LDP
/
man-pages
mirror of https://github.com/mkerrisk/man-pages
0
1
Fork 0
Code Releases Activity

semget.2: Consolidate discussion on noninitialization of semaphores 

The fact that semget() does not initialize the semaphores
in a new set was covered in two places (in DESCRIPTION
and BUGS). Consolidate these into one place (in NOTES)
and also point out that POSIX.1-2008 says that a future
version of the standard may require an implementation to
initialize the semaphores to 0.

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>
This commit is contained in:
Michael Kerrisk 2014-04-28 08:26:06 +02:00
parent fc06f38c3b
commit 143f4ec383
1 changed files with 27 additions and 30 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

57
man2/semget.2
View File

@ -33,7 +33,7 @@
.\" Rewrote BUGS note about semget()'s failure to initialize
.\" semaphore values
.\"
.TH SEMGET 2 2012-05-31 "Linux" "Linux Programmer's Manual"
.TH SEMGET 2 2014-04-28 "Linux" "Linux Programmer's Manual"
.SH NAME
semget \- get a System V semaphore set identifier
.SH SYNOPSIS
@ -97,16 +97,6 @@ argument of
not meaningful for semaphores, and write permissions mean permission
to alter semaphore values).
.PP
The values of the semaphores in a newly created set are indeterminate.
(POSIX.1-2001 is explicit on this point.)
Although Linux, like many other implementations,
initializes the semaphore values to 0,
a portable application cannot rely on this:
it should explicitly initialize the semaphores to the desired values.
.\" In truth, every one of the many implementations that I've tested sets
.\" the values to zero, but I suppose there is/was some obscure
.\" implementation out there that does not.
.PP
When creating a new semaphore set,
.BR semget ()
initializes the set's associated data structure,
@ -223,6 +213,32 @@ SVr4, POSIX.1-2001.
.\" SVr4 documents additional error conditions EFBIG, E2BIG, EAGAIN,
.\" ERANGE, EFAULT.
.SH NOTES
.PP
The values of the semaphores in a newly created set are indeterminate.
(POSIX.1-2001 and POSIX.1-2008 are explicit on this point,
although POSIX.1-2008 notes that a future version of the standard
may require an implementation to initialize the semaphores to 0.)
Although Linux, like many other implementations,
initializes the semaphore values to 0,
a portable application cannot rely on this:
it should explicitly initialize the semaphores to the desired values.
.\" In truth, every one of the many implementations that I've tested sets
.\" the values to zero, but I suppose there is/was some obscure
.\" implementation out there that does not.
Initialization can be done using
.BR semctl (2)
.B SETVAL
or
.B SETALL
operation.
(Where multiple peers do not know who will be the first to
initialize the set, checking for a nonzero
.I sem_otime
in the associated data structure retrieved by a
.BR semctl (2)
.B IPC_STAT
operation can be used to avoid races.)
The inclusion of
.I <sys/types.h>
and
@ -275,25 +291,6 @@ The name choice
was perhaps unfortunate,
.B IPC_NEW
would more clearly show its function.
.LP
The semaphores in a set are not initialized by
.BR semget ().
.\" In fact they are initialized to zero on Linux, but POSIX.1-2001
.\" does not specify this, and we can't portably rely on it.
In order to initialize the semaphores,
.BR semctl (2)
must be used to perform a
.B SETVAL
or a
.B SETALL
operation on the semaphore set.
(Where multiple peers do not know who will be the first to
initialize the set, checking for a nonzero
.I sem_otime
in the associated data structure retrieved by a
.BR semctl (2)
.B IPC_STAT
operation can be used to avoid races.)
.SH SEE ALSO
.BR semctl (2),
.BR semop (2),