AndesCore™ AX45

AndesCore™ AX45 Overview

  • 64-bit in-order dual-issue 8-stage pipeline CPU architecture
  • AndeStar™ V5 Instruction Set Architecture (ISA)
    • Compliant to RISC-V RV64 GCBP little endian:
    • RV-GC: Integer, single/double precision floating point and 16-bit extensions
    • RV-B Bit manipulation extensions 
    • RV-P (draft) DSP/SIMD extensions
    • Andes V5 performance/code size extensions
  • 16/32-bit mixable instruction format for compacting code density
  • Advanced low power branch predication to speed up control code
  • Return Address Stack (RAS) to accelerate procedure returns
  • Memory Management Unit (MMU), Physical Memory Protection(PMP), and programmable Physical Memory Attribute (PMA)
  • MemBoost for heavy memory transactions
  • Flexibly configurable Platform-Level Interrupt Controller (PLIC) for supporting wide range of system event scenarios
  • Enhancement of vectored interrupt handling for real-time performance
  • Advanced CoDense™ technology to reduce program code size

The 64-bit AX45 is an 8-stage superscalar processor that supports RISC-V specification, including “G” (“IMAFD”) standard instructions, “C” 16-bit compression instructions, “P” Packed-SIMD/DSP instructions, “B” bit manipulation, and Andes performance enhancements, plus Andes Custom Extension™ (ACE) for user-defined instructions. It features MMU for Linux based applications, dynamic branch prediction for efficient branch execution, dual-issue of common instruction pairs, instruction and data caches and local memories for low-latency accesses. AX45 is also equipped with comprehensive SIMD/DSP instructions that can boost the performance of voice, audio, image and signal processing. Its ”B” extensions provide some combination of code size reduction, performance improvement, and energy reduction, and “FD” extensions support IEEE754-compliance single and double precision floating point instructions. AX45 incorporates MemBoost to greatly enhance memory bandwidth and reduce memory latencies for applications with intensive memory accesses. In addition, Other features include ECC for memory soft error protection, Platform-Level Interrupt Controller (PLIC) with enhancements for vectored dispatch and priority-based preemption, CoDense™ and StackSafe™ for software quality improvement, PowerBrake and WFI for power management.

Applications

  • Linux Operating System Based Applications
  • AR/VR
  • Networking and Communications
  • ADAS/IVI/V2X
  • Video and Image Processing
  • Machine/Deep Learning acceleration
  • Advanced High Capacity storage

Block Diagram

Development Tools

  • AndeSight™ IDE (Eclipse-based)
    • Compiler, Debugger, Profiler, Register Bit-field Display/Update, RTOS Awareness, and more
    • Tested platforms: Windows and Linux , single- or multi-user installation
  • Andes BSP
    • Optimized DSP library
    • Demo examples and sample projects
  • AndesClarity: Processor Pipeline Analyzer and Visualizer
  • COPILOT: Automation tool for Andes Custom Extension
  • RTOSes
    • Open-source: FreeRTOS, Zephyr, RT-Thread
    • Commercial: μC/OS-II, ThreadX,
  • Linux kernel and platform drivers
  • FPGA Development Boards
    • AndeShape™ ADP-XC7 (full-featured)
    • Xilinx VCU118
  • Debugging Hardware
    • AICE-MICRO and AICE-MINI+ 

Key Features and Performance

AndeStar™ V5 Architecture

Key FeaturesBenefits
RISC-V RV64GCPB ISA
  • State-of-the art ISA from latest developments in computer architecture
  • Industry standard and open architecture
Andes Custom Extension™ (ACE) option to create custom instructions for software acceleration
  • Add custom instruction extensions to facilitate Domain-Specific Architecture/Acceleration (DSA)
  • Boost application performance significantly, at the same time maintain the programmability
  • Powerful constructs are available to define high level instruction
  • ACE design is based on Verilog and C languages which are familiar to the designers
  • The COPILOT tool automatically generates the extended CPU and software toolchain
  • Do not require expertise in processor pipeline to design ACE instructions
Andes Extended InstructionsAndes exclusive performance and functionality enhancements
16/32-bit mixable instruction formatFor compact code density
32 general-purpose registersFor better code size and performance
Machine (M), User (U) and Supervisor (S) Privilege levelsFor Linux and advanced operating systems with protection between kernel and user programs

CPU Core

Key FeaturesBenefits
5.63 Coremark/MHz, 3.27 DMIPS/MHz*Excellent performance-per-MHz
8-stage dual-issue in-order pipelineSuperior performance-efficiency, while allowing for high speeds

