CAWET: Context-Aware Worst-Case Execution Time Estimation Using Transformers - Archive ouverte HAL
Communication Dans Un Congrès Année : 2023

CAWET: Context-Aware Worst-Case Execution Time Estimation Using Transformers

Résumé

This paper presents CAWET, a hybrid worst-case program timing estimation technique. CAWET identifies the longest execution path using static techniques, whereas the worst-case execution time (WCET) of basic blocks is predicted using an advanced language processing technique called Transformer-XL. By employing Transformers-XL in CAWET, the execution context formed by previously executed basic blocks is taken into account, allowing for consideration of the microarchitecture of the processor pipeline without explicit modeling. Through a series of experiments on the TacleBench benchmarks, using different target processors (Arm Cortex M4, M7, and A53), our method is demonstrated to never underestimate WCETs and is shown to be less pessimistic than its competitors.
Fichier principal
Vignette du fichier
CAWET_ECRTS_LIPS.pdf (1.01 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04148587 , version 1 (03-07-2023)

Licence

Identifiants

Citer

Abderaouf N Amalou, Elisa Fromont, Isabelle Puaut. CAWET: Context-Aware Worst-Case Execution Time Estimation Using Transformers. ECRTS 2023 - 35th Euromicro Conference on Real-Time Systems, Jul 2023, Vienne, Austria. pp.7:1--7:20, ⟨10.4230/LIPIcs.ECRTS.2023.7⟩. ⟨hal-04148587⟩
206 Consultations
141 Téléchargements

Altmetric

Partager

More