oneshot capable broadcast device. Without that
* notification the systems stays stuck in periodic mode
* forever.
*/
if (dev->features & CLOCK_EVT_FEAT_ONESHOT)
tick_clock_notify();
return 1;
}
/*
* Check, if the device is the broadcast device
*/
int tick_is_broadcast_device(struct clock_event_device *dev)
{
return (dev && tick_broadcast_device.evtdev == dev);
}
static void err_broadcast(const struct cpumask *mask)
{
pr_crit_once("Failed to broadcast timer tick. Some CPUs may be unresponsive.\n");
}
static void tick_device_setup_broadcast_func(struct clock_event_device *dev)
{
if (!dev->broadcast)
dev->broadcast = tick_broadcast;
if (!dev->broadcast) {
pr_warn_once("%s depends on broadcast, but no broadcast function available\n",
dev->name);
dev->broadcast = err_broadcast;
}
}
/*
* Check, if the device is disfunctional and a place holder, which
* needs to be handled by the broadcast device.
*/
int tick_device_uses_broadcast(struct clock_event_device *dev, int cpu)
{
unsigned long flags;
int ret = 0;
raw_spin_lock_irqsave(&tick_broadcast_lock, flags);
/*
* Devices might be registered with both periodic and oneshot
* mode disabled. This signals, that the device needs to be
* operated from the broadcast device and is a placeholder for
* the cpu local device.
*/
if (!tick_device_is_functional(dev)) {
dev->event_handler = tick_handle_periodic;
tick_device_setup_broadcast_func(dev);
cpumask_set_cpu(cpu, tick_broadcast_mask);
tick_broadcast_start_periodic(tick_broadcast_device.evtdev);
ret = 1;
} else {
/*
* When the new device is not affected by the stop
* feature and the cpu is marked in the broadcast mask
* then clear the broadcast bit.
*/
if (!(dev->features & CLOCK_EVT_FEAT_C3STOP)) {
int cpu = smp_processor_id();
cpumask_clear_cpu(cpu, tick_broadcast_mask);
tick_broadcast_clear_oneshot(cpu);
} else {
tick_device_setup_broadcast_func(dev);
}
}
raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags);
return ret;
}
#ifdef CONFIG_GENERIC_CLOCKEVENTS_BROADCAST
int tick_receive_broadcast(void)
{
struct tick_device *td = this_cpu_ptr(&tick_cpu_device);
struct clock_event_device *evt = td->evtdev;
if (!evt)
return -ENODEV;
if (!evt->event_handler)
return -EINVAL;
evt->event_handler(evt);
return 0;
}
#endif
/*
* Broadcast the event to the cpus, which are set in the mask (mangled).
*/
static void tick_do_broadcast(struct cpumask *mask)
{
int cpu = smp_processor_id();
struct tick_device *td;
/*
* Check, if the current cpu is in the mask
*/
if (cpumask_test_cpu(cpu, mask)) {
cpumask_clear_cpu(cpu, mask);
td = &per_cpu(tick_cpu_device, cpu);
td->evtdev->event_handler(td->evtdev);
}
if (!cpumask_empty(mask)) {
/*
* It might be necessary to actually check whether the devices
* have different broadcast functions. For now, just use the
* one of the first device. This works as long as we have this
* misfeature only on x86 (lapic)
*/
td = &per_cpu(tick_cpu_device, cpumask_first(mask));
td->evtdev->broadcast(mask);
}
}
/*
* Periodic broadcast:
* - invoke the broadcast handlers
*/
static void tick_do_periodic_broadcast(void)
{
raw_spin_lock(&tick_broadcast_lock);
cpumask_and(tmpmask, cpu_online_mask, tick_broadcast_mask);
tick_do_broadcast(tmpmask);
raw_spin_unlock(&tick_broadcast_lock);
}
/*
* Event handler for periodic broadcast ticks
*/
static void tick_handle_periodic_broadcast(struct clock_event_device *dev)
{
ktime_t next;
tick_do_periodic_broadcast();
/*
* The device is in periodic mode. No reprogramming necessary:
*/
if (dev->mode == CLOCK_EVT_MODE_PERIODIC)
return;
/*
* Setup the next period for devices, which do not have
* periodic mode. We read dev->next_event first and add to it
* when the event already expired. clockevents_program_event()
* sets dev->next_event only when the event is really
* programmed to the device.
