This center brings together people to conduct research for advancing the MLIR compiler ecosystem. We perform research in fundamental and applied areas expanding the capabilities and applications of MLIR compiler technology. We are firmly commited to Open Source Software and Open Science.
- Martin Lücke, PhD student at the University of Edinburgh
- Mathieu Fehr, PhD student at the University of Edinburgh
- Michel Steuwer, Lecturer (Assistant Professor) in Compilers and Runtime Systems at the University of Edinburgh
- Tobias Grosser, Reader (Associate Professor) in Compilers and Runtime Systems at the University of Edinburgh
- Compiler infrastrucure and community
- Intermediate representations
- Program optimizations and analysis
- Domain-specific compilation
- Machine learning for compilers
- Compilers for machine learning
- Compilation for accelerators and novel hardware
- Human-compiler interface
- Compiler verification, automatic theorem proving & constraint solving
- Novel applications of MLIR-based compiler technologies
- RISE: A functional pattern-based data-parallel dialect
- ELEVATE: A Language for Describing Optimization Strategies
- Paper: Towards a Domain-Extensible Compiler: Optimizing an Image Processing Pipeline on Mobile CPUs by Thomas Koehler and Michel Steuwer published at CGO 2021
- Paper: Progressive Raising in Multi-level IR by Lorenzo Chelini, Andi Drebes, Oleksandr Zinenko, Albert Cohen, Nicolas Vasilache, Tobias Grosser and Henk Corporaal published at CGO 2021
- Paper: Integrating a functional pattern-based IR into MLIR by Martin Lücke, Michel Steuwer, and Aaron Smith published at CC 2021.
- Paper: Row-Polymorphic Types for Strategic Rewriting by Rongxiao Fu, Xueying Qin, Ornela Dardha and Michel Steuwer published on arXiv
- Paper: Declarative Loop Tactics for Domain-specific Optimization by Lorenzo Chelini, Oleksandr Zinenko, Tobias Grosser and Henk Corporaal published in ACM TACO
- Paper: Achieving high-performance the functional way: a functional pearl on expressing high-performance optimizations as rewrite strategies by Bastian Hagedorn, Johannes Lenfers, Thomas Koehler, Xueying Qin, Sergei Gorlatch and Michel Steuwer published at ICFP 2020
- Paper: LLHD: a multi-level intermediate representation for hardware description languages by Fabian Schuiki, Andreas Kurth, Tobias Grosser and Luca Benini published at PLDI 2020
- Paper: MultiLevel Tactics: Lifting loops in MLIR by Lorenzo Chelini, Andi Drebes, Oleksandr Zinenko, Albert Cohen, Henk Corporaal, Tobias Grosser and Nicolas Vasilache published at EuroLLVM 2020