Automatic implementation of arithmetic functions in high-level Synthesis

Liangwei Ge; Song Chen; Takeshi Yoshimura

doi:10.1109/ICSICT.2008.4735051

Automatic implementation of arithmetic functions in high-level Synthesis

Liangwei Ge^*, Song Chen, Takeshi Yoshimura

^*Corresponding author for this work

Graduate School of Information, Production and Systems

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

4 Citations (Scopus)

Abstract

Implementation of floating-point arithmetic functions is an essential task in high-level synthesis (HLS). However, most of the existing HLS EDA tools cannot well synthesize floating-point functions. In this paper, we present an automatic method to implement general floating-point functions using piecewise polynomial approximation. Based on the proposed hardware architecture, our method achieves compact implementations, which have low delay and high throughput. Experiment shows that compared with existing methods, the proposed method reduces register usage by about 10%.

Original language	English
Title of host publication	International Conference on Solid-State and Integrated Circuits Technology Proceedings, ICSICT
Pages	2349-2352
Number of pages	4
DOIs	https://doi.org/10.1109/ICSICT.2008.4735051
Publication status	Published - 2008
Event	2008 9th International Conference on Solid-State and Integrated-Circuit Technology, ICSICT 2008 - Beijing Duration: 2008 Oct 20 → 2008 Oct 23

Other

Other	2008 9th International Conference on Solid-State and Integrated-Circuit Technology, ICSICT 2008
City	Beijing
Period	08/10/20 → 08/10/23

ASJC Scopus subject areas

Electrical and Electronic Engineering
Condensed Matter Physics
Electronic, Optical and Magnetic Materials

Access to Document

10.1109/ICSICT.2008.4735051

Cite this

Ge, L, Chen, S & Yoshimura, T 2008, Automatic implementation of arithmetic functions in high-level Synthesis. in International Conference on Solid-State and Integrated Circuits Technology Proceedings, ICSICT., 4735051, pp. 2349-2352, 2008 9th International Conference on Solid-State and Integrated-Circuit Technology, ICSICT 2008, Beijing, 08/10/20. https://doi.org/10.1109/ICSICT.2008.4735051

@inproceedings{be04b2ccbbda4a1e9479905a1a5039be,

title = "Automatic implementation of arithmetic functions in high-level Synthesis",

abstract = "Implementation of floating-point arithmetic functions is an essential task in high-level synthesis (HLS). However, most of the existing HLS EDA tools cannot well synthesize floating-point functions. In this paper, we present an automatic method to implement general floating-point functions using piecewise polynomial approximation. Based on the proposed hardware architecture, our method achieves compact implementations, which have low delay and high throughput. Experiment shows that compared with existing methods, the proposed method reduces register usage by about 10%.",

author = "Liangwei Ge and Song Chen and Takeshi Yoshimura",

year = "2008",

doi = "10.1109/ICSICT.2008.4735051",

language = "English",

isbn = "9781424421855",

pages = "2349--2352",

booktitle = "International Conference on Solid-State and Integrated Circuits Technology Proceedings, ICSICT",

note = "2008 9th International Conference on Solid-State and Integrated-Circuit Technology, ICSICT 2008 ; Conference date: 20-10-2008 Through 23-10-2008",

}

TY - GEN

T1 - Automatic implementation of arithmetic functions in high-level Synthesis

AU - Ge, Liangwei

AU - Chen, Song

AU - Yoshimura, Takeshi

PY - 2008

Y1 - 2008

N2 - Implementation of floating-point arithmetic functions is an essential task in high-level synthesis (HLS). However, most of the existing HLS EDA tools cannot well synthesize floating-point functions. In this paper, we present an automatic method to implement general floating-point functions using piecewise polynomial approximation. Based on the proposed hardware architecture, our method achieves compact implementations, which have low delay and high throughput. Experiment shows that compared with existing methods, the proposed method reduces register usage by about 10%.

AB - Implementation of floating-point arithmetic functions is an essential task in high-level synthesis (HLS). However, most of the existing HLS EDA tools cannot well synthesize floating-point functions. In this paper, we present an automatic method to implement general floating-point functions using piecewise polynomial approximation. Based on the proposed hardware architecture, our method achieves compact implementations, which have low delay and high throughput. Experiment shows that compared with existing methods, the proposed method reduces register usage by about 10%.

UR - http://www.scopus.com/inward/record.url?scp=60649108464&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=60649108464&partnerID=8YFLogxK

U2 - 10.1109/ICSICT.2008.4735051

DO - 10.1109/ICSICT.2008.4735051

M3 - Conference contribution

AN - SCOPUS:60649108464

SN - 9781424421855

SP - 2349

EP - 2352

BT - International Conference on Solid-State and Integrated Circuits Technology Proceedings, ICSICT

T2 - 2008 9th International Conference on Solid-State and Integrated-Circuit Technology, ICSICT 2008

Y2 - 20 October 2008 through 23 October 2008

ER -

Automatic implementation of arithmetic functions in high-level Synthesis

Abstract

Other

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this