MILEPOST GCC: machine learning based research compiler

Grigori Fursin; Cupertino Miranda; Olivier Temam; Mircea Namolaru; Elad Yom-Tov; Ayal Zaks; Bilha Mendelson; Edwin Bonilla; John Thomson; Hugh Leather; Chris Williams; Michael O'Boyle; Phil Barnard; Elton Ashton; Eric Courtois; Francois Bodin

MILEPOST GCC: machine learning based research compiler

Grigori Fursin, Cupertino Miranda, Olivier Temam, Mircea Namolaru, Elad Yom-Tov, Ayal Zaks, Bilha Mendelson, Edwin Bonilla, John Thomson, Hugh Leather, Chris Williams, Michael O'Boyle, Phil Barnard, Elton Ashton, Eric Courtois, Francois Bodin

School of Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

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.

Original language	English
Title of host publication	GCC Summit
Publication status	Published - 2008

Cite this

@inproceedings{1042e5a57e9b4503a58f9ece794fd4d1,

title = "MILEPOST GCC: machine learning based research compiler",

abstract = "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.",

author = "Grigori Fursin and Cupertino Miranda and Olivier Temam and Mircea Namolaru and Elad Yom-Tov and Ayal Zaks and Bilha Mendelson and Edwin Bonilla and John Thomson and Hugh Leather and Chris Williams and Michael O'Boyle and Phil Barnard and Elton Ashton and Eric Courtois and Francois Bodin",

note = "MILEPOST project (http://www.milepost.eu)",

year = "2008",

language = "English",

booktitle = "GCC Summit",

}

TY - GEN

T1 - MILEPOST GCC: machine learning based research compiler

AU - Fursin, Grigori

AU - Miranda, Cupertino

AU - Temam, Olivier

AU - Namolaru, Mircea

AU - Yom-Tov, Elad

AU - Zaks, Ayal

AU - Mendelson, Bilha

AU - Bonilla, Edwin

AU - Thomson, John

AU - Leather, Hugh

AU - Williams, Chris

AU - O'Boyle, Michael

AU - Barnard, Phil

AU - Ashton, Elton

AU - Courtois, Eric

AU - Bodin, Francois

N1 - MILEPOST project (http://www.milepost.eu)

PY - 2008

Y1 - 2008

N2 - 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.

AB - 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.

UR - http://hal.inria.fr/inria-00294704

UR - https://www.scopus.com/pages/publications/84871331470

M3 - Conference contribution

BT - GCC Summit

ER -

MILEPOST GCC: machine learning based research compiler

Abstract

Other files and links

Fingerprint

Cite this