AN 802: Intel® Stratix® 10 SoC Device Design Guidelines

Download PDF
ID 683117
Date 12/14/2020
Public

A newer version of this document is available. Customers should click here to go to the newest version.

Document Table of Contents
Document Table of Contents x
1. Introduction to the Intel® Stratix® 10 SoC Device Design Guidelines 2. Board Design Guidelines for Stratix 10 SoC FPGAs 3. Interfacing to the FPGA for Stratix 10 SoC FPGAs 4. System Considerations for Stratix 10 SoC FPGAs 5. Embedded Software Design Guidelines for Intel® Stratix® 10 SoC FPGAs 6. Recommended Resources for Stratix 10 SoC FPGAs
1. Introduction to the Intel® Stratix® 10 SoC Device Design Guidelines x
1.1. Introduction to the Stratix 10 SoC Device Design Guidelines Revision History
2. Board Design Guidelines for Stratix 10 SoC FPGAs x
2.1. Pin Connection Considerations for Board Design 2.2. HPS Clocking and Reset Design Considerations 2.3. Design Considerations for Connecting Device I/O to HPS Peripherals and Memory 2.4. Design Guidelines for HPS Interfaces 2.5. HPS EMIF Design Considerations 2.6. HPS Memory Debug 2.7. Boundary Scan for HPS 2.8. Embedded Software Debugging and Trace 2.9. Board Design Guidelines for Intel® Stratix® 10 SoC FPGAs Revision History
2.1. Pin Connection Considerations for Board Design x
2.1.1. Board-Related Intel® Quartus® Prime Settings
2.1.1. Board-Related Intel® Quartus® Prime Settings x
2.1.1.1. Unused Pins
2.2. HPS Clocking and Reset Design Considerations x
2.2.1. HPS Clock Planning 2.2.2. Early Pin Planning and I/O Assignment Analysis 2.2.3. Pin Features and Connections for HPS Clocks, Reset and PoR 2.2.4. Direct to Factory Pin Support for Remote System Update (RSU) Feature 2.2.5. Internal Clocks 2.2.6. HPS Peripheral Reset Management
2.3. Design Considerations for Connecting Device I/O to HPS Peripherals and Memory x
2.3.1. HPS Pin Multiplexing Design Considerations 2.3.2. HPS I/O Settings: Constraints and Drive Strengths
2.4. Design Guidelines for HPS Interfaces x
2.4.1. HPS EMAC PHY Interfaces 2.4.2. USB Interface Design Guidelines 2.4.3. SD/MMC and eMMC Card Interface Design Guidelines 2.4.4. Design Guidelines for Flash Interfaces 2.4.5. UART Interface Design Guidelines 2.4.6. I2C Interface Design Guidelines
2.4.1. HPS EMAC PHY Interfaces x
2.4.1.1. PHY Interfaces 2.4.1.2. PHY Interfaces Connected Through FPGA I/O 2.4.1.3. MDIO 2.4.1.4. Common PHY Interface Design Considerations
2.4.1.1. PHY Interfaces x
2.4.1.1.1. RMII 2.4.1.1.2. RGMII
2.4.1.2. PHY Interfaces Connected Through FPGA I/O x
2.4.1.2.1. GMII/MII 2.4.1.2.2. Adapting to RGMII 2.4.1.2.3. Adapting to RMII 2.4.1.2.4. Adapting to SGMII
2.4.1.4. Common PHY Interface Design Considerations x
2.4.1.4.1. Signal Integrity
2.4.4. Design Guidelines for Flash Interfaces x
2.4.4.1. NAND Flash Interface Design Guidelines
2.5. HPS EMIF Design Considerations x
2.5.1. Considerations for Connecting HPS to SDRAM 2.5.2. HPS SDRAM I/O Locations
3. Interfacing to the FPGA for Stratix 10 SoC FPGAs x
3.1. Overview of HPS Memory-Mapped Interfaces 3.2. Recommended System Topologies 3.3. Recommended Starting Point for HPS-to-FPGA Interface Designs 3.4. Timing Closure for FPGA Accelerators 3.5. Information on How to Configure and Use the Bridges 3.6. Interfacing to the FPGA for Intel® Stratix® 10 SoC FPGAs Revision History
3.1. Overview of HPS Memory-Mapped Interfaces x
3.1.1. HPS-to-FPGA Bridge 3.1.2. Lightweight HPS-to-FPGA Bridge 3.1.3. FPGA-to-HPS Bridge 3.1.4. FPGA-to-SDRAM Ports 3.1.5. Interface Bandwidths
3.2. Recommended System Topologies x
3.2.1. HPS Accesses to FPGA Fabric 3.2.2. MPU Sharing Data with FPGA 3.2.3. Examples of Cacheable and Non-Cacheable Data Accesses From the FPGA
3.2.3. Examples of Cacheable and Non-Cacheable Data Accesses From the FPGA x
3.2.3.1. Example 1: FPGA Reading Data from HPS SDRAM Directly 3.2.3.2. Example 2: FPGA Writing Data into HPS SDRAM Directly 3.2.3.3. Example 3: FPGA Reading Cache Coherent Data from HPS 3.2.3.4. Example 4: FPGA Writing Cache Coherent Data to HPS
4. System Considerations for Stratix 10 SoC FPGAs x
4.1. Timing Considerations 4.2. Maximizing Performance 4.3. System Level Cache Coherency 4.4. System Considerations for Stratix 10 SoC FPGAs Revision History
4.1. Timing Considerations x
4.1.1. Timing Closure for FPGA Accelerators 4.1.2. USB Interface Design Guidelines
4.1.1. Timing Closure for FPGA Accelerators x
4.1.1.1. HPS First and FPGA First Boot Considerations
5. Embedded Software Design Guidelines for Intel® Stratix® 10 SoC FPGAs x
5.1. Overview 5.2. Assembling the Components of Your Software Development Platform 5.3. Golden Hardware Reference Design (GHRD) 5.4. Selecting an Operating System for Your Application 5.5. Assembling Your Software Development Platform for Linux* 5.6. Assembling your Software Development Platform for a Bare-Metal Application 5.7. Assembling your Software Development Platform for Partner OS or RTOS 5.8. Choosing the Bootloader Software 5.9. Selecting Software Tools for Development, Debug and Trace 5.10. Boot And Configuration Considerations 5.11. System Reset Considerations 5.12. Flash Considerations 5.13. Embedded Software Debugging and Trace 5.14. Embedded Software Design Guidelines for Intel® Stratix® 10 SoC FPGAs Revision History
5.4. Selecting an Operating System for Your Application x
5.4.1. Using Linux* or RTOS 5.4.2. Developing a Bare-Metal Application 5.4.3. Using the Bootloader as a Bare-Metal Framework 5.4.4. Using Symmetrical vs. Asymmetrical Multiprocessing (SMP vs. AMP) Modes
5.5. Assembling Your Software Development Platform for Linux* x
5.5.1. Golden System Reference Design (GSRD) for Linux* 5.5.2. Source Code Management Considerations
5.9. Selecting Software Tools for Development, Debug and Trace x
5.9.1. Selecting Software Build Tools 5.9.2. Selecting Software Debug Tools 5.9.3. Selecting Software Trace Tools
5.10. Boot And Configuration Considerations x
5.10.1. Configuration Sources 5.10.2. Configuration Flash 5.10.3. Configuration Clock 5.10.4. Selecting HPS Boot Options 5.10.5. HPS Boot Sources 5.10.6. Remote System Update (RSU)
5.12. Flash Considerations x
5.12.1. Flash Programming Method 5.12.2. Using a Single flash for Both FPGA Configuration and HPS Mass Storage
6. Recommended Resources for Stratix 10 SoC FPGAs x
6.1. Device Documentation 6.2. Tools and Software Web Pages 6.3. Recommended Resources for Intel® Stratix® 10 SoC FPGAs Revision History
1. Introduction to the Intel® Stratix® 10 SoC Device Design Guidelines
2. Board Design Guidelines for Stratix 10 SoC FPGAs
3. Interfacing to the FPGA for Stratix 10 SoC FPGAs
4. System Considerations for Stratix 10 SoC FPGAs
5. Embedded Software Design Guidelines for Intel® Stratix® 10 SoC FPGAs
6. Recommended Resources for Stratix 10 SoC FPGAs

