linux 管道读写规则

转自：http://blog.csdn.net/nodeathphoenix/article/details/23284157

一，管道读写规则

当没有数据可读时

• O_NONBLOCK disable：read调用阻塞，即进程暂停执行，一直等到有数据来到为止。
• O_NONBLOCK enable：read调用返回-1，errno值为EAGAIN。

当管道满的时候

• O_NONBLOCK disable： write调用阻塞，直到有进程读走数据
• O_NONBLOCK enable：调用返回-1，errno值为EAGAIN

如果所有管道写端对应的文件描述符被关闭，则read返回0

如果所有管道读端对应的文件描述符被关闭，则write操作会产生信号SIGPIPE

当要写入的数据量不大于PIPE_BUF时，linux将保证写入的原子性。

当要写入的数据量大于PIPE_BUF时，linux将不再保证写入的原子性。

Detail 参见：

http://man7.org/linux/man-pages/man7/pipe.7.html

二，验证示例

示例一：O_NONBLOCK disable：read调用阻塞，即进程暂停执行，一直等到有数据来到为止。

#include <stdio.h>#include <unistd.h>#include <stdlib.h>#include <fcntl.h> int main(void){    int fds[2];    if(pipe(fds) == -1){        perror("pipe error");        exit(EXIT_FAILURE);    }    pid_t pid;    pid = fork();    if(pid == -1){        perror("fork error");        exit(EXIT_FAILURE);    }    if(pid == 0){        close(fds[0]);//子进程关闭读端        sleep(10);        write(fds[1],"hello",5);        exit(EXIT_SUCCESS);    }    close(fds[1]);//父进程关闭写端    char buf[10] = {0};    read(fds[0],buf,10);    printf("receive datas = %s\n",buf);    return 0;}

结果：

说明：管道创建时默认打开了文件描述符，且默认是阻塞（block）模式打开

所以这里，我们让子进程先睡眠10s，父进程因为没有数据从管道中读出，被阻塞了，直到子进程睡眠结束，向管道中写入数据后，父进程才读到数据

示例二：O_NONBLOCK enable：read调用返回-1，errno值为EAGAIN。

#include <stdio.h>#include <unistd.h>#include <stdlib.h>#include <fcntl.h> int main(void){    int fds[2];    if(pipe(fds) == -1){        perror("pipe error");        exit(EXIT_FAILURE);    }    pid_t pid;    pid = fork();    if(pid == -1){        perror("fork error");        exit(EXIT_FAILURE);    }    if(pid == 0){        close(fds[0]);//子进程关闭读端        sleep(10);        write(fds[1],"hello",5);        exit(EXIT_SUCCESS);    }    close(fds[1]);//父进程关闭写端    char buf[10] = {0};    int flags = fcntl(fds[0], F_GETFL);//先获取原先的flags    fcntl(fds[0],F_SETFL,flags | O_NONBLOCK);//设置fd为阻塞模式    int ret;    ret = read(fds[0],buf,10);    if(ret == -1){        perror("read error");        exit(EXIT_FAILURE);    }    printf("receive datas = %s\n",buf);    return 0;}

结果：

示例三：如果所有管道写端对应的文件描述符被关闭，则read返回0

#include <stdio.h>#include <unistd.h>#include <stdlib.h>#include <fcntl.h> int main(void){    int fds[2];    if(pipe(fds) == -1){        perror("pipe error");        exit(EXIT_FAILURE);    }    pid_t pid;    pid = fork();    if(pid == -1){        perror("fork error");        exit(EXIT_FAILURE);    }    if(pid == 0){        close(fds[1]);//子进程关闭写端        exit(EXIT_SUCCESS);    }    close(fds[1]);//父进程关闭写端    char buf[10] = {0};    int ret;    ret = read(fds[0],buf,10);    printf("ret = %d\n", ret);    return 0;}

