Scribd
Upload
139 views

Creating Arbitrary Shellcode in Unicode Expanded Strings PDF

This document introduces the "Venetian" exploit technique for executing arbitrary shellcode when a buffer overflow occurs in a Unicode string on x86 processors. The technique uses instructions with alternating zero bytes to set arbitrary bytes in the Unicode buffer, overwriting the null bytes. This allows shellcode to be executed, with up to 146 bytes possible given a 1024 byte buffer. The technique demonstrates that Unicode overflows should not be considered non-exploitable.

Uploaded by

prabodhm
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
139 views

Creating Arbitrary Shellcode in Unicode Expanded Strings PDF

This document introduces the "Venetian" exploit technique for executing arbitrary shellcode when a buffer overflow occurs in a Unicode string on x86 processors. The technique uses instructions with alternating zero bytes to set arbitrary bytes in the Unicode buffer, overwriting the null bytes. This allows shellcode to be executed, with up to 146 bytes possible given a 1024 byte buffer. The technique demonstrates that Unicode overflows should not be considered non-exploitable.

Uploaded by

prabodhm
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

An NGSSoftware Insight Security Research (NISR) Publication

Creating Arbitrary Shellcode In Unicode


Expanded Strings
The Venetian exploit
Chris Anley (chris@nextgenss.com)
8th January 2002
www.ngssoftware.com

[Abstract]
The paper is intended to be read by the
portion of the security community
responsible for creating protective
mechanisms to guard against
shellcode type security flaws; the
intention is to remove the perception
that Unicode buffer overflows are non
exploitable and thereby improve the
general state of network security. It is
often the case that several classes of
overflow or format string bug are
labelled denial of service attacks
when in fact it is possible to execute
arbitrary code. This paper deals with
one of these classes of overflow.
This paper introduces a technique (the
Venetian exploit) that can be used to
permit the execution of a small amount
of arbitrary code in a situation where a
buffer overflow occurs in a Unicode
string on the Intel x86 processors. This
situation is common in the Windows
operating systems but the technique is
not operating system specific.
[Introduction]
It is often the case that an overflow on
the Windows platform occurs in a
string that is converted to Unicode
prior to the overflow.
This leads to a complication when
attempting to write an exploit, since
the shellcode will generally have null
bytes inserted between each byte of
the submitted string. For example
AAAA

...becomes
00 41 00 41 00 41 00 41

It is generally acknowledged that


writing meaningful shellcode in which

every alternate byte is zero is


extremely hard. This paper describes
how this problem can be overcome,
using a technique not dissimilar to the
bridge building method that can be
used to create exploit code using only
printable ASCII characters.
[Assumptions]
In order for this technique to be useful,
a number of conditions must hold:
1) It must be possible to redirect
the execution path of the target
program into the Unicode
buffer.
And either:
2) One of the registers
eax,ebx,ecx,edx,esp,ebp or esi
points to a known offset in our
Unicode buffer
Or
3) We know, or can easily obtain,
the absolute address in memory
of our buffer
[The Venetian Exploit]
The Unicode buffer can be imagined to
be somewhat similar to a Venetian
blind; there are solid bytes that we
control, and gaps containing the
alternating zeroes.
The technique described below
consists of using the solid bytes at
the start of the buffer to interleave
chosen bytes into the gaps further
down in the buffer, effectively
closing the blind, and creating a
small amount of totally arbitrary
shellcode that is the actual payload of
the exploit.

In order to do this, we must come up


with some way of modifying memory
using instructions that contain
alternating zeroes.
Instructions on the Intel processors
have variable length. Since every
other byte of our code must be zero,
we will have to insert nop equivalent
instructions (instructions that do
nothing of consequence to our code,
but which act as filler) in order to
make sure that our code is aligned
correctly on instruction boundaries.
Generally we will be using instructions
that start with a non-zero byte, since
the instructions that start with 00 are
all add, which is not especially
useful to us.
Hence, if the next instruction in our
code must start with a 00, we can
use instructions of the following form
to realign so that we can do
something more interesting with
subsequent instructions:
00 6D 00:add byte ptr [ebp],ch

