From owner-freebsd-questions  Wed Jun 12 12:47:29 1996
Return-Path: owner-questions
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.5/8.7.3) id MAA25949
          for questions-outgoing; Wed, 12 Jun 1996 12:47:29 -0700 (PDT)
Received: from fh (async1.pinrt3.telecom.at [194.118.0.98])
          by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id MAA25929
          for <FreeBSD-questions@FreeBSD.ORG>; Wed, 12 Jun 1996 12:47:20 -0700 (PDT)
Received: (from fh@localhost) by fh (8.6.12/8.6.12) id VAA02488; Wed, 12 Jun 1996 21:42:14 GMT
Date: Wed, 12 Jun 1996 21:42:14 GMT
Message-Id: <199606122142.VAA02488@fh>
To: FreeBSD-questions@FreeBSD.ORG
From: "Franz Hollerer"<eeg@telecom.at>
Subj: semaphore: SEM_UNDO
Sender: owner-questions@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

Hi,

I use semaphore to control read and write access to a
shared memory.
For the different semop() I set the SEM_UNDO flag.
So the kernel should undo the semaphore operations if the
process exits or is killed.
With Linux and SCO this works. But under FreeBSD 2.1
all other processes (which wait for the semaphore) hang.

*) Please, can someone tell me, if this is a bug in
   my program or a bug in the FreeBSD kernel?


/* ============================================================ */
/*    code fragment for locking share memory with a semaphore   */ 
/* ============================================================ */

extern gshm_t *shm;

#define SHM_UNLOCKED 0
#define SHM_RDLOCKED 1
#define SHM_WRLOCKED 2

static int shm_id, sem_id;
static int lock=SHM_UNLOCKED;

/* ============================================================ */