RISCV架构的一种指令自动对齐电路*

刘德, 魏敬和, 高营

集成电路与嵌入式系统 ›› 2022, Vol. 22 ›› Issue (8) : 32-36.

PDF(2405 KB)
PDF(2405 KB)
集成电路与嵌入式系统 ›› 2022, Vol. 22 ›› Issue (8) : 32-36.
专题论述

RISCV架构的一种指令自动对齐电路*

  • 刘德, 魏敬和, 高营
作者信息 +

An Instruction Antomatic Aligning Circuit Based on RISCV

  • Liu De, Wei Jinghe, Gao Ying
Author information +
文章历史 +

摘要

基于RISC-V指令集架构实现了一种指令自动对齐电路。该电路可以将指令缓存发送过来的32位指令数据进行分解,从中解析出正确的指令。当指令缓存发送过来的32位指令数据对应的地址不是4字节对齐时(地址的低两位不是00)时,该电路可以自动将下一拍取指令数据的地址对齐到4字节,同时给出指令是否有效的指示标识;当指令地址4字节对齐时,该电路对指令数据进行分析,给出指令有效的标识、指令和对应指令的实际地址。给出的指令对齐电路的延时为4级两输入逻辑门,适用于高频标量处理器的前端取指令电路。

Abstract

Based on RISC-V instruction architecture,this article implements an instruction aligning circuit.This circuit can recognize and output the correct instruction by analyzing the 32-bit instruction data passed from the I-Cache.When the address of the 32-bit instruction data is not aligned with 4-Byte,say the two least significant bits of the address is not equal to 00,this circuit automatically aligns the address of the next instruction-fetching with 4-Byte and output the signal that indicating whether the instruction is valid or invalid.When the address of the instruction data is aligned with 4-Byte,this circuit presents the instruction,the real address of this instruction,and indicates that the instruction is valid.The latency of the circuit for aligning the instruction address to 4-Byte presented in this article is 4 levels of 2-inputs logic gates,which is very useful for the front-end instruction fetch circuit of high frequency scalar processor.

关键词

RISC-V / 指令对齐 / 微架构 / 内核流水线

Key words

RISC-V / instruction align / architecture / pipeline of core

引用本文

导出引用
刘德, 魏敬和, 高营. RISCV架构的一种指令自动对齐电路*[J]. 集成电路与嵌入式系统. 2022, 22(8): 32-36
Liu De, Wei Jinghe, Gao Ying. An Instruction Antomatic Aligning Circuit Based on RISCV[J]. Integrated Circuits and Embedded Systems. 2022, 22(8): 32-36
中图分类号: TP322   

参考文献

[1] A Waterman,Y Lee,D A Patterson,et al.The RISCV instruction set manual: User-level ISA, version 2.0[C]//Dept. Elect. Eng. Comput. Sci.,Univ. California,Berkeley, Berkeley,CA,USA,Tech. Rep.UCB/EECS-2016-118,2014,vol.1.
[2] Zaruba F,Benini L.The Cost of Application-Class Processing: Energy and Performance Analysis of a Linux-ready 1.7GHz 64bit RISC-V Core in 22nm FDSOI Technology[J]. IEEE Transactions on Very Large Scale Integration(VLSI) Systems,2019,27(11):2629-2640.
[3] K Patsidis,D Konstantinou,C Nicopoulos,et al.A low-cost synthesizable risc-v dual-issue processor core leveraging the compressed instruction set extension[J]. Microprocess. Microsyst.,2018,61(9):1-10.
[4] Celio C,Chiu P F,Asanovic K,et al.BROOM:An open-source out-of-order processor with resilient low-voltage operation in 28 nm CMOS[J].IEEE Micro,2019(99):1.
[5] Celio C,D A Patterson,K Asanovi'c.The Berkeley Out-of-Order Machine (BOOM):An Industry-Competitive,Synthesizable, Parameterized RISC-V Processor,2015.
[6] Y Lee .A 45 nm 1.3 GHz 16.7 double-precision GFLOPS/W RISC-V processor with vector accelerators[C]//in Proc. 40th Eur. Solid State Circuits Conf. (ESSCIRC),Sep.2014:199-202.
[7] Lee Y,Zimmer B,Waterman A,et al.Raven: A 28nm RISC-V vector processor with integrated switched-capacitor DC-DC converters and adaptive clocking[C]//2015 IEEE Hot Chips 27 Symposium (HCS).IEEE, 2015.
[8] B Keller,M Cochet,B Zimmer,et al.A RISC-V processor soc with integrated power management at submicrosecond timescales in 28 nm FD-SOI[J].Solid-State Circuits,2017,52(7):1863-1875.

基金

* 江苏省产业前瞻与关键核心技术重点项目(BE2019003)。

PDF(2405 KB)

Accesses

Citation

Detail

段落导航
相关文章

/