Agile Design of a RISC-V Processor with Trigonometric Math Extensions Using SpinalHDL

The purpose of hardware agile design is to speed up chip prototype verification and shorten the design cycle. The key of it is higher hardware description level, more flexible design reuse and faster design iteration. This paper uses SpinalHDL to design a five-stage pipeline RISC-V processor and expands five trigonometric math instructions based on the CORDIC algorithm. Using SpinalHDL offers a shorter design cycle and better processor configurability. The resource of processor is evaluated on Xilinx Kintex-7 XC7K325T-2FFG900C and the precision of the extended instruction is evaluated. Compared with the unextended processor, the consumption of LUTs and Registers of the extended processor increases by 53.20% and 19.97% respectively. These functions achieve an average 382× speedup in execution cycles when implemented using extended instructions instead of the standard math library.

Authors

Institutions

Publication Details

Published
2024-11-01
DOI
https://doi.org/10.1109/asid63618.2024.10839653
Citations
2
Primary Topic
Numerical Methods and Algorithms
Type
article
Field-Weighted Citation Impact
0.72
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Agile Design of a RISC-V Processor with Trigonometric Math Extensions Using SpinalHDL

Duoli Zhang, Zhiqiang Zhou, Wei Ni, Xinlong Min et al.
2 citations
Numerical Methods and Algorithms
0.72
article

Agile Design of a RISC-V Processor with Trigonometric Math Extensions Using SpinalHDL

Duoli Zhang, Zhiqiang Zhou, Wei Ni, Xinlong Min, Yukun Song
article en
2 citations

Abstract

The purpose of hardware agile design is to speed up chip prototype verification and shorten the design cycle. The key of it is higher hardware description level, more flexible design reuse and faster design iteration. This paper uses SpinalHDL to design a five-stage pipeline RISC-V processor and expands five trigonometric math instructions based on the CORDIC algorithm. Using SpinalHDL offers a shorter design cycle and better processor configurability. The resource of processor is evaluated on Xilinx Kintex-7 XC7K325T-2FFG900C and the precision of the extended instruction is evaluated. Compared with the unextended processor, the consumption of LUTs and Registers of the extended processor increases by 53.20% and 19.97% respectively. These functions achieve an average 382× speedup in execution cycles when implemented using extended instructions instead of the standard math library.

Hefei University of Technology (CN)
Openalex Percentile: Top 24%
Numerical Methods and Algorithms
0.72
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.