Section: New Software and Platforms

APOLLO

Automatic speculative POLyhedral Loop Optimizer

Functional Description

We are developing a framework called APOLLO (Automatic speculative POLyhedral Loop Optimizer), dedicated to automatic, dynamic and speculative parallelization of loop nests that cannot be handled efficiently at compile-time. It is composed of a static part consisting of specific passes in the LLVM compiler suite, plus a modified Clang frontend, and a dynamic part consisting of a runtime system. It has been extended in 2015 to apply on-the-fly any kind of polyhedral transformations, including tiling, and to handle nonlinear loops as while-loops referencing memory through pointers and indirections.

• Participants: Aravind Sukumaran-Rajam, Juan Manuel Martinez Caamaño, Luis Esteban Campostrini, Artiom Baloian, Willy Wolff and Philippe Clauss

• Contact: Juan Manuel Martinez Caamaño