From owner-freebsd-commit  Sun Jan  7 21:05:59 1996
Return-Path: owner-commit
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.3/8.7.3) id VAA12994
          for freebsd-commit-outgoing; Sun, 7 Jan 1996 21:05:59 -0800 (PST)
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.3/8.7.3) id VAA12975
          for cvs-all-outgoing; Sun, 7 Jan 1996 21:05:52 -0800 (PST)
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.3/8.7.3) id VAA12965
          for cvs-sys-outgoing; Sun, 7 Jan 1996 21:05:48 -0800 (PST)
Received: from jhome.DIALix.COM (jhome.DIALix.COM [192.203.228.69])
          by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id VAA12954
          Sun, 7 Jan 1996 21:05:25 -0800 (PST)
Received: (from peter@localhost) by jhome.DIALix.COM (8.7.3/8.7.3) id NAA01198; Mon, 8 Jan 1996 13:04:26 +0800 (WST)
Date: Mon, 8 Jan 1996 13:04:26 +0800 (WST)
From: Peter Wemm <peter@jhome.DIALix.COM>
To: =?KOI8-R?Q?=E1=CE=C4=D2=C5=CA_=FE=C5=D2=CE=CF=D7?= <ache@astral.msk.su>
cc: CVS-committers@freefall.freebsd.org, cvs-sys@freefall.freebsd.org
Subject: Re: cvs commit: src/sys/kern syscalls.master
In-Reply-To: <cB3t9ymWJ1@ache.dialup.ru>
Message-ID: <Pine.BSF.3.91.960108124809.557B-100000@jhome.DIALix.COM>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-commit@FreeBSD.ORG
Precedence: bulk

On Mon, 8 Jan 1996, =?KOI8-R?Q?=E1=CE=C4=D2=C5=CA_=FE=C5=D2=CE=CF=D7?= wrote:
> In message <199601080413.UAA09911@freefall.freebsd.org> Peter Wemm
>     writes:
> >  Modified:    sys/i386/ibcs2  syscalls.master
> >               sys/kern  syscalls.master
> >  Log:
> >  Remove the #ifdef SYSVSHM etc.  Always call the functions, some stubs
> >  are about to go in.  This is to fix the problem with the ibcs2 and linux
> >  lkm's not being able to call the sysv ipc functions unless the build is
> >  modified.
> 
> Does it means completely nuking of SYSV* options?
> If yes, at least /sys/conf/options needs modifications,
> if no, I don't understand this idea well, please, explain...

No, not quite.  Under the original system, all the syscall vectors 
pointed to enosys() if the appropriate option was no enabled.

What I've done, is make the syscall vectors *always* call the functions 
and simply provide stubs that do a log() and enosys() if the option was 
not enabled.

So, we still have the ability to build a kernel without 99% of the 
sysv_shm/sem/msg stuff.  We simply have stubs in their place.

This means that the linux and ibcs2 lkms can be compiled to use 
the sysv functions always..  As it is at the moment, they are
#include "opt_sysvipc.h"
#ifdef SYSVxxx
..
and the lkm makefiles "touch opt_sysvipc.h", so the lkm's could not use 
ipc _at all_, regardless of whether the kernel had it.

Now, it'll work if it's there and enosys() if not, all with the same LKM 
binary. (well, that's the aim anyway.. :-)

Also, this would be one step closer to having a LKM for the sysv ipc 
stuff itself - all that's needed now is to figure out how to allocate the 
maps etc which are normally done before the VM starts up properly.

-Peter