结果：

可知确实返回0，表示读到了文件末尾，并不表示出错

示例四：如果所有管道读端对应的文件描述符被关闭，则write操作会产生信号SIGPIPE

#include <stdio.h>#include <unistd.h>#include <stdlib.h>#include <fcntl.h>#include <signal.h>void sighandler(int signo);int main(void){    int fds[2];    if(signal(SIGPIPE,sighandler) == SIG_ERR)    {        perror("signal error");        exit(EXIT_FAILURE);    }    if(pipe(fds) == -1){        perror("pipe error");        exit(EXIT_FAILURE);    }    pid_t pid;    pid = fork();    if(pid == -1){        perror("fork error");        exit(EXIT_FAILURE);    }    if(pid == 0){        close(fds[0]);//子进程关闭读端        exit(EXIT_SUCCESS);    }    close(fds[0]);//父进程关闭读端    sleep(1);//确保子进程也将读端关闭    int ret;    ret = write(fds[1],"hello",5);    if(ret == -1){        printf("write error\n");    }    return 0;}void sighandler(int signo){    printf("catch a SIGPIPE signal and signum = %d\n",signo);}

结果：

可知当所有读端都关闭时，write时确实产生SIGPIPE信号

示例五：O_NONBLOCK disable： write调用阻塞，直到有进程读走数据

#include <stdio.h>#include <unistd.h>#include <stdlib.h>#include <fcntl.h> int main(void){    int fds[2];    if(pipe(fds) == -1){        perror("pipe error");        exit(EXIT_FAILURE);    }    int ret;    int count = 0;    while(1){        ret = write(fds[1],"A",1);//fds[1]默认是阻塞模式        if(ret == -1){            perror("write error");            break;        }        count++;    }    return 0;}

结果：

说明：fd打开时默认是阻塞模式，当pipe缓冲区满时，write操作确实阻塞了，等待其他进程将数据从管道中取走

示例六：O_NONBLOCK enable：调用返回-1，errno值为EAGAIN

#include <stdio.h>#include <unistd.h>#include <stdlib.h>#include <fcntl.h> int main(void){    int fds[2];    if(pipe(fds) == -1){        perror("pipe error");        exit(EXIT_FAILURE);    }    int ret;    int count = 0;    int flags = fcntl(fds[1],F_GETFL);    fcntl(fds[1],F_SETFL,flags|O_NONBLOCK);    while(1){        ret = write(fds[1],"A",1);//fds[1]默认是阻塞模式        if(ret == -1){            perror("write error");            break;        }        count++;    }    printf("the pipe capcity is = %d\n",count);    return 0;}

结果：

可知也出现EGIN错误，管道容量是65536字节

man 7 pipe说明：

Pipe capacity

       A pipe has a limited capacity.  If the pipe is full, then a write(2)       will block or fail, depending on whether the O_NONBLOCK flag is set       (see below).  Different implementations have different limits for the       pipe capacity.  Applications should not rely on a particular       capacity: an application should be designed so that a reading process       consumes data as soon as it is available, so that a writing process       does not remain blocked.       In Linux versions before 2.6.11, the capacity of a pipe was the same       as the system page size (e.g., 4096 bytes on i386).  Since Linux       2.6.11, the pipe capacity is 65536 bytes.

