2014-01-27 05:42:01 +01:00
|
|
|
/*
|
|
|
|
* Generic entry point for the idle threads
|
|
|
|
*/
|
|
|
|
#include <linux/sched.h>
|
|
|
|
#include <linux/cpu.h>
|
|
|
|
#include <linux/cpuidle.h>
|
|
|
|
#include <linux/tick.h>
|
|
|
|
#include <linux/mm.h>
|
|
|
|
#include <linux/stackprotector.h>
|
|
|
|
|
|
|
|
#include <asm/tlb.h>
|
|
|
|
|
|
|
|
#include <trace/events/power.h>
|
|
|
|
|
|
|
|
static int __read_mostly cpu_idle_force_poll;
|
|
|
|
|
|
|
|
void cpu_idle_poll_ctrl(bool enable)
|
|
|
|
{
|
|
|
|
if (enable) {
|
|
|
|
cpu_idle_force_poll++;
|
|
|
|
} else {
|
|
|
|
cpu_idle_force_poll--;
|
|
|
|
WARN_ON_ONCE(cpu_idle_force_poll < 0);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
#ifdef CONFIG_GENERIC_IDLE_POLL_SETUP
|
|
|
|
static int __init cpu_idle_poll_setup(char *__unused)
|
|
|
|
{
|
|
|
|
cpu_idle_force_poll = 1;
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
__setup("nohlt", cpu_idle_poll_setup);
|
|
|
|
|
|
|
|
static int __init cpu_idle_nopoll_setup(char *__unused)
|
|
|
|
{
|
|
|
|
cpu_idle_force_poll = 0;
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
__setup("hlt", cpu_idle_nopoll_setup);
|
|
|
|
#endif
|
|
|
|
|
|
|
|
static inline int cpu_idle_poll(void)
|
|
|
|
{
|
|
|
|
rcu_idle_enter();
|
|
|
|
trace_cpu_idle_rcuidle(0, smp_processor_id());
|
|
|
|
local_irq_enable();
|
|
|
|
while (!tif_need_resched())
|
|
|
|
cpu_relax();
|
|
|
|
trace_cpu_idle_rcuidle(PWR_EVENT_EXIT, smp_processor_id());
|
|
|
|
rcu_idle_exit();
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Weak implementations for optional arch specific functions */
|
|
|
|
void __weak arch_cpu_idle_prepare(void) { }
|
|
|
|
void __weak arch_cpu_idle_enter(void) { }
|
|
|
|
void __weak arch_cpu_idle_exit(void) { }
|
|
|
|
void __weak arch_cpu_idle_dead(void) { }
|
|
|
|
void __weak arch_cpu_idle(void)
|
|
|
|
{
|
|
|
|
cpu_idle_force_poll = 1;
|
|
|
|
local_irq_enable();
|
|
|
|
}
|
|
|
|
|
2014-03-03 08:48:51 +01:00
|
|
|
/**
|
|
|
|
* cpuidle_idle_call - the main idle function
|
|
|
|
*
|
|
|
|
* NOTE: no locks or semaphores should be used here
|
|
|
|
* return non-zero on failure
|
|
|
|
*/
|
|
|
|
static int cpuidle_idle_call(void)
|
|
|
|
{
|
|
|
|
struct cpuidle_device *dev = __this_cpu_read(cpuidle_devices);
|
|
|
|
struct cpuidle_driver *drv = cpuidle_get_cpu_driver(dev);
|
|
|
|
int next_state, entered_state, ret;
|
|
|
|
bool broadcast;
|
|
|
|
|
2014-03-03 08:48:54 +01:00
|
|
|
/*
|
|
|
|
* Check if the idle task must be rescheduled. If it is the
|
|
|
|
* case, exit the function after re-enabling the local irq and
|
|
|
|
* set again the polling flag
|
|
|
|
*/
|
2014-03-03 08:48:53 +01:00
|
|
|
if (current_clr_polling_and_test()) {
|
|
|
|
local_irq_enable();
|
|
|
|
__current_set_polling();
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2014-03-03 08:48:54 +01:00
|
|
|
/*
|
|
|
|
* During the idle period, stop measuring the disabled irqs
|
|
|
|
* critical sections latencies
|
|
|
|
*/
|
2014-03-03 08:48:52 +01:00
|
|
|
stop_critical_timings();
|
