Embedded Design Handbook

Download PDF
ID 683689
Date 8/28/2023
Public
Document Table of Contents
Document Table of Contents x
Product Discontinuance Notification 1. Introduction 2. First Time Designer's Guide 3. Hardware System Design with Intel® Quartus® Prime and Platform Designer 4. Software System Design with a Nios® II Processor 5. Nios® II Configuration and Booting Solutions 6. Nios® II Debug, Verification, and Simulation 7. Optimizing Nios® II Based Systems and Software
1. Introduction x
1.1. Document Revision History for Embedded Design Handbook
2. First Time Designer's Guide x
2.1. FPGAs and Soft-Core Processors 2.2. Embedded System Design 2.3. Embedded Design Resources 2.4. Intel Embedded Glossary 2.5. First Time Designer's Guide Revision History
2.3. Embedded Design Resources x
2.3.1. Intel Embedded Support 2.3.2. Intel Embedded Training 2.3.3. Intel Embedded Documentation 2.3.4. Third Party Intellectual Property
3. Hardware System Design with Intel® Quartus® Prime and Platform Designer x
3.1. FPGA Hardware Design 3.2. System Design with Platform Designer 3.3. Interfacing an External Processor to an Intel FPGA 3.4. Avalon-MM Byte Ordering 3.5. Memory System Design 3.6. Nios® II Hardware Development Tutorial 3.7. Platform Designer System Design Tutorial 3.8. Hardware System Design with Intel® Quartus® Prime and Platform Designer Revision History
3.1. FPGA Hardware Design x
3.1.1. Connecting Your FPGA Design to Your Hardware 3.1.2. Connecting Signals to your Platform Designer System 3.1.3. Constraining Your FPGA-Based Design
3.2. System Design with Platform Designer x
3.2.1. Intel System on a Programmable Chip (Platform Designer) Solutions 3.2.2. Platform Designer Design
3.3. Interfacing an External Processor to an Intel FPGA x
3.3.1. Configuration Options 3.3.2. RapidIO Interface 3.3.3. PCI Express* Interface 3.3.4. PCI Interface 3.3.5. Serial Protocol Interface (SPI) 3.3.6. Custom Bridge Interfaces
3.4. Avalon-MM Byte Ordering x
3.4.1. Endianness 3.4.2. Avalon® -MM Interface Ordering 3.4.3. Nios® II Processor Data Accesses 3.4.4. Adapting Processor Masters to be Avalon-MM Compliant 3.4.5. System-Wide Design Recommendations
3.4.1. Endianness x
3.4.1.1. Hardware Endianness 3.4.1.2. Software Endianness
3.4.2. Avalon® -MM Interface Ordering x
3.4.2.1. Dynamic Bus Sizing DMA Examples
3.4.4. Adapting Processor Masters to be Avalon-MM Compliant x
3.4.4.1. PowerPC Bus Byte Ordering 3.4.4.2. ARM BE-32 Bus Byte Ordering 3.4.4.3. ARM BE-8 Bus Byte Ordering 3.4.4.4. Other Processor Bit and Byte Orders 3.4.4.5. Arithmetic Byte Reordering
3.4.5. System-Wide Design Recommendations x
3.4.5.1. System-Wide Arithmetic Byte Ordering 3.4.5.2. System-Wide Arithmetic Byte Reordering in Software
3.5. Memory System Design x
3.5.1. Memory Types 3.5.2. On-Chip Memory 3.5.3. External SRAM 3.5.4. Flash Memory 3.5.5. SDRAM 3.5.6. Case Study
3.5.1. Memory Types x
3.5.1.1. Volatile Memory 3.5.1.2. Non-Volatile Memory
3.5.2. On-Chip Memory x
3.5.2.1. Advantages 3.5.2.2. Disadvantages 3.5.2.3. Best Applications 3.5.2.4. Poor Applications 3.5.2.5. On-Chip Memory Types 3.5.2.6. Best Practices
3.5.2.3. Best Applications x
3.5.2.3.1. Cache 3.5.2.3.2. Tightly Coupled Memory 3.5.2.3.3. Look Up Tables 3.5.2.3.4. FIFO
3.5.3. External SRAM x
3.5.3.1. Advantages 3.5.3.2. Disadvantages 3.5.3.3. Best Applications 3.5.3.4. Poor Applications 3.5.3.5. External SRAM Types 3.5.3.6. Best Practices
3.5.4. Flash Memory x
3.5.4.1. Advantages 3.5.4.2. Disadvantages 3.5.4.3. Typical Applications 3.5.4.4. Poor Applications 3.5.4.5. Flash Types
3.5.5. SDRAM x
3.5.5.1. Advantages 3.5.5.2. Disadvantages 3.5.5.3. Best Applications 3.5.5.4. Poor Applications 3.5.5.5. SDRAM Types 3.5.5.6. SDRAM Controller Types Available From Intel 3.5.5.7. Best Practices
3.5.5.7. Best Practices x
3.5.5.7.1. Half-Rate Mode 3.5.5.7.2. Full-Rate Mode 3.5.5.7.3. Sequential Access 3.5.5.7.4. Bursting 3.5.5.7.5. SDRAM Minimum Frequency 3.5.5.7.6. SDRAM Device Speed
3.5.6. Case Study x
3.5.6.1. Application Description 3.5.6.2. Initial Memory Partitioning 3.5.6.3. Optimized Memory Partitioning
3.5.6.3. Optimized Memory Partitioning x
3.5.6.3.1. Add an External SRAM for Input Buffers 3.5.6.3.2. Add On-Chip Memory for Video Line Buffers
3.6. Nios® II Hardware Development Tutorial x
3.6.1. Software and Hardware Requirements 3.6.2. Intel® FPGA IP Evaluation Mode 3.6.3. Nios® II Design Example 3.6.4. Nios® II System Development Flow 3.6.5. Creating the Design Example
3.6.4. Nios® II System Development Flow x
3.6.4.1. Analyzing System Requirements 3.6.4.2. Defining and Generating the System in Platform Designer 3.6.4.3. Integrating the Platform Designer System into the Intel® Quartus® Prime Project 3.6.4.4. Developing Software with the Nios® II Software Build Tools for Eclipse 3.6.4.5. Running and Debugging Software on the Target Board 3.6.4.6. Varying the Development Flow
3.6.5. Creating the Design Example x
