May 18, 2026
Goals
- Find out more exploitable regions inside the
libtriton.solibrary.
What I worked on
- Started thinking about the design of my automated LLVM exploitation tool.
- Started reading Designing Data-Intensive Applications (DDIA)
- Read a chapter in
Fluent Pythonabout objects and their aliasing.
Key Insights
- Although
tuplesare immutable, the data that they contain can be mutable, hence the data resident within a tuple can change.- Aliasing and shallow copies of mutable data types can lead to subtle programming bugs.
- Never use mutable data types as default parameters in fucntions.
Questions / Confusions
- I still need to how predicate in ptx can be manipulated to skip instructions.
Ideas / Hypotheses
- I can generalize the attack surface of LLVM to different
targetslikeAMDGPU,RISC-Vetc.
Papers / Resources
None
Next Steps
- Create an end-to-end automated framework for finding out the exploitable flips.
Introspection
- Today was a good day in terms of my education as a gained a lot of knowledge about python internals, cloud deployments, data warehouses and data lakes.