Extensive branch predication features

  • Branch Target Buffer (BTB)
  • Branch History Table (BHT)
  • Return Address Stack (RAS)
  • Branch Target Buffer and Branch History Table to speed up control codes
  • Return Address Stack to speeds up procedure returns

Memory Management Unit

  • Sv39/Sv48 virtual-memory systems
  • 4/8-entry ITLB/DTLB
  • Up to 512-entry shared TLB
  • Virtual memory support for full address space and easy code/data sharing
  • Support for full-featured OS such as Linux
  • Protection of supervisor and user privilege
  • Hardware for fast address translation
Physical Memory Protection (PMP), configurable up to 32 regionsBasic read/write/execute memory protection with minimum cost
Programmable Physical Memory Attribute (PMA), configurable up to 16 regions

Configurable memory attributes:

  • Memory, I/O, None
  • Cacheable/Non-cacheable
  • Write-back/Write-through
  • Read/write/read & write allocate, no allocate
  • Access fault for non-existent regions
Performance monitorsProgram code performance tuning

Multiplier options

  • Fast multiplier: pipelined, 2-cycle
  • Small multipliers: producing 1, 2, 4, or 8 bits per cycle
Option to choose between speed and area according to application's requirements
PowerBrake technologyPerformance throttling to digitally reduce power consumption
StackSafe™ hardware stack protection
  • Easy identification of stack size threshold during development
  • Hardware error detection of stack overflow and underflow at runtime
QuickNap™ technologyFast power-down/wake-up support for caches

* BSP v5.1.0, DMIPS/MHZ follow Dhrystone’s no-inline ground rules, best performances 

Memory Subsystems

Key FeaturesBenefits

I-Cache & D-Cache

  • Size: 8KB to 64KB
  • Set associativity: Direct, 2-way or 4-way
  • Accelerating accesses to slow memories
  • Flexible cache configurations

ILM & DLM

  • Size: 4KB to 16M
  • SRAM or AXI interface support
  • Bus masters accesses by AXI slave port
  • For deterministic and efficient program execution
  • Flexible size selection to fit diversified needs
MemBoostOptimize cache reads and writes to achieve higher bandwidth and lower latency
Soft-error protection: parity for I-Cache, and ECC for D-Cache, ILM and DLMCode and data integrity protection
Bus master port: AXI with 64/128/256-bit data, I/D joint or separate busUser-selectable bus interface for optimal efficiency
Bus save port: AXI with 64/128/256-bit data, for ILM/DLM accessesEfficient data transfer between CPU and SoC masters
Core/bus clock ratio of N:1Simplified SoC integration
64-bit AXI Peripheral Port InterfaceFor latecny-sensitve peripheral

Platform-Level Interrupt Controller (PLIC)

Key FeaturesBenefits

Implements RISC-V PLIC specification

  • Up to 1023 PLIC interrupt sources
  • Up to 255 PLIC interrupt priority levels
  • Up to 16 PLIC interrupt targets
Allow individual interrupts to be serviced and prioritized without sharing

Enhanced interrupt features

  • Vectored interrupt dispatch
  • Priority-based preemption
  • Selectable edge trigger or level trigger
  • Faster interrupt handling for real-time applications
  • Complete hardware preemption support for faster response
  • Flexible interrupt source interface for simpler SoC design

Debug Support

Key FeaturesBenefits
Implements RISC-V debug specifications ver 0.13Supported by industry debug tool suppliers
JTAG Debug PortIndustry-standard support
Embedded Debug Module with up to 8 triggersFlexible configurations to tradeoff between gate count and debugging capabilities
Exception redirection supportEntering debugger upon selected exceptions without using breakpoints

Trace Support

Key FeaturesBenefits
Implements RISC-V Trace 1.0 Instruction Trace interfaceSupported by Andes tools

Performance

Core, Process AX45, 28nm
Frequency (MHz)1700
Dynamic power (uW/MHz)16.3
Area (mm2)0.055

TSMC 7nm FIN FET ULVT/LVT/SVT, cell height 240nm, High Speed L1 Cache Memory Compiler. Frequency condition: worst: SSGNP/0.675V/-40oc, typical: TT/0.75v/+85oc. Power and area : typical corner. Configurations: 256-entry BTB, PMP&PMA 16-entry, 32KB I/D$ (no Local Memory), Double precision FPU,  MemBoostw/o DSP, with I/O constraint; die area and power are core only, 65% utilization.

Product Package

AndesCore™ AX45 Single-core Processor with AE350 AXI Platform