From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43)
	id 1JaIPw-0000eZ-CZ
	for qemu-devel@nongnu.org; Fri, 14 Mar 2008 18:30:24 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43)
	id 1JaIPt-0000eG-VS
	for qemu-devel@nongnu.org; Fri, 14 Mar 2008 18:30:23 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.43) id 1JaIPt-0000eB-PL
	for qemu-devel@nongnu.org; Fri, 14 Mar 2008 18:30:21 -0400
Received: from vs165240.vserver.de ([62.75.165.240])
	by monty-python.gnu.org with esmtps (TLS-1.0:RSA_AES_256_CBC_SHA1:32)
	(Exim 4.60) (envelope-from <stefan@nurnware.de>) id 1JaIPs-0003Fx-IH
	for qemu-devel@nongnu.org; Fri, 14 Mar 2008 18:30:22 -0400
Received: from p57a6bc53.dip.t-dialin.net ([87.166.188.83] helo=[192.168.0.10])
	by vs165240.vserver.de with esmtpsa
	(TLS-1.0:DHE_RSA_AES_256_CBC_SHA:32) (Exim 4.50) id 1JaHde-0004dm-Ch
	for qemu-devel@nongnu.org; Fri, 14 Mar 2008 21:40:36 +0000
Message-ID: <47DAF0CC.4070806@nurnware.de>
Date: Fri, 14 Mar 2008 22:40:28 +0100
From: =?ISO-8859-15?Q?Stefan_N=FCrnberger?= <stefan@nurnware.de>
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="------------090006010904000603030307"
Subject: [Qemu-devel] Bug with 32-bit segmentation
Reply-To: qemu-devel@nongnu.org
List-Id: qemu-devel.nongnu.org
List-Unsubscribe: <http://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/qemu-devel>
List-Post: <mailto:qemu-devel@nongnu.org>
List-Help: <mailto:qemu-devel-request@nongnu.org?subject=help>
List-Subscribe: <http://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=subscribe>
To: qemu-devel@nongnu.org

This is a multi-part message in MIME format.
--------------090006010904000603030307
Content-Type: multipart/alternative;
 boundary="------------020509030009000707070404"


--------------020509030009000707070404
Content-Type: text/plain; charset=ISO-8859-15; format=flowed
Content-Transfer-Encoding: 7bit

Hi,

before talking about the bad things, I have to remark, that qemu is a 
great program. I like it for it's simplicity and I cherish the 
integrated gdb-stub, wich avoided a lot of headaches in my early 
kernel-develpment phase, when no own serial debugger stub existed. And 
secondly, I hope this is the right place to post bug reports, but I 
couldn't find an appropriate place. The forum seemed more "user oriented".
Unfortunately I encountered a little error when experimenting with 
segmentation. I'm talking about the x86 version of qemu (0.9.1 soft-mmu).
I set up a code segment in the GDT from 0x00000000 to 0x1ffffffff. The 
data segment ranges from 0x00000000 to 0xbfffcfff. All the stack for 
instance is allocated behind 0x20000000 - thus after the code segment. 
Then I tried to execute the stack, to test some security features that 
this segmentation model should provide in spite of lacking a hardware 
NX-bit for example. But unfortunately a "jmp"-instruction to an address 
 > 0x1fffffff didn't raise a #GPF, wich shouldn't be possible, I think. 
I also tested the code on real hardware (AMD Athlon 64, Athlon XP) as 
well as in VMware. On the these three machines it worked correctly - so, 
as expected.
This error doesn't always occur. Sometimes I get the "wanted" #GPF in 
qemu, but - as I already said - sometimes not. To investigate I did the 
following: I modified the demo program to execute the intentionally 
stored instructions on the stack several times in a for-loop (see the 
attached file for details). The #GPF I was waiting for then raised after 
executing the stack the 47th time, the 582th or the 200th time...etc. No 
real pattern. But sometimes, as wanted and expected, the first time. 
Another thing that's strange about this, is, that a GPF will be raised 
correctly when stepping through the code via the integrated gdb-stub of 
qemu and an attached gdb. So there wasn't really a way to determine why 
it wouldn't work.
I've also attached a screenshot showing the "show registers" on the qemu 
console. As you can see the EIP is > GDT[CS].
BTW. don't worry abount the includes in the attached c-file. This file 
is designed for another OS. But this this shouldn't matter as it's only 
to get an impression of the code.

