Apply by doing:
       cd /usr/src
       patch -p0 < 024_sysv_sem.patch

And then rebuild your kernel.

Index: sys/kern/sysv_sem.c
===================================================================
RCS file: /cvs/src/sys/kern/sysv_sem.c,v
retrieving revision 1.3
diff -u -r1.3 sysv_sem.c
--- sys/kern/sysv_sem.c 1998/06/11 18:32:16     1.3
+++ sys/kern/sysv_sem.c 2000/05/26 18:35:36
@@ -1,4 +1,4 @@
-/*     $OpenBSD: sysv_sem.c,v 1.3 1998/06/11 18:32:16 deraadt Exp $    */
+/*     $OpenBSD: sysv_sem.c,v 1.4 2000/05/01 23:12:30 deraadt Exp $    */
/*     $NetBSD: sysv_sem.c,v 1.26 1996/02/09 19:00:25 christos Exp $   */

/*
@@ -20,9 +20,7 @@
#include <sys/syscallargs.h>

int    semtot = 0;
-struct proc *semlock_holder = NULL;

-void semlock __P((struct proc *));
struct sem_undo *semu_alloc __P((struct proc *));
int semundo_adjust __P((struct proc *, struct sem_undo **, int, int, int));
void semundo_clear __P((int, int));
@@ -48,63 +46,16 @@
       semu_list = NULL;
}

-void
-semlock(p)
-       struct proc *p;
-{
-
-       while (semlock_holder != NULL && semlock_holder != p)
-               sleep((caddr_t)&semlock_holder, (PZERO - 4));
-}
-
-/*
- * Lock or unlock the entire semaphore facility.
- *
- * This will probably eventually evolve into a general purpose semaphore
- * facility status enquiry mechanism (I don't like the "read /dev/kmem"
- * approach currently taken by ipcs and the amount of info that we want
- * to be able to extract for ipcs is probably beyond the capability of
- * the getkerninfo facility.
- *
- * At the time that the current version of semconfig was written, ipcs is
- * the only user of the semconfig facility.  It uses it to ensure that the
- * semaphore facility data structures remain static while it fishes around
- * in /dev/kmem.
- */
-
int
sys_semconfig(p, v, retval)
       struct proc *p;
       void *v;
       register_t *retval;
{
-       struct sys_semconfig_args /* {
-               syscallarg(int) flag;
-       } */ *uap = v;
-       int eval = 0;
-
-       semlock(p);
-
-       switch (SCARG(uap, flag)) {
-       case SEM_CONFIG_FREEZE:
-               semlock_holder = p;
-               break;
-
-       case SEM_CONFIG_THAW:
-               semlock_holder = NULL;
-               wakeup((caddr_t)&semlock_holder);
-               break;
-
-       default:
-               printf(
-                   "semconfig: unknown flag parameter value (%d) - ignored\n",
-                   SCARG(uap, flag));
-               eval = EINVAL;
-               break;
-       }
-
-       *retval = 0;
-       return(eval);
+       /*
+        * Dummy system call, removed in 2.7
+        */
+       return (0);
}

/*
@@ -318,8 +269,6 @@
       printf("call to semctl(%d, %d, %d, %p)\n", semid, semnum, cmd, arg);
#endif

-       semlock(p);
-
       semid = IPCID_TO_IX(semid);
       if (semid < 0 || semid >= seminfo.semmsl)
               return(EINVAL);
@@ -483,8 +432,6 @@
       printf("call to semctl(%d, %d, %d, %p)\n", semid, semnum, cmd, arg);
#endif

-       semlock(p);
-
       semid = IPCID_TO_IX(semid);
       if (semid < 0 || semid >= seminfo.semmsl)
               return(EINVAL);
@@ -643,8 +590,6 @@
       printf("semget(0x%x, %d, 0%o)\n", key, nsems, semflg);
#endif

-       semlock(p);
-
       if (key != IPC_PRIVATE) {
               for (semid = 0; semid < seminfo.semmni; semid++) {
                       if ((sema[semid].sem_perm.mode & SEM_ALLOC) &&
@@ -762,8 +707,6 @@
       printf("call to semop(%d, %p, %d)\n", semid, sops, nsops);
#endif

