pthread_cond_wait()的使用方法

13 篇文章 0 订阅
3 篇文章 0 订阅


pthread_cond_wait

  编辑
条件变量是利用线程间共享的 全局变量进行同步的一种机制,主要包括两个动作:一个线程等待"条件变量的条件成立"而挂起;另一个线程使"条件成立"(给出条件成立信号)。为了防止竞争, 条件变量的使用总是和一个 互斥锁结合在一起。
中文名
多线程的 条件变量
外文名
pthread_cond_wait
动作一
条件变量的条件成立
动作二
"条件成立"(给出条件成立信号)
目    的
为了防止竞争
搭    配
和一个 互斥锁结合在一起

简介

编辑
多线程的 条件变量 [1]  

创建和注销

编辑
条件变量和 互斥锁一样,都有静态动态两种创建方式,静态方式使用PTHREAD_COND_INITIALIZER 常量,如下:
pthread_cond_t cond=PTHREAD_COND_INITIALIZER
动态方式调用 pthread_cond_init()函数,API定义如下:
int pthread_cond_init(pthread_cond_t *cond, pthread_condattr_t *cond_attr)
尽管POSIX标准中为 条件变量定义了属性,但在LinuxThreads中没有实现,因此cond_attr值通常为NULL,且被忽略。
注销一个条件变量需要调用pthread_cond_destroy(),只有在没有线程在该条件变量上等待的时候才能注销这个条件变量,否则返回EBUSY。因为Linux实现的条件变量没有分配什么资源,所以注销动作只包括检查是否有等待线程。API定义如下:
int pthread_cond_destroy(pthread_cond_t *cond)

等待和激发

编辑
int pthread_cond_wait(pthread_cond_t *cond, pthread_mutex_t *mutex)
int pthread_cond_timedwait(pthread_cond_t *cond, pthread_mutex_t *mutex, const struct timespec *abstime)
等待条件有两种方式:条件等待pthread_cond_wait()和计时等待pthread_cond_timedwait(),其中计时等待方式如果在给定时刻前条件没有满足,则返回ETIMEOUT,结束等待,其中abstime以与time() 系统调用相同意义的绝对时间形式出现,0表示 格林尼治时间1970年1月1日0时0分0秒。
无论哪种等待方式,都必须和一个 互斥锁配合,以防止多个线程同时请求pthread_cond_wait()(或pthread_cond_timedwait(),下同)的 竞争条件(Race Condition)。mutex互斥锁必须是普通锁(PTHREAD_MUTEX_TIMED_NP)或者适应锁(PTHREAD_MUTEX_ADAPTIVE_NP),且在调用pthread_cond_wait()前必须由本 线程加锁(pthread_mutex_lock()),而在更新条件 等待队列以前,mutex保持锁定状态,并在线程挂起进入等待前解锁。在条件满足从而离开pthread_cond_wait()之前,mutex将被重新加锁,以与进入pthread_cond_wait()前的加锁动作对应。
激发条件有两种形式,pthread_cond_signal()激活一个等待该条件的线程,存在多个等待线程时按入队顺序激活其中一个;而pthread_cond_broadcast()则激活所有等待线程。
现在来看一段典型的应用:看注释即可。
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
#include<pthread.h>
#include<unistd.h>
#include<stdio.h>
#include<string.h>
#include<stdlib.h>
 
static  pthread_mutex_t mtx=PTHREAD_MUTEX_INITIALIZER;
static  pthread_cond_t cond=PTHREAD_COND_INITIALIZER;
 
struct  node {
     int  n_number;
     struct  node *n_next;
} *head=NULL;  /*[thread_func]*/
 
/*释放节点内存*/
static  void  cleanup_handler( void *arg) {
     printf ( "Clean up handler of second thread.\n" );
     free (arg);
     ( void )pthread_mutex_unlock(&mtx);
}
 
static  void  *thread_func( void  *arg) {
     struct  node*p=NULL;
     pthread_cleanup_push(cleanup_handler,p);
 