3.6.5.1. Install the Design Files 3.6.5.2. Analyze System Requirements 3.6.5.3. Start the Intel® Quartus® Prime Software and Open the Example Project 3.6.5.4. Create a New Platform Designer System 3.6.5.5. Define the System in Platform Designer 3.6.5.6. Integrate the Platform Designer System into the Intel® Quartus® Prime Project 3.6.5.7. Download the Hardware Design to the Target FPGA 3.6.5.8. Develop Software Using the Nios® II SBT for Eclipse 3.6.5.9. Run the Program on Target Hardware
3.6.5.5. Define the System in Platform Designer x
3.6.5.5.1. Specify Target FPGA and Clock Settings 3.6.5.5.2. Add the On-Chip Memory 3.6.5.5.3. Add the Nios® II Processor Core 3.6.5.5.4. Add the JTAG UART 3.6.5.5.5. Add the Interval Timer 3.6.5.5.6. Add the System ID Peripheral 3.6.5.5.7. Add the PIO 3.6.5.5.8. Specify Base Addresses and Interrupt Request Priorities 3.6.5.5.9. Generate the Platform Designer System
3.6.5.6. Integrate the Platform Designer System into the Intel® Quartus® Prime Project x
3.6.5.6.1. Instantiate the Platform Designer System Module in the Intel® Quartus® Prime Project 3.6.5.6.2. Add IP Variation File 3.6.5.6.3. Assign FPGA Device 3.6.5.6.4. Assign FPGA Pin Locations 3.6.5.6.5. Set Timing 3.6.5.6.6. Compile the Intel® Quartus® Prime Project and Verify Timing
3.6.5.8. Develop Software Using the Nios® II SBT for Eclipse x
3.6.5.8.1. Create a New Nios® II Application and BSP from Template 3.6.5.8.2. Compile the Project
3.7. Platform Designer System Design Tutorial x
3.7.1. Software and Hardware Requirements 3.7.2. Download and Install the Tutorial Design Files 3.7.3. Open the Tutorial Project 3.7.4. Creating Platform Designer Systems 3.7.5. Assemble a Hierarchical System 3.7.6. Viewing the Memory Tester System in Platform Designer 3.7.7. Compiling and Downloading Software to a Development Board 3.7.8. Debugging Your Design 3.7.9. Verifying Hardware in System Console 3.7.10. Simulating Custom Components 3.7.11. View a Diagram of the Completed System
3.7.4. Creating Platform Designer Systems x
3.7.4.1. Create a Data Pattern Generator Platform Designer System 3.7.4.2. Create a Data Pattern Checker Platform Designer System
3.7.4.1. Create a Data Pattern Generator Platform Designer System x
3.7.4.1.1. Create a New Platform Designer System and Set up the Clock Source 3.7.4.1.2. Add a Pipeline Bridge 3.7.4.1.3. Add a Custom Pattern Generator 3.7.4.1.4. Add a PRBS Pattern Generator 3.7.4.1.5. Add a Two-to-One Streaming Multiplexer 3.7.4.1.6. Verify the Memory Address Map 3.7.4.1.7. Connect the Reset Signals 3.7.4.1.8. Save the System
3.7.4.2. Create a Data Pattern Checker Platform Designer System x
3.7.4.2.1. Create a New Platform Designer System and Set Up the Clock Source 3.7.4.2.2. Add a Pipeline Bridge 3.7.4.2.3. Add a Custom Pattern Checker 3.7.4.2.4. Add the PRBS Pattern Checker 3.7.4.2.5. Add a One-to-Two Streaming Demultiplexer 3.7.4.2.6. Verify the Memory Address Map 3.7.4.2.7. Connect the Reset Signals 3.7.4.2.8. Save the System
3.7.5. Assemble a Hierarchical System x
3.7.5.1. Create the Hierarchical Memory Tester System 3.7.5.2. Complete the Top-Level System
3.7.5.1. Create the Hierarchical Memory Tester System x
3.7.5.1.1. Add the Pattern Generator 3.7.5.1.2. Add the Pattern Checker 3.7.5.1.3. Add Memory Master Components 3.7.5.1.4. Connect the Reset Signals 3.7.5.1.5. Verify the Memory Address Map 3.7.5.1.6. Save the System
3.7.9. Verifying Hardware in System Console x
3.7.9.1. Open the Tutorial Project 3.7.9.2. Add the JTAG-to- Avalon® Master Bridge 3.7.9.3. Debug with System Console
3.7.10. Simulating Custom Components x
3.7.10.1. Generate a Testbench System in Platform Designer 3.7.10.2. Run Simulation In the ModelSim Software
3.7.10.1. Generate a Testbench System in Platform Designer x
3.7.10.1.1. Create a New Platform Designer System for the Design Under Test 3.7.10.1.2. Export Design Under Test 3.7.10.1.3. Generate a Testbench System 3.7.10.1.4. Generate Testbench System's Simulation Models
3.7.10.2. Run Simulation In the ModelSim Software x
3.7.10.2.1. Set Up the Simulation Environment 3.7.10.2.2. Run the Simulation
4. Software System Design with a Nios® II Processor x
4.1. Nios® II Command-Line Tools 4.2. Developing Nios® II Software 4.3. Nios® II MPU Usage 4.4. Profiling Nios® II Systems 4.5. Software System Design with a Nios® II Processor Revision History
4.1. Nios® II Command-Line Tools x
4.1.1. Intel Command-Line Tools for Board Bringup and Diagnostics 4.1.2. Intel Command-Line Tools for Flash Programming 4.1.3. Intel Command-Line Tools for Software Development and Debug 4.1.4. Intel Command-Line Nios® II Software Build Tools 4.1.5. Rebuilding Software from the Command Line 4.1.6. GNU Command-Line Tools
4.1.1. Intel Command-Line Tools for Board Bringup and Diagnostics x
4.1.1.1. jtagconfig 4.1.1.2. nios2-configure-sof 4.1.1.3. system-console
4.1.1.1. jtagconfig x
4.1.1.1.1. jtagconfig Usage Example
4.1.1.2. nios2-configure-sof x
4.1.1.2.1. nios2-configure-sof Usage Example
4.1.2. Intel Command-Line Tools for Flash Programming x
4.1.2.1. nios2-flash-programmer 4.1.2.2. elf2flash, bin2flash, and sof2flash
4.1.2.1. nios2-flash-programmer x
4.1.2.1.1. nios2-flash-programmer Usage Example
4.1.2.2. elf2flash, bin2flash, and sof2flash x
4.1.2.2.1. bin2flash Usage Example
4.1.3. Intel Command-Line Tools for Software Development and Debug x
