Communication Dans Un Congrès Année : 2025

Towards optimal reconfigurable constant multipliers

Résumé

This paper introduces a novel algorithm for generating run-time reconfigurable single constant multipliers (RSCMs) which are optimal within their model in terms of hardware cost. Optimality is ensured by an exhaustive exploration of the design space mixing constraint programming, depth-first search, and branch-and-prune techniques. The cost model of previous works is also refined. Compared to the state of the art, this approach enables much larger constant sets, and also significantly improves the performance of the resulting architectures. Applications to neural network inference and small floating point multiplication units are evaluated.

Fichier principal
Vignette du fichier
2025-DSD-RSCM.pdf (341.47 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence

Dates et versions

hal-05094796 , version 1 (03-06-2025)
hal-05094796 , version 2 (09-07-2025)

Licence

Identifiants

Citer

Bastien Barbe, Xiao Peng, Anastasia Volkova, Florent de Dinechin. Towards optimal reconfigurable constant multipliers. 28th Conference on Digital System Design (DSD), Euromicro, Sep 2025, Salerno, Italy. pp.418-425, ⟨10.1109/DSD67783.2025.00064⟩. ⟨hal-05094796v2⟩
1283 Consultations
545 Téléchargements

Altmetric

Partager

  • More