Loop aware IR-level annotation framework for performance estimation in native simulation - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2017

Loop aware IR-level annotation framework for performance estimation in native simulation

Résumé

Native simulation is an interesting virtual prototyping candidate to speed-up architecture exploration and early software developments. It however does not provide out-of-the box non-functional information needed for software performance estimation. Annotating software with information is complex as highlevel codes and binary codes have different structures due to compiler optimizations. This work proposes an annotation framework at compiler IR-level that focuses on loop structures, and reflects optimizations through a mapping scheme between the binary and the high-level IR. Experiments on instruction count show in average around 2% of error.
Fichier non déposé

Dates et versions

hal-01522712 , version 1 (15-05-2017)

Licence

Paternité - Pas d'utilisation commerciale

Identifiants

  • HAL Id : hal-01522712 , version 1

Citer

O. Matoussi, Frédéric Pétrot. Loop aware IR-level annotation framework for performance estimation in native simulation. 22nd Asia and South Pacific Design Automation Conference (ASPDAC 2017), Jan 2017, Chiba, Japan. pp.220-225. ⟨hal-01522712⟩

Collections

UGA CNRS TIMA
68 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More