Hard Processor System Technical Reference Manual: Agilex™ 5 SoCs

Download PDF
ID 814346
Date 7/19/2024
Public
Document Table of Contents
Document Table of Contents x
1. Agilex™ 5 Hard Processor System Technical Reference Manual Revision History 2. Introduction to the Hard Processor System 3. Micro Processor Unit (MPU) 4. Application Processor Subsystem 5. Peripheral Subsystem 6. System Manager 7. Clock Manager 8. Reset Manager 9. Power Management 10. Address Map 11. Bridges 12. Interfaces 13. System Interconnect and Firewalls 14. Error Checking and Correction Controller 15. CoreSight Debug and Trace 16. HPS Register Map A. Appendix
1. Agilex™ 5 Hard Processor System Technical Reference Manual Revision History x
1.1. Introduction to the HPS Revision History 1.2. MPU Revision History 1.3. CCU Revision History 1.4. GIC Revision History 1.5. SMMU Revision History 1.6. On-Chip RAM Revision History 1.7. EMAC Revision History 1.8. DMA Controller Revision History 1.9. NAND Flash Controller Revision History 1.10. SD/eMMC Revision History 1.11. Combo DLL PHY Revision History 1.12. USB 3.1 Gen1 Controller Revision History 1.13. USB 2.0 OTG Controller Revision History 1.14. I3C Controller Revision History 1.15. I2C Controller Revision History 1.16. SPI Controller Revision History 1.17. Timers Revision History 1.18. Watchdog Timers Revision History 1.19. UART Controller Revision History 1.20. GPIO Revision History 1.21. I/O Pin Multiplexing Revision History 1.22. System Manager Revision History 1.23. Clock Manager Revision History 1.24. Reset Manager Revision History 1.25. Power Management Revision History 1.26. Bridges Revision History 1.27. HPS Mailbox Revision History 1.28. MPFE and MPFE-lite Revision History 1.29. EMAC GMII through FPGA Fabric Revision History 1.30. System Interconnect and Firewalls Revision History 1.31. ECC Controller Revision History 1.32. CoreSight* Debug and Trace Revision History 1.33. HPS Register Map Revision History 1.34. Booting and Configuration Revision History 1.35. HPS Use of SDM QSPI Controller Revision History 1.36. Security Revision History 1.37. Operational Status of the HPS to the FPGA Logic Revision History
2. Introduction to the Hard Processor System x
2.1. HPS Differences Among Intel SoC Device Families 2.2. HPS Features 2.3. HPS System Integration 2.4. HPS IP Revisions 2.5. HPS Address Map and Register Definitions
2.3. HPS System Integration x
2.3.1. HPS Block Diagram 2.3.2. MPU Features 2.3.3. Application Processor Subsystem 2.3.4. Peripheral Subsystem 2.3.5. System Manager Features 2.3.6. Clock Manager Features 2.3.7. Reset Manager Features 2.3.8. Bridges Features 2.3.9. System Interconnect and Firewalls Features 2.3.10. ECC Controller Features 2.3.11. CoreSight Debug and Trace Features
2.3.2. MPU Features x
2.3.2.1. Arm Cortex-A76 Core Features 2.3.2.2. Arm* Cortex* -A55 Core Features 2.3.2.3. Arm* DynamIQ Shared Unit Features
2.3.3. Application Processor Subsystem x
2.3.3.1. CCU Features 2.3.3.2. GIC Features 2.3.3.3. SMMU Features 2.3.3.4. On-Chip RAM Features
2.3.4. Peripheral Subsystem x
2.3.4.1. EMAC Features 2.3.4.2. DMA Controller Features 2.3.4.3. NAND Flash Controller Features 2.3.4.4. Combo DLL PHY Features 2.3.4.5. USB 3.1 Gen1 Controller Features 2.3.4.6. USB 2.0 OTG Controller Features 2.3.4.7. I3C Controller Features 2.3.4.8. I2C Controller Features 2.3.4.9. SPI Controller Features 2.3.4.10. Timers Features 2.3.4.11. Watchdog Timers Features 2.3.4.12. UART Controller Features 2.3.4.13. GPIO Features 2.3.4.14. I/O Pin Multiplexing Features
2.3.4.1. EMAC Features x
2.3.4.1.1. XGMAC Core 2.3.4.1.2. Time Sensitive Networking 2.3.4.1.3. MAC Transaction Layer 2.3.4.1.4. DMA 2.3.4.1.5. Management Interface 2.3.4.1.6. Synchronized Multidrop Timestamp Gathering Hub 2.3.4.1.7. External Memory 2.3.4.1.8. PHY Interface
3. Micro Processor Unit (MPU) x
3.1. MPU Differences Among Intel SoC Device Families 3.2. MPU Use Cases 3.3. MPU Features 3.4. MPU System Integration 3.5. MPU Arm* Cortex* -A76 Core 3.6. MPU Arm* Cortex* -A55 Core 3.7. MPU Arm* DynamIQ Shared Unit 3.8. MPU Clock Domains 3.9. MPU Reset Domains 3.10. MPU Power Domains 3.11. MPU Address Map and Register Definitions
3.5. MPU Arm* Cortex* -A76 Core x
3.5.1. Arm Cortex-A76 Core Features 3.5.2. System Integration of the Arm* Cortex* -A76 Core 3.5.3. Functional Description of the Arm Cortex-A76 Core
3.5.3. Functional Description of the Arm Cortex-A76 Core x
3.5.3.1. Cortex* -A76 Core Configuration 3.5.3.2. Exception Levels 3.5.3.3. Virtualization 3.5.3.4. Memory Management Unit 3.5.3.5. Level 1 Memory System 3.5.3.6. Level 2 Memory System 3.5.3.7. Generic Interrupt Controller CPU Interface 3.5.3.8. Advanced Single Instruction Multiple Data and Floating Point Support 3.5.3.9. Cryptographic Extensions 3.5.3.10. Generic Timer 3.5.3.11. Cache Protection 3.5.3.12. Debug
3.5.3.2. Exception Levels x
3.5.3.2.1. Security State 3.5.3.2.2. Security Model
3.5.3.4. Memory Management Unit x
3.5.3.4.1. Translation Lookaside Buffer 3.5.3.4.2. Translation Match Process 3.5.3.4.3. Translation Table Walks
3.5.3.5. Level 1 Memory System x
3.5.3.5.1. Instruction Cache 3.5.3.5.2. Data Cache 3.5.3.5.3. Data Prefetching
3.5.3.11. Cache Protection x
3.5.3.11.1. Uncorrected Errors and Data Poisoning 3.5.3.11.2. Reliability, Availability, and Serviceability Error Types 3.5.3.11.3. Error Recording 3.5.3.11.4. Error Injection
3.5.3.12. Debug x
3.5.3.12.1. Breakpoints and Watchpoints 3.5.3.12.2. Performance Monitoring Unit 3.5.3.12.3. Activity Monitor Unit 3.5.3.12.4. Embedded Trace Macrocell
3.6. MPU Arm* Cortex* -A55 Core x
3.6.1. Arm* Cortex* -A55 Core Features 3.6.2. System Integration of the Arm* Cortex* -A55 Core 3.6.3. Functional Description of the Arm* Cortex* -A55 Core
3.6.3. Functional Description of the Arm* Cortex* -A55 Core x
3.6.3.1. Cortex* -A55 Core Configuration 3.6.3.2. Exception Levels 3.6.3.3. Virtualization 3.6.3.4. Memory Management Unit 3.6.3.5. Level 1 Memory System 3.6.3.6. Level 2 Memory System 3.6.3.7. Generic Interrupt Controller CPU Interface 3.6.3.8. Data Processing Unit 3.6.3.9. Generic Timer 3.6.3.10. Cache Protection 3.6.3.11. Debug
3.6.3.2. Exception Levels x
3.6.3.2.1. Security State 3.6.3.2.2. Security Model
3.6.3.4. Memory Management Unit x
3.6.3.4.1. Translation Lookaside Buffer 3.6.3.4.2. Translation Match Process 3.6.3.4.3. Translation Table Walks
3.6.3.10. Cache Protection x
3.6.3.10.1. Uncorrected Errors and Data Poisoning 3.6.3.10.2. Reliability, Availability and Serviceability Error Types 3.6.3.10.3. Error Reporting 3.6.3.10.4. Error Injection
3.6.3.11. Debug x
3.6.3.11.1. Breakpoints and Watchpoints 3.6.3.11.2. Performance Monitoring Unit 3.6.3.11.3. Embedded Trace Macrocell
3.7. MPU Arm* DynamIQ Shared Unit x
3.7.1. Arm* DynamIQ Shared Unit Features 3.7.2. System Integration of the Arm* DynamIQ Shared Unit 3.7.3. Functional Description of the Arm* DynamIQ Shared Unit
3.7.3. Functional Description of the Arm* DynamIQ Shared Unit x
3.7.3.1. DSU Configuration 3.7.3.2. Level 3 Cache 3.7.3.3. Interfaces 3.7.3.4. Debug
3.7.3.2. Level 3 Cache x
3.7.3.2.1. Cache Allocation Policy 3.7.3.2.2. Cache Partitioning 3.7.3.2.3. Cache Data RAM Latency 3.7.3.2.4. Cache Slices and Portions
3.7.3.4. Debug x
3.7.3.4.1. DebugBlock 3.7.3.4.2. Embedded Cross Trigger 3.7.3.4.3. Performance Monitoring Unit
4. Application Processor Subsystem x
4.1. Cache Coherency Unit (CCU) 4.2. Generic Interrupt Controller (GIC) 4.3. System Memory Management Unit (SMMU) 4.4. On-Chip RAM
4.1. Cache Coherency Unit (CCU) x
4.1.1. CCU Differences Among Intel SoC Device Families 4.1.2. CCU Use Cases 4.1.3. CCU Features 4.1.4. CCU System Integration 4.1.5. CCU Functional Description 4.1.6. CCU Programming Model 4.1.7. CCU Address Map and Register Definitions
4.1.5. CCU Functional Description x
4.1.5.1. Block Diagram 4.1.5.2. Ports 4.1.5.3. Cache Coherency Protocol 4.1.5.4. Addressing and Memory Regions 4.1.5.5. Connectivity 4.1.5.6. Snoop Filters 4.1.5.7. System Memory Cache 4.1.5.8. Credits and Resources 4.1.5.9. Quality of Service 4.1.5.10. Storage Protection 4.1.5.11. Exclusive Monitors 4.1.5.12. Firewall and Security 4.1.5.13. Interrupts 4.1.5.14. Clocks 4.1.5.15. Resets 4.1.5.16. Power Management 4.1.5.17. Shutdown of Interfaces 4.1.5.18. Error Handling 4.1.5.19. CCU Restrictions
4.1.5.1. Block Diagram x
4.1.5.1.1. Coherent Agent Interface Unit (CAIU) 4.1.5.1.2. Non-coherent Agent Interface Unit (NCAIU) 4.1.5.1.3. Distributed Coherence Engine (DCE) 4.1.5.1.4. Distributed Memory Interface (DMI) 4.1.5.1.5. Distributed Virtual Memory Engine (DVE) 4.1.5.1.6. Distributed I/O Interface (DII)
4.1.5.2. Ports x
4.1.5.2.1. DSU CHI-B Initiator Port 4.1.5.2.2. F2H ACE-Lite Initiator Port 4.1.5.2.3. GIC_M ACE-lite Initiator Port 4.1.5.2.4. TCU Ace-lite+DVM Initiator Port 4.1.5.2.5. CCU_IOM ACE-Lite Initiator Port 4.1.5.2.6. CCU_DMI0, CCU_DMI1 1AXI4 Target Ports 4.1.5.2.7. CCU_IOS AXI Target Port 4.1.5.2.8. MPFE CSR AXI Target Port 4.1.5.2.9. GIC AXI Target Port 4.1.5.2.10. OCRAM AXI Target Port
4.1.5.3. Cache Coherency Protocol x
4.1.5.3.1. Coherent Read Request 4.1.5.3.2. Coherent Clean Request 4.1.5.3.3. Coherent Write Request
4.1.5.4. Addressing and Memory Regions x
4.1.5.4.1. Ncore Register Space 4.1.5.4.2. General Purpose Address Space 4.1.5.4.3. Boot Address Space 4.1.5.4.4. Interleaving
4.1.5.4.2. General Purpose Address Space x
4.1.5.4.2.1. Address Map When Using a Single SDRAM Channel 4.1.5.4.2.2. Address Map When Using Multiple SDRAM Channels
