Adopting a Multi-issue, Out-of-order RISC-V Core Cluster in a Complex Modern System-on-Chip

计算机科学 芯(光纤) 芯片上的系统 星团(航天器) 订单(交换) 炸薯条 多核 多核处理器 计算机体系结构 精简计算指令集 嵌入式系统 并行计算 指令集 操作系统 电信 业务 财务
作者
S Shrinidhi Rao,R Kavya,Alok Kumar,Sanjoy Kumar Saha,Moonki Jang,Garima Srivastava
标识
DOI:10.1109/iatmsi64286.2025.10984468
摘要

The increasing need for powerful, yet power efficient SoCs has driven the VLSI industry towards extremely complex application processors while maintaining control over rigorous power saving schemes. A major change in processor technology was triggered by the introduction of RISC cores in early 1980s, which were extremely efficient, compared to the dominant CISC architecture at the time. RISC based computers were simpler, faster and power efficient compared to CISC processors. Majority of modern System-on-Chips (SoC) in mobile and automotive industry are driven by RISC based CPUs. The cost involved in the usage of advances licensed RISC processors has pushed the industry and academics towards open source hardware design. Adopting RISC-V enables the design of specialized processors for efficient handling of target applications, through a flexible, scalable and highly customized processor architecture. In this research, we analyze the feasibility of adopting RISC-V processors in a complex heterogeneous multi-master SoC by replacing ARM based CPU cluster with a RISC-V processor cluster while retaining the existing ARM based debug architecture. The RISC-V processor used is a 64-bit multi-issue, out of order, pipelined processor capable of high performance with low power consumption. We outline the difficulties, drawbacks, and advantages of working with RISC-V from design and verification perspectives, serving as a reference for engineers looking to adopt such systems in their designs. We also highlight and discuss the performance metrics observed between the two SoCs.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
1秒前
周嘉伟发布了新的文献求助10
3秒前
大个的应助被东起欧采纳,获得10
3秒前
初景的应助被橘子女王采纳,获得20
4秒前
名称不是重点完成签到,获得积分10
5秒前
羊毛发布了新的文献求助10
5秒前
Ayiiiii完成签到 ,获得积分10
5秒前
6秒前
科研小白的应助被善良的孤风采纳,获得10
7秒前
7秒前
fcjnb的应助被Sindy采纳,获得10
8秒前
万能图书馆的应助被化学废材采纳,获得10
8秒前
Owen的应助被123采纳,获得10
8秒前
xiaoxiao完成签到 ,获得积分10
8秒前
钟江完成签到 ,获得积分10
10秒前
jj发布了新的文献求助10
10秒前
田様的应助被一啊鸭采纳,获得10
11秒前
Jasper的应助被隐形静芙采纳,获得10
11秒前
yygdfw发布了新的文献求助10
11秒前
12秒前
13秒前
科研的神龙猫完成签到,获得积分10
14秒前
www发布了新的文献求助10
14秒前
他化自在天完成签到,获得积分10
14秒前
15秒前
16秒前
17秒前
明朝发布了新的文献求助10
19秒前
那就这样发布了新的文献求助10
19秒前
21秒前
称心青亦发布了新的文献求助10
21秒前
Imp完成签到,获得积分10
22秒前
汉堡包的应助被zimengzhu采纳,获得10
23秒前
samgood发布了新的文献求助10
23秒前
一啊鸭发布了新的文献求助10
24秒前
24秒前
24秒前
小蘑菇的应助被化学废材采纳,获得10
24秒前
26秒前
TT发布了新的文献求助10
27秒前
高分求助中
(应助此贴封号)通过应助OA文献获取积分 10000
Composite Materials Handbook Volume 1 - Revision H 1500
Rosenblum, Global Change Biology 800
Computational Chemical Reaction Engineering: Modeling, Simulation, and Design with MATLAB 600
Organizational Behavior 510
Management and the Arts 510
Decentring Leadership 400
热门求助领域 (近24小时)
化学 材料科学 医学 生物 计算机科学 工程类 纳米技术 内科学 物理 有机化学 化学工程 生物化学 复合材料 光电子学 细胞生物学 心理学 量子力学 催化作用 物理化学 电极
热门帖子
关注 科研通微信公众号,转发送积分 7807875
求助须知:如何正确求助?哪些是违规求助? 9340452
关于积分的说明 20501769
捐赠科研通 7400081
什么是DOI,文献DOI怎么找? 3328507
关于科研通互助平台的介绍 2475397
邀请新用户注册赠送积分活动 2346859