4.1.3.1. nios2-terminal 4.1.3.2. nios2-download 4.1.3.3. nios2-stackreport 4.1.3.4. validate_zip 4.1.3.5. nios2-gdb-server
4.1.3.2. nios2-download x
4.1.3.2.1. nios2-download Usage Example
4.1.3.3. nios2-stackreport x
4.1.3.3.1. nios2-stackreport Usage Example
4.1.3.4. validate_zip x
4.1.3.4.1. validate_zip Usage Example
4.1.3.5. nios2-gdb-server x
4.1.3.5.1. nios2-gdb-server Usage Example
4.1.4. Intel Command-Line Nios® II Software Build Tools x
4.1.4.1. BSP Related Tools 4.1.4.2. Application Related Tools
4.1.6. GNU Command-Line Tools x
4.1.6.1. nios2-elf-addr2line 4.1.6.2. nios2-elf-gdb 4.1.6.3. nios2-elf-readelf 4.1.6.4. nios2-elf-ar 4.1.6.5. Linker 4.1.6.6. nios2-elf-size 4.1.6.7. nios2-elf-strings 4.1.6.8. nios2-elf-strip 4.1.6.9. nios2-elf-gdbtui 4.1.6.10. nios2-elf-gprof 4.1.6.11. nios2-elf-gcc and g++ 4.1.6.12. nios2-elf-c++filt 4.1.6.13. nios2-elf-nm 4.1.6.14. nios2-elf-objcopy 4.1.6.15. nios2-elf-objdump 4.1.6.16. nios2-elf-ranlib
4.1.6.1. nios2-elf-addr2line x
4.1.6.1.1. nios2-elf-addr2line Usage Example
4.1.6.3. nios2-elf-readelf x
4.1.6.3.1. nios2-elf-readelf Usage Example
4.1.6.4. nios2-elf-ar x
4.1.6.4.1. nios2-elf-ar Usage Example
4.1.6.5. Linker x
4.1.6.5.1. Linker Usage Example
4.1.6.6. nios2-elf-size x
4.1.6.6.1. nios2-elf-size Usage Example
4.1.6.7. nios2-elf-strings x
4.1.6.7.1. nios2-elf-strings Usage Example
4.1.6.8. nios2-elf-strip x
4.1.6.8.1. nios2-elf-strip Usage Example 4.1.6.8.2. nios2-elf-strip Usage Notes
4.1.6.11. nios2-elf-gcc and g++ x
4.1.6.11.1. Compilation Command Usage Example 4.1.6.11.2. More Complex Compilation Example
4.1.6.12. nios2-elf-c++filt x
4.1.6.12.1. nios2-elf-c++filt Usage Example 4.1.6.12.2. More Complex nios2-elf-c++filt Example
4.1.6.13. nios2-elf-nm x
4.1.6.13.1. nios2-elf-nm Usage Example 4.1.6.13.2. More Complex nios2-elf-nm Example
4.1.6.14. nios2-elf-objcopy x
4.1.6.14.1. nios2-elf-objcopy Usage Example
4.1.6.15. nios2-elf-objdump x
4.1.6.15.1. nios2-elf-objdump Usage Description
4.2. Developing Nios® II Software x
4.2.1. Software Development Cycle 4.2.2. Software Project Mechanics 4.2.3. Developing With the Hardware Abstraction Layer 4.2.4. Linking Applications
4.2.1. Software Development Cycle x
4.2.1.1. Nios® II Software Design 4.2.1.2. Nios® II Software Development Process
4.2.1.1. Nios® II Software Design x
4.2.1.1.1. Nios® II Tools Overview 4.2.1.1.2. Nios® II Software Build Tools
4.2.2. Software Project Mechanics x
4.2.2.1. Software Tools Background 4.2.2.2. Development Flow Guidelines 4.2.2.3. Nios® II Software Build Tools 4.2.2.4. Configuring BSP and Application Projects 4.2.2.5. Ensuring Software Project Coherency
4.2.2.3. Nios® II Software Build Tools x
4.2.2.3.1. The Nios® II Software Build Tools for Eclipse 4.2.2.3.2. The Nios® II Software Build Tools Command Line
4.2.2.4. Configuring BSP and Application Projects x
4.2.2.4.1. Software Example Designs 4.2.2.4.2. Selecting the Operating System (HAL versus MicroC/OS-II RTOS) 4.2.2.4.3. Configuring the BSP Project 4.2.2.4.4. Configuring the Application Project 4.2.2.4.5. Makefiles and the Nios® II Software Build Tools for Eclipse 4.2.2.4.6. Building and Running the Software in Nios® II Software Build Tools for Eclipse
4.2.2.5. Ensuring Software Project Coherency x
4.2.2.5.1. Recommended Development Practice 4.2.2.5.2. Recommended Architecture Practice
4.2.3. Developing With the Hardware Abstraction Layer x
4.2.3.1. Overview of the HAL 4.2.3.2. System Startup in HAL-Based Applications 4.2.3.3. HAL Peripheral Services 4.2.3.4. Accessing Memory With the Nios® II Processor 4.2.3.5. Handling Exceptions 4.2.3.6. Modifying the Exception Handler
4.2.3.1. Overview of the HAL x
4.2.3.1.1. HAL Configuration Options 4.2.3.1.2. Configuring the Boot Environment 4.2.3.1.3. Controlling HAL Initialization 4.2.3.1.4. Minimizing the Code Footprint and Increasing Performance 4.2.3.1.5. Configuring Peripherals and Services
4.2.3.2. System Startup in HAL-Based Applications x
4.2.3.2.1. System Initialization 4.2.3.2.2. crt0 Initialization 4.2.3.2.3. HAL Initialization
4.2.3.3. HAL Peripheral Services x
4.2.3.3.1. Timers 4.2.3.3.2. Character Mode Devices 4.2.3.3.3. Flash Memory Devices 4.2.3.3.4. Direct Memory Access Devices 4.2.3.3.5. Files and File Systems 4.2.3.3.6. Unsupported Devices
4.2.3.4. Accessing Memory With the Nios® II Processor x
4.2.3.4.1. Creating General C/C++ Applications 4.2.3.4.2. Accessing Peripherals 4.2.3.4.3. Sharing Uncached Memory 4.2.3.4.4. Sharing Memory With Cache Performance Benefits
4.2.4. Linking Applications x
4.2.4.1. Background 4.2.4.2. Linker Sections and Application Configuration 4.2.4.3. HAL Linking Behavior
4.2.4.3. HAL Linking Behavior x
4.2.4.3.1. Default BSP Linking 4.2.4.3.2. User-Controlled BSP Linking
4.3. Nios® II MPU Usage x
4.3.1. Requirements 4.3.2. General Usage 4.3.3. Nios® II MPU Design Examples
4.3.2. General Usage x
4.3.2.1. Adding the MPU Hardware 4.3.2.2. Writing Software for the MPU
4.3.2.2. Writing Software for the MPU x
4.3.2.2.1. MPU Programming Guidelines 4.3.2.2.2. Operating Systems and the MPU 4.3.2.2.3. MPU Register Details 4.3.2.2.4. Flow Summary
4.3.3. Nios® II MPU Design Examples x
