Titre :

A detailed VM profiler for the Cog VM

Auteur(s) :

Kaleba, Sophie [Auteur]
Analyses and Languages Constructs for Object-Oriented Application Evolution [RMOD]
Bera, Clément [Auteur]
Analyses and Languages Constructs for Object-Oriented Application Evolution [RMOD]
Bergel, Alexandre [Auteur]
PLEIAD
Ducasse, Stephane [Auteur]
Analyses and Languages Constructs for Object-Oriented Application Evolution [RMOD]

Titre de la manifestation scientifique :

International Workshop on Smalltalk Technology IWST'17

Ville :

Maribor

Pays :

Slovénie

Date de début de la manifestation scientifique :

2017-09-04

Titre de la revue :

IWST '17 Proceedings of the 12th edition of the International Workshop on Smalltalk Technologies

Date de publication :

2017-09-04

Mot(s)-clé(s) en anglais :

Just in time compilation
Profiling tools
Virtual machine

Discipline(s) HAL :

Informatique [cs]/Langage de programmation [cs.PL]

Résumé en anglais : [en]

Code profiling enables a user to know where in an application or function the execution time is spent. The Pharo ecosystem offers several code profilers. However, most of the publicly available profilers (MessageTally, ...
Lire la suite >Code profiling enables a user to know where in an application or function the execution time is spent. The Pharo ecosystem offers several code profilers. However, most of the publicly available profilers (MessageTally, Spy, GadgetPro-filer) largely ignore the activity carried out by the virtual machine , thus incurring inaccuracy in the gathered information and missing important information, such as the Just-in-time compiler activity. This paper describes the motivations and the latest improvements carried out in VMProfiler, a code execution pro-filer hooked into the virtual machine, that performs its analysis by monitoring the virtual machine execution. These improvements address some limitations related to assessing the activity of native functions (resulting from a Just-in-time compiler operation): as of now, VMProfiler provides more detailed profiling reports, showing for native code functions in which bytecode range the execution time is spent.Lire moins >

Langue :

Anglais

Comité de lecture :

Oui

Audience :

Internationale

Vulgarisation :

Non

Collections :

• Centre de Recherche en Informatique, Signal et Automatique de Lille (CRIStAL) - UMR 9189

Source :

Harvested from HAL