Jacob Kreindl

Dipl.-Ing. Jacob Kreindl, BSc.

Address: Altenberger Straße 69, 4040 Linz, Austria
Building: Computer Science Building (S3) Room: 202
Phone: + 43-732-2468-4349
Fax: + 43-732-2468-4345
Email: jacob.kreindl@jku.at

Research

As part of the collaboration between SSW and Oracle Labs I am working on the GraalVM project. My work focuses mostly on the LLVM IR interpreter Sulong and its interaction with other GraalVM-based language runtimes. For my PhD I am working on supporting dynamic taint analysis on GraalVM. The concept of dynamic taint analysis is to track the propagation of specific data while a program executes. My research addresses limitations of current implementations of this, such as language support and performance, by exploiting the capabilities of a managed runtime, more precisely of GraalVM.

Research Interests

Compilers & Interpreters
Virtual Machines
Dynamic Program Analysis
Tooling Support for Virtual Machines

Projects

TruffleTaint: Efficient Polyglot Dynamic Taint Analysis (PhD Thesis)
Sulong: Efficient Execution of LLVM IR on the JVM
- LLVM IR Parser (Master Project)
- Intstrumentation Support for Debugging, Tracing, and other Tooling (Master Thesis)

Publications

Dynamic Taint Analysis with Label-Defined Semantics [doi]
Jacob Kreindl, Daniele Bonetta, Lukas Stadler, David Leopoldseder, Hanspeter Mössenböck
Proceedings of the 19th International Conference on Managed Programming Languages and Runtimes, 2022

Low-Overhead Multi-Language Dynamic Taint Analysis on Managed Runtimes through Speculative Optimization [doi, preprint]
Jacob Kreindl, Daniele Bonetta, Lukas Stadler, David Leopoldseder, Hanspeter Mössenböck
Proceedings of the 18th International Conference on Managed Programming Languages and Runtimes, 2021

Multi-Language Dynamic Taint Analysis in a Polyglot Virtual Machine [doi, preprint]
Jacob Kreindl, Daniele Bonetta, Lukas Stadler, David Leopoldseder, Hanspeter Mössenböck
Proceedings of the 17th International Conference on Managed Programming Languages and Runtimes, 2020

Towards Efficient, Multi-Language Dynamic Taint Analysis [doi, preprint]
Jacob Kreindl, Daniele Bonetta, Hanspeter Mössenböck
Proceedings of the 16th International Conference on Managed Programming Languages and Runtimes, 2019

Debugging Native Extensions of Dynamic Languages [doi, preprint]
Jacob Kreindl, Manuel Rigger, Hanspeter Mössenböck
Proceedings of the 15th International Conference on Managed Languages & Runtimes, 2018

Tool Demonstrations & Other Presentations

Polyglot, Label-Defined Dynamic Taint Analysis in TruffleTaint [doi]
Jacob Kreindl, Daniele Bonetta, Lukas Stadler, David Leopoldseder, Hanspeter Mössenböck
Proceedings of the 19th International Conference on Managed Programming Languages and Runtimes, 2022

LLVM IR in GraalVM: Multi-Level, Polyglot Debugging with Sulong [slides, video]
Jacob Kreindl
2019 European LLVM Developers' Meeting

Demos

Source-Level Debugging with Sulong on GraalVM

Teaching

Lectures

2022 Summer Term:
Praktikum aus Softwareentwicklung 2
2021 Winter Term:
Übung zu Softwareentwicklung 1
2021 Summer Term:
Praktikum aus Softwareentwicklung 2
2020 Winter Term:
Übung zu Softwareentwicklung 1
2020 Summer Term:
Praktikum aus Softwareentwicklung 2
2019 Winter Term:
Übung zu Softwareentwicklung 1
2019 Summer Term:
Praktikum aus Softwareentwicklung 2

Bachelor Thesis Advisor

Character-Level Taint Tracking for Strings in Graal.js (Finished)
Expression Evaluation for Debugging with Sulong (Finished)
Partial Support for PThreads in Sulong (Finished)