4.3.3.1. Working with the MPU Design Examples 4.3.3.2. Example Hardware 4.3.3.3. Region Layout Considerations 4.3.3.4. Software
4.3.3.4. Software x
4.3.3.4.1. MPU Utilities 4.3.3.4.2. Building the Software
4.4. Profiling Nios® II Systems x
4.4.1. Requirements 4.4.2. Tools 4.4.3. Using the GNU Profiler to Measure Code Performance 4.4.4. Using Performance Counter and Timer Components 4.4.5. Troubleshooting
4.4.1. Requirements x
4.4.1.1. Obtaining the Hardware Design 4.4.1.2. Obtaining the Software Examples
4.4.2. Tools x
4.4.2.1. GNU Profiler 4.4.2.2. Intel Performance Counter 4.4.2.3. High-Resolution Timer 4.4.2.4. Program Counter Trace Information
4.4.3. Using the GNU Profiler to Measure Code Performance x
4.4.3.1. GNU Profiler Advantages 4.4.3.2. GNU Profiler Limitations 4.4.3.3. Software Considerations 4.4.3.4. Hardware Considerations 4.4.3.5. Tutorial: Using the GNU Profiler
4.4.3.3. Software Considerations x
4.4.3.3.1. Profiler Mechanics 4.4.3.3.2. Profiler Overhead
4.4.3.5. Tutorial: Using the GNU Profiler x
4.4.3.5.1. Profiler Example with the Nios® II Command Line 4.4.3.5.2. Profiler Example with Nios® II SBT for Eclipse 4.4.3.5.3. Analyzing the GNU Profiler Report
4.4.4. Using Performance Counter and Timer Components x
4.4.4.1. Performance Counter Advantages 4.4.4.2. Timer Advantages 4.4.4.3. Performance Counter and Timer Hardware Considerations 4.4.4.4. Performance Counter and Timer Software Considerations 4.4.4.5. Performance Counter Software Considerations 4.4.4.6. The Global Counter 4.4.4.7. Hardware Considerations 4.4.4.8. Tutorial: Using Performance Counters and Timers
4.4.4.8. Tutorial: Using Performance Counters and Timers x
4.4.4.8.1. Modifying the Nios® II Hardware Design 4.4.4.8.2. Programming the Hardware Design to Your Device 4.4.4.8.3. Performance Counter Example with the Nios® II Command Line 4.4.4.8.4. Performance Counter Example with Nios® II SBT for Eclipse 4.4.4.8.5. Analyzing the Performance Counter Report
4.4.5. Troubleshooting x
4.4.5.1. nios2-elf-gprof –annotated-source Switch Has No Effect 4.4.5.2. Writing to the Registers of a Nonexistent Section Counter 4.4.5.3. Output From a printf() or perf_print_formatted_output() Call Near the End 4.4.5.4. Fitting a Performance Counter in a Hardware Design That Consumes Most 4.4.5.5. The Histogram for the gmon.out File Is Missing, Even Though My main()
5. Nios® II Configuration and Booting Solutions x
5.1. Introduction 5.2. Nios® II Processor Booting Methods 5.3. Alternative Nios® II Boot Methods 5.4. Boot Time Performance Analysis 5.5. Nios® II Configuration and Booting Solutions Revision History
5.1. Introduction x
5.1.1. Prerequisites
5.2. Nios® II Processor Booting Methods x
5.2.1. Introduction to Nios® II Booting Methods 5.2.2. Nios® II Processor Booting from On-Chip Flash (UFM) 5.2.3. Nios® II Processor Booting from EPCQ Flash 5.2.4. Nios® II Processor Booting from QSPI Flash 5.2.5. Nios® II Processor Booting from On-Chip Memory (OCRAM) 5.2.6. Nios® II Processor Booting from CFI Flash 5.2.7. Summary of Nios® II Processor Vector Configurations and BSP Settings
5.2.1. Introduction to Nios® II Booting Methods x
5.2.1.1. Nios® II Processor Application Execute-In-Place from Boot Flash 5.2.1.2. Nios® II Processor Application Copied from Boot Flash to RAM Using Boot Copier 5.2.1.3. Nios® II Processor Application Execute-In-Place from OCRAM
5.2.1.1. Nios® II Processor Application Execute-In-Place from Boot Flash x
5.2.1.1.1. alt_load ()
5.2.1.2. Nios® II Processor Application Copied from Boot Flash to RAM Using Boot Copier x
5.2.1.2.1. Memcpy-based Boot Copier
5.2.2. Nios® II Processor Booting from On-Chip Flash (UFM) x
5.2.2.1. Intel® MAX® 10 FPGA On-Chip Flash Description 5.2.2.2. Nios® II Processor Application Execute-In-Place from UFM 5.2.2.3. Nios® II Processor Application Copied from UFM to RAM using Boot Copier
5.2.2.2. Nios® II Processor Application Execute-In-Place from UFM x
5.2.2.2.1. Hardware Design 5.2.2.2.2. Application 5.2.2.2.3. Programming
5.2.2.3. Nios® II Processor Application Copied from UFM to RAM using Boot Copier x
5.2.2.3.1. Hardware Design 5.2.2.3.2. Application 5.2.2.3.3. Programming
5.2.3. Nios® II Processor Booting from EPCQ Flash x
5.2.3.1. Intel FPGA Serial Flash Controller (EPCQ) Overview 5.2.3.2. Nios® II Processor Design, Configuration, and Boot Flow 5.2.3.3. Nios® II Processor Application Execute-In-Place from EPCQ Flash 5.2.3.4. Nios® II Processor Application Copied from EPCQ Flash to RAM Using Boot Copier 5.2.3.5. EPCQ HAL Driver
5.2.3.3. Nios® II Processor Application Execute-In-Place from EPCQ Flash x
5.2.3.3.15.2.3.4.1. Hardware Design5.2.3.3.15.2.3.4.1. Hardware Design 5.2.3.3.2. Application 5.2.3.3.3. Programming
5.2.3.4. Nios® II Processor Application Copied from EPCQ Flash to RAM Using Boot Copier x
5.2.3.3.15.2.3.4.1. Hardware Design5.2.3.3.15.2.3.4.1. Hardware Design 5.2.3.4.2. Application 5.2.3.4.3. Programming
5.2.4. Nios® II Processor Booting from QSPI Flash x