2014-03-03 08:48:54 +01:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Tell the RCU framework we are entering an idle section,
|
|
|
|
* so no more rcu read side critical sections and one more
|
|
|
|
* step to the grace period
|
|
|
|
*/
|
2014-03-03 08:48:52 +01:00
|
|
|
rcu_idle_enter();
|
|
|
|
|
2014-03-03 08:48:54 +01:00
|
|
|
/*
|
|
|
|
* Check if the cpuidle framework is ready, otherwise fallback
|
|
|
|
* to the default arch specific idle method
|
|
|
|
*/
|
2014-03-03 08:48:51 +01:00
|
|
|
ret = cpuidle_enabled(drv, dev);
|
|
|
|
|
2014-03-03 08:48:53 +01:00
|
|
|
if (!ret) {
|
2014-03-03 08:48:54 +01:00
|
|
|
/*
|
|
|
|
* Ask the governor to choose an idle state it thinks
|
|
|
|
* it is convenient to go to. There is *always* a
|
|
|
|
* convenient idle state
|
|
|
|
*/
|
2014-03-03 08:48:53 +01:00
|
|
|
next_state = cpuidle_select(drv, dev);
|
2014-03-03 08:48:51 +01:00
|
|
|
|
2014-03-03 08:48:54 +01:00
|
|
|
/*
|
|
|
|
* The idle task must be scheduled, it is pointless to
|
|
|
|
* go to idle, just update no idle residency and get
|
|
|
|
* out of this function
|
|
|
|
*/
|
2014-03-03 08:48:53 +01:00
|
|
|
if (current_clr_polling_and_test()) {
|
|
|
|
dev->last_residency = 0;
|
|
|
|
entered_state = next_state;
|
|
|
|
local_irq_enable();
|
|
|
|
} else {
|
|
|
|
broadcast = !!(drv->states[next_state].flags &
|
|
|
|
CPUIDLE_FLAG_TIMER_STOP);
|
2014-03-03 08:48:51 +01:00
|
|
|
|
2014-03-03 08:48:53 +01:00
|
|
|
if (broadcast)
|
2014-03-03 08:48:54 +01:00
|
|
|
/*
|
|
|
|
* Tell the time framework to switch
|
|
|
|
* to a broadcast timer because our
|
|
|
|
* local timer will be shutdown. If a
|
|
|
|
* local timer is used from another
|
|
|
|
* cpu as a broadcast timer, this call
|
|
|
|
* may fail if it is not available
|
|
|
|
*/
|
2014-03-03 08:48:53 +01:00
|
|
|
ret = clockevents_notify(
|
|
|
|
CLOCK_EVT_NOTIFY_BROADCAST_ENTER,
|
|
|
|
&dev->cpu);
|
2014-03-03 08:48:51 +01:00
|
|
|
|
2014-03-03 08:48:53 +01:00
|
|
|
if (!ret) {
|
|
|
|
trace_cpu_idle_rcuidle(next_state, dev->cpu);
|
2014-03-03 08:48:51 +01:00
|
|
|
|
2014-03-03 08:48:54 +01:00
|
|
|
/*
|
|
|
|
* Enter the idle state previously
|
|
|
|
* returned by the governor
|
|
|
|
* decision. This function will block
|
|
|
|
* until an interrupt occurs and will
|
|
|
|
* take care of re-enabling the local
|
|
|
|
* interrupts
|
|
|
|
*/
|
2014-03-03 08:48:53 +01:00
|
|
|
entered_state = cpuidle_enter(drv, dev,
|
|
|
|
next_state);
|
2014-03-03 08:48:51 +01:00
|
|
|
|
2014-03-03 08:48:53 +01:00
|
|
|
trace_cpu_idle_rcuidle(PWR_EVENT_EXIT,
|
|
|
|
dev->cpu);
|
2014-03-03 08:48:51 +01:00
|
|
|
|
2014-03-03 08:48:53 +01:00
|
|
|
if (broadcast)
|
|
|
|
clockevents_notify(
|
|
|
|
CLOCK_EVT_NOTIFY_BROADCAST_EXIT,
|
|
|
|
&dev->cpu);
|
2014-03-03 08:48:51 +01:00
|
|
|
|
2014-03-03 08:48:54 +01:00
|
|
|
/*
|
|
|
|
* Give the governor an opportunity to reflect on the
|
|
|
|
* outcome
|
|
|
|
*/
|
2014-03-03 08:48:53 +01:00
|
|
|
cpuidle_reflect(dev, entered_state);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2014-03-03 08:48:54 +01:00
