Section: New Results

Mapping Deviation

Participant : Cédric Bastoul.

Compilers can provide a major help by automating the optimization and parallelization work. However they are very fragile black-boxes. A compiler may take a bad optimization decision because of imprecise heuristics or may turn off an optimization because of imprecise analyses, without providing much control or feedback to the end user. To address this issue, we researched and introduced mapping deviation, a new compiler technique that aims at providing a useful feedback on the semantics of a given program restructuring. Starting from a transformation intuition a user or a compiler wants to apply, our algorithm studies its correctness and can suggest changes or conditions to make it possible rather than being limited to the classical go/no-go answer. This algorithm builds on state-of-the-art polyhedral representation of programs and provides a high flexibility. We present two example applications of this technique: improving semi-automatic optimization tools for programmers and automatically designing runtime tests to check the correctness of a transformation for compilers.

This is a mid-term research on the mathematical ground of polyhedral compilation, started back to 2012. We found a solution and published it in 2016 in one of the main conferences in compilation: Compiler Construction [11]. We plan to release the tool that implements this research during the coming year.