5.2.4.1. Nios® II Processor Design, Configuration and Boot Flow 5.2.4.2. Nios® II Processor Application Executes In-Place from General Purpose QSPI Flash ( Intel® MAX® 10) 5.2.4.3. Nios® II Processor Application Copied from General Purpose QSPI Flash to RAM Using Boot Copier ( Intel® MAX® 10) 5.2.4.4. Nios® II Processor Application Executes In-Place from Configuration QSPI Flash (Other FPGA devices) 5.2.4.5. Nios® II Processor Application Copied from Configuration QSPI Flash to RAM Using Boot Copier (Other FPGA devices)
5.2.4.2. Nios® II Processor Application Executes In-Place from General Purpose QSPI Flash ( Intel® MAX® 10) x
5.2.4.2.1. Hardware Design 5.2.4.2.2. Application 5.2.4.2.3. Programming Files Generation 5.2.4.2.4. QSPI Flash Programming
5.2.4.3. Nios® II Processor Application Copied from General Purpose QSPI Flash to RAM Using Boot Copier ( Intel® MAX® 10) x
5.2.4.3.1. Hardware Design 5.2.4.3.2. Application 5.2.4.3.3. Programming Files Generation 5.2.4.3.4. QSPI Flash Programming
5.2.4.4. Nios® II Processor Application Executes In-Place from Configuration QSPI Flash (Other FPGA devices) x
5.2.4.4.1. Hardware Design 5.2.4.4.2. Application 5.2.4.4.3. Programming Files Generation 5.2.4.4.4. QSPI Flash Programming
5.2.4.5. Nios® II Processor Application Copied from Configuration QSPI Flash to RAM Using Boot Copier (Other FPGA devices) x
5.2.4.5.1. Hardware Design 5.2.4.5.2. Application 5.2.4.5.3. Programming Files Generation 5.2.4.5.4. QSPI Flash Programming
5.2.5. Nios® II Processor Booting from On-Chip Memory (OCRAM) x
5.2.5.1. Nios® II Processor Application Executes in-place from OCRAM
5.2.5.1. Nios® II Processor Application Executes in-place from OCRAM x
5.2.5.1.1. Hardware Design 5.2.5.1.2. Application 5.2.5.1.3. Programming
5.2.6. Nios® II Processor Booting from CFI Flash x
5.2.6.1. Description of the Avalon® Tri-State Conduit Components 5.2.6.2. Nios® II Processor Application Execute-In-Place from CFI Flash 5.2.6.3. Nios® II Processor Application Copied from CFI Flash to RAM using Boot Copier
5.2.6.2. Nios® II Processor Application Execute-In-Place from CFI Flash x
5.2.6.2.15.2.6.3.1. Hardware Design5.2.6.2.15.2.6.3.1. Hardware Design Application 5.2.6.2.3. Programming
5.2.6.3. Nios® II Processor Application Copied from CFI Flash to RAM using Boot Copier x
5.2.6.2.15.2.6.3.1. Hardware Design5.2.6.2.15.2.6.3.1. Hardware Design Application 5.2.6.3.3. Programming
5.3. Alternative Nios® II Boot Methods x
5.3.1. Assumptions About the Reader 5.3.2. Implementing a Custom Boot Copier 5.3.3. Default Nios® II Boot Copier 5.3.4. Advanced Boot Copier Example 5.3.5. Implementing the Advanced Boot Copier Example 5.3.6. Small Boot Copier Example 5.3.7. Implementing the Small Boot Copier Example 5.3.8. Debugging Boot Copiers 5.3.9. Externally Controlling the Nios® II Boot Process
5.3.2. Implementing a Custom Boot Copier x
5.3.2.1. Hardware Design Files 5.3.2.2. Software Files
5.3.3. Default Nios® II Boot Copier x
5.3.3.1. Overview of the Default Nios® II Boot Copier 5.3.3.2. The Default CFI Flash Boot Copier 5.3.3.3. The Default EPCS/EPCQ Boot Copier
5.3.4. Advanced Boot Copier Example x
5.3.4.1. Driver Initialization 5.3.4.2. Printing to the JTAG UART 5.3.4.3. Boot Images 5.3.4.4. Boot Methods 5.3.4.5. Preventing Overlapping Data in Flash 5.3.4.6. Boot Copier Code Size
5.3.4.2. Printing to the JTAG UART x
5.3.4.2.1. Preventing Stalls by the JTAG UART 5.3.4.2.2. Reducing Memory Use for Printing
5.3.4.3. Boot Images x
5.3.4.3.1. Boot Image Format 5.3.4.3.2. Boot Image Header Format 5.3.4.3.3. Boot Record Format 5.3.4.3.4. Choosing a Boot Image 5.3.4.3.5. Word Alignment
5.3.4.4. Boot Methods x
5.3.4.4.1. Booting Directly From CFI Flash 5.3.4.4.2. Booting From CFI Flash, Running From On-Chip Memory 5.3.4.4.3. Booting From EPCS/EPCQ Flash, Running From On-Chip Memory 5.3.4.4.4. Setting the Boot Method
5.3.4.5. Preventing Overlapping Data in Flash x
5.3.4.5.1. Overlapping Data in CFI Flash 5.3.4.5.2. Overlapping Data in EPCS/EPCQ Flash
5.3.5. Implementing the Advanced Boot Copier Example x
5.3.5.1. Setting Up the Software Tools and Development Board 5.3.5.2. Creating a Suitable Hardware Design 5.3.5.3. Building the Advanced Boot Copier 5.3.5.4. Building a Test Application to Boot 5.3.5.5. Packing the Test Application in a Boot Record 5.3.5.6. Booting Directly From CFI Flash Memory 5.3.5.7. Booting CFI or EPCS/EPCQ Flash From On-Chip Memory 5.3.5.8. Running the Advanced Boot Copier Example
5.3.5.2. Creating a Suitable Hardware Design x
5.3.5.2.1. Opening the Example Project 5.3.5.2.2. Adding On-chip Boot ROM to the System
5.3.6. Small Boot Copier Example x
5.3.6.1. Small Boot Copier Features 5.3.6.2. Implementation in Nios® II Assembly Language 5.3.6.3. System Initialization 5.3.6.4. Code Size
5.3.7. Implementing the Small Boot Copier Example x
5.3.7.1. Setting Up the Software Tools and Development Board 5.3.7.2. Creating a Suitable Hardware Design 5.3.7.3. Building the Small Boot Copier Using 'make' 5.3.7.4. Building a Test Application to Boot 5.3.7.5. Booting From On-Chip Memory 5.3.7.6. Running the Small Boot Copier Example
5.3.8. Debugging Boot Copiers x
5.3.8.1. Hardware and Software Breakpoints 5.3.8.2. Enabling Hardware Breakpoints 5.3.8.3. Breaking Before main() 5.3.8.4. Setting Up the Debugger
5.3.9. Externally Controlling the Nios® II Boot Process x