     pthread_mutex_lock(&mtx);
     //这个mutex_lock主要是用来保护wait等待临界时期的情况,
     //当在wait为放入队列时,这时,已经存在Head条件等待激活
     //的条件,此时可能会漏掉这种处理
     //这个while要特别说明一下,单个pthread_cond_wait功能很完善,
     //为何这里要有一个while(head==NULL)呢?因为pthread_cond_wait
     //里的线程可能会被意外唤醒,如果这个时候head==NULL,
     //则不是我们想要的情况。这个时候,
     //应该让线程继续进入pthread_cond_wait
     while (1) {
         while (head==NULL) {
             pthread_cond_wait(&cond,&mtx);
         }
         //pthread_cond_wait会先解除之前的pthread_mutex_lock锁定的mtx,
         //然后阻塞在等待队列里休眠,直到再次被唤醒
         //(大多数情况下是等待的条件成立而被唤醒,唤醒后,
         //该进程会先锁定先pthread_mutex_lock(&mtx);,
         //再读取资源用这个流程是比较清楚的
         /*block-->unlock-->wait()return-->lock*/
         p=head;
         head=head->n_next;
         printf ( "Got%dfromfrontofqueue\n" ,p->n_number);
         free (p);
     }
     pthread_mutex_unlock(&mtx); //临界区数据操作完毕,释放互斥锁
 
     pthread_cleanup_pop(0);
     return  0;
}
 
int  main( void ) {
     pthread_t tid;
     int  i;
     struct  node *p;
     pthread_create(&tid,NULL,thread_func,NULL);
     //子线程会一直等待资源,类似生产者和消费者,
     //但是这里的消费者可以是多个消费者,
     //而不仅仅支持普通的单个消费者,这个模型虽然简单,
     //但是很强大
     for (i=0;i<10;i++) {
         p=( struct  node*) malloc ( sizeof ( struct  node));
         p->n_number=i;
         pthread_mutex_lock(&mtx); //需要操作head这个临界资源,先加锁,
         p->n_next=head;
         head=p;
         pthread_cond_signal(&cond);
         pthread_mutex_unlock(&mtx); //解锁
         sleep(1);
     }
     printf ( "thread1wannaendthecancelthread2.\n" );
     pthread_cancel(tid);
     //关于pthread_cancel,有一点额外的说明,它是从外部终止子线程,
     //子线程会在最近的取消点,退出线程,而在我们的代码里,最近的
     //取消点肯定就是pthread_cond_wait()了。
     pthread_join(tid,NULL);
     printf ( "Alldone--exiting\n" );
     return  0;
}
参考资料

  • 0
    点赞
  • 0
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
`pthread_cond_timedwait` 函数用于在指定的时间内等待条件变量满足。以下是 `pthread_cond_timedwait` 函数的使用方法: ```c int pthread_cond_timedwait(pthread_cond_t *restrict cond, pthread_mutex_t *restrict mutex, const struct timespec *restrict abstime); ``` 参数说明: - `cond`:指向条件变量的指针。 - `mutex`:指向互斥锁的指针,用于保护共享资源。 - `abstime`:指向 `struct timespec` 结构体的指针,表示等待的绝对时间。 函数返回值: - 成功时返回 0。 - 如果等待超时,则返回 ETIMEDOUT 错误。 - 其他错误情况下返回相应的错误码。 使用 `pthread_cond_timedwait` 函数的一般步骤如下: 1. 在进入等待之前,获取互斥锁。 2. 设置等待的绝对时间。 3. 调用 `pthread_cond_timedwait` 函数,传入条件变量、互斥锁和绝对时间作为参数。 4. 根据返回值判断等待的结果,如果返回 0,则条件满足可以继续执行;如果返回 ETIMEDOUT,则表示等待超时。 5. 在条件满足或超时后,释放互斥锁。 下面是一个示例代码: ```c #include <pthread.h> #include <stdio.h> #include <stdlib.h> #include <unistd.h> pthread_mutex_t mutex = PTHREAD_MUTEX_INITIALIZER; pthread_cond_t cond = PTHREAD_COND_INITIALIZER; void* thread_func(void* arg) { sleep(5); // 模拟一些操作 pthread_mutex_lock(&mutex); pthread_cond_signal(&cond); // 发送信号通知等待的线程 pthread_mutex_unlock(&mutex); return NULL; } int main() { pthread_t thread; struct timespec timeout; // 获取当前时间并设置等待时间为 3 秒 clock_gettime(CLOCK_REALTIME, &timeout); timeout.tv_sec += 3; pthread_create(&thread, NULL, thread_func, NULL); pthread_mutex_lock(&mutex); int result = pthread_cond_timedwait(&cond, &mutex, &timeout); if (result == 0) { printf("条件满足,继续执行\n"); } else if (result == ETIMEDOUT) { printf("等待超时\n"); } else { printf("等待出错\n"); } pthread_mutex_unlock(&mutex); pthread_join(thread, NULL); return 0; } ``` 在上述示例中,我们创建了一个线程,在 5 秒后发送信号通知等待的线程。等待的线程在等待超过 3 秒后会自动返回 ETIMEDOUT 错误。 希望以上信息能够帮助到你。如果你还有其他问题,请随时提问。

“相关推荐”对你有帮助么?

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值