4.1.5.12. Firewall and Security x
4.1.5.12.1. OCRAM Firewall 4.1.5.12.2. GIC Firewall 4.1.5.12.3. Service Network Access 4.1.5.12.4. AXI APROT Tunneling
4.1.5.15. Resets x
4.1.5.15.1. Q-Channel
4.1.5.18. Error Handling x
4.1.5.18.1. Correctable Errors 4.1.5.18.2. Uncorrectable Errors
4.1.5.19. CCU Restrictions x
4.1.5.19.1. CHI Support 4.1.5.19.2. AXI and ACE Support 4.1.5.19.3. Ordering Support
4.1.6. CCU Programming Model x
4.1.6.1. CCU Initialization 4.1.6.2. Boot Region CSR Default Settings 4.1.6.3. Error Registers
4.1.6.1. CCU Initialization x
4.1.6.1.1. Address Space Configuration 4.1.6.1.2. SMC Configuration
4.2. Generic Interrupt Controller (GIC) x
4.2.1. GIC Differences Among Intel® SoC Device Families 4.2.2. GIC Features 4.2.3. GIC System Integration 4.2.4. GIC Functional Description 4.2.5. GIC Programming Model 4.2.6. GIC Address Map and Register Definitions 4.2.7. GIC Design Guidelines and Examples
4.2.4. GIC Functional Description x
4.2.4.1. Types of Interrupts 4.2.4.2. Clock Domains 4.2.4.3. Reset Domains
4.2.5. GIC Programming Model x
4.2.5.1. GIC Shared Peripheral Interrupts Map for the SoC HPS
4.2.7. GIC Design Guidelines and Examples x
4.2.7.1. Guidelines for Interrupt request from FPGA2HPS and GPIOs
4.3. System Memory Management Unit (SMMU) x
4.3.1. SMMU Differences Among Intel SoC Device Families 4.3.2. SMMU Use Cases 4.3.3. SMMU Features 4.3.4. SMMU System Integration 4.3.5. SMMU Signal Description 4.3.6. SMMU Functional Description 4.3.7. SMMU Programming Model 4.3.8. SMMU Address Map and Register Definitions 4.3.9. SMMU Design Guidelines and Examples
4.3.3. SMMU Features x
4.3.3.1. DTI Interface for TCU 4.3.3.2. Stage2 Translation 4.3.3.3. TLB 4.3.3.4. Bypass Mode 4.3.3.5. APB4 Programming Interface
4.3.4. SMMU System Integration x
4.3.4.1. TBU Instances 4.3.4.2. DTI Interconnect Topology 4.3.4.3. MMU-600 interfaces 4.3.4.4. IP Configuration
4.3.4.3. MMU-600 interfaces x
4.3.4.3.1. TCU Interfaces 4.3.4.3.2. TBU Interfaces 4.3.4.3.3. DTI Interconnect Interfaces
4.3.4.3.1. TCU Interfaces x
4.3.4.3.1.1. APB4 Programming Interface 4.3.4.3.1.2. ACE-Lite Manager for Table walk and DVM for TCU 4.3.4.3.1.3. LPI Interfaces for TCU 4.3.4.3.1.4. DTI Interface for TCU 4.3.4.3.1.5. SYSCO Interface for TCU 4.3.4.3.1.6. PMU snapshot Interface for TCU
4.3.4.3.2. TBU Interfaces x
4.3.4.3.2.1. ACE-Lite Subordinate Interface 4.3.4.3.2.2. ACE-Lite Manager Interface 4.3.4.3.2.3. DTI interface for TBU 4.3.4.3.2.4. LPI Interfaces for TBU 4.3.4.3.2.5. Interrupt Interfaces for TBU
4.3.4.3.3. DTI Interconnect Interfaces x
4.3.4.3.3.1. DTI Interconnect Switch Interfaces 4.3.4.3.3.2. DTI Interconnect Sizer Interfaces 4.3.4.3.3.3. DTI Interconnect Register Slice Interfaces
4.3.4.4. IP Configuration x
4.3.4.4.1. MMU-600 TCU Hardware Configuration 4.3.4.4.2. MMU-600 TBU Hardware Configurations 4.3.4.4.3. MMU-600 DTI Switch Hardware Configuration 4.3.4.4.4. MMU-600 DTI Async Bridge configuration
4.3.5. SMMU Signal Description x
4.3.5.1. MMU-600 TCU Configuration Signals 4.3.5.2. MMU-600 TBU Configuration Signals 4.3.5.3. TBU with Untranslated Interface
4.3.6. SMMU Functional Description x
4.3.6.1. Block Diagram 4.3.6.2. Functional Blocks 4.3.6.3. Functional Modes 4.3.6.4. Interrupts
4.3.6.3. Functional Modes x
4.3.6.3.1. Address Translation 4.3.6.3.2. Stage1 Translation 4.3.6.3.3. Stage2 Translation 4.3.6.3.4. TLB 4.3.6.3.5. Bypass Mode
4.3.6.4. Interrupts x
4.3.6.4.1. Interrupt Interfaces for TBU
4.3.7. SMMU Programming Model x
4.3.7.1. Initializing the SMMU 4.3.7.2. Assigning Stream IDs 4.3.7.3. Allocating the Command Queue 4.3.7.4. Allocating the Event Queue 4.3.7.5. Configuring the Stream Table 4.3.7.6. Initializing the Command Queue 4.3.7.7. Initializing the Event Queue 4.3.7.8. Invalidate TLBs and Configuration caches 4.3.7.9. Create Context Descriptor 4.3.7.10. Creating Stream Table Entry 4.3.7.11. Enabling the SMMU
4.3.8. SMMU Address Map and Register Definitions x
4.3.8.1. TBU Private Registers
4.3.8.1. TBU Private Registers x
4.3.8.1.1. TBU Stream Control Register[m][n] 4.3.8.1.2. TBU Stream ID Ax Register[m][n]
4.3.9. SMMU Design Guidelines and Examples x
4.3.9.1. F2SDRAM_TBU Special Handling 4.3.9.2. Translation Table Walk Sequence
4.3.9.2. Translation Table Walk Sequence x
4.3.9.2.1. Calculating Page Table Walks
4.4. On-Chip RAM x
4.4.1. On-Chip RAM Differences Among Intel SoC Device Families 4.4.2. On-Chip RAM Use Cases 4.4.3. On-Chip RAM Features 4.4.4. On-Chip RAM Functional Description 4.4.5. On-Chip RAM Address Map and Register Definitions
4.4.4. On-Chip RAM Functional Description x
4.4.4.1. Read and Write Double-Bit Bus Errors 4.4.4.2. On-Chip RAM Controller 4.4.4.3. On-Chip RAM Burst Support 4.4.4.4. Exclusive Access Support 4.4.4.5. Sub-word Accesses 4.4.4.6. On-Chip RAM Clocks 4.4.4.7. On-Chip RAM Resets 4.4.4.8. On-Chip RAM Initialization 4.4.4.9. ECC Protection
4.4.4.5. Sub-word Accesses x
4.4.4.5.1. Pipeline and Timing
5. Peripheral Subsystem x
5.1. Ethernet Media Access Controller 5.2. DMA Controller 5.3. NAND Flash Controller 5.4. SD/eMMC Host Controller 5.5. Combo DLL PHY 5.6. USB 3.1 Gen1 Controller 5.7. USB 2.0 OTG Controller 5.8. I3C Controller 5.9. I2C Controller 5.10. SPI Controller 5.11. Timers 5.12. Watchdog Timers 5.13. UART Controller 5.14. General-Purpose I/O Interface (GPIO) 5.15. Hard Processor System I/O Pin Multiplexing
5.1. Ethernet Media Access Controller x
5.1.1. EMAC Differences Among Intel SoC Device Families 5.1.2. EMAC Use Cases 5.1.3. EMAC Features 5.1.4. EMAC System Integration 5.1.5. EMAC Signal Description and Interfaces 5.1.6. EMAC Functional Description 5.1.7. EMAC Programming Model 5.1.8. EMAC Address Map and Register Definitions 5.1.9. EMAC Design Guidelines and Examples
5.1.3. EMAC Features x
5.1.3.1. XGMAC Core 5.1.3.2. Time Sensitive Networking 5.1.3.3. MAC Transaction Layer 5.1.3.4. DMA 5.1.3.5. Management Interface 5.1.3.6. Synchronized Multidrop Timestamp Gathering Hub 5.1.3.7. External Memory 5.1.3.8. PHY Interface
5.1.4. EMAC System Integration x
5.1.4.1. EMAC Block Diagram and Overview 5.1.4.2. EMAC System Integration
5.1.5. EMAC Signal Description and Interfaces x
5.1.5.1. HPS EMAC I/O Signals 5.1.5.2. FPGA EMAC I/O Signals 5.1.5.3. PHY Management Interface 5.1.5.4. SMTG Hub Signals and Interface 5.1.5.5. Timestamp Interface Signals 5.1.5.6. DMA Host Interface 5.1.5.7. System Manager Configuration Interface
5.1.5.3. PHY Management Interface x
5.1.5.3.1. MDIO Interface 5.1.5.3.2. I2C External PHY Management Interface
5.1.6. EMAC Functional Description x
5.1.6.1. External Memory 5.1.6.2. DMA Controller 5.1.6.3. Descriptor 5.1.6.4. Checksum Offload Engine (COE) 5.1.6.5. TCP Segmentation Offload 5.1.6.6. Packet Filtering 5.1.6.7. Management Counter 5.1.6.8. Flow Control 5.1.6.9. IEEE 1588-2008 Advanced Timestamp 5.1.6.10. TSN Features 5.1.6.11. SMTG Hub Time of Day Synchronization 5.1.6.12. Clocks 5.1.6.13. Resets 5.1.6.14. Interrupts
5.1.6.1. External Memory x
5.1.6.1.1. Transmit and Receive Data FIFO Buffers 5.1.6.1.2. TCP/IP Segmentation Offload Memory 5.1.6.1.3. Descriptor Cache Memory 5.1.6.1.4. Gate Control List Memory
5.1.6.2. DMA Controller x
5.1.6.2.1. Application Bus Burst Access 5.1.6.2.2. Application Data Buffer Alignment 5.1.6.2.3. Buffer Size Calculations 5.1.6.2.4. DMA Descriptor Fetch Operation 5.1.6.2.5. DMA TX Data Transfer Operation 5.1.6.2.6. DMA RX Data Transfer Operation 5.1.6.2.7. DMA Descriptor Write-Back Operation 5.1.6.2.8. DMA Start/Stop Operation 5.1.6.2.9. Memory Cache Size Requirements 5.1.6.2.10. Memory Cache Access Arbitration 5.1.6.2.11. DMA Error Handling
5.1.6.2.11. DMA Error Handling x
5.1.6.2.11.1. TX DMA Bus Error Handling Flow 5.1.6.2.11.2. RX DMA Bus Error Handling Flow
5.1.6.3. Descriptor x
5.1.6.3.1. Descriptor Structure 5.1.6.3.2. Descriptor Endianness 5.1.6.3.3. Transmit Descriptor 5.1.6.3.4. Receive Descriptor 5.1.6.3.5. Enhanced Descriptor for Time-Based Scheduling 5.1.6.3.6. Header Payload Split Support
5.1.6.5. TCP Segmentation Offload x
5.1.6.5.1. Description of the TSO Feature 5.1.6.5.2. DMA Operation with TSO Feature 5.1.6.5.3. TCP/IP Header Fields 5.1.6.5.4. Header and Payload Fields of Segmented Packets 5.1.6.5.5. Context Descriptor Sequence
5.1.6.6. Packet Filtering x
5.1.6.6.1. Source Address or Destination Address Filtering 5.1.6.6.2. VLAN Filtering 5.1.6.6.3. Extended VLAN Filtering 5.1.6.6.4. Layer 3 and Layer 4 Match Filtering
5.1.6.8. Flow Control x
5.1.6.8.1. Transmit Flow Control 5.1.6.8.2. Receive Flow Control
5.1.6.9. IEEE 1588-2008 Advanced Timestamp x
5.1.6.9.1. Delay Request-Response Mechanism 5.1.6.9.2. Peer-to-Peer PTP Transparent Clock Message Support 5.1.6.9.3. Clock Types 5.1.6.9.4. Reference Timing Source 5.1.6.9.5. System Time Register Module 5.1.6.9.6. Transmit Path Functions 5.1.6.9.7. Receive Path Functions
5.1.6.10. TSN Features x
5.1.6.10.1. Enhancements to Scheduled Traffic Implementation of Gate Control List Gate Control List Registers Transmission Gating Implementation Idle Slope Computation Updates Installing a New GCL Impact of Transmit Scheduling Algorithms on EST EST Errors 5.1.6.10.2. Frame Preemption 5.1.6.10.3. Time-Based Scheduling 5.1.6.10.4. Credit Based Shaper
5.1.6.11. SMTG Hub Time of Day Synchronization x
5.1.6.11.1. SMTG Hub Overview 5.1.6.11.2. SMTG Hub Logic 5.1.6.11.3. SMTG Hub Integration with the HPS and FPGA
5.1.6.12. Clocks x
5.1.6.12.1. Clock Domain
5.1.6.13. Resets x
5.1.6.13.1. EMAC ECC RAM Reset 5.1.6.13.2. EMAC Software Reset