*/
for (next = dev->next_event; ;) {
next = ktime_add(next, tick_period);
if (!clockevents_program_event(dev, next, false))
return;
tick_do_periodic_broadcast();
}
}
/*
* Powerstate information: The system enters/leaves a state, where
* affected devices might stop
*/
static void tick_do_broadcast_on_off(unsigned long *reason)
{
struct clock_event_device *bc, *dev;
struct tick_device *td;
unsigned long flags;
int cpu, bc_stopped;
raw_spin_lock_irqsave(&tick_broadcast_lock, flags);
cpu = smp_processor_id();
td = &per_cpu(tick_cpu_device, cpu);
dev = td->evtdev;
bc = tick_broadcast_device.evtdev;
/*
* Is the device not affected by the powerstate ?
*/
if (!dev || !(dev->features & CLOCK_EVT_FEAT_C3STOP))
goto out;
if (!tick_device_is_functional(dev))
goto out;
bc_stopped = cpumask_empty(tick_broadcast_mask);
switch (*reason) {
case CLOCK_EVT_NOTIFY_BROADCAST_ON:
case CLOCK_EVT_NOTIFY_BROADCAST_FORCE:
if (!cpumask_test_and_set_cpu(cpu, tick_broadcast_mask)) {
if (tick_broadcast_device.mode ==
TICKDEV_MODE_PERIODIC)
clockevents_shutdown(dev);
}
if (*reason == CLOCK_EVT_NOTIFY_BROADCAST_FORCE)
tick_broadcast_force = 1;
break;
case CLOCK_EVT_NOTIFY_BROADCAST_OFF:
if (!tick_broadcast_force &&
cpumask_test_and_clear_cpu(cpu, tick_broadcast_mask)) {
if (tick_broadcast_device.mode ==
TICKDEV_MODE_PERIODIC)
tick_setup_periodic(dev, 0);
}
break;
}
if (cpumask_empty(tick_broadcast_mask)) {
if (!bc_stopped)
clockevents_shutdown(bc);
} else if (bc_stopped) {
if (tick_broadcast_device.mode == TICKDEV_MODE_PERIODIC)
tick_broadcast_start_periodic(bc);
else
tick_broadcast_setup_oneshot(bc);
}
out:
raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags);
}
/*
* Powerstate information: The system enters/leaves a state, where
* affected devices might stop.
*/
void tick_broadcast_on_off(unsigned long reason, int *oncpu)
{
if (!cpumask_test_cpu(*oncpu, cpu_online_mask))
printk(KERN_ERR "tick-broadcast: ignoring broadcast for "
"offline CPU #%d\n", *oncpu);
else
tick_do_broadcast_on_off(&reason);
}
/*
* Set the periodic handler depending on broadcast on/off
*/
void tick_set_periodic_handler(struct clock_event_device *dev, int broadcast)
{
if (!broadcast)
dev->event_handler = tick_handle_periodic;
else
dev->event_handler = tick_handle_periodic_broadcast;
}
/*
* Remove a CPU from broadcasting
*/
void tick_shutdown_broadcast(unsigned int *cpup)
{
struct clock_event_device *bc;
unsigned long flags;
unsigned int cpu = *cpup;
raw_spin_lock_irqsave(&tick_broadcast_lock, flags);
bc = tick_broadcast_device.evtdev;
cpumask_clear_cpu(cpu, tick_broadcast_mask);
if (tick_broadcast_device.mode == TICKDEV_MODE_PERIODIC) {
if (bc && cpumask_empty(tick_broadcast_mask))
clockevents_shutdown(bc);
}
raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags);
}
void tick_suspend_broadcast(void)
{
struct clock_event_device *bc;
unsigned long flags;
raw_spin_lock_irqsave(&tick_broadcast_lock, flags);
bc = tick_broadcast_device.evtdev;
if (bc)
clockevents_shutdown(bc);
raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags);
}
int tick_resume_broadcast(void)
{
struct clock_event_device *bc;
unsigned long flags;
int broadcast = 0;
raw_spin_lock_irqsave(&tick_broadcast_lock, flags);
bc = tick_broadcast_device.evtdev;
if (bc) {
clockevents_set_mode(bc, CLOCK_EVT_MODE_RESUME);
