linux IPC--信号量

SEMOP(2)                                                               Linux Programmer's Manual                                                              SEMOP(2)



NAME
       semop, semtimedop - System V semaphore operations

SYNOPSIS
       #include <sys/types.h>
       #include <sys/ipc.h>
       #include <sys/sem.h>

       int semop(int semid, struct sembuf *sops, size_t nsops);

       int semtimedop(int semid, struct sembuf *sops, size_t nsops,
                      const struct timespec *timeout);

   Feature Test Macro Requirements for glibc (see feature_test_macros(7)):

       semtimedop(): _GNU_SOURCE

DESCRIPTION
       Each semaphore in a System V semaphore set has the following associated values:

           unsigned short  semval;   /* semaphore value */
           unsigned short  semzcnt;  /* # waiting for zero */
           unsigned short  semncnt;  /* # waiting for increase */
           pid_t           sempid;   /* ID of process that did last op */

       semop()  performs  operations  on  selected  semaphores in the set indicated by semid.  Each of the nsops elements in the array pointed to by sops specifies an
       operation to be performed on a single semaphore.  The elements of this structure are of type struct sembuf, containing the following members:

           unsigned short sem_num;  /* semaphore number */
           short          sem_op;   /* semaphore operation */
           short          sem_flg;  /* operation flags */

       Flags recognized in sem_flg are IPC_NOWAIT and SEM_UNDO.  If an operation specifies SEM_UNDO, it will be automatically undone when the process terminates.

       The set of operations contained in sops is performed in array order, and atomically, that is, the operations are performed either as a complete unit, or not at
       all.   The  behavior  of  the  system  call if not all operations can be performed immediately depends on the presence of the IPC_NOWAIT flag in the individual
       sem_flg fields, as noted below.

       Each operation is performed on the sem_num-th semaphore of the semaphore set, where the first semaphore of the set is numbered 0.  There  are  three  types  of
       operation, distinguished by the value of sem_op.

       If  sem_op is a positive integer, the operation adds this value to the semaphore value (semval).  Furthermore, if SEM_UNDO is specified for this operation, the
       system subtracts the value sem_op from the semaphore adjustment (semadj) value for this semaphore.  This operation can always proceed—it never forces a  thread
       to wait.  The calling process must have alter permission on the semaphore set.

       If sem_op is zero, the process must have read permission on the semaphore set.  This is a "wait-for-zero" operation: if semval is zero, the operation can imme‐
       diately proceed.  Otherwise, if IPC_NOWAIT is specified in sem_flg, semop() fails with errno set to EAGAIN (and none of the operations in sops  is  performed).
       Otherwise,  semzcnt  (the count of threads waiting until this semaphore's value becomes zero) is incremented by one and the thread sleeps until one of the fol‐
       lowing occurs:

       ·  semval becomes 0, at which time the value of semzcnt is decremented.

       ·  The semaphore set is removed: semop() fails, with errno set to EIDRM.

       ·  The calling thread catches a signal: the value of semzcnt is decremented and semop() fails, with errno set to EINTR.

       ·  The time limit specified by timeout in a semtimedop() call expires: semop() fails, with errno set to EAGAIN.

       If sem_op is less than zero, the process must have alter permission on the semaphore set.  If semval is greater than or equal to the absolute value of  sem_op,
       the  operation  can  proceed  immediately: the absolute value of sem_op is subtracted from semval, and, if SEM_UNDO is specified for this operation, the system
       adds the absolute value of sem_op to the semaphore adjustment (semadj) value for this semaphore.  If the absolute value of sem_op is greater than  semval,  and
       IPC_NOWAIT is specified in sem_flg, semop() fails, with errno set to EAGAIN (and none of the operations in sops is performed).  Otherwise, semncnt (the counter
       of threads waiting for this semaphore's value to increase) is incremented by one and the thread sleeps until one of the following occurs:

       ·  semval becomes greater than or equal to the absolute value of sem_op: the operation now proceeds, as described above.

       ·  The semaphore set is removed from the system: semop() fails, with errno set to EIDRM.

       ·  The calling thread catches a signal: the value of semncnt is decremented and semop() fails, with errno set to EINTR.

       ·  The time limit specified by timeout in a semtimedop() call expires: the system call fails, with errno set to EAGAIN.

       On successful completion, the sempid value for each semaphore specified in the array pointed to by sops is set to the caller's process ID.   In  addition,  the
       sem_otime is set to the current time.

       semtimedop()  behaves  identically  to  semop()  except  that in those cases where the calling thread would sleep, the duration of that sleep is limited by the
       amount of elapsed time specified by the timespec structure whose address is passed in the timeout argument.  (This sleep interval will be  rounded  up  to  the
       system  clock  granularity,  and kernel scheduling delays mean that the interval may overrun by a small amount.)  If the specified time limit has been reached,
       semtimedop() fails with errno set to EAGAIN (and none of the operations in sops is performed).  If the timeout argument  is  NULL,  then  semtimedop()  behaves
       exactly like semop().

