zephyr/kernel/thread.c
Ramesh Thomas 89ffd44dfb kernel: tickless: Add tickless kernel support
Adds event based scheduling logic to the kernel. Updates
management of timeouts, timers, idling etc. based on
time tracked at events rather than periodic ticks. Provides
interfaces for timers to announce and get next timer expiry
based on kernel scheduling decisions involving time slicing
of threads, timeouts and idling. Uses wall time units instead
of ticks in all scheduling activities.

The implementation involves changes in the following areas

1. Management of time in wall units like ms/us instead of ticks
The existing implementation already had an option to configure
number of ticks in a second. The new implementation builds on
top of that feature and provides option to set the size of the
scheduling granurality to mili seconds or micro seconds. This
allows most of the current implementation to be reused. Due to
this re-use and co-existence with tick based kernel, the names
of variables may contain the word "tick". However, in the
tickless kernel implementation, it represents the currently
configured time unit, which would be be mili seconds or
micro seconds. The APIs that take time as a parameter are not
impacted and they continue to pass time in mili seconds.

2. Timers would not be programmed in periodic mode
generating ticks. Instead they would be programmed in one
shot mode to generate events at the time the kernel scheduler
needs to gain control for its scheduling activities like
timers, timeouts, time slicing, idling etc.

3. The scheduler provides interfaces that the timer drivers
use to announce elapsed time and get the next time the scheduler
needs a timer event. It is possible that the scheduler may not
need another timer event, in which case the system would wait
for a non-timer event to wake it up if it is idling.

4. New APIs are defined to be implemented by timer drivers. Also
they need to handler timer events differently. These changes
have been done in the HPET timer driver. In future other timers
that support tickles kernel should implement these APIs as well.
These APIs are to re-program the timer, update and announce
elapsed time.

5. Philosopher and timer_api applications have been enabled to
test tickless kernel. Separate configuration files are created
which define the necessary CONFIG flags. Run these apps using
following command
make pristine && make BOARD=qemu_x86 CONF_FILE=prj_tickless.conf qemu

Jira: ZEP-339 ZEP-1946 ZEP-948
Change-Id: I7d950c31bf1ff929a9066fad42c2f0559a2e5983
Signed-off-by: Ramesh Thomas <ramesh.thomas@intel.com>
2017-04-27 13:46:28 +00:00

