Static Prediction of Silent Stores
| Autor: | Guilherme Vieira Leobas, Fernando Magno Quintão Pereira, Abdoulaye Gamatié |
|---|---|
| Přispěvatelé: | Universidade Federal de Minas Gerais [Belo Horizonte] (UFMG), ADAptive Computing (ADAC), Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), ANR-15-CE25-0007,CONTINUUM,Continuum de Conception pour Noeuds de Calculs Energétiquement-Efficaces de Prochaine Génération(2015) |
| Jazyk: | angličtina |
| Rok vydání: | 2019 |
| Předmět: |
[INFO.INFO-AR]Computer Science [cs]/Hardware Architecture [cs.AR]
Silent stores Computer science 02 engineering and technology [INFO.INFO-SE]Computer Science [cs]/Software Engineering [cs.SE] Machine learning computer.software_genre 01 natural sciences Non volatile memory 0103 physical sciences 0202 electrical engineering electronic engineering information engineering 010302 applied physics Profiling (computer programming) 020203 distributed computing [INFO.INFO-PL]Computer Science [cs]/Programming Languages [cs.PL] business.industry Static analysis Program optimization Syntax Automatic parallelization Hardware and Architecture [INFO.INFO-ES]Computer Science [cs]/Embedded Systems Artificial intelligence business computer Software Information Systems Coding (social sciences) |
| Zdroj: | ACM Transactions on Architecture and Code Optimization ACM Transactions on Architecture and Code Optimization, Association for Computing Machinery, 2019, 15 (4), pp.#44. ⟨10.1145/3280848⟩ |
| ISSN: | 1544-3566 1544-3973 |
| DOI: | 10.1145/3280848⟩ |
| Popis: | International audience; A Store operation is called “silent” if it writes in memory a value that is already there. e ability to detect silent stores is important, because they might indicate performance bugs, might enable code optimizations, and might reveal opportunities of automatic parallelization, for instance. Silent stores are traditionally detected via proling tools. In this paper, we depart from this methodology, and, instead, explore the following question: is it possible to predict silentness by analyzing the syntax of programs? e process of building an answer to this question is interesting in itself, given the stochastic nature of silent stores, which depend on data and coding style. To build such an answer, we have developed a methodology to classify store operations in terms of syntactic features of programs. Based on such features, we develop dierent kinds of predictors, some of which go much beyond what any trivial approach could achieve. To illustrate how static prediction can be employed in practice, we use it to optimize programs running on non-volatile memory systems. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|