三，管道写与PIPE_BUF关系

man帮助说明：

PIPE_BUF

       POSIX.1-2001 says that write(2)s of less than PIPE_BUF bytes must be       atomic: the output data is written to the pipe as a contiguous       sequence.  Writes of more than PIPE_BUF bytes may be nonatomic: the       kernel may interleave the data with data written by other processes.       POSIX.1-2001 requires PIPE_BUF to be at least 512 bytes.  (On Linux,       PIPE_BUF is 4096 bytes.)  The precise semantics depend on whether the       file descriptor is nonblocking (O_NONBLOCK), whether there are       multiple writers to the pipe, and on n, the number of bytes to be       written:       O_NONBLOCK disabled, n <= PIPE_BUF              All n bytes are written atomically; write(2) may block if              there is not room for n bytes to be written immediately       阻塞模式时且n<PIPE_BUF:写入具有原子性，如果没有足够的空间供n个字节全部写入，则阻塞直到有足够空间将n个字节全部写入管道              O_NONBLOCK enabled, n <= PIPE_BUF              If there is room to write n bytes to the pipe, then write(2)              succeeds immediately, writing all n bytes; otherwise write(2)              fails, with errno set to EAGAIN.      非阻塞模式时且n<PIPE_BUF：写入具有原子性，立即全部成功写入，否则一个都不写入，返回错误       O_NONBLOCK disabled, n > PIPE_BUF              The write is nonatomic: the data given to write(2) may be              interleaved with write(2)s by other process; the write(2)              blocks until n bytes have been written.      阻塞模式时且n>PIPE_BUF：不具有原子性，可能中间有其他进程穿插写入，直到将n字节全部写入才返回，否则阻塞等待写入       O_NONBLOCK enabled, n > PIPE_BUF              If the pipe is full, then write(2) fails, with errno set to              EAGAIN.  Otherwise, from 1 to n bytes may be written (i.e., a              "partial write" may occur; the caller should check the return              value from write(2) to see how many bytes were actually              written), and these bytes may be interleaved with writes by              other processes.

   非阻塞模式时且N>PIPE_BUF：如果管道满的，则立即失败，一个都不写入，返回错误，如果不满，则返回写入的字节数为1~n，即部分写入，写入时可能有其他进程穿插写入

• 当要写入的数据量不大于PIPE_BUF时，linux将保证写入的原子性。
• 当要写入的数据量大于PIPE_BUF时，linux将不再保证写入的原子性。

注：管道容量不一定等于PIPE_BUF

示例：当写入数据大于PIPE_BUF时

#include <stdio.h>#include <stdlib.h>#include <string.h>#include <unistd.h>#include <sys/types.h>#include <errno.h>#include <fcntl.h>#define ERR_EXIT(m) \        do \        { \                perror(m); \                exit(EXIT_FAILURE); \        } while(0)#define TEST_SIZE 68*1024int main(void){    char a[TEST_SIZE];    char b[TEST_SIZE];    char c[TEST_SIZE];    memset(a, 'A', sizeof(a));    memset(b, 'B', sizeof(b));    memset(c, 'C', sizeof(c));    int pipefd[2];    int ret = pipe(pipefd);    if (ret == -1)        ERR_EXIT("pipe error");    pid_t pid;    pid = fork();    if (pid == 0)//第一个子进程    {        close(pipefd[0]);        ret = write(pipefd[1], a, sizeof(a));        printf("apid=%d write %d bytes to pipe\n", getpid(), ret);        exit(0);    }    pid = fork();        if (pid == 0)//第二个子进程    {        close(pipefd[0]);        ret = write(pipefd[1], b, sizeof(b));        printf("bpid=%d write %d bytes to pipe\n", getpid(), ret);        exit(0);    }    pid = fork();        if (pid == 0)//第三个子进程    {        close(pipefd[0]);        ret = write(pipefd[1], c, sizeof(c));        printf("bpid=%d write %d bytes to pipe\n", getpid(), ret);        exit(0);    }    close(pipefd[1]);        sleep(1);    int fd = open("test.txt", O_WRONLY | O_CREAT | O_TRUNC, 0644);    char buf[1024*4] = {0};    int n = 1;    while (1)    {        ret = read(pipefd[0], buf, sizeof(buf));        if (ret == 0)            break;        printf("n=%02d pid=%d read %d bytes from pipe buf[4095]=%c\n", n++, getpid(), ret, buf[4095]);        write(fd, buf, ret);    }    return 0;    }

结果：

可见各子进程间出现穿插写入，并没保证原子性写入，且父进程在子进程编写时边读。