Vous pouvez vous abonner à nos annonces de séminaires:
                     http://cambium.inria.fr/seminar.html

             Nos séminaires sont accessibles en ligne en direct:
                  https://webconf.math.cnrs.fr/b/fra-ryy-fjn

			     S É M I N A I R E

                   ______                __    _
                  / ____/___ _____ ___  / /_  (_)_  ______ ___
                 / /   / __ `/ __ `__ \/ __ \/ / / / / __ `__ \
                / /___/ /_/ / / / / / / /_/ / / /_/ / / / / / /
                \____/\__,_/_/ /_/ /_/_.___/_/\__,_/_/ /_/ /_/

			     I N R I A - Paris
                 2 rue Simone Iff (ou: 41 rue du Charolais)
                            Salle A215, bâtiment A

                           Lundi 24 janvier, 10h30

                                 ------------
                                 Albert Cohen
                                 ------------

                                    Google

                =============================================
                MLIR: Compiler Construction for Heterogeneity
                =============================================

This is a new golden age for optimizing compilers. We live in a
heterogeneous world of domain-specific languages and accelerators,
freeing programming language and computer architects from the chains
of general-purpose, one-size-fits all designs. [Dave Patterson and
John Hennessy’s Turing award lecture, shamelessly adapted.] We are
moving from optimization targeting a von Neumann architecture to the
orchestration of a distributed hierarchy of computational and memory
resources, from performance through native libraries to just-in-time
code generation in active libraries, from heuristics to machine
learning compilation. Beyond performance, heterogeneity raises
challenges in compiler construction and validation across abstractions
and domain languages, and in the analysis, transformation and
debugging of programs crossing such abstractions. We will illustrate
these challenges on the design of MLIR, an open source infrastructure
derived from LLVM and inspired by many compiler construction projects.
MLIR aims to accelerate innovation in machine learning (ML) and
high-performance computing (HPC). It is built for extension and
evolution, enabling research and engineering on heterogeneous
compilation; it is also a research artifact raising its own design,
semantics and algorithmic challenges.