RETURN VALUE
       If successful, semop() and semtimedop() return 0; otherwise they return -1 with errno indicating the error.

ERRORS
       On failure, errno is set to one of the following:

       E2BIG  The argument nsops is greater than SEMOPM, the maximum number of operations allowed per system call.

       EACCES The  calling  process does not have the permissions required to perform the specified semaphore operations, and does not have the CAP_IPC_OWNER capabil‐
              ity.

       EAGAIN An operation could not proceed immediately and either IPC_NOWAIT was specified in sem_flg or the time limit specified in timeout expired.

       EFAULT An address specified in either the sops or the timeout argument isn't accessible.

       EFBIG  For some operation the value of sem_num is less than 0 or greater than or equal to the number of semaphores in the set.

       EIDRM  The semaphore set was removed.

       EINTR  While blocked in this system call, the thread caught a signal; see signal(7).

       EINVAL The semaphore set doesn't exist, or semid is less than zero, or nsops has a nonpositive value.

       ENOMEM The sem_flg of some operation specified SEM_UNDO and the system does not have enough memory to allocate the undo structure.

       ERANGE For some operation sem_op+semval is greater than SEMVMX, the implementation dependent maximum value for semval.

VERSIONS
       semtimedop() first appeared in Linux 2.5.52, and was subsequently backported into kernel 2.4.22.  Glibc support for  semtimedop()  first  appeared  in  version
       2.3.3.

CONFORMING TO
       SVr4, POSIX.1-2001.

NOTES
       The inclusion of <sys/types.h> and <sys/ipc.h> isn't required on Linux or by any version of POSIX.  However, some old implementations required the inclusion of
       these header files, and the SVID also documented their inclusion.  Applications intended to be portable to such old systems may need to  include  these  header
       files.

       The sem_undo structures of a process aren't inherited by the child produced by fork(2), but they are inherited across an execve(2) system call.

       semop() is never automatically restarted after being interrupted by a signal handler, regardless of the setting of the SA_RESTART flag when establishing a sig‐
       nal handler.

       A semaphore adjustment (semadj) value is a per-process, per-semaphore integer that is the negated sum of all operations performed on a semaphore specifying the
       SEM_UNDO flag.  Each process has a list of semadj values—one value for each semaphore on which it has operated using SEM_UNDO.  When a process terminates, each
       of its per-semaphore semadj values is added to the corresponding semaphore, thus undoing the effect of that process's operations on the semaphore (but see BUGS
       below).   When  a  semaphore's  value  is  directly  set  using the SETVAL or SETALL request to semctl(2), the corresponding semadj values in all processes are
       cleared.  The clone() CLONE_SYSVSEM flag allows more than one process to share a semadj list; see clone(2) for details.

       The semval, sempid, semzcnt, and semnct values for a semaphore can all be retrieved using appropriate semctl(2) calls.

   Semaphore limits
       The following limits on semaphore set resources affect the semop() call:

       SEMOPM Maximum number of operations allowed for one semop() call (32) (on Linux, this limit can be read and modified via  the  third  field  of  /proc/sys/ker‐
              nel/sem).

       SEMVMX Maximum allowable value for semval: implementation dependent (32767).

       The  implementation  has  no intrinsic limits for the adjust on exit maximum value (SEMAEM), the system wide maximum number of undo structures (SEMMNU) and the
       per-process maximum number of undo entries system parameters.

BUGS
       When a process terminates, its set of associated semadj structures is used to undo the effect of all of the semaphore operations it performed with the SEM_UNDO
       flag.   This  raises  a difficulty: if one (or more) of these semaphore adjustments would result in an attempt to decrease a semaphore's value below zero, what
       should an implementation do?  One possible approach would be to block until all the semaphore adjustments could be  performed.   This  is  however  undesirable
       since  it could force process termination to block for arbitrarily long periods.  Another possibility is that such semaphore adjustments could be ignored alto‐
       gether (somewhat analogously to failing when IPC_NOWAIT is specified for a semaphore operation).  Linux adopts a third approach: decreasing the semaphore value
       as far as possible (i.e., to zero) and allowing process termination to proceed immediately.

       In kernels 2.6.x, x <= 10, there is a bug that in some circumstances prevents a thread that is waiting for a semaphore value to become zero from being woken up
       when the value does actually become zero.  This bug is fixed in kernel 2.6.11.