435 lines
9.4 KiB
C

/*
* Copyright (c) 2010-2014 Wind River Systems, Inc.
*
* SPDX-License-Identifier: Apache-2.0
*/
/**
* @file
* @brief Kernel thread support
*
* This module provides general purpose thread support.
*/
#include <kernel.h>
#include <toolchain.h>
#include <sections.h>
#include <kernel_structs.h>
#include <misc/printk.h>
#include <sys_clock.h>
#include <drivers/system_timer.h>
#include <ksched.h>
#include <wait_q.h>
extern struct _static_thread_data _static_thread_data_list_start[];
extern struct _static_thread_data _static_thread_data_list_end[];
#define _FOREACH_STATIC_THREAD(thread_data) \
for (struct _static_thread_data *thread_data = \
_static_thread_data_list_start; \
thread_data < _static_thread_data_list_end; \
thread_data++)
int k_is_in_isr(void)
{
return _is_in_isr();
}
/*
* This function tags the current thread as essential to system operation.
* Exceptions raised by this thread will be treated as a fatal system error.
*/
void _thread_essential_set(void)
{
_current->base.user_options |= K_ESSENTIAL;
}
/*
* This function tags the current thread as not essential to system operation.
* Exceptions raised by this thread may be recoverable.
* (This is the default tag for a thread.)
*/
void _thread_essential_clear(void)
{
_current->base.user_options &= ~K_ESSENTIAL;
}
/*
* This routine indicates if the current thread is an essential system thread.
*
* Returns non-zero if current thread is essential, zero if it is not.
*/
int _is_thread_essential(void)
{
return _current->base.user_options & K_ESSENTIAL;
}
void k_busy_wait(u32_t usec_to_wait)
{
#if defined(CONFIG_TICKLESS_KERNEL) && \
!defined(CONFIG_BUSY_WAIT_USES_ALTERNATE_CLOCK)
int saved_always_on = k_enable_sys_clock_always_on();
#endif
/* use 64-bit math to prevent overflow when multiplying */
u32_t cycles_to_wait = (u32_t)(
(u64_t)usec_to_wait *
(u64_t)sys_clock_hw_cycles_per_sec /
(u64_t)USEC_PER_SEC
);
u32_t start_cycles = k_cycle_get_32();
for (;;) {
u32_t current_cycles = k_cycle_get_32();
/* this handles the rollover on an unsigned 32-bit value */
if ((current_cycles - start_cycles) >= cycles_to_wait) {
break;
}
}
#if defined(CONFIG_TICKLESS_KERNEL) && \
!defined(CONFIG_BUSY_WAIT_USES_ALTERNATE_CLOCK)
_sys_clock_always_on = saved_always_on;
#endif
}
#ifdef CONFIG_THREAD_CUSTOM_DATA
void k_thread_custom_data_set(void *value)
{
_current->custom_data = value;
}
void *k_thread_custom_data_get(void)
{
return _current->custom_data;
}
#endif /* CONFIG_THREAD_CUSTOM_DATA */
#if defined(CONFIG_THREAD_MONITOR)
/*
* Remove a thread from the kernel's list of active threads.
*/
void _thread_monitor_exit(struct k_thread *thread)
{
unsigned int key = irq_lock();
if (thread == _kernel.threads) {
_kernel.threads = _kernel.threads->next_thread;
} else {
struct k_thread *prev_thread;
prev_thread = _kernel.threads;
while (thread != prev_thread->next_thread) {
prev_thread = prev_thread->next_thread;
}
prev_thread->next_thread = thread->next_thread;
}
irq_unlock(key);
}
#endif /* CONFIG_THREAD_MONITOR */
/*
* Common thread entry point function (used by all threads)
*
* This routine invokes the actual thread entry point function and passes
* it three arguments. It also handles graceful termination of the thread
* if the entry point function ever returns.
*
* This routine does not return, and is marked as such so the compiler won't
* generate preamble code that is only used by functions that actually return.
*/
FUNC_NORETURN void _thread_entry(void (*entry)(void *, void *, void *),
void *p1, void *p2, void *p3)
{
entry(p1, p2, p3);
#ifdef CONFIG_MULTITHREADING
if (_is_thread_essential()) {
_k_except_reason(_NANO_ERR_INVALID_TASK_EXIT);
}
k_thread_abort(_current);
#else
for (;;) {
k_cpu_idle();
}
#endif
/*
* Compiler can't tell that k_thread_abort() won't return and issues a
* warning unless we tell it that control never gets this far.
*/
CODE_UNREACHABLE;
}
#ifdef CONFIG_MULTITHREADING
static void start_thread(struct k_thread *thread)
{
int key = irq_lock(); /* protect kernel queues */
_mark_thread_as_started(thread);
if (_is_thread_ready(thread)) {
_add_thread_to_ready_q(thread);
if (_must_switch_threads()) {
_Swap(key);
return;
}
}
irq_unlock(key);
}
#endif
#ifdef CONFIG_MULTITHREADING
static void schedule_new_thread(struct k_thread *thread, s32_t delay)
{
#ifdef CONFIG_SYS_CLOCK_EXISTS
if (delay == 0) {
start_thread(thread);
} else {
s32_t ticks = _TICK_ALIGN + _ms_to_ticks(delay);
int key = irq_lock();
_add_thread_timeout(thread, NULL, ticks);
irq_unlock(key);
}
#else
ARG_UNUSED(delay);
start_thread(thread);