5.1.7. EMAC Programming Model x
5.1.7.1. System Level EMAC configurable Registers 5.1.7.2. EMAC HPS Interface Initialization 5.1.7.3. EMAC FPGA Interface Initialization 5.1.7.4. DMA Initialization 5.1.7.5. EMAC Initialization and Configuration 5.1.7.6. Performing Normal Receive and Transmit Operation 5.1.7.7. Stopping and Starting Transmission 5.1.7.8. Reconfiguring the DMA Registers 5.1.7.9. Switching to a New Descriptor List in the Receive DMA 5.1.7.10. Handling Bus Errors and Recovery 5.1.7.11. Setting up TCP Segmentation Offload 5.1.7.12. Setting up VLAN Filtering on Receive 5.1.7.13. Setting up Extended VLAN Filtering 5.1.7.14. Setting up the L3-L4 Filtering 5.1.7.15. Programming the SMTG Hub 5.1.7.16. Setting up the IEEE 1588 PTP Timestamping 5.1.7.17. Programming Guidelines for Flexible Pulse-Per-Second (PPS) Output 5.1.7.18. Programming the GCL and GCL Linked Registers 5.1.7.19. Programming Guidelines for EST 5.1.7.20. Enabling the Frame Preemption Function 5.1.7.21. Setting up the Time-Based Scheduling Function
5.1.7.1. System Level EMAC configurable Registers x
5.1.7.1.1. System Manager Configurable Registers 5.1.7.1.2. Clock Manager Configurable Registers 5.1.7.1.3. Reset Manager Configurable Registers
5.1.7.10. Handling Bus Errors and Recovery x
5.1.7.10.1. Transmit DMA Channel 5.1.7.10.2. Receive DMA Channel
5.1.7.14. Setting up the L3-L4 Filtering x
5.1.7.14.1. Writing to the Indirect Addressed Registers 5.1.7.14.2. Reading the Indirect Addressed Registers
5.1.7.16. Setting up the IEEE 1588 PTP Timestamping x
5.1.7.16.1. Initialization Guidelines for System Time Generation 5.1.7.16.2. Coarse Correction Method 5.1.7.16.3. Fine-Correction Method
5.1.7.17. Programming Guidelines for Flexible Pulse-Per-Second (PPS) Output x
5.1.7.17.1. Generating a Single Pulse on PPS 5.1.7.17.2. Generating a Pulse Train on PPS 5.1.7.17.3. Generating an Interrupt without Affecting the PPS
5.1.7.20. Enabling the Frame Preemption Function x
5.1.7.20.1. Receive Side Programming 5.1.7.20.2. Transmit Side Programming
5.1.9. EMAC Design Guidelines and Examples x
5.1.9.1. HPS RGMII System Integration 5.1.9.2. Interface to HPS IO Pins 5.1.9.3. RGMII Clock Skew Guidelines
5.1.9.3. RGMII Clock Skew Guidelines x
5.1.9.3.1. Use Case 1 – Utilize Programmable HPS I/O Delay Feature 5.1.9.3.2. Use Case 2 – Configure External PHY RGMII-Internal Delay (RGMII-ID) 5.1.9.3.3. Use Case 3 – Introduce Board Trace Delay
5.2. DMA Controller x
5.2.1. DMA Controller Differences Among Intel SoC Device Families 5.2.2. DMA Controller Use Cases 5.2.3. DMA Controller Features 5.2.4. DMA Controller System Integration 5.2.5. DMA Controller Functional Description 5.2.6. DMA Controller Address Map and Register Definitions
5.2.5. DMA Controller Functional Description x
5.2.5.1. Flow Control 5.2.5.2. Handshaking Interface 5.2.5.3. Interrupt Outputs 5.2.5.4. DMA Controller Peripheral Request Interface
5.2.5.4. DMA Controller Peripheral Request Interface x
5.2.5.4.1. Peripheral Request Interface Mapping
5.3. NAND Flash Controller x
5.3.1. NAND Flash Controller Differences Among Intel SoC Device Families 5.3.2. NAND Flash Controller Use Cases 5.3.3. NAND Flash Controller Features 5.3.4. NAND Flash Controller System Integration 5.3.5. NAND Flash Controller Signal Description 5.3.6. NAND Flash Controller Functional Description 5.3.7. NAND Flash Controller Programming Model 5.3.8. NAND Flash Controller Address Map and Register Definitions
5.3.5. NAND Flash Controller Signal Description x
5.3.5.1. Device Discovery Interface 5.3.5.2. Write Protection Interface 5.3.5.3. Interrupt Interface 5.3.5.4. DFI Interface
5.3.6. NAND Flash Controller Functional Description x
5.3.6.1. Block Diagram 5.3.6.2. Initialization Protocol (Device Discovery) 5.3.6.3. NAND Flash Addressing and Data Layout 5.3.6.4. Command Engine Functionality 5.3.6.5. ECC Engine Functionality 5.3.6.6. Control Data Mechanism 5.3.6.7. Remapping Mechanism 5.3.6.8. Write Protection Mechanism 5.3.6.9. Error and Special Scenarios Handling 5.3.6.10. Controller Fixed Parameters and Clock Frequencies Supported
5.3.6.3. NAND Flash Addressing and Data Layout x
5.3.6.3.1. NAND Flash Addressing 5.3.6.3.2. Data Layout 5.3.6.3.3. Layout Between Pages
5.3.6.4. Command Engine Functionality x
5.3.6.4.1. Command Engine Architecture 5.3.6.4.2. CDMA Work Mode 5.3.6.4.3. PIO Work Mode 5.3.6.4.4. Generic Work Mode 5.3.6.4.5. Extended Functionality in Command Engine
5.3.6.4.2. CDMA Work Mode x
5.3.6.4.2.1. Operation in CDMA Mode 5.3.6.4.2.2. Thread Synchronization Mechanism 5.3.6.4.2.3. NOP Descriptor
5.3.6.4.3. PIO Work Mode x
5.3.6.4.3.1. Operation in PIO Work Mode 5.3.6.4.3.2. Last Operation Status
5.3.6.4.4. Generic Work Mode x
5.3.6.4.4.1. Operation in Generic Work Mode 5.3.6.4.4.2. Last Operation Status 5.3.6.4.4.3. Switching To/From Generic Work Mode
5.3.6.4.5. Extended Functionality in Command Engine x
5.3.6.4.5.1. LUN Interleaving 5.3.6.4.5.2. Multipage Commands 5.3.6.4.5.3. Cache Commands 5.3.6.4.5.4. Multi-plane Commands 5.3.6.4.5.5. Thread Reset Commands 5.3.6.4.5.6. Timeout Functionality 5.3.6.4.5.7. Support of TLC (pSLC Mode) Devices
5.3.6.5. ECC Engine Functionality x
5.3.6.5.1. Error Correction 5.3.6.5.2. Sector Sizes 5.3.6.5.3. Bad Block Marker 5.3.6.5.4. Scrambling Support 5.3.6.5.5. Erased Page Detection Support
5.3.6.9. Error and Special Scenarios Handling x
5.3.6.9.1. Access to Protected Areas 5.3.6.9.2. Bus Error 5.3.6.9.3. NAND Device Error 5.3.6.9.4. Data Integrity Errors 5.3.6.9.5. ECC Errors 5.3.6.9.6. Descriptor/Command Error 5.3.6.9.7. DQS Error 5.3.6.9.8. Error Interrupt Generation 5.3.6.9.9. Recommended Error Handling Checking Software Flow
5.3.7. NAND Flash Controller Programming Model x
5.3.7.1. NAND Controller Registers Programming Model 5.3.7.2. Status Polling Configuration 5.3.7.3. Device Layout Configuration 5.3.7.4. Configure Multiplane and Cache Operations 5.3.7.5. ECC Enabling 5.3.7.6. Interrupts Configuration 5.3.7.7. Configuring Timing Registers 5.3.7.8. Switch from SDR to DDR Operation Mode 5.3.7.9. Switch from DDR to SDR Operation Mode 5.3.7.10. Slave DMA Programming 5.3.7.11. Data Pre-Fetching Mechanism 5.3.7.12. Data Integrity Mechanism 5.3.7.13. Enabling pSLC Mode for the TLC Devices
5.3.7.1. NAND Controller Registers Programming Model x
5.3.7.1.1. Programming SFR Registers 5.3.7.1.2. Programming Command Registers
5.4. SD/eMMC Host Controller x
5.4.1. SD/eMMC Differences Among Intel SoC Device Families 5.4.2. SD/eMMC Use Cases 5.4.3. SD/eMMC Features 5.4.4. SD/eMMC System Integration 5.4.5. SD/eMMC Signal Description 5.4.6. SD/eMMC Functional Description 5.4.7. SD/eMMC Programming Model 5.4.8. SD/eMMC Address Map and Register Definitions 5.4.9. SD/eMMC Design Guidelines and Examples
5.4.3. SD/eMMC Features x
5.4.3.1. Supported Standards 5.4.3.2. SD Support 5.4.3.3. eMMC Support 5.4.3.4. DMA Support
5.4.6. SD/eMMC Functional Description x
5.4.6.1. Block Diagram 5.4.6.2. Configuration 5.4.6.3. DMA Operation 5.4.6.4. Clocks 5.4.6.5. Resets 5.4.6.6. SD Voltage Switching 5.4.6.7. Timings on the SD/eMMC Interface 5.4.6.8. Command Queuing
5.4.6.1. Block Diagram x
5.4.6.1.1. Bus Interface Unit 5.4.6.1.2. Reset Control Module 5.4.6.1.3. Synchronization Module 5.4.6.1.4. Response Module 5.4.6.1.5. FIFO Interface Unit 5.4.6.1.6. Card Interface Unit 5.4.6.1.7. System Requester Interface 5.4.6.1.8. System Manager Interface 5.4.6.1.9. Host Settings Interface 5.4.6.1.10. Command Queue Settings
5.4.6.2. Configuration x
5.4.6.2.1. Host Capabilities
5.4.6.3. DMA Operation x
5.4.6.3.1. SDMA Operation 5.4.6.3.2. ADMA2 Operation 5.4.6.3.3. ADMA3 Operation 5.4.6.3.4. DMA Errors
5.4.6.5. Resets x
5.4.6.5.1. Controller Software Reset
5.4.6.7. Timings on the SD/eMMC Interface x
5.4.6.7.1. Tuning
5.4.6.7.1. Tuning x
5.4.6.7.1.1. Tuning Sequence for SD 5.4.6.7.1.2. Tuning Sequence for eMMC
5.4.6.8. Command Queuing x
5.4.6.8.1. Optional Modes of Operation 5.4.6.8.2. Task Execution Order
5.4.7. SD/eMMC Programming Model x
5.4.7.1. Combo PHY Register Interface 5.4.7.2. Pre-Initialization Sequence 5.4.7.3. Error Handling
5.4.7.1. Combo PHY Register Interface x
5.4.7.1.1. Read Access 5.4.7.1.2. Write Access
5.4.7.3. Error Handling x
5.4.7.3.1. Generic Operation Error Recovery 5.4.7.3.2. Command Queueing Error Recovery 5.4.7.3.3. Combo PHY Underrun/Overflow Error Recovery
5.4.9. SD/eMMC Design Guidelines and Examples x
5.4.9.1. SD Card Level Shifting
5.5. Combo DLL PHY x
5.5.1. Combo DLL PHY Differences Among Intel SoC Device Families 5.5.2. Combo DLL PHY Use Cases 5.5.3. Combo DLL PHY Features 5.5.4. Combo DLL PHY System Integration 5.5.5. Combo DLL PHY Signal Description 5.5.6. Combo DLL PHY Functional Description 5.5.7. Combo DLL PHY Programming Model 5.5.8. Combo DLL PHY Address Map and Register Definitions
5.5.6. Combo DLL PHY Functional Description x
5.5.6.1. Block Diagram 5.5.6.2. Operation Modes 5.5.6.3. Clocking 5.5.6.4. DLL Functionality 5.5.6.5. Write Data Path 5.5.6.6. Read Data Path 5.5.6.7. Generation of CE#, CLE, ALE, WP#, and RE# signals 5.5.6.8. Resets
5.5.6.4. DLL Functionality x
5.5.6.4.1. DLL Locking
5.5.6.5. Write Data Path x
5.5.6.5.1. DQS Write Signal Generation 5.5.6.5.2. Output Enable Signal Generation
5.5.6.6. Read Data Path x
5.5.6.6.1. Generation of Dynamic Termination Signal 5.5.6.6.2. Input Enable Signal Generation
5.5.7. Combo DLL PHY Programming Model x
