.\" Copyright 1993 Giorgio Ciucci (giorgio@crcc.it) .\" .\" Permission is granted to make and distribute verbatim copies of this .\" manual provided the copyright notice and this permission notice are .\" preserved on all copies. .\" .\" Permission is granted to copy and distribute modified versions of this .\" manual under the conditions for verbatim copying, provided that the .\" entire resulting derived work is distributed under the terms of a .\" permission notice identical to this one. .\" .\" Since the Linux kernel and libraries are constantly changing, this .\" manual page may be incorrect or out-of-date. The author(s) assume no .\" responsibility for errors or omissions, or for damages resulting from .\" the use of the information contained herein. The author(s) may not .\" have taken the same level of care in the production of this manual, .\" which is licensed free of charge, as they might when working .\" professionally. .\" .\" Formatted or processed versions of this manual, if unaccompanied by .\" the source, must acknowledge the copyright and authors of this work. .\" .\" Modified 1996-10-22, Eric S. Raymond .\" Modified 2002-01-08, Michael Kerrisk .\" Modified 2003-04-28, Ernie Petrides .\" Modified 2004-05-27, Michael Kerrisk .\" Modified, 11 Nov 2004, Michael Kerrisk .\" Language and formatting clean-ups .\" Added notes on /proc files .\" 2005-04-08, mtk, Noted kernel version numbers for semtimedop() .\" .TH SEMOP 2 2004-11-10 "Linux 2.6.9" "Linux Programmer's Manual" .SH NAME semop, semtimedop \- semaphore operations .SH SYNOPSIS .nf .B #include .B #include .B #include .fi .sp .BI "int semop(int " semid , .BI "struct sembuf *" sops , .BI "unsigned " nsops ); .sp .BI "int semtimedop(int " semid , .BI "struct sembuf *" sops , .BI "unsigned " nsops , .BI "struct timespec *" timeout ); .SH DESCRIPTION Each semaphore in a semaphore set has the following associated values: .sp .in +4n .nf unsigned short semval; /* semaphore value */ unsigned short semzcnt; /* # waiting for zero */ unsigned short semncnt; /* # waiting for increase */ pid_t sempid; /* process that did last op */ .sp .in -4n .fi .BR semop () performs operations on selected semaphores in the set indicated by .IR semid . Each of the .I nsops elements in the array pointed to by .I sops specifies an operation to be performed on a single semaphore. The elements of this structure are of type .IR "struct sembuf" , containing the following members: .sp .in +4n .nf unsigned short sem_num; /* semaphore number */ short sem_op; /* semaphore operation */ short sem_flg; /* operation flags */ .sp .in -4n .fi Flags recognized in .I sem_flg are .B IPC_NOWAIT and .BR SEM_UNDO . If an operation specifies .BR SEM_UNDO , it will be automatically undone when the process terminates. .PP The set of operations contained in .I sops is performed .IR atomically , that is, the operations are performed at the same time, and only if they can all be simultaneously performed. The behaviour of the system call if not all operations can be performed immediately depends on the presence of the .B IPC_NOWAIT flag in the individual .I sem_flg fields, as noted below. Each operation is performed on the .IR 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 .IR sem_op . .PP If .I sem_op is a positive integer, the operation adds this value to the semaphore value .RI ( semval ). Furthermore, if .B SEM_UNDO is specified for this operation, the system updates the process undo count .RI ( semadj ) for this semaphore. This operation can always proceed \(em it never forces a process to wait. The calling process must have alter permission on the semaphore set. .PP If .I sem_op is zero, the process must have read permission on the semaphore set. This is a "wait-for-zero" operation: if .I semval is zero, the operation can immediately proceed. Otherwise, if .B IPC_NOWAIT is specified in .IR sem_flg , .BR semop () fails with .I errno set to .B EAGAIN (and none of the operations in .I sops is performed). Otherwise .I semzcnt (the count of processes waiting until this semaphore's value becomes zero) is incremented by one and the process sleeps until one of the following occurs: .IP \(bu .I semval becomes 0, at which time the value of .I semzcnt is decremented. .IP \(bu The semaphore set is removed: .BR semop () fails, with .I errno set to .BR EIDRM . .IP \(bu The calling process catches a signal: the value of .I semzcnt is decremented and .BR semop () fails, with .I errno set to .BR EINTR . .IP \(bu The time limit specified by .I timeout in a .BR semtimedop () call expires: .BR semop () fails, with .I errno set to .BR EAGAIN . .PP If .I sem_op is less than zero, the process must have alter permission on the semaphore set. If .I semval is greater than or equal to the absolute value of .IR sem_op , the operation can proceed immediately: the absolute value of .I sem_op is subtracted from .IR semval , and, if .B SEM_UNDO is specified for this operation, the system updates the process undo count .RI ( semadj ) for this semaphore. If the absolute value of .I sem_op is greater than .IR semval , and .B IPC_NOWAIT is specified in .IR sem_flg , .BR semop () fails, with .I errno set to .B EAGAIN (and none of the operations in .I sops is performed). Otherwise .I semncnt (the counter of processes waiting for this semaphore's value to increase) is