|
|
|
/*
|
|
|
|
* We can't use the cpuidle framework, let's use the default
|
|
|
|
* idle routine
|
|
|
|
*/
|
2014-03-03 08:48:53 +01:00
|
|
|
if (ret)
|
|
|
|
arch_cpu_idle();
|
|
|
|
|
|
|
|
__current_set_polling();
|
2014-03-03 08:48:51 +01:00
|
|
|
|
2014-03-03 08:48:54 +01:00
|
|
|
/*
|
|
|
|
* It is up to the idle functions to enable back the local
|
|
|
|
* interrupt
|
|
|
|
*/
|
2014-03-03 08:48:52 +01:00
|
|
|
if (WARN_ON_ONCE(irqs_disabled()))
|
|
|
|
local_irq_enable();
|
|
|
|
|
|
|
|
rcu_idle_exit();
|
|
|
|
start_critical_timings();
|
2014-03-03 08:48:51 +01:00
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2014-01-27 05:42:01 +01:00
|
|
|
/*
|
|
|
|
* Generic idle loop implementation
|
|
|
|
*/
|
|
|
|
static void cpu_idle_loop(void)
|
|
|
|
{
|
|
|
|
while (1) {
|
|
|
|
tick_nohz_idle_enter();
|
|
|
|
|
|
|
|
while (!need_resched()) {
|
|
|
|
check_pgt_cache();
|
|
|
|
rmb();
|
|
|
|
|
|
|
|
if (cpu_is_offline(smp_processor_id()))
|
|
|
|
arch_cpu_idle_dead();
|
|
|
|
|
|
|
|
local_irq_disable();
|
|
|
|
arch_cpu_idle_enter();
|
|
|
|
|
|
|
|
/*
|
|
|
|
* In poll mode we reenable interrupts and spin.
|
|
|
|
*
|
|
|
|
* Also if we detected in the wakeup from idle
|
|
|
|
* path that the tick broadcast device expired
|
|
|
|
* for us, we don't want to go deep idle as we
|
|
|
|
* know that the IPI is going to arrive right
|
|
|
|
* away
|
|
|
|
*/
|
2014-03-03 08:48:53 +01:00
|
|
|
if (cpu_idle_force_poll || tick_check_broadcast_expired())
|
2014-01-27 05:42:01 +01:00
|
|
|
cpu_idle_poll();
|
2014-03-03 08:48:53 +01:00
|
|
|
else
|
|
|
|
cpuidle_idle_call();
|
|
|
|
|
2014-01-27 05:42:01 +01:00
|
|
|
arch_cpu_idle_exit();
|
|
|
|
}
|
2014-02-24 18:22:07 +01:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Since we fell out of the loop above, we know
|
|
|
|
* TIF_NEED_RESCHED must be set, propagate it into
|
|
|
|
* PREEMPT_NEED_RESCHED.
|
|
|
|
*
|
|
|
|
* This is required because for polling idle loops we will
|
|
|
|
* not have had an IPI to fold the state for us.
|
|
|
|
*/
|
|
|
|
preempt_set_need_resched();
|
2014-01-27 05:42:01 +01:00
|
|
|
tick_nohz_idle_exit();
|
|
|
|
schedule_preempt_disabled();
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
void cpu_startup_entry(enum cpuhp_state state)
|
|
|
|
{
|
|
|
|
/*
|
|
|
|
* This #ifdef needs to die, but it's too late in the cycle to
|
|
|
|
* make this generic (arm and sh have never invoked the canary
|
|
|
|
* init for the non boot cpus!). Will be fixed in 3.11
|
|
|
|
*/
|
|
|
|
#ifdef CONFIG_X86
|
|
|
|
/*
|
|
|
|
* If we're the non-boot CPU, nothing set the stack canary up
|
|
|
|
* for us. The boot CPU already has it initialized but no harm
|
|
|
|
* in doing it again. This is a good place for updating it, as
|
|
|
|
* we wont ever return from this function (so the invalid
|
|
|
|
* canaries already on the stack wont ever trigger).
|
|
|
|
*/
|
|
|
|
boot_init_stack_canary();
|
|
|
|
#endif
|
|
|
|
__current_set_polling();
|
|
|
|
arch_cpu_idle_prepare();
|
|
|
|
cpu_idle_loop();
|
|
|
|
}
|