If you have any further questions, feel free to cantact me.
Best regards,
Stefan

--------------020509030009000707070404
Content-Type: text/html; charset=ISO-8859-15
Content-Transfer-Encoding: 7bit

<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
</head>
<body bgcolor="#ffffff" text="#000000">
<font size="-1"><font face="Helvetica, Arial, sans-serif">Hi,<br>
<br>
before talking about the bad things, I have to remark, that qemu is a
great program. I like it for it's simplicity and I cherish the
integrated gdb-stub, wich avoided a lot of headaches in my early
kernel-develpment phase, when no own serial debugger stub existed. And
secondly, I hope this is the right place to post bug reports, but I
couldn't find an appropriate place. The forum seemed more "user
oriented".<br>
Unfortunately I encountered a little error when experimenting with
segmentation. I'm talking about the x86 version of qemu (0.9.1
soft-mmu). <br>
I set up a code segment in the GDT from 0x00000000 to 0x1ffffffff. The
data segment ranges from 0x00000000 to 0xbfffcfff. All the stack for
instance is allocated behind 0x20000000 - thus after the code segment.
Then I tried to execute the stack, to test some security features that
this segmentation model should provide in spite of lacking a hardware
NX-bit for example. But unfortunately a "jmp"-instruction to an address
&gt; 0x1fffffff didn't raise a #GPF, wich shouldn't be possible, I
think. I also tested the code on real hardware (AMD Athlon 64, Athlon
XP) as well as in VMware. On the these three machines it worked
correctly - so, as expected.<br>
This error doesn't always occur. Sometimes I get the "wanted" #GPF in
qemu, but - as I already said - sometimes not. To investigate I did the
following: I modified the demo program to execute the intentionally
stored instructions on the stack several times in a for-loop (see the
attached file for details). The #GPF I was waiting for then raised
after executing the stack the 47th time, the 582th or the 200th
time...etc. No real pattern. But sometimes, as wanted and expected, the
first time. Another thing that's strange about this, is, that a GPF
will be raised correctly when stepping through the code via the
integrated gdb-stub of qemu and an attached gdb. So there wasn't really
a way to determine why it wouldn't work.<br>
I've also attached a screenshot showing the "show registers" on the
qemu console. As you can see the EIP is &gt; GDT[CS].<br>
BTW. don't worry abount the includes in the attached c-file. This file
is designed for another OS. But this this shouldn't matter as it's only
to get an impression of the code.<br>
<br>
If you have any further questions, feel free to cantact me.<br>
Best regards,<br>
Stefan<br>
</font></font>
</body>
</html>

--------------020509030009000707070404--

--------------090006010904000603030307
Content-Type: image/png;
 name="qemu_maybe_error.png"
Content-Transfer-Encoding: base64
Content-Disposition: inline;
 filename="qemu_maybe_error.png"
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--------------090006010904000603030307
Content-Type: text/x-csrc;
 name="foo.c"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
 filename="foo.c"

#include <raykernel.h>
#include <kdisplay/kprintf.h>
#include <tdm/io.h>

// x86 instructions : nop, nop, nop, ret
unsigned char stack_instructions[] = {0x90, 0x90, 0x90, 0xc3};


RAYENTRY KernelModuleEntry(char *arguments) {
	
	KPrintf("Starting stack execution test...\n");
	__asm__ __volatile__ ("call *%0\n": : "r"(stack_instructions));
	
	KPrintf ("This shouldnt happen!");
	exit(-2);
}

--------------090006010904000603030307--