Computing, Electronics and Digital Systems
Work description
- Study the current LLVM / MLIR ecosystem and determine gaps and/or potential flows that can be reused to meet the objectives - Define the structural and computational features that you want to represent with IR (e.g., vectorization, streaming, in-memory computing, etc.) - Iterate on the current syntax to represent new features and improve the current integration into flows within and/or outside the MLIR ecosystem. - Study the feasibility of using the SRL DSL as a representation from which standard vector computing can be extracted - Write the activity report.
Academic Qualifications
Master student of Informatics Engineering, Electrical and Computer Engineering or similar
Minimum profile required
- Experience in the C++ programming language- Fluency in English (spoken and written)
Preference factors
- experience with compilers (parsing, and IRs), and the LLVM and/or MLIR ecosystem - basic knowledge of assembly generation
Application Period
Since 15 Nov 2024 to 28 Nov 2024
Centre
Human-Centered Computing and Information Science