5.5.7.1. Initialization Procedure 5.5.7.2. DLL Lock Debugging 5.5.7.3. DLL Secondary Delay Configuration 5.5.7.4. Handling FIFO DQS Overflow and Underrun 5.5.7.5. Tuning for SD/eMMC Interface
5.6. USB 3.1 Gen1 Controller x
5.6.1. USB 3.1 Gen1 Controller Differences Among Intel SoC Device Families 5.6.2. USB 3.1 Gen1 Controller Use Cases 5.6.3. USB 3.1 Gen1 Controller Features 5.6.4. USB 3.1 Gen1 Controller System Integration 5.6.5. USB 3.1 Gen1 Controller Functional Description 5.6.6. USB 3.1 Gen1 Controller Programming Model 5.6.7. USB 3.1 Gen1 Controller Address Map and Register Definitions 5.6.8. USB 3.1 Gen1 Controller Design Guidelines and Examples
5.6.3. USB 3.1 Gen1 Controller Features x
5.6.3.1. Unsupported Features
5.6.5. USB 3.1 Gen1 Controller Functional Description x
5.6.5.1. Manager Interface 5.6.5.2. AHB Subordinate Interface 5.6.5.3. Application interface Unit 5.6.5.4. Bus Management Unit 5.6.5.5. Packet FIFO Controller 5.6.5.6. RAM 5.6.5.7. MAC 5.6.5.8. DMA 5.6.5.9. Loopback 5.6.5.10. PHY Interfaces 5.6.5.11. USB 3.1 Gen1 Controller Clocks 5.6.5.12. USB 3.1 Gen1 Controller Resets
5.6.5.1. Manager Interface x
5.6.5.1.1. AXI Manager Interface
5.6.5.8. DMA x
5.6.5.8.1. Reception Flow Details 5.6.5.8.2. Transmission Flow Details
5.6.5.9. Loopback x
5.6.5.9.1. Software Loopback Mode 5.6.5.9.2. BIST Loopback Mode
5.6.5.10. PHY Interfaces x
5.6.5.10.1. USB 2.0 ULPI PHY 5.6.5.10.2. PIPE Interface 5.6.5.10.3. USB 3.1 Mode-switch Controller 5.6.5.10.4. USB Type-C Receptacle 5.6.5.10.5. Performance 5.6.5.10.6. Maximum USB Bandwidth
5.6.5.10.1. USB 2.0 ULPI PHY x
5.6.5.10.1.1. Signals
5.6.5.10.2. PIPE Interface x
5.6.5.10.2.1. IO Connection
5.6.5.10.3. USB 3.1 Mode-switch Controller x
5.6.5.10.3.1. USB3.1 Mode-switch Controller Signals
5.6.5.11. USB 3.1 Gen1 Controller Clocks x
5.6.5.11.1. Clock Enable 5.6.5.11.2. Clock Gating
5.6.5.12. USB 3.1 Gen1 Controller Resets x
5.6.5.12.1. Reset Requirements 5.6.5.12.2. Software Resets 5.6.5.12.3. Warm Reset
5.6.6. USB 3.1 Gen1 Controller Programming Model x
5.6.6.1. Programming Controller in Host Mode 5.6.6.2. Programming Controller in Device Mode
5.6.6.1. Programming Controller in Host Mode x
5.6.6.1.1. Initialization Phase 5.6.6.1.2. Host Controller Capability Registers 5.6.6.1.3. Programming Model 5.6.6.1.4. Device Initialization
5.6.6.2. Programming Controller in Device Mode x
5.6.6.2.1. Device Power-on or Soft Reset 5.6.6.2.2. Initialization on USB Reset 5.6.6.2.3. Initialization on Connect Done 5.6.6.2.4. Initialization on SetAddress Request 5.6.6.2.5. Initialization on SetConfiguration or SetInterface Request 5.6.6.2.6. Agilex 5 Programming Model 5.6.6.2.7. Controller as Host
5.6.8. USB 3.1 Gen1 Controller Design Guidelines and Examples x
5.6.8.1. USB 3.1 Gen1 Controller Wakeup and Power Control
5.7. USB 2.0 OTG Controller x
5.7.1. USB 2.0 OTG Controller Differences Among Intel SoC Device Families 5.7.2. USB 2.0 OTG Controller Use Cases 5.7.3. USB 2.0 OTG Controller Features 5.7.4. USB 2.0 OTG Controller System Integration 5.7.5. USB 2.0 OTG Controller Functional Description 5.7.6. USB 2.0 OTG Controller Programming Model 5.7.7. USB 2.0 OTG Controller Address Map and Register Definitions
5.7.4. USB 2.0 OTG Controller System Integration x
5.7.4.1. USB 2.0 ULPI PHY Signal Description
5.7.5. USB 2.0 OTG Controller Functional Description x
5.7.5.1. Distributed Virtual Memory Support 5.7.5.2. Initiator Interface 5.7.5.3. Target Interface 5.7.5.4. USB OTG Controller Components 5.7.5.5. DMA 5.7.5.6. Local Memory Buffer 5.7.5.7. Clocks 5.7.5.8. Resets 5.7.5.9. Interrupts
5.7.5.4. USB OTG Controller Components x
5.7.5.4.1. Application Interface Unit 5.7.5.4.2. Packet FIFO Controller 5.7.5.4.3. SPRAM 5.7.5.4.4. MAC 5.7.5.4.5. Wakeup and Power Control 5.7.5.4.6. PHY Interface Unit
5.7.5.4.4. MAC x
5.7.5.4.4.1. USB Transactions 5.7.5.4.4.2. Host Protocol 5.7.5.4.4.3. Device Protocol 5.7.5.4.4.4. OTG Protocol
5.7.5.7. Clocks x
5.7.5.7.1. Clock Gating
5.7.5.8. Resets x
5.7.5.8.1. Reset Requirements 5.7.5.8.2. Hardware Reset 5.7.5.8.3. Software Reset 5.7.5.8.4. Taking the USB 2.0 OTG Controller Out of Reset
5.7.6. USB 2.0 OTG Controller Programming Model x
5.7.6.1. Enabling SPRAM ECCs 5.7.6.2. Host Initialization 5.7.6.3. Host Transaction 5.7.6.4. Device Initialization 5.7.6.5. Device Transaction
5.7.6.5. Device Transaction x
5.7.6.5.1. IN Transactions 5.7.6.5.2. OUT Transactions 5.7.6.5.3. Control Transfers
5.8. I3C Controller x
5.8.1. I3C Controller Differences Among Intel SoC Device Families 5.8.2. I3C Controller Use Cases 5.8.3. I3C Controller Features 5.8.4. I3C Controller System Integration 5.8.5. I3C Controller Signal Description 5.8.6. I3C Controller Functional Description 5.8.7. I3C Controller Programming Model 5.8.8. I3C Controller Address Map and Register Definitions 5.8.9. I3C Controller Design Guidelines and Examples
5.8.5. I3C Controller Signal Description x
5.8.5.1. Interface to HPS I/O 5.8.5.2. FPGA Routing
5.8.6. I3C Controller Functional Description x
5.8.6.1. Device Address 5.8.6.2. In-Band Interrupt (IBI) 5.8.6.3. Simple-Transfer DMA (SDMA) Handshake 5.8.6.4. Interrupts in I3C Controller 5.8.6.5. Overview of Master Role in I3C 5.8.6.6. Overview of Slave Role in I3C
5.8.6.5. Overview of Master Role in I3C x
5.8.6.5.1. Dynamic Address Assignment (DAA) 5.8.6.5.2. In-Band Interrupt (IBI) Detection and Handling 5.8.6.5.3. I3C Slave Interrupt Request (SIR) 5.8.6.5.4. Disabling I3C Master 5.8.6.5.5. Aborting Transfers of I3C Master 5.8.6.5.6. I3C Master Request (MR) 5.8.6.5.7. Master Command Data Structures 5.8.6.5.8. Response Data Structure 5.8.6.5.9. Operation Modes of I3C Controller 5.8.6.5.10. SCL Generation and Timings Based on Bus Configuration 5.8.6.5.11. Derivation of I3C/I2C Timing Parameters from Timing Registers 5.8.6.5.12. Error Detection 5.8.6.5.13. Defining Byte Support 5.8.6.5.14. Broadcast CCCs in I2C Speed 5.8.6.5.15. BUS RESET Generation DMA Controller Interface
5.8.6.5.1. Dynamic Address Assignment (DAA) x
5.8.6.5.1.1. SETDASA (Format 1: Primary) 5.8.6.5.1.2. SETDASA (Format 2: Point-to-Point) 5.8.6.5.1.3. ENTDAA 5.8.6.5.1.4. SETAASA
5.8.6.5.3. I3C Slave Interrupt Request (SIR) x
5.8.6.5.3.1. SIR Response Control
5.8.6.5.6. I3C Master Request (MR) x
5.8.6.5.6.1. MR Response Control
5.8.6.5.7. Master Command Data Structures x
5.8.6.5.7.1. Transfer Command Data Structure 5.8.6.5.7.2. Transfer Argument Data Structure 5.8.6.5.7.3. Short Data Argument Data Structure 5.8.6.5.7.4. Address Assignment Command Data Structure
5.8.6.5.9. Operation Modes of I3C Controller x
5.8.6.5.9.1. Single Data Rate (SDR) Transfers in Master Mode 5.8.6.5.9.2. Broadcast CCC Write Transfers 5.8.6.5.9.3. Directed Write and Read Transfers 5.8.6.5.9.4. Directed CCC Transfer Targeted to Multiple Slaves 5.8.6.5.9.5. I3C Private Write or Read Transfers 5.8.6.5.9.6. I2C Private Write or Read Transfers 5.8.6.5.9.7. Implication of TX-FIFO Empty and RX-FIFO Full Conditions 5.8.6.5.9.8. Implication of TOC and ROC Bit Settings for SDR Transfers
5.8.6.5.12. Error Detection x
5.8.6.5.12.1. Detecting Error Type in the Processed Commands 5.8.6.5.12.2. SDR Error Detection and Recovery Methods
5.8.6.6. Overview of Slave Role in I3C x
5.8.6.6.1. Description of the Slave Role in I3C 5.8.6.6.2. I3C versus I2C Role Selection 5.8.6.6.3. Slave Role Related Registers 5.8.6.6.4. Handling Address Assignment 5.8.6.6.5. CCC Transfers with I3C Slave 5.8.6.6.6. Private Data Transfers 5.8.6.6.7. Handling Private Transmit (Master Read) Transfers 5.8.6.6.8. Slave Interrupt Request Generation 5.8.6.6.9. Master Request Generation 5.8.6.6.10. Disabling I3C Slave 5.8.6.6.11. Data Structure in I3C Slave
5.8.6.6.6. Private Data Transfers x
5.8.6.6.6.1. Overview of Private Data Transfers
5.8.6.6.11. Data Structure in I3C Slave x
5.8.6.6.11.1. Transmit Command Data Structure 5.8.6.6.11.2. Response Data Structure
5.8.7. I3C Controller Programming Model x
5.8.7.1. Initializing Common Registers 5.8.7.2. Initializing Master Registers 5.8.7.3. Enabling the Controller 5.8.7.4. Master Mode Operation 5.8.7.5. Initializing Slave Registers 5.8.7.6. Slave Mode Operation 5.8.7.7. DMA Controller Operation
5.8.7.1. Initializing Common Registers x
5.8.7.1.1. Programming Threshold Control Registers 5.8.7.1.2. Programming Interrupt Related Registers
5.8.7.2. Initializing Master Registers x
5.8.7.2.1. Programming DEVICE_ADDR Register 5.8.7.2.2. Programming Timing Registers
5.8.7.4. Master Mode Operation x
5.8.7.4.1. Slave Address Assignment 5.8.7.4.2. Receiving IBI (All types) 5.8.7.4.3. CCC Read/Write Transfers 5.8.7.4.4. Private Read/Write Transfers 5.8.7.4.5. Programming Flow to Prepare the Controller to Switch to Master Mode
5.8.7.4.1. Slave Address Assignment x
5.8.7.4.1.1. ENTDAA Transfer 5.8.7.4.1.2. Issue SETDASA CCC Command
5.8.7.4.3. CCC Read/Write Transfers x
5.8.7.4.3.1. Broadcast CCC Transfer in Master Mode 5.8.7.4.3.2. Directed CCC Transfer in Master Mode
5.8.7.5. Initializing Slave Registers x
5.8.7.5.1. Programming Interrupt Related Registers
5.8.7.6. Slave Mode Operation x
5.8.7.6.1. Private Receive (Master Write) Transfers in Slave Mode 5.8.7.6.2. Private Transmit (Master Read) Transfers in Slave Mode 5.8.7.6.3. Programming Flow for Generating Slave Interrupt Request 5.8.7.6.4. Programming Flow for Generating Master Request 5.8.7.6.5. Programming Flow to Prepare the Controller to Switch to Master Mode 5.8.7.6.6. Command Pipeline and Aggregation of Response Queue Threshold Interrupt 5.8.7.6.7. Error Recovery Flow 5.8.7.6.8. CCC Updated Interrupt Flow 5.8.7.6.9. Flow for Disable and TX/RX/CMD/Response Queue Reset