This assumes that ebp points to


something that is writeable, and that
we don't care about for the purposes of
the exploit. If this is not the case, we
can use one of the following:
00
00
00
00
00
00

6E
6F
70
71
72
73

00:add
00:add
00:add
00:add
00:add
00:add

byte
byte
byte
byte
byte
byte

ptr
ptr
ptr
ptr
ptr
ptr

[esi],ch
[edi],ch
[eax],dh
[ecx],dh
[edx],dh
[ebx],dh

If none of the registers points to a


location that we can safely overwrite,
we just assign a constant pointer value
to (say) eax using these instructions:
6A 00:push 0
58
:pop

eax

(to assign 0 to eax), then we add


and sub as described below, until eax
points to a location in memory that we
can safely overwrite with our nop
equivalent alignment instructions. This
gives us a convenient way of
realigning our instructions.
As stated above, we assume that there
is a register that points to our Unicode
buffer.
What we are going to do is set every
00 byte beyond a certain point in our
buffer to the value of our choice, by
doing this:
80 00 75:add byte ptr [eax],75h

...then incrementing eax twice...


40:inc eax
00 6D 00:add byte ptr [ebp],ch
40:inc eax

Then setting the next 00 byte. This will


end up with arbitrary bytes being
placed in a part of the buffer towards
the end of our shellcode. The buffer
will be laid out like this:
0x00000000
...
[ alternate-zero byte setting code ]
[ arbitrary bytes of shellcode ]
...
0xffffffff
Of course, the first thing we have to do
is get a pointer to the part of the buffer
where we intend to start writing
arbitrary bytes.
To do this, we exchange the values of
the register that points to our shellcode
with (say) eax, using the convenient
one-byte xchg instruction - one of
the following:
93:xchg eax,ebx
91:xchg eax,ecx
92:xchg eax,edx

94:xchg
95:xchg
96:xchg
97:xchg

eax,esp
eax,ebp
eax,esi
eax,edi

We then modify the value of eax using


add and sub, to make it point to
the arbitrary byte part of our buffer:
05 00 75 00 4C:
add eax,4C007500h
2D 00 75 00 4C:
sub eax,4C007500h

Multiple add and sub operations


will probably be necessary. We can
easily add multiples of 256 by going
like this:
add eax,4C007500h
sub eax,4C007400h

We then start adding and incrementing


as described above.
Our arbitrary code gets executed due to
the fact that we just execute through
our byte setting code and into the
arbitrary code. If we get that initial
pointer offset right, we will just
continue executing into our arbitrary
code.
[Problems]
First; if the target program has a high
bit filter, this technique is very hard,
because it is probably necessary to do
the initial pointer xchg, and that
requires an opcode above 0x7f. This is
likely to create difficulty, although a
series of push and pop instructions
could be made to be equivalent.
Size is also an issue - the instruction
sequence to set a single 00 byte looks
like this:
40
:inc
00 6D 00:add
40
:inc
00 6D 00:add

eax
byte ptr [ebp],ch
eax
byte ptr [ebp],ch

80 00 75:add byte ptr [eax],75h


00 6D 00:add byte ptr [ebp],ch

... so that's 14 bytes of code to set 2


arbitrary bytes (we get one for free,
remember; the one that was already in
the string).
That means assuming a buffer of 1024
bytes that we can set, the maximum
size of the exploit code will be
(1024 * 2) / 14 = 146 bytes (since the
size of a Unicode string doubles)
Which isn't much, but it is enough to
do some harm; run an arbitrary
command, for example.
It is probable that refinements to the
technique are possible that reduce the
amount of code necessary to create the
arbitrary shellcode.
To put this in context, code that will
initiate a reverse shell fits into less than
170 bytes. This technique will
therefore probably be sufficient to
successfully exploit a Unicode
overflow in the wild.
[Conclusion]
The Venetian exploit technique
described in this paper is a somewhat
convoluted way of writing an exploit
but it handles a situation that is quite
commonly seen in the Windows family
of operating systems.
Hopefully this paper has demonstrated
that treating Unicode based
overflows as non-exploitable is
dangerous.
It is always safest to assume that if the
execution path of a program can be
affected in any way, that it is possible
to execute arbitrary code.

You might also like