#endif
}
#endif
#ifdef CONFIG_MULTITHREADING
k_tid_t k_thread_spawn(char *stack, size_t stack_size,
void (*entry)(void *, void *, void*),
void *p1, void *p2, void *p3,
int prio, u32_t options, s32_t delay)
{
__ASSERT(!_is_in_isr(), "");
struct k_thread *new_thread = (struct k_thread *)stack;
_new_thread(stack, stack_size, entry, p1, p2, p3, prio, options);
schedule_new_thread(new_thread, delay);
return new_thread;
}
#endif
int k_thread_cancel(k_tid_t tid)
{
struct k_thread *thread = tid;
int key = irq_lock();
if (_has_thread_started(thread) ||
!_is_thread_timeout_active(thread)) {
irq_unlock(key);
return -EINVAL;
}
_abort_thread_timeout(thread);
_thread_monitor_exit(thread);
irq_unlock(key);
return 0;
}
static inline int is_in_any_group(struct _static_thread_data *thread_data,
u32_t groups)
{
return !!(thread_data->init_groups & groups);
}
void _k_thread_group_op(u32_t groups, void (*func)(struct k_thread *))
{
unsigned int key;
__ASSERT(!_is_in_isr(), "");
_sched_lock();
/* Invoke func() on each static thread in the specified group set. */
_FOREACH_STATIC_THREAD(thread_data) {
if (is_in_any_group(thread_data, groups)) {
key = irq_lock();
func(thread_data->thread);
irq_unlock(key);
}
}
/*
* If the current thread is still in a ready state, then let the
* "unlock scheduler" code determine if any rescheduling is needed.
*/
if (_is_thread_ready(_current)) {
k_sched_unlock();
return;
}
/* The current thread is no longer in a ready state--reschedule. */
key = irq_lock();
_sched_unlock_no_reschedule();
_Swap(key);
}
void _k_thread_single_start(struct k_thread *thread)
{
_mark_thread_as_started(thread);
if (_is_thread_ready(thread)) {
_add_thread_to_ready_q(thread);
}
}
void _k_thread_single_suspend(struct k_thread *thread)
{
if (_is_thread_ready(thread)) {
_remove_thread_from_ready_q(thread);
}
_mark_thread_as_suspended(thread);
}
void k_thread_suspend(struct k_thread *thread)
{
unsigned int key = irq_lock();
_k_thread_single_suspend(thread);
if (thread == _current) {
_Swap(key);
} else {
irq_unlock(key);
}
}
void _k_thread_single_resume(struct k_thread *thread)
{
_mark_thread_as_not_suspended(thread);
if (_is_thread_ready(thread)) {
_add_thread_to_ready_q(thread);
}
}
void k_thread_resume(struct k_thread *thread)
{
unsigned int key = irq_lock();
_k_thread_single_resume(thread);
_reschedule_threads(key);
}
void _k_thread_single_abort(struct k_thread *thread)
{
if (thread->fn_abort != NULL) {
thread->fn_abort();
}
if (_is_thread_ready(thread)) {
_remove_thread_from_ready_q(thread);
} else {
if (_is_thread_pending(thread)) {
_unpend_thread(thread);
}
if (_is_thread_timeout_active(thread)) {
_abort_thread_timeout(thread);
}
}
_mark_thread_as_dead(thread);
}
#ifdef CONFIG_MULTITHREADING
void _init_static_threads(void)
{
unsigned int key;
_FOREACH_STATIC_THREAD(thread_data) {
_new_thread(
thread_data->init_stack,
thread_data->init_stack_size,
thread_data->init_entry,
thread_data->init_p1,
thread_data->init_p2,
thread_data->init_p3,
thread_data->init_prio,
thread_data->init_options);
thread_data->thread->init_data = thread_data;
}
_sched_lock();
/*
* Non-legacy static threads may be started immediately or after a
* previously specified delay. Even though the scheduler is locked,
* ticks can still be delivered and processed. Lock interrupts so
* that the countdown until execution begins from the same tick.
*
* Note that static threads defined using the legacy API have a
* delay of K_FOREVER.
*/
key = irq_lock();
_FOREACH_STATIC_THREAD(thread_data) {
if (thread_data->init_delay != K_FOREVER) {
schedule_new_thread(thread_data->thread,
thread_data->init_delay);
}
}
irq_unlock(key);
k_sched_unlock();
}
#endif
void _init_thread_base(struct _thread_base *thread_base, int priority,
u32_t initial_state, unsigned int options)
{
/* k_q_node is initialized upon first insertion in a list */
thread_base->user_options = (u8_t)options;
thread_base->thread_state = (u8_t)initial_state;
thread_base->prio = priority;
thread_base->sched_locked = 0;
/* swap_data does not need to be initialized */
_init_thread_timeout(thread_base);
}
u32_t _k_thread_group_mask_get(struct k_thread *thread)
{
struct _static_thread_data *thread_data = thread->init_data;
return thread_data->init_groups;
}
void _k_thread_group_join(u32_t groups, struct k_thread *thread)
{
struct _static_thread_data *thread_data = thread->init_data;
thread_data->init_groups |= groups;
}
void _k_thread_group_leave(u32_t groups, struct k_thread *thread)
{
struct _static_thread_data *thread_data = thread->init_data;
thread_data->init_groups &= groups;
}