5.8.7.6.7. Error Recovery Flow x
5.8.7.6.7.1. S0-S5 Error Handling
5.8.9. I3C Controller Design Guidelines and Examples x
5.8.9.1. Guidelines for External SDA and SCL Signals via HPS I/O in the Board Design
5.9. I2C Controller x
5.9.1. I2C Controller Differences Among Intel SoC Device Families 5.9.2. I2C Controller Use Cases 5.9.3. I2C Controller Features 5.9.4. I2C Controller System Integration 5.9.5. I2C Controller Signal Description 5.9.6. I2C Controller Functional Description 5.9.7. I2C Controller Programming Model 5.9.8. I2C Controller Address Map and Register Definitions 5.9.9. I2C Controller Design Guidelines and Examples
5.9.6. I2C Controller Functional Description x
5.9.6.1. Feature Usage 5.9.6.2. Behavior 5.9.6.3. Protocol Details 5.9.6.4. Multiple Master Arbitration 5.9.6.5. Clock Frequency Configuration 5.9.6.6. SDA Hold Time 5.9.6.7. DMA Controller Interface 5.9.6.8. Clocks 5.9.6.9. Resets
5.9.6.2. Behavior x
5.9.6.2.1. START and STOP Generation 5.9.6.2.2. Combined Formats
5.9.6.3. Protocol Details x
5.9.6.3.1. START and STOP Conditions 5.9.6.3.2. Addressing Slave Protocol 5.9.6.3.3. Transmitting and Receiving Protocol 5.9.6.3.4. START BYTE Transfer Protocol
5.9.6.4. Multiple Master Arbitration x
5.9.6.4.1. Clock Synchronization
5.9.6.5. Clock Frequency Configuration x
5.9.6.5.1. Minimum High and Low Counts
5.9.6.5.1. Minimum High and Low Counts x
5.9.6.5.1.1. Calculating High and Low Counts
5.9.6.9. Resets x
5.9.6.9.1. Taking the I2C Controller Out of Reset
5.9.7. I2C Controller Programming Model x
5.9.7.1. Slave Mode Operation 5.9.7.2. Master Mode Operation 5.9.7.3. Disabling the I2C Controller 5.9.7.4. Abort Transfer 5.9.7.5. DMA Controller Operation
5.9.7.1. Slave Mode Operation x
5.9.7.1.1. Initial Configuration 5.9.7.1.2. Slave-Transmitter Operation for a Single Byte 5.9.7.1.3. Slave-Receiver Operation for a Single Byte 5.9.7.1.4. Slave-Transfer Operation for Bulk Transfers 5.9.7.1.5. Slave Programming Model
5.9.7.2. Master Mode Operation x
5.9.7.2.1. Initial Configuration 5.9.7.2.2. Dynamic IC_TAR or IC_10BITADDR_MASTER Update 5.9.7.2.3. Master Transmit and Master Receive 5.9.7.2.4. Master Programming Model
5.9.7.5. DMA Controller Operation x
5.9.7.5.1. Transmit FIFO Underflow 5.9.7.5.2. Transmit Watermark Level 5.9.7.5.3. Transmit FIFO Overflow 5.9.7.5.4. Receive FIFO Overflow 5.9.7.5.5. Receive Watermark Level 5.9.7.5.6. Receive FIFO Underflow
5.9.9. I2C Controller Design Guidelines and Examples x
5.9.9.1. I2C Interface Design Guidelines
5.10. SPI Controller x
5.10.1. SPI Controller Differences Among Intel SoC Device Families 5.10.2. SPI Controller Use Cases 5.10.3. SPI Controller Features 5.10.4. SPI Controller System Integration 5.10.5. SPI Controller Signal Description 5.10.6. SPI Controller Functional Description 5.10.7. SPI Controller Programming Model 5.10.8. SPI Controller Address Map and Register Definitions
5.10.5. SPI Controller Signal Description x
5.10.5.1. Interface to HPS I/O 5.10.5.2. FPGA Routing
5.10.6. SPI Controller Functional Description x
5.10.6.1. Protocol Details and Standards Compliance 5.10.6.2. Overview 5.10.6.3. Serial Bit-Rate Clocks 5.10.6.4. Transmit and Receive FIFO Buffers 5.10.6.5. SPI Interrupts 5.10.6.6. Transfer Modes 5.10.6.7. SPI Master 5.10.6.8. SPI Slave 5.10.6.9. Partner Connection Interfaces 5.10.6.10. DMA Controller Interface 5.10.6.11. Slave Interface 5.10.6.12. Clocks and Resets
5.10.6.3. Serial Bit-Rate Clocks x
5.10.6.3.1. SPI Master Bit-Rate Clock 5.10.6.3.2. SPI Slave Bit-Rate Clock
5.10.6.6. Transfer Modes x
5.10.6.6.1. Transmit and Receive 5.10.6.6.2. Transmit Only 5.10.6.6.3. Receive Only 5.10.6.6.4. EEPROM Read
5.10.6.7. SPI Master x
5.10.6.7.1. Glue Logic for Master Port ss_in_n 5.10.6.7.2. Multi-Master System 5.10.6.7.3. RXD Sample Delay 5.10.6.7.4. Data Transfers 5.10.6.7.5. Master SPI and SSP Serial Transfers 5.10.6.7.6. Master Microwire Serial Transfers
5.10.6.8. SPI Slave x
5.10.6.8.1. Slave SPI and SSP Serial Transfers 5.10.6.8.2. Serial Transfers
5.10.6.9. Partner Connection Interfaces x
5.10.6.9.1. Motorola SPI Protocol 5.10.6.9.2. Texas Instruments Synchronous Serial Protocol (SSP) 5.10.6.9.3. National Semiconductor Microwire Protocol
5.10.6.11. Slave Interface x
5.10.6.11.1. Control and Status Register Access 5.10.6.11.2. Data Register Access
5.10.6.12. Clocks and Resets x
5.10.6.12.1. Clock Gating 5.10.6.12.2. Taking the SPI Controller Out of Reset
5.10.7. SPI Controller Programming Model x
5.10.7.1. Master SPI and SSP Serial Transfers 5.10.7.2. Master Microwire Serial Transfers 5.10.7.3. Slave SPI and SSP Serial Transfers 5.10.7.4. Slave Microwire Serial Transfers 5.10.7.5. Software Control for Slave Selection 5.10.7.6. DMA Controller Operation
5.10.7.5. Software Control for Slave Selection x
5.10.7.5.1. Example: Slave Selection Software Flow for SPI Master 5.10.7.5.2. Example: Slave Selection Software Flow for SPI Slave
5.10.7.6. DMA Controller Operation x
5.10.7.6.1. Transmit FIFO Buffer Underflow 5.10.7.6.2. Transmit FIFO Watermark 5.10.7.6.3. Transmit FIFO Buffer Overflow 5.10.7.6.4. Receive FIFO Buffer Overflow 5.10.7.6.5. Choosing Receive Watermark Level 5.10.7.6.6. Receive FIFO Buffer Underflow
5.10.7.6.2. Transmit FIFO Watermark x
5.10.7.6.2.1. Example 1: Transmit FIFO Watermark Level = 64 5.10.7.6.2.2. Example 2: Transmit FIFO Watermark Level = 192
5.11. Timers x
5.11.1. Timers Differences Among Intel SoC Device Families 5.11.2. Timers Use Cases 5.11.3. Timers Features 5.11.4. Timers System Integration 5.11.5. Timers Functional Description 5.11.6. Timers Programming Model 5.11.7. Timers Address Map and Register Definitions
5.11.5. Timers Functional Description x
5.11.5.1. Clocks 5.11.5.2. Resets 5.11.5.3. Interrupts
5.11.6. Timers Programming Model x
5.11.6.1. Initialization 5.11.6.2. Enabling the Timers 5.11.6.3. Disabling the Timers 5.11.6.4. Loading the Timers Countdown Value 5.11.6.5. Timers Servicing Interrupts
5.11.6.5. Timers Servicing Interrupts x
5.11.6.5.1. Clearing the Interrupt 5.11.6.5.2. Checking the Interrupt Status 5.11.6.5.3. Masking the Interrupt
5.12. Watchdog Timers x
5.12.1. Watchdog Timers Differences Among Intel SoC Device Families 5.12.2. Watchdog Timers Use Cases 5.12.3. Watchdog Timers Features 5.12.4. Watchdog Timers System Integration 5.12.5. Watchdog Timers Functional Description 5.12.6. Watchdog Timers Programming Model 5.12.7. Watchdog Timers Address Map and Register Definitions
5.12.5. Watchdog Timers Functional Description x
5.12.5.1. Watchdog Timers Counter 5.12.5.2. Watchdog Timers Pause Mode 5.12.5.3. Watchdog Timers Clocks 5.12.5.4. Watchdog Timers Resets
5.12.6. Watchdog Timers Programming Model x
5.12.6.1. Setting the Timeout Period Values 5.12.6.2. Selecting the Output Response Mode 5.12.6.3. Enabling and Initially Starting a Watchdog Timers 5.12.6.4. Reloading a Watchdog Counter 5.12.6.5. Pausing a Watchdog Timers 5.12.6.6. Disabling and Stopping a Watchdog Timers 5.12.6.7. Watchdog Timers State Machine
5.13. UART Controller x
5.13.1. UART Controller Differences Among Intel SoC Device Families 5.13.2. UART Controller Use Cases 5.13.3. UART Controller Features 5.13.4. UART Controller System Integration 5.13.5. UART Controller Signal Description 5.13.6. UART Controller Functional Description 5.13.7. UART Controller Address Map and Register Definitions 5.13.8. UART Controller Design Guidelines and Example
5.13.5. UART Controller Signal Description x
5.13.5.1. HPS I/O Pins 5.13.5.2. FPGA Routing
5.13.6. UART Controller Functional Description x
5.13.6.1. FIFO Buffer Support 5.13.6.2. UART Serial Protocol 5.13.6.3. Automatic Flow Control 5.13.6.4. Clocks 5.13.6.5. Resets 5.13.6.6. UART Controller Interrupts 5.13.6.7. UART Controller DMA Controller Operation
5.13.6.7. UART Controller DMA Controller Operation x
5.13.6.7.1. Transmit FIFO Underflow 5.13.6.7.2. Transmit Watermark Level 5.13.6.7.3. Transmit FIFO Overflow 5.13.6.7.4. Receive FIFO Overflow 5.13.6.7.5. Receive Watermark Level 5.13.6.7.6. Receive FIFO Underflow
5.14. General-Purpose I/O Interface (GPIO) x
5.14.1. GPIO Differences Among Intel SoC Device Families 5.14.2. GPIO Use Cases 5.14.3. GPIO Features 5.14.4. GPIO System Integration 5.14.5. GPIO Functional Description 5.14.6. GPIO Programming Model 5.14.7. GPIO Address Map and Register Definitions
5.14.5. GPIO Functional Description x
5.14.5.1. Interrupts 5.14.5.2. Debounce Operation 5.14.5.3. Pin Directions 5.14.5.4. Taking the GPIO Interface Out of Reset
5.15. Hard Processor System I/O Pin Multiplexing x
5.15.1. I/O Pin Multiplexing Differences Among Intel SoC Device Families 5.15.2. I/O Pin Multiplexing Use Cases 5.15.3. I/O Pin Multiplexing Features 5.15.4. I/O Pin Multiplexing System Integration 5.15.5. I/O Pin Multiplexing Functional Description 5.15.6. I/O Pin Multiplexing Address Map and Register Definitions 5.15.7. I/O Pin Multiplexing Design Guidelines and Examples
5.15.5. I/O Pin Multiplexing Functional Description x
5.15.5.1. I/O Pins 5.15.5.2. FPGA Access 5.15.5.3. I/O Control Registers
5.15.5.3. I/O Control Registers x
5.15.5.3.1. Dedicated Pin MUX Registers 5.15.5.3.2. Dedicated Configuration Registers 5.15.5.3.3. FPGA Access MUX Registers 5.15.5.3.4. HPS JTAG Pin MUX Register
5.15.7. I/O Pin Multiplexing Design Guidelines and Examples x
5.15.7.1. Boundary Scan for HPS 5.15.7.2. HPS I/O Design Considerations
6. System Manager x