5.3.9.1. Overview 5.3.9.2. Building an Appropriate Platform Designer System 5.3.9.3. The Boot Process
5.3.9.2. Building an Appropriate Platform Designer System x
5.3.9.2.1. External Processor Bridge 5.3.9.2.2. The cpu_resetrequest Signal 5.3.9.2.3. Nios® II Reset Address 5.3.9.2.4. One-Bit PIO Peripheral
5.3.9.3. The Boot Process x
5.3.9.3.1. Boot Images 5.3.9.3.2. Example C Code 5.3.9.3.3. External Boot Flow
5.4. Boot Time Performance Analysis x
5.4.1. Boot Time Performance Analysis Design Example 5.4.2. Boot Time Measurement Strategy 5.4.3. Reducing Nios® II Boot Time in Intel® MAX® 10 FPGA Design 5.4.4. Boot Time Performance and Estimation
5.4.3. Reducing Nios® II Boot Time in Intel® MAX® 10 FPGA Design x
5.4.3.1. Boot Methods 5.4.3.2. Boot Device Performance 5.4.3.3. Peripheral Initialization 5.4.3.4. Nios® II Processor Caches 5.4.3.5. System Speed 5.4.3.6. Intel® MAX® 10 FPGA Nios® II Flash Accelerator
5.4.4. Boot Time Performance and Estimation x
5.4.4.1. Boot Time Performance for Design that Boots from On Chip Flash with Flash Accelerator 5.4.4.2. Boot Time Estimation
6. Nios® II Debug, Verification, and Simulation x
6.1. Software Debugging Options 6.2. Debugging Nios® II Designs 6.3. Verification and Board Bring-Up 6.4. Additional Embedded Design Considerations 6.5. Simulating Nios® II Embedded Processor Designs 6.6. Nios® II Debug, Verification, and Simulation Revision History
6.2. Debugging Nios® II Designs x
6.2.1. Debuggers 6.2.2. Run-Time Analysis Debug Techniques 6.2.3. Using the Debug Code from Intel® MAX® 10 On-Chip Flash - User Flash Memory (UFM)
6.2.1. Debuggers x
6.2.1.1. Nios® II Software Development Tools 6.2.1.2. Signal Tap II Embedded Logic Analyzer 6.2.1.3. Lauterbach Trace32 Debugger and PowerTrace Hardware 6.2.1.4. Data Display Debuggers
6.2.1.1. Nios® II Software Development Tools x
6.2.1.1.1. Nios® II System ID 6.2.1.1.2. Project Templates 6.2.1.1.3. Configuration Options 6.2.1.1.4. Nios® II GDB Console and GDB Commands 6.2.1.1.5. Nios® II Console View and stdio Library Functions 6.2.1.1.6. Importing Projects Created Using the Nios® II Software Build Tools 6.2.1.1.7. Selecting a Processor Instance in a Multiple Processor Design
6.2.2. Run-Time Analysis Debug Techniques x
6.2.2.1. Software Profiling 6.2.2.2. Watchpoints 6.2.2.3. Stack Overflow 6.2.2.4. Hardware Abstraction Layer (HAL) 6.2.2.5. Breakpoints 6.2.2.6. Debugger Stepping and Using No Optimizations
6.2.2.3. Stack Overflow x
6.2.2.3.1. Enabling Run Time Stack Checking
6.2.3. Using the Debug Code from Intel® MAX® 10 On-Chip Flash - User Flash Memory (UFM) x
6.2.3.1. Building Hardware Design in Platform Designer (Standard) 6.2.3.2. Compiling using the Intel® Quartus® Prime 6.2.3.3. Building Software Design 6.2.3.4. Creating the POF File 6.2.3.5. Programming the POF File 6.2.3.6. Debugging the Nios II Software Running in On-Chip Flash
6.3. Verification and Board Bring-Up x
6.3.1. Verification Methods 6.3.2. Board Bring-up 6.3.3. System Verification
6.3.1. Verification Methods x
6.3.1.1. Prerequisites 6.3.1.2. System Console 6.3.1.3. Signal Tap II Embedded Logic Analyzer 6.3.1.4. External Instrumentation 6.3.1.5. Stimuli Generation
6.3.1.4. External Instrumentation x
6.3.1.4.1. Signal Probe Signal Probe 6.3.1.4.2. Logic Analyzer Interface
6.3.2. Board Bring-up x
6.3.2.1. Peripheral Testing 6.3.2.2. Board Testing 6.3.2.3. Minimal Test System
6.3.2.1. Peripheral Testing x
6.3.2.1.1. Data Trace Failure 6.3.2.1.2. Address Trace Failure 6.3.2.1.3. Device Isolation 6.3.2.1.4. JTAG
6.3.3. System Verification x
6.3.3.1. Designing with Verification in Mind 6.3.3.2. Accelerating Verification 6.3.3.3. Using Software to Verify Hardware 6.3.3.4. Environmental Testing
6.4. Additional Embedded Design Considerations x
6.4.1. JTAG Signal Integrity 6.4.2. Memory Space For System Prototyping
6.5. Simulating Nios® II Embedded Processor Designs x
6.5.1. Before You Begin 6.5.2. Setting Up and Generating Your Simulation Environment in Platform Designer 6.5.3. Creating the Nios® II Software 6.5.4. Running Simulation in the ModelSim Simulator Using Nios II SBT for Eclipse 6.5.5. Running Simulation in the ModelSim Simulator Using Command Line
6.5.2. Setting Up and Generating Your Simulation Environment in Platform Designer x
6.5.2.1. Platform Designer-Generated System Simulation Files 6.5.2.2. Memory Simulation Models 6.5.2.3. Using IP and Platform Designer Simulation Setup Scripts
6.5.3. Creating the Nios® II Software x
6.5.3.1. Creating a Software Project Using Nios® II SBT for Eclipse 6.5.3.2. Creating a Software Project Using Command Line
7. Optimizing Nios® II Based Systems and Software x
7.1. Hardware Acceleration and Coprocessing 7.2. Software Application Optimization 7.3. Memory Optimization 7.4. Accelerating Nios® II Networking Applications 7.5. Using Tightly Coupled Memory with the Nios® II Processor Tutorial 7.6. Optimizing Nios® II Based Systems and Software Revision History
7.1. Hardware Acceleration and Coprocessing x
7.1.1. Hardware Acceleration 7.1.2. Coprocessing
7.1.1. Hardware Acceleration x