4.3. System Level Cache Coherency

Cache coherency is a fundamental topic to understand any time data must be shared amongst multiple masters in a system. In the context of a SoC device these masters can be the MPU, DMA, peripherals with master interfaces, and masters in the FPGA connected to the HPS. Since the MPU contains level 1 and level 2 cache controllers, it can hold more up-to-date contents than main memory in the system. The HPS supports two mechanisms to make sure masters in the system observe a coherent view of memory: ensuring main memory contains the latest value, or have masters access a directory-based CCU fabric using the ACE-Lite interface.

The MPU can allocate buffers to be non-cacheable which ensures data is never cached by the L1 and L2 caches. The MPU can also access cacheable data and either flush it to main memory or copy it to a non-cacheable buffer before other masters attempt to access the data. Operating systems typically provide mechanisms for maintaining cache coherency both ways described above.

Masters in the system access coherent data by either relying on the MPU to place data into main memory instead of having it cached, or by having the master in the system perform a cacheable access through the CCU. The mechanism you use depends on the size of the buffer of memory the master is accessing.

For more information, refer to the "Interfacing to the FPGA" section.

GUIDELINE: Ensure that data accessed through the CCU fits in the 1 MB L2 cache to avoid thrashing overhead.

Since the L2 cache is 1 MB in size, if a master in the system frequently accesses buffers whose total size exceeds 1 MB, thrashing results.

Cache thrashing is a situation where the size of the data exceeds the size of the cache, causing the cache to perform frequent evictions and prefetches to main memory. Thrashing negates the performance benefits of caching the data.

In potential thrashing situation, it makes more sense to have the masters access non-cache coherent data and allow software executing on the MPU maintain the data coherency throughout the system.

GUIDELINE: For small buffers of data shared between the MPU and system masters, consider having the system master perform cacheable accesses to avoid overhead caused by cache flushing operations.

If a master in the system requires access to smaller coherent blocks of data then you should consider having the MPU access the buffer as cacheable memory and the master in the system perform cacheable accesses to the data. Cacheable accesses to the CCU through the ACE-Lite protocol supported by the FPGA-to-HPS bridge ensure that the master and MPU access the same copy of the data. By having the MPU use cacheable buffers and the system master performing cacheable accesses, software does not have to maintain system wide coherency ensuring both the MPU and system master observe the same copy of data.

Related Information
Level Two Title