Academic Journal

Removing checks in dynamically typed languages through efficient profiling

Bibliographic Details
Title: Removing checks in dynamically typed languages through efficient profiling
Authors: Dot, Gem, Martinez, Alejandro, González Colás, Antonio María
Contributors: Universitat Politècnica de Catalunya. Departament d'Arquitectura de Computadors, Universitat Politècnica de Catalunya. ARCO - Microarquitectura i Compiladors
Source: UPCommons. Portal del coneixement obert de la UPC
Universitat Politècnica de Catalunya (UPC)
Recercat. Dipósit de la Recerca de Catalunya
instname
Publisher Information: IEEE, 2017.
Publication Year: 2017
Subject Terms: Disseny de sistemes, System design, Llenguatges de programació, Programming languages (Electronic computers), 02 engineering and technology, Informàtica::Llenguatges de programació [Àrees temàtiques de la UPC], 01 natural sciences, Systems engineering, Program diagnostics, Àrees temàtiques de la UPC::Informàtica::Llenguatges de programació, 0103 physical sciences, Hardware-software codesign, 0202 electrical engineering, electronic engineering, information engineering, Enginyeria de sistemes, Java
Description: Dynamically typed languages increase programmer's productivity at the expense of some runtime overheads to manage the types of variables, since they are not declared at compile time and can change at runtime. One of the most important overheads is due to very frequent checks that are introduced in the specialized code to identify the type of the variables. In this paper, we present a HW/SW hybrid mechanism that allows the removal of checks executed in the optimized code by performing a HW profiling of the types of object variables. To demonstrate the benefits of the proposed technique, we implement it in a JavaScript engine and show that it produces 7.1% speedup on average for optimized JavaScript code (up to 34% for some applications) and 6.5% energy reduction.
Document Type: Article
Conference object
File Description: application/pdf
DOI: 10.1109/cgo.2017.7863745
DOI: 10.5555/3049832.3049860
Access URL: https://upcommons.upc.edu/bitstream/2117/113960/3/CGO2017.pdf
http://hdl.handle.net/2117/113960
https://dl.acm.org/doi/10.5555/3049832.3049860
https://upcommons.upc.edu/handle/2117/113960
https://dblp.uni-trier.de/db/conf/cgo/cgo2017.html#DotM017
https://upcommons.upc.edu/bitstream/2117/113960/3/CGO2017.pdf
Accession Number: edsair.doi.dedup.....110d60f02706b4ccb8d8853df82b3e5c
Database: OpenAIRE
Description
DOI:10.1109/cgo.2017.7863745