BlipsAndChitz🚀

(Peter Garba) Software Security Expert @ Thales and passionated reverse engineer. Software De-Obfuscation and novel attacks

[BlipsAndChitz🚀 @Blips_and_Chitz]

@lauriewired What about libsqlfs?

[BlipsAndChitz🚀 @Blips_and_Chitz]

Missed Total Commander on Linux so I built my own with the same look and feel on Linux and Windows
https://t.co/3pkISl6eBj

[BlipsAndChitz🚀 @Blips_and_Chitz]

New paper highlights SATURN, a framework we developed for automated malware deobfuscation!

Researchers call it the "most promising method," proving it natively clears MBA masks, opaque predicates, and flattened code via LLVM IR + SMT 😎

https://t.co/if53HGrEE1

[rev.ng @_revng]

Check out this interesting paper.
We’re thinking about doing something similar!
https://t.co/XbfVz2Bdxv

[RE//verse @REverseConf]

WebAssembly de-obfuscation with LLVM. Vikas Gupta & Peter Garba (@Blips_and_Chitz) reveal how to reverse Wasm obfuscation using compiler optimizations. Simplify complex code without pattern rules & tackle real-world scenarios. https://t.co/14afKD3EVR #REverse2025 #Wasm #ReverseEngineering

[BlipsAndChitz🚀 @Blips_and_Chitz]

DeepSeek R1 is giving correct simplifications for several MBAs I tried. Really impressive!

[RE//verse @REverseConf]

Ready to get out of the cold? Come to RE//verse and see some of our amazing lineup. https://t.co/MMaX9RganZ Also stay tuned we're going to be announcing our 2 keynote speakers very soon!

[BlipsAndChitz🚀 @Blips_and_Chitz]

@vxunderground Compile your code with movfuscator as mov is turing complete 😉

[Tim Blazytko @mr_phrazer]

The line-up for @REverseConf is impressive, but one talk I’m particularly excited about is from Vikas Gupta and @Blips_and_Chitz:

“Standing on the Shoulders of Giants: De-Obfuscating WebAssembly using LLVM”

https://t.co/0f4AlyNWfm

[BlipsAndChitz🚀 @Blips_and_Chitz]

@peter_a_goodman @HexRaysSA Congrats! Looking forward to seeing your impact on the product

[Phoronix @phoronix]

How @AMD Is Taking Standard C/C++ Code To Run Directly On GPUs

https://t.co/zamXqaGeeh

[Thalium Team @thalium_team]

Reversing virtualized binaries is no easy task. Our intern Jack took on exploring automated devirtualization techniques, and presents in our latest blog post an efficient, modular, taint-based approach that leverages LLVM IR: https://t.co/ARiXRvLV6J

[Colton Skees @ColtonSkees]

(1) Decided to release the source code for my LLVM-based static binary analysis framework (https://t.co/xEgouoAkWS). It implements, among other things, an iterative control flow graph reconstruction algorithm heavily inspired by SATURN, using Remill and Souper.

[BlipsAndChitz🚀 @Blips_and_Chitz]

SLOT: SMT-LLVM Optimizing Translation

Slides: https://t.co/AxTu1HUHdS

https://t.co/4fyqaZRPqT

[BlipsAndChitz🚀 @Blips_and_Chitz]

@Microsoft Thats the piece of code that was also applied for AMD and is skipped now.

[BlipsAndChitz🚀 @Blips_and_Chitz]

In recent ntoskrnl.exe update from KB5041587 @Microsoft introduced "Feature_Servicing_AmdDisableEarlyIBPB__private_IsEnabledDeviceUsage" and enables IBPB later on in the KiUpdateSpeculationControl function. Wondering why this gives a huge performance boost on >=Zen3 CPUs ?!

[BlipsAndChitz🚀 @Blips_and_Chitz]

@Microsoft It seems that Intel CPU specific settings were also applied in AMD CPUs and this piece of code is now wrapped with an "if" statement.

[BlipsAndChitz🚀 @Blips_and_Chitz]

@phoronix @AMDRyzen Are you planning some gaming benchmarks? Win11 vs Linux?!

[BlipsAndChitz🚀 @Blips_and_Chitz]

@lzsthw @Farenain The problem might be that it needs a correct implementation of all opcodes in unicorn/qemu which is not the case for some opcodes

[Peter Goodman @peter_a_goodman]

@h0mbre_ I open-sourced my C/C++ build indexer. It has a really cool and extensive API: https://t.co/0BXb5RWAOk