From owner-FreeBSD-tech-jp@jp.freebsd.org  Wed Feb  5 09:57:02 1997
Received: by jaz.jp.freebsd.org (8.8.5+2.7Wbeta5/8.7.3) id JAA22490
	Wed, 5 Feb 1997 09:57:02 +0900 (JST)
Received: by jaz.jp.freebsd.org (8.8.5+2.7Wbeta5/8.7.3) with ESMTP id JAA22485
	for <FreeBSD-tech-jp@jp.freebsd.org>; Wed, 5 Feb 1997 09:57:01 +0900 (JST)
Received: from sbl-gw.sbl.cl.nec.co.jp by research.gate.nec.co.jp (8.8.5+2.7Wbeta5/950912) with ESMTP id JAA16457; Wed, 5 Feb 1997 09:57:00 +0900 (JST)
Received: from sirius.sbl.cl.nec.co.jp by sbl-gw.sbl.cl.nec.co.jp (8.8.5+2.7Wbeta5/3.3W6) with ESMTP
	id JAA26692; Wed, 5 Feb 1997 09:56:55 +0900 (JST)
Received: by sirius.sbl.cl.nec.co.jp (8.7.5+2.6Wbeta6/3.3W6) with UUCP
	id JAA17656; Wed, 5 Feb 1997 09:56:44 +0900 (JST)
Date: Wed, 5 Feb 1997 09:56:44 +0900 (JST)
From: Naoki Hamada <nao@sbl.cl.nec.co.jp>
Message-Id: <199702050056.JAA17656@sirius.sbl.cl.nec.co.jp>
References: <199701311134.UAA26203@frig.mt.cs.keio.ac.jp>
To: FreeBSD-tech-jp@jp.freebsd.org
In-reply-to: HOSOKAWA Tatsumi's message of "Fri, 31 Jan 1997 20:34:47 +0900"
	     <199701311134.UAA26203@frig.mt.cs.keio.ac.jp>
Reply-To: FreeBSD-tech-jp@jp.freebsd.org
Precedence: bulk
X-Distribute: distribute [version 2.1 (Alpha) patchlevel=19]
X-Sequence: FreeBSD-tech-jp 61
Subject: [FreeBSD-tech-jp 61] Re: VM86
Errors-To: owner-FreeBSD-tech-jp@jp.freebsd.org
Sender: owner-FreeBSD-tech-jp@jp.freebsd.org

$B$[$=$+$o$5$s(B:
>$B$@$l$+!"(BVM86 $B$r$^$8$a$K$d$C$F$_$k$-$N$"$k$R$H$O$$$^$;$s$+!)(B

  $B$[$1$[$1$7$F$$$k4V$K!":rF|(B(4$BF|(B)$B$+$i(B freebsd-hackers $B$GF0$-$,$"$j$^$7(B
$B$?!#(BSubject: vm86 in current? $B$N5-;v$r$^$H$a$F$"$j$^$9!#(B

$BIMED(B $BD><y(B

$B<g$JEP>l?MJ*(B:
Timo J Rinne <tri@iki.fi>			Rinne	
Poul-Henning Kamp <phk@critter.dk.tfs.com>	PHK
jlemon@americantv.com (Jonathan Lemon)		Jonathan
Bruce Evans <bde@zeta.org.au>			Bruce

Rinne in <199702041420.QAA06516@pooh.tky.hut.fi>:
>Is the vm86 code going to make its way into the freebsd-current (3.0)
>code anytime soon?

PHK in <1135.855067749@critter.dk.tfs.com>:
|Yes, If somebody starts working on it.

Rinne in <199702041451.QAA06624@pooh.tky.hut.fi>:
>As far as I understood the stuff in emulation list, someone has.
>It's a bit of the confusion if that stuff lives only in some weird
>patch set.  Of course it's also bad, if the stability of the overall
>system is compromised because of this.

Jonathan in <Mutt.19970204101725.jlemon@right.PCS>:
]Well, I've been re-working the patch set a little but, and now have a 
]working set of vm86 patches against -current, which appear to be stable.
]
]I've eliminated a private 'return' to VM86, so now the normal kernel exit
]point is used.  The next step I want to do is move the emulation of a couple
]instructions (cli/sti) into the kernel, and then add in VME support.
]
]When I get the cli/sti stuff working, I'll submit them to someone for
]integration.  Hopefully this week.
]
]BTW, are there any objections if I grow the trapframe/intrframe/clockframe
]structures by 4 more words?  It would only affect things that do sizeof(),
]and perhaps the kernel debugger.  Otherwise, a new vm86frame structure will
]be needed, with a little bit of typecasting back and forth.

BTW $B0J2<$K$D$$$F(B:

Bruce in <199702041847.FAA32304@godzilla.zeta.org.au>:
}It would be bogus because the standard trapframe doesn't actually have
}the extra words.  However, perhaps you can fudge the extra words by
}setting tss_esp0 16 lower.

Jonathan in <Mutt.19970204125843.jlemon@right.PCS>:
]This is what I've already done.  I just wanted to see if I could get rid 
]of the (struct trapframe_vm86) and use the reg[tXXXX] constructs instead.  
]
]Besides, SS and ESP are not actually present for traps/interrupts from the
]kernel, either.