-       semlock(p);
-
       semid = IPCID_TO_IX(semid);     /* Convert back to zero origin */

       if (semid < 0 || semid >= seminfo.semmsl)
@@ -1030,96 +973,15 @@
       }

       /*
-        * There are a few possibilities to consider here ...
+        * No (i.e. we are in case 1 or 2).
        *
-        * 1) The semaphore facility isn't currently locked.  In this case,
-        *    this call should proceed normally.
-        * 2) The semaphore facility is locked by this process (i.e. the one
-        *    that is exiting).  In this case, this call should proceed as
-        *    usual and the facility should be unlocked at the end of this
-        *    routine (since the locker is exiting).
-        * 3) The semaphore facility is locked by some other process and this
-        *    process doesn't have an undo structure allocated for it.  In this
-        *    case, this call should proceed normally (i.e. not accomplish
-        *    anything and, most importantly, not block since that is
-        *    unnecessary and could result in a LOT of processes blocking in
-        *    here if the facility is locked for a long time).
-        * 4) The semaphore facility is locked by some other process and this
-        *    process has an undo structure allocated for it.  In this case,
-        *    this call should block until the facility has been unlocked since
-        *    the holder of the lock may be examining this process's proc entry
-        *    (the ipcs utility does this when printing out the information
-        *    from the allocated sem undo elements).
-        *
-        * This leads to the conclusion that we should not block unless we
-        * discover that the someone else has the semaphore facility locked and
-        * this process has an undo structure.  Let's do that...
-        *
-        * Note that we do this in a separate pass from the one that processes
-        * any existing undo structure since we don't want to risk blocking at
-        * that time (it would make the actual unlinking of the element from
-        * the chain of allocated undo structures rather messy).
+        * If there is no undo vector, skip to the end and unlock the
+        * semaphore facility if necessary.
        */
+       if (suptr == NULL)
+               return;

       /*
-        * Does someone else hold the semaphore facility's lock?
-        */
-
-       if (semlock_holder != NULL && semlock_holder != p) {
-               /*
-                * Yes (i.e. we are in case 3 or 4).
-                *
-                * If we didn't find an undo vector associated with this
-                * process than we can just return (i.e. we are in case 3).
-                *
-                * Note that we know that someone else is holding the lock so
-                * we don't even have to see if we're holding it...
-                */
-
-               if (suptr == NULL)
-                       return;
-
-               /*
-                * We are in case 4.
-                *
-                * Go to sleep as long as someone else is locking the semaphore
-                * facility (note that we won't get here if we are holding the
-                * lock so we don't need to check for that possibility).
-                */
-
-               while (semlock_holder != NULL)
-                       sleep((caddr_t)&semlock_holder, (PZERO - 4));
-
-               /*
-                * Nobody is holding the facility (i.e. we are now in case 1).
-                * We can proceed safely according to the argument outlined
-                * above.
-                *
-                * We look up the undo vector again, in case the list changed
-                * while we were asleep, and the parent is now different.
-                */
-
-               for (supptr = &semu_list; (suptr = *supptr) != NULL;
-                   supptr = &suptr->un_next) {
-                       if (suptr->un_proc == p)
-                               break;
-               }
-
-               if (suptr == NULL)
-                       panic("semexit: undo vector disappeared");
-       } else {
-               /*
-                * No (i.e. we are in case 1 or 2).
-                *
-                * If there is no undo vector, skip to the end and unlock the
-                * semaphore facility if necessary.
-                */
-
-               if (suptr == NULL)
-                       goto unlock;
-       }
-
-       /*
        * We are now in case 1 or 2, and we have an undo vector for this
        * process.
        */
@@ -1180,14 +1042,4 @@
#endif
       suptr->un_proc = NULL;
       *supptr = suptr->un_next;
-
-unlock:
-       /*
-        * If the exiting process is holding the global semaphore facility
-        * lock (i.e. we are in case 2) then release it.
-        */
-       if (semlock_holder == p) {
-               semlock_holder = NULL;
-               wakeup((caddr_t)&semlock_holder);
-       }
}