7.1.1.1. Customizing and Accelerating FPGA Designs 7.1.1.2. Accelerating Cyclic Redundancy Checking (CRC) 7.1.1.3. Creating Nios® II Custom Instructions
7.1.1.2. Accelerating Cyclic Redundancy Checking (CRC) x
7.1.1.2.1. Matching I/O Bandwidths 7.1.1.2.2. Pipelining Algorithms
7.1.2. Coprocessing x
7.1.2.1. Creating Multicore Designs 7.1.2.2. Pre- and Post-Processing 7.1.2.3. Replacing State Machines
7.1.2.1. Creating Multicore Designs x
7.1.2.1.1. Accessing Multiple Nios® II Designs
7.1.2.3. Replacing State Machines x
7.1.2.3.1. Low-Speed State Machines 7.1.2.3.2. High-Speed State Machines 7.1.2.3.3. Subdivided State Machines
7.2. Software Application Optimization x
7.2.1. Performance Tuning Background 7.2.2. Speeding Up System Processing Tasks 7.2.3. Accelerating Interrupt Service Routines 7.2.4. Reducing Code Size
7.2.2. Speeding Up System Processing Tasks x
7.2.2.1. Analyzing the Problem 7.2.2.2. Accelerating your Application
7.2.2.2. Accelerating your Application x
7.2.2.2.1. Increasing Processor Efficiency 7.2.2.2.2. Accelerating Hardware 7.2.2.2.3. Improving Data Movement
7.2.3. Accelerating Interrupt Service Routines x
7.2.3.1. Analyzing the Problem 7.2.3.2. Accelerating the Interrupt Service Routine
7.2.4. Reducing Code Size x
7.2.4.1. Analyzing the Problem 7.2.4.2. Reducing the Code Footprint
7.3. Memory Optimization x
7.3.1. Isolate Critical Memory Connections 7.3.2. Match Master and Slave Data Width 7.3.3. Use Separate Memories to Exploit Concurrency 7.3.4. Understand the Nios® II Instruction Master Address Space 7.3.5. Test Memory
7.4. Accelerating Nios® II Networking Applications x
7.4.1. Downloading the Ethernet Acceleration Design Example 7.4.2. The Structure of Networking Applications 7.4.3. The User Application 7.4.4. Structure of the NicheStack Networking Stack 7.4.5. Ethernet Device 7.4.6. Benchmarking Setup, Results, and Analysis 7.4.7. Nios® II Test Hardware and Test Results
7.4.2. The Structure of Networking Applications x
7.4.2.1. Ethernet System Hierarchy 7.4.2.2. Relationships Between Networking System Elements 7.4.2.3. Finding the Performance Bottlenecks
7.4.3. The User Application x
7.4.3.1. User Application Optimizations
7.4.3.1. User Application Optimizations x
7.4.3.1.1. Software Optimizations 7.4.3.1.2. Hardware Optimizations 7.4.3.1.3. The Sockets API
7.4.4. Structure of the NicheStack Networking Stack x
7.4.4.1. General Optimizations 7.4.4.2. NicheStack Specific Optimizations
7.4.4.2. NicheStack Specific Optimizations x
7.4.4.2.1. NicheStack Thread Priorities 7.4.4.2.2. Disabling Nonessential NicheStack Modules 7.4.4.2.3. Using Faster Packet Memory 7.4.4.2.4. Super Loop Mode 7.4.4.2.5. NicheStack Support and Licensing
7.4.5. Ethernet Device x
7.4.5.1. Link Speed 7.4.5.2. Network Interface (Intel FPGA Triple Speed Ethernet Intel® FPGA IP Function) 7.4.5.3. NicheStack Device Driver Model
7.4.6. Benchmarking Setup, Results, and Analysis x
7.4.6.1. Overview 7.4.6.2. Test Setup 7.4.6.3. Test Methodology 7.4.6.4. Nios® II System Software Configuration 7.4.6.5. Workstation System Software
7.4.6.2. Test Setup x
7.4.6.2.1. Test Systems
7.4.6.3. Test Methodology x
7.4.6.3.1. Ethernet Link Type 7.4.6.3.2. Protocols Tested 7.4.6.3.3. Data Transmission Sizes 7.4.6.3.4. Test Runs
7.4.6.4. Nios® II System Software Configuration x
7.4.6.4.1. NicheStack Networking Stack Configuration 7.4.6.4.2. MicroC/OS-II Configuration 7.4.6.4.3. Benchmark Application 7.4.6.4.4. General Application and System Library Settings
7.5. Using Tightly Coupled Memory with the Nios® II Processor Tutorial x
7.5.1. Reasons for Using Tightly Coupled Memory 7.5.2. Tradeoffs 7.5.3. Guidelines for Using Tightly Coupled Memory 7.5.4. Tightly Coupled Memory Interface 7.5.5. Building a Nios® II System with Tightly Coupled Memory 7.5.6. Generate the Platform Designer System 7.5.7. Run the Tightly Coupled Memories Examples from the Nios® II Command 7.5.8. Program and Run the Tightly Coupled Memory Project 7.5.9. Understanding the Tcl Scripts
7.5.3. Guidelines for Using Tightly Coupled Memory x
7.5.3.1. Hardware Guidelines 7.5.3.2. Software Guidelines
7.5.3.2. Software Guidelines x
7.5.3.2.1. Locating Functions in Tightly Coupled Memory
7.5.4. Tightly Coupled Memory Interface x
7.5.4.1. Restrictions 7.5.4.2. Dual Port Memories
7.5.5. Building a Nios® II System with Tightly Coupled Memory x
7.5.5.1. Hardware and Software Requirements 7.5.5.2. Modify the Example Design to Include Tightly Coupled Memories 7.5.5.3. Create the Tightly Coupled Memories 7.5.5.4. Connect and Position the Tightly Coupled Memories 7.5.5.5. Add a Performance Counter
7.5.9. Understanding the Tcl Scripts x
7.5.9.1. Timer Memory 7.5.9.2. Exception Stack 7.5.9.3. Timer Definitions
7.5.9.3. Timer Definitions x
7.5.9.3.1. peripheral_subsystem_sys_clk_timer 7.5.9.3.2. peripheral_subsystem_high_res_timer
Product Discontinuance Notification
1. Introduction
2. First Time Designer's Guide
3. Hardware System Design with Intel® Quartus® Prime and Platform Designer
4. Software System Design with a Nios® II Processor
5. Nios® II Configuration and Booting Solutions
6. Nios® II Debug, Verification, and Simulation
7. Optimizing Nios® II Based Systems and Software

