Name

semop - semaphore operations

Library

libc.lib

Synopsis

  #include <sys/types.h>
  #include <sys/ipc.h>
  #include <sys/sem.h>
  int semop (int semid, struct sembuf *array, size_t nops);

Return values

The semop function returns 0 if successful; otherwise the value -1 is returned and errno is set to indicate the error.

Detailed description

The semop system call atomically performs the array of operations indicated by array on the semaphore set indicated by semid. The length of array is indicated by nops. Each operation is encoded in a struct sembuf , which is defined as follows:
struct sembuf {
        u_short sem_num;        /* semaphore # */
        short   sem_op;         /* semaphore operation */
        short   sem_flg;        /* operation flags */
};

         

For each element in array, sem_op and sem_flg determine an operation to be performed on semaphore number sem_num in the set. The values SEM_UNDO and IPC_NOWAIT may be OR ’ed into the sem_flg member in order to modify the behavior of the given operation.

The operation performed depends as follows on the value of sem_op:

For each semaphore a process has in use, an "adjust on exit" value is maintained, as alluded to earlier. When a process exits, either voluntarily or involuntarily, the adjust on exit value for each semaphore is added to the semaphore’s value. This can be used to insure that a resource is released if a process terminates unexpectedly.


Examples

#include <sys/ipc.h>
#include <sys/sem.h>
#include <stdio.h>

         
void File_Update(char* path, int val)
{
    struct sembuf sem_op;
    FILE* fp;
    /*
     * Wait on the semaphore till the value is non-negative.
     */
    sem_op.sem_num = 0;
    sem_op.sem_op = -1;
    sem_op.sem_flg = 0;
    semop(sem_set_id, &sem_op, 1);
    /*
     * If we are here, then We have locked the semaphore,
     * and are assured exclusive access to file.
     * We can now manipulate the file
     */
    fp = fopen(path, "w");
    if (fp) {
        fprintf(fp, "%d\n", val);
        fclose(fp);
    }
    /*
     * Increase the value of the semaphore by 1 so that others blocked on
     * this semaphore get awakened.
     */
    sem_op.sem_num = 0;
    sem_op.sem_op = 1;
    sem_op.sem_flg = 0;
    semop(sem_set_id, &sem_op, 1);
}

         


Errors

The semop system call will fail if:
[EINVAL]
  No semaphore set corresponds to semid, or the process would exceed the system-defined limit for the number of per-process SEM_UNDO structures.
[EACCES]
  Permission denied due to mismatch between operation and mode of semaphore set.
[EAGAIN]
  The semaphore’s value would have resulted in the process being put to sleep and IPC_NOWAIT was specified.
[E2BIG]
  Too many operations were specified. [SEMOPM]
[EFBIG]
  sem_num was not in the range of valid semaphores for the set.
[EIDRM]
  The semaphore set was removed from the system.
[ENOSPC]
  The system SEM_UNDO pool [SEMMNU] is full.
[ERANGE]
  The requested operation would cause either the semaphore’s current value [SEMVMX] or its adjust on exit value [SEMAEM] to exceed the system-imposed limits.

See also

semctl, semget,

Bugs

The semop system call may block waiting for memory even if IPC_NOWAIT was specified.


Feedback

For additional information or queries on this page send feedback

© 2005-2007 Nokia

Top