switch (tick_broadcast_device.mode) {
case TICKDEV_MODE_PERIODIC:
if (!cpumask_empty(tick_broadcast_mask))
tick_broadcast_start_periodic(bc);
broadcast = cpumask_test_cpu(smp_processor_id(),
tick_broadcast_mask);
break;
case TICKDEV_MODE_ONESHOT:
if (!cpumask_empty(tick_broadcast_mask))
broadcast = tick_resume_broadcast_oneshot(bc);
break;
}
}
raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags);
return broadcast;
}
#ifdef CONFIG_TICK_ONESHOT
static cpumask_var_t tick_broadcast_oneshot_mask;
static cpumask_var_t tick_broadcast_pending_mask;
static cpumask_var_t tick_broadcast_force_mask;
/*
* Exposed for debugging: see timer_list.c
*/
struct cpumask *tick_get_broadcast_oneshot_mask(void)
{
return tick_broadcast_oneshot_mask;
}
/*
* Called before going idle with interrupts disabled. Checks whether a
* broadcast event from the other core is about to happen. We detected
* that in tick_broadcast_oneshot_control(). The callsite can use this
* to avoid a deep idle transition as we are about to get the
* broadcast IPI right away.
*/
int tick_check_broadcast_expired(void)
{
return cpumask_test_cpu(smp_processor_id(), tick_broadcast_force_mask);
}
/*
* Set broadcast interrupt affinity
*/
static void tick_broadcast_set_affinity(struct clock_event_device *bc,
const struct cpumask *cpumask)
{
if (!(bc->features & CLOCK_EVT_FEAT_DYNIRQ))
return;
if (cpumask_equal(bc->cpumask, cpumask))
return;
bc->cpumask = cpumask;
irq_set_affinity(bc->irq, bc->cpumask);
}
static int tick_broadcast_set_event(struct clock_event_device *bc, int cpu,
ktime_t expires, int force)
{
int ret;
if (bc->mode != CLOCK_EVT_MODE_ONESHOT)
clockevents_set_mode(bc, CLOCK_EVT_MODE_ONESHOT);
ret = clockevents_program_event(bc, expires, force);
if (!ret)
tick_broadcast_set_affinity(bc, cpumask_of(cpu));
return ret;
}
int tick_resume_broadcast_oneshot(struct clock_event_device *bc)
{
clockevents_set_mode(bc, CLOCK_EVT_MODE_ONESHOT);
return 0;
}
/*
* Called from irq_enter() when idle was interrupted to reenable the
* per cpu device.
*/
void tick_check_oneshot_broadcast(int cpu)
{
if (cpumask_test_cpu(cpu, tick_broadcast_oneshot_mask)) {
struct tick_device *td = &per_cpu(tick_cpu_device, cpu);
clockevents_set_mode(td->evtdev, CLOCK_EVT_MODE_ONESHOT);
}
}
/*
* Handle oneshot mode broadcasting
*/
static void tick_handle_oneshot_broadcast(struct clock_event_device *dev)
{
struct tick_device *td;
ktime_t now, next_event;
int cpu, next_cpu = 0;
raw_spin_lock(&tick_broadcast_lock);
again:
dev->next_event.tv64 = KTIME_MAX;
next_event.tv64 = KTIME_MAX;
cpumask_clear(tmpmask);
now = ktime_get();
/* Find all expired events */
for_each_cpu(cpu, tick_broadcast_oneshot_mask) {
td = &per_cpu(tick_cpu_device, cpu);
if (td->evtdev->next_event.tv64 <= now.tv64) {
cpumask_set_cpu(cpu, tmpmask);
/*
* Mark the remote cpu in the pending mask, so
* it can avoid reprogramming the cpu local
* timer in tick_broadcast_oneshot_control().
*/
cpumask_set_cpu(cpu, tick_broadcast_pending_mask);
} else if (td->evtdev->next_event.tv64 < next_event.tv64) {
next_event.tv64 = td->evtdev->next_event.tv64;
next_cpu = cpu;
}
}
/* Take care of enforced broadcast requests */
cpumask_or(tmpmask, tmpmask, tick_broadcast_force_mask);
cpumask_clear(tick_broadcast_force_mask);
/*
* Wakeup the cpus which have an expired event.