6.1. System Manager Differences Among Intel SoC Device Families 6.2. System Manager Use Cases 6.3. System Manager Features 6.4. System Manager System Integration 6.5. System Manager Address Map and Register Definitions 6.6. System Manager Revision History
6.4. System Manager System Integration x
6.4.1. Additional Module Control 6.4.2. FPGA Interface Enables 6.4.3. ECC and Parity Control 6.4.4. Preloader Handoff Information 6.4.5. Clocks 6.4.6. Resets
6.4.1. Additional Module Control x
6.4.1.1. DMA Controller 6.4.1.2. NAND Flash Controller 6.4.1.3. EMAC 6.4.1.4. USB 2.0 OTG Controller 6.4.1.5. USB 3.1 Controller 6.4.1.6. NOC Registers 6.4.1.7. Boot Scratch Space 6.4.1.8. I3C Controller 6.4.1.9. SD/eMMC Controller 6.4.1.10. GPIO Interconnect Between the HPS and FPGA 6.4.1.11. Watchdog Timer
7. Clock Manager x
7.1. Clock Manager Differences Among Intel SoC Device Families 7.2. Clock Manager Use Cases 7.3. Clock Manager Features 7.4. Clock Manager System Integration 7.5. Clock Manager Signal Description 7.6. Clock Manager Functional Description 7.7. Clock Manager Programming Model 7.8. Clock Manager Address Map and Register Definitions 7.9. Clock Manager Design Guidelines and Examples
7.2. Clock Manager Use Cases x
7.2.1. A76 Core Power and Performance Trade-off 7.2.2. A55 Core Power and Performance Trade-off 7.2.3. DSU Power and Performance Trade-off 7.2.4. Peripheral Clock Generation 7.2.5. Peripheral Power Optimization 7.2.6. Clock Tree Clock Sources 7.2.7. Boot Clock 7.2.8. Power-on-reset 7.2.9. Boot Mode 7.2.10. Disable Clocks Before Reset Assertion 7.2.11. PLL Configuration using JTAG
7.4. Clock Manager System Integration x
7.4.1. Top Level of the Clock Groups 7.4.2. Clock Manager Block Diagram
7.5. Clock Manager Signal Description x
7.5.1. Hardware-Managed Clocks 7.5.2. Software-Managed Clocks
7.6. Clock Manager Functional Description x
7.6.1. PLL Wrapper 7.6.2. MPU/DSU and APS/CCU Clock Groups 7.6.3. PSS Clock Group 7.6.4. MPFE Clock Group 7.6.5. EMAC and XGMAC Clock Group 7.6.6. USB31 Clock Group 7.6.7. SD/eMMC, NAND, and SoftPHY/ComboPHY 7.6.8. H2F User Clock Group 7.6.9. F2H User Clock Group 7.6.10. GPIO Debounce Clock Group 7.6.11. CoreSight Clocks 7.6.12. PSI Clock Group (to the SDM)
7.6.3. PSS Clock Group x
7.6.3.1. USB2OTG Clock Group
7.6.5. EMAC and XGMAC Clock Group x
7.6.5.1. EMAC Clocks 7.6.5.2. XGMAC Clocks
7.7. Clock Manager Programming Model x
7.7.1. Example Configuration of Registers for Default Operation (640 MHz CPU) 7.7.2. Example Configuration of Registers for Power Optimized (1000 MHz CPU) 7.7.3. Example Configuration of Registers for Performance Optimized (1800 MHz CPU) 7.7.4. Example Configuration of Registers for Power Optimized (1200 MHz CPU) 7.7.5. Summary Table of Registers Used to Program Clocks
8. Reset Manager x
8.1. Reset Manager Differences Among Intel SoC Device Families 8.2. Reset Manager Use Cases 8.3. Reset Manager Features 8.4. Reset Manager System Integration 8.5. Reset Manager Signal Description 8.6. Reset Manager Functional Description 8.7. Reset Manager Programming Model 8.8. Reset Manager Address Map and Register Definitions
8.5. Reset Manager Signal Description x
8.5.1. HPS Reset Domains 8.5.2. HPS Reset Domain to Signal Mapping 8.5.3. HPS Reset Priority
8.6. Reset Manager Functional Description x
8.6.1. System Reset High Level Flow 8.6.2. Hardware Reset Sequences 8.6.3. Software Reset Sequences
8.6.2. Hardware Reset Sequences x
8.6.2.1. POR De-assertion Sequence 8.6.2.2. Reset De-assertion Sequence 8.6.2.3. HPS Idle Sequence 8.6.2.4. Reset Assertion Sequence 8.6.2.5. Wait for Reset Requests to De-assert Sequence
8.6.3. Software Reset Sequences x
8.6.3.1. Normal Operation 8.6.3.2. Cold and Warm Reset Sequence 8.6.3.3. Debug Reset Sequence
8.7. Reset Manager Programming Model x
8.7.1. H2F Bridge Reset Sequence 8.7.2. LWH2F Bridge Reset Sequence 8.7.3. F2H Bridge Reset Sequence 8.7.4. F2SDRAM Bridge Reset Sequence 8.7.5. HPS_COLD_nRESET Pin Function
8.7.5. HPS_COLD_nRESET Pin Function x
8.7.5.1. FPGA Boot First 8.7.5.2. HPS Boot First 8.7.5.3. HPS Pin-triggered Cold Reset 8.7.5.4. HPS Mailbox-triggered Cold Reset
9. Power Management x
9.1. Power Management Differences Among Intel SoC Device Families 9.2. Power Management System Integration 9.3. Power Management Functional Description 9.4. Power Management Design Guidelines and Examples
9.3. Power Management Functional Description x
9.3.1. DSU L3 Cache Power Gating 9.3.2. CPU Core Power Gating 9.3.3. PSS Partition Power Management
10. Address Map x
10.1. Address Space Introduction 10.2. Total Address Spaces 10.3. MPU Address Space 10.4. NOC: Level 3 and Level 4 Address Space 10.5. FPGA Slaves Address Space 10.6. Lightweight FPGA Slaves Address Space 10.7. FPGA-to-HPS Address Space 10.8. FPGA-to-SDRAM Address Space
10.2. Total Address Spaces x
10.2.1. Total Address Map Graphical 10.2.2. Total Address Map Tabular 10.2.3. Total Address Map for F2H and F2SDRAM
11. Bridges x
11.1. Bridges Differences Among Intel SoC Device Families 11.2. Bridges Use Cases 11.3. Bridges Features 11.4. Bridges System Integration 11.5. Bridges Functional Description 11.6. Bridges Clocks and Resets 11.7. Bridges Address Map and Register Definitions 11.8. Bridges Design Guidelines and Examples
11.2. Bridges Use Cases x
11.2.1. FPGA-to-HPS Bridge 11.2.2. HPS-to-FPGA 11.2.3. Lightweight HPS-to-FPGA 11.2.4. FPGA-to-SDRAM
11.2.1. FPGA-to-HPS Bridge x
11.2.1.1. FPGA-to-HPS Initialization and Shutdown Procedure
11.5. Bridges Functional Description x
11.5.1. FPGA-to-HPS Bridge 11.5.2. HPS-to-FPGA Bridge 11.5.3. Lightweight HPS-to-FPGA Bridge 11.5.4. FPGA-to-SDRAM Bridge
11.5.4. FPGA-to-SDRAM Bridge x
11.5.4.1. Transaction Buffer Unit Interface
11.6. Bridges Clocks and Resets x
11.6.1. FPGA-to-HPS Bridge Clocks and Resets 11.6.2. HPS-to-FPGA Bridge Clocks and Resets 11.6.3. Lightweight HPS-to-FPGA Bridge Clocks and Resets 11.6.4. FPGA-to-SDRAM Clocks and Resets
11.8. Bridges Design Guidelines and Examples x
11.8.1. Recommended Starting Point for Interface Designs 11.8.2. Information on How to Configure and Use the Bridges 11.8.3. Bridges Example Transactions
11.8.3. Bridges Example Transactions x
11.8.3.1. FPGA-to-SDRAM Direct (Cache Non-Allocate) 11.8.3.2. FPGA-to-HPS CCU to Memory (Cache Non-Allocate) 11.8.3.3. FPGA-to-HPS CCU to Memory (Cache-Allocate) 11.8.3.4. FPGA-to-HPS CCU to Peripherals (Device Non-Bufferable) 11.8.3.5. FPGA-to-HPS Example Transactions Summary
12. Interfaces x
12.1. HPS Mailbox 12.2. MPFE and MPFE-lite 12.3. EMAC GMII through FPGA Fabric
12.1. HPS Mailbox x
12.1.1. HPS Mailbox Differences Among Intel SoC Device Families 12.1.2. HPS Mailbox Features 12.1.3. Accessing HPS Mailbox Services 12.1.4. HPS Mailbox System Integration 12.1.5. HPS Mailbox Address Map and Register Definitions
12.2. MPFE and MPFE-lite x
12.2.1. MPFE and MPFE-lite Terminology 12.2.2. MPFE and MPFE-lite Differences Among Intel SoC Families 12.2.3. MPFE and MPFE-lite Use Cases 12.2.4. MPFE and MPFE-lite Features 12.2.5. MPFE and MPFE-lite System Integration 12.2.6. MPFE and MPFE-lite Functional Description 12.2.7. MPFE and MPFE-lite Address Map and Register Definitions
12.2.3. MPFE and MPFE-lite Use Cases x
12.2.3.1. Fabric Bypass 12.2.3.2. One 16-bit SDRAM channel 12.2.3.3. One 32-bit SDRAM channel 12.2.3.4. Two 16-bit SDRAM channels utilizing a single IOBank 12.2.3.5. Two 16-bit or two 32-bit SDRAM channels utilizing two IOBank 12.2.3.6. Four 16-bit SDRAM channels utilizing two IOBank 12.2.3.7. Supports up to 512GB of memory 12.2.3.8. Supports sideband ECC on DDR4/5 12.2.3.9. Supports inband ECC on LPDDR4/5
12.2.5. MPFE and MPFE-lite System Integration x
12.2.5.1. Block Diagram (High Level) 12.2.5.2. IOBank Introduction
12.2.6. MPFE and MPFE-lite Functional Description x
12.2.6.1. Block Diagram 12.2.6.2. Interfaces 12.2.6.3. AxUSER Bit Connectivity 12.2.6.4. Fabric Bypass 12.2.6.5. FPGA-to-SDRAM bridge 12.2.6.6. FPGA-to-HPS bridge 12.2.6.7. MPFE NoC 12.2.6.8. MPFE-lite / MPFE-lite NoC 12.2.6.9. Address Gap Glue Logic 12.2.6.10. Resets Circuitry 12.2.6.11. Clocks Circuitry 12.2.6.12. Preserving SDRAM contents
12.3. EMAC GMII through FPGA Fabric x
12.3.1. GMII to RGMII through RGMII adapter via FPGA HVIOs 12.3.2. GMII to SGMII+ through Multi-Rate Ethernet PHY Adapter via FPGA IOs
12.3.1. GMII to RGMII through RGMII adapter via FPGA HVIOs x
12.3.1.1. Use Cases 12.3.1.2. System Integration 12.3.1.3. Architecture 12.3.1.4. Programming Model 12.3.1.5. RGMII Design Guidelines
12.3.1.3. Architecture x
12.3.1.3.1. HPS GMII to RGMII Adapter Intel FPGA IP 12.3.1.3.2. MDIO Management Interface 12.3.1.3.3. Timestamp Interface 12.3.1.3.4. RGMII HVIO Pin Assignment
12.3.1.5. RGMII Design Guidelines x
12.3.1.5.1. Utilize FPGA Programmable IOE Delay Feature 12.3.1.5.2. Configure External PHY RGMII-Internal Delay (RGMII-ID) 12.3.1.5.3. Introduce Board Trace Delay
13. System Interconnect and Firewalls x
13.1. System Interconnect and Firewalls Differences Among Intel SoC Device Families 13.2. System Interconnect and Firewalls Features 13.3. System Interconnect and Firewalls System Integration 13.4. System Interconnect and Firewalls Functional Description 13.5. System Interconnect and Firewalls Address Map and Register Definitions
13.4. System Interconnect and Firewalls Functional Description x
13.4.1. Initiator and Target Connectivity 13.4.2. Firewall and Security 13.4.3. Transaction Privilege 13.4.4. Arbitration and Quality-of-Service 13.4.5. Observation Network 13.4.6. System Interconnect Clocks 13.4.7. System Interconnect Resets
