MILEPOST GCC: machine learning based research compiler

Grigori Fursin (1) , Cupertino Miranda (1) , Olivier Temam (1, 2) , Mircea Namolaru (3) , Elad Yom-Tov (4) , Ayal Zaks (3) , Bilha Mendelson (3) , Edwin Bonilla (5) , John Thomson (6) , Hugh Leather (6) , Chris Williams (5) , Michael O'Boyle (6) , Phil Barnard (7) , Elton Ashton (7) , Eric Courtois (8) , François Bodin (8)

1 ALCHEMY - Architectures, Languages and Compilers to Harness the End of Moore Years
2 LRI - Laboratoire de Recherche en Informatique
3 Compiler Technology Group
4 IBM Research [HAIFA]
5 ANC - Institute for Adaptive and Neural Computation
6 ICSA - Institute for Computing Systems Architecture School of Informatics - University of Edinburgh
7 ARC International
8 CAPS - CAPS Enterprise

Tuning hardwired compiler optimizations for rapidly evolving hardware makes porting an optimizing compiler for each new platform extremely challenging. Our radical approach is to develop a modular, extensible, self-optimizing compiler that automatically learns the best optimization heuristics based on the behavior of the platform. In this paper we describe MILEPOST GCC, a machine-learning-based compiler that automatically adjusts its optimization heuristics to improve the execution time, code size, or compilation time of specific programs on different architectures. Our preliminary experimental results show that it is possible to considerably reduce execution time of the MiBench benchmark suite on a range of platforms entirely automatically.

Domaines: Intelligence artificielle [cs.AI]; Architectures Matérielles [cs.AR]; Systèmes embarqués; Informatique mobile; Performance et fiabilité [cs.PF]

Grigori Fursin, Cupertino Miranda, Olivier Temam, Mircea Namolaru, Elad Yom-Tov, et al.. MILEPOST GCC: machine learning based research compiler. GCC Summit, Jun 2008, Ottawa, Canada. ⟨inria-00294704⟩