Julian Kirsch & Clemens Jonischkeit
proudly present
-------------------
Movfuscator-Be-Gone
-------------------
a.k.a the
___ ___
/ /\ / /\
/ /::\ / /::\
/ /:/\:\ / /:/\:\ _____
/ /:/ \:\ / /::\ \:\ /____/\
/__/:/ \__\:|/__/:/\:\ \:\ \____\/
\ \:\ / /:/\ \:\ \:\_\/
___ ___ \ \:\ /:/ \ \:\ \:\ __ ___ ___ ___
/\ \ /\ \ ___ /\\ \:\/:/. \ \:\_\//__\ /\ \ /\ \ /\ \
|::\ \ /::\ \ /\ \ /:/ \ \::/ \ /:\ \:\ / / /::\ \ ___ /::\ \ /::\ \
|:::\ \:/\:\ \ \:\ \:/ /\\__\/:\ \:/ \__\// / /:/\:\ \/\__\:/\:\ \:/\:\__\
__|:|\:\ \ \:\ \ \:\ \ /:/ / \:\ \ /::\ \ / _:/ /::\ \/ // \:\ \ /:/ /
/::::|_\:\__\/ \:\__\ \:\__\:/ / \ \:\__\:/\:\__\/ /\__\:/\:\__\_//__/ \:\__\:/__/___
\:\~~\ \/__/\ /:/ / |:| |/ /\ \ /:/ // /:/ /\ /:/ // \/__/ \\ \ /:/ /::::/ /
\:\ \ \:\ /:/ / \|:| |__/\:\ /:/ // /:/ / /:/ //__/:/\:\ \\ /:/ //~~/~~~~
\:\ \ \:\/:/ /\__|:|__| \ \:\/:/ //_/:/ /:\/:/ /:\ \/__\:\ \\/:/ /:\~~\
\:\__\ \::/ /\::::/__/:\__\ \::/ / /:/ / \::/ / \:\__\ \:\__\:/ / \:\__\
\/__/ \/__/ ~~~~ \/__/ \/__/ \/__/ \/__/ \/__/ \/__/ __/ \/__/
-- Recovering from soul-crushing RE nightmares --
Since the publication of Christopher Domas' M/o/Vfuscator, we spent a great amount of time to analyze the inner workings of the famous one-instruction-compiler. We are happy to announce and release the (to our knowledge) first approach to a generic demovfuscator.
This tool constitutes a generic way of recovering the control flow of the original program from movfuscated binaries. As our approach makes zero assumptions about register allocations or a particular instruction order, but rather adheres to the high-level invariants that each movfuscated binary needs to conform to, our demovfuscator is also not affected by the proposed hardening techniques such as register renaming and instruction reordering. To achieve this, we use a combination of static taint analysis on the movfuscated code and a satisfiable modulo theory (SMT) solver. We successfully used our demovfuscator against several movfuscated binaries that emerged during several CTFs during the last months (Hackover CTF and 0CTF) proving that it already can handle real-world binaries that were not created by us.
The demovfuscator is programmed in C++ and as such has several (cool) dependencies:
- libcapstone as the core disassembler
- libz3 to reason about the semantics of the mov code
- libkeystone for re-substitution
As inconvenient this may be, we think that all three libraries should be in your RE toolchain anyway. If this is just unacceptable for you (and you trust us), there is a binary package that can be downloaded below.
After installing the dependencies, simply type make in the demov root
directory to compile. Note that you might have to adjust the library include
paths to match your distro.
The demovfuscator supports the following parameters:
./demov [-h] [-i symbols.idc] [-o patched_bin] [-g cfg.dot] obfuscated_input
-h Use for a description of the options
-i Derive symbols from the input bin and store them into symbols.idc
-o Generate a patched executable with explicit control flow and some
instructions resubstituted
-g Generate a UNIX dot compatible file containing the control flow
graph (might be easier to read than IDA's graph view)
Convert the .dot file to something usable by
cat cfg.dot | dot -Tpng > cfg.png
- Project Website: https://kirschju.re/demov
- Compiled demovfuscator (you still need the compiled dependencies): demov-compiled.tar.gz
- Bachelor's thesis describing parts of the approach: ba_jonischkeit_2016.pdf