7.4.3.1.3. The Sockets API

After tuning your application to become more computationally efficient (thereby freeing more of the processor’s time for operating the networking stack), you can optimize how the application uses the networking stack. This section describes how to select the best protocol for use by your application and the most efficient way to use the Sockets API.

Selecting the Right Networking Protocol

When using the Sockets API, you must also select which protocol to use for transporting data across the network. There are two main protocols used to transport data across networks: TCP and UDP. Both of these protocols perform the basic function of moving data across Ethernet networks, but they have very different implementations and performance implications. The table below compares the two protocols.

Table 55.  The UDP and TCP Protocols
Parameter Protocol
UDP TCP
Connection Mode Connectionless Connection-Oriented
In Order Data Guarantee No Yes
Data Integrity and Validation No Yes
Data Retransmission No Yes
Data Checksum Yes; Can be disabled Yes

In terms of just throughput performance, the UDP protocol is much faster than TCP because it has very little overhead. The UDP protocol makes no attempt to validate that the data being sent arrived at its destination (or even that the destination is capable of receiving packets), so the network stack needs to perform much less work in order to send or receive data using this protocol.

However, aside from very specialized cases where your embedded system can tolerate losing data (for example, streaming multimedia applications), use the TCP protocol.

Note: Use the UDP protocol to gain the fastest performance possible; however, use the TCP protocol when you must guarantee the transmission of the data.

Improving Send and Receive Performance

Proper use of the Sockets API in your application can also increase the overall networking throughput of your system. The following list describes several ways to optimally use the Sockets API:

  • Minimize send and receive function calls—The Sockets API provides two sets of functions for sending and receiving data through the networking stack. For the UDP protocol these functions are sendto() and recvfrom(). For the TCP protocol these functions are send() and recv().

    Depending on which transport protocol you use (TCP or UDP), your application uses one of these sets of functions. To increase overall performance, avoid calling these functions repetitively to handle small units of data. Every call to these functions incurs a fixed time penalty for execution, which can compound quickly when these functions are called multiple times in rapid succession. Combine data that you want to send (or receive) and call these functions with the largest possible amount of data at one time.

    Note: Call the Socket API’s send and receive functions with larger buffer sizes to minimize system call overhead.
  • Minimize latency when sending data—Although the TCP Sockets send() function can accept an arbitrary number of bytes, those bytes might not be immediately sent as a packet. This situation is especially likely when send() is called with a small number of bytes, because the networking stack attempts to coalesce these small data chunks into a larger packet. Small data chunks are coalesced to avoid congesting the network with many small packets (using the Nagle algorithm for congestion avoidance). There is a solution, however, through the use of the TCP_NODELAY flag.

    Setting a socket’s TCP_NODELAY flag, with the setsockopt() function call, disables the Nagle algorithm. The socket immediately sends whatever bytes are passed in as a TCP packet. Disabling the Nagle algorithm can be a useful way to increase network throughput in the case where your application must send many small chunks of data very quickly.

    Note: If you need to accelerate the transmission of small TCP packets, use the TCP_NODELAY flag on your socket. You can find an example of setting the TCP_NODELAY flag in the benchmarking application software in the Nios® II ethernet acceleration design example.

    While disabling the Nagle algorithm usually causes smaller packets to be immediately sent over the network, the networking stack might still coalesce some of the packets into larger packets. This situation is especially likely in the case of the Windows workstation platform. However, you can expect the networking stack to do so with much lower frequency than if the Nagle algorithm were enabled.

The Zero Copy API

The NicheStack networking stack provides a further optimization to accelerate the data transfers to and from the stack called the zero copy API. The zero copy API increases overall system performance by eliminating the buffer management scheme performed by the Socket API’s read and write function calls. The application manages the send and receive data buffers directly, eliminating an extra level of data copying performed by the Nios® II processor.

Using the NicheStack Zero Copy API can accelerate your network application’s throughput by eliminating an extra layer of copying.

Level Two Title