arXivMOM: Matrix Operations in MLIR

Lorenzo Chelini, Henrik Barthels, Paolo Bientinesi, Marcin Copik, Tobias Grosser, Daniele G. Spampinato, 2022

Modern research in code generators for dense linear algebra computations has shown the ability to produce optimized code with a performance which compares and often exceeds the one of state-of-the-art implementations by domain experts. However, the underlying infrastructure is often developed in isolation making the interconnection of logically combinable systems complicated if not impossible. In this paper, we propose to leverage MLIR as a unifying compiler infrastructure for the optimization of dense linear algebra operations. We propose a new MLIR dialect for expressing linear algebraic computations including matrix properties to enable high-level algorithmic transformations. The integration of this new dialect in MLIR enables end-to-end compilation of matrix computations via conversion to existing lower-level dialects already provided by the framework.

Publisher

@misc{chelini2022mom,
  author = {Chelini, Lorenzo and Barthels, Henrik and Bientinesi, Paolo and Copik, Marcin and Grosser, Tobias and Spampinato, Daniele G.},
  title = {{MOM: Matrix Operations in MLIR}},
  series = {arXiv},
  year = {2022},
  month = {8},
  url = {https://arxiv.org/abs/2208.10391},
  archiveprefix = {arXiv},
  eprint = {2208.10391}
}