13.4.2. Firewall and Security x
13.4.2.1. Initiator Firewall and Security 13.4.2.2. Target Firewall and Security 13.4.2.3. F2H Bridge Firewall 13.4.2.4. MPFE Firewalls 13.4.2.5. Other Firewalls
13.4.2.4. MPFE Firewalls x
13.4.2.4.1. CCU-to-DDR Firewalls (MPU Firewall and F2H Firewall) 13.4.2.4.2. Initiator-to-CSR Firewall 13.4.2.4.3. F2SDRAM Firewall 13.4.2.4.4. MPFE Firewall Example
13.4.4. Arbitration and Quality-of-Service x
13.4.4.1. Priority Levels 13.4.4.2. Urgency and Pressure Signals 13.4.4.3. Passing QOS across NoCs 13.4.4.4. QoS Generators 13.4.4.5. Configuring the Quality of Service Logic 13.4.4.6. Programming QoS Limiter Mode 13.4.4.7. Programming QoS Regulator Mode 13.4.4.8. Programming QoS Fixed Mode 13.4.4.9. Bandwidth and Saturation 13.4.4.10. QoS Programming Examples
13.4.4.10. QoS Programming Examples x
13.4.4.10.1. Example: One Initiator Always Takes Precedence 13.4.4.10.2. Example: Tuning for Specific Throughput Requirements
13.4.5. Observation Network x
13.4.5.1. Interconnect Probes Generators 13.4.5.2. Packet Tracing, Profiling, Statistics, Alarms, and Error Logs
14. Error Checking and Correction Controller x
14.1. ECC Differences Among Intel SoC Device Families 14.2. ECC Controller Use Cases 14.3. ECC Controller Features 14.4. ECC Supported Memories 14.5. ECC Controller System Integration 14.6. ECC Controller Functional Description 14.7. ECC Controller Address Map and Register Definitions
14.6. ECC Controller Functional Description x
14.6.1. ECC Structure 14.6.2. Memory Data Initialization 14.6.3. Indirect Memory Access 14.6.4. Error Checking and Correction Algorithm 14.6.5. Error Logging 14.6.6. ECC Controller Interrupts 14.6.7. ECC Controller Initialization and Configuration 14.6.8. ECC Controller Clocks 14.6.9. ECC Controller Reset
14.6.1. ECC Structure x
14.6.1.1. RAM and ECC Memory Organization Example
14.6.3. Indirect Memory Access x
14.6.3.1. Watchdog Timer 14.6.3.2. Data Correction 14.6.3.3. Error Injection 14.6.3.4. Memory Testing
14.6.3.4. Memory Testing x
14.6.3.4.1. Register Interface Tests
14.6.3.4.1. Register Interface Tests x
14.6.3.4.1.1. Single-Bit Error Test for Word-Writeable Memories 14.6.3.4.1.2. Double-Bit Error Test
14.6.5. Error Logging x
14.6.5.1. Recent Error Address Registers 14.6.5.2. Single-Bit Error Occurrence 14.6.5.3. Single-Bit Error Look-Up Table
14.6.6. ECC Controller Interrupts x
14.6.6.1. Single-Bit Error Interrupts 14.6.6.2. Double-Bit Error Interrupt 14.6.6.3. Interrupt Testing 14.6.6.4. Interrupt Testing
14.6.6.1. Single-Bit Error Interrupts x
14.6.6.1.1. All Single-Bit Error Interrupt 14.6.6.1.2. LUT Overflow Interrupt 14.6.6.1.3. Counter Match Interrupt
15. CoreSight Debug and Trace x
15.1. CoreSight* Debug and Trace Differences Among Intel SoC Device Families 15.2. CoreSight Debug and Trace Use Cases 15.3. CoreSight Debug and Trace Features 15.4. CoreSight* Debug and Trace System Integration 15.5. CoreSight Debug and Trace Functional Description 15.6. CoreSight Debug and Trace Programming Model 15.7. CoreSight* Debug and Trace Address Map and Register Definitions
15.2. CoreSight Debug and Trace Use Cases x
15.2.1. CPU Static Debug 15.2.2. CPU Trace using Embedded Trace Macrocell
15.3. CoreSight Debug and Trace Features x
15.3.1. Debug 15.3.2. Trace
15.3.1. Debug x
15.3.1.1. CPU Performance Events 15.3.1.2. Cross Triggers 15.3.1.3. Fabric Coresight APB
15.3.2. Trace x
15.3.2.1. System Trace Macrocell 15.3.2.2. NoC Trace 15.3.2.3. Fabric Coresight Trace
15.3.2.3. Fabric Coresight Trace x
15.3.2.3.1. Fabric Trace using CoreSight Trace Network 15.3.2.3.2. Timestamp Correlation using ATB SYNCREQ 15.3.2.3.3. Timestamp Correlation using STM HWEVENT
15.5. CoreSight Debug and Trace Functional Description x
15.5.1. Debug APB 15.5.2. Trace Subsystem 15.5.3. Cross Triggers 15.5.4. Timestamp 15.5.5. Embedded Cross Trigger System 15.5.6. HPS Debug APB Interface 15.5.7. FPGA Interface 15.5.8. Interrupts
15.5.1. Debug APB x
15.5.1.1. ROM Tables & Topology Detection 15.5.1.2. DAP SWJ-DP JTAG I/O Connectivity 15.5.1.3. DAP SWJ-DP Debug Clock Enable Request 15.5.1.4. Debug Reset 15.5.1.5. DAP SWJ-DP TargetID 15.5.1.6. DAP TDO Output to SDM in Daisy Chain Mode
15.5.1.4. Debug Reset x
15.5.1.4.1. Debug Reset Request via Reset Manager CSR 15.5.1.4.2. Debug Reset Request via DAP SWJ-DP 15.5.1.4.3. Debug Reset Limitation
15.5.2. Trace Subsystem x
15.5.2.1. Embedded Trace Macrocell (ETM) 15.5.2.2. System Trace Macrocell (STM) 15.5.2.3. PSS NOC and MPFE NOC 15.5.2.4. AMBA* Trace Bus (ATB) 15.5.2.5. Trace Port Interface Unit (TPIU) 15.5.2.6. Embedded Trace FIFO (ETF) 15.5.2.7. Embedded Trace Router (ETR) 15.5.2.8. ATB IDs 15.5.2.9. NOC Trace Observability 15.5.2.10. STM HWEVENT Connectivity
15.5.2.9. NOC Trace Observability x
15.5.2.9.1. Error Probes 15.5.2.9.2. Packet Probes 15.5.2.9.3. Transaction Probes
15.5.3. Cross Triggers x
15.5.3.1. CTI-GT 15.5.3.2. CTI-NOC 15.5.3.3. CTI-5 15.5.3.4. CTI-FPGA 15.5.3.5. CTI-MPU
15.5.4. Timestamp x
15.5.4.1. Timestamp Interpolater
15.5.5. Embedded Cross Trigger System x
15.5.5.1. Cross Trigger Interface 15.5.5.2. Cross Trigger Matrix
15.5.7. FPGA Interface x
15.5.7.1. DAP 15.5.7.2. System Trace Macrocell (STM) 15.5.7.3. CTI-FPGA 15.5.7.4. Trace Port Interface Unit (TPIU) 15.5.7.5. CoreSight* Debug and Trace Clocks 15.5.7.6. CoreSight* Debug and Trace Resets
15.6. CoreSight Debug and Trace Programming Model x
15.6.1. CoreSight Component Address 15.6.2. CTI Trigger Connections to Outside the Debug System 15.6.3. Configuring Embedded Cross-Trigger Connections
15.6.2. CTI Trigger Connections to Outside the Debug System x
15.6.2.1. CTI 15.6.2.2. FPGA-CTI 15.6.2.3. L3-CTI
15.6.3. Configuring Embedded Cross-Trigger Connections x
15.6.3.1. Configuring Trigger Input 0 15.6.3.2. Triggering a Flush of Trace Data to the TPIU 15.6.3.3. Triggering an STM Message 15.6.3.4. Triggering a Breakpoint on CPU 1
A. Appendix x
A.1. Booting and Configuration A.2. HPS Use of SDM QSPI Controller A.3. Security A.4. Operational Status of the HPS to the FPGA Logic
A.1. Booting and Configuration x
A.1.1. Booting and Configuration Features A.1.2. Booting and Configuration FPGA Configuration First Mode Overview A.1.3. Booting and Configuration HPS Boot First Mode Overview A.1.4. Booting and Configuration Device Response to External Configuration and Reset Events
A.2. HPS Use of SDM QSPI Controller x
A.2.1. HPS Use of SDM QSPI Controller Differences Among Intel SoC Device Families A.2.2. HPS Use of SDM QSPI Controller Use Cases A.2.3. HPS Use of SDM QSPI Controller Features A.2.4. HPS Use of SDM QSPI Controller System Integration A.2.5. HPS Use of SDM QSPI Controller Signal Description A.2.6. HPS Use of SDM QSPI Controller Functional Description A.2.7. HPS Use of SDM QSPI Controller Programming Model A.2.8. HPS Use of SDM QSPI Controller Address Map and Register Definitions A.2.9. HPS Use of SDM QSPI Controller Design Guidelines and Examples
A.2.6. HPS Use of SDM QSPI Controller Functional Description x
A.2.6.1. Data Target Interface A.2.6.2. Register Target Interface A.2.6.3. Local Memory Buffer A.2.6.4. Arbitration between Direct/Indirect Access Controller and STIG A.2.6.5. Configuring the Flash Device A.2.6.6. Write Protection A.2.6.7. Data Target Sequential Access Detection A.2.6.8. Clocks A.2.6.9. Resets A.2.6.10. Interrupts
A.2.6.1. Data Target Interface x
A.2.6.1.1. Direct Access Mode A.2.6.1.2. Indirect Access Mode
A.2.6.1.1. Direct Access Mode x
A.2.6.1.1.1. Data Target Remapping Example A.2.6.1.1.2. AHB
A.2.6.1.2. Indirect Access Mode x
A.2.6.1.2.1. Indirect Read Operation A.2.6.1.2.2. Indirect Write Operation A.2.6.1.2.3. Consecutive Reads and Writes
A.2.6.2. Register Target Interface x
A.2.6.2.1. STIG Operation
A.2.7. HPS Use of SDM QSPI Controller Programming Model x
A.2.7.1. Setting Up the QSPI Flash Controller A.2.7.2. Indirect Read Operation A.2.7.3. Indirect Write Operation
A.2.9. HPS Use of SDM QSPI Controller Design Guidelines and Examples x
A.2.9.1. Ownership and Control of Quad SPI Controller A.2.9.2. Using a Single Flash for both FPGA Configuration and HPS Mass Storage A.2.9.3. Pin Features and Connections for SDM QSPI
A.3. Security x
A.3.1. Security Differences Among Intel SoC Device Families A.3.2. ARM Security ISA A.3.3. Secure Device Manager A.3.4. Configuration Bitstream A.3.5. HPS Secure Boot A.3.6. HPS Debug A.3.7. Available Security Documentation
A.4. Operational Status of the HPS to the FPGA Logic x
A.4.1. Overview A.4.2. Software Requirements A.4.3. Signal Behavior Event Diagrams A.4.4. SDM Gated Signals
A.4.1. Overview x
A.4.1.1. SDM Gated Signals A.4.1.2. HPS Events A.4.1.3. FPGA Events
A.4.2. Software Requirements x
A.4.2.1. FPGA Boot First Mode A.4.2.2. HPS Boot First Mode
A.4.3. Signal Behavior Event Diagrams x
A.4.3.1. FPGA User Mode Entry A.4.3.2. HPS Warm Reset Event A.4.3.3. HPS Cold Reset Event A.4.3.4. Watchdog Causes HPS Warm Reset Event A.4.3.5. Watchdog Causes HPS Cold Reset Event
1. Agilex™ 5 Hard Processor System Technical Reference Manual Revision History
2. Introduction to the Hard Processor System
3. Micro Processor Unit (MPU)
4. Application Processor Subsystem
5. Peripheral Subsystem
6. System Manager
7. Clock Manager
8. Reset Manager
9. Power Management
10. Address Map
11. Bridges
12. Interfaces
13. System Interconnect and Firewalls
14. Error Checking and Correction Controller
15. CoreSight Debug and Trace
16. HPS Register Map
A. Appendix

