PowLu/TencentOS-tiny
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

support posix

Browse Source 
1. posix pthread.h: pthread, pthread_barrier, pthread_cond, pthread_mutex, , pthread_rwlock, pthread_spin, etc
2. posix semaphore.h: sem_*
3. posix mqueue.h: mq_*
4. posix time.h: timer_*
5. to support posix, add tos_barrier, tos_bitmap, tos_rwlock, tos_stopwatch, change name of k_task_t from a char * pointer to a char array.
6. sample, see examples\posix
7. project, see board\TencentOS_tiny_EVB_MX_Plus\KEIL\posix
This commit is contained in:
daishengdong
2020-02-28 00:11:28 +08:00
parent 7bfc998494
commit 40f55ec57b
84 changed files with 11704 additions and 158 deletions
Download Patch File Download Diff File Expand all files Collapse all files

59
kernel/core/include/tos_task.h
View File

@@ -18,6 +18,11 @@
#ifndef _TOS_TASK_H_
#define _TOS_TASK_H_
__CDECLS_BEGIN
#define K_TASK_NAME_MAX (16u)
#define K_TASK_STK_SIZE_MIN (sizeof(cpu_context_t))
// task state is just a flag, indicating which manager list we are in.
// ready to schedule
@@ -67,43 +72,43 @@ typedef void (*k_task_walker_t)(k_task_t *task);
* task control block
*/
typedef struct k_task_st {
k_stack_t *sp; /**< task stack pointer. This lady always comes first, we count on her in port_s.S for context switch. */
k_stack_t *sp; /**< task stack pointer. This lady always comes first, we count on her in port_s.S for context switch. */
knl_obj_t knl_obj; /**< just for verification, test whether current object is really a task. */
knl_obj_t knl_obj; /**< just for verification, test whether current object is really a task. */
char *name; /**< task name */
k_task_entry_t entry; /**< task entry */
void *arg; /**< argument for task entry */
k_task_state_t state; /**< just state */
k_prio_t prio; /**< just priority */
char name[K_TASK_NAME_MAX]; /**< task name */
k_task_entry_t entry; /**< task entry */
void *arg; /**< argument for task entry */
k_task_state_t state; /**< just state */
k_prio_t prio; /**< just priority */
k_stack_t *stk_base; /**< task stack base address */
size_t stk_size; /**< stack size of the task */
k_stack_t *stk_base; /**< task stack base address */
size_t stk_size; /**< stack size of the task */
#if TOS_CFG_TASK_DYNAMIC_CREATE_EN > 0u
k_list_t dead_list; /**< when a dynamic allocated task destroyed, we hook the task's dead_list to the k_dead_task_list */
k_list_t dead_list; /**< when a dynamic allocated task destroyed, we hook the task's dead_list to the k_dead_task_list */
#endif
k_list_t stat_list; /**< list for hooking us to the k_stat_list */
k_list_t stat_list; /**< list for hooking us to the k_stat_list */
k_tick_t tick_expires; /**< if we are in k_tick_list, how much time will we wait for? */
k_tick_t tick_expires; /**< if we are in k_tick_list, how much time will we wait for? */
k_list_t tick_list; /**< list for hooking us to the k_tick_list */
k_list_t pend_list; /**< when we are ready, our pend_list is in readyqueue; when pend, in a certain pend object's list. */
k_list_t tick_list; /**< list for hooking us to the k_tick_list */
k_list_t pend_list; /**< when we are ready, our pend_list is in readyqueue; when pend, in a certain pend object's list. */
#if TOS_CFG_MUTEX_EN > 0u
k_list_t mutex_own_list; /**< the list hold all the mutex we own.
When we die(tos_task_destroy), we have an obligation to wakeup all the task pending for those mutexs we own;
if not, those pending tasks may never get a chance to wakeup. */
k_prio_t prio_pending; /*< when tos_task_prio_change called, we may be just the owner of a mutex.
to avoid PRIORITY INVERSION, must make sure our priority is higher than any one who is pending for
the mutex we hold. So, if the prio_new of tos_task_prio_change is not appropriate
(may against the principle of PRIORITY INVERSION), we just mark the prio_new here, do the real priority
change in the right time(mutex_old_owner_release) later. */
k_list_t mutex_own_list; /**< the list hold all the mutex we own.
When we die(tos_task_destroy), we have an obligation to wakeup all the task pending for those mutexs we own;
if not, those pending tasks may never get a chance to wakeup. */
k_prio_t prio_pending; /*< when tos_task_prio_change called, we may be just the owner of a mutex.
to avoid PRIORITY INVERSION, must make sure our priority is higher than any one who is pending for
the mutex we hold. So, if the prio_new of tos_task_prio_change is not appropriate
(may against the principle of PRIORITY INVERSION), we just mark the prio_new here, do the real priority
change in the right time(mutex_old_owner_release) later. */
#endif
pend_obj_t *pending_obj; /**< if we are pending, which pend object's list we are in? */
pend_state_t pend_state; /**< why we wakeup from a pend */
pend_obj_t *pending_obj; /**< if we are pending, which pend object's list we are in? */
pend_state_t pend_state; /**< why we wakeup from a pend */
#if TOS_CFG_ROUND_ROBIN_EN > 0u
k_timeslice_t timeslice_reload; /**< if current time slice is used up, use time_slice_reload to reload our time slice */
@@ -406,7 +411,9 @@ __DEBUG__ __STATIC_INLINE__ void task_default_walker(k_task_t *task)
state_str = state_str;
tos_kprintln("tsk name: %s", task->name);
if (task->state == K_TASK_STATE_PENDTIMEOUT_SUSPENDED) {
if (tos_task_curr_task_get() == task) {
state_str = "RUNNING";
} else if (task->state == K_TASK_STATE_PENDTIMEOUT_SUSPENDED) {
state_str = "PENDTIMEOUT_SUSPENDED";
} else if (task->state == K_TASK_STATE_PEND_SUSPENDED) {
state_str = "PEND_SUSPENDED";
@@ -431,5 +438,7 @@ __DEBUG__ __STATIC_INLINE__ void task_default_walker(k_task_t *task)
tos_kprintf("\n");
}
__CDECLS_END
#endif /* _TOS_TASK_H_ */