*/
tick_do_broadcast(tmpmask);
/*
* Two reasons for reprogram:
*
* - The global event did not expire any CPU local
* events. This happens in dyntick mode, as the maximum PIT
* delta is quite small.
*
* - There are pending events on sleeping CPUs which were not
* in the event mask
*/
if (next_event.tv64 != KTIME_MAX) {
/*
* Rearm the broadcast device. If event expired,
* repeat the above
*/
if (tick_broadcast_set_event(dev, next_cpu, next_event, 0))
goto again;
}
raw_spin_unlock(&tick_broadcast_lock);
}
/*
* Powerstate information: The system enters/leaves a state, where
* affected devices might stop
*/
void tick_broadcast_oneshot_control(unsigned long reason)
{
struct clock_event_device *bc, *dev;
struct tick_device *td;
unsigned long flags;
ktime_t now;
int cpu;
/*
* Periodic mode does not care about the enter/exit of power
* states
*/
if (tick_broadcast_device.mode == TICKDEV_MODE_PERIODIC)
return;
/*
* We are called with preemtion disabled from the depth of the
* idle code, so we can't be moved away.
*/
cpu = smp_processor_id();
td = &per_cpu(tick_cpu_device, cpu);
dev = td->evtdev;
if (!(dev->features & CLOCK_EVT_FEAT_C3STOP))
return;
bc = tick_broadcast_device.evtdev;
raw_spin_lock_irqsave(&tick_broadcast_lock, flags);
if (reason == CLOCK_EVT_NOTIFY_BROADCAST_ENTER) {
WARN_ON_ONCE(cpumask_test_cpu(cpu, tick_broadcast_pending_mask));
if (!cpumask_test_and_set_cpu(cpu, tick_broadcast_oneshot_mask)) {
clockevents_set_mode(dev, CLOCK_EVT_MODE_SHUTDOWN);
/*
* We only reprogram the broadcast timer if we
* did not mark ourself in the force mask and
* if the cpu local event is earlier than the
* broadcast event. If the current CPU is in
* the force mask, then we are going to be
* woken by the IPI right away.
*/
if (!cpumask_test_cpu(cpu, tick_broadcast_force_mask) &&
dev->next_event.tv64 < bc->next_event.tv64)
tick_broadcast_set_event(bc, cpu, dev->next_event, 1);
}
} else {
if (cpumask_test_and_clear_cpu(cpu, tick_broadcast_oneshot_mask)) {
clockevents_set_mode(dev, CLOCK_EVT_MODE_ONESHOT);
if (dev->next_event.tv64 == KTIME_MAX)
goto out;
/*
* The cpu which was handling the broadcast
* timer marked this cpu in the broadcast
* pending mask and fired the broadcast
* IPI. So we are going to handle the expired
* event anyway via the broadcast IPI
* handler. No need to reprogram the timer
* with an already expired event.
*/
if (cpumask_test_and_clear_cpu(cpu,
tick_broadcast_pending_mask))
goto out;
/*
* If the pending bit is not set, then we are
* either the CPU handling the broadcast
* interrupt or we got woken by something else.
*
* We are not longer in the broadcast mask, so
* if the cpu local expiry time is already
* reached, we would reprogram the cpu local
* timer with an already expired event.
*
* This can lead to a ping-pong when we return
* to idle and therefor rearm the broadcast
* timer before the cpu local timer was able
* to fire. This happens because the forced
* reprogramming makes sure that the event
* will happen in the future and depending on
* the min_delta setting this might be far
* enough out that the ping-pong starts.
*
* If the cpu local next_event has expired
* then we know that the broadcast timer
* next_event has expired as well and
* broadcast is about to be handled. So we
* avoid reprogramming and enforce that the
* broadcast handler, which did not run yet,
* will invoke the cpu local handler.
*
* We cannot call the handler directly from
* here, because we might be in a NOHZ phase
* and we did not go through the irq_enter()
* nohz fixups.
*/
now = ktime_get();
if (dev->next_event.tv64 <= now.tv64) {
cpumask_set_cpu(cpu, tick_broadcast_force_mask);
goto out;
}
/*
* We got woken by something else. Reprogram
* the cpu local timer device.
