From mboxrd@z Thu Jan  1 00:00:00 1970
From: Marcelo Tosatti <mtosatti@redhat.com>
Subject: Re: [RFC] Patch - Big real mode emulation
Date: Wed, 21 May 2008 13:18:05 -0300
Message-ID: <20080521161805.GA9298@dmt>
References: <20080521113410.43ec182f@frecb000711.frec.bull.fr>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: kvm@vger.kernel.org, Avi Kivity <avi@qumranet.com>,
	Anthony Liguori <anthony@codemonkey.ws>,
	Mohammed Gamal <m.gamal005@gmail.com>,
	"Kamble, Nitin A" <nitin.a.kamble@intel.com>,
	Alexander Graf <alex@csgraf.de>
To: Guillaume Thouvenin <guillaume.thouvenin@ext.bull.net>
Return-path: <kvm-owner@vger.kernel.org>
Received: from mx1.redhat.com ([66.187.233.31]:50322 "EHLO mx1.redhat.com"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1759172AbYEUQQB (ORCPT <rfc822;kvm@vger.kernel.org>);
	Wed, 21 May 2008 12:16:01 -0400
Content-Disposition: inline
In-Reply-To: <20080521113410.43ec182f@frecb000711.frec.bull.fr>
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>

Hi Guillaume,

On Wed, May 21, 2008 at 11:34:10AM +0200, Guillaume Thouvenin wrote:
> Hello,
> 
> Opensuse 10.3 is it uses a version of gfxboot that reads SS after
> switching from real to protected mode, where SS contains an invalid
> value, which VMX does not allow. So this patch 

<snip>

>  	      add:		/* add */
> +		if ((c->d & ModRM) && c->modrm_mod == 3) {
> +			c->dst.bytes = (c->d & ByteOp) ? 1 : c->op_bytes;
> +			c->dst.ptr =  decode_register(c->modrm_rm, c->regs, c->d & ByteOp);
> +		}
>  		emulate_2op_SrcV("add", c->src, c->dst, ctxt->eflags);
>  		break;

I don't see any difference from the previous patch here (on the issue that
add result was stored in the wrong register) ?

     6486:       66 64 89 3e 72 01       mov    %edi,%fs:0x172
     648c:       66 be 8d 03 00 00       mov    $0x38d,%esi
     6492:       66 c1 e6 04             shl    $0x4,%esi
     6496:       66 b8 98 0a 00 00       mov    $0xa98,%eax
     649c:       66 03 f0                add    %eax,%esi

So "66 03 f0" stores result in eax instead of esi. And of course this
can be fatal (in the FreeDOS case the TSS data was copied to a wrong
location). Better fix that before merging.

Thanks!