5.1.6.10.1. Enhancements to Scheduled Traffic

The IEEE 802.1Qbv-2015 standard defines the schedule for each of the queues on every egress port which makes the implementation aware of traffic arrival schedule. This information can be used to block the lower priority traffic from transmission in this time window/slot. This ensures that scheduled traffic is forwarded from sender to receiver through all the network nodes with a deterministic delay.
Figure 78. Time Aware Shaper Implementation for Traffic Scheduling

One of the important requirements to achieve a low latency is to ensure there are no interfering frames during the scheduled windows that are reserved for high priority traffic. The use of scheduled traffic imposes limitations while starting a transmission.

As shown in the following figure, if an interfering frame begins transmission just before the start of a reserved time period, it can extend transmission into the reserved window, and potentially interfere with higher priority traffic. Therefore, a guard band whose size is equal to the largest possible interfering frame is required before the window starts.
Figure 79. Implementing a Guard Band to Avoid Delays Due to Interfering Frames

A larger guard band equates to a less efficient use of network bandwidth. However, with the implementation of IEEE802.1Qbu (frame preemption), this issue is addressed. Frame preemption breaks the interfering frame into smaller fragments. Therefore, the guard band needs to be only as large as the largest possible interfering fragment instead of the largest possible interfering frame.

During the guard band, only the frames that can complete the transmission of the entire frame before the next gate close event are permitted. This ensures that the high priority traffic can always start at the beginning of the window reserved for it.

The following operations are required to implement EST:
  • Implementation of gate control list (GCL)
  • Enforcing gate controls in the scheduler
  • Accounting for gate open (O) duty cycle in the computation of idleSlope (CBS)
Note:
Common EST support definitions:
  • Gate control list - The list in the hardware memory that holds the gate controls and the associated time intervals.
  • Gate controls - For a given schedule (row in gate control list) there is a gate open (O) and gate close (C) state associated with each traffic class (TC). The set of O or C values, whose width is same as configured TCs is called the gate controls. As an example, CCOCOOCO means TC7=C, TC6=C, TC5=O and so on.
  • Time interval - Time interval (in nano seconds) is a 16, 20, or 24-bit configurable field in the gate control list that indicates the time for which the associated gate controls are valid.
  • Base time register - Each gate control list is associated with a 64-bit base time register that holds the start time (in PTP format) for the list.

Implementation of Gate Control List

The gate control list (GCL) has the following two parts:
  • Time interval - Defines the time in nanoseconds for which the gate controls are valid and must be applied before reading the next gate controls from the list. Configurable width of 16, 20 or 24 to represent a maximum of 64 us, 1 ms, or 16 ms schedule interval respectively. Supports left shift of the time interval up to 7 bits to be able to apply a multiplication factor from 1 to 128 ns (in steps of 2^n). The maximum value (post shifting) of this field must be 999,999,999 ns. The time interval (TI) value must be large enough to account for the implementation delays. Any value less than that cannot be accurately implemented. To account for the implementation constraints, the minimum time interval for CGL entry must be equal to the time required to transmit fragment_size + IPG/EIPG + preamble overheads.
  • Gate control - Defines the open (O) represented by logic 1 or close (C) represented by logic 0 state for the gate of each TC.
The following figure shows a block diagram from the IEEE 802.1Qbv specification that illustrates how the gate control list governs the gate close (C) and open (O) events based on the schedule provided for each event.
Figure 80. Block Diagram from IEEE 802.1 Qbv Specification - GCL Governing Gate Close and Open Events
The implementation of GCL consists of the following two gate control lists:
  • HWOL - Hardware owned list which is a list for hardware access.
  • SWOL - Software owned list which is a list for software access.
The access to these lists is mutually exclusive. The ownership of the list is set by hardware in the SWOL field of MTL_EST_Status register. The following table provides the implementation details of the GCL.
Table 160.  External Memory Used for Holding the Two Gate-Control Lists
Gate Control (up to 8 bits) Time Interval (ns) 16, 20, or 24 bits Type
OOCCCCCC T0 HWOL or SWOL
OOOOCCCC T1
CCCCOOCC T2
CCCCCCOO T3
I I
I I
OCOCOCOO  
OOOOCCCC T last
OOOOCCCC T0 SWOL or HWOL
OOOOCCCC T1
OOCCCCCC T2
OOOOCCCC T3
I I
I I
OCOCCCCC  
OOCCCCCC T last

Gate Control List Registers

The specification implements the gate control list registers through indirect addressing using the GCL_Control and GCL_Data registers as shown in the following list and described in the table:

  1. 64-bit Base time register (BTR)
  2. 40-bit Cycle time register(CTR)
  3. m-bit Time extension register (TER)
  4. n-bit List length register (LLR) (n depends of the configured GCL depth)
Table 161.  Gate Control List Registers
Registers Name Description

Base Time Register (BTR)

Base time register is a 64-bit register that holds the start time to execute the GCL. The format of the BTR is same as the PTP format (upper 32-bits holds time in seconds and lower 32 bits hold time in nano seconds).

Once the execution of a given list begins, the implementation can update the value in BTR to indicate the next list execution begin time.

Cycle Time Register (CTR)

Cycle time register is a 40-bit register that holds the time at which the execution of the GCL must be repeated. The cycle time register consists of an 8-bit value in seconds, and a 32-bit value in nanoseconds (similar to the PTP time format with truncated seconds register).

For a given gate control list the start time is Base Time + N * Cycle Time where N is an integer value representing the iteration number starting with 0 for the first iteration. If the gate control list execution takes longer than the cycle time, then the list is truncated at the cycle time and the subsequent loop begins at cycle time.

Time Extension Register (TER) Time extension register is a m-bit (where m = configured time interval width + 7) register that holds the amount of time (in nano seconds) the current gate control list can be extended before switching to the new gate control list. This avoids smaller fragments of the current list being executed before switching to a new list.

List Length Register (LLR)

 List length register is an n-bit register (when n is 7, 8, 9, 10, or 11 for a GCL configured depth of 64, 128, 256, 512, or 1024 respectively) that holds the integer value of the length of the GCL (that is the number of valid rows in each GCL). The processing of the GCL stops after the number of rows read equals to the LLR value.

Transmission Gating Implementation

A bridge or an end station can be enhanced to allow transmission from each TC that is yet to be scheduled relative to a known timescale. To achieve this, a transmission gate is associated with each TC; the state of the transmission gate determines if the queued frames can be selected for transmission.

For a TC, the transmission gate can be in one of the following two states:
  1. OPEN - queued frames are selected for transmission, in accordance with the definition of the transmission selection algorithm associated with the TC.
  2. CLOSED - queued frames are not selected for transmission.

A frame on a traffic class queue is not available for transmission if the transmission gate is in the closed state or if there is insufficient time available to transmit the entirety of that frame before the next gate-close event associated with that queue.

The implementation has visibility to the current schedule of gate controls and the immediate next schedule of the gate controls. So the maximum gate open period does not exceed the sum of the two time intervals. This is because, a frame is selected for transmission only if the gate is currently open and the duration of gate open interval is greater than the time taken to transfer the entire frame.

The implementation must know the frame size before the transmission, so that the transmission overruns can be avoided and only the frames that can complete are scheduled at all times. The implementation adequately compensates for the implementation delays in the data transfer from the buffer to the line by offsetting the current time with all the relevant delays (provided by the CTOV field of MTL_EST_Control register). This ensures that the schedule provided is always accurately implemented at the line.
Note:
  • A 20 byte overhead is added to the packet size to account for the IFG and preamble overheads on the line. Therefore for a 128 byte frame to be transmitted, the gate open window must at the least be able to accommodate 148 bytes.
  • In 100 Mbps mode of operation, the rounding down error is about 0.4%. For example, the gate open duration must be at least 1024 bytes for transmitting 1000 byte frames at a speed of 100 Mbps (20 bytes for line overheads, and 4 bytes for rounding down error)

Idle Slope Computation Updates

When EST is enabled, credit is accumulated only when the gate is open; therefore, the effective data rate of the idleSlope must be increased to reflect the duty cycle for the transmission gate associated with the queue.

The idleSlope is computed based on the GateOpenTime and OperCycleTime values. Program the idleSlope registers (implemented one per CBS enabled TC) based on the following equation. The existing MTL register has sufficient bit width to accommodate the new values for idleSlope.

idleSlope = (operIdleSlope(N) * OperCycle/GateOpenTime)

Installing a New GCL

When a new software programmed GCL is available and must be executed at the new BTR value, the switch to the new GCL can happen in one of the following ways:

  • New Base Time Aligned with Current Schedule - If the choice of cycle time for the new gating cycle is unchanged from the cycle time for the current gating cycle, and if the BTR chosen for the new gating cycle (new BTR) is an integer multiple of the current cycle time (+ current BTR), then the new gating cycle exactly lines up with the old gating cycle, that is, the cycle start times for the new gating cycle is same as they would have been for the old configuration. This could be considered to be the ideal case and allows the new gating cycle to be installed and executed with no timing issues. The implementation completes the execution of an iteration of the current list and switches to the new list at the beginning of the BaseTime listed in the new list.
  • New Base Time Unaligned with Current Schedule - If new CycleTime differs from current CycleTime or new BaseTime is in the future and is not an integer multiple of current CycleTime, then the old and new cycles do not line up; when new BaseTime is reached (when the new configuration is installed and starts to execute), the last old cycle is normally truncated to start the first new cycle. This could be undesirable if it results in a very short last old cycle; arguably it would be better to simply extend the penultimate old cycle by that small amount, rather than starting a very short cycle. The cycle time extension register (related to the current config list) allows this extension of the last old cycle to be done in a defined way; if the last complete old cycle ends normally in less than current cycle time extension (TER) ns before the new base time, then the last complete cycle before new BaseTime is reached is extended so that it ends at new BaseTime.

Impact of Transmit Scheduling Algorithms on EST

When you use EST in isolation, the gate control list manages the final open/close state of the transmit class along with the checks carried out by the transmission selection algorithm in MTL. As the gate controls operate on a predefined repetitive schedule, it is recommended to use strict priority or credit based shaper (CBS) scheduling algorithms.

Other algorithms such as the weighted round robin (WRR), deficit weighted round robin (DWRR), and weighted fair queuing (WFQ) implement masking of the transmit channels based on the current winning transmit channel. The algorithm is applied only among the group of transmit channels that open simultaneously. To ensure transmit channels whose gates are open get priority, these algorithms are modified to treat gate open transmit channels and gate closed transmit channels as separate groups giving priority to gate open transmit channels.

For example, consider 4 transmit channels (TC3, TC2, TC1, TC0) with weights 4:3:2:1; TC3 & TC2 are in open state in one slot, while TC1 & TC0 are in open state in another slot. In this case, the scheduler works as follows:
  1. In the first slot, the TC3 & TC2 are serviced in the ratio of 4:3 for the duration the slot is open.
  2. In the second slot, the transmit channels TC1 & TC0 are serviced in the ratio of 2:1.
  3. Fresh arbitration is started every time a slot is opened.

In other words, the traffic does not get distributed in the intended ratio of 4:3:2:1; but as two groups with different ratios and only for the duration of the slot when the gates are open continuously.

When multiple queues map to one TC, EMAC has a first-stage scheduler (round robin) for the queue to TC selection. When EST is enabled, EMAC uses an enhanced round robin. Based on the open window size of EST, the scheduler masks the round robin for queues whose frames do not fit in the window size. This ensures that the scheduler picks the correct queue in the round robin, considering the frame sizes too.

EST Errors

EST can have the following errors:
  • BTR programming error - This error indicates that the value of the new BTR is less than the current time. EMAC attempts to recover from this error by recursively adding the new cycle time to the new BTR.
  • Constant gate control error - This error occurs when the list length register (LLR) is 1 and the programmed time interval (TI) value after the optional left shifting is more than or equal to the cycle time register (CTR). This implies that the gates are either always closed or always open based on the gate control values. As this is not supported, it is reported as an error.
Related Information
Level Two Title