*/
tick_program_event(dev->next_event, 1);
}
}
out:
raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags);
}
/*
* Reset the one shot broadcast for a cpu
*
* Called with tick_broadcast_lock held
*/
static void tick_broadcast_clear_oneshot(int cpu)
{
cpumask_clear_cpu(cpu, tick_broadcast_oneshot_mask);
}
static void tick_broadcast_init_next_event(struct cpumask *mask,
ktime_t expires)
{
struct tick_device *td;
int cpu;
for_each_cpu(cpu, mask) {
td = &per_cpu(tick_cpu_device, cpu);
if (td->evtdev)
td->evtdev->next_event = expires;
}
}
/**
* tick_broadcast_setup_oneshot - setup the broadcast device
*/
void tick_broadcast_setup_oneshot(struct clock_event_device *bc)
{
int cpu = smp_processor_id();
/* Set it up only once ! */
if (bc->event_handler != tick_handle_oneshot_broadcast) {
int was_periodic = bc->mode == CLOCK_EVT_MODE_PERIODIC;
bc->event_handler = tick_handle_oneshot_broadcast;
/* Take the do_timer update */
if (!tick_nohz_full_cpu(cpu))
tick_do_timer_cpu = cpu;
/*
* We must be careful here. There might be other CPUs
* waiting for periodic broadcast. We need to set the
* oneshot_mask bits for those and program the
* broadcast device to fire.
*/
cpumask_copy(tmpmask, tick_broadcast_mask);
cpumask_clear_cpu(cpu, tmpmask);
cpumask_or(tick_broadcast_oneshot_mask,
tick_broadcast_oneshot_mask, tmpmask);
if (was_periodic && !cpumask_empty(tmpmask)) {
clockevents_set_mode(bc, CLOCK_EVT_MODE_ONESHOT);
tick_broadcast_init_next_event(tmpmask,
tick_next_period);
tick_broadcast_set_event(bc, cpu, tick_next_period, 1);
} else
bc->next_event.tv64 = KTIME_MAX;
} else {
/*
* The first cpu which switches to oneshot mode sets
* the bit for all other cpus which are in the general
* (periodic) broadcast mask. So the bit is set and
* would prevent the first broadcast enter after this
* to program the bc device.
*/
tick_broadcast_clear_oneshot(cpu);
}
}
/*
* Select oneshot operating mode for the broadcast device
*/
void tick_broadcast_switch_to_oneshot(void)
{
struct clock_event_device *bc;
unsigned long flags;
raw_spin_lock_irqsave(&tick_broadcast_lock, flags);
tick_broadcast_device.mode = TICKDEV_MODE_ONESHOT;
bc = tick_broadcast_device.evtdev;
if (bc)
tick_broadcast_setup_oneshot(bc);
raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags);
}
/*
* Remove a dead CPU from broadcasting
*/
void tick_shutdown_broadcast_oneshot(unsigned int *cpup)
{
unsigned long flags;
unsigned int cpu = *cpup;
raw_spin_lock_irqsave(&tick_broadcast_lock, flags);
/*
* Clear the broadcast mask flag for the dead cpu, but do not
* stop the broadcast device!
*/
cpumask_clear_cpu(cpu, tick_broadcast_oneshot_mask);
raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags);
}
/*
* Check, whether the broadcast device is in one shot mode
*/
int tick_broadcast_oneshot_active(void)
{
return tick_broadcast_device.mode == TICKDEV_MODE_ONESHOT;
}
/*
* Check whether the broadcast device supports oneshot.
*/
bool tick_broadcast_oneshot_available(void)
{
struct clock_event_device *bc = tick_broadcast_device.evtdev;
return bc ? bc->features & CLOCK_EVT_FEAT_ONESHOT : false;
}
#endif
void __init tick_broadcast_init(void)
{
alloc_cpumask_var(&tick_broadcast_mask, GFP_NOWAIT);
alloc_cpumask_var(&tmpmask, GFP_NOWAIT);
#ifdef CONFIG_TICK_ONESHOT
alloc_cpumask_var(&tick_broadcast_oneshot_mask, GFP_NOWAIT);
alloc_cpumask_var(&tick_broadcast_pending_mask, GFP_NOWAIT);
alloc_cpumask_var(&tick_broadcast_force_mask, GFP_NOWAIT);
#endif
}