incremented by one and the process sleeps until one of the following occurs: .IP \(bu .I semval becomes greater than or equal to the absolute value of .IR sem_op , at which time the value of .I semncnt is decremented, the absolute value of .I sem_op is subtracted from .I semval and, if .B SEM_UNDO is specified for this operation, the system updates the process undo count .RI ( semadj ) for this semaphore. .IP \(bu The semaphore set is removed from the system: .BR semop () fails, with .I errno set to .BR EIDRM . .IP \(bu The calling process catches a signal: the value of .I semncnt is decremented and .BR semop () fails, with .I errno set to .BR EINTR . .IP \(bu The time limit specified by .I timeout in a .BR semtimedop () call expires: the system call fails, with .I errno set to .BR EAGAIN . .PP On successful completion, the .I sempid value for each semaphore specified in the array pointed to by .I sops is set to the process ID of the calling process. In addition, the .I sem_otime .\" and .\" .I sem_ctime is set to the current time. .PP .BR semtimedop () behaves identically to .BR semop () except that in those cases were the calling process would sleep, the duration of that sleep is limited by the amount of elapsed time specified by the .B timespec structure whose address is passed in the .I timeout parameter. If the specified time limit has been reached, .BR semtimedop () fails with .I errno set to .B EAGAIN (and none of the operations in .I sops is performed). If the .I timeout parameter is NULL, then .BR semtimedop () behaves exactly like .BR semop (). .SH "RETURN VALUE" If successful .BR semop () and .BR semtimedop () return 0; otherwise they return \-1 with .I errno indicating the error. .SH ERRORS On failure, .I errno is set to one of the following: .TP .B E2BIG The argument .I nsops is greater than .BR SEMOPM , the maximum number of operations allowed per system call. .TP .B EACCES The calling process does not have the permissions required to perform the specified semaphore operations, and does not have the .B CAP_IPC_OWNER capability. .TP .B EAGAIN An operation could not proceed immediately and either .BR IPC_NOWAIT was specified in .I sem_flg or the time limit specified in .I timeout expired. .TP .B EFAULT An address specified in either the .I sops or .I timeout parameters isn't accessible. .TP .B EFBIG For some operation the value of .I sem_num is less than 0 or greater than or equal to the number of semaphores in the set. .TP .B EIDRM The semaphore set was removed. .TP .B EINTR While blocked in this system call, the process caught a signal. .TP .B EINVAL The semaphore set doesn't exist, or .I semid is less than zero, or .I nsops has a non-positive value. .TP .B ENOMEM The .I sem_flg of some operation specified .B SEM_UNDO and the system does not have enough memory to allocate the undo structure. .TP .B ERANGE For some operation .I sem_op+semval is greater than .BR SEMVMX , the implementation dependent maximum value for .IR semval . .SH NOTES The .I sem_undo structures of a process aren't inherited across a .BR fork (2) system call, but they are inherited across an .BR execve (2) system call. .PP .BR semop () is never automatically restarted after being interrupted by a signal handler, regardless of the setting of the .B SA_RESTART flag when establishing a signal handler. .PP .I semadj is a per\-process integer which is simply the (negative) count of all semaphore operations performed specifying the .B SEM_UNDO flag. When a semaphore's value is directly set using the .B SETVAL or .B SETALL request to .BR semctl (2), the corresponding .I semadj values in all processes are cleared. .PP The \fIsemval\fP, \fIsempid\fP, \fIsemzcnt\fP, and \fIsemnct\fP values for a semaphore can all be retrieved using appropriate .BR semctl (2) calls. .PP The following limits on semaphore set resources affect the .BR semop () call: .TP .B SEMOPM Maximum number of operations allowed for one .BR semop () call (32) (on Linux, this limit can be read and modified via the third field of .IR /proc/sys/kernel/sem ). .\" This /proc file is not available in Linux 2.2 and earlier -- MTK .TP .B SEMVMX Maximum allowable value for .IR semval : implementation dependent (32767). .PP The implementation has no intrinsic limits for the adjust on exit maximum value .RB ( SEMAEM ), the system wide maximum number of undo structures .RB ( SEMMNU ) and the per\-process maximum number of undo entries system parameters. .BR semtimedop () first appeared in Linux 2.5.52, and was subsequently backported into kernel 2.4.22. .SH BUGS When a process terminates, its set of associated .I semadj structures is used to undo the effect of all of the semaphore operations it performed with the .B 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 altogether (somewhat analogously to failing when .B 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 process 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. .\" The bug report: .\" http://marc.theaimsgroup.com/?l=linux-kernel&m=110260821123863&w=2 .\" the fix: .\" http://marc.theaimsgroup.com/?l=linux-kernel&m=110261701025794&w=2 .SH "CONFORMING TO" SVr4, SVID. SVr4 documents additional error conditions EINVAL, EFBIG, ENOSPC. .SH "SEE ALSO" .BR semctl (2), .BR semget (2), .BR sigaction (2), .BR capabilities (7), .BR sem_overview (7), .BR svipc (7)