EXAMPLE
       The following code segment uses semop() to atomically wait for the value of semaphore 0 to become zero, and then increment the semaphore value by one.

           struct sembuf sops[2];
           int semid;

           /* Code to set semid omitted */

           sops[0].sem_num = 0;        /* Operate on semaphore 0 */
           sops[0].sem_op = 0;         /* Wait for value to equal 0 */
           sops[0].sem_flg = 0;

           sops[1].sem_num = 0;        /* Operate on semaphore 0 */
           sops[1].sem_op = 1;         /* Increment value by one */
           sops[1].sem_flg = 0;

           if (semop(semid, sops, 2) == -1) {
               perror("semop");
               exit(EXIT_FAILURE);
           }

SEE ALSO
       clone(2), semctl(2), semget(2), sigaction(2), capabilities(7), sem_overview(7), svipc(7), time(7)

COLOPHON
       This page is part of release 3.74 of the Linux man-pages project.  A description of the project, information about reporting bugs, and the  latest  version  of

       this page, can be found at http://www.kernel.org/doc/man-pages/.

example:

client.c

#include <sys/types.h>
#include <sys/ipc.h>
#include <sys/sem.h>
#include <sys/shm.h>
#include <sys/stat.h>
#include <stdlib.h>
#include <stdio.h>
#include <string.h>

#define SEM_KEY 4001
#define SHM_KEY 5678



int semid,shmid;
int *shmptr;
struct sembuf sembuf1;
union semun {
        int val;
};

int main(void)
{
        //create a shm
        if((shmid = shmget(SHM_KEY,0,0666))<0){
                printf("create shm error\n");
                return 1;
        }
        if((shmptr =shmat(shmid,NULL,0))==(int*)-1){
                printf("shmat error\n");
                return 1;
        }
        semid = semget(SEM_KEY,2,0666);
        union semun semun1;

        //
        while(1){
                sembuf1.sem_num=1;//这里指向第2个信号量（sem_num=1）
                sembuf1.sem_op=-1;//操作是-1，但是第二个信号量初始值为1，所以下面不会阻塞
                sembuf1.sem_flg=SEM_UNDO;

                semop(semid,&sembuf1,1);//继续
                scanf("%d",shmptr);  //用户客户端输入数据

                sembuf1.sem_num=0;//这里指向第一个信号量
                sembuf1.sem_op=1;//操作加1
                sembuf1.sem_flg=SEM_UNDO;

                semop(semid,&sembuf1,1);//执行+1后，我们发现，服务器阻塞正是由于第一个信号量为0，无法减一，而现在客户端先为其加1，那服务器就绪！客户端继续循环，发现第二个信号量已经减为0，则阻塞了，我们回到服务器
        }
    shmdt(shmptr);
        return 0;
}

server.c

#include <sys/types.h>
#include <sys/ipc.h>
#include <sys/stat.h>
#include <errno.h>
#include <stdlib.h>
#include <stdio.h>
#include <string.h>
#include <sys/sem.h>
#include <sys/shm.h>

#define SEM_KEY 4001
#define SHM_KEY 5678



int semid,shmid;
int *shmptr;
struct sembuf sembuf1[2];
union semun {
        int val;
};

int main(void)
{
        //create a shm  
        if((shmid = shmget(SHM_KEY,sizeof(int),IPC_CREAT|0666))<0){
                printf("create shm error\n");
                return 1;
        }
    if((shmptr = shmat(shmid, (int *)0, 0)) == (int *)-1){
                printf("shmat error:%s\n",strerror(errno));
                return 1;
        }
        semid = semget(SEM_KEY,2,IPC_CREAT|0666);//这里是创建一个semid，并且有两个信号量
//    if(semid == -1)
//        printf("semid err is: %s\n",strerror(errno));
        union semun semun1;//下面这四行就是初始化那两个信号量，一个val=0,另一个val=1
        semun1.val=0;
        semctl(semid,0,SETVAL,semun1);
        semun1.val=1;
        semctl(semid,1,SETVAL,semun1);

        //
        while(1){
                sembuf1[0].sem_num=0;//sem_num=0指的是下面操作指向第一个信号量，上面设置可知其                                                       //val=0
                sembuf1[0].sem_op=-1; //这里定义操作是-1，但是他已经为0 了，所以会阻塞，48行就产生                                                      //了这个阻塞
                sembuf1[0].sem_flg=SEM_UNDO;

                semop(semid,sembuf1,1);//服务器在这里会阻塞,下面看下client
                printf("the NUM:%d\n",*(shmptr));//输出结果

                sembuf1[0].sem_num=1;//这里让客户端再次就绪，就这样循环
                sembuf1[0].sem_op=1;
                sembuf1[0].sem_flg=SEM_UNDO;

                semop(semid,sembuf1,1);
        }
    semctl(semid, 0, IPC_RMID);
    shmdt(shmptr);
    shmctl(shmid, IPC_RMID, 0);
        return 0;
}